)]}'
{
  "log": [
    {
      "commit": "fd8b912165dfffccb4f652610a0a19afc51d98ee",
      "tree": "f1eb7ec443c02bd553f716b9f8b05e60a346e2e0",
      "parents": [
        "cedfc87c4c262744167947f845c2222514a8c9f1"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Thu Nov 19 06:40:07 2026 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Nov 19 09:40:07 2026 -0500"
      },
      "message": "Add ARM Lumex-C2-Ultra uarch detection, cache decode, and core scores (#463)\n\nAdd cpuinfo_uarch_lumex_c2_ultra (0x00300604) for ARM implementer 0x41\nCPU part 0xD96 (Arm C2-Ultra), format it as \"Lumex-C2-Ultra\" in\ncpu-info (keeping the lumex_ prefix for consistency with\ncpuinfo_uarch_lumex_c1_*), and score the Lumex C1 and C2 tiers in\nmidr_score_core() so prime and mid-tier clusters stay grouped even with\nnon-monotonic cluster frequencies or offline cores.\n\nReferences (public):\n- Arm C2-Ultra Core Technical Reference Manual (109736):\n  https://developer.arm.com/documentation/109736/latest\n- Arm C2-Ultra product page:\n  https://developer.arm.com/Processors/C2-Ultra\n- Upstream LLVM Host.cpp mapping 0x41 / 0xD96 to c2-ultra:\n  https://github.com/llvm/llvm-project/commit/b4badb65df4ea15fc6ecabd8f9366a96df11dcb5\n\nAlso add cpuinfo_arm_decode_cache() support for\ncpuinfo_uarch_lumex_c2_ultra using the public Arm C2-Ultra TRM 109736\ncache defaults (64 KB 4-way L1I, 128 KB 4-way L1D, 2 MB 8-way private\nL2 minimum configuration, 64-byte lines) with unit test coverage in\ntest/arm-cache.cc, and add the exynos-9975 mode to\ntest/mock/arm-linux-topology.cc using synthetic non-monotonic\nfrequencies to cover the midr_score_core() ordering."
    },
    {
      "commit": "cedfc87c4c262744167947f845c2222514a8c9f1",
      "tree": "593040f56d74c5f6d2e398b2afef8a7d9ac6e8a7",
      "parents": [
        "0bbfab7e26e1394f844689381e54b81aa135b686"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Thu Nov 19 06:39:36 2026 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Nov 19 09:39:36 2026 -0500"
      },
      "message": "Add Intel Diamond Rapids (Panther Cove) uarch and AMX-AVX512 ISA (#449)\n\nIntel Diamond Rapids reports CPUID family 0x13 (19), model 0x01 (Panther\nCove, previously reported as unknown) and introduces AMX-AVX512 in CPUID\nleaf 0x1E, subleaf 0x01, EAX bit 7 for moving and converting tile rows\ndirectly from AMX TMM registers to AVX-512 ZMM registers (tilemovrow,\ntcvtrowd2ps, tcvtrowps2bf16h/l, tcvtrowps2phh/l).\n\nAdd cpuinfo_uarch_panther_cove (\"Panther Cove\") for Intel family 0x13\nmodel 0x01, check OS AMX tile state in XCR0[18:17] (TILECFG and\nTILEDATA) alongside AVX-512 state when detecting amx_avx512 in\ncpuinfo_x86_detect_isa, expose cpuinfo_has_x86_amx_avx512(), and print\nAMX_AVX512 in isa-info.\n\nVerified under Intel SDE 10.13.1:\n  sde64 -spr -- ./isa-info                        # AMX_AVX512: no\n  sde64 -gnr -- ./isa-info                        # AMX_AVX512: no\n  sde64 -dmr -- ./cpu-info                        # 12x Panther Cove\n  sde64 -dmr -- ./isa-info                        # AMX_AVX512: yes\n  sde64 -dmr -xsave_default 0xe7 -- ./isa-info    # AMX_AVX512: no\n  sde64 -dmr -xsave_default 0x600e7 -- ./isa-info # AMX_AVX512: yes"
    },
    {
      "commit": "0bbfab7e26e1394f844689381e54b81aa135b686",
      "tree": "19251efb471adce82323a368c9ceafe428727c16",
      "parents": [
        "2ca65a8a4c0542da03f53fa31b56ae483cb36e55"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Thu Nov 19 06:26:27 2026 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Nov 19 09:26:27 2026 -0500"
      },
      "message": "Add Qualcomm Oryon V4 uarch detection for ARM64 (#462)\n\nQualcomm SM8975 (Samsung Galaxy S27 Ultra) reports implementer 0x51\nand CPU part 0x003 across its 8 cores (variants 0x0, 0x1, and 0x2),\nwhich cpuinfo previously reported as unknown.\n\nAdd cpuinfo_uarch_oryon_v4 (0x00400107), decode Qualcomm CPU part\n0x003 to cpuinfo_uarch_oryon_v4 on ARM64, and format it as \"Oryon V4\"\nin cpu-info."
    },
    {
      "commit": "2ca65a8a4c0542da03f53fa31b56ae483cb36e55",
      "tree": "da5e34a6fb4fd69fc3eb91d83bd586580d526b64",
      "parents": [
        "3ce3d87adef69f04e196e25b7f754028d12750c4"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Mon Nov 16 07:43:30 2026 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Nov 16 10:43:30 2026 -0500"
      },
      "message": "Guard processors[i].cache.l2 assignment in cpuinfo_arm_linux_init (#434)\n\nOn ARM Linux platforms where L2 is shared by cores in the same cluster,\nif arm_linux_processors[i].package_leader_id !\u003d arm_linux_processors[i].system_processor_id\n(or if cluster topology does not assign a leader on processor 0), l2_index remains\nat UINT32_MAX. Setting processors[i].cache.l2 \u003d l2 + l2_index resulted in an invalid\nout-of-bounds pointer (l2 + UINT32_MAX), leading to SIGSEGV crashes when downstream\ncallers access processor-\u003ecache.l2.\n\nGuard the assignment with an (l2 !\u003d NULL \u0026\u0026 l2_index \u003c l2_count) check,\nmatching the same bounds validation pattern already used for L3 (l3_index \u003c l3_count).\nAlso add pointer bounds checks in test/init.cc for consistent_l2 and consistent_l3.\n\nCo-authored-by: Alexandru Paiu \u003capaiu@google.com\u003e"
    },
    {
      "commit": "3ce3d87adef69f04e196e25b7f754028d12750c4",
      "tree": "fb8419db04bdb99c0b5ee6d384235640cb4fe4d3",
      "parents": [
        "731932997e55e84900e6f95aefb6edd167b0edcc"
      ],
      "author": {
        "name": "Jacob Stevens",
        "email": "stevens.jacob1492@gmail.com",
        "time": "Wed Nov 11 14:42:42 2026 -0500"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Nov 11 14:42:42 2026 -0500"
      },
      "message": "Add CPUINFO_AVOID_SELINUX_VIOLATIONS to skip non-standard Android properties (#459)\n\n* Add CPUINFO_AVOID_SELINUX_VIOLATIONS to skip non-standard Android properties\n\nOn Android ARM, cpuinfo reads several properties that are not defined by\nany domain in the Android platform sepolicy:\n\n* ro.mediatek.platform\n* ro.chipname\n* ro.hardware.chipname\n\nProcesses linking cpuinfo therefore generate SELinux denials on devices\nwhose vendor policy does not label these properties.\n\nAdd a CPUINFO_AVOID_SELINUX_VIOLATIONS compile-time flag (and matching\nCMake option, OFF by default) that skips reading them, at the cost of\nless precise chipset detection on devices that rely on them. Default\nbehavior is unchanged.\n\nRe-implementation of #242.\n\n* Use #if for CPUINFO_AVOID_SELINUX_VIOLATIONS and add Bazel define\n\nDefault CPUINFO_AVOID_SELINUX_VIOLATIONS to 0 and test it with #if rather\nthan #ifndef, matching CPUINFO_LOG_TO_STDIO and friends, so that\n-DCPUINFO_AVOID_SELINUX_VIOLATIONS\u003d0 keeps reading the properties.\n\nAdd an avoid_selinux_violations config_setting to BUILD.bazel so Bazel\nconsumers can enable the flag with\n--define\u003dcpuinfo_avoid_selinux_violations\u003dtrue."
    },
    {
      "commit": "731932997e55e84900e6f95aefb6edd167b0edcc",
      "tree": "6ca27a6f03b2608be76d9fb14bed9d4dc042b30b",
      "parents": [
        "86acbee086b4e27da17195d1593a10169540ebee"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Wed Nov 04 09:36:25 2026 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Nov 04 12:36:25 2026 -0500"
      },
      "message": "Add Qualcomm Oryon and Oryon V3 cache decoding on ARM (#451)\n\nOn Android kernels where /sys/devices/system/cpu/cpu*/cache/index*/size\nis absent, cpuinfo falls back to cpuinfo_arm_decode_cache(). Neither\ncpuinfo_uarch_oryon nor cpuinfo_uarch_oryon_v3 was handled, hitting the\ngeneric ARMv8 fallback (32 KB L1I/L1D and 256 KB per core L2, reporting\nonly 512 KB L2 on the 2-core Prime cluster instead of 12 MB) and\ncapping cpuinfo_arm_compute_max_cache_size() at 4 MB.\n\nDecode cache sizes and associativity for cpuinfo_uarch_oryon and\ncpuinfo_uarch_oryon_v3:\n- Prime cores: 192 KB 6-way L1I, 96 KB 6-way L1D, 12 MB 12-way L2.\n- Performance cores (Oryon variant 0x4, Oryon V3 variant 0x2):\n  128 KB 4-way L1I, 64 KB 4-way L1D, 12 MB 12-way L2.\n\nAlso read ways_of_associativity from sysfs when overriding L2 size so\ncache-\u003esets remains consistent on systems with non-12-way L2 caches,\nand add unit tests for Snapdragon X Elite, Snapdragon 8 Elite, and\nSnapdragon 8 Elite Gen 5."
    },
    {
      "commit": "86acbee086b4e27da17195d1593a10169540ebee",
      "tree": "94205e988d472ce4b1cf249d051606fafb77e0f9",
      "parents": [
        "e38248d114ee2df0f5ea27fca2453bceefd43d52"
      ],
      "author": {
        "name": "Jacob Stevens",
        "email": "stevens.jacob1492@gmail.com",
        "time": "Wed Nov 04 12:28:20 2026 -0500"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Nov 04 12:28:20 2026 -0500"
      },
      "message": "Fix ARM Linux topology for heterogeneous core groups (#438)"
    },
    {
      "commit": "e38248d114ee2df0f5ea27fca2453bceefd43d52",
      "tree": "088352ec8eef9bfa5ab55c2f0adb72bab93715c9",
      "parents": [
        "f1a04cb8759aacab99bf658b5459e30eb69679b3"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Wed Nov 04 08:50:40 2026 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Nov 04 11:50:40 2026 -0500"
      },
      "message": "Disable ARM64 Linux SME features when base SME is disabled (#455)\n\nOn ARM64 Linux, the kernel can report SME sub-feature hwcaps (such as\nHWCAP2_SME2, HWCAP2_SME2P1, and HWCAP2_SME_*) while base SME is\ndisabled (HWCAP2_SME is clear or prctl(PR_SME_GET_VL) fails). Clear all\nSME feature flags in cpuinfo_arm64_linux_decode_isa_from_proc_cpuinfo\nwhen isa-\u003esme is false or isa-\u003esmelen is 0 so downstream callers do not\nexecute SME instructions and trigger SIGILL.\n\nThis is the ARM64 Linux counterpart to #432 / #433 (which gated Mach/iOS\nSME feature flags on smelen \u003e 0).\n\nFixes #442"
    },
    {
      "commit": "f1a04cb8759aacab99bf658b5459e30eb69679b3",
      "tree": "1eabd627396194b6811bfbfb593f79389f521393",
      "parents": [
        "e9320ce3398899fbecc986c82f49b5d59b80dff7"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Tue Nov 03 10:45:00 2026 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Nov 03 13:45:00 2026 -0500"
      },
      "message": "Add Intel client and hybrid x86 microarchitectures with P-core/E-core detection (#443)\n\n* Add Intel client and hybrid x86 microarchitectures\n\n* Detect both P-core and E-core uarchs on hybrid x86 without threads or extra OS calls\n\n* Resolve current uarch index on x86 with CPUID leaf 0x1A instead of getcpu\n\ncpuinfo_get_current_uarch_index() and _with_default() now read the core\ntype of the calling core from CPUID(0x1A).EAX[31:24] (0x20 \u003d Atom E-core,\n0x40 \u003d Core P-core) and map it to the matching cpuinfo_uarchs entry. This\nneeds no system call, so it works on every OS (Linux, Android, Windows,\nmacOS, FreeBSD) and inside sandboxes, and is skipped entirely on\nnon-hybrid processors. The now-unused x86 Linux cpu_to_uarch_index_map is\nremoved.\n\n* Address review feedback from @digantdesai\n\n- src/x86/linux/init.c, src/x86/init.c: classify hybrid cores from\n  /sys/devices/cpu_core/cpus, /sys/devices/cpu_atom/cpus, and\n  /sys/devices/cpu_lowpower/cpus; keep a single uarch (and return\n  uarch index 0 without CPUID) when sysfs PMU lists are missing or\n  ambiguous instead of guessing from thread counts, APIC deltas, or a\n  50/50 split (@digantdesai)\n  https://github.com/pytorch/cpuinfo/pull/443#discussion_r4385174860\n- src/x86/windows/init.c: split P-core and E-core uarchs only when\n  EfficiencyClass varies across cores, falling back to a single uarch\n  when uniform (@digantdesai)\n  https://github.com/pytorch/cpuinfo/pull/443#discussion_r4385174865\n- src/x86/uarch.c: move Arrow Lake-U (0xB5) to the Meteor Lake Redwood\n  Cove and Crestmont case (@digantdesai)\n  https://github.com/pytorch/cpuinfo/pull/443#discussion_r4385174868\n- include/cpuinfo.h, src/x86/uarch.c, tools/cpu-info.c: add\n  cpuinfo_uarch_arctic_wolf for Nova Lake E-cores (@digantdesai)\n  https://github.com/pytorch/cpuinfo/pull/443#discussion_r4385174872\n\n* Detect Crestmont Low-Power E-cores on Arrow Lake-H\n\nArrow Lake-H (family 6, model 0xC5) pairs Lion Cove P-cores and\nSkymont E-cores on the compute tile with two Crestmont Low-Power\nIsland E-cores on the SoC tile (CPUID.1AH.EAX[23:0] native model ID\n0x02 vs 0x03 for Skymont, exposed under /sys/devices/cpu_lowpower/cpus\non Linux).\n\nDecode lp_e_uarch in cpuinfo_x86_init_processor using native model ID\n0x02, map /sys/devices/cpu_lowpower/cpus on Linux and the lowest\nthree-tier EfficiencyClass on Windows to lp_e_uarch, and check\nCPUID.1AH.EAX[23:0] in cpuinfo_x86_get_current_uarch_index so\nArrow Lake-H Crestmont LP-E cores report cpuinfo_uarch_crestmont\ninstead of being grouped with Skymont."
    },
    {
      "commit": "e9320ce3398899fbecc986c82f49b5d59b80dff7",
      "tree": "7c9a569ad22cec073e0b335d7d131efe170809ef",
      "parents": [
        "d53799f29aee624d8c6b99bbe9b4a5846ae82588"
      ],
      "author": {
        "name": "Ashley Feng",
        "email": "147632986+ashuie@users.noreply.github.com",
        "time": "Fri Oct 30 12:28:50 2026 -0500"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Oct 30 13:28:50 2026 -0400"
      },
      "message": "Add Arm Neoverse V3/V3AE recognition (#435)\n\n* Add Arm Neoverse V3 recognition\n\n* Add Arm Neoverse V3AE recognition\n\n* Fix Neoverse V3 minimum L2 cache size\n\n* Add Neoverse V3 cache coverage\n\n* Fix Neoverse V3 formatting"
    },
    {
      "commit": "d53799f29aee624d8c6b99bbe9b4a5846ae82588",
      "tree": "4ba3139fec7464d6fb0244d19ea8725c2c4df801",
      "parents": [
        "783a616b2197d7e48d6507210d1d0009fbf1f590"
      ],
      "author": {
        "name": "patryk-kaiser-ARM",
        "email": "patryk.kaiser@arm.com",
        "time": "Thu Oct 29 18:11:15 2026 +0000"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Oct 29 14:11:15 2026 -0400"
      },
      "message": "Add Arm SVE2.1 and SME MOP4 feature detection (#439)\n\nSigned-off-by: Patryk Kaiser \u003cpatryk.kaiser@arm.com\u003e"
    },
    {
      "commit": "783a616b2197d7e48d6507210d1d0009fbf1f590",
      "tree": "94bf3bef0c50fdfc5d99d7b3885831ed188aab51",
      "parents": [
        "b23e963e93ec65a12d7f5b65fa36f30704744294"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Tue Oct 27 04:50:54 2026 +0000"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Oct 26 23:50:54 2026 -0500"
      },
      "message": "Allow x86 Linux cpuinfo initialization to succeed without /proc/cpuinfo (#447)"
    },
    {
      "commit": "b23e963e93ec65a12d7f5b65fa36f30704744294",
      "tree": "8faafdfe709747a182ffaddc7f996c0b3437f6fe",
      "parents": [
        "66ee79c038d70dad9f08705b2c9b3e58f6d8f512"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Tue Oct 27 01:53:51 2026 +0000"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Oct 26 20:53:51 2026 -0500"
      },
      "message": "Add Sony PlayStation 5 (Oberon) uarch detection (#441)\n\nAMD family 17h extended model 0x4 now returns cpuinfo_uarch_oberon\n(previously zen2, mislabeled as Xbox Series X):\n- model 46h: Sony PlayStation 5 (Oberon), retail\n  https://github.com/ps5-linux/ps5-linux-patches/pull/29\n- model 47h: AMD 4700S Desktop Kit (Cardinal), CPUID 0x00840F70\n  http://users.atw.hu/instlatx64/AuthenticAMD/AuthenticAMD0840F70_K17_CPUID.txt\n\nAMD family 17h extended model 0x8 now returns cpuinfo_uarch_arden:\n- model 84h: Microsoft Xbox Series X/S (Arden) / AMD 4800S, CPUID 0x00880F40\n  http://users.atw.hu/instlatx64/AuthenticAMD/AuthenticAMD0880F40_K17_CPUID.txt"
    },
    {
      "commit": "66ee79c038d70dad9f08705b2c9b3e58f6d8f512",
      "tree": "6fa72253182068bce65eb8af443237641096a1dc",
      "parents": [
        "8ce83db858065145192c97af90cb668ad72a12e9"
      ],
      "author": {
        "name": "Mark Hansen",
        "email": "mark@markhansen.co.nz",
        "time": "Fri Jul 31 03:32:33 2026 +1000"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jul 30 10:32:33 2026 -0700"
      },
      "message": "Fix Clang CFI type mismatch in src/arm/linux/cpuinfo.c (#403)\n\nChange parse_line signature to accept void* and cast inside.\nRemove the cast when passing it to cpuinfo_linux_parse_multiline_file.\n\nFixes #402\n\nCo-authored-by: Nikita Shulga \u003cnikita.shulga@gmail.com\u003e"
    },
    {
      "commit": "8ce83db858065145192c97af90cb668ad72a12e9",
      "tree": "4a2ea9a2b029f7ef4483861d064856c1a222b998",
      "parents": [
        "2cd040684ee617caa6ae783372d0aadfb2918116"
      ],
      "author": {
        "name": "Nikita Shulga",
        "email": "2453524+malfet@users.noreply.github.com",
        "time": "Thu Jul 30 09:55:14 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jul 30 09:55:14 2026 -0700"
      },
      "message": "Fix C++17 array initializer warnings (#430)\n\nFixes warnings like because C++17 is not C99 compatible...\n```\n/home/runner/work/cpuinfo/cpuinfo/test/arm-cache.cc:364:5: warning: array designators are a C99 extension [-Wc99-designator]\n  364 |                                 [0] \u003d \u0027P\u0027,\n      |                                 ^~~\n/home/runner/work/cpuinfo/cpuinfo/test/arm-cache.cc:620:5: warning: array designators are a C99 extension [-Wc99-designator]\n  620 |                                 [0] \u003d \u0027P\u0027,\n      |                                 ^~~\n/home/runner/work/cpuinfo/cpuinfo/test/arm-cache.cc:867:5: warning: array designators are a C99 extension [-Wc99-designator]\n  867 |                                 [0] \u003d \u0027P\u0027,\n      |                                 ^~~\n/home/runner/work/cpuinfo/cpuinfo/test/arm-cache.cc:1240:5: warning: array designators are a C99 extension [-Wc99-designator]\n 1240 |                                 [0] \u003d \u0027C\u0027,\n      |                                 ^~~\n```"
    },
    {
      "commit": "2cd040684ee617caa6ae783372d0aadfb2918116",
      "tree": "870df3f36ddb1757941ffad1e0169437790d5834",
      "parents": [
        "d6108819d1d41c749b076e250be454c4559cb771"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Thu Jul 30 09:39:47 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jul 30 09:39:47 2026 -0700"
      },
      "message": "Add missing iOS, iPad, iPod, and AppleTV device model mappings (#428)\n\n* Add missing iOS, iPad, iPod, and AppleTV device model mappings\n\nUpdate decode_package_name_from_hw_machine to support missing iPad, iPhone, iPod, and AppleTV hardware machine model identifiers (iPad8,x through iPad16,x; iPod9,1; AppleTV2,1 through AppleTV14,1) and map them to their corresponding Apple A-series and M-series SoCs.\n\nFixes #425, #140\n\nCo-authored-by: Nikita Shulga \u003cnikita.shulga@gmail.com\u003e\nCo-authored-by: Claude Opus 4.8 \u003cnoreply@anthropic.com\u003e"
    },
    {
      "commit": "d6108819d1d41c749b076e250be454c4559cb771",
      "tree": "62bd44e5dc6ca873a9002903177e67d84acc664c",
      "parents": [
        "46fbc42c10b39bfd4ed6c3a90749080fb18103c9"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Thu Jul 23 13:19:03 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jul 23 16:19:03 2026 -0400"
      },
      "message": "Detect Apple M5 Pro / Max (Sotra) super and performance cores (#424)\n\nAdd microarchitecture detection for Apple M5 Pro and M5 Max (Sotra SoC / CPUFAMILY_ARM_SOTRA).\n\n- Add cpuinfo_uarch_sotra_super and cpuinfo_uarch_sotra_performance enums.\n- Detect 15-core (5 Super + 10 Performance) and 18-core (6 Super + 12 Performance) configurations on macOS/XNU.\n- Update tools/cpu-info.c to use consistent spacing for Apple uarch display names.\n\nResolves https://github.com/pytorch/cpuinfo/issues/423"
    },
    {
      "commit": "46fbc42c10b39bfd4ed6c3a90749080fb18103c9",
      "tree": "7d955069fc53c993aba0f8c9c356c0ceab08ea95",
      "parents": [
        "6882af581bbbdd073c946eae4fd21bc037ff39cf"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Thu Jul 23 13:18:51 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jul 23 16:18:51 2026 -0400"
      },
      "message": "Rename ARM FP8 inline helpers to NEON naming conventions (#426)\n\nRename cpuinfo_has_arm_f8dot and cpuinfo_has_arm_f8mm to\ncpuinfo_has_arm_neon_f8dot4 and cpuinfo_has_arm_neon_f8mm.\n\nARM type support refers to scalar ARM, not NEON. FP8 scalar conversions\nare checked via cpuinfo_has_arm_fp8, while the dot product and matrix\nmultiplication instructions are NEON/SIMD.\n\nDetails:\n- Add backward compatibility macros:\n  #define cpuinfo_has_arm_f8dot cpuinfo_has_arm_neon_f8dot4\n  #define cpuinfo_has_arm_f8mm cpuinfo_has_arm_neon_f8mm\n- Rename inline helpers to cpuinfo_has_arm_neon_f8dot4 and cpuinfo_has_arm_neon_f8mm\n- Update callers in tools/isa-info.c\n- Add test coverage in test/mock/pixel-8.cc\n\nFixes https://github.com/pytorch/cpuinfo/issues/427"
    },
    {
      "commit": "6882af581bbbdd073c946eae4fd21bc037ff39cf",
      "tree": "47302a8225794276af965884110e16e8ab8a29bd",
      "parents": [
        "c3766aefc0bb0287bdbb7db71d6621e453ceb48a"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Thu Jul 16 08:23:46 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jul 16 11:23:46 2026 -0400"
      },
      "message": "Add ARM v8.7 FP8 ISA feature detection and reporting to cpuinfo. (#413)\n\nExpose ARM v8.7 / Armv9.6-A FP8 instructions across public API structs, Linux HWCAP2 / /proc/cpuinfo parsing, Apple sysctls, Windows processor feature checks, and isa_info:\n- Base FP8 (cpuinfo_has_arm_fp8 / FEAT_FP8 / HWCAP2_F8CVT / F8E4M3 / F8E5M2)\n- FP8 4-way Dot Product (cpuinfo_has_arm_f8dot / FEAT_FP8DOT4 / HWCAP2_F8DP4)\n- FP8 8-way Matrix Multiply-Accumulate into FP32 (cpuinfo_has_arm_f8mm / FEAT_F8F32MM / HWCAP_F8MM8)\n\nFixes https://github.com/pytorch/cpuinfo/issues/412"
    },
    {
      "commit": "c3766aefc0bb0287bdbb7db71d6621e453ceb48a",
      "tree": "ace4f40c0b25b1e5601a9e51e40ca7790fd8f5c7",
      "parents": [
        "9c6d24853efbb689a500507693e12c1fe540aa57"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Thu Jul 16 06:34:42 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jul 16 09:34:42 2026 -0400"
      },
      "message": "Guard _DEFAULT_SOURCE with #ifndef in src/api.c (#422)\n\nGuard the definition of _DEFAULT_SOURCE in src/api.c with #ifndef to prevent macro redefinition build errors when _DEFAULT_SOURCE or _GNU_SOURCE is already defined by the build environment (e.g. -D_GNU_SOURCE\u003d1 or -D_DEFAULT_SOURCE\u003d1)."
    },
    {
      "commit": "9c6d24853efbb689a500507693e12c1fe540aa57",
      "tree": "5cb8292ae92216b524aff85a52e46ce34d6984ed",
      "parents": [
        "b1a5d63f585bdb41c97ed3cb00daa64e06f982cc"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Tue Jul 14 10:53:47 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jul 14 13:53:47 2026 -0400"
      },
      "message": "Add AMX-FP8 ISA feature detection and reporting to cpuinfo (#416)\n\nExpose Intel AMX-FP8 (`amx_fp8`) instructions in `pytorch/cpuinfo`:\n- Add `amx_fp8` boolean field to `struct cpuinfo_x86_isa`\n- Add public inline accessor `cpuinfo_has_x86_amx_fp8(void)`\n- Detect support via CPUID Leaf 0x1E, Sub-leaf 1 (ECX \u003d 1), EAX bit 4 (`0x00000010`) and AVX512/AMX registers\n- Report `AMX_FP8` support in `isa-info` tool"
    },
    {
      "commit": "b1a5d63f585bdb41c97ed3cb00daa64e06f982cc",
      "tree": "8ffff493560fedbc620689f3a44b4bd537cc41b1",
      "parents": [
        "69e20fa2bb5e4d6950e431d09db028a21c8d9487"
      ],
      "author": {
        "name": "Nikita Shulga",
        "email": "2453524+malfet@users.noreply.github.com",
        "time": "Tue Jul 07 10:45:47 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jul 07 10:45:47 2026 -0700"
      },
      "message": "Revert \"Implement `cpuinfo_deinitialize()` to free heap-allocated globals (#387)\" (#411)\n\nThis reverts commit 4628dc060ce4e82345dc166bbac875609db4ff69."
    },
    {
      "commit": "69e20fa2bb5e4d6950e431d09db028a21c8d9487",
      "tree": "1504826df37aeb50d06b8855bc805dbd601800a2",
      "parents": [
        "ae5443646e5092d42d2a5b6e202b548169bc023e"
      ],
      "author": {
        "name": "Mark Hansen",
        "email": "mark@markhansen.co.nz",
        "time": "Wed Jul 08 03:42:07 2026 +1000"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jul 07 10:42:07 2026 -0700"
      },
      "message": "Remove obsolete .travis.yml (#410)\n\nAs this hasn\u0027t been edited in 8 years, and the badge to travis 404s, I\u0027m\nguessing that this is obsolete. It was a little distracting when I was\ntrying to figure out how you run your tests (not on GitHub Actions, so I\nwondered briefly if it was on Travis)."
    },
    {
      "commit": "ae5443646e5092d42d2a5b6e202b548169bc023e",
      "tree": "02ceae7486a3ee29c9b78b23ea2779a4b18745ba",
      "parents": [
        "315d03cacc51bfabe316057b0d3466e13bce88a0"
      ],
      "author": {
        "name": "Nikita Shulga",
        "email": "2453524+malfet@users.noreply.github.com",
        "time": "Wed Jul 01 11:17:46 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Jul 01 11:17:46 2026 -0700"
      },
      "message": "[CI] Fix Win UWP builds (#401)\n\nBy bumping generator to VS2026, to match https://github.com/actions/runner-images/blob/main/images/windows/Windows2025-VS2026-Readme.md"
    },
    {
      "commit": "315d03cacc51bfabe316057b0d3466e13bce88a0",
      "tree": "0d31801ebeb45dcd85add5998181fb4c22867899",
      "parents": [
        "317b8b50ea64a17ca0d185e93ece101a22b4c326"
      ],
      "author": {
        "name": "James Y Knight",
        "email": "foom@fuhm.net",
        "time": "Wed Jun 17 14:29:55 2026 -0400"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Jun 17 11:29:55 2026 -0700"
      },
      "message": "Delete unused variable \"core_apic_id\" (#272)\n\n(Diagnosed by -Wunused-variable)"
    },
    {
      "commit": "317b8b50ea64a17ca0d185e93ece101a22b4c326",
      "tree": "7d87b9323d1811e5a3082214ab1185635c9cb0a2",
      "parents": [
        "66eb859892822ab5e0ebfd5d4ba6a841e25f431b"
      ],
      "author": {
        "name": "Dmitry Baryshkov",
        "email": "dmitry.baryshkov@oss.qualcomm.com",
        "time": "Wed Jun 17 20:09:38 2026 +0300"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Jun 17 10:09:38 2026 -0700"
      },
      "message": "Fix syscall() undeclared error if built with stricter build options (#374)\n\nIf cpuinfo is built with a stricer build flags, it fails on some systems\nwith the undeclared function error, because on Glibc syscall() is not a\npart of the headers unless one of the feature macros is defined. Add the\nleast intrusive one, _DEFAULT_SOURCE.\n\ncpuinfo/src/api.c:319:23: error: call to undeclared function \u0027syscall\u0027; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]\n                if CPUINFO_UNLIKELY(syscall(__NR_getcpu, \u0026cpu, NULL, NULL) !\u003d 0) {\n                                    ^\ncpuinfo/src/api.c:338:23: error: call to undeclared function \u0027syscall\u0027; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]\n                if CPUINFO_UNLIKELY(syscall(__NR_getcpu, \u0026cpu, NULL, NULL) !\u003d 0) {\n\nCloses #171"
    },
    {
      "commit": "66eb859892822ab5e0ebfd5d4ba6a841e25f431b",
      "tree": "fdece830b436fa80485bddb63b82d3e84620b19e",
      "parents": [
        "f6f4b3edd3d91ddb4c9a8d8fe8904106751fb10a"
      ],
      "author": {
        "name": "johnthacker",
        "email": "johnthacker@gmail.com",
        "time": "Wed Jun 17 13:00:24 2026 -0400"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Jun 17 10:00:24 2026 -0700"
      },
      "message": "Normalization of AMD brand string with \"with Radeon Graphics\" (#389)\n\nSome AMD processors with integrated graphics have \"with Radeon Graphics\"\nin the cpuid brand string without being preceded by other tokens like\n\"SOC\" or \"APU\" that are already used to erase the string. With the\ncurrent logic, everything starting with \"Radeon\" is erased, leading to\nless than ideal results. Add a check for a \"with\" token and erase\neverything starting with that.\n\nRunning cpu-info before:\n\nPackages:\n        0: AMD Ryzen 5 PRO 6650U with\n\nand after:\n\nPackages:\n        0: AMD Ryzen 5 PRO 6650U\n\nThere are plenty of other examples at\nhttps://github.com/InstLatx64/InstLatx64/tree/master/AuthenticAMD\n\nSigned-off-by: John Thacker \u003cjohnthacker@gmail.com\u003e"
    },
    {
      "commit": "f6f4b3edd3d91ddb4c9a8d8fe8904106751fb10a",
      "tree": "391fca382cf9c3981d40af458cccd18de3e40877",
      "parents": [
        "f72c5c54d1d644a9915ad89f7cef269309357b76"
      ],
      "author": {
        "name": "vozvivan",
        "email": "vozvivan@gmail.com",
        "time": "Wed Jun 17 20:00:06 2026 +0300"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Jun 17 10:00:06 2026 -0700"
      },
      "message": "arm/linux: zero-initialize stack buffers to fix Valgrind warnings (#376)\n\nOn modern arm64 kernels, /proc/cpuinfo may legitimately omit the\n\"Hardware\" field. Other optional identification sources, such as\n\"Revision\" or Android property-based identifiers, may also be absent\ndepending on the platform.\n\nDefault-initialize the corresponding stack buffers / structs so missing\noptional identifiers are treated as empty values rather than\nuninitialized memory."
    },
    {
      "commit": "f72c5c54d1d644a9915ad89f7cef269309357b76",
      "tree": "e53444dbba9a43a1aff9ca131d7d0055a7e2212d",
      "parents": [
        "f9176bdf8fd453835eac402e2b294cda3432daf7"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Wed Jun 17 09:32:56 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Jun 17 09:32:56 2026 -0700"
      },
      "message": "Add Nova Lake (Coyote Cove) uarch support (#393)\n\n* Add Nova Lake (Coyote Cove) uarch support. Resolves #392\n\n* Fix clang-format\n\n---------\n\nCo-authored-by: Gregory Comer \u003cgregoryjcomer@gmail.com\u003e"
    },
    {
      "commit": "f9176bdf8fd453835eac402e2b294cda3432daf7",
      "tree": "22d3e4b57f74138910d9121030ed3476c59e8bc3",
      "parents": [
        "0c0ab15cb0a8bafbbf71c2ae6f84128a4c2a8da6"
      ],
      "author": {
        "name": "maxim-davgalev",
        "email": "Maxim.Davgalev@gmail.com",
        "time": "Wed Jun 17 09:24:40 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Jun 17 09:24:40 2026 -0700"
      },
      "message": "Initialize vendor/uarch out-parameters in cpuinfo_arm_decode_vendor_uarch (#395)\n\nThe default case of the implementer switch only logs a warning and\nreturns without writing *vendor or *uarch. Callers such as\nget_core_info_from_midr() in src/arm/windows/init.c pass uninitialized\nstack variables, so for any unhandled implementer the reported vendor\nand uarch are whatever happened to be on the stack.\n\nInitialize both out-parameters to cpuinfo_vendor_unknown /\ncpuinfo_uarch_unknown at function entry so the decode result is\ndeterministic for every implementer, present or future.\n\nCo-authored-by: Maxim Davgalev \u003cmaxd@parallels.com\u003e"
    },
    {
      "commit": "0c0ab15cb0a8bafbbf71c2ae6f84128a4c2a8da6",
      "tree": "38e2da1d91b54b3d0a2fed4baabfe8ea154ccfdb",
      "parents": [
        "398ad9f5d798587740a92f1cdaee0c5694aeccc9"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Wed Jun 17 09:11:36 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Jun 17 09:11:36 2026 -0700"
      },
      "message": "Add Apple M1/M2/M3/M4 Linux MIDR detection (#385)\n\n* Add Apple M1/M2 Linux MIDR detection\n\nFixes #380 by adding cpuinfo_arm_decode_vendor_uarch logic for Apple\u0027s implementer \u0027a\u0027 (0x61).\n\n* Add Apple M3/M4 Linux MIDR detection\n\n* style: format src/arm/uarch.c with clang-format"
    },
    {
      "commit": "398ad9f5d798587740a92f1cdaee0c5694aeccc9",
      "tree": "b7d4387b6777eb535d63fc1b432c373759deb061",
      "parents": [
        "4628dc060ce4e82345dc166bbac875609db4ff69"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Wed Jun 17 09:08:55 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Jun 17 09:08:55 2026 -0700"
      },
      "message": "fix: remove unused import \u0027string\u0027 in android-device-dump.py (#391)"
    },
    {
      "commit": "4628dc060ce4e82345dc166bbac875609db4ff69",
      "tree": "45e80f86189d2ab72f5a12b3dee48fd92a28c128",
      "parents": [
        "ea6b9f1bb6e1001d8b21574d5bc78ddef62e499d"
      ],
      "author": {
        "name": "Vineeth Chelur",
        "email": "crvineeth97@gmail.com",
        "time": "Thu Jun 11 17:13:27 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jun 11 17:13:27 2026 -0700"
      },
      "message": "Implement `cpuinfo_deinitialize()` to free heap-allocated globals (#387)\n\n* Add support for freeing cpuinfo memory\n\n* Add deinit tests\n\n* Move the deinits to respective allocation routines\n\n* Use compound literals for guard resets\n\n---------\n\nCo-authored-by: Vineeth Chelur \u003cvchelur@microsoft.com\u003e"
    },
    {
      "commit": "ea6b9f1bb6e1001d8b21574d5bc78ddef62e499d",
      "tree": "bd3bfb950aff2c268c254fdbb8fd91e8a9607d5b",
      "parents": [
        "3681f0ce1446167d01dfe125d6db96ba2ac31c3c"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Mon May 18 11:31:26 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon May 18 11:31:26 2026 -0700"
      },
      "message": "Add Google Tensor SoC detection. (#382)\n\nIdentify Google Tensor SoCs (Tensor, GS201) in ro.soc.model Android system property.\nUpdated src/arm/api.h and src/arm/linux/chipset.c to match internal Google CL 604382821."
    },
    {
      "commit": "3681f0ce1446167d01dfe125d6db96ba2ac31c3c",
      "tree": "e7ff9a6a208587e7a4a4ec1ff377fca2a70b2668",
      "parents": [
        "e829e80faba35db623b5e272c867ad72146adcda"
      ],
      "author": {
        "name": "Nicolas Pitre",
        "email": "nico@fluxnic.net",
        "time": "Thu Apr 30 19:06:12 2026 -0400"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Apr 30 16:06:12 2026 -0700"
      },
      "message": "Add Cortex-A320 to MIDR decode table (#384)\n\nARM Cortex-A320 (MIDR part 0xD8F) is an ARMv9.2-A efficiency core.\nAdd its uarch enum and MIDR mapping so XNNPACK can select optimized\nkernels when running on this core.\n\nSigned-off-by: Nicolas Pitre \u003cnpitre@baylibre.com\u003e"
    },
    {
      "commit": "e829e80faba35db623b5e272c867ad72146adcda",
      "tree": "ea1048a5fe1636ffca70830b36b6a731addce6bb",
      "parents": [
        "d05fbcd57dc096718c4979e7c054e628f1f3520b"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Tue Apr 28 09:00:41 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Apr 28 09:00:41 2026 -0700"
      },
      "message": "Adding ro.soc.model support to cpuinfo to detect Qualcomm SM8850 SoC (#381)\n\n* Adding ro.soc.model support to cpuinfo to detect Qualcomm SM8850 SoC.\n\n* style: format src/riscv/linux/riscv-hw.c with clang-format"
    },
    {
      "commit": "d05fbcd57dc096718c4979e7c054e628f1f3520b",
      "tree": "0f25b381978e8359626dede347287818002e16ec",
      "parents": [
        "bc3c01e230c6974283e4b89421cfb0e232435589"
      ],
      "author": {
        "name": "Ken Unger",
        "email": "108287829+ken-unger@users.noreply.github.com",
        "time": "Tue Apr 14 22:20:00 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Apr 14 22:20:00 2026 -0700"
      },
      "message": "Add riscv half-precision floating point detection (#375)\n\nAdd cpuinfo_has_riscv_zfh() and cpuinfo_has_riscv_zvfh for fp16 detection.  \n\nThe motivation here is to enable this runtime detection support in xnnpack for its rvv fp16 kernels. (xnnpack uses this library)  "
    },
    {
      "commit": "bc3c01e230c6974283e4b89421cfb0e232435589",
      "tree": "0f9b497a3f298d624f160784844e566e7626385e",
      "parents": [
        "7607ca500436b37ad23fb8d18614bec7796b68a7"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Mon Apr 13 13:19:06 2026 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Apr 13 13:19:06 2026 -0700"
      },
      "message": "Add Sapphire Rapids (Golden Cove) microarchitecture detection (#377)\n\n* Add Sapphire Rapids (Golden Cove) microarchitecture detection\n\n* Add Emerald Rapids and Granite Rapids microarchitecture detection"
    },
    {
      "commit": "7607ca500436b37ad23fb8d18614bec7796b68a7",
      "tree": "68fdec874a9b8ead86a941a869ed5d9920e583c1",
      "parents": [
        "7364b490b5f78d58efe23ea76e74210fd6c3c76f"
      ],
      "author": {
        "name": "Rohanjames1997",
        "email": "rohan.james4@gmail.com",
        "time": "Wed Mar 11 20:02:42 2026 -0500"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Mar 11 18:02:42 2026 -0700"
      },
      "message": "Report correct L2 cache size on ARM (Neoverse V1/V2) (#372)\n\nBy fetching the values from `/sys/devices/system/cpu/cpuN/cache/indexM/{size,level,type,...}`, that is populate via \nhttps://github.com/torvalds/linux/blob/master/arch/arm64/kernel/cacheinfo.c\nand\nhttps://github.com/torvalds/linux/blob/master/drivers/acpi/pptt.c\n\nFixes #369 \n\nTested on Arm Neoverse V1 and V2 EC2 instances. I reused the reproducer in #369 \n\n```\n#include \u003ccstddef\u003e\n#include \u003ccstdio\u003e\n#include \u003ccpuinfo.h\u003e\nsize_t l2_bytes() {\n  if (!cpuinfo_initialize()) return 0;\n  const cpuinfo_processor* p \u003d cpuinfo_get_current_processor();\n  if (!p || !p-\u003ecache.l2) return 0;\n  return p-\u003ecache.l2-\u003esize; // bytes\n}\nint main() { std::printf(\"%zu\\n\", l2_bytes()); }\n```\n\nThis now returns the expected results on Arm Neoverse V1 and V2\n___\n\nAdditionally, here is the output of `./cache-info`.\n1. On Neoverse-V1:\n```\nMax cache size (upper bound): 4194304 bytes\nL1 instruction cache: 64 x 64 KB, 4-way set associative (256 sets), 64 byte lines, shared by 1 processors\nL1 data cache: 64 x 64 KB, 4-way set associative (256 sets), 64 byte lines, shared by 1 processors\nL2 data cache: 64 x 1 MB (inclusive), 8-way set associative (2048 sets), 64 byte lines, shared by 1 processors\n```\n2. On Neoverse-V2:\n```\nMax cache size (upper bound): 4194304 bytes\nL1 instruction cache: 64 x 64 KB, 4-way set associative (256 sets), 64 byte lines, shared by 1 processors\nL1 data cache: 64 x 64 KB, 4-way set associative (256 sets), 64 byte lines, shared by 1 processors\nL2 data cache: 64 x 2 MB (inclusive), 8-way set associative (4096 sets), 64 byte lines, shared by 1 processors\n```"
    },
    {
      "commit": "7364b490b5f78d58efe23ea76e74210fd6c3c76f",
      "tree": "fd4b8ddff8ff284110131dffa892a24b36b00b2c",
      "parents": [
        "84818a41e074779dbb00521a4731d3e14160ff15"
      ],
      "author": {
        "name": "Alexander Shaposhnikov",
        "email": "alexander.v.shaposhnikov@gmail.com",
        "time": "Wed Feb 18 13:35:49 2026 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Feb 18 13:35:49 2026 -0800"
      },
      "message": "[cpuinfo] Add linux_ppc64le config (#373)\n\n"
    },
    {
      "commit": "84818a41e074779dbb00521a4731d3e14160ff15",
      "tree": "8f75f7ab8b7d0a9a2bd5571dc3b684824068d25a",
      "parents": [
        "f9a03241f8c3d4ed0c9728f5d70bff873d43d4e0"
      ],
      "author": {
        "name": "Redwanul Haque Sourave",
        "email": "redwanul.sourav@gmail.com",
        "time": "Fri Jan 30 12:45:09 2026 -0500"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Jan 30 09:45:09 2026 -0800"
      },
      "message": "Fix buffer size for reading processor core id (#352)\n\n"
    },
    {
      "commit": "f9a03241f8c3d4ed0c9728f5d70bff873d43d4e0",
      "tree": "a46fa1d5f805fc69d910992e36b098fa5bc57c2a",
      "parents": [
        "b8ecd2f4455f593be18326b89af19be3f9959293"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Tue Jan 27 12:07:08 2026 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jan 27 12:07:08 2026 -0800"
      },
      "message": "Detect AMD Family 26 extended models for cpuinfo_uarch_zen6 (#371)\n\n* Fix cpuinfo_x86_normalize_brand_string unannotated fall-through warning\n\n* Detect AMD Family 26 extended models for cpuinfo_uarch_zen6"
    },
    {
      "commit": "b8ecd2f4455f593be18326b89af19be3f9959293",
      "tree": "5734d479a1fee4683d0d0c5ee403bca0ccc0be21",
      "parents": [
        "e7b796ccd140df3264afd3f032debd51b4a776df"
      ],
      "author": {
        "name": "Edward Chen",
        "email": "18449977+edgchen1@users.noreply.github.com",
        "time": "Tue Jan 27 07:36:42 2026 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jan 27 07:36:42 2026 -0800"
      },
      "message": "Add fallback for ARM Windows fp16 detection. (#348)\n\nBackground: On a Windows ARM system, I observed that `cpuinfo_has_arm_fp16_arith()` started to return false after upgrading to a more recent cpuinfo version.\n\nIn #333, the initialization of `cpuinfo_isa.fp16arith` was updated to use `IsProcessorFeaturePresent(PF_ARM_V82_FP16_INSTRUCTIONS_AVAILABLE)`. I suspect that this is not supported on older Windows versions.\n\nThis change adds a fallback path to set `cpuinfo_isa.fp16arith` the old way.\nhttps://github.com/pytorch/cpuinfo/blob/d3a86a813e2bb49d1eb5841ec12e2b135867ab98/src/arm/windows/init.c#L205-L208"
    },
    {
      "commit": "e7b796ccd140df3264afd3f032debd51b4a776df",
      "tree": "66919038895214c8affa3da23c8a818ac9b87c4e",
      "parents": [
        "1ce292fdeceb9a4dee168074a84d629a1f57e2fd"
      ],
      "author": {
        "name": "Alfredo Tupone",
        "email": "tupone@gentoo.org",
        "time": "Mon Jan 26 23:47:25 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Jan 26 14:47:25 2026 -0800"
      },
      "message": "cmake will require at least 3.10 compatibility (#319)\n\n"
    },
    {
      "commit": "1ce292fdeceb9a4dee168074a84d629a1f57e2fd",
      "tree": "3c5701bc9e364ff2dce86ab05a7295f2c32b28d6",
      "parents": [
        "c4b4f4bf08c0cf486fc3111d0244ebf2a48ad01b"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Mon Jan 26 14:25:29 2026 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Jan 26 14:25:29 2026 -0800"
      },
      "message": "Add ARM Lumex C1 uarch (#370)\n\n* Add ARM Lumex C1 uarch\n\nSoC name: MediaTek MT6993\nMicroarchitectures:\n\t1x Lumex-C1-Ultra\n\t3x Lumex-C1-Premium\n\t4x Lumex-C1-Pro\nCores:\n\t0: 1 processor (0), ARM Lumex-C1-Ultra\n\t1: 1 processor (1), ARM Lumex-C1-Premium\n\t2: 1 processor (2), ARM Lumex-C1-Premium\n\t3: 1 processor (3), ARM Lumex-C1-Premium\n\t4: 1 processor (4), ARM Lumex-C1-Pro\n\t5: 1 processor (5), ARM Lumex-C1-Pro\n\t6: 1 processor (6), ARM Lumex-C1-Pro\n\t7: 1 processor (7), ARM Lumex-C1-Pro\nClusters:\n\t0: 1 processor (0),\t0: 1 core (0), ARM Lumex-C1-Ultra\n\t1: 3 processors (1-3),\t1: 3 cores (1-3), ARM Lumex-C1-Premium\n\t2: 4 processors (4-7),\t2: 4 cores (4-7), ARM Lumex-C1-Pro\nLogical processors (System ID):\n\t0 (7)\n\t1 (4)\n\t2 (5)\n\t3 (6)\n\t4 (0)\n\t5 (1)\n\t6 (2)\n\t7 (3)\n\nPart numbers documented here:\nhttps://en.wikichip.org/wiki/arm_holdings/cortex\n\n* Add ARM Lumex C1 uarch\n\nSoC name: MediaTek MT6993\nMicroarchitectures:\n\t1x Lumex-C1-Ultra\n\t3x Lumex-C1-Premium\n\t4x Lumex-C1-Pro\nCores:\n\t0: 1 processor (0), ARM Lumex-C1-Ultra\n\t1: 1 processor (1), ARM Lumex-C1-Premium\n\t2: 1 processor (2), ARM Lumex-C1-Premium\n\t3: 1 processor (3), ARM Lumex-C1-Premium\n\t4: 1 processor (4), ARM Lumex-C1-Pro\n\t5: 1 processor (5), ARM Lumex-C1-Pro\n\t6: 1 processor (6), ARM Lumex-C1-Pro\n\t7: 1 processor (7), ARM Lumex-C1-Pro\nClusters:\n\t0: 1 processor (0),\t0: 1 core (0), ARM Lumex-C1-Ultra\n\t1: 3 processors (1-3),\t1: 3 cores (1-3), ARM Lumex-C1-Premium\n\t2: 4 processors (4-7),\t2: 4 cores (4-7), ARM Lumex-C1-Pro\nLogical processors (System ID):\n\t0 (7)\n\t1 (4)\n\t2 (5)\n\t3 (6)\n\t4 (0)\n\t5 (1)\n\t6 (2)\n\t7 (3)\n\nPart numbers documented here:\nhttps://en.wikichip.org/wiki/arm_holdings/cortex\n\n* remove dot and fp16\n\n* fix mock naming"
    },
    {
      "commit": "c4b4f4bf08c0cf486fc3111d0244ebf2a48ad01b",
      "tree": "7e063fe8afdc3404bd20ed7cd91822dae3e3093b",
      "parents": [
        "0fea7f5f88243ee354df0e0082b5f27d13fc9551"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Wed Jan 14 09:57:04 2026 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Jan 14 09:57:04 2026 -0800"
      },
      "message": "Add ARM Cortex A520 little core uarch (#367)\n\nExample Samsung S24 sm-s921u\n\nSoC name: Unknown\nMicroarchitectures:\n\t1x Cortex-X4\n\t5x Cortex-A720\n\t2x Cortex-A520\nCores:\n\t0: 1 processor (0), ARM Cortex-X4\n\t1: 1 processor (1), ARM Cortex-A720\n\t2: 1 processor (2), ARM Cortex-A720\n\t3: 1 processor (3), ARM Cortex-A720\n\t4: 1 processor (4), ARM Cortex-A720\n\t5: 1 processor (5), ARM Cortex-A720\n\t6: 1 processor (6), ARM Cortex-A520\n\t7: 1 processor (7), ARM Cortex-A520\nClusters:\n\t0: 1 processor (0),\t0: 1 core (0), ARM Cortex-X4\n\t1: 3 processors (1-3),\t1: 3 cores (1-3), ARM Cortex-A720\n\t2: 2 processors (4-5),\t2: 2 cores (4-5), ARM Cortex-A720\n\t3: 2 processors (6-7),\t3: 2 cores (6-7), ARM Cortex-A520\nLogical processors (System ID):\n\t0 (7)\n\t1 (2)\n\t2 (3)\n\t3 (4)\n\t4 (5)\n\t5 (6)\n\t6 (0)\n\t7 (1)"
    },
    {
      "commit": "0fea7f5f88243ee354df0e0082b5f27d13fc9551",
      "tree": "0f6e76fca4526f1cc23b0d76ef22c3e2592d591b",
      "parents": [
        "315f594e7f4d4c35f357289ae5304af8d38d9fc3"
      ],
      "author": {
        "name": "Alfredo Tupone",
        "email": "tupone@gentoo.org",
        "time": "Wed Jan 07 02:51:28 2026 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jan 06 17:51:28 2026 -0800"
      },
      "message": "Upgrade C++ standard from CXX14 to CXX17 (#355)\n\nIt is required by gtest (newest version)"
    },
    {
      "commit": "315f594e7f4d4c35f357289ae5304af8d38d9fc3",
      "tree": "c102474e5cb9bf9cfd86e6b217484ead709dfcbb",
      "parents": [
        "2846bafc7c785f274e6e5defb26dabdb05fac148"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Tue Jan 06 16:43:55 2026 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jan 06 16:43:55 2026 -0800"
      },
      "message": "Add Apple M3, M5 and A19 uarch support (#366)\n\n* Fix cpuinfo_x86_normalize_brand_string unannotated fall-through warning\n\n* Add Apple M3, M5 and A19 uarch support\n\n* fix duplicate header uarch"
    },
    {
      "commit": "2846bafc7c785f274e6e5defb26dabdb05fac148",
      "tree": "c0cf671fa8a7cea710891b52fcfbb8a1e32e4f01",
      "parents": [
        "b3b25967b5b80406304d575321e572c5f9e5e3c4"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Mon Jan 05 13:38:16 2026 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Jan 05 13:38:16 2026 -0800"
      },
      "message": "Add Apple SME isa and vector length detects (#363)\n\n* Apple SME 2p1 and vector length detect\n\n- similar to windows/init.c but use sysinfo\n  hw.optional.arm.FEAT_SME2p1\n  hw.optional.arm.sme_max_svl_b\n\nTested on Macbook Pro with M4\n\n* Add Apple SME isa and vector length detects\n\n* clang format applied"
    },
    {
      "commit": "b3b25967b5b80406304d575321e572c5f9e5e3c4",
      "tree": "afd1e9430dfbee596c1fe90bc211ef665caf2cb5",
      "parents": [
        "e3f82e8f64f81e7a6987485f6795af6bd5f639e4"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Tue Dec 30 12:40:41 2025 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Dec 30 15:40:41 2025 -0500"
      },
      "message": "Apple M4 uarch detect (#362)\n\n* Apple M4 uarch detect\n\n- Detect M4 and M4 Pro p-core and e-core\n\n* clang-format applied\n\n* fix clang-format"
    },
    {
      "commit": "e3f82e8f64f81e7a6987485f6795af6bd5f639e4",
      "tree": "c341d4d0872404afe6d94bb77a712c2d959a7672",
      "parents": [
        "38b1af9c619b4546bcfa3e66398a8cd8aa94c25a"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Tue Dec 30 10:13:14 2025 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Dec 30 10:13:14 2025 -0800"
      },
      "message": "Blizzard duplicate return removed (#360)\n\n* Fix duplicate blizzard return\n\n- Fix a copy/paste error in uarch detect for Apple blizzard/avalanche uarch\n- Add comment for A14/M1 and A15/M2 cpufamily\n\n* Fix duplicate return for Apple M2 detect uarch\n\n- Add comments for M1 and M2\n\n* Fix comments for CPUFAMILY to match what the case/return are for"
    },
    {
      "commit": "38b1af9c619b4546bcfa3e66398a8cd8aa94c25a",
      "tree": "b27aebfe81fe5f1e90b8edd5429dadc664656e11",
      "parents": [
        "002c213d16b9dc3ddfb7ab388b13183df0e81c49"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Sun Dec 28 17:27:38 2025 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sun Dec 28 20:27:38 2025 -0500"
      },
      "message": "Fix typo cpofreq (#358)\n\n- Replace cpofreq with cpufreq"
    },
    {
      "commit": "002c213d16b9dc3ddfb7ab388b13183df0e81c49",
      "tree": "682a537d3045465f9993990d0a1e597a09550c4d",
      "parents": [
        "e8b4def4e6e67bc78d1882b2d709d6f6e13097ee"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Sun Dec 28 17:27:30 2025 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Sun Dec 28 20:27:30 2025 -0500"
      },
      "message": "Add Oryon V3 uarch detect for pytorch/cpuinfo (#357)\n\nTested with One Plus 15\ndevice:\n - ro.product.device: \"OP611FL1\"\n - ro.product.name: \"CPH2747\"\n - ro.product.brand: \"OnePlus\"\n - ro.product.manufacturer: \"OnePlus\"\n - ro.product.model: \"CPH2747\"\n - ro.soc.manufacturer: \"QTI\"\n - ro.soc.model: \"SM8850\"\nprocessor:\n - chipset: \"Unknown\"\n - core: \"Oryon V3\"\n - MIDR: 0x512F0021"
    },
    {
      "commit": "e8b4def4e6e67bc78d1882b2d709d6f6e13097ee",
      "tree": "16009fd34c313340f5f2d0971d7fbbef8f3234a9",
      "parents": [
        "ff24ffee8340fbd9001cce6a9ef41cdd16aa2bd3"
      ],
      "author": {
        "name": "Vertexwahn",
        "email": "julian.amann@tum.de",
        "time": "Thu Dec 11 23:49:18 2025 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Dec 11 14:49:18 2025 -0800"
      },
      "message": "Bazel support: Sync with current status of BCR and apply fixes for Bazel 9 (#345)\n\n"
    },
    {
      "commit": "ff24ffee8340fbd9001cce6a9ef41cdd16aa2bd3",
      "tree": "97c053a1e9792c7626c775501fbb2382f011813f",
      "parents": [
        "161a9ec374884f4b3e85725cb22e05f9458fdc93"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Tue Dec 09 15:14:26 2025 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Dec 09 15:14:26 2025 -0800"
      },
      "message": "add Cortex A78C midr detect (#354)\n\n* add Cortex A78C midr detect\n\n- Detect ARM part 0xD4B (Cortex-A78C) and return cpuinfo_uarch_cortex_a78\n\n* fix clang format error for tabs"
    },
    {
      "commit": "161a9ec374884f4b3e85725cb22e05f9458fdc93",
      "tree": "d990ead5980dafd52612acd8fbebfb78e7725f8e",
      "parents": [
        "403d652dca4c1046e8145950b1c0997a9f748b57"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Wed Nov 19 09:38:05 2025 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Nov 19 09:38:05 2025 -0800"
      },
      "message": "Fix ARM build -Wlogical-op-parentheses for unisoc detect (#347)\n\n- Add parentheses around the \u0027\u0026\u0026\u0027 expression"
    },
    {
      "commit": "403d652dca4c1046e8145950b1c0997a9f748b57",
      "tree": "5dcedc8305dd86a04a0a4e2d16144762e0643965",
      "parents": [
        "f858c30bcb16f8effd5ff46996f0514539e17abc"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Tue Nov 18 11:51:03 2025 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Nov 18 11:51:03 2025 -0800"
      },
      "message": "Fix overread asan error of brand_string (#344)\n\n- Brand string is 3 rows of cpuid which is 48 bytes.\n- Allow for CPUINFO_PACKAGE_NAME_MAX to be 64 bytes for benefit of Oryon"
    },
    {
      "commit": "f858c30bcb16f8effd5ff46996f0514539e17abc",
      "tree": "996d51dc31c2268286a504ab29385e6b9e9ada4b",
      "parents": [
        "1d1dbb4cd620d9d0026872da2d85418263179f20"
      ],
      "author": {
        "name": "Gregory Comer",
        "email": "gregoryjcomer@gmail.com",
        "time": "Fri Nov 14 12:00:48 2025 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Nov 14 11:00:48 2025 -0800"
      },
      "message": "Improve unisoc vendor detection, re-enable neon dot on unknown chipsets (#342)\n\n"
    },
    {
      "commit": "1d1dbb4cd620d9d0026872da2d85418263179f20",
      "tree": "7fe6905c070abf50a81ab58612066642e3d247e1",
      "parents": [
        "952a519883239e086e9a78299a6efb8f6614022f"
      ],
      "author": {
        "name": "Bin Bao",
        "email": "binbao@meta.com",
        "time": "Fri Nov 14 10:47:39 2025 -0500"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Nov 14 10:47:39 2025 -0500"
      },
      "message": "Fix missing AMD cache info when running on a VM (#338)\n\nSummary: Fix https://github.com/pytorch/pytorch/issues/138718. This PR is verified to fix the issue and will be guarded on the PyTorch CI.\n\nWhen running on a VM, the AMD TopologyExtensions bit can be cleared, but that shouldn\u0027t stop us from reading the cache information."
    },
    {
      "commit": "952a519883239e086e9a78299a6efb8f6614022f",
      "tree": "7f41d52e6d74c277556ba8244fb1c73cf5d12802",
      "parents": [
        "f1f25cbe838f514c6bdbe762574e0e14538bdba8"
      ],
      "author": {
        "name": "Jimothy Branson",
        "email": "jimothy@bouncingsheep.org",
        "time": "Thu Nov 13 15:46:12 2025 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Nov 13 15:46:12 2025 -0800"
      },
      "message": "Allow for longer CPU names (#320)\n\nIncrease CPUINFO_PACKAGE_NAME_MAX to allow for longer strings,\nsuch as \"Snapdragon® X Elite - X1E78100 - Qualcomm® Oryon™ CPU\",\nwhich is 53 characters and WideCharToMultiByte converts to a\n57 byte string."
    },
    {
      "commit": "f1f25cbe838f514c6bdbe762574e0e14538bdba8",
      "tree": "0aa4eb896f4a9f9fe6aa5a7232a459bdffa3fb4c",
      "parents": [
        "735b2ac12f69aaf408784ed251b9bd6319cfc62d"
      ],
      "author": {
        "name": "Byoungchan Lee",
        "email": "byoungchan.lee@gmx.com",
        "time": "Fri Nov 14 08:29:49 2025 +0900"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Nov 13 15:29:49 2025 -0800"
      },
      "message": "Fix Unisoc T618 Chipset Detection (#307)\n\nAdd support for uppercase \"UNISOC T\" prefix in `match_t` function to handle\ndevices like Samsung Galaxy Tab A8 (SM-X205N) that report \"UNISOC T618\"\ninstead of the expected mixed-case \"Unisoc T618\".\n\nThe function now explicitly matches both variants:\n- \"Unisoc T\" (mixed case, existing)\n- \"UNISOC T\" (uppercase, new)\n\nThis ensures proper chipset vendor detection on affected Samsung devices\nwhere the uppercase variant caused match failures.\n\nIncludes test case for \"UNISOC T618\" detection."
    },
    {
      "commit": "735b2ac12f69aaf408784ed251b9bd6319cfc62d",
      "tree": "62ac5ec4683862470031e2edbcd40253ffddb6f4",
      "parents": [
        "9a05d3998a5197a31f24b317d86178e81c821fea"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Tue Nov 11 19:16:48 2025 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Nov 11 19:16:48 2025 -0800"
      },
      "message": "Rollback Ignore disabled hyperthreads (#340)\n\nFixes #339"
    },
    {
      "commit": "9a05d3998a5197a31f24b317d86178e81c821fea",
      "tree": "42d81f4420fd656c866bdb40eff4960d95efb568",
      "parents": [
        "f7b233b2755235c97ef1c2e650820ece1b2869a4"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Mon Nov 10 16:09:48 2025 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Nov 10 16:09:48 2025 -0800"
      },
      "message": "Fix windows arm64 build warnings (treated as error) (#332)\n\n- init.c:24:22: error: unused variable \u0027vendor\u0027\n- init-by-logical-sys-info.c:845:11: error: unused variable \u0027converted_chars\u0027\n- init-by-logical-sys-info.c:754:18: error: comparison of array \u0027chip_info-\u003euarchs\u0027 equal to a null pointer"
    },
    {
      "commit": "f7b233b2755235c97ef1c2e650820ece1b2869a4",
      "tree": "42d81f4420fd656c866bdb40eff4960d95efb568",
      "parents": [
        "d3a86a813e2bb49d1eb5841ec12e2b135867ab98"
      ],
      "author": {
        "name": "Anthony Shaw",
        "email": "anthony.p.shaw@gmail.com",
        "time": "Mon Nov 10 11:00:20 2025 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Nov 10 11:00:20 2025 -0800"
      },
      "message": "Implement missing ISA feature flags for ARM64/Windows (#333)\n\n* implement SVE/SVE2 and list out the missing feature flags\n\n* Implement most of the ISA feature flags\n\n* Update notes\n\n* Add more flags to isa info\n\n* Additional flags\n\n* Add conservative values for SVE\n\n* clang formatting\n\n* format headers"
    },
    {
      "commit": "d3a86a813e2bb49d1eb5841ec12e2b135867ab98",
      "tree": "1c3a49600adb0216748bb987e83b4e6cc7499b0d",
      "parents": [
        "f01ce870215f9e5d4c32006796994469c5334fd7"
      ],
      "author": {
        "name": "davidbienvenu",
        "email": "bienvenu@davidbienvenu.org",
        "time": "Fri Nov 07 16:47:29 2025 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Nov 07 16:47:29 2025 -0800"
      },
      "message": "Fix compile warnings in arm/windows (#323)\n\n* Update init-by-logical-sys-info.c\n\nFix compiler warnings\n\n* Update init.c\n\nfix compiler warning\n\n* reformat init.c\n\nattempt to fix clang format errors\n\n* Update init-by-logical-sys-info.c\n\n* Update init-by-logical-sys-info.c\n\nattempt to fix clang format errors"
    },
    {
      "commit": "f01ce870215f9e5d4c32006796994469c5334fd7",
      "tree": "e66676eb1ecce5588907b06e79b9f7e847cacc85",
      "parents": [
        "10a95e87a10ae7a7b108bd25ff8a606e53d5af74"
      ],
      "author": {
        "name": "Anthony Shaw",
        "email": "anthony.p.shaw@gmail.com",
        "time": "Thu Nov 06 15:01:15 2025 -0800"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Nov 06 15:01:15 2025 -0800"
      },
      "message": "Add missing ARM-Cortex uarchs (#335)\n\n* Add A720, 725 and X4 identifiers\n\n* Add the X4 successor as well\n\n* Add enumerators"
    },
    {
      "commit": "10a95e87a10ae7a7b108bd25ff8a606e53d5af74",
      "tree": "469290f884547169572237b47240962cdf8a32fd",
      "parents": [
        "0d5985d60de2512a2130eede846fc755cb222dc8"
      ],
      "author": {
        "name": "Gregory Comer",
        "email": "gregoryjcomer@gmail.com",
        "time": "Thu Nov 06 15:52:45 2025 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Nov 06 14:52:45 2025 -0800"
      },
      "message": "Update googletest and googlebench deps (#337)\n\n"
    },
    {
      "commit": "0d5985d60de2512a2130eede846fc755cb222dc8",
      "tree": "661a39dce3fa3d9e8539b1aafda51eea983655a9",
      "parents": [
        "37b484a5ce3f73f2f37977b86a74698e612936b0"
      ],
      "author": {
        "name": "Alexander Grund",
        "email": "Flamefire@users.noreply.github.com",
        "time": "Wed Nov 05 19:10:34 2025 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Nov 05 10:10:34 2025 -0800"
      },
      "message": "Ignore disabled hyperthreads (#291)\n\nOnly consider processors with an APIC_ID as valid.\n\nFor processors with Hyperthreads but disabled SMT the APIC_ID will never be set and stays zero for the \"disabled\" threads.\nAmong other inconsistencies this causes the first active thread/processor to be considered \"the same\" as all the disabled ones as they share `APIC_ID\u003d0`\n\nIt also doesn\u0027t make sense to make and decisions based on the APIC_ID if we don\u0027t have any information on it."
    },
    {
      "commit": "37b484a5ce3f73f2f37977b86a74698e612936b0",
      "tree": "947877297ee3cc8b5e3063bbc853113c87918cac",
      "parents": [
        "877328f188a3c7d1fa855871a278eb48d530c4c0"
      ],
      "author": {
        "name": "Byoungchan Lee",
        "email": "byoungchan.lee@gmx.com",
        "time": "Thu Nov 06 02:51:15 2025 +0900"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Nov 05 09:51:15 2025 -0800"
      },
      "message": "Fix Samsung Exynos 2000 series chipset detection (#309)\n\n* Fix Samsung Exynos 2000 series chipset detection\n\nAdd support for \"s5e\" prefix pattern matching in `ro.product.board`\nproperty to correctly identify Samsung Exynos 2000 series chipsets.\nThese chipsets use generic identifiers (e.g., `s5e9925`) instead of\nmarketing names like \"Exynos 2200\", but can still be identified and\nmapped to their corresponding models using external resources.\n\nImplements `match_s5e()` function to parse the s5e\u003c4-digit-model\u003e format.\n\n* Clean up whitespace in chipset.c\n\n---------\n\nCo-authored-by: Byoungchan Lee \u003cdaniel.l@hpcnt.com\u003e"
    },
    {
      "commit": "877328f188a3c7d1fa855871a278eb48d530c4c0",
      "tree": "f67e0d068d30d0cd26f9efe8cbdbfe3af0d558a4",
      "parents": [
        "8a9210069b5a37dd89ed118a783945502a30a4ae"
      ],
      "author": {
        "name": "Nikita Shulga",
        "email": "2453524+malfet@users.noreply.github.com",
        "time": "Fri Sep 05 15:55:09 2025 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Sep 05 15:55:09 2025 -0700"
      },
      "message": "Update CMake minimum requirements to 3.18 (#330)\n\nAs CMake 3.5 support has been removed in cmake-4.x"
    },
    {
      "commit": "8a9210069b5a37dd89ed118a783945502a30a4ae",
      "tree": "070437e7039f01f4feb7b7713c0ce6f38bbd9aef",
      "parents": [
        "e414c0446436ed34151de3158d18f8ae32e55d03"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Fri Aug 08 07:49:05 2025 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Aug 08 07:49:05 2025 -0700"
      },
      "message": "Add Intel Tremont uarch (#305)\n\nTremont is the e-core (10nm Atom) in Jasperlake,Elkhart Lake, Jacobville and Lakefield\n\nThe family/model are supported by linux\nhttps://github.com/torvalds/linux/blob/master/arch/x86/include/asm/intel-family.h#L176-L178\nDocumented on wikichip\nhttps://en.wikichip.org/wiki/intel/cpuid#Small_Cores\nSupported by Intel sde -tnt\n\nReordered e-core uarch by generation\ndarkmont is an e-core used in both hybrid laptops and server, so move to ecore\ngracemont model 0xBE is confirmed for Alderlake N, so redocument support\n\nPredecessor Goldmont Plus\nSuccessor Gracemont"
    },
    {
      "commit": "e414c0446436ed34151de3158d18f8ae32e55d03",
      "tree": "cd82ad00b8d244cb36675b399c0a8137e9885d8a",
      "parents": [
        "33ed0be77d7767d0e2010e2c3cf972ef36c7c307"
      ],
      "author": {
        "name": "Yuri Khrustalev",
        "email": "ykhrustalev@users.noreply.github.com",
        "time": "Tue Aug 05 13:27:21 2025 -0400"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Aug 05 10:27:21 2025 -0700"
      },
      "message": "Add aarch64 to the windows arm list (#311)\n\n"
    },
    {
      "commit": "33ed0be77d7767d0e2010e2c3cf972ef36c7c307",
      "tree": "05920c6ced80e19f2f1ead12b06f77f6f1219b3c",
      "parents": [
        "e4cadd02a8b386c38b84f0a19eddacec3f433baa"
      ],
      "author": {
        "name": "Conan Truong",
        "email": "Conarnar@users.noreply.github.com",
        "time": "Thu Jul 24 10:36:24 2025 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Jul 24 10:36:24 2025 -0700"
      },
      "message": "Added Emscripten to list of supported systems in CMakeLists.txt (#310)\n\n"
    },
    {
      "commit": "e4cadd02a8b386c38b84f0a19eddacec3f433baa",
      "tree": "15eb0f4f823bad997f22133b2f3753ddc5ac7d69",
      "parents": [
        "d7427551d6531037da216d20cd36feb19ed4905f"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Wed Jun 25 21:30:09 2025 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Wed Jun 25 23:30:09 2025 -0500"
      },
      "message": "Add Intel Gracemont uarch (#303)\n\n"
    },
    {
      "commit": "d7427551d6531037da216d20cd36feb19ed4905f",
      "tree": "2921254f801bd706116cb36996c87a42818d255e",
      "parents": [
        "bac1e85899d8a5663d56a1924aabcaee3de6d8d0"
      ],
      "author": {
        "name": "Richard Winterton",
        "email": "rrwinterton@gmail.com",
        "time": "Tue Jun 10 08:00:39 2025 -0600"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jun 10 07:00:39 2025 -0700"
      },
      "message": "Added  Willow Cove (#301)\n\n* Added  Willow Cove\n\n* Update uarch.c\n\nremoved incomplete modification in naming\n\n* Update cpu-info.c\n\nmoved willow_cove up two lines to be next to sunny_cove\n\n* Update cpuinfo.h\n\nmade the enum consistent with uarch naming following sunny_cove\n\n* Update cpuinfo.h\n\nupdated comments to refer to intel microarchitecture per request to keep consistent."
    },
    {
      "commit": "bac1e85899d8a5663d56a1924aabcaee3de6d8d0",
      "tree": "2ae375b51fd33925ae9e0e3f4651f5f92071875d",
      "parents": [
        "6c9eb84ba310f237cea13c478be50102e1128e9b"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Mon Jun 09 14:42:28 2025 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Jun 09 14:42:28 2025 -0700"
      },
      "message": "Add Intel Darkmont uarch (#298)\n\n* Add Intel Darkmont uarch\n\n- Darkmont is the uarch used in Clearwater Forest\n\n* Add Intel Darkmont uarch\n\n- Darkmont is the uarch used in Clearwater Forest\n\n* Add Intel Darkmont uarch\n\n- Darkmont is the uarch used in Clearwater Forest"
    },
    {
      "commit": "6c9eb84ba310f237cea13c478be50102e1128e9b",
      "tree": "1692e327d01d6ee93d179b94d14100e6dbcb3420",
      "parents": [
        "88b498b9d73acd0b8edb1ec1c31a8fc1cefaf2da"
      ],
      "author": {
        "name": "Gregory Comer",
        "email": "gregoryjcomer@gmail.com",
        "time": "Tue Jun 03 09:50:51 2025 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Jun 03 09:50:51 2025 -0700"
      },
      "message": "Disable neon dot on unknown chipsets on aarch32 (#300)\n\n"
    },
    {
      "commit": "88b498b9d73acd0b8edb1ec1c31a8fc1cefaf2da",
      "tree": "5bc2dbc0ed7ead85b497d5516879d8843989266d",
      "parents": [
        "c61fe919607bbc534d7a5a5707bdd7041e72c5ff"
      ],
      "author": {
        "name": "fbarchard",
        "email": "fbarchard@google.com",
        "time": "Mon Jun 02 09:54:50 2025 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Jun 02 09:54:50 2025 -0700"
      },
      "message": "Add Intel Crestmont uarch (#299)\n\nDarkmont is the uarch used in Sierra Forest\nTested:\nmake cpu-info\nsde -srf -- ./cpu-info"
    },
    {
      "commit": "c61fe919607bbc534d7a5a5707bdd7041e72c5ff",
      "tree": "e7b5ddfdaa1868cee79cecddc5efa15f5e96219a",
      "parents": [
        "de0ce7c7251372892e53ce9bc891750d2c9a4fd8"
      ],
      "author": {
        "name": "Kimish Patel",
        "email": "kimishpatel@fb.com",
        "time": "Thu May 29 13:26:30 2025 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu May 29 13:26:30 2025 -0700"
      },
      "message": "update apple soc info (#290)\n\n* [WIP] update apple soc info\n\nSummary:\nAdded support for A16, A17, A18, A18 pro. Reg values are found from ncnn and needs validation.\n\nAdditional source\nConstants are taken from https://github.com/apple-oss-distributions/xnu/blob/e3723e1f17661b24996789d8afc084c0c3303b26/osfmk/mach/machine.h#L449\n\nTest Plan:\n\nReviewers:\n\nSubscribers:\n\nTasks:\n\nTags:\n\n[ghstack-poisoned]\n\n* Update on \"[WIP] update apple soc info\"\n\n\nSummary:\n\nTest Plan:\n\nReviewers:\n\nSubscribers:\n\nTasks:\n\nTags:\n\n[ghstack-poisoned]\n\n* Update on \"[WIP] update apple soc info\"\n\n\nSummary:\n\nTest Plan:\n\nReviewers:\n\nSubscribers:\n\nTasks:\n\nTags:\n\n[ghstack-poisoned]\n\n* Update on \"[WIP] update apple soc info\"\n\n\nSummary:\n\nTest Plan:\n\nReviewers:\n\nSubscribers:\n\nTasks:\n\nTags:\n\n[ghstack-poisoned]\n\n* Update on \"update apple soc info\"\n\n\nSummary:\n\nTest Plan:\n\nReviewers:\n\nSubscribers:\n\nTasks:\n\nTags:\n\n[ghstack-poisoned]"
    },
    {
      "commit": "de0ce7c7251372892e53ce9bc891750d2c9a4fd8",
      "tree": "f717b193c93a799c8dab375041f71b2924d17a53",
      "parents": [
        "957b852c5ca521ecb848527a5dc3cc55274aa496"
      ],
      "author": {
        "name": "enh-google",
        "email": "enh@google.com",
        "time": "Thu May 22 20:45:30 2025 -0400"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu May 22 17:45:30 2025 -0700"
      },
      "message": "riscv-hw.c: match kernel type in syscall(). (#292)\n\nsyscall() doesn\u0027t care about these types at all, and the kernel uses cpu_set_t, so we\u0027re better off just removing the cast entirely."
    },
    {
      "commit": "957b852c5ca521ecb848527a5dc3cc55274aa496",
      "tree": "ca17fc4c732f43f0d97e488338e19e27cbe397d6",
      "parents": [
        "905918fe9c2a335e99a7092dc028dee927a103a0"
      ],
      "author": {
        "name": "Nikita Shulga",
        "email": "2453524+malfet@users.noreply.github.com",
        "time": "Thu May 22 16:06:17 2025 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu May 22 16:06:17 2025 -0700"
      },
      "message": "[CI] Fix riscv64-in-qemu build (#295)\n\nTried a few things, but looks like all one needs to do is to add `--platform linux/riscv64` flag to `docker run` command"
    },
    {
      "commit": "905918fe9c2a335e99a7092dc028dee927a103a0",
      "tree": "e6b8be40051710ca6daab74ba39da8b1edd69da4",
      "parents": [
        "4ab5a589708c6aa9ceb561fb2a6506374107da21"
      ],
      "author": {
        "name": "Nikita Shulga",
        "email": "2453524+malfet@users.noreply.github.com",
        "time": "Thu May 22 14:47:34 2025 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu May 22 14:47:34 2025 -0700"
      },
      "message": "Revert \"cmake: rename duplicate names\" (#296)\n\nReverts pytorch/cpuinfo#284 as it broke Bazel builds"
    },
    {
      "commit": "4ab5a589708c6aa9ceb561fb2a6506374107da21",
      "tree": "1cd0b54f3d36f19a6844ddbd6db43deeb99bbdad",
      "parents": [
        "1961df8470beebffa5d82b99b9b9e430436b8d34"
      ],
      "author": {
        "name": "Nikita Shulga",
        "email": "nshulga@meta.com",
        "time": "Thu May 22 14:08:46 2025 -0700"
      },
      "committer": {
        "name": "Nikita Shulga",
        "email": "nshulga@meta.com",
        "time": "Thu May 22 14:08:46 2025 -0700"
      },
      "message": "[CI] Update checkout action to v4\n"
    },
    {
      "commit": "1961df8470beebffa5d82b99b9b9e430436b8d34",
      "tree": "208e99d85f47c7298d7619555d3b20200ea9b090",
      "parents": [
        "48a5942c92a3d0590873f1fdfbe1ce3a66412afe"
      ],
      "author": {
        "name": "Nikita Shulga",
        "email": "2453524+malfet@users.noreply.github.com",
        "time": "Thu May 22 08:27:53 2025 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu May 22 08:27:53 2025 -0700"
      },
      "message": "Fix clang format (#294)\n\nRegressions introduced by https://github.com/pytorch/cpuinfo/pull/275 and https://github.com/pytorch/cpuinfo/pull/287"
    },
    {
      "commit": "48a5942c92a3d0590873f1fdfbe1ce3a66412afe",
      "tree": "ae00736fc1aaf34057ee88fe2ae580b6ce5c52f5",
      "parents": [
        "b728a23da00ff511275eeaec0122bf28db00fd59",
        "a8d118ce2824f33c50311cf163c1eb586a8046a5"
      ],
      "author": {
        "name": "Digant Desai",
        "email": "digantdesai@meta.com",
        "time": "Thu May 22 00:18:27 2025 -0500"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu May 22 00:18:27 2025 -0500"
      },
      "message": "Merge pull request #284 from 1480c1/xcode\n\n"
    },
    {
      "commit": "b728a23da00ff511275eeaec0122bf28db00fd59",
      "tree": "81ffe159c81447bdbcc32bff5f126d890fc4cd51",
      "parents": [
        "39ea79a3c132f4e678695c579ea9353d2bd29968",
        "2cd6d6ddb6903a1df42a477940cf306ee9225df8"
      ],
      "author": {
        "name": "Digant Desai",
        "email": "digantdesai@meta.com",
        "time": "Thu May 22 00:17:38 2025 -0500"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu May 22 00:17:38 2025 -0500"
      },
      "message": "Merge pull request #287 from fbarchard/smelen\n\n"
    },
    {
      "commit": "2cd6d6ddb6903a1df42a477940cf306ee9225df8",
      "tree": "81ffe159c81447bdbcc32bff5f126d890fc4cd51",
      "parents": [
        "077e89f9b26d2204e7714613b7a42ef21f5d4dc3",
        "81bc2e04824011c137785f19e80a58cd2a801919"
      ],
      "author": {
        "name": "Frank Barchard",
        "email": "fbarchard@google.com",
        "time": "Fri Mar 28 13:13:52 2025 -0700"
      },
      "committer": {
        "name": "Frank Barchard",
        "email": "fbarchard@google.com",
        "time": "Fri Mar 28 13:13:52 2025 -0700"
      },
      "message": "Merge remote-tracking branch \u0027refs/remotes/origin/smelen\u0027 into smelen\n"
    },
    {
      "commit": "077e89f9b26d2204e7714613b7a42ef21f5d4dc3",
      "tree": "81ffe159c81447bdbcc32bff5f126d890fc4cd51",
      "parents": [
        "39ea79a3c132f4e678695c579ea9353d2bd29968"
      ],
      "author": {
        "name": "Frank Barchard",
        "email": "fbarchard@google.com",
        "time": "Thu Mar 27 17:35:14 2025 -0700"
      },
      "committer": {
        "name": "Frank Barchard",
        "email": "fbarchard@google.com",
        "time": "Fri Mar 28 12:55:22 2025 -0700"
      },
      "message": "Add SME vector length detect\n\n- cpuinfo_get_max_arm_sme_length() returns svl vector length in bits\n- Display length of SME vectors in isa-tool\n- Upgrade cmake-linux-riscv64 ubuntu-22.04 runners to ubuntu-24.04\n\nSME may be enabled on cpus that do not have SVE\n"
    },
    {
      "commit": "81bc2e04824011c137785f19e80a58cd2a801919",
      "tree": "f6ee41a31576d9c93668dfb92c8af9b9cb6eab98",
      "parents": [
        "39ea79a3c132f4e678695c579ea9353d2bd29968"
      ],
      "author": {
        "name": "Frank Barchard",
        "email": "fbarchard@google.com",
        "time": "Thu Mar 27 17:35:14 2025 -0700"
      },
      "committer": {
        "name": "Frank Barchard",
        "email": "fbarchard@google.com",
        "time": "Thu Mar 27 17:35:14 2025 -0700"
      },
      "message": "Add SME vector length detect\n\n- cpuinfo_get_max_arm_sme_length() returns svl vector length in bits\n- Display length of SME vectors in isa-tool\n\nSME may be enabled on cpus that do not have SVE\n"
    },
    {
      "commit": "39ea79a3c132f4e678695c579ea9353d2bd29968",
      "tree": "c14f26d49ff0a34b2d3aff8db145d761100dd2f5",
      "parents": [
        "5e3d2445e6a84d9599bee2bf78edbb4d80865e1d",
        "9d278416b6a5e79d32a618cfeb6893b7652c8b02"
      ],
      "author": {
        "name": "Digant Desai",
        "email": "digantdesai@meta.com",
        "time": "Thu Mar 27 11:18:57 2025 -0500"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Mar 27 11:18:57 2025 -0500"
      },
      "message": "Merge pull request #285 from fbarchard/sme2\n\nisa-tool display SME and SME2 for arm"
    },
    {
      "commit": "9d278416b6a5e79d32a618cfeb6893b7652c8b02",
      "tree": "c14f26d49ff0a34b2d3aff8db145d761100dd2f5",
      "parents": [
        "5e3d2445e6a84d9599bee2bf78edbb4d80865e1d"
      ],
      "author": {
        "name": "Frank Barchard",
        "email": "fbarchard@google.com",
        "time": "Thu Mar 27 02:38:47 2025 -0700"
      },
      "committer": {
        "name": "Frank Barchard",
        "email": "fbarchard@google.com",
        "time": "Thu Mar 27 02:38:47 2025 -0700"
      },
      "message": "isa-tool display SME and SME2 for arm\n"
    },
    {
      "commit": "a8d118ce2824f33c50311cf163c1eb586a8046a5",
      "tree": "1c4e99a97404cc28b8c0d0c26f4cdbbdf3c42208",
      "parents": [
        "5e3d2445e6a84d9599bee2bf78edbb4d80865e1d"
      ],
      "author": {
        "name": "Christopher Degawa",
        "email": "ccom@randomderp.com",
        "time": "Wed Feb 12 11:54:08 2025 -0600"
      },
      "committer": {
        "name": "Christopher Degawa",
        "email": "ccom@randomderp.com",
        "time": "Fri Mar 21 13:33:53 2025 -0500"
      },
      "message": "cmake: rename duplicate names\n\nXcode on macOS cannot handle multiple .c files having the same name as\nit produces an object file like init-\u003chash\u003e.o, but tries to link to\ninit.o\n\nhttps://gitlab.kitware.com/cmake/cmake/-/issues/20501\n\nSigned-off-by: Christopher Degawa \u003cccom@randomderp.com\u003e\n"
    },
    {
      "commit": "5e3d2445e6a84d9599bee2bf78edbb4d80865e1d",
      "tree": "f1d21c0c73f853ebe37087c5b176d1b2cd1ddf90",
      "parents": [
        "1a02246bf5af7e9d07c181f995d2a07c07fc6e50",
        "2624a7e05611cde935da8f57feb558e8f2239bbc"
      ],
      "author": {
        "name": "Digant Desai",
        "email": "digantdesai@meta.com",
        "time": "Fri Mar 21 02:07:22 2025 -0500"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Mar 21 02:07:22 2025 -0500"
      },
      "message": "Merge pull request #246 from dlenski/main\n\nFor Apple silicon, use machdep.cpu.brand_string in preference to decoding hw.machine"
    },
    {
      "commit": "1a02246bf5af7e9d07c181f995d2a07c07fc6e50",
      "tree": "7ff872e131598df7f75cb60a60f635625a16636e",
      "parents": [
        "d6120c73cb6d02f7dac3f2a85f5690fc91f33dbd",
        "8cb442f4dd0d4ab490aa0d64f9345400c1fca3ed"
      ],
      "author": {
        "name": "Digant Desai",
        "email": "digantdesai@meta.com",
        "time": "Fri Mar 21 01:53:40 2025 -0500"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Mar 21 01:53:40 2025 -0500"
      },
      "message": "Merge pull request #275 from davidben/func-cast\n\nFix invalid function pointer cast in cpuinfo.c"
    },
    {
      "commit": "d6120c73cb6d02f7dac3f2a85f5690fc91f33dbd",
      "tree": "030edbe392e4c2161ad8bfb8b2c70bb002348590",
      "parents": [
        "b73ae6ce38d5dd0b7fe46dbe0a4b5f4bab91c7ea",
        "48cea9d42553566c7ffda0039a819bb891d3707b"
      ],
      "author": {
        "name": "Eli Uriegas",
        "email": "1700823+seemethere@users.noreply.github.com",
        "time": "Thu Mar 20 10:01:21 2025 -0700"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Thu Mar 20 10:01:21 2025 -0700"
      },
      "message": "Merge pull request #283 from ozanMSFT/users/ozanMSFT/refactor-chip-detection\n\nRefactor windows arm64 for auto cpu detection"
    },
    {
      "commit": "48cea9d42553566c7ffda0039a819bb891d3707b",
      "tree": "030edbe392e4c2161ad8bfb8b2c70bb002348590",
      "parents": [
        "7bca5efbabfd8a7f31007a8c120332d629f1645e"
      ],
      "author": {
        "name": "Ozan Aydin",
        "email": "148207261+ozanMSFT@users.noreply.github.com",
        "time": "Thu Mar 20 10:17:31 2025 +0100"
      },
      "committer": {
        "name": "Ozan Aydin",
        "email": "148207261+ozanMSFT@users.noreply.github.com",
        "time": "Thu Mar 20 10:17:31 2025 +0100"
      },
      "message": "- clang reformat\n"
    },
    {
      "commit": "7bca5efbabfd8a7f31007a8c120332d629f1645e",
      "tree": "12104357ffb86ef33027371be479b8f9e7c629f0",
      "parents": [
        "bf43522cb72ec00593237b87c196189fa4a22fe6"
      ],
      "author": {
        "name": "Ozan Aydin",
        "email": "148207261+ozanMSFT@users.noreply.github.com",
        "time": "Mon Mar 17 09:27:41 2025 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Mon Mar 17 09:27:41 2025 +0100"
      },
      "message": "- clang format"
    },
    {
      "commit": "bf43522cb72ec00593237b87c196189fa4a22fe6",
      "tree": "89b637fdecae5fa49cff04c31ae7ed0869933e36",
      "parents": [
        "1ba344eb8f4580e5f302dd57453d574ef777617e"
      ],
      "author": {
        "name": "Ozan Aydin",
        "email": "148207261+ozanMSFT@users.noreply.github.com",
        "time": "Fri Mar 14 19:39:28 2025 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Mar 14 19:39:28 2025 +0100"
      },
      "message": "- oryon string"
    },
    {
      "commit": "1ba344eb8f4580e5f302dd57453d574ef777617e",
      "tree": "7532dcf2b071ec1f382da3af64d9a79c453863c5",
      "parents": [
        "b73ae6ce38d5dd0b7fe46dbe0a4b5f4bab91c7ea"
      ],
      "author": {
        "name": "Ozan Aydin",
        "email": "148207261+ozanMSFT@users.noreply.github.com",
        "time": "Fri Mar 14 13:57:31 2025 +0100"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Fri Mar 14 13:57:31 2025 +0100"
      },
      "message": "- refactor for auto detection"
    },
    {
      "commit": "b73ae6ce38d5dd0b7fe46dbe0a4b5f4bab91c7ea",
      "tree": "c540e99b88e54b400fef9e6f56350283b942e55f",
      "parents": [
        "aaac07ee499895770c89163ce0920ef8bb41ed23",
        "0bfd15daf1d86f5227affbd2f2a9df1cf7a8296f"
      ],
      "author": {
        "name": "Digant Desai",
        "email": "digantdesai@meta.com",
        "time": "Tue Feb 18 21:41:25 2025 -0600"
      },
      "committer": {
        "name": "GitHub",
        "email": "noreply@github.com",
        "time": "Tue Feb 18 21:41:25 2025 -0600"
      },
      "message": "Merge pull request #277 from 1480c1/pkg-config-name\n\npkg-config: use PROJECT_NAME instead of CMAKE_PROJECT_NAME"
    }
  ],
  "next": "0bfd15daf1d86f5227affbd2f2a9df1cf7a8296f"
}
