From 42d4dc3f4e1ec1396371aac89d0dccfdd977191b Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Fri, 29 Apr 2005 07:40:12 -0700 Subject: [PATCH] Add suspend method to cpufreq core In order to properly fix some issues with cpufreq vs. sleep on PowerBooks, I had to add a suspend callback to the pmac_cpufreq driver. I must force a switch to full speed before sleep and I switch back to previous speed on resume. I also added a driver flag to disable the warnings in suspend/resume since it is expected in this case to have different speed (and I want it to fixup the jiffies properly). Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index b30001f..4fc0cb7 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -223,7 +223,7 @@ static inline void adjust_jiffies(unsigned long val, struct cpufreq_freqs *ci) } if ((val == CPUFREQ_PRECHANGE && ci->old < ci->new) || (val == CPUFREQ_POSTCHANGE && ci->old > ci->new) || - (val == CPUFREQ_RESUMECHANGE)) { + (val == CPUFREQ_RESUMECHANGE || val == CPUFREQ_SUSPENDCHANGE)) { loops_per_jiffy = cpufreq_scale(l_p_j_ref, l_p_j_ref_freq, ci->new); dprintk("scaling loops_per_jiffy to %lu for frequency %u kHz\n", loops_per_jiffy, ci->new); } @@ -866,11 +866,90 @@ EXPORT_SYMBOL(cpufreq_get); /** + * cpufreq_suspend - let the low level driver prepare for suspend + */ + +static int cpufreq_suspend(struct sys_device * sysdev, u32 state) +{ + int cpu = sysdev->id; + unsigned int ret = 0; + unsigned int cur_freq = 0; + struct cpufreq_policy *cpu_policy; + + dprintk("resuming cpu %u\n", cpu); + + if (!cpu_online(cpu)) + return 0; + + /* we may be lax here as interrupts are off. Nonetheless + * we need to grab the correct cpu policy, as to check + * whether we really run on this CPU. + */ + + cpu_policy = cpufreq_cpu_get(cpu); + if (!cpu_policy) + return -EINVAL; + + /* only handle each CPU group once */ + if (unlikely(cpu_policy->cpu != cpu)) { + cpufreq_cpu_put(cpu_policy); + return 0; + } + + if (cpufreq_driver->suspend) { + ret = cpufreq_driver->suspend(cpu_policy, state); + if (ret) { + printk(KERN_ERR "cpufreq: suspend failed in ->suspend " + "step on CPU %u\n", cpu_policy->cpu); + cpufreq_cpu_put(cpu_policy); + return ret; + } + } + + + if (cpufreq_driver->flags & CPUFREQ_CONST_LOOPS) + goto out; + + if (cpufreq_driver->get) + cur_freq = cpufreq_driver->get(cpu_policy->cpu); + + if (!cur_freq || !cpu_policy->cur) { + printk(KERN_ERR "cpufreq: suspend failed to assert current " + "frequency is what timing core thinks it is.\n"); + goto out; + } + + if (unlikely(cur_freq != cpu_policy->cur)) { + struct cpufreq_freqs freqs; + + if (!(cpufreq_driver->flags & CPUFREQ_PM_NO_WARN)) + printk(KERN_DEBUG "Warning: CPU frequency is %u, " + "cpufreq assumed %u kHz.\n", + cur_freq, cpu_policy->cur); + + freqs.cpu = cpu; + freqs.old = cpu_policy->cur; + freqs.new = cur_freq; + + notifier_call_chain(&cpufreq_transition_notifier_list, + CPUFREQ_SUSPENDCHANGE, &freqs); + adjust_jiffies(CPUFREQ_SUSPENDCHANGE, &freqs); + + cpu_policy->cur = cur_freq; + } + + out: + cpufreq_cpu_put(cpu_policy); + return 0; +} + +/** * cpufreq_resume - restore proper CPU frequency handling after resume * * 1.) resume CPUfreq hardware support (cpufreq_driver->resume()) * 2.) if ->target and !CPUFREQ_CONST_LOOPS: verify we're in sync - * 3.) schedule call cpufreq_update_policy() ASAP as interrupts are restored. + * 3.) schedule call cpufreq_update_policy() ASAP as interrupts are + * restored. */ static int cpufreq_resume(struct sys_device * sysdev) { @@ -915,7 +994,9 @@ static int cpufreq_resume(struct sys_device * sysdev) cur_freq = cpufreq_driver->get(cpu_policy->cpu); if (!cur_freq || !cpu_policy->cur) { - printk(KERN_ERR "cpufreq: resume failed to assert current frequency is what timing core thinks it is.\n"); + printk(KERN_ERR "cpufreq: resume failed to assert " + "current frequency is what timing core " + "thinks it is.\n"); goto out; } @@ -923,13 +1004,15 @@ static int cpufreq_resume(struct sys_device * sysdev) struct cpufreq_freqs freqs; printk(KERN_WARNING "Warning: CPU frequency is %u, " - "cpufreq assumed %u kHz.\n", cur_freq, cpu_policy->cur); + "cpufreq assumed %u kHz.\n", + cur_freq, cpu_policy->cur); freqs.cpu = cpu; freqs.old = cpu_policy->cur; freqs.new = cur_freq; - notifier_call_chain(&cpufreq_transition_notifier_list, CPUFREQ_RESUMECHANGE, &freqs); + notifier_call_chain(&cpufreq_transition_notifier_list, + CPUFREQ_RESUMECHANGE, &freqs); adjust_jiffies(CPUFREQ_RESUMECHANGE, &freqs); cpu_policy->cur = cur_freq; @@ -945,6 +1028,7 @@ out: static struct sysdev_driver cpufreq_sysdev_driver = { .add = cpufreq_add_dev, .remove = cpufreq_remove_dev, + .suspend = cpufreq_suspend, .resume = cpufreq_resume, }; diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h index 910eca3..f21af06 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -103,6 +103,7 @@ struct cpufreq_policy { #define CPUFREQ_PRECHANGE (0) #define CPUFREQ_POSTCHANGE (1) #define CPUFREQ_RESUMECHANGE (8) +#define CPUFREQ_SUSPENDCHANGE (9) struct cpufreq_freqs { unsigned int cpu; /* cpu nr */ @@ -200,6 +201,7 @@ struct cpufreq_driver { /* optional */ int (*exit) (struct cpufreq_policy *policy); + int (*suspend) (struct cpufreq_policy *policy, u32 state); int (*resume) (struct cpufreq_policy *policy); struct freq_attr **attr; }; @@ -211,7 +213,8 @@ struct cpufreq_driver { #define CPUFREQ_CONST_LOOPS 0x02 /* loops_per_jiffy or other kernel * "constants" aren't affected by * frequency transitions */ - +#define CPUFREQ_PM_NO_WARN 0x04 /* don't warn on suspend/resume speed + * mismatches */ int cpufreq_register_driver(struct cpufreq_driver *driver_data); int cpufreq_unregister_driver(struct cpufreq_driver *driver_data); -- cgit v0.10.2 From c06fec5022ebe014af876da2df4a0eee836e97c8 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Fri, 29 Apr 2005 09:37:07 -0700 Subject: Remove bogus BUG() in kernel/exit.c It's old sanity checking that may have been useful for debugging, but is just bogus these days. Noticed by Mattia Belletti. diff --git a/kernel/exit.c b/kernel/exit.c index 39d35935..93851bc 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -517,8 +517,6 @@ static inline void choose_new_parent(task_t *p, task_t *reaper, task_t *child_re */ BUG_ON(p == reaper || reaper->exit_state >= EXIT_ZOMBIE); p->real_parent = reaper; - if (p->parent == p->real_parent) - BUG(); } static inline void reparent_thread(task_t *p, task_t *father, int traced) -- cgit v0.10.2 From a879cbbb34cbecfa9707fbb6e5a00c503ac1ecb9 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Fri, 29 Apr 2005 09:38:44 -0700 Subject: x86: make traps on 'iret' be debuggable in user space This makes a trap on the 'iret' that returns us to user space cause a nice clean SIGSEGV, instead of just a hard (and silent) exit. That way a debugger can actually try to see what happened, and we also properly notify everybody who might be interested about us being gone. This loses the error code, but tells the debugger what happened with ILL_BADSTK in the siginfo. diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S index 3c73dc8..fe1918c 100644 --- a/arch/i386/kernel/entry.S +++ b/arch/i386/kernel/entry.S @@ -260,11 +260,9 @@ restore_nocheck: .section .fixup,"ax" iret_exc: sti - movl $__USER_DS, %edx - movl %edx, %ds - movl %edx, %es - movl $11,%eax - call do_exit + pushl $0 # no error code + pushl $do_iret_error + jmp error_code .previous .section __ex_table,"a" .align 4 diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 6c0e383..d708194 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -451,6 +451,7 @@ DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS) DO_ERROR(11, SIGBUS, "segment not present", segment_not_present) DO_ERROR(12, SIGBUS, "stack segment", stack_segment) DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0) +DO_ERROR_INFO(32, SIGSEGV, "iret exception", iret_error, ILL_BADSTK, 0) fastcall void do_general_protection(struct pt_regs * regs, long error_code) { -- cgit v0.10.2 From 8443b165f13d21214e5d5495eee7c3bf7f2456bf Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Fri, 29 Apr 2005 21:58:15 +0100 Subject: [PATCH] ARM: 2657/1: export ixp2000_pci_config_addr Patch from Lennert Buytenhek Export ixp2000_pci_config_addr, to be used by the IXDP2800 platform setup code to coordinate booting the master and slave NPU. Signed-off-by: Lennert Buytenhek Signed-off-by: Deepak Saxena Signed-off-by: Russell King diff --git a/arch/arm/mach-ixp2000/pci.c b/arch/arm/mach-ixp2000/pci.c index 831f8ff..36c9a94 100644 --- a/arch/arm/mach-ixp2000/pci.c +++ b/arch/arm/mach-ixp2000/pci.c @@ -37,7 +37,7 @@ static int pci_master_aborts = 0; static int clear_master_aborts(void); -static u32 * +u32 * ixp2000_pci_config_addr(unsigned int bus_nr, unsigned int devfn, int where) { u32 *paddress; diff --git a/include/asm-arm/arch-ixp2000/platform.h b/include/asm-arm/arch-ixp2000/platform.h index 509e44d..901bba6 100644 --- a/include/asm-arm/arch-ixp2000/platform.h +++ b/include/asm-arm/arch-ixp2000/platform.h @@ -121,6 +121,7 @@ unsigned long ixp2000_gettimeoffset(void); struct pci_sys_data; +u32 *ixp2000_pci_config_addr(unsigned int bus, unsigned int devfn, int where); void ixp2000_pci_preinit(void); int ixp2000_pci_setup(int, struct pci_sys_data*); struct pci_bus* ixp2000_pci_scan_bus(int, struct pci_sys_data*); -- cgit v0.10.2 From ae36bf5861e1091dd337f0b475e043ab07d4a937 Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Fri, 29 Apr 2005 21:58:15 +0100 Subject: [PATCH] ARM: 2658/1: start ixp2000 pci memory resource at 0xe0000000 Patch from Lennert Buytenhek On the IXDP2800, the bootloader does an awful job of configuring the PCI bus, so we make linux reconfigure everything. Having a 1:1 pci:phys address mapping generally simplifies everything, so try to allocate PCI addresses from the [e0000000..ffffffff] range, which is the physical address range of the outbound PCI window on the IXP2000. This does not affect any of the other IXP2000 platforms since they all use their bootloader's PCI resource assignment. Signed-off-by: Lennert Buytenhek Signed-off-by: Deepak Saxena Signed-off-by: Russell King diff --git a/arch/arm/mach-ixp2000/pci.c b/arch/arm/mach-ixp2000/pci.c index 36c9a94..3844d5c 100644 --- a/arch/arm/mach-ixp2000/pci.c +++ b/arch/arm/mach-ixp2000/pci.c @@ -208,7 +208,7 @@ ixp2000_pci_preinit(void) * use our own resource space. */ static struct resource ixp2000_pci_mem_space = { - .start = 0x00000000, + .start = 0xe0000000, .end = 0xffffffff, .flags = IORESOURCE_MEM, .name = "PCI Mem Space" -- cgit v0.10.2 From 458a83fa43e83505f9401783ce9ed41b5a8b5591 Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Fri, 29 Apr 2005 21:58:16 +0100 Subject: [PATCH] ARM: 2659/1: do not assign PCI I/O address zero on IXP2000 Patch from Lennert Buytenhek Assigning the address zero to a PCI device BAR causes some part of the PCI subsystem to believe that resource allocation for that BAR failed due to resource conflicts, which will make attempts to enable the device fail. Work around this by assigning I/O addresses starting from 00010000. While we're at it, make the PCI I/O resource end at 0001ffff, since we only have 64k of outbound I/O window on the IXP2000, and we don't do bank switching. Signed-off-by: Lennert Buytenhek Signed-off-by: Deepak Saxena Signed-off-by: Russell King diff --git a/arch/arm/mach-ixp2000/pci.c b/arch/arm/mach-ixp2000/pci.c index 3844d5c..5ff2f27 100644 --- a/arch/arm/mach-ixp2000/pci.c +++ b/arch/arm/mach-ixp2000/pci.c @@ -215,8 +215,8 @@ static struct resource ixp2000_pci_mem_space = { }; static struct resource ixp2000_pci_io_space = { - .start = 0x00000000, - .end = 0xffffffff, + .start = 0x00010000, + .end = 0x0001ffff, .flags = IORESOURCE_IO, .name = "PCI I/O Space" }; -- cgit v0.10.2 From 3a1e501511a1e2c665c566939047794dcf86466b Mon Sep 17 00:00:00 2001 From: "George G. Davis" Date: Fri, 29 Apr 2005 22:08:33 +0100 Subject: [PATCH] ARM: 2655/1: ARM1136 SWP instruction abort handler fix Patch from George G. Davis As noted in http://www.arm.com/linux/patch-2.6.9-arm1.gz, the "Faulty SWP instruction on 1136 doesn't set bit 11 in DFSR." So the v6_early_abort handler does not report the correct rd/wr direction for the SWP instruction which may result in SEGVS or hangs. In order to work around this problem, this patch merely updates the fix contained in the ARM Ltd. patch to use the macroised abort handler fixups. Signed-off-by: George G. Davis Signed-off-by: Russell King diff --git a/arch/arm/mm/abort-ev6.S b/arch/arm/mm/abort-ev6.S index 38b2cbb..8f76f3d 100644 --- a/arch/arm/mm/abort-ev6.S +++ b/arch/arm/mm/abort-ev6.S @@ -1,5 +1,6 @@ #include #include +#include "abort-macro.S" /* * Function: v6_early_abort * @@ -13,11 +14,26 @@ * : sp = pointer to registers * * Purpose : obtain information about current aborted instruction. + * Note: we read user space. This means we might cause a data + * abort here if the I-TLB and D-TLB aren't seeing the same + * picture. Unfortunately, this does happen. We live with it. */ .align 5 ENTRY(v6_early_abort) mrc p15, 0, r1, c5, c0, 0 @ get FSR mrc p15, 0, r0, c6, c0, 0 @ get FAR +/* + * Faulty SWP instruction on 1136 doesn't set bit 11 in DFSR. + * The test below covers all the write situations, including Java bytecodes + */ + bic r1, r1, #1 << 11 | 1 << 10 @ clear bits 11 and 10 of FSR + tst r3, #PSR_J_BIT @ Java? + movne pc, lr + do_thumb_abort + ldreq r3, [r2] @ read aborted ARM instruction + do_ldrd_abort + tst r3, #1 << 20 @ L = 0 -> write + orreq r1, r1, #1 << 11 @ yes. mov pc, lr -- cgit v0.10.2 From 2d2669b62984b8d76b05a6a045390a3250317d21 Mon Sep 17 00:00:00 2001 From: Nicolas Pitre Date: Fri, 29 Apr 2005 22:08:33 +0100 Subject: [PATCH] ARM: 2651/3: kernel helpers for NPTL support Patch from Nicolas Pitre This patch entirely reworks the kernel assistance for NPTL on ARM. In particular this provides an efficient way to retrieve the TLS value and perform atomic operations without any instruction emulation nor special system call. This even allows for pre ARMv6 binaries to be forward compatible with SMP systems without any penalty. The problematic and performance critical operations are performed through segment of kernel provided user code reachable from user space at a fixed address in kernel memory. Those fixed entry points are within the vector page so we basically get it for free as no extra memory page is required and nothing else may be mapped at that location anyway. This is different from (but doesn't preclude) a full blown VDSO implementation, however a VDSO would prevent some assembly tricks with constants that allows for efficient branching to those code segments. And since those code segments only use a few cycles before returning to user code, the overhead of a VDSO far call would add a significant overhead to such minimalistic operations. The ARM_NR_set_tls syscall also changed number. This is done for two reasons: 1) this patch changes the way the TLS value was previously meant to be retrieved, therefore we ensure whatever library using the old way gets fixed (they only exist in private tree at the moment since the NPTL work is still progressing). 2) the previous number was allocated in a range causing an undefined instruction trap on kernels not supporting that syscall and it was determined that allocating it in a range returning -ENOSYS would be much nicer for libraries trying to determine if the feature is present or not. Signed-off-by: Nicolas Pitre Signed-off-by: Russell King diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S index 2a5c3fe..080df90 100644 --- a/arch/arm/kernel/entry-armv.S +++ b/arch/arm/kernel/entry-armv.S @@ -269,6 +269,12 @@ __pabt_svc: add r5, sp, #S_PC ldmia r7, {r2 - r4} @ Get USR pc, cpsr +#if __LINUX_ARM_ARCH__ < 6 + @ make sure our user space atomic helper is aborted + cmp r2, #VIRT_OFFSET + bichs r3, r3, #PSR_Z_BIT +#endif + @ @ We are now ready to fill in the remaining blanks on the stack: @ @@ -499,8 +505,12 @@ ENTRY(__switch_to) mra r4, r5, acc0 stmia ip, {r4, r5} #endif +#ifdef CONFIG_HAS_TLS_REG + mcr p15, 0, r3, c13, c0, 3 @ set TLS register +#else mov r4, #0xffff0fff - str r3, [r4, #-3] @ Set TLS ptr + str r3, [r4, #-15] @ TLS val at 0xffff0ff0 +#endif mcr p15, 0, r6, c3, c0, 0 @ Set domain register #ifdef CONFIG_VFP @ Always disable VFP so we can lazily save/restore the old @@ -519,6 +529,207 @@ ENTRY(__switch_to) ldmib r2, {r4 - sl, fp, sp, pc} @ Load all regs saved previously __INIT + +/* + * User helpers. + * + * These are segment of kernel provided user code reachable from user space + * at a fixed address in kernel memory. This is used to provide user space + * with some operations which require kernel help because of unimplemented + * native feature and/or instructions in many ARM CPUs. The idea is for + * this code to be executed directly in user mode for best efficiency but + * which is too intimate with the kernel counter part to be left to user + * libraries. In fact this code might even differ from one CPU to another + * depending on the available instruction set and restrictions like on + * SMP systems. In other words, the kernel reserves the right to change + * this code as needed without warning. Only the entry points and their + * results are guaranteed to be stable. + * + * Each segment is 32-byte aligned and will be moved to the top of the high + * vector page. New segments (if ever needed) must be added in front of + * existing ones. This mechanism should be used only for things that are + * really small and justified, and not be abused freely. + * + * User space is expected to implement those things inline when optimizing + * for a processor that has the necessary native support, but only if such + * resulting binaries are already to be incompatible with earlier ARM + * processors due to the use of unsupported instructions other than what + * is provided here. In other words don't make binaries unable to run on + * earlier processors just for the sake of not using these kernel helpers + * if your compiled code is not going to use the new instructions for other + * purpose. + */ + + .align 5 + .globl __kuser_helper_start +__kuser_helper_start: + +/* + * Reference prototype: + * + * int __kernel_cmpxchg(int oldval, int newval, int *ptr) + * + * Input: + * + * r0 = oldval + * r1 = newval + * r2 = ptr + * lr = return address + * + * Output: + * + * r0 = returned value (zero or non-zero) + * C flag = set if r0 == 0, clear if r0 != 0 + * + * Clobbered: + * + * r3, ip, flags + * + * Definition and user space usage example: + * + * typedef int (__kernel_cmpxchg_t)(int oldval, int newval, int *ptr); + * #define __kernel_cmpxchg (*(__kernel_cmpxchg_t *)0xffff0fc0) + * + * Atomically store newval in *ptr if *ptr is equal to oldval for user space. + * Return zero if *ptr was changed or non-zero if no exchange happened. + * The C flag is also set if *ptr was changed to allow for assembly + * optimization in the calling code. + * + * For example, a user space atomic_add implementation could look like this: + * + * #define atomic_add(ptr, val) \ + * ({ register unsigned int *__ptr asm("r2") = (ptr); \ + * register unsigned int __result asm("r1"); \ + * asm volatile ( \ + * "1: @ atomic_add\n\t" \ + * "ldr r0, [r2]\n\t" \ + * "mov r3, #0xffff0fff\n\t" \ + * "add lr, pc, #4\n\t" \ + * "add r1, r0, %2\n\t" \ + * "add pc, r3, #(0xffff0fc0 - 0xffff0fff)\n\t" \ + * "bcc 1b" \ + * : "=&r" (__result) \ + * : "r" (__ptr), "rIL" (val) \ + * : "r0","r3","ip","lr","cc","memory" ); \ + * __result; }) + */ + +__kuser_cmpxchg: @ 0xffff0fc0 + +#if __LINUX_ARM_ARCH__ < 6 + +#ifdef CONFIG_SMP /* sanity check */ +#error "CONFIG_SMP on a machine supporting pre-ARMv6 processors?" +#endif + + /* + * Theory of operation: + * + * We set the Z flag before loading oldval. If ever an exception + * occurs we can not be sure the loaded value will still be the same + * when the exception returns, therefore the user exception handler + * will clear the Z flag whenever the interrupted user code was + * actually from the kernel address space (see the usr_entry macro). + * + * The post-increment on the str is used to prevent a race with an + * exception happening just after the str instruction which would + * clear the Z flag although the exchange was done. + */ + teq ip, ip @ set Z flag + ldr ip, [r2] @ load current val + add r3, r2, #1 @ prepare store ptr + teqeq ip, r0 @ compare with oldval if still allowed + streq r1, [r3, #-1]! @ store newval if still allowed + subs r0, r2, r3 @ if r2 == r3 the str occured + mov pc, lr + +#else + + ldrex r3, [r2] + subs r3, r3, r0 + strexeq r3, r1, [r2] + rsbs r0, r3, #0 + mov pc, lr + +#endif + + .align 5 + +/* + * Reference prototype: + * + * int __kernel_get_tls(void) + * + * Input: + * + * lr = return address + * + * Output: + * + * r0 = TLS value + * + * Clobbered: + * + * the Z flag might be lost + * + * Definition and user space usage example: + * + * typedef int (__kernel_get_tls_t)(void); + * #define __kernel_get_tls (*(__kernel_get_tls_t *)0xffff0fe0) + * + * Get the TLS value as previously set via the __ARM_NR_set_tls syscall. + * + * This could be used as follows: + * + * #define __kernel_get_tls() \ + * ({ register unsigned int __val asm("r0"); \ + * asm( "mov r0, #0xffff0fff; mov lr, pc; sub pc, r0, #31" \ + * : "=r" (__val) : : "lr","cc" ); \ + * __val; }) + */ + +__kuser_get_tls: @ 0xffff0fe0 + +#ifndef CONFIG_HAS_TLS_REG + +#ifdef CONFIG_SMP /* sanity check */ +#error "CONFIG_SMP without CONFIG_HAS_TLS_REG is wrong" +#endif + + ldr r0, [pc, #(16 - 8)] @ TLS stored at 0xffff0ff0 + mov pc, lr + +#else + + mrc p15, 0, r0, c13, c0, 3 @ read TLS register + mov pc, lr + +#endif + + .rep 5 + .word 0 @ pad up to __kuser_helper_version + .endr + +/* + * Reference declaration: + * + * extern unsigned int __kernel_helper_version; + * + * Definition and user space usage example: + * + * #define __kernel_helper_version (*(unsigned int *)0xffff0ffc) + * + * User space may read this to determine the curent number of helpers + * available. + */ + +__kuser_helper_version: @ 0xffff0ffc + .word ((__kuser_helper_end - __kuser_helper_start) >> 5) + + .globl __kuser_helper_end +__kuser_helper_end: + + /* * Vector stubs. * diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c index 0078aeb..3a001fe 100644 --- a/arch/arm/kernel/traps.c +++ b/arch/arm/kernel/traps.c @@ -450,13 +450,17 @@ asmlinkage int arm_syscall(int no, struct pt_regs *regs) case NR(set_tls): thread->tp_value = regs->ARM_r0; +#ifdef CONFIG_HAS_TLS_REG + asm ("mcr p15, 0, %0, c13, c0, 3" : : "r" (regs->ARM_r0) ); +#else /* - * Our user accessible TLS ptr is located at 0xffff0ffc. - * On SMP read access to this address must raise a fault - * and be emulated from the data abort handler. - * m + * User space must never try to access this directly. + * Expect your app to break eventually if you do so. + * The user helper at 0xffff0fe0 must be used instead. + * (see entry-armv.S for details) */ - *((unsigned long *)0xffff0ffc) = thread->tp_value; + *((unsigned int *)0xffff0ff0) = regs->ARM_r0; +#endif return 0; default: @@ -493,6 +497,41 @@ asmlinkage int arm_syscall(int no, struct pt_regs *regs) return 0; } +#if defined(CONFIG_CPU_32v6) && !defined(CONFIG_HAS_TLS_REG) + +/* + * We might be running on an ARMv6+ processor which should have the TLS + * register, but for some reason we can't use it and have to emulate it. + */ + +static int get_tp_trap(struct pt_regs *regs, unsigned int instr) +{ + int reg = (instr >> 12) & 15; + if (reg == 15) + return 1; + regs->uregs[reg] = current_thread_info()->tp_value; + regs->ARM_pc += 4; + return 0; +} + +static struct undef_hook arm_mrc_hook = { + .instr_mask = 0x0fff0fff, + .instr_val = 0x0e1d0f70, + .cpsr_mask = PSR_T_BIT, + .cpsr_val = 0, + .fn = get_tp_trap, +}; + +static int __init arm_mrc_hook_init(void) +{ + register_undef_hook(&arm_mrc_hook); + return 0; +} + +late_initcall(arm_mrc_hook_init); + +#endif + void __bad_xchg(volatile void *ptr, int size) { printk("xchg: bad data size: pc 0x%p, ptr 0x%p, size %d\n", @@ -580,14 +619,17 @@ void __init trap_init(void) { extern char __stubs_start[], __stubs_end[]; extern char __vectors_start[], __vectors_end[]; + extern char __kuser_helper_start[], __kuser_helper_end[]; + int kuser_sz = __kuser_helper_end - __kuser_helper_start; /* - * Copy the vectors and stubs (in entry-armv.S) into the - * vector page, mapped at 0xffff0000, and ensure these are - * visible to the instruction stream. + * Copy the vectors, stubs and kuser helpers (in entry-armv.S) + * into the vector page, mapped at 0xffff0000, and ensure these + * are visible to the instruction stream. */ memcpy((void *)0xffff0000, __vectors_start, __vectors_end - __vectors_start); memcpy((void *)0xffff0200, __stubs_start, __stubs_end - __stubs_start); + memcpy((void *)0xffff1000 - kuser_sz, __kuser_helper_start, kuser_sz); flush_icache_range(0xffff0000, 0xffff0000 + PAGE_SIZE); modify_domain(DOMAIN_USER, DOMAIN_CLIENT); } diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig index 5b670c9..007766a 100644 --- a/arch/arm/mm/Kconfig +++ b/arch/arm/mm/Kconfig @@ -409,3 +409,17 @@ config CPU_BPREDICT_DISABLE depends on CPU_ARM1020 help Say Y here to disable branch prediction. If unsure, say N. + +config HAS_TLS_REG + bool + depends on CPU_32v6 && !CPU_32v5 && !CPU_32v4 && !CPU_32v3 + help + This selects support for the CP15 thread register. + It is defined to be available on ARMv6 or later. However + if the kernel is configured to support multiple CPUs including + a pre-ARMv6 processors, or if a given ARMv6 processor doesn't + implement the thread register for some reason, then access to + this register from user space must be trapped and emulated. + If user space is relying on the __kuser_get_tls code then + there should not be any impact. + diff --git a/include/asm-arm/unistd.h b/include/asm-arm/unistd.h index a19ec09..ace2748 100644 --- a/include/asm-arm/unistd.h +++ b/include/asm-arm/unistd.h @@ -359,8 +359,7 @@ #define __ARM_NR_cacheflush (__ARM_NR_BASE+2) #define __ARM_NR_usr26 (__ARM_NR_BASE+3) #define __ARM_NR_usr32 (__ARM_NR_BASE+4) - -#define __ARM_NR_set_tls (__ARM_NR_BASE+0x800) +#define __ARM_NR_set_tls (__ARM_NR_BASE+5) #define __sys2(x) #x #define __sys1(x) __sys2(x) -- cgit v0.10.2 From 05f9869bf20e11bcb9b64b9ebd6a9cf89d6b71ba Mon Sep 17 00:00:00 2001 From: Olav Kongas Date: Fri, 29 Apr 2005 22:08:34 +0100 Subject: [PATCH] ARM: 2649/1: Fix 'sparse -Wbitwise' warnings from MMIO macros Patch from Olav Kongas On ARM, the outX() and writeX() families of macros take the result of cpu_to_leYY(), which is of restricted type __leYY, and feed it to __raw_writeX(), which expect an argument of unrestricted type. This results in 'sparse -Wbitwise' warnings about incorrect types in assignments. Analogous type mismatch warnings are issued for inX() and readX() counterparts. The below patch resolves these warnings by adding forced typecasts. Signed-off-by: Olav Kongas Signed-off-by: Russell King diff --git a/include/asm-arm/io.h b/include/asm-arm/io.h index 69bc7a3..658ffa3 100644 --- a/include/asm-arm/io.h +++ b/include/asm-arm/io.h @@ -99,12 +99,16 @@ extern void __readwrite_bug(const char *fn); */ #ifdef __io #define outb(v,p) __raw_writeb(v,__io(p)) -#define outw(v,p) __raw_writew(cpu_to_le16(v),__io(p)) -#define outl(v,p) __raw_writel(cpu_to_le32(v),__io(p)) +#define outw(v,p) __raw_writew((__force __u16) \ + cpu_to_le16(v),__io(p)) +#define outl(v,p) __raw_writel((__force __u32) \ + cpu_to_le32(v),__io(p)) -#define inb(p) ({ unsigned int __v = __raw_readb(__io(p)); __v; }) -#define inw(p) ({ unsigned int __v = le16_to_cpu(__raw_readw(__io(p))); __v; }) -#define inl(p) ({ unsigned int __v = le32_to_cpu(__raw_readl(__io(p))); __v; }) +#define inb(p) ({ __u8 __v = __raw_readb(__io(p)); __v; }) +#define inw(p) ({ __u16 __v = le16_to_cpu((__force __le16) \ + __raw_readw(__io(p))); __v; }) +#define inl(p) ({ __u32 __v = le32_to_cpu((__force __le32) \ + __raw_readl(__io(p))); __v; }) #define outsb(p,d,l) __raw_writesb(__io(p),d,l) #define outsw(p,d,l) __raw_writesw(__io(p),d,l) @@ -149,9 +153,11 @@ extern void _memset_io(void __iomem *, int, size_t); * IO port primitives for more information. */ #ifdef __mem_pci -#define readb(c) ({ unsigned int __v = __raw_readb(__mem_pci(c)); __v; }) -#define readw(c) ({ unsigned int __v = le16_to_cpu(__raw_readw(__mem_pci(c))); __v; }) -#define readl(c) ({ unsigned int __v = le32_to_cpu(__raw_readl(__mem_pci(c))); __v; }) +#define readb(c) ({ __u8 __v = __raw_readb(__mem_pci(c)); __v; }) +#define readw(c) ({ __u16 __v = le16_to_cpu((__force __le16) \ + __raw_readw(__mem_pci(c))); __v; }) +#define readl(c) ({ __u32 __v = le32_to_cpu((__force __le32) \ + __raw_readl(__mem_pci(c))); __v; }) #define readb_relaxed(addr) readb(addr) #define readw_relaxed(addr) readw(addr) #define readl_relaxed(addr) readl(addr) @@ -161,8 +167,10 @@ extern void _memset_io(void __iomem *, int, size_t); #define readsl(p,d,l) __raw_readsl(__mem_pci(p),d,l) #define writeb(v,c) __raw_writeb(v,__mem_pci(c)) -#define writew(v,c) __raw_writew(cpu_to_le16(v),__mem_pci(c)) -#define writel(v,c) __raw_writel(cpu_to_le32(v),__mem_pci(c)) +#define writew(v,c) __raw_writew((__force __u16) \ + cpu_to_le16(v),__mem_pci(c)) +#define writel(v,c) __raw_writel((__force __u32) \ + cpu_to_le32(v),__mem_pci(c)) #define writesb(p,d,l) __raw_writesb(__mem_pci(p),d,l) #define writesw(p,d,l) __raw_writesw(__mem_pci(p),d,l) -- cgit v0.10.2 From ca315159dfa80a2bfc7d917a717a7ee8d771bdf9 Mon Sep 17 00:00:00 2001 From: "George G. Davis" Date: Fri, 29 Apr 2005 22:08:35 +0100 Subject: [PATCH] ARM: 2656/1: Access permission bits are wrong for kernel XIP sections on ARMv6 Patch from George G. Davis This patch is required for kernel XIP support on ARMv6 machines. It ensures that the access permission bits for kernel XIP section descriptors are APX=1 and AP[1:0]=01, which is Kernel read-only/User no access permissions. Prior to this change, kernel XIP section descriptor access permissions were set to Kernel no access/User no access on ARMv6 machines and the kernel would therefore hang upon entry to userspace when set_fs(USER_DS) was executed. Signed-off-by: Steve Longerbeam Signed-off-by: George G. Davis Signed-off-by: Russell King diff --git a/arch/arm/mm/mm-armv.c b/arch/arm/mm/mm-armv.c index f5a87db8..585dfb8 100644 --- a/arch/arm/mm/mm-armv.c +++ b/arch/arm/mm/mm-armv.c @@ -411,9 +411,10 @@ static void __init build_mem_type_table(void) mem_types[MT_MEMORY].prot_sect &= ~PMD_BIT4; mem_types[MT_ROM].prot_sect &= ~PMD_BIT4; /* - * Mark cache clean areas read only from SVC mode - * and no access from userspace. + * Mark cache clean areas and XIP ROM read only + * from SVC mode and no access from userspace. */ + mem_types[MT_ROM].prot_sect |= PMD_SECT_APX|PMD_SECT_AP_WRITE; mem_types[MT_MINICLEAN].prot_sect |= PMD_SECT_APX|PMD_SECT_AP_WRITE; mem_types[MT_CACHECLEAN].prot_sect |= PMD_SECT_APX|PMD_SECT_AP_WRITE; } -- cgit v0.10.2 From 53e173f62c318e65e6ae13524b04c5cf38c1bc3c Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Fri, 29 Apr 2005 22:13:57 +0100 Subject: [PATCH] ARM: 2660/2: fix ixdp2800 boot and pci init Patch from Lennert Buytenhek The IXDP2800 is an evalution platform for the IXP2800 processor that has two IXP2800s connected to the same PCI bus. This is problematic as both CPUs will try to configure the PCI bus as they boot linux. Contrary to on the other IXP2000 platforms, the boot loader on the IXDP2800 doesn't configure the PCI bus properly, so we do want the linux instance on one of the CPUs to do that. Making one of the CPUs ignore the PCI bus (and thus act like a pure PCI slave device) is not an option because there is a 82559 NIC on the PCI bus for each of the CPUs. The chosen solution is to have the master CPU configure the PCI bus while the slave is kept in a quiescent state, and then to have the slave CPU scan the PCI bus (without assigning resources) while the master is kept in a quiescent state. After this ritual, the master deletes the slave NIC from its PCI device list, the slave deletes the master NIC from its device list, and (almost) all is well. There's still one little problem: each of the CPUs has a 1G SDRAM BAR, but the IXP2000 only has 512M of outbound PCI memory window. We solve this by hand-assigning the master and slave SDRAM BARs to a location outside each of the IXP's outbound PCI windows, and by having the rest of the BARs autoconfigured in the outbound PCI windows, in the range [e0000000..ffffffff], so that there is a 1:1 pci:phys mapping between them. Even with this patch, a number of issues still remain -- just imagine what happens if one of the CPUs is rebooted, by watchdog or by hand, but the other one isn't. But those issues are not easily fixable given the strange PCI layout of this board and the behavior of the boot loader shipped with the platform. Signed-off-by: Lennert Buytenhek Signed-off-by: Russell King diff --git a/arch/arm/configs/ixdp2800_defconfig b/arch/arm/configs/ixdp2800_defconfig index d36f991..7be3521 100644 --- a/arch/arm/configs/ixdp2800_defconfig +++ b/arch/arm/configs/ixdp2800_defconfig @@ -133,7 +133,7 @@ CONFIG_ALIGNMENT_TRAP=y # CONFIG_ZBOOT_ROM_TEXT=0x0 CONFIG_ZBOOT_ROM_BSS=0x0 -CONFIG_CMDLINE="console=ttyS0,9600 root=/dev/nfs ip=bootp mem=64M@0x0 pci=firmware" +CONFIG_CMDLINE="console=ttyS0,9600 root=/dev/nfs ip=bootp mem=64M@0x0" # CONFIG_XIP_KERNEL is not set # diff --git a/arch/arm/mach-ixp2000/ixdp2800.c b/arch/arm/mach-ixp2000/ixdp2800.c index c4683aa..aec13c7 100644 --- a/arch/arm/mach-ixp2000/ixdp2800.c +++ b/arch/arm/mach-ixp2000/ixdp2800.c @@ -65,19 +65,102 @@ static struct sys_timer ixdp2800_timer = { /************************************************************************* * IXDP2800 PCI *************************************************************************/ +static void __init ixdp2800_slave_disable_pci_master(void) +{ + *IXP2000_PCI_CMDSTAT &= ~(PCI_COMMAND_MASTER | PCI_COMMAND_MEMORY); +} + +static void __init ixdp2800_master_wait_for_slave(void) +{ + volatile u32 *addr; + + printk(KERN_INFO "IXDP2800: waiting for slave NPU to configure " + "its BAR sizes\n"); + + addr = ixp2000_pci_config_addr(0, IXDP2X00_SLAVE_NPU_DEVFN, + PCI_BASE_ADDRESS_1); + do { + *addr = 0xffffffff; + cpu_relax(); + } while (*addr != 0xfe000008); + + addr = ixp2000_pci_config_addr(0, IXDP2X00_SLAVE_NPU_DEVFN, + PCI_BASE_ADDRESS_2); + do { + *addr = 0xffffffff; + cpu_relax(); + } while (*addr != 0xc0000008); + + /* + * Configure the slave's SDRAM BAR by hand. + */ + *addr = 0x40000008; +} + +static void __init ixdp2800_slave_wait_for_master_enable(void) +{ + printk(KERN_INFO "IXDP2800: waiting for master NPU to enable us\n"); + + while ((*IXP2000_PCI_CMDSTAT & PCI_COMMAND_MASTER) == 0) + cpu_relax(); +} + void __init ixdp2800_pci_preinit(void) { printk("ixdp2x00_pci_preinit called\n"); - *IXP2000_PCI_ADDR_EXT = 0x0000e000; + *IXP2000_PCI_ADDR_EXT = 0x0001e000; + + if (!ixdp2x00_master_npu()) + ixdp2800_slave_disable_pci_master(); - *IXP2000_PCI_DRAM_BASE_ADDR_MASK = (0x40000000 - 1) & ~0xfffff; *IXP2000_PCI_SRAM_BASE_ADDR_MASK = (0x2000000 - 1) & ~0x3ffff; + *IXP2000_PCI_DRAM_BASE_ADDR_MASK = (0x40000000 - 1) & ~0xfffff; ixp2000_pci_preinit(); + + if (ixdp2x00_master_npu()) { + /* + * Wait until the slave set its SRAM/SDRAM BAR sizes + * correctly before we proceed to scan and enumerate + * the bus. + */ + ixdp2800_master_wait_for_slave(); + + /* + * We configure the SDRAM BARs by hand because they + * are 1G and fall outside of the regular allocated + * PCI address space. + */ + *IXP2000_PCI_SDRAM_BAR = 0x00000008; + } else { + /* + * Wait for the master to complete scanning the bus + * and assigning resources before we proceed to scan + * the bus ourselves. Set pci=firmware to honor the + * master's resource assignment. + */ + ixdp2800_slave_wait_for_master_enable(); + pcibios_setup("firmware"); + } } -int ixdp2800_pci_setup(int nr, struct pci_sys_data *sys) +/* + * We assign the SDRAM BARs for the two IXP2800 CPUs by hand, outside + * of the regular PCI window, because there's only 512M of outbound PCI + * memory window on each IXP, while we need 1G for each of the BARs. + */ +static void __devinit ixp2800_pci_fixup(struct pci_dev *dev) +{ + if (machine_is_ixdp2800()) { + dev->resource[2].start = 0; + dev->resource[2].end = 0; + dev->resource[2].flags = 0; + } +} +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IXP2800, ixp2800_pci_fixup); + +static int __init ixdp2800_pci_setup(int nr, struct pci_sys_data *sys) { sys->mem_offset = 0x00000000; @@ -129,22 +212,47 @@ static int __init ixdp2800_pci_map_irq(struct pci_dev *dev, u8 slot, u8 pin) } else return IRQ_IXP2000_PCIB; /* Slave NIC interrupt */ } -static void ixdp2800_pci_postinit(void) +static void __init ixdp2800_master_enable_slave(void) { - struct pci_dev *dev; + volatile u32 *addr; - if (ixdp2x00_master_npu()) { - dev = pci_find_slot(1, IXDP2800_SLAVE_ENET_DEVFN); - pci_remove_bus_device(dev); - } else { - dev = pci_find_slot(1, IXDP2800_MASTER_ENET_DEVFN); - pci_remove_bus_device(dev); + printk(KERN_INFO "IXDP2800: enabling slave NPU\n"); + + addr = (volatile u32 *)ixp2000_pci_config_addr(0, + IXDP2X00_SLAVE_NPU_DEVFN, + PCI_COMMAND); + + *addr |= PCI_COMMAND_MASTER; +} +static void __init ixdp2800_master_wait_for_slave_bus_scan(void) +{ + volatile u32 *addr; + + printk(KERN_INFO "IXDP2800: waiting for slave to finish bus scan\n"); + + addr = (volatile u32 *)ixp2000_pci_config_addr(0, + IXDP2X00_SLAVE_NPU_DEVFN, + PCI_COMMAND); + while ((*addr & PCI_COMMAND_MEMORY) == 0) + cpu_relax(); +} + +static void __init ixdp2800_slave_signal_bus_scan_completion(void) +{ + printk(KERN_INFO "IXDP2800: bus scan done, signaling master\n"); + *IXP2000_PCI_CMDSTAT |= PCI_COMMAND_MEMORY; +} + +static void __init ixdp2800_pci_postinit(void) +{ + if (!ixdp2x00_master_npu()) { ixdp2x00_slave_pci_postinit(); + ixdp2800_slave_signal_bus_scan_completion(); } } -struct hw_pci ixdp2800_pci __initdata = { +struct __initdata hw_pci ixdp2800_pci __initdata = { .nr_controllers = 1, .setup = ixdp2800_pci_setup, .preinit = ixdp2800_pci_preinit, @@ -155,8 +263,21 @@ struct hw_pci ixdp2800_pci __initdata = { int __init ixdp2800_pci_init(void) { - if (machine_is_ixdp2800()) + if (machine_is_ixdp2800()) { + struct pci_dev *dev; + pci_common_init(&ixdp2800_pci); + if (ixdp2x00_master_npu()) { + dev = pci_find_slot(1, IXDP2800_SLAVE_ENET_DEVFN); + pci_remove_bus_device(dev); + + ixdp2800_master_enable_slave(); + ixdp2800_master_wait_for_slave_bus_scan(); + } else { + dev = pci_find_slot(1, IXDP2800_MASTER_ENET_DEVFN); + pci_remove_bus_device(dev); + } + } return 0; } -- cgit v0.10.2 From 587897f51fac04988de497a553215fdf41e6d5f6 Mon Sep 17 00:00:00 2001 From: Sascha Hauer Date: Fri, 29 Apr 2005 22:46:40 +0100 Subject: [PATCH] ARM: 2654/1: i.MX UART initialization sets and honors UFCR value Patch from Sascha Hauer This patch adds UCFR_RFDIV setting into i.MX serial driver. This is required, if loader does not fully agree with Linux kernel about UART setup manner. Linux only blindly expected some values until now. This should enable to use even serial ports not recognized by boot-loader as for example third UART found in the bluethoot module. Patch also enables to detect original setup baudrate in more cases. Signed-off-by: Pavel Pisa Signed-off-by: Sascha Hauer Signed-off-by: Russell King diff --git a/drivers/serial/imx.c b/drivers/serial/imx.c index c682c63..01a8726 100644 --- a/drivers/serial/imx.c +++ b/drivers/serial/imx.c @@ -321,18 +321,39 @@ static void imx_break_ctl(struct uart_port *port, int break_state) #define TXTL 2 /* reset default */ #define RXTL 1 /* reset default */ +static int imx_setup_ufcr(struct imx_port *sport, unsigned int mode) +{ + unsigned int val; + unsigned int ufcr_rfdiv; + + /* set receiver / transmitter trigger level. + * RFDIV is set such way to satisfy requested uartclk value + */ + val = TXTL<<10 | RXTL; + ufcr_rfdiv = (imx_get_perclk1() + sport->port.uartclk / 2) / sport->port.uartclk; + + if(!ufcr_rfdiv) + ufcr_rfdiv = 1; + + if(ufcr_rfdiv >= 7) + ufcr_rfdiv = 6; + else + ufcr_rfdiv = 6 - ufcr_rfdiv; + + val |= UFCR_RFDIV & (ufcr_rfdiv << 7); + + UFCR((u32)sport->port.membase) = val; + + return 0; +} + static int imx_startup(struct uart_port *port) { struct imx_port *sport = (struct imx_port *)port; int retval; - unsigned int val; unsigned long flags; - /* set receiver / transmitter trigger level. We assume - * that RFDIV has been set by the arch setup or by the bootloader. - */ - val = (UFCR((u32)sport->port.membase) & UFCR_RFDIV) | TXTL<<10 | RXTL; - UFCR((u32)sport->port.membase) = val; + imx_setup_ufcr(sport, 0); /* disable the DREN bit (Data Ready interrupt enable) before * requesting IRQs @@ -737,9 +758,12 @@ static void __init imx_console_get_options(struct imx_port *sport, int *baud, int *parity, int *bits) { + if ( UCR1((u32)sport->port.membase) | UCR1_UARTEN ) { /* ok, the port was enabled */ unsigned int ucr2, ubir,ubmr, uartclk; + unsigned int baud_raw; + unsigned int ucfr_rfdiv; ucr2 = UCR2((u32)sport->port.membase); @@ -758,9 +782,35 @@ imx_console_get_options(struct imx_port *sport, int *baud, ubir = UBIR((u32)sport->port.membase) & 0xffff; ubmr = UBMR((u32)sport->port.membase) & 0xffff; - uartclk = sport->port.uartclk; - *baud = ((uartclk/16) * (ubir + 1)) / (ubmr + 1); + + ucfr_rfdiv = (UFCR((u32)sport->port.membase) & UFCR_RFDIV) >> 7; + if (ucfr_rfdiv == 6) + ucfr_rfdiv = 7; + else + ucfr_rfdiv = 6 - ucfr_rfdiv; + + uartclk = imx_get_perclk1(); + uartclk /= ucfr_rfdiv; + + { /* + * The next code provides exact computation of + * baud_raw = round(((uartclk/16) * (ubir + 1)) / (ubmr + 1)) + * without need of float support or long long division, + * which would be required to prevent 32bit arithmetic overflow + */ + unsigned int mul = ubir + 1; + unsigned int div = 16 * (ubmr + 1); + unsigned int rem = uartclk % div; + + baud_raw = (uartclk / div) * mul; + baud_raw += (rem * mul + div / 2) / div; + *baud = (baud_raw + 50) / 100 * 100; + } + + if(*baud != baud_raw) + printk(KERN_INFO "Serial: Console IMX rounded baud rate from %d to %d\n", + baud_raw, *baud); } } @@ -787,6 +837,8 @@ imx_console_setup(struct console *co, char *options) else imx_console_get_options(sport, &baud, &parity, &bits); + imx_setup_ufcr(sport, 0); + return uart_set_options(&sport->port, co, baud, parity, bits, flow); } -- cgit v0.10.2 From d5aa207e46ff7ee838683a7d95ecf46fe42a9a56 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sat, 30 Apr 2005 12:19:28 +0100 Subject: [PATCH] ARM: RTC: allow driver methods to return error Allow RTC drivers to return error codes from their read_time or read_alarm methods. Signed-off-by: Russell King diff --git a/arch/arm/common/rtctime.c b/arch/arm/common/rtctime.c index c397e71..72b03f2 100644 --- a/arch/arm/common/rtctime.c +++ b/arch/arm/common/rtctime.c @@ -141,10 +141,10 @@ void rtc_next_alarm_time(struct rtc_time *next, struct rtc_time *now, struct rtc next->tm_sec = alrm->tm_sec; } -static inline void rtc_read_time(struct rtc_ops *ops, struct rtc_time *tm) +static inline int rtc_read_time(struct rtc_ops *ops, struct rtc_time *tm) { memset(tm, 0, sizeof(struct rtc_time)); - ops->read_time(tm); + return ops->read_time(tm); } static inline int rtc_set_time(struct rtc_ops *ops, struct rtc_time *tm) @@ -163,8 +163,7 @@ static inline int rtc_read_alarm(struct rtc_ops *ops, struct rtc_wkalrm *alrm) int ret = -EINVAL; if (ops->read_alarm) { memset(alrm, 0, sizeof(struct rtc_wkalrm)); - ops->read_alarm(alrm); - ret = 0; + ret = ops->read_alarm(alrm); } return ret; } @@ -283,7 +282,9 @@ static int rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd, break; case RTC_RD_TIME: - rtc_read_time(ops, &tm); + ret = rtc_read_time(ops, &tm); + if (ret) + break; ret = copy_to_user(uarg, &tm, sizeof(tm)); if (ret) ret = -EFAULT; @@ -424,15 +425,15 @@ static int rtc_read_proc(char *page, char **start, off_t off, int count, int *eo struct rtc_time tm; char *p = page; - rtc_read_time(ops, &tm); - - p += sprintf(p, - "rtc_time\t: %02d:%02d:%02d\n" - "rtc_date\t: %04d-%02d-%02d\n" - "rtc_epoch\t: %04lu\n", - tm.tm_hour, tm.tm_min, tm.tm_sec, - tm.tm_year + 1900, tm.tm_mon + 1, tm.tm_mday, - rtc_epoch); + if (rtc_read_time(ops, &tm) == 0) { + p += sprintf(p, + "rtc_time\t: %02d:%02d:%02d\n" + "rtc_date\t: %04d-%02d-%02d\n" + "rtc_epoch\t: %04lu\n", + tm.tm_hour, tm.tm_min, tm.tm_sec, + tm.tm_year + 1900, tm.tm_mon + 1, tm.tm_mday, + rtc_epoch); + } if (rtc_read_alarm(ops, &alrm) == 0) { p += sprintf(p, "alrm_time\t: "); diff --git a/arch/arm/mach-integrator/time.c b/arch/arm/mach-integrator/time.c index 20729de..1a844ca 100644 --- a/arch/arm/mach-integrator/time.c +++ b/arch/arm/mach-integrator/time.c @@ -40,25 +40,32 @@ static int integrator_set_rtc(void) return 1; } -static void rtc_read_alarm(struct rtc_wkalrm *alrm) +static int rtc_read_alarm(struct rtc_wkalrm *alrm) { rtc_time_to_tm(readl(rtc_base + RTC_MR), &alrm->time); + return 0; } -static int rtc_set_alarm(struct rtc_wkalrm *alrm) +static inline int rtc_set_alarm(struct rtc_wkalrm *alrm) { unsigned long time; int ret; - ret = rtc_tm_to_time(&alrm->time, &time); + /* + * At the moment, we can only deal with non-wildcarded alarm times. + */ + ret = rtc_valid_tm(&alrm->time); + if (ret == 0) + ret = rtc_tm_to_time(&alrm->time, &time); if (ret == 0) writel(time, rtc_base + RTC_MR); return ret; } -static void rtc_read_time(struct rtc_time *tm) +static int rtc_read_time(struct rtc_time *tm) { rtc_time_to_tm(readl(rtc_base + RTC_DR), tm); + return 0; } /* @@ -69,7 +76,7 @@ static void rtc_read_time(struct rtc_time *tm) * edge of the 1Hz clock, we must write the time one second * in advance. */ -static int rtc_set_time(struct rtc_time *tm) +static inline int rtc_set_time(struct rtc_time *tm) { unsigned long time; int ret; diff --git a/drivers/char/s3c2410-rtc.c b/drivers/char/s3c2410-rtc.c index 8e61be3..ed867db 100644 --- a/drivers/char/s3c2410-rtc.c +++ b/drivers/char/s3c2410-rtc.c @@ -116,7 +116,7 @@ static void s3c2410_rtc_setfreq(int freq) /* Time read/write */ -static void s3c2410_rtc_gettime(struct rtc_time *rtc_tm) +static int s3c2410_rtc_gettime(struct rtc_time *rtc_tm) { unsigned int have_retried = 0; @@ -151,6 +151,8 @@ static void s3c2410_rtc_gettime(struct rtc_time *rtc_tm) rtc_tm->tm_year += 100; rtc_tm->tm_mon -= 1; + + return 0; } @@ -171,7 +173,7 @@ static int s3c2410_rtc_settime(struct rtc_time *tm) return 0; } -static void s3c2410_rtc_getalarm(struct rtc_wkalrm *alrm) +static int s3c2410_rtc_getalarm(struct rtc_wkalrm *alrm) { struct rtc_time *alm_tm = &alrm->time; unsigned int alm_en; @@ -231,6 +233,8 @@ static void s3c2410_rtc_getalarm(struct rtc_wkalrm *alrm) } /* todo - set alrm->enabled ? */ + + return 0; } static int s3c2410_rtc_setalarm(struct rtc_wkalrm *alrm) diff --git a/include/asm-arm/rtc.h b/include/asm-arm/rtc.h index aa7e16b..370dfe7 100644 --- a/include/asm-arm/rtc.h +++ b/include/asm-arm/rtc.h @@ -18,9 +18,9 @@ struct rtc_ops { void (*release)(void); int (*ioctl)(unsigned int, unsigned long); - void (*read_time)(struct rtc_time *); + int (*read_time)(struct rtc_time *); int (*set_time)(struct rtc_time *); - void (*read_alarm)(struct rtc_wkalrm *); + int (*read_alarm)(struct rtc_wkalrm *); int (*set_alarm)(struct rtc_wkalrm *); int (*proc)(char *buf); }; -- cgit v0.10.2 From bb9bffcbef6166cf03385fbcde97c27bc1a5e689 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sat, 30 Apr 2005 13:26:06 +0100 Subject: [PATCH] ARM: PXA I2C: add platform device Add the PXA I2C platform device. Signed-off-by: Russell King diff --git a/arch/arm/mach-pxa/generic.c b/arch/arm/mach-pxa/generic.c index b1575b8..a45aaa1 100644 --- a/arch/arm/mach-pxa/generic.c +++ b/arch/arm/mach-pxa/generic.c @@ -220,6 +220,30 @@ static struct platform_device stuart_device = { .id = 2, }; +static struct resource i2c_resources[] = { + { + .start = 0x40301680, + .end = 0x403016a3, + .flags = IORESOURCE_MEM, + }, { + .start = IRQ_I2C, + .end = IRQ_I2C, + .flags = IORESOURCE_IRQ, + }, +}; + +static struct platform_device i2c_device = { + .name = "pxa2xx-i2c", + .id = 0, + .resource = i2c_resources, + .num_resources = ARRAY_SIZE(i2c_resources), +}; + +void __init pxa_set_i2c_info(struct i2c_pxa_platform_data *info) +{ + i2c_device.dev.platform_data = info; +} + static struct platform_device *devices[] __initdata = { &pxamci_device, &udc_device, @@ -227,6 +251,7 @@ static struct platform_device *devices[] __initdata = { &ffuart_device, &btuart_device, &stuart_device, + &i2c_device, }; static int __init pxa_init(void) -- cgit v0.10.2 From 9747dd6fa98f2983f4dd09cd6dad1fa3d2a4c5f2 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Sat, 30 Apr 2005 10:01:40 -0700 Subject: [PATCH] ppc64: fix 32-bit signal frame back link When the kernel creates a signal frame on the user stack, it puts the old stack pointer value at the beginning so that the signal frame is linked into the chain of stack frames like any other frame. Unfortunately, for 32-bit processes we are writing the old stack pointer as a 64-bit value rather than a 32-bit value, and the process sees that as a null pointer, since it only looks at the first 32 bits, which are zero since ppc is bigendian and the stack pointer is below 4GB. This bug is in SLES9 and RHEL4 too, hence the ccs. This patch fixes the bug by making the signal code write the old stack pointer as a u32 instead of an unsigned long. Signed-off-by: Paul Mackerras Signed-off-by: Linus Torvalds diff --git a/arch/ppc64/kernel/signal32.c b/arch/ppc64/kernel/signal32.c index b0e167d..3c2fa5c 100644 --- a/arch/ppc64/kernel/signal32.c +++ b/arch/ppc64/kernel/signal32.c @@ -657,7 +657,7 @@ static int handle_rt_signal32(unsigned long sig, struct k_sigaction *ka, /* Save user registers on the stack */ frame = &rt_sf->uc.uc_mcontext; - if (put_user(regs->gpr[1], (unsigned long __user *)newsp)) + if (put_user(regs->gpr[1], (u32 __user *)newsp)) goto badframe; if (vdso32_rt_sigtramp && current->thread.vdso_base) { @@ -842,7 +842,7 @@ static int handle_signal32(unsigned long sig, struct k_sigaction *ka, regs->link = (unsigned long) frame->mctx.tramp; } - if (put_user(regs->gpr[1], (unsigned long __user *)newsp)) + if (put_user(regs->gpr[1], (u32 __user *)newsp)) goto badframe; regs->gpr[1] = (unsigned long) newsp; regs->gpr[3] = sig; -- cgit v0.10.2 From 68575476718ed1c6d6ddafeec8310b109e7a7a7f Mon Sep 17 00:00:00 2001 From: Steve French Date: Sat, 30 Apr 2005 11:10:57 -0700 Subject: [PATCH] cifs: append \* properly on ASCII servers For older servers which do not support Unicode Signed-off-by: Steve French (sfrench@us.ibm.com) Signed-off-by: Linus Torvalds diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index 4d24043..95483ba 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -4,7 +4,9 @@ Fix error mapping of the TOO_MANY_LINKS (hardlinks) case. Do not oops if root user kills cifs oplock kernel thread or kills the cifsd thread (NB: killing the cifs kernel threads is not recommended, unmount and rmmod cifs will kill them when they are -no longer needed). +no longer needed). Fix readdir to ASCII servers (ie older servers +which do not support Unicode) and also require asterik. + Version 1.33 ------------ diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index b004fef..741ff0c 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -2451,12 +2451,14 @@ findFirstRetry: name_len += 2; } else { /* BB add check for overrun of SMB buf BB */ name_len = strnlen(searchName, PATH_MAX); - name_len++; /* trailing null */ /* BB fix here and in unicode clause above ie if(name_len > buffersize-header) free buffer exit; BB */ strncpy(pSMB->FileName, searchName, name_len); - pSMB->FileName[name_len] = 0; /* just in case */ + pSMB->FileName[name_len] = '\\'; + pSMB->FileName[name_len+1] = '*'; + pSMB->FileName[name_len+2] = 0; + name_len += 3; } params = 12 + name_len /* includes null */ ; -- cgit v0.10.2 From 9ea1f8f505f6f770bd593e689960ac4f893509b2 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sat, 30 Apr 2005 11:10:58 -0700 Subject: [PATCH] cifs: Update cifs todo list Signed-off-by: Steve French (sfrench@us.ibm.com) Signed-off-by: Linus Torvalds diff --git a/fs/cifs/TODO b/fs/cifs/TODO index 1e8490e..8cc8816 100644 --- a/fs/cifs/TODO +++ b/fs/cifs/TODO @@ -1,4 +1,4 @@ -version 1.32 April 3, 2005 +version 1.34 April 29, 2005 A Partial List of Missing Features ================================== @@ -70,7 +70,15 @@ r) Implement O_DIRECT flag on open (already supported on mount) s) Allow remapping of last remaining character (\) to +0xF000 which (this character is valid for POSIX but not for Windows) -KNOWN BUGS (updated April 3, 2005) +t) Create UID mapping facility so server UIDs can be mapped on a per +mount or a per server basis to client UIDs or nobody if no mapping +exists. This is helpful when Unix extensions are negotiated to +allow better permission checking when UIDs differ on the server +and client. Add new protocol request to the CIFS protocol +standard for asking the server for the corresponding name of a +particular uid. + +KNOWN BUGS (updated April 29, 2005) ==================================== See http://bugzilla.samba.org - search on product "CifsVFS" for current bug list. -- cgit v0.10.2 From 1ddb8a16aa0e60e7fdc48b1f532cf43e692f8fae Mon Sep 17 00:00:00 2001 From: Russell King Date: Sat, 30 Apr 2005 22:39:51 +0100 Subject: [PATCH] ARM: AMBA CLCD: program palette for pseudocolor visuals Signed-off-by: Russell King diff --git a/drivers/video/amba-clcd.c b/drivers/video/amba-clcd.c index 3e386fd..2896a38 100644 --- a/drivers/video/amba-clcd.c +++ b/drivers/video/amba-clcd.c @@ -256,7 +256,7 @@ clcdfb_setcolreg(unsigned int regno, unsigned int red, unsigned int green, convert_bitfield(green, &fb->fb.var.green) | convert_bitfield(red, &fb->fb.var.red); - if (fb->fb.var.bits_per_pixel == 8 && regno < 256) { + if (fb->fb.fix.visual == FB_VISUAL_PSEUDOCOLOR && regno < 256) { int hw_reg = CLCD_PALETTE + ((regno * 2) & ~3); u32 val, mask, newval; -- cgit v0.10.2 From ed562ab12733ab75437b3390427d41fce9da83c3 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sat, 30 Apr 2005 23:28:47 +0100 Subject: [PATCH] ARM: IntegratorCP: 16bpp is RGB565 not RGB555 Signed-off-by: Russell King diff --git a/drivers/video/amba-clcd.c b/drivers/video/amba-clcd.c index 2896a38..321dbe9 100644 --- a/drivers/video/amba-clcd.c +++ b/drivers/video/amba-clcd.c @@ -134,16 +134,16 @@ clcdfb_set_bitfields(struct clcd_fb *fb, struct fb_var_screeninfo *var) break; case 16: var->red.length = 5; - var->green.length = 5; + var->green.length = 6; var->blue.length = 5; if (fb->panel->cntl & CNTL_BGR) { - var->red.offset = 10; + var->red.offset = 11; var->green.offset = 5; var->blue.offset = 0; } else { var->red.offset = 0; var->green.offset = 5; - var->blue.offset = 10; + var->blue.offset = 11; } break; case 32: -- cgit v0.10.2 From 4774e2260cf25c54f2188dd0407676e3af6f1f23 Mon Sep 17 00:00:00 2001 From: Russell King Date: Sat, 30 Apr 2005 23:32:38 +0100 Subject: [PATCH] ARM: IntegratorCP: Fix CLCD MUX selection values The documentation on these values seems to be rather wrong. These values have been determined by mere trial and error. Signed-off-by: Russell King diff --git a/arch/arm/mach-integrator/integrator_cp.c b/arch/arm/mach-integrator/integrator_cp.c index 68e15c3..3b948e8 100644 --- a/arch/arm/mach-integrator/integrator_cp.c +++ b/arch/arm/mach-integrator/integrator_cp.c @@ -420,7 +420,22 @@ static struct clcd_panel vga = { */ static void cp_clcd_enable(struct clcd_fb *fb) { - cm_control(CM_CTRL_LCDMUXSEL_MASK, CM_CTRL_LCDMUXSEL_VGA); + u32 val; + + if (fb->fb.var.bits_per_pixel <= 8) + val = CM_CTRL_LCDMUXSEL_VGA_8421BPP; + else if (fb->fb.var.bits_per_pixel <= 16) + val = CM_CTRL_LCDMUXSEL_VGA_16BPP; + else + val = 0; /* no idea for this, don't trust the docs */ + + cm_control(CM_CTRL_LCDMUXSEL_MASK| + CM_CTRL_LCDEN0| + CM_CTRL_LCDEN1| + CM_CTRL_STATIC1| + CM_CTRL_STATIC2| + CM_CTRL_STATIC| + CM_CTRL_n24BITEN, val); } static unsigned long framesize = SZ_1M; diff --git a/include/asm-arm/arch-integrator/cm.h b/include/asm-arm/arch-integrator/cm.h index d31c1a7..1ab353e 100644 --- a/include/asm-arm/arch-integrator/cm.h +++ b/include/asm-arm/arch-integrator/cm.h @@ -24,9 +24,9 @@ void cm_control(u32, u32); #define CM_CTRL_LCDBIASDN (1 << 10) #define CM_CTRL_LCDMUXSEL_MASK (7 << 11) #define CM_CTRL_LCDMUXSEL_GENLCD (1 << 11) -#define CM_CTRL_LCDMUXSEL_SHARPLCD1 (3 << 11) -#define CM_CTRL_LCDMUXSEL_SHARPLCD2 (4 << 11) -#define CM_CTRL_LCDMUXSEL_VGA (7 << 11) +#define CM_CTRL_LCDMUXSEL_VGA_16BPP (2 << 11) +#define CM_CTRL_LCDMUXSEL_SHARPLCD (3 << 11) +#define CM_CTRL_LCDMUXSEL_VGA_8421BPP (4 << 11) #define CM_CTRL_LCDEN0 (1 << 14) #define CM_CTRL_LCDEN1 (1 << 15) #define CM_CTRL_STATIC1 (1 << 16) -- cgit v0.10.2 From 2cacb3da620a4a93f3a77e1d2c8c06bb3c74bcb0 Mon Sep 17 00:00:00 2001 From: Sam Ravnborg Date: Sat, 30 Apr 2005 16:51:42 -0700 Subject: [PATCH] kbuild/i386: re-introduce dependency on vmlinux for install target, and add kernel_install Removing the dependency on vmlinux for the install target raised a few complaints, so instead a new target i added: kernel_install. kernel_install will install the kernel just like the ordinary install target. The only difference is that install has a dependency on vmlinux, kernel_install does not. Therefore kernel_install is the best choice when accessing the kernel over a NFS mount or as another user. kernel_install is similar to modules_install in the fact that neither does a full kernel compile before performing the install. In this way they are good for root use. Also added back the dependency on vmlinux for the install target so peoples scripts are no longer broken. Signed-off-by: Sam Ravnborg Signed-off-by: Linus Torvalds diff --git a/arch/i386/Makefile b/arch/i386/Makefile index 314c714..04783ce 100644 --- a/arch/i386/Makefile +++ b/arch/i386/Makefile @@ -123,7 +123,7 @@ AFLAGS += $(mflags-y) boot := arch/i386/boot .PHONY: zImage bzImage compressed zlilo bzlilo \ - zdisk bzdisk fdimage fdimage144 fdimage288 install + zdisk bzdisk fdimage fdimage144 fdimage288 install kernel_install all: bzImage @@ -145,8 +145,9 @@ zdisk bzdisk: vmlinux fdimage fdimage144 fdimage288: vmlinux $(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(KBUILD_IMAGE) $@ -install: - $(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(KBUILD_IMAGE) $@ +install: vmlinux +install kernel_install: + $(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(KBUILD_IMAGE) install prepare: include/asm-$(ARCH)/asm_offsets.h CLEAN_FILES += include/asm-$(ARCH)/asm_offsets.h -- cgit v0.10.2 From b3d9ae4b98f6d28481d4d4b768d860a2cfb5805d Mon Sep 17 00:00:00 2001 From: Sam Ravnborg Date: Sat, 30 Apr 2005 16:51:42 -0700 Subject: [PATCH] kbuild/ppc: tell when uimage was not built Tom Rini said: Note that there is still a trivial'ish change to make. When mkimage doesn't exist on the host we should say "uImage not made" or something similar. So I did like Tom asked. Signed-off-by: Sam Ravnborg Signed-off-by: Linus Torvalds diff --git a/arch/ppc/boot/images/Makefile b/arch/ppc/boot/images/Makefile index 774de8e..f850fb0 100644 --- a/arch/ppc/boot/images/Makefile +++ b/arch/ppc/boot/images/Makefile @@ -20,8 +20,9 @@ quiet_cmd_uimage = UIMAGE $@ targets += uImage $(obj)/uImage: $(obj)/vmlinux.gz + $(Q)rm -f $@ $(call if_changed,uimage) - @echo ' Image $@ is ready' + @echo ' Image: $@' $(if $(wildcard $@),'is ready','not made') # Files generated that shall be removed upon make clean clean-files := sImage vmapus vmlinux* miboot* zImage* uImage -- cgit v0.10.2 From e8e6993178344eb348f60f05b16d9dc30db3b9cf Mon Sep 17 00:00:00 2001 From: Sam Ravnborg Date: Sat, 30 Apr 2005 16:51:42 -0700 Subject: [PATCH] kbuild: Set NOSTDINC_FLAGS late to speed up compile (a little) Move definition of NOSTDINC_FLAGS below inclusion of arch Makefile, so any arch specific settings to $(CC) takes effect before looking up the compiler include directory. The previous solution that replaced ':=' with '=' caused gcc to be invoked one additional time for each directory visited. This decreases kernel compile time with 0.1 second (3.6 -> 3.5 seconds) when running make on a fully built kernel Signed-off-by: Sam Ravnborg Signed-off-by: Linus Torvalds diff --git a/Makefile b/Makefile index 8da3a30..4899089 100644 --- a/Makefile +++ b/Makefile @@ -332,9 +332,7 @@ KALLSYMS = scripts/kallsyms PERL = perl CHECK = sparse -NOSTDINC_FLAGS = -nostdinc -isystem $(shell $(CC) -print-file-name=include) CHECKFLAGS := -D__linux__ -Dlinux -D__STDC__ -Dunix -D__unix__ -CHECKFLAGS += $(NOSTDINC_FLAGS) MODFLAGS = -DMODULE CFLAGS_MODULE = $(MODFLAGS) AFLAGS_MODULE = $(MODFLAGS) @@ -531,6 +529,10 @@ endif include $(srctree)/arch/$(ARCH)/Makefile +# arch Makefile may override CC so keep this after arch Makefile is included +NOSTDINC_FLAGS := -nostdinc -isystem $(shell $(CC) -print-file-name=include) +CHECKFLAGS += $(NOSTDINC_FLAGS) + # warn about C99 declaration after statement CFLAGS += $(call cc-option,-Wdeclaration-after-statement,) -- cgit v0.10.2 From 69aa3f71580990f39e387d96ed1001d2f5fb04b1 Mon Sep 17 00:00:00 2001 From: "akpm@osdl.org" Date: Sun, 1 May 2005 08:58:35 -0700 Subject: [PATCH] ultrastor build fix Fix a typo. James Bottomley Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/scsi/ultrastor.c b/drivers/scsi/ultrastor.c index a00095c..97f4d91 100644 --- a/drivers/scsi/ultrastor.c +++ b/drivers/scsi/ultrastor.c @@ -945,7 +945,7 @@ static int ultrastor_abort(Scsi_Cmnd *SCpnt) config.mscp[mscp_index].SCint, SCpnt); #endif if (config.mscp[mscp_index].SCint == 0) - return FAILURE; + return FAILED; if (config.mscp[mscp_index].SCint != SCpnt) panic("Bad abort"); config.mscp[mscp_index].SCint = NULL; -- cgit v0.10.2 From f021e9210185b46e41ec3a0e78ec1621e168eacb Mon Sep 17 00:00:00 2001 From: "akpm@osdl.org" Date: Sun, 1 May 2005 08:58:35 -0700 Subject: [PATCH] generic_file_buffered_write fixes Anton Altaparmakov points out: - It calls fault_in_pages_readable() which is completely bogus if @nr_segs > 1. It needs to be replaced by a to be written "fault_in_pages_readable_iovec()". - It increments @buf even in the iovec case thus @buf can point to random memory really quickly (in the iovec case) and then it calls fault_in_pages_readable() on this random memory. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/filemap.c b/mm/filemap.c index 93595c3..9b74674 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1949,7 +1949,7 @@ generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov, buf = iov->iov_base + written; else { filemap_set_next_iovec(&cur_iov, &iov_base, written); - buf = iov->iov_base + iov_base; + buf = cur_iov->iov_base + iov_base; } do { @@ -2007,9 +2007,11 @@ generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov, count -= status; pos += status; buf += status; - if (unlikely(nr_segs > 1)) + if (unlikely(nr_segs > 1)) { filemap_set_next_iovec(&cur_iov, &iov_base, status); + buf = cur_iov->iov_base + iov_base; + } } } if (unlikely(copied != bytes)) -- cgit v0.10.2 From 119f657c72fc07d6fd28c61de59cfba1566970a9 Mon Sep 17 00:00:00 2001 From: "akpm@osdl.org" Date: Sun, 1 May 2005 08:58:35 -0700 Subject: [PATCH] RLIMIT_AS checking fix Address bug #4508: there's potential for wraparound in the various places where we perform RLIMIT_AS checking. (I'm a bit worried about acct_stack_growth(). Are we sure that vma->vm_mm is always equal to current->mm? If not, then we're comparing some other process's total_vm with the calling process's rlimits). Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/mm.h b/include/linux/mm.h index c74a74c..8b007ad 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -726,6 +726,7 @@ extern void __vma_link_rb(struct mm_struct *, struct vm_area_struct *, extern struct vm_area_struct *copy_vma(struct vm_area_struct **, unsigned long addr, unsigned long len, pgoff_t pgoff); extern void exit_mmap(struct mm_struct *); +extern int may_expand_vm(struct mm_struct *mm, unsigned long npages); extern unsigned long get_unmapped_area(struct file *, unsigned long, unsigned long, unsigned long, unsigned long); diff --git a/mm/mmap.c b/mm/mmap.c index 6ea204c..1ec0f6e 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1009,8 +1009,7 @@ munmap_back: } /* Check against address space limit. */ - if ((mm->total_vm << PAGE_SHIFT) + len - > current->signal->rlim[RLIMIT_AS].rlim_cur) + if (!may_expand_vm(mm, len >> PAGE_SHIFT)) return -ENOMEM; if (accountable && (!(flags & MAP_NORESERVE) || @@ -1421,7 +1420,7 @@ static int acct_stack_growth(struct vm_area_struct * vma, unsigned long size, un struct rlimit *rlim = current->signal->rlim; /* address space limit tests */ - if (mm->total_vm + grow > rlim[RLIMIT_AS].rlim_cur >> PAGE_SHIFT) + if (!may_expand_vm(mm, grow)) return -ENOMEM; /* Stack limit test */ @@ -1848,8 +1847,7 @@ unsigned long do_brk(unsigned long addr, unsigned long len) } /* Check against address space limits *after* clearing old maps... */ - if ((mm->total_vm << PAGE_SHIFT) + len - > current->signal->rlim[RLIMIT_AS].rlim_cur) + if (!may_expand_vm(mm, len >> PAGE_SHIFT)) return -ENOMEM; if (mm->map_count > sysctl_max_map_count) @@ -2019,3 +2017,19 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap, } return new_vma; } + +/* + * Return true if the calling process may expand its vm space by the passed + * number of pages + */ +int may_expand_vm(struct mm_struct *mm, unsigned long npages) +{ + unsigned long cur = mm->total_vm; /* pages */ + unsigned long lim; + + lim = current->signal->rlim[RLIMIT_AS].rlim_cur >> PAGE_SHIFT; + + if (cur + npages > lim) + return 0; + return 1; +} diff --git a/mm/mremap.c b/mm/mremap.c index 0d1c1b9..0dd7ace 100644 --- a/mm/mremap.c +++ b/mm/mremap.c @@ -347,10 +347,10 @@ unsigned long do_mremap(unsigned long addr, if (locked > lock_limit && !capable(CAP_IPC_LOCK)) goto out; } - ret = -ENOMEM; - if ((current->mm->total_vm << PAGE_SHIFT) + (new_len - old_len) - > current->signal->rlim[RLIMIT_AS].rlim_cur) + if (!may_expand_vm(current->mm, (new_len - old_len) >> PAGE_SHIFT)) { + ret = -ENOMEM; goto out; + } if (vma->vm_flags & VM_ACCOUNT) { charged = (new_len - old_len) >> PAGE_SHIFT; -- cgit v0.10.2 From 81b4082dc7666e2bc5ec229d8e837f3bafb96883 Mon Sep 17 00:00:00 2001 From: Nikita Danilov Date: Sun, 1 May 2005 08:58:36 -0700 Subject: [PATCH] mm: rmap.c cleanup mm/rmap.c:page_referenced_one() and mm/rmap.c:try_to_unmap_one() contain identical code that - takes mm->page_table_lock; - drills through page tables; - checks that correct pte is reached. Coalesce this into page_check_address() Signed-off-by: Nikita Danilov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/rmap.c b/mm/rmap.c index 884d6d1..378de23 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -243,6 +243,42 @@ unsigned long page_address_in_vma(struct page *page, struct vm_area_struct *vma) } /* + * Check that @page is mapped at @address into @mm. + * + * On success returns with mapped pte and locked mm->page_table_lock. + */ +static pte_t *page_check_address(struct page *page, struct mm_struct *mm, + unsigned long address) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + + /* + * We need the page_table_lock to protect us from page faults, + * munmap, fork, etc... + */ + spin_lock(&mm->page_table_lock); + pgd = pgd_offset(mm, address); + if (likely(pgd_present(*pgd))) { + pud = pud_offset(pgd, address); + if (likely(pud_present(*pud))) { + pmd = pmd_offset(pud, address); + if (likely(pmd_present(*pmd))) { + pte = pte_offset_map(pmd, address); + if (likely(pte_present(*pte) && + page_to_pfn(page) == pte_pfn(*pte))) + return pte; + pte_unmap(pte); + } + } + } + spin_unlock(&mm->page_table_lock); + return ERR_PTR(-ENOENT); +} + +/* * Subfunctions of page_referenced: page_referenced_one called * repeatedly from either page_referenced_anon or page_referenced_file. */ @@ -251,9 +287,6 @@ static int page_referenced_one(struct page *page, { struct mm_struct *mm = vma->vm_mm; unsigned long address; - pgd_t *pgd; - pud_t *pud; - pmd_t *pmd; pte_t *pte; int referenced = 0; @@ -263,39 +296,18 @@ static int page_referenced_one(struct page *page, if (address == -EFAULT) goto out; - spin_lock(&mm->page_table_lock); - - pgd = pgd_offset(mm, address); - if (!pgd_present(*pgd)) - goto out_unlock; - - pud = pud_offset(pgd, address); - if (!pud_present(*pud)) - goto out_unlock; - - pmd = pmd_offset(pud, address); - if (!pmd_present(*pmd)) - goto out_unlock; - - pte = pte_offset_map(pmd, address); - if (!pte_present(*pte)) - goto out_unmap; - - if (page_to_pfn(page) != pte_pfn(*pte)) - goto out_unmap; - - if (ptep_clear_flush_young(vma, address, pte)) - referenced++; - - if (mm != current->mm && !ignore_token && has_swap_token(mm)) - referenced++; + pte = page_check_address(page, mm, address); + if (!IS_ERR(pte)) { + if (ptep_clear_flush_young(vma, address, pte)) + referenced++; - (*mapcount)--; + if (mm != current->mm && !ignore_token && has_swap_token(mm)) + referenced++; -out_unmap: - pte_unmap(pte); -out_unlock: - spin_unlock(&mm->page_table_lock); + (*mapcount)--; + pte_unmap(pte); + spin_unlock(&mm->page_table_lock); + } out: return referenced; } @@ -502,9 +514,6 @@ static int try_to_unmap_one(struct page *page, struct vm_area_struct *vma) { struct mm_struct *mm = vma->vm_mm; unsigned long address; - pgd_t *pgd; - pud_t *pud; - pmd_t *pmd; pte_t *pte; pte_t pteval; int ret = SWAP_AGAIN; @@ -515,30 +524,9 @@ static int try_to_unmap_one(struct page *page, struct vm_area_struct *vma) if (address == -EFAULT) goto out; - /* - * We need the page_table_lock to protect us from page faults, - * munmap, fork, etc... - */ - spin_lock(&mm->page_table_lock); - - pgd = pgd_offset(mm, address); - if (!pgd_present(*pgd)) - goto out_unlock; - - pud = pud_offset(pgd, address); - if (!pud_present(*pud)) - goto out_unlock; - - pmd = pmd_offset(pud, address); - if (!pmd_present(*pmd)) - goto out_unlock; - - pte = pte_offset_map(pmd, address); - if (!pte_present(*pte)) - goto out_unmap; - - if (page_to_pfn(page) != pte_pfn(*pte)) - goto out_unmap; + pte = page_check_address(page, mm, address); + if (IS_ERR(pte)) + goto out; /* * If the page is mlock()d, we cannot swap it out. @@ -604,7 +592,6 @@ static int try_to_unmap_one(struct page *page, struct vm_area_struct *vma) out_unmap: pte_unmap(pte); -out_unlock: spin_unlock(&mm->page_table_lock); out: return ret; @@ -708,7 +695,6 @@ static void try_to_unmap_cluster(unsigned long cursor, } pte_unmap(pte); - out_unlock: spin_unlock(&mm->page_table_lock); } @@ -860,3 +846,4 @@ int try_to_unmap(struct page *page) ret = SWAP_SUCCESS; return ret; } + -- cgit v0.10.2 From 8e30f272a93ec9c1d5c305c5040dfaebc880499d Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sun, 1 May 2005 08:58:36 -0700 Subject: [PATCH] mm: pcp use non powers of 2 for batch size Jack Steiner reported this to have fixed his problem (bad colouring): "The patches fix both problems that I found - bad coloring & excessive pages in pagesets." In most workloads this is not likely to be such a pronounced problem, however it should help corner cases. And avoiding powers of 2 in these types of memory operations is always a good idea. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/page_alloc.c b/mm/page_alloc.c index c73dbbc..08e8627 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1671,6 +1671,18 @@ static void __init free_area_init_core(struct pglist_data *pgdat, if (batch < 1) batch = 1; + /* + * Clamp the batch to a 2^n - 1 value. Having a power + * of 2 value was found to be more likely to have + * suboptimal cache aliasing properties in some cases. + * + * For example if 2 tasks are alternately allocating + * batches of pages, one task can end up with a lot + * of pages of one half of the possible page colors + * and the other with pages of the other colors. + */ + batch = (1 << fls(batch + batch/2)) - 1; + for (cpu = 0; cpu < NR_CPUS; cpu++) { struct per_cpu_pages *pcp; -- cgit v0.10.2 From b84a35be0285229b0a8a5e2e04d79360c5b75562 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sun, 1 May 2005 08:58:36 -0700 Subject: [PATCH] mempool: NOMEMALLOC and NORETRY Mempools have 2 problems. The first is that mempool_alloc can possibly get stuck in __alloc_pages when they should opt to fail, and take an element from their reserved pool. The second is that it will happily eat emergency PF_MEMALLOC reserves instead of going to their reserved pools. Fix the first by passing __GFP_NORETRY in the allocation calls in mempool_alloc. Fix the second by introducing a __GFP_MEMPOOL flag which directs the page allocator not to allocate from the reserve pool. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/gfp.h b/include/linux/gfp.h index 848a1ba..af7407e 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -38,14 +38,16 @@ struct vm_area_struct; #define __GFP_NO_GROW 0x2000u /* Slab internal usage */ #define __GFP_COMP 0x4000u /* Add compound page metadata */ #define __GFP_ZERO 0x8000u /* Return zeroed page on success */ +#define __GFP_NOMEMALLOC 0x10000u /* Don't use emergency reserves */ -#define __GFP_BITS_SHIFT 16 /* Room for 16 __GFP_FOO bits */ +#define __GFP_BITS_SHIFT 20 /* Room for 20 __GFP_FOO bits */ #define __GFP_BITS_MASK ((1 << __GFP_BITS_SHIFT) - 1) /* if you forget to add the bitmask here kernel will crash, period */ #define GFP_LEVEL_MASK (__GFP_WAIT|__GFP_HIGH|__GFP_IO|__GFP_FS| \ __GFP_COLD|__GFP_NOWARN|__GFP_REPEAT| \ - __GFP_NOFAIL|__GFP_NORETRY|__GFP_NO_GROW|__GFP_COMP) + __GFP_NOFAIL|__GFP_NORETRY|__GFP_NO_GROW|__GFP_COMP| \ + __GFP_NOMEMALLOC) #define GFP_ATOMIC (__GFP_HIGH) #define GFP_NOIO (__GFP_WAIT) diff --git a/mm/mempool.c b/mm/mempool.c index b014ffe..d691b5c 100644 --- a/mm/mempool.c +++ b/mm/mempool.c @@ -198,11 +198,16 @@ void * mempool_alloc(mempool_t *pool, unsigned int __nocast gfp_mask) void *element; unsigned long flags; DEFINE_WAIT(wait); - int gfp_nowait = gfp_mask & ~(__GFP_WAIT | __GFP_IO); + int gfp_nowait; + + gfp_mask |= __GFP_NOMEMALLOC; /* don't allocate emergency reserves */ + gfp_mask |= __GFP_NORETRY; /* don't loop in __alloc_pages */ + gfp_mask |= __GFP_NOWARN; /* failures are OK */ + gfp_nowait = gfp_mask & ~(__GFP_WAIT | __GFP_IO); might_sleep_if(gfp_mask & __GFP_WAIT); repeat_alloc: - element = pool->alloc(gfp_nowait|__GFP_NOWARN, pool->pool_data); + element = pool->alloc(gfp_nowait, pool->pool_data); if (likely(element != NULL)) return element; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 08e8627..04a35b3 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -799,14 +799,18 @@ __alloc_pages(unsigned int __nocast gfp_mask, unsigned int order, } /* This allocation should allow future memory freeing. */ - if (((p->flags & PF_MEMALLOC) || unlikely(test_thread_flag(TIF_MEMDIE))) && !in_interrupt()) { - /* go through the zonelist yet again, ignoring mins */ - for (i = 0; (z = zones[i]) != NULL; i++) { - if (!cpuset_zone_allowed(z)) - continue; - page = buffered_rmqueue(z, order, gfp_mask); - if (page) - goto got_pg; + + if (((p->flags & PF_MEMALLOC) || unlikely(test_thread_flag(TIF_MEMDIE))) + && !in_interrupt()) { + if (!(gfp_mask & __GFP_NOMEMALLOC)) { + /* go through the zonelist yet again, ignoring mins */ + for (i = 0; (z = zones[i]) != NULL; i++) { + if (!cpuset_zone_allowed(z)) + continue; + page = buffered_rmqueue(z, order, gfp_mask); + if (page) + goto got_pg; + } } goto nopage; } -- cgit v0.10.2 From 20a77776c24800d1e40a73f520cfcb32239568a9 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sun, 1 May 2005 08:58:37 -0700 Subject: [PATCH] mempool: simplify alloc Mempool is pretty clever. Looks too clever for its own good :) It shouldn't really know so much about page reclaim internals. - don't guess about what effective page reclaim might involve. - don't randomly flush out all dirty data if some unlikely thing happens (alloc returns NULL). page reclaim can (sort of :P) handle it. I think the main motivation is trying to avoid pool->lock at all costs. However the first allocation is attempted with __GFP_WAIT cleared, so it will be 'can_try_harder' if it hits the page allocator. So if allocation still fails, then we can probably afford to hit the pool->lock - and what's the alternative? Try page reclaim and hit zone->lru_lock? A nice upshot is that we don't need to do any fancy memory barriers or do (intentionally) racy access to pool-> fields outside the lock. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/mempool.c b/mm/mempool.c index d691b5c..e9a0a63 100644 --- a/mm/mempool.c +++ b/mm/mempool.c @@ -198,36 +198,22 @@ void * mempool_alloc(mempool_t *pool, unsigned int __nocast gfp_mask) void *element; unsigned long flags; DEFINE_WAIT(wait); - int gfp_nowait; + int gfp_temp; + + might_sleep_if(gfp_mask & __GFP_WAIT); gfp_mask |= __GFP_NOMEMALLOC; /* don't allocate emergency reserves */ gfp_mask |= __GFP_NORETRY; /* don't loop in __alloc_pages */ gfp_mask |= __GFP_NOWARN; /* failures are OK */ - gfp_nowait = gfp_mask & ~(__GFP_WAIT | __GFP_IO); - might_sleep_if(gfp_mask & __GFP_WAIT); + gfp_temp = gfp_mask & ~(__GFP_WAIT|__GFP_IO); + repeat_alloc: - element = pool->alloc(gfp_nowait, pool->pool_data); + + element = pool->alloc(gfp_temp, pool->pool_data); if (likely(element != NULL)) return element; - /* - * If the pool is less than 50% full and we can perform effective - * page reclaim then try harder to allocate an element. - */ - mb(); - if ((gfp_mask & __GFP_FS) && (gfp_mask != gfp_nowait) && - (pool->curr_nr <= pool->min_nr/2)) { - element = pool->alloc(gfp_mask, pool->pool_data); - if (likely(element != NULL)) - return element; - } - - /* - * Kick the VM at this point. - */ - wakeup_bdflush(0); - spin_lock_irqsave(&pool->lock, flags); if (likely(pool->curr_nr)) { element = remove_element(pool); @@ -240,6 +226,8 @@ repeat_alloc: if (!(gfp_mask & __GFP_WAIT)) return NULL; + /* Now start performing page reclaim */ + gfp_temp = gfp_mask; prepare_to_wait(&pool->wait, &wait, TASK_UNINTERRUPTIBLE); mb(); if (!pool->curr_nr) -- cgit v0.10.2 From bd53b714d32a29bdf33009f812e295667e92b930 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sun, 1 May 2005 08:58:37 -0700 Subject: [PATCH] mm: use __GFP_NOMEMALLOC Use the new __GFP_NOMEMALLOC to simplify the previous handling of PF_MEMALLOC. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index 77619a5..0dd6c2b 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -331,25 +331,19 @@ crypt_alloc_buffer(struct crypt_config *cc, unsigned int size, struct bio *bio; unsigned int nr_iovecs = (size + PAGE_SIZE - 1) >> PAGE_SHIFT; int gfp_mask = GFP_NOIO | __GFP_HIGHMEM; - unsigned long flags = current->flags; unsigned int i; /* - * Tell VM to act less aggressively and fail earlier. - * This is not necessary but increases throughput. + * Use __GFP_NOMEMALLOC to tell the VM to act less aggressively and + * to fail earlier. This is not necessary but increases throughput. * FIXME: Is this really intelligent? */ - current->flags &= ~PF_MEMALLOC; - if (base_bio) - bio = bio_clone(base_bio, GFP_NOIO); + bio = bio_clone(base_bio, GFP_NOIO|__GFP_NOMEMALLOC); else - bio = bio_alloc(GFP_NOIO, nr_iovecs); - if (!bio) { - if (flags & PF_MEMALLOC) - current->flags |= PF_MEMALLOC; + bio = bio_alloc(GFP_NOIO|__GFP_NOMEMALLOC, nr_iovecs); + if (!bio) return NULL; - } /* if the last bio was not complete, continue where that one ended */ bio->bi_idx = *bio_vec_idx; @@ -386,9 +380,6 @@ crypt_alloc_buffer(struct crypt_config *cc, unsigned int size, size -= bv->bv_len; } - if (flags & PF_MEMALLOC) - current->flags |= PF_MEMALLOC; - if (!bio->bi_size) { bio_put(bio); return NULL; diff --git a/mm/swap_state.c b/mm/swap_state.c index a063a90..4f25177 100644 --- a/mm/swap_state.c +++ b/mm/swap_state.c @@ -143,7 +143,6 @@ void __delete_from_swap_cache(struct page *page) int add_to_swap(struct page * page) { swp_entry_t entry; - int pf_flags; int err; if (!PageLocked(page)) @@ -154,29 +153,19 @@ int add_to_swap(struct page * page) if (!entry.val) return 0; - /* Radix-tree node allocations are performing - * GFP_ATOMIC allocations under PF_MEMALLOC. - * They can completely exhaust the page allocator. - * - * So PF_MEMALLOC is dropped here. This causes the slab - * allocations to fail earlier, so radix-tree nodes will - * then be allocated from the mempool reserves. + /* + * Radix-tree node allocations from PF_MEMALLOC contexts could + * completely exhaust the page allocator. __GFP_NOMEMALLOC + * stops emergency reserves from being allocated. * - * We're still using __GFP_HIGH for radix-tree node - * allocations, so some of the emergency pools are available, - * just not all of them. + * TODO: this could cause a theoretical memory reclaim + * deadlock in the swap out path. */ - - pf_flags = current->flags; - current->flags &= ~PF_MEMALLOC; - /* * Add it to the swap cache and mark it dirty */ - err = __add_to_swap_cache(page, entry, GFP_ATOMIC|__GFP_NOWARN); - - if (pf_flags & PF_MEMALLOC) - current->flags |= PF_MEMALLOC; + err = __add_to_swap_cache(page, entry, + GFP_ATOMIC|__GFP_NOMEMALLOC|__GFP_NOWARN); switch (err) { case 0: /* Success */ -- cgit v0.10.2 From 2054606ad6dd6fee559fe790f190b15ed9355237 Mon Sep 17 00:00:00 2001 From: Nikita Danilov Date: Sun, 1 May 2005 08:58:37 -0700 Subject: [PATCH] doc: Locking update Make the Locking document truer. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking index a934bae..1045da5 100644 --- a/Documentation/filesystems/Locking +++ b/Documentation/filesystems/Locking @@ -219,8 +219,12 @@ This may also be done to avoid internal deadlocks, but rarely. If the filesytem is called for sync then it must wait on any in-progress I/O and then start new I/O. -The filesystem should unlock the page synchronously, before returning -to the caller. +The filesystem should unlock the page synchronously, before returning to the +caller, unless ->writepage() returns special WRITEPAGE_ACTIVATE +value. WRITEPAGE_ACTIVATE means that page cannot really be written out +currently, and VM should stop calling ->writepage() on this page for some +time. VM does this by moving page to the head of the active list, hence the +name. Unless the filesystem is going to redirty_page_for_writepage(), unlock the page and return zero, writepage *must* run set_page_writeback() against the page, -- cgit v0.10.2 From edfbe2b0038723e5699ab22695ccd62b5542a5c1 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Sun, 1 May 2005 08:58:37 -0700 Subject: [PATCH] count bounce buffer pages in vmstat This is a patch for counting the number of pages for bounce buffers. It's shown in /proc/vmstat. Currently, the number of bounce pages are not counted anywhere. So, if there are many bounce pages, it seems that there are leaked pages. And it's difficult for a user to imagine the usage of bounce pages. So, it's meaningful to show # of bouce pages. Signed-off-by: KAMEZAWA Hiroyuki Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index 6b74fcf..39ab8c6 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -131,6 +131,7 @@ struct page_state { unsigned long allocstall; /* direct reclaim calls */ unsigned long pgrotated; /* pages rotated to tail of the LRU */ + unsigned long nr_bounce; /* pages for bounce buffers */ }; extern void get_page_state(struct page_state *ret); diff --git a/mm/highmem.c b/mm/highmem.c index d012765..4009115 100644 --- a/mm/highmem.c +++ b/mm/highmem.c @@ -325,6 +325,7 @@ static void bounce_end_io(struct bio *bio, mempool_t *pool, int err) continue; mempool_free(bvec->bv_page, pool); + dec_page_state(nr_bounce); } bio_endio(bio_orig, bio_orig->bi_size, err); @@ -405,6 +406,7 @@ static void __blk_queue_bounce(request_queue_t *q, struct bio **bio_orig, to->bv_page = mempool_alloc(pool, q->bounce_gfp); to->bv_len = from->bv_len; to->bv_offset = from->bv_offset; + inc_page_state(nr_bounce); if (rw == WRITE) { char *vto, *vfrom; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 04a35b3..80ce7f2 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1897,6 +1897,7 @@ static char *vmstat_text[] = { "allocstall", "pgrotated", + "nr_bounce", }; static void *vmstat_start(struct seq_file *m, loff_t *pos) -- cgit v0.10.2 From 93ea1d0a12623dc1a693642b5758261f35f9bf96 Mon Sep 17 00:00:00 2001 From: Chris Wright Date: Sun, 1 May 2005 08:58:38 -0700 Subject: [PATCH] RLIMIT_MEMLOCK checking fix Always use page counts when doing RLIMIT_MEMLOCK checking to avoid possible overflow. Signed-off-by: Chris Wright Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/mmap.c b/mm/mmap.c index 1ec0f6e..01f9793 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -937,9 +937,10 @@ unsigned long do_mmap_pgoff(struct file * file, unsigned long addr, /* mlock MCL_FUTURE? */ if (vm_flags & VM_LOCKED) { unsigned long locked, lock_limit; - locked = mm->locked_vm << PAGE_SHIFT; + locked = len >> PAGE_SHIFT; + locked += mm->locked_vm; lock_limit = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur; - locked += len; + lock_limit >>= PAGE_SHIFT; if (locked > lock_limit && !capable(CAP_IPC_LOCK)) return -EAGAIN; } @@ -1822,9 +1823,10 @@ unsigned long do_brk(unsigned long addr, unsigned long len) */ if (mm->def_flags & VM_LOCKED) { unsigned long locked, lock_limit; - locked = mm->locked_vm << PAGE_SHIFT; + locked = len >> PAGE_SHIFT; + locked += mm->locked_vm; lock_limit = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur; - locked += len; + lock_limit >>= PAGE_SHIFT; if (locked > lock_limit && !capable(CAP_IPC_LOCK)) return -EAGAIN; } -- cgit v0.10.2 From dd1d5afca8d3bda7ff9db773fc08e648d2503dc6 Mon Sep 17 00:00:00 2001 From: William Lee Irwin III Date: Sun, 1 May 2005 08:58:38 -0700 Subject: [PATCH] sync_page() smp_mb() comment The smp_mb() is becaus sync_page() doesn't have PG_locked while it accesses page_mapping(page). The comments in the patch (the entire patch is the addition of this comment) try to explain further how and why smp_mb() is used. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/filemap.c b/mm/filemap.c index 9b74674..ee79b5d 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -139,7 +139,25 @@ static int sync_page(void *word) page = container_of((page_flags_t *)word, struct page, flags); /* - * FIXME, fercrissake. What is this barrier here for? + * page_mapping() is being called without PG_locked held. + * Some knowledge of the state and use of the page is used to + * reduce the requirements down to a memory barrier. + * The danger here is of a stale page_mapping() return value + * indicating a struct address_space different from the one it's + * associated with when it is associated with one. + * After smp_mb(), it's either the correct page_mapping() for + * the page, or an old page_mapping() and the page's own + * page_mapping() has gone NULL. + * The ->sync_page() address_space operation must tolerate + * page_mapping() going NULL. By an amazing coincidence, + * this comes about because none of the users of the page + * in the ->sync_page() methods make essential use of the + * page_mapping(), merely passing the page down to the backing + * device's unplug functions when it's non-NULL, which in turn + * ignore it for all cases but swap, where only page->private is + * of interest. When page_mapping() does go NULL, the entire + * call stack gracefully ignores the page and returns. + * -- wli */ smp_mb(); mapping = page_mapping(page); -- cgit v0.10.2 From 97e2bde47f886a317909c8a8f9bd2fcd8ce2f0b0 Mon Sep 17 00:00:00 2001 From: Manfred Spraul Date: Sun, 1 May 2005 08:58:38 -0700 Subject: [PATCH] add kmalloc_node, inline cleanup The patch makes the following function calls available to allocate memory on a specific node without changing the basic operation of the slab allocator: kmem_cache_alloc_node(kmem_cache_t *cachep, unsigned int flags, int node); kmalloc_node(size_t size, unsigned int flags, int node); in a similar way to the existing node-blind functions: kmem_cache_alloc(kmem_cache_t *cachep, unsigned int flags); kmalloc(size, flags); kmem_cache_alloc_node was changed to pass flags and the node information through the existing layers of the slab allocator (which lead to some minor rearrangements). The functions at the lowest layer (kmem_getpages, cache_grow) are already node aware. Also __alloc_percpu can call kmalloc_node now. Performance measurements (using the pageset localization patch) yields: w/o patches: Tasks jobs/min jti jobs/min/task real cpu 1 484.27 100 484.2736 12.02 1.97 Wed Mar 30 20:50:43 2005 100 25170.83 91 251.7083 23.12 150.10 Wed Mar 30 20:51:06 2005 200 34601.66 84 173.0083 33.64 294.14 Wed Mar 30 20:51:40 2005 300 37154.47 86 123.8482 46.99 436.56 Wed Mar 30 20:52:28 2005 400 39839.82 80 99.5995 58.43 580.46 Wed Mar 30 20:53:27 2005 500 40036.32 79 80.0726 72.68 728.60 Wed Mar 30 20:54:40 2005 600 44074.21 79 73.4570 79.23 872.10 Wed Mar 30 20:55:59 2005 700 44016.60 78 62.8809 92.56 1015.84 Wed Mar 30 20:57:32 2005 800 40411.05 80 50.5138 115.22 1161.13 Wed Mar 30 20:59:28 2005 900 42298.56 79 46.9984 123.83 1303.42 Wed Mar 30 21:01:33 2005 1000 40955.05 80 40.9551 142.11 1441.92 Wed Mar 30 21:03:55 2005 with pageset localization and slab API patches: Tasks jobs/min jti jobs/min/task real cpu 1 484.19 100 484.1930 12.02 1.98 Wed Mar 30 21:10:18 2005 100 27428.25 92 274.2825 21.22 149.79 Wed Mar 30 21:10:40 2005 200 37228.94 86 186.1447 31.27 293.49 Wed Mar 30 21:11:12 2005 300 41725.42 85 139.0847 41.84 434.10 Wed Mar 30 21:11:54 2005 400 43032.22 82 107.5805 54.10 582.06 Wed Mar 30 21:12:48 2005 500 42211.23 83 84.4225 68.94 722.61 Wed Mar 30 21:13:58 2005 600 40084.49 82 66.8075 87.12 873.11 Wed Mar 30 21:15:25 2005 700 44169.30 79 63.0990 92.24 1008.77 Wed Mar 30 21:16:58 2005 800 43097.94 79 53.8724 108.03 1155.88 Wed Mar 30 21:18:47 2005 900 41846.75 79 46.4964 125.17 1303.38 Wed Mar 30 21:20:52 2005 1000 40247.85 79 40.2478 144.60 1442.21 Wed Mar 30 21:23:17 2005 Signed-off-by: Christoph Lameter Signed-off-by: Manfred Spraul Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/slab.h b/include/linux/slab.h index 3e3c3ab..7d66385 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -62,16 +62,9 @@ extern kmem_cache_t *kmem_cache_create(const char *, size_t, size_t, unsigned lo extern int kmem_cache_destroy(kmem_cache_t *); extern int kmem_cache_shrink(kmem_cache_t *); extern void *kmem_cache_alloc(kmem_cache_t *, unsigned int __nocast); -#ifdef CONFIG_NUMA -extern void *kmem_cache_alloc_node(kmem_cache_t *, int); -#else -static inline void *kmem_cache_alloc_node(kmem_cache_t *cachep, int node) -{ - return kmem_cache_alloc(cachep, GFP_KERNEL); -} -#endif extern void kmem_cache_free(kmem_cache_t *, void *); extern unsigned int kmem_cache_size(kmem_cache_t *); +extern kmem_cache_t *kmem_find_general_cachep(size_t size, int gfpflags); /* Size description struct for general caches. */ struct cache_sizes { @@ -109,6 +102,20 @@ extern void *kcalloc(size_t, size_t, unsigned int __nocast); extern void kfree(const void *); extern unsigned int ksize(const void *); +#ifdef CONFIG_NUMA +extern void *kmem_cache_alloc_node(kmem_cache_t *, int flags, int node); +extern void *kmalloc_node(size_t size, int flags, int node); +#else +static inline void *kmem_cache_alloc_node(kmem_cache_t *cachep, int flags, int node) +{ + return kmem_cache_alloc(cachep, flags); +} +static inline void *kmalloc_node(size_t size, int flags, int node) +{ + return kmalloc(size, flags); +} +#endif + extern int FASTCALL(kmem_cache_reap(int)); extern int FASTCALL(kmem_ptr_validate(kmem_cache_t *cachep, void *ptr)); diff --git a/mm/slab.c b/mm/slab.c index ec660d8..771cc09 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -583,7 +583,7 @@ static inline struct array_cache *ac_data(kmem_cache_t *cachep) return cachep->array[smp_processor_id()]; } -static inline kmem_cache_t *kmem_find_general_cachep(size_t size, int gfpflags) +static inline kmem_cache_t *__find_general_cachep(size_t size, int gfpflags) { struct cache_sizes *csizep = malloc_sizes; @@ -607,6 +607,12 @@ static inline kmem_cache_t *kmem_find_general_cachep(size_t size, int gfpflags) return csizep->cs_cachep; } +kmem_cache_t *kmem_find_general_cachep(size_t size, int gfpflags) +{ + return __find_general_cachep(size, gfpflags); +} +EXPORT_SYMBOL(kmem_find_general_cachep); + /* Cal the num objs, wastage, and bytes left over for a given slab size. */ static void cache_estimate(unsigned long gfporder, size_t size, size_t align, int flags, size_t *left_over, unsigned int *num) @@ -672,14 +678,11 @@ static struct array_cache *alloc_arraycache(int cpu, int entries, int memsize = sizeof(void*)*entries+sizeof(struct array_cache); struct array_cache *nc = NULL; - if (cpu != -1) { - kmem_cache_t *cachep; - cachep = kmem_find_general_cachep(memsize, GFP_KERNEL); - if (cachep) - nc = kmem_cache_alloc_node(cachep, cpu_to_node(cpu)); - } - if (!nc) + if (cpu == -1) nc = kmalloc(memsize, GFP_KERNEL); + else + nc = kmalloc_node(memsize, GFP_KERNEL, cpu_to_node(cpu)); + if (nc) { nc->avail = 0; nc->limit = entries; @@ -2361,7 +2364,7 @@ out: * and can sleep. And it will allocate memory on the given node, which * can improve the performance for cpu bound structures. */ -void *kmem_cache_alloc_node(kmem_cache_t *cachep, int nodeid) +void *kmem_cache_alloc_node(kmem_cache_t *cachep, int flags, int nodeid) { int loop; void *objp; @@ -2393,7 +2396,7 @@ void *kmem_cache_alloc_node(kmem_cache_t *cachep, int nodeid) spin_unlock_irq(&cachep->spinlock); local_irq_disable(); - if (!cache_grow(cachep, GFP_KERNEL, nodeid)) { + if (!cache_grow(cachep, flags, nodeid)) { local_irq_enable(); return NULL; } @@ -2435,6 +2438,16 @@ got_slabp: } EXPORT_SYMBOL(kmem_cache_alloc_node); +void *kmalloc_node(size_t size, int flags, int node) +{ + kmem_cache_t *cachep; + + cachep = kmem_find_general_cachep(size, flags); + if (unlikely(cachep == NULL)) + return NULL; + return kmem_cache_alloc_node(cachep, flags, node); +} +EXPORT_SYMBOL(kmalloc_node); #endif /** @@ -2462,7 +2475,12 @@ void *__kmalloc(size_t size, unsigned int __nocast flags) { kmem_cache_t *cachep; - cachep = kmem_find_general_cachep(size, flags); + /* If you want to save a few bytes .text space: replace + * __ with kmem_. + * Then kmalloc uses the uninlined functions instead of the inline + * functions. + */ + cachep = __find_general_cachep(size, flags); if (unlikely(cachep == NULL)) return NULL; return __cache_alloc(cachep, flags); @@ -2489,9 +2507,8 @@ void *__alloc_percpu(size_t size, size_t align) for (i = 0; i < NR_CPUS; i++) { if (!cpu_possible(i)) continue; - pdata->ptrs[i] = kmem_cache_alloc_node( - kmem_find_general_cachep(size, GFP_KERNEL), - cpu_to_node(i)); + pdata->ptrs[i] = kmalloc_node(size, GFP_KERNEL, + cpu_to_node(i)); if (!pdata->ptrs[i]) goto unwind_oom; -- cgit v0.10.2 From 552fca4cbe552520d85b21e839f289c880fa48d2 Mon Sep 17 00:00:00 2001 From: Nikita Danilov Date: Sun, 1 May 2005 08:58:39 -0700 Subject: [PATCH] mpage_writepages() page locking fix When ->writepage() returns WRITEPAGE_ACTIVATE, the page is still locked. Explicitly unlock the page in mpage_writepages(). Signed-off-by: Nikita Danilov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/mpage.c b/fs/mpage.c index e7d8d1a..3923fac 100644 --- a/fs/mpage.c +++ b/fs/mpage.c @@ -727,6 +727,8 @@ retry: &last_block_in_bio, &ret, wbc, writepage_fn); } + if (unlikely(ret == WRITEPAGE_ACTIVATE)) + unlock_page(page); if (ret || (--(wbc->nr_to_write) <= 0)) done = 1; if (wbc->nonblocking && bdi_write_congested(bdi)) { -- cgit v0.10.2 From de7d5a3b6c9ff8429bf046c36b56d3192b75c3da Mon Sep 17 00:00:00 2001 From: "akpm@osdl.org" Date: Sun, 1 May 2005 08:58:39 -0700 Subject: [PATCH] drop_buffers() oops fix In rare situations, drop_buffers() can be called for a page which has buffers, but no ->mapping (it was truncated, but the buffers were left behind because ext3 was still fiddling with them). But if there was an I/O error in a buffer_head, drop_buffers() will try to get at the address_space and will oops. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/buffer.c b/fs/buffer.c index 3b12cf9..665db84 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2917,7 +2917,7 @@ drop_buffers(struct page *page, struct buffer_head **buffers_to_free) bh = head; do { - if (buffer_write_io_error(bh)) + if (buffer_write_io_error(bh) && page->mapping) set_bit(AS_EIO, &page->mapping->flags); if (buffer_busy(bh)) goto failed; -- cgit v0.10.2 From 6af963f1d6789ef20abca5696cd52a758b396e52 Mon Sep 17 00:00:00 2001 From: Stephen Smalley Date: Sun, 1 May 2005 08:58:39 -0700 Subject: [PATCH] SELinux: cleanup ipc_has_perm This patch removes the sclass argument from ipc_has_perm in the SELinux module, as it can be obtained from the ipc security structure. The use of a separate argument was a legacy of the older precondition function handling in SELinux and is obsolete. Please apply. Signed-off-by: Stephen Smalley Signed-off-by: James Morris Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 0d37814..aae1e79 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -3667,7 +3667,7 @@ static void msg_msg_free_security(struct msg_msg *msg) } static int ipc_has_perm(struct kern_ipc_perm *ipc_perms, - u16 sclass, u32 perms) + u32 perms) { struct task_security_struct *tsec; struct ipc_security_struct *isec; @@ -3679,7 +3679,7 @@ static int ipc_has_perm(struct kern_ipc_perm *ipc_perms, AVC_AUDIT_DATA_INIT(&ad, IPC); ad.u.ipc_id = ipc_perms->key; - return avc_has_perm(tsec->sid, isec->sid, sclass, perms, &ad); + return avc_has_perm(tsec->sid, isec->sid, isec->sclass, perms, &ad); } static int selinux_msg_msg_alloc_security(struct msg_msg *msg) @@ -3764,7 +3764,7 @@ static int selinux_msg_queue_msgctl(struct msg_queue *msq, int cmd) return 0; } - err = ipc_has_perm(&msq->q_perm, SECCLASS_MSGQ, perms); + err = ipc_has_perm(&msq->q_perm, perms); return err; } @@ -3916,7 +3916,7 @@ static int selinux_shm_shmctl(struct shmid_kernel *shp, int cmd) return 0; } - err = ipc_has_perm(&shp->shm_perm, SECCLASS_SHM, perms); + err = ipc_has_perm(&shp->shm_perm, perms); return err; } @@ -3935,7 +3935,7 @@ static int selinux_shm_shmat(struct shmid_kernel *shp, else perms = SHM__READ | SHM__WRITE; - return ipc_has_perm(&shp->shm_perm, SECCLASS_SHM, perms); + return ipc_has_perm(&shp->shm_perm, perms); } /* Semaphore security operations */ @@ -4024,7 +4024,7 @@ static int selinux_sem_semctl(struct sem_array *sma, int cmd) return 0; } - err = ipc_has_perm(&sma->sem_perm, SECCLASS_SEM, perms); + err = ipc_has_perm(&sma->sem_perm, perms); return err; } @@ -4038,18 +4038,13 @@ static int selinux_sem_semop(struct sem_array *sma, else perms = SEM__READ; - return ipc_has_perm(&sma->sem_perm, SECCLASS_SEM, perms); + return ipc_has_perm(&sma->sem_perm, perms); } static int selinux_ipc_permission(struct kern_ipc_perm *ipcp, short flag) { - struct ipc_security_struct *isec = ipcp->security; - u16 sclass = SECCLASS_IPC; u32 av = 0; - if (isec && isec->magic == SELINUX_MAGIC) - sclass = isec->sclass; - av = 0; if (flag & S_IRUGO) av |= IPC__UNIX_READ; @@ -4059,7 +4054,7 @@ static int selinux_ipc_permission(struct kern_ipc_perm *ipcp, short flag) if (av == 0) return 0; - return ipc_has_perm(ipcp, sclass, av); + return ipc_has_perm(ipcp, av); } /* module stacking operations */ -- cgit v0.10.2 From b207a290ea7dc83dba02e40b81cc8a29415a9c60 Mon Sep 17 00:00:00 2001 From: James Morris Date: Sun, 1 May 2005 08:58:40 -0700 Subject: [PATCH] SELinux: add finer grained permissions to Netlink audit processing This patch provides finer grained permissions for the audit family of Netlink sockets under SELinux. 1. We need a way to differentiate between privileged and unprivileged reads of kernel data maintained by the audit subsystem. The AUDIT_GET operation is unprivileged: it returns the current status of the audit subsystem (e.g. whether it's enabled etc.). The AUDIT_LIST operation however returns a list of the current audit ruleset, which is considered privileged by the audit folk. To deal with this, a new SELinux permission has been implemented and applied to the operation: nlmsg_readpriv, which can be allocated to appropriately privileged domains. Unprivileged domains would only be allocated nlmsg_read. 2. There is a requirement for certain domains to generate audit events from userspace. These events need to be collected by the kernel, collated and transmitted sequentially back to the audit daemon. An example is user level login, an auditable event under CAPP, where login-related domains generate AUDIT_USER messages via PAM which are relayed back to auditd via the kernel. To prevent handing out nlmsg_write permissions to such domains, a new permission has been added, nlmsg_relay, which is intended for this type of purpose: data is passed via the kernel back to userspace but no privileged information is written to the kernel. Also, AUDIT_LOGIN messages are now valid only for kernel->user messaging, so this value has been removed from the SELinux nlmsgtab (which is only used to check user->kernel messages). Signed-off-by: James Morris Signed-off-by: Stephen Smalley Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/security/selinux/include/av_perm_to_string.h b/security/selinux/include/av_perm_to_string.h index eb340b4..8928bb4d 100644 --- a/security/selinux/include/av_perm_to_string.h +++ b/security/selinux/include/av_perm_to_string.h @@ -220,6 +220,8 @@ S_(SECCLASS_NETLINK_XFRM_SOCKET, NETLINK_XFRM_SOCKET__NLMSG_WRITE, "nlmsg_write") S_(SECCLASS_NETLINK_AUDIT_SOCKET, NETLINK_AUDIT_SOCKET__NLMSG_READ, "nlmsg_read") S_(SECCLASS_NETLINK_AUDIT_SOCKET, NETLINK_AUDIT_SOCKET__NLMSG_WRITE, "nlmsg_write") + S_(SECCLASS_NETLINK_AUDIT_SOCKET, NETLINK_AUDIT_SOCKET__NLMSG_RELAY, "nlmsg_relay") + S_(SECCLASS_NETLINK_AUDIT_SOCKET, NETLINK_AUDIT_SOCKET__NLMSG_READPRIV, "nlmsg_readpriv") S_(SECCLASS_NETLINK_IP6FW_SOCKET, NETLINK_IP6FW_SOCKET__NLMSG_READ, "nlmsg_read") S_(SECCLASS_NETLINK_IP6FW_SOCKET, NETLINK_IP6FW_SOCKET__NLMSG_WRITE, "nlmsg_write") S_(SECCLASS_DBUS, DBUS__ACQUIRE_SVC, "acquire_svc") diff --git a/security/selinux/include/av_permissions.h b/security/selinux/include/av_permissions.h index f9de0f9..bdfce4c 100644 --- a/security/selinux/include/av_permissions.h +++ b/security/selinux/include/av_permissions.h @@ -840,6 +840,8 @@ #define NETLINK_AUDIT_SOCKET__NLMSG_READ 0x00400000UL #define NETLINK_AUDIT_SOCKET__NLMSG_WRITE 0x00800000UL +#define NETLINK_AUDIT_SOCKET__NLMSG_RELAY 0x01000000UL +#define NETLINK_AUDIT_SOCKET__NLMSG_READPRIV 0x02000000UL #define NETLINK_IP6FW_SOCKET__IOCTL 0x00000001UL #define NETLINK_IP6FW_SOCKET__READ 0x00000002UL diff --git a/security/selinux/nlmsgtab.c b/security/selinux/nlmsgtab.c index f794082..b3adb48 100644 --- a/security/selinux/nlmsgtab.c +++ b/security/selinux/nlmsgtab.c @@ -91,13 +91,12 @@ static struct nlmsg_perm nlmsg_xfrm_perms[] = static struct nlmsg_perm nlmsg_audit_perms[] = { - { AUDIT_GET, NETLINK_AUDIT_SOCKET__NLMSG_READ }, - { AUDIT_SET, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, - { AUDIT_LIST, NETLINK_AUDIT_SOCKET__NLMSG_READ }, - { AUDIT_ADD, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, - { AUDIT_DEL, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, - { AUDIT_USER, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, - { AUDIT_LOGIN, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, + { AUDIT_GET, NETLINK_AUDIT_SOCKET__NLMSG_READ }, + { AUDIT_SET, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, + { AUDIT_LIST, NETLINK_AUDIT_SOCKET__NLMSG_READPRIV }, + { AUDIT_ADD, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, + { AUDIT_DEL, NETLINK_AUDIT_SOCKET__NLMSG_WRITE }, + { AUDIT_USER, NETLINK_AUDIT_SOCKET__NLMSG_RELAY }, }; -- cgit v0.10.2 From f1c55dea0bb2df94aa2b01b0871cb02f2e206676 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Sun, 1 May 2005 08:58:40 -0700 Subject: [PATCH] ppc32: Fix errata for some G3 CPUs Some G3 CPUs can crash in funny way if a store from an FPU register instruction is executed on a register that has never been initialized since power on. This patch fixes it by making sure all FP registers have been properly initialized at kernel boot and when waking from sleep. It also makes the code that decides wether HID0_BTIC and HID0_DPM are allowed on a given CPU smarter (it can actually _clear_ them now if they are not allowed instead of just setting them when they are allowed in case the firmware got them wrong) Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ppc/kernel/cpu_setup_6xx.S b/arch/ppc/kernel/cpu_setup_6xx.S index 74f781b..468721d 100644 --- a/arch/ppc/kernel/cpu_setup_6xx.S +++ b/arch/ppc/kernel/cpu_setup_6xx.S @@ -30,12 +30,14 @@ _GLOBAL(__setup_cpu_604) blr _GLOBAL(__setup_cpu_750) mflr r4 + bl __init_fpu_registers bl setup_common_caches bl setup_750_7400_hid0 mtlr r4 blr _GLOBAL(__setup_cpu_750cx) mflr r4 + bl __init_fpu_registers bl setup_common_caches bl setup_750_7400_hid0 bl setup_750cx @@ -43,6 +45,7 @@ _GLOBAL(__setup_cpu_750cx) blr _GLOBAL(__setup_cpu_750fx) mflr r4 + bl __init_fpu_registers bl setup_common_caches bl setup_750_7400_hid0 bl setup_750fx @@ -50,6 +53,7 @@ _GLOBAL(__setup_cpu_750fx) blr _GLOBAL(__setup_cpu_7400) mflr r4 + bl __init_fpu_registers bl setup_7400_workarounds bl setup_common_caches bl setup_750_7400_hid0 @@ -57,6 +61,7 @@ _GLOBAL(__setup_cpu_7400) blr _GLOBAL(__setup_cpu_7410) mflr r4 + bl __init_fpu_registers bl setup_7410_workarounds bl setup_common_caches bl setup_750_7400_hid0 @@ -80,7 +85,7 @@ setup_common_caches: bne 1f /* don't invalidate the D-cache */ ori r8,r8,HID0_DCI /* unless it wasn't enabled */ 1: sync - mtspr SPRN_HID0,r8 /* enable and invalidate caches */ + mtspr SPRN_HID0,r8 /* enable and invalidate caches */ sync mtspr SPRN_HID0,r11 /* enable caches */ sync @@ -152,9 +157,13 @@ setup_7410_workarounds: setup_750_7400_hid0: mfspr r11,SPRN_HID0 ori r11,r11,HID0_SGE | HID0_ABE | HID0_BHTE | HID0_BTIC + oris r11,r11,HID0_DPM@h BEGIN_FTR_SECTION - oris r11,r11,HID0_DPM@h /* enable dynamic power mgmt */ -END_FTR_SECTION_IFCLR(CPU_FTR_NO_DPM) + xori r11,r11,HID0_BTIC +END_FTR_SECTION_IFSET(CPU_FTR_NO_BTIC) +BEGIN_FTR_SECTION + xoris r11,r11,HID0_DPM@h /* disable dynamic power mgmt */ +END_FTR_SECTION_IFSET(CPU_FTR_NO_DPM) li r3,HID0_SPD andc r11,r11,r3 /* clear SPD: enable speculative */ li r3,0 @@ -218,13 +227,15 @@ setup_745x_specifics: /* All of the bits we have to set..... */ - ori r11,r11,HID0_SGE | HID0_FOLD | HID0_BHTE | HID0_LRSTK | HID0_BTIC + ori r11,r11,HID0_SGE | HID0_FOLD | HID0_BHTE + ori r11,r11,HID0_LRSTK | HID0_BTIC + oris r11,r11,HID0_DPM@h BEGIN_FTR_SECTION xori r11,r11,HID0_BTIC END_FTR_SECTION_IFSET(CPU_FTR_NO_BTIC) BEGIN_FTR_SECTION - oris r11,r11,HID0_DPM@h /* enable dynamic power mgmt */ -END_FTR_SECTION_IFCLR(CPU_FTR_NO_DPM) + xoris r11,r11,HID0_DPM@h /* disable dynamic power mgmt */ +END_FTR_SECTION_IFSET(CPU_FTR_NO_DPM) /* All of the bits we have to clear.... */ @@ -248,6 +259,25 @@ END_FTR_SECTION_IFCLR(CPU_FTR_NO_DPM) isync blr +/* + * Initialize the FPU registers. This is needed to work around an errata + * in some 750 cpus where using a not yet initialized FPU register after + * power on reset may hang the CPU + */ +_GLOBAL(__init_fpu_registers) + mfmsr r10 + ori r11,r10,MSR_FP + mtmsr r11 + isync + addis r9,r3,empty_zero_page@ha + addi r9,r9,empty_zero_page@l + REST_32FPRS(0,r9) + sync + mtmsr r10 + isync + blr + + /* Definitions for the table use to save CPU states */ #define CS_HID0 0 #define CS_HID1 4 diff --git a/arch/ppc/platforms/pmac_sleep.S b/arch/ppc/platforms/pmac_sleep.S index 3139b67..f459ade 100644 --- a/arch/ppc/platforms/pmac_sleep.S +++ b/arch/ppc/platforms/pmac_sleep.S @@ -267,6 +267,10 @@ grackle_wake_up: /* Restore various CPU config stuffs */ bl __restore_cpu_setup + /* Make sure all FPRs have been initialized */ + bl reloc_offset + bl __init_fpu_registers + /* Invalidate & enable L1 cache, we don't care about * whatever the ROM may have tried to write to memory */ -- cgit v0.10.2 From 443a848cd30eb5bb5c1038e6371d83404775dcfc Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Sun, 1 May 2005 08:58:40 -0700 Subject: [PATCH] ppc32: refactor FPU exception handling Moved common FPU exception handling code out of head.S so it can be used by several of the sub-architectures that might of a full PowerPC FPU. Also, uses new CONFIG_PPC_FPU define to fix alignment exception handling for floating point load/store instructions to only occur if we have a hardware FPU. Signed-off-by: Jason McMullan Signed-off-by: Kumar Gala Signed-off-by: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ppc/Kconfig b/arch/ppc/Kconfig index 74aa1e9..c3d9413 100644 --- a/arch/ppc/Kconfig +++ b/arch/ppc/Kconfig @@ -53,6 +53,7 @@ choice config 6xx bool "6xx/7xx/74xx/52xx/82xx/83xx" + select PPC_FPU help There are four types of PowerPC chips supported. The more common types (601, 603, 604, 740, 750, 7400), the Motorola embedded @@ -86,6 +87,9 @@ config E500 endchoice +config PPC_FPU + bool + config BOOKE bool depends on E500 diff --git a/arch/ppc/Makefile b/arch/ppc/Makefile index 73cbdda..0432a25 100644 --- a/arch/ppc/Makefile +++ b/arch/ppc/Makefile @@ -53,6 +53,7 @@ head-$(CONFIG_FSL_BOOKE) := arch/ppc/kernel/head_fsl_booke.o head-$(CONFIG_6xx) += arch/ppc/kernel/idle_6xx.o head-$(CONFIG_POWER4) += arch/ppc/kernel/idle_power4.o +head-$(CONFIG_PPC_FPU) += arch/ppc/kernel/fpu.o core-y += arch/ppc/kernel/ arch/ppc/platforms/ \ arch/ppc/mm/ arch/ppc/lib/ arch/ppc/syslib/ diff --git a/arch/ppc/kernel/Makefile b/arch/ppc/kernel/Makefile index 86bc878..b284451 100644 --- a/arch/ppc/kernel/Makefile +++ b/arch/ppc/kernel/Makefile @@ -9,6 +9,7 @@ extra-$(CONFIG_FSL_BOOKE) := head_fsl_booke.o extra-$(CONFIG_8xx) := head_8xx.o extra-$(CONFIG_6xx) += idle_6xx.o extra-$(CONFIG_POWER4) += idle_power4.o +extra-$(CONFIG_PPC_FPU) += fpu.o extra-y += vmlinux.lds obj-y := entry.o traps.o irq.o idle.o time.o misc.o \ diff --git a/arch/ppc/kernel/align.c b/arch/ppc/kernel/align.c index 79c9294..40d356c 100644 --- a/arch/ppc/kernel/align.c +++ b/arch/ppc/kernel/align.c @@ -368,16 +368,24 @@ fix_alignment(struct pt_regs *regs) /* Single-precision FP load and store require conversions... */ case LD+F+S: +#ifdef CONFIG_PPC_FPU preempt_disable(); enable_kernel_fp(); cvt_fd(&data.f, &data.d, ¤t->thread.fpscr); preempt_enable(); +#else + return 0; +#endif break; case ST+F+S: +#ifdef CONFIG_PPC_FPU preempt_disable(); enable_kernel_fp(); cvt_df(&data.d, &data.f, ¤t->thread.fpscr); preempt_enable(); +#else + return 0; +#endif break; } diff --git a/arch/ppc/kernel/entry.S b/arch/ppc/kernel/entry.S index 035217d..5f075db 100644 --- a/arch/ppc/kernel/entry.S +++ b/arch/ppc/kernel/entry.S @@ -563,6 +563,65 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC) addi r1,r1,INT_FRAME_SIZE blr + .globl fast_exception_return +fast_exception_return: +#if !(defined(CONFIG_4xx) || defined(CONFIG_BOOKE)) + andi. r10,r9,MSR_RI /* check for recoverable interrupt */ + beq 1f /* if not, we've got problems */ +#endif + +2: REST_4GPRS(3, r11) + lwz r10,_CCR(r11) + REST_GPR(1, r11) + mtcr r10 + lwz r10,_LINK(r11) + mtlr r10 + REST_GPR(10, r11) + mtspr SPRN_SRR1,r9 + mtspr SPRN_SRR0,r12 + REST_GPR(9, r11) + REST_GPR(12, r11) + lwz r11,GPR11(r11) + SYNC + RFI + +#if !(defined(CONFIG_4xx) || defined(CONFIG_BOOKE)) +/* check if the exception happened in a restartable section */ +1: lis r3,exc_exit_restart_end@ha + addi r3,r3,exc_exit_restart_end@l + cmplw r12,r3 + bge 3f + lis r4,exc_exit_restart@ha + addi r4,r4,exc_exit_restart@l + cmplw r12,r4 + blt 3f + lis r3,fee_restarts@ha + tophys(r3,r3) + lwz r5,fee_restarts@l(r3) + addi r5,r5,1 + stw r5,fee_restarts@l(r3) + mr r12,r4 /* restart at exc_exit_restart */ + b 2b + + .comm fee_restarts,4 + +/* aargh, a nonrecoverable interrupt, panic */ +/* aargh, we don't know which trap this is */ +/* but the 601 doesn't implement the RI bit, so assume it's OK */ +3: +BEGIN_FTR_SECTION + b 2b +END_FTR_SECTION_IFSET(CPU_FTR_601) + li r10,-1 + stw r10,TRAP(r11) + addi r3,r1,STACK_FRAME_OVERHEAD + lis r10,MSR_KERNEL@h + ori r10,r10,MSR_KERNEL@l + bl transfer_to_handler_full + .long nonrecoverable_exception + .long ret_from_except +#endif + .globl sigreturn_exit sigreturn_exit: subi r1,r3,STACK_FRAME_OVERHEAD diff --git a/arch/ppc/kernel/fpu.S b/arch/ppc/kernel/fpu.S new file mode 100644 index 0000000..6189b26 --- /dev/null +++ b/arch/ppc/kernel/fpu.S @@ -0,0 +1,133 @@ +/* + * FPU support code, moved here from head.S so that it can be used + * by chips which use other head-whatever.S files. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* + * This task wants to use the FPU now. + * On UP, disable FP for the task which had the FPU previously, + * and save its floating-point registers in its thread_struct. + * Load up this task's FP registers from its thread_struct, + * enable the FPU for the current task and return to the task. + */ + .globl load_up_fpu +load_up_fpu: + mfmsr r5 + ori r5,r5,MSR_FP +#ifdef CONFIG_PPC64BRIDGE + clrldi r5,r5,1 /* turn off 64-bit mode */ +#endif /* CONFIG_PPC64BRIDGE */ + SYNC + MTMSRD(r5) /* enable use of fpu now */ + isync +/* + * For SMP, we don't do lazy FPU switching because it just gets too + * horrendously complex, especially when a task switches from one CPU + * to another. Instead we call giveup_fpu in switch_to. + */ +#ifndef CONFIG_SMP + tophys(r6,0) /* get __pa constant */ + addis r3,r6,last_task_used_math@ha + lwz r4,last_task_used_math@l(r3) + cmpwi 0,r4,0 + beq 1f + add r4,r4,r6 + addi r4,r4,THREAD /* want last_task_used_math->thread */ + SAVE_32FPRS(0, r4) + mffs fr0 + stfd fr0,THREAD_FPSCR-4(r4) + lwz r5,PT_REGS(r4) + add r5,r5,r6 + lwz r4,_MSR-STACK_FRAME_OVERHEAD(r5) + li r10,MSR_FP|MSR_FE0|MSR_FE1 + andc r4,r4,r10 /* disable FP for previous task */ + stw r4,_MSR-STACK_FRAME_OVERHEAD(r5) +1: +#endif /* CONFIG_SMP */ + /* enable use of FP after return */ + mfspr r5,SPRN_SPRG3 /* current task's THREAD (phys) */ + lwz r4,THREAD_FPEXC_MODE(r5) + ori r9,r9,MSR_FP /* enable FP for current */ + or r9,r9,r4 + lfd fr0,THREAD_FPSCR-4(r5) + mtfsf 0xff,fr0 + REST_32FPRS(0, r5) +#ifndef CONFIG_SMP + subi r4,r5,THREAD + sub r4,r4,r6 + stw r4,last_task_used_math@l(r3) +#endif /* CONFIG_SMP */ + /* restore registers and return */ + /* we haven't used ctr or xer or lr */ + b fast_exception_return + +/* + * FP unavailable trap from kernel - print a message, but let + * the task use FP in the kernel until it returns to user mode. + */ + .globl KernelFP +KernelFP: + lwz r3,_MSR(r1) + ori r3,r3,MSR_FP + stw r3,_MSR(r1) /* enable use of FP after return */ + lis r3,86f@h + ori r3,r3,86f@l + mr r4,r2 /* current */ + lwz r5,_NIP(r1) + bl printk + b ret_from_except +86: .string "floating point used in kernel (task=%p, pc=%x)\n" + .align 4,0 + +/* + * giveup_fpu(tsk) + * Disable FP for the task given as the argument, + * and save the floating-point registers in its thread_struct. + * Enables the FPU for use in the kernel on return. + */ + .globl giveup_fpu +giveup_fpu: + mfmsr r5 + ori r5,r5,MSR_FP + SYNC_601 + ISYNC_601 + MTMSRD(r5) /* enable use of fpu now */ + SYNC_601 + isync + cmpwi 0,r3,0 + beqlr- /* if no previous owner, done */ + addi r3,r3,THREAD /* want THREAD of task */ + lwz r5,PT_REGS(r3) + cmpwi 0,r5,0 + SAVE_32FPRS(0, r3) + mffs fr0 + stfd fr0,THREAD_FPSCR-4(r3) + beq 1f + lwz r4,_MSR-STACK_FRAME_OVERHEAD(r5) + li r3,MSR_FP|MSR_FE0|MSR_FE1 + andc r4,r4,r3 /* disable FP for previous task */ + stw r4,_MSR-STACK_FRAME_OVERHEAD(r5) +1: +#ifndef CONFIG_SMP + li r5,0 + lis r4,last_task_used_math@ha + stw r5,last_task_used_math@l(r4) +#endif /* CONFIG_SMP */ + blr diff --git a/arch/ppc/kernel/head.S b/arch/ppc/kernel/head.S index 1a89a71..a931d77 100644 --- a/arch/ppc/kernel/head.S +++ b/arch/ppc/kernel/head.S @@ -775,133 +775,6 @@ InstructionSegment: EXC_XFER_STD(0x480, UnknownException) #endif /* CONFIG_PPC64BRIDGE */ -/* - * This task wants to use the FPU now. - * On UP, disable FP for the task which had the FPU previously, - * and save its floating-point registers in its thread_struct. - * Load up this task's FP registers from its thread_struct, - * enable the FPU for the current task and return to the task. - */ -load_up_fpu: - mfmsr r5 - ori r5,r5,MSR_FP -#ifdef CONFIG_PPC64BRIDGE - clrldi r5,r5,1 /* turn off 64-bit mode */ -#endif /* CONFIG_PPC64BRIDGE */ - SYNC - MTMSRD(r5) /* enable use of fpu now */ - isync -/* - * For SMP, we don't do lazy FPU switching because it just gets too - * horrendously complex, especially when a task switches from one CPU - * to another. Instead we call giveup_fpu in switch_to. - */ -#ifndef CONFIG_SMP - tophys(r6,0) /* get __pa constant */ - addis r3,r6,last_task_used_math@ha - lwz r4,last_task_used_math@l(r3) - cmpwi 0,r4,0 - beq 1f - add r4,r4,r6 - addi r4,r4,THREAD /* want last_task_used_math->thread */ - SAVE_32FPRS(0, r4) - mffs fr0 - stfd fr0,THREAD_FPSCR-4(r4) - lwz r5,PT_REGS(r4) - add r5,r5,r6 - lwz r4,_MSR-STACK_FRAME_OVERHEAD(r5) - li r10,MSR_FP|MSR_FE0|MSR_FE1 - andc r4,r4,r10 /* disable FP for previous task */ - stw r4,_MSR-STACK_FRAME_OVERHEAD(r5) -1: -#endif /* CONFIG_SMP */ - /* enable use of FP after return */ - mfspr r5,SPRN_SPRG3 /* current task's THREAD (phys) */ - lwz r4,THREAD_FPEXC_MODE(r5) - ori r9,r9,MSR_FP /* enable FP for current */ - or r9,r9,r4 - lfd fr0,THREAD_FPSCR-4(r5) - mtfsf 0xff,fr0 - REST_32FPRS(0, r5) -#ifndef CONFIG_SMP - subi r4,r5,THREAD - sub r4,r4,r6 - stw r4,last_task_used_math@l(r3) -#endif /* CONFIG_SMP */ - /* restore registers and return */ - /* we haven't used ctr or xer or lr */ - /* fall through to fast_exception_return */ - - .globl fast_exception_return -fast_exception_return: - andi. r10,r9,MSR_RI /* check for recoverable interrupt */ - beq 1f /* if not, we've got problems */ -2: REST_4GPRS(3, r11) - lwz r10,_CCR(r11) - REST_GPR(1, r11) - mtcr r10 - lwz r10,_LINK(r11) - mtlr r10 - REST_GPR(10, r11) - mtspr SPRN_SRR1,r9 - mtspr SPRN_SRR0,r12 - REST_GPR(9, r11) - REST_GPR(12, r11) - lwz r11,GPR11(r11) - SYNC - RFI - -/* check if the exception happened in a restartable section */ -1: lis r3,exc_exit_restart_end@ha - addi r3,r3,exc_exit_restart_end@l - cmplw r12,r3 - bge 3f - lis r4,exc_exit_restart@ha - addi r4,r4,exc_exit_restart@l - cmplw r12,r4 - blt 3f - lis r3,fee_restarts@ha - tophys(r3,r3) - lwz r5,fee_restarts@l(r3) - addi r5,r5,1 - stw r5,fee_restarts@l(r3) - mr r12,r4 /* restart at exc_exit_restart */ - b 2b - - .comm fee_restarts,4 - -/* aargh, a nonrecoverable interrupt, panic */ -/* aargh, we don't know which trap this is */ -/* but the 601 doesn't implement the RI bit, so assume it's OK */ -3: -BEGIN_FTR_SECTION - b 2b -END_FTR_SECTION_IFSET(CPU_FTR_601) - li r10,-1 - stw r10,TRAP(r11) - addi r3,r1,STACK_FRAME_OVERHEAD - li r10,MSR_KERNEL - bl transfer_to_handler_full - .long nonrecoverable_exception - .long ret_from_except - -/* - * FP unavailable trap from kernel - print a message, but let - * the task use FP in the kernel until it returns to user mode. - */ -KernelFP: - lwz r3,_MSR(r1) - ori r3,r3,MSR_FP - stw r3,_MSR(r1) /* enable use of FP after return */ - lis r3,86f@h - ori r3,r3,86f@l - mr r4,r2 /* current */ - lwz r5,_NIP(r1) - bl printk - b ret_from_except -86: .string "floating point used in kernel (task=%p, pc=%x)\n" - .align 4,0 - #ifdef CONFIG_ALTIVEC /* Note that the AltiVec support is closely modeled after the FP * support. Changes to one are likely to be applicable to the @@ -1016,42 +889,6 @@ giveup_altivec: #endif /* CONFIG_ALTIVEC */ /* - * giveup_fpu(tsk) - * Disable FP for the task given as the argument, - * and save the floating-point registers in its thread_struct. - * Enables the FPU for use in the kernel on return. - */ - .globl giveup_fpu -giveup_fpu: - mfmsr r5 - ori r5,r5,MSR_FP - SYNC_601 - ISYNC_601 - MTMSRD(r5) /* enable use of fpu now */ - SYNC_601 - isync - cmpwi 0,r3,0 - beqlr- /* if no previous owner, done */ - addi r3,r3,THREAD /* want THREAD of task */ - lwz r5,PT_REGS(r3) - cmpwi 0,r5,0 - SAVE_32FPRS(0, r3) - mffs fr0 - stfd fr0,THREAD_FPSCR-4(r3) - beq 1f - lwz r4,_MSR-STACK_FRAME_OVERHEAD(r5) - li r3,MSR_FP|MSR_FE0|MSR_FE1 - andc r4,r4,r3 /* disable FP for previous task */ - stw r4,_MSR-STACK_FRAME_OVERHEAD(r5) -1: -#ifndef CONFIG_SMP - li r5,0 - lis r4,last_task_used_math@ha - stw r5,last_task_used_math@l(r4) -#endif /* CONFIG_SMP */ - blr - -/* * This code is jumped to from the startup code to copy * the kernel image to physical address 0. */ diff --git a/arch/ppc/kernel/head_44x.S b/arch/ppc/kernel/head_44x.S index 9ed8165..9b6a8e5 100644 --- a/arch/ppc/kernel/head_44x.S +++ b/arch/ppc/kernel/head_44x.S @@ -426,7 +426,11 @@ interrupt_base: PROGRAM_EXCEPTION /* Floating Point Unavailable Interrupt */ +#ifdef CONFIG_PPC_FPU + FP_UNAVAILABLE_EXCEPTION +#else EXCEPTION(0x2010, FloatingPointUnavailable, UnknownException, EXC_XFER_EE) +#endif /* System Call Interrupt */ START_EXCEPTION(SystemCall) @@ -686,8 +690,10 @@ _GLOBAL(giveup_altivec) * * The 44x core does not have an FPU. */ +#ifndef CONFIG_PPC_FPU _GLOBAL(giveup_fpu) blr +#endif /* * extern void abort(void) diff --git a/arch/ppc/kernel/head_booke.h b/arch/ppc/kernel/head_booke.h index 884dac9..f213d12 100644 --- a/arch/ppc/kernel/head_booke.h +++ b/arch/ppc/kernel/head_booke.h @@ -337,4 +337,11 @@ label: addi r3,r1,STACK_FRAME_OVERHEAD; \ EXC_XFER_LITE(0x0900, timer_interrupt) +#define FP_UNAVAILABLE_EXCEPTION \ + START_EXCEPTION(FloatingPointUnavailable) \ + NORMAL_EXCEPTION_PROLOG; \ + bne load_up_fpu; /* if from user, just load it up */ \ + addi r3,r1,STACK_FRAME_OVERHEAD; \ + EXC_XFER_EE_LITE(0x800, KernelFP) + #endif /* __HEAD_BOOKE_H__ */ diff --git a/arch/ppc/kernel/head_fsl_booke.S b/arch/ppc/kernel/head_fsl_booke.S index d64bf61..f22ddce 100644 --- a/arch/ppc/kernel/head_fsl_booke.S +++ b/arch/ppc/kernel/head_fsl_booke.S @@ -504,7 +504,11 @@ interrupt_base: PROGRAM_EXCEPTION /* Floating Point Unavailable Interrupt */ +#ifdef CONFIG_PPC_FPU + FP_UNAVAILABLE_EXCEPTION +#else EXCEPTION(0x0800, FloatingPointUnavailable, UnknownException, EXC_XFER_EE) +#endif /* System Call Interrupt */ START_EXCEPTION(SystemCall) @@ -916,10 +920,12 @@ _GLOBAL(giveup_spe) /* * extern void giveup_fpu(struct task_struct *prev) * - * The e500 core does not have an FPU. + * Not all FSL Book-E cores have an FPU */ +#ifndef CONFIG_PPC_FPU _GLOBAL(giveup_fpu) blr +#endif /* * extern void abort(void) diff --git a/arch/ppc/kernel/misc.S b/arch/ppc/kernel/misc.S index 73f7c23..e4f1615 100644 --- a/arch/ppc/kernel/misc.S +++ b/arch/ppc/kernel/misc.S @@ -1096,17 +1096,7 @@ _GLOBAL(_get_SP) * and exceptions as if the cpu had performed the load or store. */ -#if defined(CONFIG_4xx) || defined(CONFIG_E500) -_GLOBAL(cvt_fd) - lfs 0,0(r3) - stfd 0,0(r4) - blr - -_GLOBAL(cvt_df) - lfd 0,0(r3) - stfs 0,0(r4) - blr -#else +#ifdef CONFIG_PPC_FPU _GLOBAL(cvt_fd) lfd 0,-4(r5) /* load up fpscr value */ mtfsf 0xff,0 diff --git a/arch/ppc/kernel/traps.c b/arch/ppc/kernel/traps.c index 361865c..f8e7e32 100644 --- a/arch/ppc/kernel/traps.c +++ b/arch/ppc/kernel/traps.c @@ -176,7 +176,7 @@ static inline int check_io_access(struct pt_regs *regs) #else #define get_mc_reason(regs) (mfspr(SPRN_MCSR)) #endif -#define REASON_FP 0 +#define REASON_FP ESR_FP #define REASON_ILLEGAL ESR_PIL #define REASON_PRIVILEGED ESR_PPR #define REASON_TRAP ESR_PTR diff --git a/include/asm-ppc/reg_booke.h b/include/asm-ppc/reg_booke.h index e70c25f..45c5e6f 100644 --- a/include/asm-ppc/reg_booke.h +++ b/include/asm-ppc/reg_booke.h @@ -305,6 +305,7 @@ do { \ #define ESR_PIL 0x08000000 /* Program Exception - Illegal */ #define ESR_PPR 0x04000000 /* Program Exception - Priveleged */ #define ESR_PTR 0x02000000 /* Program Exception - Trap */ +#define ESR_FP 0x01000000 /* Floating Point Operation */ #define ESR_DST 0x00800000 /* Storage Exception - Data miss */ #define ESR_DIZ 0x00400000 /* Storage Exception - Zone fault */ #define ESR_ST 0x00800000 /* Store Operation */ -- cgit v0.10.2 From 630710e3f72b6a324a525f6e5db3f7985d0fc4a2 Mon Sep 17 00:00:00 2001 From: Chris Elston Date: Sun, 1 May 2005 08:58:40 -0700 Subject: [PATCH] ppc32: fix for misreported SDRAM size on Radstone PPC7D platform This patch fixes the SDRAM output from /proc/cpuinfo. The previous code assumed that there was only one bank of SDRAM, and that the size in the memory configuration register was the total size. Signed-off-by: Chris Elston Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ppc/platforms/radstone_ppc7d.c b/arch/ppc/platforms/radstone_ppc7d.c index 2a99b43..df2ea05 100644 --- a/arch/ppc/platforms/radstone_ppc7d.c +++ b/arch/ppc/platforms/radstone_ppc7d.c @@ -253,6 +253,8 @@ static int ppc7d_show_cpuinfo(struct seq_file *m) u8 val1, val2; static int flash_sizes[4] = { 64, 32, 0, 16 }; static int flash_banks[4] = { 4, 3, 2, 1 }; + static int sdram_bank_sizes[4] = { 128, 256, 512, 1 }; + int sdram_num_banks = 2; static char *pci_modes[] = { "PCI33", "PCI66", "Unknown", "Unknown", "PCIX33", "PCIX66", @@ -279,13 +281,17 @@ static int ppc7d_show_cpuinfo(struct seq_file *m) (val1 == PPC7D_CPLD_MB_TYPE_PLL_100) ? 100 : (val1 == PPC7D_CPLD_MB_TYPE_PLL_64) ? 64 : 0); + val = inb(PPC7D_CPLD_MEM_CONFIG); + if (val & PPC7D_CPLD_SDRAM_BANK_NUM_MASK) sdram_num_banks--; + val = inb(PPC7D_CPLD_MEM_CONFIG_EXTEND); - val1 = val & PPC7D_CPLD_SDRAM_BANK_SIZE_MASK; - seq_printf(m, "SDRAM\t\t: %d%c", - (val1 == PPC7D_CPLD_SDRAM_BANK_SIZE_128M) ? 128 : - (val1 == PPC7D_CPLD_SDRAM_BANK_SIZE_256M) ? 256 : - (val1 == PPC7D_CPLD_SDRAM_BANK_SIZE_512M) ? 512 : 1, - (val1 == PPC7D_CPLD_SDRAM_BANK_SIZE_1G) ? 'G' : 'M'); + val1 = (val & PPC7D_CPLD_SDRAM_BANK_SIZE_MASK) >> 6; + seq_printf(m, "SDRAM\t\t: %d banks of %d%c, total %d%c", + sdram_num_banks, + sdram_bank_sizes[val1], + (sdram_bank_sizes[val1] < 128) ? 'G' : 'M', + sdram_num_banks * sdram_bank_sizes[val1], + (sdram_bank_sizes[val1] < 128) ? 'G' : 'M'); if (val2 & PPC7D_CPLD_MB_TYPE_ECC_FITTED_MASK) { seq_printf(m, " [ECC %sabled]", (val2 & PPC7D_CPLD_MB_TYPE_ECC_ENABLE_MASK) ? "en" : diff --git a/arch/ppc/platforms/radstone_ppc7d.h b/arch/ppc/platforms/radstone_ppc7d.h index 4546fff2..9383755 100644 --- a/arch/ppc/platforms/radstone_ppc7d.h +++ b/arch/ppc/platforms/radstone_ppc7d.h @@ -240,6 +240,7 @@ #define PPC7D_CPLD_FLASH_CNTL 0x086E /* MEMORY_CONFIG_EXTEND */ +#define PPC7D_CPLD_SDRAM_BANK_NUM_MASK 0x02 #define PPC7D_CPLD_SDRAM_BANK_SIZE_MASK 0xc0 #define PPC7D_CPLD_SDRAM_BANK_SIZE_128M 0 #define PPC7D_CPLD_SDRAM_BANK_SIZE_256M 0x40 -- cgit v0.10.2 From a497aa20e5ea54fdee474192d6dc138b4832fc9e Mon Sep 17 00:00:00 2001 From: Chris Elston Date: Sun, 1 May 2005 08:58:41 -0700 Subject: [PATCH] ppc32: add rtc hooks in PPC7D platform file This patch adds the hooks into the PPC7D platforms file to support the DS1337 RTC device as the clock device for the PPC7D board. Signed-off-by: Chris Elston Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ppc/platforms/radstone_ppc7d.c b/arch/ppc/platforms/radstone_ppc7d.c index df2ea05..c30607a 100644 --- a/arch/ppc/platforms/radstone_ppc7d.c +++ b/arch/ppc/platforms/radstone_ppc7d.c @@ -68,6 +68,7 @@ #define PPC7D_RST_PIN 17 /* GPP17 */ extern u32 mv64360_irq_base; +extern spinlock_t rtc_lock; static struct mv64x60_handle bh; static int ppc7d_has_alma; @@ -75,6 +76,11 @@ static int ppc7d_has_alma; extern void gen550_progress(char *, unsigned short); extern void gen550_init(int, struct uart_port *); +/* FIXME - move to h file */ +extern int ds1337_do_command(int id, int cmd, void *arg); +#define DS1337_GET_DATE 0 +#define DS1337_SET_DATE 1 + /* residual data */ unsigned char __res[sizeof(bd_t)]; @@ -1242,6 +1248,38 @@ static void __init ppc7d_setup_arch(void) printk(KERN_INFO "Radstone Technology PPC7D\n"); if (ppc_md.progress) ppc_md.progress("ppc7d_setup_arch: exit", 0); + +} + +/* Real Time Clock support. + * PPC7D has a DS1337 accessed by I2C. + */ +static ulong ppc7d_get_rtc_time(void) +{ + struct rtc_time tm; + int result; + + spin_lock(&rtc_lock); + result = ds1337_do_command(0, DS1337_GET_DATE, &tm); + spin_unlock(&rtc_lock); + + if (result == 0) + result = mktime(tm.tm_year, tm.tm_mon, tm.tm_mday, tm.tm_hour, tm.tm_min, tm.tm_sec); + + return result; +} + +static int ppc7d_set_rtc_time(unsigned long nowtime) +{ + struct rtc_time tm; + int result; + + spin_lock(&rtc_lock); + to_tm(nowtime, &tm); + result = ds1337_do_command(0, DS1337_SET_DATE, &tm); + spin_unlock(&rtc_lock); + + return result; } /* This kernel command line parameter can be used to have the target @@ -1299,6 +1337,10 @@ static void ppc7d_init2(void) data8 |= 0x07; outb(data8, PPC7D_CPLD_LEDS); + /* Hook up RTC. We couldn't do this earlier because we need the I2C subsystem */ + ppc_md.set_rtc_time = ppc7d_set_rtc_time; + ppc_md.get_rtc_time = ppc7d_get_rtc_time; + pr_debug("%s: exit\n", __FUNCTION__); } -- cgit v0.10.2 From 616299afcebfb2228f45a95aac7e63704c3733fc Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Sun, 1 May 2005 08:58:41 -0700 Subject: [PATCH] ppc32: Fix IDE related crash on wakeup I noticed an occasional crash on wakeup from sleep on my powerbook (strangly never happened before, probably timing related) that appears to be due to a dangling interrupt while the chip is put to sleep and beeing reset on wakeup. This patch fixes is by disabling the irq in the ide pmac driver while asleep and only re-enable it after the chip has been fully reset. This is safe to do so as the interrupt of these apple IDE cells is never shared. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c index 6dc273a..569f167 100644 --- a/drivers/ide/ppc/pmac.c +++ b/drivers/ide/ppc/pmac.c @@ -1204,6 +1204,8 @@ pmac_ide_do_suspend(ide_hwif_t *hwif) } #endif /* CONFIG_BLK_DEV_IDE_PMAC_BLINK */ + disable_irq(pmif->irq); + /* The media bay will handle itself just fine */ if (pmif->mediabay) return 0; @@ -1236,7 +1238,6 @@ pmac_ide_do_resume(ide_hwif_t *hwif) ppc_md.feature_call(PMAC_FTR_IDE_ENABLE, pmif->node, pmif->aapl_bus_id, 1); msleep(10); ppc_md.feature_call(PMAC_FTR_IDE_RESET, pmif->node, pmif->aapl_bus_id, 0); - msleep(jiffies_to_msecs(IDE_WAKEUP_DELAY)); /* Kauai has it different */ if (pmif->kauai_fcr) { @@ -1244,11 +1245,15 @@ pmac_ide_do_resume(ide_hwif_t *hwif) fcr |= KAUAI_FCR_UATA_RESET_N | KAUAI_FCR_UATA_ENABLE; writel(fcr, pmif->kauai_fcr); } + + msleep(jiffies_to_msecs(IDE_WAKEUP_DELAY)); } /* Sanitize drive timings */ sanitize_timings(pmif); + enable_irq(pmif->irq); + return 0; } -- cgit v0.10.2 From 146a4b3bdfb5641bfbf975e29680b482b8b343ba Mon Sep 17 00:00:00 2001 From: Andreas Jaggi Date: Sun, 1 May 2005 08:58:41 -0700 Subject: [PATCH] macintosh/adbhid.c: adb buttons support for aluminium PowerBook G4 This patch adds support for the special adb buttons of the aluminium PowerBook G4. Signed-off-by: Andreas Jaggi Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/macintosh/adbhid.c b/drivers/macintosh/adbhid.c index 8f93d01..db654e8 100644 --- a/drivers/macintosh/adbhid.c +++ b/drivers/macintosh/adbhid.c @@ -555,6 +555,42 @@ adbhid_buttons_input(unsigned char *data, int nb, struct pt_regs *regs, int auto #endif /* CONFIG_PMAC_BACKLIGHT */ input_report_key(&adbhid[id]->input, KEY_BRIGHTNESSUP, down); break; + + case 0xc: /* videomode switch */ + input_report_key(&adbhid[id]->input, KEY_SWITCHVIDEOMODE, down); + break; + + case 0xd: /* keyboard illumination toggle */ + input_report_key(&adbhid[id]->input, KEY_KBDILLUMTOGGLE, down); + break; + + case 0xe: /* keyboard illumination decrease */ + input_report_key(&adbhid[id]->input, KEY_KBDILLUMDOWN, down); + break; + + case 0xf: + switch (data[1]) { + case 0x8f: + case 0x0f: + /* keyboard illumination increase */ + input_report_key(&adbhid[id]->input, KEY_KBDILLUMUP, down); + break; + + case 0x7f: + case 0xff: + /* keypad overlay toogle */ + break; + + default: + printk(KERN_INFO "Unhandled ADB_MISC event %02x, %02x, %02x, %02x\n", + data[0], data[1], data[2], data[3]); + break; + } + break; + default: + printk(KERN_INFO "Unhandled ADB_MISC event %02x, %02x, %02x, %02x\n", + data[0], data[1], data[2], data[3]); + break; } } break; @@ -775,6 +811,10 @@ adbhid_input_register(int id, int default_id, int original_handler_id, set_bit(KEY_BRIGHTNESSUP, adbhid[id]->input.keybit); set_bit(KEY_BRIGHTNESSDOWN, adbhid[id]->input.keybit); set_bit(KEY_EJECTCD, adbhid[id]->input.keybit); + set_bit(KEY_SWITCHVIDEOMODE, adbhid[id]->input.keybit); + set_bit(KEY_KBDILLUMTOGGLE, adbhid[id]->input.keybit); + set_bit(KEY_KBDILLUMDOWN, adbhid[id]->input.keybit); + set_bit(KEY_KBDILLUMUP, adbhid[id]->input.keybit); break; } if (adbhid[id]->name[0]) diff --git a/include/linux/input.h b/include/linux/input.h index b70df8f..72731d7 100644 --- a/include/linux/input.h +++ b/include/linux/input.h @@ -328,6 +328,11 @@ struct input_absinfo { #define KEY_BRIGHTNESSUP 225 #define KEY_MEDIA 226 +#define KEY_SWITCHVIDEOMODE 227 +#define KEY_KBDILLUMTOGGLE 228 +#define KEY_KBDILLUMDOWN 229 +#define KEY_KBDILLUMUP 230 + #define KEY_UNKNOWN 240 #define BTN_MISC 0x100 -- cgit v0.10.2 From b20cc8aff258eea5a2339107605ebea949fa6ecf Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Sun, 1 May 2005 08:58:42 -0700 Subject: [PATCH] ppc32: Fix a sleep issues on some laptops Some earlier models of aluminium powerbooks and ibook G4s have a clock chip that requires some tweaking before and after sleep. It seems that without that magic incantation to disable and re-enable clock spreading, RAM isn't properly refreshed during sleep. This fixes it. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ppc/platforms/pmac_feature.c b/arch/ppc/platforms/pmac_feature.c index 46cbf36..be41f6f 100644 --- a/arch/ppc/platforms/pmac_feature.c +++ b/arch/ppc/platforms/pmac_feature.c @@ -1590,6 +1590,112 @@ intrepid_shutdown(struct macio_chip* macio, int sleep_mode) mdelay(10); } + +static void __pmac pmac_tweak_clock_spreading(struct macio_chip* macio, int enable) +{ + /* Hack for doing clock spreading on some machines PowerBooks and + * iBooks. This implements the "platform-do-clockspreading" OF + * property as decoded manually on various models. For safety, we also + * check the product ID in the device-tree in cases we'll whack the i2c + * chip to make reasonably sure we won't set wrong values in there + * + * Of course, ultimately, we have to implement a real parser for + * the platform-do-* stuff... + */ + + if (macio->type == macio_intrepid) { + if (enable) + UN_OUT(UNI_N_CLOCK_SPREADING, 2); + else + UN_OUT(UNI_N_CLOCK_SPREADING, 0); + mdelay(40); + } + + while (machine_is_compatible("PowerBook5,2") || + machine_is_compatible("PowerBook5,3") || + machine_is_compatible("PowerBook6,2") || + machine_is_compatible("PowerBook6,3")) { + struct device_node *ui2c = of_find_node_by_type(NULL, "i2c"); + struct device_node *dt = of_find_node_by_name(NULL, "device-tree"); + u8 buffer[9]; + u32 *productID; + int i, rc, changed = 0; + + if (dt == NULL) + break; + productID = (u32 *)get_property(dt, "pid#", NULL); + if (productID == NULL) + break; + while(ui2c) { + struct device_node *p = of_get_parent(ui2c); + if (p && !strcmp(p->name, "uni-n")) + break; + ui2c = of_find_node_by_type(ui2c, "i2c"); + } + if (ui2c == NULL) + break; + DBG("Trying to bump clock speed for PID: %08x...\n", *productID); + rc = pmac_low_i2c_open(ui2c, 1); + if (rc != 0) + break; + pmac_low_i2c_setmode(ui2c, pmac_low_i2c_mode_combined); + rc = pmac_low_i2c_xfer(ui2c, 0xd2 | pmac_low_i2c_read, 0x80, buffer, 9); + DBG("read result: %d,", rc); + if (rc != 0) { + pmac_low_i2c_close(ui2c); + break; + } + for (i=0; i<9; i++) + DBG(" %02x", buffer[i]); + DBG("\n"); + + switch(*productID) { + case 0x1182: /* AlBook 12" rev 2 */ + case 0x1183: /* iBook G4 12" */ + buffer[0] = (buffer[0] & 0x8f) | 0x70; + buffer[2] = (buffer[2] & 0x7f) | 0x00; + buffer[5] = (buffer[5] & 0x80) | 0x31; + buffer[6] = (buffer[6] & 0x40) | 0xb0; + buffer[7] = (buffer[7] & 0x00) | (enable ? 0xc0 : 0xba); + buffer[8] = (buffer[8] & 0x00) | 0x30; + changed = 1; + break; + case 0x3142: /* AlBook 15" (ATI M10) */ + case 0x3143: /* AlBook 17" (ATI M10) */ + buffer[0] = (buffer[0] & 0xaf) | 0x50; + buffer[2] = (buffer[2] & 0x7f) | 0x00; + buffer[5] = (buffer[5] & 0x80) | 0x31; + buffer[6] = (buffer[6] & 0x40) | 0xb0; + buffer[7] = (buffer[7] & 0x00) | (enable ? 0xd0 : 0xc0); + buffer[8] = (buffer[8] & 0x00) | 0x30; + changed = 1; + break; + default: + DBG("i2c-hwclock: Machine model not handled\n"); + break; + } + if (!changed) { + pmac_low_i2c_close(ui2c); + break; + } + pmac_low_i2c_setmode(ui2c, pmac_low_i2c_mode_stdsub); + rc = pmac_low_i2c_xfer(ui2c, 0xd2 | pmac_low_i2c_write, 0x80, buffer, 9); + DBG("write result: %d,", rc); + pmac_low_i2c_setmode(ui2c, pmac_low_i2c_mode_combined); + rc = pmac_low_i2c_xfer(ui2c, 0xd2 | pmac_low_i2c_read, 0x80, buffer, 9); + DBG("read result: %d,", rc); + if (rc != 0) { + pmac_low_i2c_close(ui2c); + break; + } + for (i=0; i<9; i++) + DBG(" %02x", buffer[i]); + pmac_low_i2c_close(ui2c); + break; + } +} + + static int __pmac core99_sleep(void) { @@ -1601,11 +1707,8 @@ core99_sleep(void) macio->type != macio_intrepid) return -ENODEV; - /* The device-tree contains that in the hwclock node */ - if (macio->type == macio_intrepid) { - UN_OUT(UNI_N_CLOCK_SPREADING, 0); - mdelay(40); - } + /* Disable clock spreading */ + pmac_tweak_clock_spreading(macio, 0); /* We power off the wireless slot in case it was not done * by the driver. We don't power it on automatically however @@ -1749,11 +1852,8 @@ core99_wake_up(void) UN_OUT(UNI_N_CLOCK_CNTL, save_unin_clock_ctl); udelay(100); - /* Restore clock spreading */ - if (macio->type == macio_intrepid) { - UN_OUT(UNI_N_CLOCK_SPREADING, 2); - mdelay(40); - } + /* Enable clock spreading */ + pmac_tweak_clock_spreading(macio, 1); return 0; } @@ -2718,97 +2818,11 @@ set_initial_features(void) MACIO_BIC(HEATHROW_FCR, HRW_SOUND_POWER_N); } - /* Hack for bumping clock speed on the new PowerBooks and the - * iBook G4. This implements the "platform-do-clockspreading" OF - * property. For safety, we also check the product ID in the - * device-tree to make reasonably sure we won't set wrong values - * in the clock chip. - * - * Of course, ultimately, we have to implement a real parser for - * the platform-do-* stuff... + /* Some machine models need the clock chip to be properly setup for + * clock spreading now. This should be a platform function but we + * don't do these at the moment */ - while (machine_is_compatible("PowerBook5,2") || - machine_is_compatible("PowerBook5,3") || - machine_is_compatible("PowerBook6,2") || - machine_is_compatible("PowerBook6,3")) { - struct device_node *ui2c = of_find_node_by_type(NULL, "i2c"); - struct device_node *dt = of_find_node_by_name(NULL, "device-tree"); - u8 buffer[9]; - u32 *productID; - int i, rc, changed = 0; - - if (dt == NULL) - break; - productID = (u32 *)get_property(dt, "pid#", NULL); - if (productID == NULL) - break; - while(ui2c) { - struct device_node *p = of_get_parent(ui2c); - if (p && !strcmp(p->name, "uni-n")) - break; - ui2c = of_find_node_by_type(ui2c, "i2c"); - } - if (ui2c == NULL) - break; - DBG("Trying to bump clock speed for PID: %08x...\n", *productID); - rc = pmac_low_i2c_open(ui2c, 1); - if (rc != 0) - break; - pmac_low_i2c_setmode(ui2c, pmac_low_i2c_mode_combined); - rc = pmac_low_i2c_xfer(ui2c, 0xd2 | pmac_low_i2c_read, 0x80, buffer, 9); - DBG("read result: %d,", rc); - if (rc != 0) { - pmac_low_i2c_close(ui2c); - break; - } - for (i=0; i<9; i++) - DBG(" %02x", buffer[i]); - DBG("\n"); - - switch(*productID) { - case 0x1182: /* AlBook 12" rev 2 */ - case 0x1183: /* iBook G4 12" */ - buffer[0] = (buffer[0] & 0x8f) | 0x70; - buffer[2] = (buffer[2] & 0x7f) | 0x00; - buffer[5] = (buffer[5] & 0x80) | 0x31; - buffer[6] = (buffer[6] & 0x40) | 0xb0; - buffer[7] = (buffer[7] & 0x00) | 0xc0; - buffer[8] = (buffer[8] & 0x00) | 0x30; - changed = 1; - break; - case 0x3142: /* AlBook 15" (ATI M10) */ - case 0x3143: /* AlBook 17" (ATI M10) */ - buffer[0] = (buffer[0] & 0xaf) | 0x50; - buffer[2] = (buffer[2] & 0x7f) | 0x00; - buffer[5] = (buffer[5] & 0x80) | 0x31; - buffer[6] = (buffer[6] & 0x40) | 0xb0; - buffer[7] = (buffer[7] & 0x00) | 0xd0; - buffer[8] = (buffer[8] & 0x00) | 0x30; - changed = 1; - break; - default: - DBG("i2c-hwclock: Machine model not handled\n"); - break; - } - if (!changed) { - pmac_low_i2c_close(ui2c); - break; - } - pmac_low_i2c_setmode(ui2c, pmac_low_i2c_mode_stdsub); - rc = pmac_low_i2c_xfer(ui2c, 0xd2 | pmac_low_i2c_write, 0x80, buffer, 9); - DBG("write result: %d,", rc); - pmac_low_i2c_setmode(ui2c, pmac_low_i2c_mode_combined); - rc = pmac_low_i2c_xfer(ui2c, 0xd2 | pmac_low_i2c_read, 0x80, buffer, 9); - DBG("read result: %d,", rc); - if (rc != 0) { - pmac_low_i2c_close(ui2c); - break; - } - for (i=0; i<9; i++) - DBG(" %02x", buffer[i]); - pmac_low_i2c_close(ui2c); - break; - } + pmac_tweak_clock_spreading(&macio_chips[0], 1); #endif /* CONFIG_POWER4 */ -- cgit v0.10.2 From d5812a77e5803468a5033be91af978be0f7a17d9 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Sun, 1 May 2005 08:58:42 -0700 Subject: [PATCH] ppc32: Fix address checking on lmw/stmw align exception The handling of misaligned load/store multiple instructions did not check to see if the address was ok to access before using __{get,put}_user(). Signed-off-by: Kumar Gala Signed-off-by: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ppc/kernel/align.c b/arch/ppc/kernel/align.c index 40d356c..ff81da9 100644 --- a/arch/ppc/kernel/align.c +++ b/arch/ppc/kernel/align.c @@ -290,6 +290,10 @@ fix_alignment(struct pt_regs *regs) /* lwm, stmw */ nb = (32 - reg) * 4; } + + if (!access_ok((flags & ST? VERIFY_WRITE: VERIFY_READ), addr, nb+nb0)) + return -EFAULT; /* bad address */ + rptr = (unsigned char *) ®s->gpr[reg]; if (flags & LD) { for (i = 0; i < nb; ++i) -- cgit v0.10.2 From 1bdacf88ebd7969fecbbf4c5b388cc094871222e Mon Sep 17 00:00:00 2001 From: Dan Malek Date: Sun, 1 May 2005 08:58:42 -0700 Subject: [PATCH] ppc32: workaround for spurious IRQs on PQ2 There is a problem with large amounts of spurious IRQs on PowerPC 82xx systems. The problem is corrected by adding sync at the end of cpm2_mask_and_ack. This may be needed on 8xx as well but has not yet been confirmed. Signed-off-by: Eugene Surovegin Signed-off-by: Dan Malek Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ppc/syslib/cpm2_pic.c b/arch/ppc/syslib/cpm2_pic.c index 954b07f..c867be6 100644 --- a/arch/ppc/syslib/cpm2_pic.c +++ b/arch/ppc/syslib/cpm2_pic.c @@ -107,6 +107,11 @@ static void cpm2_end_irq(unsigned int irq_nr) simr = &(cpm2_immr->im_intctl.ic_simrh); ppc_cached_irq_mask[word] |= 1 << bit; simr[word] = ppc_cached_irq_mask[word]; + /* + * Work around large numbers of spurious IRQs on PowerPC 82xx + * systems. + */ + mb(); } } -- cgit v0.10.2 From 4be8dc7ff69182610b40a078b9815bcdf27e0c49 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Sun, 1 May 2005 08:58:43 -0700 Subject: [PATCH] ppc64: improve g5 sound headphone mute This patch fixes a couple more issues with the management of the GPIOs dealing with headphone and line out mute on the G5. It should fix the remaining problems of people not getting any sound out of the headphone jack. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/sound/ppc/tumbler.c b/sound/ppc/tumbler.c index c71807e..f3e0107 100644 --- a/sound/ppc/tumbler.c +++ b/sound/ppc/tumbler.c @@ -177,11 +177,22 @@ static void write_audio_gpio(pmac_gpio_t *gp, int active) if (! gp->addr) return; active = active ? gp->active_val : gp->inactive_val; - do_gpio_write(gp, active); DBG("(I) gpio %x write %d\n", gp->addr, active); } +static int check_audio_gpio(pmac_gpio_t *gp) +{ + int ret; + + if (! gp->addr) + return 0; + + ret = do_gpio_read(gp); + + return (ret & 0xd) == (gp->active_val & 0xd); +} + static int read_audio_gpio(pmac_gpio_t *gp) { int ret; @@ -683,7 +694,7 @@ static int tumbler_get_mute_switch(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_ } if (gp == NULL) return -EINVAL; - ucontrol->value.integer.value[0] = ! read_audio_gpio(gp); + ucontrol->value.integer.value[0] = !check_audio_gpio(gp); return 0; } @@ -711,7 +722,7 @@ static int tumbler_put_mute_switch(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_ } if (gp == NULL) return -EINVAL; - val = ! read_audio_gpio(gp); + val = ! check_audio_gpio(gp); if (val != ucontrol->value.integer.value[0]) { write_audio_gpio(gp, ! ucontrol->value.integer.value[0]); return 1; @@ -897,11 +908,11 @@ static int tumbler_detect_lineout(pmac_t *chip) static void check_mute(pmac_t *chip, pmac_gpio_t *gp, int val, int do_notify, snd_kcontrol_t *sw) { - //pmac_tumbler_t *mix = chip->mixer_data; - if (val != read_audio_gpio(gp)) { + if (check_audio_gpio(gp) != val) { write_audio_gpio(gp, val); if (do_notify) - snd_ctl_notify(chip->card, SNDRV_CTL_EVENT_MASK_VALUE, &sw->id); + snd_ctl_notify(chip->card, SNDRV_CTL_EVENT_MASK_VALUE, + &sw->id); } } -- cgit v0.10.2 From 1f7b49d042abfbda71f41b8aff6e1bf7685c1f00 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Sun, 1 May 2005 08:58:43 -0700 Subject: [PATCH] ppc32: add sound support for Mac Mini This patch applies on top of my previous g5 related sound patches and adds support for the Mac Mini to the PowerMac Alsa driver. However, I haven't found any kind of HW support for volume control on this machine. If it exist, it's well hidden. That means that you probably want to make sure you use software with the ability to do soft volume control, or use Alsa 0.9 pre-release with the softvol plugin. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/sound/ppc/Makefile b/sound/ppc/Makefile index 4d95c65..d6ba995 100644 --- a/sound/ppc/Makefile +++ b/sound/ppc/Makefile @@ -3,7 +3,7 @@ # Copyright (c) 2001 by Jaroslav Kysela # -snd-powermac-objs := powermac.o pmac.o awacs.o burgundy.o daca.o tumbler.o keywest.o beep.o +snd-powermac-objs := powermac.o pmac.o awacs.o burgundy.o daca.o tumbler.o toonie.o keywest.o beep.o # Toplevel Module Dependency obj-$(CONFIG_SND_POWERMAC) += snd-powermac.o diff --git a/sound/ppc/pmac.c b/sound/ppc/pmac.c index 3bf5f06..32d9475 100644 --- a/sound/ppc/pmac.c +++ b/sound/ppc/pmac.c @@ -986,7 +986,13 @@ static int __init snd_pmac_detect(pmac_t *chip) chip->num_freqs = ARRAY_SIZE(tumbler_freqs); chip->model = PMAC_SNAPPER; chip->can_byte_swap = 0; /* FIXME: check this */ - chip->control_mask = MASK_IEPC | 0x11; /* disable IEE */ + chip->control_mask = MASK_IEPC | 0x11;/* disable IEE */ + break; + case 0x3a: + chip->num_freqs = ARRAY_SIZE(tumbler_freqs); + chip->model = PMAC_TOONIE; + chip->can_byte_swap = 0; /* FIXME: check this */ + chip->control_mask = MASK_IEPC | 0x11;/* disable IEE */ break; } } diff --git a/sound/ppc/pmac.h b/sound/ppc/pmac.h index dc6c99d..0a84c05 100644 --- a/sound/ppc/pmac.h +++ b/sound/ppc/pmac.h @@ -94,7 +94,8 @@ struct snd_pmac_stream { */ enum snd_pmac_model { - PMAC_AWACS, PMAC_SCREAMER, PMAC_BURGUNDY, PMAC_DACA, PMAC_TUMBLER, PMAC_SNAPPER + PMAC_AWACS, PMAC_SCREAMER, PMAC_BURGUNDY, PMAC_DACA, PMAC_TUMBLER, + PMAC_SNAPPER, PMAC_TOONIE }; struct snd_pmac { @@ -191,6 +192,7 @@ int snd_pmac_burgundy_init(pmac_t *chip); int snd_pmac_daca_init(pmac_t *chip); int snd_pmac_tumbler_init(pmac_t *chip); int snd_pmac_tumbler_post_init(void); +int snd_pmac_toonie_init(pmac_t *chip); /* i2c functions */ typedef struct snd_pmac_keywest { diff --git a/sound/ppc/powermac.c b/sound/ppc/powermac.c index 8f1953a..231f643 100644 --- a/sound/ppc/powermac.c +++ b/sound/ppc/powermac.c @@ -95,6 +95,13 @@ static int __init snd_pmac_probe(void) if ( snd_pmac_tumbler_init(chip) < 0 || snd_pmac_tumbler_post_init() < 0) goto __error; break; + case PMAC_TOONIE: + strcpy(card->driver, "PMac Toonie"); + strcpy(card->shortname, "PowerMac Toonie"); + strcpy(card->longname, card->shortname); + if ((err = snd_pmac_toonie_init(chip)) < 0) + goto __error; + break; case PMAC_AWACS: case PMAC_SCREAMER: name_ext = chip->model == PMAC_SCREAMER ? "Screamer" : "AWACS"; diff --git a/sound/ppc/toonie.c b/sound/ppc/toonie.c new file mode 100644 index 0000000..0f90919 --- /dev/null +++ b/sound/ppc/toonie.c @@ -0,0 +1,380 @@ +/* + * Mac Mini "toonie" mixer control + * + * Copyright (c) 2005 by Benjamin Herrenschmidt + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "pmac.h" + +#undef DEBUG + +#ifdef DEBUG +#define DBG(fmt...) printk(fmt) +#else +#define DBG(fmt...) +#endif + +struct pmac_gpio { + unsigned int addr; + u8 active_val; + u8 inactive_val; + u8 active_state; +}; + +struct pmac_toonie +{ + struct pmac_gpio hp_detect_gpio; + struct pmac_gpio hp_mute_gpio; + struct pmac_gpio amp_mute_gpio; + int hp_detect_irq; + int auto_mute_notify; + struct work_struct detect_work; +}; + + +/* + * gpio access + */ +#define do_gpio_write(gp, val) \ + pmac_call_feature(PMAC_FTR_WRITE_GPIO, NULL, (gp)->addr, val) +#define do_gpio_read(gp) \ + pmac_call_feature(PMAC_FTR_READ_GPIO, NULL, (gp)->addr, 0) +#define tumbler_gpio_free(gp) /* NOP */ + +static void write_audio_gpio(struct pmac_gpio *gp, int active) +{ + if (! gp->addr) + return; + active = active ? gp->active_val : gp->inactive_val; + do_gpio_write(gp, active); + DBG("(I) gpio %x write %d\n", gp->addr, active); +} + +static int check_audio_gpio(struct pmac_gpio *gp) +{ + int ret; + + if (! gp->addr) + return 0; + + ret = do_gpio_read(gp); + + return (ret & 0xd) == (gp->active_val & 0xd); +} + +static int read_audio_gpio(struct pmac_gpio *gp) +{ + int ret; + if (! gp->addr) + return 0; + ret = ((do_gpio_read(gp) & 0x02) !=0); + return ret == gp->active_state; +} + + +enum { TOONIE_MUTE_HP, TOONIE_MUTE_AMP }; + +static int toonie_get_mute_switch(snd_kcontrol_t *kcontrol, + snd_ctl_elem_value_t *ucontrol) +{ + pmac_t *chip = snd_kcontrol_chip(kcontrol); + struct pmac_toonie *mix = chip->mixer_data; + struct pmac_gpio *gp; + + if (mix == NULL) + return -ENODEV; + switch(kcontrol->private_value) { + case TOONIE_MUTE_HP: + gp = &mix->hp_mute_gpio; + break; + case TOONIE_MUTE_AMP: + gp = &mix->amp_mute_gpio; + break; + default: + return -EINVAL;; + } + ucontrol->value.integer.value[0] = !check_audio_gpio(gp); + return 0; +} + +static int toonie_put_mute_switch(snd_kcontrol_t *kcontrol, + snd_ctl_elem_value_t *ucontrol) +{ + pmac_t *chip = snd_kcontrol_chip(kcontrol); + struct pmac_toonie *mix = chip->mixer_data; + struct pmac_gpio *gp; + int val; + + if (chip->update_automute && chip->auto_mute) + return 0; /* don't touch in the auto-mute mode */ + + if (mix == NULL) + return -ENODEV; + + switch(kcontrol->private_value) { + case TOONIE_MUTE_HP: + gp = &mix->hp_mute_gpio; + break; + case TOONIE_MUTE_AMP: + gp = &mix->amp_mute_gpio; + break; + default: + return -EINVAL;; + } + val = ! check_audio_gpio(gp); + if (val != ucontrol->value.integer.value[0]) { + write_audio_gpio(gp, ! ucontrol->value.integer.value[0]); + return 1; + } + return 0; +} + +static snd_kcontrol_new_t toonie_hp_sw __initdata = { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Headphone Playback Switch", + .info = snd_pmac_boolean_mono_info, + .get = toonie_get_mute_switch, + .put = toonie_put_mute_switch, + .private_value = TOONIE_MUTE_HP, +}; +static snd_kcontrol_new_t toonie_speaker_sw __initdata = { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "PC Speaker Playback Switch", + .info = snd_pmac_boolean_mono_info, + .get = toonie_get_mute_switch, + .put = toonie_put_mute_switch, + .private_value = TOONIE_MUTE_AMP, +}; + +/* + * auto-mute stuffs + */ +static int toonie_detect_headphone(pmac_t *chip) +{ + struct pmac_toonie *mix = chip->mixer_data; + int detect = 0; + + if (mix->hp_detect_gpio.addr) + detect |= read_audio_gpio(&mix->hp_detect_gpio); + return detect; +} + +static void toonie_check_mute(pmac_t *chip, struct pmac_gpio *gp, int val, + int do_notify, snd_kcontrol_t *sw) +{ + if (check_audio_gpio(gp) != val) { + write_audio_gpio(gp, val); + if (do_notify) + snd_ctl_notify(chip->card, SNDRV_CTL_EVENT_MASK_VALUE, + &sw->id); + } +} + +static void toonie_detect_handler(void *self) +{ + pmac_t *chip = (pmac_t*) self; + struct pmac_toonie *mix; + int headphone; + + if (!chip) + return; + + mix = chip->mixer_data; + snd_assert(mix, return); + + headphone = toonie_detect_headphone(chip); + + DBG("headphone: %d, lineout: %d\n", headphone, lineout); + + if (headphone) { + /* unmute headphone/lineout & mute speaker */ + toonie_check_mute(chip, &mix->hp_mute_gpio, 0, + mix->auto_mute_notify, chip->master_sw_ctl); + toonie_check_mute(chip, &mix->amp_mute_gpio, 1, + mix->auto_mute_notify, chip->speaker_sw_ctl); + } else { + /* unmute speaker, mute others */ + toonie_check_mute(chip, &mix->amp_mute_gpio, 0, + mix->auto_mute_notify, chip->speaker_sw_ctl); + toonie_check_mute(chip, &mix->hp_mute_gpio, 1, + mix->auto_mute_notify, chip->master_sw_ctl); + } + if (mix->auto_mute_notify) { + snd_ctl_notify(chip->card, SNDRV_CTL_EVENT_MASK_VALUE, + &chip->hp_detect_ctl->id); + } +} + +static void toonie_update_automute(pmac_t *chip, int do_notify) +{ + if (chip->auto_mute) { + struct pmac_toonie *mix; + mix = chip->mixer_data; + snd_assert(mix, return); + mix->auto_mute_notify = do_notify; + schedule_work(&mix->detect_work); + } +} + +/* interrupt - headphone plug changed */ +static irqreturn_t toonie_hp_intr(int irq, void *devid, struct pt_regs *regs) +{ + pmac_t *chip = devid; + + if (chip->update_automute && chip->initialized) { + chip->update_automute(chip, 1); + return IRQ_HANDLED; + } + return IRQ_NONE; +} + +/* look for audio gpio device */ +static int find_audio_gpio(const char *name, const char *platform, + struct pmac_gpio *gp) +{ + struct device_node *np; + u32 *base, addr; + + if (! (np = find_devices("gpio"))) + return -ENODEV; + + for (np = np->child; np; np = np->sibling) { + char *property = get_property(np, "audio-gpio", NULL); + if (property && strcmp(property, name) == 0) + break; + if (device_is_compatible(np, name)) + break; + } + if (np == NULL) + return -ENODEV; + + base = (u32 *)get_property(np, "AAPL,address", NULL); + if (! base) { + base = (u32 *)get_property(np, "reg", NULL); + if (!base) { + DBG("(E) cannot find address for device %s !\n", device); + snd_printd("cannot find address for device %s\n", device); + return -ENODEV; + } + addr = *base; + if (addr < 0x50) + addr += 0x50; + } else + addr = *base; + + gp->addr = addr & 0x0000ffff; + + /* Try to find the active state, default to 0 ! */ + base = (u32 *)get_property(np, "audio-gpio-active-state", NULL); + if (base) { + gp->active_state = *base; + gp->active_val = (*base) ? 0x5 : 0x4; + gp->inactive_val = (*base) ? 0x4 : 0x5; + } else { + u32 *prop = NULL; + gp->active_state = 0; + gp->active_val = 0x4; + gp->inactive_val = 0x5; + /* Here are some crude hacks to extract the GPIO polarity and + * open collector informations out of the do-platform script + * as we don't yet have an interpreter for these things + */ + if (platform) + prop = (u32 *)get_property(np, platform, NULL); + if (prop) { + if (prop[3] == 0x9 && prop[4] == 0x9) { + gp->active_val = 0xd; + gp->inactive_val = 0xc; + } + if (prop[3] == 0x1 && prop[4] == 0x1) { + gp->active_val = 0x5; + gp->inactive_val = 0x4; + } + } + } + + DBG("(I) GPIO device %s found, offset: %x, active state: %d !\n", + device, gp->addr, gp->active_state); + + return (np->n_intrs > 0) ? np->intrs[0].line : 0; +} + +static void toonie_cleanup(pmac_t *chip) +{ + struct pmac_toonie *mix = chip->mixer_data; + if (! mix) + return; + if (mix->hp_detect_irq >= 0) + free_irq(mix->hp_detect_irq, chip); + kfree(mix); + chip->mixer_data = NULL; +} + +int snd_pmac_toonie_init(pmac_t *chip) +{ + struct pmac_toonie *mix; + + mix = kmalloc(sizeof(*mix), GFP_KERNEL); + if (! mix) + return -ENOMEM; + + chip->mixer_data = mix; + chip->mixer_free = toonie_cleanup; + + find_audio_gpio("headphone-mute", NULL, &mix->hp_mute_gpio); + find_audio_gpio("amp-mute", NULL, &mix->amp_mute_gpio); + mix->hp_detect_irq = find_audio_gpio("headphone-detect", + NULL, &mix->hp_detect_gpio); + + strcpy(chip->card->mixername, "PowerMac Toonie"); + + chip->master_sw_ctl = snd_ctl_new1(&toonie_hp_sw, chip); + snd_ctl_add(chip->card, chip->master_sw_ctl); + + chip->speaker_sw_ctl = snd_ctl_new1(&toonie_speaker_sw, chip); + snd_ctl_add(chip->card, chip->speaker_sw_ctl); + + INIT_WORK(&mix->detect_work, toonie_detect_handler, (void *)chip); + + if (mix->hp_detect_irq >= 0) { + snd_pmac_add_automute(chip); + + chip->detect_headphone = toonie_detect_headphone; + chip->update_automute = toonie_update_automute; + toonie_update_automute(chip, 0); + + if (request_irq(mix->hp_detect_irq, toonie_hp_intr, 0, + "Sound Headphone Detection", chip) < 0) + mix->hp_detect_irq = -1; + } + + return 0; +} + -- cgit v0.10.2 From 085e6fc96bcf239cab8adc600aab2452d75590b1 Mon Sep 17 00:00:00 2001 From: Colin Leroy Date: Sun, 1 May 2005 08:58:43 -0700 Subject: [PATCH] pmac: save master volume on sleep Ben's patch that shutdowns master switch and restores it after resume ("pmac: Improve sleep code of tumbler driver") isn't enough here on an iBook (snapper chip). The master switch is correctly saved and restored, but somehow tumbler_put_master_volume() gets called just after tumbler_set_master_volume() and sets mix->master_vol[*] to 0. So, on resuming, the master switch is reenabled, but the volume is set to 0. Here's a patch that also saves and restores master_vol. Signed-off-by: Colin Leroy Cc: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/sound/ppc/tumbler.c b/sound/ppc/tumbler.c index f3e0107..9332237 100644 --- a/sound/ppc/tumbler.c +++ b/sound/ppc/tumbler.c @@ -99,6 +99,7 @@ typedef struct pmac_tumbler_t { pmac_gpio_t hp_detect; int headphone_irq; int lineout_irq; + unsigned int save_master_vol[2]; unsigned int master_vol[2]; unsigned int save_master_switch[2]; unsigned int master_switch[2]; @@ -1139,6 +1140,8 @@ static void tumbler_suspend(pmac_t *chip) disable_irq(mix->lineout_irq); mix->save_master_switch[0] = mix->master_switch[0]; mix->save_master_switch[1] = mix->master_switch[1]; + mix->save_master_vol[0] = mix->master_vol[0]; + mix->save_master_vol[1] = mix->master_vol[1]; mix->master_switch[0] = mix->master_switch[1] = 0; tumbler_set_master_volume(mix); if (!mix->anded_reset) { @@ -1166,6 +1169,8 @@ static void tumbler_resume(pmac_t *chip) mix->acs &= ~1; mix->master_switch[0] = mix->save_master_switch[0]; mix->master_switch[1] = mix->save_master_switch[1]; + mix->master_vol[0] = mix->save_master_vol[0]; + mix->master_vol[1] = mix->save_master_vol[1]; tumbler_reset_audio(chip); if (mix->i2c.client && mix->i2c.init_client) { if (mix->i2c.init_client(&mix->i2c) < 0) -- cgit v0.10.2 From 1b29f9d13e3cf0fe86bf7f82a3399c9e3caf58e5 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Sun, 1 May 2005 08:58:43 -0700 Subject: [PATCH] ppc64: add PT_NOTE section to vDSO This patch from Roland adds a PT_NOTE section to both 32 and 64 bits vDSOs to expose the kernel version to glibc, thus avoiding a uname syscall on every launch. This is equivalent to the patches Roland posted already for x86 and x86-64. Note: the 64 bits .note is actually using the 32 bits format. This is normal. The ELF spec specifies a different format for 64 bits .note, but for some reason, this was never properly implemented, the core dumps for example are all using 32 bits format .note, and binutils cannot even read a 64 bits format .note. Talking to our toolchain folks, they think we'd rather stick to 32 bits format .note everywhere and get the spec fixed some day ... Signed-off-by: Roland McGrath Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ppc64/kernel/vdso32/Makefile b/arch/ppc64/kernel/vdso32/Makefile index ede2f7e..0b1b0df 100644 --- a/arch/ppc64/kernel/vdso32/Makefile +++ b/arch/ppc64/kernel/vdso32/Makefile @@ -1,7 +1,7 @@ # List of files in the vdso, has to be asm only for now -obj-vdso32 = sigtramp.o gettimeofday.o datapage.o cacheflush.o +obj-vdso32 = sigtramp.o gettimeofday.o datapage.o cacheflush.o note.o # Build rules diff --git a/arch/ppc64/kernel/vdso32/note.S b/arch/ppc64/kernel/vdso32/note.S new file mode 100644 index 0000000..d4b5be4 --- /dev/null +++ b/arch/ppc64/kernel/vdso32/note.S @@ -0,0 +1,25 @@ +/* + * This supplies .note.* sections to go into the PT_NOTE inside the vDSO text. + * Here we can supply some information useful to userland. + */ + +#include +#include + +#define ASM_ELF_NOTE_BEGIN(name, flags, vendor, type) \ + .section name, flags; \ + .balign 4; \ + .long 1f - 0f; /* name length */ \ + .long 3f - 2f; /* data length */ \ + .long type; /* note type */ \ +0: .asciz vendor; /* vendor name */ \ +1: .balign 4; \ +2: + +#define ASM_ELF_NOTE_END \ +3: .balign 4; /* pad out section */ \ + .previous + + ASM_ELF_NOTE_BEGIN(".note.kernel-version", "a", UTS_SYSNAME, 0) + .long LINUX_VERSION_CODE + ASM_ELF_NOTE_END diff --git a/arch/ppc64/kernel/vdso32/vdso32.lds.S b/arch/ppc64/kernel/vdso32/vdso32.lds.S index cca27bd..11290c9 100644 --- a/arch/ppc64/kernel/vdso32/vdso32.lds.S +++ b/arch/ppc64/kernel/vdso32/vdso32.lds.S @@ -20,6 +20,8 @@ SECTIONS .gnu.version_d : { *(.gnu.version_d) } .gnu.version_r : { *(.gnu.version_r) } + .note : { *(.note.*) } :text :note + . = ALIGN (16); .text : { @@ -87,6 +89,7 @@ SECTIONS PHDRS { text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */ + note PT_NOTE FLAGS(4); /* PF_R */ dynamic PT_DYNAMIC FLAGS(4); /* PF_R */ eh_frame_hdr 0x6474e550; /* PT_GNU_EH_FRAME, but ld doesn't match the name */ } diff --git a/arch/ppc64/kernel/vdso64/Makefile b/arch/ppc64/kernel/vdso64/Makefile index bd3f70b..ab39988 100644 --- a/arch/ppc64/kernel/vdso64/Makefile +++ b/arch/ppc64/kernel/vdso64/Makefile @@ -1,6 +1,6 @@ # List of files in the vdso, has to be asm only for now -obj-vdso64 = sigtramp.o gettimeofday.o datapage.o cacheflush.o +obj-vdso64 = sigtramp.o gettimeofday.o datapage.o cacheflush.o note.o # Build rules diff --git a/arch/ppc64/kernel/vdso64/note.S b/arch/ppc64/kernel/vdso64/note.S new file mode 100644 index 0000000..dc2a509 --- /dev/null +++ b/arch/ppc64/kernel/vdso64/note.S @@ -0,0 +1 @@ +#include "../vdso32/note.S" diff --git a/arch/ppc64/kernel/vdso64/vdso64.lds.S b/arch/ppc64/kernel/vdso64/vdso64.lds.S index 942c815..9cb2818 100644 --- a/arch/ppc64/kernel/vdso64/vdso64.lds.S +++ b/arch/ppc64/kernel/vdso64/vdso64.lds.S @@ -18,12 +18,14 @@ SECTIONS .gnu.version_d : { *(.gnu.version_d) } .gnu.version_r : { *(.gnu.version_r) } + .note : { *(.note.*) } :text :note + . = ALIGN (16); .text : { *(.text .stub .text.* .gnu.linkonce.t.*) *(.sfpr .glink) - } + } :text PROVIDE (__etext = .); PROVIDE (_etext = .); PROVIDE (etext = .); @@ -88,6 +90,7 @@ SECTIONS PHDRS { text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */ + note PT_NOTE FLAGS(4); /* PF_R */ dynamic PT_DYNAMIC FLAGS(4); /* PF_R */ eh_frame_hdr 0x6474e550; /* PT_GNU_EH_FRAME, but ld doesn't match the name */ } -- cgit v0.10.2 From bb78cb72201985ae9269b723c82ea0f892048b9e Mon Sep 17 00:00:00 2001 From: Olof Johansson Date: Sun, 1 May 2005 08:58:44 -0700 Subject: [PATCH] ppc64: remove unused argument to create_slbe Remove vsid argument to create_slbe, since it's no longer used. Spotted by R Sharada. Signed-off-by: Olof Johansson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ppc64/mm/slb.c b/arch/ppc64/mm/slb.c index 6a20773..244150a 100644 --- a/arch/ppc64/mm/slb.c +++ b/arch/ppc64/mm/slb.c @@ -33,8 +33,8 @@ static inline unsigned long mk_vsid_data(unsigned long ea, unsigned long flags) return (get_kernel_vsid(ea) << SLB_VSID_SHIFT) | flags; } -static inline void create_slbe(unsigned long ea, unsigned long vsid, - unsigned long flags, unsigned long entry) +static inline void create_slbe(unsigned long ea, unsigned long flags, + unsigned long entry) { asm volatile("slbmte %0,%1" : : "r" (mk_vsid_data(ea, flags)), @@ -145,9 +145,8 @@ void slb_initialize(void) asm volatile("isync":::"memory"); asm volatile("slbmte %0,%0"::"r" (0) : "memory"); asm volatile("isync; slbia; isync":::"memory"); - create_slbe(KERNELBASE, get_kernel_vsid(KERNELBASE), flags, 0); - create_slbe(VMALLOCBASE, get_kernel_vsid(KERNELBASE), - SLB_VSID_KERNEL, 1); + create_slbe(KERNELBASE, flags, 0); + create_slbe(VMALLOCBASE, SLB_VSID_KERNEL, 1); /* We don't bolt the stack for the time being - we're in boot, * so the stack is in the bolted segment. By the time it goes * elsewhere, we'll call _switch() which will bolt in the new -- cgit v0.10.2 From dc3ec7503e693e05c01d85b664482d0f694429ab Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Sun, 1 May 2005 08:58:44 -0700 Subject: [PATCH] ppc64: Fix irq parsing on powermac When I tried Ben's patches to the powermac sound driver on my G5, I found that it was taking enormous numbers of sound DMA transmit interrupts. This turned out to be because it was incorrectly configured as level-sensitive instead of edge-sensitive, which in turn was because the code that parses the interrupt tree that Open Firmware gives us was incorrectly assigning another device the same irq number as the sound DMA transmit interrupt (i.e. 1). This patch fixes the problem, in a somewhat quick and dirty way for now, but one which will work for all the machines we currently run on. Ultimately Ben and I want to do something more general and robust, but this should go in for 2.6.12. Signed-off-by: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ppc64/kernel/prom.c b/arch/ppc64/kernel/prom.c index 45a4ad0..fe2946c 100644 --- a/arch/ppc64/kernel/prom.c +++ b/arch/ppc64/kernel/prom.c @@ -321,6 +321,10 @@ static int __devinit finish_node_interrupts(struct device_node *np, char *name = get_property(ic->parent, "name", NULL); if (name && !strcmp(name, "u3")) np->intrs[intrcount].line += 128; + else if (!(name && !strcmp(name, "mac-io"))) + /* ignore other cascaded controllers, such as + the k2-sata-root */ + break; } np->intrs[intrcount].sense = 1; if (n > 1) -- cgit v0.10.2 From 0339ad77c4a06fa8529db17c91f790058e18b65b Mon Sep 17 00:00:00 2001 From: "akpm@osdl.org" Date: Sun, 1 May 2005 08:58:44 -0700 Subject: [PATCH] ppc64: nvram cleanups - Fix arch/ppc64/kernel/nvram.c:342: warning: `part' might be used uninitialized in this function - Various codingstyle tweaks. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ppc64/kernel/nvram.c b/arch/ppc64/kernel/nvram.c index b9069c2..4e71781 100644 --- a/arch/ppc64/kernel/nvram.c +++ b/arch/ppc64/kernel/nvram.c @@ -339,9 +339,9 @@ static int nvram_remove_os_partition(void) static int nvram_create_os_partition(void) { struct list_head * p; - struct nvram_partition * part; - struct nvram_partition * new_part = NULL; - struct nvram_partition * free_part = NULL; + struct nvram_partition *part = NULL; + struct nvram_partition *new_part = NULL; + struct nvram_partition *free_part = NULL; int seq_init[2] = { 0, 0 }; loff_t tmp_index; long size = 0; @@ -364,13 +364,11 @@ static int nvram_create_os_partition(void) free_part = part; } } - if (!size) { + if (!size) return -ENOSPC; - } /* Create our OS partition */ - new_part = (struct nvram_partition *) - kmalloc(sizeof(struct nvram_partition), GFP_KERNEL); + new_part = kmalloc(sizeof(*new_part), GFP_KERNEL); if (!new_part) { printk(KERN_ERR "nvram_create_os_partition: kmalloc failed\n"); return -ENOMEM; @@ -379,7 +377,7 @@ static int nvram_create_os_partition(void) new_part->index = free_part->index; new_part->header.signature = NVRAM_SIG_OS; new_part->header.length = size; - sprintf(new_part->header.name, "ppc64,linux"); + strcpy(new_part->header.name, "ppc64,linux"); new_part->header.checksum = nvram_checksum(&new_part->header); rc = nvram_write_header(new_part); @@ -394,7 +392,8 @@ static int nvram_create_os_partition(void) tmp_index = new_part->index + NVRAM_HEADER_LEN; rc = ppc_md.nvram_write((char *)&seq_init, sizeof(seq_init), &tmp_index); if (rc <= 0) { - printk(KERN_ERR "nvram_create_os_partition: nvram_write failed (%d)\n", rc); + printk(KERN_ERR "nvram_create_os_partition: nvram_write " + "failed (%d)\n", rc); return rc; } -- cgit v0.10.2 From 58366af5861eee1479426380e3c91ecb334c301d Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Sun, 1 May 2005 08:58:44 -0700 Subject: [PATCH] ppc64: update to use the new 4L headers This patch converts ppc64 to use the generic pgtable-nopud.h instead of the "fixup" header. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ppc64/mm/hugetlbpage.c b/arch/ppc64/mm/hugetlbpage.c index 390296e..d3bf86a 100644 --- a/arch/ppc64/mm/hugetlbpage.c +++ b/arch/ppc64/mm/hugetlbpage.c @@ -42,7 +42,7 @@ static inline int hugepgd_index(unsigned long addr) return (addr & ~REGION_MASK) >> HUGEPGDIR_SHIFT; } -static pgd_t *hugepgd_offset(struct mm_struct *mm, unsigned long addr) +static pud_t *hugepgd_offset(struct mm_struct *mm, unsigned long addr) { int index; @@ -52,21 +52,21 @@ static pgd_t *hugepgd_offset(struct mm_struct *mm, unsigned long addr) index = hugepgd_index(addr); BUG_ON(index >= PTRS_PER_HUGEPGD); - return mm->context.huge_pgdir + index; + return (pud_t *)(mm->context.huge_pgdir + index); } -static inline pte_t *hugepte_offset(pgd_t *dir, unsigned long addr) +static inline pte_t *hugepte_offset(pud_t *dir, unsigned long addr) { int index; - if (pgd_none(*dir)) + if (pud_none(*dir)) return NULL; index = (addr >> HPAGE_SHIFT) % PTRS_PER_HUGEPTE; - return (pte_t *)pgd_page(*dir) + index; + return (pte_t *)pud_page(*dir) + index; } -static pgd_t *hugepgd_alloc(struct mm_struct *mm, unsigned long addr) +static pud_t *hugepgd_alloc(struct mm_struct *mm, unsigned long addr) { BUG_ON(! in_hugepage_area(mm->context, addr)); @@ -90,10 +90,9 @@ static pgd_t *hugepgd_alloc(struct mm_struct *mm, unsigned long addr) return hugepgd_offset(mm, addr); } -static pte_t *hugepte_alloc(struct mm_struct *mm, pgd_t *dir, - unsigned long addr) +static pte_t *hugepte_alloc(struct mm_struct *mm, pud_t *dir, unsigned long addr) { - if (! pgd_present(*dir)) { + if (! pud_present(*dir)) { pte_t *new; spin_unlock(&mm->page_table_lock); @@ -104,7 +103,7 @@ static pte_t *hugepte_alloc(struct mm_struct *mm, pgd_t *dir, * Because we dropped the lock, we should re-check the * entry, as somebody else could have populated it.. */ - if (pgd_present(*dir)) { + if (pud_present(*dir)) { if (new) kmem_cache_free(zero_cache, new); } else { @@ -115,7 +114,7 @@ static pte_t *hugepte_alloc(struct mm_struct *mm, pgd_t *dir, ptepage = virt_to_page(new); ptepage->mapping = (void *) mm; ptepage->index = addr & HUGEPGDIR_MASK; - pgd_populate(mm, dir, new); + pud_populate(mm, dir, new); } } @@ -124,28 +123,28 @@ static pte_t *hugepte_alloc(struct mm_struct *mm, pgd_t *dir, static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) { - pgd_t *pgd; + pud_t *pud; BUG_ON(! in_hugepage_area(mm->context, addr)); - pgd = hugepgd_offset(mm, addr); - if (! pgd) + pud = hugepgd_offset(mm, addr); + if (! pud) return NULL; - return hugepte_offset(pgd, addr); + return hugepte_offset(pud, addr); } static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) { - pgd_t *pgd; + pud_t *pud; BUG_ON(! in_hugepage_area(mm->context, addr)); - pgd = hugepgd_alloc(mm, addr); - if (! pgd) + pud = hugepgd_alloc(mm, addr); + if (! pud) return NULL; - return hugepte_alloc(mm, pgd, addr); + return hugepte_alloc(mm, pud, addr); } static void set_huge_pte(struct mm_struct *mm, struct vm_area_struct *vma, @@ -709,10 +708,10 @@ void hugetlb_mm_free_pgd(struct mm_struct *mm) /* cleanup any hugepte pages leftover */ for (i = 0; i < PTRS_PER_HUGEPGD; i++) { - pgd_t *pgd = pgdir + i; + pud_t *pud = (pud_t *)(pgdir + i); - if (! pgd_none(*pgd)) { - pte_t *pte = (pte_t *)pgd_page(*pgd); + if (! pud_none(*pud)) { + pte_t *pte = (pte_t *)pud_page(*pud); struct page *ptepage = virt_to_page(pte); ptepage->mapping = NULL; @@ -720,7 +719,7 @@ void hugetlb_mm_free_pgd(struct mm_struct *mm) BUG_ON(memcmp(pte, empty_zero_page, PAGE_SIZE)); kmem_cache_free(zero_cache, pte); } - pgd_clear(pgd); + pud_clear(pud); } BUG_ON(memcmp(pgdir, empty_zero_page, PAGE_SIZE)); diff --git a/arch/ppc64/mm/init.c b/arch/ppc64/mm/init.c index a7149b9..cf33d7e 100644 --- a/arch/ppc64/mm/init.c +++ b/arch/ppc64/mm/init.c @@ -136,14 +136,78 @@ void iounmap(volatile void __iomem *addr) #else +static void unmap_im_area_pte(pmd_t *pmd, unsigned long addr, + unsigned long end) +{ + pte_t *pte; + + pte = pte_offset_kernel(pmd, addr); + do { + pte_t ptent = ptep_get_and_clear(&ioremap_mm, addr, pte); + WARN_ON(!pte_none(ptent) && !pte_present(ptent)); + } while (pte++, addr += PAGE_SIZE, addr != end); +} + +static inline void unmap_im_area_pmd(pud_t *pud, unsigned long addr, + unsigned long end) +{ + pmd_t *pmd; + unsigned long next; + + pmd = pmd_offset(pud, addr); + do { + next = pmd_addr_end(addr, end); + if (pmd_none_or_clear_bad(pmd)) + continue; + unmap_im_area_pte(pmd, addr, next); + } while (pmd++, addr = next, addr != end); +} + +static inline void unmap_im_area_pud(pgd_t *pgd, unsigned long addr, + unsigned long end) +{ + pud_t *pud; + unsigned long next; + + pud = pud_offset(pgd, addr); + do { + next = pud_addr_end(addr, end); + if (pud_none_or_clear_bad(pud)) + continue; + unmap_im_area_pmd(pud, addr, next); + } while (pud++, addr = next, addr != end); +} + +static void unmap_im_area(unsigned long addr, unsigned long end) +{ + struct mm_struct *mm = &ioremap_mm; + unsigned long next; + pgd_t *pgd; + + spin_lock(&mm->page_table_lock); + + pgd = pgd_offset_i(addr); + flush_cache_vunmap(addr, end); + do { + next = pgd_addr_end(addr, end); + if (pgd_none_or_clear_bad(pgd)) + continue; + unmap_im_area_pud(pgd, addr, next); + } while (pgd++, addr = next, addr != end); + flush_tlb_kernel_range(start, end); + + spin_unlock(&mm->page_table_lock); +} + /* * map_io_page currently only called by __ioremap * map_io_page adds an entry to the ioremap page table * and adds an entry to the HPT, possibly bolting it */ -static void map_io_page(unsigned long ea, unsigned long pa, int flags) +static int map_io_page(unsigned long ea, unsigned long pa, int flags) { pgd_t *pgdp; + pud_t *pudp; pmd_t *pmdp; pte_t *ptep; unsigned long vsid; @@ -151,9 +215,15 @@ static void map_io_page(unsigned long ea, unsigned long pa, int flags) if (mem_init_done) { spin_lock(&ioremap_mm.page_table_lock); pgdp = pgd_offset_i(ea); - pmdp = pmd_alloc(&ioremap_mm, pgdp, ea); + pudp = pud_alloc(&ioremap_mm, pgdp, ea); + if (!pudp) + return -ENOMEM; + pmdp = pmd_alloc(&ioremap_mm, pudp, ea); + if (!pmdp) + return -ENOMEM; ptep = pte_alloc_kernel(&ioremap_mm, pmdp, ea); - + if (!ptep) + return -ENOMEM; pa = abs_to_phys(pa); set_pte_at(&ioremap_mm, ea, ptep, pfn_pte(pa >> PAGE_SHIFT, __pgprot(flags))); @@ -181,6 +251,7 @@ static void map_io_page(unsigned long ea, unsigned long pa, int flags) panic("map_io_page: could not insert mapping"); } } + return 0; } @@ -194,9 +265,14 @@ static void __iomem * __ioremap_com(unsigned long addr, unsigned long pa, flags |= pgprot_val(PAGE_KERNEL); for (i = 0; i < size; i += PAGE_SIZE) - map_io_page(ea+i, pa+i, flags); + if (map_io_page(ea+i, pa+i, flags)) + goto failure; return (void __iomem *) (ea + (addr & ~PAGE_MASK)); + failure: + if (mem_init_done) + unmap_im_area(ea, ea + size); + return NULL; } @@ -206,10 +282,11 @@ ioremap(unsigned long addr, unsigned long size) return __ioremap(addr, size, _PAGE_NO_CACHE | _PAGE_GUARDED); } -void __iomem * -__ioremap(unsigned long addr, unsigned long size, unsigned long flags) +void __iomem * __ioremap(unsigned long addr, unsigned long size, + unsigned long flags) { unsigned long pa, ea; + void __iomem *ret; /* * Choose an address to map it to. @@ -232,12 +309,16 @@ __ioremap(unsigned long addr, unsigned long size, unsigned long flags) if (area == NULL) return NULL; ea = (unsigned long)(area->addr); + ret = __ioremap_com(addr, pa, ea, size, flags); + if (!ret) + im_free(area->addr); } else { ea = ioremap_bot; - ioremap_bot += size; + ret = __ioremap_com(addr, pa, ea, size, flags); + if (ret) + ioremap_bot += size; } - - return __ioremap_com(addr, pa, ea, size, flags); + return ret; } #define IS_PAGE_ALIGNED(_val) ((_val) == ((_val) & PAGE_MASK)) @@ -246,6 +327,7 @@ int __ioremap_explicit(unsigned long pa, unsigned long ea, unsigned long size, unsigned long flags) { struct vm_struct *area; + void __iomem *ret; /* For now, require page-aligned values for pa, ea, and size */ if (!IS_PAGE_ALIGNED(pa) || !IS_PAGE_ALIGNED(ea) || @@ -276,7 +358,12 @@ int __ioremap_explicit(unsigned long pa, unsigned long ea, } } - if (__ioremap_com(pa, pa, ea, size, flags) != (void *) ea) { + ret = __ioremap_com(pa, pa, ea, size, flags); + if (ret == NULL) { + printk(KERN_ERR "ioremap_explicit() allocation failure !\n"); + return 1; + } + if (ret != (void *) ea) { printk(KERN_ERR "__ioremap_com() returned unexpected addr\n"); return 1; } @@ -284,69 +371,6 @@ int __ioremap_explicit(unsigned long pa, unsigned long ea, return 0; } -static void unmap_im_area_pte(pmd_t *pmd, unsigned long address, - unsigned long size) -{ - unsigned long base, end; - pte_t *pte; - - if (pmd_none(*pmd)) - return; - if (pmd_bad(*pmd)) { - pmd_ERROR(*pmd); - pmd_clear(pmd); - return; - } - - pte = pte_offset_kernel(pmd, address); - base = address & PMD_MASK; - address &= ~PMD_MASK; - end = address + size; - if (end > PMD_SIZE) - end = PMD_SIZE; - - do { - pte_t page; - page = ptep_get_and_clear(&ioremap_mm, base + address, pte); - address += PAGE_SIZE; - pte++; - if (pte_none(page)) - continue; - if (pte_present(page)) - continue; - printk(KERN_CRIT "Whee.. Swapped out page in kernel page" - " table\n"); - } while (address < end); -} - -static void unmap_im_area_pmd(pgd_t *dir, unsigned long address, - unsigned long size) -{ - unsigned long base, end; - pmd_t *pmd; - - if (pgd_none(*dir)) - return; - if (pgd_bad(*dir)) { - pgd_ERROR(*dir); - pgd_clear(dir); - return; - } - - pmd = pmd_offset(dir, address); - base = address & PGDIR_MASK; - address &= ~PGDIR_MASK; - end = address + size; - if (end > PGDIR_SIZE) - end = PGDIR_SIZE; - - do { - unmap_im_area_pte(pmd, base + address, end - address); - address = (address + PMD_SIZE) & PMD_MASK; - pmd++; - } while (address < end); -} - /* * Unmap an IO region and remove it from imalloc'd list. * Access to IO memory should be serialized by driver. @@ -356,39 +380,19 @@ static void unmap_im_area_pmd(pgd_t *dir, unsigned long address, */ void iounmap(volatile void __iomem *token) { - unsigned long address, start, end, size; - struct mm_struct *mm; - pgd_t *dir; + unsigned long address, size; void *addr; - if (!mem_init_done) { + if (!mem_init_done) return; - } addr = (void *) ((unsigned long __force) token & PAGE_MASK); - if ((size = im_free(addr)) == 0) { + if ((size = im_free(addr)) == 0) return; - } address = (unsigned long)addr; - start = address; - end = address + size; - - mm = &ioremap_mm; - spin_lock(&mm->page_table_lock); - - dir = pgd_offset_i(address); - flush_cache_vunmap(address, end); - do { - unmap_im_area_pmd(dir, address, end - address); - address = (address + PGDIR_SIZE) & PGDIR_MASK; - dir++; - } while (address && (address < end)); - flush_tlb_kernel_range(start, end); - - spin_unlock(&mm->page_table_lock); - return; + unmap_im_area(address, address + size); } static int iounmap_subset_regions(unsigned long addr, unsigned long size) diff --git a/include/asm-ppc64/pgalloc.h b/include/asm-ppc64/pgalloc.h index 16232d7..4fc4b73 100644 --- a/include/asm-ppc64/pgalloc.h +++ b/include/asm-ppc64/pgalloc.h @@ -27,7 +27,7 @@ pgd_free(pgd_t *pgd) kmem_cache_free(zero_cache, pgd); } -#define pgd_populate(MM, PGD, PMD) pgd_set(PGD, PMD) +#define pud_populate(MM, PUD, PMD) pud_set(PUD, PMD) static inline pmd_t * pmd_alloc_one(struct mm_struct *mm, unsigned long addr) diff --git a/include/asm-ppc64/pgtable.h b/include/asm-ppc64/pgtable.h index a261205..b984e27 100644 --- a/include/asm-ppc64/pgtable.h +++ b/include/asm-ppc64/pgtable.h @@ -1,8 +1,6 @@ #ifndef _PPC64_PGTABLE_H #define _PPC64_PGTABLE_H -#include - /* * This file contains the functions and defines necessary to modify and use * the ppc64 hashed page table. @@ -17,6 +15,8 @@ #include #endif /* __ASSEMBLY__ */ +#include + /* PMD_SHIFT determines what a second-level page table entry can map */ #define PMD_SHIFT (PAGE_SHIFT + PAGE_SHIFT - 3) #define PMD_SIZE (1UL << PMD_SHIFT) @@ -228,12 +228,13 @@ void hugetlb_mm_free_pgd(struct mm_struct *mm); #define pmd_page_kernel(pmd) \ (__bpn_to_ba(pmd_val(pmd) >> PMD_TO_PTEPAGE_SHIFT)) #define pmd_page(pmd) virt_to_page(pmd_page_kernel(pmd)) -#define pgd_set(pgdp, pmdp) (pgd_val(*(pgdp)) = (__ba_to_bpn(pmdp))) -#define pgd_none(pgd) (!pgd_val(pgd)) -#define pgd_bad(pgd) ((pgd_val(pgd)) == 0) -#define pgd_present(pgd) (pgd_val(pgd) != 0UL) -#define pgd_clear(pgdp) (pgd_val(*(pgdp)) = 0UL) -#define pgd_page(pgd) (__bpn_to_ba(pgd_val(pgd))) + +#define pud_set(pudp, pmdp) (pud_val(*(pudp)) = (__ba_to_bpn(pmdp))) +#define pud_none(pud) (!pud_val(pud)) +#define pud_bad(pud) ((pud_val(pud)) == 0UL) +#define pud_present(pud) (pud_val(pud) != 0UL) +#define pud_clear(pudp) (pud_val(*(pudp)) = 0UL) +#define pud_page(pud) (__bpn_to_ba(pud_val(pud))) /* * Find an entry in a page-table-directory. We combine the address region @@ -245,12 +246,13 @@ void hugetlb_mm_free_pgd(struct mm_struct *mm); #define pgd_offset(mm, address) ((mm)->pgd + pgd_index(address)) /* Find an entry in the second-level page table.. */ -#define pmd_offset(dir,addr) \ - ((pmd_t *) pgd_page(*(dir)) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1))) +#define pmd_offset(pudp,addr) \ + ((pmd_t *) pud_page(*(pudp)) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1))) /* Find an entry in the third-level page table.. */ #define pte_offset_kernel(dir,addr) \ - ((pte_t *) pmd_page_kernel(*(dir)) + (((addr) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))) + ((pte_t *) pmd_page_kernel(*(dir)) \ + + (((addr) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))) #define pte_offset_map(dir,addr) pte_offset_kernel((dir), (addr)) #define pte_offset_map_nested(dir,addr) pte_offset_kernel((dir), (addr)) @@ -582,19 +584,22 @@ extern long native_hpte_insert(unsigned long hpte_group, unsigned long va, static inline pte_t *find_linux_pte(pgd_t *pgdir, unsigned long ea) { pgd_t *pg; + pud_t *pu; pmd_t *pm; pte_t *pt = NULL; pte_t pte; pg = pgdir + pgd_index(ea); if (!pgd_none(*pg)) { - - pm = pmd_offset(pg, ea); - if (pmd_present(*pm)) { - pt = pte_offset_kernel(pm, ea); - pte = *pt; - if (!pte_present(pte)) - pt = NULL; + pu = pud_offset(pg, ea); + if (!pud_none(*pu)) { + pm = pmd_offset(pu, ea); + if (pmd_present(*pm)) { + pt = pte_offset_kernel(pm, ea); + pte = *pt; + if (!pte_present(pte)) + pt = NULL; + } } } -- cgit v0.10.2 From 66faf9845a05905d75da380767e93455f3e6d620 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Sun, 1 May 2005 08:58:45 -0700 Subject: [PATCH] ppc64: tell firmware about kernel capabilities On pSeries systems, according to the platform architecture specs, we are supposed to be supplying a structure to firmware that tells firmware about our capabilities, such as which version of the data structures that describe available memory we are expecting to see. The way we end up having to supply this data structure is a bit gross, since it was designed for AIX and doesn't suit us very well. This patch adds the code to supply this data structure to the firmware. Signed-off-by: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ppc64/boot/addnote.c b/arch/ppc64/boot/addnote.c index 66ff810..719663a 100644 --- a/arch/ppc64/boot/addnote.c +++ b/arch/ppc64/boot/addnote.c @@ -19,6 +19,7 @@ #include #include +/* CHRP note section */ char arch[] = "PowerPC"; #define N_DESCR 6 @@ -31,6 +32,29 @@ unsigned int descr[N_DESCR] = { 0x4000, /* load-base */ }; +/* RPA note section */ +char rpaname[] = "IBM,RPA-Client-Config"; + +/* + * Note: setting ignore_my_client_config *should* mean that OF ignores + * all the other fields, but there is a firmware bug which means that + * it looks at the splpar field at least. So these values need to be + * reasonable. + */ +#define N_RPA_DESCR 8 +unsigned int rpanote[N_RPA_DESCR] = { + 0, /* lparaffinity */ + 64, /* min_rmo_size */ + 0, /* min_rmo_percent */ + 40, /* max_pft_size */ + 1, /* splpar */ + -1, /* min_load */ + 0, /* new_mem_def */ + 1, /* ignore_my_client_config */ +}; + +#define ROUNDUP(len) (((len) + 3) & ~3) + unsigned char buf[512]; #define GET_16BE(off) ((buf[off] << 8) + (buf[(off)+1])) @@ -69,7 +93,7 @@ main(int ac, char **av) { int fd, n, i; int ph, ps, np; - int nnote, ns; + int nnote, nnote2, ns; if (ac != 2) { fprintf(stderr, "Usage: %s elf-file\n", av[0]); @@ -81,7 +105,8 @@ main(int ac, char **av) exit(1); } - nnote = strlen(arch) + 1 + (N_DESCR + 3) * 4; + nnote = 12 + ROUNDUP(strlen(arch) + 1) + sizeof(descr); + nnote2 = 12 + ROUNDUP(strlen(rpaname) + 1) + sizeof(rpanote); n = read(fd, buf, sizeof(buf)); if (n < 0) { @@ -104,7 +129,7 @@ main(int ac, char **av) np = GET_16BE(E_PHNUM); if (ph < E_HSIZE || ps < PH_HSIZE || np < 1) goto notelf; - if (ph + (np + 1) * ps + nnote > n) + if (ph + (np + 2) * ps + nnote + nnote2 > n) goto nospace; for (i = 0; i < np; ++i) { @@ -117,12 +142,12 @@ main(int ac, char **av) } /* XXX check that the area we want to use is all zeroes */ - for (i = 0; i < ps + nnote; ++i) + for (i = 0; i < 2 * ps + nnote + nnote2; ++i) if (buf[ph + i] != 0) goto nospace; /* fill in the program header entry */ - ns = ph + ps; + ns = ph + 2 * ps; PUT_32BE(ph + PH_TYPE, PT_NOTE); PUT_32BE(ph + PH_OFFSET, ns); PUT_32BE(ph + PH_FILESZ, nnote); @@ -134,11 +159,26 @@ main(int ac, char **av) PUT_32BE(ns + 8, 0x1275); strcpy(&buf[ns + 12], arch); ns += 12 + strlen(arch) + 1; - for (i = 0; i < N_DESCR; ++i) - PUT_32BE(ns + i * 4, descr[i]); + for (i = 0; i < N_DESCR; ++i, ns += 4) + PUT_32BE(ns, descr[i]); + + /* fill in the second program header entry and the RPA note area */ + ph += ps; + PUT_32BE(ph + PH_TYPE, PT_NOTE); + PUT_32BE(ph + PH_OFFSET, ns); + PUT_32BE(ph + PH_FILESZ, nnote2); + + /* fill in the note area we point to */ + PUT_32BE(ns, strlen(rpaname) + 1); + PUT_32BE(ns + 4, sizeof(rpanote)); + PUT_32BE(ns + 8, 0x12759999); + strcpy(&buf[ns + 12], rpaname); + ns += 12 + ROUNDUP(strlen(rpaname) + 1); + for (i = 0; i < N_RPA_DESCR; ++i, ns += 4) + PUT_32BE(ns, rpanote[i]); /* Update the number of program headers */ - PUT_16BE(E_PHNUM, np + 1); + PUT_16BE(E_PHNUM, np + 2); /* write back */ lseek(fd, (long) 0, SEEK_SET); @@ -155,11 +195,11 @@ main(int ac, char **av) exit(0); notelf: - fprintf(stderr, "%s does not appear to be an ELF file\n", av[0]); + fprintf(stderr, "%s does not appear to be an ELF file\n", av[1]); exit(1); nospace: fprintf(stderr, "sorry, I can't find space in %s to put the note\n", - av[0]); + av[1]); exit(1); } diff --git a/arch/ppc64/kernel/prom_init.c b/arch/ppc64/kernel/prom_init.c index 8dffa9a..b0b784f 100644 --- a/arch/ppc64/kernel/prom_init.c +++ b/arch/ppc64/kernel/prom_init.c @@ -493,6 +493,113 @@ static void __init early_cmdline_parse(void) } /* + * To tell the firmware what our capabilities are, we have to pass + * it a fake 32-bit ELF header containing a couple of PT_NOTE sections + * that contain structures that contain the actual values. + */ +static struct fake_elf { + Elf32_Ehdr elfhdr; + Elf32_Phdr phdr[2]; + struct chrpnote { + u32 namesz; + u32 descsz; + u32 type; + char name[8]; /* "PowerPC" */ + struct chrpdesc { + u32 real_mode; + u32 real_base; + u32 real_size; + u32 virt_base; + u32 virt_size; + u32 load_base; + } chrpdesc; + } chrpnote; + struct rpanote { + u32 namesz; + u32 descsz; + u32 type; + char name[24]; /* "IBM,RPA-Client-Config" */ + struct rpadesc { + u32 lpar_affinity; + u32 min_rmo_size; + u32 min_rmo_percent; + u32 max_pft_size; + u32 splpar; + u32 min_load; + u32 new_mem_def; + u32 ignore_me; + } rpadesc; + } rpanote; +} fake_elf = { + .elfhdr = { + .e_ident = { 0x7f, 'E', 'L', 'F', + ELFCLASS32, ELFDATA2MSB, EV_CURRENT }, + .e_type = ET_EXEC, /* yeah right */ + .e_machine = EM_PPC, + .e_version = EV_CURRENT, + .e_phoff = offsetof(struct fake_elf, phdr), + .e_phentsize = sizeof(Elf32_Phdr), + .e_phnum = 2 + }, + .phdr = { + [0] = { + .p_type = PT_NOTE, + .p_offset = offsetof(struct fake_elf, chrpnote), + .p_filesz = sizeof(struct chrpnote) + }, [1] = { + .p_type = PT_NOTE, + .p_offset = offsetof(struct fake_elf, rpanote), + .p_filesz = sizeof(struct rpanote) + } + }, + .chrpnote = { + .namesz = sizeof("PowerPC"), + .descsz = sizeof(struct chrpdesc), + .type = 0x1275, + .name = "PowerPC", + .chrpdesc = { + .real_mode = ~0U, /* ~0 means "don't care" */ + .real_base = ~0U, + .real_size = ~0U, + .virt_base = ~0U, + .virt_size = ~0U, + .load_base = ~0U + }, + }, + .rpanote = { + .namesz = sizeof("IBM,RPA-Client-Config"), + .descsz = sizeof(struct rpadesc), + .type = 0x12759999, + .name = "IBM,RPA-Client-Config", + .rpadesc = { + .lpar_affinity = 0, + .min_rmo_size = 64, /* in megabytes */ + .min_rmo_percent = 0, + .max_pft_size = 48, /* 2^48 bytes max PFT size */ + .splpar = 1, + .min_load = ~0U, + .new_mem_def = 0 + } + } +}; + +static void __init prom_send_capabilities(void) +{ + unsigned long offset = reloc_offset(); + ihandle elfloader; + int ret; + + elfloader = call_prom("open", 1, 1, ADDR("/packages/elf-loader")); + if (elfloader == 0) { + prom_printf("couldn't open /packages/elf-loader\n"); + return; + } + ret = call_prom("call-method", 3, 1, ADDR("process-elf-header"), + elfloader, ADDR(&fake_elf)); + call_prom("close", 1, 0, elfloader); +} + +/* * Memory allocation strategy... our layout is normally: * * at 14Mb or more we vmlinux, then a gap and initrd. In some rare cases, initrd -- cgit v0.10.2 From d03853d566fb32c6bb8cab4bf2ecf53e692f001c Mon Sep 17 00:00:00 2001 From: Olof Johansson Date: Sun, 1 May 2005 08:58:45 -0700 Subject: [PATCH] PPC64: Remove hot busy-wait loop in __hash_page It turns out that our current __hash_page code will do a very hot busy-wait loop waiting on _PAGE_BUSY to be cleared. It even does ldarx/stdcx in the loop, which will bounce reservations around like crazy if there's more than one CPU spinning on the same PTE (or even another PTE in the same reservation granule). The end result is that each fault takes longer when there's contention, which in turn increases the chance of another thread hitting the same fault and also piling up. Not pretty. There's two options here: 1. Do an out-of-line busy loop a'la spinlocks with just loads (no reserves) 2. Just bail and refault if needed. (2) makes sense here: If the PTE is busy, chances are it's in flux anyway and the other code path making a change might just be ready to hash it. This fixes a stampede seen on a large-ish system where a multithreaded HPC app faults in the same text pages on several cpus at the same time. Signed-off-by: Olof Johansson Cc: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ppc64/mm/hash_low.S b/arch/ppc64/mm/hash_low.S index 8c0156a..c23d469 100644 --- a/arch/ppc64/mm/hash_low.S +++ b/arch/ppc64/mm/hash_low.S @@ -85,7 +85,10 @@ _GLOBAL(__hash_page) bne- htab_wrong_access /* Check if PTE is busy */ andi. r0,r31,_PAGE_BUSY - bne- 1b + /* If so, just bail out and refault if needed. Someone else + * is changing this PTE anyway and might hash it. + */ + bne- bail_ok /* Prepare new PTE value (turn access RW into DIRTY, then * add BUSY,HASHPTE and ACCESSED) */ @@ -215,6 +218,10 @@ _GLOBAL(htab_call_hpte_remove) /* Try all again */ b htab_insert_pte +bail_ok: + li r3,0 + b bail + htab_pte_insert_ok: /* Insert slot number & secondary bit in PTE */ rldimi r30,r3,12,63-15 -- cgit v0.10.2 From a2f95a5ae99eb8209ad8d9faeaada00600bd8027 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Sun, 1 May 2005 08:58:45 -0700 Subject: [PATCH] ppc64: noexec fixes There were a few issues with the ppc64 noexec support: The 64bit ABI has a non executable stack by default. At the moment 64bit apps require a PT_GNU_STACK section in order to have a non executable stack. Disable the read implies exec workaround on the 64bit ABI. The 64bit toolchain has never had problems with incorrect mmap permissions (the 32bit has, thats why we need to retain the workaround). With these fixes as well as a gcc fix from Alan Modra (that was recently committed) 64bit apps work as expected. Signed-off-by: Anton Blanchard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-ppc64/elf.h b/include/asm-ppc64/elf.h index 8457d90..6c42d61 100644 --- a/include/asm-ppc64/elf.h +++ b/include/asm-ppc64/elf.h @@ -229,9 +229,13 @@ do { \ /* * An executable for which elf_read_implies_exec() returns TRUE will - * have the READ_IMPLIES_EXEC personality flag set automatically. + * have the READ_IMPLIES_EXEC personality flag set automatically. This + * is only required to work around bugs in old 32bit toolchains. Since + * the 64bit ABI has never had these issues dont enable the workaround + * even if we have an executable stack. */ -#define elf_read_implies_exec(ex, exec_stk) (exec_stk != EXSTACK_DISABLE_X) +#define elf_read_implies_exec(ex, exec_stk) (test_thread_flag(TIF_32BIT) ? \ + (exec_stk != EXSTACK_DISABLE_X) : 0) #endif diff --git a/include/asm-ppc64/page.h b/include/asm-ppc64/page.h index 20e0f19..8621957 100644 --- a/include/asm-ppc64/page.h +++ b/include/asm-ppc64/page.h @@ -252,10 +252,19 @@ extern u64 ppc64_pft_size; /* Log 2 of page table size */ /* * This is the default if a program doesn't have a PT_GNU_STACK - * program header entry. + * program header entry. The PPC64 ELF ABI has a non executable stack + * stack by default, so in the absense of a PT_GNU_STACK program header + * we turn execute permission off. */ -#define VM_STACK_DEFAULT_FLAGS (VM_READ | VM_WRITE | VM_EXEC | \ - VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) +#define VM_STACK_DEFAULT_FLAGS32 (VM_READ | VM_WRITE | VM_EXEC | \ + VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) + +#define VM_STACK_DEFAULT_FLAGS64 (VM_READ | VM_WRITE | \ + VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) + +#define VM_STACK_DEFAULT_FLAGS \ + (test_thread_flag(TIF_32BIT) ? \ + VM_STACK_DEFAULT_FLAGS32 : VM_STACK_DEFAULT_FLAGS64) #endif /* __KERNEL__ */ #endif /* _PPC64_PAGE_H */ -- cgit v0.10.2 From 4b88e927e8c38f4053680a3098325142017a37f0 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Sun, 1 May 2005 08:58:46 -0700 Subject: [PATCH] ppc64: remove unnecessary include We no longer use any ppcdebug stuff in a.out.h, so remove the define. Signed-off-by: Anton Blanchard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-ppc64/a.out.h b/include/asm-ppc64/a.out.h index 802338e..3871e25 100644 --- a/include/asm-ppc64/a.out.h +++ b/include/asm-ppc64/a.out.h @@ -1,8 +1,6 @@ #ifndef __PPC64_A_OUT_H__ #define __PPC64_A_OUT_H__ -#include - /* * c 2001 PPC 64 Team, IBM Corp * -- cgit v0.10.2 From c4005e4f661a9ec8d91720a6b570865d060afcaf Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Sun, 1 May 2005 08:58:46 -0700 Subject: [PATCH] ppc64: firmware workaround Recent gcc 4.0 testing uncovered a firmware issue. Some properties are larger than 31 bytes and due to gcc 4.0s better stack allocation this overflow ran over non volatile register storage. Signed-off-by: Anton Blanchard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ppc64/kernel/prom_init.c b/arch/ppc64/kernel/prom_init.c index b0b784f..35ec42d 100644 --- a/arch/ppc64/kernel/prom_init.c +++ b/arch/ppc64/kernel/prom_init.c @@ -1555,6 +1555,12 @@ static void __init scan_dt_build_strings(phandle node, unsigned long *mem_start, } } +/* + * The Open Firmware 1275 specification states properties must be 31 bytes or + * less, however not all firmwares obey this. Make it 64 bytes to be safe. + */ +#define MAX_PROPERTY_NAME 64 + static void __init scan_dt_build_struct(phandle node, unsigned long *mem_start, unsigned long *mem_end) { @@ -1564,7 +1570,7 @@ static void __init scan_dt_build_struct(phandle node, unsigned long *mem_start, unsigned long soff; unsigned char *valp; unsigned long offset = reloc_offset(); - char pname[32]; + char pname[MAX_PROPERTY_NAME]; char *path; path = RELOC(prom_scratch); -- cgit v0.10.2 From eeb24de431ac8c80fd13a2c479cd0eb51b70484e Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Sun, 1 May 2005 08:58:46 -0700 Subject: [PATCH] ppc64: enforce medium thread priority in hypervisor calls Calls into the hypervisor do not raise the thread priority. Ensure we are running at medium priority upon entry to the hypervisor. Signed-off-by: Anton Blanchard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ppc64/kernel/pSeries_hvCall.S b/arch/ppc64/kernel/pSeries_hvCall.S index 0715d30..176e8da 100644 --- a/arch/ppc64/kernel/pSeries_hvCall.S +++ b/arch/ppc64/kernel/pSeries_hvCall.S @@ -28,6 +28,8 @@ unsigned long *out3); R10 */ _GLOBAL(plpar_hcall) + HMT_MEDIUM + mfcr r0 std r8,STK_PARM(r8)(r1) /* Save out ptrs */ @@ -53,6 +55,8 @@ _GLOBAL(plpar_hcall) /* Simple interface with no output values (other than status) */ _GLOBAL(plpar_hcall_norets) + HMT_MEDIUM + mfcr r0 stw r0,8(r1) @@ -75,6 +79,8 @@ _GLOBAL(plpar_hcall_norets) unsigned long *out1); 120(R1) */ _GLOBAL(plpar_hcall_8arg_2ret) + HMT_MEDIUM + mfcr r0 ld r11,STK_PARM(r11)(r1) /* put arg8 in R11 */ stw r0,8(r1) @@ -99,6 +105,8 @@ _GLOBAL(plpar_hcall_8arg_2ret) unsigned long *out4); 112(R1) */ _GLOBAL(plpar_hcall_4out) + HMT_MEDIUM + mfcr r0 stw r0,8(r1) -- cgit v0.10.2 From 0d8d4d42f2d00eb65262b49f4edd4cf7ef4eb6fc Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Sun, 1 May 2005 08:58:47 -0700 Subject: [PATCH] ppc64: use smp_mb and smp_wmb Use smp_mb and smp_wmb. In particular smp_wmb is lighter weight than wmb. Signed-off-by: Anton Blanchard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/ppc64/kernel/smp.c b/arch/ppc64/kernel/smp.c index 1c92da3..3b906cd 100644 --- a/arch/ppc64/kernel/smp.c +++ b/arch/ppc64/kernel/smp.c @@ -125,7 +125,7 @@ void __devinit smp_generic_kick_cpu(int nr) * the processor will continue on to secondary_start */ paca[nr].cpu_start = 1; - mb(); + smp_mb(); } #endif /* CONFIG_PPC_MULTIPLATFORM */ @@ -256,7 +256,7 @@ int smp_call_function (void (*func) (void *info), void *info, int nonatomic, } call_data = &data; - wmb(); + smp_wmb(); /* Send a message to all other CPUs and wait for them to respond */ smp_ops->message_pass(MSG_ALL_BUT_SELF, PPC_MSG_CALL_FUNCTION); @@ -431,7 +431,7 @@ int generic_cpu_enable(unsigned int cpu) /* get the target out of it's holding state */ per_cpu(cpu_state, cpu) = CPU_UP_PREPARE; - wmb(); + smp_wmb(); while (!cpu_online(cpu)) cpu_relax(); @@ -447,7 +447,7 @@ void generic_cpu_die(unsigned int cpu) int i; for (i = 0; i < 100; i++) { - rmb(); + smp_rmb(); if (per_cpu(cpu_state, cpu) == CPU_DEAD) return; msleep(100); @@ -463,7 +463,7 @@ void generic_mach_cpu_die(void) cpu = smp_processor_id(); printk(KERN_DEBUG "CPU%d offline\n", cpu); __get_cpu_var(cpu_state) = CPU_DEAD; - wmb(); + smp_wmb(); while (__get_cpu_var(cpu_state) != CPU_UP_PREPARE) cpu_relax(); @@ -515,7 +515,7 @@ int __devinit __cpu_up(unsigned int cpu) * be written out to main store before we release * the processor. */ - mb(); + smp_mb(); /* wake up cpus */ DBG("smp: kicking cpu %d\n", cpu); diff --git a/arch/ppc64/kernel/time.c b/arch/ppc64/kernel/time.c index 77ded5a..772a465 100644 --- a/arch/ppc64/kernel/time.c +++ b/arch/ppc64/kernel/time.c @@ -221,15 +221,15 @@ static __inline__ void timer_recalc_offset(unsigned long cur_tb) temp_varp->tb_to_xs = do_gtod.varp->tb_to_xs; temp_varp->tb_orig_stamp = new_tb_orig_stamp; temp_varp->stamp_xsec = new_stamp_xsec; - mb(); + smp_mb(); do_gtod.varp = temp_varp; do_gtod.var_idx = temp_idx; ++(systemcfg->tb_update_count); - wmb(); + smp_wmb(); systemcfg->tb_orig_stamp = new_tb_orig_stamp; systemcfg->stamp_xsec = new_stamp_xsec; - wmb(); + smp_wmb(); ++(systemcfg->tb_update_count); } @@ -648,7 +648,7 @@ void ppc_adjtimex(void) temp_varp->tb_to_xs = new_tb_to_xs; temp_varp->stamp_xsec = new_stamp_xsec; temp_varp->tb_orig_stamp = do_gtod.varp->tb_orig_stamp; - mb(); + smp_mb(); do_gtod.varp = temp_varp; do_gtod.var_idx = temp_idx; @@ -662,10 +662,10 @@ void ppc_adjtimex(void) * loops back and reads them again until this criteria is met. */ ++(systemcfg->tb_update_count); - wmb(); + smp_wmb(); systemcfg->tb_to_xs = new_tb_to_xs; systemcfg->stamp_xsec = new_stamp_xsec; - wmb(); + smp_wmb(); ++(systemcfg->tb_update_count); write_sequnlock_irqrestore( &xtime_lock, flags ); -- cgit v0.10.2 From d59dd4620fb8d6422555a9e2b82a707718e68327 Mon Sep 17 00:00:00 2001 From: "akpm@osdl.org" Date: Sun, 1 May 2005 08:58:47 -0700 Subject: [PATCH] use smp_mb/wmb/rmb where possible Replace a number of memory barriers with smp_ variants. This means we won't take the unnecessary hit on UP machines. Signed-off-by: Anton Blanchard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/buffer.c b/fs/buffer.c index 665db84..188365c 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -218,7 +218,7 @@ struct super_block *freeze_bdev(struct block_device *bdev) sb = get_super(bdev); if (sb && !(sb->s_flags & MS_RDONLY)) { sb->s_frozen = SB_FREEZE_WRITE; - wmb(); + smp_wmb(); sync_inodes_sb(sb, 0); DQUOT_SYNC(sb); @@ -235,7 +235,7 @@ struct super_block *freeze_bdev(struct block_device *bdev) sync_inodes_sb(sb, 1); sb->s_frozen = SB_FREEZE_TRANS; - wmb(); + smp_wmb(); sync_blockdev(sb->s_bdev); @@ -263,7 +263,7 @@ void thaw_bdev(struct block_device *bdev, struct super_block *sb) if (sb->s_op->unlockfs) sb->s_op->unlockfs(sb); sb->s_frozen = SB_UNFROZEN; - wmb(); + smp_wmb(); wake_up(&sb->s_wait_unfrozen); drop_super(sb); } diff --git a/ipc/mqueue.c b/ipc/mqueue.c index cb0cd3c..33f7152 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -767,7 +767,7 @@ static inline void pipelined_send(struct mqueue_inode_info *info, list_del(&receiver->list); receiver->state = STATE_PENDING; wake_up_process(receiver->task); - wmb(); + smp_wmb(); receiver->state = STATE_READY; } @@ -786,7 +786,7 @@ static inline void pipelined_receive(struct mqueue_inode_info *info) list_del(&sender->list); sender->state = STATE_PENDING; wake_up_process(sender->task); - wmb(); + smp_wmb(); sender->state = STATE_READY; } diff --git a/kernel/kthread.c b/kernel/kthread.c index e377e22..f50f174 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -174,7 +174,7 @@ int kthread_stop(struct task_struct *k) /* Must init completion *before* thread sees kthread_stop_info.k */ init_completion(&kthread_stop_info.done); - wmb(); + smp_wmb(); /* Now set kthread_should_stop() to true, and wake it up. */ kthread_stop_info.k = k; diff --git a/kernel/profile.c b/kernel/profile.c index a38fa70..a66be46 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -522,7 +522,7 @@ static int __init create_hash_tables(void) return 0; out_cleanup: prof_on = 0; - mb(); + smp_mb(); on_each_cpu(profile_nop, NULL, 0, 1); for_each_online_cpu(cpu) { struct page *page; diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 88b306c..f5cc1ce 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -135,7 +135,7 @@ int ptrace_attach(struct task_struct *task) (current->gid != task->sgid) || (current->gid != task->gid)) && !capable(CAP_SYS_PTRACE)) goto bad; - rmb(); + smp_rmb(); if (!task->mm->dumpable && !capable(CAP_SYS_PTRACE)) goto bad; /* the same process cannot be attached many times */ diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index c39ed70..6116b25 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -33,7 +33,7 @@ static int stopmachine(void *cpu) set_cpus_allowed(current, cpumask_of_cpu((int)(long)cpu)); /* Ack: we are alive */ - mb(); /* Theoretically the ack = 0 might not be on this CPU yet. */ + smp_mb(); /* Theoretically the ack = 0 might not be on this CPU yet. */ atomic_inc(&stopmachine_thread_ack); /* Simple state machine */ @@ -43,14 +43,14 @@ static int stopmachine(void *cpu) local_irq_disable(); irqs_disabled = 1; /* Ack: irqs disabled. */ - mb(); /* Must read state first. */ + smp_mb(); /* Must read state first. */ atomic_inc(&stopmachine_thread_ack); } else if (stopmachine_state == STOPMACHINE_PREPARE && !prepared) { /* Everyone is in place, hold CPU. */ preempt_disable(); prepared = 1; - mb(); /* Must read state first. */ + smp_mb(); /* Must read state first. */ atomic_inc(&stopmachine_thread_ack); } /* Yield in first stage: migration threads need to @@ -62,7 +62,7 @@ static int stopmachine(void *cpu) } /* Ack: we are exiting. */ - mb(); /* Must read state first. */ + smp_mb(); /* Must read state first. */ atomic_inc(&stopmachine_thread_ack); if (irqs_disabled) @@ -77,7 +77,7 @@ static int stopmachine(void *cpu) static void stopmachine_set_state(enum stopmachine_state state) { atomic_set(&stopmachine_thread_ack, 0); - wmb(); + smp_wmb(); stopmachine_state = state; while (atomic_read(&stopmachine_thread_ack) != stopmachine_num_threads) cpu_relax(); diff --git a/kernel/sys.c b/kernel/sys.c index 462d78d..df2ddcc 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -525,7 +525,7 @@ asmlinkage long sys_setregid(gid_t rgid, gid_t egid) if (new_egid != old_egid) { current->mm->dumpable = 0; - wmb(); + smp_wmb(); } if (rgid != (gid_t) -1 || (egid != (gid_t) -1 && egid != old_rgid)) @@ -556,7 +556,7 @@ asmlinkage long sys_setgid(gid_t gid) if(old_egid != gid) { current->mm->dumpable=0; - wmb(); + smp_wmb(); } current->gid = current->egid = current->sgid = current->fsgid = gid; } @@ -565,7 +565,7 @@ asmlinkage long sys_setgid(gid_t gid) if(old_egid != gid) { current->mm->dumpable=0; - wmb(); + smp_wmb(); } current->egid = current->fsgid = gid; } @@ -596,7 +596,7 @@ static int set_user(uid_t new_ruid, int dumpclear) if(dumpclear) { current->mm->dumpable = 0; - wmb(); + smp_wmb(); } current->uid = new_ruid; return 0; @@ -653,7 +653,7 @@ asmlinkage long sys_setreuid(uid_t ruid, uid_t euid) if (new_euid != old_euid) { current->mm->dumpable=0; - wmb(); + smp_wmb(); } current->fsuid = current->euid = new_euid; if (ruid != (uid_t) -1 || @@ -703,7 +703,7 @@ asmlinkage long sys_setuid(uid_t uid) if (old_euid != uid) { current->mm->dumpable = 0; - wmb(); + smp_wmb(); } current->fsuid = current->euid = uid; current->suid = new_suid; @@ -748,7 +748,7 @@ asmlinkage long sys_setresuid(uid_t ruid, uid_t euid, uid_t suid) if (euid != current->euid) { current->mm->dumpable = 0; - wmb(); + smp_wmb(); } current->euid = euid; } @@ -798,7 +798,7 @@ asmlinkage long sys_setresgid(gid_t rgid, gid_t egid, gid_t sgid) if (egid != current->egid) { current->mm->dumpable = 0; - wmb(); + smp_wmb(); } current->egid = egid; } @@ -845,7 +845,7 @@ asmlinkage long sys_setfsuid(uid_t uid) if (uid != old_fsuid) { current->mm->dumpable = 0; - wmb(); + smp_wmb(); } current->fsuid = uid; } @@ -875,7 +875,7 @@ asmlinkage long sys_setfsgid(gid_t gid) if (gid != old_fsgid) { current->mm->dumpable = 0; - wmb(); + smp_wmb(); } current->fsgid = gid; key_fsgid_changed(current); diff --git a/kernel/timer.c b/kernel/timer.c index ecb3d67..207aa4f0 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -1007,7 +1007,7 @@ asmlinkage long sys_getppid(void) * Make sure we read the pid before re-reading the * parent pointer: */ - rmb(); + smp_rmb(); parent = me->group_leader->real_parent; if (old != parent) continue; diff --git a/lib/rwsem-spinlock.c b/lib/rwsem-spinlock.c index 21f0db2..40ffde9 100644 --- a/lib/rwsem-spinlock.c +++ b/lib/rwsem-spinlock.c @@ -76,7 +76,7 @@ __rwsem_do_wake(struct rw_semaphore *sem, int wakewrite) list_del(&waiter->list); tsk = waiter->task; /* Don't touch waiter after ->task has been NULLed */ - mb(); + smp_mb(); waiter->task = NULL; wake_up_process(tsk); put_task_struct(tsk); @@ -91,7 +91,7 @@ __rwsem_do_wake(struct rw_semaphore *sem, int wakewrite) list_del(&waiter->list); tsk = waiter->task; - mb(); + smp_mb(); waiter->task = NULL; wake_up_process(tsk); put_task_struct(tsk); @@ -123,7 +123,7 @@ __rwsem_wake_one_writer(struct rw_semaphore *sem) list_del(&waiter->list); tsk = waiter->task; - mb(); + smp_mb(); waiter->task = NULL; wake_up_process(tsk); put_task_struct(tsk); diff --git a/lib/rwsem.c b/lib/rwsem.c index 7644089..62fa4eb 100644 --- a/lib/rwsem.c +++ b/lib/rwsem.c @@ -74,7 +74,7 @@ __rwsem_do_wake(struct rw_semaphore *sem, int downgrading) */ list_del(&waiter->list); tsk = waiter->task; - mb(); + smp_mb(); waiter->task = NULL; wake_up_process(tsk); put_task_struct(tsk); @@ -117,7 +117,7 @@ __rwsem_do_wake(struct rw_semaphore *sem, int downgrading) waiter = list_entry(next, struct rwsem_waiter, list); next = waiter->list.next; tsk = waiter->task; - mb(); + smp_mb(); waiter->task = NULL; wake_up_process(tsk); put_task_struct(tsk); diff --git a/mm/mempool.c b/mm/mempool.c index e9a0a63..c9f3d46 100644 --- a/mm/mempool.c +++ b/mm/mempool.c @@ -229,7 +229,7 @@ repeat_alloc: /* Now start performing page reclaim */ gfp_temp = gfp_mask; prepare_to_wait(&pool->wait, &wait, TASK_UNINTERRUPTIBLE); - mb(); + smp_mb(); if (!pool->curr_nr) io_schedule(); finish_wait(&pool->wait, &wait); @@ -250,7 +250,7 @@ void mempool_free(void *element, mempool_t *pool) { unsigned long flags; - mb(); + smp_mb(); if (pool->curr_nr < pool->min_nr) { spin_lock_irqsave(&pool->lock, flags); if (pool->curr_nr < pool->min_nr) { -- cgit v0.10.2 From d637413f3f05b41f678f8004225b33b62274183f Mon Sep 17 00:00:00 2001 From: Jake Moilanen Date: Sun, 1 May 2005 08:58:47 -0700 Subject: [PATCH] ppc64: reverse prediction on spinlock busy loop code On our raw spinlocks, we currently have an attempt at the lock, and if we do not get it we enter a spin loop. This spinloop will likely continue for awhile, and we pridict likely. Shouldn't we predict that we will get out of the loop so our next instructions are already prefetched. Even when we miss because the lock is still held, it won't matter since we are waiting anyways. I did a couple quick benchmarks, but the results are inconclusive. 16-way 690 running specjbb with original code # ./specjbb 3000 16 1 1 19 30 120 ... Valid run, Score is 59282 16-way 690 running specjbb with unlikely code # ./specjbb 3000 16 1 1 19 30 120 ... Valid run, Score is 59541 I saw a smaller increase on a JS20 (~1.6%) JS20 specjbb w/ original code # ./specjbb 400 2 1 1 19 30 120 ... Valid run, Score is 20460 JS20 specjbb w/ unlikely code # ./specjbb 400 2 1 1 19 30 120 ... Valid run, Score is 20803 Anton said: Mispredicting the spinlock busy loop also means we slow down the rate at which we do the loads which can be good for heavily contended locks. Note: There are some gcc issues with our default build and branch prediction, but a CONFIG_POWER4_ONLY build should emit them correctly. I'm working with Alan Modra on it now. Signed-off-by: Jake Moilanen Signed-off-by: Anton Blanchard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-ppc64/spinlock.h b/include/asm-ppc64/spinlock.h index a9b2a11..acd1156 100644 --- a/include/asm-ppc64/spinlock.h +++ b/include/asm-ppc64/spinlock.h @@ -110,7 +110,7 @@ static void __inline__ _raw_spin_lock(spinlock_t *lock) HMT_low(); if (SHARED_PROCESSOR) __spin_yield(lock); - } while (likely(lock->lock != 0)); + } while (unlikely(lock->lock != 0)); HMT_medium(); } } @@ -128,7 +128,7 @@ static void __inline__ _raw_spin_lock_flags(spinlock_t *lock, unsigned long flag HMT_low(); if (SHARED_PROCESSOR) __spin_yield(lock); - } while (likely(lock->lock != 0)); + } while (unlikely(lock->lock != 0)); HMT_medium(); local_irq_restore(flags_dis); } @@ -194,7 +194,7 @@ static void __inline__ _raw_read_lock(rwlock_t *rw) HMT_low(); if (SHARED_PROCESSOR) __rw_yield(rw); - } while (likely(rw->lock < 0)); + } while (unlikely(rw->lock < 0)); HMT_medium(); } } @@ -251,7 +251,7 @@ static void __inline__ _raw_write_lock(rwlock_t *rw) HMT_low(); if (SHARED_PROCESSOR) __rw_yield(rw); - } while (likely(rw->lock != 0)); + } while (unlikely(rw->lock != 0)); HMT_medium(); } } -- cgit v0.10.2 From d5b63d78f1e75f6c6f04862dfb2f2a4aeffafd4c Mon Sep 17 00:00:00 2001 From: Denis Vlasenko Date: Sun, 1 May 2005 08:58:48 -0700 Subject: [PATCH] fix i386 memcpy This patch shortens non-constant memcpy() by two bytes and fixes spurious out-of-line constant memcpy(). # size vmlinux.org vmlinux text data bss dec hex filename 3954591 1553426 236544 5744561 57a7b1 vmlinux.org 3952615 1553426 236544 5742585 579ff9 vmlinux Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-i386/string.h b/include/asm-i386/string.h index 1679983..6a78ac5 100644 --- a/include/asm-i386/string.h +++ b/include/asm-i386/string.h @@ -198,47 +198,80 @@ static inline void * __memcpy(void * to, const void * from, size_t n) int d0, d1, d2; __asm__ __volatile__( "rep ; movsl\n\t" - "testb $2,%b4\n\t" - "je 1f\n\t" - "movsw\n" - "1:\ttestb $1,%b4\n\t" - "je 2f\n\t" - "movsb\n" - "2:" + "movl %4,%%ecx\n\t" + "andl $3,%%ecx\n\t" +#if 1 /* want to pay 2 byte penalty for a chance to skip microcoded rep? */ + "jz 1f\n\t" +#endif + "rep ; movsb\n\t" + "1:" : "=&c" (d0), "=&D" (d1), "=&S" (d2) - :"0" (n/4), "q" (n),"1" ((long) to),"2" ((long) from) + : "0" (n/4), "g" (n), "1" ((long) to), "2" ((long) from) : "memory"); return (to); } /* - * This looks horribly ugly, but the compiler can optimize it totally, + * This looks ugly, but the compiler can optimize it totally, * as the count is constant. */ static inline void * __constant_memcpy(void * to, const void * from, size_t n) { - if (n <= 128) - return __builtin_memcpy(to, from, n); - -#define COMMON(x) \ -__asm__ __volatile__( \ - "rep ; movsl" \ - x \ - : "=&c" (d0), "=&D" (d1), "=&S" (d2) \ - : "0" (n/4),"1" ((long) to),"2" ((long) from) \ - : "memory"); -{ - int d0, d1, d2; + long esi, edi; + if (!n) return to; +#if 1 /* want to do small copies with non-string ops? */ + switch (n) { + case 1: *(char*)to = *(char*)from; return to; + case 2: *(short*)to = *(short*)from; return to; + case 4: *(int*)to = *(int*)from; return to; +#if 1 /* including those doable with two moves? */ + case 3: *(short*)to = *(short*)from; + *((char*)to+2) = *((char*)from+2); return to; + case 5: *(int*)to = *(int*)from; + *((char*)to+4) = *((char*)from+4); return to; + case 6: *(int*)to = *(int*)from; + *((short*)to+2) = *((short*)from+2); return to; + case 8: *(int*)to = *(int*)from; + *((int*)to+1) = *((int*)from+1); return to; +#endif + } +#endif + esi = (long) from; + edi = (long) to; + if (n >= 5*4) { + /* large block: use rep prefix */ + int ecx; + __asm__ __volatile__( + "rep ; movsl" + : "=&c" (ecx), "=&D" (edi), "=&S" (esi) + : "0" (n/4), "1" (edi),"2" (esi) + : "memory" + ); + } else { + /* small block: don't clobber ecx + smaller code */ + if (n >= 4*4) __asm__ __volatile__("movsl" + :"=&D"(edi),"=&S"(esi):"0"(edi),"1"(esi):"memory"); + if (n >= 3*4) __asm__ __volatile__("movsl" + :"=&D"(edi),"=&S"(esi):"0"(edi),"1"(esi):"memory"); + if (n >= 2*4) __asm__ __volatile__("movsl" + :"=&D"(edi),"=&S"(esi):"0"(edi),"1"(esi):"memory"); + if (n >= 1*4) __asm__ __volatile__("movsl" + :"=&D"(edi),"=&S"(esi):"0"(edi),"1"(esi):"memory"); + } switch (n % 4) { - case 0: COMMON(""); return to; - case 1: COMMON("\n\tmovsb"); return to; - case 2: COMMON("\n\tmovsw"); return to; - default: COMMON("\n\tmovsw\n\tmovsb"); return to; + /* tail */ + case 0: return to; + case 1: __asm__ __volatile__("movsb" + :"=&D"(edi),"=&S"(esi):"0"(edi),"1"(esi):"memory"); + return to; + case 2: __asm__ __volatile__("movsw" + :"=&D"(edi),"=&S"(esi):"0"(edi),"1"(esi):"memory"); + return to; + default: __asm__ __volatile__("movsw\n\tmovsb" + :"=&D"(edi),"=&S"(esi):"0"(edi),"1"(esi):"memory"); + return to; } } - -#undef COMMON -} #define __HAVE_ARCH_MEMCPY -- cgit v0.10.2 From fd51f666fa591294bd7462447512666e61c56ea0 Mon Sep 17 00:00:00 2001 From: "H. J. Lu" Date: Sun, 1 May 2005 08:58:48 -0700 Subject: [PATCH] i386/x86_64 segment register access update The new i386/x86_64 assemblers no longer accept instructions for moving between a segment register and a 32bit memory location, i.e., movl (%eax),%ds movl %ds,(%eax) To generate instructions for moving between a segment register and a 16bit memory location without the 16bit operand size prefix, 0x66, mov (%eax),%ds mov %ds,(%eax) should be used. It will work with both new and old assemblers. The assembler starting from 2.16.90.0.1 will also support movw (%eax),%ds movw %ds,(%eax) without the 0x66 prefix. I am enclosing patches for 2.4 and 2.6 kernels here. The resulting kernel binaries should be unchanged as before, with old and new assemblers, if gcc never generates memory access for unsigned gsindex; asm volatile("movl %%gs,%0" : "=g" (gsindex)); If gcc does generate memory access for the code above, the upper bits in gsindex are undefined and the new assembler doesn't allow it. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index b2203e2..85bd56d 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -611,8 +611,8 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas * Save away %fs and %gs. No need to save %es and %ds, as * those are always kernel segments while inside the kernel. */ - asm volatile("movl %%fs,%0":"=m" (*(int *)&prev->fs)); - asm volatile("movl %%gs,%0":"=m" (*(int *)&prev->gs)); + asm volatile("mov %%fs,%0":"=m" (prev->fs)); + asm volatile("mov %%gs,%0":"=m" (prev->gs)); /* * Restore %fs and %gs if needed. diff --git a/arch/i386/kernel/vm86.c b/arch/i386/kernel/vm86.c index 2f3d52d..d16cd37 100644 --- a/arch/i386/kernel/vm86.c +++ b/arch/i386/kernel/vm86.c @@ -294,8 +294,8 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk */ info->regs32->eax = 0; tsk->thread.saved_esp0 = tsk->thread.esp0; - asm volatile("movl %%fs,%0":"=m" (tsk->thread.saved_fs)); - asm volatile("movl %%gs,%0":"=m" (tsk->thread.saved_gs)); + asm volatile("mov %%fs,%0":"=m" (tsk->thread.saved_fs)); + asm volatile("mov %%gs,%0":"=m" (tsk->thread.saved_gs)); tss = &per_cpu(init_tss, get_cpu()); tsk->thread.esp0 = (unsigned long) &info->VM86_TSS_ESP0; diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index 9922d2b..761b6d3 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -402,10 +402,10 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long rsp, p->thread.fs = me->thread.fs; p->thread.gs = me->thread.gs; - asm("movl %%gs,%0" : "=m" (p->thread.gsindex)); - asm("movl %%fs,%0" : "=m" (p->thread.fsindex)); - asm("movl %%es,%0" : "=m" (p->thread.es)); - asm("movl %%ds,%0" : "=m" (p->thread.ds)); + asm("mov %%gs,%0" : "=m" (p->thread.gsindex)); + asm("mov %%fs,%0" : "=m" (p->thread.fsindex)); + asm("mov %%es,%0" : "=m" (p->thread.es)); + asm("mov %%ds,%0" : "=m" (p->thread.ds)); if (unlikely(me->thread.io_bitmap_ptr != NULL)) { p->thread.io_bitmap_ptr = kmalloc(IO_BITMAP_BYTES, GFP_KERNEL); @@ -468,11 +468,11 @@ struct task_struct *__switch_to(struct task_struct *prev_p, struct task_struct * * Switch DS and ES. * This won't pick up thread selector changes, but I guess that is ok. */ - asm volatile("movl %%es,%0" : "=m" (prev->es)); + asm volatile("mov %%es,%0" : "=m" (prev->es)); if (unlikely(next->es | prev->es)) loadsegment(es, next->es); - asm volatile ("movl %%ds,%0" : "=m" (prev->ds)); + asm volatile ("mov %%ds,%0" : "=m" (prev->ds)); if (unlikely(next->ds | prev->ds)) loadsegment(ds, next->ds); diff --git a/include/asm-i386/system.h b/include/asm-i386/system.h index 6f74d4c..3db717a 100644 --- a/include/asm-i386/system.h +++ b/include/asm-i386/system.h @@ -81,7 +81,7 @@ static inline unsigned long _get_base(char * addr) #define loadsegment(seg,value) \ asm volatile("\n" \ "1:\t" \ - "movl %0,%%" #seg "\n" \ + "mov %0,%%" #seg "\n" \ "2:\n" \ ".section .fixup,\"ax\"\n" \ "3:\t" \ @@ -93,13 +93,13 @@ static inline unsigned long _get_base(char * addr) ".align 4\n\t" \ ".long 1b,3b\n" \ ".previous" \ - : :"m" (*(unsigned int *)&(value))) + : :"m" (value)) /* * Save a segment register away */ #define savesegment(seg, value) \ - asm volatile("movl %%" #seg ",%0":"=m" (*(int *)&(value))) + asm volatile("mov %%" #seg ",%0":"=m" (value)) /* * Clear and set 'TS' bit respectively -- cgit v0.10.2 From 67701ae9767534534d3710664037dfde2cc04935 Mon Sep 17 00:00:00 2001 From: Jack F Vogel Date: Sun, 1 May 2005 08:58:48 -0700 Subject: [PATCH] check nmi watchdog is broken A bug against an xSeries system showed up recently noting that the check_nmi_watchdog() test was failing. I have been investigating it and discovered in both i386 and x86_64 the recent change to the routine to use the cpu_callin_map has uncovered a problem. Prior to that change, on an SMP box, the test was trivally passing because all cpu's were found to not yet be online, but now with the callin_map they are discovered, it goes on to test the counter and they have not yet begun to increment, so it announces a CPU is stuck and bails out. On all the systems I have access to test, the announcement of failure is also bougs... by the time you can login and check /proc/interrupts, the NMI count is happily incrementing on all CPUs. Its just that the test is being done too early. I have tried moving the call to the test around a bit, and it was always too early. I finally hit on this proposed solution, it delays the routine via a late_initcall(), seems like the right solution to me. Signed-off-by: Adrian Bunk Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c index e3879f7..d509836 100644 --- a/arch/i386/kernel/apic.c +++ b/arch/i386/kernel/apic.c @@ -1265,8 +1265,6 @@ int __init APIC_init_uniprocessor (void) setup_local_APIC(); - if (nmi_watchdog == NMI_LOCAL_APIC) - check_nmi_watchdog(); #ifdef CONFIG_X86_IO_APIC if (smp_found_config) if (!skip_ioapic_setup && nr_ioapics) diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index 5e0d55b..7a324e8 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -2175,7 +2175,6 @@ static inline void check_timer(void) disable_8259A_irq(0); setup_nmi(); enable_8259A_irq(0); - check_nmi_watchdog(); } return; } @@ -2198,7 +2197,6 @@ static inline void check_timer(void) add_pin_to_irq(0, 0, pin2); if (nmi_watchdog == NMI_IO_APIC) { setup_nmi(); - check_nmi_watchdog(); } return; } diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c index 2f89d00..2c0ee9c 100644 --- a/arch/i386/kernel/nmi.c +++ b/arch/i386/kernel/nmi.c @@ -102,20 +102,21 @@ int nmi_active; (P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT| \ P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE) -int __init check_nmi_watchdog (void) +static int __init check_nmi_watchdog(void) { unsigned int prev_nmi_count[NR_CPUS]; int cpu; - printk(KERN_INFO "testing NMI watchdog ... "); + if (nmi_watchdog == NMI_NONE) + return 0; + + printk(KERN_INFO "Testing NMI watchdog ... "); for (cpu = 0; cpu < NR_CPUS; cpu++) prev_nmi_count[cpu] = per_cpu(irq_stat, cpu).__nmi_count; local_irq_enable(); mdelay((10*1000)/nmi_hz); // wait 10 ticks - /* FIXME: Only boot CPU is online at this stage. Check CPUs - as they come up. */ for (cpu = 0; cpu < NR_CPUS; cpu++) { #ifdef CONFIG_SMP /* Check cpu_callin_map here because that is set @@ -139,6 +140,8 @@ int __init check_nmi_watchdog (void) return 0; } +/* This needs to happen later in boot so counters are working */ +late_initcall(check_nmi_watchdog); static int __init setup_nmi_watchdog(char *str) { diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c index fd36d2f..cbea7ac 100644 --- a/arch/i386/kernel/smpboot.c +++ b/arch/i386/kernel/smpboot.c @@ -1089,9 +1089,6 @@ static void __init smp_boot_cpus(unsigned int max_cpus) } } - if (nmi_watchdog == NMI_LOCAL_APIC) - check_nmi_watchdog(); - smpboot_setup_io_apic(); setup_boot_APIC_clock(); diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c index 29a2572..60be586 100644 --- a/arch/x86_64/kernel/io_apic.c +++ b/arch/x86_64/kernel/io_apic.c @@ -1607,7 +1607,6 @@ static inline void check_timer(void) disable_8259A_irq(0); setup_nmi(); enable_8259A_irq(0); - check_nmi_watchdog(); } return; } @@ -1627,7 +1626,6 @@ static inline void check_timer(void) nmi_watchdog_default(); if (nmi_watchdog == NMI_IO_APIC) { setup_nmi(); - check_nmi_watchdog(); } return; } diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c index e00d4ad..61de0b3 100644 --- a/arch/x86_64/kernel/nmi.c +++ b/arch/x86_64/kernel/nmi.c @@ -112,17 +112,20 @@ static __init int cpu_has_lapic(void) } } -int __init check_nmi_watchdog (void) +static int __init check_nmi_watchdog (void) { int counts[NR_CPUS]; int cpu; + if (nmi_watchdog == NMI_NONE) + return 0; + if (nmi_watchdog == NMI_LOCAL_APIC && !cpu_has_lapic()) { nmi_watchdog = NMI_NONE; return -1; } - printk(KERN_INFO "testing NMI watchdog ... "); + printk(KERN_INFO "Testing NMI watchdog ... "); for (cpu = 0; cpu < NR_CPUS; cpu++) counts[cpu] = cpu_pda[cpu].__nmi_count; @@ -148,6 +151,8 @@ int __init check_nmi_watchdog (void) return 0; } +/* Have this called later during boot so counters are updating */ +late_initcall(check_nmi_watchdog); int __init setup_nmi_watchdog(char *str) { diff --git a/include/asm-i386/apic.h b/include/asm-i386/apic.h index e1de674..a5810cf 100644 --- a/include/asm-i386/apic.h +++ b/include/asm-i386/apic.h @@ -109,7 +109,6 @@ extern int APIC_init_uniprocessor (void); extern void disable_APIC_timer(void); extern void enable_APIC_timer(void); -extern int check_nmi_watchdog (void); extern void enable_NMI_through_LVT0 (void * dummy); extern unsigned int nmi_watchdog; diff --git a/include/asm-x86_64/apic.h b/include/asm-x86_64/apic.h index c025cc3..e4b1017 100644 --- a/include/asm-x86_64/apic.h +++ b/include/asm-x86_64/apic.h @@ -99,7 +99,6 @@ extern void disable_APIC_timer(void); extern void enable_APIC_timer(void); extern void clustered_apic_check(void); -extern int check_nmi_watchdog(void); extern void nmi_watchdog_default(void); extern int setup_nmi_watchdog(char *); -- cgit v0.10.2 From a2f7c354159b87dfbd9900f597d48d18755a9d16 Mon Sep 17 00:00:00 2001 From: Jaya Kumar Date: Sun, 1 May 2005 08:58:49 -0700 Subject: [PATCH] x86 reboot: Add reboot fixup for gx1/cs5530a This patch by Jaya Kumar introduces a generic infrastructure to deal with x86 chipsets with nonstandard reset sequences, and adds support for the Geode gx1/cs5530a chipset. Signed-off-by: Jaya Kumar Signed-off-by: H. Peter Anvin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index 17a0cbc..99b4f29 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -653,6 +653,24 @@ config I8K Say Y if you intend to run this kernel on a Dell Inspiron 8000. Say N otherwise. +config X86_REBOOTFIXUPS + bool "Enable X86 board specific fixups for reboot" + depends on X86 + default n + ---help--- + This enables chipset and/or board specific fixups to be done + in order to get reboot to work correctly. This is only needed on + some combinations of hardware and BIOS. The symptom, for which + this config is intended, is when reboot ends with a stalled/hung + system. + + Currently, the only fixup is for the Geode GX1/CS5530A/TROM2.1. + combination. + + Say Y if you want to enable the fixup. Currently, it's safe to + enable this option even if you don't need it. + Say N otherwise. + config MICROCODE tristate "/dev/cpu/microcode - Intel IA32 CPU microcode support" ---help--- diff --git a/arch/i386/kernel/Makefile b/arch/i386/kernel/Makefile index aacdae6..0fbcfe0 100644 --- a/arch/i386/kernel/Makefile +++ b/arch/i386/kernel/Makefile @@ -23,6 +23,7 @@ obj-$(CONFIG_X86_TRAMPOLINE) += trampoline.o obj-$(CONFIG_X86_MPPARSE) += mpparse.o obj-$(CONFIG_X86_LOCAL_APIC) += apic.o nmi.o obj-$(CONFIG_X86_IO_APIC) += io_apic.o +obj-$(CONFIG_X86_REBOOTFIXUPS) += reboot_fixups.o obj-$(CONFIG_X86_NUMAQ) += numaq.o obj-$(CONFIG_X86_SUMMIT_NUMA) += summit.o obj-$(CONFIG_KPROBES) += kprobes.o diff --git a/arch/i386/kernel/reboot.c b/arch/i386/kernel/reboot.c index 3d7e994..6dc27eb 100644 --- a/arch/i386/kernel/reboot.c +++ b/arch/i386/kernel/reboot.c @@ -13,6 +13,7 @@ #include #include #include "mach_reboot.h" +#include /* * Power off function, if any @@ -348,6 +349,7 @@ void machine_restart(char * __unused) /* rebooting needs to touch the page at absolute addr 0 */ *((unsigned short *)__va(0x472)) = reboot_mode; for (;;) { + mach_reboot_fixups(); /* for board specific fixups */ mach_reboot(); /* That didn't work - force a triple fault.. */ __asm__ __volatile__("lidt %0": :"m" (no_idt)); diff --git a/arch/i386/kernel/reboot_fixups.c b/arch/i386/kernel/reboot_fixups.c new file mode 100644 index 0000000..1b183b3 --- /dev/null +++ b/arch/i386/kernel/reboot_fixups.c @@ -0,0 +1,56 @@ +/* + * linux/arch/i386/kernel/reboot_fixups.c + * + * This is a good place to put board specific reboot fixups. + * + * List of supported fixups: + * geode-gx1/cs5530a - Jaya Kumar + * + */ + +#include +#include + +static void cs5530a_warm_reset(struct pci_dev *dev) +{ + /* writing 1 to the reset control register, 0x44 causes the + cs5530a to perform a system warm reset */ + pci_write_config_byte(dev, 0x44, 0x1); + udelay(50); /* shouldn't get here but be safe and spin-a-while */ + return; +} + +struct device_fixup { + unsigned int vendor; + unsigned int device; + void (*reboot_fixup)(struct pci_dev *); +}; + +static struct device_fixup fixups_table[] = { +{ PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5530_LEGACY, cs5530a_warm_reset }, +}; + +/* + * we see if any fixup is available for our current hardware. if there + * is a fixup, we call it and we expect to never return from it. if we + * do return, we keep looking and then eventually fall back to the + * standard mach_reboot on return. + */ +void mach_reboot_fixups(void) +{ + struct device_fixup *cur; + struct pci_dev *dev; + int i; + + for (i=0; i < (sizeof(fixups_table)/sizeof(fixups_table[0])); i++) { + cur = &(fixups_table[i]); + dev = pci_get_device(cur->vendor, cur->device, 0); + if (!dev) + continue; + + cur->reboot_fixup(dev); + } + + printk(KERN_WARNING "No reboot fixup found for your hardware\n"); +} + diff --git a/include/linux/reboot_fixups.h b/include/linux/reboot_fixups.h new file mode 100644 index 0000000..480ea2d --- /dev/null +++ b/include/linux/reboot_fixups.h @@ -0,0 +1,10 @@ +#ifndef _LINUX_REBOOT_FIXUPS_H +#define _LINUX_REBOOT_FIXUPS_H + +#ifdef CONFIG_X86_REBOOTFIXUPS +extern void mach_reboot_fixups(void); +#else +#define mach_reboot_fixups() ((void)(0)) +#endif + +#endif /* _LINUX_REBOOT_FIXUPS_H */ -- cgit v0.10.2 From 48c88211a65bc30f030fe228bda30fda870deb90 Mon Sep 17 00:00:00 2001 From: Stas Sergeev Date: Sun, 1 May 2005 08:58:49 -0700 Subject: [PATCH] x86: entry.S trap return fixes do_debug() and do_int3() return void. This patch fixes the CONFIG_KPROBES variant of do_int3() to return void too and adjusts entry.S accordingly. Signed-off-by: Stas Sergeev Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S index fe1918c..25bf758 100644 --- a/arch/i386/kernel/entry.S +++ b/arch/i386/kernel/entry.S @@ -514,8 +514,6 @@ debug_stack_correct: xorl %edx,%edx # error code 0 movl %esp,%eax # pt_regs pointer call do_debug - testl %eax,%eax - jnz restore_all jmp ret_from_exception /* @@ -596,8 +594,6 @@ ENTRY(int3) xorl %edx,%edx # zero error code movl %esp,%eax # pt_regs pointer call do_int3 - testl %eax,%eax - jnz restore_all jmp ret_from_exception ENTRY(overflow) diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index d708194..00c6341 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -643,16 +643,15 @@ void unset_nmi_callback(void) } #ifdef CONFIG_KPROBES -fastcall int do_int3(struct pt_regs *regs, long error_code) +fastcall void do_int3(struct pt_regs *regs, long error_code) { if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP) == NOTIFY_STOP) - return 1; + return; /* This is an interrupt gate, because kprobes wants interrupts disabled. Normal trap handlers don't. */ restore_interrupts(regs); do_trap(3, SIGTRAP, "int3", 1, regs, error_code, NULL); - return 0; } #endif -- cgit v0.10.2 From a6954ba2e8d344a07e066737827116eb7bc0fdcd Mon Sep 17 00:00:00 2001 From: Lee Revell Date: Sun, 1 May 2005 08:58:49 -0700 Subject: [PATCH] Enable write combining for server works LE rev > 6 Enable write combining for server works LE rev > 6 per http://www.ussg.iu.edu/hypermail/linux/kernel/0104.3/1007.html Signed-Off-By: Lee Revell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/cpu/mtrr/main.c b/arch/i386/kernel/cpu/mtrr/main.c index 54999e4..e1c2042 100644 --- a/arch/i386/kernel/cpu/mtrr/main.c +++ b/arch/i386/kernel/cpu/mtrr/main.c @@ -72,17 +72,21 @@ void set_mtrr_ops(struct mtrr_ops * ops) static int have_wrcomb(void) { struct pci_dev *dev; + u8 rev; if ((dev = pci_get_class(PCI_CLASS_BRIDGE_HOST << 8, NULL)) != NULL) { - /* ServerWorks LE chipsets have problems with write-combining + /* ServerWorks LE chipsets < rev 6 have problems with write-combining Don't allow it and leave room for other chipsets to be tagged */ if (dev->vendor == PCI_VENDOR_ID_SERVERWORKS && dev->device == PCI_DEVICE_ID_SERVERWORKS_LE) { - printk(KERN_INFO "mtrr: Serverworks LE detected. Write-combining disabled.\n"); - pci_dev_put(dev); - return 0; + pci_read_config_byte(dev, PCI_CLASS_REVISION, &rev); + if (rev <= 5) { + printk(KERN_INFO "mtrr: Serverworks LE rev < 6 detected. Write-combining disabled.\n"); + pci_dev_put(dev); + return 0; + } } - /* Intel 450NX errata # 23. Non ascending cachline evictions to + /* Intel 450NX errata # 23. Non ascending cacheline evictions to write combining memory may resulting in data corruption */ if (dev->vendor == PCI_VENDOR_ID_INTEL && dev->device == PCI_DEVICE_ID_INTEL_82451NX) { -- cgit v0.10.2 From 5b7abc6fdcaf103f15e06c518ef0aec02a9c00e7 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Sun, 1 May 2005 08:58:49 -0700 Subject: [PATCH] CPUID bug and inconsistency fix The recent support for K8 multicore was misported from x86-64 to i386, due to an unnecessary inconsistency between the CPUID code. Sure, there is are no x86-64 VIA chips yet, but it should happen eventually. This patch fixes the i386 bug as well as makes x86-64 match i386 in the handing of the CPUID array. Signed-off-by: H. Peter Anvin Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c index b18c114..2959ce7 100644 --- a/arch/x86_64/kernel/setup.c +++ b/arch/x86_64/kernel/setup.c @@ -977,7 +977,7 @@ void __init identify_cpu(struct cpuinfo_x86 *c) if ((xlvl & 0xffff0000) == 0x80000000) { if (xlvl >= 0x80000001) { c->x86_capability[1] = cpuid_edx(0x80000001); - c->x86_capability[5] = cpuid_ecx(0x80000001); + c->x86_capability[6] = cpuid_ecx(0x80000001); } if (xlvl >= 0x80000004) get_model_name(c); /* Default name */ @@ -1100,11 +1100,17 @@ static int show_cpuinfo(struct seq_file *m, void *v) NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, + /* VIA/Cyrix/Centaur-defined */ + NULL, NULL, "rng", "rng_en", NULL, NULL, "ace", "ace_en", + NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, + /* AMD-defined (#2) */ "lahf_lm", "cmp_legacy", NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, - NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL + NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, }; static char *x86_power_flags[] = { "ts", /* temperature sensor */ diff --git a/include/asm-i386/cpufeature.h b/include/asm-i386/cpufeature.h index e147cab..ff1187e 100644 --- a/include/asm-i386/cpufeature.h +++ b/include/asm-i386/cpufeature.h @@ -87,8 +87,8 @@ #define X86_FEATURE_XCRYPT_EN (5*32+ 7) /* on-CPU crypto enabled */ /* More extended AMD flags: CPUID level 0x80000001, ecx, word 6 */ -#define X86_FEATURE_LAHF_LM (5*32+ 0) /* LAHF/SAHF in long mode */ -#define X86_FEATURE_CMP_LEGACY (5*32+ 1) /* If yes HyperThreading not valid */ +#define X86_FEATURE_LAHF_LM (6*32+ 0) /* LAHF/SAHF in long mode */ +#define X86_FEATURE_CMP_LEGACY (6*32+ 1) /* If yes HyperThreading not valid */ #define cpu_has(c, bit) test_bit(bit, (c)->x86_capability) #define boot_cpu_has(bit) test_bit(bit, boot_cpu_data.x86_capability) diff --git a/include/asm-x86_64/cpufeature.h b/include/asm-x86_64/cpufeature.h index e68ad97..aea308c 100644 --- a/include/asm-x86_64/cpufeature.h +++ b/include/asm-x86_64/cpufeature.h @@ -7,7 +7,7 @@ #ifndef __ASM_X8664_CPUFEATURE_H #define __ASM_X8664_CPUFEATURE_H -#define NCAPINTS 6 +#define NCAPINTS 7 /* N 32-bit words worth of info */ /* Intel-defined CPU features, CPUID level 0x00000001, word 0 */ #define X86_FEATURE_FPU (0*32+ 0) /* Onboard FPU */ @@ -74,9 +74,15 @@ #define X86_FEATURE_CX16 (4*32+13) /* CMPXCHG16B */ #define X86_FEATURE_XTPR (4*32+14) /* Send Task Priority Messages */ -/* More extended AMD flags: CPUID level 0x80000001, ecx, word 5 */ -#define X86_FEATURE_LAHF_LM (5*32+ 0) /* LAHF/SAHF in long mode */ -#define X86_FEATURE_CMP_LEGACY (5*32+ 1) /* If yes HyperThreading not valid */ +/* VIA/Cyrix/Centaur-defined CPU features, CPUID level 0xC0000001, word 5 */ +#define X86_FEATURE_XSTORE (5*32+ 2) /* on-CPU RNG present (xstore insn) */ +#define X86_FEATURE_XSTORE_EN (5*32+ 3) /* on-CPU RNG enabled */ +#define X86_FEATURE_XCRYPT (5*32+ 6) /* on-CPU crypto (xcrypt insn) */ +#define X86_FEATURE_XCRYPT_EN (5*32+ 7) /* on-CPU crypto enabled */ + +/* More extended AMD flags: CPUID level 0x80000001, ecx, word 6 */ +#define X86_FEATURE_LAHF_LM (6*32+ 0) /* LAHF/SAHF in long mode */ +#define X86_FEATURE_CMP_LEGACY (6*32+ 1) /* If yes HyperThreading not valid */ #define cpu_has(c, bit) test_bit(bit, (c)->x86_capability) #define boot_cpu_has(bit) test_bit(bit, boot_cpu_data.x86_capability) -- cgit v0.10.2 From 35492df5ae0f36f717448b2aea908d3a8891d1c4 Mon Sep 17 00:00:00 2001 From: john stultz Date: Sun, 1 May 2005 08:58:50 -0700 Subject: [PATCH] i386: fix hpet for systems that don't support legacy replacement Currently the i386 HPET code assumes the entire HPET implementation from the spec is present. This breaks on boxes that do not implement the optional legacy timer replacement functionality portion of the spec. This patch, which is very similar to my x86-64 patch for the same issue, fixes the problem allowing i386 systems that cannot use the HPET for the timer interrupt and RTC to still use the HPET as a time source. I've tested this patch on a system systems without HPET, with HPET but without legacy timer replacement, as well as HPET with legacy timer replacement. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/time.c b/arch/i386/kernel/time.c index 4d75b37..a0dcb7c 100644 --- a/arch/i386/kernel/time.c +++ b/arch/i386/kernel/time.c @@ -441,7 +441,7 @@ static void __init hpet_time_init(void) set_normalized_timespec(&wall_to_monotonic, -xtime.tv_sec, -xtime.tv_nsec); - if (hpet_enable() >= 0) { + if ((hpet_enable() >= 0) && hpet_use_timer) { printk("Using HPET for base-timer\n"); } diff --git a/arch/i386/kernel/time_hpet.c b/arch/i386/kernel/time_hpet.c index 244a31b..10a0cbb 100644 --- a/arch/i386/kernel/time_hpet.c +++ b/arch/i386/kernel/time_hpet.c @@ -26,6 +26,7 @@ static unsigned long hpet_period; /* fsecs / HPET clock */ unsigned long hpet_tick; /* hpet clks count per tick */ unsigned long hpet_address; /* hpet memory map physical address */ +int hpet_use_timer; static int use_hpet; /* can be used for runtime check of hpet */ static int boot_hpet_disable; /* boottime override for HPET timer */ @@ -73,27 +74,30 @@ static int hpet_timer_stop_set_go(unsigned long tick) hpet_writel(0, HPET_COUNTER); hpet_writel(0, HPET_COUNTER + 4); - /* - * Set up timer 0, as periodic with first interrupt to happen at - * hpet_tick, and period also hpet_tick. - */ - cfg = hpet_readl(HPET_T0_CFG); - cfg |= HPET_TN_ENABLE | HPET_TN_PERIODIC | - HPET_TN_SETVAL | HPET_TN_32BIT; - hpet_writel(cfg, HPET_T0_CFG); - - /* - * The first write after writing TN_SETVAL to the config register sets - * the counter value, the second write sets the threshold. - */ - hpet_writel(tick, HPET_T0_CMP); - hpet_writel(tick, HPET_T0_CMP); + if (hpet_use_timer) { + /* + * Set up timer 0, as periodic with first interrupt to happen at + * hpet_tick, and period also hpet_tick. + */ + cfg = hpet_readl(HPET_T0_CFG); + cfg |= HPET_TN_ENABLE | HPET_TN_PERIODIC | + HPET_TN_SETVAL | HPET_TN_32BIT; + hpet_writel(cfg, HPET_T0_CFG); + /* + * The first write after writing TN_SETVAL to the config register sets + * the counter value, the second write sets the threshold. + */ + hpet_writel(tick, HPET_T0_CMP); + hpet_writel(tick, HPET_T0_CMP); + } /* * Go! */ cfg = hpet_readl(HPET_CFG); - cfg |= HPET_CFG_ENABLE | HPET_CFG_LEGACY; + if (hpet_use_timer) + cfg |= HPET_CFG_LEGACY; + cfg |= HPET_CFG_ENABLE; hpet_writel(cfg, HPET_CFG); return 0; @@ -128,12 +132,11 @@ int __init hpet_enable(void) * However, we can do with one timer otherwise using the * the single HPET timer for system time. */ - if ( #ifdef CONFIG_HPET_EMULATE_RTC - !(id & HPET_ID_NUMBER) || -#endif - !(id & HPET_ID_LEGSUP)) + if (!(id & HPET_ID_NUMBER)) return -1; +#endif + hpet_period = hpet_readl(HPET_PERIOD); if ((hpet_period < HPET_MIN_PERIOD) || (hpet_period > HPET_MAX_PERIOD)) @@ -152,6 +155,8 @@ int __init hpet_enable(void) if (hpet_tick_rem > (hpet_period >> 1)) hpet_tick++; /* rounding the result */ + hpet_use_timer = id & HPET_ID_LEGSUP; + if (hpet_timer_stop_set_go(hpet_tick)) return -1; @@ -202,7 +207,8 @@ int __init hpet_enable(void) #endif #ifdef CONFIG_X86_LOCAL_APIC - wait_timer_tick = wait_hpet_tick; + if (hpet_use_timer) + wait_timer_tick = wait_hpet_tick; #endif return 0; } diff --git a/arch/i386/kernel/timers/timer_hpet.c b/arch/i386/kernel/timers/timer_hpet.c index 713134e..f778f47 100644 --- a/arch/i386/kernel/timers/timer_hpet.c +++ b/arch/i386/kernel/timers/timer_hpet.c @@ -79,7 +79,7 @@ static unsigned long get_offset_hpet(void) eax = hpet_readl(HPET_COUNTER); eax -= hpet_last; /* hpet delta */ - + eax = min(hpet_tick, eax); /* * Time offset = (hpet delta) * ( usecs per HPET clock ) * = (hpet delta) * ( usecs per tick / HPET clocks per tick) @@ -105,9 +105,12 @@ static void mark_offset_hpet(void) last_offset = ((unsigned long long)last_tsc_high<<32)|last_tsc_low; rdtsc(last_tsc_low, last_tsc_high); - offset = hpet_readl(HPET_T0_CMP) - hpet_tick; - if (unlikely(((offset - hpet_last) > hpet_tick) && (hpet_last != 0))) { - int lost_ticks = (offset - hpet_last) / hpet_tick; + if (hpet_use_timer) + offset = hpet_readl(HPET_T0_CMP) - hpet_tick; + else + offset = hpet_readl(HPET_COUNTER); + if (unlikely(((offset - hpet_last) >= (2*hpet_tick)) && (hpet_last != 0))) { + int lost_ticks = ((offset - hpet_last) / hpet_tick) - 1; jiffies_64 += lost_ticks; } hpet_last = offset; diff --git a/arch/i386/kernel/timers/timer_tsc.c b/arch/i386/kernel/timers/timer_tsc.c index a685994..7926d96 100644 --- a/arch/i386/kernel/timers/timer_tsc.c +++ b/arch/i386/kernel/timers/timer_tsc.c @@ -477,7 +477,7 @@ static int __init init_tsc(char* override) if (cpu_has_tsc) { unsigned long tsc_quotient; #ifdef CONFIG_HPET_TIMER - if (is_hpet_enabled()){ + if (is_hpet_enabled() && hpet_use_timer) { unsigned long result, remain; printk("Using TSC for gettimeofday\n"); tsc_quotient = calibrate_tsc_hpet(NULL); diff --git a/include/asm-i386/hpet.h b/include/asm-i386/hpet.h index 6e20b07..16ef9f9 100644 --- a/include/asm-i386/hpet.h +++ b/include/asm-i386/hpet.h @@ -92,6 +92,7 @@ extern unsigned long hpet_tick; /* hpet clks count per tick */ extern unsigned long hpet_address; /* hpet memory map physical address */ +extern int hpet_use_timer; extern int hpet_rtc_timer_init(void); extern int hpet_enable(void); -- cgit v0.10.2 From 4d24a439a6b2280357d62fb30a73350cf253bdb7 Mon Sep 17 00:00:00 2001 From: Jason Gaston Date: Sun, 1 May 2005 08:58:50 -0700 Subject: [PATCH] irq and pci_ids for Intel ICH7DH & ICH7-M DH MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch adds the Intel ICH7DH and ICH7-M DH DID's to the irq.c and pci_ids.h files. Signed-off-by:  Jason Gaston Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/pci/irq.c b/arch/i386/pci/irq.c index e07589d..d6598da 100644 --- a/arch/i386/pci/irq.c +++ b/arch/i386/pci/irq.c @@ -495,6 +495,8 @@ static __init int intel_router_probe(struct irq_router *r, struct pci_dev *route case PCI_DEVICE_ID_INTEL_ICH6_1: case PCI_DEVICE_ID_INTEL_ICH7_0: case PCI_DEVICE_ID_INTEL_ICH7_1: + case PCI_DEVICE_ID_INTEL_ICH7_30: + case PCI_DEVICE_ID_INTEL_ICH7_31: case PCI_DEVICE_ID_INTEL_ESB2_0: r->name = "PIIX/ICH"; r->get = pirq_piix_get; diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 6a18974..5d5820a 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2415,6 +2415,8 @@ #define PCI_DEVICE_ID_INTEL_ICH7_1 0x27b9 #define PCI_DEVICE_ID_INTEL_ICH7_2 0x27c0 #define PCI_DEVICE_ID_INTEL_ICH7_3 0x27c1 +#define PCI_DEVICE_ID_INTEL_ICH7_30 0x27b0 +#define PCI_DEVICE_ID_INTEL_ICH7_31 0x27bd #define PCI_DEVICE_ID_INTEL_ICH7_5 0x27c4 #define PCI_DEVICE_ID_INTEL_ICH7_6 0x27c5 #define PCI_DEVICE_ID_INTEL_ICH7_7 0x27c8 -- cgit v0.10.2 From 2f1b381825411fd3d5c8323dd1be77438e19e192 Mon Sep 17 00:00:00 2001 From: Jason Gaston Date: Sun, 1 May 2005 08:58:50 -0700 Subject: [PATCH] hda_intel: Intel ESB2 support MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This adds the Intel ESB2 HD Audio DID to the hda_intel.c audio driver. Signed-off-by:  Jason Gaston Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index d89647a..959953c 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -64,7 +64,8 @@ MODULE_PARM_DESC(model, "Use the given board model."); MODULE_LICENSE("GPL"); MODULE_SUPPORTED_DEVICE("{{Intel, ICH6}," "{Intel, ICH6M}," - "{Intel, ICH7}}"); + "{Intel, ICH7}," + "{Intel, ESB2}}"); MODULE_DESCRIPTION("Intel HDA driver"); #define SFX "hda-intel: " @@ -1422,6 +1423,7 @@ static void __devexit azx_remove(struct pci_dev *pci) static struct pci_device_id azx_ids[] = { { 0x8086, 0x2668, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* ICH6 */ { 0x8086, 0x27d8, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* ICH7 */ + { 0x8086, 0x269a, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* ESB2 */ { 0, } }; MODULE_DEVICE_TABLE(pci, azx_ids); -- cgit v0.10.2 From 3c3b73b6f50a47601b1897deb9c6b4b5ecadf591 Mon Sep 17 00:00:00 2001 From: Zwane Mwaikambo Date: Sun, 1 May 2005 08:58:51 -0700 Subject: [PATCH] cpuid x87 bit on AMD falsely marked as PNI http://bugme.osdl.org/show_bug.cgi?id=4426 vendor_id : AuthenticAMD cpu family : 6 model : 10 model name : AMD Athlon(tm) XP stepping : 0 cpu MHz : 2204.807 cpuid level : 1 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 mmx fxsr sse pni syscall mmxext 3dnowext 3dnow bogomips : 4358.14 We're marking bit 0 of extended function 0x80000001 cpuid as PNI support on AMD processors, when it actually denotes x87 FPU present. Patch for i386 and x86_64 below. Signed-off-by: Zwane Mwaikambo Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/cpu/proc.c b/arch/i386/kernel/cpu/proc.c index 4f28eba..7323c19 100644 --- a/arch/i386/kernel/cpu/proc.c +++ b/arch/i386/kernel/cpu/proc.c @@ -25,7 +25,7 @@ static int show_cpuinfo(struct seq_file *m, void *v) "fxsr", "sse", "sse2", "ss", "ht", "tm", "ia64", "pbe", /* AMD-defined */ - "pni", NULL, NULL, NULL, NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, "syscall", NULL, NULL, NULL, NULL, NULL, NULL, NULL, "mp", "nx", NULL, "mmxext", NULL, NULL, "fxsr_opt", NULL, NULL, NULL, "lm", "3dnowext", "3dnow", diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c index 2959ce7..2129cf9 100644 --- a/arch/x86_64/kernel/setup.c +++ b/arch/x86_64/kernel/setup.c @@ -1076,7 +1076,7 @@ static int show_cpuinfo(struct seq_file *m, void *v) "fxsr", "sse", "sse2", "ss", "ht", "tm", "ia64", NULL, /* AMD-defined */ - "pni", NULL, NULL, NULL, NULL, NULL, NULL, NULL, + NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, "syscall", NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, "nx", NULL, "mmxext", NULL, NULL, "fxsr_opt", NULL, NULL, NULL, "lm", "3dnowext", "3dnow", -- cgit v0.10.2 From be9e68703c53fece406bc2e40976a5cfe2a8e77f Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sun, 1 May 2005 08:58:51 -0700 Subject: [PATCH] x86_64: interrupt handling fix - Initialize workmask correctly on interrupt signal handling - Readd missing cli's in the interrupt return path. Signed-off-by: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S index 3233a15..1086b5f 100644 --- a/arch/x86_64/kernel/entry.S +++ b/arch/x86_64/kernel/entry.S @@ -296,6 +296,7 @@ int_very_careful: call syscall_trace_leave popq %rdi andl $~(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP),%edi + cli jmp int_restore_rest int_signal: @@ -307,6 +308,7 @@ int_signal: 1: movl $_TIF_NEED_RESCHED,%edi int_restore_rest: RESTORE_REST + cli jmp int_with_check CFI_ENDPROC @@ -490,7 +492,8 @@ retint_signal: call do_notify_resume RESTORE_REST cli - GET_THREAD_INFO(%rcx) + GET_THREAD_INFO(%rcx) + movl $_TIF_WORK_MASK,%edi jmp retint_check #ifdef CONFIG_PREEMPT -- cgit v0.10.2 From f9ba70535dc12d9eb57d466a2ecd749e16eca866 Mon Sep 17 00:00:00 2001 From: Venkatesh Pallipadi Date: Sun, 1 May 2005 08:58:51 -0700 Subject: [PATCH] Increase number of e820 entries hard limit from 32 to 128 The specifications that talk about E820 map doesn't have an upper limit on the number of e820 entries. But, today's kernel has a hard limit of 32. With increase in memory size, we are seeing the number of E820 entries reaching close to 32. Patch below bumps the number upto 128. The patch changes the location of EDDBUF in zero-page (as it comes after E820). As, EDDBUF is not used by boot loaders, this patch should not have any effect on bootloader-setup code interface. Patch covers both i386 and x86-64. Tested on: * grub booting bzImage * lilo booting bzImage with EDID info enabled * pxeboot of bzImage Side-effect: bss increases by ~ 2K and init.data increases by ~7.5K on all systems, due to increase in size of static arrays. Signed-off-by: Venkatesh Pallipadi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/i386/zero-page.txt b/Documentation/i386/zero-page.txt index 67c053a..df28c74 100644 --- a/Documentation/i386/zero-page.txt +++ b/Documentation/i386/zero-page.txt @@ -79,6 +79,6 @@ Offset Type Description 0x22c unsigned long ramdisk_max 0x230 16 bytes trampoline 0x290 - 0x2cf EDD_MBR_SIG_BUFFER (edd.S) -0x2d0 - 0x600 E820MAP -0x600 - 0x7ff EDDBUF (edd.S) for disk signature read sector -0x600 - 0x7eb EDDBUF (edd.S) for edd data +0x2d0 - 0xd00 E820MAP +0xd00 - 0xeff EDDBUF (edd.S) for disk signature read sector +0xd00 - 0xeeb EDDBUF (edd.S) for edd data diff --git a/arch/i386/boot/setup.S b/arch/i386/boot/setup.S index a934ab3..caa1fde 100644 --- a/arch/i386/boot/setup.S +++ b/arch/i386/boot/setup.S @@ -164,7 +164,7 @@ ramdisk_max: .long (-__PAGE_OFFSET-(512 << 20)-1) & 0x7fffffff trampoline: call start_of_setup .align 16 # The offset at this point is 0x240 - .space (0x7ff-0x240+1) # E820 & EDD space (ending at 0x7ff) + .space (0xeff-0x240+1) # E820 & EDD space (ending at 0xeff) # End of setup header ##################################################### start_of_setup: @@ -333,9 +333,9 @@ jmpe820: # sizeof(e820rec). # good820: - movb (E820NR), %al # up to 32 entries + movb (E820NR), %al # up to 128 entries cmpb $E820MAX, %al - jnl bail820 + jae bail820 incb (E820NR) movw %di, %ax diff --git a/arch/x86_64/boot/setup.S b/arch/x86_64/boot/setup.S index 3e838be..75d4d2a 100644 --- a/arch/x86_64/boot/setup.S +++ b/arch/x86_64/boot/setup.S @@ -160,7 +160,7 @@ ramdisk_max: .long 0xffffffff trampoline: call start_of_setup .align 16 # The offset at this point is 0x240 - .space (0x7ff-0x240+1) # E820 & EDD space (ending at 0x7ff) + .space (0xeff-0x240+1) # E820 & EDD space (ending at 0xeff) # End of setup header ##################################################### start_of_setup: @@ -412,9 +412,9 @@ jmpe820: # sizeof(e820rec). # good820: - movb (E820NR), %al # up to 32 entries + movb (E820NR), %al # up to 128 entries cmpb $E820MAX, %al - jnl bail820 + jae bail820 incb (E820NR) movw %di, %ax diff --git a/arch/x86_64/kernel/head64.c b/arch/x86_64/kernel/head64.c index 6cad46c..77df2cc 100644 --- a/arch/x86_64/kernel/head64.c +++ b/arch/x86_64/kernel/head64.c @@ -29,8 +29,6 @@ static void __init clear_bss(void) (unsigned long) __bss_end - (unsigned long) __bss_start); } -extern char x86_boot_params[2048]; - #define NEW_CL_POINTER 0x228 /* Relative to real mode data */ #define OLD_CL_MAGIC_ADDR 0x90020 #define OLD_CL_MAGIC 0xA33F @@ -44,7 +42,7 @@ static void __init copy_bootdata(char *real_mode_data) int new_data; char * command_line; - memcpy(x86_boot_params, real_mode_data, 2048); + memcpy(x86_boot_params, real_mode_data, BOOT_PARAM_SIZE); new_data = *(int *) (x86_boot_params + NEW_CL_POINTER); if (!new_data) { if (OLD_CL_MAGIC != * (u16 *) OLD_CL_MAGIC_ADDR) { diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c index b5305b0..678b7ac 100644 --- a/arch/x86_64/kernel/setup64.c +++ b/arch/x86_64/kernel/setup64.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include @@ -26,7 +27,7 @@ #include #include -char x86_boot_params[2048] __initdata = {0,}; +char x86_boot_params[BOOT_PARAM_SIZE] __initdata = {0,}; cpumask_t cpu_initialized __initdata = CPU_MASK_NONE; diff --git a/include/asm-i386/e820.h b/include/asm-i386/e820.h index 5c285ae..edf65be 100644 --- a/include/asm-i386/e820.h +++ b/include/asm-i386/e820.h @@ -13,7 +13,7 @@ #define __E820_HEADER #define E820MAP 0x2d0 /* our map */ -#define E820MAX 32 /* number of entries in E820MAP */ +#define E820MAX 128 /* number of entries in E820MAP */ #define E820NR 0x1e8 /* # entries in E820MAP */ #define E820_RAM 1 diff --git a/include/asm-i386/setup.h b/include/asm-i386/setup.h index 8814b54..7a32184 100644 --- a/include/asm-i386/setup.h +++ b/include/asm-i386/setup.h @@ -16,7 +16,7 @@ #define MAXMEM_PFN PFN_DOWN(MAXMEM) #define MAX_NONPAE_PFN (1 << 20) -#define PARAM_SIZE 2048 +#define PARAM_SIZE 4096 #define COMMAND_LINE_SIZE 256 #define OLD_CL_MAGIC_ADDR 0x90020 diff --git a/include/asm-x86_64/bootsetup.h b/include/asm-x86_64/bootsetup.h index b570a48..b829f7b 100644 --- a/include/asm-x86_64/bootsetup.h +++ b/include/asm-x86_64/bootsetup.h @@ -2,7 +2,8 @@ #ifndef _X86_64_BOOTSETUP_H #define _X86_64_BOOTSETUP_H 1 -extern char x86_boot_params[2048]; +#define BOOT_PARAM_SIZE 4096 +extern char x86_boot_params[BOOT_PARAM_SIZE]; /* * This is set up by the setup-routine at boot-time diff --git a/include/asm-x86_64/e820.h b/include/asm-x86_64/e820.h index 08f83a4..8e94edf 100644 --- a/include/asm-x86_64/e820.h +++ b/include/asm-x86_64/e820.h @@ -14,7 +14,7 @@ #include #define E820MAP 0x2d0 /* our map */ -#define E820MAX 32 /* number of entries in E820MAP */ +#define E820MAX 128 /* number of entries in E820MAP */ #define E820NR 0x1e8 /* # entries in E820MAP */ #define E820_RAM 1 diff --git a/include/linux/edd.h b/include/linux/edd.h index c6e6747..162512b 100644 --- a/include/linux/edd.h +++ b/include/linux/edd.h @@ -32,7 +32,7 @@ #define EDDNR 0x1e9 /* addr of number of edd_info structs at EDDBUF in boot_params - treat this as 1 byte */ -#define EDDBUF 0x600 /* addr of edd_info structs in boot_params */ +#define EDDBUF 0xd00 /* addr of edd_info structs in boot_params */ #define EDDMAXNR 6 /* number of edd_info structs starting at EDDBUF */ #define EDDEXTSIZE 8 /* change these if you muck with the structures */ #define EDDPARMSIZE 74 -- cgit v0.10.2 From ad6714230f2269d5d7db2cd1900fe7bfc7aa76dc Mon Sep 17 00:00:00 2001 From: Pavel Pisa Date: Sun, 1 May 2005 08:58:52 -0700 Subject: [PATCH] Linux 2.6.x VM86 interrupt emulation fixes Patch solves VM86 interrupt emulation deadlock on SMP systems. The VM86 interrupt emulation has been heavily tested and works well on UP systems after last update, but it seems to deadlock when we have used it on SMP/HT boxes now. It seems, that disable_irq() cannot be called from interrupts, because it waits until disabled interrupt handler finishes (/kernel/irq/manage.c:synchronize_irq():while(IRQ_INPROGRESS);). This blocks one CPU after another. Solved by use disable_irq_nosync. There is the second problem. If IRQ source is fast, it is possible, that interrupt is sometimes processed and re-enabled by the second CPU, before it is disabled by the first one, but negative IRQ disable depths are not allowed. The spinlocking and disabling IRQs over call to disable_irq_nosync/enable_irq is the only solution found reliable till now. Signed-off-by: Michal Sojka Signed-off-by: Pavel Pisa Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/vm86.c b/arch/i386/kernel/vm86.c index d16cd37..d3b4c54 100644 --- a/arch/i386/kernel/vm86.c +++ b/arch/i386/kernel/vm86.c @@ -717,12 +717,12 @@ static irqreturn_t irq_handler(int intno, void *dev_id, struct pt_regs * regs) irqbits |= irq_bit; if (vm86_irqs[intno].sig) send_sig(vm86_irqs[intno].sig, vm86_irqs[intno].tsk, 1); - spin_unlock_irqrestore(&irqbits_lock, flags); /* * IRQ will be re-enabled when user asks for the irq (whether * polling or as a result of the signal) */ - disable_irq(intno); + disable_irq_nosync(intno); + spin_unlock_irqrestore(&irqbits_lock, flags); return IRQ_HANDLED; out: @@ -754,17 +754,20 @@ static inline int get_and_reset_irq(int irqnumber) { int bit; unsigned long flags; + int ret = 0; if (invalid_vm86_irq(irqnumber)) return 0; if (vm86_irqs[irqnumber].tsk != current) return 0; spin_lock_irqsave(&irqbits_lock, flags); bit = irqbits & (1 << irqnumber); irqbits &= ~bit; + if (bit) { + enable_irq(irqnumber); + ret = 1; + } + spin_unlock_irqrestore(&irqbits_lock, flags); - if (!bit) - return 0; - enable_irq(irqnumber); - return 1; + return ret; } -- cgit v0.10.2 From 8059b2a292d59b6d94f9e3af93bf711f161c9476 Mon Sep 17 00:00:00 2001 From: Venkatesh Pallipadi Date: Sun, 1 May 2005 08:58:52 -0700 Subject: [PATCH] x86-64: Handle empty E820 regions correctly Brings sanitize_e820_map() in x86-64 in sync with that of i386. x86_64 version was missing the changes from this patch. http://linux.bkbits.net:8080/linux-2.6/cset@3e5e4083Y3HevldZl5KCy94V4DcZww?nav=index.html|src/|src/arch|src/arch/i386|src/arch/i386/kernel|related/arch/i386/kernel/setup.c Signed-off-by: Venkatesh Pallipadi Acked-by: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c index 56516ac..7c154df 100644 --- a/arch/x86_64/kernel/e820.c +++ b/arch/x86_64/kernel/e820.c @@ -2,6 +2,12 @@ * Handle the memory map. * The functions here do the job until bootmem takes over. * $Id: e820.c,v 1.4 2002/09/19 19:25:32 ak Exp $ + * + * Getting sanitize_e820_map() in sync with i386 version by applying change: + * - Provisions for empty E820 memory regions (reported by certain BIOSes). + * Alex Achenbach , December 2002. + * Venkatesh Pallipadi + * */ #include #include @@ -277,7 +283,7 @@ static int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map) int chgidx, still_changing; int overlap_entries; int new_bios_entry; - int old_nr, new_nr; + int old_nr, new_nr, chg_nr; int i; /* @@ -331,20 +337,24 @@ static int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map) for (i=0; i < 2*old_nr; i++) change_point[i] = &change_point_list[i]; - /* record all known change-points (starting and ending addresses) */ + /* record all known change-points (starting and ending addresses), + omitting those that are for empty memory regions */ chgidx = 0; for (i=0; i < old_nr; i++) { - change_point[chgidx]->addr = biosmap[i].addr; - change_point[chgidx++]->pbios = &biosmap[i]; - change_point[chgidx]->addr = biosmap[i].addr + biosmap[i].size; - change_point[chgidx++]->pbios = &biosmap[i]; + if (biosmap[i].size != 0) { + change_point[chgidx]->addr = biosmap[i].addr; + change_point[chgidx++]->pbios = &biosmap[i]; + change_point[chgidx]->addr = biosmap[i].addr + biosmap[i].size; + change_point[chgidx++]->pbios = &biosmap[i]; + } } + chg_nr = chgidx; /* sort change-point list by memory addresses (low -> high) */ still_changing = 1; while (still_changing) { still_changing = 0; - for (i=1; i < 2*old_nr; i++) { + for (i=1; i < chg_nr; i++) { /* if > , swap */ /* or, if current= & last=, swap */ if ((change_point[i]->addr < change_point[i-1]->addr) || @@ -367,7 +377,7 @@ static int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map) last_type = 0; /* start with undefined memory type */ last_addr = 0; /* start with 0 as last starting address */ /* loop through change-points, determining affect on the new bios map */ - for (chgidx=0; chgidx < 2*old_nr; chgidx++) + for (chgidx=0; chgidx < chg_nr; chgidx++) { /* keep track of all overlapping bios entries */ if (change_point[chgidx]->addr == change_point[chgidx]->pbios->addr) -- cgit v0.10.2 From 429e9c5eeb66fc272c15568849ad0b23689a4a06 Mon Sep 17 00:00:00 2001 From: Alexander Nyberg Date: Sun, 1 May 2005 08:58:52 -0700 Subject: [PATCH] x86_64: saved_command_line overflow fix This strcpy can run off the end of saved_command_line, and we don't need it any more anyway. Signed-off-by: Alexander Nyberg Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/x86_64/kernel/head64.c b/arch/x86_64/kernel/head64.c index 77df2cc..0f8c78d 100644 --- a/arch/x86_64/kernel/head64.c +++ b/arch/x86_64/kernel/head64.c @@ -91,9 +91,6 @@ void __init x86_64_start_kernel(char * real_mode_data) #ifdef CONFIG_SMP cpu_set(0, cpu_online_map); #endif - /* default console: */ - if (!strstr(saved_command_line, "console=")) - strcat(saved_command_line, " console=tty0"); s = strstr(saved_command_line, "earlyprintk="); if (s != NULL) setup_early_printk(s); -- cgit v0.10.2 From 31df7b7fc6182b6c4e5c3b77084d0c026b3ac613 Mon Sep 17 00:00:00 2001 From: Pavel Machek Date: Sun, 1 May 2005 08:58:52 -0700 Subject: [PATCH] hp100: fix card names Those cards really need A in their names. Otherwise it is pretty hard to find anything about them on the net. Signed-off-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/net/hp100.c b/drivers/net/hp100.c index acb1701..b3a898c 100644 --- a/drivers/net/hp100.c +++ b/drivers/net/hp100.c @@ -13,8 +13,8 @@ ** This driver has only been tested with ** -- HP J2585B 10/100 Mbit/s PCI Busmaster ** -- HP J2585A 10/100 Mbit/s PCI -** -- HP J2970 10 Mbit/s PCI Combo 10base-T/BNC -** -- HP J2973 10 Mbit/s PCI 10base-T +** -- HP J2970A 10 Mbit/s PCI Combo 10base-T/BNC +** -- HP J2973A 10 Mbit/s PCI 10base-T ** -- HP J2573 10/100 ISA ** -- Compex ReadyLink ENET100-VG4 10/100 Mbit/s PCI / EISA ** -- Compex FreedomLine 100/VG 10/100 Mbit/s ISA / EISA / PCI -- cgit v0.10.2 From 92eac95287d75f220a8bbef6646f51a6497c4b4c Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Sun, 1 May 2005 08:58:53 -0700 Subject: [PATCH] uml: fix oops related to exception table Paolo 'Blaisorblade' Giarrusso Prevent the kernel from oopsing during the extable sorting, as it can do now, because the extable is in the readonly section of the binary. Jeff says: The exception table turned RO in 2.6.11-rc3-mm1 for some reason. Moving it causes it to land in the writable data section of the binary. Paolo says: This patch fixes a oops on startup, which can be easily triggered by compiling with CONFIG_MODE_TT disabled, and STATIC_LINK either disabled or enabled. The resulting kernel will always Oops on startup, after printing this simple output: I've verified, by binary search on the BitKeeper repository (synced up as of 2.6.12-rc2), starting from the range 2.6.11-2.6.12-rc1, that this bug shows up on BitKeeper revisions in the range [@1.1994.11.168,+inf), i.e. starting from this: [PATCH] lib/sort: Replace insertion sort in exception tables Since UML does not use the exception table, it's likely that insertion sort didn't happen to write anything on the table. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-um/common.lds.S b/include/asm-um/common.lds.S index a3d6aab..1010153 100644 --- a/include/asm-um/common.lds.S +++ b/include/asm-um/common.lds.S @@ -8,11 +8,6 @@ _sdata = .; PROVIDE (sdata = .); - . = ALIGN(16); /* Exception table */ - __start___ex_table = .; - __ex_table : { *(__ex_table) } - __stop___ex_table = .; - RODATA .unprotected : { *(.unprotected) } @@ -20,6 +15,10 @@ PROVIDE (_unprotected_end = .); . = ALIGN(4096); + __start___ex_table = .; + __ex_table : { *(__ex_table) } + __stop___ex_table = .; + __uml_setup_start = .; .uml.setup.init : { *(.uml.setup.init) } __uml_setup_end = .; -- cgit v0.10.2 From 8e7a4f69e4ed328aa09ce48601073052557fa85d Mon Sep 17 00:00:00 2001 From: "blaisorblade@yahoo.it" Date: Sun, 1 May 2005 08:58:53 -0700 Subject: [PATCH] uml: add nfsd syscall when nfsd is modular This trick is useless, because sys_ni.c will handle this problem by itself, like it does even on UML for other syscalls. Also, it does not provide the NFSD syscall when NFSD is compiled as a module, which is a big problem. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/kernel/sys_call_table.c b/arch/um/kernel/sys_call_table.c index 7fc06c8..b671a31 100644 --- a/arch/um/kernel/sys_call_table.c +++ b/arch/um/kernel/sys_call_table.c @@ -14,12 +14,6 @@ #include "sysdep/syscalls.h" #include "kern_util.h" -#ifdef CONFIG_NFSD -#define NFSSERVCTL sys_nfsservctl -#else -#define NFSSERVCTL sys_ni_syscall -#endif - #define LAST_GENERIC_SYSCALL __NR_keyctl #if LAST_GENERIC_SYSCALL > LAST_ARCH_SYSCALL @@ -189,7 +183,7 @@ syscall_handler_t *sys_call_table[] = { [ __NR_getresuid ] = (syscall_handler_t *) sys_getresuid16, [ __NR_query_module ] = (syscall_handler_t *) sys_ni_syscall, [ __NR_poll ] = (syscall_handler_t *) sys_poll, - [ __NR_nfsservctl ] = (syscall_handler_t *) NFSSERVCTL, + [ __NR_nfsservctl ] = (syscall_handler_t *) sys_nfsservctl, [ __NR_setresgid ] = (syscall_handler_t *) sys_setresgid16, [ __NR_getresgid ] = (syscall_handler_t *) sys_getresgid16, [ __NR_prctl ] = (syscall_handler_t *) sys_prctl, -- cgit v0.10.2 From 6aec3072833283a5e2ecf07318bd383a67321657 Mon Sep 17 00:00:00 2001 From: Andree Leidenfrost Date: Sun, 1 May 2005 08:58:53 -0700 Subject: [PATCH] uml: fix handling of no fpx_regs Paolo 'Blaisorblade' Giarrusso Fix the error path, which is triggered when the processor misses the fpx regs (i.e. the "fxsr" cpuinfo feature). For instance by VIA C3 Samuel2. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/os-Linux/sys-i386/registers.c b/arch/um/os-Linux/sys-i386/registers.c index 148645b..9a0ad09 100644 --- a/arch/um/os-Linux/sys-i386/registers.c +++ b/arch/um/os-Linux/sys-i386/registers.c @@ -105,14 +105,15 @@ void init_registers(int pid) panic("check_ptrace : PTRACE_GETREGS failed, errno = %d", err); + errno = 0; err = ptrace(PTRACE_GETFPXREGS, pid, 0, exec_fpx_regs); if(!err) return; + if(errno != EIO) + panic("check_ptrace : PTRACE_GETFPXREGS failed, errno = %d", + errno); have_fpx_regs = 0; - if(err != EIO) - panic("check_ptrace : PTRACE_GETFPXREGS failed, errno = %d", - err); err = ptrace(PTRACE_GETFPREGS, pid, 0, exec_fp_regs); if(err) -- cgit v0.10.2 From b05d85a87d9711f5f5f2eb05c79038d5d5ff1f44 Mon Sep 17 00:00:00 2001 From: Rob Landley Date: Sun, 1 May 2005 08:58:54 -0700 Subject: [PATCH] uml: workaround old problematic sed behaviour Old versions of sed from 1998 (predating the first release of gcc 2.95, but still in use by debian stable) don't understand the single-line version of the sed append command. Since newer versions of sed still understand the... ahem, "vintage" form of the command, change our code to use that. Signed-off-by: Rob Landley Acked-by: Ian McDonald Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/kernel/Makefile b/arch/um/kernel/Makefile index dc796c1..280d300 100644 --- a/arch/um/kernel/Makefile +++ b/arch/um/kernel/Makefile @@ -53,6 +53,7 @@ quiet_cmd_quote2 = QUOTE $@ cmd_quote2 = sed -e '/CONFIG/{' \ -e 's/"CONFIG"\;/""/' \ -e 'r $(obj)/config.tmp' \ - -e 'a""\;' \ + -e 'a \' \ + -e '""\;' \ -e '}' \ $< > $@ -- cgit v0.10.2 From c45166be3cc666ce88fe623ad79276c943e74eff Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Sun, 1 May 2005 08:58:54 -0700 Subject: [PATCH] uml: support AES i586 crypto driver We want to make possible, for the user, to enable the i586 AES implementation. This requires a restructure. - Add a CONFIG_UML_X86 to notify that we are building a UML for i386. - Rename CONFIG_64_BIT to CONFIG_64BIT as is used for all other archs - Tell crypto/Kconfig that UML_X86 is as good as X86 - Tell it that it must exclude not X86_64 but 64BIT, which will give the same results. - Tell kbuild to descend down into arch/i386/crypto/ to build what's needed. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/Kconfig b/arch/um/Kconfig index 9a23df1..c529218 100644 --- a/arch/um/Kconfig +++ b/arch/um/Kconfig @@ -244,6 +244,7 @@ config KERNEL_HALF_GIGS config HIGHMEM bool "Highmem support" + depends on !64BIT config KERNEL_STACK_ORDER int "Kernel stack size order" diff --git a/arch/um/Kconfig_i386 b/arch/um/Kconfig_i386 index 203c242..e41f374 100644 --- a/arch/um/Kconfig_i386 +++ b/arch/um/Kconfig_i386 @@ -1,4 +1,8 @@ -config 64_BIT +config UML_X86 + bool + default y + +config 64BIT bool default n diff --git a/arch/um/Kconfig_x86_64 b/arch/um/Kconfig_x86_64 index 768dc66..fd8d7e8 100644 --- a/arch/um/Kconfig_x86_64 +++ b/arch/um/Kconfig_x86_64 @@ -1,4 +1,8 @@ -config 64_BIT +config UML_X86 + bool + default y + +config 64BIT bool default y diff --git a/arch/um/Makefile-i386 b/arch/um/Makefile-i386 index 97b223b..f9e3c0f 100644 --- a/arch/um/Makefile-i386 +++ b/arch/um/Makefile-i386 @@ -1,4 +1,4 @@ -SUBARCH_CORE := arch/um/sys-i386/ +SUBARCH_CORE := arch/um/sys-i386/ arch/i386/crypto/ TOP_ADDR := $(CONFIG_TOP_ADDR) diff --git a/arch/um/defconfig b/arch/um/defconfig index fc3075c..4067c3a 100644 --- a/arch/um/defconfig +++ b/arch/um/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.12-rc1-bk1 -# Sun Mar 20 16:53:00 2005 +# Linux kernel version: 2.6.12-rc3-skas3-v9-pre2 +# Sun Apr 24 19:46:10 2005 # CONFIG_GENERIC_HARDIRQS=y CONFIG_UML=y @@ -15,7 +15,8 @@ CONFIG_GENERIC_CALIBRATE_DELAY=y # CONFIG_MODE_TT=y CONFIG_MODE_SKAS=y -# CONFIG_64_BIT is not set +CONFIG_UML_X86=y +# CONFIG_64BIT is not set CONFIG_TOP_ADDR=0xc0000000 # CONFIG_3_LEVEL_PGTABLES is not set CONFIG_ARCH_HAS_SC_SIGNALS=y @@ -41,6 +42,7 @@ CONFIG_UML_REAL_TIME_CLOCK=y CONFIG_EXPERIMENTAL=y CONFIG_CLEAN_COMPILE=y CONFIG_BROKEN_ON_SMP=y +CONFIG_INIT_ENV_ARG_LIMIT=32 # # General setup @@ -158,7 +160,6 @@ CONFIG_UML_NET_SLIRP=y # CONFIG_PACKET=y CONFIG_PACKET_MMAP=y -# CONFIG_NETLINK_DEV is not set CONFIG_UNIX=y # CONFIG_NET_KEY is not set CONFIG_INET=y @@ -412,6 +413,5 @@ CONFIG_DEBUG_INFO=y # CONFIG_DEBUG_FS is not set CONFIG_FRAME_POINTER=y CONFIG_PT_PROXY=y -# CONFIG_GPROF is not set # CONFIG_GCOV is not set # CONFIG_SYSCALL_DEBUG is not set diff --git a/crypto/Kconfig b/crypto/Kconfig index 536754f..90d6089 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -146,7 +146,7 @@ config CRYPTO_SERPENT config CRYPTO_AES tristate "AES cipher algorithms" - depends on CRYPTO && !(X86 && !X86_64) + depends on CRYPTO && !((X86 || UML_X86) && !64BIT) help AES cipher algorithms (FIPS-197). AES uses the Rijndael algorithm. @@ -166,7 +166,7 @@ config CRYPTO_AES config CRYPTO_AES_586 tristate "AES cipher algorithms (i586)" - depends on CRYPTO && (X86 && !X86_64) + depends on CRYPTO && ((X86 || UML_X86) && !64BIT) help AES cipher algorithms (FIPS-197). AES uses the Rijndael algorithm. diff --git a/include/asm-um/elf.h b/include/asm-um/elf.h index b3a7258..7908f8f 100644 --- a/include/asm-um/elf.h +++ b/include/asm-um/elf.h @@ -13,7 +13,7 @@ extern long elf_aux_hwcap; #define elf_check_arch(x) (1) -#ifdef CONFIG_64_BIT +#ifdef CONFIG_64BIT #define ELF_CLASS ELFCLASS64 #else #define ELF_CLASS ELFCLASS32 diff --git a/include/asm-um/page.h b/include/asm-um/page.h index 3620a08..102eb3d 100644 --- a/include/asm-um/page.h +++ b/include/asm-um/page.h @@ -27,7 +27,7 @@ struct page; #define clear_user_page(page, vaddr, pg) clear_page(page) #define copy_user_page(to, from, vaddr, pg) copy_page(to, from) -#if defined(CONFIG_3_LEVEL_PGTABLES) && !defined(CONFIG_64_BIT) +#if defined(CONFIG_3_LEVEL_PGTABLES) && !defined(CONFIG_64BIT) typedef struct { unsigned long pte_low, pte_high; } pte_t; typedef struct { unsigned long long pmd; } pmd_t; diff --git a/include/asm-um/pgtable-3level.h b/include/asm-um/pgtable-3level.h index bdbc3f9..d309f3a 100644 --- a/include/asm-um/pgtable-3level.h +++ b/include/asm-um/pgtable-3level.h @@ -145,7 +145,7 @@ static inline pmd_t pfn_pmd(pfn_t page_nr, pgprot_t pgprot) */ #define PTE_FILE_MAX_BITS 32 -#ifdef CONFIG_64_BIT +#ifdef CONFIG_64BIT #define pte_to_pgoff(p) ((p).pte >> 32) -- cgit v0.10.2 From c16993d9009b4311f0e6088af38844eabc8b5e5b Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Sun, 1 May 2005 08:58:54 -0700 Subject: [PATCH] uml: inline empty proc Cleanup: make an inline of this empty proc. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/kernel/process_kern.c b/arch/um/kernel/process_kern.c index 1d719d5..7a94369 100644 --- a/arch/um/kernel/process_kern.c +++ b/arch/um/kernel/process_kern.c @@ -161,10 +161,6 @@ void *get_current(void) return(current); } -void prepare_to_copy(struct task_struct *tsk) -{ -} - int copy_thread(int nr, unsigned long clone_flags, unsigned long sp, unsigned long stack_top, struct task_struct * p, struct pt_regs *regs) diff --git a/include/asm-um/processor-generic.h b/include/asm-um/processor-generic.h index 038ba6f..4d94049 100644 --- a/include/asm-um/processor-generic.h +++ b/include/asm-um/processor-generic.h @@ -89,7 +89,11 @@ extern struct task_struct *alloc_task_struct(void); extern void release_thread(struct task_struct *); extern int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags); extern void dump_thread(struct pt_regs *regs, struct user *u); -extern void prepare_to_copy(struct task_struct *tsk); + +static inline void prepare_to_copy(struct task_struct *tsk) +{ +} + extern unsigned long thread_saved_pc(struct task_struct *t); -- cgit v0.10.2 From e9c527163d31da9f616e989a90429729525c5233 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Sun, 1 May 2005 08:58:54 -0700 Subject: [PATCH] uml: move va_copy conditional def GCC 2.95 uses __va_copy instead of va_copy. Handle it inside compiler.h instead of in a casual file, and avoid the risk that this breaks with a newer compiler (which it could do). Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/kernel/skas/uaccess.c b/arch/um/kernel/skas/uaccess.c index 7575ec4..f7da9d0 100644 --- a/arch/um/kernel/skas/uaccess.c +++ b/arch/um/kernel/skas/uaccess.c @@ -3,6 +3,7 @@ * Licensed under the GPL */ +#include "linux/compiler.h" #include "linux/stddef.h" #include "linux/kernel.h" #include "linux/string.h" @@ -61,8 +62,7 @@ static void do_buffer_op(void *jmpbuf, void *arg_ptr) void *arg; int *res; - /* Some old gccs recognize __va_copy, but not va_copy */ - __va_copy(args, *(va_list *)arg_ptr); + va_copy(args, *(va_list *)arg_ptr); addr = va_arg(args, unsigned long); len = va_arg(args, int); is_write = va_arg(args, int); diff --git a/include/linux/compiler-gcc2.h b/include/linux/compiler-gcc2.h index 5a35915..ebed1766 100644 --- a/include/linux/compiler-gcc2.h +++ b/include/linux/compiler-gcc2.h @@ -22,3 +22,8 @@ # define __attribute_pure__ __attribute__((pure)) # define __attribute_const__ __attribute__((__const__)) #endif + +/* GCC 2.95.x/2.96 recognize __va_copy, but not va_copy. Actually later GCC's + * define both va_copy and __va_copy, but the latter may go away, so limit this + * to this header */ +#define va_copy __va_copy -- cgit v0.10.2 From 5e7b83ffc67e15791d9bf8b2a18e4f5fd0eb69b8 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Sun, 1 May 2005 08:58:55 -0700 Subject: [PATCH] uml: fix syscall table by including $(SUBARCH)'s one, for i386 Split the i386 entry.S files into entry.S and syscall_table.S which is included in the previous one (so actually there is no difference between them) and use the syscall_table.S in the UML build, instead of tracking by hand the syscall table changes (which is inherently error-prone). We must only insert the right #defines to inject the changes we need from the i386 syscall table (for instance some different function names); also, we don't implement some i386 syscalls, as ioperm(), nor some TLS-related ones (yet to provide). Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S index 25bf758..a991d4e 100644 --- a/arch/i386/kernel/entry.S +++ b/arch/i386/kernel/entry.S @@ -652,296 +652,6 @@ ENTRY(spurious_interrupt_bug) pushl $do_spurious_interrupt_bug jmp error_code -.data -ENTRY(sys_call_table) - .long sys_restart_syscall /* 0 - old "setup()" system call, used for restarting */ - .long sys_exit - .long sys_fork - .long sys_read - .long sys_write - .long sys_open /* 5 */ - .long sys_close - .long sys_waitpid - .long sys_creat - .long sys_link - .long sys_unlink /* 10 */ - .long sys_execve - .long sys_chdir - .long sys_time - .long sys_mknod - .long sys_chmod /* 15 */ - .long sys_lchown16 - .long sys_ni_syscall /* old break syscall holder */ - .long sys_stat - .long sys_lseek - .long sys_getpid /* 20 */ - .long sys_mount - .long sys_oldumount - .long sys_setuid16 - .long sys_getuid16 - .long sys_stime /* 25 */ - .long sys_ptrace - .long sys_alarm - .long sys_fstat - .long sys_pause - .long sys_utime /* 30 */ - .long sys_ni_syscall /* old stty syscall holder */ - .long sys_ni_syscall /* old gtty syscall holder */ - .long sys_access - .long sys_nice - .long sys_ni_syscall /* 35 - old ftime syscall holder */ - .long sys_sync - .long sys_kill - .long sys_rename - .long sys_mkdir - .long sys_rmdir /* 40 */ - .long sys_dup - .long sys_pipe - .long sys_times - .long sys_ni_syscall /* old prof syscall holder */ - .long sys_brk /* 45 */ - .long sys_setgid16 - .long sys_getgid16 - .long sys_signal - .long sys_geteuid16 - .long sys_getegid16 /* 50 */ - .long sys_acct - .long sys_umount /* recycled never used phys() */ - .long sys_ni_syscall /* old lock syscall holder */ - .long sys_ioctl - .long sys_fcntl /* 55 */ - .long sys_ni_syscall /* old mpx syscall holder */ - .long sys_setpgid - .long sys_ni_syscall /* old ulimit syscall holder */ - .long sys_olduname - .long sys_umask /* 60 */ - .long sys_chroot - .long sys_ustat - .long sys_dup2 - .long sys_getppid - .long sys_getpgrp /* 65 */ - .long sys_setsid - .long sys_sigaction - .long sys_sgetmask - .long sys_ssetmask - .long sys_setreuid16 /* 70 */ - .long sys_setregid16 - .long sys_sigsuspend - .long sys_sigpending - .long sys_sethostname - .long sys_setrlimit /* 75 */ - .long sys_old_getrlimit - .long sys_getrusage - .long sys_gettimeofday - .long sys_settimeofday - .long sys_getgroups16 /* 80 */ - .long sys_setgroups16 - .long old_select - .long sys_symlink - .long sys_lstat - .long sys_readlink /* 85 */ - .long sys_uselib - .long sys_swapon - .long sys_reboot - .long old_readdir - .long old_mmap /* 90 */ - .long sys_munmap - .long sys_truncate - .long sys_ftruncate - .long sys_fchmod - .long sys_fchown16 /* 95 */ - .long sys_getpriority - .long sys_setpriority - .long sys_ni_syscall /* old profil syscall holder */ - .long sys_statfs - .long sys_fstatfs /* 100 */ - .long sys_ioperm - .long sys_socketcall - .long sys_syslog - .long sys_setitimer - .long sys_getitimer /* 105 */ - .long sys_newstat - .long sys_newlstat - .long sys_newfstat - .long sys_uname - .long sys_iopl /* 110 */ - .long sys_vhangup - .long sys_ni_syscall /* old "idle" system call */ - .long sys_vm86old - .long sys_wait4 - .long sys_swapoff /* 115 */ - .long sys_sysinfo - .long sys_ipc - .long sys_fsync - .long sys_sigreturn - .long sys_clone /* 120 */ - .long sys_setdomainname - .long sys_newuname - .long sys_modify_ldt - .long sys_adjtimex - .long sys_mprotect /* 125 */ - .long sys_sigprocmask - .long sys_ni_syscall /* old "create_module" */ - .long sys_init_module - .long sys_delete_module - .long sys_ni_syscall /* 130: old "get_kernel_syms" */ - .long sys_quotactl - .long sys_getpgid - .long sys_fchdir - .long sys_bdflush - .long sys_sysfs /* 135 */ - .long sys_personality - .long sys_ni_syscall /* reserved for afs_syscall */ - .long sys_setfsuid16 - .long sys_setfsgid16 - .long sys_llseek /* 140 */ - .long sys_getdents - .long sys_select - .long sys_flock - .long sys_msync - .long sys_readv /* 145 */ - .long sys_writev - .long sys_getsid - .long sys_fdatasync - .long sys_sysctl - .long sys_mlock /* 150 */ - .long sys_munlock - .long sys_mlockall - .long sys_munlockall - .long sys_sched_setparam - .long sys_sched_getparam /* 155 */ - .long sys_sched_setscheduler - .long sys_sched_getscheduler - .long sys_sched_yield - .long sys_sched_get_priority_max - .long sys_sched_get_priority_min /* 160 */ - .long sys_sched_rr_get_interval - .long sys_nanosleep - .long sys_mremap - .long sys_setresuid16 - .long sys_getresuid16 /* 165 */ - .long sys_vm86 - .long sys_ni_syscall /* Old sys_query_module */ - .long sys_poll - .long sys_nfsservctl - .long sys_setresgid16 /* 170 */ - .long sys_getresgid16 - .long sys_prctl - .long sys_rt_sigreturn - .long sys_rt_sigaction - .long sys_rt_sigprocmask /* 175 */ - .long sys_rt_sigpending - .long sys_rt_sigtimedwait - .long sys_rt_sigqueueinfo - .long sys_rt_sigsuspend - .long sys_pread64 /* 180 */ - .long sys_pwrite64 - .long sys_chown16 - .long sys_getcwd - .long sys_capget - .long sys_capset /* 185 */ - .long sys_sigaltstack - .long sys_sendfile - .long sys_ni_syscall /* reserved for streams1 */ - .long sys_ni_syscall /* reserved for streams2 */ - .long sys_vfork /* 190 */ - .long sys_getrlimit - .long sys_mmap2 - .long sys_truncate64 - .long sys_ftruncate64 - .long sys_stat64 /* 195 */ - .long sys_lstat64 - .long sys_fstat64 - .long sys_lchown - .long sys_getuid - .long sys_getgid /* 200 */ - .long sys_geteuid - .long sys_getegid - .long sys_setreuid - .long sys_setregid - .long sys_getgroups /* 205 */ - .long sys_setgroups - .long sys_fchown - .long sys_setresuid - .long sys_getresuid - .long sys_setresgid /* 210 */ - .long sys_getresgid - .long sys_chown - .long sys_setuid - .long sys_setgid - .long sys_setfsuid /* 215 */ - .long sys_setfsgid - .long sys_pivot_root - .long sys_mincore - .long sys_madvise - .long sys_getdents64 /* 220 */ - .long sys_fcntl64 - .long sys_ni_syscall /* reserved for TUX */ - .long sys_ni_syscall - .long sys_gettid - .long sys_readahead /* 225 */ - .long sys_setxattr - .long sys_lsetxattr - .long sys_fsetxattr - .long sys_getxattr - .long sys_lgetxattr /* 230 */ - .long sys_fgetxattr - .long sys_listxattr - .long sys_llistxattr - .long sys_flistxattr - .long sys_removexattr /* 235 */ - .long sys_lremovexattr - .long sys_fremovexattr - .long sys_tkill - .long sys_sendfile64 - .long sys_futex /* 240 */ - .long sys_sched_setaffinity - .long sys_sched_getaffinity - .long sys_set_thread_area - .long sys_get_thread_area - .long sys_io_setup /* 245 */ - .long sys_io_destroy - .long sys_io_getevents - .long sys_io_submit - .long sys_io_cancel - .long sys_fadvise64 /* 250 */ - .long sys_ni_syscall - .long sys_exit_group - .long sys_lookup_dcookie - .long sys_epoll_create - .long sys_epoll_ctl /* 255 */ - .long sys_epoll_wait - .long sys_remap_file_pages - .long sys_set_tid_address - .long sys_timer_create - .long sys_timer_settime /* 260 */ - .long sys_timer_gettime - .long sys_timer_getoverrun - .long sys_timer_delete - .long sys_clock_settime - .long sys_clock_gettime /* 265 */ - .long sys_clock_getres - .long sys_clock_nanosleep - .long sys_statfs64 - .long sys_fstatfs64 - .long sys_tgkill /* 270 */ - .long sys_utimes - .long sys_fadvise64_64 - .long sys_ni_syscall /* sys_vserver */ - .long sys_mbind - .long sys_get_mempolicy - .long sys_set_mempolicy - .long sys_mq_open - .long sys_mq_unlink - .long sys_mq_timedsend - .long sys_mq_timedreceive /* 280 */ - .long sys_mq_notify - .long sys_mq_getsetattr - .long sys_ni_syscall /* reserved for kexec */ - .long sys_waitid - .long sys_ni_syscall /* 285 */ /* available */ - .long sys_add_key - .long sys_request_key - .long sys_keyctl +#include "syscall_table.S" syscall_table_size=(.-sys_call_table) diff --git a/arch/i386/kernel/syscall_table.S b/arch/i386/kernel/syscall_table.S new file mode 100644 index 0000000..6cd1ed3 --- /dev/null +++ b/arch/i386/kernel/syscall_table.S @@ -0,0 +1,291 @@ +.data +ENTRY(sys_call_table) + .long sys_restart_syscall /* 0 - old "setup()" system call, used for restarting */ + .long sys_exit + .long sys_fork + .long sys_read + .long sys_write + .long sys_open /* 5 */ + .long sys_close + .long sys_waitpid + .long sys_creat + .long sys_link + .long sys_unlink /* 10 */ + .long sys_execve + .long sys_chdir + .long sys_time + .long sys_mknod + .long sys_chmod /* 15 */ + .long sys_lchown16 + .long sys_ni_syscall /* old break syscall holder */ + .long sys_stat + .long sys_lseek + .long sys_getpid /* 20 */ + .long sys_mount + .long sys_oldumount + .long sys_setuid16 + .long sys_getuid16 + .long sys_stime /* 25 */ + .long sys_ptrace + .long sys_alarm + .long sys_fstat + .long sys_pause + .long sys_utime /* 30 */ + .long sys_ni_syscall /* old stty syscall holder */ + .long sys_ni_syscall /* old gtty syscall holder */ + .long sys_access + .long sys_nice + .long sys_ni_syscall /* 35 - old ftime syscall holder */ + .long sys_sync + .long sys_kill + .long sys_rename + .long sys_mkdir + .long sys_rmdir /* 40 */ + .long sys_dup + .long sys_pipe + .long sys_times + .long sys_ni_syscall /* old prof syscall holder */ + .long sys_brk /* 45 */ + .long sys_setgid16 + .long sys_getgid16 + .long sys_signal + .long sys_geteuid16 + .long sys_getegid16 /* 50 */ + .long sys_acct + .long sys_umount /* recycled never used phys() */ + .long sys_ni_syscall /* old lock syscall holder */ + .long sys_ioctl + .long sys_fcntl /* 55 */ + .long sys_ni_syscall /* old mpx syscall holder */ + .long sys_setpgid + .long sys_ni_syscall /* old ulimit syscall holder */ + .long sys_olduname + .long sys_umask /* 60 */ + .long sys_chroot + .long sys_ustat + .long sys_dup2 + .long sys_getppid + .long sys_getpgrp /* 65 */ + .long sys_setsid + .long sys_sigaction + .long sys_sgetmask + .long sys_ssetmask + .long sys_setreuid16 /* 70 */ + .long sys_setregid16 + .long sys_sigsuspend + .long sys_sigpending + .long sys_sethostname + .long sys_setrlimit /* 75 */ + .long sys_old_getrlimit + .long sys_getrusage + .long sys_gettimeofday + .long sys_settimeofday + .long sys_getgroups16 /* 80 */ + .long sys_setgroups16 + .long old_select + .long sys_symlink + .long sys_lstat + .long sys_readlink /* 85 */ + .long sys_uselib + .long sys_swapon + .long sys_reboot + .long old_readdir + .long old_mmap /* 90 */ + .long sys_munmap + .long sys_truncate + .long sys_ftruncate + .long sys_fchmod + .long sys_fchown16 /* 95 */ + .long sys_getpriority + .long sys_setpriority + .long sys_ni_syscall /* old profil syscall holder */ + .long sys_statfs + .long sys_fstatfs /* 100 */ + .long sys_ioperm + .long sys_socketcall + .long sys_syslog + .long sys_setitimer + .long sys_getitimer /* 105 */ + .long sys_newstat + .long sys_newlstat + .long sys_newfstat + .long sys_uname + .long sys_iopl /* 110 */ + .long sys_vhangup + .long sys_ni_syscall /* old "idle" system call */ + .long sys_vm86old + .long sys_wait4 + .long sys_swapoff /* 115 */ + .long sys_sysinfo + .long sys_ipc + .long sys_fsync + .long sys_sigreturn + .long sys_clone /* 120 */ + .long sys_setdomainname + .long sys_newuname + .long sys_modify_ldt + .long sys_adjtimex + .long sys_mprotect /* 125 */ + .long sys_sigprocmask + .long sys_ni_syscall /* old "create_module" */ + .long sys_init_module + .long sys_delete_module + .long sys_ni_syscall /* 130: old "get_kernel_syms" */ + .long sys_quotactl + .long sys_getpgid + .long sys_fchdir + .long sys_bdflush + .long sys_sysfs /* 135 */ + .long sys_personality + .long sys_ni_syscall /* reserved for afs_syscall */ + .long sys_setfsuid16 + .long sys_setfsgid16 + .long sys_llseek /* 140 */ + .long sys_getdents + .long sys_select + .long sys_flock + .long sys_msync + .long sys_readv /* 145 */ + .long sys_writev + .long sys_getsid + .long sys_fdatasync + .long sys_sysctl + .long sys_mlock /* 150 */ + .long sys_munlock + .long sys_mlockall + .long sys_munlockall + .long sys_sched_setparam + .long sys_sched_getparam /* 155 */ + .long sys_sched_setscheduler + .long sys_sched_getscheduler + .long sys_sched_yield + .long sys_sched_get_priority_max + .long sys_sched_get_priority_min /* 160 */ + .long sys_sched_rr_get_interval + .long sys_nanosleep + .long sys_mremap + .long sys_setresuid16 + .long sys_getresuid16 /* 165 */ + .long sys_vm86 + .long sys_ni_syscall /* Old sys_query_module */ + .long sys_poll + .long sys_nfsservctl + .long sys_setresgid16 /* 170 */ + .long sys_getresgid16 + .long sys_prctl + .long sys_rt_sigreturn + .long sys_rt_sigaction + .long sys_rt_sigprocmask /* 175 */ + .long sys_rt_sigpending + .long sys_rt_sigtimedwait + .long sys_rt_sigqueueinfo + .long sys_rt_sigsuspend + .long sys_pread64 /* 180 */ + .long sys_pwrite64 + .long sys_chown16 + .long sys_getcwd + .long sys_capget + .long sys_capset /* 185 */ + .long sys_sigaltstack + .long sys_sendfile + .long sys_ni_syscall /* reserved for streams1 */ + .long sys_ni_syscall /* reserved for streams2 */ + .long sys_vfork /* 190 */ + .long sys_getrlimit + .long sys_mmap2 + .long sys_truncate64 + .long sys_ftruncate64 + .long sys_stat64 /* 195 */ + .long sys_lstat64 + .long sys_fstat64 + .long sys_lchown + .long sys_getuid + .long sys_getgid /* 200 */ + .long sys_geteuid + .long sys_getegid + .long sys_setreuid + .long sys_setregid + .long sys_getgroups /* 205 */ + .long sys_setgroups + .long sys_fchown + .long sys_setresuid + .long sys_getresuid + .long sys_setresgid /* 210 */ + .long sys_getresgid + .long sys_chown + .long sys_setuid + .long sys_setgid + .long sys_setfsuid /* 215 */ + .long sys_setfsgid + .long sys_pivot_root + .long sys_mincore + .long sys_madvise + .long sys_getdents64 /* 220 */ + .long sys_fcntl64 + .long sys_ni_syscall /* reserved for TUX */ + .long sys_ni_syscall + .long sys_gettid + .long sys_readahead /* 225 */ + .long sys_setxattr + .long sys_lsetxattr + .long sys_fsetxattr + .long sys_getxattr + .long sys_lgetxattr /* 230 */ + .long sys_fgetxattr + .long sys_listxattr + .long sys_llistxattr + .long sys_flistxattr + .long sys_removexattr /* 235 */ + .long sys_lremovexattr + .long sys_fremovexattr + .long sys_tkill + .long sys_sendfile64 + .long sys_futex /* 240 */ + .long sys_sched_setaffinity + .long sys_sched_getaffinity + .long sys_set_thread_area + .long sys_get_thread_area + .long sys_io_setup /* 245 */ + .long sys_io_destroy + .long sys_io_getevents + .long sys_io_submit + .long sys_io_cancel + .long sys_fadvise64 /* 250 */ + .long sys_ni_syscall + .long sys_exit_group + .long sys_lookup_dcookie + .long sys_epoll_create + .long sys_epoll_ctl /* 255 */ + .long sys_epoll_wait + .long sys_remap_file_pages + .long sys_set_tid_address + .long sys_timer_create + .long sys_timer_settime /* 260 */ + .long sys_timer_gettime + .long sys_timer_getoverrun + .long sys_timer_delete + .long sys_clock_settime + .long sys_clock_gettime /* 265 */ + .long sys_clock_getres + .long sys_clock_nanosleep + .long sys_statfs64 + .long sys_fstatfs64 + .long sys_tgkill /* 270 */ + .long sys_utimes + .long sys_fadvise64_64 + .long sys_ni_syscall /* sys_vserver */ + .long sys_mbind + .long sys_get_mempolicy + .long sys_set_mempolicy + .long sys_mq_open + .long sys_mq_unlink + .long sys_mq_timedsend + .long sys_mq_timedreceive /* 280 */ + .long sys_mq_notify + .long sys_mq_getsetattr + .long sys_ni_syscall /* reserved for kexec */ + .long sys_waitid + .long sys_ni_syscall /* 285 */ /* available */ + .long sys_add_key + .long sys_request_key + .long sys_keyctl diff --git a/arch/um/include/sysdep-i386/syscalls.h b/arch/um/include/sysdep-i386/syscalls.h index 5db81ec..be0a3e3 100644 --- a/arch/um/include/sysdep-i386/syscalls.h +++ b/arch/um/include/sysdep-i386/syscalls.h @@ -22,102 +22,3 @@ extern syscall_handler_t old_mmap_i386; extern long sys_mmap2(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long pgoff); - -/* On i386 they choose a meaningless naming.*/ -#define __NR_kexec_load __NR_sys_kexec_load - -#define ARCH_SYSCALLS \ - [ __NR_waitpid ] = (syscall_handler_t *) sys_waitpid, \ - [ __NR_break ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_oldstat ] = (syscall_handler_t *) sys_stat, \ - [ __NR_umount ] = (syscall_handler_t *) sys_oldumount, \ - [ __NR_stime ] = um_stime, \ - [ __NR_oldfstat ] = (syscall_handler_t *) sys_fstat, \ - [ __NR_stty ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_gtty ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_nice ] = (syscall_handler_t *) sys_nice, \ - [ __NR_ftime ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_prof ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_signal ] = (syscall_handler_t *) sys_signal, \ - [ __NR_lock ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_mpx ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_ulimit ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_oldolduname ] = (syscall_handler_t *) sys_olduname, \ - [ __NR_sigaction ] = (syscall_handler_t *) sys_sigaction, \ - [ __NR_sgetmask ] = (syscall_handler_t *) sys_sgetmask, \ - [ __NR_ssetmask ] = (syscall_handler_t *) sys_ssetmask, \ - [ __NR_sigsuspend ] = (syscall_handler_t *) sys_sigsuspend, \ - [ __NR_sigpending ] = (syscall_handler_t *) sys_sigpending, \ - [ __NR_oldlstat ] = (syscall_handler_t *) sys_lstat, \ - [ __NR_readdir ] = old_readdir, \ - [ __NR_profil ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_socketcall ] = (syscall_handler_t *) sys_socketcall, \ - [ __NR_olduname ] = (syscall_handler_t *) sys_uname, \ - [ __NR_iopl ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_idle ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_ipc ] = (syscall_handler_t *) sys_ipc, \ - [ __NR_sigreturn ] = (syscall_handler_t *) sys_sigreturn, \ - [ __NR_sigprocmask ] = (syscall_handler_t *) sys_sigprocmask, \ - [ __NR_bdflush ] = (syscall_handler_t *) sys_bdflush, \ - [ __NR__llseek ] = (syscall_handler_t *) sys_llseek, \ - [ __NR__newselect ] = (syscall_handler_t *) sys_select, \ - [ __NR_vm86 ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_mmap ] = (syscall_handler_t *) old_mmap_i386, \ - [ __NR_ugetrlimit ] = (syscall_handler_t *) sys_getrlimit, \ - [ __NR_mmap2 ] = (syscall_handler_t *) sys_mmap2, \ - [ __NR_truncate64 ] = (syscall_handler_t *) sys_truncate64, \ - [ __NR_ftruncate64 ] = (syscall_handler_t *) sys_ftruncate64, \ - [ __NR_stat64 ] = (syscall_handler_t *) sys_stat64, \ - [ __NR_lstat64 ] = (syscall_handler_t *) sys_lstat64, \ - [ __NR_fstat64 ] = (syscall_handler_t *) sys_fstat64, \ - [ __NR_fcntl64 ] = (syscall_handler_t *) sys_fcntl64, \ - [ __NR_sendfile64 ] = (syscall_handler_t *) sys_sendfile64, \ - [ __NR_statfs64 ] = (syscall_handler_t *) sys_statfs64, \ - [ __NR_fstatfs64 ] = (syscall_handler_t *) sys_fstatfs64, \ - [ __NR_fadvise64_64 ] = (syscall_handler_t *) sys_fadvise64_64, \ - [ __NR_select ] = (syscall_handler_t *) old_select, \ - [ __NR_vm86old ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_modify_ldt ] = (syscall_handler_t *) sys_modify_ldt, \ - [ __NR_lchown32 ] = (syscall_handler_t *) sys_lchown, \ - [ __NR_getuid32 ] = (syscall_handler_t *) sys_getuid, \ - [ __NR_getgid32 ] = (syscall_handler_t *) sys_getgid, \ - [ __NR_geteuid32 ] = (syscall_handler_t *) sys_geteuid, \ - [ __NR_getegid32 ] = (syscall_handler_t *) sys_getegid, \ - [ __NR_setreuid32 ] = (syscall_handler_t *) sys_setreuid, \ - [ __NR_setregid32 ] = (syscall_handler_t *) sys_setregid, \ - [ __NR_getgroups32 ] = (syscall_handler_t *) sys_getgroups, \ - [ __NR_setgroups32 ] = (syscall_handler_t *) sys_setgroups, \ - [ __NR_fchown32 ] = (syscall_handler_t *) sys_fchown, \ - [ __NR_setresuid32 ] = (syscall_handler_t *) sys_setresuid, \ - [ __NR_getresuid32 ] = (syscall_handler_t *) sys_getresuid, \ - [ __NR_setresgid32 ] = (syscall_handler_t *) sys_setresgid, \ - [ __NR_getresgid32 ] = (syscall_handler_t *) sys_getresgid, \ - [ __NR_chown32 ] = (syscall_handler_t *) sys_chown, \ - [ __NR_setuid32 ] = (syscall_handler_t *) sys_setuid, \ - [ __NR_setgid32 ] = (syscall_handler_t *) sys_setgid, \ - [ __NR_setfsuid32 ] = (syscall_handler_t *) sys_setfsuid, \ - [ __NR_setfsgid32 ] = (syscall_handler_t *) sys_setfsgid, \ - [ __NR_pivot_root ] = (syscall_handler_t *) sys_pivot_root, \ - [ __NR_mincore ] = (syscall_handler_t *) sys_mincore, \ - [ __NR_madvise ] = (syscall_handler_t *) sys_madvise, \ - [ 222 ] = (syscall_handler_t *) sys_ni_syscall, \ - [ 223 ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_set_thread_area ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_get_thread_area ] = (syscall_handler_t *) sys_ni_syscall, \ - [ 251 ] = (syscall_handler_t *) sys_ni_syscall, \ - [ 285 ] = (syscall_handler_t *) sys_ni_syscall, - -/* 222 doesn't yet have a name in include/asm-i386/unistd.h */ - -#define LAST_ARCH_SYSCALL 285 - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/include/sysdep-x86_64/syscalls.h b/arch/um/include/sysdep-x86_64/syscalls.h index b187a41..65a01e3 100644 --- a/arch/um/include/sysdep-x86_64/syscalls.h +++ b/arch/um/include/sysdep-x86_64/syscalls.h @@ -78,14 +78,3 @@ extern syscall_handler_t sys_arch_prctl; #define NR_syscalls 1024 #endif - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/kernel/Makefile b/arch/um/kernel/Makefile index 280d300..246f0e7 100644 --- a/arch/um/kernel/Makefile +++ b/arch/um/kernel/Makefile @@ -10,7 +10,7 @@ obj-y = checksum.o config.o exec_kern.o exitcode.o \ helper.o init_task.o irq.o irq_user.o ksyms.o main.o mem.o mem_user.o \ physmem.o process.o process_kern.o ptrace.o reboot.o resource.o \ sigio_user.o sigio_kern.o signal_kern.o signal_user.o smp.o \ - syscall_kern.o sysrq.o sys_call_table.o tempfile.o time.o time_kern.o \ + syscall_kern.o sysrq.o tempfile.o time.o time_kern.o \ tlb.o trap_kern.o trap_user.o uaccess_user.o um_arch.o umid.o \ user_util.o diff --git a/arch/um/kernel/sys_call_table.c b/arch/um/kernel/sys_call_table.c deleted file mode 100644 index b671a31..0000000 --- a/arch/um/kernel/sys_call_table.c +++ /dev/null @@ -1,270 +0,0 @@ -/* - * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) - * Copyright 2003 PathScale, Inc. - * Licensed under the GPL - */ - -#include "linux/config.h" -#include "linux/unistd.h" -#include "linux/sys.h" -#include "linux/swap.h" -#include "linux/syscalls.h" -#include "linux/sysctl.h" -#include "asm/signal.h" -#include "sysdep/syscalls.h" -#include "kern_util.h" - -#define LAST_GENERIC_SYSCALL __NR_keyctl - -#if LAST_GENERIC_SYSCALL > LAST_ARCH_SYSCALL -#define LAST_SYSCALL LAST_GENERIC_SYSCALL -#else -#define LAST_SYSCALL LAST_ARCH_SYSCALL -#endif - -extern syscall_handler_t sys_fork; -extern syscall_handler_t sys_execve; -extern syscall_handler_t um_time; -extern syscall_handler_t um_stime; -extern syscall_handler_t sys_pipe; -extern syscall_handler_t sys_olduname; -extern syscall_handler_t sys_sigaction; -extern syscall_handler_t sys_sigsuspend; -extern syscall_handler_t old_readdir; -extern syscall_handler_t sys_uname; -extern syscall_handler_t sys_ipc; -extern syscall_handler_t sys_sigreturn; -extern syscall_handler_t sys_clone; -extern syscall_handler_t sys_rt_sigreturn; -extern syscall_handler_t sys_sigaltstack; -extern syscall_handler_t sys_vfork; -extern syscall_handler_t old_select; -extern syscall_handler_t sys_modify_ldt; -extern syscall_handler_t sys_rt_sigsuspend; -extern syscall_handler_t sys_mbind; -extern syscall_handler_t sys_get_mempolicy; -extern syscall_handler_t sys_set_mempolicy; -extern syscall_handler_t sys_sys_setaltroot; - -syscall_handler_t *sys_call_table[] = { - [ __NR_restart_syscall ] = (syscall_handler_t *) sys_restart_syscall, - [ __NR_exit ] = (syscall_handler_t *) sys_exit, - [ __NR_fork ] = (syscall_handler_t *) sys_fork, - [ __NR_read ] = (syscall_handler_t *) sys_read, - [ __NR_write ] = (syscall_handler_t *) sys_write, - - /* These three are declared differently in asm/unistd.h */ - [ __NR_open ] = (syscall_handler_t *) sys_open, - [ __NR_close ] = (syscall_handler_t *) sys_close, - [ __NR_creat ] = (syscall_handler_t *) sys_creat, - [ __NR_link ] = (syscall_handler_t *) sys_link, - [ __NR_unlink ] = (syscall_handler_t *) sys_unlink, - [ __NR_execve ] = (syscall_handler_t *) sys_execve, - - /* declared differently in kern_util.h */ - [ __NR_chdir ] = (syscall_handler_t *) sys_chdir, - [ __NR_time ] = um_time, - [ __NR_mknod ] = (syscall_handler_t *) sys_mknod, - [ __NR_chmod ] = (syscall_handler_t *) sys_chmod, - [ __NR_lchown ] = (syscall_handler_t *) sys_lchown16, - [ __NR_lseek ] = (syscall_handler_t *) sys_lseek, - [ __NR_getpid ] = (syscall_handler_t *) sys_getpid, - [ __NR_mount ] = (syscall_handler_t *) sys_mount, - [ __NR_setuid ] = (syscall_handler_t *) sys_setuid16, - [ __NR_getuid ] = (syscall_handler_t *) sys_getuid16, - [ __NR_ptrace ] = (syscall_handler_t *) sys_ptrace, - [ __NR_alarm ] = (syscall_handler_t *) sys_alarm, - [ __NR_pause ] = (syscall_handler_t *) sys_pause, - [ __NR_utime ] = (syscall_handler_t *) sys_utime, - [ __NR_access ] = (syscall_handler_t *) sys_access, - [ __NR_sync ] = (syscall_handler_t *) sys_sync, - [ __NR_kill ] = (syscall_handler_t *) sys_kill, - [ __NR_rename ] = (syscall_handler_t *) sys_rename, - [ __NR_mkdir ] = (syscall_handler_t *) sys_mkdir, - [ __NR_rmdir ] = (syscall_handler_t *) sys_rmdir, - - /* Declared differently in asm/unistd.h */ - [ __NR_dup ] = (syscall_handler_t *) sys_dup, - [ __NR_pipe ] = (syscall_handler_t *) sys_pipe, - [ __NR_times ] = (syscall_handler_t *) sys_times, - [ __NR_brk ] = (syscall_handler_t *) sys_brk, - [ __NR_setgid ] = (syscall_handler_t *) sys_setgid16, - [ __NR_getgid ] = (syscall_handler_t *) sys_getgid16, - [ __NR_geteuid ] = (syscall_handler_t *) sys_geteuid16, - [ __NR_getegid ] = (syscall_handler_t *) sys_getegid16, - [ __NR_acct ] = (syscall_handler_t *) sys_acct, - [ __NR_umount2 ] = (syscall_handler_t *) sys_umount, - [ __NR_ioctl ] = (syscall_handler_t *) sys_ioctl, - [ __NR_fcntl ] = (syscall_handler_t *) sys_fcntl, - [ __NR_setpgid ] = (syscall_handler_t *) sys_setpgid, - [ __NR_umask ] = (syscall_handler_t *) sys_umask, - [ __NR_chroot ] = (syscall_handler_t *) sys_chroot, - [ __NR_ustat ] = (syscall_handler_t *) sys_ustat, - [ __NR_dup2 ] = (syscall_handler_t *) sys_dup2, - [ __NR_getppid ] = (syscall_handler_t *) sys_getppid, - [ __NR_getpgrp ] = (syscall_handler_t *) sys_getpgrp, - [ __NR_setsid ] = (syscall_handler_t *) sys_setsid, - [ __NR_setreuid ] = (syscall_handler_t *) sys_setreuid16, - [ __NR_setregid ] = (syscall_handler_t *) sys_setregid16, - [ __NR_sethostname ] = (syscall_handler_t *) sys_sethostname, - [ __NR_setrlimit ] = (syscall_handler_t *) sys_setrlimit, - [ __NR_getrlimit ] = (syscall_handler_t *) sys_old_getrlimit, - [ __NR_getrusage ] = (syscall_handler_t *) sys_getrusage, - [ __NR_gettimeofday ] = (syscall_handler_t *) sys_gettimeofday, - [ __NR_settimeofday ] = (syscall_handler_t *) sys_settimeofday, - [ __NR_getgroups ] = (syscall_handler_t *) sys_getgroups16, - [ __NR_setgroups ] = (syscall_handler_t *) sys_setgroups16, - [ __NR_symlink ] = (syscall_handler_t *) sys_symlink, - [ __NR_readlink ] = (syscall_handler_t *) sys_readlink, - [ __NR_uselib ] = (syscall_handler_t *) sys_uselib, - [ __NR_swapon ] = (syscall_handler_t *) sys_swapon, - [ __NR_reboot ] = (syscall_handler_t *) sys_reboot, - [ __NR_munmap ] = (syscall_handler_t *) sys_munmap, - [ __NR_truncate ] = (syscall_handler_t *) sys_truncate, - [ __NR_ftruncate ] = (syscall_handler_t *) sys_ftruncate, - [ __NR_fchmod ] = (syscall_handler_t *) sys_fchmod, - [ __NR_fchown ] = (syscall_handler_t *) sys_fchown16, - [ __NR_getpriority ] = (syscall_handler_t *) sys_getpriority, - [ __NR_setpriority ] = (syscall_handler_t *) sys_setpriority, - [ __NR_statfs ] = (syscall_handler_t *) sys_statfs, - [ __NR_fstatfs ] = (syscall_handler_t *) sys_fstatfs, - [ __NR_ioperm ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_syslog ] = (syscall_handler_t *) sys_syslog, - [ __NR_setitimer ] = (syscall_handler_t *) sys_setitimer, - [ __NR_getitimer ] = (syscall_handler_t *) sys_getitimer, - [ __NR_stat ] = (syscall_handler_t *) sys_newstat, - [ __NR_lstat ] = (syscall_handler_t *) sys_newlstat, - [ __NR_fstat ] = (syscall_handler_t *) sys_newfstat, - [ __NR_vhangup ] = (syscall_handler_t *) sys_vhangup, - [ __NR_wait4 ] = (syscall_handler_t *) sys_wait4, - [ __NR_swapoff ] = (syscall_handler_t *) sys_swapoff, - [ __NR_sysinfo ] = (syscall_handler_t *) sys_sysinfo, - [ __NR_fsync ] = (syscall_handler_t *) sys_fsync, - [ __NR_clone ] = (syscall_handler_t *) sys_clone, - [ __NR_setdomainname ] = (syscall_handler_t *) sys_setdomainname, - [ __NR_uname ] = (syscall_handler_t *) sys_newuname, - [ __NR_adjtimex ] = (syscall_handler_t *) sys_adjtimex, - [ __NR_mprotect ] = (syscall_handler_t *) sys_mprotect, - [ __NR_create_module ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_init_module ] = (syscall_handler_t *) sys_init_module, - [ __NR_delete_module ] = (syscall_handler_t *) sys_delete_module, - [ __NR_get_kernel_syms ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_quotactl ] = (syscall_handler_t *) sys_quotactl, - [ __NR_getpgid ] = (syscall_handler_t *) sys_getpgid, - [ __NR_fchdir ] = (syscall_handler_t *) sys_fchdir, - [ __NR_sysfs ] = (syscall_handler_t *) sys_sysfs, - [ __NR_personality ] = (syscall_handler_t *) sys_personality, - [ __NR_afs_syscall ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_setfsuid ] = (syscall_handler_t *) sys_setfsuid16, - [ __NR_setfsgid ] = (syscall_handler_t *) sys_setfsgid16, - [ __NR_getdents ] = (syscall_handler_t *) sys_getdents, - [ __NR_flock ] = (syscall_handler_t *) sys_flock, - [ __NR_msync ] = (syscall_handler_t *) sys_msync, - [ __NR_readv ] = (syscall_handler_t *) sys_readv, - [ __NR_writev ] = (syscall_handler_t *) sys_writev, - [ __NR_getsid ] = (syscall_handler_t *) sys_getsid, - [ __NR_fdatasync ] = (syscall_handler_t *) sys_fdatasync, - [ __NR__sysctl ] = (syscall_handler_t *) sys_sysctl, - [ __NR_mlock ] = (syscall_handler_t *) sys_mlock, - [ __NR_munlock ] = (syscall_handler_t *) sys_munlock, - [ __NR_mlockall ] = (syscall_handler_t *) sys_mlockall, - [ __NR_munlockall ] = (syscall_handler_t *) sys_munlockall, - [ __NR_sched_setparam ] = (syscall_handler_t *) sys_sched_setparam, - [ __NR_sched_getparam ] = (syscall_handler_t *) sys_sched_getparam, - [ __NR_sched_setscheduler ] = (syscall_handler_t *) sys_sched_setscheduler, - [ __NR_sched_getscheduler ] = (syscall_handler_t *) sys_sched_getscheduler, - [ __NR_sched_yield ] = (syscall_handler_t *) yield, - [ __NR_sched_get_priority_max ] = (syscall_handler_t *) sys_sched_get_priority_max, - [ __NR_sched_get_priority_min ] = (syscall_handler_t *) sys_sched_get_priority_min, - [ __NR_sched_rr_get_interval ] = (syscall_handler_t *) sys_sched_rr_get_interval, - [ __NR_nanosleep ] = (syscall_handler_t *) sys_nanosleep, - [ __NR_mremap ] = (syscall_handler_t *) sys_mremap, - [ __NR_setresuid ] = (syscall_handler_t *) sys_setresuid16, - [ __NR_getresuid ] = (syscall_handler_t *) sys_getresuid16, - [ __NR_query_module ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_poll ] = (syscall_handler_t *) sys_poll, - [ __NR_nfsservctl ] = (syscall_handler_t *) sys_nfsservctl, - [ __NR_setresgid ] = (syscall_handler_t *) sys_setresgid16, - [ __NR_getresgid ] = (syscall_handler_t *) sys_getresgid16, - [ __NR_prctl ] = (syscall_handler_t *) sys_prctl, - [ __NR_rt_sigreturn ] = (syscall_handler_t *) sys_rt_sigreturn, - [ __NR_rt_sigaction ] = (syscall_handler_t *) sys_rt_sigaction, - [ __NR_rt_sigprocmask ] = (syscall_handler_t *) sys_rt_sigprocmask, - [ __NR_rt_sigpending ] = (syscall_handler_t *) sys_rt_sigpending, - [ __NR_rt_sigtimedwait ] = (syscall_handler_t *) sys_rt_sigtimedwait, - [ __NR_rt_sigqueueinfo ] = (syscall_handler_t *) sys_rt_sigqueueinfo, - [ __NR_rt_sigsuspend ] = (syscall_handler_t *) sys_rt_sigsuspend, - [ __NR_pread64 ] = (syscall_handler_t *) sys_pread64, - [ __NR_pwrite64 ] = (syscall_handler_t *) sys_pwrite64, - [ __NR_chown ] = (syscall_handler_t *) sys_chown16, - [ __NR_getcwd ] = (syscall_handler_t *) sys_getcwd, - [ __NR_capget ] = (syscall_handler_t *) sys_capget, - [ __NR_capset ] = (syscall_handler_t *) sys_capset, - [ __NR_sigaltstack ] = (syscall_handler_t *) sys_sigaltstack, - [ __NR_sendfile ] = (syscall_handler_t *) sys_sendfile, - [ __NR_getpmsg ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_putpmsg ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_vfork ] = (syscall_handler_t *) sys_vfork, - [ __NR_getdents64 ] = (syscall_handler_t *) sys_getdents64, - [ __NR_gettid ] = (syscall_handler_t *) sys_gettid, - [ __NR_readahead ] = (syscall_handler_t *) sys_readahead, - [ __NR_setxattr ] = (syscall_handler_t *) sys_setxattr, - [ __NR_lsetxattr ] = (syscall_handler_t *) sys_lsetxattr, - [ __NR_fsetxattr ] = (syscall_handler_t *) sys_fsetxattr, - [ __NR_getxattr ] = (syscall_handler_t *) sys_getxattr, - [ __NR_lgetxattr ] = (syscall_handler_t *) sys_lgetxattr, - [ __NR_fgetxattr ] = (syscall_handler_t *) sys_fgetxattr, - [ __NR_listxattr ] = (syscall_handler_t *) sys_listxattr, - [ __NR_llistxattr ] = (syscall_handler_t *) sys_llistxattr, - [ __NR_flistxattr ] = (syscall_handler_t *) sys_flistxattr, - [ __NR_removexattr ] = (syscall_handler_t *) sys_removexattr, - [ __NR_lremovexattr ] = (syscall_handler_t *) sys_lremovexattr, - [ __NR_fremovexattr ] = (syscall_handler_t *) sys_fremovexattr, - [ __NR_tkill ] = (syscall_handler_t *) sys_tkill, - [ __NR_futex ] = (syscall_handler_t *) sys_futex, - [ __NR_sched_setaffinity ] = (syscall_handler_t *) sys_sched_setaffinity, - [ __NR_sched_getaffinity ] = (syscall_handler_t *) sys_sched_getaffinity, - [ __NR_io_setup ] = (syscall_handler_t *) sys_io_setup, - [ __NR_io_destroy ] = (syscall_handler_t *) sys_io_destroy, - [ __NR_io_getevents ] = (syscall_handler_t *) sys_io_getevents, - [ __NR_io_submit ] = (syscall_handler_t *) sys_io_submit, - [ __NR_io_cancel ] = (syscall_handler_t *) sys_io_cancel, - [ __NR_exit_group ] = (syscall_handler_t *) sys_exit_group, - [ __NR_lookup_dcookie ] = (syscall_handler_t *) sys_lookup_dcookie, - [ __NR_epoll_create ] = (syscall_handler_t *) sys_epoll_create, - [ __NR_epoll_ctl ] = (syscall_handler_t *) sys_epoll_ctl, - [ __NR_epoll_wait ] = (syscall_handler_t *) sys_epoll_wait, - [ __NR_remap_file_pages ] = (syscall_handler_t *) sys_remap_file_pages, - [ __NR_set_tid_address ] = (syscall_handler_t *) sys_set_tid_address, - [ __NR_timer_create ] = (syscall_handler_t *) sys_timer_create, - [ __NR_timer_settime ] = (syscall_handler_t *) sys_timer_settime, - [ __NR_timer_gettime ] = (syscall_handler_t *) sys_timer_gettime, - [ __NR_timer_getoverrun ] = (syscall_handler_t *) sys_timer_getoverrun, - [ __NR_timer_delete ] = (syscall_handler_t *) sys_timer_delete, - [ __NR_clock_settime ] = (syscall_handler_t *) sys_clock_settime, - [ __NR_clock_gettime ] = (syscall_handler_t *) sys_clock_gettime, - [ __NR_clock_getres ] = (syscall_handler_t *) sys_clock_getres, - [ __NR_clock_nanosleep ] = (syscall_handler_t *) sys_clock_nanosleep, - [ __NR_tgkill ] = (syscall_handler_t *) sys_tgkill, - [ __NR_utimes ] = (syscall_handler_t *) sys_utimes, - [ __NR_fadvise64 ] = (syscall_handler_t *) sys_fadvise64, - [ __NR_vserver ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_mbind ] = (syscall_handler_t *) sys_mbind, - [ __NR_get_mempolicy ] = (syscall_handler_t *) sys_get_mempolicy, - [ __NR_set_mempolicy ] = (syscall_handler_t *) sys_set_mempolicy, - [ __NR_mq_open ] = (syscall_handler_t *) sys_mq_open, - [ __NR_mq_unlink ] = (syscall_handler_t *) sys_mq_unlink, - [ __NR_mq_timedsend ] = (syscall_handler_t *) sys_mq_timedsend, - [ __NR_mq_timedreceive ] = (syscall_handler_t *) sys_mq_timedreceive, - [ __NR_mq_notify ] = (syscall_handler_t *) sys_mq_notify, - [ __NR_mq_getsetattr ] = (syscall_handler_t *) sys_mq_getsetattr, - [ __NR_kexec_load ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_waitid ] = (syscall_handler_t *) sys_waitid, - [ __NR_add_key ] = (syscall_handler_t *) sys_add_key, - [ __NR_request_key ] = (syscall_handler_t *) sys_request_key, - [ __NR_keyctl ] = (syscall_handler_t *) sys_keyctl, - - ARCH_SYSCALLS - [ LAST_SYSCALL + 1 ... NR_syscalls ] = - (syscall_handler_t *) sys_ni_syscall -}; diff --git a/arch/um/sys-i386/Makefile b/arch/um/sys-i386/Makefile index 71b47e6..950781e 100644 --- a/arch/um/sys-i386/Makefile +++ b/arch/um/sys-i386/Makefile @@ -1,5 +1,6 @@ obj-y = bitops.o bugs.o checksum.o delay.o fault.o ksyms.o ldt.o ptrace.o \ - ptrace_user.o semaphore.o signal.o sigcontext.o syscalls.o sysrq.o + ptrace_user.o semaphore.o signal.o sigcontext.o syscalls.o sysrq.o \ + sys_call_table.o obj-$(CONFIG_HIGHMEM) += highmem.o obj-$(CONFIG_MODULES) += module.o diff --git a/arch/um/sys-i386/sys_call_table.S b/arch/um/sys-i386/sys_call_table.S new file mode 100644 index 0000000..ad75c27 --- /dev/null +++ b/arch/um/sys-i386/sys_call_table.S @@ -0,0 +1,16 @@ +#include +/* Steal i386 syscall table for our purposes, but with some slight changes.*/ + +#define sys_iopl sys_ni_syscall +#define sys_ioperm sys_ni_syscall + +#define sys_vm86old sys_ni_syscall +#define sys_vm86 sys_ni_syscall +#define sys_set_thread_area sys_ni_syscall +#define sys_get_thread_area sys_ni_syscall + +#define sys_stime um_stime +#define sys_time um_time +#define old_mmap old_mmap_i386 + +#include "../../i386/kernel/syscall_table.S" diff --git a/arch/um/sys-x86_64/Makefile b/arch/um/sys-x86_64/Makefile index 2129e31..274158e 100644 --- a/arch/um/sys-x86_64/Makefile +++ b/arch/um/sys-x86_64/Makefile @@ -6,7 +6,7 @@ lib-y = bitops.o bugs.o csum-partial.o delay.o fault.o mem.o memcpy.o \ ptrace.o ptrace_user.o semaphore.o sigcontext.o signal.o \ - syscalls.o sysrq.o thunk.o + syscalls.o sysrq.o thunk.o sys_call_table.o USER_OBJS := ptrace_user.o sigcontext.o diff --git a/arch/um/sys-x86_64/sys_call_table.c b/arch/um/sys-x86_64/sys_call_table.c new file mode 100644 index 0000000..3f5efbf --- /dev/null +++ b/arch/um/sys-x86_64/sys_call_table.c @@ -0,0 +1,276 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Copyright 2003 PathScale, Inc. + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/unistd.h" +#include "linux/sys.h" +#include "linux/swap.h" +#include "linux/syscalls.h" +#include "linux/sysctl.h" +#include "asm/signal.h" +#include "sysdep/syscalls.h" +#include "kern_util.h" + +#ifdef CONFIG_NFSD +#define NFSSERVCTL sys_nfsservctl +#else +#define NFSSERVCTL sys_ni_syscall +#endif + +#define LAST_GENERIC_SYSCALL __NR_keyctl + +#if LAST_GENERIC_SYSCALL > LAST_ARCH_SYSCALL +#define LAST_SYSCALL LAST_GENERIC_SYSCALL +#else +#define LAST_SYSCALL LAST_ARCH_SYSCALL +#endif + +extern syscall_handler_t sys_fork; +extern syscall_handler_t sys_execve; +extern syscall_handler_t um_time; +extern syscall_handler_t um_stime; +extern syscall_handler_t sys_pipe; +extern syscall_handler_t sys_olduname; +extern syscall_handler_t sys_sigaction; +extern syscall_handler_t sys_sigsuspend; +extern syscall_handler_t old_readdir; +extern syscall_handler_t sys_uname; +extern syscall_handler_t sys_ipc; +extern syscall_handler_t sys_sigreturn; +extern syscall_handler_t sys_clone; +extern syscall_handler_t sys_rt_sigreturn; +extern syscall_handler_t sys_sigaltstack; +extern syscall_handler_t sys_vfork; +extern syscall_handler_t old_select; +extern syscall_handler_t sys_modify_ldt; +extern syscall_handler_t sys_rt_sigsuspend; +extern syscall_handler_t sys_mbind; +extern syscall_handler_t sys_get_mempolicy; +extern syscall_handler_t sys_set_mempolicy; +extern syscall_handler_t sys_sys_setaltroot; + +syscall_handler_t *sys_call_table[] = { + [ __NR_restart_syscall ] = (syscall_handler_t *) sys_restart_syscall, + [ __NR_exit ] = (syscall_handler_t *) sys_exit, + [ __NR_fork ] = (syscall_handler_t *) sys_fork, + [ __NR_read ] = (syscall_handler_t *) sys_read, + [ __NR_write ] = (syscall_handler_t *) sys_write, + + /* These three are declared differently in asm/unistd.h */ + [ __NR_open ] = (syscall_handler_t *) sys_open, + [ __NR_close ] = (syscall_handler_t *) sys_close, + [ __NR_creat ] = (syscall_handler_t *) sys_creat, + [ __NR_link ] = (syscall_handler_t *) sys_link, + [ __NR_unlink ] = (syscall_handler_t *) sys_unlink, + [ __NR_execve ] = (syscall_handler_t *) sys_execve, + + /* declared differently in kern_util.h */ + [ __NR_chdir ] = (syscall_handler_t *) sys_chdir, + [ __NR_time ] = um_time, + [ __NR_mknod ] = (syscall_handler_t *) sys_mknod, + [ __NR_chmod ] = (syscall_handler_t *) sys_chmod, + [ __NR_lchown ] = (syscall_handler_t *) sys_lchown16, + [ __NR_lseek ] = (syscall_handler_t *) sys_lseek, + [ __NR_getpid ] = (syscall_handler_t *) sys_getpid, + [ __NR_mount ] = (syscall_handler_t *) sys_mount, + [ __NR_setuid ] = (syscall_handler_t *) sys_setuid16, + [ __NR_getuid ] = (syscall_handler_t *) sys_getuid16, + [ __NR_ptrace ] = (syscall_handler_t *) sys_ptrace, + [ __NR_alarm ] = (syscall_handler_t *) sys_alarm, + [ __NR_pause ] = (syscall_handler_t *) sys_pause, + [ __NR_utime ] = (syscall_handler_t *) sys_utime, + [ __NR_access ] = (syscall_handler_t *) sys_access, + [ __NR_sync ] = (syscall_handler_t *) sys_sync, + [ __NR_kill ] = (syscall_handler_t *) sys_kill, + [ __NR_rename ] = (syscall_handler_t *) sys_rename, + [ __NR_mkdir ] = (syscall_handler_t *) sys_mkdir, + [ __NR_rmdir ] = (syscall_handler_t *) sys_rmdir, + + /* Declared differently in asm/unistd.h */ + [ __NR_dup ] = (syscall_handler_t *) sys_dup, + [ __NR_pipe ] = (syscall_handler_t *) sys_pipe, + [ __NR_times ] = (syscall_handler_t *) sys_times, + [ __NR_brk ] = (syscall_handler_t *) sys_brk, + [ __NR_setgid ] = (syscall_handler_t *) sys_setgid16, + [ __NR_getgid ] = (syscall_handler_t *) sys_getgid16, + [ __NR_geteuid ] = (syscall_handler_t *) sys_geteuid16, + [ __NR_getegid ] = (syscall_handler_t *) sys_getegid16, + [ __NR_acct ] = (syscall_handler_t *) sys_acct, + [ __NR_umount2 ] = (syscall_handler_t *) sys_umount, + [ __NR_ioctl ] = (syscall_handler_t *) sys_ioctl, + [ __NR_fcntl ] = (syscall_handler_t *) sys_fcntl, + [ __NR_setpgid ] = (syscall_handler_t *) sys_setpgid, + [ __NR_umask ] = (syscall_handler_t *) sys_umask, + [ __NR_chroot ] = (syscall_handler_t *) sys_chroot, + [ __NR_ustat ] = (syscall_handler_t *) sys_ustat, + [ __NR_dup2 ] = (syscall_handler_t *) sys_dup2, + [ __NR_getppid ] = (syscall_handler_t *) sys_getppid, + [ __NR_getpgrp ] = (syscall_handler_t *) sys_getpgrp, + [ __NR_setsid ] = (syscall_handler_t *) sys_setsid, + [ __NR_setreuid ] = (syscall_handler_t *) sys_setreuid16, + [ __NR_setregid ] = (syscall_handler_t *) sys_setregid16, + [ __NR_sethostname ] = (syscall_handler_t *) sys_sethostname, + [ __NR_setrlimit ] = (syscall_handler_t *) sys_setrlimit, + [ __NR_getrlimit ] = (syscall_handler_t *) sys_old_getrlimit, + [ __NR_getrusage ] = (syscall_handler_t *) sys_getrusage, + [ __NR_gettimeofday ] = (syscall_handler_t *) sys_gettimeofday, + [ __NR_settimeofday ] = (syscall_handler_t *) sys_settimeofday, + [ __NR_getgroups ] = (syscall_handler_t *) sys_getgroups16, + [ __NR_setgroups ] = (syscall_handler_t *) sys_setgroups16, + [ __NR_symlink ] = (syscall_handler_t *) sys_symlink, + [ __NR_readlink ] = (syscall_handler_t *) sys_readlink, + [ __NR_uselib ] = (syscall_handler_t *) sys_uselib, + [ __NR_swapon ] = (syscall_handler_t *) sys_swapon, + [ __NR_reboot ] = (syscall_handler_t *) sys_reboot, + [ __NR_munmap ] = (syscall_handler_t *) sys_munmap, + [ __NR_truncate ] = (syscall_handler_t *) sys_truncate, + [ __NR_ftruncate ] = (syscall_handler_t *) sys_ftruncate, + [ __NR_fchmod ] = (syscall_handler_t *) sys_fchmod, + [ __NR_fchown ] = (syscall_handler_t *) sys_fchown16, + [ __NR_getpriority ] = (syscall_handler_t *) sys_getpriority, + [ __NR_setpriority ] = (syscall_handler_t *) sys_setpriority, + [ __NR_statfs ] = (syscall_handler_t *) sys_statfs, + [ __NR_fstatfs ] = (syscall_handler_t *) sys_fstatfs, + [ __NR_ioperm ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_syslog ] = (syscall_handler_t *) sys_syslog, + [ __NR_setitimer ] = (syscall_handler_t *) sys_setitimer, + [ __NR_getitimer ] = (syscall_handler_t *) sys_getitimer, + [ __NR_stat ] = (syscall_handler_t *) sys_newstat, + [ __NR_lstat ] = (syscall_handler_t *) sys_newlstat, + [ __NR_fstat ] = (syscall_handler_t *) sys_newfstat, + [ __NR_vhangup ] = (syscall_handler_t *) sys_vhangup, + [ __NR_wait4 ] = (syscall_handler_t *) sys_wait4, + [ __NR_swapoff ] = (syscall_handler_t *) sys_swapoff, + [ __NR_sysinfo ] = (syscall_handler_t *) sys_sysinfo, + [ __NR_fsync ] = (syscall_handler_t *) sys_fsync, + [ __NR_clone ] = (syscall_handler_t *) sys_clone, + [ __NR_setdomainname ] = (syscall_handler_t *) sys_setdomainname, + [ __NR_uname ] = (syscall_handler_t *) sys_newuname, + [ __NR_adjtimex ] = (syscall_handler_t *) sys_adjtimex, + [ __NR_mprotect ] = (syscall_handler_t *) sys_mprotect, + [ __NR_create_module ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_init_module ] = (syscall_handler_t *) sys_init_module, + [ __NR_delete_module ] = (syscall_handler_t *) sys_delete_module, + [ __NR_get_kernel_syms ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_quotactl ] = (syscall_handler_t *) sys_quotactl, + [ __NR_getpgid ] = (syscall_handler_t *) sys_getpgid, + [ __NR_fchdir ] = (syscall_handler_t *) sys_fchdir, + [ __NR_sysfs ] = (syscall_handler_t *) sys_sysfs, + [ __NR_personality ] = (syscall_handler_t *) sys_personality, + [ __NR_afs_syscall ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_setfsuid ] = (syscall_handler_t *) sys_setfsuid16, + [ __NR_setfsgid ] = (syscall_handler_t *) sys_setfsgid16, + [ __NR_getdents ] = (syscall_handler_t *) sys_getdents, + [ __NR_flock ] = (syscall_handler_t *) sys_flock, + [ __NR_msync ] = (syscall_handler_t *) sys_msync, + [ __NR_readv ] = (syscall_handler_t *) sys_readv, + [ __NR_writev ] = (syscall_handler_t *) sys_writev, + [ __NR_getsid ] = (syscall_handler_t *) sys_getsid, + [ __NR_fdatasync ] = (syscall_handler_t *) sys_fdatasync, + [ __NR__sysctl ] = (syscall_handler_t *) sys_sysctl, + [ __NR_mlock ] = (syscall_handler_t *) sys_mlock, + [ __NR_munlock ] = (syscall_handler_t *) sys_munlock, + [ __NR_mlockall ] = (syscall_handler_t *) sys_mlockall, + [ __NR_munlockall ] = (syscall_handler_t *) sys_munlockall, + [ __NR_sched_setparam ] = (syscall_handler_t *) sys_sched_setparam, + [ __NR_sched_getparam ] = (syscall_handler_t *) sys_sched_getparam, + [ __NR_sched_setscheduler ] = (syscall_handler_t *) sys_sched_setscheduler, + [ __NR_sched_getscheduler ] = (syscall_handler_t *) sys_sched_getscheduler, + [ __NR_sched_yield ] = (syscall_handler_t *) yield, + [ __NR_sched_get_priority_max ] = (syscall_handler_t *) sys_sched_get_priority_max, + [ __NR_sched_get_priority_min ] = (syscall_handler_t *) sys_sched_get_priority_min, + [ __NR_sched_rr_get_interval ] = (syscall_handler_t *) sys_sched_rr_get_interval, + [ __NR_nanosleep ] = (syscall_handler_t *) sys_nanosleep, + [ __NR_mremap ] = (syscall_handler_t *) sys_mremap, + [ __NR_setresuid ] = (syscall_handler_t *) sys_setresuid16, + [ __NR_getresuid ] = (syscall_handler_t *) sys_getresuid16, + [ __NR_query_module ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_poll ] = (syscall_handler_t *) sys_poll, + [ __NR_nfsservctl ] = (syscall_handler_t *) NFSSERVCTL, + [ __NR_setresgid ] = (syscall_handler_t *) sys_setresgid16, + [ __NR_getresgid ] = (syscall_handler_t *) sys_getresgid16, + [ __NR_prctl ] = (syscall_handler_t *) sys_prctl, + [ __NR_rt_sigreturn ] = (syscall_handler_t *) sys_rt_sigreturn, + [ __NR_rt_sigaction ] = (syscall_handler_t *) sys_rt_sigaction, + [ __NR_rt_sigprocmask ] = (syscall_handler_t *) sys_rt_sigprocmask, + [ __NR_rt_sigpending ] = (syscall_handler_t *) sys_rt_sigpending, + [ __NR_rt_sigtimedwait ] = (syscall_handler_t *) sys_rt_sigtimedwait, + [ __NR_rt_sigqueueinfo ] = (syscall_handler_t *) sys_rt_sigqueueinfo, + [ __NR_rt_sigsuspend ] = (syscall_handler_t *) sys_rt_sigsuspend, + [ __NR_pread64 ] = (syscall_handler_t *) sys_pread64, + [ __NR_pwrite64 ] = (syscall_handler_t *) sys_pwrite64, + [ __NR_chown ] = (syscall_handler_t *) sys_chown16, + [ __NR_getcwd ] = (syscall_handler_t *) sys_getcwd, + [ __NR_capget ] = (syscall_handler_t *) sys_capget, + [ __NR_capset ] = (syscall_handler_t *) sys_capset, + [ __NR_sigaltstack ] = (syscall_handler_t *) sys_sigaltstack, + [ __NR_sendfile ] = (syscall_handler_t *) sys_sendfile, + [ __NR_getpmsg ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_putpmsg ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_vfork ] = (syscall_handler_t *) sys_vfork, + [ __NR_getdents64 ] = (syscall_handler_t *) sys_getdents64, + [ __NR_gettid ] = (syscall_handler_t *) sys_gettid, + [ __NR_readahead ] = (syscall_handler_t *) sys_readahead, + [ __NR_setxattr ] = (syscall_handler_t *) sys_setxattr, + [ __NR_lsetxattr ] = (syscall_handler_t *) sys_lsetxattr, + [ __NR_fsetxattr ] = (syscall_handler_t *) sys_fsetxattr, + [ __NR_getxattr ] = (syscall_handler_t *) sys_getxattr, + [ __NR_lgetxattr ] = (syscall_handler_t *) sys_lgetxattr, + [ __NR_fgetxattr ] = (syscall_handler_t *) sys_fgetxattr, + [ __NR_listxattr ] = (syscall_handler_t *) sys_listxattr, + [ __NR_llistxattr ] = (syscall_handler_t *) sys_llistxattr, + [ __NR_flistxattr ] = (syscall_handler_t *) sys_flistxattr, + [ __NR_removexattr ] = (syscall_handler_t *) sys_removexattr, + [ __NR_lremovexattr ] = (syscall_handler_t *) sys_lremovexattr, + [ __NR_fremovexattr ] = (syscall_handler_t *) sys_fremovexattr, + [ __NR_tkill ] = (syscall_handler_t *) sys_tkill, + [ __NR_futex ] = (syscall_handler_t *) sys_futex, + [ __NR_sched_setaffinity ] = (syscall_handler_t *) sys_sched_setaffinity, + [ __NR_sched_getaffinity ] = (syscall_handler_t *) sys_sched_getaffinity, + [ __NR_io_setup ] = (syscall_handler_t *) sys_io_setup, + [ __NR_io_destroy ] = (syscall_handler_t *) sys_io_destroy, + [ __NR_io_getevents ] = (syscall_handler_t *) sys_io_getevents, + [ __NR_io_submit ] = (syscall_handler_t *) sys_io_submit, + [ __NR_io_cancel ] = (syscall_handler_t *) sys_io_cancel, + [ __NR_exit_group ] = (syscall_handler_t *) sys_exit_group, + [ __NR_lookup_dcookie ] = (syscall_handler_t *) sys_lookup_dcookie, + [ __NR_epoll_create ] = (syscall_handler_t *) sys_epoll_create, + [ __NR_epoll_ctl ] = (syscall_handler_t *) sys_epoll_ctl, + [ __NR_epoll_wait ] = (syscall_handler_t *) sys_epoll_wait, + [ __NR_remap_file_pages ] = (syscall_handler_t *) sys_remap_file_pages, + [ __NR_set_tid_address ] = (syscall_handler_t *) sys_set_tid_address, + [ __NR_timer_create ] = (syscall_handler_t *) sys_timer_create, + [ __NR_timer_settime ] = (syscall_handler_t *) sys_timer_settime, + [ __NR_timer_gettime ] = (syscall_handler_t *) sys_timer_gettime, + [ __NR_timer_getoverrun ] = (syscall_handler_t *) sys_timer_getoverrun, + [ __NR_timer_delete ] = (syscall_handler_t *) sys_timer_delete, + [ __NR_clock_settime ] = (syscall_handler_t *) sys_clock_settime, + [ __NR_clock_gettime ] = (syscall_handler_t *) sys_clock_gettime, + [ __NR_clock_getres ] = (syscall_handler_t *) sys_clock_getres, + [ __NR_clock_nanosleep ] = (syscall_handler_t *) sys_clock_nanosleep, + [ __NR_tgkill ] = (syscall_handler_t *) sys_tgkill, + [ __NR_utimes ] = (syscall_handler_t *) sys_utimes, + [ __NR_fadvise64 ] = (syscall_handler_t *) sys_fadvise64, + [ __NR_vserver ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_mbind ] = (syscall_handler_t *) sys_mbind, + [ __NR_get_mempolicy ] = (syscall_handler_t *) sys_get_mempolicy, + [ __NR_set_mempolicy ] = (syscall_handler_t *) sys_set_mempolicy, + [ __NR_mq_open ] = (syscall_handler_t *) sys_mq_open, + [ __NR_mq_unlink ] = (syscall_handler_t *) sys_mq_unlink, + [ __NR_mq_timedsend ] = (syscall_handler_t *) sys_mq_timedsend, + [ __NR_mq_timedreceive ] = (syscall_handler_t *) sys_mq_timedreceive, + [ __NR_mq_notify ] = (syscall_handler_t *) sys_mq_notify, + [ __NR_mq_getsetattr ] = (syscall_handler_t *) sys_mq_getsetattr, + [ __NR_kexec_load ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_waitid ] = (syscall_handler_t *) sys_waitid, + [ __NR_add_key ] = (syscall_handler_t *) sys_add_key, + [ __NR_request_key ] = (syscall_handler_t *) sys_request_key, + [ __NR_keyctl ] = (syscall_handler_t *) sys_keyctl, + + ARCH_SYSCALLS + [ LAST_SYSCALL + 1 ... NR_syscalls ] = + (syscall_handler_t *) sys_ni_syscall +}; -- cgit v0.10.2 From ddcd6b175793839e51fdd55debb670a9f512c448 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Sun, 1 May 2005 08:58:55 -0700 Subject: [PATCH] uml: quick fix syscall table for x86_64 Fix the moved syscall table for the x86_64 SUBARCH: - redirect __NR_chown and such to versions aware of 32-bit UIDs, - avoid the useless hack for sys_nfsservctl, - use sys_sendfile64 in the table rather than sys_sendfile. - __NR_uselib is sys_ni_syscall on x86_64 (which does not support A.OUT). - __NR_getrlimit is sys_getrlimit, not sys_old_getrlimit Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/sys-x86_64/sys_call_table.c b/arch/um/sys-x86_64/sys_call_table.c index 3f5efbf..a31e0df 100644 --- a/arch/um/sys-x86_64/sys_call_table.c +++ b/arch/um/sys-x86_64/sys_call_table.c @@ -14,12 +14,6 @@ #include "sysdep/syscalls.h" #include "kern_util.h" -#ifdef CONFIG_NFSD -#define NFSSERVCTL sys_nfsservctl -#else -#define NFSSERVCTL sys_ni_syscall -#endif - #define LAST_GENERIC_SYSCALL __NR_keyctl #if LAST_GENERIC_SYSCALL > LAST_ARCH_SYSCALL @@ -52,6 +46,7 @@ extern syscall_handler_t sys_get_mempolicy; extern syscall_handler_t sys_set_mempolicy; extern syscall_handler_t sys_sys_setaltroot; +/* On X86-64 all syscalls are aware of 32-bit [ug]ids, so I had to fix this!*/ syscall_handler_t *sys_call_table[] = { [ __NR_restart_syscall ] = (syscall_handler_t *) sys_restart_syscall, [ __NR_exit ] = (syscall_handler_t *) sys_exit, @@ -72,12 +67,12 @@ syscall_handler_t *sys_call_table[] = { [ __NR_time ] = um_time, [ __NR_mknod ] = (syscall_handler_t *) sys_mknod, [ __NR_chmod ] = (syscall_handler_t *) sys_chmod, - [ __NR_lchown ] = (syscall_handler_t *) sys_lchown16, + [ __NR_lchown ] = (syscall_handler_t *) sys_lchown, [ __NR_lseek ] = (syscall_handler_t *) sys_lseek, [ __NR_getpid ] = (syscall_handler_t *) sys_getpid, [ __NR_mount ] = (syscall_handler_t *) sys_mount, - [ __NR_setuid ] = (syscall_handler_t *) sys_setuid16, - [ __NR_getuid ] = (syscall_handler_t *) sys_getuid16, + [ __NR_setuid ] = (syscall_handler_t *) sys_setuid, + [ __NR_getuid ] = (syscall_handler_t *) sys_getuid, [ __NR_ptrace ] = (syscall_handler_t *) sys_ptrace, [ __NR_alarm ] = (syscall_handler_t *) sys_alarm, [ __NR_pause ] = (syscall_handler_t *) sys_pause, @@ -94,10 +89,10 @@ syscall_handler_t *sys_call_table[] = { [ __NR_pipe ] = (syscall_handler_t *) sys_pipe, [ __NR_times ] = (syscall_handler_t *) sys_times, [ __NR_brk ] = (syscall_handler_t *) sys_brk, - [ __NR_setgid ] = (syscall_handler_t *) sys_setgid16, - [ __NR_getgid ] = (syscall_handler_t *) sys_getgid16, - [ __NR_geteuid ] = (syscall_handler_t *) sys_geteuid16, - [ __NR_getegid ] = (syscall_handler_t *) sys_getegid16, + [ __NR_setgid ] = (syscall_handler_t *) sys_setgid, + [ __NR_getgid ] = (syscall_handler_t *) sys_getgid, + [ __NR_geteuid ] = (syscall_handler_t *) sys_geteuid, + [ __NR_getegid ] = (syscall_handler_t *) sys_getegid, [ __NR_acct ] = (syscall_handler_t *) sys_acct, [ __NR_umount2 ] = (syscall_handler_t *) sys_umount, [ __NR_ioctl ] = (syscall_handler_t *) sys_ioctl, @@ -110,26 +105,26 @@ syscall_handler_t *sys_call_table[] = { [ __NR_getppid ] = (syscall_handler_t *) sys_getppid, [ __NR_getpgrp ] = (syscall_handler_t *) sys_getpgrp, [ __NR_setsid ] = (syscall_handler_t *) sys_setsid, - [ __NR_setreuid ] = (syscall_handler_t *) sys_setreuid16, - [ __NR_setregid ] = (syscall_handler_t *) sys_setregid16, + [ __NR_setreuid ] = (syscall_handler_t *) sys_setreuid, + [ __NR_setregid ] = (syscall_handler_t *) sys_setregid, [ __NR_sethostname ] = (syscall_handler_t *) sys_sethostname, [ __NR_setrlimit ] = (syscall_handler_t *) sys_setrlimit, - [ __NR_getrlimit ] = (syscall_handler_t *) sys_old_getrlimit, + [ __NR_getrlimit ] = (syscall_handler_t *) sys_getrlimit, [ __NR_getrusage ] = (syscall_handler_t *) sys_getrusage, [ __NR_gettimeofday ] = (syscall_handler_t *) sys_gettimeofday, [ __NR_settimeofday ] = (syscall_handler_t *) sys_settimeofday, - [ __NR_getgroups ] = (syscall_handler_t *) sys_getgroups16, - [ __NR_setgroups ] = (syscall_handler_t *) sys_setgroups16, + [ __NR_getgroups ] = (syscall_handler_t *) sys_getgroups, + [ __NR_setgroups ] = (syscall_handler_t *) sys_setgroups, [ __NR_symlink ] = (syscall_handler_t *) sys_symlink, [ __NR_readlink ] = (syscall_handler_t *) sys_readlink, - [ __NR_uselib ] = (syscall_handler_t *) sys_uselib, + [ __NR_uselib ] = (syscall_handler_t *) sys_ni_syscall, [ __NR_swapon ] = (syscall_handler_t *) sys_swapon, [ __NR_reboot ] = (syscall_handler_t *) sys_reboot, [ __NR_munmap ] = (syscall_handler_t *) sys_munmap, [ __NR_truncate ] = (syscall_handler_t *) sys_truncate, [ __NR_ftruncate ] = (syscall_handler_t *) sys_ftruncate, [ __NR_fchmod ] = (syscall_handler_t *) sys_fchmod, - [ __NR_fchown ] = (syscall_handler_t *) sys_fchown16, + [ __NR_fchown ] = (syscall_handler_t *) sys_fchown, [ __NR_getpriority ] = (syscall_handler_t *) sys_getpriority, [ __NR_setpriority ] = (syscall_handler_t *) sys_setpriority, [ __NR_statfs ] = (syscall_handler_t *) sys_statfs, @@ -161,8 +156,8 @@ syscall_handler_t *sys_call_table[] = { [ __NR_sysfs ] = (syscall_handler_t *) sys_sysfs, [ __NR_personality ] = (syscall_handler_t *) sys_personality, [ __NR_afs_syscall ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_setfsuid ] = (syscall_handler_t *) sys_setfsuid16, - [ __NR_setfsgid ] = (syscall_handler_t *) sys_setfsgid16, + [ __NR_setfsuid ] = (syscall_handler_t *) sys_setfsuid, + [ __NR_setfsgid ] = (syscall_handler_t *) sys_setfsgid, [ __NR_getdents ] = (syscall_handler_t *) sys_getdents, [ __NR_flock ] = (syscall_handler_t *) sys_flock, [ __NR_msync ] = (syscall_handler_t *) sys_msync, @@ -185,13 +180,13 @@ syscall_handler_t *sys_call_table[] = { [ __NR_sched_rr_get_interval ] = (syscall_handler_t *) sys_sched_rr_get_interval, [ __NR_nanosleep ] = (syscall_handler_t *) sys_nanosleep, [ __NR_mremap ] = (syscall_handler_t *) sys_mremap, - [ __NR_setresuid ] = (syscall_handler_t *) sys_setresuid16, - [ __NR_getresuid ] = (syscall_handler_t *) sys_getresuid16, + [ __NR_setresuid ] = (syscall_handler_t *) sys_setresuid, + [ __NR_getresuid ] = (syscall_handler_t *) sys_getresuid, [ __NR_query_module ] = (syscall_handler_t *) sys_ni_syscall, [ __NR_poll ] = (syscall_handler_t *) sys_poll, - [ __NR_nfsservctl ] = (syscall_handler_t *) NFSSERVCTL, - [ __NR_setresgid ] = (syscall_handler_t *) sys_setresgid16, - [ __NR_getresgid ] = (syscall_handler_t *) sys_getresgid16, + [ __NR_nfsservctl ] = (syscall_handler_t *) sys_nfsservctl, + [ __NR_setresgid ] = (syscall_handler_t *) sys_setresgid, + [ __NR_getresgid ] = (syscall_handler_t *) sys_getresgid, [ __NR_prctl ] = (syscall_handler_t *) sys_prctl, [ __NR_rt_sigreturn ] = (syscall_handler_t *) sys_rt_sigreturn, [ __NR_rt_sigaction ] = (syscall_handler_t *) sys_rt_sigaction, @@ -202,12 +197,12 @@ syscall_handler_t *sys_call_table[] = { [ __NR_rt_sigsuspend ] = (syscall_handler_t *) sys_rt_sigsuspend, [ __NR_pread64 ] = (syscall_handler_t *) sys_pread64, [ __NR_pwrite64 ] = (syscall_handler_t *) sys_pwrite64, - [ __NR_chown ] = (syscall_handler_t *) sys_chown16, + [ __NR_chown ] = (syscall_handler_t *) sys_chown, [ __NR_getcwd ] = (syscall_handler_t *) sys_getcwd, [ __NR_capget ] = (syscall_handler_t *) sys_capget, [ __NR_capset ] = (syscall_handler_t *) sys_capset, [ __NR_sigaltstack ] = (syscall_handler_t *) sys_sigaltstack, - [ __NR_sendfile ] = (syscall_handler_t *) sys_sendfile, + [ __NR_sendfile ] = (syscall_handler_t *) sys_sendfile64, [ __NR_getpmsg ] = (syscall_handler_t *) sys_ni_syscall, [ __NR_putpmsg ] = (syscall_handler_t *) sys_ni_syscall, [ __NR_vfork ] = (syscall_handler_t *) sys_vfork, -- cgit v0.10.2 From 80f9507886076de0cadfdf2b87701e68156829c2 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Sun, 1 May 2005 08:58:55 -0700 Subject: [PATCH] uml: fix syscall table by including $(SUBARCH)'s one, for x86-64 Reuse asm-x86-64/unistd.h to build our syscall table, like x86-64 already does. Like for i386, we must add some #defines for all the (right!) changes UML does to x86-64 syscall table. Note: I noted a bogus: [ __NR_sched_yield ] = (syscall_handler_t *) yield, while doing this patch (which could only be a workaround for some strange bug, but I would ignore this possibility). I'm changing this without notice. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/include/sysdep-x86_64/syscalls.h b/arch/um/include/sysdep-x86_64/syscalls.h index 65a01e3..b56b335 100644 --- a/arch/um/include/sysdep-x86_64/syscalls.h +++ b/arch/um/include/sysdep-x86_64/syscalls.h @@ -30,51 +30,6 @@ extern syscall_handler_t wrap_sys_shmat; extern syscall_handler_t sys_modify_ldt; extern syscall_handler_t sys_arch_prctl; -#define ARCH_SYSCALLS \ - [ __NR_mmap ] = (syscall_handler_t *) old_mmap, \ - [ __NR_select ] = (syscall_handler_t *) sys_select, \ - [ __NR_mincore ] = (syscall_handler_t *) sys_mincore, \ - [ __NR_madvise ] = (syscall_handler_t *) sys_madvise, \ - [ __NR_shmget ] = (syscall_handler_t *) sys_shmget, \ - [ __NR_shmat ] = (syscall_handler_t *) wrap_sys_shmat, \ - [ __NR_shmctl ] = (syscall_handler_t *) sys_shmctl, \ - [ __NR_semop ] = (syscall_handler_t *) sys_semop, \ - [ __NR_semget ] = (syscall_handler_t *) sys_semget, \ - [ __NR_semctl ] = (syscall_handler_t *) sys_semctl, \ - [ __NR_shmdt ] = (syscall_handler_t *) sys_shmdt, \ - [ __NR_msgget ] = (syscall_handler_t *) sys_msgget, \ - [ __NR_msgsnd ] = (syscall_handler_t *) sys_msgsnd, \ - [ __NR_msgrcv ] = (syscall_handler_t *) sys_msgrcv, \ - [ __NR_msgctl ] = (syscall_handler_t *) sys_msgctl, \ - [ __NR_pivot_root ] = (syscall_handler_t *) sys_pivot_root, \ - [ __NR_tuxcall ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_security ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_epoll_ctl_old ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_epoll_wait_old ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_modify_ldt ] = (syscall_handler_t *) sys_modify_ldt, \ - [ __NR_arch_prctl ] = (syscall_handler_t *) sys_arch_prctl, \ - [ __NR_socket ] = (syscall_handler_t *) sys_socket, \ - [ __NR_connect ] = (syscall_handler_t *) sys_connect, \ - [ __NR_accept ] = (syscall_handler_t *) sys_accept, \ - [ __NR_recvfrom ] = (syscall_handler_t *) sys_recvfrom, \ - [ __NR_recvmsg ] = (syscall_handler_t *) sys_recvmsg, \ - [ __NR_sendmsg ] = (syscall_handler_t *) sys_sendmsg, \ - [ __NR_bind ] = (syscall_handler_t *) sys_bind, \ - [ __NR_listen ] = (syscall_handler_t *) sys_listen, \ - [ __NR_getsockname ] = (syscall_handler_t *) sys_getsockname, \ - [ __NR_getpeername ] = (syscall_handler_t *) sys_getpeername, \ - [ __NR_socketpair ] = (syscall_handler_t *) sys_socketpair, \ - [ __NR_sendto ] = (syscall_handler_t *) sys_sendto, \ - [ __NR_shutdown ] = (syscall_handler_t *) sys_shutdown, \ - [ __NR_setsockopt ] = (syscall_handler_t *) sys_setsockopt, \ - [ __NR_getsockopt ] = (syscall_handler_t *) sys_getsockopt, \ - [ __NR_iopl ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_set_thread_area ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_get_thread_area ] = (syscall_handler_t *) sys_ni_syscall, \ - [ __NR_semtimedop ] = (syscall_handler_t *) sys_semtimedop, \ - [ 251 ] = (syscall_handler_t *) sys_ni_syscall, - -#define LAST_ARCH_SYSCALL 251 -#define NR_syscalls 1024 +#define NR_syscalls (__NR_syscall_max + 1) #endif diff --git a/arch/um/sys-x86_64/Makefile b/arch/um/sys-x86_64/Makefile index 274158e..d7ed2f7 100644 --- a/arch/um/sys-x86_64/Makefile +++ b/arch/um/sys-x86_64/Makefile @@ -6,7 +6,7 @@ lib-y = bitops.o bugs.o csum-partial.o delay.o fault.o mem.o memcpy.o \ ptrace.o ptrace_user.o semaphore.o sigcontext.o signal.o \ - syscalls.o sysrq.o thunk.o sys_call_table.o + syscalls.o sysrq.o thunk.o syscall_table.o USER_OBJS := ptrace_user.o sigcontext.o diff --git a/arch/um/sys-x86_64/sys_call_table.c b/arch/um/sys-x86_64/sys_call_table.c deleted file mode 100644 index a31e0df..0000000 --- a/arch/um/sys-x86_64/sys_call_table.c +++ /dev/null @@ -1,271 +0,0 @@ -/* - * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) - * Copyright 2003 PathScale, Inc. - * Licensed under the GPL - */ - -#include "linux/config.h" -#include "linux/unistd.h" -#include "linux/sys.h" -#include "linux/swap.h" -#include "linux/syscalls.h" -#include "linux/sysctl.h" -#include "asm/signal.h" -#include "sysdep/syscalls.h" -#include "kern_util.h" - -#define LAST_GENERIC_SYSCALL __NR_keyctl - -#if LAST_GENERIC_SYSCALL > LAST_ARCH_SYSCALL -#define LAST_SYSCALL LAST_GENERIC_SYSCALL -#else -#define LAST_SYSCALL LAST_ARCH_SYSCALL -#endif - -extern syscall_handler_t sys_fork; -extern syscall_handler_t sys_execve; -extern syscall_handler_t um_time; -extern syscall_handler_t um_stime; -extern syscall_handler_t sys_pipe; -extern syscall_handler_t sys_olduname; -extern syscall_handler_t sys_sigaction; -extern syscall_handler_t sys_sigsuspend; -extern syscall_handler_t old_readdir; -extern syscall_handler_t sys_uname; -extern syscall_handler_t sys_ipc; -extern syscall_handler_t sys_sigreturn; -extern syscall_handler_t sys_clone; -extern syscall_handler_t sys_rt_sigreturn; -extern syscall_handler_t sys_sigaltstack; -extern syscall_handler_t sys_vfork; -extern syscall_handler_t old_select; -extern syscall_handler_t sys_modify_ldt; -extern syscall_handler_t sys_rt_sigsuspend; -extern syscall_handler_t sys_mbind; -extern syscall_handler_t sys_get_mempolicy; -extern syscall_handler_t sys_set_mempolicy; -extern syscall_handler_t sys_sys_setaltroot; - -/* On X86-64 all syscalls are aware of 32-bit [ug]ids, so I had to fix this!*/ -syscall_handler_t *sys_call_table[] = { - [ __NR_restart_syscall ] = (syscall_handler_t *) sys_restart_syscall, - [ __NR_exit ] = (syscall_handler_t *) sys_exit, - [ __NR_fork ] = (syscall_handler_t *) sys_fork, - [ __NR_read ] = (syscall_handler_t *) sys_read, - [ __NR_write ] = (syscall_handler_t *) sys_write, - - /* These three are declared differently in asm/unistd.h */ - [ __NR_open ] = (syscall_handler_t *) sys_open, - [ __NR_close ] = (syscall_handler_t *) sys_close, - [ __NR_creat ] = (syscall_handler_t *) sys_creat, - [ __NR_link ] = (syscall_handler_t *) sys_link, - [ __NR_unlink ] = (syscall_handler_t *) sys_unlink, - [ __NR_execve ] = (syscall_handler_t *) sys_execve, - - /* declared differently in kern_util.h */ - [ __NR_chdir ] = (syscall_handler_t *) sys_chdir, - [ __NR_time ] = um_time, - [ __NR_mknod ] = (syscall_handler_t *) sys_mknod, - [ __NR_chmod ] = (syscall_handler_t *) sys_chmod, - [ __NR_lchown ] = (syscall_handler_t *) sys_lchown, - [ __NR_lseek ] = (syscall_handler_t *) sys_lseek, - [ __NR_getpid ] = (syscall_handler_t *) sys_getpid, - [ __NR_mount ] = (syscall_handler_t *) sys_mount, - [ __NR_setuid ] = (syscall_handler_t *) sys_setuid, - [ __NR_getuid ] = (syscall_handler_t *) sys_getuid, - [ __NR_ptrace ] = (syscall_handler_t *) sys_ptrace, - [ __NR_alarm ] = (syscall_handler_t *) sys_alarm, - [ __NR_pause ] = (syscall_handler_t *) sys_pause, - [ __NR_utime ] = (syscall_handler_t *) sys_utime, - [ __NR_access ] = (syscall_handler_t *) sys_access, - [ __NR_sync ] = (syscall_handler_t *) sys_sync, - [ __NR_kill ] = (syscall_handler_t *) sys_kill, - [ __NR_rename ] = (syscall_handler_t *) sys_rename, - [ __NR_mkdir ] = (syscall_handler_t *) sys_mkdir, - [ __NR_rmdir ] = (syscall_handler_t *) sys_rmdir, - - /* Declared differently in asm/unistd.h */ - [ __NR_dup ] = (syscall_handler_t *) sys_dup, - [ __NR_pipe ] = (syscall_handler_t *) sys_pipe, - [ __NR_times ] = (syscall_handler_t *) sys_times, - [ __NR_brk ] = (syscall_handler_t *) sys_brk, - [ __NR_setgid ] = (syscall_handler_t *) sys_setgid, - [ __NR_getgid ] = (syscall_handler_t *) sys_getgid, - [ __NR_geteuid ] = (syscall_handler_t *) sys_geteuid, - [ __NR_getegid ] = (syscall_handler_t *) sys_getegid, - [ __NR_acct ] = (syscall_handler_t *) sys_acct, - [ __NR_umount2 ] = (syscall_handler_t *) sys_umount, - [ __NR_ioctl ] = (syscall_handler_t *) sys_ioctl, - [ __NR_fcntl ] = (syscall_handler_t *) sys_fcntl, - [ __NR_setpgid ] = (syscall_handler_t *) sys_setpgid, - [ __NR_umask ] = (syscall_handler_t *) sys_umask, - [ __NR_chroot ] = (syscall_handler_t *) sys_chroot, - [ __NR_ustat ] = (syscall_handler_t *) sys_ustat, - [ __NR_dup2 ] = (syscall_handler_t *) sys_dup2, - [ __NR_getppid ] = (syscall_handler_t *) sys_getppid, - [ __NR_getpgrp ] = (syscall_handler_t *) sys_getpgrp, - [ __NR_setsid ] = (syscall_handler_t *) sys_setsid, - [ __NR_setreuid ] = (syscall_handler_t *) sys_setreuid, - [ __NR_setregid ] = (syscall_handler_t *) sys_setregid, - [ __NR_sethostname ] = (syscall_handler_t *) sys_sethostname, - [ __NR_setrlimit ] = (syscall_handler_t *) sys_setrlimit, - [ __NR_getrlimit ] = (syscall_handler_t *) sys_getrlimit, - [ __NR_getrusage ] = (syscall_handler_t *) sys_getrusage, - [ __NR_gettimeofday ] = (syscall_handler_t *) sys_gettimeofday, - [ __NR_settimeofday ] = (syscall_handler_t *) sys_settimeofday, - [ __NR_getgroups ] = (syscall_handler_t *) sys_getgroups, - [ __NR_setgroups ] = (syscall_handler_t *) sys_setgroups, - [ __NR_symlink ] = (syscall_handler_t *) sys_symlink, - [ __NR_readlink ] = (syscall_handler_t *) sys_readlink, - [ __NR_uselib ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_swapon ] = (syscall_handler_t *) sys_swapon, - [ __NR_reboot ] = (syscall_handler_t *) sys_reboot, - [ __NR_munmap ] = (syscall_handler_t *) sys_munmap, - [ __NR_truncate ] = (syscall_handler_t *) sys_truncate, - [ __NR_ftruncate ] = (syscall_handler_t *) sys_ftruncate, - [ __NR_fchmod ] = (syscall_handler_t *) sys_fchmod, - [ __NR_fchown ] = (syscall_handler_t *) sys_fchown, - [ __NR_getpriority ] = (syscall_handler_t *) sys_getpriority, - [ __NR_setpriority ] = (syscall_handler_t *) sys_setpriority, - [ __NR_statfs ] = (syscall_handler_t *) sys_statfs, - [ __NR_fstatfs ] = (syscall_handler_t *) sys_fstatfs, - [ __NR_ioperm ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_syslog ] = (syscall_handler_t *) sys_syslog, - [ __NR_setitimer ] = (syscall_handler_t *) sys_setitimer, - [ __NR_getitimer ] = (syscall_handler_t *) sys_getitimer, - [ __NR_stat ] = (syscall_handler_t *) sys_newstat, - [ __NR_lstat ] = (syscall_handler_t *) sys_newlstat, - [ __NR_fstat ] = (syscall_handler_t *) sys_newfstat, - [ __NR_vhangup ] = (syscall_handler_t *) sys_vhangup, - [ __NR_wait4 ] = (syscall_handler_t *) sys_wait4, - [ __NR_swapoff ] = (syscall_handler_t *) sys_swapoff, - [ __NR_sysinfo ] = (syscall_handler_t *) sys_sysinfo, - [ __NR_fsync ] = (syscall_handler_t *) sys_fsync, - [ __NR_clone ] = (syscall_handler_t *) sys_clone, - [ __NR_setdomainname ] = (syscall_handler_t *) sys_setdomainname, - [ __NR_uname ] = (syscall_handler_t *) sys_newuname, - [ __NR_adjtimex ] = (syscall_handler_t *) sys_adjtimex, - [ __NR_mprotect ] = (syscall_handler_t *) sys_mprotect, - [ __NR_create_module ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_init_module ] = (syscall_handler_t *) sys_init_module, - [ __NR_delete_module ] = (syscall_handler_t *) sys_delete_module, - [ __NR_get_kernel_syms ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_quotactl ] = (syscall_handler_t *) sys_quotactl, - [ __NR_getpgid ] = (syscall_handler_t *) sys_getpgid, - [ __NR_fchdir ] = (syscall_handler_t *) sys_fchdir, - [ __NR_sysfs ] = (syscall_handler_t *) sys_sysfs, - [ __NR_personality ] = (syscall_handler_t *) sys_personality, - [ __NR_afs_syscall ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_setfsuid ] = (syscall_handler_t *) sys_setfsuid, - [ __NR_setfsgid ] = (syscall_handler_t *) sys_setfsgid, - [ __NR_getdents ] = (syscall_handler_t *) sys_getdents, - [ __NR_flock ] = (syscall_handler_t *) sys_flock, - [ __NR_msync ] = (syscall_handler_t *) sys_msync, - [ __NR_readv ] = (syscall_handler_t *) sys_readv, - [ __NR_writev ] = (syscall_handler_t *) sys_writev, - [ __NR_getsid ] = (syscall_handler_t *) sys_getsid, - [ __NR_fdatasync ] = (syscall_handler_t *) sys_fdatasync, - [ __NR__sysctl ] = (syscall_handler_t *) sys_sysctl, - [ __NR_mlock ] = (syscall_handler_t *) sys_mlock, - [ __NR_munlock ] = (syscall_handler_t *) sys_munlock, - [ __NR_mlockall ] = (syscall_handler_t *) sys_mlockall, - [ __NR_munlockall ] = (syscall_handler_t *) sys_munlockall, - [ __NR_sched_setparam ] = (syscall_handler_t *) sys_sched_setparam, - [ __NR_sched_getparam ] = (syscall_handler_t *) sys_sched_getparam, - [ __NR_sched_setscheduler ] = (syscall_handler_t *) sys_sched_setscheduler, - [ __NR_sched_getscheduler ] = (syscall_handler_t *) sys_sched_getscheduler, - [ __NR_sched_yield ] = (syscall_handler_t *) yield, - [ __NR_sched_get_priority_max ] = (syscall_handler_t *) sys_sched_get_priority_max, - [ __NR_sched_get_priority_min ] = (syscall_handler_t *) sys_sched_get_priority_min, - [ __NR_sched_rr_get_interval ] = (syscall_handler_t *) sys_sched_rr_get_interval, - [ __NR_nanosleep ] = (syscall_handler_t *) sys_nanosleep, - [ __NR_mremap ] = (syscall_handler_t *) sys_mremap, - [ __NR_setresuid ] = (syscall_handler_t *) sys_setresuid, - [ __NR_getresuid ] = (syscall_handler_t *) sys_getresuid, - [ __NR_query_module ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_poll ] = (syscall_handler_t *) sys_poll, - [ __NR_nfsservctl ] = (syscall_handler_t *) sys_nfsservctl, - [ __NR_setresgid ] = (syscall_handler_t *) sys_setresgid, - [ __NR_getresgid ] = (syscall_handler_t *) sys_getresgid, - [ __NR_prctl ] = (syscall_handler_t *) sys_prctl, - [ __NR_rt_sigreturn ] = (syscall_handler_t *) sys_rt_sigreturn, - [ __NR_rt_sigaction ] = (syscall_handler_t *) sys_rt_sigaction, - [ __NR_rt_sigprocmask ] = (syscall_handler_t *) sys_rt_sigprocmask, - [ __NR_rt_sigpending ] = (syscall_handler_t *) sys_rt_sigpending, - [ __NR_rt_sigtimedwait ] = (syscall_handler_t *) sys_rt_sigtimedwait, - [ __NR_rt_sigqueueinfo ] = (syscall_handler_t *) sys_rt_sigqueueinfo, - [ __NR_rt_sigsuspend ] = (syscall_handler_t *) sys_rt_sigsuspend, - [ __NR_pread64 ] = (syscall_handler_t *) sys_pread64, - [ __NR_pwrite64 ] = (syscall_handler_t *) sys_pwrite64, - [ __NR_chown ] = (syscall_handler_t *) sys_chown, - [ __NR_getcwd ] = (syscall_handler_t *) sys_getcwd, - [ __NR_capget ] = (syscall_handler_t *) sys_capget, - [ __NR_capset ] = (syscall_handler_t *) sys_capset, - [ __NR_sigaltstack ] = (syscall_handler_t *) sys_sigaltstack, - [ __NR_sendfile ] = (syscall_handler_t *) sys_sendfile64, - [ __NR_getpmsg ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_putpmsg ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_vfork ] = (syscall_handler_t *) sys_vfork, - [ __NR_getdents64 ] = (syscall_handler_t *) sys_getdents64, - [ __NR_gettid ] = (syscall_handler_t *) sys_gettid, - [ __NR_readahead ] = (syscall_handler_t *) sys_readahead, - [ __NR_setxattr ] = (syscall_handler_t *) sys_setxattr, - [ __NR_lsetxattr ] = (syscall_handler_t *) sys_lsetxattr, - [ __NR_fsetxattr ] = (syscall_handler_t *) sys_fsetxattr, - [ __NR_getxattr ] = (syscall_handler_t *) sys_getxattr, - [ __NR_lgetxattr ] = (syscall_handler_t *) sys_lgetxattr, - [ __NR_fgetxattr ] = (syscall_handler_t *) sys_fgetxattr, - [ __NR_listxattr ] = (syscall_handler_t *) sys_listxattr, - [ __NR_llistxattr ] = (syscall_handler_t *) sys_llistxattr, - [ __NR_flistxattr ] = (syscall_handler_t *) sys_flistxattr, - [ __NR_removexattr ] = (syscall_handler_t *) sys_removexattr, - [ __NR_lremovexattr ] = (syscall_handler_t *) sys_lremovexattr, - [ __NR_fremovexattr ] = (syscall_handler_t *) sys_fremovexattr, - [ __NR_tkill ] = (syscall_handler_t *) sys_tkill, - [ __NR_futex ] = (syscall_handler_t *) sys_futex, - [ __NR_sched_setaffinity ] = (syscall_handler_t *) sys_sched_setaffinity, - [ __NR_sched_getaffinity ] = (syscall_handler_t *) sys_sched_getaffinity, - [ __NR_io_setup ] = (syscall_handler_t *) sys_io_setup, - [ __NR_io_destroy ] = (syscall_handler_t *) sys_io_destroy, - [ __NR_io_getevents ] = (syscall_handler_t *) sys_io_getevents, - [ __NR_io_submit ] = (syscall_handler_t *) sys_io_submit, - [ __NR_io_cancel ] = (syscall_handler_t *) sys_io_cancel, - [ __NR_exit_group ] = (syscall_handler_t *) sys_exit_group, - [ __NR_lookup_dcookie ] = (syscall_handler_t *) sys_lookup_dcookie, - [ __NR_epoll_create ] = (syscall_handler_t *) sys_epoll_create, - [ __NR_epoll_ctl ] = (syscall_handler_t *) sys_epoll_ctl, - [ __NR_epoll_wait ] = (syscall_handler_t *) sys_epoll_wait, - [ __NR_remap_file_pages ] = (syscall_handler_t *) sys_remap_file_pages, - [ __NR_set_tid_address ] = (syscall_handler_t *) sys_set_tid_address, - [ __NR_timer_create ] = (syscall_handler_t *) sys_timer_create, - [ __NR_timer_settime ] = (syscall_handler_t *) sys_timer_settime, - [ __NR_timer_gettime ] = (syscall_handler_t *) sys_timer_gettime, - [ __NR_timer_getoverrun ] = (syscall_handler_t *) sys_timer_getoverrun, - [ __NR_timer_delete ] = (syscall_handler_t *) sys_timer_delete, - [ __NR_clock_settime ] = (syscall_handler_t *) sys_clock_settime, - [ __NR_clock_gettime ] = (syscall_handler_t *) sys_clock_gettime, - [ __NR_clock_getres ] = (syscall_handler_t *) sys_clock_getres, - [ __NR_clock_nanosleep ] = (syscall_handler_t *) sys_clock_nanosleep, - [ __NR_tgkill ] = (syscall_handler_t *) sys_tgkill, - [ __NR_utimes ] = (syscall_handler_t *) sys_utimes, - [ __NR_fadvise64 ] = (syscall_handler_t *) sys_fadvise64, - [ __NR_vserver ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_mbind ] = (syscall_handler_t *) sys_mbind, - [ __NR_get_mempolicy ] = (syscall_handler_t *) sys_get_mempolicy, - [ __NR_set_mempolicy ] = (syscall_handler_t *) sys_set_mempolicy, - [ __NR_mq_open ] = (syscall_handler_t *) sys_mq_open, - [ __NR_mq_unlink ] = (syscall_handler_t *) sys_mq_unlink, - [ __NR_mq_timedsend ] = (syscall_handler_t *) sys_mq_timedsend, - [ __NR_mq_timedreceive ] = (syscall_handler_t *) sys_mq_timedreceive, - [ __NR_mq_notify ] = (syscall_handler_t *) sys_mq_notify, - [ __NR_mq_getsetattr ] = (syscall_handler_t *) sys_mq_getsetattr, - [ __NR_kexec_load ] = (syscall_handler_t *) sys_ni_syscall, - [ __NR_waitid ] = (syscall_handler_t *) sys_waitid, - [ __NR_add_key ] = (syscall_handler_t *) sys_add_key, - [ __NR_request_key ] = (syscall_handler_t *) sys_request_key, - [ __NR_keyctl ] = (syscall_handler_t *) sys_keyctl, - - ARCH_SYSCALLS - [ LAST_SYSCALL + 1 ... NR_syscalls ] = - (syscall_handler_t *) sys_ni_syscall -}; diff --git a/arch/um/sys-x86_64/syscall_table.c b/arch/um/sys-x86_64/syscall_table.c new file mode 100644 index 0000000..34b2e84 --- /dev/null +++ b/arch/um/sys-x86_64/syscall_table.c @@ -0,0 +1,59 @@ +/* System call table for UML/x86-64, copied from arch/x86_64/kernel/syscall.c + * with some changes for UML. */ + +#include +#include +#include +#include + +#define __NO_STUBS + +/* Below you can see, in terms of #define's, the differences between the x86-64 + * and the UML syscall table. */ + +/* Not going to be implemented by UML, since we have no hardware. */ +#define stub_iopl sys_ni_syscall +#define sys_ioperm sys_ni_syscall + +/* The UML TLS problem. Note that x86_64 does not implement this, so the below + * is needed only for the ia32 compatibility. */ +/*#define sys_set_thread_area sys_ni_syscall +#define sys_get_thread_area sys_ni_syscall*/ + +/* For __NR_time. The x86-64 name hopefully will change from sys_time64 to + * sys_time (since the current situation is bogus). I've sent a patch to cleanup + * this. Remove below the obsoleted line. */ +#define sys_time64 um_time +#define sys_time um_time + +/* On UML we call it this way ("old" means it's not mmap2) */ +#define sys_mmap old_mmap +/* On x86-64 sys_uname is actually sys_newuname plus a compatibility trick. + * See arch/x86_64/kernel/sys_x86_64.c */ +#define sys_uname sys_uname64 + +#define stub_clone sys_clone +#define stub_fork sys_fork +#define stub_vfork sys_vfork +#define stub_execve sys_execve +#define stub_rt_sigsuspend sys_rt_sigsuspend +#define stub_sigaltstack sys_sigaltstack +#define stub_rt_sigreturn sys_rt_sigreturn + +#define __SYSCALL(nr, sym) extern asmlinkage void sym(void) ; +#undef _ASM_X86_64_UNISTD_H_ +#include + +#undef __SYSCALL +#define __SYSCALL(nr, sym) [ nr ] = sym, +#undef _ASM_X86_64_UNISTD_H_ + +typedef void (*sys_call_ptr_t)(void); + +extern void sys_ni_syscall(void); + +sys_call_ptr_t sys_call_table[__NR_syscall_max+1] __cacheline_aligned = { + /* Smells like a like a compiler bug -- it doesn't work when the & below is removed. */ + [0 ... __NR_syscall_max] = &sys_ni_syscall, +#include +}; diff --git a/arch/um/sys-x86_64/syscalls.c b/arch/um/sys-x86_64/syscalls.c index 68205a0..15768c9 100644 --- a/arch/um/sys-x86_64/syscalls.c +++ b/arch/um/sys-x86_64/syscalls.c @@ -14,6 +14,7 @@ #include "asm/prctl.h" /* XXX This should get the constants from libc */ #include "choose-mode.h" +/* XXX: copied from x86-64: arch/x86_64/kernel/sys_x86_64.c */ asmlinkage long wrap_sys_shmat(int shmid, char __user *shmaddr, int shmflg) { unsigned long raddr; @@ -21,6 +22,17 @@ asmlinkage long wrap_sys_shmat(int shmid, char __user *shmaddr, int shmflg) return do_shmat(shmid, shmaddr, shmflg, &raddr) ?: (long) raddr; } +asmlinkage long sys_uname64(struct new_utsname __user * name) +{ + int err; + down_read(&uts_sem); + err = copy_to_user(name, &system_utsname, sizeof (*name)); + up_read(&uts_sem); + if (personality(current->personality) == PER_LINUX32) + err |= copy_to_user(&name->machine, "i686", 5); + return err ? -EFAULT : 0; +} + #ifdef CONFIG_MODE_TT extern int modify_ldt(int func, void *ptr, unsigned long bytecount); -- cgit v0.10.2 From b97b77cca627b2e3e6d0015e2e14b1d1c12de0c8 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Sun, 1 May 2005 08:58:56 -0700 Subject: [PATCH] uml: redo console locking Fix some console locking problems (including scheduling in atomic) and various reorderings and cleanup in that code. Not yet ready for 2.6.12 probably. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/drivers/chan_kern.c b/arch/um/drivers/chan_kern.c index 1f77deb..0150038 100644 --- a/arch/um/drivers/chan_kern.c +++ b/arch/um/drivers/chan_kern.c @@ -22,7 +22,7 @@ #ifdef CONFIG_NOCONFIG_CHAN static void *not_configged_init(char *str, int device, struct chan_opts *opts) { - printk(KERN_ERR "Using a channel type which is configured out of " + printf(KERN_ERR "Using a channel type which is configured out of " "UML\n"); return(NULL); } @@ -30,27 +30,27 @@ static void *not_configged_init(char *str, int device, struct chan_opts *opts) static int not_configged_open(int input, int output, int primary, void *data, char **dev_out) { - printk(KERN_ERR "Using a channel type which is configured out of " + printf(KERN_ERR "Using a channel type which is configured out of " "UML\n"); return(-ENODEV); } static void not_configged_close(int fd, void *data) { - printk(KERN_ERR "Using a channel type which is configured out of " + printf(KERN_ERR "Using a channel type which is configured out of " "UML\n"); } static int not_configged_read(int fd, char *c_out, void *data) { - printk(KERN_ERR "Using a channel type which is configured out of " + printf(KERN_ERR "Using a channel type which is configured out of " "UML\n"); return(-EIO); } static int not_configged_write(int fd, const char *buf, int len, void *data) { - printk(KERN_ERR "Using a channel type which is configured out of " + printf(KERN_ERR "Using a channel type which is configured out of " "UML\n"); return(-EIO); } @@ -58,7 +58,7 @@ static int not_configged_write(int fd, const char *buf, int len, void *data) static int not_configged_console_write(int fd, const char *buf, int len, void *data) { - printk(KERN_ERR "Using a channel type which is configured out of " + printf(KERN_ERR "Using a channel type which is configured out of " "UML\n"); return(-EIO); } @@ -66,14 +66,14 @@ static int not_configged_console_write(int fd, const char *buf, int len, static int not_configged_window_size(int fd, void *data, unsigned short *rows, unsigned short *cols) { - printk(KERN_ERR "Using a channel type which is configured out of " + printf(KERN_ERR "Using a channel type which is configured out of " "UML\n"); return(-ENODEV); } static void not_configged_free(void *data) { - printk(KERN_ERR "Using a channel type which is configured out of " + printf(KERN_ERR "Using a channel type which is configured out of " "UML\n"); } diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c index 6924f27..d0f9712 100644 --- a/arch/um/drivers/line.c +++ b/arch/um/drivers/line.c @@ -39,19 +39,69 @@ static void line_timer_cb(void *arg) line_interrupt(line->driver->read_irq, arg, NULL); } -static int write_room(struct line *dev) +/* Returns the free space inside the ring buffer of this line. + * + * Should be called while holding line->lock (this does not modify datas). + */ +static int write_room(struct line *line) { int n; - if (dev->buffer == NULL) - return (LINE_BUFSIZE - 1); + if (line->buffer == NULL) + return LINE_BUFSIZE - 1; + + /* This is for the case where the buffer is wrapped! */ + n = line->head - line->tail; - n = dev->head - dev->tail; if (n <= 0) - n = LINE_BUFSIZE + n; - return (n - 1); + n = LINE_BUFSIZE + n; /* The other case */ + return n - 1; +} + +int line_write_room(struct tty_struct *tty) +{ + struct line *line = tty->driver_data; + unsigned long flags; + int room; + + if (tty->stopped) + return 0; + + spin_lock_irqsave(&line->lock, flags); + room = write_room(line); + spin_unlock_irqrestore(&line->lock, flags); + + /*XXX: Warning to remove */ + if (0 == room) + printk(KERN_DEBUG "%s: %s: no room left in buffer\n", + __FUNCTION__,tty->name); + return room; } +int line_chars_in_buffer(struct tty_struct *tty) +{ + struct line *line = tty->driver_data; + unsigned long flags; + int ret; + + spin_lock_irqsave(&line->lock, flags); + + /*write_room subtracts 1 for the needed NULL, so we readd it.*/ + ret = LINE_BUFSIZE - (write_room(line) + 1); + spin_unlock_irqrestore(&line->lock, flags); + + return ret; +} + +/* + * This copies the content of buf into the circular buffer associated with + * this line. + * The return value is the number of characters actually copied, i.e. the ones + * for which there was space: this function is not supposed to ever flush out + * the circular buffer. + * + * Must be called while holding line->lock! + */ static int buffer_data(struct line *line, const char *buf, int len) { int end, room; @@ -70,48 +120,95 @@ static int buffer_data(struct line *line, const char *buf, int len) len = (len > room) ? room : len; end = line->buffer + LINE_BUFSIZE - line->tail; - if(len < end){ + + if (len < end){ memcpy(line->tail, buf, len); line->tail += len; - } - else { + } else { + /* The circular buffer is wrapping */ memcpy(line->tail, buf, end); buf += end; memcpy(line->buffer, buf, len - end); line->tail = line->buffer + len - end; } - return(len); + return len; } +/* + * Flushes the ring buffer to the output channels. That is, write_chan is + * called, passing it line->head as buffer, and an appropriate count. + * + * On exit, returns 1 when the buffer is empty, + * 0 when the buffer is not empty on exit, + * and -errno when an error occurred. + * + * Must be called while holding line->lock!*/ static int flush_buffer(struct line *line) { int n, count; if ((line->buffer == NULL) || (line->head == line->tail)) - return(1); + return 1; if (line->tail < line->head) { + /* line->buffer + LINE_BUFSIZE is the end of the buffer! */ count = line->buffer + LINE_BUFSIZE - line->head; + n = write_chan(&line->chan_list, line->head, count, line->driver->write_irq); if (n < 0) - return(n); - if (n == count) + return n; + if (n == count) { + /* We have flushed from ->head to buffer end, now we + * must flush only from the beginning to ->tail.*/ line->head = line->buffer; - else { + } else { line->head += n; - return(0); + return 0; } } count = line->tail - line->head; n = write_chan(&line->chan_list, line->head, count, line->driver->write_irq); - if(n < 0) return(n); + + if(n < 0) + return n; line->head += n; - return(line->head == line->tail); + return line->head == line->tail; +} + +void line_flush_buffer(struct tty_struct *tty) +{ + struct line *line = tty->driver_data; + unsigned long flags; + int err; + + /*XXX: copied from line_write, verify if it is correct!*/ + if(tty->stopped) + return; + //return 0; + + spin_lock_irqsave(&line->lock, flags); + err = flush_buffer(line); + /*if (err == 1) + err = 0;*/ + spin_unlock_irqrestore(&line->lock, flags); + //return err; +} + +/* We map both ->flush_chars and ->put_char (which go in pair) onto ->flush_buffer + * and ->write. Hope it's not that bad.*/ +void line_flush_chars(struct tty_struct *tty) +{ + line_flush_buffer(tty); +} + +void line_put_char(struct tty_struct *tty, unsigned char ch) +{ + line_write(tty, &ch, sizeof(ch)); } int line_write(struct tty_struct *tty, const unsigned char *buf, int len) @@ -120,38 +217,31 @@ int line_write(struct tty_struct *tty, const unsigned char *buf, int len) unsigned long flags; int n, err, ret = 0; - if(tty->stopped) return 0; + if(tty->stopped) + return 0; - down(&line->sem); - if(line->head != line->tail){ - local_irq_save(flags); + spin_lock_irqsave(&line->lock, flags); + if (line->head != line->tail) { ret = buffer_data(line, buf, len); err = flush_buffer(line); - local_irq_restore(flags); - if(err <= 0 && (err != -EAGAIN || !ret)) + if (err <= 0 && (err != -EAGAIN || !ret)) ret = err; - } - else { + } else { n = write_chan(&line->chan_list, buf, len, line->driver->write_irq); - if(n < 0){ + if (n < 0) { ret = n; goto out_up; } len -= n; ret += n; - if(len > 0) + if (len > 0) ret += buffer_data(line, buf + n, len); } - out_up: - up(&line->sem); - return(ret); -} - -void line_put_char(struct tty_struct *tty, unsigned char ch) -{ - line_write(tty, &ch, sizeof(ch)); +out_up: + spin_unlock_irqrestore(&line->lock, flags); + return ret; } void line_set_termios(struct tty_struct *tty, struct termios * old) @@ -159,11 +249,6 @@ void line_set_termios(struct tty_struct *tty, struct termios * old) /* nothing */ } -int line_chars_in_buffer(struct tty_struct *tty) -{ - return 0; -} - static struct { int cmd; char *level; @@ -250,7 +335,7 @@ int line_ioctl(struct tty_struct *tty, struct file * file, ret = -ENOIOCTLCMD; break; } - return(ret); + return ret; } static irqreturn_t line_write_interrupt(int irq, void *data, @@ -260,18 +345,23 @@ static irqreturn_t line_write_interrupt(int irq, void *data, struct line *line = tty->driver_data; int err; + /* Interrupts are enabled here because we registered the interrupt with + * SA_INTERRUPT (see line_setup_irq).*/ + + spin_lock_irq(&line->lock); err = flush_buffer(line); - if(err == 0) - return(IRQ_NONE); - else if(err < 0){ + if (err == 0) { + return IRQ_NONE; + } else if(err < 0) { line->head = line->buffer; line->tail = line->buffer; } + spin_unlock_irq(&line->lock); if(tty == NULL) - return(IRQ_NONE); + return IRQ_NONE; - if(test_bit(TTY_DO_WRITE_WAKEUP, &tty->flags) && + if (test_bit(TTY_DO_WRITE_WAKEUP, &tty->flags) && (tty->ldisc.write_wakeup != NULL)) (tty->ldisc.write_wakeup)(tty); @@ -281,9 +371,9 @@ static irqreturn_t line_write_interrupt(int irq, void *data, * writes. */ - if(waitqueue_active(&tty->write_wait)) + if (waitqueue_active(&tty->write_wait)) wake_up_interruptible(&tty->write_wait); - return(IRQ_HANDLED); + return IRQ_HANDLED; } int line_setup_irq(int fd, int input, int output, struct tty_struct *tty) @@ -292,15 +382,18 @@ int line_setup_irq(int fd, int input, int output, struct tty_struct *tty) struct line_driver *driver = line->driver; int err = 0, flags = SA_INTERRUPT | SA_SHIRQ | SA_SAMPLE_RANDOM; - if(input) err = um_request_irq(driver->read_irq, fd, IRQ_READ, + if (input) + err = um_request_irq(driver->read_irq, fd, IRQ_READ, line_interrupt, flags, driver->read_irq_name, tty); - if(err) return(err); - if(output) err = um_request_irq(driver->write_irq, fd, IRQ_WRITE, + if (err) + return err; + if (output) + err = um_request_irq(driver->write_irq, fd, IRQ_WRITE, line_write_interrupt, flags, driver->write_irq_name, tty); line->have_irq = 1; - return(err); + return err; } void line_disable(struct tty_struct *tty, int current_irq) @@ -336,7 +429,9 @@ int line_open(struct line *lines, struct tty_struct *tty, line = &lines[tty->index]; tty->driver_data = line; - down(&line->sem); + /* The IRQ which takes this lock is not yet enabled and won't be run + * before the end, so we don't need to use spin_lock_irq.*/ + spin_lock(&line->lock); if (tty->count == 1) { if (!line->valid) { err = -ENODEV; @@ -349,6 +444,7 @@ int line_open(struct line *lines, struct tty_struct *tty, err = open_chan(&line->chan_list); if(err) goto out; } + /* Here the interrupt is registered.*/ enable_chan(&line->chan_list, tty); INIT_WORK(&line->task, line_timer_cb, tty); } @@ -362,21 +458,27 @@ int line_open(struct line *lines, struct tty_struct *tty, line->count++; out: - up(&line->sem); - return(err); + spin_unlock(&line->lock); + return err; } void line_close(struct tty_struct *tty, struct file * filp) { struct line *line = tty->driver_data; - down(&line->sem); + /* XXX: I assume this should be called in process context, not with interrupt + * disabled!*/ + spin_lock_irq(&line->lock); + + /* We ignore the error anyway! */ + flush_buffer(line); + line->count--; if (tty->count == 1) { line_disable(tty, -1); tty->driver_data = NULL; } - up(&line->sem); + spin_unlock_irq(&line->lock); } void close_lines(struct line *lines, int nlines) @@ -387,31 +489,41 @@ void close_lines(struct line *lines, int nlines) close_chan(&lines[i].chan_list); } -int line_setup(struct line *lines, int num, char *init, int all_allowed) +/* Common setup code for both startup command line and mconsole initialization. + * @lines contains the the array (of size @num) to modify; + * @init is the setup string; + * @all_allowed is a boolean saying if we can setup the whole @lines + * at once. For instance, it will be usually true for startup init. (where we + * can use con=xterm) and false for mconsole.*/ + +int line_setup(struct line *lines, unsigned int num, char *init, int all_allowed) { int i, n; char *end; - if(*init == '=') n = -1; - else { + if(*init == '=') { + /* We said con=/ssl= instead of con#=, so we are configuring all + * consoles at once.*/ + n = -1; + } else { n = simple_strtoul(init, &end, 0); if(*end != '='){ printk(KERN_ERR "line_setup failed to parse \"%s\"\n", init); - return(0); + return 0; } init = end; } init++; - if((n >= 0) && (n >= num)){ + + if (n >= (signed int) num) { printk("line_setup - %d out of range ((0 ... %d) allowed)\n", n, num - 1); - return(0); - } - else if (n >= 0){ + return 0; + } else if (n >= 0){ if (lines[n].count > 0) { printk("line_setup - device %d is open\n", n); - return(0); + return 0; } if (lines[n].init_pri <= INIT_ONE){ lines[n].init_pri = INIT_ONE; @@ -422,13 +534,11 @@ int line_setup(struct line *lines, int num, char *init, int all_allowed) lines[n].valid = 1; } } - } - else if(!all_allowed){ + } else if(!all_allowed){ printk("line_setup - can't configure all devices from " "mconsole\n"); - return(0); - } - else { + return 0; + } else { for(i = 0; i < num; i++){ if(lines[i].init_pri <= INIT_ALL){ lines[i].init_pri = INIT_ALL; @@ -440,21 +550,21 @@ int line_setup(struct line *lines, int num, char *init, int all_allowed) } } } - return(1); + return 1; } -int line_config(struct line *lines, int num, char *str) +int line_config(struct line *lines, unsigned int num, char *str) { char *new = uml_strdup(str); if(new == NULL){ printk("line_config - uml_strdup failed\n"); - return(-ENOMEM); + return -ENOMEM; } - return(!line_setup(lines, num, new, 0)); + return !line_setup(lines, num, new, 0); } -int line_get_config(char *name, struct line *lines, int num, char *str, +int line_get_config(char *name, struct line *lines, unsigned int num, char *str, int size, char **error_out) { struct line *line; @@ -464,47 +574,33 @@ int line_get_config(char *name, struct line *lines, int num, char *str, dev = simple_strtoul(name, &end, 0); if((*end != '\0') || (end == name)){ *error_out = "line_get_config failed to parse device number"; - return(0); + return 0; } if((dev < 0) || (dev >= num)){ - *error_out = "device number of of range"; - return(0); + *error_out = "device number out of range"; + return 0; } line = &lines[dev]; - down(&line->sem); + spin_lock(&line->lock); if(!line->valid) CONFIG_CHUNK(str, size, n, "none", 1); else if(line->count == 0) CONFIG_CHUNK(str, size, n, line->init_str, 1); else n = chan_config_string(&line->chan_list, str, size, error_out); - up(&line->sem); + spin_unlock(&line->lock); - return(n); + return n; } -int line_remove(struct line *lines, int num, char *str) +int line_remove(struct line *lines, unsigned int num, char *str) { char config[sizeof("conxxxx=none\0")]; sprintf(config, "%s=none", str); - return(!line_setup(lines, num, config, 0)); -} - -int line_write_room(struct tty_struct *tty) -{ - struct line *dev = tty->driver_data; - int room; - - if (tty->stopped) - return 0; - room = write_room(dev); - if (0 == room) - printk(KERN_DEBUG "%s: %s: no room left in buffer\n", - __FUNCTION__,tty->name); - return room; + return !line_setup(lines, num, config, 0); } struct tty_driver *line_register_devfs(struct lines *set, @@ -553,7 +649,7 @@ void lines_init(struct line *lines, int nlines) for(i = 0; i < nlines; i++){ line = &lines[i]; INIT_LIST_HEAD(&line->chan_list); - sema_init(&line->sem, 1); + spin_lock_init(&line->lock); if(line->init_str != NULL){ line->init_str = uml_strdup(line->init_str); if(line->init_str == NULL) @@ -587,7 +683,7 @@ irqreturn_t winch_interrupt(int irq, void *data, struct pt_regs *unused) "errno = %d\n", -err); printk("fd %d is losing SIGWINCH support\n", winch->tty_fd); - return(IRQ_HANDLED); + return IRQ_HANDLED; } goto out; } @@ -603,7 +699,7 @@ irqreturn_t winch_interrupt(int irq, void *data, struct pt_regs *unused) out: if(winch->fd != -1) reactivate_fd(winch->fd, WINCH_IRQ); - return(IRQ_HANDLED); + return IRQ_HANDLED; } DECLARE_MUTEX(winch_handler_sem); @@ -625,7 +721,7 @@ void register_winch_irq(int fd, int tty_fd, int pid, struct tty_struct *tty) .pid = pid, .tty = tty }); list_add(&winch->list, &winch_handlers); - if(um_request_irq(WINCH_IRQ, fd, IRQ_READ, winch_interrupt, + if(um_request_irq(WINCH_IRQ, fd, IRQ_READ, winch_interrupt, SA_INTERRUPT | SA_SHIRQ | SA_SAMPLE_RANDOM, "winch", winch) < 0) printk("register_winch_irq - failed to register IRQ\n"); @@ -656,26 +752,16 @@ char *add_xterm_umid(char *base) int len; umid = get_umid(1); - if(umid == NULL) return(base); + if(umid == NULL) + return base; len = strlen(base) + strlen(" ()") + strlen(umid) + 1; title = kmalloc(len, GFP_KERNEL); if(title == NULL){ printk("Failed to allocate buffer for xterm title\n"); - return(base); + return base; } snprintf(title, len, "%s (%s)", base, umid); - return(title); + return title; } - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/drivers/ssl.c b/arch/um/drivers/ssl.c index c5839c3..a2bac42 100644 --- a/arch/um/drivers/ssl.c +++ b/arch/um/drivers/ssl.c @@ -107,11 +107,6 @@ int ssl_open(struct tty_struct *tty, struct file *filp) } #if 0 -static int ssl_chars_in_buffer(struct tty_struct *tty) -{ - return(0); -} - static void ssl_flush_buffer(struct tty_struct *tty) { return; @@ -149,11 +144,11 @@ static struct tty_operations ssl_ops = { .put_char = line_put_char, .write_room = line_write_room, .chars_in_buffer = line_chars_in_buffer, + .flush_buffer = line_flush_buffer, + .flush_chars = line_flush_chars, .set_termios = line_set_termios, .ioctl = line_ioctl, #if 0 - .flush_chars = ssl_flush_chars, - .flush_buffer = ssl_flush_buffer, .throttle = ssl_throttle, .unthrottle = ssl_unthrottle, .stop = ssl_stop, @@ -171,10 +166,11 @@ static void ssl_console_write(struct console *c, const char *string, unsigned len) { struct line *line = &serial_lines[c->index]; + unsigned long flags; - down(&line->sem); + spin_lock_irqsave(&line->lock, flags); console_write_chan(&line->chan_list, string, len); - up(&line->sem); + spin_unlock_irqrestore(&line->lock, flags); } static struct tty_driver *ssl_console_device(struct console *c, int *index) @@ -238,14 +234,3 @@ static int ssl_chan_setup(char *str) __setup("ssl", ssl_chan_setup); __channel_help(ssl_chan_setup, "ssl"); - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/drivers/stdio_console.c b/arch/um/drivers/stdio_console.c index e604d7c..361d0be 100644 --- a/arch/um/drivers/stdio_console.c +++ b/arch/um/drivers/stdio_console.c @@ -116,8 +116,11 @@ static struct tty_operations console_ops = { .open = con_open, .close = line_close, .write = line_write, + .put_char = line_put_char, .write_room = line_write_room, .chars_in_buffer = line_chars_in_buffer, + .flush_buffer = line_flush_buffer, + .flush_chars = line_flush_chars, .set_termios = line_set_termios, .ioctl = line_ioctl, }; @@ -126,10 +129,11 @@ static void uml_console_write(struct console *console, const char *string, unsigned len) { struct line *line = &vts[console->index]; + unsigned long flags; - down(&line->sem); + spin_lock_irqsave(&line->lock, flags); console_write_chan(&line->chan_list, string, len); - up(&line->sem); + spin_unlock_irqrestore(&line->lock, flags); } static struct tty_driver *uml_console_device(struct console *c, int *index) @@ -192,14 +196,3 @@ static int console_chan_setup(char *str) } __setup("con", console_chan_setup); __channel_help(console_chan_setup, "con"); - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ diff --git a/arch/um/include/line.h b/arch/um/include/line.h index 6d81ecc..4c5e92c 100644 --- a/arch/um/include/line.h +++ b/arch/um/include/line.h @@ -10,7 +10,7 @@ #include "linux/workqueue.h" #include "linux/tty.h" #include "linux/interrupt.h" -#include "asm/semaphore.h" +#include "linux/spinlock.h" #include "chan_user.h" #include "mconsole_kern.h" @@ -37,10 +37,18 @@ struct line { struct list_head chan_list; int valid; int count; - struct semaphore sem; + /*This lock is actually, mostly, local to*/ + spinlock_t lock; + + /* Yes, this is a real circular buffer. + * XXX: And this should become a struct kfifo! + * + * buffer points to a buffer allocated on demand, of length + * LINE_BUFSIZE, head to the start of the ring, tail to the end.*/ char *buffer; char *head; char *tail; + int sigio; struct work_struct task; struct line_driver *driver; @@ -52,7 +60,6 @@ struct line { init_pri : INIT_STATIC, \ chan_list : { }, \ valid : 1, \ - sem : { }, \ buffer : NULL, \ head : NULL, \ tail : NULL, \ @@ -69,15 +76,18 @@ struct lines { extern void line_close(struct tty_struct *tty, struct file * filp); extern int line_open(struct line *lines, struct tty_struct *tty, struct chan_opts *opts); -extern int line_setup(struct line *lines, int num, char *init, +extern int line_setup(struct line *lines, unsigned int sizeof_lines, char *init, int all_allowed); extern int line_write(struct tty_struct *tty, const unsigned char *buf, int len); extern void line_put_char(struct tty_struct *tty, unsigned char ch); extern void line_set_termios(struct tty_struct *tty, struct termios * old); extern int line_chars_in_buffer(struct tty_struct *tty); +extern void line_flush_buffer(struct tty_struct *tty); +extern void line_flush_chars(struct tty_struct *tty); extern int line_write_room(struct tty_struct *tty); extern int line_ioctl(struct tty_struct *tty, struct file * file, unsigned int cmd, unsigned long arg); + extern char *add_xterm_umid(char *base); extern int line_setup_irq(int fd, int input, int output, struct tty_struct *tty); extern void line_close_chan(struct line *line); @@ -89,20 +99,10 @@ extern struct tty_driver * line_register_devfs(struct lines *set, int nlines); extern void lines_init(struct line *lines, int nlines); extern void close_lines(struct line *lines, int nlines); -extern int line_config(struct line *lines, int num, char *str); -extern int line_remove(struct line *lines, int num, char *str); -extern int line_get_config(char *dev, struct line *lines, int num, char *str, + +extern int line_config(struct line *lines, unsigned int sizeof_lines, char *str); +extern int line_remove(struct line *lines, unsigned int sizeof_lines, char *str); +extern int line_get_config(char *dev, struct line *lines, unsigned int sizeof_lines, char *str, int size, char **error_out); #endif - -/* - * Overrides for Emacs so that we follow Linus's tabbing style. - * Emacs will notice this stuff at the end of the file and automatically - * adjust the settings for this buffer only. This must remain at the end - * of the file. - * --------------------------------------------------------------------------- - * Local variables: - * c-file-style: "linux" - * End: - */ -- cgit v0.10.2 From ffa0aea681a5f3c8aecbb86f1cfd3486043805de Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Sun, 1 May 2005 08:58:56 -0700 Subject: [PATCH] uml - hostfs: avoid buffers Use this: .set_page_dirty = __set_page_dirty_nobuffers, We already dropped the inclusion of , and we don't have a backing block device for this FS. "Without having looked at it, I'm sure that hostfs does not use buffer_heads. So setting your ->set_page_dirty a_op to point at __set_page_dirty_nobuffers() is a reasonable thing to do - it'll provide a slight speedup." This speedup is one less spinlock held and one less conditional branch, which isn't bad. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index a88ad29..e6c63d9 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -521,7 +521,7 @@ int hostfs_commit_write(struct file *file, struct page *page, unsigned from, static struct address_space_operations hostfs_aops = { .writepage = hostfs_writepage, .readpage = hostfs_readpage, -/* .set_page_dirty = __set_page_dirty_nobuffers, */ + .set_page_dirty = __set_page_dirty_nobuffers, .prepare_write = hostfs_prepare_write, .commit_write = hostfs_commit_write }; -- cgit v0.10.2 From acef2e55d2a1b59bb5610cacc546c1d1b5de2dc9 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Sun, 1 May 2005 08:58:56 -0700 Subject: [PATCH] uml: commentary about forking flag Add some commentary about UML internals, for a strange trick. Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-um/processor-generic.h b/include/asm-um/processor-generic.h index 4d94049..b953b1a 100644 --- a/include/asm-um/processor-generic.h +++ b/include/asm-um/processor-generic.h @@ -17,6 +17,10 @@ struct task_struct; struct mm_struct; struct thread_struct { + /* This flag is set to 1 before calling do_fork (and analyzed in + * copy_thread) to mark that we are begin called from userspace (fork / + * vfork / clone), and reset to 0 after. It is left to 0 when called + * from kernelspace (i.e. kernel_thread() or fork_idle(), as of 2.6.11). */ int forking; int nsyscalls; struct pt_regs regs; -- cgit v0.10.2 From 2c49be99ede1e3685a9817a564b4b1678d48e1b9 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Sun, 1 May 2005 08:58:57 -0700 Subject: [PATCH] uml ubd: handle readonly status Use the set_disk_ro() API when the backing file is read-only, to mark the disk read-only, during the ->open(). The current hack does not work when doing a mount -o remount. Also, mark explicitly the code paths which should no more be triggerable (I've removed the WARN_ON(1) things). They should actually become BUG()s probably but I'll avoid that since I'm not so sure the change works so well. I gave it only some limited testing. Signed-off-by: Paolo 'Blaisorblade' Giarrusso CC: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c index 4d8b165..9a56ff9 100644 --- a/arch/um/drivers/ubd_kern.c +++ b/arch/um/drivers/ubd_kern.c @@ -156,6 +156,7 @@ static struct gendisk *fake_gendisk[MAX_DEV]; static struct openflags global_openflags = OPEN_FLAGS; struct cow { + /* This is the backing file, actually */ char *file; int fd; unsigned long *bitmap; @@ -927,10 +928,14 @@ static int ubd_open(struct inode *inode, struct file *filp) } } dev->count++; - if((filp->f_mode & FMODE_WRITE) && !dev->openflags.w){ + set_disk_ro(disk, !dev->openflags.w); + + /* This should no more be needed. And it didn't work anyway to exclude + * read-write remounting of filesystems.*/ + /*if((filp->f_mode & FMODE_WRITE) && !dev->openflags.w){ if(--dev->count == 0) ubd_close(dev); err = -EROFS; - } + }*/ out: return(err); } @@ -1096,6 +1101,7 @@ static int prepare_request(struct request *req, struct io_thread_req *io_req) if(req->rq_status == RQ_INACTIVE) return(1); + /* This should be impossible now */ if((rq_data_dir(req) == WRITE) && !dev->openflags.w){ printk("Write attempted on readonly ubd device %s\n", disk->disk_name); @@ -1243,6 +1249,7 @@ static int ubd_check_remapped(int fd, unsigned long address, int is_write, /* It's a write to a ubd device */ + /* This should be impossible now */ if(!dev->openflags.w){ /* It's a write access on a read-only device - probably * shouldn't happen. If the kernel is trying to change @@ -1605,8 +1612,7 @@ void do_io(struct io_thread_req *req) } } while((n < len) && (n != 0)); if (n < len) memset(&buf[n], 0, len - n); - } - else { + } else { n = os_write_file(req->fds[bit], buf, len); if(n != len){ printk("do_io - write failed err = %d " -- cgit v0.10.2 From b2c6678c858c5e54d95b996754adfb319cf65735 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Sun, 1 May 2005 08:58:57 -0700 Subject: [PATCH] s390: regenerate defconfig Regenerate the default configuration for s390. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/s390/defconfig b/arch/s390/defconfig index 1358b42..07fd041 100644 --- a/arch/s390/defconfig +++ b/arch/s390/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.11 -# Wed Mar 2 16:57:55 2005 +# Linux kernel version: 2.6.12-rc3 +# Fri Apr 22 15:30:58 2005 # CONFIG_MMU=y CONFIG_RWSEM_XCHGADD_ALGORITHM=y @@ -15,6 +15,7 @@ CONFIG_UID16=y CONFIG_EXPERIMENTAL=y CONFIG_CLEAN_COMPILE=y CONFIG_LOCK_KERNEL=y +CONFIG_INIT_ENV_ARG_LIMIT=32 # # General setup @@ -26,24 +27,25 @@ CONFIG_SYSVIPC=y # CONFIG_BSD_PROCESS_ACCT is not set CONFIG_SYSCTL=y # CONFIG_AUDIT is not set -CONFIG_LOG_BUF_SHIFT=17 CONFIG_HOTPLUG=y CONFIG_KOBJECT_UEVENT=y CONFIG_IKCONFIG=y CONFIG_IKCONFIG_PROC=y +# CONFIG_CPUSETS is not set # CONFIG_EMBEDDED is not set CONFIG_KALLSYMS=y # CONFIG_KALLSYMS_ALL is not set # CONFIG_KALLSYMS_EXTRA_PASS is not set +CONFIG_BASE_FULL=y CONFIG_FUTEX=y CONFIG_EPOLL=y -# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set CONFIG_SHMEM=y CONFIG_CC_ALIGN_FUNCTIONS=0 CONFIG_CC_ALIGN_LABELS=0 CONFIG_CC_ALIGN_LOOPS=0 CONFIG_CC_ALIGN_JUMPS=0 # CONFIG_TINY_SHMEM is not set +CONFIG_BASE_SMALL=0 # # Loadable module support @@ -261,7 +263,6 @@ CONFIG_NET=y # CONFIG_PACKET=y # CONFIG_PACKET_MMAP is not set -# CONFIG_NETLINK_DEV is not set CONFIG_UNIX=y CONFIG_NET_KEY=y CONFIG_INET=y @@ -329,6 +330,7 @@ CONFIG_NET_SCH_DSMARK=m CONFIG_NET_QOS=y CONFIG_NET_ESTIMATOR=y CONFIG_NET_CLS=y +# CONFIG_NET_CLS_BASIC is not set CONFIG_NET_CLS_TCINDEX=m CONFIG_NET_CLS_ROUTE4=m CONFIG_NET_CLS_ROUTE=y @@ -338,6 +340,7 @@ CONFIG_NET_CLS_U32=m # CONFIG_NET_CLS_IND is not set CONFIG_NET_CLS_RSVP=m CONFIG_NET_CLS_RSVP6=m +# CONFIG_NET_EMATCH is not set # CONFIG_NET_CLS_ACT is not set CONFIG_NET_CLS_POLICE=y @@ -393,6 +396,8 @@ CONFIG_CTC=m CONFIG_IUCV=m # CONFIG_NETIUCV is not set # CONFIG_SMSGIUCV is not set +# CONFIG_CLAW is not set +# CONFIG_MPC is not set CONFIG_QETH=y # @@ -532,10 +537,13 @@ CONFIG_MSDOS_PARTITION=y # # Kernel hacking # +# CONFIG_PRINTK_TIME is not set CONFIG_DEBUG_KERNEL=y CONFIG_MAGIC_SYSRQ=y +CONFIG_LOG_BUF_SHIFT=17 # CONFIG_SCHEDSTATS is not set # CONFIG_DEBUG_SLAB is not set +# CONFIG_DEBUG_SPINLOCK is not set # CONFIG_DEBUG_SPINLOCK_SLEEP is not set # CONFIG_DEBUG_KOBJECT is not set # CONFIG_DEBUG_INFO is not set @@ -560,6 +568,7 @@ CONFIG_CRYPTO=y # CONFIG_CRYPTO_SHA256 is not set # CONFIG_CRYPTO_SHA512 is not set # CONFIG_CRYPTO_WP512 is not set +# CONFIG_CRYPTO_TGR192 is not set # CONFIG_CRYPTO_DES is not set # CONFIG_CRYPTO_DES_Z990 is not set # CONFIG_CRYPTO_BLOWFISH is not set -- cgit v0.10.2 From 4b7e0706620e3947dc1685dfdbc1413404afb545 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Sun, 1 May 2005 08:58:57 -0700 Subject: [PATCH] s390: idle timer setup Fix overflow in calculation of the new tod value in stop_hz_timer and fix wrong virtual timer list idle time in case the virtual timer is already expired in stop_cpu_timer. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index 061e811..8ca4856 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -244,7 +244,7 @@ int sysctl_hz_timer = 1; */ static inline void stop_hz_timer(void) { - __u64 timer; + __u64 timer, todval; if (sysctl_hz_timer != 0) return; @@ -265,8 +265,14 @@ static inline void stop_hz_timer(void) * for the next event. */ timer = (__u64) (next_timer_interrupt() - jiffies) + jiffies_64; - timer = jiffies_timer_cc + timer * CLK_TICKS_PER_JIFFY; - asm volatile ("SCKC %0" : : "m" (timer)); + todval = -1ULL; + /* Be careful about overflows. */ + if (timer < (-1ULL / CLK_TICKS_PER_JIFFY)) { + timer = jiffies_timer_cc + timer * CLK_TICKS_PER_JIFFY; + if (timer >= jiffies_timer_cc) + todval = timer; + } + asm volatile ("SCKC %0" : : "m" (todval)); } /* diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c index bb6cf02..fa07265 100644 --- a/arch/s390/kernel/vtime.c +++ b/arch/s390/kernel/vtime.c @@ -122,12 +122,17 @@ static void start_cpu_timer(void) struct vtimer_queue *vt_list; vt_list = &per_cpu(virt_cpu_timer, smp_processor_id()); - set_vtimer(vt_list->idle); + + /* CPU timer interrupt is pending, don't reprogramm it */ + if (vt_list->idle & 1LL<<63) + return; + + if (!list_empty(&vt_list->list)) + set_vtimer(vt_list->idle); } static void stop_cpu_timer(void) { - __u64 done; struct vtimer_queue *vt_list; vt_list = &per_cpu(virt_cpu_timer, smp_processor_id()); @@ -138,21 +143,17 @@ static void stop_cpu_timer(void) goto fire; } - /* store progress */ - asm volatile ("STPT %0" : "=m" (done)); + /* store the actual expire value */ + asm volatile ("STPT %0" : "=m" (vt_list->idle)); /* - * If done is negative we do not stop the CPU timer - * because we will get instantly an interrupt that - * will start the CPU timer again. + * If the CPU timer is negative we don't reprogramm + * it because we will get instantly an interrupt. */ - if (done & 1LL<<63) + if (vt_list->idle & 1LL<<63) return; - else - vt_list->offset += vt_list->to_expire - done; - /* save the actual expire value */ - vt_list->idle = done; + vt_list->offset += vt_list->to_expire - vt_list->idle; /* * We cannot halt the CPU timer, we just write a value that -- cgit v0.10.2 From c9e3735359ac2d74ee61c6f1e5724f4a6db570bf Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Sun, 1 May 2005 08:58:57 -0700 Subject: [PATCH] s390: fix memory holes and cleanup setup_arch The memory setup didn't take care of memory holes and this makes the memory management think there would be more memory available than there is in reality. That causes the OOM killer to kill processes even if there is enough memory left that can be written to the swap space. The patch fixes this by using free_area_init_node with an array of memory holes instead of free_area_init. Further the patch cleans up the code in setup.c by splitting setup_arch into smaller pieces. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index c879c40..f0679be 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -60,6 +60,8 @@ struct { #define CHUNK_READ_WRITE 0 #define CHUNK_READ_ONLY 1 volatile int __cpu_logical_map[NR_CPUS]; /* logical cpu to cpu address */ +unsigned long __initdata zholes_size[MAX_NR_ZONES]; +static unsigned long __initdata memory_end; /* * Setup options @@ -78,11 +80,15 @@ static char command_line[COMMAND_LINE_SIZE] = { 0, }; static struct resource code_resource = { .name = "Kernel code", + .start = (unsigned long) &_text, + .end = (unsigned long) &_etext - 1, .flags = IORESOURCE_BUSY | IORESOURCE_MEM, }; static struct resource data_resource = { .name = "Kernel data", + .start = (unsigned long) &_etext, + .end = (unsigned long) &_edata - 1, .flags = IORESOURCE_BUSY | IORESOURCE_MEM, }; @@ -310,90 +316,50 @@ void machine_power_off(void) EXPORT_SYMBOL(machine_power_off); -/* - * Setup function called from init/main.c just after the banner - * was printed. - */ -extern char _pstart, _pend, _stext; - -void __init setup_arch(char **cmdline_p) +static void __init +add_memory_hole(unsigned long start, unsigned long end) { - unsigned long bootmap_size; - unsigned long memory_start, memory_end; - char c = ' ', cn, *to = command_line, *from = COMMAND_LINE; - unsigned long start_pfn, end_pfn; - static unsigned int smptrap=0; - unsigned long delay = 0; - struct _lowcore *lc; - int i; + unsigned long dma_pfn = MAX_DMA_ADDRESS >> PAGE_SHIFT; + + if (end <= dma_pfn) + zholes_size[ZONE_DMA] += end - start + 1; + else if (start > dma_pfn) + zholes_size[ZONE_NORMAL] += end - start + 1; + else { + zholes_size[ZONE_DMA] += dma_pfn - start + 1; + zholes_size[ZONE_NORMAL] += end - dma_pfn; + } +} - if (smptrap) - return; - smptrap=1; +static void __init +parse_cmdline_early(char **cmdline_p) +{ + char c = ' ', cn, *to = command_line, *from = COMMAND_LINE; + unsigned long delay = 0; - /* - * print what head.S has found out about the machine - */ -#ifndef CONFIG_ARCH_S390X - printk((MACHINE_IS_VM) ? - "We are running under VM (31 bit mode)\n" : - "We are running native (31 bit mode)\n"); - printk((MACHINE_HAS_IEEE) ? - "This machine has an IEEE fpu\n" : - "This machine has no IEEE fpu\n"); -#else /* CONFIG_ARCH_S390X */ - printk((MACHINE_IS_VM) ? - "We are running under VM (64 bit mode)\n" : - "We are running native (64 bit mode)\n"); -#endif /* CONFIG_ARCH_S390X */ + /* Save unparsed command line copy for /proc/cmdline */ + memcpy(saved_command_line, COMMAND_LINE, COMMAND_LINE_SIZE); + saved_command_line[COMMAND_LINE_SIZE-1] = '\0'; - ROOT_DEV = Root_RAM0; - memory_start = (unsigned long) &_end; /* fixit if use $CODELO etc*/ -#ifndef CONFIG_ARCH_S390X - memory_end = memory_size & ~0x400000UL; /* align memory end to 4MB */ - /* - * We need some free virtual space to be able to do vmalloc. - * On a machine with 2GB memory we make sure that we have at - * least 128 MB free space for vmalloc. - */ - if (memory_end > 1920*1024*1024) - memory_end = 1920*1024*1024; -#else /* CONFIG_ARCH_S390X */ - memory_end = memory_size & ~0x200000UL; /* detected in head.s */ -#endif /* CONFIG_ARCH_S390X */ - init_mm.start_code = PAGE_OFFSET; - init_mm.end_code = (unsigned long) &_etext; - init_mm.end_data = (unsigned long) &_edata; - init_mm.brk = (unsigned long) &_end; - - code_resource.start = (unsigned long) &_text; - code_resource.end = (unsigned long) &_etext - 1; - data_resource.start = (unsigned long) &_etext; - data_resource.end = (unsigned long) &_edata - 1; - - /* Save unparsed command line copy for /proc/cmdline */ - memcpy(saved_command_line, COMMAND_LINE, COMMAND_LINE_SIZE); - saved_command_line[COMMAND_LINE_SIZE-1] = '\0'; - - for (;;) { - /* - * "mem=XXX[kKmM]" sets memsize - */ - if (c == ' ' && strncmp(from, "mem=", 4) == 0) { - memory_end = simple_strtoul(from+4, &from, 0); - if ( *from == 'K' || *from == 'k' ) { - memory_end = memory_end << 10; - from++; - } else if ( *from == 'M' || *from == 'm' ) { - memory_end = memory_end << 20; - from++; - } - } - /* - * "ipldelay=XXX[sm]" sets ipl delay in seconds or minutes - */ - if (c == ' ' && strncmp(from, "ipldelay=", 9) == 0) { - delay = simple_strtoul(from+9, &from, 0); + for (;;) { + /* + * "mem=XXX[kKmM]" sets memsize + */ + if (c == ' ' && strncmp(from, "mem=", 4) == 0) { + memory_end = simple_strtoul(from+4, &from, 0); + if ( *from == 'K' || *from == 'k' ) { + memory_end = memory_end << 10; + from++; + } else if ( *from == 'M' || *from == 'm' ) { + memory_end = memory_end << 20; + from++; + } + } + /* + * "ipldelay=XXX[sm]" sets ipl delay in seconds or minutes + */ + if (c == ' ' && strncmp(from, "ipldelay=", 9) == 0) { + delay = simple_strtoul(from+9, &from, 0); if (*from == 's' || *from == 'S') { delay = delay*1000000; from++; @@ -403,24 +369,110 @@ void __init setup_arch(char **cmdline_p) } /* now wait for the requested amount of time */ udelay(delay); - } - cn = *(from++); - if (!cn) - break; - if (cn == '\n') - cn = ' '; /* replace newlines with space */ + } + cn = *(from++); + if (!cn) + break; + if (cn == '\n') + cn = ' '; /* replace newlines with space */ if (cn == 0x0d) cn = ' '; /* replace 0x0d with space */ - if (cn == ' ' && c == ' ') - continue; /* remove additional spaces */ - c = cn; - if (to - command_line >= COMMAND_LINE_SIZE) - break; - *(to++) = c; - } - if (c == ' ' && to > command_line) to--; - *to = '\0'; - *cmdline_p = command_line; + if (cn == ' ' && c == ' ') + continue; /* remove additional spaces */ + c = cn; + if (to - command_line >= COMMAND_LINE_SIZE) + break; + *(to++) = c; + } + if (c == ' ' && to > command_line) to--; + *to = '\0'; + *cmdline_p = command_line; +} + +static void __init +setup_lowcore(void) +{ + struct _lowcore *lc; + int lc_pages; + + /* + * Setup lowcore for boot cpu + */ + lc_pages = sizeof(void *) == 8 ? 2 : 1; + lc = (struct _lowcore *) + __alloc_bootmem(lc_pages * PAGE_SIZE, lc_pages * PAGE_SIZE, 0); + memset(lc, 0, lc_pages * PAGE_SIZE); + lc->restart_psw.mask = PSW_BASE_BITS; + lc->restart_psw.addr = + PSW_ADDR_AMODE | (unsigned long) restart_int_handler; + lc->external_new_psw.mask = PSW_KERNEL_BITS; + lc->external_new_psw.addr = + PSW_ADDR_AMODE | (unsigned long) ext_int_handler; + lc->svc_new_psw.mask = PSW_KERNEL_BITS | PSW_MASK_IO | PSW_MASK_EXT; + lc->svc_new_psw.addr = PSW_ADDR_AMODE | (unsigned long) system_call; + lc->program_new_psw.mask = PSW_KERNEL_BITS; + lc->program_new_psw.addr = + PSW_ADDR_AMODE | (unsigned long)pgm_check_handler; + lc->mcck_new_psw.mask = PSW_KERNEL_BITS; + lc->mcck_new_psw.addr = + PSW_ADDR_AMODE | (unsigned long) mcck_int_handler; + lc->io_new_psw.mask = PSW_KERNEL_BITS; + lc->io_new_psw.addr = PSW_ADDR_AMODE | (unsigned long) io_int_handler; + lc->ipl_device = S390_lowcore.ipl_device; + lc->jiffy_timer = -1LL; + lc->kernel_stack = ((unsigned long) &init_thread_union) + THREAD_SIZE; + lc->async_stack = (unsigned long) + __alloc_bootmem(ASYNC_SIZE, ASYNC_SIZE, 0) + ASYNC_SIZE; +#ifdef CONFIG_CHECK_STACK + lc->panic_stack = (unsigned long) + __alloc_bootmem(PAGE_SIZE, PAGE_SIZE, 0) + PAGE_SIZE; +#endif + lc->current_task = (unsigned long) init_thread_union.thread_info.task; + lc->thread_info = (unsigned long) &init_thread_union; +#ifdef CONFIG_ARCH_S390X + if (MACHINE_HAS_DIAG44) + lc->diag44_opcode = 0x83000044; + else + lc->diag44_opcode = 0x07000700; +#endif /* CONFIG_ARCH_S390X */ + set_prefix((u32)(unsigned long) lc); +} + +static void __init +setup_resources(void) +{ + struct resource *res; + int i; + + for (i = 0; i < MEMORY_CHUNKS && memory_chunk[i].size > 0; i++) { + res = alloc_bootmem_low(sizeof(struct resource)); + res->flags = IORESOURCE_BUSY | IORESOURCE_MEM; + switch (memory_chunk[i].type) { + case CHUNK_READ_WRITE: + res->name = "System RAM"; + break; + case CHUNK_READ_ONLY: + res->name = "System ROM"; + res->flags |= IORESOURCE_READONLY; + break; + default: + res->name = "reserved"; + } + res->start = memory_chunk[i].addr; + res->end = memory_chunk[i].addr + memory_chunk[i].size - 1; + request_resource(&iomem_resource, res); + request_resource(res, &code_resource); + request_resource(res, &data_resource); + } +} + +static void __init +setup_memory(void) +{ + unsigned long bootmap_size; + unsigned long start_pfn, end_pfn; + unsigned long last_rw_end; + int i; /* * partially used pages are not usable - thus @@ -437,6 +489,8 @@ void __init setup_arch(char **cmdline_p) /* * Register RAM areas with the bootmem allocator. */ + last_rw_end = start_pfn; + for (i = 0; i < 16 && memory_chunk[i].size > 0; i++) { unsigned long start_chunk, end_chunk; @@ -450,102 +504,91 @@ void __init setup_arch(char **cmdline_p) start_chunk = start_pfn; if (end_chunk > end_pfn) end_chunk = end_pfn; - if (start_chunk < end_chunk) + if (start_chunk < end_chunk) { free_bootmem(start_chunk << PAGE_SHIFT, (end_chunk - start_chunk) << PAGE_SHIFT); + if (last_rw_end < start_chunk) + add_memory_hole(last_rw_end, start_chunk - 1); + last_rw_end = end_chunk; + } } - /* - * Reserve the bootmem bitmap itself as well. We do this in two - * steps (first step was init_bootmem()) because this catches - * the (very unlikely) case of us accidentally initializing the - * bootmem allocator with an invalid RAM area. - */ - reserve_bootmem(start_pfn << PAGE_SHIFT, bootmap_size); + if (last_rw_end < end_pfn - 1) + add_memory_hole(last_rw_end, end_pfn - 1); + + /* + * Reserve the bootmem bitmap itself as well. We do this in two + * steps (first step was init_bootmem()) because this catches + * the (very unlikely) case of us accidentally initializing the + * bootmem allocator with an invalid RAM area. + */ + reserve_bootmem(start_pfn << PAGE_SHIFT, bootmap_size); #ifdef CONFIG_BLK_DEV_INITRD - if (INITRD_START) { + if (INITRD_START) { if (INITRD_START + INITRD_SIZE <= memory_end) { reserve_bootmem(INITRD_START, INITRD_SIZE); initrd_start = INITRD_START; initrd_end = initrd_start + INITRD_SIZE; } else { - printk("initrd extends beyond end of memory " - "(0x%08lx > 0x%08lx)\ndisabling initrd\n", - initrd_start + INITRD_SIZE, memory_end); - initrd_start = initrd_end = 0; + printk("initrd extends beyond end of memory " + "(0x%08lx > 0x%08lx)\ndisabling initrd\n", + initrd_start + INITRD_SIZE, memory_end); + initrd_start = initrd_end = 0; } - } + } #endif +} - for (i = 0; i < 16 && memory_chunk[i].size > 0; i++) { - struct resource *res; - - res = alloc_bootmem_low(sizeof(struct resource)); - res->flags = IORESOURCE_BUSY | IORESOURCE_MEM; - - switch (memory_chunk[i].type) { - case CHUNK_READ_WRITE: - res->name = "System RAM"; - break; - case CHUNK_READ_ONLY: - res->name = "System ROM"; - res->flags |= IORESOURCE_READONLY; - break; - default: - res->name = "reserved"; - } - res->start = memory_chunk[i].addr; - res->end = memory_chunk[i].addr + memory_chunk[i].size - 1; - request_resource(&iomem_resource, res); - request_resource(res, &code_resource); - request_resource(res, &data_resource); - } +/* + * Setup function called from init/main.c just after the banner + * was printed. + */ +void __init +setup_arch(char **cmdline_p) +{ /* - * Setup lowcore for boot cpu + * print what head.S has found out about the machine */ #ifndef CONFIG_ARCH_S390X - lc = (struct _lowcore *) __alloc_bootmem(PAGE_SIZE, PAGE_SIZE, 0); - memset(lc, 0, PAGE_SIZE); + printk((MACHINE_IS_VM) ? + "We are running under VM (31 bit mode)\n" : + "We are running native (31 bit mode)\n"); + printk((MACHINE_HAS_IEEE) ? + "This machine has an IEEE fpu\n" : + "This machine has no IEEE fpu\n"); #else /* CONFIG_ARCH_S390X */ - lc = (struct _lowcore *) __alloc_bootmem(2*PAGE_SIZE, 2*PAGE_SIZE, 0); - memset(lc, 0, 2*PAGE_SIZE); + printk((MACHINE_IS_VM) ? + "We are running under VM (64 bit mode)\n" : + "We are running native (64 bit mode)\n"); #endif /* CONFIG_ARCH_S390X */ - lc->restart_psw.mask = PSW_BASE_BITS; - lc->restart_psw.addr = - PSW_ADDR_AMODE | (unsigned long) restart_int_handler; - lc->external_new_psw.mask = PSW_KERNEL_BITS; - lc->external_new_psw.addr = - PSW_ADDR_AMODE | (unsigned long) ext_int_handler; - lc->svc_new_psw.mask = PSW_KERNEL_BITS | PSW_MASK_IO | PSW_MASK_EXT; - lc->svc_new_psw.addr = PSW_ADDR_AMODE | (unsigned long) system_call; - lc->program_new_psw.mask = PSW_KERNEL_BITS; - lc->program_new_psw.addr = - PSW_ADDR_AMODE | (unsigned long)pgm_check_handler; - lc->mcck_new_psw.mask = PSW_KERNEL_BITS; - lc->mcck_new_psw.addr = - PSW_ADDR_AMODE | (unsigned long) mcck_int_handler; - lc->io_new_psw.mask = PSW_KERNEL_BITS; - lc->io_new_psw.addr = PSW_ADDR_AMODE | (unsigned long) io_int_handler; - lc->ipl_device = S390_lowcore.ipl_device; - lc->jiffy_timer = -1LL; - lc->kernel_stack = ((unsigned long) &init_thread_union) + THREAD_SIZE; - lc->async_stack = (unsigned long) - __alloc_bootmem(ASYNC_SIZE, ASYNC_SIZE, 0) + ASYNC_SIZE; -#ifdef CONFIG_CHECK_STACK - lc->panic_stack = (unsigned long) - __alloc_bootmem(PAGE_SIZE, PAGE_SIZE, 0) + PAGE_SIZE; -#endif - lc->current_task = (unsigned long) init_thread_union.thread_info.task; - lc->thread_info = (unsigned long) &init_thread_union; -#ifdef CONFIG_ARCH_S390X - if (MACHINE_HAS_DIAG44) - lc->diag44_opcode = 0x83000044; - else - lc->diag44_opcode = 0x07000700; + + ROOT_DEV = Root_RAM0; +#ifndef CONFIG_ARCH_S390X + memory_end = memory_size & ~0x400000UL; /* align memory end to 4MB */ + /* + * We need some free virtual space to be able to do vmalloc. + * On a machine with 2GB memory we make sure that we have at + * least 128 MB free space for vmalloc. + */ + if (memory_end > 1920*1024*1024) + memory_end = 1920*1024*1024; +#else /* CONFIG_ARCH_S390X */ + memory_end = memory_size & ~0x200000UL; /* detected in head.s */ #endif /* CONFIG_ARCH_S390X */ - set_prefix((u32)(unsigned long) lc); + + init_mm.start_code = PAGE_OFFSET; + init_mm.end_code = (unsigned long) &_etext; + init_mm.end_data = (unsigned long) &_edata; + init_mm.brk = (unsigned long) &_end; + + parse_cmdline_early(cmdline_p); + + setup_memory(); + setup_resources(); + setup_lowcore(); + cpu_init(); __cpu_logical_map[0] = S390_lowcore.cpu_data.cpu_addr; diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c index 8e723bc..6ec5cd9 100644 --- a/arch/s390/mm/init.c +++ b/arch/s390/mm/init.c @@ -101,6 +101,7 @@ extern unsigned long _end; extern unsigned long __init_begin; extern unsigned long __init_end; +extern unsigned long __initdata zholes_size[]; /* * paging_init() sets up the page tables */ @@ -163,10 +164,13 @@ void __init paging_init(void) local_flush_tlb(); { - unsigned long zones_size[MAX_NR_ZONES] = { 0, 0, 0}; + unsigned long zones_size[MAX_NR_ZONES]; + memset(zones_size, 0, sizeof(zones_size)); zones_size[ZONE_DMA] = max_low_pfn; - free_area_init(zones_size); + free_area_init_node(0, &contig_page_data, zones_size, + __pa(PAGE_OFFSET) >> PAGE_SHIFT, + zholes_size); } return; } @@ -184,9 +188,10 @@ void __init paging_init(void) _KERN_REGION_TABLE; static const int ssm_mask = 0x04000000L; - unsigned long zones_size[MAX_NR_ZONES] = {0, 0, 0}; + unsigned long zones_size[MAX_NR_ZONES]; unsigned long dma_pfn, high_pfn; + memset(zones_size, 0, sizeof(zones_size)); dma_pfn = MAX_DMA_ADDRESS >> PAGE_SHIFT; high_pfn = max_low_pfn; @@ -198,8 +203,8 @@ void __init paging_init(void) } /* Initialize mem_map[]. */ - free_area_init(zones_size); - + free_area_init_node(0, &contig_page_data, zones_size, + __pa(PAGE_OFFSET) >> PAGE_SHIFT, zholes_size); /* * map whole physical memory to virtual memory (identity mapping) -- cgit v0.10.2 From 0b642ede47969d4180b0922d982777fe64379228 Mon Sep 17 00:00:00 2001 From: Peter Oberparleiter Date: Sun, 1 May 2005 08:58:58 -0700 Subject: [PATCH] s390: default storage key Provide an easy way to define a non-zero storage key at compile time. This is useful for debugging purposes. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/s390/kernel/s390_ksyms.c b/arch/s390/kernel/s390_ksyms.c index 11fd6d5..bee654a 100644 --- a/arch/s390/kernel/s390_ksyms.c +++ b/arch/s390/kernel/s390_ksyms.c @@ -34,7 +34,6 @@ EXPORT_SYMBOL(__clear_user_asm); EXPORT_SYMBOL(__strncpy_from_user_asm); EXPORT_SYMBOL(__strnlen_user_asm); EXPORT_SYMBOL(diag10); -EXPORT_SYMBOL(default_storage_key); /* * semaphore ops diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index f0679be..df83215 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -44,6 +44,8 @@ #include #include #include +#include +#include /* * Machine setup.. @@ -53,7 +55,6 @@ unsigned int console_devno = -1; unsigned int console_irq = -1; unsigned long memory_size = 0; unsigned long machine_flags = 0; -unsigned int default_storage_key = 0; struct { unsigned long addr, size, type; } memory_chunk[MEMORY_CHUNKS] = { { 0 } }; @@ -402,7 +403,7 @@ setup_lowcore(void) lc = (struct _lowcore *) __alloc_bootmem(lc_pages * PAGE_SIZE, lc_pages * PAGE_SIZE, 0); memset(lc, 0, lc_pages * PAGE_SIZE); - lc->restart_psw.mask = PSW_BASE_BITS; + lc->restart_psw.mask = PSW_BASE_BITS | PSW_DEFAULT_KEY; lc->restart_psw.addr = PSW_ADDR_AMODE | (unsigned long) restart_int_handler; lc->external_new_psw.mask = PSW_KERNEL_BITS; @@ -470,7 +471,7 @@ static void __init setup_memory(void) { unsigned long bootmap_size; - unsigned long start_pfn, end_pfn; + unsigned long start_pfn, end_pfn, init_pfn; unsigned long last_rw_end; int i; @@ -481,6 +482,10 @@ setup_memory(void) start_pfn = (__pa(&_end) + PAGE_SIZE - 1) >> PAGE_SHIFT; end_pfn = max_pfn = memory_end >> PAGE_SHIFT; + /* Initialize storage key for kernel pages */ + for (init_pfn = 0 ; init_pfn < start_pfn; init_pfn++) + page_set_storage_key(init_pfn << PAGE_SHIFT, PAGE_DEFAULT_KEY); + /* * Initialize the boot-time allocator (with low memory only): */ @@ -491,7 +496,7 @@ setup_memory(void) */ last_rw_end = start_pfn; - for (i = 0; i < 16 && memory_chunk[i].size > 0; i++) { + for (i = 0; i < MEMORY_CHUNKS && memory_chunk[i].size > 0; i++) { unsigned long start_chunk, end_chunk; if (memory_chunk[i].type != CHUNK_READ_WRITE) @@ -505,6 +510,11 @@ setup_memory(void) if (end_chunk > end_pfn) end_chunk = end_pfn; if (start_chunk < end_chunk) { + /* Initialize storage key for RAM pages */ + for (init_pfn = start_chunk ; init_pfn < end_chunk; + init_pfn++) + page_set_storage_key(init_pfn << PAGE_SHIFT, + PAGE_DEFAULT_KEY); free_bootmem(start_chunk << PAGE_SHIFT, (end_chunk - start_chunk) << PAGE_SHIFT); if (last_rw_end < start_chunk) @@ -513,6 +523,8 @@ setup_memory(void) } } + psw_set_key(PAGE_DEFAULT_KEY); + if (last_rw_end < end_pfn - 1) add_memory_hole(last_rw_end, end_pfn - 1); diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c index 99ce5a5..1d9b3f1 100644 --- a/drivers/s390/cio/cio.c +++ b/drivers/s390/cio/cio.c @@ -1,7 +1,7 @@ /* * drivers/s390/cio/cio.c * S/390 common I/O routines -- low level i/o calls - * $Revision: 1.131 $ + * $Revision: 1.133 $ * * Copyright (C) 1999-2002 IBM Deutschland Entwicklung GmbH, * IBM Corporation @@ -228,7 +228,7 @@ cio_start_key (struct subchannel *sch, /* subchannel structure */ int cio_start (struct subchannel *sch, struct ccw1 *cpa, __u8 lpm) { - return cio_start_key(sch, cpa, lpm, default_storage_key); + return cio_start_key(sch, cpa, lpm, PAGE_DEFAULT_KEY); } /* diff --git a/drivers/s390/cio/device_ops.c b/drivers/s390/cio/device_ops.c index 11e260e..02d01a0 100644 --- a/drivers/s390/cio/device_ops.c +++ b/drivers/s390/cio/device_ops.c @@ -1,7 +1,7 @@ /* * drivers/s390/cio/device_ops.c * - * $Revision: 1.55 $ + * $Revision: 1.56 $ * * Copyright (C) 2002 IBM Deutschland Entwicklung GmbH, * IBM Corporation @@ -128,7 +128,7 @@ ccw_device_start(struct ccw_device *cdev, struct ccw1 *cpa, unsigned long intparm, __u8 lpm, unsigned long flags) { return ccw_device_start_key(cdev, cpa, intparm, lpm, - default_storage_key, flags); + PAGE_DEFAULT_KEY, flags); } int @@ -137,7 +137,7 @@ ccw_device_start_timeout(struct ccw_device *cdev, struct ccw1 *cpa, int expires) { return ccw_device_start_timeout_key(cdev, cpa, intparm, lpm, - default_storage_key, flags, + PAGE_DEFAULT_KEY, flags, expires); } diff --git a/drivers/s390/cio/qdio.h b/drivers/s390/cio/qdio.h index 9ad14db..b6daada 100644 --- a/drivers/s390/cio/qdio.h +++ b/drivers/s390/cio/qdio.h @@ -1,7 +1,9 @@ #ifndef _CIO_QDIO_H #define _CIO_QDIO_H -#define VERSION_CIO_QDIO_H "$Revision: 1.26 $" +#include + +#define VERSION_CIO_QDIO_H "$Revision: 1.32 $" #ifdef CONFIG_QDIO_DEBUG #define QDIO_VERBOSE_LEVEL 9 @@ -42,7 +44,7 @@ #define QDIO_Q_LAPS 5 -#define QDIO_STORAGE_KEY 0 +#define QDIO_STORAGE_KEY PAGE_DEFAULT_KEY #define L2_CACHELINE_SIZE 256 #define INDICATORS_PER_CACHELINE (L2_CACHELINE_SIZE/sizeof(__u32)) diff --git a/include/asm-s390/page.h b/include/asm-s390/page.h index 614e2a9..2be287b 100644 --- a/include/asm-s390/page.h +++ b/include/asm-s390/page.h @@ -16,6 +16,8 @@ #define PAGE_SHIFT 12 #define PAGE_SIZE (1UL << PAGE_SHIFT) #define PAGE_MASK (~(PAGE_SIZE-1)) +#define PAGE_DEFAULT_ACC 0 +#define PAGE_DEFAULT_KEY (PAGE_DEFAULT_ACC << 4) #ifdef __KERNEL__ #ifndef __ASSEMBLY__ diff --git a/include/asm-s390/processor.h b/include/asm-s390/processor.h index 88c272c..fb46e90 100644 --- a/include/asm-s390/processor.h +++ b/include/asm-s390/processor.h @@ -245,7 +245,7 @@ static inline void enabled_wait(void) psw_t wait_psw; wait_psw.mask = PSW_BASE_BITS | PSW_MASK_IO | PSW_MASK_EXT | - PSW_MASK_MCHECK | PSW_MASK_WAIT; + PSW_MASK_MCHECK | PSW_MASK_WAIT | PSW_DEFAULT_KEY; #ifndef __s390x__ asm volatile ( " basr %0,0\n" diff --git a/include/asm-s390/ptrace.h b/include/asm-s390/ptrace.h index 1dc8066..4eff8f2 100644 --- a/include/asm-s390/ptrace.h +++ b/include/asm-s390/ptrace.h @@ -185,6 +185,7 @@ #include #include #include +#include typedef union { @@ -235,6 +236,7 @@ typedef struct #define PSW_ADDR_INSN 0x7FFFFFFFUL #define PSW_BASE_BITS 0x00080000UL +#define PSW_DEFAULT_KEY (((unsigned long) PAGE_DEFAULT_ACC) << 20) #define PSW_ASC_PRIMARY 0x00000000UL #define PSW_ASC_ACCREG 0x00004000UL @@ -260,6 +262,7 @@ typedef struct #define PSW_BASE_BITS 0x0000000180000000UL #define PSW_BASE32_BITS 0x0000000080000000UL +#define PSW_DEFAULT_KEY (((unsigned long) PAGE_DEFAULT_ACC) << 52) #define PSW_ASC_PRIMARY 0x0000000000000000UL #define PSW_ASC_ACCREG 0x0000400000000000UL @@ -268,14 +271,15 @@ typedef struct #define PSW_USER32_BITS (PSW_BASE32_BITS | PSW_MASK_DAT | PSW_ASC_HOME | \ PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK | \ - PSW_MASK_PSTATE) + PSW_MASK_PSTATE | PSW_DEFAULT_KEY) #endif /* __s390x__ */ -#define PSW_KERNEL_BITS (PSW_BASE_BITS | PSW_MASK_DAT | PSW_ASC_PRIMARY) +#define PSW_KERNEL_BITS (PSW_BASE_BITS | PSW_MASK_DAT | PSW_ASC_PRIMARY | \ + PSW_DEFAULT_KEY) #define PSW_USER_BITS (PSW_BASE_BITS | PSW_MASK_DAT | PSW_ASC_HOME | \ PSW_MASK_IO | PSW_MASK_EXT | PSW_MASK_MCHECK | \ - PSW_MASK_PSTATE) + PSW_MASK_PSTATE | PSW_DEFAULT_KEY) /* This macro merges a NEW PSW mask specified by the user into the currently active PSW mask CURRENT, modifying only those @@ -470,6 +474,12 @@ struct user_regs_struct extern void show_regs(struct pt_regs * regs); #endif +static inline void +psw_set_key(unsigned int key) +{ + asm volatile ( "spka 0(%0)" : : "d" (key) ); +} + #endif /* __ASSEMBLY__ */ #endif /* _S390_PTRACE_H */ -- cgit v0.10.2 From 15439d74f621ff8252b53136d259781148e6e768 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Sun, 1 May 2005 08:58:58 -0700 Subject: [PATCH] s390: cmm guest sender id An arbitrary guest must not be allowed to trigger cmm actions. Only one specific guest namely the one that serves as the resource monitor may send cmm messages. Add a parameter that allows to specify the guest that may send messages. z/VMs resource manager has the name 'VMRMSVM' which is the default. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/s390/mm/cmm.c b/arch/s390/mm/cmm.c index d30cdb4..f5a5bc0 100644 --- a/arch/s390/mm/cmm.c +++ b/arch/s390/mm/cmm.c @@ -20,6 +20,11 @@ #include #include +static char *sender = "VMRMSVM"; +module_param(sender, charp, 0); +MODULE_PARM_DESC(sender, + "Guest name that may send SMSG messages (default VMRMSVM)"); + #include "../../../drivers/s390/net/smsgiucv.h" #define CMM_NR_PAGES ((PAGE_SIZE / sizeof(unsigned long)) - 2) @@ -367,10 +372,12 @@ static struct ctl_table cmm_dir_table[] = { #ifdef CONFIG_CMM_IUCV #define SMSG_PREFIX "CMM" static void -cmm_smsg_target(char *msg) +cmm_smsg_target(char *from, char *msg) { long pages, seconds; + if (strlen(sender) > 0 && strcmp(from, sender) != 0) + return; if (!cmm_skip_blanks(msg + strlen(SMSG_PREFIX), &msg)) return; if (strncmp(msg, "SHRINK", 6) == 0) { diff --git a/drivers/s390/net/smsgiucv.c b/drivers/s390/net/smsgiucv.c index a3d2858..1e3f7f3 100644 --- a/drivers/s390/net/smsgiucv.c +++ b/drivers/s390/net/smsgiucv.c @@ -32,7 +32,7 @@ struct smsg_callback { struct list_head list; char *prefix; int len; - void (*callback)(char *str); + void (*callback)(char *from, char *str); }; MODULE_AUTHOR @@ -55,8 +55,9 @@ smsg_message_pending(iucv_MessagePending *eib, void *pgm_data) { struct smsg_callback *cb; unsigned char *msg; + unsigned char sender[9]; unsigned short len; - int rc; + int rc, i; len = eib->ln1msg2.ipbfln1f; msg = kmalloc(len + 1, GFP_ATOMIC|GFP_DMA); @@ -69,10 +70,18 @@ smsg_message_pending(iucv_MessagePending *eib, void *pgm_data) if (rc == 0) { msg[len] = 0; EBCASC(msg, len); + memcpy(sender, msg, 8); + sender[8] = 0; + /* Remove trailing whitespace from the sender name. */ + for (i = 7; i >= 0; i--) { + if (sender[i] != ' ' && sender[i] != '\t') + break; + sender[i] = 0; + } spin_lock(&smsg_list_lock); list_for_each_entry(cb, &smsg_list, list) if (strncmp(msg + 8, cb->prefix, cb->len) == 0) { - cb->callback(msg + 8); + cb->callback(sender, msg + 8); break; } spin_unlock(&smsg_list_lock); @@ -91,7 +100,7 @@ static struct device_driver smsg_driver = { }; int -smsg_register_callback(char *prefix, void (*callback)(char *str)) +smsg_register_callback(char *prefix, void (*callback)(char *from, char *str)) { struct smsg_callback *cb; @@ -108,7 +117,7 @@ smsg_register_callback(char *prefix, void (*callback)(char *str)) } void -smsg_unregister_callback(char *prefix, void (*callback)(char *str)) +smsg_unregister_callback(char *prefix, void (*callback)(char *from, char *str)) { struct smsg_callback *cb, *tmp; diff --git a/drivers/s390/net/smsgiucv.h b/drivers/s390/net/smsgiucv.h index 04cd871..67f5d4f 100644 --- a/drivers/s390/net/smsgiucv.h +++ b/drivers/s390/net/smsgiucv.h @@ -5,6 +5,6 @@ * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com) */ -int smsg_register_callback(char *, void (*)(char *)); -void smsg_unregister_callback(char *, void (*)(char *)); +int smsg_register_callback(char *, void (*)(char *, char *)); +void smsg_unregister_callback(char *, void (*)(char *, char *)); -- cgit v0.10.2 From e8f0641ef74eaa71ed9aa9d19c4b741c2143d752 Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Sun, 1 May 2005 08:58:58 -0700 Subject: [PATCH] s390: allow longer debug feature names The current limitation of 16 characters of the debug feature names turned out to be insufficient. Increase it to 64 characters. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-s390/debug.h b/include/asm-s390/debug.h index 28ef235..6bbcdea 100644 --- a/include/asm-s390/debug.h +++ b/include/asm-s390/debug.h @@ -43,7 +43,7 @@ struct __debug_entry{ #define DEBUG_OFF_LEVEL -1 /* level where debug is switched off */ #define DEBUG_FLUSH_ALL -1 /* parameter to flush all areas */ #define DEBUG_MAX_VIEWS 10 /* max number of views in proc fs */ -#define DEBUG_MAX_PROCF_LEN 16 /* max length for a proc file name */ +#define DEBUG_MAX_PROCF_LEN 64 /* max length for a proc file name */ #define DEBUG_DEFAULT_LEVEL 3 /* initial debug level */ #define DEBUG_DIR_ROOT "s390dbf" /* name of debug root directory in proc fs */ -- cgit v0.10.2 From f24acd4503270ed4c842c8fef0b71105285e0a06 Mon Sep 17 00:00:00 2001 From: Horst Hummel Date: Sun, 1 May 2005 08:58:59 -0700 Subject: [PATCH] s390: dasd readonly attribute The independent read-only flags in devmap, dasd_device and gendisk are not kept in sync. Use one bit per feature in the dasd driver and keep that bit in sync with the gendisk bit. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c index b755bac..826fd23 100644 --- a/drivers/s390/block/dasd.c +++ b/drivers/s390/block/dasd.c @@ -7,7 +7,7 @@ * Bugreports.to..: * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 1999-2001 * - * $Revision: 1.158 $ + * $Revision: 1.161 $ */ #include @@ -1131,13 +1131,17 @@ __dasd_process_blk_queue(struct dasd_device * device) request_queue_t *queue; struct request *req; struct dasd_ccw_req *cqr; - int nr_queued; + int nr_queued, feature_ro; queue = device->request_queue; /* No queue ? Then there is nothing to do. */ if (queue == NULL) return; + feature_ro = dasd_get_feature(device->cdev, DASD_FEATURE_READONLY); + if (feature_ro < 0) /* no devmap */ + return; + /* * We requeue request from the block device queue to the ccw * queue only in two states. In state DASD_STATE_READY the @@ -1157,8 +1161,8 @@ __dasd_process_blk_queue(struct dasd_device * device) elv_next_request(queue) && nr_queued < DASD_CHANQ_MAX_SIZE) { req = elv_next_request(queue); - if (test_bit(DASD_FLAG_RO, &device->flags) && - rq_data_dir(req) == WRITE) { + + if (feature_ro && rq_data_dir(req) == WRITE) { DBF_DEV_EVENT(DBF_ERR, device, "Rejecting write request %p", req); @@ -1803,13 +1807,17 @@ dasd_generic_set_online (struct ccw_device *cdev, { struct dasd_device *device; - int rc; + int feature_diag, rc; + + feature_diag = dasd_get_feature(cdev, DASD_FEATURE_USEDIAG); + if (feature_diag < 0) + return feature_diag; device = dasd_create_device(cdev); if (IS_ERR(device)) return PTR_ERR(device); - if (test_bit(DASD_FLAG_USE_DIAG, &device->flags)) { + if (feature_diag) { if (!dasd_diag_discipline_pointer) { printk (KERN_WARNING "dasd_generic couldn't online device %s " diff --git a/drivers/s390/block/dasd_devmap.c b/drivers/s390/block/dasd_devmap.c index ad1841a..1aedc48 100644 --- a/drivers/s390/block/dasd_devmap.c +++ b/drivers/s390/block/dasd_devmap.c @@ -11,7 +11,7 @@ * functions may not be called from interrupt context. In particular * dasd_get_device is a no-no from interrupt context. * - * $Revision: 1.37 $ + * $Revision: 1.40 $ */ #include @@ -513,14 +513,6 @@ dasd_create_device(struct ccw_device *cdev) if (!devmap->device) { devmap->device = device; device->devindex = devmap->devindex; - if (devmap->features & DASD_FEATURE_READONLY) - set_bit(DASD_FLAG_RO, &device->flags); - else - clear_bit(DASD_FLAG_RO, &device->flags); - if (devmap->features & DASD_FEATURE_USEDIAG) - set_bit(DASD_FLAG_USE_DIAG, &device->flags); - else - clear_bit(DASD_FLAG_USE_DIAG, &device->flags); get_device(&cdev->dev); device->cdev = cdev; rc = 0; @@ -651,14 +643,8 @@ dasd_ro_store(struct device *dev, const char *buf, size_t count) devmap->features |= DASD_FEATURE_READONLY; else devmap->features &= ~DASD_FEATURE_READONLY; - if (devmap->device) { - if (devmap->device->gdp) - set_disk_ro(devmap->device->gdp, ro_flag); - if (ro_flag) - set_bit(DASD_FLAG_RO, &devmap->device->flags); - else - clear_bit(DASD_FLAG_RO, &devmap->device->flags); - } + if (devmap->device && devmap->device->gdp) + set_disk_ro(devmap->device->gdp, ro_flag); spin_unlock(&dasd_devmap_lock); return count; } @@ -739,6 +725,45 @@ static struct attribute_group dasd_attr_group = { .attrs = dasd_attrs, }; +/* + * Return value of the specified feature. + */ +int +dasd_get_feature(struct ccw_device *cdev, int feature) +{ + struct dasd_devmap *devmap; + + devmap = dasd_find_busid(cdev->dev.bus_id); + if (IS_ERR(devmap)) + return (int) PTR_ERR(devmap); + + return ((devmap->features & feature) != 0); +} + +/* + * Set / reset given feature. + * Flag indicates wether to set (!=0) or the reset (=0) the feature. + */ +int +dasd_set_feature(struct ccw_device *cdev, int feature, int flag) +{ + struct dasd_devmap *devmap; + + devmap = dasd_find_busid(cdev->dev.bus_id); + if (IS_ERR(devmap)) + return (int) PTR_ERR(devmap); + + spin_lock(&dasd_devmap_lock); + if (flag) + devmap->features |= feature; + else + devmap->features &= ~feature; + + spin_unlock(&dasd_devmap_lock); + return 0; +} + + int dasd_add_sysfs_files(struct ccw_device *cdev) { diff --git a/drivers/s390/block/dasd_genhd.c b/drivers/s390/block/dasd_genhd.c index 1d52db4..96c4934 100644 --- a/drivers/s390/block/dasd_genhd.c +++ b/drivers/s390/block/dasd_genhd.c @@ -9,7 +9,7 @@ * * gendisk related functions for the dasd driver. * - * $Revision: 1.48 $ + * $Revision: 1.50 $ */ #include @@ -31,12 +31,16 @@ int dasd_gendisk_alloc(struct dasd_device *device) { struct gendisk *gdp; - int len; + int len, feature_ro; /* Make sure the minor for this device exists. */ if (device->devindex >= DASD_PER_MAJOR) return -EBUSY; + feature_ro = dasd_get_feature(device->cdev, DASD_FEATURE_READONLY); + if (feature_ro < 0) + return feature_ro; + gdp = alloc_disk(1 << DASD_PARTN_BITS); if (!gdp) return -ENOMEM; @@ -71,7 +75,7 @@ dasd_gendisk_alloc(struct dasd_device *device) sprintf(gdp->devfs_name, "dasd/%s", device->cdev->dev.bus_id); - if (test_bit(DASD_FLAG_RO, &device->flags)) + if (feature_ro) set_disk_ro(gdp, 1); gdp->private_data = device; gdp->queue = device->request_queue; diff --git a/drivers/s390/block/dasd_int.h b/drivers/s390/block/dasd_int.h index 4586e0e..a9f38b2 100644 --- a/drivers/s390/block/dasd_int.h +++ b/drivers/s390/block/dasd_int.h @@ -6,7 +6,7 @@ * Bugreports.to..: * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 1999,2000 * - * $Revision: 1.63 $ + * $Revision: 1.64 $ */ #ifndef DASD_INT_H @@ -329,8 +329,6 @@ struct dasd_device { #define DASD_STOPPED_DC_EIO 16 /* disconnected, return -EIO */ /* per device flags */ -#define DASD_FLAG_RO 0 /* device is read-only */ -#define DASD_FLAG_USE_DIAG 1 /* use diag disciplnie */ #define DASD_FLAG_DSC_ERROR 2 /* return -EIO when disconnected */ #define DASD_FLAG_OFFLINE 3 /* device is in offline processing */ @@ -501,6 +499,9 @@ void dasd_devmap_exit(void); struct dasd_device *dasd_create_device(struct ccw_device *); void dasd_delete_device(struct dasd_device *); +int dasd_get_feature(struct ccw_device *, int); +int dasd_set_feature(struct ccw_device *, int, int); + int dasd_add_sysfs_files(struct ccw_device *); void dasd_remove_sysfs_files(struct ccw_device *); diff --git a/drivers/s390/block/dasd_ioctl.c b/drivers/s390/block/dasd_ioctl.c index f1892ba..980c555 100644 --- a/drivers/s390/block/dasd_ioctl.c +++ b/drivers/s390/block/dasd_ioctl.c @@ -7,6 +7,8 @@ * Bugreports.to..: * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 1999-2001 * + * $Revision: 1.45 $ + * * i/o controls for the dasd driver. */ #include @@ -294,6 +296,7 @@ dasd_ioctl_format(struct block_device *bdev, int no, long args) { struct dasd_device *device; struct format_data_t fdata; + int feature_ro; if (!capable(CAP_SYS_ADMIN)) return -EACCES; @@ -304,7 +307,11 @@ dasd_ioctl_format(struct block_device *bdev, int no, long args) if (device == NULL) return -ENODEV; - if (test_bit(DASD_FLAG_RO, &device->flags)) + + feature_ro = dasd_get_feature(device->cdev, DASD_FEATURE_READONLY); + if (feature_ro < 0) + return feature_ro; + if (feature_ro) return -EROFS; if (copy_from_user(&fdata, (void __user *) args, sizeof (struct format_data_t))) @@ -377,7 +384,7 @@ dasd_ioctl_information(struct block_device *bdev, int no, long args) struct dasd_device *device; struct dasd_information2_t *dasd_info; unsigned long flags; - int rc; + int rc, feature_ro; struct ccw_device *cdev; device = bdev->bd_disk->private_data; @@ -387,6 +394,10 @@ dasd_ioctl_information(struct block_device *bdev, int no, long args) if (!device->discipline->fill_info) return -EINVAL; + feature_ro = dasd_get_feature(device->cdev, DASD_FEATURE_READONLY); + if (feature_ro < 0) + return feature_ro; + dasd_info = kmalloc(sizeof(struct dasd_information2_t), GFP_KERNEL); if (dasd_info == NULL) return -ENOMEM; @@ -415,9 +426,8 @@ dasd_ioctl_information(struct block_device *bdev, int no, long args) if ((device->state < DASD_STATE_READY) || (dasd_check_blocksize(device->bp_block))) dasd_info->format = DASD_FORMAT_NONE; - - dasd_info->features |= test_bit(DASD_FLAG_RO, &device->flags) ? - DASD_FEATURE_READONLY : DASD_FEATURE_DEFAULT; + + dasd_info->features |= feature_ro; if (device->discipline) memcpy(dasd_info->type, device->discipline->name, 4); @@ -460,7 +470,7 @@ static int dasd_ioctl_set_ro(struct block_device *bdev, int no, long args) { struct dasd_device *device; - int intval; + int intval, rc; if (!capable(CAP_SYS_ADMIN)) return -EACCES; @@ -472,12 +482,11 @@ dasd_ioctl_set_ro(struct block_device *bdev, int no, long args) device = bdev->bd_disk->private_data; if (device == NULL) return -ENODEV; + set_disk_ro(bdev->bd_disk, intval); - if (intval) - set_bit(DASD_FLAG_RO, &device->flags); - else - clear_bit(DASD_FLAG_RO, &device->flags); - return 0; + rc = dasd_set_feature(device->cdev, DASD_FEATURE_READONLY, intval); + + return rc; } /* diff --git a/drivers/s390/block/dasd_proc.c b/drivers/s390/block/dasd_proc.c index 353d411..d7f1974 100644 --- a/drivers/s390/block/dasd_proc.c +++ b/drivers/s390/block/dasd_proc.c @@ -9,7 +9,7 @@ * * /proc interface for the dasd driver. * - * $Revision: 1.30 $ + * $Revision: 1.31 $ */ #include @@ -54,6 +54,7 @@ dasd_devices_show(struct seq_file *m, void *v) { struct dasd_device *device; char *substr; + int feature; device = dasd_device_from_devindex((unsigned long) v - 1); if (IS_ERR(device)) @@ -77,7 +78,10 @@ dasd_devices_show(struct seq_file *m, void *v) else seq_printf(m, " is ????????"); /* Print devices features. */ - substr = test_bit(DASD_FLAG_RO, &device->flags) ? "(ro)" : " "; + feature = dasd_get_feature(device->cdev, DASD_FEATURE_READONLY); + if (feature < 0) + return 0; + substr = feature ? "(ro)" : " "; seq_printf(m, "%4s: ", substr); /* Print device status information. */ switch ((device != NULL) ? device->state : -1) { -- cgit v0.10.2 From 6ed93c827eea181d87ce078cd1d671128c030b5a Mon Sep 17 00:00:00 2001 From: Stefan Weinhuber Date: Sun, 1 May 2005 08:58:59 -0700 Subject: [PATCH] s390: enable write barriers in the dasd driver The DASD device driver never reorders the I/O requests and relies on the hardware to write all data to nonvolatile storage before signaling a successful write. Hence, the only thing we have to do to support write barriers is to set the queue ordered flag. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c index 826fd23..02cfe24 100644 --- a/drivers/s390/block/dasd.c +++ b/drivers/s390/block/dasd.c @@ -1635,6 +1635,7 @@ dasd_setup_queue(struct dasd_device * device) blk_queue_max_hw_segments(device->request_queue, -1L); blk_queue_max_segment_size(device->request_queue, -1L); blk_queue_segment_boundary(device->request_queue, -1L); + blk_queue_ordered(device->request_queue, 1); } /* -- cgit v0.10.2 From ec5883abebb2e249ea8d318cb58fb4b2c269cf10 Mon Sep 17 00:00:00 2001 From: Horst Hummel Date: Sun, 1 May 2005 08:58:59 -0700 Subject: [PATCH] s390: don't pad cdl blocks for write requests The first blocks on a cdl formatted dasd device are smaller than the blocksize of the device. Read requests are padded with a 'e5' pattern. Write requests should not pad the (user) buffer with 'e5' because a write request is not allowed to modify the buffer. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/s390/block/dasd_eckd.c b/drivers/s390/block/dasd_eckd.c index 838aedf..811060e 100644 --- a/drivers/s390/block/dasd_eckd.c +++ b/drivers/s390/block/dasd_eckd.c @@ -7,7 +7,7 @@ * Bugreports.to..: * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 1999,2000 * - * $Revision: 1.69 $ + * $Revision: 1.71 $ */ #include @@ -1101,7 +1101,8 @@ dasd_eckd_build_cp(struct dasd_device * device, struct request *req) if (dasd_eckd_cdl_special(blk_per_trk, recid)){ rcmd |= 0x8; count = dasd_eckd_cdl_reclen(recid); - if (count < blksize) + if (count < blksize && + rq_data_dir(req) == READ) memset(dst + count, 0xe5, blksize - count); } -- cgit v0.10.2 From 4beb37097b20b61054b15c56848e4ffcef093819 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Sun, 1 May 2005 08:58:59 -0700 Subject: [PATCH] s390: remove ioctl32 from dasdcmb The ioctl32_conversion routines will be deprecated: Remove them from dasd_cmb and handle the three cmb ioctls like all other dasd ioctls. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/s390/kernel/compat_ioctl.c b/arch/s390/kernel/compat_ioctl.c index 96571ff..03d03c6 100644 --- a/arch/s390/kernel/compat_ioctl.c +++ b/arch/s390/kernel/compat_ioctl.c @@ -16,6 +16,7 @@ #define CODE #include "../../../fs/compat_ioctl.c" #include +#include #include static int do_ioctl32_pointer(unsigned int fd, unsigned int cmd, @@ -58,7 +59,11 @@ COMPATIBLE_IOCTL(BIODASDPRRD) COMPATIBLE_IOCTL(BIODASDPSRD) COMPATIBLE_IOCTL(BIODASDGATTR) COMPATIBLE_IOCTL(BIODASDSATTR) - +#if defined(CONFIG_DASD_CMB) || defined(CONFIG_DASD_CMB_MODULE) +COMPATIBLE_IOCTL(BIODASDCMFENABLE) +COMPATIBLE_IOCTL(BIODASDCMFDISABLE) +COMPATIBLE_IOCTL(BIODASDREADALLCMB) +#endif #endif #if defined(CONFIG_S390_TAPE) || defined(CONFIG_S390_TAPE_MODULE) diff --git a/drivers/s390/block/dasd_cmb.c b/drivers/s390/block/dasd_cmb.c index ed1ab47..4f365bf 100644 --- a/drivers/s390/block/dasd_cmb.c +++ b/drivers/s390/block/dasd_cmb.c @@ -1,5 +1,5 @@ /* - * linux/drivers/s390/block/dasd_cmb.c ($Revision: 1.6 $) + * linux/drivers/s390/block/dasd_cmb.c ($Revision: 1.9 $) * * Linux on zSeries Channel Measurement Facility support * (dasd device driver interface) @@ -23,7 +23,6 @@ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ #include -#include #include #include #include @@ -84,27 +83,13 @@ dasd_ioctl_readall_cmb(struct block_device *bdev, int no, long args) static inline int ioctl_reg(unsigned int no, dasd_ioctl_fn_t handler) { - int ret; - ret = dasd_ioctl_no_register(THIS_MODULE, no, handler); -#ifdef CONFIG_COMPAT - if (ret) - return ret; - - ret = register_ioctl32_conversion(no, NULL); - if (ret) - dasd_ioctl_no_unregister(THIS_MODULE, no, handler); -#endif - return ret; + return dasd_ioctl_no_register(THIS_MODULE, no, handler); } static inline void ioctl_unreg(unsigned int no, dasd_ioctl_fn_t handler) { dasd_ioctl_no_unregister(THIS_MODULE, no, handler); -#ifdef CONFIG_COMPAT - unregister_ioctl32_conversion(no); -#endif - } static void diff --git a/include/asm-s390/cmb.h b/include/asm-s390/cmb.h index 1bfe2bd..dae1dd4 100644 --- a/include/asm-s390/cmb.h +++ b/include/asm-s390/cmb.h @@ -52,7 +52,7 @@ struct cmbdata { #define BIODASDREADALLCMB _IOWR(DASD_IOCTL_LETTER,33,struct cmbdata) #ifdef __KERNEL__ - +struct ccw_device; /** * enable_cmf() - switch on the channel measurement for a specific device * @cdev: The ccw device to be enabled -- cgit v0.10.2 From af6c8eed14a0c14791e2fbb4c7726755c1a637ca Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Sun, 1 May 2005 08:59:00 -0700 Subject: [PATCH] s390: remove ioctl32 from crypto driver The ioctl32_conversion routines will be deprecated: Remove them from the crypto driver. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/s390/crypto/z90main.c b/drivers/s390/crypto/z90main.c index a98c00c..9ec29bb 100644 --- a/drivers/s390/crypto/z90main.c +++ b/drivers/s390/crypto/z90main.c @@ -385,8 +385,8 @@ static int z90crypt_release(struct inode *, struct file *); static ssize_t z90crypt_read(struct file *, char __user *, size_t, loff_t *); static ssize_t z90crypt_write(struct file *, const char __user *, size_t, loff_t *); -static int z90crypt_ioctl(struct inode *, struct file *, - unsigned int, unsigned long); +static long z90crypt_unlocked_ioctl(struct file *, unsigned int, unsigned long); +static long z90crypt_compat_ioctl(struct file *, unsigned int, unsigned long); static void z90crypt_reader_task(unsigned long); static void z90crypt_schedule_reader_task(unsigned long); @@ -433,12 +433,15 @@ static atomic_t total_open; static atomic_t z90crypt_step; static struct file_operations z90crypt_fops = { - .owner = THIS_MODULE, - .read = z90crypt_read, - .write = z90crypt_write, - .ioctl = z90crypt_ioctl, - .open = z90crypt_open, - .release = z90crypt_release + .owner = THIS_MODULE, + .read = z90crypt_read, + .write = z90crypt_write, + .unlocked_ioctl = z90crypt_unlocked_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = z90crypt_compat_ioctl, +#endif + .open = z90crypt_open, + .release = z90crypt_release }; #ifndef Z90CRYPT_USE_HOTPLUG @@ -474,14 +477,13 @@ struct ica_rsa_modexpo_32 { // For 32-bit callers compat_uptr_t n_modulus; }; -static int -trans_modexpo32(unsigned int fd, unsigned int cmd, unsigned long arg, - struct file *file) +static long +trans_modexpo32(struct file *filp, unsigned int cmd, unsigned long arg) { struct ica_rsa_modexpo_32 __user *mex32u = compat_ptr(arg); struct ica_rsa_modexpo_32 mex32k; struct ica_rsa_modexpo __user *mex64; - int ret = 0; + long ret = 0; unsigned int i; if (!access_ok(VERIFY_WRITE, mex32u, sizeof(struct ica_rsa_modexpo_32))) @@ -498,7 +500,7 @@ trans_modexpo32(unsigned int fd, unsigned int cmd, unsigned long arg, __put_user(compat_ptr(mex32k.b_key), &mex64->b_key) || __put_user(compat_ptr(mex32k.n_modulus), &mex64->n_modulus)) return -EFAULT; - ret = sys_ioctl(fd, cmd, (unsigned long)mex64); + ret = z90crypt_unlocked_ioctl(filp, cmd, (unsigned long)mex64); if (!ret) if (__get_user(i, &mex64->outputdatalength) || __put_user(i, &mex32u->outputdatalength)) @@ -518,14 +520,13 @@ struct ica_rsa_modexpo_crt_32 { // For 32-bit callers compat_uptr_t u_mult_inv; }; -static int -trans_modexpo_crt32(unsigned int fd, unsigned int cmd, unsigned long arg, - struct file *file) +static long +trans_modexpo_crt32(struct file *filp, unsigned int cmd, unsigned long arg) { struct ica_rsa_modexpo_crt_32 __user *crt32u = compat_ptr(arg); struct ica_rsa_modexpo_crt_32 crt32k; struct ica_rsa_modexpo_crt __user *crt64; - int ret = 0; + long ret = 0; unsigned int i; if (!access_ok(VERIFY_WRITE, crt32u, @@ -546,9 +547,8 @@ trans_modexpo_crt32(unsigned int fd, unsigned int cmd, unsigned long arg, __put_user(compat_ptr(crt32k.np_prime), &crt64->np_prime) || __put_user(compat_ptr(crt32k.nq_prime), &crt64->nq_prime) || __put_user(compat_ptr(crt32k.u_mult_inv), &crt64->u_mult_inv)) - ret = -EFAULT; - if (!ret) - ret = sys_ioctl(fd, cmd, (unsigned long)crt64); + return -EFAULT; + ret = z90crypt_unlocked_ioctl(filp, cmd, (unsigned long)crt64); if (!ret) if (__get_user(i, &crt64->outputdatalength) || __put_user(i, &crt32u->outputdatalength)) @@ -556,66 +556,34 @@ trans_modexpo_crt32(unsigned int fd, unsigned int cmd, unsigned long arg, return ret; } -static int compatible_ioctls[] = { - ICAZ90STATUS, Z90QUIESCE, Z90STAT_TOTALCOUNT, Z90STAT_PCICACOUNT, - Z90STAT_PCICCCOUNT, Z90STAT_PCIXCCCOUNT, Z90STAT_PCIXCCMCL2COUNT, - Z90STAT_PCIXCCMCL3COUNT, Z90STAT_CEX2CCOUNT, Z90STAT_REQUESTQ_COUNT, - Z90STAT_PENDINGQ_COUNT, Z90STAT_TOTALOPEN_COUNT, Z90STAT_DOMAIN_INDEX, - Z90STAT_STATUS_MASK, Z90STAT_QDEPTH_MASK, Z90STAT_PERDEV_REQCNT, -}; - -static void z90_unregister_ioctl32s(void) -{ - int i; - - unregister_ioctl32_conversion(ICARSAMODEXPO); - unregister_ioctl32_conversion(ICARSACRT); - - for(i = 0; i < ARRAY_SIZE(compatible_ioctls); i++) - unregister_ioctl32_conversion(compatible_ioctls[i]); -} - -static int z90_register_ioctl32s(void) -{ - int result, i; - - result = register_ioctl32_conversion(ICARSAMODEXPO, trans_modexpo32); - if (result == -EBUSY) { - unregister_ioctl32_conversion(ICARSAMODEXPO); - result = register_ioctl32_conversion(ICARSAMODEXPO, - trans_modexpo32); - } - if (result) - return result; - result = register_ioctl32_conversion(ICARSACRT, trans_modexpo_crt32); - if (result == -EBUSY) { - unregister_ioctl32_conversion(ICARSACRT); - result = register_ioctl32_conversion(ICARSACRT, - trans_modexpo_crt32); - } - if (result) - return result; - - for(i = 0; i < ARRAY_SIZE(compatible_ioctls); i++) { - result = register_ioctl32_conversion(compatible_ioctls[i], 0); - if (result == -EBUSY) { - unregister_ioctl32_conversion(compatible_ioctls[i]); - result = register_ioctl32_conversion( - compatible_ioctls[i], 0); - } - if (result) - return result; - } - return 0; -} -#else // !CONFIG_COMPAT -static inline void z90_unregister_ioctl32s(void) -{ -} - -static inline int z90_register_ioctl32s(void) +static long +z90crypt_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { - return 0; + switch (cmd) { + case ICAZ90STATUS: + case Z90QUIESCE: + case Z90STAT_TOTALCOUNT: + case Z90STAT_PCICACOUNT: + case Z90STAT_PCICCCOUNT: + case Z90STAT_PCIXCCCOUNT: + case Z90STAT_PCIXCCMCL2COUNT: + case Z90STAT_PCIXCCMCL3COUNT: + case Z90STAT_CEX2CCOUNT: + case Z90STAT_REQUESTQ_COUNT: + case Z90STAT_PENDINGQ_COUNT: + case Z90STAT_TOTALOPEN_COUNT: + case Z90STAT_DOMAIN_INDEX: + case Z90STAT_STATUS_MASK: + case Z90STAT_QDEPTH_MASK: + case Z90STAT_PERDEV_REQCNT: + return z90crypt_unlocked_ioctl(filp, cmd, arg); + case ICARSAMODEXPO: + return trans_modexpo32(filp, cmd, arg); + case ICARSACRT: + return trans_modexpo_crt32(filp, cmd, arg); + default: + return -ENOIOCTLCMD; + } } #endif @@ -730,14 +698,9 @@ z90crypt_init_module(void) reader_timer.expires = jiffies + (READERTIME * HZ / 1000); add_timer(&reader_timer); - if ((result = z90_register_ioctl32s())) - goto init_module_cleanup; - return 0; // success init_module_cleanup: - z90_unregister_ioctl32s(); - #ifndef Z90CRYPT_USE_HOTPLUG if ((nresult = misc_deregister(&z90crypt_misc_device))) PRINTK("misc_deregister failed with %d.\n", nresult); @@ -763,8 +726,6 @@ z90crypt_cleanup_module(void) PDEBUG("PID %d\n", PID()); - z90_unregister_ioctl32s(); - remove_proc_entry("driver/z90crypt", 0); #ifndef Z90CRYPT_USE_HOTPLUG @@ -800,7 +761,7 @@ z90crypt_cleanup_module(void) * z90crypt_release * z90crypt_read * z90crypt_write - * z90crypt_ioctl + * z90crypt_unlocked_ioctl * z90crypt_status * z90crypt_status_write * disable_card @@ -1804,9 +1765,8 @@ z90crypt_rsa(struct priv_data *private_data_p, pid_t pid, * This function is a little long, but it's really just one large switch * statement. */ -static int -z90crypt_ioctl(struct inode *inode, struct file *filp, - unsigned int cmd, unsigned long arg) +static long +z90crypt_unlocked_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { struct priv_data *private_data_p = filp->private_data; unsigned char *status; -- cgit v0.10.2 From 9fc1427a01a9df3605e219c6de0c59c4639209a1 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Sun, 1 May 2005 08:59:00 -0700 Subject: [PATCH] s390: cio documentation Synchronize documentation with current interface. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/s390/cds.txt b/Documentation/s390/cds.txt index d939717..f0be389 100644 --- a/Documentation/s390/cds.txt +++ b/Documentation/s390/cds.txt @@ -56,12 +56,16 @@ read_dev_chars() read device characteristics read_conf_data() +read_conf_data_lpm() read configuration data. ccw_device_get_ciw() get commands from extended sense data. ccw_device_start() +ccw_device_start_timeout() +ccw_device_start_key() +ccw_device_start_key_timeout() initiate an I/O request. ccw_device_resume() @@ -197,19 +201,21 @@ The read_dev_chars() function returns : operational. -read_conf_data() - Read Configuration Data +read_conf_data(), read_conf_data_lpm() - Read Configuration Data Retrieve the device dependent configuration data. Please have a look at your device dependent I/O commands for the device specific layout of the node -descriptor elements. +descriptor elements. read_conf_data_lpm() will retrieve the configuration data +for a specific path. -The function is meant to be called with an irq handler in place; that is, +The function is meant to be called with the device already enabled; that is, at earliest during set_online() processing. The function may be called enabled or disabled, but the device must not be locked -int read_conf_data(struct ccw_device, void **buffer, int *length, __u8 lpm); +int read_conf_data(struct ccw_device, void **buffer, int *length); +int read_conf_data_lpm(struct ccw_device, void **buffer, int *length, __u8 lpm); cdev - the ccw_device the data is requested for. buffer - Pointer to a buffer pointer. The read_conf_data() routine @@ -263,6 +269,25 @@ int ccw_device_start(struct ccw_device *cdev, unsigned long intparm, __u8 lpm, unsigned long flags); +int ccw_device_start_timeout(struct ccw_device *cdev, + struct ccw1 *cpa, + unsigned long intparm, + __u8 lpm, + unsigned long flags, + int expires); +int ccw_device_start_key(struct ccw_device *cdev, + struct ccw1 *cpa, + unsigned long intparm, + __u8 lpm, + __u8 key, + unsigned long flags); +int ccw_device_start_key_timeout(struct ccw_device *cdev, + struct ccw1 *cpa, + unsigned long intparm, + __u8 lpm, + __u8 key, + unsigned long flags, + int expires); cdev : ccw_device the I/O is destined for cpa : logical start address of channel program @@ -272,7 +297,12 @@ user_intparm : user specific interrupt information; will be presented particular I/O request. lpm : defines the channel path to be used for a specific I/O request. A value of 0 will make cio use the opm. +key : the storage key to use for the I/O (useful for operating on a + storage with a storage key != default key) flag : defines the action to be performed for I/O processing +expires : timeout value in jiffies. The common I/O layer will terminate + the running program after this and call the interrupt handler + with ERR_PTR(-ETIMEDOUT) as irb. Possible flag values are : @@ -327,6 +357,13 @@ current (last) I/O request. In case of a delayed status notification no special interrupt will be presented to indicate I/O completion as the I/O request was never started, even though ccw_device_start() returned with successful completion. +The irb may contain an error value, and the device driver should check for this +first: + +-ETIMEDOUT: the common I/O layer terminated the request after the specified + timeout value +-EIO: the common I/O layer terminated the request due to an error state + If the concurrent sense flag in the extended status word in the irb is set, the field irb->scsw.count describes the numer of device specific sense bytes available in the extended control word irb->scsw.ecw[0]. No device sensing by -- cgit v0.10.2 From e43379f10b42194b8a6e1de342cfb44463c0f6da Mon Sep 17 00:00:00 2001 From: Matt Mackall Date: Sun, 1 May 2005 08:59:00 -0700 Subject: [PATCH] nice and rt-prio rlimits Add a pair of rlimits for allowing non-root tasks to raise nice and rt priorities. Defaults to traditional behavior. Originally written by Chris Wright. The patch implements a simple rlimit ceiling for the RT (and nice) priorities a task can set. The rlimit defaults to 0, meaning no change in behavior by default. A value of 50 means RT priority levels 1-50 are allowed. A value of 100 means all 99 privilege levels from 1 to 99 are allowed. CAP_SYS_NICE is blanket permission. (akpm: see http://www.uwsg.iu.edu/hypermail/linux/kernel/0503.1/1921.html for tips on integrating this with PAM). Signed-off-by: Matt Mackall Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-generic/resource.h b/include/asm-generic/resource.h index b1fcda9..cfe3692 100644 --- a/include/asm-generic/resource.h +++ b/include/asm-generic/resource.h @@ -41,8 +41,11 @@ #define RLIMIT_LOCKS 10 /* maximum file locks held */ #define RLIMIT_SIGPENDING 11 /* max number of pending signals */ #define RLIMIT_MSGQUEUE 12 /* maximum bytes in POSIX mqueues */ +#define RLIMIT_NICE 13 /* max nice prio allowed to raise to + 0-39 for nice level 19 .. -20 */ +#define RLIMIT_RTPRIO 14 /* maximum realtime priority */ -#define RLIM_NLIMITS 13 +#define RLIM_NLIMITS 15 /* * SuS says limits have to be unsigned. @@ -81,6 +84,8 @@ [RLIMIT_LOCKS] = { RLIM_INFINITY, RLIM_INFINITY }, \ [RLIMIT_SIGPENDING] = { 0, 0 }, \ [RLIMIT_MSGQUEUE] = { MQ_BYTES_MAX, MQ_BYTES_MAX }, \ + [RLIMIT_NICE] = { 0, 0 }, \ + [RLIMIT_RTPRIO] = { 0, 0 }, \ } #endif /* __KERNEL__ */ diff --git a/include/linux/sched.h b/include/linux/sched.h index 1cced97..8960f99 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -845,6 +845,7 @@ extern void sched_idle_next(void); extern void set_user_nice(task_t *p, long nice); extern int task_prio(const task_t *p); extern int task_nice(const task_t *p); +extern int can_nice(const task_t *p, const int nice); extern int task_curr(const task_t *p); extern int idle_cpu(int cpu); extern int sched_setscheduler(struct task_struct *, int, struct sched_param *); diff --git a/kernel/sched.c b/kernel/sched.c index 9bb7489..5dadcc6 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -3223,6 +3223,19 @@ out_unlock: EXPORT_SYMBOL(set_user_nice); +/* + * can_nice - check if a task can reduce its nice value + * @p: task + * @nice: nice value + */ +int can_nice(const task_t *p, const int nice) +{ + /* convert nice value [19,-20] to rlimit style value [0,39] */ + int nice_rlim = 19 - nice; + return (nice_rlim <= p->signal->rlim[RLIMIT_NICE].rlim_cur || + capable(CAP_SYS_NICE)); +} + #ifdef __ARCH_WANT_SYS_NICE /* @@ -3242,12 +3255,8 @@ asmlinkage long sys_nice(int increment) * We don't have to worry. Conceptually one call occurs first * and we have a single winner. */ - if (increment < 0) { - if (!capable(CAP_SYS_NICE)) - return -EPERM; - if (increment < -40) - increment = -40; - } + if (increment < -40) + increment = -40; if (increment > 40) increment = 40; @@ -3257,6 +3266,9 @@ asmlinkage long sys_nice(int increment) if (nice > 19) nice = 19; + if (increment < 0 && !can_nice(current, nice)) + return -EPERM; + retval = security_task_setnice(current, nice); if (retval) return retval; @@ -3372,6 +3384,7 @@ recheck: return -EINVAL; if ((policy == SCHED_FIFO || policy == SCHED_RR) && + param->sched_priority > p->signal->rlim[RLIMIT_RTPRIO].rlim_cur && !capable(CAP_SYS_NICE)) return -EPERM; if ((current->euid != p->euid) && (current->euid != p->uid) && diff --git a/kernel/sys.c b/kernel/sys.c index df2ddcc..7f43d6e 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -227,7 +227,7 @@ static int set_one_prio(struct task_struct *p, int niceval, int error) error = -EPERM; goto out; } - if (niceval < task_nice(p) && !capable(CAP_SYS_NICE)) { + if (niceval < task_nice(p) && !can_nice(p, niceval)) { error = -EACCES; goto out; } -- cgit v0.10.2 From c8538a7aa5527d02c7191ac5da124efadf6a2827 Mon Sep 17 00:00:00 2001 From: Matt Mackall Date: Sun, 1 May 2005 08:59:01 -0700 Subject: [PATCH] remove all kernel BUGs This patch eliminates all kernel BUGs, trims about 35k off the typical kernel, and makes the system slightly faster. Signed-off-by: Matt Mackall Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-alpha/bug.h b/include/asm-alpha/bug.h index ae1e0a5..39a3e2a 100644 --- a/include/asm-alpha/bug.h +++ b/include/asm-alpha/bug.h @@ -1,6 +1,7 @@ #ifndef _ALPHA_BUG_H #define _ALPHA_BUG_H +#ifdef CONFIG_BUG #include /* ??? Would be nice to use .gprel32 here, but we can't be sure that the @@ -10,6 +11,8 @@ : : "i" (PAL_bugchk), "i"(__LINE__), "i"(__FILE__)) #define HAVE_ARCH_BUG +#endif + #include #endif diff --git a/include/asm-arm/bug.h b/include/asm-arm/bug.h index 5e91b90..24d1167 100644 --- a/include/asm-arm/bug.h +++ b/include/asm-arm/bug.h @@ -3,6 +3,7 @@ #include +#ifdef CONFIG_BUG #ifdef CONFIG_DEBUG_BUGVERBOSE extern volatile void __bug(const char *file, int line, void *data); @@ -17,6 +18,8 @@ extern volatile void __bug(const char *file, int line, void *data); #endif #define HAVE_ARCH_BUG +#endif + #include #endif diff --git a/include/asm-arm26/bug.h b/include/asm-arm26/bug.h index 920b705..7177c73 100644 --- a/include/asm-arm26/bug.h +++ b/include/asm-arm26/bug.h @@ -3,6 +3,7 @@ #include +#ifdef CONFIG_BUG #ifdef CONFIG_DEBUG_BUGVERBOSE extern volatile void __bug(const char *file, int line, void *data); /* give file/line information */ @@ -12,6 +13,8 @@ extern volatile void __bug(const char *file, int line, void *data); #endif #define HAVE_ARCH_BUG +#endif + #include #endif diff --git a/include/asm-frv/bug.h b/include/asm-frv/bug.h index 011860b..074c0d5 100644 --- a/include/asm-frv/bug.h +++ b/include/asm-frv/bug.h @@ -13,6 +13,7 @@ #include +#ifdef CONFIG_BUG /* * Tell the user there is some problem. */ @@ -45,6 +46,7 @@ do { \ #define HAVE_ARCH_KGDB_BAD_PAGE #define kgdb_bad_page(page) do { kgdb_raise(SIGABRT); } while(0) #endif +#endif #include diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h index e5913c3..6e5aaaa 100644 --- a/include/asm-generic/bug.h +++ b/include/asm-generic/bug.h @@ -4,6 +4,7 @@ #include #include +#ifdef CONFIG_BUG #ifndef HAVE_ARCH_BUG #define BUG() do { \ printk("kernel BUG at %s:%d!\n", __FILE__, __LINE__); \ @@ -31,4 +32,22 @@ } while (0) #endif +#else /* !CONFIG_BUG */ +#ifndef HAVE_ARCH_BUG +#define BUG() +#endif + +#ifndef HAVE_ARCH_PAGE_BUG +#define PAGE_BUG(page) do { if (page) ; } while (0) +#endif + +#ifndef HAVE_ARCH_BUG_ON +#define BUG_ON(condition) do { if (condition) ; } while(0) +#endif + +#ifndef HAVE_ARCH_WARN_ON +#define WARN_ON(condition) do { if (condition) ; } while(0) +#endif +#endif + #endif diff --git a/include/asm-i386/bug.h b/include/asm-i386/bug.h index 706eb51..8f79de1 100644 --- a/include/asm-i386/bug.h +++ b/include/asm-i386/bug.h @@ -9,6 +9,8 @@ * undefined" opcode for parsing in the trap handler. */ +#ifdef CONFIG_BUG +#define HAVE_ARCH_BUG #ifdef CONFIG_DEBUG_BUGVERBOSE #define BUG() \ __asm__ __volatile__( "ud2\n" \ @@ -18,8 +20,7 @@ #else #define BUG() __asm__ __volatile__("ud2\n") #endif +#endif -#define HAVE_ARCH_BUG #include - #endif diff --git a/include/asm-ia64/bug.h b/include/asm-ia64/bug.h index 2c0cd51..3aa0a0a 100644 --- a/include/asm-ia64/bug.h +++ b/include/asm-ia64/bug.h @@ -1,6 +1,7 @@ #ifndef _ASM_IA64_BUG_H #define _ASM_IA64_BUG_H +#ifdef CONFIG_BUG #if (__GNUC__ > 3) || (__GNUC__ == 3 && __GNUC_MINOR__ >= 1) # define ia64_abort() __builtin_trap() #else @@ -8,8 +9,10 @@ #endif #define BUG() do { printk("kernel BUG at %s:%d!\n", __FILE__, __LINE__); ia64_abort(); } while (0) -/* should this BUG should be made generic? */ +/* should this BUG be made generic? */ #define HAVE_ARCH_BUG +#endif + #include #endif diff --git a/include/asm-m68k/bug.h b/include/asm-m68k/bug.h index 3e1d226..072ce27 100644 --- a/include/asm-m68k/bug.h +++ b/include/asm-m68k/bug.h @@ -3,6 +3,7 @@ #include +#ifdef CONFIG_BUG #ifdef CONFIG_DEBUG_BUGVERBOSE #ifndef CONFIG_SUN3 #define BUG() do { \ @@ -22,6 +23,8 @@ #endif #define HAVE_ARCH_BUG +#endif + #include #endif diff --git a/include/asm-mips/bug.h b/include/asm-mips/bug.h index eb94bb9..3f594b4 100644 --- a/include/asm-mips/bug.h +++ b/include/asm-mips/bug.h @@ -3,12 +3,14 @@ #include +#ifdef CONFIG_BUG +#define HAVE_ARCH_BUG #define BUG() \ do { \ __asm__ __volatile__("break %0" : : "i" (BRK_BUG)); \ } while (0) +#endif -#define HAVE_ARCH_BUG #include #endif diff --git a/include/asm-parisc/bug.h b/include/asm-parisc/bug.h index e72f6e2..695588d 100644 --- a/include/asm-parisc/bug.h +++ b/include/asm-parisc/bug.h @@ -1,12 +1,14 @@ #ifndef _PARISC_BUG_H #define _PARISC_BUG_H +#ifdef CONFIG_BUG #define HAVE_ARCH_BUG #define BUG() do { \ printk("kernel BUG at %s:%d!\n", __FILE__, __LINE__); \ dump_stack(); \ panic("BUG!"); \ } while (0) +#endif #include #endif diff --git a/include/asm-ppc/bug.h b/include/asm-ppc/bug.h index e99c6cb..8b34fd6 100644 --- a/include/asm-ppc/bug.h +++ b/include/asm-ppc/bug.h @@ -14,6 +14,7 @@ struct bug_entry { */ #define BUG_WARNING_TRAP 0x1000000 +#ifdef CONFIG_BUG #define BUG() do { \ __asm__ __volatile__( \ "1: twi 31,0,0\n" \ @@ -50,6 +51,8 @@ struct bug_entry { #define HAVE_ARCH_BUG #define HAVE_ARCH_BUG_ON #define HAVE_ARCH_WARN_ON +#endif + #include #endif diff --git a/include/asm-ppc64/bug.h b/include/asm-ppc64/bug.h index db31dd2..169868f 100644 --- a/include/asm-ppc64/bug.h +++ b/include/asm-ppc64/bug.h @@ -26,6 +26,8 @@ struct bug_entry *find_bug(unsigned long bugaddr); */ #define BUG_WARNING_TRAP 0x1000000 +#ifdef CONFIG_BUG + #define BUG() do { \ __asm__ __volatile__( \ "1: twi 31,0,0\n" \ @@ -55,11 +57,12 @@ struct bug_entry *find_bug(unsigned long bugaddr); "i" (__FILE__), "i" (__FUNCTION__)); \ } while (0) -#endif - #define HAVE_ARCH_BUG #define HAVE_ARCH_BUG_ON #define HAVE_ARCH_WARN_ON +#endif +#endif + #include #endif diff --git a/include/asm-s390/bug.h b/include/asm-s390/bug.h index 2b8d6d4..a2e7430 100644 --- a/include/asm-s390/bug.h +++ b/include/asm-s390/bug.h @@ -3,12 +3,15 @@ #include +#ifdef CONFIG_BUG #define BUG() do { \ printk("kernel BUG at %s:%d!\n", __FILE__, __LINE__); \ __asm__ __volatile__(".long 0"); \ } while (0) #define HAVE_ARCH_BUG +#endif + #include #endif diff --git a/include/asm-sh/bug.h b/include/asm-sh/bug.h index 7017221..70508a3 100644 --- a/include/asm-sh/bug.h +++ b/include/asm-sh/bug.h @@ -3,6 +3,7 @@ #include +#ifdef CONFIG_BUG /* * Tell the user there is some problem. */ @@ -12,6 +13,8 @@ } while (0) #define HAVE_ARCH_BUG +#endif + #include #endif diff --git a/include/asm-sparc/bug.h b/include/asm-sparc/bug.h index 0d30a67..0415120 100644 --- a/include/asm-sparc/bug.h +++ b/include/asm-sparc/bug.h @@ -1,6 +1,7 @@ #ifndef _SPARC_BUG_H #define _SPARC_BUG_H +#ifdef CONFIG_BUG /* Only use the inline asm until a gcc release that can handle __builtin_trap * -rob 2003-06-25 * @@ -26,6 +27,8 @@ extern void do_BUG(const char *file, int line); #endif #define HAVE_ARCH_BUG +#endif + #include #endif diff --git a/include/asm-sparc64/bug.h b/include/asm-sparc64/bug.h index 25c5b1d..516bb27 100644 --- a/include/asm-sparc64/bug.h +++ b/include/asm-sparc64/bug.h @@ -1,6 +1,7 @@ #ifndef _SPARC64_BUG_H #define _SPARC64_BUG_H +#ifdef CONFIG_BUG #include #ifdef CONFIG_DEBUG_BUGVERBOSE @@ -14,6 +15,8 @@ extern void do_BUG(const char *file, int line); #endif #define HAVE_ARCH_BUG +#endif + #include #endif diff --git a/include/asm-v850/bug.h b/include/asm-v850/bug.h index c778916..b0ed2d3 100644 --- a/include/asm-v850/bug.h +++ b/include/asm-v850/bug.h @@ -14,9 +14,12 @@ #ifndef __V850_BUG_H__ #define __V850_BUG_H__ +#ifdef CONFIG_BUG extern void __bug (void) __attribute__ ((noreturn)); #define BUG() __bug() #define HAVE_ARCH_BUG +#endif + #include #endif /* __V850_BUG_H__ */ diff --git a/include/asm-x86_64/bug.h b/include/asm-x86_64/bug.h index 19aed6e..bdbf66e 100644 --- a/include/asm-x86_64/bug.h +++ b/include/asm-x86_64/bug.h @@ -15,11 +15,13 @@ struct bug_frame { unsigned short line; } __attribute__((packed)); +#ifdef CONFIG_BUG #define HAVE_ARCH_BUG #define BUG() \ asm volatile("ud2 ; .quad %c1 ; .short %c0" :: \ "i"(__LINE__), "i" (__stringify(__FILE__))) void out_of_line_bug(void); -#include +#endif +#include #endif diff --git a/init/Kconfig b/init/Kconfig index abe2682..42dca39 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -275,6 +275,16 @@ config KALLSYMS_EXTRA_PASS reported. KALLSYMS_EXTRA_PASS is only a temporary workaround while you wait for kallsyms to be fixed. +config BUG + bool "BUG() support" if EMBEDDED + default y + help + Disabling this option eliminates support for BUG and WARN, reducing + the size of your kernel image and potentially quietly ignoring + numerous fatal conditions. You should only consider disabling this + option for embedded systems with no facilities for reporting errors. + Just say Y. + config BASE_FULL default y bool "Enable full-sized data structures for core" if EMBEDDED diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 426a0cf..ac23847 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -108,6 +108,7 @@ config DEBUG_HIGHMEM config DEBUG_BUGVERBOSE bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EMBEDDED + depends on BUG depends on ARM || ARM26 || M32R || M68K || SPARC32 || SPARC64 || (X86 && !X86_64) || FRV default !EMBEDDED help -- cgit v0.10.2 From cd7619d6bf36564cf54ff7218ef54e558a741913 Mon Sep 17 00:00:00 2001 From: Matt Mackall Date: Sun, 1 May 2005 08:59:01 -0700 Subject: [PATCH] Exterminate PAGE_BUG Remove PAGE_BUG - repalce it with BUG and BUG_ON. Signed-off-by: Matt Mackall Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/arm26/mm/small_page.c b/arch/arm26/mm/small_page.c index 77be86c..3044710 100644 --- a/arch/arm26/mm/small_page.c +++ b/arch/arm26/mm/small_page.c @@ -92,8 +92,7 @@ static unsigned long __get_small_page(int priority, struct order *order) page = list_entry(order->queue.next, struct page, lru); again: #ifdef PEDANTIC - if (USED_MAP(page) & ~order->all_used) - PAGE_BUG(page); + BUG_ON(USED_MAP(page) & ~order->all_used); #endif offset = ffz(USED_MAP(page)); SET_USED(page, offset); @@ -141,8 +140,7 @@ static void __free_small_page(unsigned long spage, struct order *order) goto non_small; #ifdef PEDANTIC - if (USED_MAP(page) & ~order->all_used) - PAGE_BUG(page); + BUG_ON(USED_MAP(page) & ~order->all_used); #endif spage = spage >> order->shift; diff --git a/fs/afs/file.c b/fs/afs/file.c index 6b6bb7c..23c1251 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -131,8 +131,7 @@ static int afs_file_readpage(struct file *file, struct page *page) vnode = AFS_FS_I(inode); - if (!PageLocked(page)) - PAGE_BUG(page); + BUG_ON(!PageLocked(page)); ret = -ESTALE; if (vnode->flags & AFS_VNODE_DELETED) diff --git a/fs/buffer.c b/fs/buffer.c index 188365c..792cbac 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2078,8 +2078,7 @@ int block_read_full_page(struct page *page, get_block_t *get_block) int nr, i; int fully_mapped = 1; - if (!PageLocked(page)) - PAGE_BUG(page); + BUG_ON(!PageLocked(page)); blocksize = 1 << inode->i_blkbits; if (!page_has_buffers(page)) create_empty_buffers(page, blocksize, 0); diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c index 0c607c1..771a554 100644 --- a/fs/jffs2/file.c +++ b/fs/jffs2/file.c @@ -79,8 +79,7 @@ static int jffs2_do_readpage_nolock (struct inode *inode, struct page *pg) D2(printk(KERN_DEBUG "jffs2_do_readpage_nolock(): ino #%lu, page at offset 0x%lx\n", inode->i_ino, pg->index << PAGE_CACHE_SHIFT)); - if (!PageLocked(pg)) - PAGE_BUG(pg); + BUG_ON(!PageLocked(pg)); pg_buf = kmap(pg); /* FIXME: Can kmap fail? */ diff --git a/fs/udf/file.c b/fs/udf/file.c index 2faa417..bb40d63 100644 --- a/fs/udf/file.c +++ b/fs/udf/file.c @@ -49,8 +49,7 @@ static int udf_adinicb_readpage(struct file *file, struct page * page) struct inode *inode = page->mapping->host; char *kaddr; - if (!PageLocked(page)) - PAGE_BUG(page); + BUG_ON(!PageLocked(page)); kaddr = kmap(page); memset(kaddr, 0, PAGE_CACHE_SIZE); @@ -67,8 +66,7 @@ static int udf_adinicb_writepage(struct page *page, struct writeback_control *wb struct inode *inode = page->mapping->host; char *kaddr; - if (!PageLocked(page)) - PAGE_BUG(page); + BUG_ON(!PageLocked(page)); kaddr = kmap(page); memcpy(UDF_I_DATA(inode) + UDF_I_LENEATTR(inode), kaddr, inode->i_size); diff --git a/fs/udf/inode.c b/fs/udf/inode.c index 0506e11..3d68de3 100644 --- a/fs/udf/inode.c +++ b/fs/udf/inode.c @@ -167,8 +167,8 @@ void udf_expand_file_adinicb(struct inode * inode, int newsize, int * err) } page = grab_cache_page(inode->i_mapping, 0); - if (!PageLocked(page)) - PAGE_BUG(page); + BUG_ON(!PageLocked(page)); + if (!PageUptodate(page)) { kaddr = kmap(page); diff --git a/include/asm-cris/page.h b/include/asm-cris/page.h index ddd8915..c767da1 100644 --- a/include/asm-cris/page.h +++ b/include/asm-cris/page.h @@ -77,10 +77,6 @@ typedef struct { unsigned long pgprot; } pgprot_t; printk("kernel BUG at %s:%d!\n", __FILE__, __LINE__); \ } while (0) -#define PAGE_BUG(page) do { \ - BUG(); \ -} while (0) - /* Pure 2^n version of get_order */ static inline int get_order(unsigned long size) { diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h index 6e5aaaa..400c2b4 100644 --- a/include/asm-generic/bug.h +++ b/include/asm-generic/bug.h @@ -12,13 +12,6 @@ } while (0) #endif -#ifndef HAVE_ARCH_PAGE_BUG -#define PAGE_BUG(page) do { \ - printk("page BUG for page at %p\n", page); \ - BUG(); \ -} while (0) -#endif - #ifndef HAVE_ARCH_BUG_ON #define BUG_ON(condition) do { if (unlikely((condition)!=0)) BUG(); } while(0) #endif @@ -37,10 +30,6 @@ #define BUG() #endif -#ifndef HAVE_ARCH_PAGE_BUG -#define PAGE_BUG(page) do { if (page) ; } while (0) -#endif - #ifndef HAVE_ARCH_BUG_ON #define BUG_ON(condition) do { if (condition) ; } while(0) #endif diff --git a/include/asm-sh64/bug.h b/include/asm-sh64/bug.h index 3acd54d..5d659ec 100644 --- a/include/asm-sh64/bug.h +++ b/include/asm-sh64/bug.h @@ -17,10 +17,6 @@ BUG(); \ } while(0) -#define PAGE_BUG(page) do { \ - BUG(); \ -} while (0) - #define WARN_ON(condition) do { \ if (unlikely((condition)!=0)) { \ printk("Badness in %s at %s:%d\n", __FUNCTION__, __FILE__, __LINE__); \ diff --git a/mm/filemap.c b/mm/filemap.c index ee79b5d..c085af2 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -123,8 +123,7 @@ void remove_from_page_cache(struct page *page) { struct address_space *mapping = page->mapping; - if (unlikely(!PageLocked(page))) - PAGE_BUG(page); + BUG_ON(!PageLocked(page)); write_lock_irq(&mapping->tree_lock); __remove_from_page_cache(page); -- cgit v0.10.2 From d59745ce3e7aa13856bca16d3bcbb95041775ff6 Mon Sep 17 00:00:00 2001 From: Matt Mackall Date: Sun, 1 May 2005 08:59:02 -0700 Subject: [PATCH] clean up kernel messages Arrange for all kernel printks to be no-ops. Only available if CONFIG_EMBEDDED. This patch saves about 375k on my laptop config and nearly 100k on minimal configs. Signed-off-by: Matt Mackall Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S index d273fd7..e966fc8 100644 --- a/arch/i386/kernel/head.S +++ b/arch/i386/kernel/head.S @@ -380,6 +380,7 @@ rp_sidt: ALIGN ignore_int: cld +#ifdef CONFIG_PRINTK pushl %eax pushl %ecx pushl %edx @@ -400,6 +401,7 @@ ignore_int: popl %edx popl %ecx popl %eax +#endif iret /* diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 7c1cba4..e25b970 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -115,10 +115,19 @@ extern int __kernel_text_address(unsigned long addr); extern int kernel_text_address(unsigned long addr); extern int session_of_pgrp(int pgrp); +#ifdef CONFIG_PRINTK asmlinkage int vprintk(const char *fmt, va_list args) __attribute__ ((format (printf, 1, 0))); asmlinkage int printk(const char * fmt, ...) __attribute__ ((format (printf, 1, 2))); +#else +static inline int vprintk(const char *s, va_list args) + __attribute__ ((format (printf, 1, 0))); +static inline int vprintk(const char *s, va_list args) { return 0; } +static inline int printk(const char *s, ...) + __attribute__ ((format (printf, 1, 2))); +static inline int printk(const char *s, ...) { return 0; } +#endif unsigned long int_sqrt(unsigned long); diff --git a/init/Kconfig b/init/Kconfig index 42dca39..40d286d 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -275,6 +275,17 @@ config KALLSYMS_EXTRA_PASS reported. KALLSYMS_EXTRA_PASS is only a temporary workaround while you wait for kallsyms to be fixed. + +config PRINTK + default y + bool "Enable support for printk" if EMBEDDED + help + This option enables normal printk support. Removing it + eliminates most of the message strings from the kernel image + and makes the kernel more or less silent. As this makes it + very difficult to diagnose system problems, saying N here is + strongly discouraged. + config BUG bool "BUG() support" if EMBEDDED default y diff --git a/kernel/printk.c b/kernel/printk.c index 1498689..290a07c 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -85,10 +85,6 @@ static int console_locked; */ static DEFINE_SPINLOCK(logbuf_lock); -static char __log_buf[__LOG_BUF_LEN]; -static char *log_buf = __log_buf; -static int log_buf_len = __LOG_BUF_LEN; - #define LOG_BUF_MASK (log_buf_len-1) #define LOG_BUF(idx) (log_buf[(idx) & LOG_BUF_MASK]) @@ -99,7 +95,6 @@ static int log_buf_len = __LOG_BUF_LEN; static unsigned long log_start; /* Index into log_buf: next char to be read by syslog() */ static unsigned long con_start; /* Index into log_buf: next char to be sent to consoles */ static unsigned long log_end; /* Index into log_buf: most-recently-written-char + 1 */ -static unsigned long logged_chars; /* Number of chars produced since last read+clear operation */ /* * Array of consoles built from command line options (console=) @@ -120,6 +115,13 @@ static int preferred_console = -1; /* Flag: console code may call schedule() */ static int console_may_schedule; +#ifdef CONFIG_PRINTK + +static char __log_buf[__LOG_BUF_LEN]; +static char *log_buf = __log_buf; +static int log_buf_len = __LOG_BUF_LEN; +static unsigned long logged_chars; /* Number of chars produced since last read+clear operation */ + /* * Setup a list of consoles. Called from init/main.c */ @@ -535,6 +537,7 @@ __setup("time", printk_time_setup); * then changes console_loglevel may break. This is because console_loglevel * is inspected when the actual printing occurs. */ + asmlinkage int printk(const char *fmt, ...) { va_list args; @@ -655,6 +658,18 @@ out: EXPORT_SYMBOL(printk); EXPORT_SYMBOL(vprintk); +#else + +asmlinkage long sys_syslog(int type, char __user * buf, int len) +{ + return 0; +} + +int do_syslog(int type, char __user * buf, int len) { return 0; } +static void call_console_drivers(unsigned long start, unsigned long end) {} + +#endif + /** * acquire_console_sem - lock the console system for exclusive use. * @@ -931,7 +946,7 @@ int unregister_console(struct console * console) return res; } EXPORT_SYMBOL(unregister_console); - + /** * tty_write_message - write a message to a certain tty, not just the console. * -- cgit v0.10.2 From 7f261b5f0dccd53ed3a9a95b55c36e24a698a92a Mon Sep 17 00:00:00 2001 From: Stas Sergeev Date: Sun, 1 May 2005 08:59:02 -0700 Subject: [PATCH] move SA_xxx defines to linux/signal.h The attached patch moves the IRQ-related SA_xxx flags (namely, SA_PROBE, SA_SAMPLE_RANDOM and SA_SHIRQ) from all the arch-specific headers to linux/signal.h. This looks like a left-over after the irq-handling code was consolidated. The code was moved to kernel/irq/*, but the flags are still left per-arch. Right now, adding a new IRQ flag to the arch-specific header, like this patch does: http://cvs.sourceforge.net/viewcvs.py/*checkout*/alsa/alsa-driver/utils/patches/pcsp-kernel-2.6.10-03.diff?rev=1.1 no longer works, it breaks the compilation for all other arches, unless you add that flag to all the other arch-specific headers too. So I think such a clean-up makes sense. Signed-off-by: Stas Sergeev Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-alpha/signal.h b/include/asm-alpha/signal.h index 25f98bc..4e0842b 100644 --- a/include/asm-alpha/signal.h +++ b/include/asm-alpha/signal.h @@ -109,20 +109,6 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 4096 #define SIGSTKSZ 16384 - -#ifdef __KERNEL__ -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x40000000 -#endif - #define SIG_BLOCK 1 /* for blocking signals */ #define SIG_UNBLOCK 2 /* for unblocking signals */ #define SIG_SETMASK 3 /* for setting the signal mask */ diff --git a/include/asm-arm/signal.h b/include/asm-arm/signal.h index b033e5f..b860dc3 100644 --- a/include/asm-arm/signal.h +++ b/include/asm-arm/signal.h @@ -114,18 +114,7 @@ typedef unsigned long sigset_t; #define SIGSTKSZ 8192 #ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE 0x80000000 -#define SA_SAMPLE_RANDOM 0x10000000 #define SA_IRQNOMASK 0x08000000 -#define SA_SHIRQ 0x04000000 #endif #define SIG_BLOCK 0 /* for blocking signals */ diff --git a/include/asm-arm26/signal.h b/include/asm-arm26/signal.h index 6f62e51..a1aacef 100644 --- a/include/asm-arm26/signal.h +++ b/include/asm-arm26/signal.h @@ -114,18 +114,7 @@ typedef unsigned long sigset_t; #define SIGSTKSZ 8192 #ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE 0x80000000 -#define SA_SAMPLE_RANDOM 0x10000000 #define SA_IRQNOMASK 0x08000000 -#define SA_SHIRQ 0x04000000 #endif #define SIG_BLOCK 0 /* for blocking signals */ diff --git a/include/asm-cris/signal.h b/include/asm-cris/signal.h index 3f187ec..2330769 100644 --- a/include/asm-cris/signal.h +++ b/include/asm-cris/signal.h @@ -108,20 +108,6 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - #define SIG_BLOCK 0 /* for blocking signals */ #define SIG_UNBLOCK 1 /* for unblocking signals */ #define SIG_SETMASK 2 /* for setting the signal mask */ diff --git a/include/asm-frv/signal.h b/include/asm-frv/signal.h index f18952f..c930bb1 100644 --- a/include/asm-frv/signal.h +++ b/include/asm-frv/signal.h @@ -107,20 +107,6 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - #define SIG_BLOCK 0 /* for blocking signals */ #define SIG_UNBLOCK 1 /* for unblocking signals */ #define SIG_SETMASK 2 /* for setting the signal mask */ diff --git a/include/asm-h8300/signal.h b/include/asm-h8300/signal.h index 3a08544..ac3e01b 100644 --- a/include/asm-h8300/signal.h +++ b/include/asm-h8300/signal.h @@ -107,19 +107,6 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - #define SIG_BLOCK 0 /* for blocking signals */ #define SIG_UNBLOCK 1 /* for unblocking signals */ #define SIG_SETMASK 2 /* for setting the signal mask */ diff --git a/include/asm-i386/signal.h b/include/asm-i386/signal.h index 7ef343b..0f082bd 100644 --- a/include/asm-i386/signal.h +++ b/include/asm-i386/signal.h @@ -110,20 +110,6 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - #define SIG_BLOCK 0 /* for blocking signals */ #define SIG_UNBLOCK 1 /* for unblocking signals */ #define SIG_SETMASK 2 /* for setting the signal mask */ diff --git a/include/asm-ia64/signal.h b/include/asm-ia64/signal.h index 660a759..85a577a 100644 --- a/include/asm-ia64/signal.h +++ b/include/asm-ia64/signal.h @@ -114,16 +114,6 @@ #define _NSIG_BPW 64 #define _NSIG_WORDS (_NSIG / _NSIG_BPW) -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 #define SA_PERCPU_IRQ 0x02000000 #endif /* __KERNEL__ */ diff --git a/include/asm-m32r/signal.h b/include/asm-m32r/signal.h index ce46eae..6e55fd4 100644 --- a/include/asm-m32r/signal.h +++ b/include/asm-m32r/signal.h @@ -114,20 +114,6 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - #define SIG_BLOCK 0 /* for blocking signals */ #define SIG_UNBLOCK 1 /* for unblocking signals */ #define SIG_SETMASK 2 /* for setting the signal mask */ diff --git a/include/asm-m68k/signal.h b/include/asm-m68k/signal.h index 6681bb6..1d016e9 100644 --- a/include/asm-m68k/signal.h +++ b/include/asm-m68k/signal.h @@ -105,19 +105,6 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - #define SIG_BLOCK 0 /* for blocking signals */ #define SIG_UNBLOCK 1 /* for unblocking signals */ #define SIG_SETMASK 2 /* for setting the signal mask */ diff --git a/include/asm-m68knommu/signal.h b/include/asm-m68knommu/signal.h index 486cbb0..37c9c8a 100644 --- a/include/asm-m68knommu/signal.h +++ b/include/asm-m68knommu/signal.h @@ -105,19 +105,6 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - #define SIG_BLOCK 0 /* for blocking signals */ #define SIG_UNBLOCK 1 /* for unblocking signals */ #define SIG_SETMASK 2 /* for setting the signal mask */ diff --git a/include/asm-mips/signal.h b/include/asm-mips/signal.h index 994987d..d813567 100644 --- a/include/asm-mips/signal.h +++ b/include/asm-mips/signal.h @@ -98,21 +98,6 @@ typedef unsigned long old_sigset_t; /* at least 32 bits */ #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ flag is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x02000000 - -#endif /* __KERNEL__ */ - #define SIG_BLOCK 1 /* for blocking signals */ #define SIG_UNBLOCK 2 /* for unblocking signals */ #define SIG_SETMASK 3 /* for setting the signal mask */ diff --git a/include/asm-parisc/signal.h b/include/asm-parisc/signal.h index 358f577..25cb23e 100644 --- a/include/asm-parisc/signal.h +++ b/include/asm-parisc/signal.h @@ -89,17 +89,6 @@ #define _NSIG_BPW BITS_PER_LONG #define _NSIG_WORDS (_NSIG / _NSIG_BPW) -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 - #endif /* __KERNEL__ */ #define SIG_BLOCK 0 /* for blocking signals */ diff --git a/include/asm-ppc/signal.h b/include/asm-ppc/signal.h index 8cc8b88..d890dab 100644 --- a/include/asm-ppc/signal.h +++ b/include/asm-ppc/signal.h @@ -99,19 +99,6 @@ typedef struct { #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif /* __KERNEL__ */ #define SIG_BLOCK 0 /* for blocking signals */ #define SIG_UNBLOCK 1 /* for unblocking signals */ diff --git a/include/asm-ppc64/signal.h b/include/asm-ppc64/signal.h index fe5401a..a2d7bbb 100644 --- a/include/asm-ppc64/signal.h +++ b/include/asm-ppc64/signal.h @@ -96,19 +96,6 @@ typedef struct { #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif #define SIG_BLOCK 0 /* for blocking signals */ #define SIG_UNBLOCK 1 /* for unblocking signals */ diff --git a/include/asm-s390/signal.h b/include/asm-s390/signal.h index f273cdc..bfed83a8 100644 --- a/include/asm-s390/signal.h +++ b/include/asm-s390/signal.h @@ -117,20 +117,6 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - #define SIG_BLOCK 0 /* for blocking signals */ #define SIG_UNBLOCK 1 /* for unblocking signals */ #define SIG_SETMASK 2 /* for setting the signal mask */ diff --git a/include/asm-sh/signal.h b/include/asm-sh/signal.h index 0a7ff71..29f1ac1 100644 --- a/include/asm-sh/signal.h +++ b/include/asm-sh/signal.h @@ -108,20 +108,6 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - #define SIG_BLOCK 0 /* for blocking signals */ #define SIG_UNBLOCK 1 /* for unblocking signals */ #define SIG_SETMASK 2 /* for setting the signal mask */ diff --git a/include/asm-sh64/signal.h b/include/asm-sh64/signal.h index 77957e9..864c94e 100644 --- a/include/asm-sh64/signal.h +++ b/include/asm-sh64/signal.h @@ -107,20 +107,6 @@ typedef struct { #define MINSIGSTKSZ 2048 #define SIGSTKSZ THREAD_SIZE -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - #define SIG_BLOCK 0 /* for blocking signals */ #define SIG_UNBLOCK 1 /* for unblocking signals */ #define SIG_SETMASK 2 /* for setting the signal mask */ diff --git a/include/asm-sparc/signal.h b/include/asm-sparc/signal.h index d8211cb..f792e10 100644 --- a/include/asm-sparc/signal.h +++ b/include/asm-sparc/signal.h @@ -143,7 +143,6 @@ struct sigstack { #define SA_ONESHOT _SV_RESET #define SA_INTERRUPT 0x10u #define SA_NOMASK 0x20u -#define SA_SHIRQ 0x40u #define SA_NOCLDWAIT 0x100u #define SA_SIGINFO 0x200u @@ -162,11 +161,6 @@ struct sigstack { #ifdef __KERNEL__ /* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * * DJHR * SA_STATIC_ALLOC is used for the SPARC system to indicate that this * interrupt handler's irq structure should be statically allocated @@ -177,8 +171,6 @@ struct sigstack { * statically allocated data.. which is NOT GOOD. * */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART #define SA_STATIC_ALLOC 0x80 #endif diff --git a/include/asm-sparc64/signal.h b/include/asm-sparc64/signal.h index 6428e36..466d021 100644 --- a/include/asm-sparc64/signal.h +++ b/include/asm-sparc64/signal.h @@ -145,7 +145,6 @@ struct sigstack { #define SA_ONESHOT _SV_RESET #define SA_INTERRUPT 0x10u #define SA_NOMASK 0x20u -#define SA_SHIRQ 0x40u #define SA_NOCLDWAIT 0x100u #define SA_SIGINFO 0x200u @@ -165,11 +164,6 @@ struct sigstack { #ifdef __KERNEL__ /* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * * DJHR * SA_STATIC_ALLOC is used for the SPARC system to indicate that this * interrupt handler's irq structure should be statically allocated @@ -180,8 +174,6 @@ struct sigstack { * statically allocated data.. which is NOT GOOD. * */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART #define SA_STATIC_ALLOC 0x80 #endif diff --git a/include/asm-v850/signal.h b/include/asm-v850/signal.h index 407db87..ec3566c 100644 --- a/include/asm-v850/signal.h +++ b/include/asm-v850/signal.h @@ -110,21 +110,6 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 - -#ifdef __KERNEL__ -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif /* __KERNEL__ */ - - #define SIG_BLOCK 0 /* for blocking signals */ #define SIG_UNBLOCK 1 /* for unblocking signals */ #define SIG_SETMASK 2 /* for setting the signal mask */ diff --git a/include/asm-x86_64/signal.h b/include/asm-x86_64/signal.h index 643a20d..4987ad8 100644 --- a/include/asm-x86_64/signal.h +++ b/include/asm-x86_64/signal.h @@ -116,20 +116,6 @@ typedef unsigned long sigset_t; #define MINSIGSTKSZ 2048 #define SIGSTKSZ 8192 -#ifdef __KERNEL__ - -/* - * These values of sa_flags are used only by the kernel as part of the - * irq handling routines. - * - * SA_INTERRUPT is also used by the irq handling routines. - * SA_SHIRQ is for shared interrupt support on PCI and EISA. - */ -#define SA_PROBE SA_ONESHOT -#define SA_SAMPLE_RANDOM SA_RESTART -#define SA_SHIRQ 0x04000000 -#endif - #define SIG_BLOCK 0 /* for blocking signals */ #define SIG_UNBLOCK 1 /* for unblocking signals */ #define SIG_SETMASK 2 /* for setting the signal mask */ diff --git a/include/linux/signal.h b/include/linux/signal.h index 99c97ad..78bfb26 100644 --- a/include/linux/signal.h +++ b/include/linux/signal.h @@ -9,6 +9,17 @@ #ifdef __KERNEL__ /* + * These values of sa_flags are used only by the kernel as part of the + * irq handling routines. + * + * SA_INTERRUPT is also used by the irq handling routines. + * SA_SHIRQ is for shared interrupt support on PCI and EISA. + */ +#define SA_PROBE SA_ONESHOT +#define SA_SAMPLE_RANDOM SA_RESTART +#define SA_SHIRQ 0x04000000 + +/* * Real Time signals may be queued. */ -- cgit v0.10.2 From bcf88e1163623e8e8ef2ba7feface9c826a890c9 Mon Sep 17 00:00:00 2001 From: Daniel Drake Date: Sun, 1 May 2005 08:59:03 -0700 Subject: [PATCH] procfs: Fix hardlink counts The pid directories in /proc/ currently return the wrong hardlink count - 3, when there are actually 4 : ".", "..", "fd", and "task". This is easy to notice using find(1): cd /proc/ find In the output, you'll see a message similar to: find: WARNING: Hard link count is wrong for .: this may be a bug in your filesystem driver. Automatically turning on find's -noleaf option. Earlier results may have failed to include directories that should have been searched. http://bugs.gentoo.org/show_bug.cgi?id=86031 I also noticed that CONFIG_SECURITY can add a 5th: attr, and performed a similar fix on the task directories too. Signed-off-by: Daniel Drake Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/proc/base.c b/fs/proc/base.c index 39fd336..4718173 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1800,8 +1800,12 @@ struct dentry *proc_pid_lookup(struct inode *dir, struct dentry * dentry, struct inode->i_mode = S_IFDIR|S_IRUGO|S_IXUGO; inode->i_op = &proc_tgid_base_inode_operations; inode->i_fop = &proc_tgid_base_operations; - inode->i_nlink = 3; inode->i_flags|=S_IMMUTABLE; +#ifdef CONFIG_SECURITY + inode->i_nlink = 5; +#else + inode->i_nlink = 4; +#endif dentry->d_op = &pid_base_dentry_operations; @@ -1855,8 +1859,12 @@ static struct dentry *proc_task_lookup(struct inode *dir, struct dentry * dentry inode->i_mode = S_IFDIR|S_IRUGO|S_IXUGO; inode->i_op = &proc_tid_base_inode_operations; inode->i_fop = &proc_tid_base_operations; - inode->i_nlink = 3; inode->i_flags|=S_IMMUTABLE; +#ifdef CONFIG_SECURITY + inode->i_nlink = 4; +#else + inode->i_nlink = 3; +#endif dentry->d_op = &pid_base_dentry_operations; -- cgit v0.10.2 From f246315e1ab96c40978777d1e159820ecca45aa8 Mon Sep 17 00:00:00 2001 From: Daniel Drake Date: Sun, 1 May 2005 08:59:03 -0700 Subject: [PATCH] procfs: Fix hardlink counts for /proc//task The current logic assumes that a /proc//task directory should have a hardlink count of 3, probably counting ".", "..", and a directory for a single child task. It's fairly obvious that this doesn't work out correctly when a PID has more than one child task, which is quite often the case. Signed-off-by: Daniel Drake Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/proc/base.c b/fs/proc/base.c index 4718173..2eac86d 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1419,6 +1419,8 @@ static struct file_operations proc_tgid_attr_operations; static struct inode_operations proc_tgid_attr_inode_operations; #endif +static int get_tid_list(int index, unsigned int *tids, struct inode *dir); + /* SMP-safe */ static struct dentry *proc_pident_lookup(struct inode *dir, struct dentry *dentry, @@ -1458,7 +1460,7 @@ static struct dentry *proc_pident_lookup(struct inode *dir, */ switch(p->type) { case PROC_TGID_TASK: - inode->i_nlink = 3; + inode->i_nlink = 2 + get_tid_list(2, NULL, dir); inode->i_op = &proc_task_inode_operations; inode->i_fop = &proc_task_operations; break; @@ -1943,7 +1945,8 @@ static int get_tid_list(int index, unsigned int *tids, struct inode *dir) if (--index >= 0) continue; - tids[nr_tids] = tid; + if (tids != NULL) + tids[nr_tids] = tid; nr_tids++; if (nr_tids >= PROC_MAXPIDS) break; @@ -2043,6 +2046,7 @@ static int proc_task_readdir(struct file * filp, void * dirent, filldir_t filldi } nr_tids = get_tid_list(pos, tid_array, inode); + inode->i_nlink = pos + nr_tids; for (i = 0; i < nr_tids; i++) { unsigned long j = PROC_NUMBUF; -- cgit v0.10.2 From 66cf8f1443301a1d5bc9c21709e4264e6919a3e1 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sun, 1 May 2005 08:59:03 -0700 Subject: [PATCH] kernel/rcupdate.c: make the exports EXPORT_SYMBOL_GPL The gpl exports need to be put back. Moving them to GPL -- but in a measured manner, as I proposed on this list some months ago -- is fine. Changing these particular exports precipitously is most definitely -not- fine. Here is my earlier proposal: http://marc.theaimsgroup.com/?l=linux-kernel&m=110520930301813&w=2 See below for a patch that puts the exports back, along with an updated version of my earlier patch that starts the process of moving them to GPL. I will also be following this message with RFC patches that introduce two (EXPORT_SYMBOL_GPL) interfaces to replace synchronize_kernel(), which then becomes deprecated. Signed-off-by: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 56627c1..d3c52dd 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -48,3 +48,18 @@ When: April 2005 Why: Replaced by ->compat_ioctl in file_operations and other method vecors. Who: Andi Kleen , Christoph Hellwig + +--------------------------- + +What: RCU API moves to EXPORT_SYMBOL_GPL +When: April 2006 +Files: include/linux/rcupdate.h, kernel/rcupdate.c +Why: Outside of Linux, the only implementations of anything even + vaguely resembling RCU that I am aware of are in DYNIX/ptx, + VM/XA, Tornado, and K42. I do not expect anyone to port binary + drivers or kernel modules from any of these, since the first two + are owned by IBM and the last two are open-source research OSes. + So these will move to GPL after a grace period to allow + people, who might be using implementations that I am not aware + of, to adjust to this upcoming change. +Who: Paul E. McKenney diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index d00eded..ad49772 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -465,6 +465,6 @@ void synchronize_kernel(void) } module_param(maxbatch, int, 0); -EXPORT_SYMBOL_GPL(call_rcu); -EXPORT_SYMBOL_GPL(call_rcu_bh); -EXPORT_SYMBOL_GPL(synchronize_kernel); +EXPORT_SYMBOL(call_rcu); /* WARNING: GPL-only in April 2006. */ +EXPORT_SYMBOL(call_rcu_bh); /* WARNING: GPL-only in April 2006. */ +EXPORT_SYMBOL(synchronize_kernel); /* WARNING: GPL-only in April 2006. */ -- cgit v0.10.2 From 512345be2549308b8ae8e85a3ff7f6d56a38e5f6 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sun, 1 May 2005 08:59:03 -0700 Subject: [PATCH] Add deprecated_for_modules Add a deprecated_for_modules macro that allows symbols to be deprecated only when used by modules, as suggested by Andrew Morton some months back. Signed-off-by: Paul E. McKenney Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/compiler.h b/include/linux/compiler.h index 487725c..d737821 100644 --- a/include/linux/compiler.h +++ b/include/linux/compiler.h @@ -90,6 +90,12 @@ extern void __chk_io_ptr(void __iomem *); # define __deprecated /* unimplemented */ #endif +#ifdef MODULE +#define __deprecated_for_modules __deprecated +#else +#define __deprecated_for_modules +#endif + #ifndef __must_check #define __must_check #endif -- cgit v0.10.2 From 9b06e818985d139fd9e82c28297f7744e1b484e1 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sun, 1 May 2005 08:59:04 -0700 Subject: [PATCH] Deprecate synchronize_kernel, GPL replacement The synchronize_kernel() primitive is used for quite a few different purposes: waiting for RCU readers, waiting for NMIs, waiting for interrupts, and so on. This makes RCU code harder to read, since synchronize_kernel() might or might not have matching rcu_read_lock()s. This patch creates a new synchronize_rcu() that is to be used for RCU readers and a new synchronize_sched() that is used for the rest. These two new primitives currently have the same implementation, but this is might well change with additional real-time support. Both new primitives are GPL-only, the old primitive is deprecated. Signed-off-by: Paul E. McKenney Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 4d74743..fd276ad 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -157,9 +157,9 @@ static inline int rcu_pending(int cpu) /** * rcu_read_lock - mark the beginning of an RCU read-side critical section. * - * When synchronize_kernel() is invoked on one CPU while other CPUs + * When synchronize_rcu() is invoked on one CPU while other CPUs * are within RCU read-side critical sections, then the - * synchronize_kernel() is guaranteed to block until after all the other + * synchronize_rcu() is guaranteed to block until after all the other * CPUs exit their critical sections. Similarly, if call_rcu() is invoked * on one CPU while other CPUs are within RCU read-side critical * sections, invocation of the corresponding RCU callback is deferred @@ -256,6 +256,21 @@ static inline int rcu_pending(int cpu) (p) = (v); \ }) +/** + * synchronize_sched - block until all CPUs have exited any non-preemptive + * kernel code sequences. + * + * This means that all preempt_disable code sequences, including NMI and + * hardware-interrupt handlers, in progress on entry will have completed + * before this primitive returns. However, this does not guarantee that + * softirq handlers will have completed, since in some kernels + * + * This primitive provides the guarantees made by the (deprecated) + * synchronize_kernel() API. In contrast, synchronize_rcu() only + * guarantees that rcu_read_lock() sections will have completed. + */ +#define synchronize_sched() synchronize_rcu() + extern void rcu_init(void); extern void rcu_check_callbacks(int cpu, int user); extern void rcu_restart_cpu(int cpu); @@ -265,7 +280,9 @@ extern void FASTCALL(call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *head))); extern void FASTCALL(call_rcu_bh(struct rcu_head *head, void (*func)(struct rcu_head *head))); -extern void synchronize_kernel(void); +extern __deprecated_for_modules void synchronize_kernel(void); +extern void synchronize_rcu(void); +void synchronize_idle(void); #endif /* __KERNEL__ */ #endif /* __LINUX_RCUPDATE_H */ diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index ad49772..f436993 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -444,15 +444,18 @@ static void wakeme_after_rcu(struct rcu_head *head) } /** - * synchronize_kernel - wait until a grace period has elapsed. + * synchronize_rcu - wait until a grace period has elapsed. * * Control will return to the caller some time after a full grace * period has elapsed, in other words after all currently executing RCU * read-side critical sections have completed. RCU read-side critical * sections are delimited by rcu_read_lock() and rcu_read_unlock(), * and may be nested. + * + * If your read-side code is not protected by rcu_read_lock(), do -not- + * use synchronize_rcu(). */ -void synchronize_kernel(void) +void synchronize_rcu(void) { struct rcu_synchronize rcu; @@ -464,7 +467,16 @@ void synchronize_kernel(void) wait_for_completion(&rcu.completion); } +/* + * Deprecated, use synchronize_rcu() or synchronize_sched() instead. + */ +void synchronize_kernel(void) +{ + synchronize_rcu(); +} + module_param(maxbatch, int, 0); EXPORT_SYMBOL(call_rcu); /* WARNING: GPL-only in April 2006. */ EXPORT_SYMBOL(call_rcu_bh); /* WARNING: GPL-only in April 2006. */ +EXPORT_SYMBOL_GPL(synchronize_rcu); EXPORT_SYMBOL(synchronize_kernel); /* WARNING: GPL-only in April 2006. */ -- cgit v0.10.2 From fbd568a3e61a7decb8a754ad952aaa5b5c82e9e5 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sun, 1 May 2005 08:59:04 -0700 Subject: [PATCH] Change synchronize_kernel to _rcu and _sched This patch changes calls to synchronize_kernel(), deprecated in the earlier "Deprecate synchronize_kernel, GPL replacement" patch to instead call the new synchronize_rcu() and synchronize_sched() APIs. Signed-off-by: Paul E. McKenney Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/oprofile/nmi_timer_int.c b/arch/i386/oprofile/nmi_timer_int.c index b2e462a..c58d0c1 100644 --- a/arch/i386/oprofile/nmi_timer_int.c +++ b/arch/i386/oprofile/nmi_timer_int.c @@ -36,7 +36,7 @@ static void timer_stop(void) { enable_timer_nmi_watchdog(); unset_nmi_callback(); - synchronize_kernel(); + synchronize_sched(); /* Allow already-started NMIs to complete. */ } diff --git a/arch/ppc64/kernel/HvLpEvent.c b/arch/ppc64/kernel/HvLpEvent.c index 9802bee..f8f1963 100644 --- a/arch/ppc64/kernel/HvLpEvent.c +++ b/arch/ppc64/kernel/HvLpEvent.c @@ -45,7 +45,7 @@ int HvLpEvent_unregisterHandler( HvLpEvent_Type eventType ) /* We now sleep until all other CPUs have scheduled. This ensures that * the deletion is seen by all other CPUs, and that the deleted handler * isn't still running on another CPU when we return. */ - synchronize_kernel(); + synchronize_rcu(); } } return rc; diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index 05a1781..ff64d33 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -838,7 +838,7 @@ int acpi_processor_cst_has_changed (struct acpi_processor *pr) /* Fall back to the default idle loop */ pm_idle = pm_idle_save; - synchronize_kernel(); + synchronize_sched(); /* Relies on interrupts forcing exit from idle. */ pr->flags.power = 0; result = acpi_processor_get_power_info(pr); diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index 29de259..44a7f13 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -2199,7 +2199,7 @@ static int init_one_smi(int intf_num, struct smi_info **smi) /* Wait until we know that we are out of any interrupt handlers might have been running before we freed the interrupt. */ - synchronize_kernel(); + synchronize_sched(); if (new_smi->si_sm) { if (new_smi->handlers) @@ -2312,7 +2312,7 @@ static void __exit cleanup_one_si(struct smi_info *to_clean) /* Wait until we know that we are out of any interrupt handlers might have been running before we freed the interrupt. */ - synchronize_kernel(); + synchronize_sched(); /* Wait for the timer to stop. This avoids problems with race conditions removing the timer here. */ diff --git a/drivers/input/keyboard/atkbd.c b/drivers/input/keyboard/atkbd.c index f7304f0..ff66ed4 100644 --- a/drivers/input/keyboard/atkbd.c +++ b/drivers/input/keyboard/atkbd.c @@ -678,7 +678,7 @@ static void atkbd_disconnect(struct serio *serio) atkbd_disable(atkbd); /* make sure we don't have a command in flight */ - synchronize_kernel(); + synchronize_sched(); /* Allow atkbd_interrupt()s to complete. */ flush_scheduled_work(); device_remove_file(&serio->dev, &atkbd_attr_extra); diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c index c9b134c..1891e49 100644 --- a/drivers/md/multipath.c +++ b/drivers/md/multipath.c @@ -355,7 +355,7 @@ static int multipath_remove_disk(mddev_t *mddev, int number) goto abort; } p->rdev = NULL; - synchronize_kernel(); + synchronize_rcu(); if (atomic_read(&rdev->nr_pending)) { /* lost the race, try later */ err = -EBUSY; diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index a389394..83380b5 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -797,7 +797,7 @@ static int raid1_remove_disk(mddev_t *mddev, int number) goto abort; } p->rdev = NULL; - synchronize_kernel(); + synchronize_rcu(); if (atomic_read(&rdev->nr_pending)) { /* lost the race, try later */ err = -EBUSY; diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index b100bfe..e9dc287 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -977,7 +977,7 @@ static int raid10_remove_disk(mddev_t *mddev, int number) goto abort; } p->rdev = NULL; - synchronize_kernel(); + synchronize_rcu(); if (atomic_read(&rdev->nr_pending)) { /* lost the race, try later */ err = -EBUSY; diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 52c3a81..e96e2a1 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -1873,7 +1873,7 @@ static int raid5_remove_disk(mddev_t *mddev, int number) goto abort; } p->rdev = NULL; - synchronize_kernel(); + synchronize_rcu(); if (atomic_read(&rdev->nr_pending)) { /* lost the race, try later */ err = -EBUSY; diff --git a/drivers/md/raid6main.c b/drivers/md/raid6main.c index 7e30ab2..8a33f35 100644 --- a/drivers/md/raid6main.c +++ b/drivers/md/raid6main.c @@ -2038,7 +2038,7 @@ static int raid6_remove_disk(mddev_t *mddev, int number) goto abort; } p->rdev = NULL; - synchronize_kernel(); + synchronize_rcu(); if (atomic_read(&rdev->nr_pending)) { /* lost the race, try later */ err = -EBUSY; diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c index 07e2df0..c59507f 100644 --- a/drivers/net/r8169.c +++ b/drivers/net/r8169.c @@ -2385,7 +2385,7 @@ core_down: } /* Give a racing hard_start_xmit a few cycles to complete. */ - synchronize_kernel(); + synchronize_sched(); /* FIXME: should this be synchronize_irq()? */ /* * And now for the 50k$ question: are IRQ disabled or not ? diff --git a/drivers/s390/cio/airq.c b/drivers/s390/cio/airq.c index 3720e77..83e6a06 100644 --- a/drivers/s390/cio/airq.c +++ b/drivers/s390/cio/airq.c @@ -45,7 +45,7 @@ s390_register_adapter_interrupt (adapter_int_handler_t handler) else ret = (cmpxchg(&adapter_handler, NULL, handler) ? -EBUSY : 0); if (!ret) - synchronize_kernel(); + synchronize_sched(); /* Allow interrupts to complete. */ sprintf (dbf_txt, "ret:%d", ret); CIO_TRACE_EVENT (4, dbf_txt); @@ -65,7 +65,7 @@ s390_unregister_adapter_interrupt (adapter_int_handler_t handler) ret = -EINVAL; else { adapter_handler = NULL; - synchronize_kernel(); + synchronize_sched(); /* Allow interrupts to complete. */ ret = 0; } sprintf (dbf_txt, "ret:%d", ret); diff --git a/kernel/module.c b/kernel/module.c index 2dbfa07..5734ab0 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -1801,7 +1801,7 @@ sys_init_module(void __user *umod, /* Init routine failed: abort. Try to protect us from buggy refcounters. */ mod->state = MODULE_STATE_GOING; - synchronize_kernel(); + synchronize_sched(); if (mod->unsafe) printk(KERN_ERR "%s: module is now stuck!\n", mod->name); diff --git a/kernel/profile.c b/kernel/profile.c index a66be46..0221a50 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -184,7 +184,7 @@ void unregister_timer_hook(int (*hook)(struct pt_regs *)) WARN_ON(hook != timer_hook); timer_hook = NULL; /* make sure all CPUs see the NULL hook */ - synchronize_kernel(); + synchronize_sched(); /* Allow ongoing interrupts to complete. */ } EXPORT_SYMBOL_GPL(register_timer_hook); diff --git a/mm/slab.c b/mm/slab.c index 771cc09..8407426 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -1666,7 +1666,7 @@ int kmem_cache_destroy(kmem_cache_t * cachep) } if (unlikely(cachep->flags & SLAB_DESTROY_BY_RCU)) - synchronize_kernel(); + synchronize_rcu(); /* no cpu_online check required here since we clear the percpu * array on cpu offline and set this to NULL. diff --git a/net/core/dev.c b/net/core/dev.c index 7bd4cd4..f5f0058 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3091,7 +3091,7 @@ void free_netdev(struct net_device *dev) void synchronize_net(void) { might_sleep(); - synchronize_kernel(); + synchronize_rcu(); } /** -- cgit v0.10.2 From a83f1fe27f7252a2b73b4f22066e92bf99bd595b Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sun, 1 May 2005 08:59:05 -0700 Subject: [PATCH] Update RCU documentation Update the RCU documentation to allow for the new synchronize_rcu() and synchronize_sched() primitives. Fix a few other nits as well. Signed-off-by: Paul E. McKenney Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/RCU/RTFP.txt b/Documentation/RCU/RTFP.txt index 12250b3..9c6d450 100644 --- a/Documentation/RCU/RTFP.txt +++ b/Documentation/RCU/RTFP.txt @@ -108,8 +108,9 @@ year saw a paper describing an RCU implementation of System V IPC 2004 has seen a Linux-Journal article on use of RCU in dcache [McKenney04a], a performance comparison of locking to RCU on several different CPUs [McKenney04b], a dissertation describing use of RCU in a -number of operating-system kernels [PaulEdwardMcKenneyPhD], and a paper -describing how to make RCU safe for soft-realtime applications [Sarma04c]. +number of operating-system kernels [PaulEdwardMcKenneyPhD], a paper +describing how to make RCU safe for soft-realtime applications [Sarma04c], +and a paper describing SELinux performance with RCU [JamesMorris04b]. Bibtex Entries @@ -341,6 +342,17 @@ Dipankar Sarma" ,pages="18-26" } +@techreport{Friedberg03a +,author="Stuart A. Friedberg" +,title="Lock-Free Wild Card Search Data Structure and Method" +,institution="US Patent and Trademark Office" +,address="Washington, DC" +,year="2003" +,number="US Patent 6,662,184 (contributed under GPL)" +,month="December" +,pages="112" +} + @article{McKenney04a ,author="Paul E. McKenney and Dipankar Sarma and Maneesh Soni" ,title="Scaling dcache with {RCU}" @@ -373,6 +385,9 @@ in Operating System Kernels" ,school="OGI School of Science and Engineering at Oregon Health and Sciences University" ,year="2004" +,note="Available: +\url{http://www.rdrop.com/users/paulmck/RCU/RCUdissertation.2004.07.14e1.pdf} +[Viewed October 15, 2004]" } @Conference{Sarma04c @@ -385,3 +400,13 @@ Oregon Health and Sciences University" ,month="June" ,pages="182-191" } + +@unpublished{JamesMorris04b +,Author="James Morris" +,Title="Recent Developments in {SELinux} Kernel Performance" +,month="December" +,year="2004" +,note="Available: +\url{http://www.livejournal.com/users/james_morris/2153.html} +[Viewed December 10, 2004]" +} diff --git a/Documentation/RCU/UP.txt b/Documentation/RCU/UP.txt index 551a803..3bfb84b 100644 --- a/Documentation/RCU/UP.txt +++ b/Documentation/RCU/UP.txt @@ -2,11 +2,11 @@ RCU on Uniprocessor Systems A common misconception is that, on UP systems, the call_rcu() primitive -may immediately invoke its function, and that the synchronize_kernel +may immediately invoke its function, and that the synchronize_rcu() primitive may return immediately. The basis of this misconception is that since there is only one CPU, it should not be necessary to wait for anything else to get done, since there are no other CPUs for -anything else to be happening on. Although this approach will sort of +anything else to be happening on. Although this approach will -sort- -of- work a surprising amount of the time, it is a very bad idea in general. This document presents two examples that demonstrate exactly how bad an idea this is. @@ -44,14 +44,14 @@ its arguments would cause it to fail to make the fundamental guarantee underlying RCU, namely that call_rcu() defers invoking its arguments until all RCU read-side critical sections currently executing have completed. -Quick Quiz: why is it -not- legal to invoke synchronize_kernel() in +Quick Quiz: why is it -not- legal to invoke synchronize_rcu() in this case? Summary Permitting call_rcu() to immediately invoke its arguments or permitting -synchronize_kernel() to immediately return breaks RCU, even on a UP system. +synchronize_rcu() to immediately return breaks RCU, even on a UP system. So do not do it! Even on a UP system, the RCU infrastructure -must- respect grace periods. diff --git a/Documentation/RCU/checklist.txt b/Documentation/RCU/checklist.txt index b3a568a..8f3fb77 100644 --- a/Documentation/RCU/checklist.txt +++ b/Documentation/RCU/checklist.txt @@ -32,7 +32,10 @@ over a rather long period of time, but improvements are always welcome! them -- even x86 allows reads to be reordered), and be prepared to explain why this added complexity is worthwhile. If you choose #c, be prepared to explain how this single task does not - become a major bottleneck on big multiprocessor machines. + become a major bottleneck on big multiprocessor machines (for + example, if the task is updating information relating to itself + that other tasks can read, there by definition can be no + bottleneck). 2. Do the RCU read-side critical sections make proper use of rcu_read_lock() and friends? These primitives are needed @@ -89,27 +92,34 @@ over a rather long period of time, but improvements are always welcome! "_rcu()" list-traversal primitives, such as the list_for_each_entry_rcu(). - b. If the list macros are being used, the list_del_rcu(), - list_add_tail_rcu(), and list_del_rcu() primitives must - be used in order to prevent weakly ordered machines from - misordering structure initialization and pointer planting. + b. If the list macros are being used, the list_add_tail_rcu() + and list_add_rcu() primitives must be used in order + to prevent weakly ordered machines from misordering + structure initialization and pointer planting. Similarly, if the hlist macros are being used, the - hlist_del_rcu() and hlist_add_head_rcu() primitives - are required. + hlist_add_head_rcu() primitive is required. - c. Updates must ensure that initialization of a given + c. If the list macros are being used, the list_del_rcu() + primitive must be used to keep list_del()'s pointer + poisoning from inflicting toxic effects on concurrent + readers. Similarly, if the hlist macros are being used, + the hlist_del_rcu() primitive is required. + + The list_replace_rcu() primitive may be used to + replace an old structure with a new one in an + RCU-protected list. + + d. Updates must ensure that initialization of a given structure happens before pointers to that structure are publicized. Use the rcu_assign_pointer() primitive when publicizing a pointer to a structure that can be traversed by an RCU read-side critical section. - [The rcu_assign_pointer() primitive is in process.] - 5. If call_rcu(), or a related primitive such as call_rcu_bh(), is used, the callback function must be written to be called from softirq context. In particular, it cannot block. -6. Since synchronize_kernel() blocks, it cannot be called from +6. Since synchronize_rcu() can block, it cannot be called from any sort of irq context. 7. If the updater uses call_rcu(), then the corresponding readers @@ -125,9 +135,9 @@ over a rather long period of time, but improvements are always welcome! such cases is a must, of course! And the jury is still out on whether the increased speed is worth it. -8. Although synchronize_kernel() is a bit slower than is call_rcu(), +8. Although synchronize_rcu() is a bit slower than is call_rcu(), it usually results in simpler code. So, unless update performance - is important or the updaters cannot block, synchronize_kernel() + is important or the updaters cannot block, synchronize_rcu() should be used in preference to call_rcu(). 9. All RCU list-traversal primitives, which include @@ -155,3 +165,14 @@ over a rather long period of time, but improvements are always welcome! you -must- use the "_rcu()" variants of the list macros. Failing to do so will break Alpha and confuse people reading your code. + +11. Note that synchronize_rcu() -only- guarantees to wait until + all currently executing rcu_read_lock()-protected RCU read-side + critical sections complete. It does -not- necessarily guarantee + that all currently running interrupts, NMIs, preempt_disable() + code, or idle loops will complete. Therefore, if you do not have + rcu_read_lock()-protected read-side critical sections, do -not- + use synchronize_rcu(). + + If you want to wait for some of these other things, you might + instead need to use synchronize_irq() or synchronize_sched(). diff --git a/Documentation/RCU/listRCU.txt b/Documentation/RCU/listRCU.txt index bda6ead..f8a54fa 100644 --- a/Documentation/RCU/listRCU.txt +++ b/Documentation/RCU/listRCU.txt @@ -32,6 +32,7 @@ implementation of audit_filter_task() might be as follows: enum audit_state state; read_lock(&auditsc_lock); + /* Note: audit_netlink_sem held by caller. */ list_for_each_entry(e, &audit_tsklist, list) { if (audit_filter_rules(tsk, &e->rule, NULL, &state)) { read_unlock(&auditsc_lock); @@ -55,6 +56,7 @@ This means that RCU can be easily applied to the read side, as follows: enum audit_state state; rcu_read_lock(); + /* Note: audit_netlink_sem held by caller. */ list_for_each_entry_rcu(e, &audit_tsklist, list) { if (audit_filter_rules(tsk, &e->rule, NULL, &state)) { rcu_read_unlock(); @@ -139,12 +141,15 @@ Normally, the write_lock() and write_unlock() would be replaced by a spin_lock() and a spin_unlock(), but in this case, all callers hold audit_netlink_sem, so no additional locking is required. The auditsc_lock can therefore be eliminated, since use of RCU eliminates the need for -writers to exclude readers. +writers to exclude readers. Normally, the write_lock() calls would +be converted into spin_lock() calls. The list_del(), list_add(), and list_add_tail() primitives have been replaced by list_del_rcu(), list_add_rcu(), and list_add_tail_rcu(). The _rcu() list-manipulation primitives add memory barriers that are -needed on weakly ordered CPUs (most of them!). +needed on weakly ordered CPUs (most of them!). The list_del_rcu() +primitive omits the pointer poisoning debug-assist code that would +otherwise cause concurrent readers to fail spectacularly. So, when readers can tolerate stale data and when entries are either added or deleted, without in-place modification, it is very easy to use RCU! @@ -166,6 +171,7 @@ otherwise, the added fields would need to be filled in): struct audit_newentry *ne; write_lock(&auditsc_lock); + /* Note: audit_netlink_sem held by caller. */ list_for_each_entry(e, list, list) { if (!audit_compare_rule(rule, &e->rule)) { e->rule.action = newaction; @@ -199,8 +205,7 @@ RCU ("read-copy update") its name. The RCU code is as follows: audit_copy_rule(&ne->rule, &e->rule); ne->rule.action = newaction; ne->rule.file_count = newfield_count; - list_add_rcu(ne, e); - list_del(e); + list_replace_rcu(e, ne); call_rcu(&e->rcu, audit_free_rule, e); return 0; } diff --git a/Documentation/RCU/rcu.txt b/Documentation/RCU/rcu.txt index 7e0c2ab..eb44400 100644 --- a/Documentation/RCU/rcu.txt +++ b/Documentation/RCU/rcu.txt @@ -43,7 +43,9 @@ o If I am running on a uniprocessor kernel, which can only do one o How can I see where RCU is currently used in the Linux kernel? - Search for "rcu_read_lock", "call_rcu", and "synchronize_kernel". + Search for "rcu_read_lock", "rcu_read_unlock", "call_rcu", + "rcu_read_lock_bh", "rcu_read_unlock_bh", "call_rcu_bh", + "synchronize_rcu", and "synchronize_net". o What guidelines should I follow when writing code that uses RCU? -- cgit v0.10.2 From 9a3bb3017383fbb6fe56431d17f60bd0d50f0717 Mon Sep 17 00:00:00 2001 From: Paolo 'Blaisorblade' Giarrusso Date: Sun, 1 May 2005 08:59:05 -0700 Subject: [PATCH] reiserfs: make resize option auto-get new device size It's trivial for the resize option to auto-get the underlying device size, while it's harder for the user. I've copied the code from jfs. Since of the different reiserfs option parser (which does not use the superior match_token used by almost every other filesystem), I've had to use the "resize=auto" and not "resize" option to specify this behaviour. Changing the option parser to the kernel one wouldn't be bad but I've no time to do this cleanup in this moment. Btw, the mount(8) man page should be updated to include this option. Cc the relevant people, please (I hope I cc'ed the right people). Cc: Cc: Cc: Cc: Alex Zarochentsev Signed-off-by: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index bcdf243..bc5e889 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -889,12 +889,18 @@ static int reiserfs_parse_options (struct super_block * s, char * options, /* st char * p; p = NULL; - /* "resize=NNN" */ - *blocks = simple_strtoul (arg, &p, 0); - if (*p != '\0') { - /* NNN does not look like a number */ - reiserfs_warning (s, "reiserfs_parse_options: bad value %s", arg); - return 0; + /* "resize=NNN" or "resize=auto" */ + + if (!strcmp(arg, "auto")) { + /* From JFS code, to auto-get the size.*/ + *blocks = s->s_bdev->bd_inode->i_size >> s->s_blocksize_bits; + } else { + *blocks = simple_strtoul (arg, &p, 0); + if (*p != '\0') { + /* NNN does not look like a number */ + reiserfs_warning (s, "reiserfs_parse_options: bad value %s", arg); + return 0; + } } } @@ -903,7 +909,8 @@ static int reiserfs_parse_options (struct super_block * s, char * options, /* st unsigned long val = simple_strtoul (arg, &p, 0); /* commit=NNN (time in seconds) */ if ( *p != '\0' || val >= (unsigned int)-1) { - reiserfs_warning (s, "reiserfs_parse_options: bad value %s", arg); return 0; + reiserfs_warning (s, "reiserfs_parse_options: bad value %s", arg); + return 0; } *commit_max_age = (unsigned int)val; } -- cgit v0.10.2 From 0c8b971ebb21dc33271c38d17fd58b8072009dc5 Mon Sep 17 00:00:00 2001 From: Peter Missel Date: Sun, 1 May 2005 08:59:05 -0700 Subject: [PATCH] LifeView FlyTV Platinum FM: Remote Control support Subject says it ... this card's IR microcontroller design and attachment are compatible to the company's previous designs, so the patch was as simple as it gets. DESC LifeView FlyTV Platinum FM: GPIO usage EDESC From: Peter Missel This is take two of a patch that should have appeared two days ago, before yesterday's "remote control" patch for the same card. This patch sets unconnected GPIO to Output to keep them from floating (just good driver writing practice, being nice to the chip), and uses GPIO16 to switch TV vs. FM - this pin switches inputs onto the tuner, as well as the audio output from the tuner into the 7135 SIF input. Consequently, FM radio support is being un-commented because it's now working (sort of, see below). These two patches get the card almost fully operational; there appears to be a bug in tda8290.c remaining that puts an offset onto the tuned frequency in FM radio mode. We're investigating. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/media/video/saa7134/saa7134-cards.c b/drivers/media/video/saa7134/saa7134-cards.c index 180d317..cfe4b57 100644 --- a/drivers/media/video/saa7134/saa7134-cards.c +++ b/drivers/media/video/saa7134/saa7134-cards.c @@ -1948,6 +1948,7 @@ int saa7134_board_init1(struct saa7134_dev *dev) dev->has_remote = 1; board_flyvideo(dev); break; + case SAA7134_BOARD_FLYTVPLATINUM_FM: case SAA7134_BOARD_CINERGY400: case SAA7134_BOARD_CINERGY600: case SAA7134_BOARD_CINERGY600_MK3: diff --git a/drivers/media/video/saa7134/saa7134-input.c b/drivers/media/video/saa7134/saa7134-input.c index 727d437..ca50cf5 100644 --- a/drivers/media/video/saa7134/saa7134-input.c +++ b/drivers/media/video/saa7134/saa7134-input.c @@ -379,6 +379,7 @@ int saa7134_input_init1(struct saa7134_dev *dev) switch (dev->board) { case SAA7134_BOARD_FLYVIDEO2000: case SAA7134_BOARD_FLYVIDEO3000: + case SAA7134_BOARD_FLYTVPLATINUM_FM: ir_codes = flyvideo_codes; mask_keycode = 0xEC00000; mask_keydown = 0x0040000; -- cgit v0.10.2 From 6c9e7376bed151d1655c12d2d5f5cc96bfb83dbd Mon Sep 17 00:00:00 2001 From: Peter Missel Date: Sun, 1 May 2005 08:59:05 -0700 Subject: [PATCH] LifeView FlyTV Platinum FM: GPIO usage This is take two of a patch that should have appeared two days ago, before yesterday's "remote control" patch for the same card. This patch sets unconnected GPIO to Output to keep them from floating (just good driver writing practice, being nice to the chip), and uses GPIO16 to switch TV vs. FM - this pin switches inputs onto the tuner, as well as the audio output from the tuner into the 7135 SIF input. Consequently, FM radio support is being un-commented because it's now working (sort of, see below). These two patches get the card almost fully operational; there appears to be a bug in tda8290.c remaining that puts an offset onto the tuned frequency in FM radio mode. We're investigating. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/media/video/saa7134/saa7134-cards.c b/drivers/media/video/saa7134/saa7134-cards.c index cfe4b57..72a7b24 100644 --- a/drivers/media/video/saa7134/saa7134-cards.c +++ b/drivers/media/video/saa7134/saa7134-cards.c @@ -183,12 +183,12 @@ struct saa7134_board saa7134_boards[] = { .name = "LifeView FlyTV Platinum FM", .audio_clock = 0x00200000, .tuner_type = TUNER_PHILIPS_TDA8290, -// .gpiomask = 0xe000, + .gpiomask = 0x1E000, /* Set GP16 and unused 15,14,13 to Output */ .inputs = {{ .name = name_tv, .vmux = 1, .amux = TV, -// .gpio = 0x0000, + .gpio = 0x10000, /* GP16=1 selects TV input */ .tv = 1, },{ /* .name = name_tv_mono, @@ -212,12 +212,12 @@ struct saa7134_board saa7134_boards[] = { .amux = LINE2, // .gpio = 0x4000, }}, -/* .radio = { + .radio = { .name = name_radio, - .amux = LINE2, - .gpio = 0x2000, + .amux = TV, + .gpio = 0x00000, /* GP16=0 selects FM radio antenna */ }, -*/ }, + }, [SAA7134_BOARD_EMPRESS] = { /* "Gert Vervoort" */ .name = "EMPRESS", -- cgit v0.10.2 From 41f11a4fa378201e902892130b11d78cf7cf8e10 Mon Sep 17 00:00:00 2001 From: Yoshinori Sato Date: Sun, 1 May 2005 08:59:06 -0700 Subject: [PATCH] kallsyms C_SYMBOL_PREFIX support kallsyms does not consider SYMBOL_PREFIX of C. Consequently it does not work on architectures using that prefix character (h8300, v850). Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/scripts/kallsyms.c b/scripts/kallsyms.c index 090ffda..fe11df8 100644 --- a/scripts/kallsyms.c +++ b/scripts/kallsyms.c @@ -69,6 +69,7 @@ static struct sym_entry *table; static int size, cnt; static unsigned long long _stext, _etext, _sinittext, _einittext; static int all_symbols = 0; +static char symbol_prefix_char = '\0'; struct token { unsigned char data[MAX_TOK_SIZE]; @@ -93,7 +94,7 @@ unsigned char best_table_len[256]; static void usage(void) { - fprintf(stderr, "Usage: kallsyms [--all-symbols] < in.map > out.S\n"); + fprintf(stderr, "Usage: kallsyms [--all-symbols] [--symbol-prefix=] < in.map > out.S\n"); exit(1); } @@ -112,6 +113,7 @@ static int read_symbol(FILE *in, struct sym_entry *s) { char str[500]; + char *sym; int rc; rc = fscanf(in, "%llx %c %499s\n", &s->addr, &s->type, str); @@ -123,27 +125,32 @@ read_symbol(FILE *in, struct sym_entry *s) return -1; } + sym = str; + /* skip prefix char */ + if (symbol_prefix_char && str[0] == symbol_prefix_char) + sym++; + /* Ignore most absolute/undefined (?) symbols. */ - if (strcmp(str, "_stext") == 0) + if (strcmp(sym, "_stext") == 0) _stext = s->addr; - else if (strcmp(str, "_etext") == 0) + else if (strcmp(sym, "_etext") == 0) _etext = s->addr; - else if (strcmp(str, "_sinittext") == 0) + else if (strcmp(sym, "_sinittext") == 0) _sinittext = s->addr; - else if (strcmp(str, "_einittext") == 0) + else if (strcmp(sym, "_einittext") == 0) _einittext = s->addr; else if (toupper(s->type) == 'A') { /* Keep these useful absolute symbols */ - if (strcmp(str, "__kernel_syscall_via_break") && - strcmp(str, "__kernel_syscall_via_epc") && - strcmp(str, "__kernel_sigtramp") && - strcmp(str, "__gp")) + if (strcmp(sym, "__kernel_syscall_via_break") && + strcmp(sym, "__kernel_syscall_via_epc") && + strcmp(sym, "__kernel_sigtramp") && + strcmp(sym, "__gp")) return -1; } else if (toupper(s->type) == 'U' || - is_arm_mapping_symbol(str)) + is_arm_mapping_symbol(sym)) return -1; /* include the type field in the symbol name, so that it gets @@ -177,6 +184,11 @@ symbol_valid(struct sym_entry *s) "_SDA2_BASE_", /* ppc */ NULL }; int i; + int offset = 1; + + /* skip prefix char */ + if (symbol_prefix_char && *(s->sym + 1) == symbol_prefix_char) + offset++; /* if --all-symbols is not specified, then symbols outside the text * and inittext sections are discarded */ @@ -190,17 +202,17 @@ symbol_valid(struct sym_entry *s) * they may get dropped in pass 2, which breaks the kallsyms * rules. */ - if ((s->addr == _etext && strcmp(s->sym + 1, "_etext")) || - (s->addr == _einittext && strcmp(s->sym + 1, "_einittext"))) + if ((s->addr == _etext && strcmp(s->sym + offset, "_etext")) || + (s->addr == _einittext && strcmp(s->sym + offset, "_einittext"))) return 0; } /* Exclude symbols which vary between passes. */ - if (strstr(s->sym + 1, "_compiled.")) + if (strstr(s->sym + offset, "_compiled.")) return 0; for (i = 0; special_symbols[i]; i++) - if( strcmp(s->sym + 1, special_symbols[i]) == 0 ) + if( strcmp(s->sym + offset, special_symbols[i]) == 0 ) return 0; return 1; @@ -225,9 +237,15 @@ read_map(FILE *in) static void output_label(char *label) { - printf(".globl %s\n",label); + if (symbol_prefix_char) + printf(".globl %c%s\n", symbol_prefix_char, label); + else + printf(".globl %s\n", label); printf("\tALGN\n"); - printf("%s:\n",label); + if (symbol_prefix_char) + printf("%c%s:\n", symbol_prefix_char, label); + else + printf("%s:\n", label); } /* uncompress a compressed symbol. When this function is called, the best table @@ -665,6 +683,13 @@ static void optimize_token_table(void) insert_real_symbols_in_table(); + /* When valid symbol is not registered, exit to error */ + if (good_head.left == good_head.right && + bad_head.left == bad_head.right) { + fprintf(stderr, "No valid symbol.\n"); + exit(1); + } + optimize_result(); } @@ -672,9 +697,21 @@ static void optimize_token_table(void) int main(int argc, char **argv) { - if (argc == 2 && strcmp(argv[1], "--all-symbols") == 0) - all_symbols = 1; - else if (argc != 1) + if (argc >= 2) { + int i; + for (i = 1; i < argc; i++) { + if(strcmp(argv[i], "--all-symbols") == 0) + all_symbols = 1; + else if (strncmp(argv[i], "--symbol-prefix=", 16) == 0) { + char *p = &argv[i][16]; + /* skip quote */ + if ((*p == '"' && *(p+2) == '"') || (*p == '\'' && *(p+2) == '\'')) + p++; + symbol_prefix_char = *p; + } else + usage(); + } + } else if (argc != 1) usage(); read_map(stdin); @@ -683,4 +720,3 @@ main(int argc, char **argv) return 0; } - -- cgit v0.10.2 From 1181c1f923c349acaa01bca40fe600584f265132 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Sun, 1 May 2005 08:59:06 -0700 Subject: [PATCH] noop-iosched: kill O(N) merge scan Profiling hit rates on merging shows that the last merge hint works extremely well for most work loads. So lets kill the linear merge scan in noop-iosched, so it provides O(1) run time for any operation. Testing credits go to Ken Chen from Intel. Signed-off-by: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/block/noop-iosched.c b/drivers/block/noop-iosched.c index 888c477..b1730b6 100644 --- a/drivers/block/noop-iosched.c +++ b/drivers/block/noop-iosched.c @@ -13,34 +13,13 @@ static int elevator_noop_merge(request_queue_t *q, struct request **req, struct bio *bio) { - struct list_head *entry = &q->queue_head; - struct request *__rq; int ret; - if ((ret = elv_try_last_merge(q, bio))) { + ret = elv_try_last_merge(q, bio); + if (ret != ELEVATOR_NO_MERGE) *req = q->last_merge; - return ret; - } - while ((entry = entry->prev) != &q->queue_head) { - __rq = list_entry_rq(entry); - - if (__rq->flags & (REQ_SOFTBARRIER | REQ_HARDBARRIER)) - break; - else if (__rq->flags & REQ_STARTED) - break; - - if (!blk_fs_request(__rq)) - continue; - - if ((ret = elv_try_merge(__rq, bio))) { - *req = __rq; - q->last_merge = __rq; - return ret; - } - } - - return ELEVATOR_NO_MERGE; + return ret; } static void elevator_noop_merge_requests(request_queue_t *q, struct request *req, -- cgit v0.10.2 From 4750e2c0c59e0c84c6c036b3d96ebd88365ae7ee Mon Sep 17 00:00:00 2001 From: Joe Korty Date: Sun, 1 May 2005 08:59:06 -0700 Subject: [PATCH] add EOWNERDEAD and ENOTRECOVERABLE version 2 Add EOWNERDEAD and ENOTRECOVERABLE to all architectures. This is to support the upcoming patches for robust mutexes. We normally don't reserve parts of the name/number space for external patches, but robust mutexes are sufficiently popular and important to justify it in this case. Signed-off-by: Joe Korty Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-alpha/errno.h b/include/asm-alpha/errno.h index c85ab6b..69e2655 100644 --- a/include/asm-alpha/errno.h +++ b/include/asm-alpha/errno.h @@ -116,4 +116,8 @@ #define EKEYREVOKED 134 /* Key has been revoked */ #define EKEYREJECTED 135 /* Key was rejected by service */ +/* for robust mutexes */ +#define EOWNERDEAD 136 /* Owner died */ +#define ENOTRECOVERABLE 137 /* State not recoverable */ + #endif diff --git a/include/asm-generic/errno.h b/include/asm-generic/errno.h index 4dd2384..e8852c0 100644 --- a/include/asm-generic/errno.h +++ b/include/asm-generic/errno.h @@ -102,4 +102,8 @@ #define EKEYREVOKED 128 /* Key has been revoked */ #define EKEYREJECTED 129 /* Key was rejected by service */ +/* for robust mutexes */ +#define EOWNERDEAD 130 /* Owner died */ +#define ENOTRECOVERABLE 131 /* State not recoverable */ + #endif diff --git a/include/asm-mips/errno.h b/include/asm-mips/errno.h index 2b458f9..3c0d840 100644 --- a/include/asm-mips/errno.h +++ b/include/asm-mips/errno.h @@ -115,6 +115,10 @@ #define EKEYREVOKED 163 /* Key has been revoked */ #define EKEYREJECTED 164 /* Key was rejected by service */ +/* for robust mutexes */ +#define EOWNERDEAD 165 /* Owner died */ +#define ENOTRECOVERABLE 166 /* State not recoverable */ + #define EDQUOT 1133 /* Quota exceeded */ #ifdef __KERNEL__ diff --git a/include/asm-parisc/errno.h b/include/asm-parisc/errno.h index a10f109..08464c4 100644 --- a/include/asm-parisc/errno.h +++ b/include/asm-parisc/errno.h @@ -115,5 +115,9 @@ #define ENOTSUP 252 /* Function not implemented (POSIX.4 / HPUX) */ #define ECANCELLED 253 /* aio request was canceled before complete (POSIX.4 / HPUX) */ +/* for robust mutexes */ +#define EOWNERDEAD 254 /* Owner died */ +#define ENOTRECOVERABLE 255 /* State not recoverable */ + #endif diff --git a/include/asm-sparc/errno.h b/include/asm-sparc/errno.h index 8c01c5f3..ed41c8b 100644 --- a/include/asm-sparc/errno.h +++ b/include/asm-sparc/errno.h @@ -107,4 +107,8 @@ #define EKEYREVOKED 130 /* Key has been revoked */ #define EKEYREJECTED 131 /* Key was rejected by service */ +/* for robust mutexes */ +#define EOWNERDEAD 132 /* Owner died */ +#define ENOTRECOVERABLE 133 /* State not recoverable */ + #endif diff --git a/include/asm-sparc64/errno.h b/include/asm-sparc64/errno.h index cc98a73..ea3509e 100644 --- a/include/asm-sparc64/errno.h +++ b/include/asm-sparc64/errno.h @@ -107,4 +107,8 @@ #define EKEYREVOKED 130 /* Key has been revoked */ #define EKEYREJECTED 131 /* Key was rejected by service */ +/* for robust mutexes */ +#define EOWNERDEAD 132 /* Owner died */ +#define ENOTRECOVERABLE 133 /* State not recoverable */ + #endif /* !(_SPARC64_ERRNO_H) */ -- cgit v0.10.2 From 40be0c28b33ff0821594a3fa7126354dfe6eccd1 Mon Sep 17 00:00:00 2001 From: Lars Marowsky-Bree Date: Sun, 1 May 2005 08:59:07 -0700 Subject: [PATCH] nbd: Don't create all MAX_NBD devices by default all the time This patches adds the "nbds_max" parameter to the nbd kernel module, which limits the number of nbds allocated. Previously, always all 128 entries were allocated unconditionally, which used to waste resources and needlessly flood the hotplug system with events. (Defaults to 16 now.) Signed-off-by: Lars Marowsky-Bree Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index efdf044..9e268dd 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -78,6 +78,7 @@ #define DBG_RX 0x0200 #define DBG_TX 0x0400 static unsigned int debugflags; +static unsigned int nbds_max = 16; #endif /* NDEBUG */ static struct nbd_device nbd_dev[MAX_NBD]; @@ -647,7 +648,13 @@ static int __init nbd_init(void) return -EIO; } - for (i = 0; i < MAX_NBD; i++) { + if (nbds_max > MAX_NBD) { + printk(KERN_CRIT "nbd: cannot allocate more than %u nbds; %u requested.\n", MAX_NBD, + nbds_max); + return -EINVAL; + } + + for (i = 0; i < nbds_max; i++) { struct gendisk *disk = alloc_disk(1); if (!disk) goto out; @@ -673,7 +680,7 @@ static int __init nbd_init(void) dprintk(DBG_INIT, "nbd: debugflags=0x%x\n", debugflags); devfs_mk_dir("nbd"); - for (i = 0; i < MAX_NBD; i++) { + for (i = 0; i < nbds_max; i++) { struct gendisk *disk = nbd_dev[i].disk; nbd_dev[i].file = NULL; nbd_dev[i].magic = LO_MAGIC; @@ -706,8 +713,9 @@ out: static void __exit nbd_cleanup(void) { int i; - for (i = 0; i < MAX_NBD; i++) { + for (i = 0; i < nbds_max; i++) { struct gendisk *disk = nbd_dev[i].disk; + nbd_dev[i].magic = 0; if (disk) { del_gendisk(disk); blk_cleanup_queue(disk->queue); @@ -725,6 +733,8 @@ module_exit(nbd_cleanup); MODULE_DESCRIPTION("Network Block Device"); MODULE_LICENSE("GPL"); +module_param(nbds_max, int, 0444); +MODULE_PARM_DESC(nbds_max, "How many network block devices to initialize."); #ifndef NDEBUG module_param(debugflags, int, 0644); MODULE_PARM_DESC(debugflags, "flags for controlling debug output"); -- cgit v0.10.2 From 127144df4ce817ad648af15a3983c8d52aacf670 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Sun, 1 May 2005 08:59:07 -0700 Subject: [PATCH] Fix rewriting on a full reiserfs filesystem Allow rewriting of a file and extending a file upto the end of the allocated block on a full filesystem. From: Chris Mason Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c index 2695011..f6860e8 100644 --- a/fs/reiserfs/file.c +++ b/fs/reiserfs/file.c @@ -1284,10 +1284,11 @@ static ssize_t reiserfs_file_write( struct file *file, /* the file we are going reiserfs_claim_blocks_to_be_allocated(inode->i_sb, num_pages << (PAGE_CACHE_SHIFT - inode->i_blkbits)); reiserfs_write_unlock(inode->i_sb); - if ( !num_pages ) { /* If we do not have enough space even for */ - res = -ENOSPC; /* single page, return -ENOSPC */ - if ( pos > (inode->i_size & (inode->i_sb->s_blocksize-1))) - break; // In case we are writing past the file end, break. + if ( !num_pages ) { /* If we do not have enough space even for a single page... */ + if ( pos > inode->i_size+inode->i_sb->s_blocksize-(pos & (inode->i_sb->s_blocksize-1))) { + res = -ENOSPC; + break; // In case we are writing past the end of the last file block, break. + } // Otherwise we are possibly overwriting the file, so // let's set write size to be equal or less than blocksize. // This way we get it correctly for file holes. -- cgit v0.10.2 From a40920b42ae232fac514cc4a1eb92996114af340 Mon Sep 17 00:00:00 2001 From: Bill Nottingham Date: Sun, 1 May 2005 08:59:07 -0700 Subject: [PATCH] vgacon: set vc_hi_font_mask correctly When allocating a new VC with vgacon_init(), the font is shared across all the VGA consoles. However, the font mask was always set to the default value of zero in visual_init(), even if we were using 512 character fonts at the time. Moreover, code in vgacon.c:vga_do_font_op() didn't reset the mask if the console driver thinks it's already in 512 character mode. This means that to *fix* it, you'd actually have to take the console out of 512 character mode and then set it back. The attached sets vc_hi_font_mask in vgacon_init() for any new consoles opened if the vgacon driver is already in 512 character mode, solving this. This bug goes back to 2.4.18 at least, probably earlier. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/console/vgacon.c b/drivers/video/console/vgacon.c index 7d1ae06..bcf59b2 100644 --- a/drivers/video/console/vgacon.c +++ b/drivers/video/console/vgacon.c @@ -337,6 +337,8 @@ static void vgacon_init(struct vc_data *c, int init) c->vc_scan_lines = vga_scan_lines; c->vc_font.height = vga_video_font_height; c->vc_complement_mask = 0x7700; + if (vga_512_chars) + c->vc_hi_font_mask = 0x0800; p = *c->vc_uni_pagedir_loc; if (c->vc_uni_pagedir_loc == &c->vc_uni_pagedir || !--c->vc_uni_pagedir_loc[1]) -- cgit v0.10.2 From 696f9486d0207d499391004f5bc9bd7c0e6ae82f Mon Sep 17 00:00:00 2001 From: Joel Becker Date: Sun, 1 May 2005 08:59:08 -0700 Subject: [PATCH] hangcheck-timer: Update to 0.9.0. I recently realized that the in-kernel copy of hangcheck-timer was quite stale. Here's the latest. It adds support for s390, ppc64, and ia64 too. Signed-off-by: Joel Becker Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig index 97ac4ed..e162dab 100644 --- a/drivers/char/Kconfig +++ b/drivers/char/Kconfig @@ -982,7 +982,7 @@ config MAX_RAW_DEVS config HANGCHECK_TIMER tristate "Hangcheck timer" - depends on X86_64 || X86 + depends on X86_64 || X86 || IA64 || PPC64 || ARCH_S390 help The hangcheck-timer module detects when the system has gone out to lunch past a certain margin. It can reboot the system diff --git a/drivers/char/hangcheck-timer.c b/drivers/char/hangcheck-timer.c index 83d6b37..78e650f 100644 --- a/drivers/char/hangcheck-timer.c +++ b/drivers/char/hangcheck-timer.c @@ -3,7 +3,7 @@ * * Driver for a little io fencing timer. * - * Copyright (C) 2002 Oracle Corporation. All rights reserved. + * Copyright (C) 2002, 2003 Oracle. All rights reserved. * * Author: Joel Becker * @@ -44,11 +44,14 @@ #include #include #include +#include #include +#include #include +#include -#define VERSION_STR "0.5.0" +#define VERSION_STR "0.9.0" #define DEFAULT_IOFENCE_MARGIN 60 /* Default fudge factor, in seconds */ #define DEFAULT_IOFENCE_TICK 180 /* Default timer timeout, in seconds */ @@ -56,18 +59,89 @@ static int hangcheck_tick = DEFAULT_IOFENCE_TICK; static int hangcheck_margin = DEFAULT_IOFENCE_MARGIN; static int hangcheck_reboot; /* Defaults to not reboot */ +static int hangcheck_dump_tasks; /* Defaults to not dumping SysRQ T */ -/* Driver options */ +/* options - modular */ module_param(hangcheck_tick, int, 0); MODULE_PARM_DESC(hangcheck_tick, "Timer delay."); module_param(hangcheck_margin, int, 0); MODULE_PARM_DESC(hangcheck_margin, "If the hangcheck timer has been delayed more than hangcheck_margin seconds, the driver will fire."); module_param(hangcheck_reboot, int, 0); MODULE_PARM_DESC(hangcheck_reboot, "If nonzero, the machine will reboot when the timer margin is exceeded."); +module_param(hangcheck_dump_tasks, int, 0); +MODULE_PARM_DESC(hangcheck_dump_tasks, "If nonzero, the machine will dump the system task state when the timer margin is exceeded."); -MODULE_AUTHOR("Joel Becker"); +MODULE_AUTHOR("Oracle"); MODULE_DESCRIPTION("Hangcheck-timer detects when the system has gone out to lunch past a certain margin."); MODULE_LICENSE("GPL"); +MODULE_VERSION(VERSION_STR); + +/* options - nonmodular */ +#ifndef MODULE + +static int __init hangcheck_parse_tick(char *str) +{ + int par; + if (get_option(&str,&par)) + hangcheck_tick = par; + return 1; +} + +static int __init hangcheck_parse_margin(char *str) +{ + int par; + if (get_option(&str,&par)) + hangcheck_margin = par; + return 1; +} + +static int __init hangcheck_parse_reboot(char *str) +{ + int par; + if (get_option(&str,&par)) + hangcheck_reboot = par; + return 1; +} + +static int __init hangcheck_parse_dump_tasks(char *str) +{ + int par; + if (get_option(&str,&par)) + hangcheck_dump_tasks = par; + return 1; +} + +__setup("hcheck_tick", hangcheck_parse_tick); +__setup("hcheck_margin", hangcheck_parse_margin); +__setup("hcheck_reboot", hangcheck_parse_reboot); +__setup("hcheck_dump_tasks", hangcheck_parse_dump_tasks); +#endif /* not MODULE */ + +#if defined(CONFIG_X86) || defined(CONFIG_X86_64) +# define HAVE_MONOTONIC +# define TIMER_FREQ 1000000000ULL +#elif defined(CONFIG_ARCH_S390) +/* FA240000 is 1 Second in the IBM time universe (Page 4-38 Principles of Op for zSeries */ +# define TIMER_FREQ 0xFA240000ULL +#elif defined(CONFIG_IA64) +# define TIMER_FREQ ((unsigned long long)local_cpu_data->itc_freq) +#elif defined(CONFIG_PPC64) +# define TIMER_FREQ (HZ*loops_per_jiffy) +#endif + +#ifdef HAVE_MONOTONIC +extern unsigned long long monotonic_clock(void); +#else +static inline unsigned long long monotonic_clock(void) +{ +# ifdef __s390__ + /* returns the TOD. see 4-38 Principles of Op of zSeries */ + return get_clock(); +# else + return get_cycles(); +# endif /* __s390__ */ +} +#endif /* HAVE_MONOTONIC */ /* Last time scheduled */ @@ -78,7 +152,6 @@ static void hangcheck_fire(unsigned long); static struct timer_list hangcheck_ticktock = TIMER_INITIALIZER(hangcheck_fire, 0, 0); -extern unsigned long long monotonic_clock(void); static void hangcheck_fire(unsigned long data) { @@ -92,6 +165,12 @@ static void hangcheck_fire(unsigned long data) tsc_diff = (cur_tsc + (~0ULL - hangcheck_tsc)); /* or something */ if (tsc_diff > hangcheck_tsc_margin) { + if (hangcheck_dump_tasks) { + printk(KERN_CRIT "Hangcheck: Task state:\n"); +#ifdef CONFIG_MAGIC_SYSRQ + handle_sysrq('t', NULL, NULL); +#endif /* CONFIG_MAGIC_SYSRQ */ + } if (hangcheck_reboot) { printk(KERN_CRIT "Hangcheck: hangcheck is restarting the machine.\n"); machine_restart(NULL); @@ -108,10 +187,16 @@ static int __init hangcheck_init(void) { printk("Hangcheck: starting hangcheck timer %s (tick is %d seconds, margin is %d seconds).\n", VERSION_STR, hangcheck_tick, hangcheck_margin); - - hangcheck_tsc_margin = hangcheck_margin + hangcheck_tick; - hangcheck_tsc_margin *= 1000000000; - +#if defined (HAVE_MONOTONIC) + printk("Hangcheck: Using monotonic_clock().\n"); +#elif defined(__s390__) + printk("Hangcheck: Using TOD.\n"); +#else + printk("Hangcheck: Using get_cycles().\n"); +#endif /* HAVE_MONOTONIC */ + hangcheck_tsc_margin = + (unsigned long long)(hangcheck_margin + hangcheck_tick); + hangcheck_tsc_margin *= (unsigned long long)TIMER_FREQ; hangcheck_tsc = monotonic_clock(); mod_timer(&hangcheck_ticktock, jiffies + (hangcheck_tick*HZ)); @@ -123,6 +208,7 @@ static int __init hangcheck_init(void) static void __exit hangcheck_exit(void) { del_timer_sync(&hangcheck_ticktock); + printk("Hangcheck: Stopped hangcheck timer.\n"); } module_init(hangcheck_init); -- cgit v0.10.2 From a71c1ab50a2a0f4dd9834bf5a917a2f064535c6b Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Sun, 1 May 2005 08:59:08 -0700 Subject: [PATCH] consolidate SIGEV_PAD_SIZE Discussing with Matthew Wilcox some of his outstanding patches lead me to this patch (among others). The preamble in struct sigevent can be expressed independently of the architecture. Also use __ARCH_SI_PREAMBLE_SIZE on ia64. Signed-off-by: Stephen Rothwell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/asm-alpha/siginfo.h b/include/asm-alpha/siginfo.h index 86bcab5..9822362 100644 --- a/include/asm-alpha/siginfo.h +++ b/include/asm-alpha/siginfo.h @@ -4,8 +4,6 @@ #define __ARCH_SI_PREAMBLE_SIZE (4 * sizeof(int)) #define __ARCH_SI_TRAPNO -#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE/sizeof(int)) - 4) - #include #endif diff --git a/include/asm-generic/siginfo.h b/include/asm-generic/siginfo.h index 9cac8e8..8786e01 100644 --- a/include/asm-generic/siginfo.h +++ b/include/asm-generic/siginfo.h @@ -236,11 +236,18 @@ typedef struct siginfo { #define SIGEV_THREAD 2 /* deliver via thread creation */ #define SIGEV_THREAD_ID 4 /* deliver to thread */ -#define SIGEV_MAX_SIZE 64 -#ifndef SIGEV_PAD_SIZE -#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE/sizeof(int)) - 3) +/* + * This works because the alignment is ok on all current architectures + * but we leave open this being overridden in the future + */ +#ifndef __ARCH_SIGEV_PREAMBLE_SIZE +#define __ARCH_SIGEV_PREAMBLE_SIZE (sizeof(int) * 2 + sizeof(sigval_t)) #endif +#define SIGEV_MAX_SIZE 64 +#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE - __ARCH_SIGEV_PREAMBLE_SIZE) \ + / sizeof(int)) + typedef struct sigevent { sigval_t sigev_value; int sigev_signo; diff --git a/include/asm-ia64/siginfo.h b/include/asm-ia64/siginfo.h index d55f139..9294e4b 100644 --- a/include/asm-ia64/siginfo.h +++ b/include/asm-ia64/siginfo.h @@ -8,9 +8,7 @@ * David Mosberger-Tang , Hewlett-Packard Co */ -#define SI_PAD_SIZE ((SI_MAX_SIZE/sizeof(int)) - 4) - -#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE/sizeof(int)) - 4) +#define __ARCH_SI_PREAMBLE_SIZE (4 * sizeof(int)) #define HAVE_ARCH_SIGINFO_T #define HAVE_ARCH_COPY_SIGINFO diff --git a/include/asm-mips/siginfo.h b/include/asm-mips/siginfo.h index 8ddd3c9..a0e26e6 100644 --- a/include/asm-mips/siginfo.h +++ b/include/asm-mips/siginfo.h @@ -11,8 +11,6 @@ #include -#define SIGEV_HEAD_SIZE (sizeof(long) + 2*sizeof(int)) -#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE-SIGEV_HEAD_SIZE) / sizeof(int)) #undef __ARCH_SI_TRAPNO /* exception code needs to fill this ... */ #define HAVE_ARCH_SIGINFO_T diff --git a/include/asm-s390/siginfo.h b/include/asm-s390/siginfo.h index 7230353..e0ff1ab 100644 --- a/include/asm-s390/siginfo.h +++ b/include/asm-s390/siginfo.h @@ -13,12 +13,6 @@ #define __ARCH_SI_PREAMBLE_SIZE (4 * sizeof(int)) #endif -#ifdef CONFIG_ARCH_S390X -#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE/sizeof(int)) - 4) -#else -#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE/sizeof(int)) - 3) -#endif - #include #endif diff --git a/include/asm-sparc64/siginfo.h b/include/asm-sparc64/siginfo.h index 7160449..df17e47 100644 --- a/include/asm-sparc64/siginfo.h +++ b/include/asm-sparc64/siginfo.h @@ -3,8 +3,6 @@ #define SI_PAD_SIZE32 ((SI_MAX_SIZE/sizeof(int)) - 3) -#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE/sizeof(int)) - 4) - #define __ARCH_SI_PREAMBLE_SIZE (4 * sizeof(int)) #define __ARCH_SI_TRAPNO #define __ARCH_SI_BAND_T int diff --git a/include/asm-x86_64/siginfo.h b/include/asm-x86_64/siginfo.h index 7bc1598..d09a1e6 100644 --- a/include/asm-x86_64/siginfo.h +++ b/include/asm-x86_64/siginfo.h @@ -3,8 +3,6 @@ #define __ARCH_SI_PREAMBLE_SIZE (4 * sizeof(int)) -#define SIGEV_PAD_SIZE ((SIGEV_MAX_SIZE/sizeof(int)) - 4) - #include #endif -- cgit v0.10.2 From e49332bd12e92da2df6d002f857ec62675ba2648 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Sun, 1 May 2005 08:59:08 -0700 Subject: [PATCH] misc verify_area cleanups There were still a few comments left refering to verify_area, and two functions, verify_area_skas & verify_area_tt that just wrap corresponding access_ok_skas & access_ok_tt functions, just like verify_area does for access_ok - deprecate those. There was also a few places that still used verify_area in commented-out code, fix those up to use access_ok. After applying this one there should not be anything left but finally removing verify_area completely, which will happen after a kernel release or two. Signed-off-by: Jesper Juhl Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/kernel/vm86.c b/arch/i386/kernel/vm86.c index d3b4c54..ec0f68c 100644 --- a/arch/i386/kernel/vm86.c +++ b/arch/i386/kernel/vm86.c @@ -222,7 +222,7 @@ asmlinkage int sys_vm86(struct pt_regs regs) goto out; case VM86_PLUS_INSTALL_CHECK: /* NOTE: on old vm86 stuff this will return the error - from verify_area(), because the subfunction is + from access_ok(), because the subfunction is interpreted as (invalid) address to vm86_struct. So the installation check works. */ diff --git a/arch/um/kernel/skas/include/uaccess-skas.h b/arch/um/kernel/skas/include/uaccess-skas.h index 11986c9..c356203 100644 --- a/arch/um/kernel/skas/include/uaccess-skas.h +++ b/arch/um/kernel/skas/include/uaccess-skas.h @@ -18,7 +18,7 @@ ((unsigned long) (addr) + (size) <= FIXADDR_USER_END) && \ ((unsigned long) (addr) + (size) >= (unsigned long)(addr)))) -static inline int verify_area_skas(int type, const void * addr, +static inline int __deprecated verify_area_skas(int type, const void * addr, unsigned long size) { return(access_ok_skas(type, addr, size) ? 0 : -EFAULT); diff --git a/arch/um/kernel/tt/include/uaccess-tt.h b/arch/um/kernel/tt/include/uaccess-tt.h index f0bad01..bb69d6b 100644 --- a/arch/um/kernel/tt/include/uaccess-tt.h +++ b/arch/um/kernel/tt/include/uaccess-tt.h @@ -33,7 +33,7 @@ extern unsigned long uml_physmem; (((unsigned long) (addr) <= ((unsigned long) (addr) + (size))) && \ (under_task_size(addr, size) || is_stack(addr, size)))) -static inline int verify_area_tt(int type, const void * addr, +static inline int __deprecated verify_area_tt(int type, const void * addr, unsigned long size) { return(access_ok_tt(type, addr, size) ? 0 : -EFAULT); diff --git a/drivers/char/dtlk.c b/drivers/char/dtlk.c index 903e4c3..a229915 100644 --- a/drivers/char/dtlk.c +++ b/drivers/char/dtlk.c @@ -52,7 +52,7 @@ #define KERNEL #include #include -#include /* for verify_area */ +#include #include /* for -EBUSY */ #include /* for request_region */ #include /* for loops_per_jiffy */ diff --git a/drivers/char/specialix.c b/drivers/char/specialix.c index c789d5c..50e0b61 100644 --- a/drivers/char/specialix.c +++ b/drivers/char/specialix.c @@ -1987,10 +1987,9 @@ static inline int sx_set_serial_info(struct specialix_port * port, func_enter(); /* - error = verify_area(VERIFY_READ, (void *) newinfo, sizeof(tmp)); - if (error) { + if (!access_ok(VERIFY_READ, (void *) newinfo, sizeof(tmp))) { func_exit(); - return error; + return -EFAULT; } */ if (copy_from_user(&tmp, newinfo, sizeof(tmp))) { @@ -2046,14 +2045,12 @@ static inline int sx_get_serial_info(struct specialix_port * port, { struct serial_struct tmp; struct specialix_board *bp = port_Board(port); - // int error; func_enter(); /* - error = verify_area(VERIFY_WRITE, (void *) retinfo, sizeof(tmp)); - if (error) - return error; + if (!access_ok(VERIFY_WRITE, (void *) retinfo, sizeof(tmp))) + return -EFAULT; */ memset(&tmp, 0, sizeof(tmp)); diff --git a/include/asm-frv/pgtable.h b/include/asm-frv/pgtable.h index 3c6d42a..d0a9c2f 100644 --- a/include/asm-frv/pgtable.h +++ b/include/asm-frv/pgtable.h @@ -349,9 +349,9 @@ static inline pmd_t *pmd_offset(pud_t *dir, unsigned long address) /* * Define this to warn about kernel memory accesses that are - * done without a 'verify_area(VERIFY_WRITE,..)' + * done without a 'access_ok(VERIFY_WRITE,..)' */ -#undef TEST_VERIFY_AREA +#undef TEST_ACCESS_OK #define pte_present(x) (pte_val(x) & _PAGE_PRESENT) #define pte_clear(mm,addr,xp) do { set_pte_at(mm, addr, xp, __pte(0)); } while (0) diff --git a/include/asm-i386/checksum.h b/include/asm-i386/checksum.h index d76a5f0..6413420 100644 --- a/include/asm-i386/checksum.h +++ b/include/asm-i386/checksum.h @@ -33,7 +33,7 @@ asmlinkage unsigned int csum_partial_copy_generic(const unsigned char *src, unsi * passed in an incorrect kernel address to one of these functions. * * If you use these functions directly please don't forget the - * verify_area(). + * access_ok(). */ static __inline__ unsigned int csum_partial_copy_nocheck (const unsigned char *src, unsigned char *dst, diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h index 5c72542..8d60c2b 100644 --- a/include/asm-i386/pgtable.h +++ b/include/asm-i386/pgtable.h @@ -193,9 +193,9 @@ extern unsigned long long __PAGE_KERNEL, __PAGE_KERNEL_EXEC; /* * Define this if things work differently on an i386 and an i486: * it will (on an i486) warn about kernel memory accesses that are - * done without a 'verify_area(VERIFY_WRITE,..)' + * done without a 'access_ok(VERIFY_WRITE,..)' */ -#undef TEST_VERIFY_AREA +#undef TEST_ACCESS_OK /* The boot page tables (all created as a single array) */ extern unsigned long pg0[]; diff --git a/include/asm-parisc/uaccess.h b/include/asm-parisc/uaccess.h index 8a08423..c1b5bde 100644 --- a/include/asm-parisc/uaccess.h +++ b/include/asm-parisc/uaccess.h @@ -24,7 +24,7 @@ /* * Note that since kernel addresses are in a separate address space on - * parisc, we don't need to do anything for access_ok() or verify_area(). + * parisc, we don't need to do anything for access_ok(). * We just let the page fault handler do the right thing. This also means * that put_user is the same as __put_user, etc. */ diff --git a/include/asm-sh/checksum.h b/include/asm-sh/checksum.h index 5113c7f..5ebd0f2 100644 --- a/include/asm-sh/checksum.h +++ b/include/asm-sh/checksum.h @@ -42,7 +42,7 @@ asmlinkage unsigned int csum_partial_copy_generic(const unsigned char *src, unsi * passed in an incorrect kernel address to one of these functions. * * If you use these functions directly please don't forget the - * verify_area(). + * access_ok(). */ static __inline__ unsigned int csum_partial_copy_nocheck (const unsigned char *src, unsigned char *dst, diff --git a/include/asm-sh64/checksum.h b/include/asm-sh64/checksum.h index aa3911a..fd034e9 100644 --- a/include/asm-sh64/checksum.h +++ b/include/asm-sh64/checksum.h @@ -34,7 +34,7 @@ asmlinkage unsigned int csum_partial(const unsigned char *buff, int len, * passed in an incorrect kernel address to one of these functions. * * If you use these functions directly please don't forget the - * verify_area(). + * access_ok(). */ diff --git a/include/asm-sparc/uaccess.h b/include/asm-sparc/uaccess.h index 3f47889..f461144 100644 --- a/include/asm-sparc/uaccess.h +++ b/include/asm-sparc/uaccess.h @@ -18,7 +18,7 @@ #ifndef __ASSEMBLY__ -/* Sparc is not segmented, however we need to be able to fool verify_area() +/* Sparc is not segmented, however we need to be able to fool access_ok() * when doing system calls from kernel mode legitimately. * * "For historical reasons, these macros are grossly misnamed." -Linus diff --git a/net/8021q/vlanproc.c b/net/8021q/vlanproc.c index c32d27a..7b214cf 100644 --- a/net/8021q/vlanproc.c +++ b/net/8021q/vlanproc.c @@ -23,7 +23,7 @@ #include /* return codes */ #include #include /* kmalloc(), kfree() */ -#include /* verify_area(), etc. */ +#include #include /* inline mem*, str* functions */ #include /* __initfunc et al. */ #include /* htons(), etc. */ diff --git a/net/atm/common.c b/net/atm/common.c index 6d16be3..e93e838 100644 --- a/net/atm/common.c +++ b/net/atm/common.c @@ -12,7 +12,7 @@ #include /* SOL_SOCKET */ #include /* error codes */ #include -#include /* verify_area */ +#include #include #include /* struct timeval */ #include @@ -540,7 +540,7 @@ int vcc_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *m, error = -EMSGSIZE; goto out; } - /* verify_area is done by net/socket.c */ + eff = (size+3) & ~3; /* align to word boundary */ prepare_to_wait(sk->sk_sleep, &wait, TASK_INTERRUPTIBLE); error = 0; diff --git a/net/core/iovec.c b/net/core/iovec.c index d57ace9..65e4b56 100644 --- a/net/core/iovec.c +++ b/net/core/iovec.c @@ -33,7 +33,7 @@ * Verify iovec. The caller must ensure that the iovec is big enough * to hold the message iovec. * - * Save time not doing verify_area. copy_*_user will make this work + * Save time not doing access_ok. copy_*_user will make this work * in any case. */ diff --git a/net/wanrouter/wanmain.c b/net/wanrouter/wanmain.c index 956c17f..d6844ac 100644 --- a/net/wanrouter/wanmain.c +++ b/net/wanrouter/wanmain.c @@ -48,8 +48,8 @@ #include #include #include /* support for loadable modules */ -#include /* kmalloc(), kfree() */ -#include /* verify_area(), etc. */ +#include /* kmalloc(), kfree() */ +#include #include /* inline mem*, str* functions */ #include /* htons(), etc. */ -- cgit v0.10.2 From f78fc874f42f63a460bcebc2aeb98db526280d1c Mon Sep 17 00:00:00 2001 From: Vinay K Nallamothu Date: Sun, 1 May 2005 08:59:09 -0700 Subject: [PATCH] __attribute__ placement fixes The variable attributes "packed" and "align" when used with struct, should have the following order: struct ... {...} __attribute__((packed)) var; This patch fixes few instances where the variable and attributes are placed the other way around and had no effect. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/net/gt96100eth.h b/drivers/net/gt96100eth.h index 2f4bfd4..395869c 100644 --- a/drivers/net/gt96100eth.h +++ b/drivers/net/gt96100eth.h @@ -214,7 +214,7 @@ typedef struct { u32 cmdstat; u32 next; u32 buff_ptr; -} gt96100_td_t __attribute__ ((packed)); +} __attribute__ ((packed)) gt96100_td_t; typedef struct { #ifdef DESC_BE @@ -227,7 +227,7 @@ typedef struct { u32 cmdstat; u32 next; u32 buff_ptr; -} gt96100_rd_t __attribute__ ((packed)); +} __attribute__ ((packed)) gt96100_rd_t; /* Values for the Tx command-status descriptor entry. */ diff --git a/include/asm-m68knommu/MC68328.h b/include/asm-m68knommu/MC68328.h index 4f5a984..a337e56 100644 --- a/include/asm-m68knommu/MC68328.h +++ b/include/asm-m68knommu/MC68328.h @@ -993,7 +993,7 @@ typedef volatile struct { volatile unsigned short int pad1; volatile unsigned short int pad2; volatile unsigned short int pad3; -} m68328_uart __attribute__((packed)); +} __attribute__((packed)) m68328_uart; /********** diff --git a/include/asm-m68knommu/MC68EZ328.h b/include/asm-m68knommu/MC68EZ328.h index 801933d..69b7f91 100644 --- a/include/asm-m68knommu/MC68EZ328.h +++ b/include/asm-m68knommu/MC68EZ328.h @@ -815,7 +815,7 @@ typedef volatile struct { volatile unsigned short int nipr; volatile unsigned short int pad1; volatile unsigned short int pad2; -} m68328_uart __attribute__((packed)); +} __attribute__((packed)) m68328_uart; /********** diff --git a/include/asm-m68knommu/MC68VZ328.h b/include/asm-m68knommu/MC68VZ328.h index df74322..2b9bf62 100644 --- a/include/asm-m68knommu/MC68VZ328.h +++ b/include/asm-m68knommu/MC68VZ328.h @@ -909,7 +909,7 @@ typedef struct { volatile unsigned short int nipr; volatile unsigned short int hmark; volatile unsigned short int unused; -} m68328_uart __attribute__((packed)); +} __attribute__((packed)) m68328_uart; -- cgit v0.10.2 From 439bd3708792e31057db424d1165473122d531a3 Mon Sep 17 00:00:00 2001 From: Nicolas Boichat Date: Sun, 1 May 2005 08:59:09 -0700 Subject: [PATCH] Leadtek Winfast remote controls Add missing button codes for the Leadtek Winfast remote controls. Signed-off-by: Nicolas Boichat Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/media/common/ir-common.c b/drivers/media/common/ir-common.c index 8c842e2..84a49d2 100644 --- a/drivers/media/common/ir-common.c +++ b/drivers/media/common/ir-common.c @@ -131,10 +131,10 @@ IR_KEYTAB_TYPE ir_codes_winfast[IR_KEYTAB_SIZE] = { [ 18 ] = KEY_KP0, [ 0 ] = KEY_POWER, -// [ 27 ] = MTS button + [ 27 ] = KEY_LANGUAGE, //MTS button [ 2 ] = KEY_TUNER, // TV/FM [ 30 ] = KEY_VIDEO, -// [ 22 ] = display button + [ 22 ] = KEY_INFO, //display button [ 4 ] = KEY_VOLUMEUP, [ 8 ] = KEY_VOLUMEDOWN, [ 12 ] = KEY_CHANNELUP, @@ -142,7 +142,7 @@ IR_KEYTAB_TYPE ir_codes_winfast[IR_KEYTAB_SIZE] = { [ 3 ] = KEY_ZOOM, // fullscreen [ 31 ] = KEY_SUBTITLE, // closed caption/teletext [ 32 ] = KEY_SLEEP, -// [ 41 ] = boss key + [ 41 ] = KEY_SEARCH, //boss key [ 20 ] = KEY_MUTE, [ 43 ] = KEY_RED, [ 44 ] = KEY_GREEN, @@ -150,17 +150,17 @@ IR_KEYTAB_TYPE ir_codes_winfast[IR_KEYTAB_SIZE] = { [ 46 ] = KEY_BLUE, [ 24 ] = KEY_KPPLUS, //fine tune + [ 25 ] = KEY_KPMINUS, //fine tune - -// [ 42 ] = picture in picture - [ 33 ] = KEY_KPDOT, + [ 42 ] = KEY_ANGLE, //picture in picture + [ 33 ] = KEY_KPDOT, [ 19 ] = KEY_KPENTER, -// [ 17 ] = recall + [ 17 ] = KEY_AGAIN, //recall [ 34 ] = KEY_BACK, [ 35 ] = KEY_PLAYPAUSE, [ 36 ] = KEY_NEXT, -// [ 37 ] = time shifting + [ 37 ] = KEY_T, //time shifting [ 38 ] = KEY_STOP, - [ 39 ] = KEY_RECORD -// [ 40 ] = snapshot + [ 39 ] = KEY_RECORD, + [ 40 ] = KEY_SHUFFLE //snapshot }; EXPORT_SYMBOL_GPL(ir_codes_winfast); -- cgit v0.10.2 From 74f9f974a64dc3de554aa1977bf108334436e47b Mon Sep 17 00:00:00 2001 From: Edward Shishkin Date: Sun, 1 May 2005 08:59:09 -0700 Subject: [PATCH] reiserfs: journal_init fix This fixes segmentation fault when specifying bad journal device via a mount option. Don't pass a zero pointer to bdevname() if filp_open() returns error. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index c9ad3a7..b16d65a 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -2306,13 +2306,16 @@ static int journal_init_dev( struct super_block *super, if( !IS_ERR( journal -> j_dev_file ) ) { struct inode *jdev_inode = journal->j_dev_file->f_mapping->host; if( !S_ISBLK( jdev_inode -> i_mode ) ) { - reiserfs_warning (super, "journal_init_dev: '%s' is " - "not a block device", jdev_name ); + reiserfs_warning(super, "journal_init_dev: '%s' is " + "not a block device", jdev_name ); result = -ENOTBLK; + release_journal_dev( super, journal ); } else { /* ok */ journal->j_dev_bd = I_BDEV(jdev_inode); set_blocksize(journal->j_dev_bd, super->s_blocksize); + reiserfs_info(super, "journal_init_dev: journal device: %s\n", + bdevname(journal->j_dev_bd, b)); } } else { result = PTR_ERR( journal -> j_dev_file ); @@ -2321,11 +2324,6 @@ static int journal_init_dev( struct super_block *super, "journal_init_dev: Cannot open '%s': %i", jdev_name, result ); } - if( result != 0 ) { - release_journal_dev( super, journal ); - } - reiserfs_info(super, "journal_init_dev: journal device: %s\n", - bdevname(journal->j_dev_bd, b)); return result; } -- cgit v0.10.2 From 5717ffbe16815db229a66fee824c2190cb306734 Mon Sep 17 00:00:00 2001 From: "aquynh@gmail.com" Date: Sun, 1 May 2005 08:59:10 -0700 Subject: [PATCH] dontdiff file sorted in alphabet order Documentation/dontdiff is a little messy. Here is a patch to sort the content of that file in alphabetical Signed-off-by: Nguyen Anh Quynh Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/dontdiff b/Documentation/dontdiff index 734fcc8..7c24964 100644 --- a/Documentation/dontdiff +++ b/Documentation/dontdiff @@ -1,137 +1,137 @@ -.* +*.a +*.aux +*.bin +*.cpio +*.css +*.dvi +*.eps +*.gif +*.grep +*.grp +*.gz +*.html +*.jpeg +*.ko +*.log +*.lst +*.mod.c +*.o +*.orig +*.out +*.pdf +*.png +*.ps +*.rej +*.s +*.sgml +*.so +*.tex +*.ver +*_MODULES +*_vga16.c +*cscope* *~ +.* +.cscope +53c700_d.h 53c8xx_d.h* -*.a +BitKeeper +COPYING +CREDITS +CVS +ChangeSet +Kerntypes +MODS.txt +Module.symvers +PENDING +SCCS +System.map* +TAGS aic7*reg.h* -aic7*seq.h* aic7*reg_print.c* -53c700_d.h +aic7*seq.h* aicasm aicdb.h* asm asm_offsets.* autoconf.h* -*.aux bbootsect -*.bin bin2c binkernel.spec -BitKeeper bootsect bsetup btfixupprep build bvmlinux bzImage* -ChangeSet classlist.h* -compile.h* comp*.log +compile.h* config config-* config_data.h* conmakehash consolemap_deftbl.c* -COPYING -CREDITS -.cscope -*cscope* +crc32table.h* cscope.* -*.out -*.css -CVS defkeymap.c* devlist.h* docproc dummy_sym.c* -*.dvi -*.eps +elfconfig.h* filelist fixdep fore200e_mkfirm fore200e_pca_fw.c* gen-devlist -gen_init_cpio -gen_crc32table -crc32table.h* -*.cpio gen-kdb_cmds.c* -gentbl +gen_crc32table +gen_init_cpio genksyms -*.gif -*.gz -*.html +gentbl ikconfig.h* initramfs_list -*.jpeg +kallsyms kconfig kconfig.tk -Kerntypes keywords.c* ksym.c* ksym.h* -kallsyms -mk_elfconfig -elfconfig.h* -modpost -pnmtologo -logo_*.c -*.log lex.c* +logo_*.c logo_*_clut224.c logo_*_mono.c lxdialog make_times_h map +maui_boot.h +mk_elfconfig mkdep -*_MODULES -MODS.txt +mktables +modpost modversions.h* -Module.symvers -*.mod.c -*.o -*.ko -*.orig -*.lst -*.grp -*.grep oui.c* -mktables -raid6tables.c -raid6int*.c -raid6altivec*.c -wanxlfw.inc -maui_boot.h -pss_boot.h -trix_boot.h -*.pdf parse.c* parse.h* -PENDING +pnmtologo ppc_defs.h* promcon_tbl.c* -*.png -*.ps -*.rej -SCCS +pss_boot.h +raid6altivec*.c +raid6int*.c +raid6tables.c setup -*.s -*.so -*.sgml sim710_d.h* sm_tbl* split-include -System.map* tags -TAGS -*.tex times.h* tkparse -*.ver +trix_boot.h version.h* -*_vga16.c vmlinux -vmlinux.lds vmlinux-* +vmlinux.lds vsyscall.lds +wanxlfw.inc zImage -- cgit v0.10.2 From 9206880198589670dfc33feb9d2e903c3492701f Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Sun, 1 May 2005 08:59:10 -0700 Subject: [PATCH] IPMI: fix for handling bad IPMI DMI data Ignore the bottom bit of the base address from the DMI data. It is supposed to be set to 1 if it is I/O space. Few systems do this, but this enables the ones that do set it to work properly. Signed-off-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index 44a7f13..3522723 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -1623,7 +1623,13 @@ static int decode_dmi(dmi_header_t *dm, int intf_num) } } else { /* Old DMI spec. */ - ipmi_data->base_addr = base_addr; + /* Note that technically, the lower bit of the base + * address should be 1 if the address is I/O and 0 if + * the address is in memory. So many systems get that + * wrong (and all that I have seen are I/O) so we just + * ignore that bit and assume I/O. Systems that use + * memory should use the newer spec, anyway. */ + ipmi_data->base_addr = base_addr & 0xfffe; ipmi_data->addr_space = IPMI_IO_ADDR_SPACE; ipmi_data->offset = 1; } -- cgit v0.10.2 From 35bc37a0e0979a091bcf5d9161ffe935b3aa998c Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Sun, 1 May 2005 08:59:10 -0700 Subject: [PATCH] IPMI: fix for handling bad ACPI data If the ACPI register bit width is zero (an invalid value) assume it is the default spacing. This avoids some coredumps on invalid data and makes some systems work that have broken ACPI data. Signed-off-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index 3522723..7522bd5 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -1526,8 +1526,17 @@ static int try_init_acpi(int intf_num, struct smi_info **new_info) info->irq_setup = NULL; } - regspacings[intf_num] = spmi->addr.register_bit_width / 8; - info->io.regspacing = spmi->addr.register_bit_width / 8; + if (spmi->addr.register_bit_width) { + /* A (hopefully) properly formed register bit width. */ + regspacings[intf_num] = spmi->addr.register_bit_width / 8; + info->io.regspacing = spmi->addr.register_bit_width / 8; + } else { + /* Some broken systems get this wrong and set the value + * to zero. Assume it is the default spacing. If that + * is wrong, too bad, the vendor should fix the tables. */ + regspacings[intf_num] = DEFAULT_REGSPACING; + info->io.regspacing = DEFAULT_REGSPACING; + } regsizes[intf_num] = regspacings[intf_num]; info->io.regsize = regsizes[intf_num]; regshifts[intf_num] = spmi->addr.register_bit_offset; -- cgit v0.10.2 From ec26d79f4f5822283e0bffa44a542fd13c5146e4 Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Sun, 1 May 2005 08:59:11 -0700 Subject: [PATCH] ipmi: fix watchdog so the device can be reopened on an unexpected close If there is an unexpected close, still allow the watchdog interface to be re-opened on the IPMI watchdog. Signed-off-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/ipmi/ipmi_watchdog.c b/drivers/char/ipmi/ipmi_watchdog.c index fd70938..fcd1c02 100644 --- a/drivers/char/ipmi/ipmi_watchdog.c +++ b/drivers/char/ipmi/ipmi_watchdog.c @@ -709,11 +709,11 @@ static int ipmi_close(struct inode *ino, struct file *filep) if (expect_close == 42) { ipmi_watchdog_state = WDOG_TIMEOUT_NONE; ipmi_set_timeout(IPMI_SET_TIMEOUT_NO_HB); - clear_bit(0, &ipmi_wdog_open); } else { printk(KERN_CRIT PFX "Unexpected close, not stopping watchdog!\n"); ipmi_heartbeat(); } + clear_bit(0, &ipmi_wdog_open); } ipmi_fasync (-1, filep, 0); -- cgit v0.10.2 From 9dbf68f97d585265eaadd15aea308efd9ae39d34 Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Sun, 1 May 2005 08:59:11 -0700 Subject: [PATCH] ipmi: enable interrupts on the BT driver Enable interrupts for a BT interface. There is a specific register that needs to be set up to enable interrupts that also must be modified to clear the irq. Also, don't reset the BMC on a BT interface. That's probably not a good idea as the BMC may be performing other important functions and a reset should only be a last resort. Also, that register is also used to enable/disable interrupts to the BT; modifying it may screw up the interrupts. Signed-off-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/ipmi/ipmi_bt_sm.c b/drivers/char/ipmi/ipmi_bt_sm.c index 225b330..5ce9c62 100644 --- a/drivers/char/ipmi/ipmi_bt_sm.c +++ b/drivers/char/ipmi/ipmi_bt_sm.c @@ -235,7 +235,6 @@ static void reset_flags(struct si_sm_data *bt) if (BT_STATUS & BT_B_BUSY) BT_CONTROL(BT_B_BUSY); BT_CONTROL(BT_CLR_WR_PTR); BT_CONTROL(BT_SMS_ATN); - BT_INTMASK_W(BT_BMC_HWRST); #ifdef DEVELOPMENT_ONLY_NOT_FOR_PRODUCTION if (BT_STATUS & BT_B2H_ATN) { int i; diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index 7522bd5..5419440 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -100,6 +100,11 @@ enum si_intf_state { /* FIXME - add watchdog stuff. */ }; +/* Some BT-specific defines we need here. */ +#define IPMI_BT_INTMASK_REG 2 +#define IPMI_BT_INTMASK_CLEAR_IRQ_BIT 2 +#define IPMI_BT_INTMASK_ENABLE_IRQ_BIT 1 + enum si_type { SI_KCS, SI_SMIC, SI_BT }; @@ -875,6 +880,17 @@ static irqreturn_t si_irq_handler(int irq, void *data, struct pt_regs *regs) return IRQ_HANDLED; } +static irqreturn_t si_bt_irq_handler(int irq, void *data, struct pt_regs *regs) +{ + struct smi_info *smi_info = data; + /* We need to clear the IRQ flag for the BT interface. */ + smi_info->io.outputb(&smi_info->io, IPMI_BT_INTMASK_REG, + IPMI_BT_INTMASK_CLEAR_IRQ_BIT + | IPMI_BT_INTMASK_ENABLE_IRQ_BIT); + return si_irq_handler(irq, data, regs); +} + + static struct ipmi_smi_handlers handlers = { .owner = THIS_MODULE, @@ -1001,11 +1017,22 @@ static int std_irq_setup(struct smi_info *info) if (!info->irq) return 0; - rv = request_irq(info->irq, - si_irq_handler, - SA_INTERRUPT, - DEVICE_NAME, - info); + if (info->si_type == SI_BT) { + rv = request_irq(info->irq, + si_bt_irq_handler, + SA_INTERRUPT, + DEVICE_NAME, + info); + if (!rv) + /* Enable the interrupt in the BT interface. */ + info->io.outputb(&info->io, IPMI_BT_INTMASK_REG, + IPMI_BT_INTMASK_ENABLE_IRQ_BIT); + } else + rv = request_irq(info->irq, + si_irq_handler, + SA_INTERRUPT, + DEVICE_NAME, + info); if (rv) { printk(KERN_WARNING "ipmi_si: %s unable to claim interrupt %d," @@ -1024,6 +1051,9 @@ static void std_irq_cleanup(struct smi_info *info) if (!info->irq) return; + if (info->si_type == SI_BT) + /* Disable the interrupt in the BT interface. */ + info->io.outputb(&info->io, IPMI_BT_INTMASK_REG, 0); free_irq(info->irq, info); } -- cgit v0.10.2 From 882fe011a92fa4fc31ca6cc95b279f7e4e52935c Mon Sep 17 00:00:00 2001 From: Corey Minyard Date: Sun, 1 May 2005 08:59:12 -0700 Subject: [PATCH] ipmi: fix a deadlock Correct an issue with the IPMI message layer taking a lock and calling lower layer driver. If an error occrues at the lower layer the lock can be taken again causing a deadlock. The lock is released before calling the lower layer. Signed-off-by: David Griego Signed-off-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index a6606a1..d7fb452 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -2588,28 +2588,20 @@ handle_msg_timeout(struct ipmi_recv_msg *msg) deliver_response(msg); } -static void -send_from_recv_msg(ipmi_smi_t intf, struct ipmi_recv_msg *recv_msg, - struct ipmi_smi_msg *smi_msg, - unsigned char seq, long seqid) +static struct ipmi_smi_msg * +smi_from_recv_msg(ipmi_smi_t intf, struct ipmi_recv_msg *recv_msg, + unsigned char seq, long seqid) { - if (!smi_msg) - smi_msg = ipmi_alloc_smi_msg(); + struct ipmi_smi_msg *smi_msg = ipmi_alloc_smi_msg(); if (!smi_msg) /* If we can't allocate the message, then just return, we get 4 retries, so this should be ok. */ - return; + return NULL; memcpy(smi_msg->data, recv_msg->msg.data, recv_msg->msg.data_len); smi_msg->data_size = recv_msg->msg.data_len; smi_msg->msgid = STORE_SEQ_IN_MSGID(seq, seqid); - /* Send the new message. We send with a zero priority. It - timed out, I doubt time is that critical now, and high - priority messages are really only for messages to the local - MC, which don't get resent. */ - intf->handlers->sender(intf->send_info, smi_msg, 0); - #ifdef DEBUG_MSGING { int m; @@ -2619,6 +2611,7 @@ send_from_recv_msg(ipmi_smi_t intf, struct ipmi_recv_msg *recv_msg, printk("\n"); } #endif + return smi_msg; } static void @@ -2683,14 +2676,13 @@ ipmi_timeout_handler(long timeout_period) intf->timed_out_ipmb_commands++; spin_unlock(&intf->counter_lock); } else { + struct ipmi_smi_msg *smi_msg; /* More retries, send again. */ /* Start with the max timer, set to normal timer after the message is sent. */ ent->timeout = MAX_MSG_TIMEOUT; ent->retries_left--; - send_from_recv_msg(intf, ent->recv_msg, NULL, - j, ent->seqid); spin_lock(&intf->counter_lock); if (ent->recv_msg->addr.addr_type == IPMI_LAN_ADDR_TYPE) @@ -2698,6 +2690,20 @@ ipmi_timeout_handler(long timeout_period) else intf->retransmitted_ipmb_commands++; spin_unlock(&intf->counter_lock); + smi_msg = smi_from_recv_msg(intf, + ent->recv_msg, j, ent->seqid); + if(!smi_msg) + continue; + + spin_unlock_irqrestore(&(intf->seq_lock),flags); + /* Send the new message. We send with a zero + * priority. It timed out, I doubt time is + * that critical now, and high priority + * messages are really only for messages to the + * local MC, which don't get resent. */ + intf->handlers->sender(intf->send_info, + smi_msg, 0); + spin_lock_irqsave(&(intf->seq_lock), flags); } } spin_unlock_irqrestore(&(intf->seq_lock), flags); -- cgit v0.10.2 From 434498d5323445b59167fd7aa5633b74ebbce901 Mon Sep 17 00:00:00 2001 From: Andreas Schwab Date: Sun, 1 May 2005 08:59:12 -0700 Subject: [PATCH] sn_console: make sal_console_uart static again Signed-off-by: Andreas Schwab Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/serial/sn_console.c b/drivers/serial/sn_console.c index ffaab9b..fee6418 100644 --- a/drivers/serial/sn_console.c +++ b/drivers/serial/sn_console.c @@ -787,7 +787,7 @@ static void __init sn_sal_switch_to_interrupts(struct sn_cons_port *port) static void sn_sal_console_write(struct console *, const char *, unsigned); static int __init sn_sal_console_setup(struct console *, char *); -extern struct uart_driver sal_console_uart; +static struct uart_driver sal_console_uart; extern struct tty_driver *uart_console_device(struct console *, int *); static struct console sal_console = { @@ -801,7 +801,7 @@ static struct console sal_console = { #define SAL_CONSOLE &sal_console -struct uart_driver sal_console_uart = { +static struct uart_driver sal_console_uart = { .owner = THIS_MODULE, .driver_name = "sn_console", .dev_name = DEVICE_NAME, -- cgit v0.10.2 From 7d87e14c236d6c4cab66d87cf0bc1e0f0375d308 Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Sun, 1 May 2005 08:59:12 -0700 Subject: [PATCH] consolidate sys_shmat Signed-off-by: Stephen Rothwell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c index b5d0fd2..64e450d 100644 --- a/arch/alpha/kernel/osf_sys.c +++ b/arch/alpha/kernel/osf_sys.c @@ -457,22 +457,6 @@ osf_getdomainname(char __user *name, int namelen) return 0; } -asmlinkage long -osf_shmat(int shmid, void __user *shmaddr, int shmflg) -{ - unsigned long raddr; - long err; - - err = do_shmat(shmid, shmaddr, shmflg, &raddr); - - /* - * This works because all user-level addresses are - * non-negative longs! - */ - return err ? err : (long)raddr; -} - - /* * The following stuff should move into a header file should it ever * be labeled "officially supported." Right now, there is just enough diff --git a/arch/alpha/kernel/systbls.S b/arch/alpha/kernel/systbls.S index 3864b33..0521208 100644 --- a/arch/alpha/kernel/systbls.S +++ b/arch/alpha/kernel/systbls.S @@ -227,7 +227,7 @@ sys_call_table: .quad sys_semop .quad osf_utsname .quad sys_lchown - .quad osf_shmat + .quad sys_shmat .quad sys_shmctl /* 210 */ .quad sys_shmdt .quad sys_shmget diff --git a/arch/arm/kernel/sys_arm.c b/arch/arm/kernel/sys_arm.c index 7ba6342..ef32577 100644 --- a/arch/arm/kernel/sys_arm.c +++ b/arch/arm/kernel/sys_arm.c @@ -227,18 +227,6 @@ asmlinkage int sys_ipc(uint call, int first, int second, int third, } } -asmlinkage long sys_shmat(int shmid, char __user *shmaddr, int shmflg, - unsigned long __user *addr) -{ - unsigned long ret; - long err; - - err = do_shmat(shmid, shmaddr, shmflg, &ret); - if (err == 0) - err = put_user(ret, addr); - return err; -} - /* Fork a new task - this creates a new program thread. * This is called indirectly via a small wrapper */ diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S index bd86fea..d3f0938 100644 --- a/arch/ia64/kernel/entry.S +++ b/arch/ia64/kernel/entry.S @@ -1417,7 +1417,7 @@ sys_call_table: data8 sys_msgrcv data8 sys_msgctl data8 sys_shmget - data8 ia64_shmat + data8 sys_shmat data8 sys_shmdt // 1115 data8 sys_shmctl data8 sys_syslog diff --git a/arch/ia64/kernel/sys_ia64.c b/arch/ia64/kernel/sys_ia64.c index 3ac216e..a8cf6d8 100644 --- a/arch/ia64/kernel/sys_ia64.c +++ b/arch/ia64/kernel/sys_ia64.c @@ -93,20 +93,6 @@ sys_getpagesize (void) } asmlinkage unsigned long -ia64_shmat (int shmid, void __user *shmaddr, int shmflg) -{ - unsigned long raddr; - int retval; - - retval = do_shmat(shmid, shmaddr, shmflg, &raddr); - if (retval < 0) - return retval; - - force_successful_syscall_return(); - return raddr; -} - -asmlinkage unsigned long ia64_brk (unsigned long brk) { unsigned long rlim, retval, newbrk, oldbrk; diff --git a/arch/mips/kernel/syscall.c b/arch/mips/kernel/syscall.c index 598bfe7..ae2a131 100644 --- a/arch/mips/kernel/syscall.c +++ b/arch/mips/kernel/syscall.c @@ -374,22 +374,6 @@ asmlinkage int sys_ipc (uint call, int first, int second, } /* - * Native ABI that is O32 or N64 version - */ -asmlinkage long sys_shmat(int shmid, char __user *shmaddr, - int shmflg, unsigned long *addr) -{ - unsigned long raddr; - int err; - - err = do_shmat(shmid, shmaddr, shmflg, &raddr); - if (err) - return err; - - return put_user(raddr, addr); -} - -/* * No implemented yet ... */ asmlinkage int sys_cachectl(char *addr, int nbytes, int op) diff --git a/arch/parisc/kernel/sys_parisc.c b/arch/parisc/kernel/sys_parisc.c index 7958cd8..d15a1d5 100644 --- a/arch/parisc/kernel/sys_parisc.c +++ b/arch/parisc/kernel/sys_parisc.c @@ -161,17 +161,6 @@ asmlinkage unsigned long sys_mmap(unsigned long addr, unsigned long len, } } -long sys_shmat_wrapper(int shmid, char __user *shmaddr, int shmflag) -{ - unsigned long raddr; - int r; - - r = do_shmat(shmid, shmaddr, shmflag, &raddr); - if (r < 0) - return r; - return raddr; -} - /* Fucking broken ABI */ #ifdef CONFIG_64BIT diff --git a/arch/parisc/kernel/syscall_table.S b/arch/parisc/kernel/syscall_table.S index 779b537..dcfa4d3 100644 --- a/arch/parisc/kernel/syscall_table.S +++ b/arch/parisc/kernel/syscall_table.S @@ -297,7 +297,7 @@ ENTRY_DIFF(msgrcv) ENTRY_SAME(msgget) /* 190 */ ENTRY_SAME(msgctl) - ENTRY_SAME(shmat_wrapper) + ENTRY_SAME(shmat) ENTRY_SAME(shmdt) ENTRY_SAME(shmget) ENTRY_SAME(shmctl) /* 195 */ diff --git a/arch/sh64/kernel/sys_sh64.c b/arch/sh64/kernel/sys_sh64.c index 4546845..58ff7d5 100644 --- a/arch/sh64/kernel/sys_sh64.c +++ b/arch/sh64/kernel/sys_sh64.c @@ -283,18 +283,3 @@ asmlinkage int sys_uname(struct old_utsname * name) up_read(&uts_sem); return err?-EFAULT:0; } - -/* Copy from mips version */ -asmlinkage long sys_shmatcall(int shmid, char __user *shmaddr, - int shmflg) -{ - unsigned long raddr; - int err; - - err = do_shmat(shmid, shmaddr, shmflg, &raddr); - if (err) - return err; - - err = raddr; - return err; -} diff --git a/arch/sh64/kernel/syscalls.S b/arch/sh64/kernel/syscalls.S index 8ed417d..6aabc63 100644 --- a/arch/sh64/kernel/syscalls.S +++ b/arch/sh64/kernel/syscalls.S @@ -268,7 +268,7 @@ sys_call_table: .long sys_msgrcv .long sys_msgget .long sys_msgctl - .long sys_shmatcall + .long sys_shmat .long sys_shmdt /* 245 */ .long sys_shmget .long sys_shmctl diff --git a/arch/um/include/sysdep-x86_64/syscalls.h b/arch/um/include/sysdep-x86_64/syscalls.h index b56b335..67923cc 100644 --- a/arch/um/include/sysdep-x86_64/syscalls.h +++ b/arch/um/include/sysdep-x86_64/syscalls.h @@ -26,7 +26,6 @@ extern syscall_handler_t *ia32_sys_call_table[]; extern long old_mmap(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long pgoff); -extern syscall_handler_t wrap_sys_shmat; extern syscall_handler_t sys_modify_ldt; extern syscall_handler_t sys_arch_prctl; diff --git a/arch/um/sys-x86_64/syscalls.c b/arch/um/sys-x86_64/syscalls.c index 15768c9..ab4b0ab 100644 --- a/arch/um/sys-x86_64/syscalls.c +++ b/arch/um/sys-x86_64/syscalls.c @@ -14,14 +14,6 @@ #include "asm/prctl.h" /* XXX This should get the constants from libc */ #include "choose-mode.h" -/* XXX: copied from x86-64: arch/x86_64/kernel/sys_x86_64.c */ -asmlinkage long wrap_sys_shmat(int shmid, char __user *shmaddr, int shmflg) -{ - unsigned long raddr; - - return do_shmat(shmid, shmaddr, shmflg, &raddr) ?: (long) raddr; -} - asmlinkage long sys_uname64(struct new_utsname __user * name) { int err; diff --git a/arch/x86_64/kernel/sys_x86_64.c b/arch/x86_64/kernel/sys_x86_64.c index 477d8be..dbebd5c 100644 --- a/arch/x86_64/kernel/sys_x86_64.c +++ b/arch/x86_64/kernel/sys_x86_64.c @@ -152,12 +152,6 @@ asmlinkage long sys_uname(struct new_utsname __user * name) return err ? -EFAULT : 0; } -asmlinkage long wrap_sys_shmat(int shmid, char __user *shmaddr, int shmflg) -{ - unsigned long raddr; - return do_shmat(shmid,shmaddr,shmflg,&raddr) ?: (long)raddr; -} - asmlinkage long sys_time64(long __user * tloc) { struct timeval now; diff --git a/include/asm-x86_64/unistd.h b/include/asm-x86_64/unistd.h index 3d65d24..3c9af6f 100644 --- a/include/asm-x86_64/unistd.h +++ b/include/asm-x86_64/unistd.h @@ -76,7 +76,7 @@ __SYSCALL(__NR_madvise, sys_madvise) #define __NR_shmget 29 __SYSCALL(__NR_shmget, sys_shmget) #define __NR_shmat 30 -__SYSCALL(__NR_shmat, wrap_sys_shmat) +__SYSCALL(__NR_shmat, sys_shmat) #define __NR_shmctl 31 __SYSCALL(__NR_shmctl, sys_shmctl) diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 757cd9b..c39f6f7 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -456,8 +456,7 @@ asmlinkage long sys_semctl(int semid, int semnum, int cmd, union semun arg); asmlinkage long sys_semtimedop(int semid, struct sembuf __user *sops, unsigned nsops, const struct timespec __user *timeout); -asmlinkage long sys_shmat(int shmid, char __user *shmaddr, - int shmflg, unsigned long __user *addr); +asmlinkage long sys_shmat(int shmid, char __user *shmaddr, int shmflg); asmlinkage long sys_shmget(key_t key, size_t size, int flag); asmlinkage long sys_shmdt(char __user *shmaddr); asmlinkage long sys_shmctl(int shmid, int cmd, struct shmid_ds __user *buf); diff --git a/ipc/shm.c b/ipc/shm.c index 06cd5c91..cce02243 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -28,6 +28,8 @@ #include #include #include +#include + #include #include "util.h" @@ -771,6 +773,18 @@ out: return err; } +asmlinkage long sys_shmat(int shmid, char __user *shmaddr, int shmflg) +{ + unsigned long ret; + long err; + + err = do_shmat(shmid, shmaddr, shmflg, &ret); + if (err) + return err; + force_successful_syscall_return(); + return (long)ret; +} + /* * detach and kill segment if marked destroyed. * The work is done in shm_close. diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index 1802a31..0dda70e 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -52,6 +52,7 @@ cond_syscall(sys_msgsnd); cond_syscall(sys_msgrcv); cond_syscall(sys_msgctl); cond_syscall(sys_shmget); +cond_syscall(sys_shmat); cond_syscall(sys_shmdt); cond_syscall(sys_shmctl); cond_syscall(sys_mq_open); -- cgit v0.10.2 From 1c72d46d98e4eb16f1a1b38eba94cc3aa8022cfa Mon Sep 17 00:00:00 2001 From: Kylene Hall Date: Sun, 1 May 2005 08:59:13 -0700 Subject: [PATCH] Fix Tpm driver -- Maintainers entry This patch adds the maintainers entry. Signed-off-by: Kylene Hall Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/MAINTAINERS b/MAINTAINERS index 4333b69..f384a97 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2113,6 +2113,13 @@ M: perex@suse.cz L: alsa-devel@alsa-project.org S: Maintained +TPM DEVICE DRIVER +P: Kylene Hall +M: kjhall@us.ibm.com +W: http://tpmdd.sourceforge.net +L: tpmdd-devel@lists.sourceforge.net +S: Maintained + UltraSPARC (sparc64): P: David S. Miller M: davem@davemloft.net -- cgit v0.10.2 From e5bdd883a189243541e7a132385580703b049102 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Sun, 1 May 2005 08:59:13 -0700 Subject: [PATCH] new valid_signal() function This patch adds a new function valid_signal() that tests if its argument is a valid signal number. The reasons for adding this new function are: - some code currently testing _NSIG directly has off-by-one errors. Using this function instead avoids such errors. - some code currently tests unsigned signal numbers for <0 which is pointless and generates warnings when building with gcc -W. Using this function instead avoids such warnings. I considered various places to add this function but eventually settled on include/linux/signal.h as the most logical place for it. If there's some reason this is a bad choice then please let me know (hints as to a better location are then welcome of course). Signed-off-by: Jesper Juhl Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/signal.h b/include/linux/signal.h index 78bfb26..0a98f5e 100644 --- a/include/linux/signal.h +++ b/include/linux/signal.h @@ -220,6 +220,12 @@ static inline void init_sigpending(struct sigpending *sig) INIT_LIST_HEAD(&sig->list); } +/* Test if 'sig' is valid signal. Use this instead of testing _NSIG directly */ +static inline int valid_signal(unsigned long sig) +{ + return sig <= _NSIG ? 1 : 0; +} + extern int group_send_sig_info(int sig, struct siginfo *info, struct task_struct *p); extern int __group_send_sig_info(int, struct siginfo *, struct task_struct *); extern long do_sigpending(void __user *, unsigned long); -- cgit v0.10.2 From 7ed20e1ad521b5f5df61bf6559ae60738e393741 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Sun, 1 May 2005 08:59:14 -0700 Subject: [PATCH] convert that currently tests _NSIG directly to use valid_signal() Convert most of the current code that uses _NSIG directly to instead use valid_signal(). This avoids gcc -W warnings and off-by-one errors. Signed-off-by: Jesper Juhl Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/alpha/kernel/ptrace.c b/arch/alpha/kernel/ptrace.c index d005831..bbd3753 100644 --- a/arch/alpha/kernel/ptrace.c +++ b/arch/alpha/kernel/ptrace.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include @@ -335,7 +336,7 @@ do_sys_ptrace(long request, long pid, long addr, long data, /* continue and stop at next (return from) syscall */ case PTRACE_CONT: /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -365,7 +366,7 @@ do_sys_ptrace(long request, long pid, long addr, long data, case PTRACE_SINGLESTEP: /* execute single instruction. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; /* Mark single stepping. */ child->thread_info->bpt_nsaved = -1; diff --git a/arch/arm/kernel/ptrace.c b/arch/arm/kernel/ptrace.c index efd7a34..cd99b83 100644 --- a/arch/arm/kernel/ptrace.c +++ b/arch/arm/kernel/ptrace.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include @@ -693,7 +694,7 @@ static int do_ptrace(int request, struct task_struct *child, long addr, long dat case PTRACE_SYSCALL: case PTRACE_CONT: ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -728,7 +729,7 @@ static int do_ptrace(int request, struct task_struct *child, long addr, long dat */ case PTRACE_SINGLESTEP: ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; child->ptrace |= PT_SINGLESTEP; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); diff --git a/arch/arm26/kernel/ptrace.c b/arch/arm26/kernel/ptrace.c index 2a13714..8a52124 100644 --- a/arch/arm26/kernel/ptrace.c +++ b/arch/arm26/kernel/ptrace.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include @@ -591,7 +592,7 @@ static int do_ptrace(int request, struct task_struct *child, long addr, long dat case PTRACE_SYSCALL: case PTRACE_CONT: ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -626,7 +627,7 @@ static int do_ptrace(int request, struct task_struct *child, long addr, long dat */ case PTRACE_SINGLESTEP: ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; child->ptrace |= PT_SINGLESTEP; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); diff --git a/arch/cris/arch-v10/kernel/ptrace.c b/arch/cris/arch-v10/kernel/ptrace.c index da15db8..581ecab 100644 --- a/arch/cris/arch-v10/kernel/ptrace.c +++ b/arch/cris/arch-v10/kernel/ptrace.c @@ -10,6 +10,7 @@ #include #include #include +#include #include #include @@ -184,7 +185,7 @@ sys_ptrace(long request, long pid, long addr, long data) case PTRACE_CONT: ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) { @@ -219,7 +220,7 @@ sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SINGLESTEP: ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); diff --git a/arch/frv/kernel/ptrace.c b/arch/frv/kernel/ptrace.c index 2a0efb7..cbe03cb 100644 --- a/arch/frv/kernel/ptrace.c +++ b/arch/frv/kernel/ptrace.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include @@ -239,7 +240,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -267,7 +268,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SINGLESTEP: /* set the trap flag. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); ptrace_enable(child); diff --git a/arch/h8300/kernel/ptrace.c b/arch/h8300/kernel/ptrace.c index 5f19d77..05c15e8 100644 --- a/arch/h8300/kernel/ptrace.c +++ b/arch/h8300/kernel/ptrace.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include @@ -171,7 +172,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned long) data >= _NSIG) + if (!valid_signal(data)) break ; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -202,7 +203,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SINGLESTEP: { /* set the trap flag. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); child->exit_code = data; diff --git a/arch/i386/kernel/ptrace.c b/arch/i386/kernel/ptrace.c index b2f1764..e8c965c 100644 --- a/arch/i386/kernel/ptrace.c +++ b/arch/i386/kernel/ptrace.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include @@ -511,7 +512,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) { set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -543,7 +544,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SINGLESTEP: /* set the trap flag. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); set_singlestep(child); diff --git a/arch/ia64/kernel/ptrace.c b/arch/ia64/kernel/ptrace.c index 55789fc..c253fd5 100644 --- a/arch/ia64/kernel/ptrace.c +++ b/arch/ia64/kernel/ptrace.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include @@ -1481,7 +1482,7 @@ sys_ptrace (long request, pid_t pid, unsigned long addr, unsigned long data) case PTRACE_CONT: /* restart after signal. */ ret = -EIO; - if (data > _NSIG) + if (!valid_signal(data)) goto out_tsk; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -1520,7 +1521,7 @@ sys_ptrace (long request, pid_t pid, unsigned long addr, unsigned long data) /* let child execute for one instruction */ case PTRACE_SINGLEBLOCK: ret = -EIO; - if (data > _NSIG) + if (!valid_signal(data)) goto out_tsk; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); diff --git a/arch/m32r/kernel/ptrace.c b/arch/m32r/kernel/ptrace.c index 8b40f36..124f7c1 100644 --- a/arch/m32r/kernel/ptrace.c +++ b/arch/m32r/kernel/ptrace.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include @@ -665,7 +666,7 @@ do_ptrace(long request, struct task_struct *child, long addr, long data) case PTRACE_SYSCALL: case PTRACE_CONT: ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -700,7 +701,7 @@ do_ptrace(long request, struct task_struct *child, long addr, long data) unsigned long pc, insn; ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); if ((child->ptrace & PT_DTRACE) == 0) { diff --git a/arch/m68k/kernel/ptrace.c b/arch/m68k/kernel/ptrace.c index 0beb533..f4e1e5e 100644 --- a/arch/m68k/kernel/ptrace.c +++ b/arch/m68k/kernel/ptrace.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include @@ -251,7 +252,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) long tmp; ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) { child->thread.work.syscall_trace = ~0; @@ -292,7 +293,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) long tmp; ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; child->thread.work.syscall_trace = 0; tmp = get_reg(child, PT_SR) | (TRACE_BITS << 16); diff --git a/arch/m68knommu/kernel/ptrace.c b/arch/m68knommu/kernel/ptrace.c index 15cf790..9724e1c 100644 --- a/arch/m68knommu/kernel/ptrace.c +++ b/arch/m68knommu/kernel/ptrace.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include @@ -240,7 +241,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) long tmp; ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -278,7 +279,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) long tmp; ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); tmp = get_reg(child, PT_SR) | (TRACE_BITS << 16); diff --git a/arch/mips/kernel/ptrace.c b/arch/mips/kernel/ptrace.c index 92f2c39..a2f899c 100644 --- a/arch/mips/kernel/ptrace.c +++ b/arch/mips/kernel/ptrace.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include @@ -257,7 +258,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) { set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); diff --git a/arch/mips/kernel/ptrace32.c b/arch/mips/kernel/ptrace32.c index 611dee9..eee2079 100644 --- a/arch/mips/kernel/ptrace32.c +++ b/arch/mips/kernel/ptrace32.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include @@ -241,7 +242,7 @@ asmlinkage int sys32_ptrace(int request, int pid, int addr, int data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned int) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) { set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); diff --git a/arch/parisc/kernel/ptrace.c b/arch/parisc/kernel/ptrace.c index 2937a92..c07db9d 100644 --- a/arch/parisc/kernel/ptrace.c +++ b/arch/parisc/kernel/ptrace.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include @@ -285,7 +286,7 @@ long sys_ptrace(long request, pid_t pid, long addr, long data) ret = -EIO; DBG("sys_ptrace(%s)\n", request == PTRACE_SYSCALL ? "SYSCALL" : "CONT"); - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) goto out_tsk; child->ptrace &= ~(PT_SINGLESTEP|PT_BLOCKSTEP); if (request == PTRACE_SYSCALL) { @@ -311,7 +312,7 @@ long sys_ptrace(long request, pid_t pid, long addr, long data) case PTRACE_SINGLEBLOCK: DBG("sys_ptrace(SINGLEBLOCK)\n"); ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) goto out_tsk; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); child->ptrace &= ~PT_SINGLESTEP; @@ -328,7 +329,7 @@ long sys_ptrace(long request, pid_t pid, long addr, long data) case PTRACE_SINGLESTEP: DBG("sys_ptrace(SINGLESTEP)\n"); ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) goto out_tsk; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); diff --git a/arch/ppc/kernel/ptrace.c b/arch/ppc/kernel/ptrace.c index 426b6f7..59d59a8 100644 --- a/arch/ppc/kernel/ptrace.c +++ b/arch/ppc/kernel/ptrace.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include @@ -356,7 +357,7 @@ int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) { set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -389,7 +390,7 @@ int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SINGLESTEP: { /* set the trap flag. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); set_single_step(child); diff --git a/arch/ppc64/kernel/ptrace.c b/arch/ppc64/kernel/ptrace.c index 354a287..5a84632 100644 --- a/arch/ppc64/kernel/ptrace.c +++ b/arch/ppc64/kernel/ptrace.c @@ -28,6 +28,7 @@ #include #include #include +#include #include #include @@ -162,7 +163,7 @@ int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -194,7 +195,7 @@ int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SINGLESTEP: { /* set the trap flag. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); set_single_step(child); diff --git a/arch/ppc64/kernel/ptrace32.c b/arch/ppc64/kernel/ptrace32.c index ee81b1b..1643642 100644 --- a/arch/ppc64/kernel/ptrace32.c +++ b/arch/ppc64/kernel/ptrace32.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include @@ -293,7 +294,7 @@ int sys32_ptrace(long request, long pid, unsigned long addr, unsigned long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -325,7 +326,7 @@ int sys32_ptrace(long request, long pid, unsigned long addr, unsigned long data) case PTRACE_SINGLESTEP: { /* set the trap flag. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); set_single_step(child); diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c index 647233c..9f0d73e 100644 --- a/arch/s390/kernel/ptrace.c +++ b/arch/s390/kernel/ptrace.c @@ -32,6 +32,7 @@ #include #include #include +#include #include #include @@ -609,7 +610,7 @@ do_ptrace(struct task_struct *child, long request, long addr, long data) /* continue and stop at next (return from) syscall */ case PTRACE_CONT: /* restart after signal. */ - if ((unsigned long) data >= _NSIG) + if (!valid_signal(data)) return -EIO; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -637,7 +638,7 @@ do_ptrace(struct task_struct *child, long request, long addr, long data) case PTRACE_SINGLESTEP: /* set the trap flag. */ - if ((unsigned long) data >= _NSIG) + if (!valid_signal(data)) return -EIO; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); child->exit_code = data; diff --git a/arch/sh/kernel/ptrace.c b/arch/sh/kernel/ptrace.c index 1b0dfb4..b28919b 100644 --- a/arch/sh/kernel/ptrace.c +++ b/arch/sh/kernel/ptrace.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include @@ -197,7 +198,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -228,7 +229,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) struct pt_regs *dummy = NULL; ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); if ((child->ptrace & PT_DTRACE) == 0) { diff --git a/arch/sh64/kernel/ptrace.c b/arch/sh64/kernel/ptrace.c index 800288c..fd20009 100644 --- a/arch/sh64/kernel/ptrace.c +++ b/arch/sh64/kernel/ptrace.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include @@ -255,7 +256,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); @@ -285,7 +286,7 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data) struct pt_regs *regs; ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); if ((child->ptrace & PT_DTRACE) == 0) { diff --git a/arch/sparc/kernel/ptrace.c b/arch/sparc/kernel/ptrace.c index c4f93bd..475c4c1 100644 --- a/arch/sparc/kernel/ptrace.c +++ b/arch/sparc/kernel/ptrace.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include @@ -526,7 +527,7 @@ asmlinkage void do_ptrace(struct pt_regs *regs) addr = 1; case PTRACE_CONT: { /* restart after signal. */ - if (data > _NSIG) { + if (!valid_signal(data)) { pt_error_return(regs, EIO); goto out_tsk; } diff --git a/arch/sparc64/kernel/ptrace.c b/arch/sparc64/kernel/ptrace.c index 5f080cf..80a76e2 100644 --- a/arch/sparc64/kernel/ptrace.c +++ b/arch/sparc64/kernel/ptrace.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include @@ -559,7 +560,7 @@ asmlinkage void do_ptrace(struct pt_regs *regs) addr = 1; case PTRACE_CONT: { /* restart after signal. */ - if (data > _NSIG) { + if (!valid_signal(data)) { pt_error_return(regs, EIO); goto out_tsk; } diff --git a/arch/um/kernel/ptrace.c b/arch/um/kernel/ptrace.c index 3a99ee6..e50e60f 100644 --- a/arch/um/kernel/ptrace.c +++ b/arch/um/kernel/ptrace.c @@ -143,7 +143,7 @@ long sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ case PTRACE_CONT: { /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; child->ptrace &= ~PT_DTRACE; @@ -179,7 +179,7 @@ long sys_ptrace(long request, long pid, long addr, long data) case PTRACE_SINGLESTEP: { /* set the trap flag. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); child->ptrace |= PT_DTRACE; diff --git a/arch/v850/kernel/ptrace.c b/arch/v850/kernel/ptrace.c index 8fa7807..4726b87 100644 --- a/arch/v850/kernel/ptrace.c +++ b/arch/v850/kernel/ptrace.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include @@ -208,7 +209,7 @@ int sys_ptrace(long request, long pid, long addr, long data) /* Execute a single instruction. */ case PTRACE_SINGLESTEP: rval = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; /* Turn CHILD's single-step flag on or off. */ diff --git a/arch/x86_64/kernel/ptrace.c b/arch/x86_64/kernel/ptrace.c index c701167..c64b9c9 100644 --- a/arch/x86_64/kernel/ptrace.c +++ b/arch/x86_64/kernel/ptrace.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include @@ -467,7 +468,7 @@ asmlinkage long sys_ptrace(long request, long pid, unsigned long addr, long data case PTRACE_CONT: /* restart after signal. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; if (request == PTRACE_SYSCALL) set_tsk_thread_flag(child,TIF_SYSCALL_TRACE); @@ -529,7 +530,7 @@ asmlinkage long sys_ptrace(long request, long pid, unsigned long addr, long data case PTRACE_SINGLESTEP: /* set the trap flag. */ ret = -EIO; - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) break; clear_tsk_thread_flag(child,TIF_SYSCALL_TRACE); set_singlestep(child); diff --git a/drivers/char/vt_ioctl.c b/drivers/char/vt_ioctl.c index 5d386f4..8971484 100644 --- a/drivers/char/vt_ioctl.c +++ b/drivers/char/vt_ioctl.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include @@ -641,7 +642,7 @@ int vt_ioctl(struct tty_struct *tty, struct file * file, extern int spawnpid, spawnsig; if (!perm || !capable(CAP_KILL)) return -EPERM; - if (arg < 1 || arg > _NSIG || arg == SIGKILL) + if (!valid_signal(arg) || arg < 1 || arg == SIGKILL) return -EINVAL; spawnpid = current->pid; spawnsig = arg; diff --git a/fs/fcntl.c b/fs/fcntl.c index 3e7ab16..286a9f8 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include @@ -308,7 +309,7 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg, break; case F_SETSIG: /* arg == 0 restores default behaviour. */ - if (arg < 0 || arg > _NSIG) { + if (!valid_signal(arg)) { break; } err = 0; diff --git a/ipc/mqueue.c b/ipc/mqueue.c index 33f7152..0acf245 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include "util.h" @@ -976,8 +977,7 @@ asmlinkage long sys_mq_notify(mqd_t mqdes, notification.sigev_notify != SIGEV_THREAD)) return -EINVAL; if (notification.sigev_notify == SIGEV_SIGNAL && - (notification.sigev_signo < 0 || - notification.sigev_signo > _NSIG)) { + !valid_signal(notification.sigev_signo)) { return -EINVAL; } if (notification.sigev_notify == SIGEV_THREAD) { diff --git a/kernel/exit.c b/kernel/exit.c index 93851bc..eb8da36 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include @@ -277,7 +278,7 @@ void set_special_pids(pid_t session, pid_t pgrp) */ int allow_signal(int sig) { - if (sig < 1 || sig > _NSIG) + if (!valid_signal(sig) || sig < 1) return -EINVAL; spin_lock_irq(¤t->sighand->siglock); @@ -298,7 +299,7 @@ EXPORT_SYMBOL(allow_signal); int disallow_signal(int sig) { - if (sig < 1 || sig > _NSIG) + if (!valid_signal(sig) || sig < 1) return -EINVAL; spin_lock_irq(¤t->sighand->siglock); diff --git a/kernel/futex.c b/kernel/futex.c index 7b54a67..c7130f8 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -39,6 +39,7 @@ #include #include #include +#include #define FUTEX_HASHBITS (CONFIG_BASE_SMALL ? 4 : 8) @@ -654,7 +655,7 @@ static int futex_fd(unsigned long uaddr, int signal) int ret, err; ret = -EINVAL; - if (signal < 0 || signal > _NSIG) + if (!valid_signal(signal)) goto out; ret = get_unused_fd(); diff --git a/kernel/ptrace.c b/kernel/ptrace.c index f5cc1ce..8dcb8f6 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include @@ -166,7 +167,7 @@ bad: int ptrace_detach(struct task_struct *child, unsigned int data) { - if ((unsigned long) data > _NSIG) + if (!valid_signal(data)) return -EIO; /* Architecture-specific hardware disable .. */ diff --git a/kernel/signal.c b/kernel/signal.c index e6567d7..8f3debc 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include #include @@ -646,7 +647,7 @@ static int check_kill_permission(int sig, struct siginfo *info, struct task_struct *t) { int error = -EINVAL; - if (sig < 0 || sig > _NSIG) + if (!valid_signal(sig)) return error; error = -EPERM; if ((!info || ((unsigned long)info != 1 && @@ -1245,7 +1246,7 @@ send_sig_info(int sig, struct siginfo *info, struct task_struct *p) * Make sure legacy kernel users don't send in bad values * (normal paths check this in check_kill_permission). */ - if (sig < 0 || sig > _NSIG) + if (!valid_signal(sig)) return -EINVAL; /* @@ -1520,7 +1521,7 @@ void do_notify_parent(struct task_struct *tsk, int sig) if (psig->action[SIGCHLD-1].sa.sa_handler == SIG_IGN) sig = 0; } - if (sig > 0 && sig <= _NSIG) + if (valid_signal(sig) && sig > 0) __group_send_sig_info(sig, &info, tsk->parent); __wake_up_parent(tsk, tsk->parent); spin_unlock_irqrestore(&psig->siglock, flags); @@ -2364,7 +2365,7 @@ do_sigaction(int sig, const struct k_sigaction *act, struct k_sigaction *oact) { struct k_sigaction *k; - if (sig < 1 || sig > _NSIG || (act && sig_kernel_only(sig))) + if (!valid_signal(sig) || sig < 1 || (act && sig_kernel_only(sig))) return -EINVAL; k = ¤t->sighand->action[sig-1]; diff --git a/kernel/sys.c b/kernel/sys.c index 7f43d6e..f64e97c 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -25,6 +25,7 @@ #include #include #include +#include #include #include @@ -1637,7 +1638,7 @@ asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3, switch (option) { case PR_SET_PDEATHSIG: sig = arg2; - if (sig < 0 || sig > _NSIG) { + if (!valid_signal(sig)) { error = -EINVAL; break; } -- cgit v0.10.2 From dbcf31ba68c92b23ac3f874c4d516bacd5d7518a Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Sun, 1 May 2005 08:59:14 -0700 Subject: [PATCH] fix include order in mthca_memfree.c Fix order of #include lines in mthca_memfree.c Signed-off-by: Roland Dreier Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/infiniband/hw/mthca/mthca_memfree.c b/drivers/infiniband/hw/mthca/mthca_memfree.c index 986f218..637b30e 100644 --- a/drivers/infiniband/hw/mthca/mthca_memfree.c +++ b/drivers/infiniband/hw/mthca/mthca_memfree.c @@ -32,10 +32,11 @@ * $Id$ */ +#include + #include "mthca_memfree.h" #include "mthca_dev.h" #include "mthca_cmd.h" -#include /* * We allocate in as big chunks as we can, up to a maximum of 256 KB -- cgit v0.10.2 From 16f31113a614a35acc2da64f4ae39ab0ed7e194f Mon Sep 17 00:00:00 2001 From: Yum Rayan Date: Sun, 1 May 2005 08:59:14 -0700 Subject: [PATCH] serial_cs: Reduce stack usage in serial_event() This patch reduces the stack usage of the function serial_event() in serial_cs from 2212 to 228. I used a patched version of gcc 3.4.3 on i386 with -fno-unit-at-a-time disabled. This patch is only compile tested. Acked-by: Randy Dunlap Signed-off-by: Yum Rayan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/serial/serial_cs.c b/drivers/serial/serial_cs.c index 9034f9a..6eeb48f 100644 --- a/drivers/serial/serial_cs.c +++ b/drivers/serial/serial_cs.c @@ -107,6 +107,13 @@ struct serial_info { int line[4]; }; +struct serial_cfg_mem { + tuple_t tuple; + cisparse_t parse; + u_char buf[256]; +}; + + static void serial_config(dev_link_t * link); static int serial_event(event_t event, int priority, event_callback_args_t * args); @@ -357,14 +364,24 @@ static int simple_config(dev_link_t *link) static int size_table[2] = { 8, 16 }; client_handle_t handle = link->handle; struct serial_info *info = link->priv; - tuple_t tuple; - u_char buf[256]; - cisparse_t parse; - cistpl_cftable_entry_t *cf = &parse.cftable_entry; + struct serial_cfg_mem *cfg_mem; + tuple_t *tuple; + u_char *buf; + cisparse_t *parse; + cistpl_cftable_entry_t *cf; config_info_t config; int i, j, try; int s; + cfg_mem = kmalloc(sizeof(struct serial_cfg_mem), GFP_KERNEL); + if (!cfg_mem) + return -1; + + tuple = &cfg_mem->tuple; + parse = &cfg_mem->parse; + cf = &parse->cftable_entry; + buf = cfg_mem->buf; + /* If the card is already configured, look up the port and irq */ i = pcmcia_get_configuration_info(handle, &config); if ((i == CS_SUCCESS) && (config.Attributes & CONF_VALID_CLIENT)) { @@ -377,21 +394,23 @@ static int simple_config(dev_link_t *link) port = config.BasePort1 + 0x28; info->slave = 1; } - if (info->slave) + if (info->slave) { + kfree(cfg_mem); return setup_serial(handle, info, port, config.AssignedIRQ); + } } link->conf.Vcc = config.Vcc; /* First pass: look for a config entry that looks normal. */ - tuple.TupleData = (cisdata_t *) buf; - tuple.TupleOffset = 0; - tuple.TupleDataMax = 255; - tuple.Attributes = 0; - tuple.DesiredTuple = CISTPL_CFTABLE_ENTRY; + tuple->TupleData = (cisdata_t *) buf; + tuple->TupleOffset = 0; + tuple->TupleDataMax = 255; + tuple->Attributes = 0; + tuple->DesiredTuple = CISTPL_CFTABLE_ENTRY; /* Two tries: without IO aliases, then with aliases */ for (s = 0; s < 2; s++) { for (try = 0; try < 2; try++) { - i = first_tuple(handle, &tuple, &parse); + i = first_tuple(handle, tuple, parse); while (i != CS_NO_MORE_ITEMS) { if (i != CS_SUCCESS) goto next_entry; @@ -409,14 +428,14 @@ static int simple_config(dev_link_t *link) goto found_port; } next_entry: - i = next_tuple(handle, &tuple, &parse); + i = next_tuple(handle, tuple, parse); } } } /* Second pass: try to find an entry that isn't picky about its base address, then try to grab any standard serial port address, and finally try to get any free port. */ - i = first_tuple(handle, &tuple, &parse); + i = first_tuple(handle, tuple, parse); while (i != CS_NO_MORE_ITEMS) { if ((i == CS_SUCCESS) && (cf->io.nwin > 0) && ((cf->io.flags & CISTPL_IO_LINES_MASK) <= 3)) { @@ -429,7 +448,7 @@ next_entry: goto found_port; } } - i = next_tuple(handle, &tuple, &parse); + i = next_tuple(handle, tuple, parse); } found_port: @@ -437,6 +456,7 @@ next_entry: printk(KERN_NOTICE "serial_cs: no usable port range found, giving up\n"); cs_error(link->handle, RequestIO, i); + kfree(cfg_mem); return -1; } @@ -450,9 +470,10 @@ next_entry: i = pcmcia_request_configuration(link->handle, &link->conf); if (i != CS_SUCCESS) { cs_error(link->handle, RequestConfiguration, i); + kfree(cfg_mem); return -1; } - + kfree(cfg_mem); return setup_serial(handle, info, link->io.BasePort1, link->irq.AssignedIRQ); } @@ -460,29 +481,39 @@ static int multi_config(dev_link_t * link) { client_handle_t handle = link->handle; struct serial_info *info = link->priv; - tuple_t tuple; - u_char buf[256]; - cisparse_t parse; - cistpl_cftable_entry_t *cf = &parse.cftable_entry; + struct serial_cfg_mem *cfg_mem; + tuple_t *tuple; + u_char *buf; + cisparse_t *parse; + cistpl_cftable_entry_t *cf; config_info_t config; - int i, base2 = 0; + int i, rc, base2 = 0; + + cfg_mem = kmalloc(sizeof(struct serial_cfg_mem), GFP_KERNEL); + if (!cfg_mem) + return -1; + tuple = &cfg_mem->tuple; + parse = &cfg_mem->parse; + cf = &parse->cftable_entry; + buf = cfg_mem->buf; i = pcmcia_get_configuration_info(handle, &config); if (i != CS_SUCCESS) { cs_error(handle, GetConfigurationInfo, i); - return -1; + rc = -1; + goto free_cfg_mem; } link->conf.Vcc = config.Vcc; - tuple.TupleData = (cisdata_t *) buf; - tuple.TupleOffset = 0; - tuple.TupleDataMax = 255; - tuple.Attributes = 0; - tuple.DesiredTuple = CISTPL_CFTABLE_ENTRY; + tuple->TupleData = (cisdata_t *) buf; + tuple->TupleOffset = 0; + tuple->TupleDataMax = 255; + tuple->Attributes = 0; + tuple->DesiredTuple = CISTPL_CFTABLE_ENTRY; /* First, look for a generic full-sized window */ link->io.NumPorts1 = info->multi * 8; - i = first_tuple(handle, &tuple, &parse); + i = first_tuple(handle, tuple, parse); while (i != CS_NO_MORE_ITEMS) { /* The quad port cards have bad CIS's, so just look for a window larger than 8 ports and assume it will be right */ @@ -497,14 +528,14 @@ static int multi_config(dev_link_t * link) if (i == CS_SUCCESS) break; } - i = next_tuple(handle, &tuple, &parse); + i = next_tuple(handle, tuple, parse); } /* If that didn't work, look for two windows */ if (i != CS_SUCCESS) { link->io.NumPorts1 = link->io.NumPorts2 = 8; info->multi = 2; - i = first_tuple(handle, &tuple, &parse); + i = first_tuple(handle, tuple, parse); while (i != CS_NO_MORE_ITEMS) { if ((i == CS_SUCCESS) && (cf->io.nwin == 2)) { link->conf.ConfigIndex = cf->index; @@ -517,13 +548,14 @@ static int multi_config(dev_link_t * link) if (i == CS_SUCCESS) break; } - i = next_tuple(handle, &tuple, &parse); + i = next_tuple(handle, tuple, parse); } } if (i != CS_SUCCESS) { cs_error(link->handle, RequestIO, i); - return -1; + rc = -1; + goto free_cfg_mem; } i = pcmcia_request_irq(link->handle, &link->irq); @@ -541,7 +573,8 @@ static int multi_config(dev_link_t * link) i = pcmcia_request_configuration(link->handle, &link->conf); if (i != CS_SUCCESS) { cs_error(link->handle, RequestConfiguration, i); - return -1; + rc = -1; + goto free_cfg_mem; } /* The Oxford Semiconductor OXCF950 cards are in fact single-port: @@ -554,17 +587,23 @@ static int multi_config(dev_link_t * link) setup_serial(handle, info, link->io.BasePort1, link->irq.AssignedIRQ); outb(12, base2 + 1); } - return 0; + rc = 0; + goto free_cfg_mem; } setup_serial(handle, info, link->io.BasePort1, link->irq.AssignedIRQ); /* The Nokia cards are not really multiport cards */ - if (info->manfid == MANFID_NOKIA) - return 0; + if (info->manfid == MANFID_NOKIA) { + rc = 0; + goto free_cfg_mem; + } for (i = 0; i < info->multi - 1; i++) - setup_serial(handle, info, base2 + (8 * i), link->irq.AssignedIRQ); - - return 0; + setup_serial(handle, info, base2 + (8 * i), + link->irq.AssignedIRQ); + rc = 0; +free_cfg_mem: + kfree(cfg_mem); + return rc; } /*====================================================================== @@ -579,39 +618,49 @@ void serial_config(dev_link_t * link) { client_handle_t handle = link->handle; struct serial_info *info = link->priv; - tuple_t tuple; - u_short buf[128]; - cisparse_t parse; - cistpl_cftable_entry_t *cf = &parse.cftable_entry; + struct serial_cfg_mem *cfg_mem; + tuple_t *tuple; + u_char *buf; + cisparse_t *parse; + cistpl_cftable_entry_t *cf; int i, last_ret, last_fn; DEBUG(0, "serial_config(0x%p)\n", link); - tuple.TupleData = (cisdata_t *) buf; - tuple.TupleOffset = 0; - tuple.TupleDataMax = 255; - tuple.Attributes = 0; + cfg_mem = kmalloc(sizeof(struct serial_cfg_mem), GFP_KERNEL); + if (!cfg_mem) + goto failed; + + tuple = &cfg_mem->tuple; + parse = &cfg_mem->parse; + cf = &parse->cftable_entry; + buf = cfg_mem->buf; + + tuple->TupleData = (cisdata_t *) buf; + tuple->TupleOffset = 0; + tuple->TupleDataMax = 255; + tuple->Attributes = 0; /* Get configuration register information */ - tuple.DesiredTuple = CISTPL_CONFIG; - last_ret = first_tuple(handle, &tuple, &parse); + tuple->DesiredTuple = CISTPL_CONFIG; + last_ret = first_tuple(handle, tuple, parse); if (last_ret != CS_SUCCESS) { last_fn = ParseTuple; goto cs_failed; } - link->conf.ConfigBase = parse.config.base; - link->conf.Present = parse.config.rmask[0]; + link->conf.ConfigBase = parse->config.base; + link->conf.Present = parse->config.rmask[0]; /* Configure card */ link->state |= DEV_CONFIG; /* Is this a compliant multifunction card? */ - tuple.DesiredTuple = CISTPL_LONGLINK_MFC; - tuple.Attributes = TUPLE_RETURN_COMMON | TUPLE_RETURN_LINK; - info->multi = (first_tuple(handle, &tuple, &parse) == CS_SUCCESS); + tuple->DesiredTuple = CISTPL_LONGLINK_MFC; + tuple->Attributes = TUPLE_RETURN_COMMON | TUPLE_RETURN_LINK; + info->multi = (first_tuple(handle, tuple, parse) == CS_SUCCESS); /* Is this a multiport card? */ - tuple.DesiredTuple = CISTPL_MANFID; - if (first_tuple(handle, &tuple, &parse) == CS_SUCCESS) { + tuple->DesiredTuple = CISTPL_MANFID; + if (first_tuple(handle, tuple, parse) == CS_SUCCESS) { info->manfid = le16_to_cpu(buf[0]); for (i = 0; i < MULTI_COUNT; i++) if ((info->manfid == multi_id[i].manfid) && @@ -623,13 +672,13 @@ void serial_config(dev_link_t * link) /* Another check for dual-serial cards: look for either serial or multifunction cards that ask for appropriate IO port ranges */ - tuple.DesiredTuple = CISTPL_FUNCID; + tuple->DesiredTuple = CISTPL_FUNCID; if ((info->multi == 0) && - ((first_tuple(handle, &tuple, &parse) != CS_SUCCESS) || - (parse.funcid.func == CISTPL_FUNCID_MULTI) || - (parse.funcid.func == CISTPL_FUNCID_SERIAL))) { - tuple.DesiredTuple = CISTPL_CFTABLE_ENTRY; - if (first_tuple(handle, &tuple, &parse) == CS_SUCCESS) { + ((first_tuple(handle, tuple, parse) != CS_SUCCESS) || + (parse->funcid.func == CISTPL_FUNCID_MULTI) || + (parse->funcid.func == CISTPL_FUNCID_SERIAL))) { + tuple->DesiredTuple = CISTPL_CFTABLE_ENTRY; + if (first_tuple(handle, tuple, parse) == CS_SUCCESS) { if ((cf->io.nwin == 1) && (cf->io.win[0].len % 8 == 0)) info->multi = cf->io.win[0].len >> 3; if ((cf->io.nwin == 2) && (cf->io.win[0].len == 8) && @@ -664,6 +713,7 @@ void serial_config(dev_link_t * link) link->dev = &info->node[0]; link->state &= ~DEV_CONFIG_PENDING; + kfree(cfg_mem); return; cs_failed: @@ -671,6 +721,7 @@ void serial_config(dev_link_t * link) failed: serial_remove(link); link->state &= ~DEV_CONFIG_PENDING; + kfree(cfg_mem); } /*====================================================================== -- cgit v0.10.2 From efcd5e3ab089496865571d22443dd3f514dae60c Mon Sep 17 00:00:00 2001 From: Emanuele Giaquinta Date: Sun, 1 May 2005 08:59:15 -0700 Subject: [PATCH] Makefile: fix for compatibility with *emacs ctags I've noticed that, starting from linux-2.6.12-rc1, in the top Makefile the "cmd_tags" variable has been changed in a way incompatible with *emacs ctags. Since the "--extra" option exists only in "exuberant ctags", it should be included in the CTAGSF shell variable. Cc: Sam Ravnborg Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Makefile b/Makefile index 4899089..f7eb558 100644 --- a/Makefile +++ b/Makefile @@ -1190,8 +1190,8 @@ cmd_TAGS = $(all-sources) | etags - quiet_cmd_tags = MAKE $@ define cmd_tags rm -f $@; \ - CTAGSF=`ctags --version | grep -i exuberant >/dev/null && echo "-I __initdata,__exitdata,EXPORT_SYMBOL,EXPORT_SYMBOL_GPL"`; \ - $(all-sources) | xargs ctags $$CTAGSF -a --extra=+f + CTAGSF=`ctags --version | grep -i exuberant >/dev/null && echo "-I __initdata,__exitdata,EXPORT_SYMBOL,EXPORT_SYMBOL_GPL --extra=+f"`; \ + $(all-sources) | xargs ctags $$CTAGSF -a endef TAGS: FORCE -- cgit v0.10.2 From 212079cf4ee99e492a57b817e796825d423a30bb Mon Sep 17 00:00:00 2001 From: Ken Chen Date: Sun, 1 May 2005 08:59:15 -0700 Subject: [PATCH] aio: remove superfluous kiocb member initialization This patch removes superfluous kiocb member initialization in the AIO allocation and deallocation path. For example, in really_put_req(), right before kiocb is returned to slab, 5 variables are reset to NULL. The same variables will be initialized at the kiocb allocation time, so why bother reset them knowing that they will be set to valid data at alloc time? Another example: ki_retry is initialized in __aio_get_req, but is initialized again in io_submit_one. Signed-off-by: Ken Chen Cc: Benjamin LaHaise Cc: Suparna Bhattacharya Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/aio.c b/fs/aio.c index a82214d..9f807a5 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -405,7 +405,6 @@ static struct kiocb fastcall *__aio_get_req(struct kioctx *ctx) req->ki_ctx = ctx; req->ki_cancel = NULL; req->ki_retry = NULL; - req->ki_obj.user = NULL; req->ki_dtor = NULL; req->private = NULL; INIT_LIST_HEAD(&req->ki_run_list); @@ -451,11 +450,6 @@ static inline void really_put_req(struct kioctx *ctx, struct kiocb *req) { if (req->ki_dtor) req->ki_dtor(req); - req->ki_ctx = NULL; - req->ki_filp = NULL; - req->ki_obj.user = NULL; - req->ki_dtor = NULL; - req->private = NULL; kmem_cache_free(kiocb_cachep, req); ctx->reqs_active--; @@ -1515,8 +1509,7 @@ int fastcall io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, } req->ki_filp = file; - iocb->aio_key = req->ki_key; - ret = put_user(iocb->aio_key, &user_iocb->aio_key); + ret = put_user(req->ki_key, &user_iocb->aio_key); if (unlikely(ret)) { dprintk("EFAULT: aio_key\n"); goto out_put_req; @@ -1531,8 +1524,6 @@ int fastcall io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, req->ki_opcode = iocb->aio_lio_opcode; init_waitqueue_func_entry(&req->ki_wait, aio_wake_function); INIT_LIST_HEAD(&req->ki_wait.task_list); - req->ki_run_list.next = req->ki_run_list.prev = NULL; - req->ki_retry = NULL; req->ki_retried = 0; req->ki_kicked = 0; req->ki_queued = 0; -- cgit v0.10.2 From 4bf69b2a06090c01c27f25ea5cd1440f7bf9256f Mon Sep 17 00:00:00 2001 From: Ken Chen Date: Sun, 1 May 2005 08:59:15 -0700 Subject: [PATCH] aio: ring wrapping simplification Since the tail pointer in aio_ring structure never wrap ring size more than once, so a simple compare is sufficient to wrap the index around. This avoid a more expensive mod operation. Signed-off-by: Ken Chen Cc: Benjamin LaHaise Cc: Suparna Bhattacharya Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/aio.c b/fs/aio.c index 9f807a5..40517f3 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -978,7 +978,8 @@ int fastcall aio_complete(struct kiocb *iocb, long res, long res2) tail = info->tail; event = aio_ring_event(info, tail, KM_IRQ0); - tail = (tail + 1) % info->nr; + if (++tail >= info->nr) + tail = 0; event->obj = (u64)(unsigned long)iocb->ki_obj.user; event->data = iocb->ki_user_data; -- cgit v0.10.2 From 644d3a088a3b862ed0b57c286cf58a6bd338ce08 Mon Sep 17 00:00:00 2001 From: Ken Chen Date: Sun, 1 May 2005 08:59:15 -0700 Subject: [PATCH] aio: clean up debug code Clean up code that was previously used for debug purpose. Remove aio_run, aio_wakeups, iocb->ki_queued and iocb->ki_kicked. Also clean up unused variable count in __aio_run_iocbs() and debug code in read_events(). Signed-off-by: Ken Chen Cc: Benjamin LaHaise Cc: Suparna Bhattacharya Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/aio.c b/fs/aio.c index 40517f3..674bb47 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -40,9 +40,6 @@ #define dprintk(x...) do { ; } while (0) #endif -static long aio_run = 0; /* for testing only */ -static long aio_wakeups = 0; /* for testing only */ - /*------ sysctl variables----*/ atomic_t aio_nr = ATOMIC_INIT(0); /* current system wide number of aio requests */ unsigned aio_max_nr = 0x10000; /* system wide maximum number of aio requests */ @@ -617,7 +614,6 @@ static inline int __queue_kicked_iocb(struct kiocb *iocb) if (list_empty(&iocb->ki_run_list)) { list_add_tail(&iocb->ki_run_list, &ctx->run_list); - iocb->ki_queued++; return 1; } return 0; @@ -658,10 +654,8 @@ static ssize_t aio_run_iocb(struct kiocb *iocb) } if (!(iocb->ki_retried & 0xff)) { - pr_debug("%ld retry: %d of %d (kick %ld, Q %ld run %ld, wake %ld)\n", - iocb->ki_retried, - iocb->ki_nbytes - iocb->ki_left, iocb->ki_nbytes, - iocb->ki_kicked, iocb->ki_queued, aio_run, aio_wakeups); + pr_debug("%ld retry: %d of %d\n", iocb->ki_retried, + iocb->ki_nbytes - iocb->ki_left, iocb->ki_nbytes); } if (!(retry = iocb->ki_retry)) { @@ -768,7 +762,6 @@ out: static int __aio_run_iocbs(struct kioctx *ctx) { struct kiocb *iocb; - int count = 0; LIST_HEAD(run_list); list_splice_init(&ctx->run_list, &run_list); @@ -783,9 +776,7 @@ static int __aio_run_iocbs(struct kioctx *ctx) aio_run_iocb(iocb); if (__aio_put_req(ctx, iocb)) /* drop extra ref */ put_ioctx(ctx); - count++; } - aio_run++; if (!list_empty(&ctx->run_list)) return 1; return 0; @@ -884,10 +875,8 @@ static void queue_kicked_iocb(struct kiocb *iocb) spin_lock_irqsave(&ctx->ctx_lock, flags); run = __queue_kicked_iocb(iocb); spin_unlock_irqrestore(&ctx->ctx_lock, flags); - if (run) { + if (run) aio_queue_work(ctx); - aio_wakeups++; - } } /* @@ -907,7 +896,6 @@ void fastcall kick_iocb(struct kiocb *iocb) return; } - iocb->ki_kicked++; /* If its already kicked we shouldn't queue it again */ if (!kiocbTryKick(iocb)) { queue_kicked_iocb(iocb); @@ -1003,10 +991,8 @@ int fastcall aio_complete(struct kiocb *iocb, long res, long res2) pr_debug("added to ring %p at [%lu]\n", iocb, tail); - pr_debug("%ld retries: %d of %d (kicked %ld, Q %ld run %ld wake %ld)\n", - iocb->ki_retried, - iocb->ki_nbytes - iocb->ki_left, iocb->ki_nbytes, - iocb->ki_kicked, iocb->ki_queued, aio_run, aio_wakeups); + pr_debug("%ld retries: %d of %d\n", iocb->ki_retried, + iocb->ki_nbytes - iocb->ki_left, iocb->ki_nbytes); put_rq: /* everything turned out well, dispose of the aiocb. */ ret = __aio_put_req(ctx, iocb); @@ -1114,7 +1100,6 @@ static int read_events(struct kioctx *ctx, int i = 0; struct io_event ent; struct aio_timeout to; - int event_loop = 0; /* testing only */ int retry = 0; /* needed to zero any padding within an entry (there shouldn't be @@ -1181,7 +1166,6 @@ retry: if (to.timed_out) /* Only check after read evt */ break; schedule(); - event_loop++; if (signal_pending(tsk)) { ret = -EINTR; break; @@ -1209,9 +1193,6 @@ retry: if (timeout) clear_timeout(&to); out: - pr_debug("event loop executed %d times\n", event_loop); - pr_debug("aio_run %ld\n", aio_run); - pr_debug("aio_wakeups %ld\n", aio_wakeups); return i ? i : ret; } @@ -1526,10 +1507,6 @@ int fastcall io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, init_waitqueue_func_entry(&req->ki_wait, aio_wake_function); INIT_LIST_HEAD(&req->ki_wait.task_list); req->ki_retried = 0; - req->ki_kicked = 0; - req->ki_queued = 0; - aio_run = 0; - aio_wakeups = 0; ret = aio_setup_iocb(req); -- cgit v0.10.2 From 954d3e95369cf73b4bc1e570729f68264a0e6fe0 Mon Sep 17 00:00:00 2001 From: Ken Chen Date: Sun, 1 May 2005 08:59:16 -0700 Subject: [PATCH] aio: optimize io_submit_one() This patch optimizes io_submit_one to call aio_run_iocb() directly if ctx->run_list is empty. When the list is empty, the operation of adding to the list, then call to __aio_run_iocbs() is unnecessary because these operations are done in one atomic step. ctx->run_list always has only one element in this case. This optimization speeds up industry standard db transaction processing benchmark by 0.2%. Signed-off-by: Ken Chen Cc: Benjamin LaHaise Cc: Suparna Bhattacharya Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/aio.c b/fs/aio.c index 674bb47..7afa222 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -1514,10 +1514,14 @@ int fastcall io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, goto out_put_req; spin_lock_irq(&ctx->ctx_lock); - list_add_tail(&req->ki_run_list, &ctx->run_list); - /* drain the run list */ - while (__aio_run_iocbs(ctx)) - ; + if (likely(list_empty(&ctx->run_list))) { + aio_run_iocb(req); + } else { + list_add_tail(&req->ki_run_list, &ctx->run_list); + /* drain the run list */ + while (__aio_run_iocbs(ctx)) + ; + } spin_unlock_irq(&ctx->ctx_lock); aio_put_req(req); /* drop extra ref to req */ return 0; -- cgit v0.10.2 From 945b092011c6af71a0107be96e119c8c08776f3f Mon Sep 17 00:00:00 2001 From: Colin Leroy Date: Sun, 1 May 2005 08:59:16 -0700 Subject: [PATCH] hfs, hfsplus: don't leak s_fs_info and fix an oops This patch fixes the leak of sb->s_fs_info in both the HFS and HFS+ modules. In addition to this, it fixes an oops happening when trying to mount a non-hfsplus filesystem using hfsplus. This patch is from Roman Zippel, based off patches sent by myself. Signed-off-by: Colin Leroy Signed-off-by: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/hfs/mdb.c b/fs/hfs/mdb.c index 4efb640..217e32f 100644 --- a/fs/hfs/mdb.c +++ b/fs/hfs/mdb.c @@ -333,6 +333,8 @@ void hfs_mdb_close(struct super_block *sb) * Release the resources associated with the in-core MDB. */ void hfs_mdb_put(struct super_block *sb) { + if (!HFS_SB(sb)) + return; /* free the B-trees */ hfs_btree_close(HFS_SB(sb)->ext_tree); hfs_btree_close(HFS_SB(sb)->cat_tree); @@ -340,4 +342,7 @@ void hfs_mdb_put(struct super_block *sb) /* free the buffers holding the primary and alternate MDBs */ brelse(HFS_SB(sb)->mdb_bh); brelse(HFS_SB(sb)->alt_mdb_bh); + + kfree(HFS_SB(sb)); + sb->s_fs_info = NULL; } diff --git a/fs/hfs/super.c b/fs/hfs/super.c index 1e2c193..ab783f6 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -297,7 +297,7 @@ static int hfs_fill_super(struct super_block *sb, void *data, int silent) res = -EINVAL; if (!parse_options((char *)data, sbi)) { hfs_warn("hfs_fs: unable to parse mount options.\n"); - goto bail3; + goto bail; } sb->s_op = &hfs_super_operations; @@ -310,7 +310,7 @@ static int hfs_fill_super(struct super_block *sb, void *data, int silent) hfs_warn("VFS: Can't find a HFS filesystem on dev %s.\n", hfs_mdb_name(sb)); res = -EINVAL; - goto bail2; + goto bail; } /* try to get the root inode */ @@ -340,10 +340,8 @@ bail_iput: iput(root_inode); bail_no_root: hfs_warn("hfs_fs: get root inode failed.\n"); +bail: hfs_mdb_put(sb); -bail2: -bail3: - kfree(sbi); return res; } diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 5f80446..d55ad67 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -208,7 +208,9 @@ static void hfsplus_write_super(struct super_block *sb) static void hfsplus_put_super(struct super_block *sb) { dprint(DBG_SUPER, "hfsplus_put_super\n"); - if (!(sb->s_flags & MS_RDONLY)) { + if (!sb->s_fs_info) + return; + if (!(sb->s_flags & MS_RDONLY) && HFSPLUS_SB(sb).s_vhdr) { struct hfsplus_vh *vhdr = HFSPLUS_SB(sb).s_vhdr; vhdr->modify_date = hfsp_now2mt(); @@ -226,6 +228,8 @@ static void hfsplus_put_super(struct super_block *sb) brelse(HFSPLUS_SB(sb).s_vhbh); if (HFSPLUS_SB(sb).nls) unload_nls(HFSPLUS_SB(sb).nls); + kfree(sb->s_fs_info); + sb->s_fs_info = NULL; } static int hfsplus_statfs(struct super_block *sb, struct kstatfs *buf) -- cgit v0.10.2 From 4dcd00b18118d174c4b8d838c11f437f0af3c20c Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Sun, 1 May 2005 08:59:16 -0700 Subject: [PATCH] autofs4: wait order fix It's possible for an event wait request to arive before the event requestor. If this happens the daemon never gets notified and autofs hangs. Signed-off-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index f5a52c8..9789877 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -84,6 +84,7 @@ struct autofs_wait_queue { char *name; /* This is for status reporting upon return */ int status; + atomic_t notified; atomic_t wait_ctr; }; diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index 1ab24a6..5a40d36 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -210,17 +210,8 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, wq->len = len; wq->status = -EINTR; /* Status return if interrupted */ atomic_set(&wq->wait_ctr, 2); + atomic_set(&wq->notified, 1); up(&sbi->wq_sem); - - DPRINTK("new wait id = 0x%08lx, name = %.*s, nfy=%d", - (unsigned long) wq->wait_queue_token, wq->len, wq->name, notify); - /* autofs4_notify_daemon() may block */ - if (notify != NFY_NONE) { - autofs4_notify_daemon(sbi,wq, - notify == NFY_MOUNT ? - autofs_ptype_missing : - autofs_ptype_expire_multi); - } } else { atomic_inc(&wq->wait_ctr); up(&sbi->wq_sem); @@ -229,6 +220,17 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, (unsigned long) wq->wait_queue_token, wq->len, wq->name, notify); } + if (notify != NFY_NONE && atomic_dec_and_test(&wq->notified)) { + int type = (notify == NFY_MOUNT ? + autofs_ptype_missing : autofs_ptype_expire_multi); + + DPRINTK(("new wait id = 0x%08lx, name = %.*s, nfy=%d\n", + (unsigned long) wq->wait_queue_token, wq->len, wq->name, notify)); + + /* autofs4_notify_daemon() may block */ + autofs4_notify_daemon(sbi, wq, type); + } + /* wq->name is NULL if and only if the lock is already released */ if ( sbi->catatonic ) { -- cgit v0.10.2 From 3a9720ce73c9247e5262922d65e90444ea75eb50 Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Sun, 1 May 2005 08:59:17 -0700 Subject: [PATCH] autofs4: tree race fix For tree mount maps, a call to chdir or chroot, to a directory above the moint point directories at a certain time during the expire results in the expire incorrectly thinking the tree is not busy. This patch adds a check to see if the filesystem above the tree mount points is busy and also locks the filesystem during the tree mount expire to prevent the race. Signed-off-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index 9789877..c7b2b88 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -102,6 +102,7 @@ struct autofs_sb_info { int needs_reghost; struct super_block *sb; struct semaphore wq_sem; + spinlock_t fs_lock; struct autofs_wait_queue *queues; /* Wait queue pointer */ }; @@ -127,9 +128,18 @@ static inline int autofs4_oz_mode(struct autofs_sb_info *sbi) { static inline int autofs4_ispending(struct dentry *dentry) { struct autofs_info *inf = autofs4_dentry_ino(dentry); + int pending = 0; - return (dentry->d_flags & DCACHE_AUTOFS_PENDING) || - (inf != NULL && inf->flags & AUTOFS_INF_EXPIRING); + if (dentry->d_flags & DCACHE_AUTOFS_PENDING) + return 1; + + if (inf) { + spin_lock(&inf->sbi->fs_lock); + pending = inf->flags & AUTOFS_INF_EXPIRING; + spin_unlock(&inf->sbi->fs_lock); + } + + return pending; } static inline void autofs4_copy_atime(struct file *src, struct file *dst) diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c index 31540a6..500425e 100644 --- a/fs/autofs4/expire.c +++ b/fs/autofs4/expire.c @@ -99,6 +99,10 @@ static int autofs4_check_tree(struct vfsmount *mnt, if (!autofs4_can_expire(top, timeout, do_now)) return 0; + /* Is someone visiting anywhere in the tree ? */ + if (may_umount_tree(mnt)) + return 0; + spin_lock(&dcache_lock); repeat: next = this_parent->d_subdirs.next; @@ -270,10 +274,18 @@ static struct dentry *autofs4_expire(struct super_block *sb, /* Case 2: tree mount, expire iff entire tree is not busy */ if (!exp_leaves) { + /* Lock the tree as we must expire as a whole */ + spin_lock(&sbi->fs_lock); if (autofs4_check_tree(mnt, dentry, timeout, do_now)) { - expired = dentry; - break; + struct autofs_info *inf = autofs4_dentry_ino(dentry); + + /* Set this flag early to catch sys_chdir and the like */ + inf->flags |= AUTOFS_INF_EXPIRING; + spin_unlock(&sbi->fs_lock); + expired = dentry; + break; } + spin_unlock(&sbi->fs_lock); /* Case 3: direct mount, expire individual leaves */ } else { expired = autofs4_check_leaves(mnt, dentry, timeout, do_now); diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index a525607..4bb14cc 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -206,6 +206,7 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent) sbi->version = 0; sbi->sub_version = 0; init_MUTEX(&sbi->wq_sem); + spin_lock_init(&sbi->fs_lock); sbi->queues = NULL; s->s_blocksize = 1024; s->s_blocksize_bits = 10; -- cgit v0.10.2 From d46aa455dd5457fdbebad17db4ff4df655cbfbae Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Sun, 1 May 2005 08:59:17 -0700 Subject: [PATCH] autofs4: bump version number Bump autofs4 version so we know what's going on. Signed-off-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/auto_fs4.h b/include/linux/auto_fs4.h index d1c7b0e..a1657fb 100644 --- a/include/linux/auto_fs4.h +++ b/include/linux/auto_fs4.h @@ -23,7 +23,7 @@ #define AUTOFS_MIN_PROTO_VERSION 3 #define AUTOFS_MAX_PROTO_VERSION 4 -#define AUTOFS_PROTO_SUBVERSION 5 +#define AUTOFS_PROTO_SUBVERSION 6 /* Mask for expire behaviour */ #define AUTOFS_EXP_IMMEDIATE 1 -- cgit v0.10.2 From 6a3a16f2ef6f335286e2b2bf8284b0ab4ff38ec0 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 1 May 2005 08:59:17 -0700 Subject: [PATCH] reiserfs endianness: clone struct reiserfs_key struct reiserfs_key cloned; (currently) identical struct in_core_key added. Places that expect host-endian data in reiserfs_key switched to in_core_key. Basically, we get annotation of reiserfs_key users and keep the resulting tree obviously equivalent to original. Signed-off-by: Al Viro Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/reiserfs/bitmap.c b/fs/reiserfs/bitmap.c index a4e2ed5..f4f16fa 100644 --- a/fs/reiserfs/bitmap.c +++ b/fs/reiserfs/bitmap.c @@ -736,7 +736,7 @@ static inline int this_blocknr_allocation_would_make_it_a_large_file(reiserfs_bl #ifdef DISPLACE_NEW_PACKING_LOCALITIES static inline void displace_new_packing_locality (reiserfs_blocknr_hint_t *hint) { - struct reiserfs_key * key = &hint->key; + struct in_core_key * key = &hint->key; hint->th->displace_new_blocks = 0; hint->search_start = hint->beg + keyed_hash((char*)(&key->k_objectid),4) % (hint->end - hint->beg); @@ -777,7 +777,7 @@ static inline int old_way (reiserfs_blocknr_hint_t * hint) static inline void hundredth_slices (reiserfs_blocknr_hint_t * hint) { - struct reiserfs_key * key = &hint->key; + struct in_core_key * key = &hint->key; b_blocknr_t slice_start; slice_start = (keyed_hash((char*)(&key->k_dir_id),4) % 100) * (hint->end / 100); diff --git a/fs/reiserfs/stree.c b/fs/reiserfs/stree.c index 73ec521..1d380a5 100644 --- a/fs/reiserfs/stree.c +++ b/fs/reiserfs/stree.c @@ -229,6 +229,7 @@ const struct reiserfs_key MIN_KEY = {0, 0, {{0, 0},}}; /* Maximal possible key. It is never in the tree. */ const struct reiserfs_key MAX_KEY = {0xffffffff, 0xffffffff, {{0xffffffff, 0xffffffff},}}; +const struct in_core_key MAX_IN_CORE_KEY = {0xffffffff, 0xffffffff, {{0xffffffff, 0xffffffff},}}; /* Get delimiting key of the buffer by looking for it in the buffers in the path, starting from the bottom diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index bc5e889..d6d1d7e 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -110,7 +110,7 @@ static void reiserfs_unlockfs(struct super_block *s) { reiserfs_allow_writes(s) ; } -extern const struct reiserfs_key MAX_KEY; +extern const struct in_core_key MAX_IN_CORE_KEY; /* this is used to delete "save link" when there are no items of a @@ -164,7 +164,7 @@ static int finish_unfinished (struct super_block * s) /* compose key to look for "save" links */ max_cpu_key.version = KEY_FORMAT_3_5; - max_cpu_key.on_disk_key = MAX_KEY; + max_cpu_key.on_disk_key = MAX_IN_CORE_KEY; max_cpu_key.key_length = 3; #ifdef CONFIG_QUOTA diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h index bccff8b..d086787 100644 --- a/include/linux/reiserfs_fs.h +++ b/include/linux/reiserfs_fs.h @@ -433,6 +433,23 @@ static inline void set_offset_v2_k_offset( struct offset_v2 *v2, loff_t offset ) # define set_offset_v2_k_offset(v2,val) (offset_v2_k_offset(v2) = (val)) #endif +struct in_core_offset_v1 { + __u32 k_offset; + __u32 k_uniqueness; +} __attribute__ ((__packed__)); + +struct in_core_offset_v2 { +#ifdef __LITTLE_ENDIAN + /* little endian version */ + __u64 k_offset:60; + __u64 k_type: 4; +#else + /* big endian version */ + __u64 k_type: 4; + __u64 k_offset:60; +#endif +} __attribute__ ((__packed__)); + /* Key of an item determines its location in the S+tree, and is composed of 4 components */ struct reiserfs_key { @@ -445,9 +462,18 @@ struct reiserfs_key { } __attribute__ ((__packed__)) u; } __attribute__ ((__packed__)); +struct in_core_key { + __u32 k_dir_id; /* packing locality: by default parent + directory object id */ + __u32 k_objectid; /* object identifier */ + union { + struct in_core_offset_v1 k_offset_v1; + struct in_core_offset_v2 k_offset_v2; + } __attribute__ ((__packed__)) u; +} __attribute__ ((__packed__)); struct cpu_key { - struct reiserfs_key on_disk_key; + struct in_core_key on_disk_key; int version; int key_length; /* 3 in all cases but direct2indirect and indirect2direct conversion */ @@ -1476,7 +1502,7 @@ struct tree_balance int fs_gen; /* saved value of `reiserfs_generation' counter see FILESYSTEM_CHANGED() macro in reiserfs_fs.h */ #ifdef DISPLACE_NEW_PACKING_LOCALITIES - struct reiserfs_key key; /* key pointer, to pass to block allocator or + struct in_core_key key; /* key pointer, to pass to block allocator or another low-level subsystem */ #endif } ; @@ -2117,7 +2143,7 @@ struct buffer_head * get_FEB (struct tree_balance *); struct __reiserfs_blocknr_hint { struct inode * inode; /* inode passed to allocator, if we allocate unf. nodes */ long block; /* file offset, in blocks */ - struct reiserfs_key key; + struct in_core_key key; struct path * path; /* search path, used by allocator to deternine search_start by * various ways */ struct reiserfs_transaction_handle * th; /* transaction handle is needed to log super blocks and -- cgit v0.10.2 From 3e8962be915bacc1d70e4849a075041838d60a3f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 1 May 2005 08:59:18 -0700 Subject: [PATCH] reiserfs endianness: annotate little-endian objects little-endian objects annotated as such; again, obviously no changes of resulting code, we only replace __u16 with __le16, etc. in relevant places. Signed-off-by: Al Viro Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/reiserfs/bitmap.c b/fs/reiserfs/bitmap.c index f4f16fa..49c479c 100644 --- a/fs/reiserfs/bitmap.c +++ b/fs/reiserfs/bitmap.c @@ -260,8 +260,9 @@ static inline int block_group_used(struct super_block *s, u32 id) { /* * the packing is returned in disk byte order */ -u32 reiserfs_choose_packing(struct inode *dir) { - u32 packing; +__le32 reiserfs_choose_packing(struct inode *dir) +{ + __le32 packing; if (TEST_OPTION(packing_groups, dir->i_sb)) { u32 parent_dir = le32_to_cpu(INODE_PKEY(dir)->k_dir_id); /* @@ -655,7 +656,7 @@ static int get_left_neighbor(reiserfs_blocknr_hint_t *hint) struct buffer_head * bh; struct item_head * ih; int pos_in_item; - __u32 * item; + __le32 * item; int ret = 0; if (!hint->path) /* reiserfs code can call this function w/o pointer to path diff --git a/fs/reiserfs/dir.c b/fs/reiserfs/dir.c index d1514a9..fbde4b0 100644 --- a/fs/reiserfs/dir.c +++ b/fs/reiserfs/dir.c @@ -209,8 +209,8 @@ static int reiserfs_readdir (struct file * filp, void * dirent, filldir_t filldi /* compose directory item containing "." and ".." entries (entries are not aligned to 4 byte boundary) */ /* the last four params are LE */ -void make_empty_dir_item_v1 (char * body, __u32 dirid, __u32 objid, - __u32 par_dirid, __u32 par_objid) +void make_empty_dir_item_v1 (char * body, __le32 dirid, __le32 objid, + __le32 par_dirid, __le32 par_objid) { struct reiserfs_de_head * deh; @@ -242,8 +242,8 @@ void make_empty_dir_item_v1 (char * body, __u32 dirid, __u32 objid, } /* compose directory item containing "." and ".." entries */ -void make_empty_dir_item (char * body, __u32 dirid, __u32 objid, - __u32 par_dirid, __u32 par_objid) +void make_empty_dir_item (char * body, __le32 dirid, __le32 objid, + __le32 par_dirid, __le32 par_objid) { struct reiserfs_de_head * deh; diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c index f6860e8..2230aff 100644 --- a/fs/reiserfs/file.c +++ b/fs/reiserfs/file.c @@ -166,7 +166,7 @@ static int reiserfs_allocate_blocks_for_region( struct cpu_key key; // cpu key of item that we are going to deal with struct item_head *ih; // pointer to item head that we are going to deal with struct buffer_head *bh; // Buffer head that contains items that we are going to deal with - __u32 * item; // pointer to item we are going to deal with + __le32 * item; // pointer to item we are going to deal with INITIALIZE_PATH(path); // path to item, that we are going to deal with. b_blocknr_t *allocated_blocks; // Pointer to a place where allocated blocknumbers would be stored. reiserfs_blocknr_hint_t hint; // hint structure for block allocator. @@ -891,7 +891,7 @@ static int reiserfs_prepare_file_region_for_write( struct item_head *ih = NULL; // pointer to item head that we are going to deal with struct buffer_head *itembuf=NULL; // Buffer head that contains items that we are going to deal with INITIALIZE_PATH(path); // path to item, that we are going to deal with. - __u32 * item=NULL; // pointer to item we are going to deal with + __le32 * item=NULL; // pointer to item we are going to deal with int item_pos=-1; /* Position in indirect item */ diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 7543031..5fdb9f9 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -173,7 +173,7 @@ static inline void fix_tail_page_for_writing(struct page *page) { done already or non-hole position has been found in the indirect item */ static inline int allocation_needed (int retval, b_blocknr_t allocated, struct item_head * ih, - __u32 * item, int pos_in_item) + __le32 * item, int pos_in_item) { if (allocated) return 0; @@ -278,7 +278,7 @@ research: bh = get_last_bh (&path); ih = get_ih (&path); if (is_indirect_le_ih (ih)) { - __u32 * ind_item = (__u32 *)B_I_PITEM (bh, ih); + __le32 * ind_item = (__le32 *)B_I_PITEM (bh, ih); /* FIXME: here we could cache indirect item or part of it in the inode to avoid search_by_key in case of subsequent @@ -581,7 +581,7 @@ int reiserfs_get_block (struct inode * inode, sector_t block, struct cpu_key key; struct buffer_head * bh, * unbh = NULL; struct item_head * ih, tmp_ih; - __u32 * item; + __le32 * item; int done; int fs_gen; struct reiserfs_transaction_handle *th = NULL; @@ -746,7 +746,7 @@ start_trans: done = 0; do { if (is_statdata_le_ih (ih)) { - __u32 unp = 0; + __le32 unp = 0; struct cpu_key tmp_key; /* indirect item has to be inserted */ @@ -2067,7 +2067,7 @@ static int map_block_for_writepage(struct inode *inode, struct item_head tmp_ih ; struct item_head *ih ; struct buffer_head *bh ; - __u32 *item ; + __le32 *item ; struct cpu_key key ; INITIALIZE_PATH(path) ; int pos_in_item ; diff --git a/fs/reiserfs/item_ops.c b/fs/reiserfs/item_ops.c index 9cf7c13..0ce33db 100644 --- a/fs/reiserfs/item_ops.c +++ b/fs/reiserfs/item_ops.c @@ -296,10 +296,11 @@ static void print_sequence (__u32 start, int len) static void indirect_print_item (struct item_head * ih, char * item) { int j; - __u32 * unp, prev = INT_MAX; + __le32 * unp; + __u32 prev = INT_MAX; int num; - unp = (__u32 *)item; + unp = (__le32 *)item; if (ih_item_len(ih) % UNFM_P_SIZE) reiserfs_warning (NULL, "indirect_print_item: invalid item len"); diff --git a/fs/reiserfs/objectid.c b/fs/reiserfs/objectid.c index 0785c43..bfe8e25 100644 --- a/fs/reiserfs/objectid.c +++ b/fs/reiserfs/objectid.c @@ -11,13 +11,13 @@ // find where objectid map starts #define objectid_map(s,rs) (old_format_only (s) ? \ - (__u32 *)((struct reiserfs_super_block_v1 *)(rs) + 1) :\ - (__u32 *)((rs) + 1)) + (__le32 *)((struct reiserfs_super_block_v1 *)(rs) + 1) :\ + (__le32 *)((rs) + 1)) #ifdef CONFIG_REISERFS_CHECK -static void check_objectid_map (struct super_block * s, __u32 * map) +static void check_objectid_map (struct super_block * s, __le32 * map) { if (le32_to_cpu (map[0]) != 1) reiserfs_panic (s, "vs-15010: check_objectid_map: map corrupted: %lx", @@ -27,7 +27,7 @@ static void check_objectid_map (struct super_block * s, __u32 * map) } #else -static void check_objectid_map (struct super_block * s, __u32 * map) +static void check_objectid_map (struct super_block * s, __le32 * map) {;} #endif @@ -52,7 +52,7 @@ __u32 reiserfs_get_unused_objectid (struct reiserfs_transaction_handle *th) { struct super_block * s = th->t_super; struct reiserfs_super_block * rs = SB_DISK_SUPER_BLOCK (s); - __u32 * map = objectid_map (s, rs); + __le32 * map = objectid_map (s, rs); __u32 unused_objectid; BUG_ON (!th->t_trans_id); @@ -97,7 +97,7 @@ void reiserfs_release_objectid (struct reiserfs_transaction_handle *th, { struct super_block * s = th->t_super; struct reiserfs_super_block * rs = SB_DISK_SUPER_BLOCK (s); - __u32 * map = objectid_map (s, rs); + __le32 * map = objectid_map (s, rs); int i = 0; BUG_ON (!th->t_trans_id); @@ -172,12 +172,12 @@ int reiserfs_convert_objectid_map_v1(struct super_block *s) { int new_size = (s->s_blocksize - SB_SIZE) / sizeof(__u32) / 2 * 2 ; int old_max = sb_oid_maxsize(disk_sb); struct reiserfs_super_block_v1 *disk_sb_v1 ; - __u32 *objectid_map, *new_objectid_map ; + __le32 *objectid_map, *new_objectid_map ; int i ; disk_sb_v1=(struct reiserfs_super_block_v1 *)(SB_BUFFER_WITH_SB(s)->b_data); - objectid_map = (__u32 *)(disk_sb_v1 + 1) ; - new_objectid_map = (__u32 *)(disk_sb + 1) ; + objectid_map = (__le32 *)(disk_sb_v1 + 1) ; + new_objectid_map = (__le32 *)(disk_sb + 1) ; if (cur_size > new_size) { /* mark everyone used that was listed as free at the end of the objectid diff --git a/fs/reiserfs/procfs.c b/fs/reiserfs/procfs.c index f4ea81a..e242ebc 100644 --- a/fs/reiserfs/procfs.c +++ b/fs/reiserfs/procfs.c @@ -73,8 +73,8 @@ int reiserfs_global_version_in_proc( char *buffer, char **start, off_t offset, #define DFL( x ) D4C( rs -> s_v1.x ) #define objectid_map( s, rs ) (old_format_only (s) ? \ - (__u32 *)((struct reiserfs_super_block_v1 *)rs + 1) : \ - (__u32 *)(rs + 1)) + (__le32 *)((struct reiserfs_super_block_v1 *)rs + 1) : \ + (__le32 *)(rs + 1)) #define MAP( i ) D4C( objectid_map( sb, rs )[ i ] ) #define DJF( x ) le32_to_cpu( rs -> x ) diff --git a/fs/reiserfs/stree.c b/fs/reiserfs/stree.c index 1d380a5..15fa4cb 100644 --- a/fs/reiserfs/stree.c +++ b/fs/reiserfs/stree.c @@ -87,10 +87,11 @@ inline void copy_item_head(struct item_head * p_v_to, inline int comp_short_keys (const struct reiserfs_key * le_key, const struct cpu_key * cpu_key) { - __u32 * p_s_le_u32, * p_s_cpu_u32; + __le32 * p_s_le_u32; + __u32 * p_s_cpu_u32; int n_key_length = REISERFS_SHORT_KEY_LEN; - p_s_le_u32 = (__u32 *)le_key; + p_s_le_u32 = (__le32 *)le_key; p_s_cpu_u32 = (__u32 *)&cpu_key->on_disk_key; for( ; n_key_length--; ++p_s_le_u32, ++p_s_cpu_u32 ) { if ( le32_to_cpu (*p_s_le_u32) < *p_s_cpu_u32 ) @@ -228,7 +229,12 @@ extern struct tree_balance * cur_tb; const struct reiserfs_key MIN_KEY = {0, 0, {{0, 0},}}; /* Maximal possible key. It is never in the tree. */ -const struct reiserfs_key MAX_KEY = {0xffffffff, 0xffffffff, {{0xffffffff, 0xffffffff},}}; +const struct reiserfs_key MAX_KEY = { + __constant_cpu_to_le32(0xffffffff), + __constant_cpu_to_le32(0xffffffff), + {{__constant_cpu_to_le32(0xffffffff), + __constant_cpu_to_le32(0xffffffff)},} +}; const struct in_core_key MAX_IN_CORE_KEY = {0xffffffff, 0xffffffff, {{0xffffffff, 0xffffffff},}}; @@ -998,7 +1004,7 @@ static char prepare_for_delete_or_cut( int n_unfm_number, /* Number of the item unformatted nodes. */ n_counter, n_blk_size; - __u32 * p_n_unfm_pointer; /* Pointer to the unformatted node number. */ + __le32 * p_n_unfm_pointer; /* Pointer to the unformatted node number. */ __u32 tmp; struct item_head s_ih; /* Item header. */ char c_mode; /* Returned mode of the balance. */ @@ -1060,7 +1066,7 @@ static char prepare_for_delete_or_cut( /* pointers to be cut */ n_unfm_number -= pos_in_item (p_s_path); /* Set pointer to the last unformatted node pointer that is to be cut. */ - p_n_unfm_pointer = (__u32 *)B_I_PITEM(p_s_bh, &s_ih) + I_UNFM_NUM(&s_ih) - 1 - *p_n_removed; + p_n_unfm_pointer = (__le32 *)B_I_PITEM(p_s_bh, &s_ih) + I_UNFM_NUM(&s_ih) - 1 - *p_n_removed; /* We go through the unformatted nodes pointers of the indirect @@ -1082,8 +1088,8 @@ static char prepare_for_delete_or_cut( need_research = 1 ; break; } - RFALSE( p_n_unfm_pointer < (__u32 *)B_I_PITEM(p_s_bh, &s_ih) || - p_n_unfm_pointer > (__u32 *)B_I_PITEM(p_s_bh, &s_ih) + I_UNFM_NUM(&s_ih) - 1, + RFALSE( p_n_unfm_pointer < (__le32 *)B_I_PITEM(p_s_bh, &s_ih) || + p_n_unfm_pointer > (__le32 *)B_I_PITEM(p_s_bh, &s_ih) + I_UNFM_NUM(&s_ih) - 1, "vs-5265: pointer out of range"); /* Hole, nothing to remove. */ @@ -1432,7 +1438,7 @@ int reiserfs_delete_object (struct reiserfs_transaction_handle *th, struct inode #if defined( USE_INODE_GENERATION_COUNTER ) if( !old_format_only ( th -> t_super ) ) { - __u32 *inode_generation; + __le32 *inode_generation; inode_generation = &REISERFS_SB(th -> t_super) -> s_rs -> s_inode_generation; diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index d6d1d7e..2283f18 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -216,7 +216,7 @@ static int finish_unfinished (struct super_block * s) /* reiserfs_iget needs k_dirid and k_objectid only */ item = B_I_PITEM (bh, ih); - obj_key.on_disk_key.k_dir_id = le32_to_cpu (*(__u32 *)item); + obj_key.on_disk_key.k_dir_id = le32_to_cpu (*(__le32 *)item); obj_key.on_disk_key.k_objectid = le32_to_cpu (ih->ih_key.k_objectid); obj_key.on_disk_key.u.k_offset_v1.k_offset = 0; obj_key.on_disk_key.u.k_offset_v1.k_uniqueness = 0; @@ -304,7 +304,7 @@ void add_save_link (struct reiserfs_transaction_handle * th, int retval; struct cpu_key key; struct item_head ih; - __u32 link; + __le32 link; BUG_ON (!th->t_trans_id); @@ -1336,7 +1336,7 @@ static int read_super_block (struct super_block * s, int offset) return 1; } - if ( rs->s_v1.s_root_block == -1 ) { + if ( rs->s_v1.s_root_block == cpu_to_le32(-1) ) { brelse(bh) ; reiserfs_warning (s, "Unfinished reiserfsck --rebuild-tree run detected. Please run\n" "reiserfsck --rebuild-tree and wait for a completion. If that fails\n" diff --git a/include/linux/reiserfs_acl.h b/include/linux/reiserfs_acl.h index a57e973..2aef9c3 100644 --- a/include/linux/reiserfs_acl.h +++ b/include/linux/reiserfs_acl.h @@ -5,18 +5,18 @@ #define REISERFS_ACL_VERSION 0x0001 typedef struct { - __u16 e_tag; - __u16 e_perm; - __u32 e_id; + __le16 e_tag; + __le16 e_perm; + __le32 e_id; } reiserfs_acl_entry; typedef struct { - __u16 e_tag; - __u16 e_perm; + __le16 e_tag; + __le16 e_perm; } reiserfs_acl_entry_short; typedef struct { - __u32 a_version; + __le32 a_version; } reiserfs_acl_header; static inline size_t reiserfs_acl_size(int count) diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h index d086787..cc39c53 100644 --- a/include/linux/reiserfs_fs.h +++ b/include/linux/reiserfs_fs.h @@ -114,47 +114,47 @@ if( !( cond ) ) \ struct journal_params { - __u32 jp_journal_1st_block; /* where does journal start from on its + __le32 jp_journal_1st_block; /* where does journal start from on its * device */ - __u32 jp_journal_dev; /* journal device st_rdev */ - __u32 jp_journal_size; /* size of the journal */ - __u32 jp_journal_trans_max; /* max number of blocks in a transaction. */ - __u32 jp_journal_magic; /* random value made on fs creation (this + __le32 jp_journal_dev; /* journal device st_rdev */ + __le32 jp_journal_size; /* size of the journal */ + __le32 jp_journal_trans_max; /* max number of blocks in a transaction. */ + __le32 jp_journal_magic; /* random value made on fs creation (this * was sb_journal_block_count) */ - __u32 jp_journal_max_batch; /* max number of blocks to batch into a + __le32 jp_journal_max_batch; /* max number of blocks to batch into a * trans */ - __u32 jp_journal_max_commit_age; /* in seconds, how old can an async + __le32 jp_journal_max_commit_age; /* in seconds, how old can an async * commit be */ - __u32 jp_journal_max_trans_age; /* in seconds, how old can a transaction + __le32 jp_journal_max_trans_age; /* in seconds, how old can a transaction * be */ }; /* this is the super from 3.5.X, where X >= 10 */ struct reiserfs_super_block_v1 { - __u32 s_block_count; /* blocks count */ - __u32 s_free_blocks; /* free blocks count */ - __u32 s_root_block; /* root block number */ + __le32 s_block_count; /* blocks count */ + __le32 s_free_blocks; /* free blocks count */ + __le32 s_root_block; /* root block number */ struct journal_params s_journal; - __u16 s_blocksize; /* block size */ - __u16 s_oid_maxsize; /* max size of object id array, see + __le16 s_blocksize; /* block size */ + __le16 s_oid_maxsize; /* max size of object id array, see * get_objectid() commentary */ - __u16 s_oid_cursize; /* current size of object id array */ - __u16 s_umount_state; /* this is set to 1 when filesystem was + __le16 s_oid_cursize; /* current size of object id array */ + __le16 s_umount_state; /* this is set to 1 when filesystem was * umounted, to 2 - when not */ char s_magic[10]; /* reiserfs magic string indicates that * file system is reiserfs: * "ReIsErFs" or "ReIsEr2Fs" or "ReIsEr3Fs" */ - __u16 s_fs_state; /* it is set to used by fsck to mark which + __le16 s_fs_state; /* it is set to used by fsck to mark which * phase of rebuilding is done */ - __u32 s_hash_function_code; /* indicate, what hash function is being use + __le32 s_hash_function_code; /* indicate, what hash function is being use * to sort names in a directory*/ - __u16 s_tree_height; /* height of disk tree */ - __u16 s_bmap_nr; /* amount of bitmap blocks needed to address + __le16 s_tree_height; /* height of disk tree */ + __le16 s_bmap_nr; /* amount of bitmap blocks needed to address * each block of file system */ - __u16 s_version; /* this field is only reliable on filesystem + __le16 s_version; /* this field is only reliable on filesystem * with non-standard journal */ - __u16 s_reserved_for_journal; /* size in blocks of journal area on main + __le16 s_reserved_for_journal; /* size in blocks of journal area on main * device, we need to keep after * making fs with non-standard journal */ } __attribute__ ((__packed__)); @@ -165,8 +165,8 @@ struct reiserfs_super_block_v1 struct reiserfs_super_block { struct reiserfs_super_block_v1 s_v1; - __u32 s_inode_generation; - __u32 s_flags; /* Right now used only by inode-attributes, if enabled */ + __le32 s_inode_generation; + __le32 s_flags; /* Right now used only by inode-attributes, if enabled */ unsigned char s_uuid[16]; /* filesystem unique identifier */ unsigned char s_label[16]; /* filesystem volume label */ char s_unused[88] ; /* zero filled by mkreiserfs and @@ -269,7 +269,7 @@ int is_reiserfs_jr (struct reiserfs_super_block * rs); #define QUOTA_EXCEEDED -6 typedef __u32 b_blocknr_t; -typedef __u32 unp_t; +typedef __le32 unp_t; struct unfm_nodeinfo { unp_t unfm_nodenum; @@ -376,8 +376,8 @@ static inline struct reiserfs_sb_info *REISERFS_SB(const struct super_block *sb) // directories use this key as well as old files // struct offset_v1 { - __u32 k_offset; - __u32 k_uniqueness; + __le32 k_offset; + __le32 k_uniqueness; } __attribute__ ((__packed__)); struct offset_v2 { @@ -453,9 +453,9 @@ struct in_core_offset_v2 { /* Key of an item determines its location in the S+tree, and is composed of 4 components */ struct reiserfs_key { - __u32 k_dir_id; /* packing locality: by default parent + __le32 k_dir_id; /* packing locality: by default parent directory object id */ - __u32 k_objectid; /* object identifier */ + __le32 k_objectid; /* object identifier */ union { struct offset_v1 k_offset_v1; struct offset_v2 k_offset_v2; @@ -534,15 +534,15 @@ struct item_head item. Note that the key, not this field, is used to determine the item type, and thus which field this union contains. */ - __u16 ih_free_space_reserved; + __le16 ih_free_space_reserved; /* Iff this is a directory item, this field equals the number of directory entries in the directory item. */ - __u16 ih_entry_count; + __le16 ih_entry_count; } __attribute__ ((__packed__)) u; - __u16 ih_item_len; /* total size of the item body */ - __u16 ih_item_location; /* an offset to the item body + __le16 ih_item_len; /* total size of the item body */ + __le16 ih_item_location; /* an offset to the item body * within the block */ - __u16 ih_version; /* 0 for all old items, 2 for new + __le16 ih_version; /* 0 for all old items, 2 for new ones. Highest bit is set by fsck temporary, cleaned after all done */ @@ -778,10 +778,10 @@ extern struct reiserfs_key root_key; /* Header of a disk block. More precisely, header of a formatted leaf or internal node, and not the header of an unformatted node. */ struct block_head { - __u16 blk_level; /* Level of a block in the tree. */ - __u16 blk_nr_item; /* Number of keys/items in a block. */ - __u16 blk_free_space; /* Block free space in bytes. */ - __u16 blk_reserved; + __le16 blk_level; /* Level of a block in the tree. */ + __le16 blk_nr_item; /* Number of keys/items in a block. */ + __le16 blk_free_space; /* Block free space in bytes. */ + __le16 blk_reserved; /* dump this in v4/planA */ struct reiserfs_key blk_right_delim_key; /* kept only for compatibility */ }; @@ -845,19 +845,19 @@ struct block_head { // struct stat_data_v1 { - __u16 sd_mode; /* file type, permissions */ - __u16 sd_nlink; /* number of hard links */ - __u16 sd_uid; /* owner */ - __u16 sd_gid; /* group */ - __u32 sd_size; /* file size */ - __u32 sd_atime; /* time of last access */ - __u32 sd_mtime; /* time file was last modified */ - __u32 sd_ctime; /* time inode (stat data) was last changed (except changes to sd_atime and sd_mtime) */ + __le16 sd_mode; /* file type, permissions */ + __le16 sd_nlink; /* number of hard links */ + __le16 sd_uid; /* owner */ + __le16 sd_gid; /* group */ + __le32 sd_size; /* file size */ + __le32 sd_atime; /* time of last access */ + __le32 sd_mtime; /* time file was last modified */ + __le32 sd_ctime; /* time inode (stat data) was last changed (except changes to sd_atime and sd_mtime) */ union { - __u32 sd_rdev; - __u32 sd_blocks; /* number of blocks file uses */ + __le32 sd_rdev; + __le32 sd_blocks; /* number of blocks file uses */ } __attribute__ ((__packed__)) u; - __u32 sd_first_direct_byte; /* first byte of file which is stored + __le32 sd_first_direct_byte; /* first byte of file which is stored in a direct item: except that if it equals 1 it is a symlink and if it equals ~(__u32)0 there is no @@ -923,20 +923,20 @@ struct stat_data_v1 /* Stat Data on disk (reiserfs version of UFS disk inode minus the address blocks) */ struct stat_data { - __u16 sd_mode; /* file type, permissions */ - __u16 sd_attrs; /* persistent inode flags */ - __u32 sd_nlink; /* number of hard links */ - __u64 sd_size; /* file size */ - __u32 sd_uid; /* owner */ - __u32 sd_gid; /* group */ - __u32 sd_atime; /* time of last access */ - __u32 sd_mtime; /* time file was last modified */ - __u32 sd_ctime; /* time inode (stat data) was last changed (except changes to sd_atime and sd_mtime) */ - __u32 sd_blocks; + __le16 sd_mode; /* file type, permissions */ + __le16 sd_attrs; /* persistent inode flags */ + __le32 sd_nlink; /* number of hard links */ + __le64 sd_size; /* file size */ + __le32 sd_uid; /* owner */ + __le32 sd_gid; /* group */ + __le32 sd_atime; /* time of last access */ + __le32 sd_mtime; /* time file was last modified */ + __le32 sd_ctime; /* time inode (stat data) was last changed (except changes to sd_atime and sd_mtime) */ + __le32 sd_blocks; union { - __u32 sd_rdev; - __u32 sd_generation; - //__u32 sd_first_direct_byte; + __le32 sd_rdev; + __le32 sd_generation; + //__le32 sd_first_direct_byte; /* first byte of file which is stored in a direct item: except that if it equals 1 it is a symlink and if it equals @@ -1019,12 +1019,12 @@ struct stat_data { struct reiserfs_de_head { - __u32 deh_offset; /* third component of the directory entry key */ - __u32 deh_dir_id; /* objectid of the parent directory of the object, that is referenced + __le32 deh_offset; /* third component of the directory entry key */ + __le32 deh_dir_id; /* objectid of the parent directory of the object, that is referenced by directory entry */ - __u32 deh_objectid; /* objectid of the object, that is referenced by directory entry */ - __u16 deh_location; /* offset of name in the whole item */ - __u16 deh_state; /* whether 1) entry contains stat data (for future), and 2) whether + __le32 deh_objectid; /* objectid of the object, that is referenced by directory entry */ + __le16 deh_location; /* offset of name in the whole item */ + __le16 deh_state; /* whether 1) entry contains stat data (for future), and 2) whether entry is hidden (unlinked) */ } __attribute__ ((__packed__)); #define DEH_SIZE sizeof(struct reiserfs_de_head) @@ -1084,10 +1084,10 @@ struct reiserfs_de_head #define de_visible(deh) test_bit_unaligned (DEH_Visible, &((deh)->deh_state)) #define de_hidden(deh) !test_bit_unaligned (DEH_Visible, &((deh)->deh_state)) -extern void make_empty_dir_item_v1 (char * body, __u32 dirid, __u32 objid, - __u32 par_dirid, __u32 par_objid); -extern void make_empty_dir_item (char * body, __u32 dirid, __u32 objid, - __u32 par_dirid, __u32 par_objid); +extern void make_empty_dir_item_v1 (char * body, __le32 dirid, __le32 objid, + __le32 par_dirid, __le32 par_objid); +extern void make_empty_dir_item (char * body, __le32 dirid, __le32 objid, + __le32 par_dirid, __le32 par_objid); /* array of the entry headers */ /* get item body */ @@ -1186,9 +1186,9 @@ struct reiserfs_dir_entry /* Disk child pointer: The pointer from an internal node of the tree to a node that is on disk. */ struct disk_child { - __u32 dc_block_number; /* Disk child's block number. */ - __u16 dc_size; /* Disk child's used space. */ - __u16 dc_reserved; + __le32 dc_block_number; /* Disk child's block number. */ + __le16 dc_size; /* Disk child's used space. */ + __le16 dc_reserved; }; #define DC_SIZE (sizeof(struct disk_child)) @@ -1656,10 +1656,10 @@ struct reiserfs_iget_args { /* first block written in a commit. */ struct reiserfs_journal_desc { - __u32 j_trans_id ; /* id of commit */ - __u32 j_len ; /* length of commit. len +1 is the commit block */ - __u32 j_mount_id ; /* mount id of this trans*/ - __u32 j_realblock[1] ; /* real locations for each block */ + __le32 j_trans_id ; /* id of commit */ + __le32 j_len ; /* length of commit. len +1 is the commit block */ + __le32 j_mount_id ; /* mount id of this trans*/ + __le32 j_realblock[1] ; /* real locations for each block */ } ; #define get_desc_trans_id(d) le32_to_cpu((d)->j_trans_id) @@ -1672,9 +1672,9 @@ struct reiserfs_journal_desc { /* last block written in a commit */ struct reiserfs_journal_commit { - __u32 j_trans_id ; /* must match j_trans_id from the desc block */ - __u32 j_len ; /* ditto */ - __u32 j_realblock[1] ; /* real locations for each block */ + __le32 j_trans_id ; /* must match j_trans_id from the desc block */ + __le32 j_len ; /* ditto */ + __le32 j_realblock[1] ; /* real locations for each block */ } ; #define get_commit_trans_id(c) le32_to_cpu((c)->j_trans_id) @@ -1689,9 +1689,9 @@ struct reiserfs_journal_commit { ** and this transaction does not need to be replayed. */ struct reiserfs_journal_header { - __u32 j_last_flush_trans_id ; /* id of last fully flushed transaction */ - __u32 j_first_unflushed_offset ; /* offset in the log of where to start replay after a crash */ - __u32 j_mount_id ; + __le32 j_last_flush_trans_id ; /* id of last fully flushed transaction */ + __le32 j_first_unflushed_offset ; /* offset in the log of where to start replay after a crash */ + __le32 j_mount_id ; /* 12 */ struct journal_params jh_journal; } ; @@ -2170,7 +2170,7 @@ void reiserfs_init_alloc_options (struct super_block *s); * to use for a new object underneat it. The locality is returned * in disk byte order (le). */ -u32 reiserfs_choose_packing(struct inode *dir); +__le32 reiserfs_choose_packing(struct inode *dir); int is_reusable (struct super_block * s, b_blocknr_t block, int bit_value); void reiserfs_free_block (struct reiserfs_transaction_handle *th, struct inode *, b_blocknr_t, int for_unformatted); diff --git a/include/linux/reiserfs_xattr.h b/include/linux/reiserfs_xattr.h index 1eaa48e..9244c57 100644 --- a/include/linux/reiserfs_xattr.h +++ b/include/linux/reiserfs_xattr.h @@ -10,8 +10,8 @@ #define REISERFS_XATTR_MAGIC 0x52465841 /* "RFXA" */ struct reiserfs_xattr_header { - __u32 h_magic; /* magic number for identification */ - __u32 h_hash; /* hash of the value */ + __le32 h_magic; /* magic number for identification */ + __le32 h_hash; /* hash of the value */ }; #ifdef __KERNEL__ -- cgit v0.10.2 From b8cc936f6295bba23513a49d858ea82f64982faf Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 1 May 2005 08:59:18 -0700 Subject: [PATCH] reiserfs endianness: fix endianness bugs fixes for a couple of bugs exposed by the above: le32_to_cpu() used on 16bit value and missing conversion in comparison of host- and little-endian values. Signed-off-by: Al Viro Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index b16d65a..3072cfd 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -2391,7 +2391,7 @@ int journal_init(struct super_block *p_s_sb, const char * j_dev_name, int old_fo jh = (struct reiserfs_journal_header *)(bhjh->b_data); /* make sure that journal matches to the super block */ - if (is_reiserfs_jr(rs) && (jh->jh_journal.jp_journal_magic != sb_jp_journal_magic(rs))) { + if (is_reiserfs_jr(rs) && (le32_to_cpu(jh->jh_journal.jp_journal_magic) != sb_jp_journal_magic(rs))) { reiserfs_warning (p_s_sb, "sh-460: journal header magic %x " "(device %s) does not match to magic found in super " "block %x", diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h index cc39c53..2f7a34d 100644 --- a/include/linux/reiserfs_fs.h +++ b/include/linux/reiserfs_fs.h @@ -225,7 +225,7 @@ struct reiserfs_super_block #define SB_ONDISK_JOURNAL_DEVICE(s) \ le32_to_cpu ((SB_ONDISK_JP(s)->jp_journal_dev)) #define SB_ONDISK_RESERVED_FOR_JOURNAL(s) \ - le32_to_cpu ((SB_V1_DISK_SUPER_BLOCK(s)->s_reserved_for_journal)) + le16_to_cpu ((SB_V1_DISK_SUPER_BLOCK(s)->s_reserved_for_journal)) #define is_block_in_log_or_reserved_area(s, block) \ block >= SB_JOURNAL_1st_RESERVED_BLOCK(s) \ -- cgit v0.10.2 From 6b9f5829e6e3af44f20c681e26524c637d4f82ff Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 1 May 2005 08:59:19 -0700 Subject: [PATCH] reiserfs endianness: comp_short_keys() cleanup comp_short_keys() massaged into sane form, which kills the last place where pointer to in_core_key (or any object containing such) would be cast to or from something else. At that point we are free to change layout of in_core_key - nothing depends on it anymore. So we drop the mess with union in there and simply use (unconditional) __u64 k_offset and __u8 k_type instead; places using in_core_key switched to those. That gives _far_ better code than current mess - on all platforms. Signed-off-by: Al Viro Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 5fdb9f9..2711dff 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -1341,8 +1341,8 @@ void reiserfs_read_locked_inode (struct inode * inode, struct reiserfs_iget_args key.version = KEY_FORMAT_3_5; key.on_disk_key.k_dir_id = dirino; key.on_disk_key.k_objectid = inode->i_ino; - key.on_disk_key.u.k_offset_v1.k_offset = SD_OFFSET; - key.on_disk_key.u.k_offset_v1.k_uniqueness = SD_UNIQUENESS; + key.on_disk_key.k_offset = 0; + key.on_disk_key.k_type = 0; /* look for the object's stat data */ retval = search_item (inode->i_sb, &key, &path_to_sd); diff --git a/fs/reiserfs/stree.c b/fs/reiserfs/stree.c index 15fa4cb..da23ba7 100644 --- a/fs/reiserfs/stree.c +++ b/fs/reiserfs/stree.c @@ -87,23 +87,20 @@ inline void copy_item_head(struct item_head * p_v_to, inline int comp_short_keys (const struct reiserfs_key * le_key, const struct cpu_key * cpu_key) { - __le32 * p_s_le_u32; - __u32 * p_s_cpu_u32; - int n_key_length = REISERFS_SHORT_KEY_LEN; - - p_s_le_u32 = (__le32 *)le_key; - p_s_cpu_u32 = (__u32 *)&cpu_key->on_disk_key; - for( ; n_key_length--; ++p_s_le_u32, ++p_s_cpu_u32 ) { - if ( le32_to_cpu (*p_s_le_u32) < *p_s_cpu_u32 ) + __u32 n; + n = le32_to_cpu(le_key->k_dir_id); + if (n < cpu_key->on_disk_key.k_dir_id) return -1; - if ( le32_to_cpu (*p_s_le_u32) > *p_s_cpu_u32 ) + if (n > cpu_key->on_disk_key.k_dir_id) + return 1; + n = le32_to_cpu(le_key->k_objectid); + if (n < cpu_key->on_disk_key.k_objectid) + return -1; + if (n > cpu_key->on_disk_key.k_objectid) return 1; - } - return 0; } - /* k1 is pointer to on-disk structure which is stored in little-endian form. k2 is pointer to cpu variable. Compare keys using all 4 key fields. @@ -153,18 +150,15 @@ inline int comp_short_le_keys (const struct reiserfs_key * key1, const struct re inline void le_key2cpu_key (struct cpu_key * to, const struct reiserfs_key * from) { + int version; to->on_disk_key.k_dir_id = le32_to_cpu (from->k_dir_id); to->on_disk_key.k_objectid = le32_to_cpu (from->k_objectid); // find out version of the key - to->version = le_key_version (from); - if (to->version == KEY_FORMAT_3_5) { - to->on_disk_key.u.k_offset_v1.k_offset = le32_to_cpu (from->u.k_offset_v1.k_offset); - to->on_disk_key.u.k_offset_v1.k_uniqueness = le32_to_cpu (from->u.k_offset_v1.k_uniqueness); - } else { - to->on_disk_key.u.k_offset_v2.k_offset = offset_v2_k_offset(&from->u.k_offset_v2); - to->on_disk_key.u.k_offset_v2.k_type = offset_v2_k_type(&from->u.k_offset_v2); - } + version = le_key_version (from); + to->version = version; + to->on_disk_key.k_offset = le_key_k_offset(version, from); + to->on_disk_key.k_type = le_key_k_type(version, from); } @@ -235,8 +229,8 @@ const struct reiserfs_key MAX_KEY = { {{__constant_cpu_to_le32(0xffffffff), __constant_cpu_to_le32(0xffffffff)},} }; -const struct in_core_key MAX_IN_CORE_KEY = {0xffffffff, 0xffffffff, {{0xffffffff, 0xffffffff},}}; +const struct in_core_key MAX_IN_CORE_KEY = {~0U, ~0U, ~0ULL>>4, 15}; /* Get delimiting key of the buffer by looking for it in the buffers in the path, starting from the bottom of the path, and going upwards. We must check the path's validity at each step. If the key is not in diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 2283f18..31e7512 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -218,8 +218,8 @@ static int finish_unfinished (struct super_block * s) item = B_I_PITEM (bh, ih); obj_key.on_disk_key.k_dir_id = le32_to_cpu (*(__le32 *)item); obj_key.on_disk_key.k_objectid = le32_to_cpu (ih->ih_key.k_objectid); - obj_key.on_disk_key.u.k_offset_v1.k_offset = 0; - obj_key.on_disk_key.u.k_offset_v1.k_uniqueness = 0; + obj_key.on_disk_key.k_offset = 0; + obj_key.on_disk_key.k_type = 0; pathrelse (&path); diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h index 2f7a34d..d445b68 100644 --- a/include/linux/reiserfs_fs.h +++ b/include/linux/reiserfs_fs.h @@ -433,23 +433,6 @@ static inline void set_offset_v2_k_offset( struct offset_v2 *v2, loff_t offset ) # define set_offset_v2_k_offset(v2,val) (offset_v2_k_offset(v2) = (val)) #endif -struct in_core_offset_v1 { - __u32 k_offset; - __u32 k_uniqueness; -} __attribute__ ((__packed__)); - -struct in_core_offset_v2 { -#ifdef __LITTLE_ENDIAN - /* little endian version */ - __u64 k_offset:60; - __u64 k_type: 4; -#else - /* big endian version */ - __u64 k_type: 4; - __u64 k_offset:60; -#endif -} __attribute__ ((__packed__)); - /* Key of an item determines its location in the S+tree, and is composed of 4 components */ struct reiserfs_key { @@ -466,11 +449,9 @@ struct in_core_key { __u32 k_dir_id; /* packing locality: by default parent directory object id */ __u32 k_objectid; /* object identifier */ - union { - struct in_core_offset_v1 k_offset_v1; - struct in_core_offset_v2 k_offset_v2; - } __attribute__ ((__packed__)) u; -} __attribute__ ((__packed__)); + __u64 k_offset; + __u8 k_type; +}; struct cpu_key { struct in_core_key on_disk_key; @@ -696,43 +677,29 @@ static inline void set_le_ih_k_type (struct item_head * ih, int type) // static inline loff_t cpu_key_k_offset (const struct cpu_key * key) { - return (key->version == KEY_FORMAT_3_5) ? - key->on_disk_key.u.k_offset_v1.k_offset : - key->on_disk_key.u.k_offset_v2.k_offset; + return key->on_disk_key.k_offset; } static inline loff_t cpu_key_k_type (const struct cpu_key * key) { - return (key->version == KEY_FORMAT_3_5) ? - uniqueness2type (key->on_disk_key.u.k_offset_v1.k_uniqueness) : - key->on_disk_key.u.k_offset_v2.k_type; + return key->on_disk_key.k_type; } static inline void set_cpu_key_k_offset (struct cpu_key * key, loff_t offset) { - (key->version == KEY_FORMAT_3_5) ? - (key->on_disk_key.u.k_offset_v1.k_offset = offset) : - (key->on_disk_key.u.k_offset_v2.k_offset = offset); + key->on_disk_key.k_offset = offset; } - static inline void set_cpu_key_k_type (struct cpu_key * key, int type) { - (key->version == KEY_FORMAT_3_5) ? - (key->on_disk_key.u.k_offset_v1.k_uniqueness = type2uniqueness (type)): - (key->on_disk_key.u.k_offset_v2.k_type = type); + key->on_disk_key.k_type = type; } - static inline void cpu_key_k_offset_dec (struct cpu_key * key) { - if (key->version == KEY_FORMAT_3_5) - key->on_disk_key.u.k_offset_v1.k_offset --; - else - key->on_disk_key.u.k_offset_v2.k_offset --; + key->on_disk_key.k_offset --; } - #define is_direntry_cpu_key(key) (cpu_key_k_type (key) == TYPE_DIRENTRY) #define is_direct_cpu_key(key) (cpu_key_k_type (key) == TYPE_DIRECT) #define is_indirect_cpu_key(key) (cpu_key_k_type (key) == TYPE_INDIRECT) -- cgit v0.10.2 From f8e08a8466c4ac5f61b4bdb6338fd97eedb9c9e8 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 1 May 2005 08:59:19 -0700 Subject: [PATCH] reiserfs endianness: sanitize reiserfs_key union Since we only access reiserfs_key ->u.k_offset_v2 guts in four helper functions, we are free to sanitize those, as long as - layout of the structure is unchanged (it's on-disk object) - behaviour of these helpers is same as before. Patch kills the mess with endianness-dependent bitfields and replaces them with a single __le64. Helpers are switched to straightforward shift/and/or. Benefits: - exact same definitions for little- and big-endian architectures; no ifdefs in sight. - generate the same code on little-endian and improved on big-endian. - doesn't rely on lousy bitfields handling in gcc codegenerator. - happens to be standard C (unsigned long long is not a valid type for a bitfield; it's a gccism and not well-implemented one, at that). Signed-off-by: Al Viro Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h index d445b68..3214862 100644 --- a/include/linux/reiserfs_fs.h +++ b/include/linux/reiserfs_fs.h @@ -381,57 +381,29 @@ struct offset_v1 { } __attribute__ ((__packed__)); struct offset_v2 { -#ifdef __LITTLE_ENDIAN - /* little endian version */ - __u64 k_offset:60; - __u64 k_type: 4; -#else - /* big endian version */ - __u64 k_type: 4; - __u64 k_offset:60; -#endif + __le64 v; } __attribute__ ((__packed__)); -#ifndef __LITTLE_ENDIAN -typedef union { - struct offset_v2 offset_v2; - __u64 linear; -} __attribute__ ((__packed__)) offset_v2_esafe_overlay; - static inline __u16 offset_v2_k_type( const struct offset_v2 *v2 ) { - offset_v2_esafe_overlay tmp = *(const offset_v2_esafe_overlay *)v2; - tmp.linear = le64_to_cpu( tmp.linear ); - return (tmp.offset_v2.k_type <= TYPE_MAXTYPE)?tmp.offset_v2.k_type:TYPE_ANY; + __u8 type = le64_to_cpu(v2->v) >> 60; + return (type <= TYPE_MAXTYPE)?type:TYPE_ANY; } static inline void set_offset_v2_k_type( struct offset_v2 *v2, int type ) { - offset_v2_esafe_overlay *tmp = (offset_v2_esafe_overlay *)v2; - tmp->linear = le64_to_cpu(tmp->linear); - tmp->offset_v2.k_type = type; - tmp->linear = cpu_to_le64(tmp->linear); + v2->v = (v2->v & cpu_to_le64(~0ULL>>4)) | cpu_to_le64((__u64)type<<60); } static inline loff_t offset_v2_k_offset( const struct offset_v2 *v2 ) { - offset_v2_esafe_overlay tmp = *(const offset_v2_esafe_overlay *)v2; - tmp.linear = le64_to_cpu( tmp.linear ); - return tmp.offset_v2.k_offset; + return le64_to_cpu(v2->v) & (~0ULL>>4); } static inline void set_offset_v2_k_offset( struct offset_v2 *v2, loff_t offset ){ - offset_v2_esafe_overlay *tmp = (offset_v2_esafe_overlay *)v2; - tmp->linear = le64_to_cpu(tmp->linear); - tmp->offset_v2.k_offset = offset; - tmp->linear = cpu_to_le64(tmp->linear); + offset &= (~0ULL>>4); + v2->v = (v2->v & cpu_to_le64(15ULL<<60)) | cpu_to_le64(offset); } -#else -# define offset_v2_k_type(v2) ((v2)->k_type) -# define set_offset_v2_k_type(v2,val) (offset_v2_k_type(v2) = (val)) -# define offset_v2_k_offset(v2) ((v2)->k_offset) -# define set_offset_v2_k_offset(v2,val) (offset_v2_k_offset(v2) = (val)) -#endif /* Key of an item determines its location in the S+tree, and is composed of 4 components */ -- cgit v0.10.2 From 1622c3fcf4c1c6c8267938f366ecd2a348d67d93 Mon Sep 17 00:00:00 2001 From: Gerd Knorr Date: Sun, 1 May 2005 08:59:19 -0700 Subject: [PATCH] cx88-dvb oops fix Fixup error path, without that one the driver kills the machine by oopsing in the IRQ handler in case the frontend initialization fails. Signed-off-by: Gerd Knorr Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/media/video/cx88/cx88-dvb.c b/drivers/media/video/cx88/cx88-dvb.c index bc6f18c..84d1743 100644 --- a/drivers/media/video/cx88/cx88-dvb.c +++ b/drivers/media/video/cx88/cx88-dvb.c @@ -243,10 +243,8 @@ static int dvb_register(struct cx8802_dev *dev) break; #endif default: - printk("%s: The frontend of your DVB/ATSC card isn't supported yet\n" - "%s: you might want to look out for patches here:\n" - "%s: http://dl.bytesex.org/patches/\n", - dev->core->name, dev->core->name, dev->core->name); + printk("%s: The frontend of your DVB/ATSC card isn't supported yet\n", + dev->core->name); break; } if (NULL == dev->dvb.frontend) { @@ -308,9 +306,11 @@ static int __devinit dvb_probe(struct pci_dev *pci_dev, dev); err = dvb_register(dev); if (0 != err) - goto fail_free; + goto fail_fini; return 0; + fail_fini: + cx8802_fini_common(dev); fail_free: kfree(dev); fail_core: -- cgit v0.10.2 From 9990d744bea7d28e83c420e2c9d524c7a8a2d136 Mon Sep 17 00:00:00 2001 From: Gerd Knorr Date: Sun, 1 May 2005 08:59:20 -0700 Subject: [PATCH] dvb: cx22702 frontend driver update update cx22702 fe driver, add support for using the dvb pll lib, enable cx22702 support in cx88-dvb. Signed-off-by: Gerd Knorr Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/media/dvb/frontends/cx22702.c b/drivers/media/dvb/frontends/cx22702.c index 1930b51..011860c 100644 --- a/drivers/media/dvb/frontends/cx22702.c +++ b/drivers/media/dvb/frontends/cx22702.c @@ -32,6 +32,7 @@ #include #include #include "dvb_frontend.h" +#include "dvb-pll.h" #include "cx22702.h" @@ -203,7 +204,19 @@ static int cx22702_set_tps (struct dvb_frontend* fe, struct dvb_frontend_paramet /* set PLL */ cx22702_writereg (state, 0x0D, cx22702_readreg(state,0x0D) &0xfe); - state->config->pll_set(fe, p); + if (state->config->pll_set) { + state->config->pll_set(fe, p); + } else if (state->config->pll_desc) { + u8 pllbuf[4]; + struct i2c_msg msg = { .addr = state->config->pll_address, + .buf = pllbuf, .len = 4 }; + dvb_pll_configure(state->config->pll_desc, pllbuf, + p->frequency, + p->u.ofdm.bandwidth); + i2c_transfer(state->i2c, &msg, 1); + } else { + BUG(); + } cx22702_writereg (state, 0x0D, cx22702_readreg(state,0x0D) | 1); /* set inversion */ diff --git a/drivers/media/dvb/frontends/cx22702.h b/drivers/media/dvb/frontends/cx22702.h index 6e34f99..559fdb9 100644 --- a/drivers/media/dvb/frontends/cx22702.h +++ b/drivers/media/dvb/frontends/cx22702.h @@ -36,6 +36,9 @@ struct cx22702_config u8 demod_address; /* PLL maintenance */ + u8 pll_address; + struct dvb_pll_desc *pll_desc; + int (*pll_init)(struct dvb_frontend* fe); int (*pll_set)(struct dvb_frontend* fe, struct dvb_frontend_parameters* params); }; diff --git a/drivers/media/video/Kconfig b/drivers/media/video/Kconfig index c1b3542..d3dd422 100644 --- a/drivers/media/video/Kconfig +++ b/drivers/media/video/Kconfig @@ -252,6 +252,7 @@ config VIDEO_SAA7134_DVB depends on VIDEO_SAA7134 && DVB_CORE select VIDEO_BUF_DVB select DVB_MT352 + select DVB_CX22702 ---help--- This adds support for DVB cards based on the Philips saa7134 chip. diff --git a/drivers/media/video/cx88/cx88-dvb.c b/drivers/media/video/cx88/cx88-dvb.c index 84d1743..1ca3731 100644 --- a/drivers/media/video/cx88/cx88-dvb.c +++ b/drivers/media/video/cx88/cx88-dvb.c @@ -31,7 +31,7 @@ #include /* those two frontends need merging via linuxtv cvs ... */ -#define HAVE_CX22702 0 +#define HAVE_CX22702 1 #define HAVE_OR51132 1 #include "cx88.h" -- cgit v0.10.2 From faf8b24968ce6392ea68d9afc7de1ffbc38c1f6c Mon Sep 17 00:00:00 2001 From: Gerd Knorr Date: Sun, 1 May 2005 08:59:20 -0700 Subject: [PATCH] v4l: msp3400 update msp3400 update: Fix and enable "simpler" mode, some other minor fixes. Signed-off-by: Gerd Knorr Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/media/video/msp3400.c b/drivers/media/video/msp3400.c index c97df70..7fbb858 100644 --- a/drivers/media/video/msp3400.c +++ b/drivers/media/video/msp3400.c @@ -380,7 +380,9 @@ static void msp3400c_setvolume(struct i2c_client *client, int val = 0, bal = 0; if (!muted) { - val = (volume * 0x7F / 65535) << 8; + /* 0x7f instead if 0x73 here has sound quality issues, + * probably due to overmodulation + clipping ... */ + val = (volume * 0x73 / 65535) << 8; } if (val) { bal = (balance / 256) - 128; @@ -997,7 +999,13 @@ static int msp34xx_modus(int norm) { switch (norm) { case VIDEO_MODE_PAL: +#if 1 + /* experimental: not sure this works with all chip versions */ + return 0x7003; +#else + /* previous value, try this if it breaks ... */ return 0x1003; +#endif case VIDEO_MODE_NTSC: /* BTSC */ return 0x2003; case VIDEO_MODE_SECAM: @@ -1264,6 +1272,7 @@ static int msp34xxg_thread(void *data) int val, std, i; printk("msp34xxg: daemon started\n"); + msp->source = 1; /* default */ for (;;) { d2printk(KERN_DEBUG "msp34xxg: thread: sleep\n"); msp34xx_sleep(msp,-1); @@ -1334,8 +1343,9 @@ static void msp34xxg_set_source(struct i2c_client *client, int source) /* fix matrix mode to stereo and let the msp choose what * to output according to 'source', as recommended + * for MONO (source==0) downmixing set bit[7:0] to 0x30 */ - int value = (source&0x07)<<8|(source==0 ? 0x00:0x20); + int value = (source&0x07)<<8|(source==0 ? 0x30:0x20); dprintk("msp34xxg: set source to %d (0x%x)\n", source, value); msp3400c_write(client, I2C_MSP3400C_DFP, @@ -1359,7 +1369,7 @@ static void msp34xxg_set_source(struct i2c_client *client, int source) msp3400c_write(client, I2C_MSP3400C_DEM, 0x22, /* a2 threshold for stereo/bilingual */ - source==0 ? 0x7f0:stereo_threshold); + stereo_threshold); msp->source=source; } @@ -1394,7 +1404,7 @@ static void msp34xxg_detect_stereo(struct i2c_client *client) static void msp34xxg_set_audmode(struct i2c_client *client, int audmode) { struct msp3400c *msp = i2c_get_clientdata(client); - int source = 0; + int source; switch (audmode) { case V4L2_TUNER_MODE_MONO: @@ -1410,9 +1420,10 @@ static void msp34xxg_set_audmode(struct i2c_client *client, int audmode) case V4L2_TUNER_MODE_LANG2: source=4; /* stereo or B */ break; - default: /* doing nothing: a safe, sane default */ + default: audmode = 0; - return; + source = 1; + break; } msp->audmode = audmode; msp34xxg_set_source(client, source); @@ -1514,12 +1525,9 @@ static int msp_attach(struct i2c_adapter *adap, int addr, int kind) msp->opmode = opmode; if (OPMODE_AUTO == msp->opmode) { -#if 0 /* seems to work for ivtv only, disable by default for now ... */ if (HAVE_SIMPLER(msp)) msp->opmode = OPMODE_SIMPLER; - else -#endif - if (HAVE_SIMPLE(msp)) + else if (HAVE_SIMPLE(msp)) msp->opmode = OPMODE_SIMPLE; else msp->opmode = OPMODE_MANUAL; diff --git a/drivers/media/video/tvaudio.c b/drivers/media/video/tvaudio.c index 065eb40..80dc34f 100644 --- a/drivers/media/video/tvaudio.c +++ b/drivers/media/video/tvaudio.c @@ -991,7 +991,7 @@ static int tda9874a_initialize(struct CHIPSTATE *chip) { if (tda9874a_SIF > 2) tda9874a_SIF = 1; - if (tda9874a_STD >= 8) + if (tda9874a_STD > 8) tda9874a_STD = 0; if(tda9874a_AMSEL > 1) tda9874a_AMSEL = 0; -- cgit v0.10.2 From fe55c452368af263a9beec38ed29f6be85280524 Mon Sep 17 00:00:00 2001 From: Mingming Cao Date: Sun, 1 May 2005 08:59:20 -0700 Subject: [PATCH] ext3: remove unnecessary race then retry in ext3_get_block The extra race-with-truncate-then-retry logic around ext3_get_block_handle(), which was inherited from ext2, becomes unecessary for ext3, since we have already obtained the ei->truncate_sem in ext3_get_block_handle() before calling ext3_alloc_branch(). The ei->truncate_sem is already there to block concurrent truncate and block allocation on the same inode. So the inode's indirect addressing tree won't be changed after we grab that semaphore. We could, after get the semaphore, re-verify the branch is up-to-date or not. If it has been changed, then get the updated branch. If we still need block allocation, we will have a safe version of the branch to work with in the ext3_find_goal()/ext3_splice_branch(). The code becomes more readable after remove those retry logic. The patch also clean up some gotos in ext3_get_block_handle() to make it more readable. Signed-off-by: Mingming Cao Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index 040eb28..ea58886 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -455,12 +455,11 @@ static unsigned long ext3_find_near(struct inode *inode, Indirect *ind) * @goal: place to store the result. * * Normally this function find the prefered place for block allocation, - * stores it in *@goal and returns zero. If the branch had been changed - * under us we return -EAGAIN. + * stores it in *@goal and returns zero. */ -static int ext3_find_goal(struct inode *inode, long block, Indirect chain[4], - Indirect *partial, unsigned long *goal) +static unsigned long ext3_find_goal(struct inode *inode, long block, + Indirect chain[4], Indirect *partial) { struct ext3_block_alloc_info *block_i = EXT3_I(inode)->i_block_alloc_info; @@ -470,15 +469,10 @@ static int ext3_find_goal(struct inode *inode, long block, Indirect chain[4], */ if (block_i && (block == block_i->last_alloc_logical_block + 1) && (block_i->last_alloc_physical_block != 0)) { - *goal = block_i->last_alloc_physical_block + 1; - return 0; + return block_i->last_alloc_physical_block + 1; } - if (verify_chain(chain, partial)) { - *goal = ext3_find_near(inode, partial); - return 0; - } - return -EAGAIN; + return ext3_find_near(inode, partial); } /** @@ -582,12 +576,9 @@ static int ext3_alloc_branch(handle_t *handle, struct inode *inode, * @where: location of missing link * @num: number of blocks we are adding * - * This function verifies that chain (up to the missing link) had not - * changed, fills the missing link and does all housekeeping needed in + * This function fills the missing link and does all housekeeping needed in * inode (->i_blocks, etc.). In case of success we end up with the full - * chain to new block and return 0. Otherwise (== chain had been changed) - * we free the new blocks (forgetting their buffer_heads, indeed) and - * return -EAGAIN. + * chain to new block and return 0. */ static int ext3_splice_branch(handle_t *handle, struct inode *inode, long block, @@ -608,12 +599,6 @@ static int ext3_splice_branch(handle_t *handle, struct inode *inode, long block, if (err) goto err_out; } - /* Verify that place we are splicing to is still there and vacant */ - - if (!verify_chain(chain, where-1) || *where->p) - /* Writer: end */ - goto changed; - /* That's it */ *where->p = where->key; @@ -657,26 +642,11 @@ static int ext3_splice_branch(handle_t *handle, struct inode *inode, long block, } return err; -changed: - /* - * AKPM: if where[i].bh isn't part of the current updating - * transaction then we explode nastily. Test this code path. - */ - jbd_debug(1, "the chain changed: try again\n"); - err = -EAGAIN; - err_out: for (i = 1; i < num; i++) { BUFFER_TRACE(where[i].bh, "call journal_forget"); ext3_journal_forget(handle, where[i].bh); } - /* For the normal collision cleanup case, we free up the blocks. - * On genuine filesystem errors we don't even think about doing - * that. */ - if (err == -EAGAIN) - for (i = 0; i < num; i++) - ext3_free_blocks(handle, inode, - le32_to_cpu(where[i].key), 1); return err; } @@ -708,7 +678,7 @@ ext3_get_block_handle(handle_t *handle, struct inode *inode, sector_t iblock, unsigned long goal; int left; int boundary = 0; - int depth = ext3_block_to_path(inode, iblock, offsets, &boundary); + const int depth = ext3_block_to_path(inode, iblock, offsets, &boundary); struct ext3_inode_info *ei = EXT3_I(inode); J_ASSERT(handle != NULL || create == 0); @@ -716,54 +686,55 @@ ext3_get_block_handle(handle_t *handle, struct inode *inode, sector_t iblock, if (depth == 0) goto out; -reread: partial = ext3_get_branch(inode, depth, offsets, chain, &err); /* Simplest case - block found, no allocation needed */ if (!partial) { clear_buffer_new(bh_result); -got_it: - map_bh(bh_result, inode->i_sb, le32_to_cpu(chain[depth-1].key)); - if (boundary) - set_buffer_boundary(bh_result); - /* Clean up and exit */ - partial = chain+depth-1; /* the whole chain */ - goto cleanup; + goto got_it; } /* Next simple case - plain lookup or failed read of indirect block */ - if (!create || err == -EIO) { -cleanup: + if (!create || err == -EIO) + goto cleanup; + + down(&ei->truncate_sem); + + /* + * If the indirect block is missing while we are reading + * the chain(ext3_get_branch() returns -EAGAIN err), or + * if the chain has been changed after we grab the semaphore, + * (either because another process truncated this branch, or + * another get_block allocated this branch) re-grab the chain to see if + * the request block has been allocated or not. + * + * Since we already block the truncate/other get_block + * at this point, we will have the current copy of the chain when we + * splice the branch into the tree. + */ + if (err == -EAGAIN || !verify_chain(chain, partial)) { while (partial > chain) { - BUFFER_TRACE(partial->bh, "call brelse"); brelse(partial->bh); partial--; } - BUFFER_TRACE(bh_result, "returned"); -out: - return err; + partial = ext3_get_branch(inode, depth, offsets, chain, &err); + if (!partial) { + up(&ei->truncate_sem); + if (err) + goto cleanup; + clear_buffer_new(bh_result); + goto got_it; + } } /* - * Indirect block might be removed by truncate while we were - * reading it. Handling of that case (forget what we've got and - * reread) is taken out of the main path. - */ - if (err == -EAGAIN) - goto changed; - - goal = 0; - down(&ei->truncate_sem); - - /* lazy initialize the block allocation info here if necessary */ - if (S_ISREG(inode->i_mode) && (!ei->i_block_alloc_info)) { + * Okay, we need to do block allocation. Lazily initialize the block + * allocation info here if necessary + */ + if (S_ISREG(inode->i_mode) && (!ei->i_block_alloc_info)) ext3_init_block_alloc_info(inode); - } - if (ext3_find_goal(inode, iblock, chain, partial, &goal) < 0) { - up(&ei->truncate_sem); - goto changed; - } + goal = ext3_find_goal(inode, iblock, chain, partial); left = (chain + depth) - partial; @@ -771,38 +742,45 @@ out: * Block out ext3_truncate while we alter the tree */ err = ext3_alloc_branch(handle, inode, left, goal, - offsets+(partial-chain), partial); + offsets + (partial - chain), partial); - /* The ext3_splice_branch call will free and forget any buffers + /* + * The ext3_splice_branch call will free and forget any buffers * on the new chain if there is a failure, but that risks using * up transaction credits, especially for bitmaps where the * credits cannot be returned. Can we handle this somehow? We - * may need to return -EAGAIN upwards in the worst case. --sct */ + * may need to return -EAGAIN upwards in the worst case. --sct + */ if (!err) err = ext3_splice_branch(handle, inode, iblock, chain, partial, left); - /* i_disksize growing is protected by truncate_sem - * don't forget to protect it if you're about to implement - * concurrent ext3_get_block() -bzzz */ + /* + * i_disksize growing is protected by truncate_sem. Don't forget to + * protect it if you're about to implement concurrent + * ext3_get_block() -bzzz + */ if (!err && extend_disksize && inode->i_size > ei->i_disksize) ei->i_disksize = inode->i_size; up(&ei->truncate_sem); - if (err == -EAGAIN) - goto changed; if (err) goto cleanup; set_buffer_new(bh_result); - goto got_it; - -changed: +got_it: + map_bh(bh_result, inode->i_sb, le32_to_cpu(chain[depth-1].key)); + if (boundary) + set_buffer_boundary(bh_result); + /* Clean up and exit */ + partial = chain + depth - 1; /* the whole chain */ +cleanup: while (partial > chain) { - jbd_debug(1, "buffer chain changed, retrying\n"); - BUFFER_TRACE(partial->bh, "brelsing"); + BUFFER_TRACE(partial->bh, "call brelse"); brelse(partial->bh); partial--; } - goto reread; + BUFFER_TRACE(bh_result, "returned"); +out: + return err; } static int ext3_get_block(struct inode *inode, sector_t iblock, -- cgit v0.10.2 From 10e92060295ded7b8feac1ee57bb2135ea9ac207 Mon Sep 17 00:00:00 2001 From: Peter Missel Date: Sun, 1 May 2005 08:59:21 -0700 Subject: [PATCH] saa7134: Add OEM version of already supported card Add device table support for the LR214WF card. The driver will say it's a FlyTV, simply because the name strings are stored with the card design data, not the device ID data. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/media/video/saa7134/saa7134-cards.c b/drivers/media/video/saa7134/saa7134-cards.c index 72a7b24..c51eb7f 100644 --- a/drivers/media/video/saa7134/saa7134-cards.c +++ b/drivers/media/video/saa7134/saa7134-cards.c @@ -1628,11 +1628,17 @@ struct pci_device_id saa7134_pci_tbl[] = { },{ .vendor = PCI_VENDOR_ID_PHILIPS, .device = PCI_DEVICE_ID_PHILIPS_SAA7133, - .subvendor = 0x5168, + .subvendor = 0x5168, /* Animation Technologies (LifeView) */ .subdevice = 0x0214, /* Standard PCI, LR214WF */ .driver_data = SAA7134_BOARD_FLYTVPLATINUM_FM, },{ .vendor = PCI_VENDOR_ID_PHILIPS, + .device = PCI_DEVICE_ID_PHILIPS_SAA7133, + .subvendor = 0x1489, /* KYE */ + .subdevice = 0x0214, /* Genius VideoWonder ProTV */ + .driver_data = SAA7134_BOARD_FLYTVPLATINUM_FM, /* is an LR214WF actually */ + },{ + .vendor = PCI_VENDOR_ID_PHILIPS, .device = PCI_DEVICE_ID_PHILIPS_SAA7134, .subvendor = 0x16be, .subdevice = 0x0003, -- cgit v0.10.2 From 149733d4e274ce1add3399cabdb243116ebfdf79 Mon Sep 17 00:00:00 2001 From: Patrick Gefre Date: Sun, 1 May 2005 08:59:21 -0700 Subject: [PATCH] Altix ioc4 serial - set hfc from ioctl Allow hardware flow control to be set from an ioctl. Signed-off-by: Patrick Gefre Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/serial/ioc4_serial.c b/drivers/serial/ioc4_serial.c index d054f12..f94d2e7 100644 --- a/drivers/serial/ioc4_serial.c +++ b/drivers/serial/ioc4_serial.c @@ -1765,8 +1765,11 @@ ioc4_change_speed(struct uart_port *the_port, the_port->ignore_status_mask &= ~N_DATA_READY; } - if (cflag & CRTSCTS) + if (cflag & CRTSCTS) { info->flags |= ASYNC_CTS_FLOW; + port->ip_sscr |= IOC4_SSCR_HFC_EN; + writel(port->ip_sscr, &port->ip_serial_regs->sscr); + } else info->flags &= ~ASYNC_CTS_FLOW; @@ -1825,12 +1828,6 @@ static inline int ic4_startup_local(struct uart_port *the_port) /* set the speed of the serial port */ ioc4_change_speed(the_port, info->tty->termios, (struct termios *)0); - /* enable hardware flow control - after ioc4_change_speed because - * ASYNC_CTS_FLOW is set there */ - if (info->flags & ASYNC_CTS_FLOW) { - port->ip_sscr |= IOC4_SSCR_HFC_EN; - writel(port->ip_sscr, &port->ip_serial_regs->sscr); - } info->flags |= UIF_INITIALIZED; return 0; } -- cgit v0.10.2 From 6cb2875f8d7e392982229094ad3b6d2f213f48da Mon Sep 17 00:00:00 2001 From: Patrick Gefre Date: Sun, 1 May 2005 08:59:21 -0700 Subject: [PATCH] Altix ioc4 serial - set a better timeout/threshold Set the timeout and threshold to better values. Signed-off-by: Patrick Gefre Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/serial/ioc4_serial.c b/drivers/serial/ioc4_serial.c index f94d2e7..d3f4542 100644 --- a/drivers/serial/ioc4_serial.c +++ b/drivers/serial/ioc4_serial.c @@ -1272,8 +1272,9 @@ static inline int set_rx_timeout(struct ioc4_port *port, int timeout) * and set the rx threshold to that amount. There are 4 chars * per ring entry, so we'll divide the number of chars that will * arrive in timeout by 4. + * So .... timeout * baud / 10 / HZ / 4, with HZ = 100. */ - threshold = timeout * port->ip_baud / 10 / HZ / 4; + threshold = timeout * port->ip_baud / 4000; if (threshold == 0) threshold = 1; /* otherwise we'll intr all the time! */ @@ -1285,8 +1286,10 @@ static inline int set_rx_timeout(struct ioc4_port *port, int timeout) writel(port->ip_sscr, &port->ip_serial_regs->sscr); - /* Now set the rx timeout to the given value */ - timeout = timeout * IOC4_SRTR_HZ / HZ; + /* Now set the rx timeout to the given value + * again timeout * IOC4_SRTR_HZ / HZ + */ + timeout = timeout * IOC4_SRTR_HZ / 100; if (timeout > IOC4_SRTR_CNT) timeout = IOC4_SRTR_CNT; @@ -1380,7 +1383,7 @@ config_port(struct ioc4_port *port, if (port->ip_tx_lowat == 0) port->ip_tx_lowat = 1; - set_rx_timeout(port, port->ip_rx_timeout); + set_rx_timeout(port, 2); return 0; } -- cgit v0.10.2 From 68985e486b708671a7c7d0b4de6d94a1547d1351 Mon Sep 17 00:00:00 2001 From: Patrick Gefre Date: Sun, 1 May 2005 08:59:21 -0700 Subject: [PATCH] Altix ioc4 serial - small uart setup mods Small mods for setting up the uart - parity, flow control Signed-off-by: Patrick Gefre Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/serial/ioc4_serial.c b/drivers/serial/ioc4_serial.c index d3f4542..d9629ee 100644 --- a/drivers/serial/ioc4_serial.c +++ b/drivers/serial/ioc4_serial.c @@ -1688,8 +1688,8 @@ ioc4_change_speed(struct uart_port *the_port, { struct ioc4_port *port = get_ioc4_port(the_port); int baud, bits; - unsigned cflag, cval; - int new_parity = 0, new_parity_enable = 0, new_stop = 1, new_data = 8; + unsigned cflag; + int new_parity = 0, new_parity_enable = 0, new_stop = 0, new_data = 8; struct uart_info *info = the_port->info; cflag = new_termios->c_cflag; @@ -1697,48 +1697,35 @@ ioc4_change_speed(struct uart_port *the_port, switch (cflag & CSIZE) { case CS5: new_data = 5; - cval = 0x00; bits = 7; break; case CS6: new_data = 6; - cval = 0x01; bits = 8; break; case CS7: new_data = 7; - cval = 0x02; bits = 9; break; case CS8: new_data = 8; - cval = 0x03; bits = 10; break; default: /* cuz we always need a default ... */ new_data = 5; - cval = 0x00; bits = 7; break; } if (cflag & CSTOPB) { - cval |= 0x04; bits++; new_stop = 1; } if (cflag & PARENB) { - cval |= UART_LCR_PARITY; bits++; new_parity_enable = 1; - } - if (cflag & PARODD) { - cval |= UART_LCR_EPAR; - new_parity = 1; - } - if (cflag & IGNPAR) { - cval &= ~UART_LCR_PARITY; - new_parity_enable = 0; + if (cflag & PARODD) + new_parity = 1; } baud = uart_get_baud_rate(the_port, new_termios, old_termios, MIN_BAUD_SUPPORTED, MAX_BAUD_SUPPORTED); @@ -1771,10 +1758,12 @@ ioc4_change_speed(struct uart_port *the_port, if (cflag & CRTSCTS) { info->flags |= ASYNC_CTS_FLOW; port->ip_sscr |= IOC4_SSCR_HFC_EN; - writel(port->ip_sscr, &port->ip_serial_regs->sscr); } - else + else { info->flags &= ~ASYNC_CTS_FLOW; + port->ip_sscr &= ~IOC4_SSCR_HFC_EN; + } + writel(port->ip_sscr, &port->ip_serial_regs->sscr); /* Set the configuration and proper notification call */ DPRINT_CONFIG(("%s : port 0x%p cflag 0%o " @@ -1847,7 +1836,6 @@ static void ioc4_cb_output_lowat(struct ioc4_port *port) } } - /** * handle_intr - service any interrupts for the given port - 2nd level * called via sd_intr -- cgit v0.10.2 From 5b052d8bb3ad9108489e7475868e14372774ca08 Mon Sep 17 00:00:00 2001 From: Patrick Gefre Date: Sun, 1 May 2005 08:59:22 -0700 Subject: [PATCH] Altix ioc4 serial - Arm the read timeout timer before the first read Arm the read timeout timer before the first read. Signed-off-by: Patrick Gefre Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/serial/ioc4_serial.c b/drivers/serial/ioc4_serial.c index d9629ee..ba4e13a 100644 --- a/drivers/serial/ioc4_serial.c +++ b/drivers/serial/ioc4_serial.c @@ -838,7 +838,7 @@ static int inline port_init(struct ioc4_port *port) port->ip_tx_prod = readl(&port->ip_serial_regs->stcir) & PROD_CONS_MASK; writel(port->ip_tx_prod, &port->ip_serial_regs->stpir); port->ip_rx_cons = readl(&port->ip_serial_regs->srpir) & PROD_CONS_MASK; - writel(port->ip_rx_cons, &port->ip_serial_regs->srcir); + writel(port->ip_rx_cons | IOC4_SRCIR_ARM, &port->ip_serial_regs->srcir); /* Disable interrupts for this 16550 */ uart = port->ip_uart_regs; -- cgit v0.10.2 From 7149437669f79b497830e643a2b13d26a017b038 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Sun, 1 May 2005 08:59:22 -0700 Subject: [PATCH] fbdev: Batch cmap changes at driver level This patch adds to the fbdev interface a set_cmap callback that allow the driver to "batch" palette changes. This is useful for drivers like radeonfb which might require lenghtly workarounds on palette accesses, thus allowing to factor out those workarounds efficiently. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/aty/radeon_base.c b/drivers/video/aty/radeon_base.c index e8eb124..ee25b9e 100644 --- a/drivers/video/aty/radeon_base.c +++ b/drivers/video/aty/radeon_base.c @@ -1057,13 +1057,14 @@ static int radeonfb_blank (int blank, struct fb_info *info) return radeon_screen_blank(rinfo, blank, 0); } -static int radeonfb_setcolreg (unsigned regno, unsigned red, unsigned green, - unsigned blue, unsigned transp, struct fb_info *info) +static int radeon_setcolreg (unsigned regno, unsigned red, unsigned green, + unsigned blue, unsigned transp, + struct radeonfb_info *rinfo) { - struct radeonfb_info *rinfo = info->par; u32 pindex; unsigned int i; - + + if (regno > 255) return 1; @@ -1078,20 +1079,7 @@ static int radeonfb_setcolreg (unsigned regno, unsigned red, unsigned green, pindex = regno; if (!rinfo->asleep) { - u32 dac_cntl2, vclk_cntl = 0; - radeon_fifo_wait(9); - if (rinfo->is_mobility) { - vclk_cntl = INPLL(VCLK_ECP_CNTL); - OUTPLL(VCLK_ECP_CNTL, vclk_cntl & ~PIXCLK_DAC_ALWAYS_ONb); - } - - /* Make sure we are on first palette */ - if (rinfo->has_CRTC2) { - dac_cntl2 = INREG(DAC_CNTL2); - dac_cntl2 &= ~DAC2_PALETTE_ACCESS_CNTL; - OUTREG(DAC_CNTL2, dac_cntl2); - } if (rinfo->bpp == 16) { pindex = regno * 8; @@ -1101,24 +1089,27 @@ static int radeonfb_setcolreg (unsigned regno, unsigned red, unsigned green, if (rinfo->depth == 15 && regno > 31) return 1; - /* For 565, the green component is mixed one order below */ + /* For 565, the green component is mixed one order + * below + */ if (rinfo->depth == 16) { OUTREG(PALETTE_INDEX, pindex>>1); - OUTREG(PALETTE_DATA, (rinfo->palette[regno>>1].red << 16) | - (green << 8) | (rinfo->palette[regno>>1].blue)); + OUTREG(PALETTE_DATA, + (rinfo->palette[regno>>1].red << 16) | + (green << 8) | + (rinfo->palette[regno>>1].blue)); green = rinfo->palette[regno<<1].green; } } if (rinfo->depth != 16 || regno < 32) { OUTREG(PALETTE_INDEX, pindex); - OUTREG(PALETTE_DATA, (red << 16) | (green << 8) | blue); + OUTREG(PALETTE_DATA, (red << 16) | + (green << 8) | blue); } - if (rinfo->is_mobility) - OUTPLL(VCLK_ECP_CNTL, vclk_cntl); } if (regno < 16) { - u32 *pal = info->pseudo_palette; + u32 *pal = rinfo->info->pseudo_palette; switch (rinfo->depth) { case 15: pal[regno] = (regno << 10) | (regno << 5) | regno; @@ -1138,6 +1129,84 @@ static int radeonfb_setcolreg (unsigned regno, unsigned red, unsigned green, return 0; } +static int radeonfb_setcolreg (unsigned regno, unsigned red, unsigned green, + unsigned blue, unsigned transp, + struct fb_info *info) +{ + struct radeonfb_info *rinfo = info->par; + u32 dac_cntl2, vclk_cntl = 0; + int rc; + + if (!rinfo->asleep) { + if (rinfo->is_mobility) { + vclk_cntl = INPLL(VCLK_ECP_CNTL); + OUTPLL(VCLK_ECP_CNTL, + vclk_cntl & ~PIXCLK_DAC_ALWAYS_ONb); + } + + /* Make sure we are on first palette */ + if (rinfo->has_CRTC2) { + dac_cntl2 = INREG(DAC_CNTL2); + dac_cntl2 &= ~DAC2_PALETTE_ACCESS_CNTL; + OUTREG(DAC_CNTL2, dac_cntl2); + } + } + + rc = radeon_setcolreg (regno, red, green, blue, transp, rinfo); + + if (!rinfo->asleep && rinfo->is_mobility) + OUTPLL(VCLK_ECP_CNTL, vclk_cntl); + + return rc; +} + +static int radeonfb_setcmap(struct fb_cmap *cmap, struct fb_info *info) +{ + struct radeonfb_info *rinfo = info->par; + u16 *red, *green, *blue, *transp; + u32 dac_cntl2, vclk_cntl = 0; + int i, start, rc = 0; + + if (!rinfo->asleep) { + if (rinfo->is_mobility) { + vclk_cntl = INPLL(VCLK_ECP_CNTL); + OUTPLL(VCLK_ECP_CNTL, + vclk_cntl & ~PIXCLK_DAC_ALWAYS_ONb); + } + + /* Make sure we are on first palette */ + if (rinfo->has_CRTC2) { + dac_cntl2 = INREG(DAC_CNTL2); + dac_cntl2 &= ~DAC2_PALETTE_ACCESS_CNTL; + OUTREG(DAC_CNTL2, dac_cntl2); + } + } + + red = cmap->red; + green = cmap->green; + blue = cmap->blue; + transp = cmap->transp; + start = cmap->start; + + for (i = 0; i < cmap->len; i++) { + u_int hred, hgreen, hblue, htransp = 0xffff; + + hred = *red++; + hgreen = *green++; + hblue = *blue++; + if (transp) + htransp = *transp++; + rc = radeon_setcolreg (start++, hred, hgreen, hblue, htransp, + rinfo); + if (rc) + break; + } + + if (!rinfo->asleep && rinfo->is_mobility) + OUTPLL(VCLK_ECP_CNTL, vclk_cntl); + + return rc; +} static void radeon_save_state (struct radeonfb_info *rinfo, struct radeon_regs *save) @@ -1796,6 +1865,7 @@ static struct fb_ops radeonfb_ops = { .fb_check_var = radeonfb_check_var, .fb_set_par = radeonfb_set_par, .fb_setcolreg = radeonfb_setcolreg, + .fb_setcmap = radeonfb_setcmap, .fb_pan_display = radeonfb_pan_display, .fb_blank = radeonfb_blank, .fb_ioctl = radeonfb_ioctl, diff --git a/drivers/video/fbcmap.c b/drivers/video/fbcmap.c index c51f8fb..4e5ce8f 100644 --- a/drivers/video/fbcmap.c +++ b/drivers/video/fbcmap.c @@ -222,8 +222,11 @@ int fb_set_cmap(struct fb_cmap *cmap, struct fb_info *info) transp = cmap->transp; start = cmap->start; - if (start < 0 || !info->fbops->fb_setcolreg) + if (start < 0 || (!info->fbops->fb_setcolreg && + !info->fbops->fb_setcmap)) return -EINVAL; + if (info->fbops->fb_setcmap) + return info->fbops->fb_setcmap(cmap, info); for (i = 0; i < cmap->len; i++) { hred = *red++; hgreen = *green++; @@ -250,8 +253,33 @@ int fb_set_user_cmap(struct fb_cmap_user *cmap, struct fb_info *info) transp = cmap->transp; start = cmap->start; - if (start < 0 || !info->fbops->fb_setcolreg) + if (start < 0 || (!info->fbops->fb_setcolreg && + !info->fbops->fb_setcmap)) return -EINVAL; + + /* If we can batch, do it */ + if (info->fbops->fb_setcmap && cmap->len > 1) { + struct fb_cmap umap; + int size = cmap->len * sizeof(u16); + int rc; + + memset(&umap, 0, sizeof(struct fb_cmap)); + rc = fb_alloc_cmap(&umap, cmap->len, transp != NULL); + if (rc) + return rc; + if (copy_from_user(umap.red, red, size) || + copy_from_user(umap.green, green, size) || + copy_from_user(umap.blue, blue, size) || + (transp && copy_from_user(umap.transp, transp, size))) { + rc = -EFAULT; + } + umap.start = start; + if (rc == 0) + rc = info->fbops->fb_setcmap(&umap, info); + fb_dealloc_cmap(&umap); + return rc; + } + for (i = 0; i < cmap->len; i++, red++, blue++, green++) { if (get_user(hred, red) || get_user(hgreen, green) || diff --git a/include/linux/fb.h b/include/linux/fb.h index b45d3e2..b468bf4 100644 --- a/include/linux/fb.h +++ b/include/linux/fb.h @@ -563,6 +563,9 @@ struct fb_ops { int (*fb_setcolreg)(unsigned regno, unsigned red, unsigned green, unsigned blue, unsigned transp, struct fb_info *info); + /* set color registers in batch */ + int (*fb_setcmap)(struct fb_cmap *cmap, struct fb_info *info); + /* blank display */ int (*fb_blank)(int blank, struct fb_info *info); -- cgit v0.10.2 From 917bb0771aa077f62a3de75028a45f243d3954a8 Mon Sep 17 00:00:00 2001 From: "Antonino A. Daplas" Date: Sun, 1 May 2005 08:59:22 -0700 Subject: [PATCH] nvidiafb: ioremap and i2c fixes - Add 'vram' option to specify amount of video RAM to remap - Limit remap size to 64 MIB - Use info->screen_size for remapped RAM - Fix misplaced label in failure path Signed-off-by: Antonino Daplas Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/nvidia/nvidia.c b/drivers/video/nvidia/nvidia.c index 3a6555a..47733f5 100644 --- a/drivers/video/nvidia/nvidia.c +++ b/drivers/video/nvidia/nvidia.c @@ -408,6 +408,7 @@ static int hwcur __devinitdata = 0; static int noaccel __devinitdata = 0; static int noscale __devinitdata = 0; static int paneltweak __devinitdata = 0; +static int vram __devinitdata = 0; #ifdef CONFIG_MTRR static int nomtrr __devinitdata = 0; #endif @@ -1180,7 +1181,7 @@ static int nvidiafb_check_var(struct fb_var_screeninfo *var, var->xres_virtual = (var->xres_virtual + 63) & ~63; - vramlen = info->fix.smem_len; + vramlen = info->screen_size; pitch = ((var->xres_virtual * var->bits_per_pixel) + 7) / 8; memlen = pitch * var->yres_virtual; @@ -1343,7 +1344,7 @@ static int __devinit nvidia_set_fbinfo(struct fb_info *info) /* maximize virtual vertical length */ lpitch = info->var.xres_virtual * ((info->var.bits_per_pixel + 7) >> 3); - info->var.yres_virtual = info->fix.smem_len / lpitch; + info->var.yres_virtual = info->screen_size / lpitch; info->pixmap.scan_align = 4; info->pixmap.buf_align = 4; @@ -1507,12 +1508,20 @@ static int __devinit nvidiafb_probe(struct pci_dev *pd, par->FbAddress = nvidiafb_fix.smem_start; par->FbMapSize = par->RamAmountKBytes * 1024; + if (vram && vram * 1024 * 1024 < par->FbMapSize) + par->FbMapSize = vram * 1024 * 1024; + + /* Limit amount of vram to 64 MB */ + if (par->FbMapSize > 64 * 1024 * 1024) + par->FbMapSize = 64 * 1024 * 1024; + par->FbUsableSize = par->FbMapSize - (128 * 1024); par->ScratchBufferSize = (par->Architecture < NV_ARCH_10) ? 8 * 1024 : 16 * 1024; par->ScratchBufferStart = par->FbUsableSize - par->ScratchBufferSize; info->screen_base = ioremap(nvidiafb_fix.smem_start, par->FbMapSize); - nvidiafb_fix.smem_len = par->FbUsableSize; + info->screen_size = par->FbUsableSize; + nvidiafb_fix.smem_len = par->RamAmountKBytes * 1024; if (!info->screen_base) { printk(KERN_ERR PFX "cannot ioremap FB base\n"); @@ -1524,7 +1533,8 @@ static int __devinit nvidiafb_probe(struct pci_dev *pd, #ifdef CONFIG_MTRR if (!nomtrr) { par->mtrr.vram = mtrr_add(nvidiafb_fix.smem_start, - par->FbMapSize, MTRR_TYPE_WRCOMB, 1); + par->RamAmountKBytes * 1024, + MTRR_TYPE_WRCOMB, 1); if (par->mtrr.vram < 0) { printk(KERN_ERR PFX "unable to setup MTRR\n"); } else { @@ -1566,9 +1576,9 @@ static int __devinit nvidiafb_probe(struct pci_dev *pd, err_out_iounmap_fb: iounmap(info->screen_base); + err_out_free_base1: fb_destroy_modedb(info->monspecs.modedb); nvidia_delete_i2c_busses(par); - err_out_free_base1: iounmap(par->REGS); err_out_free_base0: pci_release_regions(pd); @@ -1645,6 +1655,8 @@ static int __devinit nvidiafb_setup(char *options) noscale = 1; } else if (!strncmp(this_opt, "paneltweak:", 11)) { paneltweak = simple_strtoul(this_opt+11, NULL, 0); + } else if (!strncmp(this_opt, "vram:", 5)) { + vram = simple_strtoul(this_opt+5, NULL, 0); #ifdef CONFIG_MTRR } else if (!strncmp(this_opt, "nomtrr", 6)) { nomtrr = 1; @@ -1716,6 +1728,10 @@ module_param(forceCRTC, int, 0); MODULE_PARM_DESC(forceCRTC, "Forces usage of a particular CRTC in case autodetection " "fails. (0 or 1) (default=autodetect)"); +module_param(vram, int, 0); +MODULE_PARM_DESC(vram, + "amount of framebuffer memory to remap in MiB" + "(default=0 - remap entire memory)"); #ifdef CONFIG_MTRR module_param(nomtrr, bool, 0); MODULE_PARM_DESC(nomtrr, "Disables MTRR support (0 or 1=disabled) " -- cgit v0.10.2 From 5f76be80d96f60adfc91f2acf22b146ce0e3072f Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Sun, 1 May 2005 08:59:23 -0700 Subject: [PATCH] fbdev: edid.h cleanups This patch removes some completely unused code. Signed-off-by: Adrian Bunk Signed-off-by: Antonino Daplas Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/i386/boot/compressed/misc.c b/arch/i386/boot/compressed/misc.c index fa67045..cedc55c 100644 --- a/arch/i386/boot/compressed/misc.c +++ b/arch/i386/boot/compressed/misc.c @@ -12,7 +12,6 @@ #include #include #include -#include