)]}'
{
  "log": [
    {
      "commit": "bb9ff576fdff48c242876f55098a3ee20a29df5d",
      "tree": "65a9ef062fdcab046fa9d4624fcafdbdbe42f150",
      "parents": [
        "3f9f0252130e7dd60d41be0802bf58f6471c691d"
      ],
      "author": {
        "name": "Vitaly Kuznetsov",
        "email": "vkuznets@redhat.com",
        "time": "Mon Dec 22 15:46:46 2025 +0100"
      },
      "committer": {
        "name": "Jason A. Donenfeld",
        "email": "Jason@zx2c4.com",
        "time": "Mon Dec 22 16:18:54 2025 +0100"
      },
      "message": "virt: vmgenid: remap memory as decrypted\n\nIt was found that AWS SEV-SNP enabled instances are not able to boot with\ncommit 81256a50aa0f (\"x86/mm: Make memremap(MEMREMAP_WB) map memory as\nencrypted by default\") applied and the reason seems to be the vmgenid\ndevice which requires unencrypted writeable memory.\n\nA similar problem was previously fixed in DRM with commit\n7dfede7d7edd (\"drm/vmwgfx: Fix guests running with TDX/SEV\").\n\nNote, trusting vmgenid device in a Confidential VM is questionable: the\nmalicious host may intentionally avoid notifying the guest when a copy is\ncreated.\n\nFixes: 81256a50aa0f (\"x86/mm: Make memremap(MEMREMAP_WB) map memory as encrypted by default\")\nSigned-off-by: Vitaly Kuznetsov \u003cvkuznets@redhat.com\u003e\nCc: stable@vger.kernel.org # 6.15+\nSigned-off-by: Jason A. Donenfeld \u003cJason@zx2c4.com\u003e\n"
    },
    {
      "commit": "3f9f0252130e7dd60d41be0802bf58f6471c691d",
      "tree": "eb5adc3fb2c5db98145b53c3fffe23c4dccd2ecc",
      "parents": [
        "f617d24606553159a271f43e36d1c71a4c317e48",
        "90fb9b98fcf5e668a13676d6e8cd546b6990d002"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 19:00:26 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 19:00:26 2025 -0800"
      },
      "message": "Merge tag \u0027random-6.19-rc1-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/crng/random\n\nPull random number generator updates from Jason Donenfeld:\n\n - Dynamically allocate cpumasks off of the stack if the kernel is\n   configured for a lot of CPUs, to handle a -Wframe-larger-than case\n\n - The removal of next_pseudo_random32() after the last user was\n   switched over to the prandom interface\n\n - The removal of get_random_u{8,16,32,64}_wait() functions, as there\n   were no users of those at all\n\n - Some house keeping changes - a few grammar cleanups in the\n   comments, system_unbound_wq was renamed to system_dfl_wq, and\n   static_key_initialized no longer needs to be checked\n\n* tag \u0027random-6.19-rc1-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/crng/random:\n  random: complete sentence of comment\n  random: drop check for static_key_initialized\n  random: remove unused get_random_var_wait functions\n  random: replace use of system_unbound_wq with system_dfl_wq\n  random: use offstack cpumask when necessary\n  prandom: remove next_pseudo_random32\n  media: vivid: use prandom\n  random: add missing words in function comments\n"
    },
    {
      "commit": "f617d24606553159a271f43e36d1c71a4c317e48",
      "tree": "f39957a87580dcc1835e2c1f0bfad5adeef93483",
      "parents": [
        "906003e15160642658358153e7598302d1b38166",
        "5dc8d277520be6f0be11f36712e557167b3964c8"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 18:53:50 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 18:53:50 2025 -0800"
      },
      "message": "Merge tag \u0027fpsimd-on-stack-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux\n\nPull arm64 FPSIMD on-stack buffer updates from Eric Biggers:\n \"This is a core arm64 change. However, I was asked to take this because\n  most uses of kernel-mode FPSIMD are in crypto or CRC code.\n\n  In v6.8, the size of task_struct on arm64 increased by 528 bytes due\n  to the new \u0027kernel_fpsimd_state\u0027 field. This field was added to allow\n  kernel-mode FPSIMD code to be preempted.\n\n  Unfortunately, 528 bytes is kind of a lot for task_struct. This\n  regression in the task_struct size was noticed and reported.\n\n  Recover that space by making this state be allocated on the stack at\n  the beginning of each kernel-mode FPSIMD section.\n\n  To make it easier for all the users of kernel-mode FPSIMD to do that\n  correctly, introduce and use a \u0027scoped_ksimd\u0027 abstraction\"\n\n* tag \u0027fpsimd-on-stack-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux: (23 commits)\n  lib/crypto: arm64: Move remaining algorithms to scoped ksimd API\n  lib/crypto: arm/blake2b: Move to scoped ksimd API\n  arm64/fpsimd: Allocate kernel mode FP/SIMD buffers on the stack\n  arm64/fpu: Enforce task-context only for generic kernel mode FPU\n  net/mlx5: Switch to more abstract scoped ksimd guard API on arm64\n  arm64/xorblocks:  Switch to \u0027ksimd\u0027 scoped guard API\n  crypto/arm64: sm4 - Switch to \u0027ksimd\u0027 scoped guard API\n  crypto/arm64: sm3 - Switch to \u0027ksimd\u0027 scoped guard API\n  crypto/arm64: sha3 - Switch to \u0027ksimd\u0027 scoped guard API\n  crypto/arm64: polyval - Switch to \u0027ksimd\u0027 scoped guard API\n  crypto/arm64: nhpoly1305 - Switch to \u0027ksimd\u0027 scoped guard API\n  crypto/arm64: aes-gcm - Switch to \u0027ksimd\u0027 scoped guard API\n  crypto/arm64: aes-blk - Switch to \u0027ksimd\u0027 scoped guard API\n  crypto/arm64: aes-ccm - Switch to \u0027ksimd\u0027 scoped guard API\n  raid6: Move to more abstract \u0027ksimd\u0027 guard API\n  crypto: aegis128-neon - Move to more abstract \u0027ksimd\u0027 guard API\n  crypto/arm64: sm4-ce-gcm - Avoid pointless yield of the NEON unit\n  crypto/arm64: sm4-ce-ccm - Avoid pointless yield of the NEON unit\n  crypto/arm64: aes-ce-ccm - Avoid pointless yield of the NEON unit\n  lib/crc: Switch ARM and arm64 to \u0027ksimd\u0027 scoped guard API\n  ...\n"
    },
    {
      "commit": "906003e15160642658358153e7598302d1b38166",
      "tree": "ee36ed29e43e6038c69cadc5c82eddcb9d139d83",
      "parents": [
        "8f4c9978de91a9a3b37df1e74d6201acfba6cefd",
        "4f0382b0901b43552b600f8e5f806295778b0fb0"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 18:26:54 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 18:26:54 2025 -0800"
      },
      "message": "Merge tag \u0027libcrypto-at-least-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux\n\nPull \u0027at_least\u0027 array size update from Eric Biggers:\n \"C supports lower bounds on the sizes of array parameters, using the\n  static keyword as follows: \u0027void f(int a[static 32]);\u0027. This allows\n  the compiler to warn about a too-small array being passed.\n\n  As discussed, this reuse of the \u0027static\u0027 keyword, while standard, is a\n  bit obscure. Therefore, add an alias \u0027at_least\u0027 to compiler_types.h.\n\n  Then, add this \u0027at_least\u0027 annotation to the array parameters of\n  various crypto library functions\"\n\n* tag \u0027libcrypto-at-least-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux:\n  lib/crypto: sha2: Add at_least decoration to fixed-size array params\n  lib/crypto: sha1: Add at_least decoration to fixed-size array params\n  lib/crypto: poly1305: Add at_least decoration to fixed-size array params\n  lib/crypto: md5: Add at_least decoration to fixed-size array params\n  lib/crypto: curve25519: Add at_least decoration to fixed-size array params\n  lib/crypto: chacha: Add at_least decoration to fixed-size array params\n  lib/crypto: chacha20poly1305: Statically check fixed array lengths\n  compiler_types: introduce at_least parameter decoration pseudo keyword\n  wifi: iwlwifi: trans: rename at_least variable to min_mode\n"
    },
    {
      "commit": "8f4c9978de91a9a3b37df1e74d6201acfba6cefd",
      "tree": "ab4ab2ba48e3890805c82a7b60f910d0491fbdcf",
      "parents": [
        "db425f7a0b158d0dbb07c4f4653795aaad3a7a15",
        "0e253e250ed0e46f5ff6962c840157da9dab48cd"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 18:24:35 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 18:24:35 2025 -0800"
      },
      "message": "Merge tag \u0027aes-gcm-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux\n\nPull AES-GCM optimizations from Eric Biggers:\n \"More optimizations and cleanups for the x86_64 AES-GCM code:\n\n   - Add a VAES+AVX2 optimized implementation of AES-GCM. This is very\n     helpful on CPUs that have VAES but not AVX512, such as AMD Zen 3.\n\n   - Make the VAES+AVX512 optimized implementation of AES-GCM handle\n     large amounts of associated data efficiently.\n\n   - Remove the \"avx10_256\" implementation of AES-GCM. It\u0027s superseded\n     by the VAES+AVX2 optimized implementation.\n\n   - Rename the \"avx10_512\" implementation to \"avx512\"\n\n  Overall, this fills in a gap where AES-GCM wasn\u0027t fully optimized on\n  some recent CPUs. It also drops code that won\u0027t be as useful as\n  initially expected due to AVX10/256 being dropped from the AVX10 spec\"\n\n* tag \u0027aes-gcm-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux:\n  crypto: x86/aes-gcm-vaes-avx2 - initialize full %rax return register\n  crypto: x86/aes-gcm - optimize long AAD processing with AVX512\n  crypto: x86/aes-gcm - optimize AVX512 precomputation of H^2 from H^1\n  crypto: x86/aes-gcm - revise some comments in AVX512 code\n  crypto: x86/aes-gcm - reorder AVX512 precompute and aad_update functions\n  crypto: x86/aes-gcm - clean up AVX512 code to assume 512-bit vectors\n  crypto: x86/aes-gcm - rename avx10 and avx10_512 to avx512\n  crypto: x86/aes-gcm - remove VAES+AVX10/256 optimized code\n  crypto: x86/aes-gcm - add VAES+AVX2 optimized code\n"
    },
    {
      "commit": "db425f7a0b158d0dbb07c4f4653795aaad3a7a15",
      "tree": "b4bcb5b5ef2092de302ae96f8a4b155c96b9ab15",
      "parents": [
        "5abe8d8efc022cc78b6273d01e4a453242b9f4d8",
        "578fe3ff3d5bc9d851d597bb12dd74c22ad55ff8"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 18:20:06 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 18:20:06 2025 -0800"
      },
      "message": "Merge tag \u0027libcrypto-tests-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux\n\nPull crypto library test updates from Eric Biggers:\n\n - Add KUnit test suites for SHA-3, BLAKE2b, and POLYVAL. These are the\n   algorithms that have new crypto library interfaces this cycle.\n\n - Remove the crypto_shash POLYVAL tests. They\u0027re no longer needed\n   because POLYVAL support was removed from crypto_shash. Better POLYVAL\n   test coverage is now provided via the KUnit test suite.\n\n* tag \u0027libcrypto-tests-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux:\n  crypto: testmgr - Remove polyval tests\n  lib/crypto: tests: Add KUnit tests for POLYVAL\n  lib/crypto: tests: Add additional SHAKE tests\n  lib/crypto: tests: Add SHA3 kunit tests\n  lib/crypto: tests: Add KUnit tests for BLAKE2b\n"
    },
    {
      "commit": "5abe8d8efc022cc78b6273d01e4a453242b9f4d8",
      "tree": "3b3dc060e8a1a60ac5924afd505aba50bd3c4dd5",
      "parents": [
        "619f4edc8d4f785dcabf564786e787fa8108deb4",
        "2dbb6f4a25d38fcf7d6c1c682e45a13e6bbe9562"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 18:01:03 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 18:01:03 2025 -0800"
      },
      "message": "Merge tag \u0027libcrypto-updates-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux\n\nPull crypto library updates from Eric Biggers:\n \"This is the main crypto library pull request for 6.19. It includes:\n\n   - Add SHA-3 support to lib/crypto/, including support for both the\n     hash functions and the extendable-output functions. Reimplement the\n     existing SHA-3 crypto_shash support on top of the library.\n\n     This is motivated mainly by the upcoming support for the ML-DSA\n     signature algorithm, which needs the SHAKE128 and SHAKE256\n     functions. But even on its own it\u0027s a useful cleanup.\n\n     This also fixes the longstanding issue where the\n     architecture-optimized SHA-3 code was disabled by default.\n\n   - Add BLAKE2b support to lib/crypto/, and reimplement the existing\n     BLAKE2b crypto_shash support on top of the library.\n\n     This is motivated mainly by btrfs, which supports BLAKE2b\n     checksums. With this change, all btrfs checksum algorithms now have\n     library APIs. btrfs is planned to start just using the library\n     directly.\n\n     This refactor also improves consistency between the BLAKE2b code\n     and BLAKE2s code. And as usual, it also fixes the issue where the\n     architecture-optimized BLAKE2b code was disabled by default.\n\n   - Add POLYVAL support to lib/crypto/, replacing the existing POLYVAL\n     support in crypto_shash. Reimplement HCTR2 on top of the library.\n\n     This simplifies the code and improves HCTR2 performance. As usual,\n     it also makes the architecture-optimized code be enabled by\n     default. The generic implementation of POLYVAL is greatly improved\n     as well.\n\n   - Clean up the BLAKE2s code\n\n   - Add FIPS self-tests for SHA-1, SHA-2, and SHA-3\"\n\n* tag \u0027libcrypto-updates-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ebiggers/linux: (37 commits)\n  fscrypt: Drop obsolete recommendation to enable optimized POLYVAL\n  crypto: polyval - Remove the polyval crypto_shash\n  crypto: hctr2 - Convert to use POLYVAL library\n  lib/crypto: x86/polyval: Migrate optimized code into library\n  lib/crypto: arm64/polyval: Migrate optimized code into library\n  lib/crypto: polyval: Add POLYVAL library\n  crypto: polyval - Rename conflicting functions\n  lib/crypto: x86/blake2s: Use vpternlogd for 3-input XORs\n  lib/crypto: x86/blake2s: Avoid writing back unchanged \u0027f\u0027 value\n  lib/crypto: x86/blake2s: Improve readability\n  lib/crypto: x86/blake2s: Use local labels for data\n  lib/crypto: x86/blake2s: Drop check for nblocks \u003d\u003d 0\n  lib/crypto: x86/blake2s: Fix 32-bit arg treated as 64-bit\n  lib/crypto: arm, arm64: Drop filenames from file comments\n  lib/crypto: arm/blake2s: Fix some comments\n  crypto: s390/sha3 - Remove superseded SHA-3 code\n  crypto: sha3 - Reimplement using library API\n  crypto: jitterentropy - Use default sha3 implementation\n  lib/crypto: s390/sha3: Add optimized one-shot SHA-3 digest functions\n  lib/crypto: sha3: Support arch overrides of one-shot digest functions\n  ...\n"
    },
    {
      "commit": "619f4edc8d4f785dcabf564786e787fa8108deb4",
      "tree": "385104683e716826750b83101779fa6c585dd311",
      "parents": [
        "d348c22394ad3c8eaf7bc693cb0ca0edc2ec5246",
        "63d26c3811421ceeb5b82a85489b88bf545e33c7"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 17:49:12 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 17:49:12 2025 -0800"
      },
      "message": "Merge tag \u0027thermal-6.19-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull thermal control updates from Rafael Wysocki:\n \"These add Nova Lake processor support to the Intel thermal drivers and\n  DPTF code, update thermal control documentation, simplify the ACPI\n  DPTF code related to thermal control, add QCS8300 compatible to the\n  tsens thermal DT bindings, add DT bindings for NXP i.MX91 thermal\n  module and add support for it to the imx91 thermal driver, update a\n  few other thermal drivers and fix a format string issue in a thermal\n  utility:\n\n   - Add Nova Lake processor thermal device to the int340x\n     processor_thermal driver, add DLVR support for Nova Lake to it, add\n     Nova Lake support to the ACPI DPTF code, document thermal\n     throttling on Intel platforms, and update workload type hint\n     interface documentation (Srinivas Pandruvada)\n\n   - Remove int340x thermal scan handler from the ACPI DPTF code because\n     it turned out to be unnecessary (Slawomir Rosek)\n\n   - Clean up the Intel int340x thermal driver (Kaushlendra Kumar)\n\n   - Document the RZ/V2H TSU DT bindings (Ovidiu Panait)\n\n   - Document the Kaanapali Temperature Sensor (Manaf Meethalavalappu\n     Pallikunhi)\n\n   - Document R-Car Gen4 and RZ/G2 support in driver comment (Marek\n     Vasut)\n\n   - Convert to DEFINE_SIMPLE_DEV_PM_OPS() in R-Car [Gen3] (Geert\n     Uytterhoeven)\n\n   - Fix format string bug in thermal-engine (Malaya Kumar Rout)\n\n   - Make ipq5018 tsens standalone compatible (George Moussalem)\n\n   - Add the QCS8300 compatible for QCom Tsens (Gaurav Kohli)\n\n   - Add support for the NXP i.MX91 thermal module, including the DT\n     bindings (Pengfei Li)\"\n\n* tag \u0027thermal-6.19-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n  thermal/drivers/imx91: Add support for i.MX91 thermal monitoring unit\n  dt-bindings: thermal: fsl,imx91-tmu: add bindings for NXP i.MX91 thermal module\n  dt-bindings: thermal: tsens: Add QCS8300 compatible\n  dt-bindings: thermal: qcom-tsens: make ipq5018 tsens standalone compatible\n  tools/thermal/thermal-engine: Fix format string bug in thermal-engine\n  docs: driver-api/thermal/intel_dptf: Add new workload type hint\n  thermal/drivers/rcar_gen3: Convert to DEFINE_SIMPLE_DEV_PM_OPS()\n  thermal/drivers/rcar: Convert to DEFINE_SIMPLE_DEV_PM_OPS()\n  Documentation: thermal: Document thermal throttling on Intel platforms\n  ACPI: DPTF: Support Nova Lake\n  thermal: intel: int340x: Add DLVR support for Nova Lake\n  thermal: int340x: processor_thermal: Add Nova Lake processor thermal device\n  thermal: intel: int340x: Replace sprintf() with sysfs_emit()\n  thermal: intel: int340x: Use symbolic constant for UUID comparison\n  thermal/drivers/rcar_gen3: Document R-Car Gen4 and RZ/G2 support in driver comment\n  dt-bindings: thermal: qcom-tsens: document the Kaanapali Temperature Sensor\n  dt-bindings: thermal: r9a09g047-tsu: Document RZ/V2H TSU\n  ACPI: DPTF: Remove int340x thermal scan handler\n  thermal: intel: Select INT340X_THERMAL from INTEL_SOC_DTS_THERMAL\n"
    },
    {
      "commit": "d348c22394ad3c8eaf7bc693cb0ca0edc2ec5246",
      "tree": "57cbdede16efa659dc64f5ab78b82a8e78596272",
      "parents": [
        "959bfe496bbaf3daa5dca32d397e29ea12471779",
        "7cede21e9f04f16a456d3c3c8a9a8899c8d84757"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 17:31:22 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 17:31:22 2025 -0800"
      },
      "message": "Merge tag \u0027pm-6.19-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull power management updates from Rafael Wysocki:\n \"There are quite a few interesting things here, including new hardware\n  support, new features, some bug fixes and documentation updates. In\n  addition, there are a usual bunch of minor fixes and cleanups all\n  over.\n\n  In the new hardware support category, there are intel_pstate and\n  intel_rapl driver updates to support new processors, Panther Lake,\n  Wildcat Lake, Noval Lake, and Diamond Rapids in the OOB mode, OPP and\n  bandwidth allocation support in the tegra186 cpufreq driver, and\n  JH7110S SOC support in dt-platdev cpufreq.\n\n  The new features are the PM QoS CPU latency limit for suspend-to-idle,\n  the netlink support for the energy model management, support for\n  terminating system suspend via a wakeup event during the sync of file\n  systems, configurable number of hibernation compression threads, the\n  runtime PM auto-cleanup macros, and the \"poweroff\" PM event that is\n  expected to be used during system shutdown.\n\n  Bugs are mostly fixed in cpuidle governors, but there are also fixes\n  elsewhere, like in the amd-pstate cpufreq driver.\n\n  Documentation updates include, but are not limited to, a new doc on\n  debugging shutdown hangs, cross-referencing fixes and cleanups in the\n  intel_pstate documentation, and updates of comments in the core\n  hibernation code.\n\n  Specifics:\n\n   - Introduce and document a QoS limit on CPU exit latency during\n     wakeup from suspend-to-idle (Ulf Hansson)\n\n   - Add support for building libcpupower statically (Zuo An)\n\n   - Add support for sending netlink notifications to user space on\n     energy model updates (Changwoo Mini, Peng Fan)\n\n   - Minor improvements to the Rust OPP interface (Tamir Duberstein)\n\n   - Fixes to scope-based pointers in the OPP library (Viresh Kumar)\n\n   - Use residency threshold in polling state override decisions in the\n     menu cpuidle governor (Aboorva Devarajan)\n\n   - Add sanity check for exit latency and target residency in the\n     cpufreq core (Rafael Wysocki)\n\n   - Use this_cpu_ptr() where possible in the teo governor (Christian\n     Loehle)\n\n   - Rework the handling of tick wakeups in the teo cpuidle governor to\n     increase the likelihood of stopping the scheduler tick in the cases\n     when tick wakeups can be counted as non-timer ones (Rafael Wysocki)\n\n   - Fix a reverse condition in the teo cpuidle governor and drop a\n     misguided target residency check from it (Rafael Wysocki)\n\n   - Clean up multiple minor defects in the teo cpuidle governor (Rafael\n     Wysocki)\n\n   - Update header inclusion to make it follow the Include What You Use\n     principle (Andy Shevchenko)\n\n   - Enable MSR-based RAPL PMU support in the intel_rapl power capping\n     driver and arrange for using it on the Panther Lake and Wildcat\n     Lake processors (Kuppuswamy Sathyanarayanan)\n\n   - Add support for Nova Lake and Wildcat Lake processors to the\n     intel_rapl power capping driver (Kaushlendra Kumar, Srinivas\n     Pandruvada)\n\n   - Add OPP and bandwidth support for Tegra186 (Aaron Kling)\n\n   - Optimizations for parameter array handling in the amd-pstate\n     cpufreq driver (Mario Limonciello)\n\n   - Fix for mode changes with offline CPUs in the amd-pstate cpufreq\n     driver (Gautham Shenoy)\n\n   - Preserve freq_table_sorted across suspend/hibernate in the cpufreq\n     core (Zihuan Zhang)\n\n   - Adjust energy model rules for Intel hybrid platforms in the\n     intel_pstate cpufreq driver and improve printing of debug messages\n     in it (Rafael Wysocki)\n\n   - Replace deprecated strcpy() in cpufreq_unregister_governor()\n     (Thorsten Blum)\n\n   - Fix duplicate hyperlink target errors in the intel_pstate cpufreq\n     driver documentation and use :ref: directive for internal linking\n     in it (Swaraj Gaikwad, Bagas Sanjaya)\n\n   - Add Diamond Rapids OOB mode support to the intel_pstate cpufreq\n     driver (Kuppuswamy Sathyanarayanan)\n\n   - Use mutex guard for driver locking in the intel_pstate driver and\n     eliminate some code duplication from it (Rafael Wysocki)\n\n   - Replace udelay() with usleep_range() in ACPI cpufreq (Kaushlendra\n     Kumar)\n\n   - Minor improvements to various cpufreq drivers (Christian Marangi,\n     Hal Feng, Jie Zhan, Marco Crivellari, Miaoqian Lin, and Shuhao Fu)\n\n   - Replace snprintf() with scnprintf() in show_trace_dev_match()\n     (Kaushlendra Kumar)\n\n   - Fix memory allocation error handling in pm_vt_switch_required()\n     (Malaya Kumar Rout)\n\n   - Introduce CALL_PM_OP() macro and use it to simplify code in generic\n     PM operations (Kaushlendra Kumar)\n\n   - Add module param to backtrace all CPUs in the device power\n     management watchdog (Sergey Senozhatsky)\n\n   - Rework message printing in swsusp_save() (Rafael Wysocki)\n\n   - Make it possible to change the number of hibernation compression\n     threads (Xueqin Luo)\n\n   - Clarify that only cgroup1 freezer uses PM freezer (Tejun Heo)\n\n   - Add document on debugging shutdown hangs to PM documentation and\n     correct a mistaken configuration option in it (Mario Limonciello)\n\n   - Shut down wakeup source timer before removing the wakeup source\n     from the list (Kaushlendra Kumar, Rafael Wysocki)\n\n   - Introduce new PMSG_POWEROFF event for system shutdown handling with\n     the help of PM device callbacks (Mario Limonciello)\n\n   - Make pm_test delay interruptible by wakeup events (Riwen Lu)\n\n   - Clean up kernel-doc comment style usage in the core hibernation\n     code and remove unuseful comments from it (Sunday Adelodun, Rafael\n     Wysocki)\n\n   - Add support for handling wakeup events and aborting the suspend\n     process while it is syncing file systems (Samuel Wu, Rafael\n     Wysocki)\n\n   - Add WQ_UNBOUND to pm_wq workqueue (Marco Crivellari)\n\n   - Add runtime PM wrapper macros for ACQUIRE()/ACQUIRE_ERR() and use\n     them in the PCI core and the ACPI TAD driver (Rafael Wysocki)\n\n   - Improve runtime PM in the ACPI TAD driver (Rafael Wysocki)\n\n   - Update pm_runtime_allow/forbid() documentation (Rafael Wysocki)\n\n   - Fix typos in runtime.c comments (Malaya Kumar Rout)\n\n   - Move governor.h from devfreq under include/linux/ and rename to\n     devfreq-governor.h to allow devfreq governor definitions in out of\n     drivers/devfreq/ (Dmitry Baryshkov)\n\n   - Use min() to improve readability in tegra30-devfreq.c (Thorsten\n     Blum)\n\n   - Fix potential use-after-free issue of OPP handling in\n     hisi_uncore_freq.c (Pengjie Zhang)\n\n   - Fix typo in DFSO_DOWNDIFFERENTIAL macro name in\n     governor_simpleondemand.c in devfreq (Riwen Lu)\"\n\n* tag \u0027pm-6.19-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: (96 commits)\n  PM / devfreq: Fix typo in DFSO_DOWNDIFFERENTIAL macro name\n  cpuidle: Warn instead of bailing out if target residency check fails\n  cpuidle: Update header inclusion\n  Documentation: power/cpuidle: Document the CPU system wakeup latency QoS\n  cpuidle: Respect the CPU system wakeup QoS limit for cpuidle\n  sched: idle: Respect the CPU system wakeup QoS limit for s2idle\n  pmdomain: Respect the CPU system wakeup QoS limit for cpuidle\n  pmdomain: Respect the CPU system wakeup QoS limit for s2idle\n  PM: QoS: Introduce a CPU system wakeup QoS limit\n  cpuidle: governors: teo: Add missing space to the description\n  PM: hibernate: Extra cleanup of comments in swap handling code\n  PM / devfreq: tegra30: use min to simplify actmon_cpu_to_emc_rate\n  PM / devfreq: hisi: Fix potential UAF in OPP handling\n  PM / devfreq: Move governor.h to a public header location\n  powercap: intel_rapl: Enable MSR-based RAPL PMU support\n  powercap: intel_rapl: Prepare read_raw() interface for atomic-context callers\n  cpufreq: qcom-nvmem: fix compilation warning for qcom_cpufreq_ipq806x_match_list\n  PM: sleep: Call pm_sleep_fs_sync() instead of ksys_sync_helper()\n  PM: sleep: Add support for wakeup during filesystem sync\n  cpufreq: ACPI: Replace udelay() with usleep_range()\n  ...\n"
    },
    {
      "commit": "959bfe496bbaf3daa5dca32d397e29ea12471779",
      "tree": "999b147c1d37a8eef2f0fb1453c195eccedc2712",
      "parents": [
        "44fc84337b6eae580a51cf6f7ca6a22ef1349556",
        "af47d98064e900f279c98bb3bb57445a1e7a8f6b"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 17:24:03 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 17:24:03 2025 -0800"
      },
      "message": "Merge tag \u0027acpi-6.19-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull ACPI updates from Rafael Wysocki:\n \"These add Microsoft fan extensions support to the ACPI fan driver, fix\n  a bug in ACPICA, update other ACPI drivers (processor, time and alarm\n  device), update ACPI power management code and ACPI device properties\n  management, and fix an ACPI utility:\n\n   - Avoid walking the ACPI namespace in the AML interpreter if the\n     starting node cannot be determined (Cryolitia PukNgae)\n\n   - Use min() instead of min_t() in the ACPI device properties handling\n     code to avoid discarding significant bits (David Laight)\n\n   - Fix potential fwnode refcount leak in\n     acpi_fwnode_graph_parse_endpoint() that may prevent the parent\n     fwnode from being released (Haotian Zhang)\n\n   - Rework acpi_graph_get_next_endpoint() to use ACPI functions only,\n     remove unnecessary conditionals from it to make it easier to\n     follow, and make acpi_get_next_subnode() static (Sakari Ailus)\n\n   - Drop unused function acpi_get_lps0_constraint(), make some\n     Low-Power S0 callback functions for suspend-to-idle static, and\n     rearrange the code retrieving Low-Power S0 constraints so it only\n     runs when the constraints are actually used (Rafael Wysocki)\n\n   - Drop redundant locking from the ACPI battery driver (Rafael\n     Wysocki)\n\n   - Improve runtime PM in the ACPI time and alarm device (TAD) driver\n     using guard macros and rearrange code related to runtime PM in\n     acpi_tad_remove() (Rafael Wysocki)\n\n   - Add support for Microsoft fan extensions to the ACPI fan driver\n     along with notification support and work around a 64-bit firmware\n     bug in that driver (Armin Wolf)\n\n   - Use ACPI_FREE() to free ACPI buffer in the ACPI DPTF code\n     (Kaushlendra Kumar)\n\n   - Fix a memory leak and a resource leak in the ACPI pfrut utility\n     (Malaya Kumar Rout)\n\n   - Replace `core::mem::zeroed` with `pin_init::zeroed` in the ACPI\n     Rust code (Siyuan Huang)\n\n   - Update the ACPI code to use the new style of allocating workqueues\n     and new global workqueues (Marco Crivellari)\n\n   - Fix two spelling mistakes in the ACPI code (Chu Guangqing)\n\n   - Fix ISAPNP to generate uevents to auto-load modules (René Rebe)\n\n   - Relocate the state flags initialization in the ACPI processor idle\n     driver and drop redundant C-state count checks from it (Huisong Li)\n\n   - Fix map_x2apic_id() in the ACPI processor core driver for\n     amd-pstate on am4 (René Rebe)\"\n\n* tag \u0027acpi-6.19-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: (30 commits)\n  ACPI: PM: Fix a spelling mistake\n  ACPI: LPSS: Fix a spelling mistake\n  ACPI: processor_core: fix map_x2apic_id for amd-pstate on am4\n  ACPICA: Avoid walking the Namespace if start_node is NULL\n  ACPI: tools: pfrut: fix memory leak and resource leak in pfrut.c\n  ACPI: property: use min() instead of min_t()\n  PNP: Fix ISAPNP to generate uevents to auto-load modules\n  ACPI: property: Fix fwnode refcount leak in acpi_fwnode_graph_parse_endpoint()\n  ACPI: DPTF: Use ACPI_FREE() for ACPI buffer deallocation\n  ACPI: processor: idle: Drop redundant C-state count checks\n  ACPI: thermal: Add WQ_PERCPU to alloc_workqueue() users\n  ACPI: OSL: Add WQ_PERCPU to alloc_workqueue() users\n  ACPI: EC: Add WQ_PERCPU to alloc_workqueue() users\n  ACPI: OSL: replace use of system_wq with system_percpu_wq\n  ACPI: scan: replace use of system_unbound_wq with system_dfl_wq\n  ACPI: fan: Add support for Microsoft fan extensions\n  ACPI: fan: Add hwmon notification support\n  ACPI: fan: Add basic notification support\n  ACPI: TAD: Improve runtime PM using guard macros\n  ACPI: TAD: Rearrange runtime PM operations in acpi_tad_remove()\n  ...\n"
    },
    {
      "commit": "44fc84337b6eae580a51cf6f7ca6a22ef1349556",
      "tree": "7a6f802b2d35ddbb30c8591159877a3bffede6c0",
      "parents": [
        "2547f79b0b0cd969ae6f736890af4ebd9368cda5",
        "edde060637b92607f3522252c03d64ad06369933"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 17:03:55 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 17:03:55 2025 -0800"
      },
      "message": "Merge tag \u0027arm64-upstream\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux\n\nPull arm64 updates from Catalin Marinas:\n \"These are the arm64 updates for 6.19.\n\n  The biggest part is the Arm MPAM driver under drivers/resctrl/.\n  There\u0027s a patch touching mm/ to handle spurious faults for huge pmd\n  (similar to the pte version). The corresponding arm64 part allows us\n  to avoid the TLB maintenance if a (huge) page is reused after a write\n  fault. There\u0027s EFI refactoring to allow runtime services with\n  preemption enabled and the rest is the usual perf/PMU updates and\n  several cleanups/typos.\n\n  Summary:\n\n  Core features:\n\n   - Basic Arm MPAM (Memory system resource Partitioning And Monitoring)\n     driver under drivers/resctrl/ which makes use of the fs/rectrl/ API\n\n  Perf and PMU:\n\n   - Avoid cycle counter on multi-threaded CPUs\n\n   - Extend CSPMU device probing and add additional filtering support\n     for NVIDIA implementations\n\n   - Add support for the PMUs on the NoC S3 interconnect\n\n   - Add additional compatible strings for new Cortex and C1 CPUs\n\n   - Add support for data source filtering to the SPE driver\n\n   - Add support for i.MX8QM and \"DB\" PMU in the imx PMU driver\n\n  Memory managemennt:\n\n   - Avoid broadcast TLBI if page reused in write fault\n\n   - Elide TLB invalidation if the old PTE was not valid\n\n   - Drop redundant cpu_set_*_tcr_t0sz() macros\n\n   - Propagate pgtable_alloc() errors outside of __create_pgd_mapping()\n\n   - Propagate return value from __change_memory_common()\n\n  ACPI and EFI:\n\n   - Call EFI runtime services without disabling preemption\n\n   - Remove unused ACPI function\n\n  Miscellaneous:\n\n   - ptrace support to disable streaming on SME-only systems\n\n   - Improve sysreg generation to include a \u0027Prefix\u0027 descriptor\n\n   - Replace __ASSEMBLY__ with __ASSEMBLER__\n\n   - Align register dumps in the kselftest zt-test\n\n   - Remove some no longer used macros/functions\n\n   - Various spelling corrections\"\n\n* tag \u0027arm64-upstream\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux: (94 commits)\n  arm64/mm: Document why linear map split failure upon vm_reset_perms is not problematic\n  arm64/pageattr: Propagate return value from __change_memory_common\n  arm64/sysreg: Remove unused define ARM64_FEATURE_FIELD_BITS\n  KVM: arm64: selftests: Consider all 7 possible levels of cache\n  KVM: arm64: selftests: Remove ARM64_FEATURE_FIELD_BITS and its last user\n  arm64: atomics: lse: Remove unused parameters from ATOMIC_FETCH_OP_AND macros\n  Documentation/arm64: Fix the typo of register names\n  ACPI: GTDT: Get rid of acpi_arch_timer_mem_init()\n  perf: arm_spe: Add support for filtering on data source\n  perf: Add perf_event_attr::config4\n  perf/imx_ddr: Add support for PMU in DB (system interconnects)\n  perf/imx_ddr: Get and enable optional clks\n  perf/imx_ddr: Move ida_alloc() from ddr_perf_init() to ddr_perf_probe()\n  dt-bindings: perf: fsl-imx-ddr: Add compatible string for i.MX8QM, i.MX8QXP and i.MX8DXL\n  arm64: remove duplicate ARCH_HAS_MEM_ENCRYPT\n  arm64: mm: use untagged address to calculate page index\n  MAINTAINERS: new entry for MPAM Driver\n  arm_mpam: Add kunit tests for props_mismatch()\n  arm_mpam: Add kunit test for bitmap reset\n  arm_mpam: Add helper to reset saved mbwu state\n  ...\n"
    },
    {
      "commit": "2547f79b0b0cd969ae6f736890af4ebd9368cda5",
      "tree": "6d60bc0250913c95850582401eac37bffaafb784",
      "parents": [
        "4a21d1b33f47b7fb330f8e7af69192493e3125d1",
        "283f90b50de077970c429e5b9b2745d5e94a5a45"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 16:37:00 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 16:37:00 2025 -0800"
      },
      "message": "Merge tag \u0027s390-6.19-1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux\n\nPull s390 updates from Heiko Carstens:\n\n - Provide a new interface for dynamic configuration and deconfiguration\n   of hotplug memory, allowing with and without memmap_on_memory\n   support. This makes the way memory hotplug is handled on s390 much\n   more similar to other architectures\n\n - Remove compat support. There shouldn\u0027t be any compat user space\n   around anymore, therefore get rid of a lot of code which also doesn\u0027t\n   need to be tested anymore\n\n - Add stackprotector support. GCC 16 will get new compiler options,\n   which allow to generate code required for kernel stackprotector\n   support\n\n - Merge pai_crypto and pai_ext PMU drivers into a new driver. This\n   removes a lot of duplicated code. The new driver is also extendable\n   and allows to support new PMUs\n\n - Add driver override support for AP queues\n\n - Rework and extend zcrypt and AP trace events to allow for tracing of\n   crypto requests\n\n - Support block sizes larger than 65535 bytes for CCW tape devices\n\n - Since the rework of the virtual kernel address space the module area\n   and the kernel image are within the same 4GB area. This eliminates\n   the need of weak per cpu variables. Get rid of\n   ARCH_MODULE_NEEDS_WEAK_PER_CPU\n\n - Various other small improvements and fixes\n\n* tag \u0027s390-6.19-1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (92 commits)\n  watchdog: diag288_wdt: Remove KMSG_COMPONENT macro\n  s390/entry: Use lay instead of aghik\n  s390/vdso: Get rid of -m64 flag handling\n  s390/vdso: Rename vdso64 to vdso\n  s390: Rename head64.S to head.S\n  s390/vdso: Use common STABS_DEBUG and DWARF_DEBUG macros\n  s390: Add stackprotector support\n  s390/modules: Simplify module_finalize() slightly\n  s390: Remove KMSG_COMPONENT macro\n  s390/percpu: Get rid of ARCH_MODULE_NEEDS_WEAK_PER_CPU\n  s390/ap: Restrict driver_override versus apmask and aqmask use\n  s390/ap: Rename mutex ap_perms_mutex to ap_attr_mutex\n  s390/ap: Support driver_override for AP queue devices\n  s390/ap: Use all-bits-one apmask/aqmask for vfio in_use() checks\n  s390/debug: Update description of resize operation\n  s390/syscalls: Switch to generic system call table generation\n  s390/syscalls: Remove system call table pointer from thread_struct\n  s390/uapi: Remove 31 bit support from uapi header files\n  s390: Remove compat support\n  tools: Remove s390 compat support\n  ...\n"
    },
    {
      "commit": "4a21d1b33f47b7fb330f8e7af69192493e3125d1",
      "tree": "f2f603ff32a3fe6b26c17880aca66dfaecff50c4",
      "parents": [
        "d61f1cc5db799f4e44a63418b2dc19396787427b",
        "aaf4e92341b081bac1230f22619b0a298e285dab"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 16:32:02 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 16:32:02 2025 -0800"
      },
      "message": "Merge tag \u0027m68k-for-v6.19-tag1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k\n\nPull m68k update from Geert Uytterhoeven:\n\n - defconfig update\n\n* tag \u0027m68k-for-v6.19-tag1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k:\n  m68k: defconfig: Update defconfigs for v6.18-rc1\n"
    },
    {
      "commit": "d61f1cc5db799f4e44a63418b2dc19396787427b",
      "tree": "cace86d8f679ae5dc509a304c7e2f7b24351d4af",
      "parents": [
        "a7610b8465393edc936c2640b903cb7758d5d155",
        "d5cb9574391cc0a4683c22944d00d0ad76a224d3"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 14:48:08 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 14:48:08 2025 -0800"
      },
      "message": "Merge tag \u0027x86_cpu_for_6.19-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 CPU feature updates from Dave Hansen:\n \"The biggest thing of note here is Linear Address Space Separation\n  (LASS). It represents the first time I can think of that the\n  upper\u003d\u003ekernel/lower\u003d\u003euser address space convention is actually\n  recognized by the hardware on x86. It ensures that userspace can not\n  even get the hardware to _start_ page walks for the kernel address\n  space. This, of course, is a really nice generic side channel defense.\n\n  This is really only a down payment on LASS support. There are still\n  some details to work out in its interaction with EFI calls and\n  vsyscall emulation. For now, LASS is disabled if either of those\n  features is compiled in (which is almost always the case).\n\n  There\u0027s also one straggler commit in here which converts an\n  under-utilized AMD CPU feature leaf into a generic Linux-defined leaf\n  so more feature can be packed in there.\n\n  Summary:\n\n   - Enable Linear Address Space Separation (LASS)\n\n   - Change X86_FEATURE leaf 17 from an AMD leaf to Linux-defined\"\n\n* tag \u0027x86_cpu_for_6.19-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  x86/cpu: Enable LASS during CPU initialization\n  selftests/x86: Update the negative vsyscall tests to expect a #GP\n  x86/traps: Communicate a LASS violation in #GP message\n  x86/kexec: Disable LASS during relocate kernel\n  x86/alternatives: Disable LASS when patching kernel code\n  x86/asm: Introduce inline memcpy and memset\n  x86/cpu: Add an LASS dependency on SMAP\n  x86/cpufeatures: Enumerate the LASS feature bits\n  x86/cpufeatures: Make X86_FEATURE leaf 17 Linux-specific\n"
    },
    {
      "commit": "a7610b8465393edc936c2640b903cb7758d5d155",
      "tree": "1884aa2b3643138bd4365559775d45547a847811",
      "parents": [
        "e2aa39b368bb147afe8f6bd63d962494354f6498",
        "4ab13be5ed12f4954d1f46cc6298e1adb2d6681b"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 14:24:21 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 14:24:21 2025 -0800"
      },
      "message": "Merge tag \u0027x86_entry_for_6.19-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 entry update from Dave Hansen:\n \"This one is pretty trivial: fix a badly-named FRED data structure\n  member\"\n\n* tag \u0027x86_entry_for_6.19-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  x86/fred: Fix 64bit identifier in fred_ss\n"
    },
    {
      "commit": "e2aa39b368bb147afe8f6bd63d962494354f6498",
      "tree": "e67312ffac4acb437918cf0b887d333fbe629c3a",
      "parents": [
        "54de197c9a5e8f522cb0a472e68e3e9888c91aa3",
        "6276c67f2bc4aeaf350a7cf889c33c38b3330ea9"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 14:16:42 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 14:16:42 2025 -0800"
      },
      "message": "Merge tag \u0027x86_misc_for_6.19-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull misc x86 updates from Dave Hansen:\n \"The most significant are some changes to ensure that symbols exported\n  for KVM are used only by KVM modules themselves, along with some\n  related cleanups.\n\n  In true x86/misc fashion, the other patch is completely unrelated and\n  just enhances an existing pr_warn() to make it clear to users how they\n  have tainted their kernel when something is mucking with MSRs.\n\n  Summary:\n\n   - Make MSR-induced taint easier for users to track down\n\n   - Restrict KVM-specific exports to KVM itself\"\n\n* tag \u0027x86_misc_for_6.19-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  x86: Restrict KVM-induced symbol exports to KVM modules where obvious/possible\n  x86/mm: Drop unnecessary export of \"ptdump_walk_pgd_level_debugfs\"\n  x86/mtrr: Drop unnecessary export of \"mtrr_state\"\n  x86/bugs: Drop unnecessary export of \"x86_spec_ctrl_base\"\n  x86/msr: Add CPU_OUT_OF_SPEC taint name to \"unrecognized\" pr_warn(msg)\n"
    },
    {
      "commit": "54de197c9a5e8f522cb0a472e68e3e9888c91aa3",
      "tree": "9d144a34a075bc75ae6989f22be7ea4a3accada0",
      "parents": [
        "c76431e3b53a426de565fa75f8278c59ca6f5ed6",
        "f2f22721aca46cebb63c589eefda843721908833"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 14:03:05 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 14:03:05 2025 -0800"
      },
      "message": "Merge tag \u0027x86_sgx_for_6.19-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 SGX updates from Dave HansenL\n \"The main content here is adding support for the new EUPDATESVN SGX\n  ISA. Before this, folks who updated microcode had to reboot before\n  enclaves could attest to the new microcode. The new functionality lets\n  them do this without a reboot.\n\n  The rest are some nice, but relatively mundane comment and kernel-doc\n  fixups.\n\n  Summary:\n\n   - Allow security version (SVN) updates so enclaves can attest to new\n     microcode\n\n   - Fix kernel docs typos\"\n\n* tag \u0027x86_sgx_for_6.19-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  x86/sgx: Fix a typo in the kernel-doc comment for enum sgx_attribute\n  x86/sgx: Remove superfluous asterisk from copyright comment in asm/sgx.h\n  x86/sgx: Document structs and enums with \u0027@\u0027, not \u0027%\u0027\n  x86/sgx: Add kernel-doc descriptions for params passed to vDSO user handler\n  x86/sgx: Add a missing colon in kernel-doc markup for \"struct sgx_enclave_run\"\n  x86/sgx: Enable automatic SVN updates for SGX enclaves\n  x86/sgx: Implement ENCLS[EUPDATESVN]\n  x86/sgx: Define error codes for use by ENCLS[EUPDATESVN]\n  x86/cpufeatures: Add X86_FEATURE_SGX_EUPDATESVN feature flag\n  x86/sgx: Introduce functions to count the sgx_(vepc_)open()\n"
    },
    {
      "commit": "c76431e3b53a426de565fa75f8278c59ca6f5ed6",
      "tree": "a1f6ed07398ea65a587720c02144fef80d0344ed",
      "parents": [
        "a9a10e920e2e1ece946c584aa68f8a4fcedab791",
        "84361123413efc84b06f3441c6c827b95d902732"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 13:32:52 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 13:32:52 2025 -0800"
      },
      "message": "Merge tag \u0027x86_mm_for_v6.19_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 mm updates from Borislav Petkov:\n\n - Use the proper accessors when reading CR3 as part of the page level\n   transitions (5-level to 4-level, the use case being kexec) so that\n   only the physical address in CR3 is picked up and not flags which are\n   above the physical mask shift\n\n - Clean up and unify __phys_addr_symbol() definitions\n\n* tag \u0027x86_mm_for_v6.19_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  efi/libstub: Fix page table access in 5-level to 4-level paging transition\n  x86/boot: Fix page table access in 5-level to 4-level paging transition\n  x86/mm: Unify __phys_addr_symbol()\n"
    },
    {
      "commit": "a9a10e920e2e1ece946c584aa68f8a4fcedab791",
      "tree": "7dfe00ca154b274c9ce7836e435fa3bad4f44e9d",
      "parents": [
        "cb502f0e5e87f769417691c7060a358a8b117c22",
        "e67997021fd0d73f230ee0473da3ad4e3d3ce37c"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 13:27:09 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 13:27:09 2025 -0800"
      },
      "message": "Merge tag \u0027x86_bugs_for_v6.19_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 CPU mitigation updates from Borislav Petkov:\n\n - Convert the tsx\u003d cmdline parsing to use early_param()\n\n - Cleanup forward declarations gunk in bugs.c\n\n* tag \u0027x86_bugs_for_v6.19_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  x86/bugs: Get rid of the forward declarations\n  x86/tsx: Get the tsx\u003d command line parameter with early_param()\n  x86/tsx: Make tsx_ctrl_state static\n"
    },
    {
      "commit": "cb502f0e5e87f769417691c7060a358a8b117c22",
      "tree": "6b53851f072f46a7b49fea35049736870a22100e",
      "parents": [
        "d7489818348e5374a8a12a309183fa44b3df41cc",
        "73029e73ccd07b64905f441d4f474a9bb91e7027"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 13:07:53 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 13:07:53 2025 -0800"
      },
      "message": "Merge tag \u0027x86_sev_for_v6.19_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 SEV updates from Borislav Petkov:\n\n - Largely cleanups along with a change to save XSS to the GHCB\n   (Guest-Host Communication Block) in SEV-ES guests so that the\n   hypervisor can determine the guest\u0027s XSAVES buffer size properly\n   and thus support shadow stacks in AMD confidential guests\n\n* tag \u0027x86_sev_for_v6.19_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  x86/cc: Fix enum spelling to fix kernel-doc warnings\n  x86/boot: Drop unused sev_enable() fallback\n  x86/coco/sev: Convert has_cpuflag() to use cpu_feature_enabled()\n  x86/sev: Include XSS value in GHCB CPUID request\n  x86/boot: Move boot_*msr helpers to asm/shared/msr.h\n"
    },
    {
      "commit": "d7489818348e5374a8a12a309183fa44b3df41cc",
      "tree": "e0ea80f7b5f7a272bb6542e62c35a1b3d4add6d6",
      "parents": [
        "2ae20d651091c71ef182d28cbf10ce6f8be79c99",
        "d911fe6e942e60900577314dc1f1529b90e4da07"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 12:17:47 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 12:17:47 2025 -0800"
      },
      "message": "Merge tag \u0027x86_cleanups_for_v6.19_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 cleanups from Borislav Petkov:\n\n - The mandatory pile of cleanups the cat drags in every merge window\n\n* tag \u0027x86_cleanups_for_v6.19_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  x86/boot: Clean up whitespace in a20.c\n  x86/mm: Delete disabled debug code\n  x86/{boot,mtrr}: Remove unused function declarations\n  x86/percpu: Use BIT_WORD() and BIT_MASK() macros\n  x86/cpufeatures: Correct LKGS feature flag description\n  x86/idtentry: Add missing \u0027*\u0027 to kernel-doc lines\n"
    },
    {
      "commit": "2ae20d651091c71ef182d28cbf10ce6f8be79c99",
      "tree": "29099e23cd97968f4b3a2c144b4e5f20a93924dd",
      "parents": [
        "2a47c26e55a2bc085a2349ed1d4e065ee298155f",
        "ac7de456a37f9b126eb53b89c2bb27d625dc5fd9"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 11:55:58 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 11:55:58 2025 -0800"
      },
      "message": "Merge tag \u0027x86_cache_for_v6.19_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 resource control updates from Borislav Petkov:\n\n - Add support for AMD\u0027s Smart Data Cache Injection feature which allows\n   for direct insertion of data from I/O devices into the L3 cache, thus\n   bypassing DRAM and saving its bandwidth; the resctrl side of the\n   feature allows the size of the L3 used for data injection to be\n   controlled\n\n - Add Intel Clearwater Forest to the list of CPUs which support\n   Sub-NUMA clustering\n\n - Other fixes and cleanups\n\n* tag \u0027x86_cache_for_v6.19_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  fs/resctrl: Update bit_usage to reflect io_alloc\n  fs/resctrl: Introduce interface to modify io_alloc capacity bitmasks\n  fs/resctrl: Modify struct rdt_parse_data to pass mode and CLOSID\n  fs/resctrl: Introduce interface to display io_alloc CBMs\n  fs/resctrl: Add user interface to enable/disable io_alloc feature\n  fs/resctrl: Introduce interface to display \"io_alloc\" support\n  x86,fs/resctrl: Implement \"io_alloc\" enable/disable handlers\n  x86,fs/resctrl: Detect io_alloc feature\n  x86/resctrl: Add SDCIAE feature in the command line options\n  x86/cpufeatures: Add support for L3 Smart Data Cache Injection Allocation Enforcement\n  fs/resctrl: Consider sparse masks when initializing new group\u0027s allocation\n  x86/resctrl: Support Sub-NUMA Cluster (SNC) mode on Clearwater Forest\n"
    },
    {
      "commit": "2a47c26e55a2bc085a2349ed1d4e065ee298155f",
      "tree": "5d9b464e577b3218ae93f1e99fc8fe357908751b",
      "parents": [
        "a61288200e8b6f42bff116508dc72ebcc206f10a",
        "ca8313fd83399ea1d18e695c2ae9b259985c9e1f"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 11:35:49 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 11:35:49 2025 -0800"
      },
      "message": "Merge tag \u0027x86_microcode_for_v6.19_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 microcode loading updates from Borislav Petkov:\n\n - Add microcode staging support on Intel: it moves the sole microcode\n   blobs loading to a non-critical path so that microcode loading\n   latencies are kept at minimum. The actual \"directing\" the hardware to\n   load microcode is the only step which is done on the critical path.\n\n   This scheme is also opportunistic as in: on a failure, the machinery\n   falls back to normal loading\n\n - Add the capability to the AMD side of the loader to select one of two\n   per-family/model/stepping patches: one is pre-Entrysign and the other\n   is post-Entrysign; with the goal to take care of machines which\n   haven\u0027t updated their BIOS yet - something they should absolutely do\n   as this is the only proper Entrysign fix\n\n - Other small cleanups and fixlets\n\n* tag \u0027x86_microcode_for_v6.19_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  x86/microcode: Mark early_parse_cmdline() as __init\n  x86/microcode/AMD: Select which microcode patch to load\n  x86/microcode/intel: Enable staging when available\n  x86/microcode/intel: Support mailbox transfer\n  x86/microcode/intel: Implement staging handler\n  x86/microcode/intel: Define staging state struct\n  x86/microcode/intel: Establish staging control logic\n  x86/microcode: Introduce staging step to reduce late-loading time\n  x86/cpu/topology: Make primary thread mask available with SMP\u003dn\n"
    },
    {
      "commit": "a61288200e8b6f42bff116508dc72ebcc206f10a",
      "tree": "a959547d2ce66db015808d7b2df4fee754d4cab8",
      "parents": [
        "49219bba0149157774b7091c3ea9ad22b2114285",
        "5c4663ed1eac01987a1421f059380db48ab7b1a3"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 11:04:37 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 11:04:37 2025 -0800"
      },
      "message": "Merge tag \u0027ras_core_for_v6.19_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 RAS updates from Borislav Petkov:\n\n - The second part of the AMD MCA interrupts rework after the\n   last-minute show-stopper from the last merge window was sorted out.\n   After this, the AMD MCA deferred errors, thresholding and corrected\n   errors interrupt handlers use common MCA code and are tightly\n   integrated into the core MCA code, thereby getting rid of\n   considerable duplication. All culminating into allowing CMCI error\n   thresholding storms to be detected at AMD too, using the common\n   infrastructure\n\n - Add support for two new MCA bank bits on AMD Zen6 which denote\n   whether the error address logged is a system physical address, which\n   obviates the need for it to be translated before further error\n   recovery can be done\n\n* tag \u0027ras_core_for_v6.19_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  x86/mce: Handle AMD threshold interrupt storms\n  x86/mce: Do not clear bank\u0027s poll bit in mce_poll_banks on AMD SMCA systems\n  x86/mce: Add support for physical address valid bit\n  x86/mce: Save and use APEI corrected threshold limit\n  x86/mce/amd: Define threshold restart function for banks\n  x86/mce/amd: Remove redundant reset_block()\n  x86/mce/amd: Support SMCA Corrected Error Interrupt\n  x86/mce/amd: Enable interrupt vectors once per-CPU on SMCA systems\n  x86/mce: Unify AMD DFR handler with MCA Polling\n  x86/mce: Unify AMD THR handler with MCA Polling\n"
    },
    {
      "commit": "49219bba0149157774b7091c3ea9ad22b2114285",
      "tree": "6a310ca45d8e7e426e457d41899396ff12670062",
      "parents": [
        "7f8d5f70fffe2177afcc62f02feead5827dfe8dd",
        "e2349c5811ae642f6d948d1ee4bfc4a93566c178"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 10:45:50 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 10:45:50 2025 -0800"
      },
      "message": "Merge tag \u0027edac_updates_for_v6.19_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras\n\nPull EDAC updates from Borislav Petkov:\n\n - imh_edac: Add a new EDAC driver for Intel Diamond Rapids and future\n   incarnations of this memory controllers architecture\n\n - amd64_edac: Remove the legacy csrow sysfs interface which has been\n   deprecated and unused (we assume) for at least a decade\n\n - Add the capability to fallback to BIOS-provided address translation\n   functionality (ACPI PRM) which can be used on systems unsupported by\n   the current AMD address translation library\n\n - The usual fixes, fixlets, cleanups and improvements all over the\n   place\n\n* tag \u0027edac_updates_for_v6.19_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras:\n  RAS/AMD/ATL: Replace bitwise_xor_bits() with hweight16()\n  EDAC/igen6: Fix error handling in igen6_edac driver\n  EDAC/imh: Setup \u0027imh_test\u0027 debugfs testing node\n  EDAC/{skx_comm,imh}: Detect 2-level memory configuration\n  EDAC/skx_common: Extend the maximum number of DRAM chip row bits\n  EDAC/{skx_common,imh}: Add EDAC driver for Intel Diamond Rapids servers\n  EDAC/skx_common: Prepare for skx_set_hi_lo()\n  EDAC/skx_common: Prepare for skx_get_edac_list()\n  EDAC/{skx_common,skx,i10nm}: Make skx_register_mci() independent of pci_dev\n  EDAC/ghes: Replace deprecated strcpy() in ghes_edac_report_mem_error()\n  EDAC/ie31200: Fix error handling in ie31200_register_mci\n  RAS/CEC: Replace use of system_wq with system_percpu_wq\n  EDAC: Remove the legacy EDAC sysfs interface\n  EDAC/amd64: Remove NUM_CONTROLLERS macro\n  EDAC/amd64: Generate ctl_name string at runtime\n  RAS/AMD/ATL: Require PRM support for future systems\n  ACPI: PRM: Add acpi_prm_handler_available()\n  RAS/AMD/ATL: Return error codes from helper functions\n"
    },
    {
      "commit": "7f8d5f70fffe2177afcc62f02feead5827dfe8dd",
      "tree": "8375d14b04532154664dc137a15c50d5f9022a50",
      "parents": [
        "d42e504a555d0da2a10001e697f0c8a7f633fb05",
        "70e0a80a1f3580ccf5bc1f34dbb433c67d9d8d00"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 10:18:49 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 10:18:49 2025 -0800"
      },
      "message": "Merge tag \u0027core-core-2025-12-03\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull core irq cleanup from Thomas Gleixner:\n \"Tree wide cleanup of the remaining users of in_irq() which got\n  replaced by in_hardirq() and marked deprecated in 2020\"\n\n* tag \u0027core-core-2025-12-03\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  treewide: Remove in_irq()\n"
    },
    {
      "commit": "d42e504a555d0da2a10001e697f0c8a7f633fb05",
      "tree": "0163b100b5e983cd10040b5bc6e277d48a503c30",
      "parents": [
        "5028f42416eaec08d3f6aa4f98ccca669b3f8ab3",
        "7dec062cfcf27808dbb70a0b231d1a698792743d"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 09:58:33 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 09:58:33 2025 -0800"
      },
      "message": "Merge tag \u0027timers-core-2025-11-30\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull timer core updates from Thomas Gleixner:\n\n - Prevent a thundering herd problem when the timekeeper CPU is delayed\n   and a large number of CPUs compete to acquire jiffies_lock to do the\n   update. Limit it to one CPU with a separate \"uncontended\" atomic\n   variable.\n\n - A set of improvements for the timer migration mechanism:\n\n     - Support imbalanced NUMA trees correctly\n\n     - Support dynamic exclusion of CPUs from the migrator duty to allow\n       the cpuset/isolation mechanism to exclude them from handling\n       timers of remote idle CPUs\n\n - The usual small updates, cleanups and enhancements\n\n* tag \u0027timers-core-2025-11-30\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  timers/migration: Exclude isolated cpus from hierarchy\n  cpumask: Add initialiser to use cleanup helpers\n  sched/isolation: Force housekeeping if isolcpus and nohz_full don\u0027t leave any\n  cgroup/cpuset: Rename update_unbound_workqueue_cpumask() to update_isolation_cpumasks()\n  timers/migration: Use scoped_guard on available flag set/clear\n  timers/migration: Add mask for CPUs available in the hierarchy\n  timers/migration: Rename \u0027online\u0027 bit to \u0027available\u0027\n  selftests/timers/nanosleep: Add tests for return of remaining time\n  selftests/timers: Clean up kernel version check in posix_timers\n  time: Fix a few typos in time[r] related code comments\n  time: tick-oneshot: Add missing Return and parameter descriptions to kernel-doc\n  hrtimer: Store time as ktime_t in restart block\n  timers/migration: Remove dead code handling idle CPU checking for remote timers\n  timers/migration: Remove unused \"cpu\" parameter from tmigr_get_group()\n  timers/migration: Assert that hotplug preparing CPU is part of stable active hierarchy\n  timers/migration: Fix imbalanced NUMA trees\n  timers/migration: Remove locking on group connection\n  timers/migration: Convert \"while\" loops to use \"for\"\n  tick/sched: Limit non-timekeeper CPUs calling jiffies update\n"
    },
    {
      "commit": "5028f42416eaec08d3f6aa4f98ccca669b3f8ab3",
      "tree": "7c18162469033014f80c30b3acfaedd7eb520178",
      "parents": [
        "9ce62ebbb7b24c28a1781f6b3bc6fdcd0b961392",
        "2437f798809d4420350b0118e4723024ce8d203b"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 09:54:27 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 09:54:27 2025 -0800"
      },
      "message": "Merge tag \u0027timers-clocksource-2025-11-30\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull clocksource updates from Thomas Gleixner:\n \"Updates for clocksource and clockevent drivers:\n\n   - A new driver for the Realtel system timer\n\n   - Prevent the unbinding of timers when the drivers do not support\n     that\n\n   - Expand the timer counter readout for the SPRD driver to 64 bit\n     to allow IOT devices suspend times of more than 36 hours, which\n     is the current limit of the 32-bi readout\n\n   - The usual small cleanups, fixes and enhancements all over the\n     place\"\n\n* tag \u0027timers-clocksource-2025-11-30\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  clocksource/drivers: Add Realtek system timer driver\n  dt-bindings: timer: Add Realtek SYSTIMER\n  clocksource/drivers/stm32-lp: Drop unused module alias\n  clocksource/drivers/rda: Add sched_clock_register for RDA8810PL SoC\n  clocksource/drivers/nxp-stm: Prevent driver unbind\n  clocksource/drivers/nxp-pit: Prevent driver unbind\n  clocksource/drivers/arm_arch_timer_mmio: Prevent driver unbind\n  clocksource/drivers/nxp-stm: Fix section mismatches\n  clocksource/drivers/sh_cmt: Always leave device running after probe\n  clocksource/drivers/stm: Fix double deregistration on probe failure\n  clocksource/drivers/ralink: Fix resource leaks in init error path\n  clocksource/drivers/timer-sp804: Fix read_current_timer() issue when clock source is not registered\n  clocksource/drivers/sprd: Enable register for timer counter from 32 bit to 64 bit\n"
    },
    {
      "commit": "9ce62ebbb7b24c28a1781f6b3bc6fdcd0b961392",
      "tree": "371ad0a3ea2ecd465acaa191bcf835f20f3732d4",
      "parents": [
        "15b87bec89cb227b55b3689bf5de31b85cf88559",
        "9c1fbc56ca0a98113e75dcc5030103a02eff8897"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 09:35:59 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 09:35:59 2025 -0800"
      },
      "message": "Merge tag \u0027irq-msi-2025-11-30\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull MSI updates from Thomas Gleixner:\n \"Updates for [PCI] MSI related code:\n\n   - Remove one variant of PCI/MSI management as all users have been\n     converted to use per device domains. That reduces the variants to\n     two:\n\n     The modern and the real archaic legacy variant, which keeps the\n     usual suspects in the museum category alive.\n\n   - Rework the platform MSI device ID detection mechanism in the ARM\n     GIC world to address resource leaks, duplicated code and other\n     details. This requires a corresponding preparatory step in the\n     PCI/iproc driver.\n\n   - Trivial core code cleanups\"\n\n* tag \u0027irq-msi-2025-11-30\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  irqchip/gic-its: Rework platform MSI deviceID detection\n  PCI: iproc: Implement MSI controller node detection with of_msi_xlate()\n  genirq/msi: Slightly simplify msi_domain_alloc()\n  PCI/MSI: Delete pci_msi_create_irq_domain()\n"
    },
    {
      "commit": "15b87bec89cb227b55b3689bf5de31b85cf88559",
      "tree": "91cee4d4693d5e48612660ac829aeb420b7fa1d2",
      "parents": [
        "6863c8385c28ba4d30391f1d1fce39f62581d362",
        "fc584d871c1641949594ccf12a48bb226636b189"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 09:32:53 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 09:32:53 2025 -0800"
      },
      "message": "Merge tag \u0027irq-drivers-2025-11-30\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull irq driver updates from Thomas Gleixner:\n \"Boring updates for interrupt drivers:\n\n   - Support for a couple of new ARM64 and RISCV SoC variants and their\n     magic interrupt controllers which either can reuse existing code or\n     require quirks due to a botched hardware implementation\n\n   - More section mismatch fixes\n\n   - The usual cleanups and fixes all over the place\"\n\n* tag \u0027irq-drivers-2025-11-30\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (32 commits)\n  irqchip/meson-gpio: Add support for Amlogic S6 S7 and S7D SoCs\n  dt-bindings: interrupt-controller: Add support for Amlogic S6 S7 and S7D SoCs\n  dt-bindings: interrupt-controller: aspeed,ast2700: Correct #interrupt-cells and interrupts count\n  irqchip/aclint-sswi: Add Nuclei UX900 support\n  dt-bindings: interrupt-controller: Add Anlogic DR1V90 ACLINT SSWI\n  dt-bindings: interrupt-controller: Add Anlogic DR1V90 ACLINT MSWI\n  dt-bindings: interrupt-controller: Add Anlogic DR1V90 PLIC\n  irqchip/irq-bcm7038-l1: Remove unused reg_mask_status()\n  irqchip/sifive-plic: Fix call to __plic_toggle() in M-Mode code path\n  irqchip/sifive-plic: Add support for UltraRISC DP1000 PLIC\n  irqchip/sifive-plic: Cache the interrupt enable state\n  dt-bindings: interrupt-controller: Add UltraRISC DP1000 PLIC\n  dt-bindings: vendor-prefixes: Add UltraRISC\n  irqchip/qcom-irq-combiner: Rename driver structure\n  irqchip/riscv-imsic: Inline imsic_vector_from_local_id()\n  irqchip/riscv-imsic: Embed the vector array in lpriv\n  irqchip/riscv-imsic: Remove redundant irq_data lookups\n  irqchip/ts4800: Drop unused module alias\n  irqchip/mvebu-pic: Drop unused module alias\n  irqchip/meson-gpio: Drop unused module alias\n  ...\n"
    },
    {
      "commit": "6863c8385c28ba4d30391f1d1fce39f62581d362",
      "tree": "57dbeb682d2d1992657cb463a4c66e5e90a35303",
      "parents": [
        "312f5b18663a6dd8d8e6ad5a3fce0059a11686b2",
        "3de5e46e50abc01a1cee7e12b657e083fc5ed638"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 09:14:26 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 09:14:26 2025 -0800"
      },
      "message": "Merge tag \u0027irq-core-2025-11-30\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull irq core updates from Thomas Gleixner:\n \"Updates for the interrupt core and treewide cleanups:\n\n   - Rework of the Per Processor Interrupt (PPI) management on ARM[64]\n\n     PPI support was built under the assumption that the systems are\n     homogenous so that the same CPU local device types are connected to\n     them. That\u0027s unfortunately wishful thinking and created horrible\n     workarounds.\n\n     This rework provides affinity management for PPIs so that they can\n     be individually configured in the firmware tables and mops up the\n     related drivers all over the place.\n\n   - Prevent CPUSET/isolation changes to arbitrarily affine interrupt\n     threads to random CPUs, which ignores user or driver settings.\n\n   - Plug a harmless race in the interrupt affinity proc interface,\n     which allows to see a half updated mask\n\n   - Adjust the priority of secondary interrupt threads on RT, so that\n     the combination of primary and secondary thread emulates the\n     hardware interrupt plus thread scenario. Having them at the same\n     priority can cause starvation issues in some drivers\"\n\n* tag \u0027irq-core-2025-11-30\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (33 commits)\n  genirq: Remove cpumask availability check on kthread affinity setting\n  genirq: Fix interrupt threads affinity vs. cpuset isolated partitions\n  genirq: Prevent early spurious wake-ups of interrupt threads\n  genirq: Use raw_spinlock_irq() in irq_set_affinity_notifier()\n  genirq/manage: Reduce priority of forced secondary interrupt handler\n  genirq/proc: Fix race in show_irq_affinity()\n  genirq: Fix percpu_devid irq affinity documentation\n  perf: arm_pmu: Kill last use of per-CPU cpu_armpmu pointer\n  irqdomain: Kill of_node_to_fwnode() helper\n  genirq: Kill irq_{g,s}et_percpu_devid_partition()\n  irqchip: Kill irq-partition-percpu\n  irqchip/apple-aic: Drop support for custom PMU irq partitions\n  irqchip/gic-v3: Drop support for custom PPI partitions\n  coresight: trbe: Request specific affinities for per CPU interrupts\n  perf: arm_spe_pmu: Request specific affinities for per CPU interrupts\n  perf: arm_pmu: Request specific affinities for per CPU NMIs/interrupts\n  genirq: Add request_percpu_irq_affinity() helper\n  genirq: Allow per-cpu interrupt sharing for non-overlapping affinities\n  genirq: Update request_percpu_nmi() to take an affinity\n  genirq: Add affinity to percpu_devid interrupt requests\n  ...\n"
    },
    {
      "commit": "312f5b18663a6dd8d8e6ad5a3fce0059a11686b2",
      "tree": "808231b89754e9a758b74aa0b1c28d22e239a3b3",
      "parents": [
        "2b09f480f0a1e68111ae36a7be9aa1c93e067255",
        "37de2dbc318ee10577c1c2704de5a803e75e55a2"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 09:07:48 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 09:07:48 2025 -0800"
      },
      "message": "Merge tag \u0027core-debugobjects-2025-11-30\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull debugobjects update from Thomas Gleixner:\n \"Two small updates for debugobjects:\n\n   - Allow pool refill on RT enabled kernels before the scheduler is up\n     and running to prevent pool exhaustion\n\n   - Correct the lockdep override to prevent false positives\"\n\n* tag \u0027core-debugobjects-2025-11-30\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  debugobjects: Use LD_WAIT_CONFIG instead of LD_WAIT_SLEEP\n  debugobjects: Allow to refill the pool before SYSTEM_SCHEDULING\n"
    },
    {
      "commit": "2b09f480f0a1e68111ae36a7be9aa1c93e067255",
      "tree": "1121380528c9a1b89bced4ccf64bf6340b61a87a",
      "parents": [
        "1dce50698a5ceedaca806e0a78573886a363dc95",
        "653fda7ae73d8033dedb65537acac0c2c287dc3f"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 08:48:53 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 08:48:53 2025 -0800"
      },
      "message": "Merge tag \u0027core-rseq-2025-11-30\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull rseq updates from Thomas Gleixner:\n \"A large overhaul of the restartable sequences and CID management:\n\n  The recent enablement of RSEQ in glibc resulted in regressions which\n  are caused by the related overhead. It turned out that the decision to\n  invoke the exit to user work was not really a decision. More or less\n  each context switch caused that. There is a long list of small issues\n  which sums up nicely and results in a 3-4% regression in I/O\n  benchmarks.\n\n  The other detail which caused issues due to extra work in context\n  switch and task migration is the CID (memory context ID) management.\n  It also requires to use a task work to consolidate the CID space,\n  which is executed in the context of an arbitrary task and results in\n  sporadic uncontrolled exit latencies.\n\n  The rewrite addresses this by:\n\n   - Removing deprecated and long unsupported functionality\n\n   - Moving the related data into dedicated data structures which are\n     optimized for fast path processing.\n\n   - Caching values so actual decisions can be made\n\n   - Replacing the current implementation with a optimized inlined\n     variant.\n\n   - Separating fast and slow path for architectures which use the\n     generic entry code, so that only fault and error handling goes into\n     the TIF_NOTIFY_RESUME handler.\n\n   - Rewriting the CID management so that it becomes mostly invisible in\n     the context switch path. That moves the work of switching modes\n     into the fork/exit path, which is a reasonable tradeoff. That work\n     is only required when a process creates more threads than the\n     cpuset it is allowed to run on or when enough threads exit after\n     that. An artificial thread pool benchmarks which triggers this did\n     not degrade, it actually improved significantly.\n\n     The main effect in migration heavy scenarios is that runqueue lock\n     held time and therefore contention goes down significantly\"\n\n* tag \u0027core-rseq-2025-11-30\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (54 commits)\n  sched/mmcid: Switch over to the new mechanism\n  sched/mmcid: Implement deferred mode change\n  irqwork: Move data struct to a types header\n  sched/mmcid: Provide CID ownership mode fixup functions\n  sched/mmcid: Provide new scheduler CID mechanism\n  sched/mmcid: Introduce per task/CPU ownership infrastructure\n  sched/mmcid: Serialize sched_mm_cid_fork()/exit() with a mutex\n  sched/mmcid: Provide precomputed maximal value\n  sched/mmcid: Move initialization out of line\n  signal: Move MMCID exit out of sighand lock\n  sched/mmcid: Convert mm CID mask to a bitmap\n  cpumask: Cache num_possible_cpus()\n  sched/mmcid: Use cpumask_weighted_or()\n  cpumask: Introduce cpumask_weighted_or()\n  sched/mmcid: Prevent pointless work in mm_update_cpus_allowed()\n  sched/mmcid: Move scheduler code out of global header\n  sched: Fixup whitespace damage\n  sched/mmcid: Cacheline align MM CID storage\n  sched/mmcid: Use proper data structures\n  sched/mmcid: Revert the complex CID management\n  ...\n"
    },
    {
      "commit": "1dce50698a5ceedaca806e0a78573886a363dc95",
      "tree": "6ce44b381ce5935c0f5546534b87666ab1140bf5",
      "parents": [
        "4a26e7032d7d57c998598c08a034872d6f0d3945",
        "4322c8f81c58da493a3c46eda32f0e7534a350a0"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 08:01:39 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Tue Dec 02 08:01:39 2025 -0800"
      },
      "message": "Merge tag \u0027core-uaccess-2025-11-30\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull scoped user access updates from Thomas Gleixner:\n \"Scoped user mode access and related changes:\n\n   - Implement the missing u64 user access function on ARM when\n     CONFIG_CPU_SPECTRE\u003dn.\n\n     This makes it possible to access a 64bit value in generic code with\n     [unsafe_]get_user(). All other architectures and ARM variants\n     provide the relevant accessors already.\n\n   - Ensure that ASM GOTO jump label usage in the user mode access\n     helpers always goes through a local C scope label indirection\n     inside the helpers.\n\n     This is required because compilers are not supporting that a ASM\n     GOTO target leaves a auto cleanup scope. GCC silently fails to emit\n     the cleanup invocation and CLANG fails the build.\n\n     [ Editor\u0027s note: gcc-16 will have fixed the code generation issue\n       in commit f68fe3ddda4 (\"eh: Invoke cleanups/destructors in asm\n       goto jumps [PR122835]\"). But we obviously have to deal with clang\n       and older versions of gcc, so.. - Linus ]\n\n     This provides generic wrapper macros and the conversion of affected\n     architecture code to use them.\n\n   - Scoped user mode access with auto cleanup\n\n     Access to user mode memory can be required in hot code paths, but\n     if it has to be done with user controlled pointers, the access is\n     shielded with a speculation barrier, so that the CPU cannot\n     speculate around the address range check. Those speculation\n     barriers impact performance quite significantly.\n\n     This cost can be avoided by \"masking\" the provided pointer so it is\n     guaranteed to be in the valid user memory access range and\n     otherwise to point to a guaranteed unpopulated address space. This\n     has to be done without branches so it creates an address dependency\n     for the access, which the CPU cannot speculate ahead.\n\n     This results in repeating and error prone programming patterns:\n\n       \t    if (can_do_masked_user_access())\n                      from \u003d masked_user_read_access_begin((from));\n              else if (!user_read_access_begin(from, sizeof(*from)))\n                      return -EFAULT;\n              unsafe_get_user(val, from, Efault);\n              user_read_access_end();\n              return 0;\n        Efault:\n              user_read_access_end();\n              return -EFAULT;\n\n      which can be replaced with scopes and automatic cleanup:\n\n              scoped_user_read_access(from, Efault)\n                      unsafe_get_user(val, from, Efault);\n              return 0;\n         Efault:\n              return -EFAULT;\n\n   - Convert code which implements the above pattern over to\n     scope_user.*.access(). This also corrects a couple of imbalanced\n     masked_*_begin() instances which are harmless on most\n     architectures, but prevent PowerPC from implementing the masking\n     optimization.\n\n   - Add a missing speculation barrier in copy_from_user_iter()\"\n\n* tag \u0027core-uaccess-2025-11-30\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  lib/strn*,uaccess: Use masked_user_{read/write}_access_begin when required\n  scm: Convert put_cmsg() to scoped user access\n  iov_iter: Add missing speculation barrier to copy_from_user_iter()\n  iov_iter: Convert copy_from_user_iter() to masked user access\n  select: Convert to scoped user access\n  x86/futex: Convert to scoped user access\n  futex: Convert to get/put_user_inline()\n  uaccess: Provide put/get_user_inline()\n  uaccess: Provide scoped user access regions\n  arm64: uaccess: Use unsafe wrappers for ASM GOTO\n  s390/uaccess: Use unsafe wrappers for ASM GOTO\n  riscv/uaccess: Use unsafe wrappers for ASM GOTO\n  powerpc/uaccess: Use unsafe wrappers for ASM GOTO\n  x86/uaccess: Use unsafe wrappers for ASM GOTO\n  uaccess: Provide ASM GOTO safe wrappers for unsafe_*_user()\n  ARM: uaccess: Implement missing __get_user_asm_dword()\n"
    },
    {
      "commit": "4a26e7032d7d57c998598c08a034872d6f0d3945",
      "tree": "0b2f0b2508c4cbdaa05b4cec2b77e8bfc846945a",
      "parents": [
        "dcd8637edb873bd940e6aa82417dfb33ae980778",
        "b0a848f4a47a91a6b4d9a9fd739d047254df5a91"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 21:33:01 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 21:33:01 2025 -0800"
      },
      "message": "Merge tag \u0027core-bugs-2025-12-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull bug handling infrastructure updates from Ingo Molnar:\n \"Core updates:\n\n   - Improve WARN(), which has vararg printf like arguments, to work\n     with the x86 #UD based WARN-optimizing infrastructure by hiding the\n     format in the bug_table and replacing this first argument with the\n     address of the bug-table entry, while making the actual function\n     that\u0027s called a UD1 instruction (Peter Zijlstra)\n\n   - Introduce the CONFIG_DEBUG_BUGVERBOSE_DETAILED Kconfig switch (Ingo\n     Molnar, s390 support by Heiko Carstens)\n\n  Fixes and cleanups:\n\n   - bugs/s390: Remove private WARN_ON() implementation (Heiko Carstens)\n\n   - \u003casm/bugs.h\u003e: Make i386 use GENERIC_BUG_RELATIVE_POINTERS (Peter\n     Zijlstra)\"\n\n* tag \u0027core-bugs-2025-12-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (31 commits)\n  x86/bugs: Make i386 use GENERIC_BUG_RELATIVE_POINTERS\n  x86/bug: Fix BUG_FORMAT vs KASLR\n  x86_64/bug: Inline the UD1\n  x86/bug: Implement WARN_ONCE()\n  x86_64/bug: Implement __WARN_printf()\n  x86/bug: Use BUG_FORMAT for DEBUG_BUGVERBOSE_DETAILED\n  x86/bug: Add BUG_FORMAT basics\n  bug: Allow architectures to provide __WARN_printf()\n  bug: Implement WARN_ON() using __WARN_FLAGS()\n  bug: Add report_bug_entry()\n  bug: Add BUG_FORMAT_ARGS infrastructure\n  bug: Clean up CONFIG_GENERIC_BUG_RELATIVE_POINTERS\n  bug: Add BUG_FORMAT infrastructure\n  x86: Rework __bug_table helpers\n  bugs/s390: Remove private WARN_ON() implementation\n  bugs/core: Reorganize fields in the first line of WARNING output, add -\u003ecomm[] output\n  bugs/sh: Concatenate \u0027cond_str\u0027 with \u0027__FILE__\u0027 in __WARN_FLAGS(), to extend WARN_ON/BUG_ON output\n  bugs/parisc: Concatenate \u0027cond_str\u0027 with \u0027__FILE__\u0027 in __WARN_FLAGS(), to extend WARN_ON/BUG_ON output\n  bugs/riscv: Concatenate \u0027cond_str\u0027 with \u0027__FILE__\u0027 in __BUG_FLAGS(), to extend WARN_ON/BUG_ON output\n  bugs/riscv: Pass in \u0027cond_str\u0027 to __BUG_FLAGS()\n  ...\n"
    },
    {
      "commit": "dcd8637edb873bd940e6aa82417dfb33ae980778",
      "tree": "6de52669a1e868094ed7e50fd1edfa44e3e7c2cb",
      "parents": [
        "e7d81c1ed665e72c55ae5cf2d7601c09842b4013",
        "ced37e9ceae50e4cb6cd058963bd315ec9afa651"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 21:31:02 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 21:31:02 2025 -0800"
      },
      "message": "Merge tag \u0027x86-core-2025-12-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull core x86 updates from Ingo Molnar:\n\n - x86/alternatives: Drop unnecessary test after call to\n   alt_replace_call() (Juergen Gross)\n\n - x86/dumpstack: Prevent KASAN false positive warnings in\n   __show_regs() (Tengda Wu)\n\n* tag \u0027x86-core-2025-12-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  x86/dumpstack: Prevent KASAN false positive warnings in __show_regs()\n  x86/alternative: Drop not needed test after call of alt_replace_call()\n"
    },
    {
      "commit": "e7d81c1ed665e72c55ae5cf2d7601c09842b4013",
      "tree": "1826abfc28a9f3518f127fe321ea3ed64844e55f",
      "parents": [
        "de2f75d55ef85861ffc81c7583e0715453eb7235",
        "780813d7018067f2796f023b56b51385970be460"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 21:28:23 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 21:28:23 2025 -0800"
      },
      "message": "Merge tag \u0027x86-build-2025-12-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 math-emu fix from Ingo Molnar:\n \"A single fix for an ancient prototype in the math-emu code, by Arnd\n  Bergmann\"\n\n* tag \u0027x86-build-2025-12-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  x86/math-emu: Fix div_Xsig() prototype\n"
    },
    {
      "commit": "de2f75d55ef85861ffc81c7583e0715453eb7235",
      "tree": "9256bc83e02b97dff5a6209329b16992afd243d4",
      "parents": [
        "6d2c10e889db596938bb7b4d3cdd42d67208439a",
        "ed4f9638d905a97cebd49ecea85cc9b706b73761"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 21:26:35 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 21:26:35 2025 -0800"
      },
      "message": "Merge tag \u0027x86-apic-2025-12-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 apic updates from Ingo Molnar:\n\n - x86/apic: Fix the frequency in apic\u003dverbose log output (Julian\n   Stecklina)\n\n - Simplify mp_irqdomain_alloc() slightly (Christophe JAILLET)\n\n* tag \u0027x86-apic-2025-12-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  x86/apic: Fix frequency in apic\u003dverbose log output\n  x86/ioapic: Simplify mp_irqdomain_alloc() slightly\n"
    },
    {
      "commit": "6d2c10e889db596938bb7b4d3cdd42d67208439a",
      "tree": "c8adec6da2f3c89e60bf30397c7ebff7ee85e288",
      "parents": [
        "6c26fbe8c9d3e932dce6afe2505b19b4b261cae9",
        "c04507ac500e2cc8048000c2a849588227554e06"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 21:04:45 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 21:04:45 2025 -0800"
      },
      "message": "Merge tag \u0027sched-core-2025-12-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull scheduler updates from Ingo Molnar:\n \"Scalability and load-balancing improvements:\n\n   - Enable scheduler feature NEXT_BUDDY (Mel Gorman)\n\n   - Reimplement NEXT_BUDDY to align with EEVDF goals (Mel Gorman)\n\n   - Skip sched_balance_running cmpxchg when balance is not due (Tim\n     Chen)\n\n   - Implement generic code for architecture specific sched domain NUMA\n     distances (Tim Chen)\n\n   - Optimize the NUMA distances of the sched-domains builds of Intel\n     Granite Rapids (GNR) and Clearwater Forest (CWF) platforms (Tim\n     Chen)\n\n   - Implement proportional newidle balance: a randomized algorithm that\n     runs newidle balancing proportional to its success rate. (Peter\n     Zijlstra)\n\n  Scheduler infrastructure changes:\n\n   - Implement the \u0027sched_change\u0027 scoped_guard() pattern for the entire\n     scheduler (Peter Zijlstra)\n\n   - More broadly utilize the sched_change guard (Peter Zijlstra)\n\n   - Add support to pick functions to take runqueue-flags (Joel\n     Fernandes)\n\n   - Provide and use set_need_resched_current() (Peter Zijlstra)\n\n  Fair scheduling enhancements:\n\n   - Forfeit vruntime on yield (Fernand Sieber)\n\n   - Only update stats for allowed CPUs when looking for dst group (Adam\n     Li)\n\n  CPU-core scheduling enhancements:\n\n   - Optimize core cookie matching check (Fernand Sieber)\n\n  Deadline scheduler fixes:\n\n   - Only set free_cpus for online runqueues (Doug Berger)\n\n   - Fix dl_server time accounting (Peter Zijlstra)\n\n   - Fix dl_server stop condition (Peter Zijlstra)\n\n  Proxy scheduling fixes:\n\n   - Yield the donor task (Fernand Sieber)\n\n  Fixes and cleanups:\n\n   - Fix do_set_cpus_allowed() locking (Peter Zijlstra)\n\n   - Fix migrate_disable_switch() locking (Peter Zijlstra)\n\n   - Remove double update_rq_clock() in __set_cpus_allowed_ptr_locked()\n     (Hao Jia)\n\n   - Increase sched_tick_remote timeout (Phil Auld)\n\n   - sched/deadline: Use cpumask_weight_and() in dl_bw_cpus() (Shrikanth\n     Hegde)\n\n   - sched/deadline: Clean up select_task_rq_dl() (Shrikanth Hegde)\"\n\n* tag \u0027sched-core-2025-12-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (44 commits)\n  sched: Provide and use set_need_resched_current()\n  sched/fair: Proportional newidle balance\n  sched/fair: Small cleanup to update_newidle_cost()\n  sched/fair: Small cleanup to sched_balance_newidle()\n  sched/fair: Revert max_newidle_lb_cost bump\n  sched/fair: Reimplement NEXT_BUDDY to align with EEVDF goals\n  sched/fair: Enable scheduler feature NEXT_BUDDY\n  sched: Increase sched_tick_remote timeout\n  sched/fair: Have SD_SERIALIZE affect newidle balancing\n  sched/fair: Skip sched_balance_running cmpxchg when balance is not due\n  sched/deadline: Minor cleanup in select_task_rq_dl()\n  sched/deadline: Use cpumask_weight_and() in dl_bw_cpus\n  sched/deadline: Document dl_server\n  sched/deadline: Fix dl_server stop condition\n  sched/deadline: Fix dl_server time accounting\n  sched/core: Remove double update_rq_clock() in __set_cpus_allowed_ptr_locked()\n  sched/eevdf: Fix min_vruntime vs avg_vruntime\n  sched/core: Add comment explaining force-idle vruntime snapshots\n  sched/core: Optimize core cookie matching check\n  sched/proxy: Yield the donor task\n  ...\n"
    },
    {
      "commit": "6c26fbe8c9d3e932dce6afe2505b19b4b261cae9",
      "tree": "81cc40ecd2cde95b1b37937cf270cc0fa3832c43",
      "parents": [
        "63e6995005be8ceb8a1d56a18df1a1a40c28356d",
        "9929dffce5ed7e2988e0274f4db98035508b16d9"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 20:42:01 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 20:42:01 2025 -0800"
      },
      "message": "Merge tag \u0027perf-core-2025-12-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull performance events updates from Ingo Molnar:\n \"Callchain support:\n\n   - Add support for deferred user-space stack unwinding for perf,\n     enabled on x86. (Peter Zijlstra, Steven Rostedt)\n\n   - unwind_user/x86: Enable frame pointer unwinding on x86 (Josh\n     Poimboeuf)\n\n  x86 PMU support and infrastructure:\n\n   - x86/insn: Simplify for_each_insn_prefix() (Peter Zijlstra)\n\n   - x86/insn,uprobes,alternative: Unify insn_is_nop() (Peter Zijlstra)\n\n  Intel PMU driver:\n\n   - Large series to prepare for and implement architectural PEBS\n     support for Intel platforms such as Clearwater Forest (CWF) and\n     Panther Lake (PTL). (Dapeng Mi, Kan Liang)\n\n   - Check dynamic constraints (Kan Liang)\n\n   - Optimize PEBS extended config (Peter Zijlstra)\n\n   - cstates:\n      - Remove PC3 support from LunarLake (Zhang Rui)\n      - Add Pantherlake support (Zhang Rui)\n      - Clearwater Forest support (Zide Chen)\n\n  AMD PMU driver:\n\n   - x86/amd: Check event before enable to avoid GPF (George Kennedy)\n\n  Fixes and cleanups:\n\n   - task_work: Fix NMI race condition (Peter Zijlstra)\n\n   - perf/x86: Fix NULL event access and potential PEBS record loss\n     (Dapeng Mi)\n\n   - Misc other fixes and cleanups (Dapeng Mi, Ingo Molnar, Peter\n     Zijlstra)\"\n\n* tag \u0027perf-core-2025-12-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (38 commits)\n  perf/x86/intel: Fix and clean up intel_pmu_drain_arch_pebs() type use\n  perf/x86/intel: Optimize PEBS extended config\n  perf/x86/intel: Check PEBS dyn_constraints\n  perf/x86/intel: Add a check for dynamic constraints\n  perf/x86/intel: Add counter group support for arch-PEBS\n  perf/x86/intel: Setup PEBS data configuration and enable legacy groups\n  perf/x86/intel: Update dyn_constraint base on PEBS event precise level\n  perf/x86/intel: Allocate arch-PEBS buffer and initialize PEBS_BASE MSR\n  perf/x86/intel: Process arch-PEBS records or record fragments\n  perf/x86/intel/ds: Factor out PEBS group processing code to functions\n  perf/x86/intel/ds: Factor out PEBS record processing code to functions\n  perf/x86/intel: Initialize architectural PEBS\n  perf/x86/intel: Correct large PEBS flag check\n  perf/x86/intel: Replace x86_pmu.drain_pebs calling with static call\n  perf/x86: Fix NULL event access and potential PEBS record loss\n  perf/x86: Remove redundant is_x86_event() prototype\n  entry,unwind/deferred: Fix unwind_reset_info() placement\n  unwind_user/x86: Fix arch\u003dum build\n  perf: Support deferred user unwind\n  unwind_user/x86: Teach FP unwind about start of function\n  ...\n"
    },
    {
      "commit": "63e6995005be8ceb8a1d56a18df1a1a40c28356d",
      "tree": "9dc6af0e1ab47710fe21cf7cfd95b1d28b6bec27",
      "parents": [
        "b53440f8e5a1466870d7a1d255e0f9966e0041fb",
        "6ec33db1aaf06a76fb063610e668f8e12f32ebbf"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 20:18:59 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 20:18:59 2025 -0800"
      },
      "message": "Merge tag \u0027objtool-core-2025-12-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull objtool updates from Ingo Molnar:\n\n - klp-build livepatch module generation (Josh Poimboeuf)\n\n   Introduce new objtool features and a klp-build script to generate\n   livepatch modules using a source .patch as input.\n\n   This builds on concepts from the longstanding out-of-tree kpatch\n   project which began in 2012 and has been used for many years to\n   generate livepatch modules for production kernels. However, this is a\n   complete rewrite which incorporates hard-earned lessons from 12+\n   years of maintaining kpatch.\n\n   Key improvements compared to kpatch-build:\n\n    - Integrated with objtool: Leverages objtool\u0027s existing control-flow\n      graph analysis to help detect changed functions.\n\n    - Works on vmlinux.o: Supports late-linked objects, making it\n      compatible with LTO, IBT, and similar.\n\n    - Simplified code base: ~3k fewer lines of code.\n\n    - Upstream: No more out-of-tree #ifdef hacks, far less cruft.\n\n    - Cleaner internals: Vastly simplified logic for\n      symbol/section/reloc inclusion and special section extraction.\n\n    - Robust __LINE__ macro handling: Avoids false positive binary diffs\n      caused by the __LINE__ macro by introducing a fix-patch-lines\n      script which injects #line directives into the source .patch to\n      preserve the original line numbers at compile time.\n\n - Disassemble code with libopcodes instead of running objdump\n   (Alexandre Chartre)\n\n - Disassemble support (-d option to objtool) by Alexandre Chartre,\n   which supports the decoding of various Linux kernel code generation\n   specials such as alternatives:\n\n      17ef:  sched_balance_find_dst_group+0x62f                 mov    0x34(%r9),%edx\n      17f3:  sched_balance_find_dst_group+0x633               | \u003calternative.17f3\u003e             | X86_FEATURE_POPCNT\n      17f3:  sched_balance_find_dst_group+0x633               | call   0x17f8 \u003c__sw_hweight64\u003e | popcnt %rdi,%rax\n      17f8:  sched_balance_find_dst_group+0x638                 cmp    %eax,%edx\n\n   ... jump table alternatives:\n\n      1895:  sched_use_asym_prio+0x5                            test   $0x8,%ch\n      1898:  sched_use_asym_prio+0x8                            je     0x18a9 \u003csched_use_asym_prio+0x19\u003e\n      189a:  sched_use_asym_prio+0xa                          | \u003cjump_table.189a\u003e                        | JUMP\n      189a:  sched_use_asym_prio+0xa                          | jmp    0x18ae \u003csched_use_asym_prio+0x1e\u003e | nop2\n      189c:  sched_use_asym_prio+0xc                            mov    $0x1,%eax\n      18a1:  sched_use_asym_prio+0x11                           and    $0x80,%ecx\n\n   ... exception table alternatives:\n\n    native_read_msr:\n      5b80:  native_read_msr+0x0                                                     mov    %edi,%ecx\n      5b82:  native_read_msr+0x2                                                   | \u003cex_table.5b82\u003e | EXCEPTION\n      5b82:  native_read_msr+0x2                                                   | rdmsr           | resume at 0x5b84 \u003cnative_read_msr+0x4\u003e\n      5b84:  native_read_msr+0x4                                                     shl    $0x20,%rdx\n\n   .... x86 feature flag decoding (also see the X86_FEATURE_POPCNT\n        example in sched_balance_find_dst_group() above):\n\n      2faaf:  start_thread_common.constprop.0+0x1f                                    jne    0x2fba4 \u003cstart_thread_common.constprop.0+0x114\u003e\n      2fab5:  start_thread_common.constprop.0+0x25                                  | \u003calternative.2fab5\u003e                  | X86_FEATURE_ALWAYS                                  | X86_BUG_NULL_SEG\n      2fab5:  start_thread_common.constprop.0+0x25                                  | jmp    0x2faba \u003c.altinstr_aux+0x2f4\u003e | jmp    0x4b0 \u003cstart_thread_common.constprop.0+0x3f\u003e | nop5\n      2faba:  start_thread_common.constprop.0+0x2a                                    mov    $0x2b,%eax\n\n   ... NOP sequence shortening:\n\n      1048e2:  snapshot_write_finalize+0xc2                                            je     0x104917 \u003csnapshot_write_finalize+0xf7\u003e\n      1048e4:  snapshot_write_finalize+0xc4                                            nop6\n      1048ea:  snapshot_write_finalize+0xca                                            nop11\n      1048f5:  snapshot_write_finalize+0xd5                                            nop11\n      104900:  snapshot_write_finalize+0xe0                                            mov    %rax,%rcx\n      104903:  snapshot_write_finalize+0xe3                                            mov    0x10(%rdx),%rax\n\n   ... and much more.\n\n - Function validation tracing support (Alexandre Chartre)\n\n - Various -ffunction-sections fixes (Josh Poimboeuf)\n\n - Clang AutoFDO (Automated Feedback-Directed Optimizations) support\n   (Josh Poimboeuf)\n\n - Misc fixes and cleanups (Borislav Petkov, Chen Ni, Dylan Hatch, Ingo\n   Molnar, John Wang, Josh Poimboeuf, Pankaj Raghav, Peter Zijlstra,\n   Thorsten Blum)\n\n* tag \u0027objtool-core-2025-12-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (129 commits)\n  objtool: Fix segfault on unknown alternatives\n  objtool: Build with disassembly can fail when including bdf.h\n  objtool: Trim trailing NOPs in alternative\n  objtool: Add wide output for disassembly\n  objtool: Compact output for alternatives with one instruction\n  objtool: Improve naming of group alternatives\n  objtool: Add Function to get the name of a CPU feature\n  objtool: Provide access to feature and flags of group alternatives\n  objtool: Fix address references in alternatives\n  objtool: Disassemble jump table alternatives\n  objtool: Disassemble exception table alternatives\n  objtool: Print addresses with alternative instructions\n  objtool: Disassemble group alternatives\n  objtool: Print headers for alternatives\n  objtool: Preserve alternatives order\n  objtool: Add the --disas\u003d\u003cfunction-pattern\u003e action\n  objtool: Do not validate IBT for .return_sites and .call_sites\n  objtool: Improve tracing of alternative instructions\n  objtool: Add functions to better name alternatives\n  objtool: Identify the different types of alternatives\n  ...\n"
    },
    {
      "commit": "b53440f8e5a1466870d7a1d255e0f9966e0041fb",
      "tree": "72dd9cb31730d6fd3fea65ca49ba6b29009637b8",
      "parents": [
        "1b5dd29869b1e63f7e5c37d7552e2dcf22de3c26",
        "43decb6b628eb033a1b6188e5018773c0d38be1d"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 19:50:58 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 19:50:58 2025 -0800"
      },
      "message": "Merge tag \u0027locking-core-2025-12-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull locking updates from Ingo Molnar:\n \"Mutexes:\n\n   - Redo __mutex_init() to reduce generated code size (Sebastian\n     Andrzej Siewior)\n\n  Seqlocks:\n\n   - Introduce scoped_seqlock_read() (Peter Zijlstra)\n\n   - Change thread_group_cputime() to use scoped_seqlock_read() (Oleg\n     Nesterov)\n\n   - Change do_task_stat() to use scoped_seqlock_read() (Oleg Nesterov)\n\n   - Change do_io_accounting() to use scoped_seqlock_read() (Oleg\n     Nesterov)\n\n   - Fix the incorrect documentation of read_seqbegin_or_lock() /\n     need_seqretry() (Oleg Nesterov)\n\n   - Allow KASAN to fail optimizing (Peter Zijlstra)\n\n  Local lock updates:\n\n   - Fix all kernel-doc warnings (Randy Dunlap)\n\n   - Add the \u003clinux/local_lock*.h\u003e headers to MAINTAINERS (Sebastian\n     Andrzej Siewior)\n\n   - Reduce the risk of shadowing via s/l/__l/ and s/tl/__tl/ (Vincent\n     Mailhol)\n\n  Lock debugging:\n\n   - spinlock/debug: Fix data-race in do_raw_write_lock (Alexander\n     Sverdlin)\n\n  Atomic primitives infrastructure:\n\n   - atomic: Skip alignment check for try_cmpxchg() old arg (Arnd\n     Bergmann)\n\n  Rust runtime integration:\n\n   - sync: atomic: Enable generated Atomic\u003cT\u003e usage (Boqun Feng)\n\n   - sync: atomic: Implement Debug for Atomic\u003cDebug\u003e (Boqun Feng)\n\n   - debugfs: Remove Rust native atomics and replace them with Linux\n     versions (Boqun Feng)\n\n   - debugfs: Implement Reader for Mutex\u003cT\u003e only when T is Unpin (Boqun\n     Feng)\n\n   - lock: guard: Add T: Unpin bound to DerefMut (Daniel Almeida)\n\n   - lock: Pin the inner data (Daniel Almeida)\n\n   - lock: Add a Pin\u003c\u0026mut T\u003e accessor (Daniel Almeida)\"\n\n* tag \u0027locking-core-2025-12-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  locking/local_lock: Fix all kernel-doc warnings\n  locking/local_lock: s/l/__l/ and s/tl/__tl/ to reduce the risk of shadowing\n  locking/local_lock: Add the \u003clinux/local_lock*.h\u003e headers to MAINTAINERS\n  locking/mutex: Redo __mutex_init() to reduce generated code size\n  rust: debugfs: Replace the usage of Rust native atomics\n  rust: sync: atomic: Implement Debug for Atomic\u003cDebug\u003e\n  rust: sync: atomic: Make Atomic*Ops pub(crate)\n  seqlock: Allow KASAN to fail optimizing\n  rust: debugfs: Implement Reader for Mutex\u003cT\u003e only when T is Unpin\n  seqlock: Change do_io_accounting() to use scoped_seqlock_read()\n  seqlock: Change do_task_stat() to use scoped_seqlock_read()\n  seqlock: Change thread_group_cputime() to use scoped_seqlock_read()\n  seqlock: Introduce scoped_seqlock_read()\n  documentation: seqlock: fix the wrong documentation of read_seqbegin_or_lock/need_seqretry\n  atomic: Skip alignment check for try_cmpxchg() old arg\n  rust: lock: Add a Pin\u003c\u0026mut T\u003e accessor\n  rust: lock: Pin the inner data\n  rust: lock: guard: Add T: Unpin bound to DerefMut\n  locking/spinlock/debug: Fix data-race in do_raw_write_lock\n"
    },
    {
      "commit": "1b5dd29869b1e63f7e5c37d7552e2dcf22de3c26",
      "tree": "11785d66fcfc49273da7448bb28fcb64238c2afd",
      "parents": [
        "ffbf700df204dd25a48a19979a126e37f5dd1e6a",
        "0512bf9701f339c8fee2cc82b6fc35f0a8f6be7a"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 17:32:07 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 17:32:07 2025 -0800"
      },
      "message": "Merge tag \u0027vfs-6.19-rc1.fd_prepare.fs\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull fd prepare updates from Christian Brauner:\n \"This adds the FD_ADD() and FD_PREPARE() primitive. They simplify the\n  common pattern of get_unused_fd_flags() + create file + fd_install()\n  that is used extensively throughout the kernel and currently requires\n  cumbersome cleanup paths.\n\n  FD_ADD() - For simple cases where a file is installed immediately:\n\n      fd \u003d FD_ADD(O_CLOEXEC, vfio_device_open_file(device));\n      if (fd \u003c 0)\n          vfio_device_put_registration(device);\n      return fd;\n\n  FD_PREPARE() - For cases requiring access to the fd or file, or\n  additional work before publishing:\n\n      FD_PREPARE(fdf, O_CLOEXEC, sync_file-\u003efile);\n      if (fdf.err) {\n          fput(sync_file-\u003efile);\n          return fdf.err;\n      }\n\n      data.fence \u003d fd_prepare_fd(fdf);\n      if (copy_to_user((void __user *)arg, \u0026data, sizeof(data)))\n          return -EFAULT;\n\n      return fd_publish(fdf);\n\n  The primitives are centered around struct fd_prepare. FD_PREPARE()\n  encapsulates all allocation and cleanup logic and must be followed by\n  a call to fd_publish() which associates the fd with the file and\n  installs it into the caller\u0027s fdtable. If fd_publish() isn\u0027t called,\n  both are deallocated automatically. FD_ADD() is a shorthand that does\n  fd_publish() immediately and never exposes the struct to the caller.\n\n  I\u0027ve implemented this in a way that it\u0027s compatible with the cleanup\n  infrastructure while also being usable separately. IOW, it\u0027s centered\n  around struct fd_prepare which is aliased to class_fd_prepare_t and so\n  we can make use of all the basica guard infrastructure\"\n\n* tag \u0027vfs-6.19-rc1.fd_prepare.fs\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: (42 commits)\n  io_uring: convert io_create_mock_file() to FD_PREPARE()\n  file: convert replace_fd() to FD_PREPARE()\n  vfio: convert vfio_group_ioctl_get_device_fd() to FD_ADD()\n  tty: convert ptm_open_peer() to FD_ADD()\n  ntsync: convert ntsync_obj_get_fd() to FD_PREPARE()\n  media: convert media_request_alloc() to FD_PREPARE()\n  hv: convert mshv_ioctl_create_partition() to FD_ADD()\n  gpio: convert linehandle_create() to FD_PREPARE()\n  pseries: port papr_rtas_setup_file_interface() to FD_ADD()\n  pseries: convert papr_platform_dump_create_handle() to FD_ADD()\n  spufs: convert spufs_gang_open() to FD_PREPARE()\n  papr-hvpipe: convert papr_hvpipe_dev_create_handle() to FD_PREPARE()\n  spufs: convert spufs_context_open() to FD_PREPARE()\n  net/socket: convert __sys_accept4_file() to FD_ADD()\n  net/socket: convert sock_map_fd() to FD_ADD()\n  net/kcm: convert kcm_ioctl() to FD_PREPARE()\n  net/handshake: convert handshake_nl_accept_doit() to FD_PREPARE()\n  secretmem: convert memfd_secret() to FD_ADD()\n  memfd: convert memfd_create() to FD_ADD()\n  bpf: convert bpf_token_create() to FD_PREPARE()\n  ...\n"
    },
    {
      "commit": "ffbf700df204dd25a48a19979a126e37f5dd1e6a",
      "tree": "d92bf07cda161c226d90baef25423faa65a8acbf",
      "parents": [
        "d0deeb803cd65c41c37ac106063c46c51d5d43ab",
        "922a6f34c1756d2b0c35d9b2d915b8af19e85965"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 16:38:21 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 16:38:21 2025 -0800"
      },
      "message": "Merge tag \u0027vfs-6.19-rc1.autofs\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull autofs update from Christian Brauner:\n \"Prevent futile mount triggers in private mount namespaces.\n\n  Fix a problematic loop in autofs when a mount namespace contains\n  autofs mounts that are propagation private and there is no\n  namespace-specific automount daemon to handle possible automounting.\n\n  Previously, attempted path resolution would loop until MAXSYMLINKS was\n  reached before failing, causing significant noise in the log.\n\n The fix adds a check in autofs -\u003ed_automount() so that the VFS can\n immediately return EPERM in this case. Since the mount is propagation\n private, EPERM is the most appropriate error code\"\n\n* tag \u0027vfs-6.19-rc1.autofs\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:\n  autofs: dont trigger mount if it cant succeed\n"
    },
    {
      "commit": "d0deeb803cd65c41c37ac106063c46c51d5d43ab",
      "tree": "5ed8daa4b88c3bfe05ba8460b8aa41554c354eb2",
      "parents": [
        "a8058f8442df3150fa58154672f4a62a13e833e5",
        "2579e21be532457742d4100bbda1c2a5b81cbdef"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 16:31:21 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 16:31:21 2025 -0800"
      },
      "message": "Merge tag \u0027vfs-6.19-rc1.ovl\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull overlayfs cred guard conversion from Christian Brauner:\n \"This converts all of overlayfs to use credential guards, eliminating\n  manual credential management throughout the filesystem.\n\n  Credential guard conversion:\n\n   - Convert all of overlayfs to use credential guards, replacing the\n     manual ovl_override_creds()/ovl_revert_creds() pattern with scoped\n     guards.\n\n     This makes credential handling visually explicit and eliminates a\n     class of potential bugs from mismatched override/revert calls.\n\n     (1) Basic credential guard (with_ovl_creds)\n     (2) Creator credential guard (ovl_override_creator_creds):\n\n         Introduced a specialized guard for file creation operations\n         that handles the two-phase credential override (mounter\n         credentials, then fs{g,u}id override). The new pattern is much\n         clearer:\n\n         with_ovl_creds(dentry-\u003ed_sb) {\n                 scoped_class(prepare_creds_ovl, cred, dentry, inode, mode) {\n                         if (IS_ERR(cred))\n                                 return PTR_ERR(cred);\n                         /* creation operations */\n                 }\n         }\n\n     (3) Copy-up credential guard (ovl_cu_creds):\n\n         Introduced a specialized guard for copy-up operations,\n         simplifying the previous struct ovl_cu_creds helper and\n         associated functions.\n\n         Ported ovl_copy_up_workdir() and ovl_copy_up_tmpfile() to this\n         pattern.\n\n  Cleanups:\n\n   - Remove ovl_revert_creds() after all callers converted to guards\n\n   - Remove struct ovl_cu_creds and associated functions\n\n   - Drop ovl_setup_cred_for_create() after conversion\n\n   - Refactor ovl_fill_super(), ovl_lookup(), ovl_iterate(),\n     ovl_rename() for cleaner credential guard scope\n\n   - Introduce struct ovl_renamedata to simplify rename handling\n\n   - Don\u0027t override credentials for ovl_check_whiteouts() (unnecessary)\n\n   - Remove unneeded semicolon\"\n\n* tag \u0027vfs-6.19-rc1.ovl\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: (54 commits)\n  ovl: remove unneeded semicolon\n  ovl: remove struct ovl_cu_creds and associated functions\n  ovl: port ovl_copy_up_tmpfile() to cred guard\n  ovl: mark *_cu_creds() as unused temporarily\n  ovl: port ovl_copy_up_workdir() to cred guard\n  ovl: add copy up credential guard\n  ovl: drop ovl_setup_cred_for_create()\n  ovl: port ovl_create_or_link() to new ovl_override_creator_creds cleanup guard\n  ovl: mark ovl_setup_cred_for_create() as unused temporarily\n  ovl: reflow ovl_create_or_link()\n  ovl: port ovl_create_tmpfile() to new ovl_override_creator_creds cleanup guard\n  ovl: add ovl_override_creator_creds cred guard\n  ovl: remove ovl_revert_creds()\n  ovl: port ovl_fill_super() to cred guard\n  ovl: refactor ovl_fill_super()\n  ovl: port ovl_lower_positive() to cred guard\n  ovl: port ovl_lookup() to cred guard\n  ovl: refactor ovl_lookup()\n  ovl: port ovl_copyfile() to cred guard\n  ovl: port ovl_rename() to cred guard\n  ...\n"
    },
    {
      "commit": "a8058f8442df3150fa58154672f4a62a13e833e5",
      "tree": "a35d60f6e1861614a5227f9667287a447adc9291",
      "parents": [
        "db74a7d02ae244ec0552d18f51054f9ae0d921ad",
        "eeec741ee0df36e79a847bb5423f9eef4ed96071"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 16:13:46 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 16:13:46 2025 -0800"
      },
      "message": "Merge tag \u0027vfs-6.19-rc1.directory.locking\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull directory locking updates from Christian Brauner:\n \"This contains the work to add centralized APIs for directory locking\n  operations.\n\n  This series is part of a larger effort to change directory operation\n  locking to allow multiple concurrent operations in a directory. The\n  ultimate goal is to lock the target dentry(s) rather than the whole\n  parent directory.\n\n  To help with changing the locking protocol, this series centralizes\n  locking and lookup in new helper functions. The helpers establish a\n  pattern where it is the dentry that is being locked and unlocked\n  (currently the lock is held on dentry-\u003ed_parent-\u003ed_inode, but that can\n  change in the future).\n\n  This also changes vfs_mkdir() to unlock the parent on failure, as well\n  as dput()ing the dentry. This allows end_creating() to only require\n  the target dentry (which may be IS_ERR() after vfs_mkdir()), not the\n  parent\"\n\n* tag \u0027vfs-6.19-rc1.directory.locking\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:\n  nfsd: fix end_creating() conversion\n  VFS: introduce end_creating_keep()\n  VFS: change vfs_mkdir() to unlock on failure.\n  ecryptfs: use new start_creating/start_removing APIs\n  Add start_renaming_two_dentries()\n  VFS/ovl/smb: introduce start_renaming_dentry()\n  VFS/nfsd/ovl: introduce start_renaming() and end_renaming()\n  VFS: add start_creating_killable() and start_removing_killable()\n  VFS: introduce start_removing_dentry()\n  smb/server: use end_removing_noperm for for target of smb2_create_link()\n  VFS: introduce start_creating_noperm() and start_removing_noperm()\n  VFS/nfsd/cachefiles/ovl: introduce start_removing() and end_removing()\n  VFS/nfsd/cachefiles/ovl: add start_creating() and end_creating()\n  VFS: tidy up do_unlinkat()\n  VFS: introduce start_dirop() and end_dirop()\n  debugfs: rename end_creating() to debugfs_end_creating()\n"
    },
    {
      "commit": "db74a7d02ae244ec0552d18f51054f9ae0d921ad",
      "tree": "5f5de7b84f050774b6ae8593e2535be03c9aeaa6",
      "parents": [
        "4664fb427c8fd0080f40109f5e2b2090a6fb0c84",
        "4be9e04ebf75a5c4478c1c6295e2122e5dc98f5f"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 15:34:41 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 15:34:41 2025 -0800"
      },
      "message": "Merge tag \u0027vfs-6.19-rc1.directory.delegations\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull directory delegations update from Christian Brauner:\n \"This contains the work for recall-only directory delegations for\n  knfsd.\n\n  Add support for simple, recallable-only directory delegations. This\n  was decided at the fall NFS Bakeathon where the NFS client and server\n  maintainers discussed how to merge directory delegation support.\n\n  The approach starts with recallable-only delegations for several reasons:\n\n   1. RFC8881 has gaps that are being addressed in RFC8881bis. In\n      particular, it requires directory position information for\n      CB_NOTIFY callbacks, which is difficult to implement properly\n      under Linux. The spec is being extended to allow that information\n      to be omitted.\n\n   2. Client-side support for CB_NOTIFY still lags. The client side\n      involves heuristics about when to request a delegation.\n\n   3. Early indication shows simple, recallable-only delegations can\n      help performance. Anna Schumaker mentioned seeing a multi-minute\n      speedup in xfstests runs with them enabled.\n\n  With these changes, userspace can also request a read lease on a\n  directory that will be recalled on conflicting accesses. This may be\n  useful for applications like Samba. Users can disable leases\n  altogether via the fs.leases-enable sysctl if needed.\n\n  VFS changes:\n\n   - Dedicated Type for Delegations\n\n     Introduce struct delegated_inode to track inodes that may have\n     delegations that need to be broken. This replaces the previous\n     approach of passing raw inode pointers through the delegation\n     breaking code paths, providing better type safety and clearer\n     semantics for the delegation machinery.\n\n   - Break parent directory delegations in open(..., O_CREAT) codepath\n\n   - Allow mkdir to wait for delegation break on parent\n\n   - Allow rmdir to wait for delegation break on parent\n\n   - Add try_break_deleg calls for parents to vfs_link(), vfs_rename(),\n     and vfs_unlink()\n\n   - Make vfs_create(), vfs_mknod(), and vfs_symlink() break delegations\n     on parent directory\n\n   - Clean up argument list for vfs_create()\n\n   - Expose delegation support to userland\n\n  Filelock changes:\n\n   - Make lease_alloc() take a flags argument\n\n   - Rework the __break_lease API to use flags\n\n   - Add struct delegated_inode\n\n   - Push the S_ISREG check down to -\u003esetlease handlers\n\n   - Lift the ban on directory leases in generic_setlease\n\n  NFSD changes:\n\n   - Allow filecache to hold S_IFDIR files\n\n   - Allow DELEGRETURN on directories\n\n   - Wire up GET_DIR_DELEGATION handling\n\n  Fixes:\n\n   - Fix kernel-doc warnings in __fcntl_getlease\n\n   - Add needed headers for new struct delegation definition\"\n\n* tag \u0027vfs-6.19-rc1.directory.delegations\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:\n  vfs: add needed headers for new struct delegation definition\n  filelock: __fcntl_getlease: fix kernel-doc warnings\n  vfs: expose delegation support to userland\n  nfsd: wire up GET_DIR_DELEGATION handling\n  nfsd: allow DELEGRETURN on directories\n  nfsd: allow filecache to hold S_IFDIR files\n  filelock: lift the ban on directory leases in generic_setlease\n  vfs: make vfs_symlink break delegations on parent dir\n  vfs: make vfs_mknod break delegations on parent directory\n  vfs: make vfs_create break delegations on parent directory\n  vfs: clean up argument list for vfs_create()\n  vfs: break parent dir delegations in open(..., O_CREAT) codepath\n  vfs: allow rmdir to wait for delegation break on parent\n  vfs: allow mkdir to wait for delegation break on parent\n  vfs: add try_break_deleg calls for parents to vfs_{link,rename,unlink}\n  filelock: push the S_ISREG check down to -\u003esetlease handlers\n  filelock: add struct delegated_inode\n  filelock: rework the __break_lease API to use flags\n  filelock: make lease_alloc() take a flags argument\n"
    },
    {
      "commit": "4664fb427c8fd0080f40109f5e2b2090a6fb0c84",
      "tree": "194e3ffbe9775535ebe843b954c25734b941a4cd",
      "parents": [
        "978d337c2ed6e5313ee426871a410eddc796ccfd",
        "0d534518ce87317e884dbd1485111b0f1606a194"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 15:22:40 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 15:22:40 2025 -0800"
      },
      "message": "Merge tag \u0027vfs-6.19-rc1.minix\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull minix fixes from Christian Brauner:\n \"Fix two syzbot corruption bugs in the minix filesystem.\n\n  Syzbot fuzzes filesystems by trying to mount and manipulate\n  deliberately corrupted images. This should not lead to BUG_ONs and\n  WARN_ONs for easy to detect corruptions.\n\n   - Add error handling to minix filesystem for inode corruption\n     detection, enabling the filesystem to report such corruptions\n     cleanly.\n\n   - Fix a drop_nlink warning in minix_rmdir() triggered by corrupted\n     directory link counts.\n\n   - Fix a drop_nlink warning in minix_rename() triggered by corrupted\n     inode link counts\"\n\n* tag \u0027vfs-6.19-rc1.minix\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:\n  Fix a drop_nlink warning in minix_rename\n  Fix a drop_nlink warning in minix_rmdir\n  Add error handling to minix filesystem for inode corruption detection\n"
    },
    {
      "commit": "978d337c2ed6e5313ee426871a410eddc796ccfd",
      "tree": "b12791ca01da7a745ed8728f43462fabfeecf352",
      "parents": [
        "afdf0fb340948a8c0f581ed1dc42828af89b80b6",
        "73fd0dba0beb1d2d1695ee5452eac8dfabce3f9e"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 14:39:03 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 14:39:03 2025 -0800"
      },
      "message": "Merge tag \u0027vfs-6.19-rc1.guards\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull superblock lock guard updates from Christian Brauner:\n \"This starts the work of introducing guards for superblock related\n  locks.\n\n  Introduce super_write_guard for scoped superblock write protection.\n\n  This provides a guard-based alternative to the manual sb_start_write()\n  and sb_end_write() pattern, allowing the compiler to automatically\n  handle the cleanup\"\n\n* tag \u0027vfs-6.19-rc1.guards\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:\n  xfs: use super write guard in xfs_file_ioctl()\n  open: use super write guard in do_ftruncate()\n  btrfs: use super write guard in relocating_repair_kthread()\n  ext4: use super write guard in write_mmp_block()\n  btrfs: use super write guard in sb_start_write()\n  btrfs: use super write guard btrfs_run_defrag_inode()\n  btrfs: use super write guard in btrfs_reclaim_bgs_work()\n  fs: add super_write_guard\n"
    },
    {
      "commit": "afdf0fb340948a8c0f581ed1dc42828af89b80b6",
      "tree": "aedc3946c80e6125506290c186981c77518535f1",
      "parents": [
        "1d18101a644e6ece450d5b0a93f21a71a21b6222",
        "dca3aa666fbd71118905d88bb1c353881002b647"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 14:18:01 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 14:18:01 2025 -0800"
      },
      "message": "Merge tag \u0027vfs-6.19-rc1.fs_header\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull fs header updates from Christian Brauner:\n \"This contains initial work to start splitting up fs.h.\n\n  Begin the long-overdue work of splitting up the monolithic fs.h\n  header. The header has grown to over 3000 lines and includes types and\n  functions for many different subsystems, making it difficult to\n  navigate and causing excessive compilation dependencies.\n\n  This series introduces new focused headers for superblock-related\n  code:\n\n   - Rename fs_types.h to fs_dirent.h to better reflect its actual\n     content (directory entry types)\n\n   - Add fs/super_types.h containing superblock type definitions\n\n   - Add fs/super.h containing superblock function declarations\n\n  This is the first step in a longer effort to modularize the VFS\n  headers.\n\n  Cleanups:\n\n   - Inode Field Layout Optimization (Mateusz Guzik)\n\n     Move inode fields used during fast path lookup closer together to\n     improve cache locality during path resolution.\n\n   - current_umask() Optimization (Mateusz Guzik)\n\n     Inline current_umask() and move it to fs_struct.h. This improves\n     performance by avoiding function call overhead for this\n     frequently-used function, and places it in a more appropriate\n     header since it operates on fs_struct\"\n\n* tag \u0027vfs-6.19-rc1.fs_header\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:\n  fs: move inode fields used during fast path lookup closer together\n  fs: inline current_umask() and move it to fs_struct.h\n  fs: add fs/super.h header\n  fs: add fs/super_types.h header\n  fs: rename fs_types.h to fs_dirent.h\n"
    },
    {
      "commit": "1d18101a644e6ece450d5b0a93f21a71a21b6222",
      "tree": "1b4470c3b046837e8f3f94ca7436056824376875",
      "parents": [
        "f2e74ecfba1b0d407f04b671a240cc65e309e529",
        "c8e00cdc7425d5c60fd1ce6e7f71e5fb1b236991"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 13:45:41 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 13:45:41 2025 -0800"
      },
      "message": "Merge tag \u0027kernel-6.19-rc1.cred\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull cred guard updates from Christian Brauner:\n \"This contains substantial credential infrastructure improvements\n  adding guard-based credential management that simplifies code and\n  eliminates manual reference counting in many subsystems.\n\n  Features:\n\n   - Kernel Credential Guards\n\n     Add with_kernel_creds() and scoped_with_kernel_creds() guards that\n     allow using the kernel credentials without allocating and copying\n     them. This was requested by Linus after seeing repeated\n     prepare_kernel_creds() calls that duplicate the kernel credentials\n     only to drop them again later.\n\n     The new guards completely avoid the allocation and never expose the\n     temporary variable to hold the kernel credentials anywhere in\n     callers.\n\n   - Generic Credential Guards\n\n     Add scoped_with_creds() guards for the common override_creds() and\n     revert_creds() pattern. This builds on earlier work that made\n     override_creds()/revert_creds() completely reference count free.\n\n   - Prepare Credential Guards\n\n     Add prepare credential guards for the more complex pattern of\n     preparing a new set of credentials and overriding the current\n     credentials with them:\n      - prepare_creds()\n      - modify new creds\n      - override_creds()\n      - revert_creds()\n      - put_cred()\n\n  Cleanups:\n\n   - Make init_cred static since it should not be directly accessed\n\n   - Add kernel_cred() helper to properly access the kernel credentials\n\n   - Fix scoped_class() macro that was introduced two cycles ago\n\n   - coredump: split out do_coredump() from vfs_coredump() for cleaner\n     credential handling\n\n   - coredump: move revert_cred() before coredump_cleanup()\n\n   - coredump: mark struct mm_struct as const\n\n   - coredump: pass struct linux_binfmt as const\n\n   - sev-dev: use guard for path\"\n\n* tag \u0027kernel-6.19-rc1.cred\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: (36 commits)\n  trace: use override credential guard\n  trace: use prepare credential guard\n  coredump: use override credential guard\n  coredump: use prepare credential guard\n  coredump: split out do_coredump() from vfs_coredump()\n  coredump: mark struct mm_struct as const\n  coredump: pass struct linux_binfmt as const\n  coredump: move revert_cred() before coredump_cleanup()\n  sev-dev: use override credential guards\n  sev-dev: use prepare credential guard\n  sev-dev: use guard for path\n  cred: add prepare credential guard\n  net/dns_resolver: use credential guards in dns_query()\n  cgroup: use credential guards in cgroup_attach_permissions()\n  act: use credential guards in acct_write_process()\n  smb: use credential guards in cifs_get_spnego_key()\n  nfs: use credential guards in nfs_idmap_get_key()\n  nfs: use credential guards in nfs_local_call_write()\n  nfs: use credential guards in nfs_local_call_read()\n  erofs: use credential guards\n  ...\n"
    },
    {
      "commit": "f2e74ecfba1b0d407f04b671a240cc65e309e529",
      "tree": "127d6d64e1e4e94f969124009c0248f87fde5227",
      "parents": [
        "212c4053a1502e5117d8cbbbd1c15579ce1839bb",
        "37d369fa97cc0774ea4eab726d16bcb5fbe3a104"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 10:26:38 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 10:26:38 2025 -0800"
      },
      "message": "Merge tag \u0027vfs-6.19-rc1.folio\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull folio updates from Christian Brauner:\n \"Add a new folio_next_pos() helper function that returns the file\n  position of the first byte after the current folio. This is a common\n  operation in filesystems when needing to know the end of the current\n  folio.\n\n  The helper is lifted from btrfs which already had its own version, and\n  is now used across multiple filesystems and subsystems:\n   - btrfs\n   - buffer\n   - ext4\n   - f2fs\n   - gfs2\n   - iomap\n   - netfs\n   - xfs\n   - mm\n\n  This fixes a long-standing bug in ocfs2 on 32-bit systems with files\n  larger than 2GiB. Presumably this is not a common configuration, but\n  the fix is backported anyway. The other filesystems did not have bugs,\n  they were just mildly inefficient.\n\n  This also introduce uoff_t as the unsigned version of loff_t. A recent\n  commit inadvertently changed a comparison from being unsigned (on\n  64-bit systems) to being signed (which it had always been on 32-bit\n  systems), leading to sporadic fstests failures.\n\n  Generally file sizes are restricted to being a signed integer, but in\n  places where -1 is passed to indicate \"up to the end of the file\", it\n  is convenient to have an unsigned type to ensure comparisons are\n  always unsigned regardless of architecture\"\n\n* tag \u0027vfs-6.19-rc1.folio\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:\n  fs: Add uoff_t\n  mm: Use folio_next_pos()\n  xfs: Use folio_next_pos()\n  netfs: Use folio_next_pos()\n  iomap: Use folio_next_pos()\n  gfs2: Use folio_next_pos()\n  f2fs: Use folio_next_pos()\n  ext4: Use folio_next_pos()\n  buffer: Use folio_next_pos()\n  btrfs: Use folio_next_pos()\n  filemap: Add folio_next_pos()\n"
    },
    {
      "commit": "212c4053a1502e5117d8cbbbd1c15579ce1839bb",
      "tree": "3cde8f505314577d9261ef45585e949298f46b0e",
      "parents": [
        "415d34b92c1f921a9ff3c38f56319cbc5536f642",
        "390d967653e17205f0e519f691b7d6be0a05ff45"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 10:17:39 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 10:17:39 2025 -0800"
      },
      "message": "Merge tag \u0027vfs-6.19-rc1.coredump\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull pidfd and coredump updates from Christian Brauner:\n \"Features:\n\n   - Expose coredump signal via pidfd\n\n     Expose the signal that caused the coredump through the pidfd\n     interface. The recent changes to rework coredump handling to rely\n     on unix sockets are in the process of being used in systemd. The\n     previous systemd coredump container interface requires the coredump\n     file descriptor and basic information including the signal number\n     to be sent to the container. This means the signal number needs to\n     be available before sending the coredump to the container.\n\n   - Add supported_mask field to pidfd\n\n     Add a new supported_mask field to struct pidfd_info that indicates\n     which information fields are supported by the running kernel. This\n     allows userspace to detect feature availability without relying on\n     error codes or kernel version checks.\n\n  Cleanups:\n\n   - Drop struct pidfs_exit_info and prepare to drop exit_info pointer,\n     simplifying the internal publication mechanism for exit and\n     coredump information retrievable via the pidfd ioctl\n\n   - Use guard() for task_lock in pidfs\n\n   - Reduce wait_pidfd lock scope\n\n   - Add missing PIDFD_INFO_SIZE_VER1 constant\n\n   - Add missing BUILD_BUG_ON() assert on struct pidfd_info\n\n  Fixes:\n\n   - Fix PIDFD_INFO_COREDUMP handling\n\n  Selftests:\n\n   - Split out coredump socket tests and common helpers into separate\n     files for better organization\n\n   - Fix userspace coredump client detection issues\n\n   - Handle edge-triggered epoll correctly\n\n   - Ignore ENOSPC errors in tests\n\n   - Add debug logging to coredump socket tests, socket protocol tests,\n     and test helpers\n\n   - Add tests for PIDFD_INFO_COREDUMP_SIGNAL\n\n   - Add tests for supported_mask field\n\n   - Update pidfd header for selftests\"\n\n* tag \u0027vfs-6.19-rc1.coredump\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: (23 commits)\n  pidfs: reduce wait_pidfd lock scope\n  selftests/coredump: add second PIDFD_INFO_COREDUMP_SIGNAL test\n  selftests/coredump: add first PIDFD_INFO_COREDUMP_SIGNAL test\n  selftests/coredump: ignore ENOSPC errors\n  selftests/coredump: add debug logging to coredump socket protocol tests\n  selftests/coredump: add debug logging to coredump socket tests\n  selftests/coredump: add debug logging to test helpers\n  selftests/coredump: handle edge-triggered epoll correctly\n  selftests/coredump: fix userspace coredump client detection\n  selftests/coredump: fix userspace client detection\n  selftests/coredump: split out coredump socket tests\n  selftests/coredump: split out common helpers\n  selftests/pidfd: add second supported_mask test\n  selftests/pidfd: add first supported_mask test\n  selftests/pidfd: update pidfd header\n  pidfs: expose coredump signal\n  pidfs: drop struct pidfs_exit_info\n  pidfs: prepare to drop exit_info pointer\n  pidfd: add a new supported_mask field\n  pidfs: add missing BUILD_BUG_ON() assert on struct pidfd_info\n  ...\n"
    },
    {
      "commit": "415d34b92c1f921a9ff3c38f56319cbc5536f642",
      "tree": "461dc8621de93dcc175f8bef9233a41d1a47e23f",
      "parents": [
        "ebaeabfa5ab711a9b69b686d58329e258fdae75f",
        "a71e4f103aed69e7a11ea913312726bb194c76ee"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 09:47:41 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 09:47:41 2025 -0800"
      },
      "message": "Merge tag \u0027namespace-6.19-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull namespace updates from Christian Brauner:\n \"This contains substantial namespace infrastructure changes including a new\n  system call, active reference counting, and extensive header cleanups.\n  The branch depends on the shared kbuild branch for -fms-extensions support.\n\n  Features:\n\n   - listns() system call\n\n     Add a new listns() system call that allows userspace to iterate\n     through namespaces in the system. This provides a programmatic\n     interface to discover and inspect namespaces, addressing\n     longstanding limitations:\n\n     Currently, there is no direct way for userspace to enumerate\n     namespaces. Applications must resort to scanning /proc/*/ns/ across\n     all processes, which is:\n      - Inefficient - requires iterating over all processes\n      - Incomplete - misses namespaces not attached to any running\n        process but kept alive by file descriptors, bind mounts, or\n        parent references\n      - Permission-heavy - requires access to /proc for many processes\n      - No ordering or ownership information\n      - No filtering per namespace type\n\n     The listns() system call solves these problems:\n\n       ssize_t listns(const struct ns_id_req *req, u64 *ns_ids,\n                      size_t nr_ns_ids, unsigned int flags);\n\n       struct ns_id_req {\n             __u32 size;\n             __u32 spare;\n             __u64 ns_id;\n             struct /* listns */ {\n                     __u32 ns_type;\n                     __u32 spare2;\n                     __u64 user_ns_id;\n             };\n       };\n\n     Features include:\n      - Pagination support for large namespace sets\n      - Filtering by namespace type (MNT_NS, NET_NS, USER_NS, etc.)\n      - Filtering by owning user namespace\n      - Permission checks respecting namespace isolation\n\n   - Active Reference Counting\n\n     Introduce an active reference count that tracks namespace\n     visibility to userspace. A namespace is visible in the following\n     cases:\n      - The namespace is in use by a task\n      - The namespace is persisted through a VFS object (namespace file\n        descriptor or bind-mount)\n      - The namespace is a hierarchical type and is the parent of child\n        namespaces\n\n     The active reference count does not regulate lifetime (that\u0027s still\n     done by the normal reference count) - it only regulates visibility\n     to namespace file handles and listns().\n\n     This prevents resurrection of namespaces that are pinned only for\n     internal kernel reasons (e.g., user namespaces held by\n     file-\u003ef_cred, lazy TLB references on idle CPUs, etc.) which should\n     not be accessible via (1)-(3).\n\n   - Unified Namespace Tree\n\n     Introduce a unified tree structure for all namespaces with:\n      - Fixed IDs assigned to initial namespaces\n      - Lookup based solely on inode number\n      - Maintained list of owned namespaces per user namespace\n      - Simplified rbtree comparison helpers\n\n   Cleanups\n\n    - Header Reorganization:\n      - Move namespace types into separate header (ns_common_types.h)\n      - Decouple nstree from ns_common header\n      - Move nstree types into separate header\n      - Switch to new ns_tree_{node,root} structures with helper functions\n      - Use guards for ns_tree_lock\n\n   - Initial Namespace Reference Count Optimization\n      - Make all reference counts on initial namespaces a nop to avoid\n        pointless cacheline ping-pong for namespaces that can never go\n        away\n      - Drop custom reference count initialization for initial namespaces\n      - Add NS_COMMON_INIT() macro and use it for all namespaces\n      - pid: rely on common reference count behavior\n\n   - Miscellaneous Cleanups\n      - Rename exit_task_namespaces() to exit_nsproxy_namespaces()\n      - Rename is_initial_namespace() and make argument const\n      - Use boolean to indicate anonymous mount namespace\n      - Simplify owner list iteration in nstree\n      - nsfs: raise SB_I_NODEV, SB_I_NOEXEC, and DCACHE_DONTCACHE explicitly\n      - nsfs: use inode_just_drop()\n      - pidfs: raise DCACHE_DONTCACHE explicitly\n      - pidfs: simplify PIDFD_GET__NAMESPACE ioctls\n      - libfs: allow to specify s_d_flags\n      - cgroup: add cgroup namespace to tree after owner is set\n      - nsproxy: fix free_nsproxy() and simplify create_new_namespaces()\n\n  Fixes:\n\n   - setns(pidfd, ...) race condition\n\n     Fix a subtle race when using pidfds with setns(). When the target\n     task exits after prepare_nsset() but before commit_nsset(), the\n     namespace\u0027s active reference count might have been dropped. If\n     setns() then installs the namespaces, it would bump the active\n     reference count from zero without taking the required reference on\n     the owner namespace, leading to underflow when later decremented.\n\n     The fix resurrects the ownership chain if necessary - if the caller\n     succeeded in grabbing passive references, the setns() should\n     succeed even if the target task exits or gets reaped.\n\n   - Return EFAULT on put_user() error instead of success\n\n   - Make sure references are dropped outside of RCU lock (some\n     namespaces like mount namespace sleep when putting the last\n     reference)\n\n   - Don\u0027t skip active reference count initialization for network\n     namespace\n\n   - Add asserts for active refcount underflow\n\n   - Add asserts for initial namespace reference counts (both passive\n     and active)\n\n   - ipc: enable is_ns_init_id() assertions\n\n   - Fix kernel-doc comments for internal nstree functions\n\n   - Selftests\n      - 15 active reference count tests\n      - 9 listns() functionality tests\n      - 7 listns() permission tests\n      - 12 inactive namespace resurrection tests\n      - 3 threaded active reference count tests\n      - commit_creds() active reference tests\n      - Pagination and stress tests\n      - EFAULT handling test\n      - nsid tests fixes\"\n\n* tag \u0027namespace-6.19-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: (103 commits)\n  pidfs: simplify PIDFD_GET_\u003ctype\u003e_NAMESPACE ioctls\n  nstree: fix kernel-doc comments for internal functions\n  nsproxy: fix free_nsproxy() and simplify create_new_namespaces()\n  selftests/namespaces: fix nsid tests\n  ns: drop custom reference count initialization for initial namespaces\n  pid: rely on common reference count behavior\n  ns: add asserts for initial namespace active reference counts\n  ns: add asserts for initial namespace reference counts\n  ns: make all reference counts on initial namespace a nop\n  ipc: enable is_ns_init_id() assertions\n  fs: use boolean to indicate anonymous mount namespace\n  ns: rename is_initial_namespace()\n  ns: make is_initial_namespace() argument const\n  nstree: use guards for ns_tree_lock\n  nstree: simplify owner list iteration\n  nstree: switch to new structures\n  nstree: add helper to operate on struct ns_tree_{node,root}\n  nstree: move nstree types into separate header\n  nstree: decouple from ns_common header\n  ns: move namespace types into separate header\n  ...\n"
    },
    {
      "commit": "ebaeabfa5ab711a9b69b686d58329e258fdae75f",
      "tree": "c4b0915701c65abed0bc3fe9f351f42daaaa9089",
      "parents": [
        "9368f0f9419cde028a6e58331065900ff089bc36",
        "4952f35f0545f3b53dab8d5fd727c4827c2a2778"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 09:20:51 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 09:20:51 2025 -0800"
      },
      "message": "Merge tag \u0027vfs-6.19-rc1.writeback\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull writeback updates from Christian Brauner:\n \"Features:\n\n   - Allow file systems to increase the minimum writeback chunk size.\n\n     The relatively low minimal writeback size of 4MiB means that\n     written back inodes on rotational media are switched a lot. Besides\n     introducing additional seeks, this also can lead to extreme file\n     fragmentation on zoned devices when a lot of files are cached\n     relative to the available writeback bandwidth.\n\n     This adds a superblock field that allows the file system to\n     override the default size, and sets it to the zone size for zoned\n     XFS.\n\n   - Add logging for slow writeback when it exceeds\n     sysctl_hung_task_timeout_secs. This helps identify tasks waiting\n     for a long time and pinpoint potential issues. Recording the\n     starting jiffies is also useful when debugging a crashed vmcore.\n\n   - Wake up waiting tasks when finishing the writeback of a chunk\n\n  Cleanups:\n\n   - filemap_* writeback interface cleanups.\n\n     Adding filemap_fdatawrite_wbc ended up being a mistake, as all but\n     the original btrfs caller should be using better high level\n     interfaces instead.\n\n     This series removes all these low-level interfaces, switches btrfs\n     to a more specific interface, and cleans up other too low-level\n     interfaces. With this the writeback_control that is passed to the\n     writeback code is only initialized in three places.\n\n   - Remove __filemap_fdatawrite, __filemap_fdatawrite_range, and\n     filemap_fdatawrite_wbc\n\n   - Add filemap_flush_nr helper for btrfs\n\n   - Push struct writeback_control into start_delalloc_inodes in btrfs\n\n   - Rename filemap_fdatawrite_range_kick to filemap_flush_range\n\n   - Stop opencoding filemap_fdatawrite_range in 9p, ocfs2, and mm\n\n   - Make wbc_to_tag() inline and use it in fs\"\n\n* tag \u0027vfs-6.19-rc1.writeback\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:\n  fs: Make wbc_to_tag() inline and use it in fs.\n  xfs: set s_min_writeback_pages for zoned file systems\n  writeback: allow the file system to override MIN_WRITEBACK_PAGES\n  writeback: cleanup writeback_chunk_size\n  mm: rename filemap_fdatawrite_range_kick to filemap_flush_range\n  mm: remove __filemap_fdatawrite_range\n  mm: remove filemap_fdatawrite_wbc\n  mm: remove __filemap_fdatawrite\n  mm,btrfs: add a filemap_flush_nr helper\n  btrfs: push struct writeback_control into start_delalloc_inodes\n  btrfs: use the local tmp_inode variable in start_delalloc_inodes\n  ocfs2: don\u0027t opencode filemap_fdatawrite_range in ocfs2_journal_submit_inode_data_buffers\n  9p: don\u0027t opencode filemap_fdatawrite_range in v9fs_mmap_vm_close\n  mm: don\u0027t opencode filemap_fdatawrite_range in filemap_invalidate_inode\n  writeback: Add logging for slow writeback (exceeds sysctl_hung_task_timeout_secs)\n  writeback: Wake up waiting tasks when finishing the writeback of a chunk.\n"
    },
    {
      "commit": "9368f0f9419cde028a6e58331065900ff089bc36",
      "tree": "b7e3a8910f9493e8422cd661e1375e5421a91bf0",
      "parents": [
        "b04b2e7a61830cabd00c6f95308a8e2f5d82fa52",
        "ca0d620b0afae20a7bcd5182606eba6860b2dbf2"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 09:02:34 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 09:02:34 2025 -0800"
      },
      "message": "Merge tag \u0027vfs-6.19-rc1.inode\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull vfs inode updates from Christian Brauner:\n \"Features:\n\n   - Hide inode-\u003ei_state behind accessors. Open-coded accesses prevent\n     asserting they are done correctly. One obvious aspect is locking,\n     but significantly more can be checked. For example it can be\n     detected when the code is clearing flags which are already missing,\n     or is setting flags when it is illegal (e.g., I_FREEING when\n     -\u003ei_count \u003e 0)\n\n   - Provide accessors for -\u003ei_state, converts all filesystems using\n     coccinelle and manual conversions (btrfs, ceph, smb, f2fs, gfs2,\n     overlayfs, nilfs2, xfs), and makes plain -\u003ei_state access fail to\n     compile\n\n   - Rework I_NEW handling to operate without fences, simplifying the\n     code after the accessor infrastructure is in place\n\n  Cleanups:\n\n   - Move wait_on_inode() from writeback.h to fs.h\n\n   - Spell out fenced -\u003ei_state accesses with explicit smp_wmb/smp_rmb\n     for clarity\n\n   - Cosmetic fixes to LRU handling\n\n   - Push list presence check into inode_io_list_del()\n\n   - Touch up predicts in __d_lookup_rcu()\n\n   - ocfs2: retire ocfs2_drop_inode() and I_WILL_FREE usage\n\n   - Assert on -\u003ei_count in iput_final()\n\n   - Assert -\u003ei_lock held in __iget()\n\n  Fixes:\n\n   - Add missing fences to I_NEW handling\"\n\n* tag \u0027vfs-6.19-rc1.inode\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: (22 commits)\n  dcache: touch up predicts in __d_lookup_rcu()\n  fs: push list presence check into inode_io_list_del()\n  fs: cosmetic fixes to lru handling\n  fs: rework I_NEW handling to operate without fences\n  fs: make plain -\u003ei_state access fail to compile\n  xfs: use the new -\u003ei_state accessors\n  nilfs2: use the new -\u003ei_state accessors\n  overlayfs: use the new -\u003ei_state accessors\n  gfs2: use the new -\u003ei_state accessors\n  f2fs: use the new -\u003ei_state accessors\n  smb: use the new -\u003ei_state accessors\n  ceph: use the new -\u003ei_state accessors\n  btrfs: use the new -\u003ei_state accessors\n  Manual conversion to use -\u003ei_state accessors of all places not covered by coccinelle\n  Coccinelle-based conversion to use -\u003ei_state accessors\n  fs: provide accessors for -\u003ei_state\n  fs: spell out fenced -\u003ei_state accesses with explicit smp_wmb/smp_rmb\n  fs: move wait_on_inode() from writeback.h to fs.h\n  fs: add missing fences to I_NEW handling\n  ocfs2: retire ocfs2_drop_inode() and I_WILL_FREE usage\n  ...\n"
    },
    {
      "commit": "b04b2e7a61830cabd00c6f95308a8e2f5d82fa52",
      "tree": "a228d66ed097ffd37e05e4bcc4d37337be731d46",
      "parents": [
        "1885cdbfbb51ede3637166c895d0b8040c9899cc",
        "ebf8538979101ef879742dcfaf04b684f5461e12"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 08:44:26 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 08:44:26 2025 -0800"
      },
      "message": "Merge tag \u0027vfs-6.19-rc1.misc\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull misc vfs updates from Christian Brauner:\n \"Features:\n\n   - Cheaper MAY_EXEC handling for path lookup. This elides MAY_WRITE\n     permission checks during path lookup and adds the\n     IOP_FASTPERM_MAY_EXEC flag so filesystems like btrfs can avoid\n     expensive permission work.\n\n   - Hide dentry_cache behind runtime const machinery.\n\n   - Add German Maglione as virtiofs co-maintainer.\n\n  Cleanups:\n\n   - Tidy up and inline step_into() and walk_component() for improved\n     code generation.\n\n   - Re-enable IOCB_NOWAIT writes to files. This refactors file\n     timestamp update logic, fixing a layering bypass in btrfs when\n     updating timestamps on device files and improving FMODE_NOCMTIME\n     handling in VFS now that nfsd started using it.\n\n   - Path lookup optimizations extracting slowpaths into dedicated\n     routines and adding branch prediction hints for mntput_no_expire(),\n     fd_install(), lookup_slow(), and various other hot paths.\n\n   - Enable clang\u0027s -fms-extensions flag, requiring a JFS rename to\n     avoid conflicts.\n\n   - Remove spurious exports in fs/file_attr.c.\n\n   - Stop duplicating union pipe_index declaration. This depends on the\n     shared kbuild branch that brings in -fms-extensions support which\n     is merged into this branch.\n\n   - Use MD5 library instead of crypto_shash in ecryptfs.\n\n   - Use largest_zero_folio() in iomap_dio_zero().\n\n   - Replace simple_strtol/strtoul with kstrtoint/kstrtouint in init and\n     initrd code.\n\n   - Various typo fixes.\n\n  Fixes:\n\n   - Fix emergency sync for btrfs. Btrfs requires an explicit sync_fs()\n     call with wait \u003d\u003d 1 to commit super blocks. The emergency sync path\n     never passed this, leaving btrfs data uncommitted during emergency\n     sync.\n\n   - Use local kmap in watch_queue\u0027s post_one_notification().\n\n   - Add hint prints in sb_set_blocksize() for LBS dependency on THP\"\n\n* tag \u0027vfs-6.19-rc1.misc\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: (35 commits)\n  MAINTAINERS: add German Maglione as virtiofs co-maintainer\n  fs: inline step_into() and walk_component()\n  fs: tidy up step_into() \u0026 friends before inlining\n  orangefs: use inode_update_timestamps directly\n  btrfs: fix the comment on btrfs_update_time\n  btrfs: use vfs_utimes to update file timestamps\n  fs: export vfs_utimes\n  fs: lift the FMODE_NOCMTIME check into file_update_time_flags\n  fs: refactor file timestamp update logic\n  include/linux/fs.h: trivial fix: regualr -\u003e regular\n  fs/splice.c: trivial fix: pipes -\u003e pipe\u0027s\n  fs: mark lookup_slow() as noinline\n  fs: add predicts based on nd-\u003edepth\n  fs: move mntput_no_expire() slowpath into a dedicated routine\n  fs: remove spurious exports in fs/file_attr.c\n  watch_queue: Use local kmap in post_one_notification()\n  fs: touch up predicts in path lookup\n  fs: move fd_install() slowpath into a dedicated routine and provide commentary\n  fs: hide dentry_cache behind runtime const machinery\n  fs: touch predicts in do_dentry_open()\n  ...\n"
    },
    {
      "commit": "1885cdbfbb51ede3637166c895d0b8040c9899cc",
      "tree": "257c3f32cff62bff6a3acfe8c76d39981cc52faa",
      "parents": [
        "7d0a66e4bb9081d75c82ec4957c50034cb0ea449",
        "7fd8720dff2d9c70cf5a1a13b7513af01952ec02"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 08:14:00 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Dec 01 08:14:00 2025 -0800"
      },
      "message": "Merge tag \u0027vfs-6.19-rc1.iomap\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull iomap updates from Christian Brauner:\n \"FUSE iomap Support for Buffered Reads:\n\n    This adds iomap support for FUSE buffered reads and readahead. This\n    enables granular uptodate tracking with large folios so only\n    non-uptodate portions need to be read. Also fixes a race condition\n    with large folios + writeback cache that could cause data corruption\n    on partial writes followed by reads.\n\n     - Refactored iomap read/readahead bio logic into helpers\n     - Added caller-provided callbacks for read operations\n     - Moved buffered IO bio logic into new file\n     - FUSE now uses iomap for read_folio and readahead\n\n  Zero Range Folio Batch Support:\n\n    Add folio batch support for iomap_zero_range() to handle dirty\n    folios over unwritten mappings. Fix raciness issues where dirty data\n    could be lost during zero range operations.\n\n     - filemap_get_folios_tag_range() helper for dirty folio lookup\n     - Optional zero range dirty folio processing\n     - XFS fills dirty folios on zero range of unwritten mappings\n     - Removed old partial EOF zeroing optimization\n\n  DIO Write Completions from Interrupt Context:\n\n    Restore pre-iomap behavior where pure overwrite completions run\n    inline rather than being deferred to workqueue. Reduces context\n    switches for high-performance workloads like ScyllaDB.\n\n     - Removed unused IOCB_DIO_CALLER_COMP code\n     - Error completions always run in user context (fixes zonefs)\n     - Reworked REQ_FUA selection logic\n     - Inverted IOMAP_DIO_INLINE_COMP to IOMAP_DIO_OFFLOAD_COMP\n\n  Buffered IO Cleanups:\n\n    Some performance and code clarity improvements:\n\n     - Replace manual bitmap scanning with find_next_bit()\n     - Simplify read skip logic for writes\n     - Optimize pending async writeback accounting\n     - Better variable naming\n     - Documentation for iomap_finish_folio_write() requirements\n\n  Misaligned Vectors for Zoned XFS:\n\n    Enables sub-block aligned vectors in XFS always-COW mode for zoned\n    devices via new IOMAP_DIO_FSBLOCK_ALIGNED flag.\n\n  Bug Fixes:\n\n     - Allocate s_dio_done_wq for async reads (fixes syzbot report after\n       error completion changes)\n     - Fix iomap_read_end() for already uptodate folios (regression fix)\"\n\n* tag \u0027vfs-6.19-rc1.iomap\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: (40 commits)\n  iomap: allocate s_dio_done_wq for async reads as well\n  iomap: fix iomap_read_end() for already uptodate folios\n  iomap: invert the polarity of IOMAP_DIO_INLINE_COMP\n  iomap: support write completions from interrupt context\n  iomap: rework REQ_FUA selection\n  iomap: always run error completions in user context\n  fs, iomap: remove IOCB_DIO_CALLER_COMP\n  iomap: use find_next_bit() for uptodate bitmap scanning\n  iomap: use find_next_bit() for dirty bitmap scanning\n  iomap: simplify when reads can be skipped for writes\n  iomap: simplify -\u003eread_folio_range() error handling for reads\n  iomap: optimize pending async writeback accounting\n  docs: document iomap writeback\u0027s iomap_finish_folio_write() requirement\n  iomap: account for unaligned end offsets when truncating read range\n  iomap: rename bytes_pending/bytes_accounted to bytes_submitted/bytes_not_submitted\n  xfs: support sub-block aligned vectors in always COW mode\n  iomap: add IOMAP_DIO_FSBLOCK_ALIGNED flag\n  xfs: error tag to force zeroing on debug kernels\n  iomap: remove old partial eof zeroing optimization\n  xfs: fill dirty folios on zero range of unwritten mappings\n  ...\n"
    },
    {
      "commit": "e2349c5811ae642f6d948d1ee4bfc4a93566c178",
      "tree": "fea218c9e7374ac2e851064d8526f4a14eb36f3a",
      "parents": [
        "69acbdbbefbda7b7b32faa706a8f68c399c9e47b",
        "ef1b6d904993d3a21baa7d4105e1a4e4ba9dd6de",
        "24e3848a2e903fe614fbeef3048fa4fe91e46482"
      ],
      "author": {
        "name": "Borislav Petkov (AMD)",
        "email": "bp@alien8.de",
        "time": "Mon Dec 01 12:06:08 2025 +0100"
      },
      "committer": {
        "name": "Borislav Petkov (AMD)",
        "email": "bp@alien8.de",
        "time": "Mon Dec 01 12:06:08 2025 +0100"
      },
      "message": "Merge remote-tracking branches \u0027ras/edac-amd-atl\u0027, \u0027ras/edac-drivers\u0027 and \u0027ras/edac-misc\u0027 into edac-updates\n\nSigned-off-by: Borislav Petkov (AMD) \u003cbp@alien8.de\u003e\n"
    },
    {
      "commit": "6ec33db1aaf06a76fb063610e668f8e12f32ebbf",
      "tree": "0aaf477b1e67a97bde3171664ded5bc2c3d8cbdf",
      "parents": [
        "59bfa6408214b6533d8691715cf5459e89b45b89"
      ],
      "author": {
        "name": "Ingo Molnar",
        "email": "mingo@kernel.org",
        "time": "Mon Dec 01 10:42:27 2025 +0100"
      },
      "committer": {
        "name": "Ingo Molnar",
        "email": "mingo@kernel.org",
        "time": "Mon Dec 01 10:42:27 2025 +0100"
      },
      "message": "objtool: Fix segfault on unknown alternatives\n\nSo \u0027objtool --link -d vmlinux.o\u0027 gets surprised by this endbr64+endbr64 pattern\nin ___bpf_prog_run():\n\n\t___bpf_prog_run:\n\t1e7680:  ___bpf_prog_run+0x0                                                     push   %r12\n\t1e7682:  ___bpf_prog_run+0x2                                                     mov    %rdi,%r12\n\t1e7685:  ___bpf_prog_run+0x5                                                     push   %rbp\n\t1e7686:  ___bpf_prog_run+0x6                                                     xor    %ebp,%ebp\n\t1e7688:  ___bpf_prog_run+0x8                                                     push   %rbx\n\t1e7689:  ___bpf_prog_run+0x9                                                     mov    %rsi,%rbx\n\t1e768c:  ___bpf_prog_run+0xc                                                     movzbl (%rbx),%esi\n\t1e768f:  ___bpf_prog_run+0xf                                                     movzbl %sil,%edx\n\t1e7693:  ___bpf_prog_run+0x13                                                    mov    %esi,%eax\n\t1e7695:  ___bpf_prog_run+0x15                                                    mov    0x0(,%rdx,8),%rdx\n\t1e769d:  ___bpf_prog_run+0x1d                                                    jmp    0x1e76a2 \u003c__x86_indirect_thunk_rdx\u003e\n\t1e76a2:  ___bpf_prog_run+0x22                                                    endbr64\n\t1e76a6:  ___bpf_prog_run+0x26                                                    endbr64\n\t1e76aa:  ___bpf_prog_run+0x2a                                                    mov    0x4(%rbx),%edx\n\nAnd crashes due to blindly dereferencing alt-\u003einsn-\u003ealt_group.\n\nBail out on NULL -\u003ealt_group, which produces this warning and continues\nwith the disassembly, instead of a segfault:\n\n  .git/O/vmlinux.o: warning: objtool: \u003calternative.1e769d\u003e: failed to disassemble alternative\n\nCc: Alexandre Chartre \u003calexandre.chartre@oracle.com\u003e\nCc: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\nCc: Josh Poimboeuf \u003cjpoimboe@kernel.org\u003e\nCc: linux-kernel@vger.kernel.org\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\n"
    },
    {
      "commit": "43decb6b628eb033a1b6188e5018773c0d38be1d",
      "tree": "38541f715cdd470670425c0a9ba0f9cb5557851e",
      "parents": [
        "719e357fc09c63238956eb7cd546627f9e050640"
      ],
      "author": {
        "name": "Randy Dunlap",
        "email": "rdunlap@infradead.org",
        "time": "Thu Nov 27 22:59:25 2025 -0800"
      },
      "committer": {
        "name": "Ingo Molnar",
        "email": "mingo@kernel.org",
        "time": "Mon Dec 01 06:56:16 2025 +0100"
      },
      "message": "locking/local_lock: Fix all kernel-doc warnings\n\nModify kernel-doc comments in local_lock.h to prevent warnings:\n\n  Warning: include/linux/local_lock.h:9 function parameter \u0027lock\u0027 not described in \u0027local_lock_init\u0027\n  Warning: include/linux/local_lock.h:56 function parameter \u0027lock\u0027 not described in \u0027local_trylock_init\u0027\n  Warning: include/linux/local_lock.h:56 expecting prototype for local_lock_init(). Prototype was for local_trylock_init() instead\n\nSigned-off-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\nSigned-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nLink: https://patch.msgid.link/20251128065925.917917-1-rdunlap@infradead.org\n"
    },
    {
      "commit": "719e357fc09c63238956eb7cd546627f9e050640",
      "tree": "0e5e4e656e0bf61a76ebf5545cfbab16f0e9d1e6",
      "parents": [
        "52ed746147140e30419ee852c1916531b4ef9b0a"
      ],
      "author": {
        "name": "Vincent Mailhol",
        "email": "mailhol@kernel.org",
        "time": "Thu Nov 27 15:41:40 2025 +0100"
      },
      "committer": {
        "name": "Ingo Molnar",
        "email": "mingo@kernel.org",
        "time": "Mon Dec 01 06:56:16 2025 +0100"
      },
      "message": "locking/local_lock: s/l/__l/ and s/tl/__tl/ to reduce the risk of shadowing\n\nThe Linux kernel coding style advises to avoid common variable\nnames in function-like macros to reduce the risk of namespace\ncollisions.\n\nThroughout local_lock_internal.h, several macros use the rather common\nvariable names \u0027l\u0027 and \u0027tl\u0027. This already resulted in an actual\ncollision: the __local_lock_acquire() function like macro is currently\nshadowing the parameter \u0027l\u0027 of the:\n\n  class_##_name##_t class_##_name##_constructor(_type *l)\n\nfunction factory from \u003clinux/cleanup.h\u003e.\n\nRename the variable \u0027l\u0027 to \u0027__l\u0027 and the variable \u0027tl\u0027 to \u0027__tl\u0027\nthroughout the file to fix the current namespace collision and\nto prevent future ones.\n\n[ bigeasy: Rebase, update all l and tl instances in macros ]\n\nSigned-off-by: Vincent Mailhol \u003cmailhol@kernel.org\u003e\nSigned-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nSigned-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nAcked-by: Waiman Long \u003clongman@redhat.com\u003e\nLink: https://patch.msgid.link/20251127144140.215722-3-bigeasy@linutronix.de\n"
    },
    {
      "commit": "52ed746147140e30419ee852c1916531b4ef9b0a",
      "tree": "ce653b88d5265ce68f8628612ea86ec4e1aa4e1c",
      "parents": [
        "51d7a054521de7085783a9a1ba15c3530863409a"
      ],
      "author": {
        "name": "Sebastian Andrzej Siewior",
        "email": "bigeasy@linutronix.de",
        "time": "Thu Nov 27 15:41:39 2025 +0100"
      },
      "committer": {
        "name": "Ingo Molnar",
        "email": "mingo@kernel.org",
        "time": "Mon Dec 01 06:56:10 2025 +0100"
      },
      "message": "locking/local_lock: Add the \u003clinux/local_lock*.h\u003e headers to MAINTAINERS\n\nThe local_lock_t was never added to the MAINTAINERS file since its\ninclusion.\n\nAdd local_lock_t to the locking primitives section.\n\nSigned-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nSigned-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nAcked-by: Waiman Long \u003clongman@redhat.com\u003e\nLink: https://patch.msgid.link/20251127144140.215722-2-bigeasy@linutronix.de\n"
    },
    {
      "commit": "51d7a054521de7085783a9a1ba15c3530863409a",
      "tree": "8390bf7fd24ec364b4af0150a4e74703e5f618ec",
      "parents": [
        "f74cf399e02e24c544b0bd4b1fe8fa2c5ae30b18"
      ],
      "author": {
        "name": "Sebastian Andrzej Siewior",
        "email": "bigeasy@linutronix.de",
        "time": "Wed Nov 05 15:23:50 2025 +0100"
      },
      "committer": {
        "name": "Ingo Molnar",
        "email": "mingo@kernel.org",
        "time": "Mon Dec 01 06:51:57 2025 +0100"
      },
      "message": "locking/mutex: Redo __mutex_init() to reduce generated code size\n\nmutex_init() invokes __mutex_init() providing the name of the lock and\na pointer to a the lock class. With LOCKDEP enabled this information is\nuseful but without LOCKDEP it not used at all. Passing the pointer\ninformation of the lock class might be considered negligible but the\nname of the lock is passed as well and the string is stored. This\ninformation is wasting storage.\n\nSplit __mutex_init() into a _genereic() variant doing the initialisation\nof the lock and a _lockdep() version which does _genereic() plus the\nlockdep bits. Restrict the lockdep version to lockdep enabled builds\nallowing the compiler to remove the unused parameter.\n\nThis results in the following size reduction:\n\n        text     data       bss        dec  filename\n  | 30237599  8161430   1176624   39575653  vmlinux.defconfig\n  | 30233269  8149142   1176560   39558971  vmlinux.defconfig.patched\n     -4.2KiB   -12KiB\n\n  | 32455099  8471098  12934684   53860881  vmlinux.defconfig.lockdep\n  | 32455100  8471098  12934684   53860882  vmlinux.defconfig.patched.lockdep\n\n  | 27152407  7191822   2068040   36412269  vmlinux.defconfig.preempt_rt\n  | 27145937  7183630   2067976   36397543  vmlinux.defconfig.patched.preempt_rt\n     -6.3KiB    -8KiB\n\n  | 29382020  7505742  13784608   50672370  vmlinux.defconfig.preempt_rt.lockdep\n  | 29376229  7505742  13784544   50666515  vmlinux.defconfig.patched.preempt_rt.lockdep\n     -5.6KiB\n\n[peterz: folded fix from boqun]\n\nSigned-off-by: Sebastian Andrzej Siewior \u003cbigeasy@linutronix.de\u003e\nSigned-off-by: Boqun Feng \u003cboqun.feng@gmail.com\u003e\nSigned-off-by: Peter Zijlstra (Intel) \u003cpeterz@infradead.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nReviewed-by: Waiman Long \u003clongman@redhat.com\u003e\nLink: https://lkml.kernel.org/r/20251125145425.68319-1-boqun.feng@gmail.com\nLink: https://patch.msgid.link/20251105142350.Tfeevs2N@linutronix.de\n"
    },
    {
      "commit": "7d0a66e4bb9081d75c82ec4957c50034cb0ea449",
      "tree": "61e7217a84de6106ad5cc06aa62164e1862d2bdc",
      "parents": [
        "e69c7c175115c51c7f95394fc55425a395b3af59"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Nov 30 14:42:10 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Nov 30 14:42:10 2025 -0800"
      },
      "message": "Linux 6.18\n"
    },
    {
      "commit": "e69c7c175115c51c7f95394fc55425a395b3af59",
      "tree": "eda96dee5b666e2cba563107d8546a85cada2ac6",
      "parents": [
        "6bda50f4333fa61c07f04f790fdd4e2c9f4ca610",
        "c7418164b463056bf4327b6a2abe638b78250f13"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Nov 30 08:47:10 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sun Nov 30 08:47:10 2025 -0800"
      },
      "message": "Merge tag \u0027timers_urgent_for_v6.18_rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull timer fix from Borislav Petkov:\n\n - Have timekeeping aux clocks sysfs interface setup function return an\n   error code on failure instead of success\n\n* tag \u0027timers_urgent_for_v6.18_rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n  timekeeping: Fix error code in tk_aux_sysfs_init()\n"
    },
    {
      "commit": "6bda50f4333fa61c07f04f790fdd4e2c9f4ca610",
      "tree": "f87e3105d67989e66964dcd4c08144cf999850e7",
      "parents": [
        "19eef1d98eeda3745df35839190b7d4a4adea656",
        "841ecc979b18d3227fad5e2d6a1e6f92688776b5"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sat Nov 29 15:15:14 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Sat Nov 29 15:15:14 2025 -0800"
      },
      "message": "Merge tag \u0027mips-fixes_6.18_2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux\n\nPull MIPS fix from Thomas Bogendoerfer:\n \"Fix TLB unification for cores with more than 64 TLB entries\"\n\n* tag \u0027mips-fixes_6.18_2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mips/linux:\n  MIPS: mm: kmalloc tlb_vpn array to avoid stack overflow\n"
    },
    {
      "commit": "841ecc979b18d3227fad5e2d6a1e6f92688776b5",
      "tree": "a9e37859e45da68bdad9a19eaea3b3c25b78563e",
      "parents": [
        "ac3fd01e4c1efce8f2c054cdeb2ddd2fc0fb150d"
      ],
      "author": {
        "name": "Thomas Bogendoerfer",
        "email": "tsbogend@alpha.franken.de",
        "time": "Fri Nov 28 16:53:46 2025 +0000"
      },
      "committer": {
        "name": "Thomas Bogendoerfer",
        "email": "tsbogend@alpha.franken.de",
        "time": "Sat Nov 29 13:36:05 2025 +0100"
      },
      "message": "MIPS: mm: kmalloc tlb_vpn array to avoid stack overflow\n\nOwing to Config4.MMUSizeExt and VTLB/FTLB MMU features later MIPSr2+\ncores can have more than 64 TLB entries.  Therefore allocate an array\nfor uniquification instead of placing too an small array on the stack.\n\nFixes: 35ad7e181541 (\"MIPS: mm: tlb-r4k: Uniquify TLB entries on init\")\nCo-developed-by: Maciej W. Rozycki \u003cmacro@orcam.me.uk\u003e\nSigned-off-by: Maciej W. Rozycki \u003cmacro@orcam.me.uk\u003e\nCc: stable@vger.kernel.org # v6.17+: 9f048fa48740: MIPS: mm: Prevent a TLB shutdown on initial uniquification\nCc: stable@vger.kernel.org # v6.17+\nTested-by: Gregory CLEMENT \u003cgregory.clement@bootlin.com\u003e\nTested-by: Klara Modin \u003cklarasmodin@gmail.com\u003e\nSigned-off-by: Thomas Bogendoerfer \u003ctsbogend@alpha.franken.de\u003e\n"
    },
    {
      "commit": "19eef1d98eeda3745df35839190b7d4a4adea656",
      "tree": "8d0b29825fa335ad0b98a8fc9e09d34d87d5e96b",
      "parents": [
        "e6640487845061255af9614ec0a192e4fafa486e"
      ],
      "author": {
        "name": "David Howells",
        "email": "dhowells@redhat.com",
        "time": "Sat Nov 29 00:40:11 2025 +0000"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 16:48:18 2025 -0800"
      },
      "message": "afs: Fix uninit var in afs_alloc_anon_key()\n\nFix an uninitialised variable (key) in afs_alloc_anon_key() by setting it\nto cell-\u003eanonymous_key.  Without this change, the error check may return a\nfalse failure with a bad error number.\n\nMost of the time this is unlikely to happen because the first encounter\nwith afs_alloc_anon_key() will usually be from (auto)mount, for which all\nsubsequent operations must wait - apart from other (auto)mounts.  Once the\ncall-\u003eanonymous_key is allocated, all further calls to afs_request_key()\nwill skip the call to afs_alloc_anon_key() for that cell.\n\nFixes: d27c71257825 (\"afs: Fix delayed allocation of a cell\u0027s anonymous key\")\nReported-by: Paulo Alcantra \u003cpc@manguebit.org\u003e\nSigned-off-by: David Howells \u003cdhowells@redhat.com\u003e\nReviewed-by: Paulo Alcantara \u003cpc@manguebit.org\u003e\ncc: Marc Dionne \u003cmarc.dionne@auristor.com\u003e\ncc: syzbot+41c68824eefb67cdf00c@syzkaller.appspotmail.com\ncc: linux-afs@lists.infradead.org\ncc: linux-fsdevel@vger.kernel.org\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "e6640487845061255af9614ec0a192e4fafa486e",
      "tree": "39cb744281ad49ea6fba0310683d0c340381e32f",
      "parents": [
        "82ebd4e32029f299011a6ef68d96f5ddcea0723a",
        "40ad64ac25bb736740f895d99a4aebbda9b80991"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 14:08:09 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 14:08:09 2025 -0800"
      },
      "message": "Merge tag \u0027spi-fix-v6.18-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi\n\nPull spi fixes from Mark Brown:\n \"A disappointingly large set of device specific fixes that have built\n  up since I\u0027ve been a bit tardy with sending a pull requests as people\n  kept sending me new new fixes.\n\n  The bcm63xx and lpspi issues could lead to corruption so the fixes are\n  fairly important for the affected parts, the other issues should all\n  be relatively minor\"\n\n* tag \u0027spi-fix-v6.18-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:\n  spi: nxp-fspi: Propagate fwnode in ACPI case as well\n  spi: tegra114: remove Kconfig dependency on TEGRA20_APB_DMA\n  spi: amlogic-spifc-a1: Handle devm_pm_runtime_enable() errors\n  spi: spi-fsl-lpspi: fix watermark truncation caused by type cast\n  spi: cadence-quadspi: Fix cqspi_probe() error handling for runtime pm\n  spi: bcm63xx: fix premature CS deassertion on RX-only transactions\n  spi: spi-cadence-quadspi: Remove duplicate pm_runtime_put_autosuspend() call\n  spi: spi-cadence-quadspi: Enable pm runtime earlier to avoid imbalance\n"
    },
    {
      "commit": "82ebd4e32029f299011a6ef68d96f5ddcea0723a",
      "tree": "a9f7a54f25e42ec6a417368a3e5ab6dcc98794bc",
      "parents": [
        "9917bf8e7f5a9efbab844cf06cfd8da8eb7e13b6",
        "8684229e19c4185d53d6fb7004d733907c865a91"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 14:06:05 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 14:06:05 2025 -0800"
      },
      "message": "Merge tag \u0027regulator-fix-v6.18-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator\n\nPull regulator fixes from Mark Brown:\n \"A couple of fixes for incorrect device descriptions in the rtq2208\n  driver\"\n\n* tag \u0027regulator-fix-v6.18-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator:\n  regulator: rtq2208: Correct LDO2 logic judgment bits\n  regulator: rtq2208: Correct buck group2 phase mapping logic\n"
    },
    {
      "commit": "9917bf8e7f5a9efbab844cf06cfd8da8eb7e13b6",
      "tree": "a390d32ccf610e7929e58fad97ee5ff5e42662c4",
      "parents": [
        "f3b17337b943949d0f3d12835d10d866210aeee8",
        "f6dc5a36195d3f5be769f60d6987150192dfb099"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 12:42:11 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 12:42:11 2025 -0800"
      },
      "message": "Merge tag \u0027io_uring-6.18-20251128\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux\n\nPull io_uring fixes from Jens Axboe:\n\n - Ensure that vectored registered buffer imports ties the lifetime of\n   those to the zero-copy send notification, not the parent request\n\n - Fix a bug introduced in this merge window, with the introduction of\n   mixed sized CQE support\n\n* tag \u0027io_uring-6.18-20251128\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux:\n  io_uring: fix mixed cqe overflow handling\n  io_uring/net: ensure vectored buffer node import is tied to notification\n"
    },
    {
      "commit": "d911fe6e942e60900577314dc1f1529b90e4da07",
      "tree": "ef0b7b0d06b9226db55b2397605fefa2852c2880",
      "parents": [
        "3d1f1088455d9a9bce51f0c1e6a81f518a5cb468"
      ],
      "author": {
        "name": "Harry Fellowes",
        "email": "harryfellowes1@gmail.com",
        "time": "Mon Aug 25 20:28:34 2025 +0100"
      },
      "committer": {
        "name": "Borislav Petkov (AMD)",
        "email": "bp@alien8.de",
        "time": "Fri Nov 28 20:29:52 2025 +0100"
      },
      "message": "x86/boot: Clean up whitespace in a20.c\n\nRemove trailing whitespace on empty lines.\n\nNo functional changes.\n\n  [ bp: Massage commit message. ]\n\nSigned-off-by: Harry Fellowes \u003charryfellowes1@gmail.com\u003e\nSigned-off-by: Borislav Petkov (AMD) \u003cbp@alien8.de\u003e\nLink: https://patch.msgid.link/20250825192832.6444-3-harryfellowes1@gmail.com\n"
    },
    {
      "commit": "f3b17337b943949d0f3d12835d10d866210aeee8",
      "tree": "1eab6626c88ea639473bf6d522e008c903679466",
      "parents": [
        "7fa0d7744c28267142287a75134de58790999e72",
        "d27c71257825dced46104eefe42e4d9964bd032e"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 10:01:24 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 10:01:24 2025 -0800"
      },
      "message": "Merge tag \u0027vfs-6.18-rc8.fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull vfs fixes from Christian Brauner:\n\n - afs: Fix delayed allocation of a cell\u0027s anonymous key\n\n   The allocation of a cell\u0027s anonymous key is done in a background\n   thread along with other cell setup such as doing a DNS upcall. The\n   normal key lookup tries to use the key description on the anonymous\n   authentication key as the reference for request_key() - but it may\n   not yet be set, causing an oops\n\n - ovl: fail ovl_lock_rename_workdir() if either target is unhashed\n\n   As well as checking that the parent hasn\u0027t changed after getting the\n   lock, the code needs to check that the dentry hasn\u0027t been unhashed.\n   Otherwise overlayfs might try to rename something that has been\n   removed\n\n - namespace: fix a reference leak in grab_requested_mnt_ns\n\n   lookup_mnt_ns() already takes a reference on mnt_ns, and so\n   grab_requested_mnt_ns() doesn\u0027t need to take an extra reference\n\n* tag \u0027vfs-6.18-rc8.fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:\n  afs: Fix delayed allocation of a cell\u0027s anonymous key\n  ovl: fail ovl_lock_rename_workdir() if either target is unhashed\n  fs/namespace: fix reference leak in grab_requested_mnt_ns\n"
    },
    {
      "commit": "7fa0d7744c28267142287a75134de58790999e72",
      "tree": "3005f185ddd899780bd3645adf6b8d473f6e486b",
      "parents": [
        "6cf62f0174de64e4161e301bb0ed52e198ce25dc",
        "3ecfcf34f0992d11e5eb2458a2108b3320c61ba4"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 09:57:31 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 09:57:31 2025 -0800"
      },
      "message": "Merge tag \u0027soc-fixes-6.18-4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc\n\nPull SoC fixes from Arnd Bergmann:\n \"A few last minute fixes came in this week:\n\n   - interrupt and gpio numbers in foud separate i.MX8 specific\n     devicetree files were wrong\n\n   - The vector length property in the C906 CPU description used the\n     wrong unit\n\n   - Two bugs with uninitialized stack variables in the tee subsystem\n\n   - Alexander Stein now maintains additional devicetree files\"\n\n* tag \u0027soc-fixes-6.18-4\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc:\n  riscv: dts: allwinner: d1: fix vlenb property\n  MAINTAINERS: Add entry for TQ-Systems AM335 device trees\n  tee: qcomtee: initialize result before use in release worker\n  arm64: dts: imx8qm-mek: fix mux-controller select/enable-gpios polarity\n  tee: qcomtee: fix uninitialized pointers with free attribute\n  ARM: dts: nxp: imx6ul: correct SAI3 interrupt line\n  arm64: dts: imx8dxl-ss-conn: swap interrupts number of eqos\n  arm64: dts: imx8dxl: Correct pcie-ep interrupt number\n"
    },
    {
      "commit": "6cf62f0174de64e4161e301bb0ed52e198ce25dc",
      "tree": "f697250b484083386113e8d1d3e8aaa3cd36f76d",
      "parents": [
        "dabf127d641e43d5fbb72a1f48209818349db7ef",
        "03bc4831ef064e114328dea906101cff7c6fb8b3"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 09:44:48 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 09:44:48 2025 -0800"
      },
      "message": "Merge tag \u0027char-misc-6.18-rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc\n\nPull char / misc / IIO fixes from Greg KH:\n \"Here are some much-delayed char/misc/iio driver fixes for 6.18-rc8.\n\n  Fixes in here include:\n\n   - lots of iio driver bugfixes for reported issues.\n\n   - counter driver bugfix\n\n   - slimbus driver bugfix\n\n   - mei tiny bugfix\n\n   - nvmem layout uevent bugfix\n\n  All of these have been in linux-next for a while, but due to travel on\n  my side, I haven\u0027t had a chance to get them to you\"\n\n* tag \u0027char-misc-6.18-rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc: (23 commits)\n  nvmem: layouts: fix nvmem_layout_bus_uevent\n  iio: accel: bmc150: Fix irq assumption regression\n  most: usb: fix double free on late probe failure\n  slimbus: ngd: Fix reference count leak in qcom_slim_ngd_notify_slaves\n  firmware: stratix10-svc: fix bug in saving controller data\n  mei: fix error flow in probe\n  iio: st_lsm6dsx: Fixed calibrated timestamp calculation\n  iio: humditiy: hdc3020: fix units for thresholds and hysteresis\n  iio: humditiy: hdc3020: fix units for temperature and humidity measurement\n  iio: imu: st_lsm6dsx: fix array size for st_lsm6dsx_settings fields\n  iio: accel: fix ADXL355 startup race condition\n  iio: adc: ad7124: fix temperature channel\n  iio:common:ssp_sensors: Fix an error handling path ssp_probe()\n  iio: adc: ad7280a: fix ad7280_store_balance_timer()\n  iio: buffer-dmaengine: enable .get_dma_dev()\n  iio: buffer-dma: support getting the DMA channel\n  iio: buffer: support getting dma channel from the buffer\n  iio: pressure: bmp280: correct meas_time_us calculation\n  iio: adc: stm32-dfsdm: fix st,adc-alt-channel property handling\n  iio: adc: ad7380: fix SPI offload trigger rate\n  ...\n"
    },
    {
      "commit": "dabf127d641e43d5fbb72a1f48209818349db7ef",
      "tree": "0db4788f6376b39c5d1c3f266acea52643a4408e",
      "parents": [
        "5d324e5159d9e6a1e6678007ce3f24e569650db6",
        "2bf95a9bcb50002ca9d47403d60aedaeb2e19abe"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 09:16:20 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 09:16:20 2025 -0800"
      },
      "message": "Merge tag \u0027tty-6.18-rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty\n\nPull serial driver fixes from Greg KH:\n \"Here are two serial driver fixes for reported issues for 6.18-rc8.\n\n  These are:\n\n   - fix for a much reported symbol build loop that broke the build for\n     some kernel configurations\n\n   - amba-pl011 driver bugfix for a reported issue\n\n  Both have been in linux next (the last for weeks, the first for a\n  shorter amount of time), with no reported issues\"\n\n* tag \u0027tty-6.18-rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:\n  serial: 8250: Fix 8250_rsa symbol loop\n  serial: amba-pl011: prefer dma_mapping_error() over explicit address checking\n"
    },
    {
      "commit": "5d324e5159d9e6a1e6678007ce3f24e569650db6",
      "tree": "cbdb974120a5f26c551072014ba5ab3fd2e6a61f",
      "parents": [
        "24a84ea4eef6ec6245e1efdc64764d60d17321bb",
        "74851fbb6d647304f8a7dc491434d3a335ef4b8d"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 09:12:40 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 09:12:40 2025 -0800"
      },
      "message": "Merge tag \u0027usb-6.18-rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb\n\nPull USB/Thunderbolt fixes from Greg KH:\n \"Here are some last-minutes USB and Thunderbolt driver fixes and new\n  device ids for 6.18-rc8. Included in here are:\n\n   - usb storage quirk fixup\n\n   - xhci driver fixes for reported issues\n\n   - usb gadget driver fixes\n\n   - dwc3 driver fixes\n\n   - UAS driver fixup\n\n   - thunderbolt new device ids\n\n   - usb-serial driver new ids\n\n  All of these have been in linux-next with no reported issues, many for\n  many weeks\"\n\n* tag \u0027usb-6.18-rc8\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (21 commits)\n  usb: gadget: renesas_usbf: Handle devm_pm_runtime_enable() errors\n  USB: storage: Remove subclass and protocol overrides from Novatek quirk\n  usb: uas: fix urb unmapping issue when the uas device is remove during ongoing data transfer\n  usb: dwc3: Fix race condition between concurrent dwc3_remove_requests() call paths\n  xhci: dbgtty: fix device unregister\n  usb: storage: sddr55: Reject out-of-bound new_pba\n  USB: serial: option: add support for Rolling RW101R-GL\n  usb: typec: ucsi: psy: Set max current to zero when disconnected\n  usb: gadget: f_eem: Fix memory leak in eem_unwrap\n  usb: dwc3: pci: Sort out the Intel device IDs\n  usb: dwc3: pci: add support for the Intel Nova Lake -S\n  drivers/usb/dwc3: fix PCI parent check\n  usb: storage: Fix memory leak in USB bulk transport\n  xhci: sideband: Fix race condition in sideband unregister\n  xhci: dbgtty: Fix data corruption when transmitting data form DbC to host\n  xhci: fix stale flag preventig URBs after link state error is cleared\n  USB: serial: ftdi_sio: add support for u-blox EVK-M101\n  usb: cdns3: Fix double resource release in cdns3_pci_probe\n  usb: gadget: udc: fix use-after-free in usb_gadget_state_work\n  usb: renesas_usbhs: Fix synchronous external abort on unbind\n  ...\n"
    },
    {
      "commit": "24a84ea4eef6ec6245e1efdc64764d60d17321bb",
      "tree": "6e962208454ce59b006c8865e17955a4b5ac66b7",
      "parents": [
        "4331989728da4e868fcda75bde872d032f5ce6ac",
        "e3cee98f2fcc2234be7813c0e55a7f7a5e77aaf2"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 09:09:33 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 09:09:33 2025 -0800"
      },
      "message": "Merge tag \u0027mailbox-fixes-v6.18-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jassibrar/mailbox\n\nPull mailbox fixes from Jassi Brar:\n\n - omap: check for pending msgs only when mbox is exclusive\n\n - mailbox-test: debugfs_create_dir error checking\n\n - mtk:\n     - cmdq: fix DMA address handling\n     - gpueb: Add missing \u0027static\u0027 to mailbox ops struct\n\n - pcc: don\u0027t zero error register\n\n - th1520: fix clock imbalance on probe failure\n\n* tag \u0027mailbox-fixes-v6.18-rc7\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jassibrar/mailbox:\n  mailbox: th1520: fix clock imbalance on probe failure\n  mailbox: pcc: don\u0027t zero error register\n  mailbox: mtk-gpueb: Add missing \u0027static\u0027 to mailbox ops struct\n  mailbox: mtk-cmdq: Refine DMA address handling for the command buffer\n  mailbox: mailbox-test: Fix debugfs_create_dir error checking\n  mailbox: omap-mailbox: Check for pending msgs only when mbox is exclusive\n"
    },
    {
      "commit": "3ecfcf34f0992d11e5eb2458a2108b3320c61ba4",
      "tree": "168af3a321808723abcd5115c8e20a17e411fac7",
      "parents": [
        "a6737fe620315db4a47d577703d516cdea960991",
        "9f393d8e757f79060baf4b2e703bd6b2d0d8d323"
      ],
      "author": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Fri Nov 28 17:37:13 2025 +0100"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Fri Nov 28 17:37:13 2025 +0100"
      },
      "message": "Merge tag \u0027sunxi-fixes-for-6.18\u0027 of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux into arm/fixes\n\nAllwinner fixes for 6.18\n\nJust one fix to correct the \"thead,vlenb\" property for the RISC-V based\nD1 SoC family.\n\n* tag \u0027sunxi-fixes-for-6.18\u0027 of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux:\n  riscv: dts: allwinner: d1: fix vlenb property\n"
    },
    {
      "commit": "a6737fe620315db4a47d577703d516cdea960991",
      "tree": "e51ef53ff070fd24a9871bb9f52001d213c9a5d0",
      "parents": [
        "fe7cdcd1bacec3a54ec94131292b5c883cec50cf",
        "517d066dc080fc297f1ab5e8c6fda7c59d48ef8d"
      ],
      "author": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Fri Nov 21 22:14:04 2025 +0100"
      },
      "committer": {
        "name": "Arnd Bergmann",
        "email": "arnd@arndb.de",
        "time": "Fri Nov 28 17:36:52 2025 +0100"
      },
      "message": "Merge tag \u0027omap-for-v6.19/maintainers-signed\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/khilman/linux-omap into arm/fixes\n\nMAINTAINERS: Add entry for TQ-Systems AM335 device trees\n\n* tag \u0027omap-for-v6.19/maintainers-signed\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/khilman/linux-omap:\n  MAINTAINERS: Add entry for TQ-Systems AM335 device trees\n"
    },
    {
      "commit": "4331989728da4e868fcda75bde872d032f5ce6ac",
      "tree": "40c3aca46921f9bd85c91e33e0c29f60f64498b5",
      "parents": [
        "f849f26f77205e25f6ad0b9011b1b68a560d35ff",
        "747528729c9b6733839f9c95f300d5bef95ee52c"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 08:20:14 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 08:20:14 2025 -0800"
      },
      "message": "Merge tag \u0027mmc-v6.18-rc2-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc\n\nPull MMC fix from Ulf Hansson:\n\n - sdhci-of-dwcmshc: Fix reset handling for some variants\n\n* tag \u0027mmc-v6.18-rc2-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc:\n  mmc: sdhci-of-dwcmshc: Promote the th1520 reset handling to ip level\n"
    },
    {
      "commit": "f849f26f77205e25f6ad0b9011b1b68a560d35ff",
      "tree": "d19e745f2bbfa91b2074eff192f77f9822fa73af",
      "parents": [
        "e538109ac71d801d26776af5f3c54f548296c29c",
        "c98c99d5dbdf9fb0063650594edfd7d49b5f4e29"
      ],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 08:08:02 2025 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Fri Nov 28 08:08:02 2025 -0800"
      },
      "message": "Merge tag \u0027pmdomain-v6.18-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/linux-pm\n\nPull pmdomain fixes from Ulf Hansson:\n\n - mediatek: Fix spinlock recursion in probe\n\n - tegra: Use GENPD_FLAG_NO_STAY_ON to restore old behaviour\n\n* tag \u0027pmdomain-v6.18-rc6\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/linux-pm:\n  pmdomain: tegra: Add GENPD_FLAG_NO_STAY_ON flag\n  pmdomains: mtk-pm-domains: Fix spinlock recursion in probe\n"
    },
    {
      "commit": "7cede21e9f04f16a456d3c3c8a9a8899c8d84757",
      "tree": "b1fefc4f8dfee5a03984ea5d65bf82f6342bcf00",
      "parents": [
        "638757c9c9e5a823671367150a9f48e93d115b48",
        "c19dfb267c28032293515a635eaefbf9194629ac",
        "30a8e0a32e7c52c75208aaf5243414b1a6930e6d"
      ],
      "author": {
        "name": "Rafael J. Wysocki",
        "email": "rafael.j.wysocki@intel.com",
        "time": "Fri Nov 28 16:50:45 2025 +0100"
      },
      "committer": {
        "name": "Rafael J. Wysocki",
        "email": "rafael.j.wysocki@intel.com",
        "time": "Fri Nov 28 16:50:45 2025 +0100"
      },
      "message": "Merge branches \u0027pm-qos\u0027 and \u0027pm-tools\u0027\n\nMerge PM QoS updates and a cpupower utility update for 6.19-rc1:\n\n - Introduce and document a QoS limit on CPU exit latency during wakeup\n   from suspend-to-idle (Ulf Hansson)\n\n - Add support for building libcpupower statically (Zuo An)\n\n* pm-qos:\n  Documentation: power/cpuidle: Document the CPU system wakeup latency QoS\n  cpuidle: Respect the CPU system wakeup QoS limit for cpuidle\n  sched: idle: Respect the CPU system wakeup QoS limit for s2idle\n  pmdomain: Respect the CPU system wakeup QoS limit for cpuidle\n  pmdomain: Respect the CPU system wakeup QoS limit for s2idle\n  PM: QoS: Introduce a CPU system wakeup QoS limit\n\n* pm-tools:\n  tools/power/cpupower: Support building libcpupower statically\n"
    },
    {
      "commit": "edde060637b92607f3522252c03d64ad06369933",
      "tree": "ba2bb8a07fa71722c1ecce063d1d4eeceaddb9b8",
      "parents": [
        "52c4d1d62424c574fa048235598ce37bd4ed1e32",
        "0c2988aaa4d3eda94b738d5a7acae7838d52fe4d"
      ],
      "author": {
        "name": "Catalin Marinas",
        "email": "catalin.marinas@arm.com",
        "time": "Fri Nov 28 15:48:03 2025 +0000"
      },
      "committer": {
        "name": "Catalin Marinas",
        "email": "catalin.marinas@arm.com",
        "time": "Fri Nov 28 15:48:03 2025 +0000"
      },
      "message": "Merge branch \u0027for-next/set_memory\u0027 into for-next/core\n\n* for-next/set_memory:\n  : Fix + documentation for the arm64 change_memory_common()\n  arm64/mm: Document why linear map split failure upon vm_reset_perms is not problematic\n  arm64/pageattr: Propagate return value from __change_memory_common\n"
    },
    {
      "commit": "52c4d1d62424c574fa048235598ce37bd4ed1e32",
      "tree": "a63e8649e6258a9fa49a651c8f7227a8f45eda5c",
      "parents": [
        "17c05cb0ef212d1a3c1bbbb67693a89cd9f0fb44",
        "27abb1ee5a4e02a5314423371dafaf41499314a5"
      ],
      "author": {
        "name": "Catalin Marinas",
        "email": "catalin.marinas@arm.com",
        "time": "Fri Nov 28 15:47:53 2025 +0000"
      },
      "committer": {
        "name": "Catalin Marinas",
        "email": "catalin.marinas@arm.com",
        "time": "Fri Nov 28 15:47:53 2025 +0000"
      },
      "message": "Merge branch \u0027for-next/sysreg\u0027 into for-next/core\n\n* for-next/sysreg:\n  : arm64 sysreg updates/cleanups\n  arm64/sysreg: Remove unused define ARM64_FEATURE_FIELD_BITS\n  KVM: arm64: selftests: Consider all 7 possible levels of cache\n  KVM: arm64: selftests: Remove ARM64_FEATURE_FIELD_BITS and its last user\n  arm64/sysreg: Add ICH_VMCR_EL2\n  arm64/sysreg: Move generation of RES0/RES1/UNKN to function\n  arm64/sysreg: Support feature-specific fields with \u0027Prefix\u0027 descriptor\n  arm64/sysreg: Fix checks for incomplete sysreg definitions\n  arm64/sysreg: Replace TCR_EL1 field macros\n"
    },
    {
      "commit": "e3cee98f2fcc2234be7813c0e55a7f7a5e77aaf2",
      "tree": "ee6c5d21fba8bb6ec03aea8658ee25e0c712dc3f",
      "parents": [
        "ff0e4d4c97c94af34cc9cad37b5a5cdbe597a3b0"
      ],
      "author": {
        "name": "Johan Hovold",
        "email": "johan@kernel.org",
        "time": "Fri Oct 17 07:54:14 2025 +0200"
      },
      "committer": {
        "name": "Jassi Brar",
        "email": "jassisinghbrar@gmail.com",
        "time": "Fri Nov 28 09:47:44 2025 -0600"
      },
      "message": "mailbox: th1520: fix clock imbalance on probe failure\n\nThe purpose of the devm_add_action_or_reset() helper is to call the\naction function in case adding an action ever fails so drop the clock\ndisable from the error path to avoid disabling the clocks twice.\n\nFixes: 5d4d263e1c6b (\"mailbox: Introduce support for T-head TH1520 Mailbox driver\")\nCc: Michal Wilczynski \u003cm.wilczynski@samsung.com\u003e\nSigned-off-by: Johan Hovold \u003cjohan@kernel.org\u003e\nReviewed-by: Michal Wilczynski \u003cm.wilczynski@samsung.com\u003e\nSigned-off-by: Jassi Brar \u003cjassisinghbrar@gmail.com\u003e\n"
    },
    {
      "commit": "17c05cb0ef212d1a3c1bbbb67693a89cd9f0fb44",
      "tree": "a33a174126e4e2519f89f6622e0f36518b1fdc00",
      "parents": [
        "e6a27290d8001538fec94e91ff8c7f956ee7e3e5",
        "c86d9f8764ba2ffa4e19cca40918c12ccc3ad909",
        "a7717cad615f2b169ccca93418a07eaa526b4a1a",
        "a5baf582f4c026c25a206ac121bceade926aec74",
        "287d163322b743a50adcad25c851600c004f59e3",
        "337f7e3a4b4d60b41d3e9a389675bb9353925183",
        "a0245b42f881be6f0ddf678ce7a0d150362938c8",
        "cb1fa2e999558fd93b519f7c4c16e75e805af1e6",
        "ce1e1421f8d8cdb5e05e13dbb516caedd67e5ee8",
        "155f8d4ef0b78afbf25b1449bbd654fd1327cc7a",
        "4b7a59fa700f422217d83a7212ccc6074ebe9cbc"
      ],
      "author": {
        "name": "Catalin Marinas",
        "email": "catalin.marinas@arm.com",
        "time": "Fri Nov 28 15:47:12 2025 +0000"
      },
      "committer": {
        "name": "Catalin Marinas",
        "email": "catalin.marinas@arm.com",
        "time": "Fri Nov 28 15:47:12 2025 +0000"
      },
      "message": "Merge branches \u0027for-next/misc\u0027, \u0027for-next/kselftest\u0027, \u0027for-next/efi-preempt\u0027, \u0027for-next/assembler-macro\u0027, \u0027for-next/typos\u0027, \u0027for-next/sme-ptrace-disable\u0027, \u0027for-next/local-tlbi-page-reused\u0027, \u0027for-next/mpam\u0027, \u0027for-next/acpi\u0027 and \u0027for-next/documentation\u0027, remote-tracking branch \u0027arm64/for-next/perf\u0027 into for-next/core\n\n* arm64/for-next/perf:\n  perf: arm_spe: Add support for filtering on data source\n  perf: Add perf_event_attr::config4\n  perf/imx_ddr: Add support for PMU in DB (system interconnects)\n  perf/imx_ddr: Get and enable optional clks\n  perf/imx_ddr: Move ida_alloc() from ddr_perf_init() to ddr_perf_probe()\n  dt-bindings: perf: fsl-imx-ddr: Add compatible string for i.MX8QM, i.MX8QXP and i.MX8DXL\n  arch_topology: Provide a stub topology_core_has_smt() for !CONFIG_GENERIC_ARCH_TOPOLOGY\n  perf/arm-ni: Fix and optimise register offset calculation\n  perf: arm_pmuv3: Add new Cortex and C1 CPU PMUs\n  perf: arm_cspmu: fix error handling in arm_cspmu_impl_unregister()\n  perf/arm-ni: Add NoC S3 support\n  perf/arm_cspmu: nvidia: Add pmevfiltr2 support\n  perf/arm_cspmu: nvidia: Add revision id matching\n  perf/arm_cspmu: Add pmpidr support\n  perf/arm_cspmu: Add callback to reset filter config\n  perf: arm_pmuv3: Don\u0027t use PMCCNTR_EL0 on SMT cores\n\n* for-next/misc:\n  : Miscellaneous patches\n  arm64: atomics: lse: Remove unused parameters from ATOMIC_FETCH_OP_AND macros\n  arm64: remove duplicate ARCH_HAS_MEM_ENCRYPT\n  arm64: mm: use untagged address to calculate page index\n  arm64: mm: make linear mapping permission update more robust for patial range\n  arm64/mm: Elide TLB flush in certain pte protection transitions\n  arm64/mm: Rename try_pgd_pgtable_alloc_init_mm\n  arm64/mm: Allow __create_pgd_mapping() to propagate pgtable_alloc() errors\n  arm64: add unlikely hint to MTE async fault check in el0_svc_common\n  arm64: acpi: add newline to deferred APEI warning\n  arm64: entry: Clean out some indirection\n  arm64/mm: Ensure PGD_SIZE is aligned to 64 bytes when PA_BITS \u003d 52\n  arm64/mm: Drop cpu_set_[default|idmap]_tcr_t0sz()\n  arm64: remove unused ARCH_PFN_OFFSET\n  arm64: use SOFTIRQ_ON_OWN_STACK for enabling softirq stack\n  arm64: Remove assertion on CONFIG_VMAP_STACK\n\n* for-next/kselftest:\n  : arm64 kselftest patches\n  kselftest/arm64: Align zt-test register dumps\n\n* for-next/efi-preempt:\n  : arm64: Make EFI calls preemptible\n  arm64/efi: Call EFI runtime services without disabling preemption\n  arm64/efi: Move uaccess en/disable out of efi_set_pgd()\n  arm64/efi: Drop efi_rt_lock spinlock from EFI arch wrapper\n  arm64/fpsimd: Permit kernel mode NEON with IRQs off\n  arm64/fpsimd: Don\u0027t warn when EFI execution context is preemptible\n  efi/runtime-wrappers: Keep track of the efi_runtime_lock owner\n  efi: Add missing static initializer for efi_mm::cpus_allowed_lock\n\n* for-next/assembler-macro:\n  : arm64: Replace __ASSEMBLY__ with __ASSEMBLER__ in headers\n  arm64: Replace __ASSEMBLY__ with __ASSEMBLER__ in non-uapi headers\n  arm64: Replace __ASSEMBLY__ with __ASSEMBLER__ in uapi headers\n\n* for-next/typos:\n  : Random typo/spelling fixes\n  arm64: Fix double word in comments\n  arm64: Fix typos and spelling errors in comments\n\n* for-next/sme-ptrace-disable:\n  : Support disabling streaming mode via ptrace on SME only systems\n  kselftest/arm64: Cover disabling streaming mode without SVE in fp-ptrace\n  kselftst/arm64: Test NT_ARM_SVE FPSIMD format writes on non-SVE systems\n  arm64/sme: Support disabling streaming mode via ptrace on SME only systems\n\n* for-next/local-tlbi-page-reused:\n  : arm64, mm: avoid TLBI broadcast if page reused in write fault\n  arm64, tlbflush: don\u0027t TLBI broadcast if page reused in write fault\n  mm: add spurious fault fixing support for huge pmd\n\n* for-next/mpam: (34 commits)\n  : Basic Arm MPAM driver (more to follow)\n  MAINTAINERS: new entry for MPAM Driver\n  arm_mpam: Add kunit tests for props_mismatch()\n  arm_mpam: Add kunit test for bitmap reset\n  arm_mpam: Add helper to reset saved mbwu state\n  arm_mpam: Use long MBWU counters if supported\n  arm_mpam: Probe for long/lwd mbwu counters\n  arm_mpam: Consider overflow in bandwidth counter state\n  arm_mpam: Track bandwidth counter state for power management\n  arm_mpam: Add mpam_msmon_read() to read monitor value\n  arm_mpam: Add helpers to allocate monitors\n  arm_mpam: Probe and reset the rest of the features\n  arm_mpam: Allow configuration to be applied and restored during cpu online\n  arm_mpam: Use a static key to indicate when mpam is enabled\n  arm_mpam: Register and enable IRQs\n  arm_mpam: Extend reset logic to allow devices to be reset any time\n  arm_mpam: Add a helper to touch an MSC from any CPU\n  arm_mpam: Reset MSC controls from cpuhp callbacks\n  arm_mpam: Merge supported features during mpam_enable() into mpam_class\n  arm_mpam: Probe the hardware features resctrl supports\n  arm_mpam: Add helpers for managing the locking around the mon_sel registers\n  ...\n\n* for-next/acpi:\n  : arm64 acpi updates\n  ACPI: GTDT: Get rid of acpi_arch_timer_mem_init()\n\n* for-next/documentation:\n  : arm64 Documentation updates\n  Documentation/arm64: Fix the typo of register names\n"
    },
    {
      "commit": "638757c9c9e5a823671367150a9f48e93d115b48",
      "tree": "37a552a1211edfc7af6422180933067db220bc0e",
      "parents": [
        "bf7ae1773ea79417d430a2d216abdb9857db8ec1",
        "65df3a9629c10d70593bc90b2ca6b235b7a24909",
        "8dfa8bb6525453f7b63379be54738440d7c908ea"
      ],
      "author": {
        "name": "Rafael J. Wysocki",
        "email": "rafael.j.wysocki@intel.com",
        "time": "Fri Nov 28 16:44:00 2025 +0100"
      },
      "committer": {
        "name": "Rafael J. Wysocki",
        "email": "rafael.j.wysocki@intel.com",
        "time": "Fri Nov 28 16:44:00 2025 +0100"
      },
      "message": "Merge branches \u0027pm-em\u0027 and \u0027pm-opp\u0027\n\nMerge energy model management updates and operating performance points\n(OPP) library changes for 6.19-rc1:\n\n - Add support for sending netlink notifications to user space on energy\n   model updates (Changwoo Mini, Peng Fan)\n\n - Minor improvements to the Rust OPP interface (Tamir Duberstein)\n\n - Fixes to scope-based pointers in the OPP library (Viresh Kumar)\n\n* pm-em:\n  PM: EM: Add to em_pd_list only when no failure\n  PM: EM: Notify an event when the performance domain changes\n  PM: EM: Implement em_notify_pd_created/updated()\n  PM: EM: Implement em_notify_pd_deleted()\n  PM: EM: Implement em_nl_get_pd_table_doit()\n  PM: EM: Implement em_nl_get_pds_doit()\n  PM: EM: Add an iterator and accessor for the performance domain\n  PM: EM: Add a skeleton code for netlink notification\n  PM: EM: Add em.yaml and autogen files\n  PM: EM: Expose the ID of a performance domain via debugfs\n  PM: EM: Assign a unique ID when creating a performance domain\n\n* pm-opp:\n  rust: opp: simplify callers of `to_c_str_array`\n  OPP: Initialize scope-based pointers inline\n  rust: opp: fix broken rustdoc link\n"
    },
    {
      "commit": "0c2988aaa4d3eda94b738d5a7acae7838d52fe4d",
      "tree": "ff06d71c97e5559554bc21defe34cdc67db8cfc5",
      "parents": [
        "e5efd56fa157d2e7d789949d1d64eccbac18a897"
      ],
      "author": {
        "name": "Dev Jain",
        "email": "dev.jain@arm.com",
        "time": "Wed Nov 12 11:57:16 2025 +0530"
      },
      "committer": {
        "name": "Catalin Marinas",
        "email": "catalin.marinas@arm.com",
        "time": "Fri Nov 28 15:36:40 2025 +0000"
      },
      "message": "arm64/mm: Document why linear map split failure upon vm_reset_perms is not problematic\n\nConsider the following code path:\n\n(1) vmalloc -\u003e (2) set_vm_flush_reset_perms -\u003e (3) set_memory_ro/set_memory_rox\n-\u003e .... (4) use the mapping .... -\u003e (5) vfree -\u003e (6) vm_reset_perms\n-\u003e (7) set_area_direct_map.\nOr, it may happen that we encounter failure at (3) and directly jump to (5).\n\nIn both cases, (7) may fail due to linear map split failure. But, we care\nabout its success *only* for the region which got successfully changed by\n(3). Such a region is guaranteed to be pte-mapped.\n\nThe TLDR is that (7) will surely succeed for the regions we care about.\n\nSigned-off-by: Dev Jain \u003cdev.jain@arm.com\u003e\nReviewed-by: Ryan Roberts \u003cryan.roberts@arm.com\u003e\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n"
    },
    {
      "commit": "e5efd56fa157d2e7d789949d1d64eccbac18a897",
      "tree": "84669e766f3f7a68c2fd7432e513027be0002e7d",
      "parents": [
        "3a8660878839faadb4f1a6dd72c3179c1df56787"
      ],
      "author": {
        "name": "Dev Jain",
        "email": "dev.jain@arm.com",
        "time": "Wed Nov 12 11:57:15 2025 +0530"
      },
      "committer": {
        "name": "Catalin Marinas",
        "email": "catalin.marinas@arm.com",
        "time": "Fri Nov 28 15:36:40 2025 +0000"
      },
      "message": "arm64/pageattr: Propagate return value from __change_memory_common\n\nThe rodata\u003don security measure requires that any code path which does\nvmalloc -\u003e set_memory_ro/set_memory_rox must protect the linear map alias\ntoo. Therefore, if such a call fails, we must abort set_memory_* and caller\nmust take appropriate action; currently we are suppressing the error, and\nthere is a real chance of such an error arising post commit a166563e7ec3\n(\"arm64: mm: support large block mapping when rodata\u003dfull\"). Therefore,\npropagate any error to the caller.\n\nFixes: a166563e7ec3 (\"arm64: mm: support large block mapping when rodata\u003dfull\")\nSigned-off-by: Dev Jain \u003cdev.jain@arm.com\u003e\nReviewed-by: Ryan Roberts \u003cryan.roberts@arm.com\u003e\nSigned-off-by: Catalin Marinas \u003ccatalin.marinas@arm.com\u003e\n"
    },
    {
      "commit": "ff0e4d4c97c94af34cc9cad37b5a5cdbe597a3b0",
      "tree": "e11d8ce81d4ed1e58bc775401fe5642267200548",
      "parents": [
        "094b53ecaa3ef227f9c800f8de90a9b44f1c1bb4"
      ],
      "author": {
        "name": "Jamie Iles",
        "email": "jamie.iles@oss.qualcomm.com",
        "time": "Wed Nov 05 14:42:29 2025 +0000"
      },
      "committer": {
        "name": "Jassi Brar",
        "email": "jassisinghbrar@gmail.com",
        "time": "Fri Nov 28 09:31:53 2025 -0600"
      },
      "message": "mailbox: pcc: don\u0027t zero error register\n\nThe error status mask for a type 3/4 subspace is used for reading the\nerror status, and the bitwise inverse is used for clearing the error\nwith the intent being to preserve any of the non-error bits.  However,\nwe were previously applying the mask to extract the status and then\napplying the inverse to the result which ended up clearing all bits.\n\nInstead, store the inverse mask in the preserve mask and then use that\non the original value read from the error status so that only the error\nis cleared.\n\nFixes: c45ded7e1135 (\"mailbox: pcc: Add support for PCCT extended PCC subspaces(type 3/4)\")\nSigned-off-by: Jamie Iles \u003cjamie.iles@oss.qualcomm.com\u003e\nSigned-off-by: Punit Agrawal \u003cpunit.agrawal@oss.qualcomm.com\u003e\nSigned-off-by: Jassi Brar \u003cjassisinghbrar@gmail.com\u003e\n"
    },
    {
      "commit": "094b53ecaa3ef227f9c800f8de90a9b44f1c1bb4",
      "tree": "d42011d4945d70f78c613f5d14e4f60b7ff92c30",
      "parents": [
        "a195c7ccfb7a21b8118139835e25936ec8722596"
      ],
      "author": {
        "name": "Nicolas Frattaroli",
        "email": "nicolas.frattaroli@collabora.com",
        "time": "Mon Oct 13 10:09:01 2025 +0200"
      },
      "committer": {
        "name": "Jassi Brar",
        "email": "jassisinghbrar@gmail.com",
        "time": "Fri Nov 28 09:31:53 2025 -0600"
      },
      "message": "mailbox: mtk-gpueb: Add missing \u0027static\u0027 to mailbox ops struct\n\nmtk_gpueb_mbox_ops should be declared static. However, due to its const\nnature, this specifier was missed, as it compiled fine without it and\nwith no warning by the compiler.\n\narc-linux-gcc (GCC) 12.5.0 doesn\u0027t seem to like it however, so add the\nstatic to fix that.\n\nReported-by: kernel test robot \u003clkp@intel.com\u003e\nCloses: https://lore.kernel.org/oe-kbuild-all/202510100629.3nGvrhEU-lkp@intel.com/\nFixes: dbca0eabb821 (\"mailbox: add MediaTek GPUEB IPI mailbox\")\nSigned-off-by: Nicolas Frattaroli \u003cnicolas.frattaroli@collabora.com\u003e\nReviewed-by: AngeloGioacchino Del Regno \u003cangelogioacchino.delregno@collabora.com\u003e\nSigned-off-by: Jassi Brar \u003cjassisinghbrar@gmail.com\u003e\n"
    },
    {
      "commit": "a195c7ccfb7a21b8118139835e25936ec8722596",
      "tree": "11c4b59ec59d236f3fdc1340341124ef21a51fc1",
      "parents": [
        "3acf1028f5003731977f750a7070f3321a9cb740"
      ],
      "author": {
        "name": "Jason-JH Lin",
        "email": "jason-jh.lin@mediatek.com",
        "time": "Thu Oct 23 01:16:30 2025 +0800"
      },
      "committer": {
        "name": "Jassi Brar",
        "email": "jassisinghbrar@gmail.com",
        "time": "Fri Nov 28 09:31:53 2025 -0600"
      },
      "message": "mailbox: mtk-cmdq: Refine DMA address handling for the command buffer\n\nGCE can only fetch the command buffer address from a 32-bit register.\nSome SoCs support a 35-bit command buffer address for GCE, which\nrequires a right shift of 3 bits before setting the address into\nthe 32-bit register. A comment has been added to the header of\ncmdq_get_shift_pa() to explain this requirement.\n\nTo prevent the GCE command buffer address from being DMA mapped beyond\nits supported bit range, the DMA bit mask for the device is set during\ninitialization.\n\nAdditionally, to ensure the correct shift is applied when setting or\nreading the register that stores the GCE command buffer address,\nnew APIs, cmdq_convert_gce_addr() and cmdq_revert_gce_addr(), have\nbeen introduced for consistent operations on this register.\n\nThe variable type for the command buffer address has been standardized\nto dma_addr_t to prevent handling issues caused by type mismatches.\n\nFixes: 0858fde496f8 (\"mailbox: cmdq: variablize address shift in platform\")\nSigned-off-by: Jason-JH Lin \u003cjason-jh.lin@mediatek.com\u003e\nReviewed-by: AngeloGioacchino Del Regno \u003cangelogioacchino.delregno@collabora.com\u003e\nSigned-off-by: Jassi Brar \u003cjassisinghbrar@gmail.com\u003e\n"
    },
    {
      "commit": "3acf1028f5003731977f750a7070f3321a9cb740",
      "tree": "ee621fe6c53cfed3ab0b3eeb777f9cfbef79ab85",
      "parents": [
        "060e4e835f9394816584942511f22d771f05100c"
      ],
      "author": {
        "name": "Haotian Zhang",
        "email": "vulab@iscas.ac.cn",
        "time": "Thu Nov 20 10:40:39 2025 +0800"
      },
      "committer": {
        "name": "Jassi Brar",
        "email": "jassisinghbrar@gmail.com",
        "time": "Fri Nov 28 09:31:52 2025 -0600"
      },
      "message": "mailbox: mailbox-test: Fix debugfs_create_dir error checking\n\nThe debugfs_create_dir() function returns ERR_PTR() on error, not NULL.\nThe current null-check fails to catch errors.\n\nUse IS_ERR() to correctly check for errors.\n\nFixes: 8ea4484d0c2b (\"mailbox: Add generic mechanism for testing Mailbox Controllers\")\nSigned-off-by: Haotian Zhang \u003cvulab@iscas.ac.cn\u003e\nSigned-off-by: Jassi Brar \u003cjassisinghbrar@gmail.com\u003e\n"
    },
    {
      "commit": "060e4e835f9394816584942511f22d771f05100c",
      "tree": "43feb4e5a3681f3e59cff582c36680703916108b",
      "parents": [
        "ac3fd01e4c1efce8f2c054cdeb2ddd2fc0fb150d"
      ],
      "author": {
        "name": "Beleswar Padhi",
        "email": "b-padhi@ti.com",
        "time": "Tue Nov 04 01:41:11 2025 +0530"
      },
      "committer": {
        "name": "Jassi Brar",
        "email": "jassisinghbrar@gmail.com",
        "time": "Fri Nov 28 09:31:52 2025 -0600"
      },
      "message": "mailbox: omap-mailbox: Check for pending msgs only when mbox is exclusive\n\nOn TI K3 devices, the mailbox resides in the Always-On power domain\n(LPSC_main_alwayson) and is shared among multiple processors. The\nmailbox is not solely exclusive to Linux.\n\nCurrently, the suspend path checks all FIFO queues for pending messages\nand blocks suspend if any are present. This behavior is unnecessary for\nK3 devices, since some of the FIFOs are used for RTOS\u003c-\u003eRTOS\ncommunication and are independent of Linux.\n\nFor FIFOs used in Linux\u003c-\u003eRTOS communication, any pending message would\ntrigger an interrupt, which naturally prevents suspend from completing.\nHence, there is no need for the mailbox driver to explicitly check for\npending messages on K3 platforms.\n\nIntroduce a device match flag to indicate whether the mailbox instance\nis exclusive to Linux, and skip the pending message check for\nnon-exclusive instances (such as in K3).\n\nFixes: a49f991e740f (\"arm64: dts: ti: k3-am62-verdin: Add missing cfg for TI IPC Firmware\")\nCloses: https://lore.kernel.org/all/sid7gtg5vay5qgicsl6smnzwg5mnneoa35cempt5ddwjvedaio@hzsgcx6oo74l/\nSigned-off-by: Beleswar Padhi \u003cb-padhi@ti.com\u003e\nTested-by: Hiago De Franco \u003chiago.franco@toradex.com\u003e\nReviewed-by: Andrew Davis \u003cafd@ti.com\u003e\nSigned-off-by: Jassi Brar \u003cjassisinghbrar@gmail.com\u003e\n"
    },
    {
      "commit": "bf7ae1773ea79417d430a2d216abdb9857db8ec1",
      "tree": "d0132ffe3d2e66bc8b508b8d1cc3c0a8995f8c50",
      "parents": [
        "1fe25237137c13aeb1a1bd17c0aa6fed6787e444",
        "4bf944f3fcb6c192af1ea73e3d183b6364458b25",
        "748d6ba43afde7e9ac27443233203995cc15d235"
      ],
      "author": {
        "name": "Rafael J. Wysocki",
        "email": "rafael.j.wysocki@intel.com",
        "time": "Fri Nov 28 16:29:41 2025 +0100"
      },
      "committer": {
        "name": "Rafael J. Wysocki",
        "email": "rafael.j.wysocki@intel.com",
        "time": "Fri Nov 28 16:29:41 2025 +0100"
      },
      "message": "Merge branches \u0027pm-cpuidle\u0027 and \u0027pm-powercap\u0027\n\nMerge cpuidle and power capping updates for 6.19-rc1:\n\n - Use residency threshold in polling state override decisions in the\n   menu cpuidle governor (Aboorva Devarajan)\n\n - Add sanity check for exit latency and target residency in the cpufreq\n   core (Rafael Wysocki)\n\n - Use this_cpu_ptr() where possible in the teo governor (Christian\n   Loehle)\n\n - Rework the handling of tick wakeups in the teo cpuidle governor to\n   increase the likelihood of stopping the scheduler tick in the cases\n   when tick wakeups can be counted as non-timer ones (Rafael Wysocki)\n\n - Fix a reverse condition in the teo cpuidle governor and drop a\n   misguided target residency check from it (Rafael Wysocki)\n\n - Clean up muliple minor defects in the teo cpuidle governor (Rafael\n   Wysocki)\n\n - Update header inclusion to make it follow the Include What You Use\n   principle (Andy Shevchenko)\n\n - Enable MSR-based RAPL PMU support in the intel_rapl power capping\n   driver and arrange for using it on the Panther Lake and Wildcat Lake\n   processors (Kuppuswamy Sathyanarayanan)\n\n - Add support for Nova Lake and Wildcat Lake processors to the\n   intel_rapl power capping driver (Kaushlendra Kumar, Srinivas\n   Pandruvada)\n\n* pm-cpuidle:\n  cpuidle: Warn instead of bailing out if target residency check fails\n  cpuidle: Update header inclusion\n  cpuidle: governors: teo: Add missing space to the description\n  cpuidle: governors: teo: Simplify intercepts-based state lookup\n  cpuidle: governors: teo: Fix tick_intercepts handling in teo_update()\n  cpuidle: governors: teo: Rework the handling of tick wakeups\n  cpuidle: governors: teo: Decay metrics below DECAY_SHIFT threshold\n  cpuidle: governors: teo: Use s64 consistently in teo_update()\n  cpuidle: governors: teo: Drop redundant function parameter\n  cpuidle: governors: teo: Drop misguided target residency check\n  cpuidle: teo: Use this_cpu_ptr() where possible\n  cpuidle: Add sanity check for exit latency and target residency\n  cpuidle: menu: Use residency threshold in polling state override decisions\n\n* pm-powercap:\n  powercap: intel_rapl: Enable MSR-based RAPL PMU support\n  powercap: intel_rapl: Prepare read_raw() interface for atomic-context callers\n  powercap: intel_rapl: Add support for Nova Lake processors\n  powercap: intel_rapl: Add support for Wildcat Lake platform\n"
    },
    {
      "commit": "1fe25237137c13aeb1a1bd17c0aa6fed6787e444",
      "tree": "511e08e5e5e7be214e50664d2f64b78150c7b747",
      "parents": [
        "f086594adb34c9db31d764efc6471570ec6a5f1c",
        "ded4feb14d222ff77b94c09629eae2882693247d"
      ],
      "author": {
        "name": "Rafael J. Wysocki",
        "email": "rafael.j.wysocki@intel.com",
        "time": "Fri Nov 28 16:15:38 2025 +0100"
      },
      "committer": {
        "name": "Rafael J. Wysocki",
        "email": "rafael.j.wysocki@intel.com",
        "time": "Fri Nov 28 16:15:38 2025 +0100"
      },
      "message": "Merge branch \u0027pm-cpufreq\u0027\n\nMerge cpufreq updates for 6.19-rc1:\n\n - Add OPP and bandwidth support for Tegra186 (Aaron Kling)\n\n - Optimizations for parameter array handling in the amd-pstate cpufreq\n   driver (Mario Limonciello)\n\n - Fix for mode changes with offline CPUs in the amd-pstate cpufreq\n   driver (Gautham Shenoy)\n\n - Preserve freq_table_sorted across suspend/hibernate in the cpufreq\n   core (Zihuan Zhang)\n\n - Adjust energy model rules for Intel hybrid platforms in the\n   intel_pstate cpufreq driver and improve printing of debug messages\n   in it (Rafael Wysocki)\n\n - Replace deprecated strcpy() in cpufreq_unregister_governor()\n   (Thorsten Blum)\n\n - Fix duplicate hyperlink target errors in the intel_pstate cpufreq\n   driver documentation and use :ref: directive for internal linking in\n   it (Swaraj Gaikwad, Bagas Sanjaya)\n\n - Add Diamond Rapids OOB mode support to the intel_pstate cpufreq\n   driver (Kuppuswamy Sathyanarayanan)\n\n - Use mutex guard for driver locking in the intel_pstate driver and\n   eliminate some code duplication from it (Rafael Wysocki)\n\n - Replace udelay() with usleep_range() in ACPI cpufreq (Kaushlendra\n   Kumar)\n\n - Minor improvements to various cpufreq drivers (Christian Marangi, Hal\n   Feng, Jie Zhan, Marco Crivellari, Miaoqian Lin, and Shuhao Fu)\n\n* pm-cpufreq: (27 commits)\n  cpufreq: qcom-nvmem: fix compilation warning for qcom_cpufreq_ipq806x_match_list\n  cpufreq: ACPI: Replace udelay() with usleep_range()\n  cpufreq: intel_pstate: Eliminate some code duplication\n  cpufreq: intel_pstate: Use mutex guard for driver locking\n  cpufreq/amd-pstate: Call cppc_set_auto_sel() only for online CPUs\n  cpufreq/amd-pstate: Add static asserts for EPP indices\n  cpufreq/amd-pstate: Fix some whitespace issues\n  cpufreq/amd-pstate: Adjust return values in amd_pstate_update_status()\n  cpufreq/amd-pstate: Make amd_pstate_get_mode_string() never return NULL\n  cpufreq/amd-pstate: Drop NULL value from amd_pstate_mode_string\n  cpufreq/amd-pstate: Use sysfs_match_string() for epp\n  cpufreq: tegra194: add WQ_PERCPU to alloc_workqueue users\n  cpufreq: qcom-nvmem: add compatible fallback for ipq806x for no SMEM\n  Documentation: intel-pstate: Use :ref: directive for internal linking\n  cpufreq: intel_pstate: Add Diamond Rapids OOB mode support\n  Documentation: intel_pstate: fix duplicate hyperlink target errors\n  cpufreq: CPPC: Don\u0027t warn if FIE init fails to read counters\n  cpufreq: nforce2: fix reference count leak in nforce2\n  cpufreq: tegra186: add OPP support and set bandwidth\n  cpufreq: dt-platdev: Add JH7110S SOC to the allowlist\n  ...\n"
    },
    {
      "commit": "f086594adb34c9db31d764efc6471570ec6a5f1c",
      "tree": "e53fd9d7e33137a5b2b9923cdc036403a73f0f3a",
      "parents": [
        "60d69a7ed1e5ffa94f788e35b2ca86c6a750f2f2",
        "c03aef8833597f184cc3439d2d336596f63bd709"
      ],
      "author": {
        "name": "Rafael J. Wysocki",
        "email": "rafael.j.wysocki@intel.com",
        "time": "Fri Nov 28 16:01:13 2025 +0100"
      },
      "committer": {
        "name": "Rafael J. Wysocki",
        "email": "rafael.j.wysocki@intel.com",
        "time": "Fri Nov 28 16:01:13 2025 +0100"
      },
      "message": "Merge branch \u0027pm-sleep\u0027\n\nMerge updates related to system suspend and hibernation for 6.19-rc1:\n\n - Replace snprintf() with scnprintf() in show_trace_dev_match()\n   (Kaushlendra Kumar)\n\n - Fix memory allocation error handling in pm_vt_switch_required()\n   (Malaya Kumar Rout)\n\n - Introduce CALL_PM_OP() macro and use it to simplify code in\n   generic PM operations (Kaushlendra Kumar)\n\n - Add module param to backtrace all CPUs in the device power management\n   watchdog (Sergey Senozhatsky)\n\n - Rework message printing in swsusp_save() (Rafael Wysocki)\n\n - Make it possible to change the number of hibernation compression\n   threads (Xueqin Luo)\n\n - Clarify that only cgroup1 freezer uses PM freezer (Tejun Heo)\n\n - Add document on debugging shutdown hangs to PM documentation and\n   correct a mistaken configuration option in it (Mario Limonciello)\n\n - Shut down wakeup source timer before removing the wakeup source from\n   the list (Kaushlendra Kumar, Rafael Wysocki)\n\n - Introduce new PMSG_POWEROFF event for system shutdown handling with\n   the help of PM device callbacks (Mario Limonciello)\n\n - Make pm_test delay interruptible by wakeup events (Riwen Lu)\n\n - Clean up kernel-doc comment style usage in the core hibernation\n   code and remove unuseful comments from it (Sunday Adelodun, Rafael\n   Wysocki)\n\n - Add support for handling wakeup events and aborting the suspend\n   process while it is syncing file systems (Samuel Wu, Rafael Wysocki)\n\n* pm-sleep: (21 commits)\n  PM: hibernate: Extra cleanup of comments in swap handling code\n  PM: sleep: Call pm_sleep_fs_sync() instead of ksys_sync_helper()\n  PM: sleep: Add support for wakeup during filesystem sync\n  PM: hibernate: Clean up kernel-doc comment style usage\n  PM: suspend: Make pm_test delay interruptible by wakeup events\n  usb: sl811-hcd: Add PM_EVENT_POWEROFF into suspend callbacks\n  scsi: Add PM_EVENT_POWEROFF into suspend callbacks\n  PM: Introduce new PMSG_POWEROFF event\n  PM: wakeup: Update after recent wakeup source removal ordering change\n  PM: wakeup: Delete timer before removing wakeup source from list\n  Documentation: power: Correct a mistaken configuration option\n  Documentation: power: Add document on debugging shutdown hangs\n  freezer: Clarify that only cgroup1 freezer uses PM freezer\n  PM: hibernate: add sysfs interface for hibernate_compression_threads\n  PM: hibernate: make compression threads configurable\n  PM: hibernate: dynamically allocate crc-\u003eunc_len/unc for configurable threads\n  PM: hibernate: Rework message printing in swsusp_save()\n  PM: dpm_watchdog: add module param to backtrace all CPUs\n  PM: sleep: Introduce CALL_PM_OP() macro to simplify code\n  PM: console: Fix memory allocation error handling in pm_vt_switch_required()\n  ...\n"
    },
    {
      "commit": "60d69a7ed1e5ffa94f788e35b2ca86c6a750f2f2",
      "tree": "e30951094f35f6f7b16fd0be253ea253ab465845",
      "parents": [
        "5e8b7b58b205da3c6c41f1e9bc66db999ed95595",
        "c9ff3637386c6eb72eac55a8b4c9a4972215dbcb",
        "07f42f8290e927a38ee4248505fc39ed0518519e"
      ],
      "author": {
        "name": "Rafael J. Wysocki",
        "email": "rafael.j.wysocki@intel.com",
        "time": "Fri Nov 28 15:56:09 2025 +0100"
      },
      "committer": {
        "name": "Rafael J. Wysocki",
        "email": "rafael.j.wysocki@intel.com",
        "time": "Fri Nov 28 15:56:09 2025 +0100"
      },
      "message": "Merge branches \u0027pm-core\u0027 and \u0027pm-runtime\u0027\n\nMerge a core power management update and runtime PM framework updates\nfor 6.19-rc1:\n\n - Add WQ_UNBOUND to pm_wq workqueue (Marco Crivellari)\n\n - Add runtime PM wrapper macros for ACQUIRE()/ACQUIRE_ERR() and use\n   them in the PCI core and the ACPI TAD driver (Rafael Wysocki)\n\n - Improve runtime PM in the ACPI TAD driver (Rafael Wysocki)\n\n - Update pm_runtime_allow/forbid() documentation (Rafael Wysocki)\n\n - Fix typos in runtime.c comments (Malaya Kumar Rout)\n\n* pm-core:\n  PM: WQ_UNBOUND added to pm_wq workqueue\n\n* pm-runtime:\n  PCI/sysfs: Use PM_RUNTIME_ACQUIRE()/PM_RUNTIME_ACQUIRE_ERR()\n  ACPI: TAD: Use PM_RUNTIME_ACQUIRE()/PM_RUNTIME_ACQUIRE_ERR()\n  PM: runtime: Wrapper macros for ACQUIRE()/ACQUIRE_ERR()\n  PM: runtime: fix typos in runtime.c comments\n  ACPI: TAD: Improve runtime PM using guard macros\n  ACPI: TAD: Rearrange runtime PM operations in acpi_tad_remove()\n  PM: runtime: docs: Update pm_runtime_allow/forbid() documentation\n"
    }
  ],
  "next": "af47d98064e900f279c98bb3bb57445a1e7a8f6b"
}