Author | Tokens | Token Proportion | Commits | Commit Proportion |
---|---|---|---|---|
Martin Schwidefsky | 1546 | 51.48% | 7 | 13.73% |
Heiko Carstens | 1132 | 37.70% | 20 | 39.22% |
Ekaterina Tumanova | 103 | 3.43% | 1 | 1.96% |
Viktor Mihajlovski | 90 | 3.00% | 1 | 1.96% |
Linus Torvalds | 56 | 1.86% | 6 | 11.76% |
Linus Torvalds (pre-git) | 49 | 1.63% | 7 | 13.73% |
Hendrik Brueckner | 12 | 0.40% | 1 | 1.96% |
Christoph Hellwig | 5 | 0.17% | 3 | 5.88% |
Christian Bornträger | 3 | 0.10% | 1 | 1.96% |
Frank Blaschka | 3 | 0.10% | 1 | 1.96% |
Andrew Morton | 2 | 0.07% | 1 | 1.96% |
Greg Kroah-Hartman | 1 | 0.03% | 1 | 1.96% |
Paul Gortmaker | 1 | 0.03% | 1 | 1.96% |
Total | 3003 | 51 |
// SPDX-License-Identifier: GPL-2.0 /* * Copyright IBM Corp. 2001, 2009 * Author(s): Ulrich Weigand <Ulrich.Weigand@de.ibm.com>, * Martin Schwidefsky <schwidefsky@de.ibm.com>, */ #include <linux/debugfs.h> #include <linux/kernel.h> #include <linux/mm.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> #include <linux/init.h> #include <linux/delay.h> #include <linux/export.h> #include <linux/slab.h> #include <asm/asm-extable.h> #include <asm/ebcdic.h> #include <asm/debug.h> #include <asm/sysinfo.h> #include <asm/cpcmd.h> #include <asm/topology.h> #include <asm/fpu/api.h> int topology_max_mnest; static inline int __stsi(void *sysinfo, int fc, int sel1, int sel2, int *lvl) { int r0 = (fc << 28) | sel1; int rc = 0; asm volatile( " lr 0,%[r0]\n" " lr 1,%[r1]\n" " stsi 0(%[sysinfo])\n" "0: jz 2f\n" "1: lhi %[rc],%[retval]\n" "2: lr %[r0],0\n" EX_TABLE(0b, 1b) : [r0] "+d" (r0), [rc] "+d" (rc) : [r1] "d" (sel2), [sysinfo] "a" (sysinfo), [retval] "K" (-EOPNOTSUPP) : "cc", "0", "1", "memory"); *lvl = ((unsigned int) r0) >> 28; return rc; } /* * stsi - store system information * * Returns the current configuration level if function code 0 was specified. * Otherwise returns 0 on success or a negative value on error. */ int stsi(void *sysinfo, int fc, int sel1, int sel2) { int lvl, rc; rc = __stsi(sysinfo, fc, sel1, sel2, &lvl); if (rc) return rc; return fc ? 0 : lvl; } EXPORT_SYMBOL(stsi); #ifdef CONFIG_PROC_FS static bool convert_ext_name(unsigned char encoding, char *name, size_t len) { switch (encoding) { case 1: /* EBCDIC */ EBCASC(name, len); break; case 2: /* UTF-8 */ break; default: return false; } return true; } static void stsi_1_1_1(struct seq_file *m, struct sysinfo_1_1_1 *info) { int i; if (stsi(info, 1, 1, 1)) return; EBCASC(info->manufacturer, sizeof(info->manufacturer)); EBCASC(info->type, sizeof(info->type)); EBCASC(info->model, sizeof(info->model)); EBCASC(info->sequence, sizeof(info->sequence)); EBCASC(info->plant, sizeof(info->plant)); EBCASC(info->model_capacity, sizeof(info->model_capacity)); EBCASC(info->model_perm_cap, sizeof(info->model_perm_cap)); EBCASC(info->model_temp_cap, sizeof(info->model_temp_cap)); seq_printf(m, "Manufacturer: %-16.16s\n", info->manufacturer); seq_printf(m, "Type: %-4.4s\n", info->type); if (info->lic) seq_printf(m, "LIC Identifier: %016lx\n", info->lic); /* * Sigh: the model field has been renamed with System z9 * to model_capacity and a new model field has been added * after the plant field. To avoid confusing older programs * the "Model:" prints "model_capacity model" or just * "model_capacity" if the model string is empty . */ seq_printf(m, "Model: %-16.16s", info->model_capacity); if (info->model[0] != '\0') seq_printf(m, " %-16.16s", info->model); seq_putc(m, '\n'); seq_printf(m, "Sequence Code: %-16.16s\n", info->sequence); seq_printf(m, "Plant: %-4.4s\n", info->plant); seq_printf(m, "Model Capacity: %-16.16s %08u\n", info->model_capacity, info->model_cap_rating); if (info->model_perm_cap_rating) seq_printf(m, "Model Perm. Capacity: %-16.16s %08u\n", info->model_perm_cap, info->model_perm_cap_rating); if (info->model_temp_cap_rating) seq_printf(m, "Model Temp. Capacity: %-16.16s %08u\n", info->model_temp_cap, info->model_temp_cap_rating); if (info->ncr) seq_printf(m, "Nominal Cap. Rating: %08u\n", info->ncr); if (info->npr) seq_printf(m, "Nominal Perm. Rating: %08u\n", info->npr); if (info->ntr) seq_printf(m, "Nominal Temp. Rating: %08u\n", info->ntr); if (info->cai) { seq_printf(m, "Capacity Adj. Ind.: %d\n", info->cai); seq_printf(m, "Capacity Ch. Reason: %d\n", info->ccr); seq_printf(m, "Capacity Transient: %d\n", info->t); } if (info->p) { for (i = 1; i <= ARRAY_SIZE(info->typepct); i++) { seq_printf(m, "Type %d Percentage: %d\n", i, info->typepct[i - 1]); } } } static void stsi_15_1_x(struct seq_file *m, struct sysinfo_15_1_x *info) { int i; seq_putc(m, '\n'); if (!MACHINE_HAS_TOPOLOGY) return; if (stsi(info, 15, 1, topology_max_mnest)) return; seq_printf(m, "CPU Topology HW: "); for (i = 0; i < TOPOLOGY_NR_MAG; i++) seq_printf(m, " %d", info->mag[i]); seq_putc(m, '\n'); #ifdef CONFIG_SCHED_TOPOLOGY store_topology(info); seq_printf(m, "CPU Topology SW: "); for (i = 0; i < TOPOLOGY_NR_MAG; i++) seq_printf(m, " %d", info->mag[i]); seq_putc(m, '\n'); #endif } static void stsi_1_2_2(struct seq_file *m, struct sysinfo_1_2_2 *info) { struct sysinfo_1_2_2_extension *ext; int i; if (stsi(info, 1, 2, 2)) return; ext = (struct sysinfo_1_2_2_extension *) ((unsigned long) info + info->acc_offset); seq_printf(m, "CPUs Total: %d\n", info->cpus_total); seq_printf(m, "CPUs Configured: %d\n", info->cpus_configured); seq_printf(m, "CPUs Standby: %d\n", info->cpus_standby); seq_printf(m, "CPUs Reserved: %d\n", info->cpus_reserved); if (info->mt_installed) { seq_printf(m, "CPUs G-MTID: %d\n", info->mt_gtid); seq_printf(m, "CPUs S-MTID: %d\n", info->mt_stid); } /* * Sigh 2. According to the specification the alternate * capability field is a 32 bit floating point number * if the higher order 8 bits are not zero. Printing * a floating point number in the kernel is a no-no, * always print the number as 32 bit unsigned integer. * The user-space needs to know about the strange * encoding of the alternate cpu capability. */ seq_printf(m, "Capability: %u", info->capability); if (info->format == 1) seq_printf(m, " %u", ext->alt_capability); seq_putc(m, '\n'); if (info->nominal_cap) seq_printf(m, "Nominal Capability: %d\n", info->nominal_cap); if (info->secondary_cap) seq_printf(m, "Secondary Capability: %d\n", info->secondary_cap); for (i = 2; i <= info->cpus_total; i++) { seq_printf(m, "Adjustment %02d-way: %u", i, info->adjustment[i-2]); if (info->format == 1) seq_printf(m, " %u", ext->alt_adjustment[i-2]); seq_putc(m, '\n'); } } static void stsi_2_2_2(struct seq_file *m, struct sysinfo_2_2_2 *info) { if (stsi(info, 2, 2, 2)) return; EBCASC(info->name, sizeof(info->name)); seq_putc(m, '\n'); seq_printf(m, "LPAR Number: %d\n", info->lpar_number); seq_printf(m, "LPAR Characteristics: "); if (info->characteristics & LPAR_CHAR_DEDICATED) seq_printf(m, "Dedicated "); if (info->characteristics & LPAR_CHAR_SHARED) seq_printf(m, "Shared "); if (info->characteristics & LPAR_CHAR_LIMITED) seq_printf(m, "Limited "); seq_putc(m, '\n'); seq_printf(m, "LPAR Name: %-8.8s\n", info->name); seq_printf(m, "LPAR Adjustment: %d\n", info->caf); seq_printf(m, "LPAR CPUs Total: %d\n", info->cpus_total); seq_printf(m, "LPAR CPUs Configured: %d\n", info->cpus_configured); seq_printf(m, "LPAR CPUs Standby: %d\n", info->cpus_standby); seq_printf(m, "LPAR CPUs Reserved: %d\n", info->cpus_reserved); seq_printf(m, "LPAR CPUs Dedicated: %d\n", info->cpus_dedicated); seq_printf(m, "LPAR CPUs Shared: %d\n", info->cpus_shared); if (info->mt_installed) { seq_printf(m, "LPAR CPUs G-MTID: %d\n", info->mt_gtid); seq_printf(m, "LPAR CPUs S-MTID: %d\n", info->mt_stid); seq_printf(m, "LPAR CPUs PS-MTID: %d\n", info->mt_psmtid); } if (convert_ext_name(info->vsne, info->ext_name, sizeof(info->ext_name))) { seq_printf(m, "LPAR Extended Name: %-.256s\n", info->ext_name); seq_printf(m, "LPAR UUID: %pUb\n", &info->uuid); } } static void print_ext_name(struct seq_file *m, int lvl, struct sysinfo_3_2_2 *info) { size_t len = sizeof(info->ext_names[lvl]); if (!convert_ext_name(info->vm[lvl].evmne, info->ext_names[lvl], len)) return; seq_printf(m, "VM%02d Extended Name: %-.256s\n", lvl, info->ext_names[lvl]); } static void print_uuid(struct seq_file *m, int i, struct sysinfo_3_2_2 *info) { if (uuid_is_null(&info->vm[i].uuid)) return; seq_printf(m, "VM%02d UUID: %pUb\n", i, &info->vm[i].uuid); } static void stsi_3_2_2(struct seq_file *m, struct sysinfo_3_2_2 *info) { int i; if (stsi(info, 3, 2, 2)) return; for (i = 0; i < info->count; i++) { EBCASC(info->vm[i].name, sizeof(info->vm[i].name)); EBCASC(info->vm[i].cpi, sizeof(info->vm[i].cpi)); seq_putc(m, '\n'); seq_printf(m, "VM%02d Name: %-8.8s\n", i, info->vm[i].name); seq_printf(m, "VM%02d Control Program: %-16.16s\n", i, info->vm[i].cpi); seq_printf(m, "VM%02d Adjustment: %d\n", i, info->vm[i].caf); seq_printf(m, "VM%02d CPUs Total: %d\n", i, info->vm[i].cpus_total); seq_printf(m, "VM%02d CPUs Configured: %d\n", i, info->vm[i].cpus_configured); seq_printf(m, "VM%02d CPUs Standby: %d\n", i, info->vm[i].cpus_standby); seq_printf(m, "VM%02d CPUs Reserved: %d\n", i, info->vm[i].cpus_reserved); print_ext_name(m, i, info); print_uuid(m, i, info); } } static int sysinfo_show(struct seq_file *m, void *v) { void *info = (void *)get_zeroed_page(GFP_KERNEL); int level; if (!info) return 0; level = stsi(NULL, 0, 0, 0); if (level >= 1) stsi_1_1_1(m, info); if (level >= 1) stsi_15_1_x(m, info); if (level >= 1) stsi_1_2_2(m, info); if (level >= 2) stsi_2_2_2(m, info); if (level >= 3) stsi_3_2_2(m, info); free_page((unsigned long)info); return 0; } static int __init sysinfo_create_proc(void) { proc_create_single("sysinfo", 0444, NULL, sysinfo_show); return 0; } device_initcall(sysinfo_create_proc); #endif /* CONFIG_PROC_FS */ /* * Service levels interface. */ static DECLARE_RWSEM(service_level_sem); static LIST_HEAD(service_level_list); int register_service_level(struct service_level *slr) { struct service_level *ptr; down_write(&service_level_sem); list_for_each_entry(ptr, &service_level_list, list) if (ptr == slr) { up_write(&service_level_sem); return -EEXIST; } list_add_tail(&slr->list, &service_level_list); up_write(&service_level_sem); return 0; } EXPORT_SYMBOL(register_service_level); int unregister_service_level(struct service_level *slr) { struct service_level *ptr, *next; int rc = -ENOENT; down_write(&service_level_sem); list_for_each_entry_safe(ptr, next, &service_level_list, list) { if (ptr != slr) continue; list_del(&ptr->list); rc = 0; break; } up_write(&service_level_sem); return rc; } EXPORT_SYMBOL(unregister_service_level); static void *service_level_start(struct seq_file *m, loff_t *pos) { down_read(&service_level_sem); return seq_list_start(&service_level_list, *pos); } static void *service_level_next(struct seq_file *m, void *p, loff_t *pos) { return seq_list_next(p, &service_level_list, pos); } static void service_level_stop(struct seq_file *m, void *p) { up_read(&service_level_sem); } static int service_level_show(struct seq_file *m, void *p) { struct service_level *slr; slr = list_entry(p, struct service_level, list); slr->seq_print(m, slr); return 0; } static const struct seq_operations service_level_seq_ops = { .start = service_level_start, .next = service_level_next, .stop = service_level_stop, .show = service_level_show }; static void service_level_vm_print(struct seq_file *m, struct service_level *slr) { char *query_buffer, *str; query_buffer = kmalloc(1024, GFP_KERNEL | GFP_DMA); if (!query_buffer) return; cpcmd("QUERY CPLEVEL", query_buffer, 1024, NULL); str = strchr(query_buffer, '\n'); if (str) *str = 0; seq_printf(m, "VM: %s\n", query_buffer); kfree(query_buffer); } static struct service_level service_level_vm = { .seq_print = service_level_vm_print }; static __init int create_proc_service_level(void) { proc_create_seq("service_levels", 0, NULL, &service_level_seq_ops); if (MACHINE_IS_VM) register_service_level(&service_level_vm); return 0; } subsys_initcall(create_proc_service_level); /* * CPU capability might have changed. Therefore recalculate loops_per_jiffy. */ void s390_adjust_jiffies(void) { struct sysinfo_1_2_2 *info; unsigned long capability; struct kernel_fpu fpu; info = (void *) get_zeroed_page(GFP_KERNEL); if (!info) return; if (stsi(info, 1, 2, 2) == 0) { /* * Major sigh. The cpu capability encoding is "special". * If the first 9 bits of info->capability are 0 then it * is a 32 bit unsigned integer in the range 0 .. 2^23. * If the first 9 bits are != 0 then it is a 32 bit float. * In addition a lower value indicates a proportionally * higher cpu capacity. Bogomips are the other way round. * To get to a halfway suitable number we divide 1e7 * by the cpu capability number. Yes, that means a floating * point division .. */ kernel_fpu_begin(&fpu, KERNEL_FPR); asm volatile( " sfpc %3\n" " l %0,%1\n" " tmlh %0,0xff80\n" " jnz 0f\n" " cefbr %%f2,%0\n" " j 1f\n" "0: le %%f2,%1\n" "1: cefbr %%f0,%2\n" " debr %%f0,%%f2\n" " cgebr %0,5,%%f0\n" : "=&d" (capability) : "Q" (info->capability), "d" (10000000), "d" (0) : "cc" ); kernel_fpu_end(&fpu, KERNEL_FPR); } else /* * Really old machine without stsi block for basic * cpu information. Report 42.0 bogomips. */ capability = 42; loops_per_jiffy = capability * (500000/HZ); free_page((unsigned long) info); } /* * calibrate the delay loop */ void calibrate_delay(void) { s390_adjust_jiffies(); /* Print the good old Bogomips line .. */ printk(KERN_DEBUG "Calibrating delay loop (skipped)... " "%lu.%02lu BogoMIPS preset\n", loops_per_jiffy/(500000/HZ), (loops_per_jiffy/(5000/HZ)) % 100); } #ifdef CONFIG_DEBUG_FS #define STSI_FILE(fc, s1, s2) \ static int stsi_open_##fc##_##s1##_##s2(struct inode *inode, struct file *file)\ { \ file->private_data = (void *) get_zeroed_page(GFP_KERNEL); \ if (!file->private_data) \ return -ENOMEM; \ if (stsi(file->private_data, fc, s1, s2)) { \ free_page((unsigned long)file->private_data); \ file->private_data = NULL; \ return -EACCES; \ } \ return nonseekable_open(inode, file); \ } \ \ static const struct file_operations stsi_##fc##_##s1##_##s2##_fs_ops = { \ .open = stsi_open_##fc##_##s1##_##s2, \ .release = stsi_release, \ .read = stsi_read, \ .llseek = no_llseek, \ }; static int stsi_release(struct inode *inode, struct file *file) { free_page((unsigned long)file->private_data); return 0; } static ssize_t stsi_read(struct file *file, char __user *buf, size_t size, loff_t *ppos) { return simple_read_from_buffer(buf, size, ppos, file->private_data, PAGE_SIZE); } STSI_FILE( 1, 1, 1); STSI_FILE( 1, 2, 1); STSI_FILE( 1, 2, 2); STSI_FILE( 2, 2, 1); STSI_FILE( 2, 2, 2); STSI_FILE( 3, 2, 2); STSI_FILE(15, 1, 2); STSI_FILE(15, 1, 3); STSI_FILE(15, 1, 4); STSI_FILE(15, 1, 5); STSI_FILE(15, 1, 6); struct stsi_file { const struct file_operations *fops; char *name; }; static struct stsi_file stsi_file[] __initdata = { {.fops = &stsi_1_1_1_fs_ops, .name = "1_1_1"}, {.fops = &stsi_1_2_1_fs_ops, .name = "1_2_1"}, {.fops = &stsi_1_2_2_fs_ops, .name = "1_2_2"}, {.fops = &stsi_2_2_1_fs_ops, .name = "2_2_1"}, {.fops = &stsi_2_2_2_fs_ops, .name = "2_2_2"}, {.fops = &stsi_3_2_2_fs_ops, .name = "3_2_2"}, {.fops = &stsi_15_1_2_fs_ops, .name = "15_1_2"}, {.fops = &stsi_15_1_3_fs_ops, .name = "15_1_3"}, {.fops = &stsi_15_1_4_fs_ops, .name = "15_1_4"}, {.fops = &stsi_15_1_5_fs_ops, .name = "15_1_5"}, {.fops = &stsi_15_1_6_fs_ops, .name = "15_1_6"}, }; static u8 stsi_0_0_0; static __init int stsi_init_debugfs(void) { struct dentry *stsi_root; struct stsi_file *sf; int lvl, i; stsi_root = debugfs_create_dir("stsi", arch_debugfs_dir); lvl = stsi(NULL, 0, 0, 0); if (lvl > 0) stsi_0_0_0 = lvl; debugfs_create_u8("0_0_0", 0400, stsi_root, &stsi_0_0_0); for (i = 0; i < ARRAY_SIZE(stsi_file); i++) { sf = &stsi_file[i]; debugfs_create_file(sf->name, 0400, stsi_root, NULL, sf->fops); } if (IS_ENABLED(CONFIG_SCHED_TOPOLOGY) && MACHINE_HAS_TOPOLOGY) { char link_to[10]; sprintf(link_to, "15_1_%d", topology_mnest_limit()); debugfs_create_symlink("topology", stsi_root, link_to); } return 0; } device_initcall(stsi_init_debugfs); #endif /* CONFIG_DEBUG_FS */
Information contained on this website is for historical information purposes only and does not indicate or represent copyright ownership.
Created with Cregit http://github.com/cregit/cregit
Version 2.0-RC1