diff options
Diffstat (limited to 'arch/s390')
28 files changed, 1680 insertions, 281 deletions
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 01c5c082f97..821a141889d 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -446,6 +446,14 @@ config NO_IDLE_HZ_INIT The HZ timer is switched off in idle by default. That means the HZ timer is already disabled at boot time. +config S390_HYPFS_FS + bool "s390 hypervisor file system support" + select SYS_HYPERVISOR + default y + help + This is a virtual file system intended to provide accounting + information in an s390 hypervisor environment. + config KEXEC bool "kexec system call (EXPERIMENTAL)" depends on EXPERIMENTAL diff --git a/arch/s390/Makefile b/arch/s390/Makefile index 7bb16fb97d4..b3791fb094a 100644 --- a/arch/s390/Makefile +++ b/arch/s390/Makefile @@ -76,7 +76,7 @@ LDFLAGS_vmlinux := -e start head-y := arch/$(ARCH)/kernel/head.o arch/$(ARCH)/kernel/init_task.o core-y += arch/$(ARCH)/mm/ arch/$(ARCH)/kernel/ arch/$(ARCH)/crypto/ \ - arch/$(ARCH)/appldata/ + arch/$(ARCH)/appldata/ arch/$(ARCH)/hypfs/ libs-y += arch/$(ARCH)/lib/ drivers-y += drivers/s390/ drivers-$(CONFIG_MATHEMU) += arch/$(ARCH)/math-emu/ diff --git a/arch/s390/appldata/appldata.h b/arch/s390/appldata/appldata.h index e806a8922bb..71d65eb3065 100644 --- a/arch/s390/appldata/appldata.h +++ b/arch/s390/appldata/appldata.h @@ -3,9 +3,9 @@ * * Definitions and interface for Linux - z/VM Monitor Stream. * - * Copyright (C) 2003 IBM Corporation, IBM Deutschland Entwicklung GmbH. + * Copyright (C) 2003,2006 IBM Corporation, IBM Deutschland Entwicklung GmbH. * - * Author: Gerald Schaefer <geraldsc@de.ibm.com> + * Author: Gerald Schaefer <gerald.schaefer@de.ibm.com> */ //#define APPLDATA_DEBUG /* Debug messages on/off */ @@ -29,6 +29,22 @@ #define CTL_APPLDATA_NET_SUM 2125 #define CTL_APPLDATA_PROC 2126 +#ifndef CONFIG_64BIT + +#define APPLDATA_START_INTERVAL_REC 0x00 /* Function codes for */ +#define APPLDATA_STOP_REC 0x01 /* DIAG 0xDC */ +#define APPLDATA_GEN_EVENT_RECORD 0x02 +#define APPLDATA_START_CONFIG_REC 0x03 + +#else + +#define APPLDATA_START_INTERVAL_REC 0x80 +#define APPLDATA_STOP_REC 0x81 +#define APPLDATA_GEN_EVENT_RECORD 0x82 +#define APPLDATA_START_CONFIG_REC 0x83 + +#endif /* CONFIG_64BIT */ + #define P_INFO(x...) printk(KERN_INFO MY_PRINT_NAME " info: " x) #define P_ERROR(x...) printk(KERN_ERR MY_PRINT_NAME " error: " x) #define P_WARNING(x...) printk(KERN_WARNING MY_PRINT_NAME " status: " x) @@ -53,7 +69,11 @@ struct appldata_ops { void *data; /* record data */ unsigned int size; /* size of record */ struct module *owner; /* THIS_MODULE */ + char mod_lvl[2]; /* modification level, EBCDIC */ }; extern int appldata_register_ops(struct appldata_ops *ops); extern void appldata_unregister_ops(struct appldata_ops *ops); +extern int appldata_diag(char record_nr, u16 function, unsigned long buffer, + u16 length, char *mod_lvl); + diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c index 9a22434a580..61bc44626c0 100644 --- a/arch/s390/appldata/appldata_base.c +++ b/arch/s390/appldata/appldata_base.c @@ -5,9 +5,9 @@ * Exports appldata_register_ops() and appldata_unregister_ops() for the * data gathering modules. * - * Copyright (C) 2003 IBM Corporation, IBM Deutschland Entwicklung GmbH. + * Copyright (C) 2003,2006 IBM Corporation, IBM Deutschland Entwicklung GmbH. * - * Author: Gerald Schaefer <geraldsc@de.ibm.com> + * Author: Gerald Schaefer <gerald.schaefer@de.ibm.com> */ #include <linux/config.h> @@ -40,22 +40,6 @@ #define TOD_MICRO 0x01000 /* nr. of TOD clock units for 1 microsecond */ -#ifndef CONFIG_64BIT - -#define APPLDATA_START_INTERVAL_REC 0x00 /* Function codes for */ -#define APPLDATA_STOP_REC 0x01 /* DIAG 0xDC */ -#define APPLDATA_GEN_EVENT_RECORD 0x02 -#define APPLDATA_START_CONFIG_REC 0x03 - -#else - -#define APPLDATA_START_INTERVAL_REC 0x80 -#define APPLDATA_STOP_REC 0x81 -#define APPLDATA_GEN_EVENT_RECORD 0x82 -#define APPLDATA_START_CONFIG_REC 0x83 - -#endif /* CONFIG_64BIT */ - /* * Parameter list for DIAGNOSE X'DC' @@ -195,8 +179,8 @@ static void appldata_work_fn(void *data) * * prepare parameter list, issue DIAG 0xDC */ -static int appldata_diag(char record_nr, u16 function, unsigned long buffer, - u16 length) +int appldata_diag(char record_nr, u16 function, unsigned long buffer, + u16 length, char *mod_lvl) { unsigned long ry; struct appldata_product_id { @@ -214,7 +198,7 @@ static int appldata_diag(char record_nr, u16 function, unsigned long buffer, .record_nr = record_nr, .version_nr = {0xF2, 0xF6}, /* "26" */ .release_nr = {0xF0, 0xF1}, /* "01" */ - .mod_lvl = {0xF0, 0xF0}, /* "00" */ + .mod_lvl = {mod_lvl[0], mod_lvl[1]}, }; struct appldata_parameter_list appldata_parameter_list = { .diag = 0xDC, @@ -467,24 +451,25 @@ appldata_generic_handler(ctl_table *ctl, int write, struct file *filp, module_put(ops->owner); return -ENODEV; } - ops->active = 1; ops->callback(ops->data); // init record rc = appldata_diag(ops->record_nr, APPLDATA_START_INTERVAL_REC, - (unsigned long) ops->data, ops->size); + (unsigned long) ops->data, ops->size, + ops->mod_lvl); if (rc != 0) { P_ERROR("START DIAG 0xDC for %s failed, " "return code: %d\n", ops->name, rc); module_put(ops->owner); - ops->active = 0; } else { P_INFO("Monitoring %s data enabled, " "DIAG 0xDC started.\n", ops->name); + 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); + (unsigned long) ops->data, ops->size, + ops->mod_lvl); if (rc != 0) { P_ERROR("STOP DIAG 0xDC for %s failed, " "return code: %d\n", ops->name, rc); @@ -633,7 +618,7 @@ appldata_offline_cpu(int cpu) spin_unlock(&appldata_timer_lock); } -static int +static int __cpuinit appldata_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { @@ -652,7 +637,7 @@ appldata_cpu_notify(struct notifier_block *self, return NOTIFY_OK; } -static struct notifier_block appldata_nb = { +static struct notifier_block __devinitdata appldata_nb = { .notifier_call = appldata_cpu_notify, }; @@ -710,7 +695,8 @@ static void __exit appldata_exit(void) list_for_each(lh, &appldata_ops_list) { ops = list_entry(lh, struct appldata_ops, list); rc = appldata_diag(ops->record_nr, APPLDATA_STOP_REC, - (unsigned long) ops->data, ops->size); + (unsigned long) ops->data, ops->size, + ops->mod_lvl); if (rc != 0) { P_ERROR("STOP DIAG 0xDC for %s failed, " "return code: %d\n", ops->name, rc); @@ -739,6 +725,7 @@ MODULE_DESCRIPTION("Linux-VM Monitor Stream, base infrastructure"); EXPORT_SYMBOL_GPL(appldata_register_ops); EXPORT_SYMBOL_GPL(appldata_unregister_ops); +EXPORT_SYMBOL_GPL(appldata_diag); #ifdef MODULE /* @@ -779,7 +766,6 @@ unsigned long nr_iowait(void) #endif /* MODULE */ EXPORT_SYMBOL_GPL(si_swapinfo); EXPORT_SYMBOL_GPL(nr_threads); -EXPORT_SYMBOL_GPL(avenrun); EXPORT_SYMBOL_GPL(get_full_page_state); 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 f0e2fbed3d4..7915a197d96 100644 --- a/arch/s390/appldata/appldata_mem.c +++ b/arch/s390/appldata/appldata_mem.c @@ -4,9 +4,9 @@ * Data gathering module for Linux-VM Monitor Stream, Stage 1. * Collects data related to memory management. * - * Copyright (C) 2003 IBM Corporation, IBM Deutschland Entwicklung GmbH. + * Copyright (C) 2003,2006 IBM Corporation, IBM Deutschland Entwicklung GmbH. * - * Author: Gerald Schaefer <geraldsc@de.ibm.com> + * Author: Gerald Schaefer <gerald.schaefer@de.ibm.com> */ #include <linux/config.h> @@ -152,6 +152,7 @@ static struct appldata_ops ops = { .callback = &appldata_get_mem_data, .data = &appldata_mem_data, .owner = THIS_MODULE, + .mod_lvl = {0xF0, 0xF0}, /* EBCDIC "00" */ }; diff --git a/arch/s390/appldata/appldata_net_sum.c b/arch/s390/appldata/appldata_net_sum.c index 2a4c7432db4..39b7bdecbf0 100644 --- a/arch/s390/appldata/appldata_net_sum.c +++ b/arch/s390/appldata/appldata_net_sum.c @@ -5,9 +5,9 @@ * Collects accumulated network statistics (Packets received/transmitted, * dropped, errors, ...). * - * Copyright (C) 2003 IBM Corporation, IBM Deutschland Entwicklung GmbH. + * Copyright (C) 2003,2006 IBM Corporation, IBM Deutschland Entwicklung GmbH. * - * Author: Gerald Schaefer <geraldsc@de.ibm.com> + * Author: Gerald Schaefer <gerald.schaefer@de.ibm.com> */ #include <linux/config.h> @@ -152,6 +152,7 @@ static struct appldata_ops ops = { .callback = &appldata_get_net_sum_data, .data = &appldata_net_sum_data, .owner = THIS_MODULE, + .mod_lvl = {0xF0, 0xF0}, /* EBCDIC "00" */ }; diff --git a/arch/s390/appldata/appldata_os.c b/arch/s390/appldata/appldata_os.c index 99ddd3bf2fb..f2b44a2f1de 100644 --- a/arch/s390/appldata/appldata_os.c +++ b/arch/s390/appldata/appldata_os.c @@ -4,9 +4,9 @@ * Data gathering module for Linux-VM Monitor Stream, Stage 1. * Collects misc. OS related data (CPU utilization, running processes). * - * Copyright (C) 2003 IBM Corporation, IBM Deutschland Entwicklung GmbH. + * Copyright (C) 2003,2006 IBM Corporation, IBM Deutschland Entwicklung GmbH. * - * Author: Gerald Schaefer <geraldsc@de.ibm.com> + * Author: Gerald Schaefer <gerald.schaefer@de.ibm.com> */ #include <linux/config.h> @@ -44,11 +44,14 @@ struct appldata_os_per_cpu { u32 per_cpu_system; /* ... spent in kernel mode */ u32 per_cpu_idle; /* ... spent in idle mode */ -// New in 2.6 --> + /* New in 2.6 */ u32 per_cpu_irq; /* ... spent in interrupts */ u32 per_cpu_softirq; /* ... spent in softirqs */ u32 per_cpu_iowait; /* ... spent while waiting for I/O */ -// <-- New in 2.6 + + /* New in modification level 01 */ + u32 per_cpu_steal; /* ... stolen by hypervisor */ + u32 cpu_id; /* number of this CPU */ } __attribute__((packed)); struct appldata_os_data { @@ -68,10 +71,9 @@ struct appldata_os_data { u32 avenrun[3]; /* average nr. of running processes during */ /* the last 1, 5 and 15 minutes */ -// New in 2.6 --> + /* New in 2.6 */ u32 nr_iowait; /* number of blocked threads (waiting for I/O) */ -// <-- New in 2.6 /* per cpu data */ struct appldata_os_per_cpu os_cpu[0]; @@ -79,6 +81,14 @@ struct appldata_os_data { static struct appldata_os_data *appldata_os_data; +static struct appldata_ops ops = { + .ctl_nr = CTL_APPLDATA_OS, + .name = "os", + .record_nr = APPLDATA_RECORD_OS_ID, + .owner = THIS_MODULE, + .mod_lvl = {0xF0, 0xF1}, /* EBCDIC "01" */ +}; + static inline void appldata_print_debug(struct appldata_os_data *os_data) { @@ -100,15 +110,17 @@ static inline void appldata_print_debug(struct appldata_os_data *os_data) 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\n", - i, + "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_iowait, + os_data->os_cpu[i].per_cpu_steal); } P_DEBUG("sync_count_1 = %u\n", os_data->sync_count_1); @@ -123,14 +135,13 @@ static inline void appldata_print_debug(struct appldata_os_data *os_data) */ static void appldata_get_os_data(void *data) { - int i, j; + int i, j, rc; struct appldata_os_data *os_data; + unsigned int new_size; os_data = data; os_data->sync_count_1++; - os_data->nr_cpus = num_online_cpus(); - os_data->nr_threads = nr_threads; os_data->nr_running = nr_running(); os_data->nr_iowait = nr_iowait(); @@ -154,9 +165,44 @@ static void appldata_get_os_data(void *data) cputime_to_jiffies(kstat_cpu(i).cpustat.softirq); os_data->os_cpu[j].per_cpu_iowait = cputime_to_jiffies(kstat_cpu(i).cpustat.iowait); + os_data->os_cpu[j].per_cpu_steal = + cputime_to_jiffies(kstat_cpu(i).cpustat.steal); + os_data->os_cpu[j].cpu_id = i; j++; } + os_data->nr_cpus = j; + + new_size = sizeof(struct appldata_os_data) + + (os_data->nr_cpus * sizeof(struct appldata_os_per_cpu)); + if (ops.size != new_size) { + if (ops.active) { + rc = appldata_diag(APPLDATA_RECORD_OS_ID, + APPLDATA_START_INTERVAL_REC, + (unsigned long) ops.data, new_size, + ops.mod_lvl); + 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, + (unsigned long) ops.data, ops.size, + ops.mod_lvl); + if (rc != 0) + P_ERROR("os: STOP OLD DIAG 0xDC failed, " + "return code: %d, old size = %i\n", rc, + ops.size); + else + P_INFO("os: old record size = %i stopped\n", + ops.size); + } + ops.size = new_size; + } os_data->timestamp = get_clock(); os_data->sync_count_2++; #ifdef APPLDATA_DEBUG @@ -165,15 +211,6 @@ static void appldata_get_os_data(void *data) } -static struct appldata_ops ops = { - .ctl_nr = CTL_APPLDATA_OS, - .name = "os", - .record_nr = APPLDATA_RECORD_OS_ID, - .callback = &appldata_get_os_data, - .owner = THIS_MODULE, -}; - - /* * appldata_os_init() * @@ -181,26 +218,25 @@ static struct appldata_ops ops = { */ static int __init appldata_os_init(void) { - int rc, size; + int rc, max_size; - size = sizeof(struct appldata_os_data) + - (NR_CPUS * sizeof(struct appldata_os_per_cpu)); - if (size > APPLDATA_MAX_REC_SIZE) { - P_ERROR("Size of record = %i, bigger than maximum (%i)!\n", - size, APPLDATA_MAX_REC_SIZE); + max_size = sizeof(struct appldata_os_data) + + (NR_CPUS * sizeof(struct appldata_os_per_cpu)); + if (max_size > APPLDATA_MAX_REC_SIZE) { + P_ERROR("Max. size of OS record = %i, bigger than maximum " + "record size (%i)\n", max_size, APPLDATA_MAX_REC_SIZE); rc = -ENOMEM; goto out; } - P_DEBUG("sizeof(os) = %i, sizeof(os_cpu) = %lu\n", size, + P_DEBUG("max. sizeof(os) = %i, sizeof(os_cpu) = %lu\n", max_size, sizeof(struct appldata_os_per_cpu)); - appldata_os_data = kmalloc(size, GFP_DMA); + 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; } - memset(appldata_os_data, 0, size); appldata_os_data->per_cpu_size = sizeof(struct appldata_os_per_cpu); appldata_os_data->cpu_offset = offsetof(struct appldata_os_data, @@ -208,7 +244,7 @@ static int __init appldata_os_init(void) P_DEBUG("cpu offset = %u\n", appldata_os_data->cpu_offset); ops.data = appldata_os_data; - ops.size = size; + ops.callback = &appldata_get_os_data; rc = appldata_register_ops(&ops); if (rc != 0) { P_ERROR("Error registering ops, rc = %i\n", rc); diff --git a/arch/s390/crypto/aes_s390.c b/arch/s390/crypto/aes_s390.c index c5ca2dc5d42..5713c7e5bd1 100644 --- a/arch/s390/crypto/aes_s390.c +++ b/arch/s390/crypto/aes_s390.c @@ -37,10 +37,10 @@ struct s390_aes_ctx { int key_len; }; -static int aes_set_key(void *ctx, const u8 *in_key, unsigned int key_len, - u32 *flags) +static int aes_set_key(struct crypto_tfm *tfm, const u8 *in_key, + unsigned int key_len, u32 *flags) { - struct s390_aes_ctx *sctx = ctx; + struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); switch (key_len) { case 16: @@ -70,9 +70,9 @@ fail: return -EINVAL; } -static void aes_encrypt(void *ctx, u8 *out, const u8 *in) +static void aes_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in) { - const struct s390_aes_ctx *sctx = ctx; + const struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); switch (sctx->key_len) { case 16: @@ -90,9 +90,9 @@ static void aes_encrypt(void *ctx, u8 *out, const u8 *in) } } -static void aes_decrypt(void *ctx, u8 *out, const u8 *in) +static void aes_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in) { - const struct s390_aes_ctx *sctx = ctx; + const struct s390_aes_ctx *sctx = crypto_tfm_ctx(tfm); switch (sctx->key_len) { case 16: diff --git a/arch/s390/crypto/des_s390.c b/arch/s390/crypto/des_s390.c index e3c37aa0a19..b3f7496a79b 100644 --- a/arch/s390/crypto/des_s390.c +++ b/arch/s390/crypto/des_s390.c @@ -44,10 +44,10 @@ struct crypt_s390_des3_192_ctx { u8 key[DES3_192_KEY_SIZE]; }; -static int des_setkey(void *ctx, const u8 *key, unsigned int keylen, - u32 *flags) +static int des_setkey(struct crypto_tfm *tfm, const u8 *key, + unsigned int keylen, u32 *flags) { - struct crypt_s390_des_ctx *dctx = ctx; + struct crypt_s390_des_ctx *dctx = crypto_tfm_ctx(tfm); int ret; /* test if key is valid (not a weak key) */ @@ -57,16 +57,16 @@ static int des_setkey(void *ctx, const u8 *key, unsigned int keylen, return ret; } -static void des_encrypt(void *ctx, u8 *out, const u8 *in) +static void des_encrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in) { - struct crypt_s390_des_ctx *dctx = ctx; + struct crypt_s390_des_ctx *dctx = crypto_tfm_ctx(tfm); crypt_s390_km(KM_DEA_ENCRYPT, dctx->key, out, in, DES_BLOCK_SIZE); } -static void des_decrypt(void *ctx, u8 *out, const u8 *in) +static void des_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in) { - struct crypt_s390_des_ctx *dctx = ctx; + struct crypt_s390_des_ctx *dctx = crypto_tfm_ctx(tfm); crypt_s390_km(KM_DEA_DECRYPT, dctx->key, out, in, DES_BLOCK_SIZE); } @@ -166,11 +166,11 @@ static struct crypto_alg des_alg = { * Implementers MUST reject keys that exhibit this property. * */ -static int des3_128_setkey(void *ctx, const u8 *key, unsigned int keylen, - u32 *flags) +static int des3_128_setkey(struct crypto_tfm *tfm, const u8 *key, + unsigned int keylen, u32 *flags) { int i, ret; - struct crypt_s390_des3_128_ctx *dctx = ctx; + struct crypt_s390_des3_128_ctx *dctx = crypto_tfm_ctx(tfm); const u8* temp_key = key; if (!(memcmp(key, &key[DES_KEY_SIZE], DES_KEY_SIZE))) { @@ -186,17 +186,17 @@ static int des3_128_setkey(void *ctx, const u8 *key, unsigned int keylen, return 0; } -static void des3_128_encrypt(void *ctx, u8 *dst, const u8 *src) +static void des3_128_encrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) { - struct crypt_s390_des3_128_ctx *dctx = ctx; + struct crypt_s390_des3_128_ctx *dctx = crypto_tfm_ctx(tfm); crypt_s390_km(KM_TDEA_128_ENCRYPT, dctx->key, dst, (void*)src, DES3_128_BLOCK_SIZE); } -static void des3_128_decrypt(void *ctx, u8 *dst, const u8 *src) +static void des3_128_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) { - struct crypt_s390_des3_128_ctx *dctx = ctx; + struct crypt_s390_des3_128_ctx *dctx = crypto_tfm_ctx(tfm); crypt_s390_km(KM_TDEA_128_DECRYPT, dctx->key, dst, (void*)src, DES3_128_BLOCK_SIZE); @@ -302,11 +302,11 @@ static struct crypto_alg des3_128_alg = { * property. * */ -static int des3_192_setkey(void *ctx, const u8 *key, unsigned int keylen, - u32 *flags) +static int des3_192_setkey(struct crypto_tfm *tfm, const u8 *key, + unsigned int keylen, u32 *flags) { int i, ret; - struct crypt_s390_des3_192_ctx *dctx = ctx; + struct crypt_s390_des3_192_ctx *dctx = crypto_tfm_ctx(tfm); const u8* temp_key = key; if (!(memcmp(key, &key[DES_KEY_SIZE], DES_KEY_SIZE) && @@ -325,17 +325,17 @@ static int des3_192_setkey(void *ctx, const u8 *key, unsigned int keylen, return 0; } -static void des3_192_encrypt(void *ctx, u8 *dst, const u8 *src) +static void des3_192_encrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) { - struct crypt_s390_des3_192_ctx *dctx = ctx; + struct crypt_s390_des3_192_ctx *dctx = crypto_tfm_ctx(tfm); crypt_s390_km(KM_TDEA_192_ENCRYPT, dctx->key, dst, (void*)src, DES3_192_BLOCK_SIZE); } -static void des3_192_decrypt(void *ctx, u8 *dst, const u8 *src) +static void des3_192_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) { - struct crypt_s390_des3_192_ctx *dctx = ctx; + struct crypt_s390_des3_192_ctx *dctx = crypto_tfm_ctx(tfm); crypt_s390_km(KM_TDEA_192_DECRYPT, dctx->key, dst, (void*)src, DES3_192_BLOCK_SIZE); diff --git a/arch/s390/crypto/sha1_s390.c b/arch/s390/crypto/sha1_s390.c index 98c896b86dc..9d34a35b1aa 100644 --- a/arch/s390/crypto/sha1_s390.c +++ b/arch/s390/crypto/sha1_s390.c @@ -40,28 +40,29 @@ struct crypt_s390_sha1_ctx { u8 buffer[2 * SHA1_BLOCK_SIZE]; }; -static void -sha1_init(void *ctx) +static void sha1_init(struct crypto_tfm *tfm) { - static const struct crypt_s390_sha1_ctx initstate = { - .state = { - 0x67452301, - 0xEFCDAB89, - 0x98BADCFE, - 0x10325476, - 0xC3D2E1F0 - }, + struct crypt_s390_sha1_ctx *ctx = crypto_tfm_ctx(tfm); + static const u32 initstate[5] = { + 0x67452301, + 0xEFCDAB89, + 0x98BADCFE, + 0x10325476, + 0xC3D2E1F0 }; - memcpy(ctx, &initstate, sizeof(initstate)); + + ctx->count = 0; + memcpy(ctx->state, &initstate, sizeof(initstate)); + ctx->buf_len = 0; } -static void -sha1_update(void *ctx, const u8 *data, unsigned int len) +static void sha1_update(struct crypto_tfm *tfm, const u8 *data, + unsigned int len) { struct crypt_s390_sha1_ctx *sctx; long imd_len; - sctx = ctx; + sctx = crypto_tfm_ctx(tfm); sctx->count += len * 8; //message bit length //anything in buffer yet? -> must be completed @@ -110,10 +111,9 @@ pad_message(struct crypt_s390_sha1_ctx* sctx) } /* Add padding and return the message digest. */ -static void -sha1_final(void* ctx, u8 *out) +static void sha1_final(struct crypto_tfm *tfm, u8 *out) { - struct crypt_s390_sha1_ctx *sctx = ctx; + struct crypt_s390_sha1_ctx *sctx = crypto_tfm_ctx(tfm); //must perform manual padding pad_message(sctx); diff --git a/arch/s390/crypto/sha256_s390.c b/arch/s390/crypto/sha256_s390.c index 1ec5e92b345..f573df30f31 100644 --- a/arch/s390/crypto/sha256_s390.c +++ b/arch/s390/crypto/sha256_s390.c @@ -31,9 +31,9 @@ struct s390_sha256_ctx { u8 buf[2 * SHA256_BLOCK_SIZE]; }; -static void sha256_init(void *ctx) +static void sha256_init(struct crypto_tfm *tfm) { - struct s390_sha256_ctx *sctx = ctx; + struct s390_sha256_ctx *sctx = crypto_tfm_ctx(tfm); sctx->state[0] = 0x6a09e667; sctx->state[1] = 0xbb67ae85; @@ -44,12 +44,12 @@ static void sha256_init(void *ctx) sctx->state[6] = 0x1f83d9ab; sctx->state[7] = 0x5be0cd19; sctx->count = 0; - memset(sctx->buf, 0, sizeof(sctx->buf)); } -static void sha256_update(void *ctx, const u8 *data, unsigned int len) +static void sha256_update(struct crypto_tfm *tfm, const u8 *data, + unsigned int len) { - struct s390_sha256_ctx *sctx = ctx; + struct s390_sha256_ctx *sctx = crypto_tfm_ctx(tfm); unsigned int index; int ret; @@ -108,9 +108,9 @@ static void pad_message(struct s390_sha256_ctx* sctx) } /* Add padding and return the message digest */ -static void sha256_final(void* ctx, u8 *out) +static void sha256_final(struct crypto_tfm *tfm, u8 *out) { - struct s390_sha256_ctx *sctx = ctx; + struct s390_sha256_ctx *sctx = crypto_tfm_ctx(tfm); /* must perform manual padding */ pad_message(sctx); diff --git a/arch/s390/hypfs/Makefile b/arch/s390/hypfs/Makefile new file mode 100644 index 00000000000..f4b00cd81f7 --- /dev/null +++ b/arch/s390/hypfs/Makefile @@ -0,0 +1,7 @@ +# +# Makefile for the linux hypfs filesystem routines. +# + +obj-$(CONFIG_S390_HYPFS_FS) += s390_hypfs.o + +s390_hypfs-objs := inode.o hypfs_diag.o diff --git a/arch/s390/hypfs/hypfs.h b/arch/s390/hypfs/hypfs.h new file mode 100644 index 00000000000..ea5567be00f --- /dev/null +++ b/arch/s390/hypfs/hypfs.h @@ -0,0 +1,30 @@ +/* + * fs/hypfs/hypfs.h + * Hypervisor filesystem for Linux on s390. + * + * Copyright (C) IBM Corp. 2006 + * Author(s): Michael Holzheu <holzheu@de.ibm.com> + */ + +#ifndef _HYPFS_H_ +#define _HYPFS_H_ + +#include <linux/fs.h> +#include <linux/types.h> + +#define REG_FILE_MODE 0440 +#define UPDATE_FILE_MODE 0220 +#define DIR_MODE 0550 + +extern struct dentry *hypfs_mkdir(struct super_block *sb, struct dentry *parent, + const char *name); + +extern struct dentry *hypfs_create_u64(struct super_block *sb, + struct dentry *dir, const char *name, + __u64 value); + +extern struct dentry *hypfs_create_str(struct super_block *sb, + struct dentry *dir, const char *name, + char *string); + +#endif /* _HYPFS_H_ */ diff --git a/arch/s390/hypfs/hypfs_diag.c b/arch/s390/hypfs/hypfs_diag.c new file mode 100644 index 00000000000..efa74af7f04 --- /dev/null +++ b/arch/s390/hypfs/hypfs_diag.c @@ -0,0 +1,696 @@ +/* + * fs/hypfs/hypfs_diag.c + * Hypervisor filesystem for Linux on s390. Diag 204 and 224 + * implementation. + * + * Copyright (C) IBM Corp. 2006 + * Author(s): Michael Holzheu <holzheu@de.ibm.com> + */ + +#include <linux/types.h> +#include <linux/errno.h> +#include <linux/string.h> +#include <linux/vmalloc.h> +#include <asm/ebcdic.h> +#include "hypfs.h" + +#define LPAR_NAME_LEN 8 /* lpar name len in diag 204 data */ +#define CPU_NAME_LEN 16 /* type name len of cpus in diag224 name table */ +#define TMP_SIZE 64 /* size of temporary buffers */ + +/* diag 204 subcodes */ +enum diag204_sc { + SUBC_STIB4 = 4, + SUBC_RSI = 5, + SUBC_STIB6 = 6, + SUBC_STIB7 = 7 +}; + +/* The two available diag 204 data formats */ +enum diag204_format { + INFO_SIMPLE = 0, + INFO_EXT = 0x00010000 +}; + +/* bit is set in flags, when physical cpu info is included in diag 204 data */ +#define LPAR_PHYS_FLG 0x80 + +static char *diag224_cpu_names; /* diag 224 name table */ +static enum diag204_sc diag204_store_sc; /* used subcode for store */ +static enum diag204_format diag204_info_type; /* used diag 204 data format */ + +static void *diag204_buf; /* 4K aligned buffer for diag204 data */ +static void *diag204_buf_vmalloc; /* vmalloc pointer for diag204 data */ +static int diag204_buf_pages; /* number of pages for diag204 data */ + +/* + * DIAG 204 data structures and member access functions. + * + * Since we have two different diag 204 data formats for old and new s390 + * machines, we do not access the structs directly, but use getter functions for + * each struct member instead. This should make the code more readable. + */ + +/* Time information block */ + +struct info_blk_hdr { + __u8 npar; + __u8 flags; + __u16 tslice; + __u16 phys_cpus; + __u16 this_part; + __u64 curtod; +} __attribute__ ((packed)); + +struct x_info_blk_hdr { + __u8 npar; + __u8 flags; + __u16 tslice; + __u16 phys_cpus; + __u16 this_part; + __u64 curtod1; + __u64 curtod2; + char reserved[40]; +} __attribute__ ((packed)); + +static inline int info_blk_hdr__size(enum diag204_format type) +{ + if (type == INFO_SIMPLE) + return sizeof(struct info_blk_hdr); + else /* INFO_EXT */ + return sizeof(struct x_info_blk_hdr); +} + +static inline __u8 info_blk_hdr__npar(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct info_blk_hdr *)hdr)->npar; + else /* INFO_EXT */ + return ((struct x_info_blk_hdr *)hdr)->npar; +} + +static inline __u8 info_blk_hdr__flags(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct info_blk_hdr *)hdr)->flags; + else /* INFO_EXT */ + return ((struct x_info_blk_hdr *)hdr)->flags; +} + +static inline __u16 info_blk_hdr__pcpus(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct info_blk_hdr *)hdr)->phys_cpus; + else /* INFO_EXT */ + return ((struct x_info_blk_hdr *)hdr)->phys_cpus; +} + +/* Partition header */ + +struct part_hdr { + __u8 pn; + __u8 cpus; + char reserved[6]; + char part_name[LPAR_NAME_LEN]; +} __attribute__ ((packed)); + +struct x_part_hdr { + __u8 pn; + __u8 cpus; + __u8 rcpus; + __u8 pflag; + __u32 mlu; + char part_name[LPAR_NAME_LEN]; + char lpc_name[8]; + char os_name[8]; + __u64 online_cs; + __u64 online_es; + __u8 upid; + char reserved1[3]; + __u32 group_mlu; + char group_name[8]; + char reserved2[32]; +} __attribute__ ((packed)); + +static inline int part_hdr__size(enum diag204_format type) +{ + if (type == INFO_SIMPLE) + return sizeof(struct part_hdr); + else /* INFO_EXT */ + return sizeof(struct x_part_hdr); +} + +static inline __u8 part_hdr__rcpus(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct part_hdr *)hdr)->cpus; + else /* INFO_EXT */ + return ((struct x_part_hdr *)hdr)->rcpus; +} + +static inline void part_hdr__part_name(enum diag204_format type, void *hdr, + char *name) +{ + if (type == INFO_SIMPLE) + memcpy(name, ((struct part_hdr *)hdr)->part_name, + LPAR_NAME_LEN); + else /* INFO_EXT */ + memcpy(name, ((struct x_part_hdr *)hdr)->part_name, + LPAR_NAME_LEN); + EBCASC(name, LPAR_NAME_LEN); + name[LPAR_NAME_LEN] = 0; + strstrip(name); +} + +struct cpu_info { + __u16 cpu_addr; + char reserved1[2]; + __u8 ctidx; + __u8 cflag; + __u16 weight; + __u64 acc_time; + __u64 lp_time; +} __attribute__ ((packed)); + +struct x_cpu_info { + __u16 cpu_addr; + char reserved1[2]; + __u8 ctidx; + __u8 cflag; + __u16 weight; + __u64 acc_time; + __u64 lp_time; + __u16 min_weight; + __u16 cur_weight; + __u16 max_weight; + char reseved2[2]; + __u64 online_time; + __u64 wait_time; + __u32 pma_weight; + __u32 polar_weight; + char reserved3[40]; +} __attribute__ ((packed)); + +/* CPU info block */ + +static inline int cpu_info__size(enum diag204_format type) +{ + if (type == INFO_SIMPLE) + return sizeof(struct cpu_info); + else /* INFO_EXT */ + return sizeof(struct x_cpu_info); +} + +static inline __u8 cpu_info__ctidx(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct cpu_info *)hdr)->ctidx; + else /* INFO_EXT */ + return ((struct x_cpu_info *)hdr)->ctidx; +} + +static inline __u16 cpu_info__cpu_addr(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct cpu_info *)hdr)->cpu_addr; + else /* INFO_EXT */ + return ((struct x_cpu_info *)hdr)->cpu_addr; +} + +static inline __u64 cpu_info__acc_time(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct cpu_info *)hdr)->acc_time; + else /* INFO_EXT */ + return ((struct x_cpu_info *)hdr)->acc_time; +} + +static inline __u64 cpu_info__lp_time(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct cpu_info *)hdr)->lp_time; + else /* INFO_EXT */ + return ((struct x_cpu_info *)hdr)->lp_time; +} + +static inline __u64 cpu_info__online_time(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return 0; /* online_time not available in simple info */ + else /* INFO_EXT */ + return ((struct x_cpu_info *)hdr)->online_time; +} + +/* Physical header */ + +struct phys_hdr { + char reserved1[1]; + __u8 cpus; + char reserved2[6]; + char mgm_name[8]; +} __attribute__ ((packed)); + +struct x_phys_hdr { + char reserved1[1]; + __u8 cpus; + char reserved2[6]; + char mgm_name[8]; + char reserved3[80]; +} __attribute__ ((packed)); + +static inline int phys_hdr__size(enum diag204_format type) +{ + if (type == INFO_SIMPLE) + return sizeof(struct phys_hdr); + else /* INFO_EXT */ + return sizeof(struct x_phys_hdr); +} + +static inline __u8 phys_hdr__cpus(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct phys_hdr *)hdr)->cpus; + else /* INFO_EXT */ + return ((struct x_phys_hdr *)hdr)->cpus; +} + +/* Physical CPU info block */ + +struct phys_cpu { + __u16 cpu_addr; + char reserved1[2]; + __u8 ctidx; + char reserved2[3]; + __u64 mgm_time; + char reserved3[8]; +} __attribute__ ((packed)); + +struct x_phys_cpu { + __u16 cpu_addr; + char reserved1[2]; + __u8 ctidx; + char reserved2[3]; + __u64 mgm_time; + char reserved3[80]; +} __attribute__ ((packed)); + +static inline int phys_cpu__size(enum diag204_format type) +{ + if (type == INFO_SIMPLE) + return sizeof(struct phys_cpu); + else /* INFO_EXT */ + return sizeof(struct x_phys_cpu); +} + +static inline __u16 phys_cpu__cpu_addr(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct phys_cpu *)hdr)->cpu_addr; + else /* INFO_EXT */ + return ((struct x_phys_cpu *)hdr)->cpu_addr; +} + +static inline __u64 phys_cpu__mgm_time(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct phys_cpu *)hdr)->mgm_time; + else /* INFO_EXT */ + return ((struct x_phys_cpu *)hdr)->mgm_time; +} + +static inline __u64 phys_cpu__ctidx(enum diag204_format type, void *hdr) +{ + if (type == INFO_SIMPLE) + return ((struct phys_cpu *)hdr)->ctidx; + else /* INFO_EXT */ + return ((struct x_phys_cpu *)hdr)->ctidx; +} + +/* Diagnose 204 functions */ + +static int diag204(unsigned long subcode, unsigned long size, void *addr) +{ + register unsigned long _subcode asm("0") = subcode; + register unsigned long _size asm("1") = size; + + asm volatile (" diag %2,%0,0x204\n" + "0: \n" ".section __ex_table,\"a\"\n" +#ifndef __s390x__ + " .align 4\n" + " .long 0b,0b\n" +#else + " .align 8\n" + " .quad 0b,0b\n" +#endif + ".previous":"+d" (_subcode), "+d"(_size) + :"d"(addr) + :"memory"); + if (_subcode) + return -1; + else + return _size; +} + +/* + * For the old diag subcode 4 with simple data format we have to use real + * memory. If we use subcode 6 or 7 with extended data format, we can (and + * should) use vmalloc, since we need a lot of memory in that case. Currently + * up to 93 pages! + */ + +static void diag204_free_buffer(void) +{ + if (!diag204_buf) + return; + if (diag204_buf_vmalloc) { + vfree(diag204_buf_vmalloc); + diag204_buf_vmalloc = NULL; + } else { + free_pages((unsigned long) diag204_buf, 0); + } + diag204_buf_pages = 0; + diag204_buf = NULL; +} + +static void *diag204_alloc_vbuf(int pages) +{ + /* The buffer has to be page aligned! */ + diag204_buf_vmalloc = vmalloc(PAGE_SIZE * (pages + 1)); + if (!diag204_buf_vmalloc) + return ERR_PTR(-ENOMEM); + diag204_buf = (void*)((unsigned long)diag204_buf_vmalloc + & ~0xfffUL) + 0x1000; + diag204_buf_pages = pages; + return diag204_buf; +} + +static void *diag204_alloc_rbuf(void) +{ + diag204_buf = (void*)__get_free_pages(GFP_KERNEL,0); + if (diag204_buf) + return ERR_PTR(-ENOMEM); + diag204_buf_pages = 1; + return diag204_buf; +} + +static void *diag204_get_buffer(enum diag204_format fmt, int *pages) +{ + if (diag204_buf) { + *pages = diag204_buf_pages; + return diag204_buf; + } + if (fmt == INFO_SIMPLE) { + *pages = 1; + return diag204_alloc_rbuf(); + } else {/* INFO_EXT */ + *pages = diag204(SUBC_RSI | INFO_EXT, 0, 0); + if (*pages <= 0) + return ERR_PTR(-ENOSYS); + else + return diag204_alloc_vbuf(*pages); + } +} + +/* + * diag204_probe() has to find out, which type of diagnose 204 implementation + * we have on our machine. Currently there are three possible scanarios: + * - subcode 4 + simple data format (only one page) + * - subcode 4-6 + extended data format + * - subcode 4-7 + extended data format + * + * Subcode 5 is used to retrieve the size of the data, provided by subcodes + * 6 and 7. Subcode 7 basically has the same function as subcode 6. In addition + * to subcode 6 it provides also information about secondary cpus. + * In order to get as much information as possible, we first try + * subcode 7, then 6 and if both fail, we use subcode 4. + */ + +static int diag204_probe(void) +{ + void *buf; + int pages, rc; + + buf = diag204_get_buffer(INFO_EXT, &pages); + if (!IS_ERR(buf)) { + if (diag204(SUBC_STIB7 | INFO_EXT, pages, buf) >= 0) { + diag204_store_sc = SUBC_STIB7; + diag204_info_type = INFO_EXT; + goto out; + } + if (diag204(SUBC_STIB6 | INFO_EXT, pages, buf) >= 0) { + diag204_store_sc = SUBC_STIB7; + diag204_info_type = INFO_EXT; + goto out; + } + diag204_free_buffer(); + } + + /* subcodes 6 and 7 failed, now try subcode 4 */ + + buf = diag204_get_buffer(INFO_SIMPLE, &pages); + if (IS_ERR(buf)) { + rc = PTR_ERR(buf); + goto fail_alloc; + } + if (diag204(SUBC_STIB4 | INFO_SIMPLE, pages, buf) >= 0) { + diag204_store_sc = SUBC_STIB4; + diag204_info_type = INFO_SIMPLE; + goto out; + } else { + rc = -ENOSYS; + goto fail_store; + } +out: + rc = 0; +fail_store: + diag204_free_buffer(); +fail_alloc: + return rc; +} + +static void *diag204_store(void) +{ + void *buf; + int pages; + + buf = diag204_get_buffer(diag204_info_type, &pages); + if (IS_ERR(buf)) + goto out; + if (diag204(diag204_store_sc | diag204_info_type, pages, buf) < 0) + return ERR_PTR(-ENOSYS); +out: + return buf; +} + +/* Diagnose 224 functions */ + +static void diag224(void *ptr) +{ + asm volatile(" diag %0,%1,0x224\n" + : :"d" (0), "d"(ptr) : "memory"); +} + +static int diag224_get_name_table(void) +{ + /* memory must be below 2GB */ + diag224_cpu_names = kmalloc(PAGE_SIZE, GFP_KERNEL | GFP_DMA); + if (!diag224_cpu_names) + return -ENOMEM; + diag224(diag224_cpu_names); + EBCASC(diag224_cpu_names + 16, (*diag224_cpu_names + 1) * 16); + return 0; +} + +static void diag224_delete_name_table(void) +{ + kfree(diag224_cpu_names); +} + +static int diag224_idx2name(int index, char *name) +{ + memcpy(name, diag224_cpu_names + ((index + 1) * CPU_NAME_LEN), + CPU_NAME_LEN); + name[CPU_NAME_LEN] = 0; + strstrip(name); + return 0; +} + +__init int hypfs_diag_init(void) +{ + int rc; + + if (diag204_probe()) { + printk(KERN_ERR "hypfs: diag 204 not working."); + return -ENODATA; + } + rc = diag224_get_name_table(); + if (rc) { + diag224_delete_name_table(); + printk(KERN_ERR "hypfs: could not get name table.\n"); + } + return rc; +} + +__exit void hypfs_diag_exit(void) +{ + diag224_delete_name_table(); + diag204_free_buffer(); +} + +/* + * Functions to create the directory structure + * ******************************************* + */ + +static int hypfs_create_cpu_files(struct super_block *sb, + struct dentry *cpus_dir, void *cpu_info) +{ + struct dentry *cpu_dir; + char buffer[TMP_SIZE]; + void *rc; + + snprintf(buffer, TMP_SIZE, "%d", cpu_info__cpu_addr(diag204_info_type, + cpu_info)); + cpu_dir = hypfs_mkdir(sb, cpus_dir, buffer); + rc = hypfs_create_u64(sb, cpu_dir, "mgmtime", + cpu_info__acc_time(diag204_info_type, cpu_info) - + cpu_info__lp_time(diag204_info_type, cpu_info)); + if (IS_ERR(rc)) + return PTR_ERR(rc); + rc = hypfs_create_u64(sb, cpu_dir, "cputime", + cpu_info__lp_time(diag204_info_type, cpu_info)); + if (IS_ERR(rc)) + return PTR_ERR(rc); + if (diag204_info_type == INFO_EXT) { + rc = hypfs_create_u64(sb, cpu_dir, "onlinetime", + cpu_info__online_time(diag204_info_type, + cpu_info)); + if (IS_ERR(rc)) + return PTR_ERR(rc); + } + diag224_idx2name(cpu_info__ctidx(diag204_info_type, cpu_info), buffer); + rc = hypfs_create_str(sb, cpu_dir, "type", buffer); + if (IS_ERR(rc)) + return PTR_ERR(rc); + return 0; +} + +static void *hypfs_create_lpar_files(struct super_block *sb, + struct dentry *systems_dir, void *part_hdr) +{ + struct dentry *cpus_dir; + struct dentry *lpar_dir; + char lpar_name[LPAR_NAME_LEN + 1]; + void *cpu_info; + int i; + + part_hdr__part_name(diag204_info_type, part_hdr, lpar_name); + lpar_name[LPAR_NAME_LEN] = 0; + lpar_dir = hypfs_mkdir(sb, systems_dir, lpar_name); + if (IS_ERR(lpar_dir)) + return lpar_dir; + cpus_dir = hypfs_mkdir(sb, lpar_dir, "cpus"); + if (IS_ERR(cpus_dir)) + return cpus_dir; + cpu_info = part_hdr + part_hdr__size(diag204_info_type); + for (i = 0; i < part_hdr__rcpus(diag204_info_type, part_hdr); i++) { + int rc; + rc = hypfs_create_cpu_files(sb, cpus_dir, cpu_info); + if (rc) + return ERR_PTR(rc); + cpu_info += cpu_info__size(diag204_info_type); + } + return cpu_info; +} + +static int hypfs_create_phys_cpu_files(struct super_block *sb, + struct dentry *cpus_dir, void *cpu_info) +{ + struct dentry *cpu_dir; + char buffer[TMP_SIZE]; + void *rc; + + snprintf(buffer, TMP_SIZE, "%i", phys_cpu__cpu_addr(diag204_info_type, + cpu_info)); + cpu_dir = hypfs_mkdir(sb, cpus_dir, buffer); + if (IS_ERR(cpu_dir)) + return PTR_ERR(cpu_dir); + rc = hypfs_create_u64(sb, cpu_dir, "mgmtime", + phys_cpu__mgm_time(diag204_info_type, cpu_info)); + if (IS_ERR(rc)) + return PTR_ERR(rc); + diag224_idx2name(phys_cpu__ctidx(diag204_info_type, cpu_info), buffer); + rc = hypfs_create_str(sb, cpu_dir, "type", buffer); + if (IS_ERR(rc)) + return PTR_ERR(rc); + return 0; +} + +static void *hypfs_create_phys_files(struct super_block *sb, + struct dentry *parent_dir, void *phys_hdr) +{ + int i; + void *cpu_info; + struct dentry *cpus_dir; + + cpus_dir = hypfs_mkdir(sb, parent_dir, "cpus"); + if (IS_ERR(cpus_dir)) + return cpus_dir; + cpu_info = phys_hdr + phys_hdr__size(diag204_info_type); + for (i = 0; i < phys_hdr__cpus(diag204_info_type, phys_hdr); i++) { + int rc; + rc = hypfs_create_phys_cpu_files(sb, cpus_dir, cpu_info); + if (rc) + return ERR_PTR(rc); + cpu_info += phys_cpu__size(diag204_info_type); + } + return cpu_info; +} + +int hypfs_diag_create_files(struct super_block *sb, struct dentry *root) +{ + struct dentry *systems_dir, *hyp_dir; + void *time_hdr, *part_hdr; + int i, rc; + void *buffer, *ptr; + + buffer = diag204_store(); + if (IS_ERR(buffer)) + return PTR_ERR(buffer); + + systems_dir = hypfs_mkdir(sb, root, "systems"); + if (IS_ERR(systems_dir)) { + rc = PTR_ERR(systems_dir); + goto err_out; + } + time_hdr = (struct x_info_blk_hdr *)buffer; + part_hdr = time_hdr + info_blk_hdr__size(diag204_info_type); + for (i = 0; i < info_blk_hdr__npar(diag204_info_type, time_hdr); i++) { + part_hdr = hypfs_create_lpar_files(sb, systems_dir, part_hdr); + if (IS_ERR(part_hdr)) { + rc = PTR_ERR(part_hdr); + goto err_out; + } + } + if (info_blk_hdr__flags(diag204_info_type, time_hdr) & LPAR_PHYS_FLG) { + ptr = hypfs_create_phys_files(sb, root, part_hdr); + if (IS_ERR(ptr)) { + rc = PTR_ERR(ptr); + goto err_out; + } + } + hyp_dir = hypfs_mkdir(sb, root, "hyp"); + if (IS_ERR(hyp_dir)) { + rc = PTR_ERR(hyp_dir); + goto err_out; + } + ptr = hypfs_create_str(sb, hyp_dir, "type", "LPAR Hypervisor"); + if (IS_ERR(ptr)) { + rc = PTR_ERR(ptr); + goto err_out; + } + rc = 0; + +err_out: + return rc; +} diff --git a/arch/s390/hypfs/hypfs_diag.h b/arch/s390/hypfs/hypfs_diag.h new file mode 100644 index 00000000000..793dea6b9bb --- /dev/null +++ b/arch/s390/hypfs/hypfs_diag.h @@ -0,0 +1,16 @@ +/* + * fs/hypfs/hypfs_diag.h + * Hypervisor filesystem for Linux on s390. + * + * Copyright (C) IBM Corp. 2006 + * Author(s): Michael Holzheu <holzheu@de.ibm.com> + */ + +#ifndef _HYPFS_DIAG_H_ +#define _HYPFS_DIAG_H_ + +extern int hypfs_diag_init(void); +extern void hypfs_diag_exit(void); +extern int hypfs_diag_create_files(struct super_block *sb, struct dentry *root); + +#endif /* _HYPFS_DIAG_H_ */ diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c new file mode 100644 index 00000000000..18c091925ea --- /dev/null +++ b/arch/s390/hypfs/inode.c @@ -0,0 +1,491 @@ +/* + * fs/hypfs/inode.c + * Hypervisor filesystem for Linux on s390. + * + * Copyright (C) IBM Corp. 2006 + * Author(s): Michael Holzheu <holzheu@de.ibm.com> + */ + +#include <linux/types.h> +#include <linux/errno.h> +#include <linux/fs.h> +#include <linux/namei.h> +#include <linux/vfs.h> +#include <linux/pagemap.h> +#include <linux/gfp.h> +#include <linux/time.h> +#include <linux/parser.h> +#include <linux/sysfs.h> +#include <linux/module.h> +#include <asm/ebcdic.h> +#include "hypfs.h" +#include "hypfs_diag.h" + +#define HYPFS_MAGIC 0x687970 /* ASCII 'hyp' */ +#define TMP_SIZE 64 /* size of temporary buffers */ + +static struct dentry *hypfs_create_update_file(struct super_block *sb, + struct dentry *dir); + +struct hypfs_sb_info { + uid_t uid; /* uid used for files and dirs */ + gid_t gid; /* gid used for files and dirs */ + struct dentry *update_file; /* file to trigger update */ + time_t last_update; /* last update time in secs since 1970 */ + struct mutex lock; /* lock to protect update process */ +}; + +static struct file_operations hypfs_file_ops; +static struct file_system_type hypfs_type; +static struct super_operations hypfs_s_ops; + +/* start of list of all dentries, which have to be deleted on update */ +static struct dentry *hypfs_last_dentry; + +static void hypfs_update_update(struct super_block *sb) +{ + struct hypfs_sb_info *sb_info = sb->s_fs_info; + struct inode *inode = sb_info->update_file->d_inode; + + sb_info->last_update = get_seconds(); + inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; +} + +/* directory tree removal functions */ + +static void hypfs_add_dentry(struct dentry *dentry) +{ + dentry->d_fsdata = hypfs_last_dentry; + hypfs_last_dentry = dentry; +} + +static void hypfs_remove(struct dentry *dentry) +{ + struct dentry *parent; + + parent = dentry->d_parent; + if (S_ISDIR(dentry->d_inode->i_mode)) + simple_rmdir(parent->d_inode, dentry); + else + simple_unlink(parent->d_inode, dentry); + d_delete(dentry); + dput(dentry); +} + +static void hypfs_delete_tree(struct dentry *root) +{ + while (hypfs_last_dentry) { + struct dentry *next_dentry; + next_dentry = hypfs_last_dentry->d_fsdata; + hypfs_remove(hypfs_last_dentry); + hypfs_last_dentry = next_dentry; + } +} + +static struct inode *hypfs_make_inode(struct super_block *sb, int mode) +{ + struct inode *ret = new_inode(sb); + + if (ret) { + struct hypfs_sb_info *hypfs_info = sb->s_fs_info; + ret->i_mode = mode; + ret->i_uid = hypfs_info->uid; + ret->i_gid = hypfs_info->gid; + ret->i_blksize = PAGE_CACHE_SIZE; + ret->i_blocks = 0; + ret->i_atime = ret->i_mtime = ret->i_ctime = CURRENT_TIME; + if (mode & S_IFDIR) + ret->i_nlink = 2; + else + ret->i_nlink = 1; + } + return ret; +} + +static void hypfs_drop_inode(struct inode *inode) +{ + kfree(inode->u.generic_ip); + generic_delete_inode(inode); +} + +static int hypfs_open(struct inode *inode, struct file *filp) +{ + char *data = filp->f_dentry->d_inode->u.generic_ip; + struct hypfs_sb_info *fs_info; + + if (filp->f_mode & FMODE_WRITE) { + if (!(inode->i_mode & S_IWUGO)) + return -EACCES; + } + if (filp->f_mode & FMODE_READ) { + if (!(inode->i_mode & S_IRUGO)) + return -EACCES; + } + + fs_info = inode->i_sb->s_fs_info; + if(data) { + mutex_lock(&fs_info->lock); + filp->private_data = kstrdup(data, GFP_KERNEL); + if (!filp->private_data) { + mutex_unlock(&fs_info->lock); + return -ENOMEM; + } + mutex_unlock(&fs_info->lock); + } + return 0; +} + +static ssize_t hypfs_aio_read(struct kiocb *iocb, __user char *buf, + size_t count, loff_t offset) +{ + char *data; + size_t len; + struct file *filp = iocb->ki_filp; + + 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; + file_accessed(filp); +out: + return count; +} +static ssize_t hypfs_aio_write(struct kiocb *iocb, const char __user *buf, + size_t count, loff_t pos) +{ + int rc; + struct super_block *sb; + struct hypfs_sb_info *fs_info; + + sb = iocb->ki_filp->f_dentry->d_inode->i_sb; + fs_info = sb->s_fs_info; + /* + * Currently we only allow one update per second for two reasons: + * 1. diag 204 is VERY expensive + * 2. If several processes do updates in parallel and then read the + * hypfs data, the likelihood of collisions is reduced, if we restrict + * the minimum update interval. A collision occurs, if during the + * data gathering of one process another process triggers an update + * If the first process wants to ensure consistent data, it has + * to restart data collection in this case. + */ + mutex_lock(&fs_info->lock); + if (fs_info->last_update == get_seconds()) { + rc = -EBUSY; + goto out; + } + hypfs_delete_tree(sb->s_root); + rc = hypfs_diag_create_files(sb, sb->s_root); + if (rc) { + printk(KERN_ERR "hypfs: Update failed\n"); + hypfs_delete_tree(sb->s_root); + goto out; + } + hypfs_update_update(sb); + rc = count; +out: + mutex_unlock(&fs_info->lock); + return rc; +} + +static int hypfs_release(struct inode *inode, struct file *filp) +{ + kfree(filp->private_data); + return 0; +} + +enum { opt_uid, opt_gid, opt_err }; + +static match_table_t hypfs_tokens = { + {opt_uid, "uid=%u"}, + {opt_gid, "gid=%u"}, + {opt_err, NULL} +}; + +static int hypfs_parse_options(char *options, struct super_block *sb) +{ + char *str; + substring_t args[MAX_OPT_ARGS]; + + if (!options) + return 0; + while ((str = strsep(&options, ",")) != NULL) { + int token, option; + struct hypfs_sb_info *hypfs_info = sb->s_fs_info; + + if (!*str) + continue; + token = match_token(str, hypfs_tokens, args); + switch (token) { + case opt_uid: + if (match_int(&args[0], &option)) + return -EINVAL; + hypfs_info->uid = option; + break; + case opt_gid: + if (match_int(&args[0], &option)) + return -EINVAL; + hypfs_info->gid = option; + break; + case opt_err: + default: + printk(KERN_ERR "hypfs: Unrecognized mount option " + "\"%s\" or missing value\n", str); + return -EINVAL; + } + } + return 0; +} + +static int hypfs_fill_super(struct super_block *sb, void *data, int silent) +{ + struct inode *root_inode; + struct dentry *root_dentry; + int rc = 0; + struct hypfs_sb_info *sbi; + + sbi = kzalloc(sizeof(struct hypfs_sb_info), GFP_KERNEL); + if (!sbi) + return -ENOMEM; + mutex_init(&sbi->lock); + sbi->uid = current->uid; + sbi->gid = current->gid; + sb->s_fs_info = sbi; + sb->s_blocksize = PAGE_CACHE_SIZE; + sb->s_blocksize_bits = PAGE_CACHE_SHIFT; + sb->s_magic = HYPFS_MAGIC; + sb->s_op = &hypfs_s_ops; + if (hypfs_parse_options(data, sb)) { + rc = -EINVAL; + goto err_alloc; + } + root_inode = hypfs_make_inode(sb, S_IFDIR | 0755); + if (!root_inode) { + rc = -ENOMEM; + goto err_alloc; + } + root_inode->i_op = &simple_dir_inode_operations; + root_inode->i_fop = &simple_dir_operations; + root_dentry = d_alloc_root(root_inode); + if (!root_dentry) { + iput(root_inode); + rc = -ENOMEM; + goto err_alloc; + } + rc = hypfs_diag_create_files(sb, root_dentry); + if (rc) + goto err_tree; + sbi->update_file = hypfs_create_update_file(sb, root_dentry); + if (IS_ERR(sbi->update_file)) { + rc = PTR_ERR(sbi->update_file); + goto err_tree; + } + hypfs_update_update(sb); + sb->s_root = root_dentry; + return 0; + +err_tree: + hypfs_delete_tree(root_dentry); + d_genocide(root_dentry); + dput(root_dentry); +err_alloc: + kfree(sbi); + return rc; +} + +static int hypfs_get_super(struct file_system_type *fst, int flags, + const char *devname, void *data, struct vfsmount *mnt) +{ + return get_sb_single(fst, flags, data, hypfs_fill_super, mnt); +} + +static void hypfs_kill_super(struct super_block *sb) +{ + struct hypfs_sb_info *sb_info = sb->s_fs_info; + + hypfs_delete_tree(sb->s_root); + hypfs_remove(sb_info->update_file); + kfree(sb->s_fs_info); + sb->s_fs_info = NULL; + kill_litter_super(sb); +} + +static struct dentry *hypfs_create_file(struct super_block *sb, + struct dentry *parent, const char *name, + char *data, mode_t mode) +{ + struct dentry *dentry; + struct inode *inode; + struct qstr qname; + + qname.name = name; + qname.len = strlen(name); + qname.hash = full_name_hash(name, qname.len); + dentry = lookup_one_len(name, parent, strlen(name)); + if (IS_ERR(dentry)) + return ERR_PTR(-ENOMEM); + inode = hypfs_make_inode(sb, mode); + if (!inode) { + dput(dentry); + return ERR_PTR(-ENOMEM); + } + if (mode & S_IFREG) { + inode->i_fop = &hypfs_file_ops; + if (data) + inode->i_size = strlen(data); + else + inode->i_size = 0; + } else if (mode & S_IFDIR) { + inode->i_op = &simple_dir_inode_operations; + inode->i_fop = &simple_dir_operations; + parent->d_inode->i_nlink++; + } else + BUG(); + inode->u.generic_ip = data; + d_instantiate(dentry, inode); + dget(dentry); + return dentry; +} + +struct dentry *hypfs_mkdir(struct super_block *sb, struct dentry *parent, + const char *name) +{ + struct dentry *dentry; + + dentry = hypfs_create_file(sb, parent, name, NULL, S_IFDIR | DIR_MODE); + if (IS_ERR(dentry)) + return dentry; + hypfs_add_dentry(dentry); + parent->d_inode->i_nlink++; + return dentry; +} + +static struct dentry *hypfs_create_update_file(struct super_block *sb, + struct dentry *dir) +{ + struct dentry *dentry; + + dentry = hypfs_create_file(sb, dir, "update", NULL, + S_IFREG | UPDATE_FILE_MODE); + /* + * We do not put the update file on the 'delete' list with + * hypfs_add_dentry(), since it should not be removed when the tree + * is updated. + */ + return dentry; +} + +struct dentry *hypfs_create_u64(struct super_block *sb, struct dentry *dir, + const char *name, __u64 value) +{ + char *buffer; + char tmp[TMP_SIZE]; + struct dentry *dentry; + + snprintf(tmp, TMP_SIZE, "%lld\n", (unsigned long long int)value); + buffer = kstrdup(tmp, GFP_KERNEL); + if (!buffer) + return ERR_PTR(-ENOMEM); + dentry = + hypfs_create_file(sb, dir, name, buffer, S_IFREG | REG_FILE_MODE); + if (IS_ERR(dentry)) { + kfree(buffer); + return ERR_PTR(-ENOMEM); + } + hypfs_add_dentry(dentry); + return dentry; +} + +struct dentry *hypfs_create_str(struct super_block *sb, struct dentry *dir, + const char *name, char *string) +{ + char *buffer; + struct dentry *dentry; + + buffer = kmalloc(strlen(string) + 2, GFP_KERNEL); + if (!buffer) + return ERR_PTR(-ENOMEM); + sprintf(buffer, "%s\n", string); + dentry = + hypfs_create_file(sb, dir, name, buffer, S_IFREG | REG_FILE_MODE); + if (IS_ERR(dentry)) { + kfree(buffer); + return ERR_PTR(-ENOMEM); + } + hypfs_add_dentry(dentry); + return dentry; +} + +static struct file_operations hypfs_file_ops = { + .open = hypfs_open, + .release = hypfs_release, + .read = do_sync_read, + .write = do_sync_write, + .aio_read = hypfs_aio_read, + .aio_write = hypfs_aio_write, +}; + +static struct file_system_type hypfs_type = { + .owner = THIS_MODULE, + .name = "s390_hypfs", + .get_sb = hypfs_get_super, + .kill_sb = hypfs_kill_super +}; + +static struct super_operations hypfs_s_ops = { + .statfs = simple_statfs, + .drop_inode = hypfs_drop_inode, +}; + +static decl_subsys(s390, NULL, NULL); + +static int __init hypfs_init(void) +{ + int rc; + + if (MACHINE_IS_VM) + return -ENODATA; + if (hypfs_diag_init()) { + rc = -ENODATA; + goto fail_diag; + } + kset_set_kset_s(&s390_subsys, hypervisor_subsys); + rc = subsystem_register(&s390_subsys); + if (rc) + goto fail_sysfs; + rc = register_filesystem(&hypfs_type); + if (rc) + goto fail_filesystem; + return 0; + +fail_filesystem: + subsystem_unregister(&s390_subsys); +fail_sysfs: + hypfs_diag_exit(); +fail_diag: + printk(KERN_ERR "hypfs: Initialization failed with rc = %i.\n", rc); + return rc; +} + +static void __exit hypfs_exit(void) +{ + hypfs_diag_exit(); + unregister_filesystem(&hypfs_type); + subsystem_unregister(&s390_subsys); +} + +module_init(hypfs_init) +module_exit(hypfs_exit) + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Michael Holzheu <holzheu@de.ibm.com>"); +MODULE_DESCRIPTION("s390 Hypervisor Filesystem"); diff --git a/arch/s390/kernel/binfmt_elf32.c b/arch/s390/kernel/binfmt_elf32.c index 1f451c2cb07..12a6311e983 100644 --- a/arch/s390/kernel/binfmt_elf32.c +++ b/arch/s390/kernel/binfmt_elf32.c @@ -177,11 +177,6 @@ struct elf_prpsinfo32 #include <linux/highuid.h> -#undef NEW_TO_OLD_UID -#undef NEW_TO_OLD_GID -#define NEW_TO_OLD_UID(uid) ((uid) > 65535) ? (u16)overflowuid : (u16)(uid) -#define NEW_TO_OLD_GID(gid) ((gid) > 65535) ? (u16)overflowgid : (u16)(gid) - #define elf_addr_t u32 /* #define init_elf_binfmt init_elf32_binfmt diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index b2448487854..aa8b52c2140 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -93,13 +93,22 @@ STACK_SIZE = 1 << STACK_SHIFT l %r13,__LC_SVC_NEW_PSW+4 # load &system_call to %r13 .endm - .macro SAVE_ALL psworg,savearea,sync + .macro SAVE_ALL_SYNC psworg,savearea la %r12,\psworg - .if \sync tm \psworg+1,0x01 # test problem state bit bz BASED(2f) # skip stack setup save l %r15,__LC_KERNEL_STACK # problem state -> load ksp - .else +#ifdef CONFIG_CHECK_STACK + b BASED(3f) +2: tml %r15,STACK_SIZE - CONFIG_STACK_GUARD + bz BASED(stack_overflow) +3: +#endif +2: + .endm + + .macro SAVE_ALL_ASYNC psworg,savearea + la %r12,\psworg tm \psworg+1,0x01 # test problem state bit bnz BASED(1f) # from user -> load async stack clc \psworg+4(4),BASED(.Lcritical_end) @@ -115,7 +124,6 @@ STACK_SIZE = 1 << STACK_SHIFT sra %r14,STACK_SHIFT be BASED(2f) 1: l %r15,__LC_ASYNC_STACK - .endif #ifdef CONFIG_CHECK_STACK b BASED(3f) 2: tml %r15,STACK_SIZE - CONFIG_STACK_GUARD @@ -196,7 +204,7 @@ system_call: STORE_TIMER __LC_SYNC_ENTER_TIMER sysc_saveall: SAVE_ALL_BASE __LC_SAVE_AREA - SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1 + SAVE_ALL_SYNC __LC_SVC_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA lh %r7,0x8a # get svc number from lowcore #ifdef CONFIG_VIRT_CPU_ACCOUNTING @@ -425,7 +433,7 @@ pgm_check_handler: SAVE_ALL_BASE __LC_SAVE_AREA tm __LC_PGM_INT_CODE+1,0x80 # check whether we got a per exception bnz BASED(pgm_per) # got per exception -> special case - SAVE_ALL __LC_PGM_OLD_PSW,__LC_SAVE_AREA,1 + SAVE_ALL_SYNC __LC_PGM_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_PGM_OLD_PSW,__LC_SAVE_AREA #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? @@ -464,7 +472,7 @@ pgm_per: # Normal per exception # pgm_per_std: - SAVE_ALL __LC_PGM_OLD_PSW,__LC_SAVE_AREA,1 + SAVE_ALL_SYNC __LC_PGM_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_PGM_OLD_PSW,__LC_SAVE_AREA #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? @@ -490,7 +498,7 @@ pgm_no_vtime2: # it was a single stepped SVC that is causing all the trouble # pgm_svcper: - SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1 + SAVE_ALL_SYNC __LC_SVC_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? @@ -519,7 +527,7 @@ io_int_handler: STORE_TIMER __LC_ASYNC_ENTER_TIMER stck __LC_INT_CLOCK SAVE_ALL_BASE __LC_SAVE_AREA+16 - SAVE_ALL __LC_IO_OLD_PSW,__LC_SAVE_AREA+16,0 + SAVE_ALL_ASYNC __LC_IO_OLD_PSW,__LC_SAVE_AREA+16 CREATE_STACK_FRAME __LC_IO_OLD_PSW,__LC_SAVE_AREA+16 #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? @@ -631,7 +639,7 @@ ext_int_handler: STORE_TIMER __LC_ASYNC_ENTER_TIMER stck __LC_INT_CLOCK SAVE_ALL_BASE __LC_SAVE_AREA+16 - SAVE_ALL __LC_EXT_OLD_PSW,__LC_SAVE_AREA+16,0 + SAVE_ALL_ASYNC __LC_EXT_OLD_PSW,__LC_SAVE_AREA+16 CREATE_STACK_FRAME __LC_EXT_OLD_PSW,__LC_SAVE_AREA+16 #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? @@ -657,21 +665,31 @@ __critical_end: .globl mcck_int_handler mcck_int_handler: spt __LC_CPU_TIMER_SAVE_AREA # revalidate cpu timer - mvc __LC_ASYNC_ENTER_TIMER(8),__LC_CPU_TIMER_SAVE_AREA lm %r0,%r15,__LC_GPREGS_SAVE_AREA # revalidate gprs SAVE_ALL_BASE __LC_SAVE_AREA+32 la %r12,__LC_MCK_OLD_PSW tm __LC_MCCK_CODE,0x80 # system damage? bo BASED(mcck_int_main) # yes -> rest of mcck code invalid - tm __LC_MCCK_CODE+5,0x02 # stored cpu timer value valid? - bo BASED(0f) - spt __LC_LAST_UPDATE_TIMER # revalidate cpu timer #ifdef CONFIG_VIRT_CPU_ACCOUNTING - mvc __LC_ASYNC_ENTER_TIMER(8),__LC_LAST_UPDATE_TIMER - mvc __LC_SYNC_ENTER_TIMER(8),__LC_LAST_UPDATE_TIMER - mvc __LC_EXIT_TIMER(8),__LC_LAST_UPDATE_TIMER + mvc __LC_SAVE_AREA+52(8),__LC_ASYNC_ENTER_TIMER + mvc __LC_ASYNC_ENTER_TIMER(8),__LC_CPU_TIMER_SAVE_AREA + tm __LC_MCCK_CODE+5,0x02 # stored cpu timer value valid? + bo BASED(1f) + la %r14,__LC_SYNC_ENTER_TIMER + clc 0(8,%r14),__LC_ASYNC_ENTER_TIMER + bl BASED(0f) + la %r14,__LC_ASYNC_ENTER_TIMER +0: clc 0(8,%r14),__LC_EXIT_TIMER + bl BASED(0f) + la %r14,__LC_EXIT_TIMER +0: clc 0(8,%r14),__LC_LAST_UPDATE_TIMER + bl BASED(0f) + la %r14,__LC_LAST_UPDATE_TIMER +0: spt 0(%r14) + mvc __LC_ASYNC_ENTER_TIMER(8),0(%r14) +1: #endif -0: tm __LC_MCCK_CODE+2,0x09 # mwp + ia of old psw valid? + tm __LC_MCCK_CODE+2,0x09 # mwp + ia of old psw valid? bno BASED(mcck_int_main) # no -> skip cleanup critical tm __LC_MCK_OLD_PSW+1,0x01 # test problem state bit bnz BASED(mcck_int_main) # from user -> load async stack @@ -691,7 +709,7 @@ mcck_int_main: #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm __LC_MCCK_CODE+2,0x08 # mwp of old psw valid? bno BASED(mcck_no_vtime) # no -> skip cleanup critical - tm __LC_MCK_OLD_PSW+1,0x01 # interrupting from user ? + tm SP_PSW+1(%r15),0x01 # interrupting from user ? bz BASED(mcck_no_vtime) UPDATE_VTIME __LC_EXIT_TIMER,__LC_ASYNC_ENTER_TIMER,__LC_USER_TIMER UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER @@ -715,6 +733,20 @@ mcck_no_vtime: l %r1,BASED(.Ls390_handle_mcck) basr %r14,%r1 # call machine check handler mcck_return: + mvc __LC_RETURN_MCCK_PSW(8),SP_PSW(%r15) # move return PSW + ni __LC_RETURN_MCCK_PSW+1,0xfd # clear wait state bit +#ifdef CONFIG_VIRT_CPU_ACCOUNTING + mvc __LC_ASYNC_ENTER_TIMER(8),__LC_SAVE_AREA+52 + tm __LC_RETURN_MCCK_PSW+1,0x01 # returning to user ? + bno BASED(0f) + lm %r0,%r15,SP_R0(%r15) # load gprs 0-15 + stpt __LC_EXIT_TIMER + lpsw __LC_RETURN_MCCK_PSW # back to caller +0: +#endif + lm %r0,%r15,SP_R0(%r15) # load gprs 0-15 + lpsw __LC_RETURN_MCCK_PSW # back to caller + RESTORE_ALL __LC_RETURN_MCCK_PSW,0 #ifdef CONFIG_SMP @@ -781,6 +813,8 @@ cleanup_table_sysc_leave: .long sysc_leave + 0x80000000, sysc_work_loop + 0x80000000 cleanup_table_sysc_work_loop: .long sysc_work_loop + 0x80000000, sysc_reschedule + 0x80000000 +cleanup_table_io_return: + .long io_return + 0x80000000, io_leave + 0x80000000 cleanup_table_io_leave: .long io_leave + 0x80000000, io_done + 0x80000000 cleanup_table_io_work_loop: @@ -807,6 +841,11 @@ cleanup_critical: clc 4(4,%r12),BASED(cleanup_table_sysc_work_loop+4) bl BASED(cleanup_sysc_return) 0: + clc 4(4,%r12),BASED(cleanup_table_io_return) + bl BASED(0f) + clc 4(4,%r12),BASED(cleanup_table_io_return+4) + bl BASED(cleanup_io_return) +0: clc 4(4,%r12),BASED(cleanup_table_io_leave) bl BASED(0f) clc 4(4,%r12),BASED(cleanup_table_io_leave+4) @@ -839,7 +878,7 @@ cleanup_system_call: mvc __LC_SAVE_AREA(16),0(%r12) 0: st %r13,4(%r12) st %r12,__LC_SAVE_AREA+48 # argh - SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1 + SAVE_ALL_SYNC __LC_SVC_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA l %r12,__LC_SAVE_AREA+48 # argh st %r15,12(%r12) @@ -980,7 +1019,6 @@ cleanup_io_leave_insn: .long cleanup_critical #define SYSCALL(esa,esame,emu) .long esa - .globl sys_call_table sys_call_table: #include "syscalls.S" #undef SYSCALL diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S index 2ac095bc0e2..f3222a1b286 100644 --- a/arch/s390/kernel/entry64.S +++ b/arch/s390/kernel/entry64.S @@ -87,13 +87,22 @@ _TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK | _TIF_NEED_RESCHED | \ larl %r13,system_call .endm - .macro SAVE_ALL psworg,savearea,sync + .macro SAVE_ALL_SYNC psworg,savearea la %r12,\psworg - .if \sync tm \psworg+1,0x01 # test problem state bit jz 2f # skip stack setup save lg %r15,__LC_KERNEL_STACK # problem state -> load ksp - .else +#ifdef CONFIG_CHECK_STACK + j 3f +2: tml %r15,STACK_SIZE - CONFIG_STACK_GUARD + jz stack_overflow +3: +#endif +2: + .endm + + .macro SAVE_ALL_ASYNC psworg,savearea + la %r12,\psworg tm \psworg+1,0x01 # test problem state bit jnz 1f # from user -> load kernel stack clc \psworg+8(8),BASED(.Lcritical_end) @@ -108,7 +117,6 @@ _TIF_WORK_INT = (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK | _TIF_NEED_RESCHED | \ srag %r14,%r14,STACK_SHIFT jz 2f 1: lg %r15,__LC_ASYNC_STACK # load async stack - .endif #ifdef CONFIG_CHECK_STACK j 3f 2: tml %r15,STACK_SIZE - CONFIG_STACK_GUARD @@ -187,7 +195,7 @@ system_call: STORE_TIMER __LC_SYNC_ENTER_TIMER sysc_saveall: SAVE_ALL_BASE __LC_SAVE_AREA - SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1 + SAVE_ALL_SYNC __LC_SVC_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA llgh %r7,__LC_SVC_INT_CODE # get svc number from lowcore #ifdef CONFIG_VIRT_CPU_ACCOUNTING @@ -446,7 +454,7 @@ pgm_check_handler: SAVE_ALL_BASE __LC_SAVE_AREA tm __LC_PGM_INT_CODE+1,0x80 # check whether we got a per exception jnz pgm_per # got per exception -> special case - SAVE_ALL __LC_PGM_OLD_PSW,__LC_SAVE_AREA,1 + SAVE_ALL_SYNC __LC_PGM_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_PGM_OLD_PSW,__LC_SAVE_AREA #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? @@ -485,7 +493,7 @@ pgm_per: # Normal per exception # pgm_per_std: - SAVE_ALL __LC_PGM_OLD_PSW,__LC_SAVE_AREA,1 + SAVE_ALL_SYNC __LC_PGM_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_PGM_OLD_PSW,__LC_SAVE_AREA #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? @@ -511,7 +519,7 @@ pgm_no_vtime2: # it was a single stepped SVC that is causing all the trouble # pgm_svcper: - SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1 + SAVE_ALL_SYNC __LC_SVC_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? @@ -539,7 +547,7 @@ io_int_handler: STORE_TIMER __LC_ASYNC_ENTER_TIMER stck __LC_INT_CLOCK SAVE_ALL_BASE __LC_SAVE_AREA+32 - SAVE_ALL __LC_IO_OLD_PSW,__LC_SAVE_AREA+32,0 + SAVE_ALL_ASYNC __LC_IO_OLD_PSW,__LC_SAVE_AREA+32 CREATE_STACK_FRAME __LC_IO_OLD_PSW,__LC_SAVE_AREA+32 #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? @@ -647,7 +655,7 @@ ext_int_handler: STORE_TIMER __LC_ASYNC_ENTER_TIMER stck __LC_INT_CLOCK SAVE_ALL_BASE __LC_SAVE_AREA+32 - SAVE_ALL __LC_EXT_OLD_PSW,__LC_SAVE_AREA+32,0 + SAVE_ALL_ASYNC __LC_EXT_OLD_PSW,__LC_SAVE_AREA+32 CREATE_STACK_FRAME __LC_EXT_OLD_PSW,__LC_SAVE_AREA+32 #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm SP_PSW+1(%r15),0x01 # interrupting from user ? @@ -672,21 +680,32 @@ __critical_end: mcck_int_handler: la %r1,4095 # revalidate r1 spt __LC_CPU_TIMER_SAVE_AREA-4095(%r1) # revalidate cpu timer - mvc __LC_ASYNC_ENTER_TIMER(8),__LC_CPU_TIMER_SAVE_AREA-4095(%r1) lmg %r0,%r15,__LC_GPREGS_SAVE_AREA-4095(%r1)# revalidate gprs SAVE_ALL_BASE __LC_SAVE_AREA+64 la %r12,__LC_MCK_OLD_PSW tm __LC_MCCK_CODE,0x80 # system damage? jo mcck_int_main # yes -> rest of mcck code invalid - tm __LC_MCCK_CODE+5,0x02 # stored cpu timer value valid? - jo 0f - spt __LC_LAST_UPDATE_TIMER #ifdef CONFIG_VIRT_CPU_ACCOUNTING - mvc __LC_ASYNC_ENTER_TIMER(8),__LC_LAST_UPDATE_TIMER - mvc __LC_SYNC_ENTER_TIMER(8),__LC_LAST_UPDATE_TIMER - mvc __LC_EXIT_TIMER(8),__LC_LAST_UPDATE_TIMER + la %r14,4095 + mvc __LC_SAVE_AREA+104(8),__LC_ASYNC_ENTER_TIMER + mvc __LC_ASYNC_ENTER_TIMER(8),__LC_CPU_TIMER_SAVE_AREA-4095(%r14) + tm __LC_MCCK_CODE+5,0x02 # stored cpu timer value valid? + jo 1f + la %r14,__LC_SYNC_ENTER_TIMER + clc 0(8,%r14),__LC_ASYNC_ENTER_TIMER + jl 0f + la %r14,__LC_ASYNC_ENTER_TIMER +0: clc 0(8,%r14),__LC_EXIT_TIMER + jl 0f + la %r14,__LC_EXIT_TIMER +0: clc 0(8,%r14),__LC_LAST_UPDATE_TIMER + jl 0f + la %r14,__LC_LAST_UPDATE_TIMER +0: spt 0(%r14) + mvc __LC_ASYNC_ENTER_TIMER(8),0(%r14) +1: #endif -0: tm __LC_MCCK_CODE+2,0x09 # mwp + ia of old psw valid? + tm __LC_MCCK_CODE+2,0x09 # mwp + ia of old psw valid? jno mcck_int_main # no -> skip cleanup critical tm __LC_MCK_OLD_PSW+1,0x01 # test problem state bit jnz mcck_int_main # from user -> load kernel stack @@ -705,7 +724,7 @@ mcck_int_main: #ifdef CONFIG_VIRT_CPU_ACCOUNTING tm __LC_MCCK_CODE+2,0x08 # mwp of old psw valid? jno mcck_no_vtime # no -> no timer update - tm __LC_MCK_OLD_PSW+1,0x01 # interrupting from user ? + tm SP_PSW+1(%r15),0x01 # interrupting from user ? jz mcck_no_vtime UPDATE_VTIME __LC_EXIT_TIMER,__LC_ASYNC_ENTER_TIMER,__LC_USER_TIMER UPDATE_VTIME __LC_LAST_UPDATE_TIMER,__LC_EXIT_TIMER,__LC_SYSTEM_TIMER @@ -727,7 +746,17 @@ mcck_no_vtime: jno mcck_return brasl %r14,s390_handle_mcck mcck_return: - RESTORE_ALL __LC_RETURN_MCCK_PSW,0 + mvc __LC_RETURN_MCCK_PSW(16),SP_PSW(%r15) # move return PSW + ni __LC_RETURN_MCCK_PSW+1,0xfd # clear wait state bit + lmg %r0,%r15,SP_R0(%r15) # load gprs 0-15 +#ifdef CONFIG_VIRT_CPU_ACCOUNTING + mvc __LC_ASYNC_ENTER_TIMER(8),__LC_SAVE_AREA+104 + tm __LC_RETURN_MCCK_PSW+1,0x01 # returning to user ? + jno 0f + stpt __LC_EXIT_TIMER +0: +#endif + lpswe __LC_RETURN_MCCK_PSW # back to caller #ifdef CONFIG_SMP /* @@ -789,6 +818,8 @@ cleanup_table_sysc_leave: .quad sysc_leave, sysc_work_loop cleanup_table_sysc_work_loop: .quad sysc_work_loop, sysc_reschedule +cleanup_table_io_return: + .quad io_return, io_leave cleanup_table_io_leave: .quad io_leave, io_done cleanup_table_io_work_loop: @@ -815,6 +846,11 @@ cleanup_critical: clc 8(8,%r12),BASED(cleanup_table_sysc_work_loop+8) jl cleanup_sysc_return 0: + clc 8(8,%r12),BASED(cleanup_table_io_return) + jl 0f + clc 8(8,%r12),BASED(cleanup_table_io_return+8) + jl cleanup_io_return +0: clc 8(8,%r12),BASED(cleanup_table_io_leave) jl 0f clc 8(8,%r12),BASED(cleanup_table_io_leave+8) @@ -847,7 +883,7 @@ cleanup_system_call: mvc __LC_SAVE_AREA(32),0(%r12) 0: stg %r13,8(%r12) stg %r12,__LC_SAVE_AREA+96 # argh - SAVE_ALL __LC_SVC_OLD_PSW,__LC_SAVE_AREA,1 + SAVE_ALL_SYNC __LC_SVC_OLD_PSW,__LC_SAVE_AREA CREATE_STACK_FRAME __LC_SVC_OLD_PSW,__LC_SAVE_AREA lg %r12,__LC_SAVE_AREA+96 # argh stg %r15,24(%r12) @@ -957,7 +993,6 @@ cleanup_io_leave_insn: .quad __critical_end #define SYSCALL(esa,esame,emu) .long esame - .globl sys_call_table sys_call_table: #include "syscalls.S" #undef SYSCALL @@ -965,7 +1000,6 @@ sys_call_table: #ifdef CONFIG_COMPAT #define SYSCALL(esa,esame,emu) .long emu - .globl sys_call_table_emu sys_call_table_emu: #include "syscalls.S" #undef SYSCALL diff --git a/arch/s390/kernel/head.S b/arch/s390/kernel/head.S index ea88d066bf0..538c82da49b 100644 --- a/arch/s390/kernel/head.S +++ b/arch/s390/kernel/head.S @@ -1,7 +1,7 @@ /* * arch/s390/kernel/head.S * - * (C) Copyright IBM Corp. 1999, 2005 + * Copyright (C) IBM Corp. 1999,2006 * * Author(s): Hartmut Penner <hp@de.ibm.com> * Martin Schwidefsky <schwidefsky@de.ibm.com> @@ -482,24 +482,23 @@ start: .macro GET_IPL_DEVICE .Lget_ipl_device: - basr %r12,0 -.LGID: l %r1,0xb8 # get sid + l %r1,0xb8 # get sid sll %r1,15 # test if subchannel is enabled srl %r1,31 ltr %r1,%r1 - bz 0(%r14) # subchannel disabled + bz 2f-.LPG1(%r13) # subchannel disabled l %r1,0xb8 - la %r5,.Lipl_schib-.LGID(%r12) + la %r5,.Lipl_schib-.LPG1(%r13) stsch 0(%r5) # get schib of subchannel - bnz 0(%r14) # schib not available + bnz 2f-.LPG1(%r13) # schib not available tm 5(%r5),0x01 # devno valid? - bno 0(%r14) - la %r6,ipl_parameter_flags-.LGID(%r12) + bno 2f-.LPG1(%r13) + la %r6,ipl_parameter_flags-.LPG1(%r13) oi 3(%r6),0x01 # set flag - la %r2,ipl_devno-.LGID(%r12) + la %r2,ipl_devno-.LPG1(%r13) mvc 0(2,%r2),6(%r5) # store devno tm 4(%r5),0x80 # qdio capable device? - bno 0(%r14) + bno 2f-.LPG1(%r13) oi 3(%r6),0x02 # set flag # copy ipl parameters @@ -523,7 +522,7 @@ start: ar %r2,%r1 sr %r0,%r4 jne 1b - b 0(%r14) + b 2f-.LPG1(%r13) .align 4 .Lipl_schib: @@ -537,6 +536,7 @@ ipl_parameter_flags: .globl ipl_devno ipl_devno: .word 0 +2: .endm #ifdef CONFIG_64BIT diff --git a/arch/s390/kernel/head31.S b/arch/s390/kernel/head31.S index 2d3b089bfb8..d00de17b377 100644 --- a/arch/s390/kernel/head31.S +++ b/arch/s390/kernel/head31.S @@ -1,7 +1,7 @@ /* * arch/s390/kernel/head31.S * - * (C) Copyright IBM Corp. 2005 + * Copyright (C) IBM Corp. 2005,2006 * * Author(s): Hartmut Penner <hp@de.ibm.com> * Martin Schwidefsky <schwidefsky@de.ibm.com> @@ -16,12 +16,31 @@ # or linload or SALIPL # .org 0x10000 -startup:basr %r13,0 # get base -.LPG1: l %r1, .Lget_ipl_device_addr-.LPG1(%r13) - basr %r14, %r1 +startup:basr %r13,0 # get base +.LPG0: l %r13,0f-.LPG0(%r13) + b 0(%r13) +0: .long startup_continue + +# +# params at 10400 (setup.h) +# + .org PARMAREA + .long 0,0 # IPL_DEVICE + .long 0,RAMDISK_ORIGIN # INITRD_START + .long 0,RAMDISK_SIZE # INITRD_SIZE + + .org COMMAND_LINE + .byte "root=/dev/ram0 ro" + .byte 0 + + .org 0x11000 + +startup_continue: + basr %r13,0 # get base +.LPG1: GET_IPL_DEVICE lctl %c0,%c15,.Lctl-.LPG1(%r13) # load control registers - la %r12,_pstart-.LPG1(%r13) # pointer to parameter area - # move IPL device to lowcore + l %r12,.Lparmaddr-.LPG1(%r13) # pointer to parameter area + # move IPL device to lowcore mvc __LC_IPLDEV(4),IPL_DEVICE-PARMAREA(%r12) # @@ -51,8 +70,8 @@ startup:basr %r13,0 # get base a %r1,__LC_EXT_NEW_PSW+4 # set handler st %r1,__LC_EXT_NEW_PSW+4 - la %r4,_pstart-.LPG1(%r13) # %r4 is our index for sccb stuff - la %r1, .Lsccb-PARMAREA(%r4) # our sccb + l %r4,.Lsccbaddr-.LPG1(%r13) # %r4 is our index for sccb stuff + lr %r1,%r4 # our sccb .insn rre,0xb2200000,%r2,%r1 # service call ipm %r1 srl %r1,28 # get cc code @@ -63,7 +82,7 @@ startup:basr %r13,0 # get base be .Lservicecall-.LPG1(%r13) lpsw .Lwaitsclp-.LPG1(%r13) .Lsclph: - lh %r1,.Lsccbr-PARMAREA(%r4) + lh %r1,.Lsccbr-.Lsccb(%r4) chi %r1,0x10 # 0x0010 is the sucess code je .Lprocsccb # let's process the sccb chi %r1,0x1f0 @@ -74,7 +93,7 @@ startup:basr %r13,0 # get base b .Lservicecall-.LPG1(%r13) .Lprocsccb: lhi %r1,0 - icm %r1,3,.Lscpincr1-PARMAREA(%r4) # use this one if != 0 + icm %r1,3,.Lscpincr1-.Lsccb(%r4) # use this one if != 0 jnz .Lscnd lhi %r1,0x800 # otherwise report 2GB .Lscnd: @@ -84,10 +103,10 @@ startup:basr %r13,0 # get base lr %r1,%r3 .Lno2gb: xr %r3,%r3 # same logic - ic %r3,.Lscpa1-PARMAREA(%r4) + ic %r3,.Lscpa1-.Lsccb(%r4) chi %r3,0x00 jne .Lcompmem - l %r3,.Lscpa2-PARMAREA(%r13) + l %r3,.Lscpa2-.Lsccb(%r4) .Lcompmem: mr %r2,%r1 # mem in MB on 128-bit l %r1,.Lonemb-.LPG1(%r13) @@ -95,8 +114,6 @@ startup:basr %r13,0 # get base b .Lfchunk-.LPG1(%r13) .align 4 -.Lget_ipl_device_addr: - .long .Lget_ipl_device .Lpmask: .byte 0 .align 8 @@ -242,6 +259,8 @@ startup:basr %r13,0 # get base .long 0 # cr13: home space segment table .long 0xc0000000 # cr14: machine check handling off .long 0 # cr15: linkage stack operations +.Lduct: .long 0,0,0,0,0,0,0,0 + .long 0,0,0,0,0,0,0,0 .Lpcmem:.long 0x00080000,0x80000000 + .Lchkmem .Lpcfpu:.long 0x00080000,0x80000000 + .Lchkfpu .Lpccsp:.long 0x00080000,0x80000000 + .Lchkcsp @@ -252,25 +271,9 @@ startup:basr %r13,0 # get base .Lmflags:.long machine_flags .Lbss_bgn: .long __bss_start .Lbss_end: .long _end - - .org PARMAREA-64 -.Lduct: .long 0,0,0,0,0,0,0,0 - .long 0,0,0,0,0,0,0,0 - -# -# params at 10400 (setup.h) -# - .org PARMAREA - .global _pstart -_pstart: - .long 0,0 # IPL_DEVICE - .long 0,RAMDISK_ORIGIN # INITRD_START - .long 0,RAMDISK_SIZE # INITRD_SIZE - - .org COMMAND_LINE - .byte "root=/dev/ram0 ro" - .byte 0 - .org 0x11000 +.Lparmaddr: .long PARMAREA +.Lsccbaddr: .long .Lsccb + .align 4096 .Lsccb: .hword 0x1000 # length, one page .byte 0x00,0x00,0x00 @@ -287,18 +290,14 @@ _pstart: .Lscpincr2: .quad 0x00 .fill 3984,1,0 - .org 0x12000 - .global _pend -_pend: - - GET_IPL_DEVICE + .align 4096 #ifdef CONFIG_SHARED_KERNEL .org 0x100000 #endif # -# startup-code, running in virtual mode +# startup-code, running in absolute addressing mode # .globl _stext _stext: basr %r13,0 # get base diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S index f08c06f45d5..47744fcca93 100644 --- a/arch/s390/kernel/head64.S +++ b/arch/s390/kernel/head64.S @@ -1,7 +1,7 @@ /* * arch/s390/kernel/head64.S * - * (C) Copyright IBM Corp. 1999,2005 + * Copyright (C) IBM Corp. 1999,2006 * * Author(s): Hartmut Penner <hp@de.ibm.com> * Martin Schwidefsky <schwidefsky@de.ibm.com> @@ -15,18 +15,37 @@ # this is called either by the ipl loader or directly by PSW restart # or linload or SALIPL # - .org 0x10000 -startup:basr %r13,0 # get base + .org 0x10000 +startup:basr %r13,0 # get base +.LPG0: l %r13,0f-.LPG0(%r13) + b 0(%r13) +0: .long startup_continue + +# +# params at 10400 (setup.h) +# + .org PARMAREA + .quad 0 # IPL_DEVICE + .quad RAMDISK_ORIGIN # INITRD_START + .quad RAMDISK_SIZE # INITRD_SIZE + + .org COMMAND_LINE + .byte "root=/dev/ram0 ro" + .byte 0 + + .org 0x11000 + +startup_continue: + basr %r13,0 # get base .LPG1: sll %r13,1 # remove high order bit srl %r13,1 - l %r1,.Lget_ipl_device_addr-.LPG1(%r13) - basr %r14,%r1 + GET_IPL_DEVICE lhi %r1,1 # mode 1 = esame slr %r0,%r0 # set cpuid to zero sigp %r1,%r0,0x12 # switch to esame mode sam64 # switch to 64 bit mode lctlg %c0,%c15,.Lctl-.LPG1(%r13) # load control registers - larl %r12,_pstart # pointer to parameter area + lg %r12,.Lparmaddr-.LPG1(%r13)# pointer to parameter area # move IPL device to lowcore mvc __LC_IPLDEV(4),IPL_DEVICE+4-PARMAREA(%r12) @@ -55,8 +74,8 @@ startup:basr %r13,0 # get base larl %r1,.Lsclph stg %r1,__LC_EXT_NEW_PSW+8 # set handler - larl %r4,_pstart # %r4 is our index for sccb stuff - la %r1,.Lsccb-PARMAREA(%r4) # our sccb + larl %r4,.Lsccb # %r4 is our index for sccb stuff + lgr %r1,%r4 # our sccb .insn rre,0xb2200000,%r2,%r1 # service call ipm %r1 srl %r1,28 # get cc code @@ -67,7 +86,7 @@ startup:basr %r13,0 # get base be .Lservicecall-.LPG1(%r13) lpswe .Lwaitsclp-.LPG1(%r13) .Lsclph: - lh %r1,.Lsccbr-PARMAREA(%r4) + lh %r1,.Lsccbr-.Lsccb(%r4) chi %r1,0x10 # 0x0010 is the sucess code je .Lprocsccb # let's process the sccb chi %r1,0x1f0 @@ -78,15 +97,15 @@ startup:basr %r13,0 # get base b .Lservicecall-.LPG1(%r13) .Lprocsccb: lghi %r1,0 - icm %r1,3,.Lscpincr1-PARMAREA(%r4) # use this one if != 0 + icm %r1,3,.Lscpincr1-.Lsccb(%r4) # use this one if != 0 jnz .Lscnd - lg %r1,.Lscpincr2-PARMAREA(%r4) # otherwise use this one + lg %r1,.Lscpincr2-.Lsccb(%r4) # otherwise use this one .Lscnd: xr %r3,%r3 # same logic - ic %r3,.Lscpa1-PARMAREA(%r4) + ic %r3,.Lscpa1-.Lsccb(%r4) chi %r3,0x00 jne .Lcompmem - l %r3,.Lscpa2-PARMAREA(%r13) + l %r3,.Lscpa2-.Lsccb(%r4) .Lcompmem: mlgr %r2,%r1 # mem in MB on 128-bit l %r1,.Lonemb-.LPG1(%r13) @@ -94,8 +113,6 @@ startup:basr %r13,0 # get base b .Lfchunk-.LPG1(%r13) .align 4 -.Lget_ipl_device_addr: - .long .Lget_ipl_device .Lpmask: .byte 0 .align 8 @@ -242,29 +259,16 @@ startup:basr %r13,0 # get base .quad 0 # cr13: home space segment table .quad 0xc0000000 # cr14: machine check handling off .quad 0 # cr15: linkage stack operations +.Lduct: .long 0,0,0,0,0,0,0,0 + .long 0,0,0,0,0,0,0,0 .Lpcmsk:.quad 0x0000000180000000 .L4malign:.quad 0xffffffffffc00000 .Lscan2g:.quad 0x80000000 + 0x20000 - 8 # 2GB + 128K - 8 .Lnop: .long 0x07000700 +.Lparmaddr: + .quad PARMAREA - .org PARMAREA-64 -.Lduct: .long 0,0,0,0,0,0,0,0 - .long 0,0,0,0,0,0,0,0 - -# -# params at 10400 (setup.h) -# - .org PARMAREA - .global _pstart -_pstart: - .quad 0 # IPL_DEVICE - .quad RAMDISK_ORIGIN # INITRD_START - .quad RAMDISK_SIZE # INITRD_SIZE - - .org COMMAND_LINE - .byte "root=/dev/ram0 ro" - .byte 0 - .org 0x11000 + .align 4096 .Lsccb: .hword 0x1000 # length, one page .byte 0x00,0x00,0x00 @@ -281,18 +285,14 @@ _pstart: .Lscpincr2: .quad 0x00 .fill 3984,1,0 - .org 0x12000 - .global _pend -_pend: - - GET_IPL_DEVICE + .align 4096 #ifdef CONFIG_SHARED_KERNEL .org 0x100000 #endif # -# startup-code, running in virtual mode +# startup-code, running in absolute addressing mode # .globl _stext _stext: basr %r13,0 # get base @@ -326,4 +326,3 @@ _stext: basr %r13,0 # get base .align 8 .Ldw: .quad 0x0002000180000000,0x0000000000000000 .Laregs: .long 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 - diff --git a/arch/s390/kernel/machine_kexec.c b/arch/s390/kernel/machine_kexec.c index bad81b5832d..fbde6a91526 100644 --- a/arch/s390/kernel/machine_kexec.c +++ b/arch/s390/kernel/machine_kexec.c @@ -27,8 +27,8 @@ static void kexec_halt_all_cpus(void *); typedef void (*relocate_kernel_t) (kimage_entry_t *, unsigned long); -const extern unsigned char relocate_kernel[]; -const extern unsigned long long relocate_kernel_len; +extern const unsigned char relocate_kernel[]; +extern const unsigned long long relocate_kernel_len; int machine_kexec_prepare(struct kimage *image) diff --git a/arch/s390/kernel/s390_ksyms.c b/arch/s390/kernel/s390_ksyms.c index 4176c77670c..0886e739d12 100644 --- a/arch/s390/kernel/s390_ksyms.c +++ b/arch/s390/kernel/s390_ksyms.c @@ -46,8 +46,6 @@ EXPORT_SYMBOL(__down_interruptible); */ extern int dump_fpu (struct pt_regs * regs, s390_fp_regs *fpregs); EXPORT_SYMBOL(dump_fpu); -EXPORT_SYMBOL(overflowuid); -EXPORT_SYMBOL(overflowgid); EXPORT_SYMBOL(empty_zero_page); /* diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 0a04e4a564b..2b2551e3510 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -37,6 +37,7 @@ #include <linux/seq_file.h> #include <linux/kernel_stat.h> #include <linux/device.h> +#include <linux/notifier.h> #include <asm/uaccess.h> #include <asm/system.h> @@ -47,6 +48,7 @@ #include <asm/irq.h> #include <asm/page.h> #include <asm/ptrace.h> +#include <asm/sections.h> /* * Machine setup.. @@ -66,11 +68,6 @@ unsigned long __initdata zholes_size[MAX_NR_ZONES]; static unsigned long __initdata memory_end; /* - * Setup options - */ -extern int _text,_etext, _edata, _end; - -/* * This is set up by the setup-routine at boot-time * for S390 need to find out, what we have to setup * using address 0x10400 ... @@ -80,15 +77,11 @@ extern int _text,_etext, _edata, _end; static struct resource code_resource = { .name = "Kernel code", - .start = (unsigned long) &_text, - .end = (unsigned long) &_etext - 1, .flags = IORESOURCE_BUSY | IORESOURCE_MEM, }; static struct resource data_resource = { .name = "Kernel data", - .start = (unsigned long) &_etext, - .end = (unsigned long) &_edata - 1, .flags = IORESOURCE_BUSY | IORESOURCE_MEM, }; @@ -123,6 +116,7 @@ void __devinit cpu_init (void) */ char vmhalt_cmd[128] = ""; char vmpoff_cmd[128] = ""; +char vmpanic_cmd[128] = ""; static inline void strncpy_skip_quote(char *dst, char *src, int n) { @@ -154,6 +148,38 @@ static int __init vmpoff_setup(char *str) __setup("vmpoff=", vmpoff_setup); +static int vmpanic_notify(struct notifier_block *self, unsigned long event, + void *data) +{ + if (MACHINE_IS_VM && strlen(vmpanic_cmd) > 0) + cpcmd(vmpanic_cmd, NULL, 0, NULL); + + return NOTIFY_OK; +} + +#define PANIC_PRI_VMPANIC 0 + +static struct notifier_block vmpanic_nb = { + .notifier_call = vmpanic_notify, + .priority = PANIC_PRI_VMPANIC +}; + +static int __init vmpanic_setup(char *str) +{ + static int register_done __initdata = 0; + + strncpy_skip_quote(vmpanic_cmd, str, 127); + vmpanic_cmd[127] = 0; + if (!register_done) { + register_done = 1; + atomic_notifier_chain_register(&panic_notifier_list, + &vmpanic_nb); + } + return 1; +} + +__setup("vmpanic=", vmpanic_setup); + /* * condev= and conmode= setup parameter. */ @@ -297,19 +323,34 @@ void (*_machine_power_off)(void) = do_machine_power_off_nonsmp; void machine_restart(char *command) { - console_unblank(); + if (!in_interrupt() || oops_in_progress) + /* + * Only unblank the console if we are called in enabled + * context or a bust_spinlocks cleared the way for us. + */ + console_unblank(); _machine_restart(command); } void machine_halt(void) { - console_unblank(); + if (!in_interrupt() || oops_in_progress) + /* + * Only unblank the console if we are called in enabled + * context or a bust_spinlocks cleared the way for us. + */ + console_unblank(); _machine_halt(); } void machine_power_off(void) { - console_unblank(); + if (!in_interrupt() || oops_in_progress) + /* + * Only unblank the console if we are called in enabled + * context or a bust_spinlocks cleared the way for us. + */ + console_unblank(); _machine_power_off(); } @@ -422,6 +463,11 @@ setup_resources(void) struct resource *res; int i; + code_resource.start = (unsigned long) &_text; + code_resource.end = (unsigned long) &_etext - 1; + data_resource.start = (unsigned long) &_etext; + data_resource.end = (unsigned long) &_edata - 1; + for (i = 0; i < MEMORY_CHUNKS && memory_chunk[i].size > 0; i++) { res = alloc_bootmem_low(sizeof(struct resource)); res->flags = IORESOURCE_BUSY | IORESOURCE_MEM; diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index 343120c9223..8e03219eea7 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -869,7 +869,7 @@ static int __init topology_init(void) int ret; for_each_possible_cpu(cpu) { - ret = register_cpu(&per_cpu(cpu_devices, cpu), cpu, NULL); + ret = register_cpu(&per_cpu(cpu_devices, cpu), cpu); if (ret) printk(KERN_WARNING "topology_init: register_cpu %d " "failed (%d)\n", cpu, ret); diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c index a46793beedd..b7630436f69 100644 --- a/arch/s390/kernel/traps.c +++ b/arch/s390/kernel/traps.c @@ -150,13 +150,11 @@ void show_stack(struct task_struct *task, unsigned long *sp) unsigned long *stack; int i; - // debugging aid: "show_stack(NULL);" prints the - // back trace for this cpu. - if (!sp) - sp = task ? (unsigned long *) task->thread.ksp : __r15; + stack = task ? (unsigned long *) task->thread.ksp : __r15; + else + stack = sp; - stack = sp; for (i = 0; i < kstack_depth_to_print; i++) { if (((addr_t) stack & (THREAD_SIZE-1)) == 0) break; diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c index dfe6f085661..1f0439dc245 100644 --- a/arch/s390/kernel/vtime.c +++ b/arch/s390/kernel/vtime.c @@ -356,7 +356,7 @@ static void internal_add_vtimer(struct vtimer_list *timer) set_vtimer(event->expires); spin_unlock_irqrestore(&vt_list->lock, flags); - /* release CPU aquired in prepare_vtimer or mod_virt_timer() */ + /* release CPU acquired in prepare_vtimer or mod_virt_timer() */ put_cpu(); } |