diff options
Diffstat (limited to 'tools')
26 files changed, 606 insertions, 234 deletions
diff --git a/tools/iio/iio_event_monitor.c b/tools/iio/iio_event_monitor.c index 8073c9e4fe46..cccf62ea2b8f 100644 --- a/tools/iio/iio_event_monitor.c +++ b/tools/iio/iio_event_monitor.c @@ -63,6 +63,7 @@ static const char * const iio_chan_type_name_spec[] = { [IIO_DELTA_VELOCITY] = "deltavelocity", [IIO_COLORTEMP] = "colortemp", [IIO_CHROMATICITY] = "chromaticity", + [IIO_ATTENTION] = "attention", }; static const char * const iio_ev_type_text[] = { @@ -183,6 +184,7 @@ static bool event_is_known(struct iio_event_data *event) case IIO_DELTA_VELOCITY: case IIO_COLORTEMP: case IIO_CHROMATICITY: + case IIO_ATTENTION: break; default: return false; @@ -449,6 +451,7 @@ error_free_chrdev_name: enable_events(dev_dir_name, 0); free(chrdev_name); + free(dev_dir_name); return ret; } diff --git a/tools/objtool/check.c b/tools/objtool/check.c index f7586f82b967..4ce176ad411f 100644 --- a/tools/objtool/check.c +++ b/tools/objtool/check.c @@ -4573,6 +4573,8 @@ static int validate_ibt(struct objtool_file *file) !strcmp(sec->name, "__jump_table") || !strcmp(sec->name, "__mcount_loc") || !strcmp(sec->name, ".kcfi_traps") || + !strcmp(sec->name, ".llvm.call-graph-profile") || + !strcmp(sec->name, ".llvm_bb_addr_map") || !strcmp(sec->name, "__tracepoints") || strstr(sec->name, "__patchable_function_entries")) continue; diff --git a/tools/objtool/elf.c b/tools/objtool/elf.c index 3d27983dc908..6f64d611faea 100644 --- a/tools/objtool/elf.c +++ b/tools/objtool/elf.c @@ -224,12 +224,17 @@ int find_symbol_hole_containing(const struct section *sec, unsigned long offset) if (n) return 0; /* not a hole */ - /* didn't find a symbol for which @offset is after it */ - if (!hole.sym) - return 0; /* not a hole */ + /* + * @offset >= sym->offset + sym->len, find symbol after it. + * When hole.sym is empty, use the first node to compute the hole. + * If there is no symbol in the section, the first node will be NULL, + * in which case, -1 is returned to skip the whole section. + */ + if (hole.sym) + n = rb_next(&hole.sym->node); + else + n = rb_first_cached(&sec->symbol_tree); - /* @offset >= sym->offset + sym->len, find symbol after it */ - n = rb_next(&hole.sym->node); if (!n) return -1; /* until end of address space */ diff --git a/tools/power/x86/turbostat/turbostat.8 b/tools/power/x86/turbostat/turbostat.8 index 067717bce1d4..a7f7ed01421c 100644 --- a/tools/power/x86/turbostat/turbostat.8 +++ b/tools/power/x86/turbostat/turbostat.8 @@ -33,6 +33,9 @@ name as necessary to disambiguate it from others is necessary. Note that option msr0xXXX is a hex offset, eg. msr0x10 /sys/path... is an absolute path to a sysfs attribute <device> is a perf device from /sys/bus/event_source/devices/<device> eg. cstate_core + On Intel hybrid platforms, instead of one "cpu" perf device there are two, "cpu_core" and "cpu_atom" devices for P and E cores respectively. + Turbostat, in this case, allow user to use "cpu" device and will automatically detect the type of a CPU and translate it to "cpu_core" and "cpu_atom" accordingly. + For a complete example see "ADD PERF COUNTER EXAMPLE #2 (using virtual "cpu" device)". <event> is a perf event for given device from /sys/bus/event_source/devices/<device>/events/<event> eg. c1-residency perf/cstate_core/c1-residency would then use /sys/bus/event_source/devices/cstate_core/events/c1-residency @@ -187,6 +190,8 @@ The system configuration dump (if --quiet is not used) is followed by statistics .PP \fBRAMWatt\fP Watts consumed by the DRAM DIMMS -- available only on server processors. .PP +\fBSysWatt\fP Watts consumed by the whole platform (RAPL PSYS). Disabled by default. Enable with --enable SysWatt. +.PP \fBPKG_%\fP percent of the interval that RAPL throttling was active on the Package. Note that the system summary is the sum of the package throttling time, and thus may be higher than 100% on a multi-package system. Note that the meaning of this field is model specific. For example, some hardware increments this counter when RAPL responds to thermal limits, but does not increment this counter when RAPL responds to power limits. Comparing PkgWatt and PkgTmp to system limits is necessary. .PP \fBRAM_%\fP percent of the interval that RAPL throttling was active on DRAM. @@ -387,6 +392,28 @@ CPU pCPU%c1 CPU%c1 .fi +.SH ADD PERF COUNTER EXAMPLE #2 (using virtual cpu device) +Here we run on hybrid, Raptor Lake platform. +We limit turbostat to show output for just cpu0 (pcore) and cpu12 (ecore). +We add a counter showing number of L3 cache misses, using virtual "cpu" device, +labeling it with the column header, "VCMISS". +We add a counter showing number of L3 cache misses, using virtual "cpu_core" device, +labeling it with the column header, "PCMISS". This will fail on ecore cpu12. +We add a counter showing number of L3 cache misses, using virtual "cpu_atom" device, +labeling it with the column header, "ECMISS". This will fail on pcore cpu0. +We display it only once, after the conclusion of 0.1 second sleep. +.nf +sudo ./turbostat --quiet --cpu 0,12 --show CPU --add perf/cpu/cache-misses,cpu,delta,raw,VCMISS --add perf/cpu_core/cache-misses,cpu,delta,raw,PCMISS --add perf/cpu_atom/cache-misses,cpu,delta,raw,ECMISS sleep .1 +turbostat: added_perf_counters_init_: perf/cpu_atom/cache-misses: failed to open counter on cpu0 +turbostat: added_perf_counters_init_: perf/cpu_core/cache-misses: failed to open counter on cpu12 +0.104630 sec +CPU ECMISS PCMISS VCMISS +- 0x0000000000000000 0x0000000000000000 0x0000000000000000 +0 0x0000000000000000 0x0000000000007951 0x0000000000007796 +12 0x000000000001137a 0x0000000000000000 0x0000000000011392 + +.fi + .SH ADD PMT COUNTER EXAMPLE Here we limit turbostat to showing just the CPU number 0. We add two counters, showing crystal clock count and the DC6 residency. diff --git a/tools/power/x86/turbostat/turbostat.c b/tools/power/x86/turbostat/turbostat.c index 089220aaa5c9..58a487c225a7 100644 --- a/tools/power/x86/turbostat/turbostat.c +++ b/tools/power/x86/turbostat/turbostat.c @@ -31,6 +31,9 @@ ) // end copied section +#define CPUID_LEAF_MODEL_ID 0x1A +#define CPUID_LEAF_MODEL_ID_CORE_TYPE_SHIFT 24 + #define X86_VENDOR_INTEL 0 #include INTEL_FAMILY_HEADER @@ -89,6 +92,9 @@ #define PERF_DEV_NAME_BYTES 32 #define PERF_EVT_NAME_BYTES 32 +#define INTEL_ECORE_TYPE 0x20 +#define INTEL_PCORE_TYPE 0x40 + enum counter_scope { SCOPE_CPU, SCOPE_CORE, SCOPE_PACKAGE }; enum counter_type { COUNTER_ITEMS, COUNTER_CYCLES, COUNTER_SECONDS, COUNTER_USEC, COUNTER_K2M }; enum counter_format { FORMAT_RAW, FORMAT_DELTA, FORMAT_PERCENT, FORMAT_AVERAGE }; @@ -194,6 +200,8 @@ struct msr_counter bic[] = { { 0x0, "SAMMHz", NULL, 0, 0, 0, NULL, 0 }, { 0x0, "SAMAMHz", NULL, 0, 0, 0, NULL, 0 }, { 0x0, "Die%c6", NULL, 0, 0, 0, NULL, 0 }, + { 0x0, "SysWatt", NULL, 0, 0, 0, NULL, 0 }, + { 0x0, "Sys_J", NULL, 0, 0, 0, NULL, 0 }, }; #define MAX_BIC (sizeof(bic) / sizeof(struct msr_counter)) @@ -256,6 +264,8 @@ struct msr_counter bic[] = { #define BIC_SAMMHz (1ULL << 56) #define BIC_SAMACTMHz (1ULL << 57) #define BIC_Diec6 (1ULL << 58) +#define BIC_SysWatt (1ULL << 59) +#define BIC_Sys_J (1ULL << 60) #define BIC_TOPOLOGY (BIC_Package | BIC_Node | BIC_CoreCnt | BIC_PkgCnt | BIC_Core | BIC_CPU | BIC_Die ) #define BIC_THERMAL_PWR ( BIC_CoreTmp | BIC_PkgTmp | BIC_PkgWatt | BIC_CorWatt | BIC_GFXWatt | BIC_RAMWatt | BIC_PKG__ | BIC_RAM__) @@ -263,7 +273,7 @@ struct msr_counter bic[] = { #define BIC_IDLE (BIC_sysfs | BIC_CPU_c1 | BIC_CPU_c3 | BIC_CPU_c6 | BIC_CPU_c7 | BIC_GFX_rc6 | BIC_Pkgpc2 | BIC_Pkgpc3 | BIC_Pkgpc6 | BIC_Pkgpc7 | BIC_Pkgpc8 | BIC_Pkgpc9 | BIC_Pkgpc10 | BIC_CPU_LPI | BIC_SYS_LPI | BIC_Mod_c6 | BIC_Totl_c0 | BIC_Any_c0 | BIC_GFX_c0 | BIC_CPUGFX | BIC_SAM_mc6 | BIC_Diec6) #define BIC_OTHER ( BIC_IRQ | BIC_SMI | BIC_ThreadC | BIC_CoreTmp | BIC_IPC) -#define BIC_DISABLED_BY_DEFAULT (BIC_USEC | BIC_TOD | BIC_APIC | BIC_X2APIC) +#define BIC_DISABLED_BY_DEFAULT (BIC_USEC | BIC_TOD | BIC_APIC | BIC_X2APIC | BIC_SysWatt | BIC_Sys_J) unsigned long long bic_enabled = (0xFFFFFFFFFFFFFFFFULL & ~BIC_DISABLED_BY_DEFAULT); unsigned long long bic_present = BIC_USEC | BIC_TOD | BIC_sysfs | BIC_APIC | BIC_X2APIC; @@ -370,7 +380,6 @@ enum gfx_sysfs_idx { }; struct gfx_sysfs_info { - const char *path; FILE *fp; unsigned int val; unsigned long long val_ull; @@ -502,12 +511,15 @@ enum rapl_msrs { RAPL_AMD_PWR_UNIT = BIT(14), /* 0xc0010299 MSR_AMD_RAPL_POWER_UNIT */ RAPL_AMD_CORE_ENERGY_STAT = BIT(15), /* 0xc001029a MSR_AMD_CORE_ENERGY_STATUS */ RAPL_AMD_PKG_ENERGY_STAT = BIT(16), /* 0xc001029b MSR_AMD_PKG_ENERGY_STATUS */ + RAPL_PLATFORM_ENERGY_LIMIT = BIT(17), /* 0x64c MSR_PLATFORM_ENERGY_LIMIT */ + RAPL_PLATFORM_ENERGY_STATUS = BIT(18), /* 0x64d MSR_PLATFORM_ENERGY_STATUS */ }; #define RAPL_PKG (RAPL_PKG_ENERGY_STATUS | RAPL_PKG_POWER_LIMIT) #define RAPL_DRAM (RAPL_DRAM_ENERGY_STATUS | RAPL_DRAM_POWER_LIMIT) #define RAPL_CORE (RAPL_CORE_ENERGY_STATUS | RAPL_CORE_POWER_LIMIT) #define RAPL_GFX (RAPL_GFX_POWER_LIMIT | RAPL_GFX_ENERGY_STATUS) +#define RAPL_PSYS (RAPL_PLATFORM_ENERGY_STATUS | RAPL_PLATFORM_ENERGY_LIMIT) #define RAPL_PKG_ALL (RAPL_PKG | RAPL_PKG_PERF_STATUS | RAPL_PKG_POWER_INFO) #define RAPL_DRAM_ALL (RAPL_DRAM | RAPL_DRAM_PERF_STATUS | RAPL_DRAM_POWER_INFO) @@ -708,7 +720,7 @@ static const struct platform_features skl_features = { .has_ext_cst_msrs = 1, .trl_msrs = TRL_BASE, .tcc_offset_bits = 6, - .rapl_msrs = RAPL_PKG_ALL | RAPL_CORE_ALL | RAPL_DRAM | RAPL_DRAM_PERF_STATUS | RAPL_GFX, + .rapl_msrs = RAPL_PKG_ALL | RAPL_CORE_ALL | RAPL_DRAM | RAPL_DRAM_PERF_STATUS | RAPL_GFX | RAPL_PSYS, .enable_tsc_tweak = 1, }; @@ -725,42 +737,44 @@ static const struct platform_features cnl_features = { .has_ext_cst_msrs = 1, .trl_msrs = TRL_BASE, .tcc_offset_bits = 6, - .rapl_msrs = RAPL_PKG_ALL | RAPL_CORE_ALL | RAPL_DRAM | RAPL_DRAM_PERF_STATUS | RAPL_GFX, + .rapl_msrs = RAPL_PKG_ALL | RAPL_CORE_ALL | RAPL_DRAM | RAPL_DRAM_PERF_STATUS | RAPL_GFX | RAPL_PSYS, .enable_tsc_tweak = 1, }; +/* Copied from cnl_features, with PC7/PC9 removed */ static const struct platform_features adl_features = { - .has_msr_misc_feature_control = 1, - .has_msr_misc_pwr_mgmt = 1, - .has_nhm_msrs = 1, - .has_config_tdp = 1, - .bclk_freq = BCLK_100MHZ, - .supported_cstates = CC1 | CC6 | CC7 | PC2 | PC3 | PC6 | PC8 | PC10, - .cst_limit = CST_LIMIT_HSW, - .has_irtl_msrs = 1, - .has_msr_core_c1_res = 1, - .has_ext_cst_msrs = 1, - .trl_msrs = TRL_BASE, - .tcc_offset_bits = 6, - .rapl_msrs = RAPL_PKG_ALL | RAPL_CORE_ALL | RAPL_DRAM | RAPL_DRAM_PERF_STATUS | RAPL_GFX, - .enable_tsc_tweak = 1, + .has_msr_misc_feature_control = cnl_features.has_msr_misc_feature_control, + .has_msr_misc_pwr_mgmt = cnl_features.has_msr_misc_pwr_mgmt, + .has_nhm_msrs = cnl_features.has_nhm_msrs, + .has_config_tdp = cnl_features.has_config_tdp, + .bclk_freq = cnl_features.bclk_freq, + .supported_cstates = CC1 | CC6 | CC7 | PC2 | PC3 | PC6 | PC8 | PC10, + .cst_limit = cnl_features.cst_limit, + .has_irtl_msrs = cnl_features.has_irtl_msrs, + .has_msr_core_c1_res = cnl_features.has_msr_core_c1_res, + .has_ext_cst_msrs = cnl_features.has_ext_cst_msrs, + .trl_msrs = cnl_features.trl_msrs, + .tcc_offset_bits = cnl_features.tcc_offset_bits, + .rapl_msrs = cnl_features.rapl_msrs, + .enable_tsc_tweak = cnl_features.enable_tsc_tweak, }; -static const struct platform_features arl_features = { - .has_msr_misc_feature_control = 1, - .has_msr_misc_pwr_mgmt = 1, - .has_nhm_msrs = 1, - .has_config_tdp = 1, - .bclk_freq = BCLK_100MHZ, - .supported_cstates = CC1 | CC6 | CC7 | PC2 | PC3 | PC6 | PC10, - .cst_limit = CST_LIMIT_HSW, - .has_irtl_msrs = 1, - .has_msr_core_c1_res = 1, - .has_ext_cst_msrs = 1, - .trl_msrs = TRL_BASE, - .tcc_offset_bits = 6, - .rapl_msrs = RAPL_PKG_ALL | RAPL_CORE_ALL | RAPL_DRAM | RAPL_DRAM_PERF_STATUS | RAPL_GFX, - .enable_tsc_tweak = 1, +/* Copied from adl_features, with PC3/PC8 removed */ +static const struct platform_features lnl_features = { + .has_msr_misc_feature_control = adl_features.has_msr_misc_feature_control, + .has_msr_misc_pwr_mgmt = adl_features.has_msr_misc_pwr_mgmt, + .has_nhm_msrs = adl_features.has_nhm_msrs, + .has_config_tdp = adl_features.has_config_tdp, + .bclk_freq = adl_features.bclk_freq, + .supported_cstates = CC1 | CC6 | CC7 | PC2 | PC6 | PC10, + .cst_limit = adl_features.cst_limit, + .has_irtl_msrs = adl_features.has_irtl_msrs, + .has_msr_core_c1_res = adl_features.has_msr_core_c1_res, + .has_ext_cst_msrs = adl_features.has_ext_cst_msrs, + .trl_msrs = adl_features.trl_msrs, + .tcc_offset_bits = adl_features.tcc_offset_bits, + .rapl_msrs = adl_features.rapl_msrs, + .enable_tsc_tweak = adl_features.enable_tsc_tweak, }; static const struct platform_features skx_features = { @@ -790,7 +804,7 @@ static const struct platform_features icx_features = { .has_irtl_msrs = 1, .has_cst_prewake_bit = 1, .trl_msrs = TRL_BASE | TRL_CORECOUNT, - .rapl_msrs = RAPL_PKG_ALL | RAPL_DRAM_ALL, + .rapl_msrs = RAPL_PKG_ALL | RAPL_DRAM_ALL | RAPL_PSYS, .has_fixed_rapl_unit = 1, }; @@ -806,7 +820,7 @@ static const struct platform_features spr_features = { .has_irtl_msrs = 1, .has_cst_prewake_bit = 1, .trl_msrs = TRL_BASE | TRL_CORECOUNT, - .rapl_msrs = RAPL_PKG_ALL | RAPL_DRAM_ALL, + .rapl_msrs = RAPL_PKG_ALL | RAPL_DRAM_ALL | RAPL_PSYS, }; static const struct platform_features srf_features = { @@ -822,7 +836,7 @@ static const struct platform_features srf_features = { .has_irtl_msrs = 1, .has_cst_prewake_bit = 1, .trl_msrs = TRL_BASE | TRL_CORECOUNT, - .rapl_msrs = RAPL_PKG_ALL | RAPL_DRAM_ALL, + .rapl_msrs = RAPL_PKG_ALL | RAPL_DRAM_ALL | RAPL_PSYS, }; static const struct platform_features grr_features = { @@ -838,7 +852,7 @@ static const struct platform_features grr_features = { .has_irtl_msrs = 1, .has_cst_prewake_bit = 1, .trl_msrs = TRL_BASE | TRL_CORECOUNT, - .rapl_msrs = RAPL_PKG_ALL | RAPL_DRAM_ALL, + .rapl_msrs = RAPL_PKG_ALL | RAPL_DRAM_ALL | RAPL_PSYS, }; static const struct platform_features slv_features = { @@ -997,18 +1011,19 @@ static const struct platform_data turbostat_pdata[] = { { INTEL_SAPPHIRERAPIDS_X, &spr_features }, { INTEL_EMERALDRAPIDS_X, &spr_features }, { INTEL_GRANITERAPIDS_X, &spr_features }, + { INTEL_GRANITERAPIDS_D, &spr_features }, { INTEL_LAKEFIELD, &cnl_features }, { INTEL_ALDERLAKE, &adl_features }, { INTEL_ALDERLAKE_L, &adl_features }, { INTEL_RAPTORLAKE, &adl_features }, { INTEL_RAPTORLAKE_P, &adl_features }, { INTEL_RAPTORLAKE_S, &adl_features }, - { INTEL_METEORLAKE, &cnl_features }, - { INTEL_METEORLAKE_L, &cnl_features }, - { INTEL_ARROWLAKE_H, &arl_features }, - { INTEL_ARROWLAKE_U, &arl_features }, - { INTEL_ARROWLAKE, &arl_features }, - { INTEL_LUNARLAKE_M, &arl_features }, + { INTEL_METEORLAKE, &adl_features }, + { INTEL_METEORLAKE_L, &adl_features }, + { INTEL_ARROWLAKE_H, &adl_features }, + { INTEL_ARROWLAKE_U, &adl_features }, + { INTEL_ARROWLAKE, &adl_features }, + { INTEL_LUNARLAKE_M, &lnl_features }, { INTEL_ATOM_SILVERMONT, &slv_features }, { INTEL_ATOM_SILVERMONT_D, &slvd_features }, { INTEL_ATOM_AIRMONT, &amt_features }, @@ -1100,6 +1115,7 @@ enum rapl_rci_index { RAPL_RCI_INDEX_PKG_PERF_STATUS = 4, RAPL_RCI_INDEX_DRAM_PERF_STATUS = 5, RAPL_RCI_INDEX_CORE_ENERGY = 6, + RAPL_RCI_INDEX_ENERGY_PLATFORM = 7, NUM_RAPL_COUNTERS, }; @@ -1126,6 +1142,7 @@ struct rapl_counter_info_t { struct rapl_counter_info_t *rapl_counter_info_perdomain; unsigned int rapl_counter_info_perdomain_size; +#define RAPL_COUNTER_FLAG_PLATFORM_COUNTER (1u << 0) #define RAPL_COUNTER_FLAG_USE_MSR_SUM (1u << 1) struct rapl_counter_arch_info { @@ -1247,6 +1264,19 @@ static const struct rapl_counter_arch_info rapl_counter_arch_infos[] = { .compat_scale = 1.0, .flags = 0, }, + { + .feature_mask = RAPL_PSYS, + .perf_subsys = "power", + .perf_name = "energy-psys", + .msr = MSR_PLATFORM_ENERGY_STATUS, + .msr_mask = 0x00000000FFFFFFFF, + .msr_shift = 0, + .platform_rapl_msr_scale = &rapl_energy_units, + .rci_index = RAPL_RCI_INDEX_ENERGY_PLATFORM, + .bic = BIC_SysWatt | BIC_Sys_J, + .compat_scale = 1.0, + .flags = RAPL_COUNTER_FLAG_PLATFORM_COUNTER | RAPL_COUNTER_FLAG_USE_MSR_SUM, + }, }; struct rapl_counter { @@ -1674,6 +1704,7 @@ enum { IDX_PP1_ENERGY, IDX_PKG_PERF, IDX_DRAM_PERF, + IDX_PSYS_ENERGY, IDX_COUNT, }; @@ -1718,6 +1749,9 @@ off_t idx_to_offset(int idx) case IDX_DRAM_PERF: offset = MSR_DRAM_PERF_STATUS; break; + case IDX_PSYS_ENERGY: + offset = MSR_PLATFORM_ENERGY_STATUS; + break; default: offset = -1; } @@ -1748,6 +1782,9 @@ int offset_to_idx(off_t offset) case MSR_DRAM_PERF_STATUS: idx = IDX_DRAM_PERF; break; + case MSR_PLATFORM_ENERGY_STATUS: + idx = IDX_PSYS_ENERGY; + break; default: idx = -1; } @@ -1769,6 +1806,8 @@ int idx_valid(int idx) return platform->rapl_msrs & RAPL_PKG_PERF_STATUS; case IDX_DRAM_PERF: return platform->rapl_msrs & RAPL_DRAM_PERF_STATUS; + case IDX_PSYS_ENERGY: + return platform->rapl_msrs & RAPL_PSYS; default: return 0; } @@ -1840,6 +1879,10 @@ struct system_summary { struct pkg_data packages; } average; +struct platform_counters { + struct rapl_counter energy_psys; /* MSR_PLATFORM_ENERGY_STATUS */ +} platform_counters_odd, platform_counters_even; + struct cpu_topology { int physical_package_id; int die_id; @@ -1848,6 +1891,7 @@ struct cpu_topology { int logical_node_id; /* 0-based count within the package */ int physical_core_id; int thread_id; + int type; cpu_set_t *put_ids; /* Processing Unit/Thread IDs */ } *cpus; @@ -2291,7 +2335,7 @@ void print_header(char *delim) break; case PMT_TYPE_XTAL_TIME: - outp += sprintf(outp, "%s%s", delim, ppmt->name); + outp += sprintf(outp, "%s%s", (printed++ ? delim : ""), ppmt->name); break; } @@ -2365,7 +2409,7 @@ void print_header(char *delim) break; case PMT_TYPE_XTAL_TIME: - outp += sprintf(outp, "%s%s", delim, ppmt->name); + outp += sprintf(outp, "%s%s", (printed++ ? delim : ""), ppmt->name); break; } @@ -2496,13 +2540,18 @@ void print_header(char *delim) break; case PMT_TYPE_XTAL_TIME: - outp += sprintf(outp, "%s%s", delim, ppmt->name); + outp += sprintf(outp, "%s%s", (printed++ ? delim : ""), ppmt->name); break; } ppmt = ppmt->next; } + if (DO_BIC(BIC_SysWatt)) + outp += sprintf(outp, "%sSysWatt", (printed++ ? delim : "")); + if (DO_BIC(BIC_Sys_J)) + outp += sprintf(outp, "%sSys_J", (printed++ ? delim : "")); + outp += sprintf(outp, "\n"); } @@ -2510,6 +2559,7 @@ int dump_counters(struct thread_data *t, struct core_data *c, struct pkg_data *p { int i; struct msr_counter *mp; + struct platform_counters *pplat_cnt = p == package_odd ? &platform_counters_odd : &platform_counters_even; outp += sprintf(outp, "t %p, c %p, p %p\n", t, c, p); @@ -2581,6 +2631,7 @@ int dump_counters(struct thread_data *t, struct core_data *c, struct pkg_data *p outp += sprintf(outp, "Joules COR: %0llX\n", p->energy_cores.raw_value); outp += sprintf(outp, "Joules GFX: %0llX\n", p->energy_gfx.raw_value); outp += sprintf(outp, "Joules RAM: %0llX\n", p->energy_dram.raw_value); + outp += sprintf(outp, "Joules PSYS: %0llX\n", pplat_cnt->energy_psys.raw_value); outp += sprintf(outp, "Throttle PKG: %0llX\n", p->rapl_pkg_perf_status.raw_value); outp += sprintf(outp, "Throttle RAM: %0llX\n", p->rapl_dram_perf_status.raw_value); outp += sprintf(outp, "PTM: %dC\n", p->pkg_temp_c); @@ -2619,6 +2670,9 @@ double rapl_counter_get_value(const struct rapl_counter *c, enum rapl_unit desir */ int format_counters(struct thread_data *t, struct core_data *c, struct pkg_data *p) { + static int count; + + struct platform_counters *pplat_cnt = NULL; double interval_float, tsc; char *fmt8; int i; @@ -2628,6 +2682,11 @@ int format_counters(struct thread_data *t, struct core_data *c, struct pkg_data char *delim = "\t"; int printed = 0; + if (t == &average.threads) { + pplat_cnt = count & 1 ? &platform_counters_odd : &platform_counters_even; + ++count; + } + /* if showing only 1st thread in core and this isn't one, bail out */ if (show_core_only && !is_cpu_first_thread_in_core(t, c, p)) return 0; @@ -2788,6 +2847,8 @@ int format_counters(struct thread_data *t, struct core_data *c, struct pkg_data } for (i = 0, ppmt = sys.pmt_tp; ppmt; i++, ppmt = ppmt->next) { + const unsigned long value_raw = t->pmt_counter[i]; + const double value_converted = 100.0 * value_raw / crystal_hz / interval_float; switch (ppmt->type) { case PMT_TYPE_RAW: if (pmt_counter_get_width(ppmt) <= 32) @@ -2799,9 +2860,6 @@ int format_counters(struct thread_data *t, struct core_data *c, struct pkg_data break; case PMT_TYPE_XTAL_TIME: - const unsigned long value_raw = t->pmt_counter[i]; - const double value_converted = 100.0 * value_raw / crystal_hz / interval_float; - outp += sprintf(outp, "%s%.2f", (printed++ ? delim : ""), value_converted); break; } @@ -2869,6 +2927,8 @@ int format_counters(struct thread_data *t, struct core_data *c, struct pkg_data } for (i = 0, ppmt = sys.pmt_cp; ppmt; i++, ppmt = ppmt->next) { + const unsigned long value_raw = c->pmt_counter[i]; + const double value_converted = 100.0 * value_raw / crystal_hz / interval_float; switch (ppmt->type) { case PMT_TYPE_RAW: if (pmt_counter_get_width(ppmt) <= 32) @@ -2880,9 +2940,6 @@ int format_counters(struct thread_data *t, struct core_data *c, struct pkg_data break; case PMT_TYPE_XTAL_TIME: - const unsigned long value_raw = c->pmt_counter[i]; - const double value_converted = 100.0 * value_raw / crystal_hz / interval_float; - outp += sprintf(outp, "%s%.2f", (printed++ ? delim : ""), value_converted); break; } @@ -3068,6 +3125,8 @@ int format_counters(struct thread_data *t, struct core_data *c, struct pkg_data } for (i = 0, ppmt = sys.pmt_pp; ppmt; i++, ppmt = ppmt->next) { + const unsigned long value_raw = p->pmt_counter[i]; + const double value_converted = 100.0 * value_raw / crystal_hz / interval_float; switch (ppmt->type) { case PMT_TYPE_RAW: if (pmt_counter_get_width(ppmt) <= 32) @@ -3079,14 +3138,18 @@ int format_counters(struct thread_data *t, struct core_data *c, struct pkg_data break; case PMT_TYPE_XTAL_TIME: - const unsigned long value_raw = p->pmt_counter[i]; - const double value_converted = 100.0 * value_raw / crystal_hz / interval_float; - outp += sprintf(outp, "%s%.2f", (printed++ ? delim : ""), value_converted); break; } } + if (DO_BIC(BIC_SysWatt) && (t == &average.threads)) + outp += sprintf(outp, fmt8, (printed++ ? delim : ""), + rapl_counter_get_value(&pplat_cnt->energy_psys, RAPL_UNIT_WATTS, interval_float)); + if (DO_BIC(BIC_Sys_J) && (t == &average.threads)) + outp += sprintf(outp, fmt8, (printed++ ? delim : ""), + rapl_counter_get_value(&pplat_cnt->energy_psys, RAPL_UNIT_JOULES, interval_float)); + done: if (*(outp - 1) != '\n') outp += sprintf(outp, "\n"); @@ -3394,6 +3457,11 @@ int delta_cpu(struct thread_data *t, struct core_data *c, return retval; } +void delta_platform(struct platform_counters *new, struct platform_counters *old) +{ + old->energy_psys.raw_value = new->energy_psys.raw_value - old->energy_psys.raw_value; +} + void rapl_counter_clear(struct rapl_counter *c) { c->raw_value = 0; @@ -4123,6 +4191,9 @@ static size_t cstate_counter_info_count_perf(const struct cstate_counter_info_t void write_rapl_counter(struct rapl_counter *rc, struct rapl_counter_info_t *rci, unsigned int idx) { + if (rci->source[idx] == COUNTER_SOURCE_NONE) + return; + rc->raw_value = rci->data[idx]; rc->unit = rci->unit[idx]; rc->scale = rci->scale[idx]; @@ -4130,6 +4201,7 @@ void write_rapl_counter(struct rapl_counter *rc, struct rapl_counter_info_t *rci int get_rapl_counters(int cpu, unsigned int domain, struct core_data *c, struct pkg_data *p) { + struct platform_counters *pplat_cnt = p == package_odd ? &platform_counters_odd : &platform_counters_even; unsigned long long perf_data[NUM_RAPL_COUNTERS + 1]; struct rapl_counter_info_t *rci; @@ -4157,6 +4229,7 @@ int get_rapl_counters(int cpu, unsigned int domain, struct core_data *c, struct for (unsigned int i = 0, pi = 1; i < NUM_RAPL_COUNTERS; ++i) { switch (rci->source[i]) { case COUNTER_SOURCE_NONE: + rci->data[i] = 0; break; case COUNTER_SOURCE_PERF: @@ -4195,7 +4268,7 @@ int get_rapl_counters(int cpu, unsigned int domain, struct core_data *c, struct } } - BUILD_BUG_ON(NUM_RAPL_COUNTERS != 7); + BUILD_BUG_ON(NUM_RAPL_COUNTERS != 8); write_rapl_counter(&p->energy_pkg, rci, RAPL_RCI_INDEX_ENERGY_PKG); write_rapl_counter(&p->energy_cores, rci, RAPL_RCI_INDEX_ENERGY_CORES); write_rapl_counter(&p->energy_dram, rci, RAPL_RCI_INDEX_DRAM); @@ -4203,6 +4276,7 @@ int get_rapl_counters(int cpu, unsigned int domain, struct core_data *c, struct write_rapl_counter(&p->rapl_pkg_perf_status, rci, RAPL_RCI_INDEX_PKG_PERF_STATUS); write_rapl_counter(&p->rapl_dram_perf_status, rci, RAPL_RCI_INDEX_DRAM_PERF_STATUS); write_rapl_counter(&c->core_energy, rci, RAPL_RCI_INDEX_CORE_ENERGY); + write_rapl_counter(&pplat_cnt->energy_psys, rci, RAPL_RCI_INDEX_ENERGY_PLATFORM); return 0; } @@ -5385,6 +5459,9 @@ static int parse_cpu_str(char *cpu_str, cpu_set_t *cpu_set, int cpu_set_size) if (*next == '-') /* no negative cpu numbers */ return 1; + if (*next == '\0' || *next == '\n') + break; + start = strtoul(next, &next, 10); if (start >= CPU_SUBSET_MAXCPUS) @@ -5656,6 +5733,32 @@ int init_thread_id(int cpu) return 0; } +int set_my_cpu_type(void) +{ + unsigned int eax, ebx, ecx, edx; + unsigned int max_level; + + __cpuid(0, max_level, ebx, ecx, edx); + + if (max_level < CPUID_LEAF_MODEL_ID) + return 0; + + __cpuid(CPUID_LEAF_MODEL_ID, eax, ebx, ecx, edx); + + return (eax >> CPUID_LEAF_MODEL_ID_CORE_TYPE_SHIFT); +} + +int set_cpu_hybrid_type(int cpu) +{ + if (cpu_migrate(cpu)) + return -1; + + int type = set_my_cpu_type(); + + cpus[cpu].type = type; + return 0; +} + /* * snapshot_proc_interrupts() * @@ -5728,28 +5831,21 @@ int snapshot_proc_interrupts(void) */ int snapshot_graphics(int idx) { - FILE *fp; int retval; + rewind(gfx_info[idx].fp); + switch (idx) { case GFX_rc6: case SAM_mc6: - fp = fopen_or_die(gfx_info[idx].path, "r"); - retval = fscanf(fp, "%lld", &gfx_info[idx].val_ull); + retval = fscanf(gfx_info[idx].fp, "%lld", &gfx_info[idx].val_ull); if (retval != 1) err(1, "rc6"); - fclose(fp); return 0; case GFX_MHz: case GFX_ACTMHz: case SAM_MHz: case SAM_ACTMHz: - if (gfx_info[idx].fp == NULL) { - gfx_info[idx].fp = fopen_or_die(gfx_info[idx].path, "r"); - } else { - rewind(gfx_info[idx].fp); - fflush(gfx_info[idx].fp); - } retval = fscanf(gfx_info[idx].fp, "%d", &gfx_info[idx].val); if (retval != 1) err(1, "MHz"); @@ -6116,6 +6212,7 @@ restart: re_initialize(); goto restart; } + delta_platform(&platform_counters_odd, &platform_counters_even); compute_average(EVEN_COUNTERS); format_all_counters(EVEN_COUNTERS); flush_output_stdout(); @@ -6139,6 +6236,7 @@ restart: re_initialize(); goto restart; } + delta_platform(&platform_counters_even, &platform_counters_odd); compute_average(ODD_COUNTERS); format_all_counters(ODD_COUNTERS); flush_output_stdout(); @@ -6442,14 +6540,25 @@ static void probe_intel_uncore_frequency(void) probe_intel_uncore_frequency_legacy(); } +static void set_graphics_fp(char *path, int idx) +{ + if (!access(path, R_OK)) + gfx_info[idx].fp = fopen_or_die(path, "r"); +} + +/* Enlarge this if there are /sys/class/drm/card2 ... */ +#define GFX_MAX_CARDS 2 + static void probe_graphics(void) { + char path[PATH_MAX]; + int i; + /* Xe graphics sysfs knobs */ if (!access("/sys/class/drm/card0/device/tile0/gt0/gtidle/idle_residency_ms", R_OK)) { FILE *fp; char buf[8]; bool gt0_is_gt; - int idx; fp = fopen("/sys/class/drm/card0/device/tile0/gt0/gtidle/name", "r"); if (!fp) @@ -6468,81 +6577,76 @@ static void probe_graphics(void) else goto next; - idx = gt0_is_gt ? GFX_rc6 : SAM_mc6; - gfx_info[idx].path = "/sys/class/drm/card0/device/tile0/gt0/gtidle/idle_residency_ms"; + set_graphics_fp("/sys/class/drm/card0/device/tile0/gt0/gtidle/idle_residency_ms", gt0_is_gt ? GFX_rc6 : SAM_mc6); - idx = gt0_is_gt ? GFX_MHz : SAM_MHz; - if (!access("/sys/class/drm/card0/device/tile0/gt0/freq0/cur_freq", R_OK)) - gfx_info[idx].path = "/sys/class/drm/card0/device/tile0/gt0/freq0/cur_freq"; + set_graphics_fp("/sys/class/drm/card0/device/tile0/gt0/freq0/cur_freq", gt0_is_gt ? GFX_MHz : SAM_MHz); - idx = gt0_is_gt ? GFX_ACTMHz : SAM_ACTMHz; - if (!access("/sys/class/drm/card0/device/tile0/gt0/freq0/act_freq", R_OK)) - gfx_info[idx].path = "/sys/class/drm/card0/device/tile0/gt0/freq0/act_freq"; + set_graphics_fp("/sys/class/drm/card0/device/tile0/gt0/freq0/act_freq", gt0_is_gt ? GFX_ACTMHz : SAM_ACTMHz); - idx = gt0_is_gt ? SAM_mc6 : GFX_rc6; - if (!access("/sys/class/drm/card0/device/tile0/gt1/gtidle/idle_residency_ms", R_OK)) - gfx_info[idx].path = "/sys/class/drm/card0/device/tile0/gt1/gtidle/idle_residency_ms"; + set_graphics_fp("/sys/class/drm/card0/device/tile0/gt1/gtidle/idle_residency_ms", gt0_is_gt ? SAM_mc6 : GFX_rc6); - idx = gt0_is_gt ? SAM_MHz : GFX_MHz; - if (!access("/sys/class/drm/card0/device/tile0/gt1/freq0/cur_freq", R_OK)) - gfx_info[idx].path = "/sys/class/drm/card0/device/tile0/gt1/freq0/cur_freq"; + set_graphics_fp("/sys/class/drm/card0/device/tile0/gt1/freq0/cur_freq", gt0_is_gt ? SAM_MHz : GFX_MHz); - idx = gt0_is_gt ? SAM_ACTMHz : GFX_ACTMHz; - if (!access("/sys/class/drm/card0/device/tile0/gt1/freq0/act_freq", R_OK)) - gfx_info[idx].path = "/sys/class/drm/card0/device/tile0/gt1/freq0/act_freq"; + set_graphics_fp("/sys/class/drm/card0/device/tile0/gt1/freq0/act_freq", gt0_is_gt ? SAM_ACTMHz : GFX_ACTMHz); goto end; } next: /* New i915 graphics sysfs knobs */ - if (!access("/sys/class/drm/card0/gt/gt0/rc6_residency_ms", R_OK)) { - gfx_info[GFX_rc6].path = "/sys/class/drm/card0/gt/gt0/rc6_residency_ms"; + for (i = 0; i < GFX_MAX_CARDS; i++) { + snprintf(path, PATH_MAX, "/sys/class/drm/card%d/gt/gt0/rc6_residency_ms", i); + if (!access(path, R_OK)) + break; + } - if (!access("/sys/class/drm/card0/gt/gt0/rps_cur_freq_mhz", R_OK)) - gfx_info[GFX_MHz].path = "/sys/class/drm/card0/gt/gt0/rps_cur_freq_mhz"; + if (i == GFX_MAX_CARDS) + goto legacy_i915; - if (!access("/sys/class/drm/card0/gt/gt0/rps_act_freq_mhz", R_OK)) - gfx_info[GFX_ACTMHz].path = "/sys/class/drm/card0/gt/gt0/rps_act_freq_mhz"; + snprintf(path, PATH_MAX, "/sys/class/drm/card%d/gt/gt0/rc6_residency_ms", i); + set_graphics_fp(path, GFX_rc6); - if (!access("/sys/class/drm/card0/gt/gt1/rc6_residency_ms", R_OK)) - gfx_info[SAM_mc6].path = "/sys/class/drm/card0/gt/gt1/rc6_residency_ms"; + snprintf(path, PATH_MAX, "/sys/class/drm/card%d/gt/gt0/rps_cur_freq_mhz", i); + set_graphics_fp(path, GFX_MHz); - if (!access("/sys/class/drm/card0/gt/gt1/rps_cur_freq_mhz", R_OK)) - gfx_info[SAM_MHz].path = "/sys/class/drm/card0/gt/gt1/rps_cur_freq_mhz"; + snprintf(path, PATH_MAX, "/sys/class/drm/card%d/gt/gt0/rps_act_freq_mhz", i); + set_graphics_fp(path, GFX_ACTMHz); - if (!access("/sys/class/drm/card0/gt/gt1/rps_act_freq_mhz", R_OK)) - gfx_info[SAM_ACTMHz].path = "/sys/class/drm/card0/gt/gt1/rps_act_freq_mhz"; + snprintf(path, PATH_MAX, "/sys/class/drm/card%d/gt/gt1/rc6_residency_ms", i); + set_graphics_fp(path, SAM_mc6); - goto end; - } + snprintf(path, PATH_MAX, "/sys/class/drm/card%d/gt/gt1/rps_cur_freq_mhz", i); + set_graphics_fp(path, SAM_MHz); + + snprintf(path, PATH_MAX, "/sys/class/drm/card%d/gt/gt1/rps_act_freq_mhz", i); + set_graphics_fp(path, SAM_ACTMHz); + goto end; + +legacy_i915: /* Fall back to traditional i915 graphics sysfs knobs */ - if (!access("/sys/class/drm/card0/power/rc6_residency_ms", R_OK)) - gfx_info[GFX_rc6].path = "/sys/class/drm/card0/power/rc6_residency_ms"; + set_graphics_fp("/sys/class/drm/card0/power/rc6_residency_ms", GFX_rc6); - if (!access("/sys/class/drm/card0/gt_cur_freq_mhz", R_OK)) - gfx_info[GFX_MHz].path = "/sys/class/drm/card0/gt_cur_freq_mhz"; - else if (!access("/sys/class/graphics/fb0/device/drm/card0/gt_cur_freq_mhz", R_OK)) - gfx_info[GFX_MHz].path = "/sys/class/graphics/fb0/device/drm/card0/gt_cur_freq_mhz"; + set_graphics_fp("/sys/class/drm/card0/gt_cur_freq_mhz", GFX_MHz); + if (!gfx_info[GFX_MHz].fp) + set_graphics_fp("/sys/class/graphics/fb0/device/drm/card0/gt_cur_freq_mhz", GFX_MHz); - if (!access("/sys/class/drm/card0/gt_act_freq_mhz", R_OK)) - gfx_info[GFX_ACTMHz].path = "/sys/class/drm/card0/gt_act_freq_mhz"; - else if (!access("/sys/class/graphics/fb0/device/drm/card0/gt_act_freq_mhz", R_OK)) - gfx_info[GFX_ACTMHz].path = "/sys/class/graphics/fb0/device/drm/card0/gt_act_freq_mhz"; + set_graphics_fp("/sys/class/drm/card0/gt_act_freq_mhz", GFX_ACTMHz); + if (!gfx_info[GFX_ACTMHz].fp) + set_graphics_fp("/sys/class/graphics/fb0/device/drm/card0/gt_act_freq_mhz", GFX_ACTMHz); end: - if (gfx_info[GFX_rc6].path) + if (gfx_info[GFX_rc6].fp) BIC_PRESENT(BIC_GFX_rc6); - if (gfx_info[GFX_MHz].path) + if (gfx_info[GFX_MHz].fp) BIC_PRESENT(BIC_GFXMHz); - if (gfx_info[GFX_ACTMHz].path) + if (gfx_info[GFX_ACTMHz].fp) BIC_PRESENT(BIC_GFXACTMHz); - if (gfx_info[SAM_mc6].path) + if (gfx_info[SAM_mc6].fp) BIC_PRESENT(BIC_SAM_mc6); - if (gfx_info[SAM_MHz].path) + if (gfx_info[SAM_MHz].fp) BIC_PRESENT(BIC_SAMMHz); - if (gfx_info[SAM_ACTMHz].path) + if (gfx_info[SAM_ACTMHz].fp) BIC_PRESENT(BIC_SAMACTMHz); } @@ -6911,8 +7015,8 @@ void rapl_probe_intel(void) unsigned long long msr; unsigned int time_unit; double tdp; - const unsigned long long bic_watt_bits = BIC_PkgWatt | BIC_CorWatt | BIC_RAMWatt | BIC_GFXWatt; - const unsigned long long bic_joules_bits = BIC_Pkg_J | BIC_Cor_J | BIC_RAM_J | BIC_GFX_J; + const unsigned long long bic_watt_bits = BIC_SysWatt | BIC_PkgWatt | BIC_CorWatt | BIC_RAMWatt | BIC_GFXWatt; + const unsigned long long bic_joules_bits = BIC_Sys_J | BIC_Pkg_J | BIC_Cor_J | BIC_RAM_J | BIC_GFX_J; if (rapl_joules) bic_enabled &= ~bic_watt_bits; @@ -7572,6 +7676,9 @@ void rapl_perf_init(void) domain_visited[next_domain] = 1; + if ((cai->flags & RAPL_COUNTER_FLAG_PLATFORM_COUNTER) && (cpu != base_cpu)) + continue; + struct rapl_counter_info_t *rci = &rapl_counter_info_perdomain[next_domain]; /* Check if the counter is enabled and accessible */ @@ -8196,7 +8303,7 @@ void topology_probe(bool startup) set_max_cpu_num(); topo.num_cpus = 0; for_all_proc_cpus(count_cpus); - if (!summary_only && topo.num_cpus > 1) + if (!summary_only) BIC_PRESENT(BIC_CPU); if (debug > 1) @@ -8284,6 +8391,8 @@ void topology_probe(bool startup) for_all_proc_cpus(init_thread_id); + for_all_proc_cpus(set_cpu_hybrid_type); + /* * For online cpus * find max_core_id, max_package_id @@ -8332,7 +8441,7 @@ void topology_probe(bool startup) topo.cores_per_node = max_core_id + 1; if (debug > 1) fprintf(outf, "max_core_id %d, sizing for %d cores per package\n", max_core_id, topo.cores_per_node); - if (!summary_only && topo.cores_per_node > 1) + if (!summary_only) BIC_PRESENT(BIC_Core); topo.num_die = topo.max_die_id + 1; @@ -8548,6 +8657,35 @@ void check_perf_access(void) bic_enabled &= ~BIC_IPC; } +bool perf_has_hybrid_devices(void) +{ + /* + * 0: unknown + * 1: has separate perf device for p and e core + * -1: doesn't have separate perf device for p and e core + */ + static int cached; + + if (cached > 0) + return true; + + if (cached < 0) + return false; + + if (access("/sys/bus/event_source/devices/cpu_core", F_OK)) { + cached = -1; + return false; + } + + if (access("/sys/bus/event_source/devices/cpu_atom", F_OK)) { + cached = -1; + return false; + } + + cached = 1; + return true; +} + int added_perf_counters_init_(struct perf_counter_info *pinfo) { size_t num_domains = 0; @@ -8604,29 +8742,56 @@ int added_perf_counters_init_(struct perf_counter_info *pinfo) if (domain_visited[next_domain]) continue; - perf_type = read_perf_type(pinfo->device); + /* + * Intel hybrid platforms expose different perf devices for P and E cores. + * Instead of one, "/sys/bus/event_source/devices/cpu" device, there are + * "/sys/bus/event_source/devices/{cpu_core,cpu_atom}". + * + * This makes it more complicated to the user, because most of the counters + * are available on both and have to be handled manually, otherwise. + * + * Code below, allow user to use the old "cpu" name, which is translated accordingly. + */ + const char *perf_device = pinfo->device; + + if (strcmp(perf_device, "cpu") == 0 && perf_has_hybrid_devices()) { + switch (cpus[cpu].type) { + case INTEL_PCORE_TYPE: + perf_device = "cpu_core"; + break; + + case INTEL_ECORE_TYPE: + perf_device = "cpu_atom"; + break; + + default: /* Don't change, we will probably fail and report a problem soon. */ + break; + } + } + + perf_type = read_perf_type(perf_device); if (perf_type == (unsigned int)-1) { warnx("%s: perf/%s/%s: failed to read %s", - __func__, pinfo->device, pinfo->event, "type"); + __func__, perf_device, pinfo->event, "type"); continue; } - perf_config = read_perf_config(pinfo->device, pinfo->event); + perf_config = read_perf_config(perf_device, pinfo->event); if (perf_config == (unsigned int)-1) { warnx("%s: perf/%s/%s: failed to read %s", - __func__, pinfo->device, pinfo->event, "config"); + __func__, perf_device, pinfo->event, "config"); continue; } /* Scale is not required, some counters just don't have it. */ - perf_scale = read_perf_scale(pinfo->device, pinfo->event); + perf_scale = read_perf_scale(perf_device, pinfo->event); if (perf_scale == 0.0) perf_scale = 1.0; fd_perf = open_perf_counter(cpu, perf_type, perf_config, -1, 0); if (fd_perf == -1) { warnx("%s: perf/%s/%s: failed to open counter on cpu%d", - __func__, pinfo->device, pinfo->event, cpu); + __func__, perf_device, pinfo->event, cpu); continue; } @@ -8636,7 +8801,7 @@ int added_perf_counters_init_(struct perf_counter_info *pinfo) if (debug) fprintf(stderr, "Add perf/%s/%s cpu%d: %d\n", - pinfo->device, pinfo->event, cpu, pinfo->fd_perf_per_domain[next_domain]); + perf_device, pinfo->event, cpu, pinfo->fd_perf_per_domain[next_domain]); } pinfo = pinfo->next; @@ -9071,7 +9236,7 @@ int get_and_dump_counters(void) void print_version() { - fprintf(outf, "turbostat version 2024.07.26 - Len Brown <lenb@kernel.org>\n"); + fprintf(outf, "turbostat version 2024.11.30 - Len Brown <lenb@kernel.org>\n"); } #define COMMAND_LINE_SIZE 2048 @@ -9781,7 +9946,7 @@ void cmdline(int argc, char **argv) * Parse some options early, because they may make other options invalid, * like adding the MSR counter with --add and at the same time using --no-msr. */ - while ((opt = getopt_long_only(argc, argv, "MPn:", long_options, &option_index)) != -1) { + while ((opt = getopt_long_only(argc, argv, "+MPn:", long_options, &option_index)) != -1) { switch (opt) { case 'M': no_msr = 1; @@ -9805,6 +9970,12 @@ void cmdline(int argc, char **argv) break; case 'D': dump_only++; + /* + * Force the no_perf early to prevent using it as a source. + * User asks for raw values, but perf returns them relative + * to the opening of the file descriptor. + */ + no_perf = 1; break; case 'e': /* --enable specified counter */ diff --git a/tools/testing/cxl/cxl_core_exports.c b/tools/testing/cxl/cxl_core_exports.c index 077e6883921d..f088792a8925 100644 --- a/tools/testing/cxl/cxl_core_exports.c +++ b/tools/testing/cxl/cxl_core_exports.c @@ -4,4 +4,4 @@ #include "cxl.h" /* Exporting of cxl_core symbols that are only used by cxl_test */ -EXPORT_SYMBOL_NS_GPL(cxl_num_decoders_committed, CXL); +EXPORT_SYMBOL_NS_GPL(cxl_num_decoders_committed, "CXL"); diff --git a/tools/testing/cxl/test/cxl.c b/tools/testing/cxl/test/cxl.c index 050725afa45d..d0337c11f9ee 100644 --- a/tools/testing/cxl/test/cxl.c +++ b/tools/testing/cxl/test/cxl.c @@ -1531,5 +1531,5 @@ MODULE_PARM_DESC(interleave_arithmetic, "Modulo:0, XOR:1"); module_init(cxl_test_init); module_exit(cxl_test_exit); MODULE_LICENSE("GPL v2"); -MODULE_IMPORT_NS(ACPI); -MODULE_IMPORT_NS(CXL); +MODULE_IMPORT_NS("ACPI"); +MODULE_IMPORT_NS("CXL"); diff --git a/tools/testing/cxl/test/mem.c b/tools/testing/cxl/test/mem.c index 71916e0e1546..347c1e7b37bd 100644 --- a/tools/testing/cxl/test/mem.c +++ b/tools/testing/cxl/test/mem.c @@ -1679,4 +1679,4 @@ static struct platform_driver cxl_mock_mem_driver = { module_platform_driver(cxl_mock_mem_driver); MODULE_LICENSE("GPL v2"); -MODULE_IMPORT_NS(CXL); +MODULE_IMPORT_NS("CXL"); diff --git a/tools/testing/cxl/test/mock.c b/tools/testing/cxl/test/mock.c index f4ce96cc11d4..450c7566c33f 100644 --- a/tools/testing/cxl/test/mock.c +++ b/tools/testing/cxl/test/mock.c @@ -76,7 +76,7 @@ int __wrap_acpi_table_parse_cedt(enum acpi_cedt_type id, return rc; } -EXPORT_SYMBOL_NS_GPL(__wrap_acpi_table_parse_cedt, ACPI); +EXPORT_SYMBOL_NS_GPL(__wrap_acpi_table_parse_cedt, "ACPI"); acpi_status __wrap_acpi_evaluate_integer(acpi_handle handle, acpi_string pathname, @@ -147,7 +147,7 @@ struct cxl_hdm *__wrap_devm_cxl_setup_hdm(struct cxl_port *port, return cxlhdm; } -EXPORT_SYMBOL_NS_GPL(__wrap_devm_cxl_setup_hdm, CXL); +EXPORT_SYMBOL_NS_GPL(__wrap_devm_cxl_setup_hdm, "CXL"); int __wrap_devm_cxl_add_passthrough_decoder(struct cxl_port *port) { @@ -162,7 +162,7 @@ int __wrap_devm_cxl_add_passthrough_decoder(struct cxl_port *port) return rc; } -EXPORT_SYMBOL_NS_GPL(__wrap_devm_cxl_add_passthrough_decoder, CXL); +EXPORT_SYMBOL_NS_GPL(__wrap_devm_cxl_add_passthrough_decoder, "CXL"); int __wrap_devm_cxl_enumerate_decoders(struct cxl_hdm *cxlhdm, struct cxl_endpoint_dvsec_info *info) @@ -179,7 +179,7 @@ int __wrap_devm_cxl_enumerate_decoders(struct cxl_hdm *cxlhdm, return rc; } -EXPORT_SYMBOL_NS_GPL(__wrap_devm_cxl_enumerate_decoders, CXL); +EXPORT_SYMBOL_NS_GPL(__wrap_devm_cxl_enumerate_decoders, "CXL"); int __wrap_devm_cxl_port_enumerate_dports(struct cxl_port *port) { @@ -194,7 +194,7 @@ int __wrap_devm_cxl_port_enumerate_dports(struct cxl_port *port) return rc; } -EXPORT_SYMBOL_NS_GPL(__wrap_devm_cxl_port_enumerate_dports, CXL); +EXPORT_SYMBOL_NS_GPL(__wrap_devm_cxl_port_enumerate_dports, "CXL"); int __wrap_cxl_await_media_ready(struct cxl_dev_state *cxlds) { @@ -209,7 +209,7 @@ int __wrap_cxl_await_media_ready(struct cxl_dev_state *cxlds) return rc; } -EXPORT_SYMBOL_NS_GPL(__wrap_cxl_await_media_ready, CXL); +EXPORT_SYMBOL_NS_GPL(__wrap_cxl_await_media_ready, "CXL"); int __wrap_cxl_hdm_decode_init(struct cxl_dev_state *cxlds, struct cxl_hdm *cxlhdm, @@ -226,7 +226,7 @@ int __wrap_cxl_hdm_decode_init(struct cxl_dev_state *cxlds, return rc; } -EXPORT_SYMBOL_NS_GPL(__wrap_cxl_hdm_decode_init, CXL); +EXPORT_SYMBOL_NS_GPL(__wrap_cxl_hdm_decode_init, "CXL"); int __wrap_cxl_dvsec_rr_decode(struct device *dev, struct cxl_port *port, struct cxl_endpoint_dvsec_info *info) @@ -242,7 +242,7 @@ int __wrap_cxl_dvsec_rr_decode(struct device *dev, struct cxl_port *port, return rc; } -EXPORT_SYMBOL_NS_GPL(__wrap_cxl_dvsec_rr_decode, CXL); +EXPORT_SYMBOL_NS_GPL(__wrap_cxl_dvsec_rr_decode, "CXL"); struct cxl_dport *__wrap_devm_cxl_add_rch_dport(struct cxl_port *port, struct device *dport_dev, @@ -266,7 +266,7 @@ struct cxl_dport *__wrap_devm_cxl_add_rch_dport(struct cxl_port *port, return dport; } -EXPORT_SYMBOL_NS_GPL(__wrap_devm_cxl_add_rch_dport, CXL); +EXPORT_SYMBOL_NS_GPL(__wrap_devm_cxl_add_rch_dport, "CXL"); resource_size_t __wrap_cxl_rcd_component_reg_phys(struct device *dev, struct cxl_dport *dport) @@ -283,7 +283,7 @@ resource_size_t __wrap_cxl_rcd_component_reg_phys(struct device *dev, return component_reg_phys; } -EXPORT_SYMBOL_NS_GPL(__wrap_cxl_rcd_component_reg_phys, CXL); +EXPORT_SYMBOL_NS_GPL(__wrap_cxl_rcd_component_reg_phys, "CXL"); void __wrap_cxl_endpoint_parse_cdat(struct cxl_port *port) { @@ -297,7 +297,7 @@ void __wrap_cxl_endpoint_parse_cdat(struct cxl_port *port) cxl_endpoint_parse_cdat(port); put_cxl_mock_ops(index); } -EXPORT_SYMBOL_NS_GPL(__wrap_cxl_endpoint_parse_cdat, CXL); +EXPORT_SYMBOL_NS_GPL(__wrap_cxl_endpoint_parse_cdat, "CXL"); void __wrap_cxl_dport_init_ras_reporting(struct cxl_dport *dport, struct device *host) { @@ -309,8 +309,8 @@ void __wrap_cxl_dport_init_ras_reporting(struct cxl_dport *dport, struct device put_cxl_mock_ops(index); } -EXPORT_SYMBOL_NS_GPL(__wrap_cxl_dport_init_ras_reporting, CXL); +EXPORT_SYMBOL_NS_GPL(__wrap_cxl_dport_init_ras_reporting, "CXL"); MODULE_LICENSE("GPL v2"); -MODULE_IMPORT_NS(ACPI); -MODULE_IMPORT_NS(CXL); +MODULE_IMPORT_NS("ACPI"); +MODULE_IMPORT_NS("CXL"); diff --git a/tools/testing/nvdimm/test/ndtest.c b/tools/testing/nvdimm/test/ndtest.c index 892e990c034a..68a064ce598c 100644 --- a/tools/testing/nvdimm/test/ndtest.c +++ b/tools/testing/nvdimm/test/ndtest.c @@ -883,7 +883,7 @@ static const struct platform_device_id ndtest_id[] = { static struct platform_driver ndtest_driver = { .probe = ndtest_probe, - .remove_new = ndtest_remove, + .remove = ndtest_remove, .driver = { .name = KBUILD_MODNAME, }, diff --git a/tools/testing/selftests/alsa/.gitignore b/tools/testing/selftests/alsa/.gitignore index 12dc3fcd3456..3dd8e1176b89 100644 --- a/tools/testing/selftests/alsa/.gitignore +++ b/tools/testing/selftests/alsa/.gitignore @@ -1,3 +1,5 @@ +global-timer mixer-test pcm-test test-pcmtest-driver +utimer-test diff --git a/tools/testing/selftests/drivers/net/hw/rss_ctx.py b/tools/testing/selftests/drivers/net/hw/rss_ctx.py index 0b49ce7ae678..ca8a7edff3dd 100755 --- a/tools/testing/selftests/drivers/net/hw/rss_ctx.py +++ b/tools/testing/selftests/drivers/net/hw/rss_ctx.py @@ -3,7 +3,8 @@ import datetime import random -from lib.py import ksft_run, ksft_pr, ksft_exit, ksft_eq, ksft_ne, ksft_ge, ksft_lt +import re +from lib.py import ksft_run, ksft_pr, ksft_exit, ksft_eq, ksft_ne, ksft_ge, ksft_lt, ksft_true from lib.py import NetDrvEpEnv from lib.py import EthtoolFamily, NetdevFamily from lib.py import KsftSkipEx, KsftFailEx @@ -96,6 +97,13 @@ def _send_traffic_check(cfg, port, name, params): f"traffic on inactive queues ({name}): " + str(cnts)) +def _ntuple_rule_check(cfg, rule_id, ctx_id): + """Check that ntuple rule references RSS context ID""" + text = ethtool(f"-n {cfg.ifname} rule {rule_id}").stdout + pattern = f"RSS Context (ID: )?{ctx_id}" + ksft_true(re.search(pattern, text), "RSS context not referenced in ntuple rule") + + def test_rss_key_indir(cfg): """Test basics like updating the main RSS key and indirection table.""" @@ -459,6 +467,8 @@ def test_rss_context(cfg, ctx_cnt=1, create_with_cfg=None): ntuple = ethtool_create(cfg, "-N", flow) defer(ethtool, f"-N {cfg.ifname} delete {ntuple}") + _ntuple_rule_check(cfg, ntuple, ctx_id) + for i in range(ctx_cnt): _send_traffic_check(cfg, ports[i], f"context {i}", { 'target': (2+i*2, 3+i*2), diff --git a/tools/testing/selftests/filesystems/binderfs/binderfs_test.c b/tools/testing/selftests/filesystems/binderfs/binderfs_test.c index 319567f0fae1..81db85a5cc16 100644 --- a/tools/testing/selftests/filesystems/binderfs/binderfs_test.c +++ b/tools/testing/selftests/filesystems/binderfs/binderfs_test.c @@ -57,7 +57,6 @@ static int __do_binderfs_test(struct __test_metadata *_metadata) { int fd, ret, saved_errno, result = 1; size_t len; - ssize_t wret; struct binderfs_device device = { 0 }; struct binder_version version = { 0 }; char binderfs_mntpt[] = P_tmpdir "/binderfs_XXXXXX", diff --git a/tools/testing/selftests/hid/progs/hid_bpf_helpers.h b/tools/testing/selftests/hid/progs/hid_bpf_helpers.h index e5db897586bb..531228b849da 100644 --- a/tools/testing/selftests/hid/progs/hid_bpf_helpers.h +++ b/tools/testing/selftests/hid/progs/hid_bpf_helpers.h @@ -22,6 +22,9 @@ #define HID_REQ_SET_IDLE HID_REQ_SET_IDLE___not_used #define HID_REQ_SET_PROTOCOL HID_REQ_SET_PROTOCOL___not_used +/* do not define kfunc through vmlinux.h as this messes up our custom hack */ +#define BPF_NO_KFUNC_PROTOTYPES + #include "vmlinux.h" #undef hid_bpf_ctx @@ -91,31 +94,31 @@ struct hid_bpf_ops { /* following are kfuncs exported by HID for HID-BPF */ extern __u8 *hid_bpf_get_data(struct hid_bpf_ctx *ctx, unsigned int offset, - const size_t __sz) __ksym; -extern struct hid_bpf_ctx *hid_bpf_allocate_context(unsigned int hid_id) __ksym; -extern void hid_bpf_release_context(struct hid_bpf_ctx *ctx) __ksym; + const size_t __sz) __weak __ksym; +extern struct hid_bpf_ctx *hid_bpf_allocate_context(unsigned int hid_id) __weak __ksym; +extern void hid_bpf_release_context(struct hid_bpf_ctx *ctx) __weak __ksym; extern int hid_bpf_hw_request(struct hid_bpf_ctx *ctx, __u8 *data, size_t buf__sz, enum hid_report_type type, - enum hid_class_request reqtype) __ksym; + enum hid_class_request reqtype) __weak __ksym; extern int hid_bpf_hw_output_report(struct hid_bpf_ctx *ctx, - __u8 *buf, size_t buf__sz) __ksym; + __u8 *buf, size_t buf__sz) __weak __ksym; extern int hid_bpf_input_report(struct hid_bpf_ctx *ctx, enum hid_report_type type, __u8 *data, - size_t buf__sz) __ksym; + size_t buf__sz) __weak __ksym; extern int hid_bpf_try_input_report(struct hid_bpf_ctx *ctx, enum hid_report_type type, __u8 *data, - size_t buf__sz) __ksym; + size_t buf__sz) __weak __ksym; /* bpf_wq implementation */ extern int bpf_wq_init(struct bpf_wq *wq, void *p__map, unsigned int flags) __weak __ksym; extern int bpf_wq_start(struct bpf_wq *wq, unsigned int flags) __weak __ksym; extern int bpf_wq_set_callback_impl(struct bpf_wq *wq, int (callback_fn)(void *map, int *key, void *wq), - unsigned int flags__k, void *aux__ign) __ksym; + unsigned int flags__k, void *aux__ign) __weak __ksym; #define bpf_wq_set_callback(timer, cb, flags) \ bpf_wq_set_callback_impl(timer, cb, flags, NULL) diff --git a/tools/testing/selftests/hid/run-hid-tools-tests.sh b/tools/testing/selftests/hid/run-hid-tools-tests.sh index bdae8464da86..af1682a53c27 100755 --- a/tools/testing/selftests/hid/run-hid-tools-tests.sh +++ b/tools/testing/selftests/hid/run-hid-tools-tests.sh @@ -2,24 +2,26 @@ # SPDX-License-Identifier: GPL-2.0 # Runs tests for the HID subsystem +KSELFTEST_SKIP_TEST=4 + if ! command -v python3 > /dev/null 2>&1; then echo "hid-tools: [SKIP] python3 not installed" - exit 77 + exit $KSELFTEST_SKIP_TEST fi if ! python3 -c "import pytest" > /dev/null 2>&1; then - echo "hid: [SKIP/ pytest module not installed" - exit 77 + echo "hid: [SKIP] pytest module not installed" + exit $KSELFTEST_SKIP_TEST fi if ! python3 -c "import pytest_tap" > /dev/null 2>&1; then - echo "hid: [SKIP/ pytest_tap module not installed" - exit 77 + echo "hid: [SKIP] pytest_tap module not installed" + exit $KSELFTEST_SKIP_TEST fi if ! python3 -c "import hidtools" > /dev/null 2>&1; then - echo "hid: [SKIP/ hid-tools module not installed" - exit 77 + echo "hid: [SKIP] hid-tools module not installed" + exit $KSELFTEST_SKIP_TEST fi TARGET=${TARGET:=.} diff --git a/tools/testing/selftests/kvm/riscv/get-reg-list.c b/tools/testing/selftests/kvm/riscv/get-reg-list.c index 54ab484d0000..4bc1051848e5 100644 --- a/tools/testing/selftests/kvm/riscv/get-reg-list.c +++ b/tools/testing/selftests/kvm/riscv/get-reg-list.c @@ -47,6 +47,8 @@ bool filter_reg(__u64 reg) case KVM_REG_RISCV_ISA_EXT | KVM_REG_RISCV_ISA_SINGLE | KVM_RISCV_ISA_EXT_SSCOFPMF: case KVM_REG_RISCV_ISA_EXT | KVM_REG_RISCV_ISA_SINGLE | KVM_RISCV_ISA_EXT_SSNPM: case KVM_REG_RISCV_ISA_EXT | KVM_REG_RISCV_ISA_SINGLE | KVM_RISCV_ISA_EXT_SSTC: + case KVM_REG_RISCV_ISA_EXT | KVM_REG_RISCV_ISA_SINGLE | KVM_RISCV_ISA_EXT_SVADE: + case KVM_REG_RISCV_ISA_EXT | KVM_REG_RISCV_ISA_SINGLE | KVM_RISCV_ISA_EXT_SVADU: case KVM_REG_RISCV_ISA_EXT | KVM_REG_RISCV_ISA_SINGLE | KVM_RISCV_ISA_EXT_SVINVAL: case KVM_REG_RISCV_ISA_EXT | KVM_REG_RISCV_ISA_SINGLE | KVM_RISCV_ISA_EXT_SVNAPOT: case KVM_REG_RISCV_ISA_EXT | KVM_REG_RISCV_ISA_SINGLE | KVM_RISCV_ISA_EXT_SVPBMT: @@ -422,6 +424,8 @@ static const char *isa_ext_single_id_to_str(__u64 reg_off) KVM_ISA_EXT_ARR(SSCOFPMF), KVM_ISA_EXT_ARR(SSNPM), KVM_ISA_EXT_ARR(SSTC), + KVM_ISA_EXT_ARR(SVADE), + KVM_ISA_EXT_ARR(SVADU), KVM_ISA_EXT_ARR(SVINVAL), KVM_ISA_EXT_ARR(SVNAPOT), KVM_ISA_EXT_ARR(SVPBMT), @@ -955,6 +959,8 @@ KVM_ISA_EXT_SUBLIST_CONFIG(smstateen, SMSTATEEN); KVM_ISA_EXT_SIMPLE_CONFIG(sscofpmf, SSCOFPMF); KVM_ISA_EXT_SIMPLE_CONFIG(ssnpm, SSNPM); KVM_ISA_EXT_SIMPLE_CONFIG(sstc, SSTC); +KVM_ISA_EXT_SIMPLE_CONFIG(svade, SVADE); +KVM_ISA_EXT_SIMPLE_CONFIG(svadu, SVADU); KVM_ISA_EXT_SIMPLE_CONFIG(svinval, SVINVAL); KVM_ISA_EXT_SIMPLE_CONFIG(svnapot, SVNAPOT); KVM_ISA_EXT_SIMPLE_CONFIG(svpbmt, SVPBMT); @@ -1020,6 +1026,8 @@ struct vcpu_reg_list *vcpu_configs[] = { &config_sscofpmf, &config_ssnpm, &config_sstc, + &config_svade, + &config_svadu, &config_svinval, &config_svnapot, &config_svpbmt, diff --git a/tools/testing/selftests/module/find_symbol.sh b/tools/testing/selftests/module/find_symbol.sh index 140364d3c49f..2c56805c9b6e 100755 --- a/tools/testing/selftests/module/find_symbol.sh +++ b/tools/testing/selftests/module/find_symbol.sh @@ -44,10 +44,10 @@ load_mod() local ARCH="$(uname -m)" case "${ARCH}" in x86_64) - perf stat $STATS $MODPROBE test_kallsyms_b + perf stat $STATS $MODPROBE $MOD ;; *) - time $MODPROBE test_kallsyms_b + time $MODPROBE $MOD exit 1 ;; esac diff --git a/tools/testing/vsock/control.c b/tools/testing/vsock/control.c index d2deb4b15b94..0066e0324d35 100644 --- a/tools/testing/vsock/control.c +++ b/tools/testing/vsock/control.c @@ -27,6 +27,7 @@ #include "timeout.h" #include "control.h" +#include "util.h" static int control_fd = -1; @@ -50,7 +51,6 @@ void control_init(const char *control_host, for (ai = result; ai; ai = ai->ai_next) { int fd; - int val = 1; fd = socket(ai->ai_family, ai->ai_socktype, ai->ai_protocol); if (fd < 0) @@ -65,11 +65,8 @@ void control_init(const char *control_host, break; } - if (setsockopt(fd, SOL_SOCKET, SO_REUSEADDR, - &val, sizeof(val)) < 0) { - perror("setsockopt"); - exit(EXIT_FAILURE); - } + setsockopt_int_check(fd, SOL_SOCKET, SO_REUSEADDR, 1, + "setsockopt SO_REUSEADDR"); if (bind(fd, ai->ai_addr, ai->ai_addrlen) < 0) goto next; diff --git a/tools/testing/vsock/msg_zerocopy_common.c b/tools/testing/vsock/msg_zerocopy_common.c index 5a4bdf7b5132..8622e5a0f8b7 100644 --- a/tools/testing/vsock/msg_zerocopy_common.c +++ b/tools/testing/vsock/msg_zerocopy_common.c @@ -14,16 +14,6 @@ #include "msg_zerocopy_common.h" -void enable_so_zerocopy(int fd) -{ - int val = 1; - - if (setsockopt(fd, SOL_SOCKET, SO_ZEROCOPY, &val, sizeof(val))) { - perror("setsockopt"); - exit(EXIT_FAILURE); - } -} - void vsock_recv_completion(int fd, const bool *zerocopied) { struct sock_extended_err *serr; diff --git a/tools/testing/vsock/msg_zerocopy_common.h b/tools/testing/vsock/msg_zerocopy_common.h index 3763c5ccedb9..ad14139e93ca 100644 --- a/tools/testing/vsock/msg_zerocopy_common.h +++ b/tools/testing/vsock/msg_zerocopy_common.h @@ -12,7 +12,6 @@ #define VSOCK_RECVERR 1 #endif -void enable_so_zerocopy(int fd); void vsock_recv_completion(int fd, const bool *zerocopied); #endif /* MSG_ZEROCOPY_COMMON_H */ diff --git a/tools/testing/vsock/util.c b/tools/testing/vsock/util.c index a3d448a075e3..34e9dac0a105 100644 --- a/tools/testing/vsock/util.c +++ b/tools/testing/vsock/util.c @@ -651,3 +651,145 @@ void free_test_iovec(const struct iovec *test_iovec, free(iovec); } + +/* Set "unsigned long long" socket option and check that it's indeed set */ +void setsockopt_ull_check(int fd, int level, int optname, + unsigned long long val, char const *errmsg) +{ + unsigned long long chkval; + socklen_t chklen; + int err; + + err = setsockopt(fd, level, optname, &val, sizeof(val)); + if (err) { + fprintf(stderr, "setsockopt err: %s (%d)\n", + strerror(errno), errno); + goto fail; + } + + chkval = ~val; /* just make storage != val */ + chklen = sizeof(chkval); + + err = getsockopt(fd, level, optname, &chkval, &chklen); + if (err) { + fprintf(stderr, "getsockopt err: %s (%d)\n", + strerror(errno), errno); + goto fail; + } + + if (chklen != sizeof(chkval)) { + fprintf(stderr, "size mismatch: set %zu got %d\n", sizeof(val), + chklen); + goto fail; + } + + if (chkval != val) { + fprintf(stderr, "value mismatch: set %llu got %llu\n", val, + chkval); + goto fail; + } + return; +fail: + fprintf(stderr, "%s val %llu\n", errmsg, val); + exit(EXIT_FAILURE); +; +} + +/* Set "int" socket option and check that it's indeed set */ +void setsockopt_int_check(int fd, int level, int optname, int val, + char const *errmsg) +{ + int chkval; + socklen_t chklen; + int err; + + err = setsockopt(fd, level, optname, &val, sizeof(val)); + if (err) { + fprintf(stderr, "setsockopt err: %s (%d)\n", + strerror(errno), errno); + goto fail; + } + + chkval = ~val; /* just make storage != val */ + chklen = sizeof(chkval); + + err = getsockopt(fd, level, optname, &chkval, &chklen); + if (err) { + fprintf(stderr, "getsockopt err: %s (%d)\n", + strerror(errno), errno); + goto fail; + } + + if (chklen != sizeof(chkval)) { + fprintf(stderr, "size mismatch: set %zu got %d\n", sizeof(val), + chklen); + goto fail; + } + + if (chkval != val) { + fprintf(stderr, "value mismatch: set %d got %d\n", val, chkval); + goto fail; + } + return; +fail: + fprintf(stderr, "%s val %d\n", errmsg, val); + exit(EXIT_FAILURE); +} + +static void mem_invert(unsigned char *mem, size_t size) +{ + size_t i; + + for (i = 0; i < size; i++) + mem[i] = ~mem[i]; +} + +/* Set "timeval" socket option and check that it's indeed set */ +void setsockopt_timeval_check(int fd, int level, int optname, + struct timeval val, char const *errmsg) +{ + struct timeval chkval; + socklen_t chklen; + int err; + + err = setsockopt(fd, level, optname, &val, sizeof(val)); + if (err) { + fprintf(stderr, "setsockopt err: %s (%d)\n", + strerror(errno), errno); + goto fail; + } + + /* just make storage != val */ + chkval = val; + mem_invert((unsigned char *)&chkval, sizeof(chkval)); + chklen = sizeof(chkval); + + err = getsockopt(fd, level, optname, &chkval, &chklen); + if (err) { + fprintf(stderr, "getsockopt err: %s (%d)\n", + strerror(errno), errno); + goto fail; + } + + if (chklen != sizeof(chkval)) { + fprintf(stderr, "size mismatch: set %zu got %d\n", sizeof(val), + chklen); + goto fail; + } + + if (memcmp(&chkval, &val, sizeof(val)) != 0) { + fprintf(stderr, "value mismatch: set %ld:%ld got %ld:%ld\n", + val.tv_sec, val.tv_usec, chkval.tv_sec, chkval.tv_usec); + goto fail; + } + return; +fail: + fprintf(stderr, "%s val %ld:%ld\n", errmsg, val.tv_sec, val.tv_usec); + exit(EXIT_FAILURE); +} + +void enable_so_zerocopy_check(int fd) +{ + setsockopt_int_check(fd, SOL_SOCKET, SO_ZEROCOPY, 1, + "setsockopt SO_ZEROCOPY"); +} diff --git a/tools/testing/vsock/util.h b/tools/testing/vsock/util.h index fff22d4a14c0..ba84d296d8b7 100644 --- a/tools/testing/vsock/util.h +++ b/tools/testing/vsock/util.h @@ -68,4 +68,11 @@ unsigned long iovec_hash_djb2(const struct iovec *iov, size_t iovnum); struct iovec *alloc_test_iovec(const struct iovec *test_iovec, int iovnum); void free_test_iovec(const struct iovec *test_iovec, struct iovec *iovec, int iovnum); +void setsockopt_ull_check(int fd, int level, int optname, + unsigned long long val, char const *errmsg); +void setsockopt_int_check(int fd, int level, int optname, int val, + char const *errmsg); +void setsockopt_timeval_check(int fd, int level, int optname, + struct timeval val, char const *errmsg); +void enable_so_zerocopy_check(int fd); #endif /* UTIL_H */ diff --git a/tools/testing/vsock/vsock_perf.c b/tools/testing/vsock/vsock_perf.c index 4e8578f815e0..75971ac708c9 100644 --- a/tools/testing/vsock/vsock_perf.c +++ b/tools/testing/vsock/vsock_perf.c @@ -33,7 +33,7 @@ static unsigned int port = DEFAULT_PORT; static unsigned long buf_size_bytes = DEFAULT_BUF_SIZE_BYTES; -static unsigned long vsock_buf_bytes = DEFAULT_VSOCK_BUF_BYTES; +static unsigned long long vsock_buf_bytes = DEFAULT_VSOCK_BUF_BYTES; static bool zerocopy; static void error(const char *s) @@ -133,7 +133,7 @@ static float get_gbps(unsigned long bits, time_t ns_delta) ((float)ns_delta / NSEC_PER_SEC); } -static void run_receiver(unsigned long rcvlowat_bytes) +static void run_receiver(int rcvlowat_bytes) { unsigned int read_cnt; time_t rx_begin_ns; @@ -162,8 +162,8 @@ static void run_receiver(unsigned long rcvlowat_bytes) printf("Run as receiver\n"); printf("Listen port %u\n", port); printf("RX buffer %lu bytes\n", buf_size_bytes); - printf("vsock buffer %lu bytes\n", vsock_buf_bytes); - printf("SO_RCVLOWAT %lu bytes\n", rcvlowat_bytes); + printf("vsock buffer %llu bytes\n", vsock_buf_bytes); + printf("SO_RCVLOWAT %d bytes\n", rcvlowat_bytes); fd = socket(AF_VSOCK, SOCK_STREAM, 0); @@ -251,6 +251,16 @@ static void run_receiver(unsigned long rcvlowat_bytes) close(fd); } +static void enable_so_zerocopy(int fd) +{ + int val = 1; + + if (setsockopt(fd, SOL_SOCKET, SO_ZEROCOPY, &val, sizeof(val))) { + perror("setsockopt"); + exit(EXIT_FAILURE); + } +} + static void run_sender(int peer_cid, unsigned long to_send_bytes) { time_t tx_begin_ns; @@ -439,7 +449,7 @@ static long strtolx(const char *arg) int main(int argc, char **argv) { unsigned long to_send_bytes = DEFAULT_TO_SEND_BYTES; - unsigned long rcvlowat_bytes = DEFAULT_RCVLOWAT_BYTES; + int rcvlowat_bytes = DEFAULT_RCVLOWAT_BYTES; int peer_cid = -1; bool sender = false; diff --git a/tools/testing/vsock/vsock_test.c b/tools/testing/vsock/vsock_test.c index 8d38dbf8f41f..48f17641ca50 100644 --- a/tools/testing/vsock/vsock_test.c +++ b/tools/testing/vsock/vsock_test.c @@ -429,7 +429,7 @@ static void test_seqpacket_msg_bounds_client(const struct test_opts *opts) static void test_seqpacket_msg_bounds_server(const struct test_opts *opts) { - unsigned long sock_buf_size; + unsigned long long sock_buf_size; unsigned long remote_hash; unsigned long curr_hash; int fd; @@ -444,17 +444,13 @@ static void test_seqpacket_msg_bounds_server(const struct test_opts *opts) sock_buf_size = SOCK_BUF_SIZE; - if (setsockopt(fd, AF_VSOCK, SO_VM_SOCKETS_BUFFER_MAX_SIZE, - &sock_buf_size, sizeof(sock_buf_size))) { - perror("setsockopt(SO_VM_SOCKETS_BUFFER_MAX_SIZE)"); - exit(EXIT_FAILURE); - } + setsockopt_ull_check(fd, AF_VSOCK, SO_VM_SOCKETS_BUFFER_MAX_SIZE, + sock_buf_size, + "setsockopt(SO_VM_SOCKETS_BUFFER_MAX_SIZE)"); - if (setsockopt(fd, AF_VSOCK, SO_VM_SOCKETS_BUFFER_SIZE, - &sock_buf_size, sizeof(sock_buf_size))) { - perror("setsockopt(SO_VM_SOCKETS_BUFFER_SIZE)"); - exit(EXIT_FAILURE); - } + setsockopt_ull_check(fd, AF_VSOCK, SO_VM_SOCKETS_BUFFER_SIZE, + sock_buf_size, + "setsockopt(SO_VM_SOCKETS_BUFFER_SIZE)"); /* Ready to receive data. */ control_writeln("SRVREADY"); @@ -586,10 +582,8 @@ static void test_seqpacket_timeout_client(const struct test_opts *opts) tv.tv_sec = RCVTIMEO_TIMEOUT_SEC; tv.tv_usec = 0; - if (setsockopt(fd, SOL_SOCKET, SO_RCVTIMEO, (void *)&tv, sizeof(tv)) == -1) { - perror("setsockopt(SO_RCVTIMEO)"); - exit(EXIT_FAILURE); - } + setsockopt_timeval_check(fd, SOL_SOCKET, SO_RCVTIMEO, tv, + "setsockopt(SO_RCVTIMEO)"); read_enter_ns = current_nsec(); @@ -634,7 +628,8 @@ static void test_seqpacket_timeout_server(const struct test_opts *opts) static void test_seqpacket_bigmsg_client(const struct test_opts *opts) { - unsigned long sock_buf_size; + unsigned long long sock_buf_size; + size_t buf_size; socklen_t len; void *data; int fd; @@ -655,13 +650,20 @@ static void test_seqpacket_bigmsg_client(const struct test_opts *opts) sock_buf_size++; - data = malloc(sock_buf_size); + /* size_t can be < unsigned long long */ + buf_size = (size_t)sock_buf_size; + if (buf_size != sock_buf_size) { + fprintf(stderr, "Returned BUFFER_SIZE too large\n"); + exit(EXIT_FAILURE); + } + + data = malloc(buf_size); if (!data) { perror("malloc"); exit(EXIT_FAILURE); } - send_buf(fd, data, sock_buf_size, 0, -EMSGSIZE); + send_buf(fd, data, buf_size, 0, -EMSGSIZE); control_writeln("CLISENT"); @@ -835,7 +837,7 @@ static void test_stream_poll_rcvlowat_server(const struct test_opts *opts) static void test_stream_poll_rcvlowat_client(const struct test_opts *opts) { - unsigned long lowat_val = RCVLOWAT_BUF_SIZE; + int lowat_val = RCVLOWAT_BUF_SIZE; char buf[RCVLOWAT_BUF_SIZE]; struct pollfd fds; short poll_flags; @@ -847,11 +849,8 @@ static void test_stream_poll_rcvlowat_client(const struct test_opts *opts) exit(EXIT_FAILURE); } - if (setsockopt(fd, SOL_SOCKET, SO_RCVLOWAT, - &lowat_val, sizeof(lowat_val))) { - perror("setsockopt(SO_RCVLOWAT)"); - exit(EXIT_FAILURE); - } + setsockopt_int_check(fd, SOL_SOCKET, SO_RCVLOWAT, + lowat_val, "setsockopt(SO_RCVLOWAT)"); control_expectln("SRVSENT"); @@ -1357,9 +1356,10 @@ static void test_stream_rcvlowat_def_cred_upd_client(const struct test_opts *opt static void test_stream_credit_update_test(const struct test_opts *opts, bool low_rx_bytes_test) { - size_t recv_buf_size; + int recv_buf_size; struct pollfd fds; size_t buf_size; + unsigned long long sock_buf_size; void *buf; int fd; @@ -1371,11 +1371,12 @@ static void test_stream_credit_update_test(const struct test_opts *opts, buf_size = RCVLOWAT_CREDIT_UPD_BUF_SIZE; - if (setsockopt(fd, AF_VSOCK, SO_VM_SOCKETS_BUFFER_SIZE, - &buf_size, sizeof(buf_size))) { - perror("setsockopt(SO_VM_SOCKETS_BUFFER_SIZE)"); - exit(EXIT_FAILURE); - } + /* size_t can be < unsigned long long */ + sock_buf_size = buf_size; + + setsockopt_ull_check(fd, AF_VSOCK, SO_VM_SOCKETS_BUFFER_SIZE, + sock_buf_size, + "setsockopt(SO_VM_SOCKETS_BUFFER_SIZE)"); if (low_rx_bytes_test) { /* Set new SO_RCVLOWAT here. This enables sending credit @@ -1384,11 +1385,8 @@ static void test_stream_credit_update_test(const struct test_opts *opts, */ recv_buf_size = 1 + VIRTIO_VSOCK_MAX_PKT_BUF_SIZE; - if (setsockopt(fd, SOL_SOCKET, SO_RCVLOWAT, - &recv_buf_size, sizeof(recv_buf_size))) { - perror("setsockopt(SO_RCVLOWAT)"); - exit(EXIT_FAILURE); - } + setsockopt_int_check(fd, SOL_SOCKET, SO_RCVLOWAT, + recv_buf_size, "setsockopt(SO_RCVLOWAT)"); } /* Send one dummy byte here, because 'setsockopt()' above also @@ -1430,11 +1428,8 @@ static void test_stream_credit_update_test(const struct test_opts *opts, recv_buf_size++; /* Updating SO_RCVLOWAT will send credit update. */ - if (setsockopt(fd, SOL_SOCKET, SO_RCVLOWAT, - &recv_buf_size, sizeof(recv_buf_size))) { - perror("setsockopt(SO_RCVLOWAT)"); - exit(EXIT_FAILURE); - } + setsockopt_int_check(fd, SOL_SOCKET, SO_RCVLOWAT, + recv_buf_size, "setsockopt(SO_RCVLOWAT)"); } fds.fd = fd; diff --git a/tools/testing/vsock/vsock_test_zerocopy.c b/tools/testing/vsock/vsock_test_zerocopy.c index 04c376b6937f..9d9a6cb9614a 100644 --- a/tools/testing/vsock/vsock_test_zerocopy.c +++ b/tools/testing/vsock/vsock_test_zerocopy.c @@ -162,7 +162,7 @@ static void test_client(const struct test_opts *opts, } if (test_data->so_zerocopy) - enable_so_zerocopy(fd); + enable_so_zerocopy_check(fd); iovec = alloc_test_iovec(test_data->vecs, test_data->vecs_cnt); diff --git a/tools/testing/vsock/vsock_uring_test.c b/tools/testing/vsock/vsock_uring_test.c index 6c3e6f70c457..5c3078969659 100644 --- a/tools/testing/vsock/vsock_uring_test.c +++ b/tools/testing/vsock/vsock_uring_test.c @@ -73,7 +73,7 @@ static void vsock_io_uring_client(const struct test_opts *opts, } if (msg_zerocopy) - enable_so_zerocopy(fd); + enable_so_zerocopy_check(fd); iovec = alloc_test_iovec(test_data->vecs, test_data->vecs_cnt); |