From 03109e2f0d18dcb84218bd91c4fbf864193ca934 Mon Sep 17 00:00:00 2001 From: Zhang Rui Date: Thu, 14 Nov 2024 15:59:46 +0800 Subject: [PATCH 01/16] tools/power turbostat: Add support for /sys/class/drm/card1 On some machines, the graphics device is enumerated as /sys/class/drm/card1 instead of /sys/class/drm/card0. The current implementation does not handle this scenario, resulting in the loss of graphics C6 residency and frequency information. Add support for /sys/class/drm/card1, ensuring that turbostat can retrieve and display the graphics columns for these platforms. Signed-off-by: Zhang Rui Signed-off-by: Len Brown --- tools/power/x86/turbostat/turbostat.c | 38 ++++++++++++++++++++------- 1 file changed, 29 insertions(+), 9 deletions(-) diff --git a/tools/power/x86/turbostat/turbostat.c b/tools/power/x86/turbostat/turbostat.c index 28513172ffce..b250676c174e 100644 --- a/tools/power/x86/turbostat/turbostat.c +++ b/tools/power/x86/turbostat/turbostat.c @@ -6476,8 +6476,14 @@ static void set_graphics_fp(char *path, int idx) 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; @@ -6518,22 +6524,36 @@ static void probe_graphics(void) next: /* New i915 graphics sysfs knobs */ - if (!access("/sys/class/drm/card0/gt/gt0/rc6_residency_ms", R_OK)) { - set_graphics_fp("/sys/class/drm/card0/gt/gt0/rc6_residency_ms", GFX_rc6); + 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; + } - set_graphics_fp("/sys/class/drm/card0/gt/gt0/rps_cur_freq_mhz", GFX_MHz); + if (i == GFX_MAX_CARDS) + goto legacy_i915; - set_graphics_fp("/sys/class/drm/card0/gt/gt0/rps_act_freq_mhz", GFX_ACTMHz); + snprintf(path, PATH_MAX, "/sys/class/drm/card%d/gt/gt0/rc6_residency_ms", i); + set_graphics_fp(path, GFX_rc6); - set_graphics_fp("/sys/class/drm/card0/gt/gt1/rc6_residency_ms", SAM_mc6); + snprintf(path, PATH_MAX, "/sys/class/drm/card%d/gt/gt0/rps_cur_freq_mhz", i); + set_graphics_fp(path, GFX_MHz); - set_graphics_fp("/sys/class/drm/card0/gt/gt1/rps_cur_freq_mhz", SAM_MHz); + snprintf(path, PATH_MAX, "/sys/class/drm/card%d/gt/gt0/rps_act_freq_mhz", i); + set_graphics_fp(path, GFX_ACTMHz); - set_graphics_fp("/sys/class/drm/card0/gt/gt1/rps_act_freq_mhz", SAM_ACTMHz); + 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 */ set_graphics_fp("/sys/class/drm/card0/power/rc6_residency_ms", GFX_rc6); -- 2.51.0 From bcfab87108b33f20d847fd71a2a93114dd2ce83e Mon Sep 17 00:00:00 2001 From: Patryk Wlazlyn Date: Thu, 24 Oct 2024 15:17:45 +0200 Subject: [PATCH 02/16] tools/power turbostat: Force --no-perf in --dump mode 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. Signed-off-by: Patryk Wlazlyn Signed-off-by: Len Brown --- tools/power/x86/turbostat/turbostat.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/tools/power/x86/turbostat/turbostat.c b/tools/power/x86/turbostat/turbostat.c index b250676c174e..1fed799a5537 100644 --- a/tools/power/x86/turbostat/turbostat.c +++ b/tools/power/x86/turbostat/turbostat.c @@ -9897,6 +9897,12 @@ 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 */ -- 2.51.0 From 1da0daf746342dfdc114e4dc8fbf3ece28666d4f Mon Sep 17 00:00:00 2001 From: Patryk Wlazlyn Date: Wed, 13 Nov 2024 15:48:22 +0100 Subject: [PATCH 03/16] tools/power turbostat: Fix child's argument forwarding Add '+' to optstring when early scanning for --no-msr and --no-perf. It causes option processing to stop as soon as a nonoption argument is encountered, effectively skipping child's arguments. Fixes: 3e4048466c39 ("tools/power turbostat: Add --no-msr option") Signed-off-by: Patryk Wlazlyn Signed-off-by: Len Brown --- tools/power/x86/turbostat/turbostat.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/power/x86/turbostat/turbostat.c b/tools/power/x86/turbostat/turbostat.c index 1fed799a5537..9025c2945737 100644 --- a/tools/power/x86/turbostat/turbostat.c +++ b/tools/power/x86/turbostat/turbostat.c @@ -9873,7 +9873,7 @@ 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; -- 2.51.0 From e5f687b89bc2a892256f48e14a568970b59a4812 Mon Sep 17 00:00:00 2001 From: Patryk Wlazlyn Date: Wed, 2 Oct 2024 15:05:15 +0200 Subject: [PATCH 04/16] tools/power turbostat: Add RAPL psys as a built-in counter Introduce the counter as a part of global, platform counters structure. We open the counter for only one cpu, but otherwise treat it as an ordinary RAPL counter, allowing for grouped perf read. The counter is disabled by default, because it's interpretation may require additional, platform specific information, making it unsuitable for general use. Signed-off-by: Patryk Wlazlyn Signed-off-by: Len Brown --- tools/power/x86/turbostat/turbostat.8 | 2 + tools/power/x86/turbostat/turbostat.c | 93 ++++++++++++++++++++++++--- 2 files changed, 85 insertions(+), 10 deletions(-) diff --git a/tools/power/x86/turbostat/turbostat.8 b/tools/power/x86/turbostat/turbostat.8 index 56c7ff6efcda..95eb02346d3a 100644 --- a/tools/power/x86/turbostat/turbostat.8 +++ b/tools/power/x86/turbostat/turbostat.8 @@ -190,6 +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. May require platform specific information to interpret the data, making it not suitable for general use. +.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. diff --git a/tools/power/x86/turbostat/turbostat.c b/tools/power/x86/turbostat/turbostat.c index 9025c2945737..88c7f896c5b2 100644 --- a/tools/power/x86/turbostat/turbostat.c +++ b/tools/power/x86/turbostat/turbostat.c @@ -200,6 +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)) @@ -262,6 +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__) @@ -269,7 +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; @@ -507,12 +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) @@ -713,7 +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, }; @@ -730,7 +737,7 @@ 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, }; @@ -797,7 +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, }; @@ -813,7 +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 = { @@ -829,7 +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 = { @@ -845,7 +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 = { @@ -1108,6 +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, }; @@ -1134,6 +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 { @@ -1255,6 +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 { @@ -1682,6 +1704,7 @@ enum { IDX_PP1_ENERGY, IDX_PKG_PERF, IDX_DRAM_PERF, + IDX_PSYS_ENERGY, IDX_COUNT, }; @@ -1726,6 +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; } @@ -1756,6 +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; } @@ -1777,6 +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; } @@ -1848,6 +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; @@ -2512,6 +2547,11 @@ void print_header(char *delim) 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"); } @@ -2519,6 +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); @@ -2590,6 +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); @@ -2628,6 +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; @@ -2637,6 +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; @@ -3093,6 +3143,13 @@ int format_counters(struct thread_data *t, struct core_data *c, struct pkg_data } } + 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"); @@ -3400,6 +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; @@ -4129,6 +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]; @@ -4136,6 +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; @@ -4163,6 +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: @@ -4201,7 +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); @@ -4209,6 +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; } @@ -6144,6 +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(); @@ -6167,6 +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(); @@ -6945,8 +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; @@ -7606,6 +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 */ -- 2.51.0 From 86d237734091201d2ab2c1d2e1063893621c770f Mon Sep 17 00:00:00 2001 From: Len Brown Date: Sat, 30 Nov 2024 16:22:00 -0500 Subject: [PATCH 05/16] tools/power turbostat: 2024.11.30 since 2024.07.26: assorted minor bug fixes assorted platform specific tweaks initial RAPL PSYS (SysWatt) support Signed-off-by: Len Brown --- tools/power/x86/turbostat/turbostat.8 | 2 +- tools/power/x86/turbostat/turbostat.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/power/x86/turbostat/turbostat.8 b/tools/power/x86/turbostat/turbostat.8 index 95eb02346d3a..a7f7ed01421c 100644 --- a/tools/power/x86/turbostat/turbostat.8 +++ b/tools/power/x86/turbostat/turbostat.8 @@ -190,7 +190,7 @@ 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. May require platform specific information to interpret the data, making it not suitable for general use. +\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 diff --git a/tools/power/x86/turbostat/turbostat.c b/tools/power/x86/turbostat/turbostat.c index 88c7f896c5b2..58a487c225a7 100644 --- a/tools/power/x86/turbostat/turbostat.c +++ b/tools/power/x86/turbostat/turbostat.c @@ -9236,7 +9236,7 @@ int get_and_dump_counters(void) void print_version() { - fprintf(outf, "turbostat version 2024.07.26 - Len Brown \n"); + fprintf(outf, "turbostat version 2024.11.30 - Len Brown \n"); } #define COMMAND_LINE_SIZE 2048 -- 2.51.0 From f69e63756f7822fcdad8a34f9967e8b243e883ee Mon Sep 17 00:00:00 2001 From: "Dr. David Alan Gilbert" Date: Wed, 2 Oct 2024 18:31:47 +0100 Subject: [PATCH 06/16] printf: Remove unused 'bprintf' bprintf() is unused. Remove it. It was added in the commit 4370aa4aa753 ("vsprintf: add binary printf") but as far as I can see was never used, unlike the other two functions in that patch. Link: https://lore.kernel.org/20241002173147.210107-1-linux@treblig.org Reviewed-by: Andy Shevchenko Acked-by: Petr Mladek Signed-off-by: Dr. David Alan Gilbert Signed-off-by: Steven Rostedt (Google) --- include/linux/string.h | 1 - lib/vsprintf.c | 23 ----------------------- 2 files changed, 24 deletions(-) diff --git a/include/linux/string.h b/include/linux/string.h index 0dd27afcfaf7..493ac4862c77 100644 --- a/include/linux/string.h +++ b/include/linux/string.h @@ -335,7 +335,6 @@ int __sysfs_match_string(const char * const *array, size_t n, const char *s); #ifdef CONFIG_BINARY_PRINTF int vbin_printf(u32 *bin_buf, size_t size, const char *fmt, va_list args); int bstr_printf(char *buf, size_t size, const char *fmt, const u32 *bin_buf); -int bprintf(u32 *bin_buf, size_t size, const char *fmt, ...) __printf(3, 4); #endif extern ssize_t memory_read_from_buffer(void *to, size_t count, loff_t *ppos, diff --git a/lib/vsprintf.c b/lib/vsprintf.c index 6ac02bbb7df1..9d3dac38a3f4 100644 --- a/lib/vsprintf.c +++ b/lib/vsprintf.c @@ -3428,29 +3428,6 @@ out: } EXPORT_SYMBOL_GPL(bstr_printf); -/** - * bprintf - Parse a format string and place args' binary value in a buffer - * @bin_buf: The buffer to place args' binary value - * @size: The size of the buffer(by words(32bits), not characters) - * @fmt: The format string to use - * @...: Arguments for the format string - * - * The function returns the number of words(u32) written - * into @bin_buf. - */ -int bprintf(u32 *bin_buf, size_t size, const char *fmt, ...) -{ - va_list args; - int ret; - - va_start(args, fmt); - ret = vbin_printf(bin_buf, size, fmt, args); - va_end(args); - - return ret; -} -EXPORT_SYMBOL_GPL(bprintf); - #endif /* CONFIG_BINARY_PRINTF */ /** -- 2.51.0 From 9022ed0e7e65734d83a0648648589b9fbea8e8c9 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 1 Dec 2024 09:23:33 -0800 Subject: [PATCH 07/16] strscpy: write destination buffer only once The point behind strscpy() was to once and for all avoid all the problems with 'strncpy()' and later broken "fixed" versions like strlcpy() that just made things worse. So strscpy not only guarantees NUL-termination (unlike strncpy), it also doesn't do unnecessary padding at the destination. But at the same time also avoids byte-at-a-time reads and writes by _allowing_ some extra NUL writes - within the size, of course - so that the whole copy can be done with word operations. It is also stable in the face of a mutable source string: it explicitly does not read the source buffer multiple times (so an implementation using "strnlen()+memcpy()" would be wrong), and does not read the source buffer past the size (like the mis-design that is strlcpy does). Finally, the return value is designed to be simple and unambiguous: if the string cannot be copied fully, it returns an actual negative error, making error handling clearer and simpler (and the caller already knows the size of the buffer). Otherwise it returns the string length of the result. However, there was one final stability issue that can be important to callers: the stability of the destination buffer. In particular, the same way we shouldn't read the source buffer more than once, we should avoid doing multiple writes to the destination buffer: first writing a potentially non-terminated string, and then terminating it with NUL at the end does not result in a stable result buffer. Yes, it gives the right result in the end, but if the rule for the destination buffer was that it is _always_ NUL-terminated even when accessed concurrently with updates, the final byte of the buffer needs to always _stay_ as a NUL byte. [ Note that "final byte is NUL" here is literally about the final byte in the destination array, not the terminating NUL at the end of the string itself. There is no attempt to try to make concurrent reads and writes give any kind of consistent string length or contents, but we do want to guarantee that there is always at least that final terminating NUL character at the end of the destination array if it existed before ] This is relevant in the kernel for the tsk->comm[] array, for example. Even without locking (for either readers or writers), we want to know that while the buffer contents may be garbled, it is always a valid C string and always has a NUL character at 'comm[TASK_COMM_LEN-1]' (and never has any "out of thin air" data). So avoid any "copy possibly non-terminated string, and terminate later" behavior, and write the destination buffer only once. Signed-off-by: Linus Torvalds --- lib/string.c | 23 +++++++++++++++++------ 1 file changed, 17 insertions(+), 6 deletions(-) diff --git a/lib/string.c b/lib/string.c index 76327b51e36f..eb4486ed40d2 100644 --- a/lib/string.c +++ b/lib/string.c @@ -104,6 +104,12 @@ char *strncpy(char *dest, const char *src, size_t count) EXPORT_SYMBOL(strncpy); #endif +#ifdef __BIG_ENDIAN +# define ALLBUTLAST_BYTE_MASK (~255ul) +#else +# define ALLBUTLAST_BYTE_MASK (~0ul >> 8) +#endif + ssize_t sized_strscpy(char *dest, const char *src, size_t count) { const struct word_at_a_time constants = WORD_AT_A_TIME_CONSTANTS; @@ -147,13 +153,18 @@ ssize_t sized_strscpy(char *dest, const char *src, size_t count) *(unsigned long *)(dest+res) = c & zero_bytemask(data); return res + find_zero(data); } + count -= sizeof(unsigned long); + if (unlikely(!count)) { + c &= ALLBUTLAST_BYTE_MASK; + *(unsigned long *)(dest+res) = c; + return -E2BIG; + } *(unsigned long *)(dest+res) = c; res += sizeof(unsigned long); - count -= sizeof(unsigned long); max -= sizeof(unsigned long); } - while (count) { + while (count > 1) { char c; c = src[res]; @@ -164,11 +175,11 @@ ssize_t sized_strscpy(char *dest, const char *src, size_t count) count--; } - /* Hit buffer length without finding a NUL; force NUL-termination. */ - if (res) - dest[res-1] = '\0'; + /* Force NUL-termination. */ + dest[res] = '\0'; - return -E2BIG; + /* Return E2BIG if the source didn't stop */ + return src[res] ? -E2BIG : res; } EXPORT_SYMBOL(sized_strscpy); -- 2.51.0 From 40384c840ea1944d7c5a392e8975ed088ecf0b37 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 1 Dec 2024 14:28:56 -0800 Subject: [PATCH 08/16] Linux 6.13-rc1 --- Makefile | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Makefile b/Makefile index e34a97473fb6..93ab62cef244 100644 --- a/Makefile +++ b/Makefile @@ -1,8 +1,8 @@ # SPDX-License-Identifier: GPL-2.0 VERSION = 6 -PATCHLEVEL = 12 +PATCHLEVEL = 13 SUBLEVEL = 0 -EXTRAVERSION = +EXTRAVERSION = -rc1 NAME = Baby Opossum Posse # *DOCUMENTATION* -- 2.51.0 From 77a214317a6adcbfaf0781a2d950247106c6bc1f Mon Sep 17 00:00:00 2001 From: Simon Wunderlich Date: Wed, 16 Oct 2024 09:16:07 +0200 Subject: [PATCH 09/16] batman-adv: Start new development cycle This version will contain all the (major or even only minor) changes for Linux 6.14. The version number isn't a semantic version number with major and minor information. It is just encoding the year of the expected publishing as Linux -rc1 and the number of published versions this year (starting at 0). Signed-off-by: Simon Wunderlich --- net/batman-adv/main.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h index 97ea71a052f8..1fbe3a4dd965 100644 --- a/net/batman-adv/main.h +++ b/net/batman-adv/main.h @@ -13,7 +13,7 @@ #define BATADV_DRIVER_DEVICE "batman-adv" #ifndef BATADV_SOURCE_VERSION -#define BATADV_SOURCE_VERSION "2024.3" +#define BATADV_SOURCE_VERSION "2025.0" #endif /* B.A.T.M.A.N. parameters */ -- 2.51.0 From a7d5100ed0099a9f21b796017908cc7ece5332e4 Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Wed, 16 Oct 2024 09:23:06 +0200 Subject: [PATCH 10/16] batman-adv: Reorder includes for distributed-arp-table.c The commit 5f60d5f6bbc1 ("move asm/unaligned.h to linux/unaligned.h") changed the include without adjusting the order (to match the rest of the file). Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- net/batman-adv/distributed-arp-table.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c index 801eff8a40e5..48b72c2be098 100644 --- a/net/batman-adv/distributed-arp-table.c +++ b/net/batman-adv/distributed-arp-table.c @@ -7,7 +7,6 @@ #include "distributed-arp-table.h" #include "main.h" -#include #include #include #include @@ -32,6 +31,7 @@ #include #include #include +#include #include #include #include -- 2.51.0 From 8587e0e3f562b09bbfe0df1b0c506066095043e4 Mon Sep 17 00:00:00 2001 From: Remi Pommarel Date: Fri, 22 Nov 2024 16:52:51 +0100 Subject: [PATCH 11/16] batman-adv: Remove atomic usage for tt.local_changes The tt.local_changes atomic is either written with tt.changes_list_lock or close to it (see batadv_tt_local_event()). Thus the performance gain using an atomic was limited (or because of atomic_read() impact even negative). Using atomic also comes with the need to be wary of potential negative tt.local_changes value. Simplify the tt.local_changes usage by removing the atomic property and modifying it only with tt.changes_list_lock held. Signed-off-by: Remi Pommarel Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- net/batman-adv/soft-interface.c | 2 +- net/batman-adv/translation-table.c | 24 +++++++++++------------- net/batman-adv/types.h | 4 ++-- 3 files changed, 14 insertions(+), 16 deletions(-) diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c index 2758aba47a2f..5666c268cead 100644 --- a/net/batman-adv/soft-interface.c +++ b/net/batman-adv/soft-interface.c @@ -783,13 +783,13 @@ static int batadv_softif_init_late(struct net_device *dev) atomic_set(&bat_priv->mesh_state, BATADV_MESH_INACTIVE); atomic_set(&bat_priv->bcast_seqno, 1); atomic_set(&bat_priv->tt.vn, 0); - atomic_set(&bat_priv->tt.local_changes, 0); atomic_set(&bat_priv->tt.ogm_append_cnt, 0); #ifdef CONFIG_BATMAN_ADV_BLA atomic_set(&bat_priv->bla.num_requests, 0); #endif atomic_set(&bat_priv->tp_num, 0); + WRITE_ONCE(bat_priv->tt.local_changes, 0); bat_priv->tt.last_changeset = NULL; bat_priv->tt.last_changeset_len = 0; bat_priv->isolation_mark = 0; diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index b44c382226a1..6e0345b91ece 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -423,8 +423,8 @@ static void batadv_tt_local_event(struct batadv_priv *bat_priv, struct batadv_tt_change_node *tt_change_node, *entry, *safe; struct batadv_tt_common_entry *common = &tt_local_entry->common; u8 flags = common->flags | event_flags; - bool event_removed = false; bool del_op_requested, del_op_entry; + size_t changes; tt_change_node = kmem_cache_alloc(batadv_tt_change_cache, GFP_ATOMIC); if (!tt_change_node) @@ -440,6 +440,7 @@ static void batadv_tt_local_event(struct batadv_priv *bat_priv, /* check for ADD+DEL or DEL+ADD events */ spin_lock_bh(&bat_priv->tt.changes_list_lock); + changes = READ_ONCE(bat_priv->tt.local_changes); list_for_each_entry_safe(entry, safe, &bat_priv->tt.changes_list, list) { if (!batadv_compare_eth(entry->change.addr, common->addr)) @@ -468,21 +469,18 @@ static void batadv_tt_local_event(struct batadv_priv *bat_priv, del: list_del(&entry->list); kmem_cache_free(batadv_tt_change_cache, entry); + changes--; kmem_cache_free(batadv_tt_change_cache, tt_change_node); - event_removed = true; - goto unlock; + goto update_changes; } /* track the change in the OGMinterval list */ list_add_tail(&tt_change_node->list, &bat_priv->tt.changes_list); + changes++; -unlock: +update_changes: + WRITE_ONCE(bat_priv->tt.local_changes, changes); spin_unlock_bh(&bat_priv->tt.changes_list_lock); - - if (event_removed) - atomic_dec(&bat_priv->tt.local_changes); - else - atomic_inc(&bat_priv->tt.local_changes); } /** @@ -950,7 +948,7 @@ static void batadv_tt_tvlv_container_update(struct batadv_priv *bat_priv) int tt_diff_entries_count = 0; u16 tvlv_len; - tt_diff_entries_num = atomic_read(&bat_priv->tt.local_changes); + tt_diff_entries_num = READ_ONCE(bat_priv->tt.local_changes); tt_diff_len = batadv_tt_len(tt_diff_entries_num); /* if we have too many changes for one packet don't send any @@ -970,7 +968,7 @@ static void batadv_tt_tvlv_container_update(struct batadv_priv *bat_priv) goto container_register; spin_lock_bh(&bat_priv->tt.changes_list_lock); - atomic_set(&bat_priv->tt.local_changes, 0); + WRITE_ONCE(bat_priv->tt.local_changes, 0); list_for_each_entry_safe(entry, safe, &bat_priv->tt.changes_list, list) { @@ -1380,7 +1378,7 @@ static void batadv_tt_changes_list_free(struct batadv_priv *bat_priv) kmem_cache_free(batadv_tt_change_cache, entry); } - atomic_set(&bat_priv->tt.local_changes, 0); + WRITE_ONCE(bat_priv->tt.local_changes, 0); spin_unlock_bh(&bat_priv->tt.changes_list_lock); } @@ -3634,7 +3632,7 @@ static void batadv_tt_local_commit_changes_nolock(struct batadv_priv *bat_priv) { lockdep_assert_held(&bat_priv->tt.commit_lock); - if (atomic_read(&bat_priv->tt.local_changes) < 1) { + if (READ_ONCE(bat_priv->tt.local_changes) == 0) { if (!batadv_atomic_dec_not_zero(&bat_priv->tt.ogm_append_cnt)) batadv_tt_tvlv_container_update(bat_priv); return; diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index 04f6398b3a40..f491bff8c51b 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -1022,7 +1022,7 @@ struct batadv_priv_tt { atomic_t ogm_append_cnt; /** @local_changes: changes registered in an originator interval */ - atomic_t local_changes; + size_t local_changes; /** * @changes_list: tracks tt local changes within an originator interval @@ -1044,7 +1044,7 @@ struct batadv_priv_tt { */ struct list_head roam_list; - /** @changes_list_lock: lock protecting changes_list */ + /** @changes_list_lock: lock protecting changes_list & local_changes */ spinlock_t changes_list_lock; /** @req_list_lock: lock protecting req_list */ -- 2.51.0 From fca81aa3e653d68ac696f3ee994384cecacf3845 Mon Sep 17 00:00:00 2001 From: Remi Pommarel Date: Fri, 22 Nov 2024 16:52:52 +0100 Subject: [PATCH 12/16] batman-adv: Don't keep redundant TT change events When adding a local TT twice within the same OGM interval (e.g. happens when flag get updated), the flags of the first TT change entry is updated with the second one and both change events is added to the change list. This leads to having the same ADD change entry twice. Similarly, a DEL+DEL scenario is also creating twice the same event. Deduplicate ADD+ADD or DEL+DEL scenarios to reduce the TT change events that need to be sent in both OGM and TT response. Signed-off-by: Remi Pommarel Co-developed-by: Sven Eckelmann Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- net/batman-adv/translation-table.c | 40 ++++++++++++++---------------- 1 file changed, 18 insertions(+), 22 deletions(-) diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c index 6e0345b91ece..76d5517bb507 100644 --- a/net/batman-adv/translation-table.c +++ b/net/batman-adv/translation-table.c @@ -438,7 +438,7 @@ static void batadv_tt_local_event(struct batadv_priv *bat_priv, del_op_requested = flags & BATADV_TT_CLIENT_DEL; - /* check for ADD+DEL or DEL+ADD events */ + /* check for ADD+DEL, DEL+ADD, ADD+ADD or DEL+DEL events */ spin_lock_bh(&bat_priv->tt.changes_list_lock); changes = READ_ONCE(bat_priv->tt.local_changes); list_for_each_entry_safe(entry, safe, &bat_priv->tt.changes_list, @@ -446,30 +446,26 @@ static void batadv_tt_local_event(struct batadv_priv *bat_priv, if (!batadv_compare_eth(entry->change.addr, common->addr)) continue; - /* DEL+ADD in the same orig interval have no effect and can be - * removed to avoid silly behaviour on the receiver side. The - * other way around (ADD+DEL) can happen in case of roaming of - * a client still in the NEW state. Roaming of NEW clients is - * now possible due to automatically recognition of "temporary" - * clients - */ del_op_entry = entry->change.flags & BATADV_TT_CLIENT_DEL; - if (!del_op_requested && del_op_entry) - goto del; - if (del_op_requested && !del_op_entry) - goto del; - - /* this is a second add in the same originator interval. It - * means that flags have been changed: update them! - */ - if (!del_op_requested && !del_op_entry) + if (del_op_requested != del_op_entry) { + /* DEL+ADD in the same orig interval have no effect and + * can be removed to avoid silly behaviour on the + * receiver side. The other way around (ADD+DEL) can + * happen in case of roaming of a client still in the + * NEW state. Roaming of NEW clients is now possible due + * to automatically recognition of "temporary" clients + */ + list_del(&entry->list); + kmem_cache_free(batadv_tt_change_cache, entry); + changes--; + } else { + /* this is a second add or del in the same originator + * interval. It could mean that flags have been changed + * (e.g. double add): update them + */ entry->change.flags = flags; + } - continue; -del: - list_del(&entry->list); - kmem_cache_free(batadv_tt_change_cache, entry); - changes--; kmem_cache_free(batadv_tt_change_cache, tt_change_node); goto update_changes; } -- 2.51.0 From bf2a5a622a50b47560e4b29387b2fd4cfb33881a Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Mon, 16 Dec 2024 19:37:12 +0100 Subject: [PATCH 13/16] batman-adv: Map VID 0 to untagged TT VLAN MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit VID 0 is not a valid VLAN according to "802.1Q-2011" "Table 9-2—Reserved VID values". It is only used to indicate "priority tag" frames which only contain priority information and no VID. The 8021q is also redirecting the priority tagged frames to the underlying interface since commit ad1afb003939 ("vlan_dev: VLAN 0 should be treated as "no vlan tag" (802.1p packet)"). But at the same time, it automatically adds the VID 0 to all devices to ensure that VID 0 is in the allowed list of the HW filter. This resulted in a VLAN 0 which was always announced in OGM messages. batman-adv should therefore not create a new batadv_softif_vlan for VID 0 and handle all VID 0 related frames using the "untagged" global/local translation tables. Signed-off-by: Sven Eckelmann Acked-by: Antonio Quartulli Signed-off-by: Simon Wunderlich --- net/batman-adv/main.c | 7 +++++++ net/batman-adv/soft-interface.c | 14 ++++++++++++++ 2 files changed, 21 insertions(+) diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c index 8e0f44c71696..333e947afcce 100644 --- a/net/batman-adv/main.c +++ b/net/batman-adv/main.c @@ -637,6 +637,13 @@ unsigned short batadv_get_vid(struct sk_buff *skb, size_t header_len) vhdr = (struct vlan_ethhdr *)(skb->data + header_len); vid = ntohs(vhdr->h_vlan_TCI) & VLAN_VID_MASK; + + /* VID 0 is only used to indicate "priority tag" frames which only + * contain priority information and no VID. + */ + if (vid == 0) + return BATADV_NO_FLAGS; + vid |= BATADV_VLAN_HAS_TAG; return vid; diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c index 5666c268cead..822d788a5f86 100644 --- a/net/batman-adv/soft-interface.c +++ b/net/batman-adv/soft-interface.c @@ -637,6 +637,14 @@ static int batadv_interface_add_vid(struct net_device *dev, __be16 proto, if (proto != htons(ETH_P_8021Q)) return -EINVAL; + /* VID 0 is only used to indicate "priority tag" frames which only + * contain priority information and no VID. No management structures + * should be created for this VID and it should be handled like an + * untagged frame. + */ + if (vid == 0) + return 0; + vid |= BATADV_VLAN_HAS_TAG; /* if a new vlan is getting created and it already exists, it means that @@ -684,6 +692,12 @@ static int batadv_interface_kill_vid(struct net_device *dev, __be16 proto, if (proto != htons(ETH_P_8021Q)) return -EINVAL; + /* "priority tag" frames are handled like "untagged" frames + * and no softif_vlan needs to be destroyed + */ + if (vid == 0) + return 0; + vlan = batadv_softif_vlan_get(bat_priv, vid | BATADV_VLAN_HAS_TAG); if (!vlan) return -ENOENT; -- 2.51.0 From 7bce3f75189c1c2d7b43af8184b284d783e51375 Mon Sep 17 00:00:00 2001 From: Marek Lindner Date: Fri, 27 Dec 2024 10:19:12 +0100 Subject: [PATCH 14/16] MAINTAINERS: update email address of Marek Linder Signed-off-by: Marek Lindner Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- .mailmap | 2 ++ Documentation/networking/batman-adv.rst | 2 +- MAINTAINERS | 2 +- net/batman-adv/main.h | 2 +- 4 files changed, 5 insertions(+), 3 deletions(-) diff --git a/.mailmap b/.mailmap index 5ff0e5d681e7..53e038f5f654 100644 --- a/.mailmap +++ b/.mailmap @@ -427,6 +427,8 @@ Marcin Nowakowski Marc Zyngier Marek Behún Marek Behún Marek Behun +Marek Lindner +Marek Lindner Mark Brown Mark Starovoytov Markus Schneider-Pargmann diff --git a/Documentation/networking/batman-adv.rst b/Documentation/networking/batman-adv.rst index 8a0dcb1894b4..44b9b5cc0e24 100644 --- a/Documentation/networking/batman-adv.rst +++ b/Documentation/networking/batman-adv.rst @@ -164,5 +164,5 @@ Mailing-list: You can also contact the Authors: -* Marek Lindner +* Marek Lindner * Simon Wunderlich diff --git a/MAINTAINERS b/MAINTAINERS index 1e930c7a58b1..c5e909a759e6 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3867,7 +3867,7 @@ S: Maintained F: drivers/platform/x86/barco-p50-gpio.c BATMAN ADVANCED -M: Marek Lindner +M: Marek Lindner M: Simon Wunderlich M: Antonio Quartulli M: Sven Eckelmann diff --git a/net/batman-adv/main.h b/net/batman-adv/main.h index 1fbe3a4dd965..964f3088af5b 100644 --- a/net/batman-adv/main.h +++ b/net/batman-adv/main.h @@ -7,7 +7,7 @@ #ifndef _NET_BATMAN_ADV_MAIN_H_ #define _NET_BATMAN_ADV_MAIN_H_ -#define BATADV_DRIVER_AUTHOR "Marek Lindner , " \ +#define BATADV_DRIVER_AUTHOR "Marek Lindner , " \ "Simon Wunderlich " #define BATADV_DRIVER_DESC "B.A.T.M.A.N. advanced" #define BATADV_DRIVER_DEVICE "batman-adv" -- 2.51.0 From 1f5f7ff46435226f34112e0d179f6cbd9eb30b04 Mon Sep 17 00:00:00 2001 From: Simon Wunderlich Date: Tue, 31 Dec 2024 10:31:29 +0100 Subject: [PATCH 15/16] mailmap: add entries for Simon Wunderlich Map the defunc mail addresses to the currently used mail address (listed in MAINTAINERS). Signed-off-by: Simon Wunderlich --- .mailmap | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/.mailmap b/.mailmap index 53e038f5f654..efabe03f7849 100644 --- a/.mailmap +++ b/.mailmap @@ -642,6 +642,11 @@ Simona Vetter Simon Horman Simon Horman Simon Kelley +Simon Wunderlich +Simon Wunderlich +Simon Wunderlich +Simon Wunderlich +Simon Wunderlich Sricharan Ramabadhran Srinivas Ramana Sriram R -- 2.51.0 From 285c72be944089e333ae5f49211b38e460f9528d Mon Sep 17 00:00:00 2001 From: Sven Eckelmann Date: Tue, 31 Dec 2024 10:31:29 +0100 Subject: [PATCH 16/16] mailmap: add entries for Sven Eckelmann Map the defunc mail addresses to the currently used mail address (listed in MAINTAINERS). Signed-off-by: Sven Eckelmann Signed-off-by: Simon Wunderlich --- .mailmap | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/.mailmap b/.mailmap index efabe03f7849..41aca254671d 100644 --- a/.mailmap +++ b/.mailmap @@ -667,6 +667,11 @@ Sudarshan Rajagopalan Sudeep Holla Sudeep KarkadaNagesha Sumit Semwal Surabhi Vishnoi +Sven Eckelmann +Sven Eckelmann +Sven Eckelmann +Sven Eckelmann +Sven Eckelmann Takashi YOSHII Tamizh Chelvam Raja Taniya Das -- 2.51.0