diff options
Diffstat (limited to 'arch/s390')
34 files changed, 1584 insertions, 1028 deletions
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 1d03508..5dc8f80 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -146,6 +146,7 @@ config MATHEMU config COMPAT bool "Kernel support for 31 bit emulation" depends on 64BIT + select COMPAT_BINFMT_ELF help Select this option if you want to enable your system kernel to handle system-calls from ELF binaries for 31 bit ESA. This option @@ -304,10 +305,18 @@ config ARCH_SPARSEMEM_ENABLE def_bool y select SPARSEMEM_VMEMMAP_ENABLE select SPARSEMEM_VMEMMAP + select SPARSEMEM_STATIC if !64BIT config ARCH_SPARSEMEM_DEFAULT def_bool y +config ARCH_SELECT_MEMORY_MODEL + def_bool y + +config ARCH_ENABLE_MEMORY_HOTPLUG + def_bool y + depends on SPARSEMEM + source "mm/Kconfig" comment "I/O subsystem configuration" @@ -340,6 +349,22 @@ config QDIO_DEBUG If unsure, say N. +config CHSC_SCH + tristate "Support for CHSC subchannels" + help + This driver allows usage of CHSC subchannels. A CHSC subchannel + is usually present on LPAR only. + The driver creates a device /dev/chsc, which may be used to + obtain I/O configuration information about the machine and + to issue asynchronous chsc commands (DANGEROUS). + You will usually only want to use this interface on a special + LPAR designated for system management. + + To compile this driver as a module, choose M here: the + module will be called chsc_sch. + + If unsure, say N. + comment "Misc" config IPL diff --git a/arch/s390/appldata/appldata.h b/arch/s390/appldata/appldata.h index db3ae850..17a2636 100644 --- a/arch/s390/appldata/appldata.h +++ b/arch/s390/appldata/appldata.h @@ -3,13 +3,11 @@ * * Definitions and interface for Linux - z/VM Monitor Stream. * - * Copyright (C) 2003,2006 IBM Corporation, IBM Deutschland Entwicklung GmbH. + * Copyright IBM Corp. 2003, 2008 * * Author: Gerald Schaefer <gerald.schaefer@de.ibm.com> */ -//#define APPLDATA_DEBUG /* Debug messages on/off */ - #define APPLDATA_MAX_REC_SIZE 4024 /* Maximum size of the */ /* data buffer */ #define APPLDATA_MAX_PROCS 100 @@ -32,12 +30,6 @@ #define P_ERROR(x...) printk(KERN_ERR MY_PRINT_NAME " error: " x) #define P_WARNING(x...) printk(KERN_WARNING MY_PRINT_NAME " status: " x) -#ifdef APPLDATA_DEBUG -#define P_DEBUG(x...) printk(KERN_DEBUG MY_PRINT_NAME " debug: " x) -#else -#define P_DEBUG(x...) do {} while (0) -#endif - struct appldata_ops { struct list_head list; struct ctl_table_header *sysctl_header; diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c index 655d525..9cb3d92 100644 --- a/arch/s390/appldata/appldata_base.c +++ b/arch/s390/appldata/appldata_base.c @@ -5,7 +5,7 @@ * Exports appldata_register_ops() and appldata_unregister_ops() for the * data gathering modules. * - * Copyright (C) 2003,2006 IBM Corporation, IBM Deutschland Entwicklung GmbH. + * Copyright IBM Corp. 2003, 2008 * * Author: Gerald Schaefer <gerald.schaefer@de.ibm.com> */ @@ -108,9 +108,6 @@ static LIST_HEAD(appldata_ops_list); */ static void appldata_timer_function(unsigned long data) { - P_DEBUG(" -= Timer =-\n"); - P_DEBUG("CPU: %i, expire_count: %i\n", smp_processor_id(), - atomic_read(&appldata_expire_count)); if (atomic_dec_and_test(&appldata_expire_count)) { atomic_set(&appldata_expire_count, num_online_cpus()); queue_work(appldata_wq, (struct work_struct *) data); @@ -128,18 +125,17 @@ static void appldata_work_fn(struct work_struct *work) struct appldata_ops *ops; int i; - P_DEBUG(" -= Work Queue =-\n"); i = 0; + get_online_cpus(); spin_lock(&appldata_ops_lock); list_for_each(lh, &appldata_ops_list) { ops = list_entry(lh, struct appldata_ops, list); - P_DEBUG("list_for_each loop: %i) active = %u, name = %s\n", - ++i, ops->active, ops->name); if (ops->active == 1) { ops->callback(ops->data); } } spin_unlock(&appldata_ops_lock); + put_online_cpus(); } /* @@ -210,7 +206,6 @@ __appldata_vtimer_setup(int cmd) 0, 1); } appldata_timer_active = 1; - P_INFO("Monitoring timer started.\n"); break; case APPLDATA_DEL_TIMER: for_each_online_cpu(i) @@ -219,7 +214,6 @@ __appldata_vtimer_setup(int cmd) break; appldata_timer_active = 0; atomic_set(&appldata_expire_count, num_online_cpus()); - P_INFO("Monitoring timer stopped.\n"); break; case APPLDATA_MOD_TIMER: per_cpu_interval = (u64) (appldata_interval*1000 / @@ -266,12 +260,14 @@ appldata_timer_handler(ctl_table *ctl, int write, struct file *filp, len = *lenp; if (copy_from_user(buf, buffer, len > sizeof(buf) ? sizeof(buf) : len)) return -EFAULT; + get_online_cpus(); spin_lock(&appldata_timer_lock); if (buf[0] == '1') __appldata_vtimer_setup(APPLDATA_ADD_TIMER); else if (buf[0] == '0') __appldata_vtimer_setup(APPLDATA_DEL_TIMER); spin_unlock(&appldata_timer_lock); + put_online_cpus(); out: *lenp = len; *ppos += len; @@ -309,18 +305,15 @@ appldata_interval_handler(ctl_table *ctl, int write, struct file *filp, } interval = 0; sscanf(buf, "%i", &interval); - if (interval <= 0) { - P_ERROR("Timer CPU interval has to be > 0!\n"); + if (interval <= 0) return -EINVAL; - } + get_online_cpus(); spin_lock(&appldata_timer_lock); appldata_interval = interval; __appldata_vtimer_setup(APPLDATA_MOD_TIMER); spin_unlock(&appldata_timer_lock); - - P_INFO("Monitoring CPU interval set to %u milliseconds.\n", - interval); + put_online_cpus(); out: *lenp = len; *ppos += len; @@ -400,23 +393,16 @@ appldata_generic_handler(ctl_table *ctl, int write, struct file *filp, P_ERROR("START DIAG 0xDC for %s failed, " "return code: %d\n", ops->name, rc); module_put(ops->owner); - } else { - P_INFO("Monitoring %s data enabled, " - "DIAG 0xDC started.\n", ops->name); + } else ops->active = 1; - } } else if ((buf[0] == '0') && (ops->active == 1)) { ops->active = 0; rc = appldata_diag(ops->record_nr, APPLDATA_STOP_REC, (unsigned long) ops->data, ops->size, ops->mod_lvl); - if (rc != 0) { + if (rc != 0) P_ERROR("STOP DIAG 0xDC for %s failed, " "return code: %d\n", ops->name, rc); - } else { - P_INFO("Monitoring %s data disabled, " - "DIAG 0xDC stopped.\n", ops->name); - } module_put(ops->owner); } spin_unlock(&appldata_ops_lock); @@ -462,7 +448,6 @@ int appldata_register_ops(struct appldata_ops *ops) ops->sysctl_header = register_sysctl_table(ops->ctl_table); if (!ops->sysctl_header) goto out; - P_INFO("%s-ops registered!\n", ops->name); return 0; out: spin_lock(&appldata_ops_lock); @@ -484,7 +469,6 @@ void appldata_unregister_ops(struct appldata_ops *ops) spin_unlock(&appldata_ops_lock); unregister_sysctl_table(ops->sysctl_header); kfree(ops->ctl_table); - P_INFO("%s-ops unregistered!\n", ops->name); } /********************** module-ops management <END> **************************/ @@ -547,24 +531,19 @@ static int __init appldata_init(void) { int i; - P_DEBUG("sizeof(parameter_list) = %lu\n", - sizeof(struct appldata_parameter_list)); - appldata_wq = create_singlethread_workqueue("appldata"); - if (!appldata_wq) { - P_ERROR("Could not create work queue\n"); + if (!appldata_wq) return -ENOMEM; - } + get_online_cpus(); for_each_online_cpu(i) appldata_online_cpu(i); + put_online_cpus(); /* Register cpu hotplug notifier */ register_hotcpu_notifier(&appldata_nb); appldata_sysctl_header = register_sysctl_table(appldata_dir_table); - - P_DEBUG("Base interface initialized.\n"); return 0; } @@ -576,7 +555,9 @@ EXPORT_SYMBOL_GPL(appldata_register_ops); EXPORT_SYMBOL_GPL(appldata_unregister_ops); EXPORT_SYMBOL_GPL(appldata_diag); +#ifdef CONFIG_SWAP EXPORT_SYMBOL_GPL(si_swapinfo); +#endif EXPORT_SYMBOL_GPL(nr_threads); EXPORT_SYMBOL_GPL(nr_running); EXPORT_SYMBOL_GPL(nr_iowait); diff --git a/arch/s390/appldata/appldata_mem.c b/arch/s390/appldata/appldata_mem.c index 51181cc..3ed56b7 100644 --- a/arch/s390/appldata/appldata_mem.c +++ b/arch/s390/appldata/appldata_mem.c @@ -14,14 +14,13 @@ #include <linux/slab.h> #include <linux/errno.h> #include <linux/kernel_stat.h> -#include <asm/io.h> #include <linux/pagemap.h> #include <linux/swap.h> +#include <asm/io.h> #include "appldata.h" -#define MY_PRINT_NAME "appldata_mem" /* for debug messages, etc. */ #define P2K(x) ((x) << (PAGE_SHIFT - 10)) /* Converts #Pages to KB */ /* @@ -70,30 +69,6 @@ static struct appldata_mem_data { } __attribute__((packed)) appldata_mem_data; -static inline void appldata_debug_print(struct appldata_mem_data *mem_data) -{ - P_DEBUG("--- MEM - RECORD ---\n"); - P_DEBUG("pgpgin = %8lu KB\n", mem_data->pgpgin); - P_DEBUG("pgpgout = %8lu KB\n", mem_data->pgpgout); - P_DEBUG("pswpin = %8lu Pages\n", mem_data->pswpin); - P_DEBUG("pswpout = %8lu Pages\n", mem_data->pswpout); - P_DEBUG("pgalloc = %8lu \n", mem_data->pgalloc); - P_DEBUG("pgfault = %8lu \n", mem_data->pgfault); - P_DEBUG("pgmajfault = %8lu \n", mem_data->pgmajfault); - P_DEBUG("sharedram = %8lu KB\n", mem_data->sharedram); - P_DEBUG("totalram = %8lu KB\n", mem_data->totalram); - P_DEBUG("freeram = %8lu KB\n", mem_data->freeram); - P_DEBUG("totalhigh = %8lu KB\n", mem_data->totalhigh); - P_DEBUG("freehigh = %8lu KB\n", mem_data->freehigh); - P_DEBUG("bufferram = %8lu KB\n", mem_data->bufferram); - P_DEBUG("cached = %8lu KB\n", mem_data->cached); - P_DEBUG("totalswap = %8lu KB\n", mem_data->totalswap); - P_DEBUG("freeswap = %8lu KB\n", mem_data->freeswap); - P_DEBUG("sync_count_1 = %u\n", mem_data->sync_count_1); - P_DEBUG("sync_count_2 = %u\n", mem_data->sync_count_2); - P_DEBUG("timestamp = %lX\n", mem_data->timestamp); -} - /* * appldata_get_mem_data() * @@ -140,9 +115,6 @@ static void appldata_get_mem_data(void *data) mem_data->timestamp = get_clock(); mem_data->sync_count_2++; -#ifdef APPLDATA_DEBUG - appldata_debug_print(mem_data); -#endif } @@ -164,17 +136,7 @@ static struct appldata_ops ops = { */ static int __init appldata_mem_init(void) { - int rc; - - P_DEBUG("sizeof(mem) = %lu\n", sizeof(struct appldata_mem_data)); - - rc = appldata_register_ops(&ops); - if (rc != 0) { - P_ERROR("Error registering ops, rc = %i\n", rc); - } else { - P_DEBUG("%s-ops registered!\n", ops.name); - } - return rc; + return appldata_register_ops(&ops); } /* @@ -185,7 +147,6 @@ static int __init appldata_mem_init(void) static void __exit appldata_mem_exit(void) { appldata_unregister_ops(&ops); - P_DEBUG("%s-ops unregistered!\n", ops.name); } diff --git a/arch/s390/appldata/appldata_net_sum.c b/arch/s390/appldata/appldata_net_sum.c index 4d83443..3b74655 100644 --- a/arch/s390/appldata/appldata_net_sum.c +++ b/arch/s390/appldata/appldata_net_sum.c @@ -21,9 +21,6 @@ #include "appldata.h" -#define MY_PRINT_NAME "appldata_net_sum" /* for debug messages, etc. */ - - /* * Network data * @@ -60,26 +57,6 @@ static struct appldata_net_sum_data { } __attribute__((packed)) appldata_net_sum_data; -static inline void appldata_print_debug(struct appldata_net_sum_data *net_data) -{ - P_DEBUG("--- NET - RECORD ---\n"); - - P_DEBUG("nr_interfaces = %u\n", net_data->nr_interfaces); - P_DEBUG("rx_packets = %8lu\n", net_data->rx_packets); - P_DEBUG("tx_packets = %8lu\n", net_data->tx_packets); - P_DEBUG("rx_bytes = %8lu\n", net_data->rx_bytes); - P_DEBUG("tx_bytes = %8lu\n", net_data->tx_bytes); - P_DEBUG("rx_errors = %8lu\n", net_data->rx_errors); - P_DEBUG("tx_errors = %8lu\n", net_data->tx_errors); - P_DEBUG("rx_dropped = %8lu\n", net_data->rx_dropped); - P_DEBUG("tx_dropped = %8lu\n", net_data->tx_dropped); - P_DEBUG("collisions = %8lu\n", net_data->collisions); - - P_DEBUG("sync_count_1 = %u\n", net_data->sync_count_1); - P_DEBUG("sync_count_2 = %u\n", net_data->sync_count_2); - P_DEBUG("timestamp = %lX\n", net_data->timestamp); -} - /* * appldata_get_net_sum_data() * @@ -135,9 +112,6 @@ static void appldata_get_net_sum_data(void *data) net_data->timestamp = get_clock(); net_data->sync_count_2++; -#ifdef APPLDATA_DEBUG - appldata_print_debug(net_data); -#endif } @@ -159,17 +133,7 @@ static struct appldata_ops ops = { */ static int __init appldata_net_init(void) { - int rc; - - P_DEBUG("sizeof(net) = %lu\n", sizeof(struct appldata_net_sum_data)); - - rc = appldata_register_ops(&ops); - if (rc != 0) { - P_ERROR("Error registering ops, rc = %i\n", rc); - } else { - P_DEBUG("%s-ops registered!\n", ops.name); - } - return rc; + return appldata_register_ops(&ops); } /* @@ -180,7 +144,6 @@ static int __init appldata_net_init(void) static void __exit appldata_net_exit(void) { appldata_unregister_ops(&ops); - P_DEBUG("%s-ops unregistered!\n", ops.name); } diff --git a/arch/s390/appldata/appldata_os.c b/arch/s390/appldata/appldata_os.c index 6b3eafe..eb44f9f 100644 --- a/arch/s390/appldata/appldata_os.c +++ b/arch/s390/appldata/appldata_os.c @@ -89,44 +89,6 @@ static struct appldata_ops ops = { }; -static inline void appldata_print_debug(struct appldata_os_data *os_data) -{ - int a0, a1, a2, i; - - P_DEBUG("--- OS - RECORD ---\n"); - P_DEBUG("nr_threads = %u\n", os_data->nr_threads); - P_DEBUG("nr_running = %u\n", os_data->nr_running); - P_DEBUG("nr_iowait = %u\n", os_data->nr_iowait); - P_DEBUG("avenrun(int) = %8x / %8x / %8x\n", os_data->avenrun[0], - os_data->avenrun[1], os_data->avenrun[2]); - a0 = os_data->avenrun[0]; - a1 = os_data->avenrun[1]; - a2 = os_data->avenrun[2]; - P_DEBUG("avenrun(float) = %d.%02d / %d.%02d / %d.%02d\n", - LOAD_INT(a0), LOAD_FRAC(a0), LOAD_INT(a1), LOAD_FRAC(a1), - LOAD_INT(a2), LOAD_FRAC(a2)); - - P_DEBUG("nr_cpus = %u\n", os_data->nr_cpus); - for (i = 0; i < os_data->nr_cpus; i++) { - P_DEBUG("cpu%u : user = %u, nice = %u, system = %u, " - "idle = %u, irq = %u, softirq = %u, iowait = %u, " - "steal = %u\n", - os_data->os_cpu[i].cpu_id, - os_data->os_cpu[i].per_cpu_user, - os_data->os_cpu[i].per_cpu_nice, - os_data->os_cpu[i].per_cpu_system, - os_data->os_cpu[i].per_cpu_idle, - os_data->os_cpu[i].per_cpu_irq, - os_data->os_cpu[i].per_cpu_softirq, - os_data->os_cpu[i].per_cpu_iowait, - os_data->os_cpu[i].per_cpu_steal); - } - - P_DEBUG("sync_count_1 = %u\n", os_data->sync_count_1); - P_DEBUG("sync_count_2 = %u\n", os_data->sync_count_2); - P_DEBUG("timestamp = %lX\n", os_data->timestamp); -} - /* * appldata_get_os_data() * @@ -180,13 +142,10 @@ static void appldata_get_os_data(void *data) APPLDATA_START_INTERVAL_REC, (unsigned long) ops.data, new_size, ops.mod_lvl); - if (rc != 0) { + if (rc != 0) P_ERROR("os: START NEW DIAG 0xDC failed, " "return code: %d, new size = %i\n", rc, new_size); - P_INFO("os: stopping old record now\n"); - } else - P_INFO("os: new record size = %i\n", new_size); rc = appldata_diag(APPLDATA_RECORD_OS_ID, APPLDATA_STOP_REC, @@ -204,9 +163,6 @@ static void appldata_get_os_data(void *data) } os_data->timestamp = get_clock(); os_data->sync_count_2++; -#ifdef APPLDATA_DEBUG - appldata_print_debug(os_data); -#endif } @@ -227,12 +183,9 @@ static int __init appldata_os_init(void) rc = -ENOMEM; goto out; } - P_DEBUG("max. sizeof(os) = %i, sizeof(os_cpu) = %lu\n", max_size, - sizeof(struct appldata_os_per_cpu)); appldata_os_data = kzalloc(max_size, GFP_DMA); if (appldata_os_data == NULL) { - P_ERROR("No memory for %s!\n", ops.name); rc = -ENOMEM; goto out; } @@ -240,17 +193,12 @@ static int __init appldata_os_init(void) appldata_os_data->per_cpu_size = sizeof(struct appldata_os_per_cpu); appldata_os_data->cpu_offset = offsetof(struct appldata_os_data, os_cpu); - P_DEBUG("cpu offset = %u\n", appldata_os_data->cpu_offset); ops.data = appldata_os_data; ops.callback = &appldata_get_os_data; rc = appldata_register_ops(&ops); - if (rc != 0) { - P_ERROR("Error registering ops, rc = %i\n", rc); + if (rc != 0) kfree(appldata_os_data); - } else { - P_DEBUG("%s-ops registered!\n", ops.name); - } out: return rc; } @@ -264,7 +212,6 @@ static void __exit appldata_os_exit(void) { appldata_unregister_ops(&ops); kfree(appldata_os_data); - P_DEBUG("%s-ops unregistered!\n", ops.name); } diff --git a/arch/s390/crypto/crypt_s390.h b/arch/s390/crypto/crypt_s390.h index 9992f95..0ef9829 100644 --- a/arch/s390/crypto/crypt_s390.h +++ b/arch/s390/crypto/crypt_s390.h @@ -296,6 +296,10 @@ static inline int crypt_s390_func_available(int func) unsigned char status[16]; int ret; + /* check if CPACF facility (bit 17) is available */ + if (!(stfl() & 1ULL << (31 - 17))) + return 0; + switch (func & CRYPT_S390_OP_MASK) { case CRYPT_S390_KM: ret = crypt_s390_km(KM_QUERY, &status, NULL, NULL, 0); diff --git a/arch/s390/crypto/prng.c b/arch/s390/crypto/prng.c index 0cfefdd..eca724d 100644 --- a/arch/s390/crypto/prng.c +++ b/arch/s390/crypto/prng.c @@ -6,6 +6,7 @@ #include <linux/fs.h> #include <linux/init.h> #include <linux/kernel.h> +#include <linux/smp_lock.h> #include <linux/miscdevice.h> #include <linux/module.h> #include <linux/moduleparam.h> @@ -48,6 +49,7 @@ static unsigned char parm_block[32] = { static int prng_open(struct inode *inode, struct file *file) { + cycle_kernel_lock(); return nonseekable_open(inode, file); } @@ -185,11 +187,8 @@ static int __init prng_init(void) prng_seed(16); ret = misc_register(&prng_dev); - if (ret) { - printk(KERN_WARNING - "Could not register misc device for PRNG.\n"); + if (ret) goto out_buf; - } return 0; out_buf: diff --git a/arch/s390/defconfig b/arch/s390/defconfig index aa341d0e..c5cdb97 100644 --- a/arch/s390/defconfig +++ b/arch/s390/defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.25 -# Wed Apr 30 11:07:45 2008 +# Linux kernel version: 2.6.26-rc4 +# Fri May 30 09:49:33 2008 # CONFIG_SCHED_MC=y CONFIG_MMU=y @@ -103,6 +103,7 @@ CONFIG_RT_MUTEXES=y # CONFIG_TINY_SHMEM is not set CONFIG_BASE_SMALL=0 CONFIG_MODULES=y +# CONFIG_MODULE_FORCE_LOAD is not set CONFIG_MODULE_UNLOAD=y # CONFIG_MODULE_FORCE_UNLOAD is not set CONFIG_MODVERSIONS=y @@ -173,6 +174,7 @@ CONFIG_PREEMPT=y # CONFIG_PREEMPT_RCU is not set CONFIG_ARCH_SPARSEMEM_ENABLE=y CONFIG_ARCH_SPARSEMEM_DEFAULT=y +CONFIG_ARCH_SELECT_MEMORY_MODEL=y CONFIG_SELECT_MEMORY_MODEL=y # CONFIG_FLATMEM_MANUAL is not set # CONFIG_DISCONTIGMEM_MANUAL is not set @@ -210,6 +212,7 @@ CONFIG_FORCE_MAX_ZONEORDER=9 CONFIG_PFAULT=y # CONFIG_SHARED_KERNEL is not set # CONFIG_CMM is not set +# CONFIG_PAGE_STATES is not set CONFIG_VIRT_TIMER=y CONFIG_VIRT_CPU_ACCOUNTING=y # CONFIG_APPLDATA_BASE is not set @@ -620,6 +623,7 @@ CONFIG_S390_VMUR=m # # CONFIG_MEMSTICK is not set # CONFIG_NEW_LEDS is not set +CONFIG_ACCESSIBILITY=y # # File systems @@ -754,11 +758,12 @@ CONFIG_FRAME_WARN=2048 CONFIG_MAGIC_SYSRQ=y # CONFIG_UNUSED_SYMBOLS is not set CONFIG_DEBUG_FS=y -CONFIG_HEADERS_CHECK=y +# CONFIG_HEADERS_CHECK is not set CONFIG_DEBUG_KERNEL=y # CONFIG_SCHED_DEBUG is not set # CONFIG_SCHEDSTATS is not set # CONFIG_TIMER_STATS is not set +# CONFIG_DEBUG_OBJECTS is not set # CONFIG_DEBUG_SLAB is not set CONFIG_DEBUG_PREEMPT=y # CONFIG_DEBUG_RT_MUTEXES is not set diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c index 4b010ff..7383781 100644 --- a/arch/s390/hypfs/inode.c +++ b/arch/s390/hypfs/inode.c @@ -150,33 +150,24 @@ static ssize_t hypfs_aio_read(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t offset) { char *data; - size_t len; + ssize_t ret; struct file *filp = iocb->ki_filp; /* XXX: temporary */ char __user *buf = iov[0].iov_base; size_t count = iov[0].iov_len; - if (nr_segs != 1) { - count = -EINVAL; - goto out; - } + if (nr_segs != 1) + return -EINVAL; data = filp->private_data; - len = strlen(data); - if (offset > len) { - count = 0; - goto out; - } - if (count > len - offset) - count = len - offset; - if (copy_to_user(buf, data + offset, count)) { - count = -EFAULT; - goto out; - } - iocb->ki_pos += count; + ret = simple_read_from_buffer(buf, count, &offset, data, strlen(data)); + if (ret <= 0) + return ret; + + iocb->ki_pos += ret; file_accessed(filp); -out: - return count; + + return ret; } static ssize_t hypfs_aio_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, loff_t offset) diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile index 6302f50..50f657e 100644 --- a/arch/s390/kernel/Makefile +++ b/arch/s390/kernel/Makefile @@ -7,9 +7,14 @@ # CFLAGS_smp.o := -Wno-nonnull +# +# Pass UTS_MACHINE for user_regset definition +# +CFLAGS_ptrace.o += -DUTS_MACHINE='"$(UTS_MACHINE)"' + 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 \ - 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 mem_detect.o obj-y += $(if $(CONFIG_64BIT),entry64.o,entry.o) obj-y += $(if $(CONFIG_64BIT),reipl64.o,reipl.o) @@ -23,7 +28,7 @@ obj-$(CONFIG_AUDIT) += audit.o compat-obj-$(CONFIG_AUDIT) += compat_audit.o obj-$(CONFIG_COMPAT) += compat_linux.o compat_signal.o \ compat_wrapper.o compat_exec_domain.o \ - binfmt_elf32.o $(compat-obj-y) + $(compat-obj-y) obj-$(CONFIG_VIRT_TIMER) += vtime.o obj-$(CONFIG_STACKTRACE) += stacktrace.o diff --git a/arch/s390/kernel/binfmt_elf32.c b/arch/s390/kernel/binfmt_elf32.c deleted file mode 100644 index 3e1c315..0000000 --- a/arch/s390/kernel/binfmt_elf32.c +++ /dev/null @@ -1,214 +0,0 @@ -/* - * Support for 32-bit Linux for S390 ELF binaries. - * - * Copyright (C) 2000 IBM Deutschland Entwicklung GmbH, IBM Corporation - * Author(s): Gerhard Tonn (ton@de.ibm.com) - * - * Heavily inspired by the 32-bit Sparc compat code which is - * Copyright (C) 1995, 1996, 1997, 1998 David S. Miller (davem@redhat.com) - * Copyright (C) 1995, 1996, 1997, 1998 Jakub Jelinek (jj@ultra.linux.cz) - */ - -#define __ASMS390_ELF_H - -#include <linux/time.h> - -/* - * These are used to set parameters in the core dumps. - */ -#define ELF_CLASS ELFCLASS32 -#define ELF_DATA ELFDATA2MSB -#define ELF_ARCH EM_S390 - -/* - * This is used to ensure we don't load something for the wrong architecture. - */ -#define elf_check_arch(x) \ - (((x)->e_machine == EM_S390 || (x)->e_machine == EM_S390_OLD) \ - && (x)->e_ident[EI_CLASS] == ELF_CLASS) - -/* ELF register definitions */ -#define NUM_GPRS 16 -#define NUM_FPRS 16 -#define NUM_ACRS 16 - -/* For SVR4/S390 the function pointer to be registered with `atexit` is - passed in R14. */ -#define ELF_PLAT_INIT(_r, load_addr) \ - do { \ - _r->gprs[14] = 0; \ - } while(0) - -#define USE_ELF_CORE_DUMP -#define ELF_EXEC_PAGESIZE 4096 - -/* This is the location that an ET_DYN program is loaded if exec'ed. Typical - use of this is to invoke "./ld.so someprog" to test out a new version of - the loader. We need to make sure that it is out of the way of the program - that it will "exec", and that there is sufficient room for the brk. */ - -#define ELF_ET_DYN_BASE (TASK_SIZE / 3 * 2) - -/* Wow, the "main" arch needs arch dependent functions too.. :) */ - -/* regs is struct pt_regs, pr_reg is elf_gregset_t (which is - now struct_user_regs, they are different) */ - -#define ELF_CORE_COPY_REGS(pr_reg, regs) dump_regs32(regs, &pr_reg); - -#define ELF_CORE_COPY_TASK_REGS(tsk, regs) dump_task_regs32(tsk, regs) - -#define ELF_CORE_COPY_FPREGS(tsk, fpregs) dump_task_fpu(tsk, fpregs) - -/* This yields a mask that user programs can use to figure out what - instruction set this CPU supports. */ - -#define ELF_HWCAP (0) - -/* This yields a string that ld.so will use to load implementation - specific libraries for optimization. This is more specific in - intent than poking at uname or /proc/cpuinfo. - - For the moment, we have only optimizations for the Intel generations, - but that could change... */ - -#define ELF_PLATFORM (NULL) - -#define SET_PERSONALITY(ex, ibcs2) \ -do { \ - if (ibcs2) \ - set_personality(PER_SVR4); \ - else if (current->personality != PER_LINUX32) \ - set_personality(PER_LINUX); \ - set_thread_flag(TIF_31BIT); \ -} while (0) - -#include "compat_linux.h" - -typedef _s390_fp_regs32 elf_fpregset_t; - -typedef struct -{ - - _psw_t32 psw; - __u32 gprs[__NUM_GPRS]; - __u32 acrs[__NUM_ACRS]; - __u32 orig_gpr2; -} s390_regs32; -typedef s390_regs32 elf_gregset_t; - -static inline int dump_regs32(struct pt_regs *ptregs, elf_gregset_t *regs) -{ - int i; - - memcpy(®s->psw.mask, &ptregs->psw.mask, 4); - memcpy(®s->psw.addr, (char *)&ptregs->psw.addr + 4, 4); - for (i = 0; i < NUM_GPRS; i++) - regs->gprs[i] = ptregs->gprs[i]; - save_access_regs(regs->acrs); - regs->orig_gpr2 = ptregs->orig_gpr2; - return 1; -} - -static inline int dump_task_regs32(struct task_struct *tsk, elf_gregset_t *regs) -{ - struct pt_regs *ptregs = task_pt_regs(tsk); - int i; - - memcpy(®s->psw.mask, &ptregs->psw.mask, 4); - memcpy(®s->psw.addr, (char *)&ptregs->psw.addr + 4, 4); - for (i = 0; i < NUM_GPRS; i++) - regs->gprs[i] = ptregs->gprs[i]; - memcpy(regs->acrs, tsk->thread.acrs, sizeof(regs->acrs)); - regs->orig_gpr2 = ptregs->orig_gpr2; - return 1; -} - -static inline int dump_task_fpu(struct task_struct *tsk, elf_fpregset_t *fpregs) -{ - if (tsk == current) - save_fp_regs((s390_fp_regs *) fpregs); - else - memcpy(fpregs, &tsk->thread.fp_regs, sizeof(elf_fpregset_t)); - return 1; -} - -#include <asm/processor.h> -#include <asm/pgalloc.h> -#include <linux/module.h> -#include <linux/elfcore.h> -#include <linux/binfmts.h> -#include <linux/compat.h> - -#define elf_prstatus elf_prstatus32 -struct elf_prstatus32 -{ - struct elf_siginfo pr_info; /* Info associated with signal */ - short pr_cursig; /* Current signal */ - u32 pr_sigpend; /* Set of pending signals */ - u32 pr_sighold; /* Set of held signals */ - pid_t pr_pid; - pid_t pr_ppid; - pid_t pr_pgrp; - pid_t pr_sid; - struct compat_timeval pr_utime; /* User time */ - struct compat_timeval pr_stime; /* System time */ - struct compat_timeval pr_cutime; /* Cumulative user time */ - struct compat_timeval pr_cstime; /* Cumulative system time */ - elf_gregset_t pr_reg; /* GP registers */ - int pr_fpvalid; /* True if math co-processor being used. */ -}; - -#define elf_prpsinfo elf_prpsinfo32 -struct elf_prpsinfo32 -{ - char pr_state; /* numeric process state */ - char pr_sname; /* char for pr_state */ - char pr_zomb; /* zombie */ - char pr_nice; /* nice val */ - u32 pr_flag; /* flags */ - u16 pr_uid; - u16 pr_gid; - pid_t pr_pid, pr_ppid, pr_pgrp, pr_sid; - /* Lots missing */ - char pr_fname[16]; /* filename of executable */ - char pr_psargs[ELF_PRARGSZ]; /* initial part of arg list */ -}; - -#include <linux/highuid.h> - -/* -#define init_elf_binfmt init_elf32_binfmt -*/ - -#undef start_thread -#define start_thread start_thread31 - -static inline void start_thread31(struct pt_regs *regs, unsigned long new_psw, - unsigned long new_stackp) -{ - set_fs(USER_DS); - regs->psw.mask = psw_user32_bits; - regs->psw.addr = new_psw; - regs->gprs[15] = new_stackp; - crst_table_downgrade(current->mm, 1UL << 31); -} - -MODULE_DESCRIPTION("Binary format loader for compatibility with 32bit Linux for S390 binaries," - " Copyright 2000 IBM Corporation"); -MODULE_AUTHOR("Gerhard Tonn <ton@de.ibm.com>"); - -#undef MODULE_DESCRIPTION -#undef MODULE_AUTHOR - -#undef cputime_to_timeval -#define cputime_to_timeval cputime_to_compat_timeval -static inline void -cputime_to_compat_timeval(const cputime_t cputime, struct compat_timeval *value) -{ - value->tv_usec = cputime % 1000000; - value->tv_sec = cputime / 1000000; -} - -#include "../../../fs/binfmt_elf.c" - diff --git a/arch/s390/kernel/compat_ptrace.h b/arch/s390/kernel/compat_ptrace.h index 419aef9..cde81fa 100644 --- a/arch/s390/kernel/compat_ptrace.h +++ b/arch/s390/kernel/compat_ptrace.h @@ -1,7 +1,7 @@ #ifndef _PTRACE32_H #define _PTRACE32_H -#include "compat_linux.h" /* needed for _psw_t32 */ +#include "compat_linux.h" /* needed for psw_compat_t */ typedef struct { __u32 cr[3]; @@ -38,7 +38,7 @@ typedef struct { struct user_regs_struct32 { - _psw_t32 psw; + psw_compat_t psw; u32 gprs[NUM_GPRS]; u32 acrs[NUM_ACRS]; u32 orig_gpr2; diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c index c93d129..d80fcd4 100644 --- a/arch/s390/kernel/debug.c +++ b/arch/s390/kernel/debug.c @@ -1079,7 +1079,6 @@ __init debug_init(void) s390dbf_sysctl_header = register_sysctl_table(s390dbf_dir_table); mutex_lock(&debug_mutex); debug_debugfs_root_entry = debugfs_create_dir(DEBUG_DIR_ROOT,NULL); - printk(KERN_INFO "debug: Initialization complete\n"); initialized = 1; mutex_unlock(&debug_mutex); @@ -1193,7 +1192,6 @@ debug_get_uint(char *buf) for(; isspace(*buf); buf++); rc = simple_strtoul(buf, &buf, 10); if(*buf){ - printk("debug: no integer specified!\n"); rc = -EINVAL; } return rc; @@ -1340,19 +1338,12 @@ static void debug_flush(debug_info_t* id, int area) memset(id->areas[i][j], 0, PAGE_SIZE); } } - printk(KERN_INFO "debug: %s: all areas flushed\n",id->name); } else if(area >= 0 && area < id->nr_areas) { id->active_entries[area] = 0; id->active_pages[area] = 0; for(i = 0; i < id->pages_per_area; i++) { memset(id->areas[area][i],0,PAGE_SIZE); } - printk(KERN_INFO "debug: %s: area %i has been flushed\n", - id->name, area); - } else { - printk(KERN_INFO - "debug: %s: area %i cannot be flushed (range: %i - %i)\n", - id->name, area, 0, id->nr_areas-1); } spin_unlock_irqrestore(&id->lock,flags); } diff --git a/arch/s390/kernel/dis.c b/arch/s390/kernel/dis.c index c14a336..d2f270c 100644 --- a/arch/s390/kernel/dis.c +++ b/arch/s390/kernel/dis.c @@ -208,7 +208,7 @@ static const unsigned char formats[][7] = { [INSTR_RRF_F0FF] = { 0xff, F_16,F_24,F_28,0,0,0 }, /* e.g. madbr */ [INSTR_RRF_FUFF] = { 0xff, F_24,F_16,F_28,U4_20,0,0 },/* e.g. didbr */ [INSTR_RRF_RURR] = { 0xff, R_24,R_28,R_16,U4_20,0,0 },/* e.g. .insn */ - [INSTR_RRF_R0RR] = { 0xff, R_24,R_28,R_16,0,0,0 }, /* e.g. idte */ + [INSTR_RRF_R0RR] = { 0xff, R_24,R_16,R_28,0,0,0 }, /* e.g. idte */ [INSTR_RRF_U0FF] = { 0xff, F_24,U4_16,F_28,0,0,0 }, /* e.g. fixr */ [INSTR_RRF_U0RF] = { 0xff, R_24,U4_16,F_28,0,0,0 }, /* e.g. cfebr */ [INSTR_RRF_M0RR] = { 0xff, R_24,R_28,M_16,0,0,0 }, /* e.g. sske */ diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c index d0e0968..2a2ca26 100644 --- a/arch/s390/kernel/early.c +++ b/arch/s390/kernel/early.c @@ -14,6 +14,7 @@ #include <linux/module.h> #include <linux/pfn.h> #include <linux/uaccess.h> +#include <asm/ebcdic.h> #include <asm/ipl.h> #include <asm/lowcore.h> #include <asm/processor.h> @@ -26,12 +27,40 @@ /* * Create a Kernel NSS if the SAVESYS= parameter is defined */ -#define DEFSYS_CMD_SIZE 96 +#define DEFSYS_CMD_SIZE 128 #define SAVESYS_CMD_SIZE 32 char kernel_nss_name[NSS_NAME_SIZE + 1]; +static void __init setup_boot_command_line(void); + + #ifdef CONFIG_SHARED_KERNEL +int __init savesys_ipl_nss(char *cmd, const int cmdlen); + +asm( + " .section .init.text,\"ax\",@progbits\n" + " .align 4\n" + " .type savesys_ipl_nss, @function\n" + "savesys_ipl_nss:\n" +#ifdef CONFIG_64BIT + " stmg 6,15,48(15)\n" + " lgr 14,3\n" + " sam31\n" + " diag 2,14,0x8\n" + " sam64\n" + " lgr 2,14\n" + " lmg 6,15,48(15)\n" +#else + " stm 6,15,24(15)\n" + " lr 14,3\n" + " diag 2,14,0x8\n" + " lr 2,14\n" + " lm 6,15,24(15)\n" +#endif + " br 14\n" + " .size savesys_ipl_nss, .-savesys_ipl_nss\n"); + static noinline __init void create_kernel_nss(void) { unsigned int i, stext_pfn, eshared_pfn, end_pfn, min_size; @@ -39,6 +68,7 @@ static noinline __init void create_kernel_nss(void) unsigned int sinitrd_pfn, einitrd_pfn; #endif int response; + size_t len; char *savesys_ptr; char upper_command_line[COMMAND_LINE_SIZE]; char defsys_cmd[DEFSYS_CMD_SIZE]; @@ -49,8 +79,8 @@ static noinline __init void create_kernel_nss(void) return; /* Convert COMMAND_LINE to upper case */ - for (i = 0; i < strlen(COMMAND_LINE); i++) - upper_command_line[i] = toupper(COMMAND_LINE[i]); + for (i = 0; i < strlen(boot_command_line); i++) + upper_command_line[i] = toupper(boot_command_line[i]); savesys_ptr = strstr(upper_command_line, "SAVESYS="); @@ -83,7 +113,8 @@ static noinline __init void create_kernel_nss(void) } #endif - sprintf(defsys_cmd, "%s EW MINSIZE=%.7iK", defsys_cmd, min_size); + sprintf(defsys_cmd, "%s EW MINSIZE=%.7iK PARMREGS=0-13", + defsys_cmd, min_size); sprintf(savesys_cmd, "SAVESYS %s \n IPL %s", kernel_nss_name, kernel_nss_name); @@ -94,13 +125,24 @@ static noinline __init void create_kernel_nss(void) return; } - __cpcmd(savesys_cmd, NULL, 0, &response); + len = strlen(savesys_cmd); + ASCEBC(savesys_cmd, len); + response = savesys_ipl_nss(savesys_cmd, len); - if (response != strlen(savesys_cmd)) { + /* On success: response is equal to the command size, + * max SAVESYS_CMD_SIZE + * On error: response contains the numeric portion of cp error message. + * for SAVESYS it will be >= 263 + */ + if (response > SAVESYS_CMD_SIZE) { kernel_nss_name[0] = '\0'; return; } + /* re-setup boot command line with new ipl vm parms */ + ipl_update_parameters(); + setup_boot_command_line(); + ipl_flags = IPL_NSS_VALID; } @@ -141,109 +183,11 @@ static noinline __init void detect_machine_type(void) if (cpuinfo->cpu_id.version == 0xff) machine_flags |= MACHINE_FLAG_VM; - /* Running on a P/390 ? */ - if (cpuinfo->cpu_id.machine == 0x7490) - machine_flags |= MACHINE_FLAG_P390; - /* Running under KVM ? */ if (cpuinfo->cpu_id.version == 0xfe) machine_flags |= MACHINE_FLAG_KVM; } -#ifdef CONFIG_64BIT -static noinline __init int memory_fast_detect(void) -{ - unsigned long val0 = 0; - unsigned long val1 = 0xc; - int ret = -ENOSYS; - - if (ipl_flags & IPL_NSS_VALID) - return -ENOSYS; - - asm volatile( - " diag %1,%2,0x260\n" - "0: lhi %0,0\n" - "1:\n" - EX_TABLE(0b,1b) - : "+d" (ret), "+d" (val0), "+d" (val1) : : "cc"); - - if (ret || val0 != val1) - return -ENOSYS; - - memory_chunk[0].size = val0 + 1; - return 0; -} -#else -static inline int memory_fast_detect(void) -{ - return -ENOSYS; -} -#endif - -static inline __init unsigned long __tprot(unsigned long addr) -{ - int cc = -1; - - asm volatile( - " tprot 0(%1),0\n" - "0: ipm %0\n" - " srl %0,28\n" - "1:\n" - EX_TABLE(0b,1b) - : "+d" (cc) : "a" (addr) : "cc"); - return (unsigned long)cc; -} - -/* Checking memory in 128KB increments. */ -#define CHUNK_INCR (1UL << 17) -#define ADDR2G (1UL << 31) - -static noinline __init void find_memory_chunks(unsigned long memsize) -{ - unsigned long addr = 0, old_addr = 0; - unsigned long old_cc = CHUNK_READ_WRITE; - unsigned long cc; - int chunk = 0; - - while (chunk < MEMORY_CHUNKS) { - cc = __tprot(addr); - while (cc == old_cc) { - addr += CHUNK_INCR; - if (memsize && addr >= memsize) - break; -#ifndef CONFIG_64BIT - if (addr == ADDR2G) - break; -#endif - cc = __tprot(addr); - } - - if (old_addr != addr && - (old_cc == CHUNK_READ_WRITE || old_cc == CHUNK_READ_ONLY)) { - memory_chunk[chunk].addr = old_addr; - memory_chunk[chunk].size = addr - old_addr; - memory_chunk[chunk].type = old_cc; - chunk++; - } - - old_addr = addr; - old_cc = cc; - -#ifndef CONFIG_64BIT - if (addr == ADDR2G) - break; -#endif - /* - * Finish memory detection at the first hole - * if storage size is unknown. - */ - if (cc == -1UL && !memsize) - break; - if (memsize && addr >= memsize) - break; - } -} - static __init void early_pgm_check_handler(void) { unsigned long addr; @@ -380,23 +324,61 @@ static __init void detect_machine_facilities(void) #endif } +static __init void rescue_initrd(void) +{ +#ifdef CONFIG_BLK_DEV_INITRD + /* + * Move the initrd right behind the bss section in case it starts + * within the bss section. So we don't overwrite it when the bss + * section gets cleared. + */ + if (!INITRD_START || !INITRD_SIZE) + return; + if (INITRD_START >= (unsigned long) __bss_stop) + return; + memmove(__bss_stop, (void *) INITRD_START, INITRD_SIZE); + INITRD_START = (unsigned long) __bss_stop; +#endif +} + +/* Set up boot command line */ +static void __init setup_boot_command_line(void) +{ + char *parm = NULL; + + /* copy arch command line */ + strlcpy(boot_command_line, COMMAND_LINE, ARCH_COMMAND_LINE_SIZE); + boot_command_line[ARCH_COMMAND_LINE_SIZE - 1] = 0; + + /* append IPL PARM data to the boot command line */ + if (MACHINE_IS_VM) { + parm = boot_command_line + strlen(boot_command_line); + *parm++ = ' '; + get_ipl_vmparm(parm); + if (parm[0] == '=') + memmove(boot_command_line, parm + 1, strlen(parm)); + } +} + + /* * Save ipl parameters, clear bss memory, initialize storage keys * and create a kernel NSS at startup if the SAVESYS= parm is defined */ void __init startup_init(void) { - unsigned long long memsize; - ipl_save_parameters(); + rescue_initrd(); clear_bss_section(); init_kernel_storage_key(); lockdep_init(); lockdep_off(); - detect_machine_type(); - create_kernel_nss(); sort_main_extable(); setup_lowcore_early(); + detect_machine_type(); + ipl_update_parameters(); + setup_boot_command_line(); + create_kernel_nss(); detect_mvpg(); detect_ieee(); detect_csp(); @@ -404,18 +386,7 @@ void __init startup_init(void) detect_diag44(); detect_machine_facilities(); setup_hpage(); - sclp_read_info_early(); sclp_facilities_detect(); - memsize = sclp_memory_detect(); -#ifndef CONFIG_64BIT - /* - * Can't deal with more than 2G in 31 bit addressing mode, so - * limit the value in order to avoid strange side effects. - */ - if (memsize > ADDR2G) - memsize = ADDR2G; -#endif - if (memory_fast_detect() < 0) - find_memory_chunks((unsigned long) memsize); + detect_memory_layout(memory_chunk); lockdep_on(); } diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c index 5325424..54b2779 100644 --- a/arch/s390/kernel/ipl.c +++ b/arch/s390/kernel/ipl.c @@ -14,6 +14,7 @@ #include <linux/delay.h> #include <linux/reboot.h> #include <linux/ctype.h> +#include <linux/fs.h> #include <asm/ipl.h> #include <asm/smp.h> #include <asm/setup.h> @@ -22,6 +23,7 @@ #include <asm/ebcdic.h> #include <asm/reset.h> #include <asm/sclp.h> +#include <asm/setup.h> #define IPL_PARM_BLOCK_VERSION 0 @@ -121,6 +123,7 @@ enum ipl_method { REIPL_METHOD_FCP_RO_VM, REIPL_METHOD_FCP_DUMP, REIPL_METHOD_NSS, + REIPL_METHOD_NSS_DIAG, REIPL_METHOD_DEFAULT, }; @@ -134,14 +137,15 @@ enum dump_method { static int diag308_set_works = 0; +static struct ipl_parameter_block ipl_block; + static int reipl_capabilities = IPL_TYPE_UNKNOWN; static enum ipl_type reipl_type = IPL_TYPE_UNKNOWN; static enum ipl_method reipl_method = REIPL_METHOD_DEFAULT; static struct ipl_parameter_block *reipl_block_fcp; static struct ipl_parameter_block *reipl_block_ccw; - -static char reipl_nss_name[NSS_NAME_SIZE + 1]; +static struct ipl_parameter_block *reipl_block_nss; static int dump_capabilities = DUMP_TYPE_NONE; static enum dump_type dump_type = DUMP_TYPE_NONE; @@ -263,6 +267,56 @@ static ssize_t ipl_type_show(struct kobject *kobj, struct kobj_attribute *attr, static struct kobj_attribute sys_ipl_type_attr = __ATTR_RO(ipl_type); +/* VM IPL PARM routines */ +static void reipl_get_ascii_vmparm(char *dest, + const struct ipl_parameter_block *ipb) +{ + int i; + int len = 0; + char has_lowercase = 0; + + if ((ipb->ipl_info.ccw.vm_flags & DIAG308_VM_FLAGS_VP_VALID) && + (ipb->ipl_info.ccw.vm_parm_len > 0)) { + + len = ipb->ipl_info.ccw.vm_parm_len; + memcpy(dest, ipb->ipl_info.ccw.vm_parm, len); + /* If at least one character is lowercase, we assume mixed + * case; otherwise we convert everything to lowercase. + */ + for (i = 0; i < len; i++) + if ((dest[i] > 0x80 && dest[i] < 0x8a) || /* a-i */ + (dest[i] > 0x90 && dest[i] < 0x9a) || /* j-r */ + (dest[i] > 0xa1 && dest[i] < 0xaa)) { /* s-z */ + has_lowercase = 1; + break; + } + if (!has_lowercase) + EBC_TOLOWER(dest, len); + EBCASC(dest, len); + } + dest[len] = 0; +} + +void get_ipl_vmparm(char *dest) +{ + if (diag308_set_works && (ipl_block.hdr.pbt == DIAG308_IPL_TYPE_CCW)) + reipl_get_ascii_vmparm(dest, &ipl_block); + else + dest[0] = 0; +} + +static ssize_t ipl_vm_parm_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + char parm[DIAG308_VMPARM_SIZE + 1] = {}; + + get_ipl_vmparm(parm); + return sprintf(page, "%s\n", parm); +} + +static struct kobj_attribute sys_ipl_vm_parm_attr = + __ATTR(parm, S_IRUGO, ipl_vm_parm_show, NULL); + static ssize_t sys_ipl_device_show(struct kobject *kobj, struct kobj_attribute *attr, char *page) { @@ -285,14 +339,8 @@ static struct kobj_attribute sys_ipl_device_attr = static ssize_t ipl_parameter_read(struct kobject *kobj, struct bin_attribute *attr, char *buf, loff_t off, size_t count) { - unsigned int size = IPL_PARMBLOCK_SIZE; - - if (off > size) - return 0; - if (off + count > size) - count = size - off; - memcpy(buf, (void *)IPL_PARMBLOCK_START + off, count); - return count; + return memory_read_from_buffer(buf, count, &off, IPL_PARMBLOCK_START, + IPL_PARMBLOCK_SIZE); } static struct bin_attribute ipl_parameter_attr = { @@ -310,12 +358,7 @@ static ssize_t ipl_scp_data_read(struct kobject *kobj, struct bin_attribute *att unsigned int size = IPL_PARMBLOCK_START->ipl_info.fcp.scp_data_len; void *scp_data = &IPL_PARMBLOCK_START->ipl_info.fcp.scp_data; - if (off > size) - return 0; - if (off + count > size) - count = size - off; - memcpy(buf, scp_data + off, count); - return count; + return memory_read_from_buffer(buf, count, &off, scp_data, size); } static struct bin_attribute ipl_scp_data_attr = { @@ -370,15 +413,27 @@ static ssize_t ipl_ccw_loadparm_show(struct kobject *kobj, static struct kobj_attribute sys_ipl_ccw_loadparm_attr = __ATTR(loadparm, 0444, ipl_ccw_loadparm_show, NULL); -static struct attribute *ipl_ccw_attrs[] = { +static struct attribute *ipl_ccw_attrs_vm[] = { &sys_ipl_type_attr.attr, &sys_ipl_device_attr.attr, &sys_ipl_ccw_loadparm_attr.attr, + &sys_ipl_vm_parm_attr.attr, NULL, }; -static struct attribute_group ipl_ccw_attr_group = { - .attrs = ipl_ccw_attrs, +static struct attribute *ipl_ccw_attrs_lpar[] = { + &sys_ipl_type_attr.attr, + &sys_ipl_device_attr.attr, + &sys_ipl_ccw_loadparm_attr.attr, + NULL, +}; + +static struct attribute_group ipl_ccw_attr_group_vm = { + .attrs = ipl_ccw_attrs_vm, +}; + +static struct attribute_group ipl_ccw_attr_group_lpar = { + .attrs = ipl_ccw_attrs_lpar }; /* NSS ipl device attributes */ @@ -388,6 +443,8 @@ DEFINE_IPL_ATTR_RO(ipl_nss, name, "%s\n", kernel_nss_name); static struct attribute *ipl_nss_attrs[] = { &sys_ipl_type_attr.attr, &sys_ipl_nss_name_attr.attr, + &sys_ipl_ccw_loadparm_attr.attr, + &sys_ipl_vm_parm_attr.attr, NULL, }; @@ -450,7 +507,12 @@ static int __init ipl_init(void) } switch (ipl_info.type) { case IPL_TYPE_CCW: - rc = sysfs_create_group(&ipl_kset->kobj, &ipl_ccw_attr_group); + if (MACHINE_IS_VM) + rc = sysfs_create_group(&ipl_kset->kobj, + &ipl_ccw_attr_group_vm); + else + rc = sysfs_create_group(&ipl_kset->kobj, + &ipl_ccw_attr_group_lpar); break; case IPL_TYPE_FCP: case IPL_TYPE_FCP_DUMP: @@ -481,6 +543,83 @@ static struct shutdown_action __refdata ipl_action = { * reipl shutdown action: Reboot Linux on shutdown. */ +/* VM IPL PARM attributes */ +static ssize_t reipl_generic_vmparm_show(struct ipl_parameter_block *ipb, + char *page) +{ + char vmparm[DIAG308_VMPARM_SIZE + 1] = {}; + + reipl_get_ascii_vmparm(vmparm, ipb); + return sprintf(page, "%s\n", vmparm); +} + +static ssize_t reipl_generic_vmparm_store(struct ipl_parameter_block *ipb, + size_t vmparm_max, + const char *buf, size_t len) +{ + int i, ip_len; + + /* ignore trailing newline */ + ip_len = len; + if ((len > 0) && (buf[len - 1] == '\n')) + ip_len--; + + if (ip_len > vmparm_max) + return -EINVAL; + + /* parm is used to store kernel options, check for common chars */ + for (i = 0; i < ip_len; i++) + if (!(isalnum(buf[i]) || isascii(buf[i]) || isprint(buf[i]))) + return -EINVAL; + + memset(ipb->ipl_info.ccw.vm_parm, 0, DIAG308_VMPARM_SIZE); + ipb->ipl_info.ccw.vm_parm_len = ip_len; + if (ip_len > 0) { + ipb->ipl_info.ccw.vm_flags |= DIAG308_VM_FLAGS_VP_VALID; + memcpy(ipb->ipl_info.ccw.vm_parm, buf, ip_len); + ASCEBC(ipb->ipl_info.ccw.vm_parm, ip_len); + } else { + ipb->ipl_info.ccw.vm_flags &= ~DIAG308_VM_FLAGS_VP_VALID; + } + + return len; +} + +/* NSS wrapper */ +static ssize_t reipl_nss_vmparm_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + return reipl_generic_vmparm_show(reipl_block_nss, page); +} + +static ssize_t reipl_nss_vmparm_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + return reipl_generic_vmparm_store(reipl_block_nss, 56, buf, len); +} + +/* CCW wrapper */ +static ssize_t reipl_ccw_vmparm_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + return reipl_generic_vmparm_show(reipl_block_ccw, page); +} + +static ssize_t reipl_ccw_vmparm_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + return reipl_generic_vmparm_store(reipl_block_ccw, 64, buf, len); +} + +static struct kobj_attribute sys_reipl_nss_vmparm_attr = + __ATTR(parm, S_IRUGO | S_IWUSR, reipl_nss_vmparm_show, + reipl_nss_vmparm_store); +static struct kobj_attribute sys_reipl_ccw_vmparm_attr = + __ATTR(parm, S_IRUGO | S_IWUSR, reipl_ccw_vmparm_show, + reipl_ccw_vmparm_store); + /* FCP reipl device attributes */ DEFINE_IPL_ATTR_RW(reipl_fcp, wwpn, "0x%016llx\n", "%016llx\n", @@ -513,27 +652,26 @@ static struct attribute_group reipl_fcp_attr_group = { DEFINE_IPL_ATTR_RW(reipl_ccw, device, "0.0.%04llx\n", "0.0.%llx\n", reipl_block_ccw->ipl_info.ccw.devno); -static void reipl_get_ascii_loadparm(char *loadparm) +static void reipl_get_ascii_loadparm(char *loadparm, + struct ipl_parameter_block *ibp) { - memcpy(loadparm, &reipl_block_ccw->ipl_info.ccw.load_param, - LOADPARM_LEN); + memcpy(loadparm, ibp->ipl_info.ccw.load_parm, LOADPARM_LEN); EBCASC(loadparm, LOADPARM_LEN); loadparm[LOADPARM_LEN] = 0; strstrip(loadparm); } -static ssize_t reipl_ccw_loadparm_show(struct kobject *kobj, - struct kobj_attribute *attr, char *page) +static ssize_t reipl_generic_loadparm_show(struct ipl_parameter_block *ipb, + char *page) { char buf[LOADPARM_LEN + 1]; - reipl_get_ascii_loadparm(buf); + reipl_get_ascii_loadparm(buf, ipb); return sprintf(page, "%s\n", buf); } -static ssize_t reipl_ccw_loadparm_store(struct kobject *kobj, - struct kobj_attribute *attr, - const char *buf, size_t len) +static ssize_t reipl_generic_loadparm_store(struct ipl_parameter_block *ipb, + const char *buf, size_t len) { int i, lp_len; @@ -552,35 +690,128 @@ static ssize_t reipl_ccw_loadparm_store(struct kobject *kobj, return -EINVAL; } /* initialize loadparm with blanks */ - memset(&reipl_block_ccw->ipl_info.ccw.load_param, ' ', LOADPARM_LEN); + memset(ipb->ipl_info.ccw.load_parm, ' ', LOADPARM_LEN); /* copy and convert to ebcdic */ - memcpy(&reipl_block_ccw->ipl_info.ccw.load_param, buf, lp_len); - ASCEBC(reipl_block_ccw->ipl_info.ccw.load_param, LOADPARM_LEN); + memcpy(ipb->ipl_info.ccw.load_parm, buf, lp_len); + ASCEBC(ipb->ipl_info.ccw.load_parm, LOADPARM_LEN); return len; } +/* NSS wrapper */ +static ssize_t reipl_nss_loadparm_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + return reipl_generic_loadparm_show(reipl_block_nss, page); +} + +static ssize_t reipl_nss_loadparm_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + return reipl_generic_loadparm_store(reipl_block_nss, buf, len); +} + +/* CCW wrapper */ +static ssize_t reipl_ccw_loadparm_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + return reipl_generic_loadparm_show(reipl_block_ccw, page); +} + +static ssize_t reipl_ccw_loadparm_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + return reipl_generic_loadparm_store(reipl_block_ccw, buf, len); +} + static struct kobj_attribute sys_reipl_ccw_loadparm_attr = - __ATTR(loadparm, 0644, reipl_ccw_loadparm_show, - reipl_ccw_loadparm_store); + __ATTR(loadparm, S_IRUGO | S_IWUSR, reipl_ccw_loadparm_show, + reipl_ccw_loadparm_store); -static struct attribute *reipl_ccw_attrs[] = { +static struct attribute *reipl_ccw_attrs_vm[] = { &sys_reipl_ccw_device_attr.attr, &sys_reipl_ccw_loadparm_attr.attr, + &sys_reipl_ccw_vmparm_attr.attr, NULL, }; -static struct attribute_group reipl_ccw_attr_group = { +static struct attribute *reipl_ccw_attrs_lpar[] = { + &sys_reipl_ccw_device_attr.attr, + &sys_reipl_ccw_loadparm_attr.attr, + NULL, +}; + +static struct attribute_group reipl_ccw_attr_group_vm = { + .name = IPL_CCW_STR, + .attrs = reipl_ccw_attrs_vm, +}; + +static struct attribute_group reipl_ccw_attr_group_lpar = { .name = IPL_CCW_STR, - .attrs = reipl_ccw_attrs, + .attrs = reipl_ccw_attrs_lpar, }; /* NSS reipl device attributes */ +static void reipl_get_ascii_nss_name(char *dst, + struct ipl_parameter_block *ipb) +{ + memcpy(dst, ipb->ipl_info.ccw.nss_name, NSS_NAME_SIZE); + EBCASC(dst, NSS_NAME_SIZE); + dst[NSS_NAME_SIZE] = 0; +} + +static ssize_t reipl_nss_name_show(struct kobject *kobj, + struct kobj_attribute *attr, char *page) +{ + char nss_name[NSS_NAME_SIZE + 1] = {}; -DEFINE_IPL_ATTR_STR_RW(reipl_nss, name, "%s\n", "%s\n", reipl_nss_name); + reipl_get_ascii_nss_name(nss_name, reipl_block_nss); + return sprintf(page, "%s\n", nss_name); +} + +static ssize_t reipl_nss_name_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t len) +{ + int nss_len; + + /* ignore trailing newline */ + nss_len = len; + if ((len > 0) && (buf[len - 1] == '\n')) + nss_len--; + + if (nss_len > NSS_NAME_SIZE) + return -EINVAL; + + memset(reipl_block_nss->ipl_info.ccw.nss_name, 0x40, NSS_NAME_SIZE); + if (nss_len > 0) { + reipl_block_nss->ipl_info.ccw.vm_flags |= + DIAG308_VM_FLAGS_NSS_VALID; + memcpy(reipl_block_nss->ipl_info.ccw.nss_name, buf, nss_len); + ASCEBC(reipl_block_nss->ipl_info.ccw.nss_name, nss_len); + EBC_TOUPPER(reipl_block_nss->ipl_info.ccw.nss_name, nss_len); + } else { + reipl_block_nss->ipl_info.ccw.vm_flags &= + ~DIAG308_VM_FLAGS_NSS_VALID; + } + + return len; +} + +static struct kobj_attribute sys_reipl_nss_name_attr = + __ATTR(name, S_IRUGO | S_IWUSR, reipl_nss_name_show, + reipl_nss_name_store); + +static struct kobj_attribute sys_reipl_nss_loadparm_attr = + __ATTR(loadparm, S_IRUGO | S_IWUSR, reipl_nss_loadparm_show, + reipl_nss_loadparm_store); static struct attribute *reipl_nss_attrs[] = { &sys_reipl_nss_name_attr.attr, + &sys_reipl_nss_loadparm_attr.attr, + &sys_reipl_nss_vmparm_attr.attr, NULL, }; @@ -617,7 +848,10 @@ static int reipl_set_type(enum ipl_type type) reipl_method = REIPL_METHOD_FCP_DUMP; break; case IPL_TYPE_NSS: - reipl_method = REIPL_METHOD_NSS; + if (diag308_set_works) + reipl_method = REIPL_METHOD_NSS_DIAG; + else + reipl_method = REIPL_METHOD_NSS; break; case IPL_TYPE_UNKNOWN: reipl_method = REIPL_METHOD_DEFAULT; @@ -655,11 +889,38 @@ static struct kobj_attribute reipl_type_attr = static struct kset *reipl_kset; +static void get_ipl_string(char *dst, struct ipl_parameter_block *ipb, + const enum ipl_method m) +{ + char loadparm[LOADPARM_LEN + 1] = {}; + char vmparm[DIAG308_VMPARM_SIZE + 1] = {}; + char nss_name[NSS_NAME_SIZE + 1] = {}; + size_t pos = 0; + + reipl_get_ascii_loadparm(loadparm, ipb); + reipl_get_ascii_nss_name(nss_name, ipb); + reipl_get_ascii_vmparm(vmparm, ipb); + + switch (m) { + case REIPL_METHOD_CCW_VM: + pos = sprintf(dst, "IPL %X CLEAR", ipb->ipl_info.ccw.devno); + break; + case REIPL_METHOD_NSS: + pos = sprintf(dst, "IPL %s", nss_name); + break; + default: + break; + } + if (strlen(loadparm) > 0) + pos += sprintf(dst + pos, " LOADPARM '%s'", loadparm); + if (strlen(vmparm) > 0) + sprintf(dst + pos, " PARM %s", vmparm); +} + static void reipl_run(struct shutdown_trigger *trigger) { struct ccw_dev_id devid; - static char buf[100]; - char loadparm[LOADPARM_LEN + 1]; + static char buf[128]; switch (reipl_method) { case REIPL_METHOD_CCW_CIO: @@ -668,13 +929,7 @@ static void reipl_run(struct shutdown_trigger *trigger) reipl_ccw_dev(&devid); break; case REIPL_METHOD_CCW_VM: - reipl_get_ascii_loadparm(loadparm); - if (strlen(loadparm) == 0) - sprintf(buf, "IPL %X CLEAR", - reipl_block_ccw->ipl_info.ccw.devno); - else - sprintf(buf, "IPL %X CLEAR LOADPARM '%s'", - reipl_block_ccw->ipl_info.ccw.devno, loadparm); + get_ipl_string(buf, reipl_block_ccw, REIPL_METHOD_CCW_VM); __cpcmd(buf, NULL, 0, NULL); break; case REIPL_METHOD_CCW_DIAG: @@ -691,8 +946,12 @@ static void reipl_run(struct shutdown_trigger *trigger) case REIPL_METHOD_FCP_RO_VM: __cpcmd("IPL", NULL, 0, NULL); break; + case REIPL_METHOD_NSS_DIAG: + diag308(DIAG308_SET, reipl_block_nss); + diag308(DIAG308_IPL, NULL); + break; case REIPL_METHOD_NSS: - sprintf(buf, "IPL %s", reipl_nss_name); + get_ipl_string(buf, reipl_block_nss, REIPL_METHOD_NSS); __cpcmd(buf, NULL, 0, NULL); break; case REIPL_METHOD_DEFAULT: @@ -707,16 +966,36 @@ static void reipl_run(struct shutdown_trigger *trigger) disabled_wait((unsigned long) __builtin_return_address(0)); } -static void __init reipl_probe(void) +static void reipl_block_ccw_init(struct ipl_parameter_block *ipb) { - void *buffer; + ipb->hdr.len = IPL_PARM_BLK_CCW_LEN; + ipb->hdr.version = IPL_PARM_BLOCK_VERSION; + ipb->hdr.blk0_len = IPL_PARM_BLK0_CCW_LEN; + ipb->hdr.pbt = DIAG308_IPL_TYPE_CCW; +} - buffer = (void *) get_zeroed_page(GFP_KERNEL); - if (!buffer) - return; - if (diag308(DIAG308_STORE, buffer) == DIAG308_RC_OK) - diag308_set_works = 1; - free_page((unsigned long)buffer); +static void reipl_block_ccw_fill_parms(struct ipl_parameter_block *ipb) +{ + /* LOADPARM */ + /* check if read scp info worked and set loadparm */ + if (sclp_ipl_info.is_valid) + memcpy(ipb->ipl_info.ccw.load_parm, + &sclp_ipl_info.loadparm, LOADPARM_LEN); + else + /* read scp info failed: set empty loadparm (EBCDIC blanks) */ + memset(ipb->ipl_info.ccw.load_parm, 0x40, LOADPARM_LEN); + ipb->hdr.flags = DIAG308_FLAGS_LP_VALID; + + /* VM PARM */ + if (MACHINE_IS_VM && diag308_set_works && + (ipl_block.ipl_info.ccw.vm_flags & DIAG308_VM_FLAGS_VP_VALID)) { + + ipb->ipl_info.ccw.vm_flags |= DIAG308_VM_FLAGS_VP_VALID; + ipb->ipl_info.ccw.vm_parm_len = + ipl_block.ipl_info.ccw.vm_parm_len; + memcpy(ipb->ipl_info.ccw.vm_parm, + ipl_block.ipl_info.ccw.vm_parm, DIAG308_VMPARM_SIZE); + } } static int __init reipl_nss_init(void) @@ -725,10 +1004,31 @@ static int __init reipl_nss_init(void) if (!MACHINE_IS_VM) return 0; + + reipl_block_nss = (void *) get_zeroed_page(GFP_KERNEL); + if (!reipl_block_nss) + return -ENOMEM; + + if (!diag308_set_works) + sys_reipl_nss_vmparm_attr.attr.mode = S_IRUGO; + rc = sysfs_create_group(&reipl_kset->kobj, &reipl_nss_attr_group); if (rc) return rc; - strncpy(reipl_nss_name, kernel_nss_name, NSS_NAME_SIZE + 1); + + reipl_block_ccw_init(reipl_block_nss); + if (ipl_info.type == IPL_TYPE_NSS) { + memset(reipl_block_nss->ipl_info.ccw.nss_name, + ' ', NSS_NAME_SIZE); + memcpy(reipl_block_nss->ipl_info.ccw.nss_name, + kernel_nss_name, strlen(kernel_nss_name)); + ASCEBC(reipl_block_nss->ipl_info.ccw.nss_name, NSS_NAME_SIZE); + reipl_block_nss->ipl_info.ccw.vm_flags |= + DIAG308_VM_FLAGS_NSS_VALID; + + reipl_block_ccw_fill_parms(reipl_block_nss); + } + reipl_capabilities |= IPL_TYPE_NSS; return 0; } @@ -740,28 +1040,27 @@ static int __init reipl_ccw_init(void) reipl_block_ccw = (void *) get_zeroed_page(GFP_KERNEL); if (!reipl_block_ccw) return -ENOMEM; - rc = sysfs_create_group(&reipl_kset->kobj, &reipl_ccw_attr_group); - if (rc) { - free_page((unsigned long)reipl_block_ccw); - return rc; + + if (MACHINE_IS_VM) { + if (!diag308_set_works) + sys_reipl_ccw_vmparm_attr.attr.mode = S_IRUGO; + rc = sysfs_create_group(&reipl_kset->kobj, + &reipl_ccw_attr_group_vm); + } else { + if(!diag308_set_works) + sys_reipl_ccw_loadparm_attr.attr.mode = S_IRUGO; + rc = sysfs_create_group(&reipl_kset->kobj, + &reipl_ccw_attr_group_lpar); } - reipl_block_ccw->hdr.len = IPL_PARM_BLK_CCW_LEN; - reipl_block_ccw->hdr.version = IPL_PARM_BLOCK_VERSION; - reipl_block_ccw->hdr.blk0_len = IPL_PARM_BLK0_CCW_LEN; - reipl_block_ccw->hdr.pbt = DIAG308_IPL_TYPE_CCW; - reipl_block_ccw->hdr.flags = DIAG308_FLAGS_LP_VALID; - /* check if read scp info worked and set loadparm */ - if (sclp_ipl_info.is_valid) - memcpy(reipl_block_ccw->ipl_info.ccw.load_param, - &sclp_ipl_info.loadparm, LOADPARM_LEN); - else - /* read scp info failed: set empty loadparm (EBCDIC blanks) */ - memset(reipl_block_ccw->ipl_info.ccw.load_param, 0x40, - LOADPARM_LEN); - if (!MACHINE_IS_VM && !diag308_set_works) - sys_reipl_ccw_loadparm_attr.attr.mode = S_IRUGO; - if (ipl_info.type == IPL_TYPE_CCW) + if (rc) + return rc; + + reipl_block_ccw_init(reipl_block_ccw); + if (ipl_info.type == IPL_TYPE_CCW) { reipl_block_ccw->ipl_info.ccw.devno = ipl_devno; + reipl_block_ccw_fill_parms(reipl_block_ccw); + } + reipl_capabilities |= IPL_TYPE_CCW; return 0; } @@ -1298,7 +1597,6 @@ static void __init shutdown_actions_init(void) static int __init s390_ipl_init(void) { - reipl_probe(); sclp_get_ipl_info(&sclp_ipl_info); shutdown_actions_init(); shutdown_triggers_init(); @@ -1405,6 +1703,12 @@ void __init setup_ipl(void) atomic_notifier_chain_register(&panic_notifier_list, &on_panic_nb); } +void __init ipl_update_parameters(void) +{ + if (diag308(DIAG308_STORE, &ipl_block) == DIAG308_RC_OK) + diag308_set_works = 1; +} + void __init ipl_save_parameters(void) { struct cio_iplinfo iplinfo; diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c index ed04d13..288ad49 100644 --- a/arch/s390/kernel/kprobes.c +++ b/arch/s390/kernel/kprobes.c @@ -41,10 +41,8 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p) if (is_prohibited_opcode((kprobe_opcode_t *) p->addr)) return -EINVAL; - if ((unsigned long)p->addr & 0x01) { - printk("Attempt to register kprobe at an unaligned address\n"); + if ((unsigned long)p->addr & 0x01) return -EINVAL; - } /* Use the get_insn_slot() facility for correctness */ if (!(p->ainsn.insn = get_insn_slot())) diff --git a/arch/s390/kernel/machine_kexec.c b/arch/s390/kernel/machine_kexec.c index 3c77dd3..131d7ee 100644 --- a/arch/s390/kernel/machine_kexec.c +++ b/arch/s390/kernel/machine_kexec.c @@ -52,7 +52,6 @@ void machine_kexec_cleanup(struct kimage *image) void machine_shutdown(void) { - printk(KERN_INFO "kexec: machine_shutdown called\n"); } void machine_kexec(struct kimage *image) diff --git a/arch/s390/kernel/mem_detect.c b/arch/s390/kernel/mem_detect.c new file mode 100644 index 0000000..18ed7ab --- /dev/null +++ b/arch/s390/kernel/mem_detect.c @@ -0,0 +1,100 @@ +/* + * Copyright IBM Corp. 2008 + * Author(s): Heiko Carstens <heiko.carstens@de.ibm.com> + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <asm/ipl.h> +#include <asm/sclp.h> +#include <asm/setup.h> + +static int memory_fast_detect(struct mem_chunk *chunk) +{ + unsigned long val0 = 0; + unsigned long val1 = 0xc; + int rc = -EOPNOTSUPP; + + if (ipl_flags & IPL_NSS_VALID) + return -EOPNOTSUPP; + asm volatile( + " diag %1,%2,0x260\n" + "0: lhi %0,0\n" + "1:\n" + EX_TABLE(0b,1b) + : "+d" (rc), "+d" (val0), "+d" (val1) : : "cc"); + + if (rc || val0 != val1) + return -EOPNOTSUPP; + chunk->size = val0 + 1; + return 0; +} + +static inline int tprot(unsigned long addr) +{ + int rc = -EFAULT; + + asm volatile( + " tprot 0(%1),0\n" + "0: ipm %0\n" + " srl %0,28\n" + "1:\n" + EX_TABLE(0b,1b) + : "+d" (rc) : "a" (addr) : "cc"); + return rc; +} + +#define ADDR2G (1ULL << 31) + +static void find_memory_chunks(struct mem_chunk chunk[]) +{ + unsigned long long memsize, rnmax, rzm; + unsigned long addr = 0, size; + int i = 0, type; + + rzm = sclp_get_rzm(); + rnmax = sclp_get_rnmax(); + memsize = rzm * rnmax; + if (!rzm) + rzm = 1ULL << 17; + if (sizeof(long) == 4) { + rzm = min(ADDR2G, rzm); + memsize = memsize ? min(ADDR2G, memsize) : ADDR2G; + } + do { + size = 0; + type = tprot(addr); + do { + size += rzm; + if (memsize && addr + size >= memsize) + break; + } while (type == tprot(addr + size)); + if (type == CHUNK_READ_WRITE || type == CHUNK_READ_ONLY) { + chunk[i].addr = addr; + chunk[i].size = size; + chunk[i].type = type; + i++; + } + addr += size; + } while (addr < memsize && i < MEMORY_CHUNKS); +} + +void detect_memory_layout(struct mem_chunk chunk[]) +{ + unsigned long flags, cr0; + + memset(chunk, 0, MEMORY_CHUNKS * sizeof(struct mem_chunk)); + if (memory_fast_detect(&chunk[0]) == 0) + return; + /* Disable IRQs, DAT and low address protection so tprot does the + * right thing and we don't get scheduled away with low address + * protection disabled. + */ + flags = __raw_local_irq_stnsm(0xf8); + __ctl_store(cr0, 0, 0); + __ctl_clear_bit(0, 28); + find_memory_chunks(chunk); + __ctl_load(cr0, 0, 0); + __raw_local_irq_ssm(flags); +} +EXPORT_SYMBOL(detect_memory_layout); diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c index 7920861..85defd0 100644 --- a/arch/s390/kernel/process.c +++ b/arch/s390/kernel/process.c @@ -75,46 +75,19 @@ unsigned long thread_saved_pc(struct task_struct *tsk) return sf->gprs[8]; } -/* - * 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) -{ - return atomic_notifier_chain_register(&idle_chain, nb); -} -EXPORT_SYMBOL(register_idle_notifier); - -int unregister_idle_notifier(struct notifier_block *nb) -{ - return atomic_notifier_chain_unregister(&idle_chain, nb); -} -EXPORT_SYMBOL(unregister_idle_notifier); - 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); + vtime_stop_cpu_timer(); return NOTIFY_OK; } @@ -122,13 +95,12 @@ void s390_idle_leave(void) { struct s390_idle_data *idle; + vtime_start_cpu_timer(); idle = &__get_cpu_var(s390_idle); spin_lock(&idle->lock); idle->idle_time += get_clock() - idle->idle_enter; idle->in_idle = 0; spin_unlock(&idle->lock); - atomic_notifier_call_chain(&idle_chain, S390_CPU_NOT_IDLE, - (void *)(long) smp_processor_id()); } extern void s390_handle_mcck(void); diff --git a/arch/s390/kernel/ptrace.c b/arch/s390/kernel/ptrace.c index 35827b9..2815bfe 100644 --- a/arch/s390/kernel/ptrace.c +++ b/arch/s390/kernel/ptrace.c @@ -33,6 +33,8 @@ #include <linux/security.h> #include <linux/audit.h> #include <linux/signal.h> +#include <linux/elf.h> +#include <linux/regset.h> #include <asm/segment.h> #include <asm/page.h> @@ -47,6 +49,11 @@ #include "compat_ptrace.h" #endif +enum s390_regset { + REGSET_GENERAL, + REGSET_FP, +}; + static void FixPerRegisters(struct task_struct *task) { @@ -126,24 +133,10 @@ ptrace_disable(struct task_struct *child) * struct user contain pad bytes that should be read as zeroes. * Lovely... */ -static int -peek_user(struct task_struct *child, addr_t addr, addr_t data) +static unsigned long __peek_user(struct task_struct *child, addr_t addr) { struct user *dummy = NULL; - addr_t offset, tmp, mask; - - /* - * Stupid gdb peeks/pokes the access registers in 64 bit with - * an alignment of 4. Programmers from hell... - */ - mask = __ADDR_MASK; -#ifdef CONFIG_64BIT - if (addr >= (addr_t) &dummy->regs.acrs && - addr < (addr_t) &dummy->regs.orig_gpr2) - mask = 3; -#endif - if ((addr & mask) || addr > sizeof(struct user) - __ADDR_MASK) - return -EIO; + addr_t offset, tmp; if (addr < (addr_t) &dummy->regs.acrs) { /* @@ -197,24 +190,18 @@ peek_user(struct task_struct *child, addr_t addr, addr_t data) } else tmp = 0; - return put_user(tmp, (addr_t __user *) data); + return tmp; } -/* - * Write a word to the user area of a process at location addr. This - * operation does have an additional problem compared to peek_user. - * Stores to the program status word and on the floating point - * control register needs to get checked for validity. - */ static int -poke_user(struct task_struct *child, addr_t addr, addr_t data) +peek_user(struct task_struct *child, addr_t addr, addr_t data) { struct user *dummy = NULL; - addr_t offset, mask; + addr_t tmp, mask; /* * Stupid gdb peeks/pokes the access registers in 64 bit with - * an alignment of 4. Programmers from hell indeed... + * an alignment of 4. Programmers from hell... */ mask = __ADDR_MASK; #ifdef CONFIG_64BIT @@ -225,6 +212,21 @@ poke_user(struct task_struct *child, addr_t addr, addr_t data) if ((addr & mask) || addr > sizeof(struct user) - __ADDR_MASK) return -EIO; + tmp = __peek_user(child, addr); + return put_user(tmp, (addr_t __user *) data); +} + +/* + * Write a word to the user area of a process at location addr. This + * operation does have an additional problem compared to peek_user. + * Stores to the program status word and on the floating point + * control register needs to get checked for validity. + */ +static int __poke_user(struct task_struct *child, addr_t addr, addr_t data) +{ + struct user *dummy = NULL; + addr_t offset; + if (addr < (addr_t) &dummy->regs.acrs) { /* * psw and gprs are stored on the stack @@ -292,6 +294,28 @@ poke_user(struct task_struct *child, addr_t addr, addr_t data) return 0; } +static int +poke_user(struct task_struct *child, addr_t addr, addr_t data) +{ + struct user *dummy = NULL; + addr_t mask; + + /* + * Stupid gdb peeks/pokes the access registers in 64 bit with + * an alignment of 4. Programmers from hell indeed... + */ + mask = __ADDR_MASK; +#ifdef CONFIG_64BIT + if (addr >= (addr_t) &dummy->regs.acrs && + addr < (addr_t) &dummy->regs.orig_gpr2) + mask = 3; +#endif + if ((addr & mask) || addr > sizeof(struct user) - __ADDR_MASK) + return -EIO; + + return __poke_user(child, addr, data); +} + long arch_ptrace(struct task_struct *child, long request, long addr, long data) { ptrace_area parea; @@ -367,18 +391,13 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) /* * Same as peek_user but for a 31 bit program. */ -static int -peek_user_emu31(struct task_struct *child, addr_t addr, addr_t data) +static u32 __peek_user_compat(struct task_struct *child, addr_t addr) { struct user32 *dummy32 = NULL; per_struct32 *dummy_per32 = NULL; addr_t offset; __u32 tmp; - if (!test_thread_flag(TIF_31BIT) || - (addr & 3) || addr > sizeof(struct user) - 3) - return -EIO; - if (addr < (addr_t) &dummy32->regs.acrs) { /* * psw and gprs are stored on the stack @@ -435,25 +454,32 @@ peek_user_emu31(struct task_struct *child, addr_t addr, addr_t data) } else tmp = 0; + return tmp; +} + +static int peek_user_compat(struct task_struct *child, + addr_t addr, addr_t data) +{ + __u32 tmp; + + if (!test_thread_flag(TIF_31BIT) || + (addr & 3) || addr > sizeof(struct user) - 3) + return -EIO; + + tmp = __peek_user_compat(child, addr); return put_user(tmp, (__u32 __user *) data); } /* * Same as poke_user but for a 31 bit program. */ -static int -poke_user_emu31(struct task_struct *child, addr_t addr, addr_t data) +static int __poke_user_compat(struct task_struct *child, + addr_t addr, addr_t data) { struct user32 *dummy32 = NULL; per_struct32 *dummy_per32 = NULL; + __u32 tmp = (__u32) data; addr_t offset; - __u32 tmp; - - if (!test_thread_flag(TIF_31BIT) || - (addr & 3) || addr > sizeof(struct user32) - 3) - return -EIO; - - tmp = (__u32) data; if (addr < (addr_t) &dummy32->regs.acrs) { /* @@ -528,6 +554,16 @@ poke_user_emu31(struct task_struct *child, addr_t addr, addr_t data) return 0; } +static int poke_user_compat(struct task_struct *child, + addr_t addr, addr_t data) +{ + if (!test_thread_flag(TIF_31BIT) || + (addr & 3) || addr > sizeof(struct user32) - 3) + return -EIO; + + return __poke_user_compat(child, addr, data); +} + long compat_arch_ptrace(struct task_struct *child, compat_long_t request, compat_ulong_t caddr, compat_ulong_t cdata) { @@ -539,11 +575,11 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request, switch (request) { case PTRACE_PEEKUSR: /* read the word at location addr in the USER area. */ - return peek_user_emu31(child, addr, data); + return peek_user_compat(child, addr, data); case PTRACE_POKEUSR: /* write the word at location addr in the USER area */ - return poke_user_emu31(child, addr, data); + return poke_user_compat(child, addr, data); case PTRACE_PEEKUSR_AREA: case PTRACE_POKEUSR_AREA: @@ -555,13 +591,13 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request, copied = 0; while (copied < parea.len) { if (request == PTRACE_PEEKUSR_AREA) - ret = peek_user_emu31(child, addr, data); + ret = peek_user_compat(child, addr, data); else { __u32 utmp; if (get_user(utmp, (__u32 __force __user *) data)) return -EFAULT; - ret = poke_user_emu31(child, addr, utmp); + ret = poke_user_compat(child, addr, utmp); } if (ret) return ret; @@ -610,3 +646,240 @@ syscall_trace(struct pt_regs *regs, int entryexit) regs->gprs[2], regs->orig_gpr2, regs->gprs[3], regs->gprs[4], regs->gprs[5]); } + +/* + * user_regset definitions. + */ + +static int s390_regs_get(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + void *kbuf, void __user *ubuf) +{ + if (target == current) + save_access_regs(target->thread.acrs); + + if (kbuf) { + unsigned long *k = kbuf; + while (count > 0) { + *k++ = __peek_user(target, pos); + count -= sizeof(*k); + pos += sizeof(*k); + } + } else { + unsigned long __user *u = ubuf; + while (count > 0) { + if (__put_user(__peek_user(target, pos), u++)) + return -EFAULT; + count -= sizeof(*u); + pos += sizeof(*u); + } + } + return 0; +} + +static int s390_regs_set(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + int rc = 0; + + if (target == current) + save_access_regs(target->thread.acrs); + + if (kbuf) { + const unsigned long *k = kbuf; + while (count > 0 && !rc) { + rc = __poke_user(target, pos, *k++); + count -= sizeof(*k); + pos += sizeof(*k); + } + } else { + const unsigned long __user *u = ubuf; + while (count > 0 && !rc) { + unsigned long word; + rc = __get_user(word, u++); + if (rc) + break; + rc = __poke_user(target, pos, word); + count -= sizeof(*u); + pos += sizeof(*u); + } + } + + if (rc == 0 && target == current) + restore_access_regs(target->thread.acrs); + + return rc; +} + +static int s390_fpregs_get(struct task_struct *target, + const struct user_regset *regset, unsigned int pos, + unsigned int count, void *kbuf, void __user *ubuf) +{ + if (target == current) + save_fp_regs(&target->thread.fp_regs); + + return user_regset_copyout(&pos, &count, &kbuf, &ubuf, + &target->thread.fp_regs, 0, -1); +} + +static int s390_fpregs_set(struct task_struct *target, + const struct user_regset *regset, unsigned int pos, + unsigned int count, const void *kbuf, + const void __user *ubuf) +{ + int rc = 0; + + if (target == current) + save_fp_regs(&target->thread.fp_regs); + + /* If setting FPC, must validate it first. */ + if (count > 0 && pos < offsetof(s390_fp_regs, fprs)) { + u32 fpc[2] = { target->thread.fp_regs.fpc, 0 }; + rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf, &fpc, + 0, offsetof(s390_fp_regs, fprs)); + if (rc) + return rc; + if ((fpc[0] & ~FPC_VALID_MASK) != 0 || fpc[1] != 0) + return -EINVAL; + target->thread.fp_regs.fpc = fpc[0]; + } + + if (rc == 0 && count > 0) + rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf, + target->thread.fp_regs.fprs, + offsetof(s390_fp_regs, fprs), -1); + + if (rc == 0 && target == current) + restore_fp_regs(&target->thread.fp_regs); + + return rc; +} + +static const struct user_regset s390_regsets[] = { + [REGSET_GENERAL] = { + .core_note_type = NT_PRSTATUS, + .n = sizeof(s390_regs) / sizeof(long), + .size = sizeof(long), + .align = sizeof(long), + .get = s390_regs_get, + .set = s390_regs_set, + }, + [REGSET_FP] = { + .core_note_type = NT_PRFPREG, + .n = sizeof(s390_fp_regs) / sizeof(long), + .size = sizeof(long), + .align = sizeof(long), + .get = s390_fpregs_get, + .set = s390_fpregs_set, + }, +}; + +static const struct user_regset_view user_s390_view = { + .name = UTS_MACHINE, + .e_machine = EM_S390, + .regsets = s390_regsets, + .n = ARRAY_SIZE(s390_regsets) +}; + +#ifdef CONFIG_COMPAT +static int s390_compat_regs_get(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + void *kbuf, void __user *ubuf) +{ + if (target == current) + save_access_regs(target->thread.acrs); + + if (kbuf) { + compat_ulong_t *k = kbuf; + while (count > 0) { + *k++ = __peek_user_compat(target, pos); + count -= sizeof(*k); + pos += sizeof(*k); + } + } else { + compat_ulong_t __user *u = ubuf; + while (count > 0) { + if (__put_user(__peek_user_compat(target, pos), u++)) + return -EFAULT; + count -= sizeof(*u); + pos += sizeof(*u); + } + } + return 0; +} + +static int s390_compat_regs_set(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + int rc = 0; + + if (target == current) + save_access_regs(target->thread.acrs); + + if (kbuf) { + const compat_ulong_t *k = kbuf; + while (count > 0 && !rc) { + rc = __poke_user_compat(target, pos, *k++); + count -= sizeof(*k); + pos += sizeof(*k); + } + } else { + const compat_ulong_t __user *u = ubuf; + while (count > 0 && !rc) { + compat_ulong_t word; + rc = __get_user(word, u++); + if (rc) + break; + rc = __poke_user_compat(target, pos, word); + count -= sizeof(*u); + pos += sizeof(*u); + } + } + + if (rc == 0 && target == current) + restore_access_regs(target->thread.acrs); + + return rc; +} + +static const struct user_regset s390_compat_regsets[] = { + [REGSET_GENERAL] = { + .core_note_type = NT_PRSTATUS, + .n = sizeof(s390_compat_regs) / sizeof(compat_long_t), + .size = sizeof(compat_long_t), + .align = sizeof(compat_long_t), + .get = s390_compat_regs_get, + .set = s390_compat_regs_set, + }, + [REGSET_FP] = { + .core_note_type = NT_PRFPREG, + .n = sizeof(s390_fp_regs) / sizeof(compat_long_t), + .size = sizeof(compat_long_t), + .align = sizeof(compat_long_t), + .get = s390_fpregs_get, + .set = s390_fpregs_set, + }, +}; + +static const struct user_regset_view user_s390_compat_view = { + .name = "s390", + .e_machine = EM_S390, + .regsets = s390_compat_regsets, + .n = ARRAY_SIZE(s390_compat_regsets) +}; +#endif + +const struct user_regset_view *task_user_regset_view(struct task_struct *task) +{ +#ifdef CONFIG_COMPAT + if (test_tsk_thread_flag(task, TIF_31BIT)) + return &user_s390_compat_view; +#endif + return &user_s390_view; +} diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 2bc70b6..b358e18 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -77,7 +77,7 @@ unsigned long machine_flags; unsigned long elf_hwcap = 0; char elf_platform[ELF_PLATFORM_SIZE]; -struct mem_chunk __meminitdata memory_chunk[MEMORY_CHUNKS]; +struct mem_chunk __initdata memory_chunk[MEMORY_CHUNKS]; volatile int __cpu_logical_map[NR_CPUS]; /* logical cpu to cpu address */ static unsigned long __initdata memory_end; @@ -205,12 +205,6 @@ static void __init conmode_default(void) SET_CONSOLE_SCLP; #endif } - } else if (MACHINE_IS_P390) { -#if defined(CONFIG_TN3215_CONSOLE) - SET_CONSOLE_3215; -#elif defined(CONFIG_TN3270_CONSOLE) - SET_CONSOLE_3270; -#endif } else { #if defined(CONFIG_SCLP_CONSOLE) || defined(CONFIG_SCLP_VT220_CONSOLE) SET_CONSOLE_SCLP; @@ -221,18 +215,17 @@ static void __init conmode_default(void) #if defined(CONFIG_ZFCPDUMP) || defined(CONFIG_ZFCPDUMP_MODULE) static void __init setup_zfcpdump(unsigned int console_devno) { - static char str[64]; + static char str[41]; if (ipl_info.type != IPL_TYPE_FCP_DUMP) return; if (console_devno != -1) - sprintf(str, "cio_ignore=all,!0.0.%04x,!0.0.%04x", + sprintf(str, " cio_ignore=all,!0.0.%04x,!0.0.%04x", ipl_info.data.fcp.dev_id.devno, console_devno); else - sprintf(str, "cio_ignore=all,!0.0.%04x", + sprintf(str, " cio_ignore=all,!0.0.%04x", ipl_info.data.fcp.dev_id.devno); - strcat(COMMAND_LINE, " "); - strcat(COMMAND_LINE, str); + strcat(boot_command_line, str); console_loglevel = 2; } #else @@ -289,32 +282,6 @@ static int __init early_parse_mem(char *p) } early_param("mem", early_parse_mem); -/* - * "ipldelay=XXX[sm]" sets ipl delay in seconds or minutes - */ -static int __init early_parse_ipldelay(char *p) -{ - unsigned long delay = 0; - - delay = simple_strtoul(p, &p, 0); - - switch (*p) { - case 's': - case 'S': - delay *= 1000000; - break; - case 'm': - case 'M': - delay *= 60 * 1000000; - } - - /* now wait for the requested amount of time */ - udelay(delay); - - return 0; -} -early_param("ipldelay", early_parse_ipldelay); - #ifdef CONFIG_S390_SWITCH_AMODE #ifdef CONFIG_PGSTE unsigned int switch_amode = 1; @@ -804,11 +771,9 @@ setup_arch(char **cmdline_p) printk("We are running native (64 bit mode)\n"); #endif /* CONFIG_64BIT */ - /* Save unparsed command line copy for /proc/cmdline */ - strlcpy(boot_command_line, COMMAND_LINE, COMMAND_LINE_SIZE); - - *cmdline_p = COMMAND_LINE; - *(*cmdline_p + COMMAND_LINE_SIZE - 1) = '\0'; + /* Have one command line that is parsed and saved in /proc/cmdline */ + /* boot_command_line has been already set up in early.c */ + *cmdline_p = boot_command_line; ROOT_DEV = Root_RAM0; diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 1f42289..5d4fa4b 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -711,7 +711,7 @@ int __cpuinit __cpu_up(unsigned int cpu) memset(sf, 0, sizeof(struct stack_frame)); sf->gprs[9] = (unsigned long) sf; cpu_lowcore->save_area[15] = (unsigned long) sf; - __ctl_store(cpu_lowcore->cregs_save_area[0], 0, 15); + __ctl_store(cpu_lowcore->cregs_save_area, 0, 15); asm volatile( " stam 0,15,0(%0)" : : "a" (&cpu_lowcore->access_regs_save_area) : "memory"); @@ -1089,7 +1089,7 @@ out: #ifdef CONFIG_HOTPLUG_CPU -int smp_rescan_cpus(void) +int __ref smp_rescan_cpus(void) { cpumask_t newcpus; int cpu; diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index 7aec676..7418beb 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -3,7 +3,7 @@ * Time of day based timer functions. * * S390 version - * Copyright (C) 1999 IBM Deutschland Entwicklung GmbH, IBM Corporation + * Copyright IBM Corp. 1999, 2008 * Author(s): Hartmut Penner (hp@de.ibm.com), * Martin Schwidefsky (schwidefsky@de.ibm.com), * Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com) @@ -31,6 +31,7 @@ #include <linux/notifier.h> #include <linux/clocksource.h> #include <linux/clockchips.h> +#include <linux/bootmem.h> #include <asm/uaccess.h> #include <asm/delay.h> #include <asm/s390_ext.h> @@ -162,7 +163,7 @@ void init_cpu_timer(void) /* Enable clock comparator timer interrupt. */ __ctl_set_bit(0,11); - /* Always allow ETR external interrupts, even without an ETR. */ + /* Always allow the timing alert external interrupt. */ __ctl_set_bit(0, 4); } @@ -170,8 +171,21 @@ static void clock_comparator_interrupt(__u16 code) { } +static void etr_timing_alert(struct etr_irq_parm *); +static void stp_timing_alert(struct stp_irq_parm *); + +static void timing_alert_interrupt(__u16 code) +{ + if (S390_lowcore.ext_params & 0x00c40000) + etr_timing_alert((struct etr_irq_parm *) + &S390_lowcore.ext_params); + if (S390_lowcore.ext_params & 0x00038000) + stp_timing_alert((struct stp_irq_parm *) + &S390_lowcore.ext_params); +} + static void etr_reset(void); -static void etr_ext_handler(__u16); +static void stp_reset(void); /* * Get the TOD clock running. @@ -181,6 +195,7 @@ static u64 __init reset_tod_clock(void) u64 time; etr_reset(); + stp_reset(); if (store_clock(&time) == 0) return time; /* TOD clock not running. Set the clock to Unix Epoch. */ @@ -231,8 +246,9 @@ void __init time_init(void) if (clocksource_register(&clocksource_tod) != 0) panic("Could not register TOD clock source"); - /* request the etr external interrupt */ - if (register_early_external_interrupt(0x1406, etr_ext_handler, + /* request the timing alert external interrupt */ + if (register_early_external_interrupt(0x1406, + timing_alert_interrupt, &ext_int_etr_cc) != 0) panic("Couldn't request external interrupt 0x1406"); @@ -245,10 +261,112 @@ void __init time_init(void) } /* + * 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 unsigned long long adjust_time(unsigned long long old, + unsigned long long clock, + unsigned long long delay) +{ + unsigned long long delta, ticks; + struct timex adjust; + + if (clock > old) { + /* It is later than we thought. */ + delta = ticks = clock - old; + delta = ticks = (delta < delay) ? 0 : delta - delay; + delta -= do_div(ticks, CLK_TICKS_PER_JIFFY); + adjust.offset = ticks * (1000000 / HZ); + } else { + /* It is earlier than we thought. */ + delta = ticks = old - clock; + delta -= do_div(ticks, CLK_TICKS_PER_JIFFY); + 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 DEFINE_PER_CPU(atomic_t, clock_sync_word); +static unsigned long clock_sync_flags; + +#define CLOCK_SYNC_HAS_ETR 0 +#define CLOCK_SYNC_HAS_STP 1 +#define CLOCK_SYNC_ETR 2 +#define CLOCK_SYNC_STP 3 + +/* + * The synchronous get_clock function. It will write the current clock + * value to the clock pointer and return 0 if the clock is in sync with + * the external time source. If the clock mode is local it will return + * -ENOSYS and -EAGAIN if the clock is not in sync with the external + * reference. + */ +int get_sync_clock(unsigned long long *clock) +{ + atomic_t *sw_ptr; + unsigned int sw0, sw1; + + sw_ptr = &get_cpu_var(clock_sync_word); + sw0 = atomic_read(sw_ptr); + *clock = get_clock(); + sw1 = atomic_read(sw_ptr); + put_cpu_var(clock_sync_sync); + if (sw0 == sw1 && (sw0 & 0x80000000U)) + /* Success: time is in sync. */ + return 0; + if (!test_bit(CLOCK_SYNC_HAS_ETR, &clock_sync_flags) && + !test_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags)) + return -ENOSYS; + if (!test_bit(CLOCK_SYNC_ETR, &clock_sync_flags) && + !test_bit(CLOCK_SYNC_STP, &clock_sync_flags)) + return -EACCES; + return -EAGAIN; +} +EXPORT_SYMBOL(get_sync_clock); + +/* + * Make get_sync_clock return -EAGAIN. + */ +static void disable_sync_clock(void *dummy) +{ + atomic_t *sw_ptr = &__get_cpu_var(clock_sync_word); + /* + * Clear the in-sync bit 2^31. All get_sync_clock calls will + * fail until the sync bit is turned back on. In addition + * increase the "sequence" counter to avoid the race of an + * etr event and the complete recovery against get_sync_clock. + */ + atomic_clear_mask(0x80000000, sw_ptr); + atomic_inc(sw_ptr); +} + +/* + * Make get_sync_clock return 0 again. + * Needs to be called from a context disabled for preemption. + */ +static void enable_sync_clock(void) +{ + atomic_t *sw_ptr = &__get_cpu_var(clock_sync_word); + atomic_set_mask(0x80000000, sw_ptr); +} + +/* * External Time Reference (ETR) code. */ static int etr_port0_online; static int etr_port1_online; +static int etr_steai_available; static int __init early_parse_etr(char *p) { @@ -273,12 +391,6 @@ enum etr_event { ETR_EVENT_UPDATE, }; -enum etr_flags { - ETR_FLAG_ENOSYS, - ETR_FLAG_EACCES, - ETR_FLAG_STEAI, -}; - /* * Valid bit combinations of the eacr register are (x = don't care): * e0 e1 dp p0 p1 ea es sl @@ -305,74 +417,18 @@ enum etr_flags { */ static struct etr_eacr etr_eacr; static u64 etr_tolec; /* time of last eacr update */ -static unsigned long etr_flags; static struct etr_aib etr_port0; static int etr_port0_uptodate; static struct etr_aib etr_port1; static int etr_port1_uptodate; static unsigned long etr_events; static struct timer_list etr_timer; -static DEFINE_PER_CPU(atomic_t, etr_sync_word); static void etr_timeout(unsigned long dummy); static void etr_work_fn(struct work_struct *work); static DECLARE_WORK(etr_work, etr_work_fn); /* - * The etr get_clock function. It will write the current clock value - * to the clock pointer and return 0 if the clock is in sync with the - * external time source. If the clock mode is local it will return - * -ENOSYS and -EAGAIN if the clock is not in sync with the external - * reference. This function is what ETR is all about.. - */ -int get_sync_clock(unsigned long long *clock) -{ - atomic_t *sw_ptr; - unsigned int sw0, sw1; - - sw_ptr = &get_cpu_var(etr_sync_word); - sw0 = atomic_read(sw_ptr); - *clock = get_clock(); - sw1 = atomic_read(sw_ptr); - put_cpu_var(etr_sync_sync); - if (sw0 == sw1 && (sw0 & 0x80000000U)) - /* Success: time is in sync. */ - return 0; - if (test_bit(ETR_FLAG_ENOSYS, &etr_flags)) - return -ENOSYS; - if (test_bit(ETR_FLAG_EACCES, &etr_flags)) - return -EACCES; - return -EAGAIN; -} -EXPORT_SYMBOL(get_sync_clock); - -/* - * Make get_sync_clock return -EAGAIN. - */ -static void etr_disable_sync_clock(void *dummy) -{ - atomic_t *sw_ptr = &__get_cpu_var(etr_sync_word); - /* - * Clear the in-sync bit 2^31. All get_sync_clock calls will - * fail until the sync bit is turned back on. In addition - * increase the "sequence" counter to avoid the race of an - * etr event and the complete recovery against get_sync_clock. - */ - atomic_clear_mask(0x80000000, sw_ptr); - atomic_inc(sw_ptr); -} - -/* - * Make get_sync_clock return 0 again. - * Needs to be called from a context disabled for preemption. - */ -static void etr_enable_sync_clock(void) -{ - atomic_t *sw_ptr = &__get_cpu_var(etr_sync_word); - atomic_set_mask(0x80000000, sw_ptr); -} - -/* * Reset ETR attachment. */ static void etr_reset(void) @@ -381,15 +437,13 @@ static void etr_reset(void) .e0 = 0, .e1 = 0, ._pad0 = 4, .dp = 0, .p0 = 0, .p1 = 0, ._pad1 = 0, .ea = 0, .es = 0, .sl = 0 }; - if (etr_setr(&etr_eacr) == 0) + if (etr_setr(&etr_eacr) == 0) { etr_tolec = get_clock(); - else { - set_bit(ETR_FLAG_ENOSYS, &etr_flags); - if (etr_port0_online || etr_port1_online) { - printk(KERN_WARNING "Running on non ETR capable " - "machine, only local mode available.\n"); - etr_port0_online = etr_port1_online = 0; - } + set_bit(CLOCK_SYNC_HAS_ETR, &clock_sync_flags); + } else if (etr_port0_online || etr_port1_online) { + printk(KERN_WARNING "Running on non ETR capable " + "machine, only local mode available.\n"); + etr_port0_online = etr_port1_online = 0; } } @@ -397,14 +451,12 @@ static int __init etr_init(void) { struct etr_aib aib; - if (test_bit(ETR_FLAG_ENOSYS, &etr_flags)) + if (!test_bit(CLOCK_SYNC_HAS_ETR, &clock_sync_flags)) return 0; /* Check if this machine has the steai instruction. */ if (etr_steai(&aib, ETR_STEAI_STEPPING_PORT) == 0) - set_bit(ETR_FLAG_STEAI, &etr_flags); + etr_steai_available = 1; setup_timer(&etr_timer, etr_timeout, 0UL); - if (!etr_port0_online && !etr_port1_online) - set_bit(ETR_FLAG_EACCES, &etr_flags); if (etr_port0_online) { set_bit(ETR_EVENT_PORT0_CHANGE, &etr_events); schedule_work(&etr_work); @@ -435,7 +487,8 @@ void etr_switch_to_local(void) { if (!etr_eacr.sl) return; - etr_disable_sync_clock(NULL); + if (test_bit(CLOCK_SYNC_ETR, &clock_sync_flags)) + disable_sync_clock(NULL); set_bit(ETR_EVENT_SWITCH_LOCAL, &etr_events); schedule_work(&etr_work); } @@ -450,23 +503,21 @@ void etr_sync_check(void) { if (!etr_eacr.es) return; - etr_disable_sync_clock(NULL); + if (test_bit(CLOCK_SYNC_ETR, &clock_sync_flags)) + disable_sync_clock(NULL); set_bit(ETR_EVENT_SYNC_CHECK, &etr_events); schedule_work(&etr_work); } /* - * ETR external interrupt. There are two causes: + * ETR timing alert. There are two causes: * 1) port state change, check the usability of the port * 2) port alert, one of the ETR-data-validity bits (v1-v2 bits of the * sldr-status word) or ETR-data word 1 (edf1) or ETR-data word 3 (edf3) * or ETR-data word 4 (edf4) has changed. */ -static void etr_ext_handler(__u16 code) +static void etr_timing_alert(struct etr_irq_parm *intparm) { - struct etr_interruption_parameter *intparm = - (struct etr_interruption_parameter *) &S390_lowcore.ext_params; - if (intparm->pc0) /* ETR port 0 state change. */ set_bit(ETR_EVENT_PORT0_CHANGE, &etr_events); @@ -591,58 +642,23 @@ static int etr_aib_follows(struct etr_aib *a1, struct etr_aib *a2, int p) return 1; } -/* - * 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 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; - - if (clock > old) { - /* It is later than we thought. */ - delta = ticks = clock - old; - delta = ticks = (delta < delay) ? 0 : delta - delay; - delta -= do_div(ticks, CLK_TICKS_PER_JIFFY); - adjust.offset = ticks * (1000000 / HZ); - } else { - /* It is earlier than we thought. */ - delta = ticks = old - clock; - delta -= do_div(ticks, CLK_TICKS_PER_JIFFY); - 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 { +struct clock_sync_data { int in_sync; unsigned long long fixup_cc; -} etr_sync; +}; -static void etr_sync_cpu_start(void *dummy) +static void clock_sync_cpu_start(void *dummy) { - etr_enable_sync_clock(); + struct clock_sync_data *sync = dummy; + + enable_sync_clock(); /* * This looks like a busy wait loop but it isn't. etr_sync_cpus * is called on all other cpus while the TOD clocks is stopped. * __udelay will stop the cpu on an enabled wait psw until the * TOD is running again. */ - while (etr_sync.in_sync == 0) { + while (sync->in_sync == 0) { __udelay(1); /* * A different cpu changes *in_sync. Therefore use @@ -650,17 +666,17 @@ static void etr_sync_cpu_start(void *dummy) */ barrier(); } - if (etr_sync.in_sync != 1) + if (sync->in_sync != 1) /* Didn't work. Clear per-cpu in sync bit again. */ - etr_disable_sync_clock(NULL); + disable_sync_clock(NULL); /* * This round of TOD syncing is done. Set the clock comparator * to the next tick and let the processor continue. */ - fixup_clock_comparator(etr_sync.fixup_cc); + fixup_clock_comparator(sync->fixup_cc); } -static void etr_sync_cpu_end(void *dummy) +static void clock_sync_cpu_end(void *dummy) { } @@ -672,6 +688,7 @@ static void etr_sync_cpu_end(void *dummy) static int etr_sync_clock(struct etr_aib *aib, int port) { struct etr_aib *sync_port; + struct clock_sync_data etr_sync; unsigned long long clock, old_clock, delay, delta; int follows; int rc; @@ -690,9 +707,9 @@ static int etr_sync_clock(struct etr_aib *aib, int port) */ memset(&etr_sync, 0, sizeof(etr_sync)); preempt_disable(); - smp_call_function(etr_sync_cpu_start, NULL, 0, 0); + smp_call_function(clock_sync_cpu_start, &etr_sync, 0, 0); local_irq_disable(); - etr_enable_sync_clock(); + enable_sync_clock(); /* Set clock to next OTE. */ __ctl_set_bit(14, 21); @@ -707,13 +724,13 @@ 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; - delta = etr_adjust_time(old_clock, clock, delay); + delta = 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); + disable_sync_clock(NULL); etr_sync.in_sync = -EAGAIN; rc = -EAGAIN; } else { @@ -724,12 +741,12 @@ static int etr_sync_clock(struct etr_aib *aib, int port) /* Could not set the clock ?!? */ __ctl_clear_bit(0, 29); __ctl_clear_bit(14, 21); - etr_disable_sync_clock(NULL); + disable_sync_clock(NULL); etr_sync.in_sync = -EAGAIN; rc = -EAGAIN; } local_irq_enable(); - smp_call_function(etr_sync_cpu_end,NULL,0,0); + smp_call_function(clock_sync_cpu_end, NULL, 0, 0); preempt_enable(); return rc; } @@ -832,7 +849,7 @@ static struct etr_eacr etr_handle_update(struct etr_aib *aib, * Do not try to get the alternate port aib if the clock * is not in sync yet. */ - if (!eacr.es) + if (!test_bit(CLOCK_SYNC_STP, &clock_sync_flags) && !eacr.es) return eacr; /* @@ -840,7 +857,7 @@ static struct etr_eacr etr_handle_update(struct etr_aib *aib, * the other port immediately. If only stetr is available the * data-port bit toggle has to be used. */ - if (test_bit(ETR_FLAG_STEAI, &etr_flags)) { + if (etr_steai_available) { if (eacr.p0 && !etr_port0_uptodate) { etr_steai_cv(&etr_port0, ETR_STEAI_PORT_0); etr_port0_uptodate = 1; @@ -909,10 +926,10 @@ static void etr_work_fn(struct work_struct *work) if (!eacr.ea) { /* Both ports offline. Reset everything. */ eacr.dp = eacr.es = eacr.sl = 0; - on_each_cpu(etr_disable_sync_clock, NULL, 0, 1); + on_each_cpu(disable_sync_clock, NULL, 0, 1); del_timer_sync(&etr_timer); etr_update_eacr(eacr); - set_bit(ETR_FLAG_EACCES, &etr_flags); + clear_bit(CLOCK_SYNC_ETR, &clock_sync_flags); return; } @@ -953,7 +970,6 @@ static void etr_work_fn(struct work_struct *work) eacr.e1 = 1; sync_port = (etr_port0_uptodate && etr_port_valid(&etr_port0, 0)) ? 0 : -1; - clear_bit(ETR_FLAG_EACCES, &etr_flags); } else if (eacr.p1 && aib.esw.psc1 == etr_lpsc_pps_mode) { eacr.sl = 0; eacr.e0 = 0; @@ -962,7 +978,6 @@ static void etr_work_fn(struct work_struct *work) eacr.es = 0; sync_port = (etr_port1_uptodate && etr_port_valid(&etr_port1, 1)) ? 1 : -1; - clear_bit(ETR_FLAG_EACCES, &etr_flags); } else if (eacr.p0 && aib.esw.psc0 == etr_lpsc_operational_step) { eacr.sl = 1; eacr.e0 = 1; @@ -976,7 +991,6 @@ static void etr_work_fn(struct work_struct *work) eacr.e1 = 1; sync_port = (etr_port0_uptodate && etr_port_valid(&etr_port0, 0)) ? 0 : -1; - clear_bit(ETR_FLAG_EACCES, &etr_flags); } else if (eacr.p1 && aib.esw.psc1 == etr_lpsc_operational_step) { eacr.sl = 1; eacr.e0 = 0; @@ -985,19 +999,22 @@ static void etr_work_fn(struct work_struct *work) eacr.es = 0; sync_port = (etr_port1_uptodate && etr_port_valid(&etr_port1, 1)) ? 1 : -1; - clear_bit(ETR_FLAG_EACCES, &etr_flags); } else { /* Both ports not usable. */ eacr.es = eacr.sl = 0; sync_port = -1; - set_bit(ETR_FLAG_EACCES, &etr_flags); + clear_bit(CLOCK_SYNC_ETR, &clock_sync_flags); } + if (!test_bit(CLOCK_SYNC_ETR, &clock_sync_flags)) + eacr.es = 0; + /* * If the clock is in sync just update the eacr and return. * If there is no valid sync port wait for a port update. */ - if (eacr.es || sync_port < 0) { + if (test_bit(CLOCK_SYNC_STP, &clock_sync_flags) || + eacr.es || sync_port < 0) { etr_update_eacr(eacr); etr_set_tolec_timeout(now); return; @@ -1018,11 +1035,13 @@ static void etr_work_fn(struct work_struct *work) * and set up a timer to try again after 0.5 seconds */ etr_update_eacr(eacr); + set_bit(CLOCK_SYNC_ETR, &clock_sync_flags); if (now < etr_tolec + (1600000 << 12) || etr_sync_clock(&aib, sync_port) != 0) { /* Sync failed. Try again in 1/2 second. */ eacr.es = 0; etr_update_eacr(eacr); + clear_bit(CLOCK_SYNC_ETR, &clock_sync_flags); etr_set_sync_timeout(); } else etr_set_tolec_timeout(now); @@ -1097,8 +1116,8 @@ static ssize_t etr_online_store(struct sys_device *dev, value = simple_strtoul(buf, NULL, 0); if (value != 0 && value != 1) return -EINVAL; - if (test_bit(ETR_FLAG_ENOSYS, &etr_flags)) - return -ENOSYS; + if (!test_bit(CLOCK_SYNC_HAS_ETR, &clock_sync_flags)) + return -EOPNOTSUPP; if (dev == &etr_port0_dev) { if (etr_port0_online == value) return count; /* Nothing to do. */ @@ -1292,3 +1311,318 @@ out: } device_initcall(etr_init_sysfs); + +/* + * Server Time Protocol (STP) code. + */ +static int stp_online; +static struct stp_sstpi stp_info; +static void *stp_page; + +static void stp_work_fn(struct work_struct *work); +static DECLARE_WORK(stp_work, stp_work_fn); + +static int __init early_parse_stp(char *p) +{ + if (strncmp(p, "off", 3) == 0) + stp_online = 0; + else if (strncmp(p, "on", 2) == 0) + stp_online = 1; + return 0; +} +early_param("stp", early_parse_stp); + +/* + * Reset STP attachment. + */ +static void stp_reset(void) +{ + int rc; + + stp_page = alloc_bootmem_pages(PAGE_SIZE); + rc = chsc_sstpc(stp_page, STP_OP_CTRL, 0x0000); + if (rc == 1) + set_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags); + else if (stp_online) { + printk(KERN_WARNING "Running on non STP capable machine.\n"); + free_bootmem((unsigned long) stp_page, PAGE_SIZE); + stp_page = NULL; + stp_online = 0; + } +} + +static int __init stp_init(void) +{ + if (test_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags) && stp_online) + schedule_work(&stp_work); + return 0; +} + +arch_initcall(stp_init); + +/* + * STP timing alert. There are three causes: + * 1) timing status change + * 2) link availability change + * 3) time control parameter change + * In all three cases we are only interested in the clock source state. + * If a STP clock source is now available use it. + */ +static void stp_timing_alert(struct stp_irq_parm *intparm) +{ + if (intparm->tsc || intparm->lac || intparm->tcpc) + schedule_work(&stp_work); +} + +/* + * STP sync check machine check. This is called when the timing state + * changes from the synchronized state to the unsynchronized state. + * After a STP sync check the clock is not in sync. The machine check + * is broadcasted to all cpus at the same time. + */ +void stp_sync_check(void) +{ + if (!test_bit(CLOCK_SYNC_STP, &clock_sync_flags)) + return; + disable_sync_clock(NULL); + schedule_work(&stp_work); +} + +/* + * STP island condition machine check. This is called when an attached + * server attempts to communicate over an STP link and the servers + * have matching CTN ids and have a valid stratum-1 configuration + * but the configurations do not match. + */ +void stp_island_check(void) +{ + if (!test_bit(CLOCK_SYNC_STP, &clock_sync_flags)) + return; + disable_sync_clock(NULL); + schedule_work(&stp_work); +} + +/* + * STP tasklet. Check for the STP state and take over the clock + * synchronization if the STP clock source is usable. + */ +static void stp_work_fn(struct work_struct *work) +{ + struct clock_sync_data stp_sync; + unsigned long long old_clock, delta; + int rc; + + if (!stp_online) { + chsc_sstpc(stp_page, STP_OP_CTRL, 0x0000); + return; + } + + rc = chsc_sstpc(stp_page, STP_OP_CTRL, 0xb0e0); + if (rc) + return; + + rc = chsc_sstpi(stp_page, &stp_info, sizeof(struct stp_sstpi)); + if (rc || stp_info.c == 0) + return; + + /* + * Catch all other cpus and make them wait until we have + * successfully synced the clock. smp_call_function will + * return after all other cpus are in clock_sync_cpu_start. + */ + memset(&stp_sync, 0, sizeof(stp_sync)); + preempt_disable(); + smp_call_function(clock_sync_cpu_start, &stp_sync, 0, 0); + local_irq_disable(); + enable_sync_clock(); + + set_bit(CLOCK_SYNC_STP, &clock_sync_flags); + if (test_and_clear_bit(CLOCK_SYNC_ETR, &clock_sync_flags)) + schedule_work(&etr_work); + + rc = 0; + if (stp_info.todoff[0] || stp_info.todoff[1] || + stp_info.todoff[2] || stp_info.todoff[3] || + stp_info.tmd != 2) { + old_clock = get_clock(); + rc = chsc_sstpc(stp_page, STP_OP_SYNC, 0); + if (rc == 0) { + delta = adjust_time(old_clock, get_clock(), 0); + fixup_clock_comparator(delta); + rc = chsc_sstpi(stp_page, &stp_info, + sizeof(struct stp_sstpi)); + if (rc == 0 && stp_info.tmd != 2) + rc = -EAGAIN; + } + } + if (rc) { + disable_sync_clock(NULL); + stp_sync.in_sync = -EAGAIN; + clear_bit(CLOCK_SYNC_STP, &clock_sync_flags); + if (etr_port0_online || etr_port1_online) + schedule_work(&etr_work); + } else + stp_sync.in_sync = 1; + + local_irq_enable(); + smp_call_function(clock_sync_cpu_end, NULL, 0, 0); + preempt_enable(); +} + +/* + * STP class sysfs interface functions + */ +static struct sysdev_class stp_sysclass = { + .name = "stp", +}; + +static ssize_t stp_ctn_id_show(struct sysdev_class *class, char *buf) +{ + if (!stp_online) + return -ENODATA; + return sprintf(buf, "%016llx\n", + *(unsigned long long *) stp_info.ctnid); +} + +static SYSDEV_CLASS_ATTR(ctn_id, 0400, stp_ctn_id_show, NULL); + +static ssize_t stp_ctn_type_show(struct sysdev_class *class, char *buf) +{ + if (!stp_online) + return -ENODATA; + return sprintf(buf, "%i\n", stp_info.ctn); +} + +static SYSDEV_CLASS_ATTR(ctn_type, 0400, stp_ctn_type_show, NULL); + +static ssize_t stp_dst_offset_show(struct sysdev_class *class, char *buf) +{ + if (!stp_online || !(stp_info.vbits & 0x2000)) + return -ENODATA; + return sprintf(buf, "%i\n", (int)(s16) stp_info.dsto); +} + +static SYSDEV_CLASS_ATTR(dst_offset, 0400, stp_dst_offset_show, NULL); + +static ssize_t stp_leap_seconds_show(struct sysdev_class *class, char *buf) +{ + if (!stp_online || !(stp_info.vbits & 0x8000)) + return -ENODATA; + return sprintf(buf, "%i\n", (int)(s16) stp_info.leaps); +} + +static SYSDEV_CLASS_ATTR(leap_seconds, 0400, stp_leap_seconds_show, NULL); + +static ssize_t stp_stratum_show(struct sysdev_class *class, char *buf) +{ + if (!stp_online) + return -ENODATA; + return sprintf(buf, "%i\n", (int)(s16) stp_info.stratum); +} + +static SYSDEV_CLASS_ATTR(stratum, 0400, stp_stratum_show, NULL); + +static ssize_t stp_time_offset_show(struct sysdev_class *class, char *buf) +{ + if (!stp_online || !(stp_info.vbits & 0x0800)) + return -ENODATA; + return sprintf(buf, "%i\n", (int) stp_info.tto); +} + +static SYSDEV_CLASS_ATTR(time_offset, 0400, stp_time_offset_show, NULL); + +static ssize_t stp_time_zone_offset_show(struct sysdev_class *class, char *buf) +{ + if (!stp_online || !(stp_info.vbits & 0x4000)) + return -ENODATA; + return sprintf(buf, "%i\n", (int)(s16) stp_info.tzo); +} + +static SYSDEV_CLASS_ATTR(time_zone_offset, 0400, + stp_time_zone_offset_show, NULL); + +static ssize_t stp_timing_mode_show(struct sysdev_class *class, char *buf) +{ + if (!stp_online) + return -ENODATA; + return sprintf(buf, "%i\n", stp_info.tmd); +} + +static SYSDEV_CLASS_ATTR(timing_mode, 0400, stp_timing_mode_show, NULL); + +static ssize_t stp_timing_state_show(struct sysdev_class *class, char *buf) +{ + if (!stp_online) + return -ENODATA; + return sprintf(buf, "%i\n", stp_info.tst); +} + +static SYSDEV_CLASS_ATTR(timing_state, 0400, stp_timing_state_show, NULL); + +static ssize_t stp_online_show(struct sysdev_class *class, char *buf) +{ + return sprintf(buf, "%i\n", stp_online); +} + +static ssize_t stp_online_store(struct sysdev_class *class, + const char *buf, size_t count) +{ + unsigned int value; + + value = simple_strtoul(buf, NULL, 0); + if (value != 0 && value != 1) + return -EINVAL; + if (!test_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags)) + return -EOPNOTSUPP; + stp_online = value; + schedule_work(&stp_work); + return count; +} + +/* + * Can't use SYSDEV_CLASS_ATTR because the attribute should be named + * stp/online but attr_online already exists in this file .. + */ +static struct sysdev_class_attribute attr_stp_online = { + .attr = { .name = "online", .mode = 0600 }, + .show = stp_online_show, + .store = stp_online_store, +}; + +static struct sysdev_class_attribute *stp_attributes[] = { + &attr_ctn_id, + &attr_ctn_type, + &attr_dst_offset, + &attr_leap_seconds, + &attr_stp_online, + &attr_stratum, + &attr_time_offset, + &attr_time_zone_offset, + &attr_timing_mode, + &attr_timing_state, + NULL +}; + +static int __init stp_init_sysfs(void) +{ + struct sysdev_class_attribute **attr; + int rc; + + rc = sysdev_class_register(&stp_sysclass); + if (rc) + goto out; + for (attr = stp_attributes; *attr; attr++) { + rc = sysdev_class_create_file(&stp_sysclass, *attr); + if (rc) + goto out_unreg; + } + return 0; +out_unreg: + for (; attr >= stp_attributes; attr--) + sysdev_class_remove_file(&stp_sysclass, *attr); + sysdev_class_unregister(&stp_sysclass); +out: + return rc; +} + +device_initcall(stp_init_sysfs); diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c index 661a072..212d618 100644 --- a/arch/s390/kernel/topology.c +++ b/arch/s390/kernel/topology.c @@ -313,8 +313,6 @@ void __init s390_init_cpu_topology(void) 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); diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S index b460715..76c1e60 100644 --- a/arch/s390/kernel/vmlinux.lds.S +++ b/arch/s390/kernel/vmlinux.lds.S @@ -40,7 +40,6 @@ SECTIONS _etext = .; /* End of text section */ NOTES :text :note - BUG_TABLE :text RODATA diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c index ca90ee3..0fa5dc5 100644 --- a/arch/s390/kernel/vtime.c +++ b/arch/s390/kernel/vtime.c @@ -136,7 +136,7 @@ static inline void set_vtimer(__u64 expires) } #endif -static void start_cpu_timer(void) +void vtime_start_cpu_timer(void) { struct vtimer_queue *vt_list; @@ -150,7 +150,7 @@ static void start_cpu_timer(void) set_vtimer(vt_list->idle); } -static void stop_cpu_timer(void) +void vtime_stop_cpu_timer(void) { struct vtimer_queue *vt_list; @@ -318,8 +318,7 @@ static void internal_add_vtimer(struct vtimer_list *timer) vt_list = &per_cpu(virt_cpu_timer, timer->cpu); spin_lock_irqsave(&vt_list->lock, flags); - if (timer->cpu != smp_processor_id()) - printk("internal_add_vtimer: BUG, running on wrong CPU"); + BUG_ON(timer->cpu != smp_processor_id()); /* if list is empty we only have to set the timer */ if (list_empty(&vt_list->list)) { @@ -353,25 +352,12 @@ static void internal_add_vtimer(struct vtimer_list *timer) put_cpu(); } -static inline int prepare_vtimer(struct vtimer_list *timer) +static inline void prepare_vtimer(struct vtimer_list *timer) { - if (!timer->function) { - printk("add_virt_timer: uninitialized timer\n"); - return -EINVAL; - } - - if (!timer->expires || timer->expires > VTIMER_MAX_SLICE) { - printk("add_virt_timer: invalid timer expire value!\n"); - return -EINVAL; - } - - if (vtimer_pending(timer)) { - printk("add_virt_timer: timer pending\n"); - return -EBUSY; - } - + BUG_ON(!timer->function); + BUG_ON(!timer->expires || timer->expires > VTIMER_MAX_SLICE); + BUG_ON(vtimer_pending(timer)); timer->cpu = get_cpu(); - return 0; } /* @@ -382,10 +368,7 @@ void add_virt_timer(void *new) struct vtimer_list *timer; timer = (struct vtimer_list *)new; - - if (prepare_vtimer(timer) < 0) - return; - + prepare_vtimer(timer); timer->interval = 0; internal_add_vtimer(timer); } @@ -399,10 +382,7 @@ void add_virt_timer_periodic(void *new) struct vtimer_list *timer; timer = (struct vtimer_list *)new; - - if (prepare_vtimer(timer) < 0) - return; - + prepare_vtimer(timer); timer->interval = timer->expires; internal_add_vtimer(timer); } @@ -423,15 +403,8 @@ int mod_virt_timer(struct vtimer_list *timer, __u64 expires) unsigned long flags; int cpu; - if (!timer->function) { - printk("mod_virt_timer: uninitialized timer\n"); - return -EINVAL; - } - - if (!expires || expires > VTIMER_MAX_SLICE) { - printk("mod_virt_timer: invalid expire range\n"); - return -EINVAL; - } + BUG_ON(!timer->function); + BUG_ON(!expires || expires > VTIMER_MAX_SLICE); /* * This is a common optimization triggered by the @@ -444,6 +417,9 @@ int mod_virt_timer(struct vtimer_list *timer, __u64 expires) cpu = get_cpu(); vt_list = &per_cpu(virt_cpu_timer, cpu); + /* check if we run on the right CPU */ + BUG_ON(timer->cpu != cpu); + /* disable interrupts before test if timer is pending */ spin_lock_irqsave(&vt_list->lock, flags); @@ -458,14 +434,6 @@ int mod_virt_timer(struct vtimer_list *timer, __u64 expires) return 0; } - /* check if we run on the right CPU */ - if (timer->cpu != cpu) { - printk("mod_virt_timer: running on wrong CPU, check your code\n"); - spin_unlock_irqrestore(&vt_list->lock, flags); - put_cpu(); - return -EINVAL; - } - list_del_init(&timer->entry); timer->expires = expires; @@ -536,24 +504,6 @@ void init_cpu_vtimer(void) } -static int vtimer_idle_notify(struct notifier_block *self, - unsigned long action, void *hcpu) -{ - switch (action) { - case S390_CPU_IDLE: - stop_cpu_timer(); - break; - case S390_CPU_NOT_IDLE: - start_cpu_timer(); - break; - } - return NOTIFY_OK; -} - -static struct notifier_block vtimer_idle_nb = { - .notifier_call = vtimer_idle_notify, -}; - void __init vtime_init(void) { /* request the cpu timer external interrupt */ @@ -561,9 +511,6 @@ void __init vtime_init(void) &ext_int_info_timer) != 0) panic("Couldn't request external interrupt 0x1005"); - if (register_idle_notifier(&vtimer_idle_nb)) - panic("Couldn't register idle notifier"); - /* Enable cpu timer interrupts on the boot cpu. */ init_cpu_vtimer(); } diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c index f639a15..a0775e1 100644 --- a/arch/s390/kvm/diag.c +++ b/arch/s390/kvm/diag.c @@ -20,7 +20,7 @@ static int __diag_time_slice_end(struct kvm_vcpu *vcpu) VCPU_EVENT(vcpu, 5, "%s", "diag time slice end"); vcpu->stat.diagnose_44++; vcpu_put(vcpu); - schedule(); + yield(); vcpu_load(vcpu); return 0; } diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c index fcd1ed8..84a7fed 100644 --- a/arch/s390/kvm/interrupt.c +++ b/arch/s390/kvm/interrupt.c @@ -339,6 +339,11 @@ int kvm_s390_handle_wait(struct kvm_vcpu *vcpu) if (kvm_cpu_has_interrupt(vcpu)) return 0; + __set_cpu_idle(vcpu); + spin_lock_bh(&vcpu->arch.local_int.lock); + vcpu->arch.local_int.timer_due = 0; + spin_unlock_bh(&vcpu->arch.local_int.lock); + if (psw_interrupts_disabled(vcpu)) { VCPU_EVENT(vcpu, 3, "%s", "disabled wait"); __unset_cpu_idle(vcpu); @@ -366,8 +371,6 @@ int kvm_s390_handle_wait(struct kvm_vcpu *vcpu) no_timer: spin_lock_bh(&vcpu->arch.local_int.float_int->lock); spin_lock_bh(&vcpu->arch.local_int.lock); - __set_cpu_idle(vcpu); - vcpu->arch.local_int.timer_due = 0; add_wait_queue(&vcpu->arch.local_int.wq, &wait); while (list_empty(&vcpu->arch.local_int.list) && list_empty(&vcpu->arch.local_int.float_int->list) && diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 0ac36a6..6558b09 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -423,6 +423,8 @@ int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu, return -EINVAL; /* not implemented yet */ } +extern void s390_handle_mcck(void); + static void __vcpu_run(struct kvm_vcpu *vcpu) { memcpy(&vcpu->arch.sie_block->gg14, &vcpu->arch.guest_gprs[14], 16); @@ -430,13 +432,21 @@ static void __vcpu_run(struct kvm_vcpu *vcpu) if (need_resched()) schedule(); + if (test_thread_flag(TIF_MCCK_PENDING)) + s390_handle_mcck(); + + kvm_s390_deliver_pending_interrupts(vcpu); + vcpu->arch.sie_block->icptcode = 0; local_irq_disable(); kvm_guest_enter(); local_irq_enable(); VCPU_EVENT(vcpu, 6, "entering sie flags %x", atomic_read(&vcpu->arch.sie_block->cpuflags)); - sie64a(vcpu->arch.sie_block, vcpu->arch.guest_gprs); + if (sie64a(vcpu->arch.sie_block, vcpu->arch.guest_gprs)) { + VCPU_EVENT(vcpu, 3, "%s", "fault in sie instruction"); + kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + } VCPU_EVENT(vcpu, 6, "exit sie icptcode %d", vcpu->arch.sie_block->icptcode); local_irq_disable(); @@ -475,7 +485,6 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) might_sleep(); do { - kvm_s390_deliver_pending_interrupts(vcpu); __vcpu_run(vcpu); rc = kvm_handle_sie_intercept(vcpu); } while (!signal_pending(current) && !rc); diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c index 29f3a63..388cc74 100644 --- a/arch/s390/mm/init.c +++ b/arch/s390/mm/init.c @@ -44,37 +44,34 @@ char empty_zero_page[PAGE_SIZE] __attribute__((__aligned__(PAGE_SIZE))); void show_mem(void) { - int i, total = 0, reserved = 0; - int shared = 0, cached = 0; + unsigned long i, total = 0, reserved = 0; + unsigned long shared = 0, cached = 0; + unsigned long flags; struct page *page; + pg_data_t *pgdat; printk("Mem-info:\n"); show_free_areas(); - i = max_mapnr; - while (i-- > 0) { - if (!pfn_valid(i)) - continue; - page = pfn_to_page(i); - total++; - if (PageReserved(page)) - reserved++; - else if (PageSwapCache(page)) - cached++; - else if (page_count(page)) - shared += page_count(page) - 1; + for_each_online_pgdat(pgdat) { + pgdat_resize_lock(pgdat, &flags); + for (i = 0; i < pgdat->node_spanned_pages; i++) { + if (!pfn_valid(pgdat->node_start_pfn + i)) + continue; + page = pfn_to_page(pgdat->node_start_pfn + i); + total++; + if (PageReserved(page)) + reserved++; + else if (PageSwapCache(page)) + cached++; + else if (page_count(page)) + shared += page_count(page) - 1; + } + pgdat_resize_unlock(pgdat, &flags); } - printk("%d pages of RAM\n", total); - printk("%d reserved pages\n", reserved); - printk("%d pages shared\n", shared); - printk("%d pages swap cached\n", cached); - - printk("%lu pages dirty\n", global_page_state(NR_FILE_DIRTY)); - printk("%lu pages writeback\n", global_page_state(NR_WRITEBACK)); - printk("%lu pages mapped\n", global_page_state(NR_FILE_MAPPED)); - printk("%lu pages slab\n", - global_page_state(NR_SLAB_RECLAIMABLE) + - global_page_state(NR_SLAB_UNRECLAIMABLE)); - printk("%lu pages pagetables\n", global_page_state(NR_PAGETABLE)); + printk("%ld pages of RAM\n", total); + printk("%ld reserved pages\n", reserved); + printk("%ld pages shared\n", shared); + printk("%ld pages swap cached\n", cached); } /* @@ -205,3 +202,22 @@ void free_initrd_mem(unsigned long start, unsigned long end) } } #endif + +#ifdef CONFIG_MEMORY_HOTPLUG +int arch_add_memory(int nid, u64 start, u64 size) +{ + struct pglist_data *pgdat; + struct zone *zone; + int rc; + + pgdat = NODE_DATA(nid); + zone = pgdat->node_zones + ZONE_NORMAL; + rc = vmem_add_mapping(start, size); + if (rc) + return rc; + rc = __add_pages(zone, PFN_DOWN(start), PFN_DOWN(size)); + if (rc) + vmem_remove_mapping(start, size); + return rc; +} +#endif /* CONFIG_MEMORY_HOTPLUG */ diff --git a/arch/s390/mm/pgtable.c b/arch/s390/mm/pgtable.c index 5c1aea9..3d98ba8 100644 --- a/arch/s390/mm/pgtable.c +++ b/arch/s390/mm/pgtable.c @@ -254,36 +254,46 @@ void disable_noexec(struct mm_struct *mm, struct task_struct *tsk) int s390_enable_sie(void) { struct task_struct *tsk = current; - struct mm_struct *mm; - int rc; + struct mm_struct *mm, *old_mm; - task_lock(tsk); - - rc = 0; + /* Do we have pgstes? if yes, we are done */ if (tsk->mm->context.pgstes) - goto unlock; + return 0; - rc = -EINVAL; + /* lets check if we are allowed to replace the mm */ + task_lock(tsk); if (!tsk->mm || atomic_read(&tsk->mm->mm_users) > 1 || - tsk->mm != tsk->active_mm || tsk->mm->ioctx_list) - goto unlock; + tsk->mm != tsk->active_mm || tsk->mm->ioctx_list) { + task_unlock(tsk); + return -EINVAL; + } + task_unlock(tsk); - tsk->mm->context.pgstes = 1; /* dirty little tricks .. */ + /* we copy the mm with pgstes enabled */ + tsk->mm->context.pgstes = 1; mm = dup_mm(tsk); tsk->mm->context.pgstes = 0; - - rc = -ENOMEM; if (!mm) - goto unlock; - mmput(tsk->mm); + return -ENOMEM; + + /* Now lets check again if somebody attached ptrace etc */ + task_lock(tsk); + if (!tsk->mm || atomic_read(&tsk->mm->mm_users) > 1 || + tsk->mm != tsk->active_mm || tsk->mm->ioctx_list) { + mmput(mm); + task_unlock(tsk); + return -EINVAL; + } + + /* ok, we are alone. No ptrace, no threads, etc. */ + old_mm = tsk->mm; tsk->mm = tsk->active_mm = mm; preempt_disable(); update_mm(mm, tsk); cpu_set(smp_processor_id(), mm->cpu_vm_mask); preempt_enable(); - rc = 0; -unlock: task_unlock(tsk); - return rc; + mmput(old_mm); + return 0; } EXPORT_SYMBOL_GPL(s390_enable_sie); diff --git a/arch/s390/mm/vmem.c b/arch/s390/mm/vmem.c index ea28048..e4868bf 100644 --- a/arch/s390/mm/vmem.c +++ b/arch/s390/mm/vmem.c @@ -27,12 +27,19 @@ struct memory_segment { static LIST_HEAD(mem_segs); -static pud_t *vmem_pud_alloc(void) +static void __ref *vmem_alloc_pages(unsigned int order) +{ + if (slab_is_available()) + return (void *)__get_free_pages(GFP_KERNEL, order); + return alloc_bootmem_pages((1 << order) * PAGE_SIZE); +} + +static inline pud_t *vmem_pud_alloc(void) { pud_t *pud = NULL; #ifdef CONFIG_64BIT - pud = vmemmap_alloc_block(PAGE_SIZE * 4, 0); + pud = vmem_alloc_pages(2); if (!pud) return NULL; clear_table((unsigned long *) pud, _REGION3_ENTRY_EMPTY, PAGE_SIZE * 4); @@ -40,12 +47,12 @@ static pud_t *vmem_pud_alloc(void) return pud; } -static pmd_t *vmem_pmd_alloc(void) +static inline pmd_t *vmem_pmd_alloc(void) { pmd_t *pmd = NULL; #ifdef CONFIG_64BIT - pmd = vmemmap_alloc_block(PAGE_SIZE * 4, 0); + pmd = vmem_alloc_pages(2); if (!pmd) return NULL; clear_table((unsigned long *) pmd, _SEGMENT_ENTRY_EMPTY, PAGE_SIZE * 4); @@ -207,13 +214,14 @@ int __meminit vmemmap_populate(struct page *start, unsigned long nr, int node) if (pte_none(*pt_dir)) { unsigned long new_page; - new_page =__pa(vmemmap_alloc_block(PAGE_SIZE, 0)); + new_page =__pa(vmem_alloc_pages(0)); if (!new_page) goto out; pte = pfn_pte(new_page >> PAGE_SHIFT, PAGE_KERNEL); *pt_dir = pte; } } + memset(start, 0, nr * sizeof(struct page)); ret = 0; out: flush_tlb_kernel_range(start_addr, end_addr); @@ -228,7 +236,7 @@ static int insert_memory_segment(struct memory_segment *seg) { struct memory_segment *tmp; - if (seg->start + seg->size >= VMEM_MAX_PHYS || + if (seg->start + seg->size > VMEM_MAX_PHYS || seg->start + seg->size < seg->start) return -ERANGE; |