diff options
Diffstat (limited to 'arch/s390')
32 files changed, 877 insertions, 488 deletions
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 1831833..f6a68e1 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -3,6 +3,10 @@ # see Documentation/kbuild/kconfig-language.txt. # +config SCHED_MC + def_bool y + depends on SMP + config MMU def_bool y @@ -39,6 +43,9 @@ config GENERIC_HWEIGHT config GENERIC_TIME def_bool y +config GENERIC_CLOCKEVENTS + def_bool y + config GENERIC_BUG bool depends on BUG @@ -69,6 +76,8 @@ menu "Base setup" comment "Processor type and features" +source "kernel/time/Kconfig" + config 64BIT bool "64 bit kernel" help @@ -301,10 +310,7 @@ config QDIO tristate "QDIO support" ---help--- This driver provides the Queued Direct I/O base support for - IBM mainframes. - - For details please refer to the documentation provided by IBM at - <http://www10.software.ibm.com/developerworks/opensource/linux390> + IBM System z. To compile this driver as a module, choose M here: the module will be called qdio. @@ -486,25 +492,6 @@ config APPLDATA_NET_SUM source kernel/Kconfig.hz -config NO_IDLE_HZ - bool "No HZ timer ticks in idle" - help - Switches the regular HZ timer off when the system is going idle. - This helps z/VM to detect that the Linux system is idle. VM can - then "swap-out" this guest which reduces memory usage. It also - reduces the overhead of idle systems. - - The HZ timer can be switched on/off via /proc/sys/kernel/hz_timer. - hz_timer=0 means HZ timer is disabled. hz_timer=1 means HZ - timer is active. - -config NO_IDLE_HZ_INIT - bool "HZ timer in idle off by default" - depends on NO_IDLE_HZ - help - The HZ timer is switched off in idle by default. That means the - HZ timer is already disabled at boot time. - config S390_HYPFS_FS bool "s390 hypervisor file system support" select SYS_HYPERVISOR diff --git a/arch/s390/crypto/aes_s390.c b/arch/s390/crypto/aes_s390.c index a3f67f8..e33f32b 100644 --- a/arch/s390/crypto/aes_s390.c +++ b/arch/s390/crypto/aes_s390.c @@ -499,7 +499,7 @@ static struct crypto_alg cbc_aes_alg = { } }; -static int __init aes_init(void) +static int __init aes_s390_init(void) { int ret; @@ -542,15 +542,15 @@ aes_err: goto out; } -static void __exit aes_fini(void) +static void __exit aes_s390_fini(void) { crypto_unregister_alg(&cbc_aes_alg); crypto_unregister_alg(&ecb_aes_alg); crypto_unregister_alg(&aes_alg); } -module_init(aes_init); -module_exit(aes_fini); +module_init(aes_s390_init); +module_exit(aes_s390_fini); MODULE_ALIAS("aes"); diff --git a/arch/s390/crypto/des_s390.c b/arch/s390/crypto/des_s390.c index ea22707..4aba83b 100644 --- a/arch/s390/crypto/des_s390.c +++ b/arch/s390/crypto/des_s390.c @@ -550,7 +550,7 @@ static struct crypto_alg cbc_des3_192_alg = { } }; -static int init(void) +static int des_s390_init(void) { int ret = 0; @@ -612,7 +612,7 @@ des_err: goto out; } -static void __exit fini(void) +static void __exit des_s390_fini(void) { crypto_unregister_alg(&cbc_des3_192_alg); crypto_unregister_alg(&ecb_des3_192_alg); @@ -625,8 +625,8 @@ static void __exit fini(void) crypto_unregister_alg(&des_alg); } -module_init(init); -module_exit(fini); +module_init(des_s390_init); +module_exit(des_s390_fini); MODULE_ALIAS("des"); MODULE_ALIAS("des3_ede"); diff --git a/arch/s390/crypto/sha1_s390.c b/arch/s390/crypto/sha1_s390.c index 5a834f6..9cf9eca 100644 --- a/arch/s390/crypto/sha1_s390.c +++ b/arch/s390/crypto/sha1_s390.c @@ -137,7 +137,7 @@ static struct crypto_alg alg = { .dia_final = sha1_final } } }; -static int __init init(void) +static int __init sha1_s390_init(void) { if (!crypt_s390_func_available(KIMD_SHA_1)) return -EOPNOTSUPP; @@ -145,13 +145,13 @@ static int __init init(void) return crypto_register_alg(&alg); } -static void __exit fini(void) +static void __exit sha1_s390_fini(void) { crypto_unregister_alg(&alg); } -module_init(init); -module_exit(fini); +module_init(sha1_s390_init); +module_exit(sha1_s390_fini); MODULE_ALIAS("sha1"); diff --git a/arch/s390/crypto/sha256_s390.c b/arch/s390/crypto/sha256_s390.c index ccf8633..2a3d756 100644 --- a/arch/s390/crypto/sha256_s390.c +++ b/arch/s390/crypto/sha256_s390.c @@ -133,7 +133,7 @@ static struct crypto_alg alg = { .dia_final = sha256_final } } }; -static int init(void) +static int sha256_s390_init(void) { if (!crypt_s390_func_available(KIMD_SHA_256)) return -EOPNOTSUPP; @@ -141,13 +141,13 @@ static int init(void) return crypto_register_alg(&alg); } -static void __exit fini(void) +static void __exit sha256_s390_fini(void) { crypto_unregister_alg(&alg); } -module_init(init); -module_exit(fini); +module_init(sha256_s390_init); +module_exit(sha256_s390_fini); MODULE_ALIAS("sha256"); diff --git a/arch/s390/defconfig b/arch/s390/defconfig index 62f6b5a..a72f208 100644 --- a/arch/s390/defconfig +++ b/arch/s390/defconfig @@ -3,6 +3,7 @@ # Linux kernel version: 2.6.25-rc4 # Wed Mar 5 11:22:59 2008 # +CONFIG_SCHED_MC=y CONFIG_MMU=y CONFIG_ZONE_DMA=y CONFIG_LOCKDEP_SUPPORT=y @@ -537,11 +538,9 @@ CONFIG_CTC=m # CONFIG_SMSGIUCV is not set # CONFIG_CLAW is not set CONFIG_QETH=y - -# -# Gigabit Ethernet default settings -# -# CONFIG_QETH_IPV6 is not set +CONFIG_QETH_L2=y +CONFIG_QETH_L3=y +CONFIG_QETH_IPV6=y CONFIG_CCWGROUP=y # CONFIG_PPP is not set # CONFIG_SLIP is not set diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile index 4d3e383..77051cd 100644 --- a/arch/s390/kernel/Makefile +++ b/arch/s390/kernel/Makefile @@ -11,7 +11,7 @@ CFLAGS_smp.o := -Wno-nonnull obj-y := bitmap.o traps.o time.o process.o base.o early.o \ setup.o sys_s390.o ptrace.o signal.o cpcmd.o ebcdic.o \ - semaphore.o s390_ext.o debug.o irq.o ipl.o dis.o diag.o + s390_ext.o debug.o irq.o ipl.o dis.o diag.o obj-y += $(if $(CONFIG_64BIT),entry64.o,entry.o) obj-y += $(if $(CONFIG_64BIT),reipl64.o,reipl.o) @@ -19,7 +19,7 @@ obj-y += $(if $(CONFIG_64BIT),reipl64.o,reipl.o) extra-y += head.o init_task.o vmlinux.lds obj-$(CONFIG_MODULES) += s390_ksyms.o module.o -obj-$(CONFIG_SMP) += smp.o +obj-$(CONFIG_SMP) += smp.o topology.o obj-$(CONFIG_AUDIT) += audit.o compat-obj-$(CONFIG_AUDIT) += compat_audit.o diff --git a/arch/s390/kernel/compat_linux.h b/arch/s390/kernel/compat_linux.h index e89f8c0..20723a0 100644 --- a/arch/s390/kernel/compat_linux.h +++ b/arch/s390/kernel/compat_linux.h @@ -162,4 +162,77 @@ struct ucontext32 { compat_sigset_t uc_sigmask; /* mask last for extensibility */ }; +struct __sysctl_args32; +struct stat64_emu31; +struct mmap_arg_struct_emu31; +struct fadvise64_64_args; +struct old_sigaction32; +struct old_sigaction32; + +long sys32_chown16(const char __user * filename, u16 user, u16 group); +long sys32_lchown16(const char __user * filename, u16 user, u16 group); +long sys32_fchown16(unsigned int fd, u16 user, u16 group); +long sys32_setregid16(u16 rgid, u16 egid); +long sys32_setgid16(u16 gid); +long sys32_setreuid16(u16 ruid, u16 euid); +long sys32_setuid16(u16 uid); +long sys32_setresuid16(u16 ruid, u16 euid, u16 suid); +long sys32_getresuid16(u16 __user *ruid, u16 __user *euid, u16 __user *suid); +long sys32_setresgid16(u16 rgid, u16 egid, u16 sgid); +long sys32_getresgid16(u16 __user *rgid, u16 __user *egid, u16 __user *sgid); +long sys32_setfsuid16(u16 uid); +long sys32_setfsgid16(u16 gid); +long sys32_getgroups16(int gidsetsize, u16 __user *grouplist); +long sys32_setgroups16(int gidsetsize, u16 __user *grouplist); +long sys32_getuid16(void); +long sys32_geteuid16(void); +long sys32_getgid16(void); +long sys32_getegid16(void); +long sys32_ipc(u32 call, int first, int second, int third, u32 ptr); +long sys32_truncate64(const char __user * path, unsigned long high, + unsigned long low); +long sys32_ftruncate64(unsigned int fd, unsigned long high, unsigned long low); +long sys32_sched_rr_get_interval(compat_pid_t pid, + struct compat_timespec __user *interval); +long sys32_rt_sigprocmask(int how, compat_sigset_t __user *set, + compat_sigset_t __user *oset, size_t sigsetsize); +long sys32_rt_sigpending(compat_sigset_t __user *set, size_t sigsetsize); +long sys32_rt_sigqueueinfo(int pid, int sig, compat_siginfo_t __user *uinfo); +long sys32_execve(void); +long sys32_init_module(void __user *umod, unsigned long len, + const char __user *uargs); +long sys32_delete_module(const char __user *name_user, unsigned int flags); +long sys32_gettimeofday(struct compat_timeval __user *tv, + struct timezone __user *tz); +long sys32_settimeofday(struct compat_timeval __user *tv, + struct timezone __user *tz); +long sys32_pause(void); +long sys32_pread64(unsigned int fd, char __user *ubuf, size_t count, + u32 poshi, u32 poslo); +long sys32_pwrite64(unsigned int fd, const char __user *ubuf, + size_t count, u32 poshi, u32 poslo); +compat_ssize_t sys32_readahead(int fd, u32 offhi, u32 offlo, s32 count); +long sys32_sendfile(int out_fd, int in_fd, compat_off_t __user *offset, + size_t count); +long sys32_sendfile64(int out_fd, int in_fd, compat_loff_t __user *offset, + s32 count); +long sys32_sysctl(struct __sysctl_args32 __user *args); +long sys32_stat64(char __user * filename, struct stat64_emu31 __user * statbuf); +long sys32_lstat64(char __user * filename, + struct stat64_emu31 __user * statbuf); +long sys32_fstat64(unsigned long fd, struct stat64_emu31 __user * statbuf); +long sys32_fstatat64(unsigned int dfd, char __user *filename, + struct stat64_emu31 __user* statbuf, int flag); +unsigned long old32_mmap(struct mmap_arg_struct_emu31 __user *arg); +long sys32_mmap2(struct mmap_arg_struct_emu31 __user *arg); +long sys32_read(unsigned int fd, char __user * buf, size_t count); +long sys32_write(unsigned int fd, char __user * buf, size_t count); +long sys32_clone(void); +long sys32_fadvise64(int fd, loff_t offset, size_t len, int advise); +long sys32_fadvise64_64(struct fadvise64_64_args __user *args); +long sys32_sigaction(int sig, const struct old_sigaction32 __user *act, + struct old_sigaction32 __user *oact); +long sys32_rt_sigaction(int sig, const struct sigaction32 __user *act, + struct sigaction32 __user *oact, size_t sigsetsize); +long sys32_sigaltstack(const stack_t32 __user *uss, stack_t32 __user *uoss); #endif /* _ASM_S390X_S390_H */ diff --git a/arch/s390/kernel/compat_signal.c b/arch/s390/kernel/compat_signal.c index a5692c4..c7f02e7 100644 --- a/arch/s390/kernel/compat_signal.c +++ b/arch/s390/kernel/compat_signal.c @@ -29,6 +29,7 @@ #include <asm/lowcore.h> #include "compat_linux.h" #include "compat_ptrace.h" +#include "entry.h" #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP))) @@ -428,6 +429,10 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs * regs, size_t frame_size) /* Default to using normal stack */ sp = (unsigned long) A(regs->gprs[15]); + /* Overflow on alternate signal stack gives SIGSEGV. */ + if (on_sig_stack(sp) && !on_sig_stack((sp - frame_size) & -8UL)) + return (void __user *) -1UL; + /* This is the X/Open sanctioned signal stack switching. */ if (ka->sa.sa_flags & SA_ONSTACK) { if (! sas_ss_flags(sp)) @@ -461,6 +466,9 @@ static int setup_frame32(int sig, struct k_sigaction *ka, if (!access_ok(VERIFY_WRITE, frame, sizeof(sigframe32))) goto give_sigsegv; + if (frame == (void __user *) -1UL) + goto give_sigsegv; + if (__copy_to_user(&frame->sc.oldmask, &set->sig, _SIGMASK_COPY_SIZE32)) goto give_sigsegv; @@ -514,6 +522,9 @@ static int setup_rt_frame32(int sig, struct k_sigaction *ka, siginfo_t *info, if (!access_ok(VERIFY_WRITE, frame, sizeof(rt_sigframe32))) goto give_sigsegv; + if (frame == (void __user *) -1UL) + goto give_sigsegv; + if (copy_siginfo_to_user32(&frame->info, info)) goto give_sigsegv; diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c index 1b2f5ce..1e7d4ac 100644 --- a/arch/s390/kernel/debug.c +++ b/arch/s390/kernel/debug.c @@ -73,7 +73,7 @@ static ssize_t debug_input(struct file *file, const char __user *user_buf, static int debug_open(struct inode *inode, struct file *file); static int debug_close(struct inode *inode, struct file *file); static debug_info_t* debug_info_create(char *name, int pages_per_area, - int nr_areas, int buf_size); + int nr_areas, int buf_size, mode_t mode); static void debug_info_get(debug_info_t *); static void debug_info_put(debug_info_t *); static int debug_prolog_level_fn(debug_info_t * id, @@ -157,7 +157,7 @@ struct debug_view debug_sprintf_view = { }; /* used by dump analysis tools to determine version of debug feature */ -unsigned int debug_feature_version = __DEBUG_FEATURE_VERSION; +static unsigned int __used debug_feature_version = __DEBUG_FEATURE_VERSION; /* static globals */ @@ -327,7 +327,8 @@ debug_info_free(debug_info_t* db_info){ */ static debug_info_t* -debug_info_create(char *name, int pages_per_area, int nr_areas, int buf_size) +debug_info_create(char *name, int pages_per_area, int nr_areas, int buf_size, + mode_t mode) { debug_info_t* rc; @@ -336,6 +337,8 @@ debug_info_create(char *name, int pages_per_area, int nr_areas, int buf_size) if(!rc) goto out; + rc->mode = mode & ~S_IFMT; + /* create root directory */ rc->debugfs_root_entry = debugfs_create_dir(rc->name, debug_debugfs_root_entry); @@ -676,23 +679,30 @@ debug_close(struct inode *inode, struct file *file) } /* - * debug_register: - * - creates and initializes debug area for the caller - * - returns handle for debug area + * debug_register_mode: + * - Creates and initializes debug area for the caller + * The mode parameter allows to specify access rights for the s390dbf files + * - Returns handle for debug area */ -debug_info_t* -debug_register (char *name, int pages_per_area, int nr_areas, int buf_size) +debug_info_t *debug_register_mode(char *name, int pages_per_area, int nr_areas, + int buf_size, mode_t mode, uid_t uid, + gid_t gid) { debug_info_t *rc = NULL; + /* Since debugfs currently does not support uid/gid other than root, */ + /* we do not allow gid/uid != 0 until we get support for that. */ + if ((uid != 0) || (gid != 0)) + printk(KERN_WARNING "debug: Warning - Currently only uid/gid " + "= 0 are supported. Using root as owner now!"); if (!initialized) BUG(); mutex_lock(&debug_mutex); /* create new debug_info */ - rc = debug_info_create(name, pages_per_area, nr_areas, buf_size); + rc = debug_info_create(name, pages_per_area, nr_areas, buf_size, mode); if(!rc) goto out; debug_register_view(rc, &debug_level_view); @@ -705,6 +715,20 @@ out: mutex_unlock(&debug_mutex); return rc; } +EXPORT_SYMBOL(debug_register_mode); + +/* + * debug_register: + * - creates and initializes debug area for the caller + * - returns handle for debug area + */ + +debug_info_t *debug_register(char *name, int pages_per_area, int nr_areas, + int buf_size) +{ + return debug_register_mode(name, pages_per_area, nr_areas, buf_size, + S_IRUSR | S_IWUSR, 0, 0); +} /* * debug_unregister: @@ -1073,15 +1097,16 @@ debug_register_view(debug_info_t * id, struct debug_view *view) int rc = 0; int i; unsigned long flags; - mode_t mode = S_IFREG; + mode_t mode; struct dentry *pde; if (!id) goto out; - if (view->prolog_proc || view->format_proc || view->header_proc) - mode |= S_IRUSR; - if (view->input_proc) - mode |= S_IWUSR; + mode = (id->mode | S_IFREG) & ~S_IXUGO; + if (!(view->prolog_proc || view->format_proc || view->header_proc)) + mode &= ~(S_IRUSR | S_IRGRP | S_IROTH); + if (!view->input_proc) + mode &= ~(S_IWUSR | S_IWGRP | S_IWOTH); pde = debugfs_create_file(view->name, mode, id->debugfs_root_entry, id , &debug_file_ops); if (!pde){ diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c index 01832c4..540a67f 100644 --- a/arch/s390/kernel/early.c +++ b/arch/s390/kernel/early.c @@ -21,6 +21,7 @@ #include <asm/setup.h> #include <asm/cpcmd.h> #include <asm/sclp.h> +#include "entry.h" /* * Create a Kernel NSS if the SAVESYS= parameter is defined diff --git a/arch/s390/kernel/entry.h b/arch/s390/kernel/entry.h new file mode 100644 index 0000000..6b18963 --- /dev/null +++ b/arch/s390/kernel/entry.h @@ -0,0 +1,60 @@ +#ifndef _ENTRY_H +#define _ENTRY_H + +#include <linux/types.h> +#include <linux/signal.h> +#include <asm/ptrace.h> + +typedef void pgm_check_handler_t(struct pt_regs *, long); +extern pgm_check_handler_t *pgm_check_table[128]; +pgm_check_handler_t do_protection_exception; +pgm_check_handler_t do_dat_exception; + +extern int sysctl_userprocess_debug; + +void do_single_step(struct pt_regs *regs); +void syscall_trace(struct pt_regs *regs, int entryexit); +void kernel_stack_overflow(struct pt_regs * regs); +void do_signal(struct pt_regs *regs); +int handle_signal32(unsigned long sig, struct k_sigaction *ka, + siginfo_t *info, sigset_t *oldset, struct pt_regs *regs); + +void do_extint(struct pt_regs *regs, unsigned short code); +int __cpuinit start_secondary(void *cpuvoid); +void __init startup_init(void); +void die(const char * str, struct pt_regs * regs, long err); + +struct new_utsname; +struct mmap_arg_struct; +struct fadvise64_64_args; +struct old_sigaction; +struct sel_arg_struct; + +long sys_pipe(unsigned long __user *fildes); +long sys_mmap2(struct mmap_arg_struct __user *arg); +long old_mmap(struct mmap_arg_struct __user *arg); +long sys_ipc(uint call, int first, unsigned long second, + unsigned long third, void __user *ptr); +long s390x_newuname(struct new_utsname __user *name); +long s390x_personality(unsigned long personality); +long s390_fadvise64(int fd, u32 offset_high, u32 offset_low, + size_t len, int advice); +long s390_fadvise64_64(struct fadvise64_64_args __user *args); +long s390_fallocate(int fd, int mode, loff_t offset, u32 len_high, u32 len_low); +long sys_fork(void); +long sys_clone(void); +long sys_vfork(void); +void execve_tail(void); +long sys_execve(void); +int sys_sigsuspend(int history0, int history1, old_sigset_t mask); +long sys_sigaction(int sig, const struct old_sigaction __user *act, + struct old_sigaction __user *oact); +long sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss); +long sys_sigreturn(void); +long sys_rt_sigreturn(void); +long sys32_sigreturn(void); +long sys32_rt_sigreturn(void); +long old_select(struct sel_arg_struct __user *arg); +long sys_ptrace(long request, long pid, long addr, long data); + +#endif /* _ENTRY_H */ diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S index efde6e1..cd959c0 100644 --- a/arch/s390/kernel/entry64.S +++ b/arch/s390/kernel/entry64.S @@ -475,6 +475,7 @@ pgm_check_handler: pgm_no_vtime: #endif lg %r9,__LC_THREAD_INFO # load pointer to thread_info struct + mvc SP_ARGS(8,%r15),__LC_LAST_BREAK TRACE_IRQS_OFF lgf %r3,__LC_PGM_ILC # load program interruption code lghi %r8,0x7f @@ -847,6 +848,7 @@ stack_overflow: je 0f la %r1,__LC_SAVE_AREA+32 0: mvc SP_R12(32,%r15),0(%r1) # move %r12-%r15 to stack + mvc SP_ARGS(8,%r15),__LC_LAST_BREAK xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15) # clear back chain la %r2,SP_PTREGS(%r15) # load pt_regs jg kernel_stack_overflow diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c index 375232c..5325424 100644 --- a/arch/s390/kernel/ipl.c +++ b/arch/s390/kernel/ipl.c @@ -655,7 +655,7 @@ static struct kobj_attribute reipl_type_attr = static struct kset *reipl_kset; -void reipl_run(struct shutdown_trigger *trigger) +static void reipl_run(struct shutdown_trigger *trigger) { struct ccw_dev_id devid; static char buf[100]; diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c index c5549a2..ed04d13 100644 --- a/arch/s390/kernel/kprobes.c +++ b/arch/s390/kernel/kprobes.c @@ -360,7 +360,7 @@ no_kprobe: * - When the probed function returns, this probe * causes the handlers to fire */ -void kretprobe_trampoline_holder(void) +static void __used kretprobe_trampoline_holder(void) { asm volatile(".global kretprobe_trampoline\n" "kretprobe_trampoline: bcr 0,0\n"); diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c index ce20315..c1aff19 100644 --- a/arch/s390/kernel/process.c +++ b/arch/s390/kernel/process.c @@ -36,6 +36,8 @@ #include <linux/module.h> #include <linux/notifier.h> #include <linux/utsname.h> +#include <linux/tick.h> +#include <linux/elfcore.h> #include <asm/uaccess.h> #include <asm/pgtable.h> #include <asm/system.h> @@ -44,6 +46,7 @@ #include <asm/irq.h> #include <asm/timer.h> #include <asm/cpu.h> +#include "entry.h" asmlinkage void ret_from_fork(void) asm ("ret_from_fork"); @@ -76,6 +79,7 @@ unsigned long thread_saved_pc(struct task_struct *tsk) * Need to know about CPUs going idle? */ static ATOMIC_NOTIFIER_HEAD(idle_chain); +DEFINE_PER_CPU(struct s390_idle_data, s390_idle); int register_idle_notifier(struct notifier_block *nb) { @@ -89,9 +93,33 @@ int unregister_idle_notifier(struct notifier_block *nb) } EXPORT_SYMBOL(unregister_idle_notifier); -void do_monitor_call(struct pt_regs *regs, long interruption_code) +static int s390_idle_enter(void) +{ + struct s390_idle_data *idle; + int nr_calls = 0; + void *hcpu; + int rc; + + hcpu = (void *)(long)smp_processor_id(); + rc = __atomic_notifier_call_chain(&idle_chain, S390_CPU_IDLE, hcpu, -1, + &nr_calls); + if (rc == NOTIFY_BAD) { + nr_calls--; + __atomic_notifier_call_chain(&idle_chain, S390_CPU_NOT_IDLE, + hcpu, nr_calls, NULL); + return rc; + } + idle = &__get_cpu_var(s390_idle); + spin_lock(&idle->lock); + idle->idle_count++; + idle->in_idle = 1; + idle->idle_enter = get_clock(); + spin_unlock(&idle->lock); + return NOTIFY_OK; +} + +void s390_idle_leave(void) { -#ifdef CONFIG_SMP struct s390_idle_data *idle; idle = &__get_cpu_var(s390_idle); @@ -99,10 +127,6 @@ void do_monitor_call(struct pt_regs *regs, long interruption_code) idle->idle_time += get_clock() - idle->idle_enter; idle->in_idle = 0; spin_unlock(&idle->lock); -#endif - /* disable monitor call class 0 */ - __ctl_clear_bit(8, 15); - atomic_notifier_call_chain(&idle_chain, S390_CPU_NOT_IDLE, (void *)(long) smp_processor_id()); } @@ -113,61 +137,30 @@ extern void s390_handle_mcck(void); */ static void default_idle(void) { - int cpu, rc; - int nr_calls = 0; - void *hcpu; -#ifdef CONFIG_SMP - struct s390_idle_data *idle; -#endif - /* CPU is going idle. */ - cpu = smp_processor_id(); - hcpu = (void *)(long)cpu; local_irq_disable(); if (need_resched()) { local_irq_enable(); return; } - - rc = __atomic_notifier_call_chain(&idle_chain, S390_CPU_IDLE, hcpu, -1, - &nr_calls); - if (rc == NOTIFY_BAD) { - nr_calls--; - __atomic_notifier_call_chain(&idle_chain, S390_CPU_NOT_IDLE, - hcpu, nr_calls, NULL); + if (s390_idle_enter() == NOTIFY_BAD) { local_irq_enable(); return; } - - /* enable monitor call class 0 */ - __ctl_set_bit(8, 15); - #ifdef CONFIG_HOTPLUG_CPU - if (cpu_is_offline(cpu)) { + if (cpu_is_offline(smp_processor_id())) { preempt_enable_no_resched(); cpu_die(); } #endif - local_mcck_disable(); if (test_thread_flag(TIF_MCCK_PENDING)) { local_mcck_enable(); - /* disable monitor call class 0 */ - __ctl_clear_bit(8, 15); - atomic_notifier_call_chain(&idle_chain, S390_CPU_NOT_IDLE, - hcpu); + s390_idle_leave(); local_irq_enable(); s390_handle_mcck(); return; } -#ifdef CONFIG_SMP - idle = &__get_cpu_var(s390_idle); - spin_lock(&idle->lock); - idle->idle_count++; - idle->in_idle = 1; - idle->idle_enter = get_clock(); - spin_unlock(&idle->lock); -#endif trace_hardirqs_on(); /* Wait for external, I/O or machine check interrupt. */ __load_psw_mask(psw_kernel_bits | PSW_MASK_WAIT | @@ -177,9 +170,10 @@ static void default_idle(void) void cpu_idle(void) { for (;;) { + tick_nohz_stop_sched_tick(); while (!need_resched()) default_idle(); - + tick_nohz_restart_sched_tick(); preempt_enable_no_resched(); schedule(); preempt_disable(); @@ -201,6 +195,7 @@ void show_regs(struct pt_regs *regs) /* Show stack backtrace if pt_regs is from kernel mode */ if (!(regs->psw.mask & PSW_MASK_PSTATE)) show_trace(NULL, (unsigned long *) regs->gprs[15]); + show_last_breaking_event(regs); } extern void kernel_thread_starter(void); diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c index 6e036ba..58a0642 100644 --- a/arch/s390/kernel/ptrace.c +++ b/arch/s390/kernel/ptrace.c @@ -41,6 +41,7 @@ #include <asm/system.h> #include <asm/uaccess.h> #include <asm/unistd.h> +#include "entry.h" #ifdef CONFIG_COMPAT #include "compat_ptrace.h" diff --git a/arch/s390/kernel/s390_ext.c b/arch/s390/kernel/s390_ext.c index acf93db..e019b41 100644 --- a/arch/s390/kernel/s390_ext.c +++ b/arch/s390/kernel/s390_ext.c @@ -13,11 +13,12 @@ #include <linux/errno.h> #include <linux/kernel_stat.h> #include <linux/interrupt.h> - +#include <asm/cpu.h> #include <asm/lowcore.h> #include <asm/s390_ext.h> #include <asm/irq_regs.h> #include <asm/irq.h> +#include "entry.h" /* * ext_int_hash[index] is the start of the list for all external interrupts @@ -119,13 +120,10 @@ void do_extint(struct pt_regs *regs, unsigned short code) old_regs = set_irq_regs(regs); irq_enter(); - asm volatile ("mc 0,0"); - if (S390_lowcore.int_clock >= S390_lowcore.jiffy_timer) - /** - * Make sure that the i/o interrupt did not "overtake" - * the last HZ timer interrupt. - */ - account_ticks(S390_lowcore.int_clock); + s390_idle_check(); + if (S390_lowcore.int_clock >= S390_lowcore.clock_comparator) + /* Serve timer interrupts first. */ + clock_comparator_work(); kstat_cpu(smp_processor_id()).irqs[EXTERNAL_INTERRUPT]++; index = ext_hash(code); for (p = ext_int_hash[index]; p; p = p->next) { diff --git a/arch/s390/kernel/s390_ksyms.c b/arch/s390/kernel/s390_ksyms.c index 7234c73..48238a1 100644 --- a/arch/s390/kernel/s390_ksyms.c +++ b/arch/s390/kernel/s390_ksyms.c @@ -27,13 +27,6 @@ EXPORT_SYMBOL(_zb_findmap); EXPORT_SYMBOL(_sb_findmap); /* - * semaphore ops - */ -EXPORT_SYMBOL(__up); -EXPORT_SYMBOL(__down); -EXPORT_SYMBOL(__down_interruptible); - -/* * binfmt_elf loader */ extern int dump_fpu (struct pt_regs * regs, s390_fp_regs *fpregs); diff --git a/arch/s390/kernel/semaphore.c b/arch/s390/kernel/semaphore.c deleted file mode 100644 index 191303f..0000000 --- a/arch/s390/kernel/semaphore.c +++ /dev/null @@ -1,108 +0,0 @@ -/* - * linux/arch/s390/kernel/semaphore.c - * - * S390 version - * Copyright (C) 1998-2000 IBM Corporation - * Author(s): Martin Schwidefsky - * - * Derived from "linux/arch/i386/kernel/semaphore.c - * Copyright (C) 1999, Linus Torvalds - * - */ -#include <linux/sched.h> -#include <linux/errno.h> -#include <linux/init.h> - -#include <asm/semaphore.h> - -/* - * Atomically update sem->count. Equivalent to: - * old_val = sem->count.counter; - * new_val = ((old_val >= 0) ? old_val : 0) + incr; - * sem->count.counter = new_val; - * return old_val; - */ -static inline int __sem_update_count(struct semaphore *sem, int incr) -{ - int old_val, new_val; - - asm volatile( - " l %0,0(%3)\n" - "0: ltr %1,%0\n" - " jhe 1f\n" - " lhi %1,0\n" - "1: ar %1,%4\n" - " cs %0,%1,0(%3)\n" - " jl 0b\n" - : "=&d" (old_val), "=&d" (new_val), "=m" (sem->count) - : "a" (&sem->count), "d" (incr), "m" (sem->count) - : "cc"); - return old_val; -} - -/* - * The inline function up() incremented count but the result - * was <= 0. This indicates that some process is waiting on - * the semaphore. The semaphore is free and we'll wake the - * first sleeping process, so we set count to 1 unless some - * other cpu has called up in the meantime in which case - * we just increment count by 1. - */ -void __up(struct semaphore *sem) -{ - __sem_update_count(sem, 1); - wake_up(&sem->wait); -} - -/* - * The inline function down() decremented count and the result - * was < 0. The wait loop will atomically test and update the - * semaphore counter following the rules: - * count > 0: decrement count, wake up queue and exit. - * count <= 0: set count to -1, go to sleep. - */ -void __sched __down(struct semaphore * sem) -{ - struct task_struct *tsk = current; - DECLARE_WAITQUEUE(wait, tsk); - - __set_task_state(tsk, TASK_UNINTERRUPTIBLE); - add_wait_queue_exclusive(&sem->wait, &wait); - while (__sem_update_count(sem, -1) <= 0) { - schedule(); - set_task_state(tsk, TASK_UNINTERRUPTIBLE); - } - remove_wait_queue(&sem->wait, &wait); - __set_task_state(tsk, TASK_RUNNING); - wake_up(&sem->wait); -} - -/* - * Same as __down() with an additional test for signals. - * If a signal is pending the count is updated as follows: - * count > 0: wake up queue and exit. - * count <= 0: set count to 0, wake up queue and exit. - */ -int __sched __down_interruptible(struct semaphore * sem) -{ - int retval = 0; - struct task_struct *tsk = current; - DECLARE_WAITQUEUE(wait, tsk); - - __set_task_state(tsk, TASK_INTERRUPTIBLE); - add_wait_queue_exclusive(&sem->wait, &wait); - while (__sem_update_count(sem, -1) <= 0) { - if (signal_pending(current)) { - __sem_update_count(sem, 0); - retval = -EINTR; - break; - } - schedule(); - set_task_state(tsk, TASK_INTERRUPTIBLE); - } - remove_wait_queue(&sem->wait, &wait); - __set_task_state(tsk, TASK_RUNNING); - wake_up(&sem->wait); - return retval; -} - diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 290e504..7141147 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -39,6 +39,7 @@ #include <linux/pfn.h> #include <linux/ctype.h> #include <linux/reboot.h> +#include <linux/topology.h> #include <asm/ipl.h> #include <asm/uaccess.h> @@ -427,7 +428,7 @@ setup_lowcore(void) 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->clock_comparator = -1ULL; lc->kernel_stack = ((unsigned long) &init_thread_union) + THREAD_SIZE; lc->async_stack = (unsigned long) __alloc_bootmem(ASYNC_SIZE, ASYNC_SIZE, 0) + ASYNC_SIZE; @@ -687,7 +688,7 @@ static __init unsigned int stfl(void) return S390_lowcore.stfl_fac_list; } -static __init int stfle(unsigned long long *list, int doublewords) +static int __init __stfle(unsigned long long *list, int doublewords) { typedef struct { unsigned long long _[doublewords]; } addrtype; register unsigned long __nr asm("0") = doublewords - 1; @@ -697,6 +698,13 @@ static __init int stfle(unsigned long long *list, int doublewords) return __nr + 1; } +int __init stfle(unsigned long long *list, int doublewords) +{ + if (!(stfl() & (1UL << 24))) + return -EOPNOTSUPP; + return __stfle(list, doublewords); +} + /* * Setup hardware capabilities. */ @@ -741,7 +749,7 @@ static void __init setup_hwcaps(void) * HWCAP_S390_DFP bit 6. */ if ((elf_hwcap & (1UL << 2)) && - stfle(&facility_list_extended, 1) > 0) { + __stfle(&facility_list_extended, 1) > 0) { if (facility_list_extended & (1ULL << (64 - 43))) elf_hwcap |= 1UL << 6; } @@ -823,6 +831,7 @@ setup_arch(char **cmdline_p) cpu_init(); __cpu_logical_map[0] = S390_lowcore.cpu_data.cpu_addr; + s390_init_cpu_topology(); /* * Setup capabilities (ELF_HWCAP & ELF_PLATFORM). diff --git a/arch/s390/kernel/signal.c b/arch/s390/kernel/signal.c index 4449bf3..b976820 100644 --- a/arch/s390/kernel/signal.c +++ b/arch/s390/kernel/signal.c @@ -27,6 +27,7 @@ #include <asm/ucontext.h> #include <asm/uaccess.h> #include <asm/lowcore.h> +#include "entry.h" #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP))) @@ -235,6 +236,10 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs * regs, size_t frame_size) /* Default to using normal stack */ sp = regs->gprs[15]; + /* Overflow on alternate signal stack gives SIGSEGV. */ + if (on_sig_stack(sp) && !on_sig_stack((sp - frame_size) & -8UL)) + return (void __user *) -1UL; + /* This is the X/Open sanctioned signal stack switching. */ if (ka->sa.sa_flags & SA_ONSTACK) { if (! sas_ss_flags(sp)) @@ -270,6 +275,9 @@ static int setup_frame(int sig, struct k_sigaction *ka, if (!access_ok(VERIFY_WRITE, frame, sizeof(sigframe))) goto give_sigsegv; + if (frame == (void __user *) -1UL) + goto give_sigsegv; + if (__copy_to_user(&frame->sc.oldmask, &set->sig, _SIGMASK_COPY_SIZE)) goto give_sigsegv; @@ -327,6 +335,9 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, if (!access_ok(VERIFY_WRITE, frame, sizeof(rt_sigframe))) goto give_sigsegv; + if (frame == (void __user *) -1UL) + goto give_sigsegv; + if (copy_siginfo_to_user(&frame->info, info)) goto give_sigsegv; @@ -474,11 +485,6 @@ void do_signal(struct pt_regs *regs) int ret; #ifdef CONFIG_COMPAT if (test_thread_flag(TIF_31BIT)) { - extern int handle_signal32(unsigned long sig, - struct k_sigaction *ka, - siginfo_t *info, - sigset_t *oldset, - struct pt_regs *regs); ret = handle_signal32(signr, &ka, &info, oldset, regs); } else diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 8f894d3..0dfa988 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -44,6 +44,7 @@ #include <asm/lowcore.h> #include <asm/sclp.h> #include <asm/cpu.h> +#include "entry.h" /* * An array with a pointer the lowcore of every CPU. @@ -67,13 +68,12 @@ enum s390_cpu_state { CPU_STATE_CONFIGURED, }; -#ifdef CONFIG_HOTPLUG_CPU -static DEFINE_MUTEX(smp_cpu_state_mutex); -#endif +DEFINE_MUTEX(smp_cpu_state_mutex); +int smp_cpu_polarization[NR_CPUS]; static int smp_cpu_state[NR_CPUS]; +static int cpu_management; static DEFINE_PER_CPU(struct cpu, cpu_devices); -DEFINE_PER_CPU(struct s390_idle_data, s390_idle); static void smp_ext_bitcall(int, ec_bit_sig); @@ -298,7 +298,7 @@ static void smp_ext_bitcall(int cpu, ec_bit_sig sig) /* * this function sends a 'purge tlb' signal to another CPU. */ -void smp_ptlb_callback(void *info) +static void smp_ptlb_callback(void *info) { __tlb_flush_local(); } @@ -456,6 +456,7 @@ static int smp_rescan_cpus_sigp(cpumask_t avail) if (cpu_known(cpu_id)) continue; __cpu_logical_map[logical_cpu] = cpu_id; + smp_cpu_polarization[logical_cpu] = POLARIZATION_UNKNWN; if (!cpu_stopped(logical_cpu)) continue; cpu_set(logical_cpu, cpu_present_map); @@ -489,6 +490,7 @@ static int smp_rescan_cpus_sclp(cpumask_t avail) if (cpu_known(cpu_id)) continue; __cpu_logical_map[logical_cpu] = cpu_id; + smp_cpu_polarization[logical_cpu] = POLARIZATION_UNKNWN; cpu_set(logical_cpu, cpu_present_map); if (cpu >= info->configured) smp_cpu_state[logical_cpu] = CPU_STATE_STANDBY; @@ -846,6 +848,7 @@ void __init smp_prepare_boot_cpu(void) S390_lowcore.percpu_offset = __per_cpu_offset[0]; current_set[0] = current; smp_cpu_state[0] = CPU_STATE_CONFIGURED; + smp_cpu_polarization[0] = POLARIZATION_UNKNWN; spin_lock_init(&(&__get_cpu_var(s390_idle))->lock); } @@ -897,15 +900,19 @@ static ssize_t cpu_configure_store(struct sys_device *dev, const char *buf, case 0: if (smp_cpu_state[cpu] == CPU_STATE_CONFIGURED) { rc = sclp_cpu_deconfigure(__cpu_logical_map[cpu]); - if (!rc) + if (!rc) { smp_cpu_state[cpu] = CPU_STATE_STANDBY; + smp_cpu_polarization[cpu] = POLARIZATION_UNKNWN; + } } break; case 1: if (smp_cpu_state[cpu] == CPU_STATE_STANDBY) { rc = sclp_cpu_configure(__cpu_logical_map[cpu]); - if (!rc) + if (!rc) { smp_cpu_state[cpu] = CPU_STATE_CONFIGURED; + smp_cpu_polarization[cpu] = POLARIZATION_UNKNWN; + } } break; default: @@ -919,6 +926,34 @@ out: static SYSDEV_ATTR(configure, 0644, cpu_configure_show, cpu_configure_store); #endif /* CONFIG_HOTPLUG_CPU */ +static ssize_t cpu_polarization_show(struct sys_device *dev, char *buf) +{ + int cpu = dev->id; + ssize_t count; + + mutex_lock(&smp_cpu_state_mutex); + switch (smp_cpu_polarization[cpu]) { + case POLARIZATION_HRZ: + count = sprintf(buf, "horizontal\n"); + break; + case POLARIZATION_VL: + count = sprintf(buf, "vertical:low\n"); + break; + case POLARIZATION_VM: + count = sprintf(buf, "vertical:medium\n"); + break; + case POLARIZATION_VH: + count = sprintf(buf, "vertical:high\n"); + break; + default: + count = sprintf(buf, "unknown\n"); + break; + } + mutex_unlock(&smp_cpu_state_mutex); + return count; +} +static SYSDEV_ATTR(polarization, 0444, cpu_polarization_show, NULL); + static ssize_t show_cpu_address(struct sys_device *dev, char *buf) { return sprintf(buf, "%d\n", __cpu_logical_map[dev->id]); @@ -931,6 +966,7 @@ static struct attribute *cpu_common_attrs[] = { &attr_configure.attr, #endif &attr_address.attr, + &attr_polarization.attr, NULL, }; @@ -1075,11 +1111,48 @@ static ssize_t __ref rescan_store(struct sys_device *dev, out: put_online_cpus(); mutex_unlock(&smp_cpu_state_mutex); + if (!cpus_empty(newcpus)) + topology_schedule_update(); return rc ? rc : count; } static SYSDEV_ATTR(rescan, 0200, NULL, rescan_store); #endif /* CONFIG_HOTPLUG_CPU */ +static ssize_t dispatching_show(struct sys_device *dev, char *buf) +{ + ssize_t count; + + mutex_lock(&smp_cpu_state_mutex); + count = sprintf(buf, "%d\n", cpu_management); + mutex_unlock(&smp_cpu_state_mutex); + return count; +} + +static ssize_t dispatching_store(struct sys_device *dev, const char *buf, + size_t count) +{ + int val, rc; + char delim; + + if (sscanf(buf, "%d %c", &val, &delim) != 1) + return -EINVAL; + if (val != 0 && val != 1) + return -EINVAL; + rc = 0; + mutex_lock(&smp_cpu_state_mutex); + get_online_cpus(); + if (cpu_management == val) + goto out; + rc = topology_set_cpu_management(val); + if (!rc) + cpu_management = val; +out: + put_online_cpus(); + mutex_unlock(&smp_cpu_state_mutex); + return rc ? rc : count; +} +static SYSDEV_ATTR(dispatching, 0644, dispatching_show, dispatching_store); + static int __init topology_init(void) { int cpu; @@ -1093,6 +1166,10 @@ static int __init topology_init(void) if (rc) return rc; #endif + rc = sysfs_create_file(&cpu_sysdev_class.kset.kobj, + &attr_dispatching.attr); + if (rc) + return rc; for_each_present_cpu(cpu) { rc = smp_add_present_cpu(cpu); if (rc) diff --git a/arch/s390/kernel/sys_s390.c b/arch/s390/kernel/sys_s390.c index fefee99..988d0d6 100644 --- a/arch/s390/kernel/sys_s390.c +++ b/arch/s390/kernel/sys_s390.c @@ -29,8 +29,8 @@ #include <linux/personality.h> #include <linux/unistd.h> #include <linux/ipc.h> - #include <asm/uaccess.h> +#include "entry.h" /* * sys_pipe() is the normal C calling standard for creating diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index cb232c1..7aec676 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -30,7 +30,7 @@ #include <linux/timex.h> #include <linux/notifier.h> #include <linux/clocksource.h> - +#include <linux/clockchips.h> #include <asm/uaccess.h> #include <asm/delay.h> #include <asm/s390_ext.h> @@ -39,6 +39,7 @@ #include <asm/irq_regs.h> #include <asm/timer.h> #include <asm/etr.h> +#include <asm/cio.h> /* change this if you have some constant time drift */ #define USECS_PER_JIFFY ((unsigned long) 1000000/HZ) @@ -57,16 +58,16 @@ static ext_int_info_t ext_int_info_cc; static ext_int_info_t ext_int_etr_cc; -static u64 init_timer_cc; static u64 jiffies_timer_cc; -static u64 xtime_cc; + +static DEFINE_PER_CPU(struct clock_event_device, comparators); /* * Scheduler clock - returns current time in nanosec units. */ unsigned long long sched_clock(void) { - return ((get_clock() - jiffies_timer_cc) * 125) >> 9; + return ((get_clock_xt() - jiffies_timer_cc) * 125) >> 9; } /* @@ -95,162 +96,40 @@ void tod_to_timeval(__u64 todval, struct timespec *xtime) #define s390_do_profile() do { ; } while(0) #endif /* CONFIG_PROFILING */ -/* - * Advance the per cpu tick counter up to the time given with the - * "time" argument. The per cpu update consists of accounting - * the virtual cpu time, calling update_process_times and calling - * the profiling hook. If xtime is before time it is advanced as well. - */ -void account_ticks(u64 time) +void clock_comparator_work(void) { - __u32 ticks; - __u64 tmp; - - /* Calculate how many ticks have passed. */ - if (time < S390_lowcore.jiffy_timer) - return; - tmp = time - S390_lowcore.jiffy_timer; - if (tmp >= 2*CLK_TICKS_PER_JIFFY) { /* more than two ticks ? */ - ticks = __div(tmp, CLK_TICKS_PER_JIFFY) + 1; - S390_lowcore.jiffy_timer += - CLK_TICKS_PER_JIFFY * (__u64) ticks; - } else if (tmp >= CLK_TICKS_PER_JIFFY) { - ticks = 2; - S390_lowcore.jiffy_timer += 2*CLK_TICKS_PER_JIFFY; - } else { - ticks = 1; - S390_lowcore.jiffy_timer += CLK_TICKS_PER_JIFFY; - } - -#ifdef CONFIG_SMP - /* - * Do not rely on the boot cpu to do the calls to do_timer. - * Spread it over all cpus instead. - */ - write_seqlock(&xtime_lock); - if (S390_lowcore.jiffy_timer > xtime_cc) { - __u32 xticks; - tmp = S390_lowcore.jiffy_timer - xtime_cc; - if (tmp >= 2*CLK_TICKS_PER_JIFFY) { - xticks = __div(tmp, CLK_TICKS_PER_JIFFY); - xtime_cc += (__u64) xticks * CLK_TICKS_PER_JIFFY; - } else { - xticks = 1; - xtime_cc += CLK_TICKS_PER_JIFFY; - } - do_timer(xticks); - } - write_sequnlock(&xtime_lock); -#else - do_timer(ticks); -#endif - - while (ticks--) - update_process_times(user_mode(get_irq_regs())); + struct clock_event_device *cd; + S390_lowcore.clock_comparator = -1ULL; + set_clock_comparator(S390_lowcore.clock_comparator); + cd = &__get_cpu_var(comparators); + cd->event_handler(cd); s390_do_profile(); } -#ifdef CONFIG_NO_IDLE_HZ - -#ifdef CONFIG_NO_IDLE_HZ_INIT -int sysctl_hz_timer = 0; -#else -int sysctl_hz_timer = 1; -#endif - -/* - * Stop the HZ tick on the current CPU. - * Only cpu_idle may call this function. - */ -static void stop_hz_timer(void) -{ - unsigned long flags; - unsigned long seq, next; - __u64 timer, todval; - int cpu = smp_processor_id(); - - if (sysctl_hz_timer != 0) - return; - - cpu_set(cpu, nohz_cpu_mask); - - /* - * Leave the clock comparator set up for the next timer - * tick if either rcu or a softirq is pending. - */ - if (rcu_needs_cpu(cpu) || local_softirq_pending()) { - cpu_clear(cpu, nohz_cpu_mask); - return; - } - - /* - * This cpu is going really idle. Set up the clock comparator - * for the next event. - */ - next = next_timer_interrupt(); - do { - seq = read_seqbegin_irqsave(&xtime_lock, flags); - timer = ((__u64) next) - ((__u64) jiffies) + jiffies_64; - } while (read_seqretry_irqrestore(&xtime_lock, seq, flags)); - 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; - } - set_clock_comparator(todval); -} - /* - * Start the HZ tick on the current CPU. - * Only cpu_idle may call this function. + * Fixup the clock comparator. */ -static void start_hz_timer(void) +static void fixup_clock_comparator(unsigned long long delta) { - if (!cpu_isset(smp_processor_id(), nohz_cpu_mask)) + /* If nobody is waiting there's nothing to fix. */ + if (S390_lowcore.clock_comparator == -1ULL) return; - account_ticks(get_clock()); - set_clock_comparator(S390_lowcore.jiffy_timer + CPU_DEVIATION); - cpu_clear(smp_processor_id(), nohz_cpu_mask); -} - -static int nohz_idle_notify(struct notifier_block *self, - unsigned long action, void *hcpu) -{ - switch (action) { - case S390_CPU_IDLE: - stop_hz_timer(); - break; - case S390_CPU_NOT_IDLE: - start_hz_timer(); - break; - } - return NOTIFY_OK; + S390_lowcore.clock_comparator += delta; + set_clock_comparator(S390_lowcore.clock_comparator); } -static struct notifier_block nohz_idle_nb = { - .notifier_call = nohz_idle_notify, -}; - -static void __init nohz_init(void) +static int s390_next_event(unsigned long delta, + struct clock_event_device *evt) { - if (register_idle_notifier(&nohz_idle_nb)) - panic("Couldn't register idle notifier"); + S390_lowcore.clock_comparator = get_clock() + delta; + set_clock_comparator(S390_lowcore.clock_comparator); + return 0; } -#endif - -/* - * Set up per cpu jiffy timer and set the clock comparator. - */ -static void setup_jiffy_timer(void) +static void s390_set_mode(enum clock_event_mode mode, + struct clock_event_device *evt) { - /* Set up clock comparator to next jiffy. */ - S390_lowcore.jiffy_timer = - jiffies_timer_cc + (jiffies_64 + 1) * CLK_TICKS_PER_JIFFY; - set_clock_comparator(S390_lowcore.jiffy_timer + CPU_DEVIATION); } /* @@ -259,7 +138,26 @@ static void setup_jiffy_timer(void) */ void init_cpu_timer(void) { - setup_jiffy_timer(); + struct clock_event_device *cd; + int cpu; + + S390_lowcore.clock_comparator = -1ULL; + set_clock_comparator(S390_lowcore.clock_comparator); + + cpu = smp_processor_id(); + cd = &per_cpu(comparators, cpu); + cd->name = "comparator"; + cd->features = CLOCK_EVT_FEAT_ONESHOT; + cd->mult = 16777; + cd->shift = 12; + cd->min_delta_ns = 1; + cd->max_delta_ns = LONG_MAX; + cd->rating = 400; + cd->cpumask = cpumask_of_cpu(cpu); + cd->set_next_event = s390_next_event; + cd->set_mode = s390_set_mode; + + clockevents_register_device(cd); /* Enable clock comparator timer interrupt. */ __ctl_set_bit(0,11); @@ -270,8 +168,6 @@ void init_cpu_timer(void) static void clock_comparator_interrupt(__u16 code) { - /* set clock comparator for next tick */ - set_clock_comparator(S390_lowcore.jiffy_timer + CPU_DEVIATION); } static void etr_reset(void); @@ -316,8 +212,9 @@ static struct clocksource clocksource_tod = { */ void __init time_init(void) { + u64 init_timer_cc; + init_timer_cc = reset_tod_clock(); - xtime_cc = init_timer_cc + CLK_TICKS_PER_JIFFY; jiffies_timer_cc = init_timer_cc - jiffies_64 * CLK_TICKS_PER_JIFFY; /* set xtime */ @@ -342,10 +239,6 @@ void __init time_init(void) /* Enable TOD clock interrupts on the boot cpu. */ init_cpu_timer(); -#ifdef CONFIG_NO_IDLE_HZ - nohz_init(); -#endif - #ifdef CONFIG_VIRT_TIMER vtime_init(); #endif @@ -699,53 +592,49 @@ static int etr_aib_follows(struct etr_aib *a1, struct etr_aib *a2, int p) } /* - * The time is "clock". xtime is what we think the time is. + * The time is "clock". old is what we think the time is. * Adjust the value by a multiple of jiffies and add the delta to ntp. * "delay" is an approximation how long the synchronization took. If * the time correction is positive, then "delay" is subtracted from * the time difference and only the remaining part is passed to ntp. */ -static void etr_adjust_time(unsigned long long clock, unsigned long long delay) +static unsigned long long etr_adjust_time(unsigned long long old, + unsigned long long clock, + unsigned long long delay) { unsigned long long delta, ticks; struct timex adjust; - /* - * We don't have to take the xtime lock because the cpu - * executing etr_adjust_time is running disabled in - * tasklet context and all other cpus are looping in - * etr_sync_cpu_start. - */ - if (clock > xtime_cc) { + if (clock > old) { /* It is later than we thought. */ - delta = ticks = clock - xtime_cc; + delta = ticks = clock - old; delta = ticks = (delta < delay) ? 0 : delta - delay; delta -= do_div(ticks, CLK_TICKS_PER_JIFFY); - init_timer_cc = init_timer_cc + delta; - jiffies_timer_cc = jiffies_timer_cc + delta; - xtime_cc = xtime_cc + delta; adjust.offset = ticks * (1000000 / HZ); } else { /* It is earlier than we thought. */ - delta = ticks = xtime_cc - clock; + delta = ticks = old - clock; delta -= do_div(ticks, CLK_TICKS_PER_JIFFY); - init_timer_cc = init_timer_cc - delta; - jiffies_timer_cc = jiffies_timer_cc - delta; - xtime_cc = xtime_cc - delta; + delta = -delta; adjust.offset = -ticks * (1000000 / HZ); } + jiffies_timer_cc += delta; if (adjust.offset != 0) { printk(KERN_NOTICE "etr: time adjusted by %li micro-seconds\n", adjust.offset); adjust.modes = ADJ_OFFSET_SINGLESHOT; do_adjtimex(&adjust); } + return delta; } +static struct { + int in_sync; + unsigned long long fixup_cc; +} etr_sync; + static void etr_sync_cpu_start(void *dummy) { - int *in_sync = dummy; - etr_enable_sync_clock(); /* * This looks like a busy wait loop but it isn't. etr_sync_cpus @@ -753,7 +642,7 @@ static void etr_sync_cpu_start(void *dummy) * __udelay will stop the cpu on an enabled wait psw until the * TOD is running again. */ - while (*in_sync == 0) { + while (etr_sync.in_sync == 0) { __udelay(1); /* * A different cpu changes *in_sync. Therefore use @@ -761,14 +650,14 @@ static void etr_sync_cpu_start(void *dummy) */ barrier(); } - if (*in_sync != 1) + if (etr_sync.in_sync != 1) /* Didn't work. Clear per-cpu in sync bit again. */ etr_disable_sync_clock(NULL); /* * This round of TOD syncing is done. Set the clock comparator * to the next tick and let the processor continue. */ - setup_jiffy_timer(); + fixup_clock_comparator(etr_sync.fixup_cc); } static void etr_sync_cpu_end(void *dummy) @@ -783,8 +672,8 @@ static void etr_sync_cpu_end(void *dummy) static int etr_sync_clock(struct etr_aib *aib, int port) { struct etr_aib *sync_port; - unsigned long long clock, delay; - int in_sync, follows; + unsigned long long clock, old_clock, delay, delta; + int follows; int rc; /* Check if the current aib is adjacent to the sync port aib. */ @@ -799,9 +688,9 @@ static int etr_sync_clock(struct etr_aib *aib, int port) * successfully synced the clock. smp_call_function will * return after all other cpus are in etr_sync_cpu_start. */ - in_sync = 0; + memset(&etr_sync, 0, sizeof(etr_sync)); preempt_disable(); - smp_call_function(etr_sync_cpu_start,&in_sync,0,0); + smp_call_function(etr_sync_cpu_start, NULL, 0, 0); local_irq_disable(); etr_enable_sync_clock(); @@ -809,6 +698,7 @@ static int etr_sync_clock(struct etr_aib *aib, int port) __ctl_set_bit(14, 21); __ctl_set_bit(0, 29); clock = ((unsigned long long) (aib->edf2.etv + 1)) << 32; + old_clock = get_clock(); if (set_clock(clock) == 0) { __udelay(1); /* Wait for the clock to start. */ __ctl_clear_bit(0, 29); @@ -817,16 +707,17 @@ static int etr_sync_clock(struct etr_aib *aib, int port) /* Adjust Linux timing variables. */ delay = (unsigned long long) (aib->edf2.etv - sync_port->edf2.etv) << 32; - etr_adjust_time(clock, delay); - setup_jiffy_timer(); + delta = etr_adjust_time(old_clock, clock, delay); + etr_sync.fixup_cc = delta; + fixup_clock_comparator(delta); /* Verify that the clock is properly set. */ if (!etr_aib_follows(sync_port, aib, port)) { /* Didn't work. */ etr_disable_sync_clock(NULL); - in_sync = -EAGAIN; + etr_sync.in_sync = -EAGAIN; rc = -EAGAIN; } else { - in_sync = 1; + etr_sync.in_sync = 1; rc = 0; } } else { @@ -834,7 +725,7 @@ static int etr_sync_clock(struct etr_aib *aib, int port) __ctl_clear_bit(0, 29); __ctl_clear_bit(14, 21); etr_disable_sync_clock(NULL); - in_sync = -EAGAIN; + etr_sync.in_sync = -EAGAIN; rc = -EAGAIN; } local_irq_enable(); diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c new file mode 100644 index 0000000..12b39b3 --- /dev/null +++ b/arch/s390/kernel/topology.c @@ -0,0 +1,314 @@ +/* + * Copyright IBM Corp. 2007 + * Author(s): Heiko Carstens <heiko.carstens@de.ibm.com> + */ + +#include <linux/kernel.h> +#include <linux/mm.h> +#include <linux/init.h> +#include <linux/device.h> +#include <linux/bootmem.h> +#include <linux/sched.h> +#include <linux/workqueue.h> +#include <linux/cpu.h> +#include <linux/smp.h> +#include <asm/delay.h> +#include <asm/s390_ext.h> +#include <asm/sysinfo.h> + +#define CPU_BITS 64 +#define NR_MAG 6 + +#define PTF_HORIZONTAL (0UL) +#define PTF_VERTICAL (1UL) +#define PTF_CHECK (2UL) + +struct tl_cpu { + unsigned char reserved0[4]; + unsigned char :6; + unsigned char pp:2; + unsigned char reserved1; + unsigned short origin; + unsigned long mask[CPU_BITS / BITS_PER_LONG]; +}; + +struct tl_container { + unsigned char reserved[8]; +}; + +union tl_entry { + unsigned char nl; + struct tl_cpu cpu; + struct tl_container container; +}; + +struct tl_info { + unsigned char reserved0[2]; + unsigned short length; + unsigned char mag[NR_MAG]; + unsigned char reserved1; + unsigned char mnest; + unsigned char reserved2[4]; + union tl_entry tle[0]; +}; + +struct core_info { + struct core_info *next; + cpumask_t mask; +}; + +static void topology_work_fn(struct work_struct *work); +static struct tl_info *tl_info; +static struct core_info core_info; +static int machine_has_topology; +static int machine_has_topology_irq; +static struct timer_list topology_timer; +static void set_topology_timer(void); +static DECLARE_WORK(topology_work, topology_work_fn); + +cpumask_t cpu_coregroup_map(unsigned int cpu) +{ + struct core_info *core = &core_info; + cpumask_t mask; + + cpus_clear(mask); + if (!machine_has_topology) + return cpu_present_map; + mutex_lock(&smp_cpu_state_mutex); + while (core) { + if (cpu_isset(cpu, core->mask)) { + mask = core->mask; + break; + } + core = core->next; + } + mutex_unlock(&smp_cpu_state_mutex); + if (cpus_empty(mask)) + mask = cpumask_of_cpu(cpu); + return mask; +} + +static void add_cpus_to_core(struct tl_cpu *tl_cpu, struct core_info *core) +{ + unsigned int cpu; + + for (cpu = find_first_bit(&tl_cpu->mask[0], CPU_BITS); + cpu < CPU_BITS; + cpu = find_next_bit(&tl_cpu->mask[0], CPU_BITS, cpu + 1)) + { + unsigned int rcpu, lcpu; + + rcpu = CPU_BITS - 1 - cpu + tl_cpu->origin; + for_each_present_cpu(lcpu) { + if (__cpu_logical_map[lcpu] == rcpu) { + cpu_set(lcpu, core->mask); + smp_cpu_polarization[lcpu] = tl_cpu->pp; + } + } + } +} + +static void clear_cores(void) +{ + struct core_info *core = &core_info; + + while (core) { + cpus_clear(core->mask); + core = core->next; + } +} + +static union tl_entry *next_tle(union tl_entry *tle) +{ + if (tle->nl) + return (union tl_entry *)((struct tl_container *)tle + 1); + else + return (union tl_entry *)((struct tl_cpu *)tle + 1); +} + +static void tl_to_cores(struct tl_info *info) +{ + union tl_entry *tle, *end; + struct core_info *core = &core_info; + + mutex_lock(&smp_cpu_state_mutex); + clear_cores(); + tle = info->tle; + end = (union tl_entry *)((unsigned long)info + info->length); + while (tle < end) { + switch (tle->nl) { + case 5: + case 4: + case 3: + case 2: + break; + case 1: + core = core->next; + break; + case 0: + add_cpus_to_core(&tle->cpu, core); + break; + default: + clear_cores(); + machine_has_topology = 0; + return; + } + tle = next_tle(tle); + } + mutex_unlock(&smp_cpu_state_mutex); +} + +static void topology_update_polarization_simple(void) +{ + int cpu; + + mutex_lock(&smp_cpu_state_mutex); + for_each_present_cpu(cpu) + smp_cpu_polarization[cpu] = POLARIZATION_HRZ; + mutex_unlock(&smp_cpu_state_mutex); +} + +static int ptf(unsigned long fc) +{ + int rc; + + asm volatile( + " .insn rre,0xb9a20000,%1,%1\n" + " ipm %0\n" + " srl %0,28\n" + : "=d" (rc) + : "d" (fc) : "cc"); + return rc; +} + +int topology_set_cpu_management(int fc) +{ + int cpu; + int rc; + + if (!machine_has_topology) + return -EOPNOTSUPP; + if (fc) + rc = ptf(PTF_VERTICAL); + else + rc = ptf(PTF_HORIZONTAL); + if (rc) + return -EBUSY; + for_each_present_cpu(cpu) + smp_cpu_polarization[cpu] = POLARIZATION_UNKNWN; + return rc; +} + +void arch_update_cpu_topology(void) +{ + struct tl_info *info = tl_info; + struct sys_device *sysdev; + int cpu; + + if (!machine_has_topology) { + topology_update_polarization_simple(); + return; + } + stsi(info, 15, 1, 2); + tl_to_cores(info); + for_each_online_cpu(cpu) { + sysdev = get_cpu_sysdev(cpu); + kobject_uevent(&sysdev->kobj, KOBJ_CHANGE); + } +} + +static void topology_work_fn(struct work_struct *work) +{ + arch_reinit_sched_domains(); +} + +void topology_schedule_update(void) +{ + schedule_work(&topology_work); +} + +static void topology_timer_fn(unsigned long ignored) +{ + if (ptf(PTF_CHECK)) + topology_schedule_update(); + set_topology_timer(); +} + +static void set_topology_timer(void) +{ + topology_timer.function = topology_timer_fn; + topology_timer.data = 0; + topology_timer.expires = jiffies + 60 * HZ; + add_timer(&topology_timer); +} + +static void topology_interrupt(__u16 code) +{ + schedule_work(&topology_work); +} + +static int __init init_topology_update(void) +{ + int rc; + + if (!machine_has_topology) { + topology_update_polarization_simple(); + return 0; + } + init_timer_deferrable(&topology_timer); + if (machine_has_topology_irq) { + rc = register_external_interrupt(0x2005, topology_interrupt); + if (rc) + return rc; + ctl_set_bit(0, 8); + } + else + set_topology_timer(); + return 0; +} +__initcall(init_topology_update); + +void __init s390_init_cpu_topology(void) +{ + unsigned long long facility_bits; + struct tl_info *info; + struct core_info *core; + int nr_cores; + int i; + + if (stfle(&facility_bits, 1) <= 0) + return; + if (!(facility_bits & (1ULL << 52)) || !(facility_bits & (1ULL << 61))) + return; + machine_has_topology = 1; + + if (facility_bits & (1ULL << 51)) + machine_has_topology_irq = 1; + + tl_info = alloc_bootmem_pages(PAGE_SIZE); + if (!tl_info) + goto error; + info = tl_info; + stsi(info, 15, 1, 2); + + nr_cores = info->mag[NR_MAG - 2]; + for (i = 0; i < info->mnest - 2; i++) + nr_cores *= info->mag[NR_MAG - 3 - i]; + + printk(KERN_INFO "CPU topology:"); + for (i = 0; i < NR_MAG; i++) + printk(" %d", info->mag[i]); + printk(" / %d\n", info->mnest); + + core = &core_info; + for (i = 0; i < nr_cores; i++) { + core->next = alloc_bootmem(sizeof(struct core_info)); + core = core->next; + if (!core) + goto error; + } + return; +error: + machine_has_topology = 0; + machine_has_topology_irq = 0; +} diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c index 60f728a..57b607b 100644 --- a/arch/s390/kernel/traps.c +++ b/arch/s390/kernel/traps.c @@ -42,11 +42,8 @@ #include <asm/s390_ext.h> #include <asm/lowcore.h> #include <asm/debug.h> +#include "entry.h" -/* Called from entry.S only */ -extern void handle_per_exception(struct pt_regs *regs); - -typedef void pgm_check_handler_t(struct pt_regs *, long); pgm_check_handler_t *pgm_check_table[128]; #ifdef CONFIG_SYSCTL @@ -59,7 +56,6 @@ int sysctl_userprocess_debug = 0; extern pgm_check_handler_t do_protection_exception; extern pgm_check_handler_t do_dat_exception; -extern pgm_check_handler_t do_monitor_call; extern pgm_check_handler_t do_asce_exception; #define stack_pointer ({ void **sp; asm("la %0,0(15)" : "=&d" (sp)); sp; }) @@ -138,7 +134,6 @@ void show_trace(struct task_struct *task, unsigned long *stack) else __show_trace(sp, S390_lowcore.thread_info, S390_lowcore.thread_info + THREAD_SIZE); - printk("\n"); if (!task) task = current; debug_show_held_locks(task); @@ -166,6 +161,15 @@ void show_stack(struct task_struct *task, unsigned long *sp) show_trace(task, sp); } +#ifdef CONFIG_64BIT +void show_last_breaking_event(struct pt_regs *regs) +{ + printk("Last Breaking-Event-Address:\n"); + printk(" [<%016lx>] ", regs->args[0] & PSW_ADDR_INSN); + print_symbol("%s\n", regs->args[0] & PSW_ADDR_INSN); +} +#endif + /* * The architecture-independent dump_stack generator */ @@ -739,6 +743,5 @@ void __init trap_init(void) pgm_check_table[0x15] = &operand_exception; pgm_check_table[0x1C] = &space_switch_exception; pgm_check_table[0x1D] = &hfp_sqrt_exception; - pgm_check_table[0x40] = &do_monitor_call; pfault_irq_init(); } diff --git a/arch/s390/lib/delay.c b/arch/s390/lib/delay.c index 70f2a86..eae21a8 100644 --- a/arch/s390/lib/delay.c +++ b/arch/s390/lib/delay.c @@ -34,7 +34,7 @@ void __delay(unsigned long loops) */ void __udelay(unsigned long usecs) { - u64 end, time, jiffy_timer = 0; + u64 end, time, old_cc = 0; unsigned long flags, cr0, mask, dummy; int irq_context; @@ -43,8 +43,8 @@ void __udelay(unsigned long usecs) local_bh_disable(); local_irq_save(flags); if (raw_irqs_disabled_flags(flags)) { - jiffy_timer = S390_lowcore.jiffy_timer; - S390_lowcore.jiffy_timer = -1ULL - (4096 << 12); + old_cc = S390_lowcore.clock_comparator; + S390_lowcore.clock_comparator = -1ULL; __ctl_store(cr0, 0, 0); dummy = (cr0 & 0xffff00e0) | 0x00000800; __ctl_load(dummy , 0, 0); @@ -55,8 +55,8 @@ void __udelay(unsigned long usecs) end = get_clock() + ((u64) usecs << 12); do { - time = end < S390_lowcore.jiffy_timer ? - end : S390_lowcore.jiffy_timer; + time = end < S390_lowcore.clock_comparator ? + end : S390_lowcore.clock_comparator; set_clock_comparator(time); trace_hardirqs_on(); __load_psw_mask(mask); @@ -65,10 +65,10 @@ void __udelay(unsigned long usecs) if (raw_irqs_disabled_flags(flags)) { __ctl_load(cr0, 0, 0); - S390_lowcore.jiffy_timer = jiffy_timer; + S390_lowcore.clock_comparator = old_cc; } if (!irq_context) _local_bh_enable(); - set_clock_comparator(S390_lowcore.jiffy_timer); + set_clock_comparator(S390_lowcore.clock_comparator); local_irq_restore(flags); } diff --git a/arch/s390/lib/uaccess_pt.c b/arch/s390/lib/uaccess_pt.c index 5efdfe9..d66215b 100644 --- a/arch/s390/lib/uaccess_pt.c +++ b/arch/s390/lib/uaccess_pt.c @@ -302,6 +302,10 @@ static size_t copy_in_user_pt(size_t n, void __user *to, pte_t *pte_from, *pte_to; int write_user; + if (segment_eq(get_fs(), KERNEL_DS)) { + memcpy((void __force *) to, (void __force *) from, n); + return 0; + } done = 0; retry: spin_lock(&mm->page_table_lock); @@ -361,18 +365,10 @@ fault: : "0" (-EFAULT), "d" (oparg), "a" (uaddr), \ "m" (*uaddr) : "cc" ); -int futex_atomic_op_pt(int op, int __user *uaddr, int oparg, int *old) +static int __futex_atomic_op_pt(int op, int __user *uaddr, int oparg, int *old) { int oldval = 0, newval, ret; - spin_lock(¤t->mm->page_table_lock); - uaddr = (int __user *) __dat_user_addr((unsigned long) uaddr); - if (!uaddr) { - spin_unlock(¤t->mm->page_table_lock); - return -EFAULT; - } - get_page(virt_to_page(uaddr)); - spin_unlock(¤t->mm->page_table_lock); switch (op) { case FUTEX_OP_SET: __futex_atomic_op("lr %2,%5\n", @@ -397,17 +393,17 @@ int futex_atomic_op_pt(int op, int __user *uaddr, int oparg, int *old) default: ret = -ENOSYS; } - put_page(virt_to_page(uaddr)); - *old = oldval; + if (ret == 0) + *old = oldval; return ret; } -int futex_atomic_cmpxchg_pt(int __user *uaddr, int oldval, int newval) +int futex_atomic_op_pt(int op, int __user *uaddr, int oparg, int *old) { int ret; - if (!current->mm) - return -EFAULT; + if (segment_eq(get_fs(), KERNEL_DS)) + return __futex_atomic_op_pt(op, uaddr, oparg, old); spin_lock(¤t->mm->page_table_lock); uaddr = (int __user *) __dat_user_addr((unsigned long) uaddr); if (!uaddr) { @@ -416,13 +412,40 @@ int futex_atomic_cmpxchg_pt(int __user *uaddr, int oldval, int newval) } get_page(virt_to_page(uaddr)); spin_unlock(¤t->mm->page_table_lock); - asm volatile(" cs %1,%4,0(%5)\n" - "0: lr %0,%1\n" - "1:\n" - EX_TABLE(0b,1b) + ret = __futex_atomic_op_pt(op, uaddr, oparg, old); + put_page(virt_to_page(uaddr)); + return ret; +} + +static int __futex_atomic_cmpxchg_pt(int __user *uaddr, int oldval, int newval) +{ + int ret; + + asm volatile("0: cs %1,%4,0(%5)\n" + "1: lr %0,%1\n" + "2:\n" + EX_TABLE(0b,2b) EX_TABLE(1b,2b) : "=d" (ret), "+d" (oldval), "=m" (*uaddr) : "0" (-EFAULT), "d" (newval), "a" (uaddr), "m" (*uaddr) : "cc", "memory" ); + return ret; +} + +int futex_atomic_cmpxchg_pt(int __user *uaddr, int oldval, int newval) +{ + int ret; + + if (segment_eq(get_fs(), KERNEL_DS)) + return __futex_atomic_cmpxchg_pt(uaddr, oldval, newval); + spin_lock(¤t->mm->page_table_lock); + uaddr = (int __user *) __dat_user_addr((unsigned long) uaddr); + if (!uaddr) { + spin_unlock(¤t->mm->page_table_lock); + return -EFAULT; + } + get_page(virt_to_page(uaddr)); + spin_unlock(¤t->mm->page_table_lock); + ret = __futex_atomic_cmpxchg_pt(uaddr, oldval, newval); put_page(virt_to_page(uaddr)); return ret; } diff --git a/arch/s390/mm/extmem.c b/arch/s390/mm/extmem.c index 880b0eb..ed2af0a 100644 --- a/arch/s390/mm/extmem.c +++ b/arch/s390/mm/extmem.c @@ -289,22 +289,8 @@ __segment_load (char *name, int do_nonshared, unsigned long *addr, unsigned long rc = add_shared_memory(seg->start_addr, seg->end - seg->start_addr + 1); - switch (rc) { - case 0: - break; - case -ENOSPC: - PRINT_WARN("segment_load: not loading segment %s - overlaps " - "storage/segment\n", name); - goto out_free; - case -ERANGE: - PRINT_WARN("segment_load: not loading segment %s - exceeds " - "kernel mapping range\n", name); - goto out_free; - default: - PRINT_WARN("segment_load: not loading segment %s (rc: %d)\n", - name, rc); + if (rc) goto out_free; - } seg->res = kzalloc(sizeof(struct resource), GFP_KERNEL); if (seg->res == NULL) { @@ -582,8 +568,59 @@ out: mutex_unlock(&dcss_lock); } +/* + * print appropriate error message for segment_load()/segment_type() + * return code + */ +void segment_warning(int rc, char *seg_name) +{ + switch (rc) { + case -ENOENT: + PRINT_WARN("cannot load/query segment %s, " + "does not exist\n", seg_name); + break; + case -ENOSYS: + PRINT_WARN("cannot load/query segment %s, " + "not running on VM\n", seg_name); + break; + case -EIO: + PRINT_WARN("cannot load/query segment %s, " + "hardware error\n", seg_name); + break; + case -ENOTSUPP: + PRINT_WARN("cannot load/query segment %s, " + "is a multi-part segment\n", seg_name); + break; + case -ENOSPC: + PRINT_WARN("cannot load/query segment %s, " + "overlaps with storage\n", seg_name); + break; + case -EBUSY: + PRINT_WARN("cannot load/query segment %s, " + "overlaps with already loaded dcss\n", seg_name); + break; + case -EPERM: + PRINT_WARN("cannot load/query segment %s, " + "already loaded in incompatible mode\n", seg_name); + break; + case -ENOMEM: + PRINT_WARN("cannot load/query segment %s, " + "out of memory\n", seg_name); + break; + case -ERANGE: + PRINT_WARN("cannot load/query segment %s, " + "exceeds kernel mapping range\n", seg_name); + break; + default: + PRINT_WARN("cannot load/query segment %s, " + "return value %i\n", seg_name, rc); + break; + } +} + EXPORT_SYMBOL(segment_load); EXPORT_SYMBOL(segment_unload); EXPORT_SYMBOL(segment_save); EXPORT_SYMBOL(segment_type); EXPORT_SYMBOL(segment_modify_shared); +EXPORT_SYMBOL(segment_warning); diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c index ed13d42..2650f46 100644 --- a/arch/s390/mm/fault.c +++ b/arch/s390/mm/fault.c @@ -28,11 +28,11 @@ #include <linux/hardirq.h> #include <linux/kprobes.h> #include <linux/uaccess.h> - #include <asm/system.h> #include <asm/pgtable.h> #include <asm/s390_ext.h> #include <asm/mmu_context.h> +#include "../kernel/entry.h" #ifndef CONFIG_64BIT #define __FAIL_ADDR_MASK 0x7ffff000 @@ -50,8 +50,6 @@ extern int sysctl_userprocess_debug; #endif -extern void die(const char *,struct pt_regs *,long); - #ifdef CONFIG_KPROBES static inline int notify_page_fault(struct pt_regs *regs, long err) { @@ -245,11 +243,6 @@ static void do_sigbus(struct pt_regs *regs, unsigned long error_code, } #ifdef CONFIG_S390_EXEC_PROTECT -extern long sys_sigreturn(struct pt_regs *regs); -extern long sys_rt_sigreturn(struct pt_regs *regs); -extern long sys32_sigreturn(struct pt_regs *regs); -extern long sys32_rt_sigreturn(struct pt_regs *regs); - static int signal_return(struct mm_struct *mm, struct pt_regs *regs, unsigned long address, unsigned long error_code) { @@ -270,15 +263,15 @@ static int signal_return(struct mm_struct *mm, struct pt_regs *regs, #ifdef CONFIG_COMPAT compat = test_tsk_thread_flag(current, TIF_31BIT); if (compat && instruction == 0x0a77) - sys32_sigreturn(regs); + sys32_sigreturn(); else if (compat && instruction == 0x0aad) - sys32_rt_sigreturn(regs); + sys32_rt_sigreturn(); else #endif if (instruction == 0x0a77) - sys_sigreturn(regs); + sys_sigreturn(); else if (instruction == 0x0aad) - sys_rt_sigreturn(regs); + sys_rt_sigreturn(); else { current->thread.prot_addr = address; current->thread.trap_no = error_code; @@ -424,7 +417,7 @@ no_context: } void __kprobes do_protection_exception(struct pt_regs *regs, - unsigned long error_code) + long error_code) { /* Protection exception is supressing, decrement psw address. */ regs->psw.addr -= (error_code >> 16); @@ -440,7 +433,7 @@ void __kprobes do_protection_exception(struct pt_regs *regs, do_exception(regs, 4, 1); } -void __kprobes do_dat_exception(struct pt_regs *regs, unsigned long error_code) +void __kprobes do_dat_exception(struct pt_regs *regs, long error_code) { do_exception(regs, error_code & 0xff, 0); } diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c index 8053245..202c952 100644 --- a/arch/s390/mm/init.c +++ b/arch/s390/mm/init.c @@ -50,7 +50,6 @@ void show_mem(void) printk("Mem-info:\n"); show_free_areas(); - printk("Free swap: %6ldkB\n", nr_swap_pages << (PAGE_SHIFT - 10)); i = max_mapnr; while (i-- > 0) { if (!pfn_valid(i)) |