)]}' { "log": [ { "commit": "a2cc6ff5ec8f91bc463fd3b0c26b61166a07eb11", "tree": "bd4a570c5717634d670151c2035848f14e86ba44", "parents": [ "5014bebee0cffda14fafae5a2534d08120b7b9e8", "ca2c736554c105897d67a015a97973af315e1c32" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 22:41:04 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 22:41:04 2025 -0700" }, "message": "Merge tag \u0027firewire-updates-6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394\n\nPull firewire update from Takashi Sakamoto:\n \"A single commit to use the common helper function for on-stack\n trailing array to enqueue any isochronous packet by the requests\n from userspace applications\"\n\n* tag \u0027firewire-updates-6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394:\n firewire: core: avoid -Wflex-array-member-not-at-end warning\n" }, { "commit": "5014bebee0cffda14fafae5a2534d08120b7b9e8", "tree": "cdeb23d10c019240dd207d86ed0b9a3024972838", "parents": [ "447d2d272e4e0c7cd9dfc6aeeadad9d70b3fb1ef", "9c565428788fb9b49066f94ab7b10efc686a0a4c" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 21:27:59 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 21:27:59 2025 -0700" }, "message": "Merge tag \u0027for-6.15/dm-changes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm\n\nPull device mapper updates from Mikulas Patocka:\n\n - dm-crypt: switch to using the crc32 library\n\n - dm-verity, dm-integrity, dm-crypt: documentation improvement\n\n - dm-vdo fixes\n\n - dm-stripe: enable inline crypto passthrough\n\n - dm-integrity: set ti-\u003eerror on memory allocation failure\n\n - dm-bufio: remove unused return value\n\n - dm-verity: do forward error correction on metadata I/O errors\n\n - dm: fix unconditional IO throttle caused by REQ_PREFLUSH\n\n - dm cache: prevent BUG_ON by blocking retries on failed device resumes\n\n - dm cache: support shrinking the origin device\n\n - dm: restrict dm device size to 2^63-512 bytes\n\n - dm-delay: support zoned devices\n\n - dm-verity: support block number limits for different ioprio classes\n\n - dm-integrity: fix non-constant-time tag verification (security bug)\n\n - dm-verity, dm-ebs: fix prefetch-vs-suspend race\n\n* tag \u0027for-6.15/dm-changes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm: (27 commits)\n dm-ebs: fix prefetch-vs-suspend race\n dm-verity: fix prefetch-vs-suspend race\n dm-integrity: fix non-constant-time tag verification\n dm-verity: support block number limits for different ioprio classes\n dm-delay: support zoned devices\n dm: restrict dm device size to 2^63-512 bytes\n dm cache: support shrinking the origin device\n dm cache: prevent BUG_ON by blocking retries on failed device resumes\n dm vdo indexer: reorder uds_request to reduce padding\n dm: fix unconditional IO throttle caused by REQ_PREFLUSH\n dm vdo: rework processing of loaded refcount byte arrays\n dm vdo: remove remaining ring references\n dm-verity: do forward error correction on metadata I/O errors\n dm-bufio: remove unused return value\n dm-integrity: set ti-\u003eerror on memory allocation failure\n dm: Enable inline crypto passthrough for striped target\n dm vdo slab-depot: read refcount blocks in large chunks at load time\n dm vdo vio-pool: allow variable-sized metadata vios\n dm vdo vio-pool: support pools with multiple data blocks per vio\n dm vdo vio-pool: add a pool pointer to pooled_vio\n ...\n" }, { "commit": "447d2d272e4e0c7cd9dfc6aeeadad9d70b3fb1ef", "tree": "d71bd77244b6e2b52364a02253bee1abc1642a85", "parents": [ "01ecadbe09b6c685de413ada8ba6688e9467c4b3", "ef1d3455bbc1922f94a91ed58d3d7db440652959" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 20:27:18 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 20:27:18 2025 -0700" }, "message": "Merge tag \u0027libnvdimm-for-6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm\n\nPull libnvdimm updates from Ira Weiny:\n \"Most of the code changes are to remove dead code.\n\n The bug fixes are minor, Syzkaller and one for broken devices which\n are unlikely to be in the field. So no need to backport them.\n\n - two patches to remove dead code: nd_attach_ndns() and\n nd_region_conflict() have not been used since 2017 and 2019\n respectively\n\n - Fix divide-by-0 if device returns a broken LSA value\n\n - Fix Syzkaller reported bug\"\n\n* tag \u0027libnvdimm-for-6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:\n libnvdimm/labels: Fix divide error in nd_label_data_init()\n libnvdimm: Remove unused nd_attach_ndns\n libnvdimm: Remove unused nd_region_conflict\n acpi: nfit: fix narrowing conversion in acpi_nfit_ctl\n" }, { "commit": "01ecadbe09b6c685de413ada8ba6688e9467c4b3", "tree": "4281c55f68437c789d1f148c22e9b03f8e601526", "parents": [ "a1b5bd45d4ee58af4f56e49497b8c3db96d8f8a3", "aae0594a7053c60b82621136257c8b648c67b512" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 20:04:43 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 20:04:43 2025 -0700" }, "message": "Merge tag \u0027cxl-for-6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/cxl/cxl\n\nPull Compute Express Link (CXL) updates from Dave Jiang:\n\n - Add support for Global Persistent Flush (GPF)\n\n - Cleanup of DPA partition metadata handling:\n - Remove the CXL_DECODER_MIXED enum that\u0027s not needed anymore\n - Introduce helpers to access resource and perf meta data\n - Introduce \u0027struct cxl_dpa_partition\u0027 and \u0027struct cxl_range_info\u0027\n - Make cxl_dpa_alloc() DPA partition number agnostic\n - Remove cxl_decoder_mode\n - Cleanup partition size and perf helpers\n\n - Remove unused CXL partition values\n\n - Add logging support for CXL CPER endpoint and port protocol errors:\n - Prefix protocol error struct and function names with cxl_\n - Move protocol error definitions and structures to a common location\n - Remove drivers/firmware/efi/cper_cxl.h to include/linux/cper.h\n - Add support in GHES to process CXL CPER protocol errors\n - Process CXL CPER protocol errors\n - Add trace logging for CXL PCIe port RAS errors\n\n - Remove redundant gp_port init\n\n - Add validation of cxl device serial number\n\n - CXL ABI documentation updates/fixups\n\n - A series that uses guard() to clean up open coded mutex lockings and\n remove gotos for error handling.\n\n - Some followup patches to support dirty shutdown accounting:\n - Add helper to retrieve DVSEC offset for dirty shutdown registers\n - Rename cxl_get_dirty_shutdown() to cxl_arm_dirty_shutdown()\n - Add support for dirty shutdown count via sysfs\n - cxl_test support for dirty shutdown\n\n - A series to support CXL mailbox Features commands.\n\n Mostly in preparation for CXL EDAC code to utilize the Features\n commands. It\u0027s also in preparation for CXL fwctl support to utilize\n the CXL Features. The commands include \"Get Supported Features\", \"Get\n Feature\", and \"Set Feature\".\n\n - A series to support extended linear cache support described by the\n ACPI HMAT table.\n\n The addition helps enumerate the cache and also provides additional\n RAS reporting support for configuration with extended linear cache.\n (and related fixes for the series).\n\n - An update to cxl_test to support a 3-way capable CFMWS\n\n - A documentation fix to remove unused \"mixed mode\"\n\n* tag \u0027cxl-for-6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/cxl/cxl: (39 commits)\n cxl/region: Fix the first aliased address miscalculation\n cxl/region: Quiet some dev_warn()s in extended linear cache setup\n cxl/Documentation: Remove \u0027mixed\u0027 from sysfs mode doc\n cxl: Fix warning from emitting resource_size_t as long long int on 32bit systems\n cxl/test: Define a CFMWS capable of a 3 way HB interleave\n cxl/mem: Do not return error if CONFIG_CXL_MCE unset\n tools/testing/cxl: Set Shutdown State support\n cxl/pmem: Export dirty shutdown count via sysfs\n cxl/pmem: Rename cxl_dirty_shutdown_state()\n cxl/pci: Introduce cxl_gpf_get_dvsec()\n cxl/pci: Support Global Persistent Flush (GPF)\n cxl: Document missing sysfs files\n cxl: Plug typos in ABI doc\n cxl/pmem: debug invalid serial number data\n cxl/cdat: Remove redundant gp_port initialization\n cxl/memdev: Remove unused partition values\n cxl/region: Drop goto pattern of construct_region()\n cxl/region: Drop goto pattern in cxl_dax_region_alloc()\n cxl/core: Use guard() to drop goto pattern of cxl_dpa_alloc()\n cxl/core: Use guard() to drop the goto pattern of cxl_dpa_free()\n ...\n" }, { "commit": "a1b5bd45d4ee58af4f56e49497b8c3db96d8f8a3", "tree": "0dd2b82215f3444020353af3625b5d5efd4b8581", "parents": [ "ddd0172f182e3e869a3a960e433578aeedcb37c9", "ed492c95f13a7a27f1c0426b33c920ad343232c4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 18:23:31 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 18:23:31 2025 -0700" }, "message": "Merge tag \u0027usb-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb\n\nPull USB / Thunderbolt updates from Greg KH:\n \"Here is the big set of USB and Thunderbolt driver updates for\n 6.15-rc1. Included in here are:\n\n - Thunderbolt driver and core api updates for new hardware and\n features\n\n - usb-storage const array cleanups\n\n - typec driver updates\n\n - dwc3 driver updates\n\n - xhci driver updates and bugfixes\n\n - small USB documentation updates\n\n - usb cdns3 driver updates\n\n - usb gadget driver updates\n\n - other small driver updates and fixes\n\n All of these have been in linux-next for a while with no reported\n issues\"\n\n* tag \u0027usb-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (92 commits)\n thunderbolt: Do not add non-active NVM if NVM upgrade is disabled for retimer\n thunderbolt: Scan retimers after device router has been enumerated\n usb: host: cdns3: forward lost power information to xhci\n usb: host: xhci-plat: allow upper layers to signal power loss\n usb: xhci: change xhci_resume() parameters to explicit the desired info\n usb: cdns3-ti: run HW init at resume() if HW was reset\n usb: cdns3-ti: move reg writes to separate function\n usb: cdns3: call cdns_power_is_lost() only once in cdns_resume()\n usb: cdns3: rename hibernated argument of role-\u003eresume() to lost_power\n usb: xhci: tegra: rename `runtime` boolean to `is_auto_runtime`\n usb: host: xhci-plat: mvebu: use -\u003equirks instead of -\u003einit_quirk() func\n usb: dwc3: Don\u0027t use %pK through printk\n usb: core: Don\u0027t use %pK through printk\n usb: gadget: aspeed: Add NULL pointer check in ast_vhub_init_dev()\n dt-bindings: usb: qcom,dwc3: Synchronize minItems for interrupts and -names\n usb: common: usb-conn-gpio: switch psy_cfg from of_node to fwnode\n usb: xhci: Avoid Stop Endpoint retry loop if the endpoint seems Running\n usb: xhci: Don\u0027t change the status of stalled TDs on failed Stop EP\n xhci: Avoid queuing redundant Stop Endpoint command for stalled endpoint\n xhci: Handle spurious events on Etron host isoc enpoints\n ...\n" }, { "commit": "ddd0172f182e3e869a3a960e433578aeedcb37c9", "tree": "eb2b5c3ea0d6c4de0ff45eb61bfdf9fac622b0e6", "parents": [ "25757984d77da731922bed5001431673b6daf5ac", "9f8fe348ac9544f6855f82565e754bf085d81f88" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 18:17:33 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 18:17:33 2025 -0700" }, "message": "Merge tag \u0027tty-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty\n\nPull tty/serial driver updates from Greg KH:\n \"Here is the big set of serial and tty driver updates for 6.15-rc1.\n Include in here are the following:\n\n - more great tty layer cleanups from Jiri. Someday this will be done,\n but that\u0027s not going to be any year soon...\n\n - kdb debug driver reverts to fix a reported issue\n\n - lots of .dts binding updates for different devices with serial\n devices\n\n - lots of tiny updates and tweaks and a few bugfixes for different\n serial drivers.\n\n All of these have been in linux-next for a while with no reported\n issues\"\n\n* tag \u0027tty-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty: (79 commits)\n tty: serial: fsl_lpuart: Fix unused variable \u0027sport\u0027 build warning\n serial: stm32: do not deassert RS485 RTS GPIO prematurely\n serial: 8250: add driver for NI UARTs\n dt-bindings: serial: snps-dw-apb-uart: document RZ/N1 binding without DMA\n serial: icom: fix code format problems\n serial: sh-sci: Save and restore more registers\n tty: serial: pl011: remove incorrect of_match_ptr annotation\n dt-bindings: serial: snps-dw-apb-uart: Add support for rk3562\n tty: serial: lpuart: only disable CTS instead of overwriting the whole UARTMODIR register\n tty: caif: removed unused function debugfs_tx()\n serial: 8250_dma: terminate correct DMA in tx_dma_flush()\n tty: serial: fsl_lpuart: rename register variables more specifically\n tty: serial: fsl_lpuart: use port struct directly to simply code\n tty: serial: fsl_lpuart: Use u32 and u8 for register variables\n tty: serial: fsl_lpuart: disable transmitter before changing RS485 related registers\n tty: serial: 8250: Add Brainboxes XC devices\n dt-bindings: serial: fsl-lpuart: support i.MX94\n tty: serial: 8250: Add some more device IDs\n dt-bindings: serial: samsung: add exynos7870-uart compatible\n serial: 8250_dw: Comment possible corner cases in serial_out() implementation\n ...\n" }, { "commit": "25757984d77da731922bed5001431673b6daf5ac", "tree": "4a3841f68d039617829e0f7b845d25f8e5a5e65f", "parents": [ "a0935e4ca3e1f2d5176a5b60bf6092cd7eba9914", "3b23d31e569ca19b7bf1059760b879f61cd71ded" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 18:09:17 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 18:09:17 2025 -0700" }, "message": "Merge tag \u0027staging-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging\n\nPull staging driver updates from Greg KH:\n \"Here is the big set of staging driver cleanups and updates for\n 6.15-rc1.\n\n As expected, with the introduction of the gpib drivers, loads of\n cleanups and fixes showed up, with the huge majority of changes being\n for that chunk of drivers. This is good and shows that the community\n can fix up things in public when asked to. Also included in here are:\n\n - small sm750fb cleanups\n\n - tiny rtl8723bs cleanups\n\n - more vchiq_arm cleanups and changes, hopefully this will get out of\n staging soon\n\n All of these have been in linux-next for almost 2 weeks now with no\n reported issues\"\n\n* tag \u0027staging-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging: (76 commits)\n staging: rtl8723bs: fixed a unnecessary parentheses coding style issue\n staging: vchiq_arm: Improve initial VCHIQ connect\n staging: vchiq_arm: Create keep-alive thread during probe\n staging: vchiq_arm: Stop kthreads if vchiq cdev register fails\n staging: vchiq_arm: Fix possible NPR of keep-alive thread\n staging: vchiq_arm: Register debugfs after cdev\n staging: vchiq_arm: Don\u0027t use %pK through printk\n staging: rtl8723bs: select CONFIG_CRYPTO_LIB_AES\n staging: rtl8723bs: Remove some unused functions, macros, and structs\n staging: gpib: change return type of t1_delay function to report errors\n staging: gpib: remove commented-out lines\n staging: gpib: fix kernel-doc section for usb_gpib_line_status() function\n staging: gpib: fix kernel-doc section for function usb_gpib_interface_clear()\n staging: gpib: fix kernel-doc section for write_loop() function\n staging: gpib: Removing typedef for gpib_board\n staging: gpib: struct typing for gpib_gboard_t\n staging: gpib: tnt4882: struct gpib_board\n staging: gpib: tms9914: struct gpib_board\n staging: gpib: pc2: struct gpib_board\n staging: gpib: ni_usb_gpib: struct gpib_board\n ...\n" }, { "commit": "a0935e4ca3e1f2d5176a5b60bf6092cd7eba9914", "tree": "d1253980d2ce12a72b0f1ccdb31cc53f715f9274", "parents": [ "94d471a4f42862bf53dc3776bde4b4c9175acbd4", "2dc25093218f5d42391549de6fe45e1aa9325676" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 18:03:34 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 18:03:34 2025 -0700" }, "message": "Merge tag \u0027char-misc-6.15-rc1-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc\n\nPull char/misc fixes from Greg KH:\n \"Here are two counter driver fixes that I realized I never sent to you\n for 6.14-final.\n\n They have been in my for weeks, as well as linux-next, my fault for\n not sending them earlier. They are:\n\n - bugfix for stm32-lptimer-cnt counter driver\n\n - bugfix for microchip-tcb-capture counter driver\n\n Again, these have been in linux-next for weeks with no reported\n issues\"\n\n* tag \u0027char-misc-6.15-rc1-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:\n counter: microchip-tcb-capture: Fix undefined counter channel state on probe\n counter: stm32-lptimer-cnt: fix error handling when enabling\n" }, { "commit": "94d471a4f42862bf53dc3776bde4b4c9175acbd4", "tree": "18efca8676d8aaeb6cd80961e5184144a187fb5c", "parents": [ "5e17b5c71729d8ce936c83a579ed45f65efcb456", "8e5419d6542fdf2dca9a0acdef2b8255f0e4ba69" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 17:06:31 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 17:06:31 2025 -0700" }, "message": "Merge tag \u0027nfs-for-6.15-1\u0027 of git://git.linux-nfs.org/projects/trondmy/linux-nfs\n\nPull NFS client updates from Trond Myklebust:\n \"Bugfixes:\n\n - Three fixes for looping in the NFSv4 state manager delegation code\n\n - Fix for the NFSv4 state XDR code (Neil Brown)\n\n - Fix a leaked reference in nfs_lock_and_join_requests()\n\n - Fix a use-after-free in the delegation return code\n\n Features:\n\n - Implement the NFSv4.2 copy offload OFFLOAD_STATUS operation to\n allow monitoring of an in-progress copy\n\n - Add a mount option to force NFSv3/NFSv4 to use READDIRPLUS in a\n getdents() call\n\n - SUNRPC now allows some basic management of an existing RPC client\u0027s\n connections using sysfs\n\n - Improvements to the automated teardown of a NFS client when the\n container it was initiated from gets killed\n\n - Improvements to prevent tasks from getting stuck in a killable wait\n state after calling exit_signals()\"\n\n* tag \u0027nfs-for-6.15-1\u0027 of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (29 commits)\n nfs: Add missing release on error in nfs_lock_and_join_requests()\n NFSv4: Check for delegation validity in nfs_start_delegation_return_locked()\n NFS: Don\u0027t allow waiting for exiting tasks\n SUNRPC: Don\u0027t allow waiting for exiting tasks\n NFSv4: Treat ENETUNREACH errors as fatal for state recovery\n NFSv4: clp-\u003ecl_cons_state \u003c 0 signifies an invalid nfs_client\n NFSv4: Further cleanups to shutdown loops\n NFS: Shut down the nfs_client only after all the superblocks\n SUNRPC: rpc_clnt_set_transport() must not change the autobind setting\n SUNRPC: rpcbind should never reset the port to the value \u00270\u0027\n pNFS/flexfiles: Report ENETDOWN as a connection error\n pNFS/flexfiles: Treat ENETUNREACH errors as fatal in containers\n NFS: Treat ENETUNREACH errors as fatal in containers\n NFS: Add a mount option to make ENETUNREACH errors fatal\n sunrpc: Add a sysfs file for one-step xprt deletion\n sunrpc: Add a sysfs file for adding a new xprt\n sunrpc: Add a sysfs files for rpc_clnt information\n sunrpc: Add a sysfs attr for xprtsec\n NFS: Add implid to sysfs\n NFS: Extend rdirplus mount option with \"force|none\"\n ...\n" }, { "commit": "5e17b5c71729d8ce936c83a579ed45f65efcb456", "tree": "7335bdd764e0f0f3a093291f3f6ed3985812279d", "parents": [ "0cc5543fad527878b311dcc361ec505f613da4e2", "2d066800a4276340a97acc75c148892eb6f8781a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 16:36:59 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 16:36:59 2025 -0700" }, "message": "Merge tag \u0027fuse-update-6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse\n\nPull fuse updates from Miklos Szeredi:\n\n - Allow connection to server to time out (Joanne Koong)\n\n - If server doesn\u0027t support creating a hard link, return EPERM rather\n than ENOSYS (Matt Johnston)\n\n - Allow file names longer than 1024 chars (Bernd Schubert)\n\n - Fix a possible race if request on io_uring queue is interrupted\n (Bernd Schubert)\n\n - Misc fixes and cleanups\n\n* tag \u0027fuse-update-6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:\n fuse: remove unneeded atomic set in uring creation\n fuse: fix uring race condition for null dereference of fc\n fuse: Increase FUSE_NAME_MAX to PATH_MAX\n fuse: Allocate only namelen buf memory in fuse_notify_\n fuse: add default_request_timeout and max_request_timeout sysctls\n fuse: add kernel-enforced timeout option for requests\n fuse: optmize missing FUSE_LINK support\n fuse: Return EPERM rather than ENOSYS from link()\n fuse: removed unused function fuse_uring_create() from header\n fuse: {io-uring} Fix a possible req cancellation race\n" }, { "commit": "0cc5543fad527878b311dcc361ec505f613da4e2", "tree": "d1e72171de7899950f343f08ae7cca3437e69e94", "parents": [ "4b06c990c106d0341357175b229277a90da6583e", "8b12017c1b9582db8c5833cf08d610e8f810f4b3" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 16:30:02 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 16:30:02 2025 -0700" }, "message": "Merge tag \u0027ntfs3_for_6.15\u0027 of https://github.com/Paragon-Software-Group/linux-ntfs3\n\nPull ntfs3 updates from Konstantin Komarov:\n\n - Fix integer overflows on 32-bit systems and in hdr_first_de()\n\n - Fix \u0027proc_info_root\u0027 leak on NTFS initialization failure\n\n - Remove unused functions ni_load_attr, ntfs_sb_read, ntfs_flush_inodes\n\n - update inode-\u003ei_mapping-\u003ea_ops on compression state\n\n - ensure atomicity of write operations\n\n - refactor ntfs_{create/remove}_{procdir,proc_root}()\n\n* tag \u0027ntfs3_for_6.15\u0027 of https://github.com/Paragon-Software-Group/linux-ntfs3:\n fs/ntfs3: Remove unused ntfs_flush_inodes\n fs/ntfs3: Remove unused ntfs_sb_read\n fs/ntfs3: Remove unused ni_load_attr\n fs/ntfs3: Prevent integer overflow in hdr_first_de()\n fs/ntfs3: Fix a couple integer overflows on 32bit systems\n fs/ntfs3: Update inode-\u003ei_mapping-\u003ea_ops on compression state\n fs/ntfs3: Fix WARNING in ntfs_extend_initialized_size\n fs/ntfs3: Fix \u0027proc_info_root\u0027 leak when init ntfs failed\n fs/ntfs3: Factor out ntfs_{create/remove}_proc_root()\n fs/ntfs3: Factor out ntfs_{create/remove}_procdir()\n fs/ntfs3: Keep write operations atomic\n" }, { "commit": "4b06c990c106d0341357175b229277a90da6583e", "tree": "fc84ca05e93b5c59c71d8e27030260b5f9fae68e", "parents": [ "e0a02923c2cdfc4b8f1ef19181bf40f6529812d3", "923936efeb74b3f42e5ad283a0b9110bda102601" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 16:05:21 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 16:05:21 2025 -0700" }, "message": "Merge tag \u0027vfs-6.15-rc1.fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs\n\nPull vfs fixes from Christian Brauner:\n\n - Add a new maintainer for configfs\n\n - Fix exportfs module description\n\n - Place flexible array memeber at the end of an internal struct in the\n mount code\n\n - Add new maintainer for netfslib as Jeff Layton is stepping down as\n current co-maintainer\n\n - Fix error handling in cachefiles_get_directory()\n\n - Cleanup do_notify_pidfd()\n\n - Fix syscall number definitions in pidfd selftests\n\n - Fix racy usage of fs_struct-\u003ein exec during multi-threaded exec\n\n - Ensure correct exit code is reported when pidfs_exit() is called from\n release_task() for a delayed thread-group leader exit\n\n - Fix conflicting iomap flag definitions\n\n* tag \u0027vfs-6.15-rc1.fixes\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs:\n iomap: Fix conflicting values of iomap flags\n fs: namespace: Avoid -Wflex-array-member-not-at-end warning\n MAINTAINERS: configfs: add Andreas Hindborg as maintainer\n exportfs: add module description\n exit: fix the usage of delay_group_leader-\u003eexit_code in do_notify_parent() and pidfs_exit()\n netfs: add Paulo as maintainer and remove myself as Reviewer\n cachefiles: Fix oops in vfs_mkdir from cachefiles_get_directory\n exec: fix the racy usage of fs_struct-\u003ein_exec\n selftests/pidfd: fixes syscall number defines\n pidfs: cleanup the usage of do_notify_pidfd()\n" }, { "commit": "e0a02923c2cdfc4b8f1ef19181bf40f6529812d3", "tree": "1a5e5cc72ab5801db4a40d16106d97e99913ce00", "parents": [ "e69e6e468a5c273132e9f87094fea1c62e347899", "9e4e249018d208678888bdf22f6b652728106528" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 15:22:22 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 15:22:22 2025 -0700" }, "message": "Merge tag \u0027pm-6.15-rc1-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull power management fix from Rafael Wysocki:\n \"Prevent cpufreq_update_limits() from crashing the kernel due to a NULL\n pointer dereference when it is called before registering a cpufreq\n driver, for instance as a result of a notification triggered by the\n platform firmware (Rafael Wysocki)\"\n\n* tag \u0027pm-6.15-rc1-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n cpufreq: Reference count policy in cpufreq_update_limits()\n" }, { "commit": "e69e6e468a5c273132e9f87094fea1c62e347899", "tree": "1f13bfc6ba32097af600a196ae7213f7e842b63d", "parents": [ "0a87d6bb6fd274cde3bf217a821153714374198f", "5bf46fe2b84cda662062f7aca73e15602c76a844" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 15:16:30 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 15:16:30 2025 -0700" }, "message": "Merge tag \u0027acpi-6.15-rc1-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull more ACPI updates from Rafael Wysocki:\n \"These are fixes and cleanups on top of the previous ACPI material for\n 6.15-rc1 merged recently:\n\n - Extend the Lenovo Yoga Tab 3 ACPI quirk to skip GPIO event-handlers\n along with ACPI AC and battery which makes it work with Linux when\n started in the Windows mode (Hans de Goede)\n\n - Prevent the ACPI processor idle driver from being used on systems\n without _CST and with invalid C2/C3 in FADT in order to restore its\n previous (and expected) behavior that has been altered\n inadvertently by a recent code change (Giovanni Gherdovich)\n\n - Skip ACPI IRQ override on ASUS Vivobook 14 X1404VAP to make the\n internal keyboard work on it (Paul Menzel)\n\n - Make the ACPI backlight driver handle fetching EDID passed as\n ACPI_TYPE_PACKAGE which is not specification-compliant, but has\n been encountered in the field (Gergo Koteles)\n\n - Simplify the aggregation of choices in the ACPI platform-profile\n driver which has been unlocked by recent modifications of that\n driver (Kurt Borja)\n\n - Use str_enabled_disabled() instead of hardcoded strings in the ACPI\n code related to NUMA (Thorsten Blum)\n\n - Add Intel OC Watchdog device IDs to non-PNP device list to prevent\n PNP from claiming the devices that carry these IDs in which case\n non-PNP drivers cannot bind to them (Diogo Ivo)\"\n\n* tag \u0027acpi-6.15-rc1-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n ACPI: video: Handle fetching EDID as ACPI_TYPE_PACKAGE\n ACPI: processor: idle: Return an error if both P_LVL{2,3} idle states are invalid\n ACPI: platform_profile: Optimize _aggregate_choices()\n ACPI: x86: Extend Lenovo Yoga Tab 3 quirk with skip GPIO event-handlers\n ACPI: resource: Skip IRQ override on ASUS Vivobook 14 X1404VAP\n ACPI: PNP: Add Intel OC Watchdog IDs to non-PNP device list\n ACPI: NUMA: Use str_enabled_disabled() helper function\n" }, { "commit": "0a87d6bb6fd274cde3bf217a821153714374198f", "tree": "e39aa8be37fbc72bc601ac2d6724dae086d26a04", "parents": [ "8a6b94032e0077ae15b21e7b42a9599bdc18ea22", "892c4e465c360d07f529bc3668fde7cbd4ea6b32" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 12:35:49 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 12:35:49 2025 -0700" }, "message": "Merge tag \u0027powerpc-6.15-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux\n\nPull more powerpc updates from Michael Ellerman:\n\n - Remove the IBM CAPI (cxl) driver\n\nThanks to Andrew Donnellan.\n\n* tag \u0027powerpc-6.15-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:\n docs: Fix references to IBM CAPI (cxl) removal version\n cxl: Remove driver\n" }, { "commit": "8a6b94032e0077ae15b21e7b42a9599bdc18ea22", "tree": "25d45e87121f0a2b71f10c03652b375b38d963f9", "parents": [ "1c241cba196decd73b6959e851bfb71d0a1c1767", "33c9da5dfb18c2ff5a88d01aca2cf253cd0ac3bc" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 12:25:03 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 12:25:03 2025 -0700" }, "message": "Merge tag \u0027uml-for-linux-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/uml/linux\n\nPull UML updates from Johannes Berg:\n\n - proper nofault accesses and read-only rodata\n\n - hostfs fix for host inode number reuse\n\n - fixes for host errno handling\n\n - various cleanups/small fixes\n\n* tag \u0027uml-for-linux-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/uml/linux:\n um: Rewrite the sigio workaround based on epoll and tgkill\n um: Prohibit the VM_CLONE flag in run_helper_thread()\n um: Switch to the pthread-based helper in sigio workaround\n um: ubd: Switch to the pthread-based helper\n um: Add pthread-based helper support\n um: x86: clean up elf specific definitions\n um: Store full CSGSFS and SS register from mcontext\n um: virt-pci: Refactor virtio_pcidev into its own module\n um: work around sched_yield not yielding in time-travel mode\n um/locking: Remove semicolon from \"lock\" prefix\n um: Update min_low_pfn to match changes in uml_reserved\n um: use str_yes_no() to remove hardcoded \"yes\" and \"no\"\n um: hostfs: avoid issues on inode number reuse by host\n um: Allocate vdso page pointer statically\n um: remove copy_from_kernel_nofault_allowed\n um: mark rodata read-only and implement _nofault accesses\n um: Pass the correct Rust target and options with gcc\n" }, { "commit": "1c241cba196decd73b6959e851bfb71d0a1c1767", "tree": "953c27d10eaeb5b07429c573a69198f099841605", "parents": [ "6cb094583aef6533d0fd47ae7d134718da3399da", "17ba839c3c6c95562f329340e67da432309dd0d4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 12:15:01 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 12:15:01 2025 -0700" }, "message": "Merge tag \u0027loongarch-6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson\n\nPull LoongArch updates from Huacai Chen:\n\n - Always select HAVE_VIRT_CPU_ACCOUNTING_GEN\n\n - Enable UBSAN (Undefined Behavior Sanitizer)\n\n - Increase MAX_IO_PICS up to 8\n\n - Increase ARCH_DMA_MINALIGN up to 16\n\n - Fix and improve BPF JIT\n\n - Fix and improve vDSO implementation\n\n - Update the default config file\n\n - Some bug fixes and other small changes\n\n* tag \u0027loongarch-6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson:\n LoongArch: Update Loongson-3 default config file\n LoongArch: vDSO: Make use of the t8 register for vgetrandom-chacha\n LoongArch: vDSO: Remove --hash-style\u003dsysv\n LoongArch: BPF: Don\u0027t override subprog\u0027s return value\n LoongArch: BPF: Use move_addr() for BPF_PSEUDO_FUNC\n LoongArch: BPF: Fix off-by-one error in build_prologue()\n LoongArch: Rework the arch_kgdb_breakpoint() implementation\n LoongArch: Fix device node refcount leak in fdt_cpu_clk_init()\n LoongArch: Increase ARCH_DMA_MINALIGN up to 16\n LoongArch: Increase MAX_IO_PICS up to 8\n LoongArch: Fix help text of CMDLINE_EXTEND in Kconfig\n LoongArch: Enable UBSAN (Undefined Behavior Sanitizer)\n LoongArch: Always select HAVE_VIRT_CPU_ACCOUNTING_GEN\n rust: Fix enabling Rust and building with GCC for LoongArch\n" }, { "commit": "6cb094583aef6533d0fd47ae7d134718da3399da", "tree": "4c96b5e4526287ee36875785c60a2840f1207064", "parents": [ "92b71befc349587d58fdbbe6cdd68fb67f4933a8", "e8f45927ee5d99fa52f14205a2c7ac3820c64457" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 11:33:20 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 11:33:20 2025 -0700" }, "message": "Merge tag \u0027x86_tdx_for_6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull x86 TDX updates from Dave Hansen:\n \"Avoid direct HLT instruction execution in TDX guests.\n\n TDX guests aren\u0027t expected to use the HLT instruction directly. It\n causes a virtualization exception (#VE). While the #VE _can_ be\n handled, the current handling is slow and buggy and the easiest thing\n is just to avoid HLT in the first place. Plus, the kernel already has\n paravirt infrastructure that makes it relatively painless.\n\n Make TDX guests require paravirt and add some TDX-specific paravirt\n handlers which avoid HLT in the normal halt routines. Also add a\n warning in case another HLT sneaks in.\n\n There was a report that this leads to a \"major performance\n improvement\" on specjbb2015, probably because of the extra #VE\n overhead or missed wakeups from the buggy HLT handling\"\n\n* tag \u0027x86_tdx_for_6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:\n x86/tdx: Emit warning if IRQs are enabled during HLT #VE handling\n x86/tdx: Fix arch_safe_halt() execution for TDX VMs\n x86/paravirt: Move halt paravirt calls under CONFIG_PARAVIRT\n" }, { "commit": "92b71befc349587d58fdbbe6cdd68fb67f4933a8", "tree": "8950b3714f8fb1e5c2e78d3d4db85d5c4ea511d9", "parents": [ "af54a3a151691a969b04396cff15afe70d4da824", "7c977393b8277ed319e92e4b598b26598c9d30c0" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 10:30:10 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 10:30:10 2025 -0700" }, "message": "Merge tag \u0027objtool-urgent-2025-04-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull objtool fixes from Ingo Molnar:\n \"These are objtool fixes and updates by Josh Poimboeuf, centered around\n the fallout from the new CONFIG_OBJTOOL_WERROR\u003dy feature, which,\n despite its default-off nature, increased the profile/impact of\n objtool warnings:\n\n - Improve error handling and the presentation of warnings/errors\n\n - Revert the new summary warning line that some test-bot tools\n interpreted as new regressions\n\n - Fix a number of objtool warnings in various drivers, core kernel\n code and architecture code. About half of them are potential\n problems related to out-of-bounds accesses or potential undefined\n behavior, the other half are additional objtool annotations\n\n - Update objtool to latest (known) compiler quirks and objtool bugs\n triggered by compiler code generation\n\n - Misc fixes\"\n\n* tag \u0027objtool-urgent-2025-04-01\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (36 commits)\n objtool/loongarch: Add unwind hints in prepare_frametrace()\n rcu-tasks: Always inline rcu_irq_work_resched()\n context_tracking: Always inline ct_{nmi,irq}_{enter,exit}()\n sched/smt: Always inline sched_smt_active()\n objtool: Fix verbose disassembly if CROSS_COMPILE isn\u0027t set\n objtool: Change \"warning:\" to \"error: \" for fatal errors\n objtool: Always fail on fatal errors\n Revert \"objtool: Increase per-function WARN_FUNC() rate limit\"\n objtool: Append \"()\" to function name in \"unexpected end of section\" warning\n objtool: Ignore end-of-section jumps for KCOV/GCOV\n objtool: Silence more KCOV warnings, part 2\n objtool, drm/vmwgfx: Don\u0027t ignore vmw_send_msg() for ORC\n objtool: Fix STACK_FRAME_NON_STANDARD for cold subfunctions\n objtool: Fix segfault in ignore_unreachable_insn()\n objtool: Fix NULL printf() \u0027%s\u0027 argument in builtin-check.c:save_argv()\n objtool, lkdtm: Obfuscate the do_nothing() pointer\n objtool, regulator: rk808: Remove potential undefined behavior in rk806_set_mode_dcdc()\n objtool, ASoC: codecs: wcd934x: Remove potential undefined behavior in wcd934x_slim_irq_handler()\n objtool, Input: cyapa - Remove undefined behavior in cyapa_update_fw_store()\n objtool, panic: Disable SMAP in __stack_chk_fail()\n ...\n" }, { "commit": "af54a3a151691a969b04396cff15afe70d4da824", "tree": "a73a616c13178d638b2710ece0fd388e4476a0cb", "parents": [ "da0512b2a3b4124b1483751f090d7ac13a713f0e", "bd67c1c3c353b6560f2983bdd23c665e26cf83f9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 10:05:55 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 10:05:55 2025 -0700" }, "message": "Merge tag \u0027printk-for-6.15-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/printk/linux\n\nPull more printk updates from Petr Mladek:\n\n - Silence warnings about candidates for ‘gnu_print’ format attribute\n\n* tag \u0027printk-for-6.15-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/printk/linux:\n vsnprintf: Silence false positive GCC warning for va_format()\n vsnprintf: Drop unused const char fmt * in va_format()\n vsnprintf: Mark binary printing functions with __printf() attribute\n tracing: Mark binary printing functions with __printf() attribute\n seq_file: Mark binary printing functions with __printf() attribute\n seq_buf: Mark binary printing functions with __printf() attribute\n" }, { "commit": "da0512b2a3b4124b1483751f090d7ac13a713f0e", "tree": "42ba2c0c1d11d6f51b9cfe6e9b3f0b08ffd6f7ac", "parents": [ "002dcfd05740801ab30f88350de47e47c916af1c", "1dc1e0b9d694eb9016d3105ca4ba8bd90eba888a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 10:04:48 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 10:04:48 2025 -0700" }, "message": "Merge tag \u0027rcu-fixes-v6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rcu/linux\n\nPull RCU fix from Boqun Feng:\n\n - srcu: Make FORCE_NEED_SRCU_NMI_SAFE depend on RCU_EXPERT\n\n* tag \u0027rcu-fixes-v6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rcu/linux:\n srcu: Make FORCE_NEED_SRCU_NMI_SAFE depend on RCU_EXPERT\n" }, { "commit": "002dcfd05740801ab30f88350de47e47c916af1c", "tree": "42e1685548658d446c688ad80f72ddf1ecab10ce", "parents": [ "2985dae1e521ee1464130902415f5863ea05dc34", "afdbe49276accb87a0c7414e75864c78289ece2f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 09:55:51 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 09:55:51 2025 -0700" }, "message": "Merge tag \u0027kgdb-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/danielt/linux\n\nPull kgdb updates from Daniel Thompson:\n \"Two cleanups this cycle. The larger of which is the removal of a\n private allocator within kdb and replacing it with regular memory\n allocation. The other adopts the simplified version of strscpy() in a\n couple of places in kdb\"\n\n* tag \u0027kgdb-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/danielt/linux:\n kdb: Remove optional size arguments from strscpy() calls\n kdb: remove usage of static environment buffer\n" }, { "commit": "2985dae1e521ee1464130902415f5863ea05dc34", "tree": "aeb14046f01895dd65d8700ae5ef80a27ffa3561", "parents": [ "5de0afb422e18a22968ac3d4f700e26fa15a20f9" ], "author": { "name": "Alexei Starovoitov", "email": "ast@kernel.org", "time": "Mon Mar 31 20:23:36 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 09:34:31 2025 -0700" }, "message": "mm/page_alloc: Fix try_alloc_pages\n\nFix an obvious bug. try_alloc_pages() should set_page_refcounted.\n\n[ Not so obvious: it was probably correct at the time it was written but\n was at some point then rebased on top of v6.14-rc1.\n\n And at that point there was a semantic conflict with commit\n efabfe1420f5 (\"mm/page_alloc: move set_page_refcounted() to callers\n of get_page_from_freelist()\") and became buggy.\n\t\t\t\t\t\t\t- Linus ]\n\nFixes: 97769a53f117 (\"mm, bpf: Introduce try_alloc_pages() for opportunistic page allocation\")\nSigned-off-by: Alexei Starovoitov \u003cast@kernel.org\u003e\nReviewed-by: Shakeel Butt \u003cshakeel.butt@linux.dev\u003e\nAcked-by: Michal Hocko \u003cmhocko@suse.com\u003e\nAcked-by: Vlastimil BAbka \u003cvbabka@suse.cz\u003e\nReviewed-by: Harry Yoo \u003charry.yoo@oracle.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "5de0afb422e18a22968ac3d4f700e26fa15a20f9", "tree": "a291a03d5beab1be6af0745d760185228b36c2e8", "parents": [ "acc4d5ff0b61eb1715c498b6536c38c1feb7f3c1", "9764d5b0cd0ea4846fd46c7d0b4238ea122075a9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 09:14:59 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Wed Apr 02 09:14:59 2025 -0700" }, "message": "Merge tag \u0027v6.15-p2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6\n\nPull crypto fix from Herbert Xu:\n\n - revert the multibuffer hash testing as it is buggy\n\n* tag \u0027v6.15-p2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:\n Revert \"crypto: testmgr - Add multibuffer hash testing\"\n" }, { "commit": "8e5419d6542fdf2dca9a0acdef2b8255f0e4ba69", "tree": "43fede2d741cf889d4ece352f6b65f6632f29ef4", "parents": [ "9e8f324bd44c1fe026b582b75213de4eccfa1163" ], "author": { "name": "Dan Carpenter", "email": "dan.carpenter@linaro.org", "time": "Wed Apr 02 14:02:40 2025 +0300" }, "committer": { "name": "Trond Myklebust", "email": "trond.myklebust@hammerspace.com", "time": "Wed Apr 02 09:53:16 2025 -0400" }, "message": "nfs: Add missing release on error in nfs_lock_and_join_requests()\n\nCall nfs_release_request() on this error path before returning.\n\nFixes: c3f2235782c3 (\"nfs: fold nfs_folio_find_and_lock_request into nfs_lock_and_join_requests\")\nSigned-off-by: Dan Carpenter \u003cdan.carpenter@linaro.org\u003e\nLink: https://lore.kernel.org/r/3aaaa3d5-1c8a-41e4-98c7-717801ddd171@stanley.mountain\nSigned-off-by: Trond Myklebust \u003ctrond.myklebust@hammerspace.com\u003e\n" }, { "commit": "5bf46fe2b84cda662062f7aca73e15602c76a844", "tree": "0ff5bcf61994c1613495d9f23fe0d37e0783bb97", "parents": [ "a984d060b41d99518ff569ba0c867b29dbf34c1a", "ebca08fef88febdb0a898cefa7c99b9e25b3a984", "3035a6dd2d4736738949600b8abebbdb181e86ef", "359595b20a3617da9fe611f35f2197023bdbda62" ], "author": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Wed Apr 02 15:29:08 2025 +0200" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Wed Apr 02 15:29:08 2025 +0200" }, "message": "Merge branches \u0027acpi-video\u0027, \u0027acpi-platform-profile\u0027 and \u0027acpi-misc\u0027\n\nMerge an ACPI backlight (video) driver fix, an ACPI platform-profile\ndriver optimization, and a miscellaneous ACPI-related cleanup for\n6.15-rc1:\n\n - Make the ACPI backlight driver handle fetching EDID as\n ACPI_TYPE_PACKAGE which is not specification-compliant, but\n has been encountered in the field (Gergo Koteles).\n\n - Simplify the aggregation of choices in the ACPI platform-profile\n driver which has become possible after recent modifications of that\n driver (Kurt Borja).\n\n - Use str_enabled_disabled() instead of hardcoded strings in the ACPI\n code related to NUMA (Thorsten Blum).\n\n* acpi-video:\n ACPI: video: Handle fetching EDID as ACPI_TYPE_PACKAGE\n\n* acpi-platform-profile:\n ACPI: platform_profile: Optimize _aggregate_choices()\n\n* acpi-misc:\n ACPI: NUMA: Use str_enabled_disabled() helper function\n" }, { "commit": "a984d060b41d99518ff569ba0c867b29dbf34c1a", "tree": "58e067b81d258f26458d9f6f99e61f7579a2bc0b", "parents": [ "f06777cf2bbc21dd8c71d6e3906934e56b4e18e4", "2fa87c71d2adb4b82c105f9191e6120340feff00", "9e9b893404d43894d69a18dd2fc8fcf1c36abb7e", "2da31ea2a085cd189857f2db0f7b78d0162db87a" ], "author": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Wed Apr 02 15:27:48 2025 +0200" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Wed Apr 02 15:27:48 2025 +0200" }, "message": "Merge branches \u0027acpi-x86\u0027, \u0027acpi-processor\u0027 and \u0027acpi-resource\u0027\n\nMerge an x86-specific ACPI fix, an ACPI processor driver fix, and a new\nACPI resources management quirk for 6.15-rc1:\n\n - Extend the Lenovo Yoga Tab 3 ACPI quirk to skip GPIO event-handlers\n along with ACPI AC and battery which makes it work with Linux when\n started in the Windows mode (Hans de Goede).\n\n - Prevent the ACPI processor idle driver from being used on systems\n without _CST and with invalid C2/C3 in FADT in order to restore its\n previous (and expected) behavior that has been altered inadvertently\n by a recent code change (Giovanni Gherdovich).\n\n - Skip ACPI IRQ override on ASUS Vivobook 14 X1404VAP to make the\n internal keyboard work on it (Paul Menzel).\n\n* acpi-x86:\n ACPI: x86: Extend Lenovo Yoga Tab 3 quirk with skip GPIO event-handlers\n\n* acpi-processor:\n ACPI: processor: idle: Return an error if both P_LVL{2,3} idle states are invalid\n\n* acpi-resource:\n ACPI: resource: Skip IRQ override on ASUS Vivobook 14 X1404VAP\n" }, { "commit": "892c4e465c360d07f529bc3668fde7cbd4ea6b32", "tree": "f84fb2d36960e2618ee10072ad5ee3d50054ab3e", "parents": [ "64f7efb0f536d4800f38df785f2b372f7ba8a405" ], "author": { "name": "Michael Ellerman", "email": "mpe@ellerman.id.au", "time": "Wed Apr 02 23:09:52 2025 +1100" }, "committer": { "name": "Michael Ellerman", "email": "mpe@ellerman.id.au", "time": "Wed Apr 02 23:09:52 2025 +1100" }, "message": "docs: Fix references to IBM CAPI (cxl) removal version\n\nThe IBM CAPI (cxl) driver was removed in 6.15, not 6.14.\n\nSigned-off-by: Michael Ellerman \u003cmpe@ellerman.id.au\u003e\n" }, { "commit": "acc4d5ff0b61eb1715c498b6536c38c1feb7f3c1", "tree": "ac43eb64286548bc95a4da9431dceb8034a51812", "parents": [ "3491aa04787f4d7e00da98d94b1b10001c398b5a", "f278b6d5bb465c7fd66f3d103812947e55b376ed" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 20:00:51 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 20:00:51 2025 -0700" }, "message": "Merge tag \u0027net-6.15-rc0\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net\n\nPull networking fixes from Jakub Kicinski:\n \"Rather tiny pull request, mostly so that we can get into our trees\n your fix to the x86 Makefile.\n\n Current release - regressions:\n\n - Revert \"tcp: avoid atomic operations on sk-\u003esk_rmem_alloc\", error\n queue accounting was missed\n\n Current release - new code bugs:\n\n - 5 fixes for the netdevice instance locking work\n\n Previous releases - regressions:\n\n - usbnet: restore usb%d name exception for local mac addresses\n\n Previous releases - always broken:\n\n - rtnetlink: allocate vfinfo size for VF GUIDs when supported, avoid\n spurious GET_LINK failures\n\n - eth: mana: Switch to page pool for jumbo frames\n\n - phy: broadcom: Correct BCM5221 PHY model detection\n\n Misc:\n\n - selftests: drv-net: replace helpers for referring to other files\"\n\n* tag \u0027net-6.15-rc0\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (22 commits)\n Revert \"tcp: avoid atomic operations on sk-\u003esk_rmem_alloc\"\n bnxt_en: bring back rtnl lock in bnxt_shutdown\n eth: gve: add missing netdev locks on reset and shutdown paths\n selftests: mptcp: ignore mptcp_diag binary\n selftests: mptcp: close fd_in before returning in main_loop\n selftests: mptcp: fix incorrect fd checks in main_loop\n mptcp: fix NULL pointer in can_accept_new_subflow\n octeontx2-af: Free NIX_AF_INT_VEC_GEN irq\n octeontx2-af: Fix mbox INTR handler when num VFs \u003e 64\n net: fix use-after-free in the netdev_nl_sock_priv_destroy()\n selftests: net: use Path helpers in ping\n selftests: net: use the dummy bpf from net/lib\n selftests: drv-net: replace the rpath helper with Path objects\n net: lapbether: use netdev_lockdep_set_classes() helper\n net: phy: broadcom: Correct BCM5221 PHY model detection\n net: usb: usbnet: restore usb%d name exception for local mac addresses\n net/mlx5e: SHAMPO, Make reserved size independent of page size\n net: mana: Switch to page pool for jumbo frames\n MAINTAINERS: Add dedicated entries for phy_link_topology\n net: move replay logic to tc_modify_qdisc\n ...\n" }, { "commit": "3491aa04787f4d7e00da98d94b1b10001c398b5a", "tree": "fd94fec4e9c4e08df8e919e57fcc974c52c88c3c", "parents": [ "4b98d5dcd145aab10219b9f259b70110cd34f01a", "860be250fc32de9cb24154bf21b4e36f40925707" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 19:35:19 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 19:35:19 2025 -0700" }, "message": "Merge tag \u0027vfio-v6.15-rc1\u0027 of https://github.com/awilliam/linux-vfio\n\nPull VFIO updates from Alex Williamson:\n\n - Relax IGD support code to match display class device rather than\n specifically requiring a VGA device (Tomita Moeko)\n\n - Accelerate DMA mapping of device MMIO by iterating at PMD and PUD\n levels to take advantage of huge pfnmap support added in v6.12\n (Alex Williamson)\n\n - Extend virtio vfio-pci variant driver to include migration support\n for block devices where enabled by the PF (Yishai Hadas)\n\n - Virtualize INTx PIN register for devices where the platform does not\n route legacy PCI interrupts for the device and the interrupt is\n reported as IRQ_NOTCONNECTED (Alex Williamson)\n\n* tag \u0027vfio-v6.15-rc1\u0027 of https://github.com/awilliam/linux-vfio:\n vfio/pci: Handle INTx IRQ_NOTCONNECTED\n vfio/virtio: Enable support for virtio-block live migration\n vfio/type1: Use mapping page mask for pfnmaps\n mm: Provide address mask in struct follow_pfnmap_args\n vfio/type1: Use consistent types for page counts\n vfio/type1: Use vfio_batch for vaddr_get_pfns()\n vfio/type1: Convert all vaddr_get_pfns() callers to use vfio_batch\n vfio/type1: Catch zero from pin_user_pages_remote()\n vfio/pci: match IGD devices in display controller class\n" }, { "commit": "4b98d5dcd145aab10219b9f259b70110cd34f01a", "tree": "9fbbb25159553340c59cd4f4d2a891e27a8dfd75", "parents": [ "48552153cf49e252071f28e45d770b3741040e4e", "9d8960672d63db4b3b04542f5622748b345c637a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 18:52:54 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 18:52:54 2025 -0700" }, "message": "Merge tag \u0027for_linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost\n\nPull virtio updates from Michael Tsirkin:\n \"A small number of improvements all over the place:\n\n - shutdown has been reworked to reset devices\n\n - virtio fs is now allowed in vduse\n\n - vhost-scsi memory use has been reduced\n\n - cleanups, fixes all over the place\n\n A couple more fixes are being tested and will be merged after rc1\"\n\n* tag \u0027for_linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:\n vhost-scsi: Reduce response iov mem use\n vhost-scsi: Allocate iov_iter used for unaligned copies when needed\n vhost-scsi: Stop duplicating se_cmd fields\n vhost-scsi: Dynamically allocate scatterlists\n vhost-scsi: Return queue full for page alloc failures during copy\n vhost-scsi: Add better resource allocation failure handling\n vhost-scsi: Allocate T10 PI structs only when enabled\n vhost-scsi: Reduce mem use by moving upages to per queue\n vduse: add virtio_fs to allowed dev id\n sound/virtio: Fix cancel_sync warnings on uninitialized work_structs\n vdpa/mlx5: Fix oversized null mkey longer than 32bit\n vdpa/mlx5: Fix mlx5_vdpa_get_config() endianness on big-endian machines\n vhost-scsi: Fix handling of multiple calls to vhost_scsi_set_endpoint\n tools: virtio/linux/module.h add MODULE_DESCRIPTION() define.\n tools: virtio/linux/compiler.h: Add data_race() define.\n tools/virtio: Add DMA_MAPPING_ERROR and sg_dma_len api define for virtio test\n virtio: break and reset virtio devices on device_shutdown()\n" }, { "commit": "48552153cf49e252071f28e45d770b3741040e4e", "tree": "de68c5245933e2cd1464595c66ca84b3e62573bb", "parents": [ "792b8307ecd237ba719736c5310430cff3dd2296", "7be11d34f660bfa6583f3d6e2032d5dcbff56081" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 18:03:46 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 18:03:46 2025 -0700" }, "message": "Merge tag \u0027for-linus-iommufd\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jgg/iommufd\n\nPull iommufd updates from Jason Gunthorpe:\n \"Two significant new items:\n\n - Allow reporting IOMMU HW events to userspace when the events are\n clearly linked to a device.\n\n This is linked to the VIOMMU object and is intended to be used by a\n VMM to forward HW events to the virtual machine as part of\n emulating a vIOMMU. ARM SMMUv3 is the first driver to use this\n mechanism. Like the existing fault events the data is delivered\n through a simple FD returning event records on read().\n\n - PASID support in VFIO.\n\n The \"Process Address Space ID\" is a PCI feature that allows the\n device to tag all PCI DMA operations with an ID. The IOMMU will\n then use the ID to select a unique translation for those DMAs. This\n is part of Intel\u0027s vIOMMU support as VT-D HW requires the\n hypervisor to manage each PASID entry.\n\n The support is generic so any VFIO user could attach any\n translation to a PASID, and the support should work on ARM SMMUv3\n as well. AMD requires additional driver work.\n\n Some minor updates, along with fixes:\n\n - Prevent using nested parents with fault\u0027s, no driver support today\n\n - Put a single \"cookie_type\" value in the iommu_domain to indicate\n what owns the various opaque owner fields\"\n\n* tag \u0027for-linus-iommufd\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jgg/iommufd: (49 commits)\n iommufd: Test attach before detaching pasid\n iommufd: Fix iommu_vevent_header tables markup\n iommu: Convert unreachable() to BUG()\n iommufd: Balance veventq-\u003enum_events inc/dec\n iommufd: Initialize the flags of vevent in iommufd_viommu_report_event()\n iommufd/selftest: Add coverage for reporting max_pasid_log2 via IOMMU_HW_INFO\n iommufd: Extend IOMMU_GET_HW_INFO to report PASID capability\n vfio: VFIO_DEVICE_[AT|DE]TACH_IOMMUFD_PT support pasid\n vfio-iommufd: Support pasid [at|de]tach for physical VFIO devices\n ida: Add ida_find_first_range()\n iommufd/selftest: Add coverage for iommufd pasid attach/detach\n iommufd/selftest: Add test ops to test pasid attach/detach\n iommufd/selftest: Add a helper to get test device\n iommufd/selftest: Add set_dev_pasid in mock iommu\n iommufd: Allow allocating PASID-compatible domain\n iommu/vt-d: Add IOMMU_HWPT_ALLOC_PASID support\n iommufd: Enforce PASID-compatible domain for RID\n iommufd: Support pasid attach/replace\n iommufd: Enforce PASID-compatible domain in PASID path\n iommufd/device: Add pasid_attach array to track per-PASID attach\n ...\n" }, { "commit": "792b8307ecd237ba719736c5310430cff3dd2296", "tree": "32c25338fdb5f65d08e300475d3df5f6791e629f", "parents": [ "8868485d6b13e6e12b9a90fb5d8cb2f26eb1264e", "212120a164d59fd534148d315f13db3d296efb0f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 17:46:45 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 17:46:45 2025 -0700" }, "message": "Merge tag \u0027edac_urgent_for_v6.15_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras\n\nPull EDAC documentation fix from Borislav Petkov:\n\n - A single fix making sure the EDAC subtree is included in the\n documentation table of contents\n\n* tag \u0027edac_urgent_for_v6.15_rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras:\n Documentation/EDAC: Fix warning document isn\u0027t included in any toctree\n" }, { "commit": "64f7efb0f536d4800f38df785f2b372f7ba8a405", "tree": "074d383101b389dce3b847fba9418ec241ba0397", "parents": [ "7b667acd69e316c2ed1b47e5dcd9d093be4a843f", "5a0fcb0ef5584caf7da3f31896e08650c532e4c1" ], "author": { "name": "Michael Ellerman", "email": "mpe@ellerman.id.au", "time": "Wed Apr 02 11:07:44 2025 +1100" }, "committer": { "name": "Michael Ellerman", "email": "mpe@ellerman.id.au", "time": "Wed Apr 02 11:07:44 2025 +1100" }, "message": "Merge branch \u0027topic/cxl\u0027 into next\n\nThis merges in the removal of the IBM CAPI \"cxl\" driver.\n" }, { "commit": "8868485d6b13e6e12b9a90fb5d8cb2f26eb1264e", "tree": "f06ae8d598042d4b4b05353e775944095f09b991", "parents": [ "1df7752800b41de14dd268bf53d854bc81017973", "12da0fee4543ffbd50ff01b16153cc6ca2f7453a" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 16:51:44 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 16:51:44 2025 -0700" }, "message": "Merge tag \u0027thermal-6.15-rc1-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm\n\nPull more thermal control updates from Rafael Wysocki:\n \"These are mostly assorted updates of thermal drivers used on ARM\n platforms:\n\n - Use dev_err_probe() helpers to simplify the init code in the Qoriq\n thermal driver (Frank Li)\n\n - Power down the Qoriq\u0027s TMU at suspend time (Alice Guo)\n\n - Add ipq5332, ipq5424 compatible to the QCom\u0027s tsens thermal driver\n and TSENS enable / calibration support for V2 (Praveenkumar I)\n\n - Add missing rk3328 mapping entry (Trevor Woerner)\n\n - Remove duplicate struct declaration from the thermal core header\n file (Xueqin Luo)\n\n - Disable the monitoring mode during suspend in the LVTS Mediatek\n driver to prevent temperature acquisition glitches (Nícolas F. R.\n A. Prado)\n\n - Disable Stage 3 thermal threshold in the LVTS Mediatek driver\n because it disables the suspend ability and does not have an\n interrupt handler (Nícolas F. R. A. Prado)\n\n - Fix low temperature offset interrupt in the LVTS Mediatek driver to\n prevent multiple interrupts from triggering when the system is at\n its normal functionning temperature (Nícolas F. R. A. Prado)\n\n - Enable interrupts in the LVTS Mediatek driver only on sensors that\n are in use (Nícolas F. R. A. Prado)\n\n - Add the BCM74110 compatible DT binding and the corresponding code\n to support a chip based on a different process node than previous\n chips (Florian Fainelli)\n\n - Correct indentation and style in DTS example (Krzysztof Kozlowski)\n\n - Unify hexadecimal annotatation in the rcar_gen3 driver (Niklas\n Söderlund)\n\n - Factor out the code logic to read fuses on Gen3 and Gen4 in the\n rcar_gen3 thermal driver (Niklas Söderlund)\n\n - Drop unused driver data from the QCom\u0027s spmi temperature alarm\n driver (Johan Hovold)\"\n\n* tag \u0027thermal-6.15-rc1-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:\n thermal/drivers/qcom-spmi-temp-alarm: Drop unused driver data\n thermal: rcar_gen3: Reuse logic to read fuses on Gen3 and Gen4\n thermal: rcar_gen3: Use lowercase hex constants\n dt-bindings: thermal: Correct indentation and style in DTS example\n thermal/drivers/brcmstb_thermal: Add support for BCM74110\n dt-bindings: thermal: Update for BCM74110\n thermal/drivers/mediatek/lvts: Only update IRQ enable for valid sensors\n thermal/drivers/mediatek/lvts: Start sensor interrupts disabled\n thermal/drivers/mediatek/lvts: Disable low offset IRQ for minimum threshold\n thermal/drivers/mediatek/lvts: Disable Stage 3 thermal threshold\n thermal/drivers/mediatek/lvts: Disable monitor mode during suspend\n thermal: core: Remove duplicate struct declaration\n thermal/drivers/rockchip: Add missing rk3328 mapping entry\n thermal/drivers/tsens: Add TSENS enable and calibration support for V2\n dt-bindings: thermal: tsens: Add ipq5332, ipq5424 compatible\n thermal/drivers/qoriq: Power down TMU on system suspend\n thermal/drivers/qoriq: Use dev_err_probe() simplify the code\n" }, { "commit": "1df7752800b41de14dd268bf53d854bc81017973", "tree": "032902d3799e7ce55fca8a64269ba6fa462fa069", "parents": [ "696c45bcc3c35486578fd741d8551865aee42915", "bd496a44f041da9ef3afe14d1d6193d460424e91" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 16:47:47 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 16:47:47 2025 -0700" }, "message": "Merge tag \u0027i3c/for-6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/i3c/linux\n\nPull i3c updates from Alexandre Belloni:\n \"The silvaco driver gets support for the integration of the IP in the\n Nuvoton npcm845 SoC. There is also a fix for a possible NULL pointer\n dereference that can happen with early IBIs. Summary:\n\n Core:\n\n - Fix a possible NULL pointer dereference due to IBI coming when the\n target driver is not yet probed.\n\n Drivers:\n\n - mipi-i3c-hci: Use I2C DMA-safe api\n\n - svc: add Nuvoton npcm845 support\"\n\n* tag \u0027i3c/for-6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/i3c/linux:\n i3c: Add NULL pointer check in i3c_master_queue_ibi()\n i3c: master: Drop duplicate check before calling OF APIs\n i3c: master: svc: Fix implicit fallthrough in svc_i3c_master_ibi_work()\n i3c: master: svc: Fix missing STOP for master request\n i3c: master: svc: Use readsb helper for reading MDB\n i3c: master: svc: Fix missing the IBI rules\n i3c: master: svc: Fix i3c_master_get_free_addr return check\n i3c: master: svc: Fix npcm845 DAA process corruption\n i3c: master: svc: Fix npcm845 invalid slvstart event\n i3c: master: svc: Fix npcm845 FIFO empty issue\n i3c: master: svc: Add support for Nuvoton npcm845 i3c\n dt-bindings: i3c: silvaco: Add npcm845 compatible string\n dt-bindings: i3c: dw: Add power-domains\n i3c: master: svc: Flush FIFO before sending Dynamic Address Assignment(DAA)\n i3c: mipi-i3c-hci: Use I2C DMA-safe api\n i3c: Remove the const qualifier from i2c_msg pointer in i2c_xfers API\n MAINTAINERS: Add Frank Li to Silvaco I3C\n MAINTAINERS: Remove Conor Culhane from Silvaco I3C\n" }, { "commit": "696c45bcc3c35486578fd741d8551865aee42915", "tree": "289c1fcd75c9f38ada7893b99dc10640a4a724aa", "parents": [ "28a1b05678f4e88de90b0987b06e13c454ad9bd6", "9bc64d338b0b4b2061049df8b701f9786857690e" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 16:33:36 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 16:33:36 2025 -0700" }, "message": "Merge tag \u0027linux-watchdog-6.15-rc1\u0027 of git://www.linux-watchdog.org/linux-watchdog\n\nPull watchdog updates from Wim Van Sebroeck:\n\n - Add watchdog driver for Lenovo SE30 platform\n\n - Add support for Allwinner A523\n\n - Add i.MX94 support\n\n - watchdog framework: Convert to use device property\n\n - renesas,wdt: Document RZ/G3E support\n\n - Various other fixes and improvemenents\n\n* tag \u0027linux-watchdog-6.15-rc1\u0027 of git://www.linux-watchdog.org/linux-watchdog:\n watchdog: sunxi_wdt: Add support for Allwinner A523\n dt-bindings: watchdog: sunxi: add Allwinner A523 compatible string\n watchdog: aspeed: fix 64-bit division\n watchdog: npcm: Remove unnecessary NULL check before clk_prepare_enable/clk_disable_unprepare\n dt-bindings: watchdog: renesas,wdt: Document RZ/G3E support\n watchdog: Convert to use device property\n watchdog: lenovo_se30_wdt: include io.h for devm_ioremap()\n dt-bindings: watchdog: fsl-imx7ulp-wdt: Add i.MX94 support\n watchdog: nic7018_wdt: tidy up ACPI ID table\n watchdog: s3c2410_wdt: Fix PMU register bits for ExynosAutoV920 SoC\n watchdog: lenovo_se30_wdt: Watchdog driver for Lenovo SE30 platform\n watchdog: Enable RZV2HWDT driver depend on ARCH_RENESAS\n watchdog: cros-ec: Add newlines to printks\n watchdog: aspeed: Update bootstatus handling\n" }, { "commit": "28a1b05678f4e88de90b0987b06e13c454ad9bd6", "tree": "c02b70e6890b5b650451df683c6ffbdfab4ad94f", "parents": [ "91e5bfe317d8f8471fbaa3e70cf66cae1314a516", "31396626eaf0be0e8edc87b801fcd205016e42d9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 14:21:02 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 14:21:02 2025 -0700" }, "message": "Merge tag \u0027i2c-for-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux\n\nPull i2c updates from Wolfram Sang:\n \"i2c-core updates (collected by Wolfram):\n - remove last user and unexport i2c_of_match_device()\n - irq usage cleanup from Jiri\n\n i2c-host updates (collected by Andi):\n\n Refactoring and cleanups:\n - octeon, cadence, i801, pasemi, mlxbf, bcm-iproc: general\n refactorings\n - octeon: remove 10-bit address support\n\n Improvements:\n - amd-asf: improved error handling\n - designware: use guard(mutex)\n - amd-asf, designware: update naming to follow latest specs\n - cadence: fix cleanup path in probe\n - i801: use MMIO and I/O mapping helpers to access registers\n - pxa: handle error after clk_prepare_enable\n\n New features:\n - added i2c_10bit_addr_*_from_msg() and updated multiple drivers\n - omap: added multiplexer state handling\n - qcom-geni: update frequency configuration\n - qup: introduce DMA usage policy\n\n New hardware support:\n - exynos: add support for Samsung exynos7870\n - k1: add support for spacemit k1 (new driver)\n - imx: add support for i.mx94 lpi2c\n - rk3x: add support for rk3562\n - designware: add support for Renesas RZ/N1D\n\n Multiplexers:\n - ltc4306, reg: fix assignment in platform_driver structure\n\n at24 eeprom updates (collected by Bartosz):\n - add two new compatible entries to the DT binding document\n - drop of_match_ptr() and ACPI_PTR() macros\"\n\n* tag \u0027i2c-for-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux: (50 commits)\n dt-bindings: i2c: snps,designware-i2c: describe Renesas RZ/N1D variant\n irqdomain: i2c: Switch to irq_find_mapping()\n i2c: iproc: Refactor prototype and remove redundant error checks\n i2c: qcom-geni: Update i2c frequency table to match hardware guidance\n i2c: mlxbf: Use readl_poll_timeout_atomic() for polling\n i2c: pasemi: Add registers bits and switch to BIT()\n i2c: k1: Initialize variable before use\n i2c: spacemit: add support for SpacemiT K1 SoC\n dt-bindings: i2c: spacemit: add support for K1 SoC\n i2c: omap: Add support for setting mux\n dt-bindings: i2c: omap: Add mux-states property\n i2c: octeon: remove 10-bit addressing support\n i2c: octeon: fix return commenting\n i2c: i801: Use MMIO if available\n i2c: i801: Switch to iomapped register access\n i2c: i801: Improve too small kill wait time in i801_check_post\n i2c: i801: Move i801_wait_intr and i801_wait_byte_done in the code\n i2c: i801: Cosmetic improvements\n i2c: cadence: Move reset_control_assert after pm_runtime_set_suspended in probe error path\n i2c: cadence: Simplify using devm_clk_get_enabled()\n ...\n" }, { "commit": "212120a164d59fd534148d315f13db3d296efb0f", "tree": "a199cdd37719b887b7cc19c018b29db1de3b736d", "parents": [ "ae8371a46e59d768a5540a237309962b50b8bec7" ], "author": { "name": "Shiju Jose", "email": "shiju.jose@huawei.com", "time": "Tue Apr 01 12:58:23 2025 +0100" }, "committer": { "name": "Borislav Petkov (AMD)", "email": "bp@alien8.de", "time": "Tue Apr 01 22:26:47 2025 +0200" }, "message": "Documentation/EDAC: Fix warning document isn\u0027t included in any toctree\n\nFix the build (htmldocs) warning:\n\n Documentation/edac/index.rst: WARNING: document isn\u0027t included in any toctree.\n\nFixes: db99ea5f2c03 (\"EDAC: Add support for EDAC device features control\")\nCloses: https://lore.kernel.org/all/20250228185102.15842f8b@canb.auug.org.au/\nReported-by: Stephen Rothwell \u003csfr@canb.auug.org.au\u003e\nSigned-off-by: Shiju Jose \u003cshiju.jose@huawei.com\u003e\nSigned-off-by: Borislav Petkov (AMD) \u003cbp@alien8.de\u003e\nLink: https://lore.kernel.org/r/20250401115823.573-1-shiju.jose@huawei.com\n" }, { "commit": "91e5bfe317d8f8471fbaa3e70cf66cae1314a516", "tree": "489cf03545521f5eb6fa421f3dd33cce9795687c", "parents": [ "e63a165308468d0dce39e07c97279152b043875b", "6565439894570a07b00dba0b739729fe6b56fba4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 12:57:14 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 12:57:14 2025 -0700" }, "message": "Merge tag \u0027dmaengine-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine\n\nPull dmaengine updates from Vinod Koul:\n \"The dmaengine subsystem updates for this cycle consist of a new driver\n (Microchip) along with couple of yaml binding conversions, core api\n updates and bunch of driver updates etc.\n\n New HW support:\n\n - Microchip sama7d65 dma controller\n\n - Yaml conversion of atmel dma binding and Freescale Elo DMA\n Controller binding\n\n Core:\n\n - Remove device_prep_dma_imm_data() API as users are removed\n\n - Reduce scope of some less frequently used DMA request channel APIs\n with aim to cleanup these in future\n\n Updates:\n\n - Drop Fenghua Yu from idxd maintainers, as he changed jobs\n\n - AMD ptdma support for multiqueue and ae4dma deprecated PCI IDs\n removal\"\n\n* tag \u0027dmaengine-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine: (29 commits)\n dmaengine: ptdma: Utilize the AE4DMA engine\u0027s multi-queue functionality\n dmaengine: ae4dma: Use the MSI count and its corresponding IRQ number\n dmaengine: ae4dma: Remove deprecated PCI IDs\n dmaengine: Remove device_prep_dma_imm_data from struct dma_device\n dmaengine: ti: edma: support sw triggered chans in of_edma_xlate()\n dmaengine: ti: k3-udma: Enable second resource range for BCDMA and PKTDMA\n dmaengine: fsl-edma: free irq correctly in remove path\n dmaengine: fsl-edma: cleanup chan after dma_async_device_unregister\n dt-bindings: dma: snps,dw-axi-dmac: Allow devices to be marked as noncoherent\n dmaengine: dmatest: Fix dmatest waiting less when interrupted\n dt-bindings: dma: Convert fsl,elo*-dma to YAML\n dt-bindings: dma: fsl-mxs-dma: Add compatible string for i.MX8 chips\n dmaengine: Fix typo in comment\n dmaengine: ti: k3-udma-glue: Drop skip_fdq argument from k3_udma_glue_reset_rx_chn\n dmaengine: bcm2835-dma: fix warning when CONFIG_PM\u003dn\n dt-bindings: dma: fsl,edma: Add i.MX94 support\n dt-bindings: dma: atmel: add microchip,sama7d65-dma\n dmaengine: img-mdc: remove incorrect of_match_ptr annotation\n dmaengine: idxd: Delete unnecessary NULL check\n dmaengine: pxa: Enable compile test\n ...\n" }, { "commit": "e63a165308468d0dce39e07c97279152b043875b", "tree": "b74ef577e6737ff0082da28c43582f69761b835f", "parents": [ "4d31167e844bac8be7587781c52af450b6451f01", "f13bff25161b8a0a9d716764ebe57334d496c6d9" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 12:47:11 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 12:47:11 2025 -0700" }, "message": "Merge tag \u0027phy-for-6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/phy/linux-phy\n\nPull phy updates from Vinod Koul:\n \"A fairly moderate sized request for the generic phy subsystem with\n some new device and driver support along with driver updates with\n Samsung and Qualcomm ones being major ones.\n\n New HW Support:\n\n - Qualcomm X1P42100 PCIe Gen4x4, QCS615 qmp usbc, PCIe UNIPHY 28LP\n driver, SM8750 QMP UFS PHY\n\n - Rockchip rk3576 hdptx, rk3562 naneng-combo support\n\n - Samsung MIPI D-/C-PHY driver, ExynosAutov920 ufs phy driver\n\n Updates:\n\n - Samsung USB3 Type-C lane orientation detection and configuration\n for Google gs101\n\n - Qualcomm support for dual lane PHY support for QCS8300 SoC\"\n\n* tag \u0027phy-for-6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/phy/linux-phy: (47 commits)\n phy: rockchip-naneng-combo: Support rk3562\n dt-bindings: phy: rockchip: Add rk3562 naneng-combophy compatible\n phy: rockchip: Add Samsung MIPI D-/C-PHY driver\n dt-bindings: phy: Add Rockchip MIPI C-/D-PHY schema\n phy: qcom: uniphy-28lp: add COMMON_CLK dependency\n phy: rockchip: usbdp: Remove unnecessary bool conversion\n phy: rockchip: usbdp: Avoid call hpd_event_trigger in dp_phy_init\n phy: rockchip: usbdp: Only verify link rates/lanes/voltage when the corresponding set flags are set\n phy: qcom-qmp-pcie: add dual lane PHY support for QCS8300\n dt-bindings: phy: qcom,sc8280xp-qmp-pcie-phy: Document the QCS8300 QMP PCIe PHY Gen4 x2\n phy: qcom-qmp-ufs: Add PHY Configuration support for sm8750\n dt-bindings: phy: qcom,sc8280xp-qmp-ufs-phy: document the SM8750 QMP UFS PHY\n phy: qcom: Introduce PCIe UNIPHY 28LP driver\n dt-bindings: phy: qcom,uniphy-pcie: Document PCIe uniphy\n phy: qcom: qmp-usbc: Add qmp configuration for QCS615\n phy: freescale: imx8m-pcie: assert phy reset and perst in power off\n phy: freescale: imx8m-pcie: cleanup reset logic\n phy: core: Remove unused phy_pm_runtime_(allow|forbid)\n dt-bindings: phy: document Allwinner A523 USB-2.0 PHY\n phy: phy-rockchip-samsung-hdptx: Add support for RK3576\n ...\n" }, { "commit": "4d31167e844bac8be7587781c52af450b6451f01", "tree": "53c16596a96fe51544318cf0bae7c1e7d62bc747", "parents": [ "25601e85441dd91cf7973b002f27af4c5b8691ea", "08ae0d61c3d79bb5d52ae30ad4fc12442e966a23" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 12:43:13 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 12:43:13 2025 -0700" }, "message": "Merge tag \u0027soundwire-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/soundwire\n\nPull soundwire updates from Vinod Koul:\n\n - Support for SoundWire Bulk Register Access (BRA) protocol in core\n along with Intel driver support and ASoC bits required\n\n - AMD driver updates and support for ACP 7.0 and 7.1 platforms\n\n* tag \u0027soundwire-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/soundwire: (28 commits)\n soundwire: take in count the bandwidth of a prepared stream\n ASoC: rt711-sdca: add DP0 support\n soundwire: debugfs: add interface for BPT/BRA transfers\n ASoC: SOF: Intel: hda-sdw-bpt: add CHAIN_DMA support\n soundwire: intel_ace2x: add BPT send_async/wait callbacks\n soundwire: intel: add BPT context definition\n ASoC: SOF: Intel: hda-sdw-bpt: add helpers for SoundWire BPT DMA\n soundwire: intel_auxdevice: add indirection for BPT send_async/wait\n soundwire: cadence: add BTP/BRA helpers to format data\n soundwire: bus: add bpt_stream pointer\n soundwire: bus: add send_async/wait APIs for BPT protocol\n soundwire: stream: reuse existing code for BPT stream\n soundwire: stream: special-case the bus compute_params() routine\n soundwire: stream: extend sdw_alloc_stream() to take \u0027type\u0027 parameter\n soundwire: extend sdw_stream_type to BPT\n soundwire: cadence: add BTP support for DP0\n Documentation: driver: add SoundWire BRA description\n soundwire: amd: change the log level for command response log\n soundwire: slave: fix an OF node reference leak in soundwire slave device\n soundwire: Use str_enable_disable-like helpers\n ...\n" }, { "commit": "25601e85441dd91cf7973b002f27af4c5b8691ea", "tree": "821267764ed1dbdfb7d08839f7bf0054358eb4f8", "parents": [ "2cd5769fb0b78b8ef583ab4c0015c2c48d525dac", "5d1a5c4f121f0ec50327e899c9450978505f1560" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 11:26:08 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 11:26:08 2025 -0700" }, "message": "Merge tag \u0027char-misc-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc\n\nPull char / misc / IIO driver updates from Greg KH:\n \"Here is the big set of char, misc, iio, and other smaller driver\n subsystems for 6.15-rc1. Lots of stuff in here, including:\n\n - loads of IIO changes and driver updates\n\n - counter driver updates\n\n - w1 driver updates\n\n - faux conversions for some drivers that were abusing the platform\n bus interface\n\n - coresight driver updates\n\n - rust miscdevice binding updates based on real-world-use\n\n - other minor driver updates\n\n All of these have been in linux-next with no reported issues for quite\n a while\"\n\n* tag \u0027char-misc-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc: (292 commits)\n samples: rust_misc_device: fix markup in top-level docs\n Coresight: Fix a NULL vs IS_ERR() bug in probe\n misc: lis3lv02d: convert to use faux_device\n tlclk: convert to use faux_device\n regulator: dummy: convert to use the faux device interface\n bus: mhi: host: Fix race between unprepare and queue_buf\n coresight: configfs: Constify struct config_item_type\n doc: iio: ad7380: describe offload support\n iio: ad7380: add support for SPI offload\n iio: light: Add check for array bounds in veml6075_read_int_time_ms\n iio: adc: ti-ads7924 Drop unnecessary function parameters\n staging: iio: ad9834: Use devm_regulator_get_enable()\n staging: iio: ad9832: Use devm_regulator_get_enable()\n iio: gyro: bmg160_spi: add of_match_table\n dt-bindings: iio: adc: Add i.MX94 and i.MX95 support\n iio: adc: ad7768-1: remove unnecessary locking\n Documentation: ABI: add wideband filter type to sysfs-bus-iio\n iio: adc: ad7768-1: set MOSI idle state to prevent accidental reset\n iio: adc: ad7768-1: Fix conversion result sign\n iio: adc: ad7124: Benefit of dev \u003d indio_dev-\u003edev.parent in ad7124_parse_channel_config()\n ...\n" }, { "commit": "2cd5769fb0b78b8ef583ab4c0015c2c48d525dac", "tree": "e553ca1215f21488b9d2be5eadef668fb8b50bf7", "parents": [ "d6b02199cde4b9cb99b311eeab1cdbe23165082c", "51d0de7596a458096756c895cfed6bc4a7ecac10" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 11:02:03 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 11:02:03 2025 -0700" }, "message": "Merge tag \u0027driver-core-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core\n\nPull driver core updatesk from Greg KH:\n \"Here is the big set of driver core updates for 6.15-rc1. Lots of stuff\n happened this development cycle, including:\n\n - kernfs scaling changes to make it even faster thanks to rcu\n\n - bin_attribute constify work in many subsystems\n\n - faux bus minor tweaks for the rust bindings\n\n - rust binding updates for driver core, pci, and platform busses,\n making more functionaliy available to rust drivers. These are all\n due to people actually trying to use the bindings that were in\n 6.14.\n\n - make Rafael and Danilo full co-maintainers of the driver core\n codebase\n\n - other minor fixes and updates\"\n\n* tag \u0027driver-core-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core: (52 commits)\n rust: platform: require Send for Driver trait implementers\n rust: pci: require Send for Driver trait implementers\n rust: platform: impl Send + Sync for platform::Device\n rust: pci: impl Send + Sync for pci::Device\n rust: platform: fix unrestricted \u0026mut platform::Device\n rust: pci: fix unrestricted \u0026mut pci::Device\n rust: device: implement device context marker\n rust: pci: use to_result() in enable_device_mem()\n MAINTAINERS: driver core: mark Rafael and Danilo as co-maintainers\n rust/kernel/faux: mark Registration methods inline\n driver core: faux: only create the device if probe() succeeds\n rust/faux: Add missing parent argument to Registration::new()\n rust/faux: Drop #[repr(transparent)] from faux::Registration\n rust: io: fix devres test with new io accessor functions\n rust: io: rename `io::Io` accessors\n kernfs: Move dput() outside of the RCU section.\n efi: rci2: mark bin_attribute as __ro_after_init\n rapidio: constify \u0027struct bin_attribute\u0027\n firmware: qemu_fw_cfg: constify \u0027struct bin_attribute\u0027\n powerpc/perf/hv-24x7: Constify \u0027struct bin_attribute\u0027\n ...\n" }, { "commit": "d6b02199cde4b9cb99b311eeab1cdbe23165082c", "tree": "408f91cb2a6241098bd42f04bd15b2848953ea28", "parents": [ "eb0ece16027f8223d5dc9aaf90124f70577bd22a", "434333dd3f66f9d1ad387dabd2a565182a823f31" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 10:06:52 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 10:06:52 2025 -0700" }, "message": "Merge tag \u0027mm-nonmm-stable-2025-03-30-18-23\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm\n\nPull non-MM updates from Andrew Morton:\n\n - The series \"powerpc/crash: use generic crashkernel reservation\" from\n Sourabh Jain changes powerpc\u0027s kexec code to use more of the generic\n layers.\n\n - The series \"get_maintainer: report subsystem status separately\" from\n Vlastimil Babka makes some long-requested improvements to the\n get_maintainer output.\n\n - The series \"ucount: Simplify refcounting with rcuref_t\" from\n Sebastian Siewior cleans up and optimizing the refcounting in the\n ucount code.\n\n - The series \"reboot: support runtime configuration of emergency\n hw_protection action\" from Ahmad Fatoum improves the ability for a\n driver to perform an emergency system shutdown or reboot.\n\n - The series \"Converge on using secs_to_jiffies() part two\" from Easwar\n Hariharan performs further migrations from msecs_to_jiffies() to\n secs_to_jiffies().\n\n - The series \"lib/interval_tree: add some test cases and cleanup\" from\n Wei Yang permits more userspace testing of kernel library code, adds\n some more tests and performs some cleanups.\n\n - The series \"hung_task: Dump the blocking task stacktrace\" from Masami\n Hiramatsu arranges for the hung_task detector to dump the stack of\n the blocking task and not just that of the blocked task.\n\n - The series \"resource: Split and use DEFINE_RES*() macros\" from Andy\n Shevchenko provides some cleanups to the resource definition macros.\n\n - Plus the usual shower of singleton patches - please see the\n individual changelogs for details.\n\n* tag \u0027mm-nonmm-stable-2025-03-30-18-23\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (77 commits)\n mailmap: consolidate email addresses of Alexander Sverdlin\n fs/procfs: fix the comment above proc_pid_wchan()\n relay: use kasprintf() instead of fixed buffer formatting\n resource: replace open coded variant of DEFINE_RES()\n resource: replace open coded variants of DEFINE_RES_*_NAMED()\n resource: replace open coded variant of DEFINE_RES_NAMED_DESC()\n resource: split DEFINE_RES_NAMED_DESC() out of DEFINE_RES_NAMED()\n samples: add hung_task detector mutex blocking sample\n hung_task: show the blocker task if the task is hung on mutex\n kexec_core: accept unaccepted kexec segments\u0027 destination addresses\n watchdog/perf: optimize bytes copied and remove manual NUL-termination\n lib/interval_tree: fix the comment of interval_tree_span_iter_next_gap()\n lib/interval_tree: skip the check before go to the right subtree\n lib/interval_tree: add test case for span iteration\n lib/interval_tree: add test case for interval_tree_iter_xxx() helpers\n lib/rbtree: add random seed\n lib/rbtree: split tests\n lib/rbtree: enable userland test suite for rbtree related data structure\n checkpatch: describe --min-conf-desc-length\n scripts/gdb/symbols: determine KASLR offset on s390\n ...\n" }, { "commit": "9e4e249018d208678888bdf22f6b652728106528", "tree": "ec8a7d37973e058126a960f5c7d35f4f700a5ea3", "parents": [ "7d20aa5c32ac8bd272b5470ddbd7ac6e0cb35714" ], "author": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Fri Mar 28 21:39:08 2025 +0100" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Tue Apr 01 18:44:12 2025 +0200" }, "message": "cpufreq: Reference count policy in cpufreq_update_limits()\n\nSince acpi_processor_notify() can be called before registering a cpufreq\ndriver or even in cases when a cpufreq driver is not registered at all,\ncpufreq_update_limits() needs to check if a cpufreq driver is present\nand prevent it from being unregistered.\n\nFor this purpose, make it call cpufreq_cpu_get() to obtain a cpufreq\npolicy pointer for the given CPU and reference count the corresponding\npolicy object, if present.\n\nFixes: 5a25e3f7cc53 (\"cpufreq: intel_pstate: Driver-specific handling of _PPC updates\")\nCloses: https://lore.kernel.org/linux-acpi/Z-ShAR59cTow0KcR@mail-itl\nReported-by: Marek Marczykowski-Górecki \u003cmarmarek@invisiblethingslab.com\u003e\nCc: All applicable \u003cstable@vger.kernel.org\u003e\nSigned-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\nAcked-by: Viresh Kumar \u003cviresh.kumar@linaro.org\u003e\nLink: https://patch.msgid.link/1928789.tdWV9SEqCh@rjwysocki.net\n" }, { "commit": "eb0ece16027f8223d5dc9aaf90124f70577bd22a", "tree": "1e2214cacd123b940ceca684322203643d5e9bc7", "parents": [ "08733088b566b58283f0f12fb73f5db6a9a9de30", "0a1e082b64ccce165e7307a7b49d22b2504f9d1f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 09:29:18 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Tue Apr 01 09:29:18 2025 -0700" }, "message": "Merge tag \u0027mm-stable-2025-03-30-16-52\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm\n\nPull MM updates from Andrew Morton:\n\n - The series \"Enable strict percpu address space checks\" from Uros\n Bizjak uses x86 named address space qualifiers to provide\n compile-time checking of percpu area accesses.\n\n This has caused a small amount of fallout - two or three issues were\n reported. In all cases the calling code was found to be incorrect.\n\n - The series \"Some cleanup for memcg\" from Chen Ridong implements some\n relatively monir cleanups for the memcontrol code.\n\n - The series \"mm: fixes for device-exclusive entries (hmm)\" from David\n Hildenbrand fixes a boatload of issues which David found then using\n device-exclusive PTE entries when THP is enabled. More work is\n needed, but this makes thins better - our own HMM selftests now\n succeed.\n\n - The series \"mm: zswap: remove z3fold and zbud\" from Yosry Ahmed\n remove the z3fold and zbud implementations. They have been deprecated\n for half a year and nobody has complained.\n\n - The series \"mm: further simplify VMA merge operation\" from Lorenzo\n Stoakes implements numerous simplifications in this area. No runtime\n effects are anticipated.\n\n - The series \"mm/madvise: remove redundant mmap_lock operations from\n process_madvise()\" from SeongJae Park rationalizes the locking in the\n madvise() implementation. Performance gains of 20-25% were observed\n in one MADV_DONTNEED microbenchmark.\n\n - The series \"Tiny cleanup and improvements about SWAP code\" from\n Baoquan He contains a number of touchups to issues which Baoquan\n noticed when working on the swap code.\n\n - The series \"mm: kmemleak: Usability improvements\" from Catalin\n Marinas implements a couple of improvements to the kmemleak\n user-visible output.\n\n - The series \"mm/damon/paddr: fix large folios access and schemes\n handling\" from Usama Arif provides a couple of fixes for DAMON\u0027s\n handling of large folios.\n\n - The series \"mm/damon/core: fix wrong and/or useless damos_walk()\n behaviors\" from SeongJae Park fixes a few issues with the accuracy of\n kdamond\u0027s walking of DAMON regions.\n\n - The series \"expose mapping wrprotect, fix fb_defio use\" from Lorenzo\n Stoakes changes the interaction between framebuffer deferred-io and\n core MM. No functional changes are anticipated - this is preparatory\n work for the future removal of page structure fields.\n\n - The series \"mm/damon: add support for hugepage_size DAMOS filter\"\n from Usama Arif adds a DAMOS filter which permits the filtering by\n huge page sizes.\n\n - The series \"mm: permit guard regions for file-backed/shmem mappings\"\n from Lorenzo Stoakes extends the guard region feature from its\n present \"anon mappings only\" state. The feature now covers shmem and\n file-backed mappings.\n\n - The series \"mm: batched unmap lazyfree large folios during\n reclamation\" from Barry Song cleans up and speeds up the unmapping\n for pte-mapped large folios.\n\n - The series \"reimplement per-vma lock as a refcount\" from Suren\n Baghdasaryan puts the vm_lock back into the vma. Our reasons for\n pulling it out were largely bogus and that change made the code more\n messy. This patchset provides small (0-10%) improvements on one\n microbenchmark.\n\n - The series \"Docs/mm/damon: misc DAMOS filters documentation fixes and\n improves\" from SeongJae Park does some maintenance work on the DAMON\n docs.\n\n - The series \"hugetlb/CMA improvements for large systems\" from Frank\n van der Linden addresses a pile of issues which have been observed\n when using CMA on large machines.\n\n - The series \"mm/damon: introduce DAMOS filter type for unmapped pages\"\n from SeongJae Park enables users of DMAON/DAMOS to filter my the\n page\u0027s mapped/unmapped status.\n\n - The series \"zsmalloc/zram: there be preemption\" from Sergey\n Senozhatsky teaches zram to run its compression and decompression\n operations preemptibly.\n\n - The series \"selftests/mm: Some cleanups from trying to run them\" from\n Brendan Jackman fixes a pile of unrelated issues which Brendan\n encountered while runnimg our selftests.\n\n - The series \"fs/proc/task_mmu: add guard region bit to pagemap\" from\n Lorenzo Stoakes permits userspace to use /proc/pid/pagemap to\n determine whether a particular page is a guard page.\n\n - The series \"mm, swap: remove swap slot cache\" from Kairui Song\n removes the swap slot cache from the allocation path - it simply\n wasn\u0027t being effective.\n\n - The series \"mm: cleanups for device-exclusive entries (hmm)\" from\n David Hildenbrand implements a number of unrelated cleanups in this\n code.\n\n - The series \"mm: Rework generic PTDUMP configs\" from Anshuman Khandual\n implements a number of preparatoty cleanups to the GENERIC_PTDUMP\n Kconfig logic.\n\n - The series \"mm/damon: auto-tune aggregation interval\" from SeongJae\n Park implements a feedback-driven automatic tuning feature for\n DAMON\u0027s aggregation interval tuning.\n\n - The series \"Fix lazy mmu mode\" from Ryan Roberts fixes some issues in\n powerpc, sparc and x86 lazy MMU implementations. Ryan did this in\n preparation for implementing lazy mmu mode for arm64 to optimize\n vmalloc.\n\n - The series \"mm/page_alloc: Some clarifications for migratetype\n fallback\" from Brendan Jackman reworks some commentary to make the\n code easier to follow.\n\n - The series \"page_counter cleanup and size reduction\" from Shakeel\n Butt cleans up the page_counter code and fixes a size increase which\n we accidentally added late last year.\n\n - The series \"Add a command line option that enables control of how\n many threads should be used to allocate huge pages\" from Thomas\n Prescher does that. It allows the careful operator to significantly\n reduce boot time by tuning the parallalization of huge page\n initialization.\n\n - The series \"Fix calculations in trace_balance_dirty_pages() for cgwb\"\n from Tang Yizhou fixes the tracing output from the dirty page\n balancing code.\n\n - The series \"mm/damon: make allow filters after reject filters useful\n and intuitive\" from SeongJae Park improves the handling of allow and\n reject filters. Behaviour is made more consistent and the documention\n is updated accordingly.\n\n - The series \"Switch zswap to object read/write APIs\" from Yosry Ahmed\n updates zswap to the new object read/write APIs and thus permits the\n removal of some legacy code from zpool and zsmalloc.\n\n - The series \"Some trivial cleanups for shmem\" from Baolin Wang does as\n it claims.\n\n - The series \"fs/dax: Fix ZONE_DEVICE page reference counts\" from\n Alistair Popple regularizes the weird ZONE_DEVICE page refcount\n handling in DAX, permittig the removal of a number of special-case\n checks.\n\n - The series \"refactor mremap and fix bug\" from Lorenzo Stoakes is a\n preparatoty refactoring and cleanup of the mremap() code.\n\n - The series \"mm: MM owner tracking for large folios (!hugetlb) +\n CONFIG_NO_PAGE_MAPCOUNT\" from David Hildenbrand reworks the manner in\n which we determine whether a large folio is known to be mapped\n exclusively into a single MM.\n\n - The series \"mm/damon: add sysfs dirs for managing DAMOS filters based\n on handling layers\" from SeongJae Park adds a couple of new sysfs\n directories to ease the management of DAMON/DAMOS filters.\n\n - The series \"arch, mm: reduce code duplication in mem_init()\" from\n Mike Rapoport consolidates many per-arch implementations of\n mem_init() into code generic code, where that is practical.\n\n - The series \"mm/damon/sysfs: commit parameters online via\n damon_call()\" from SeongJae Park continues the cleaning up of sysfs\n access to DAMON internal data.\n\n - The series \"mm: page_ext: Introduce new iteration API\" from Luiz\n Capitulino reworks the page_ext initialization to fix a boot-time\n crash which was observed with an unusual combination of compile and\n cmdline options.\n\n - The series \"Buddy allocator like (or non-uniform) folio split\" from\n Zi Yan reworks the code to split a folio into smaller folios. The\n main benefit is lessened memory consumption: fewer post-split folios\n are generated.\n\n - The series \"Minimize xa_node allocation during xarry split\" from Zi\n Yan reduces the number of xarray xa_nodes which are generated during\n an xarray split.\n\n - The series \"drivers/base/memory: Two cleanups\" from Gavin Shan\n performs some maintenance work on the drivers/base/memory code.\n\n - The series \"Add tracepoints for lowmem reserves, watermarks and\n totalreserve_pages\" from Martin Liu adds some more tracepoints to the\n page allocator code.\n\n - The series \"mm/madvise: cleanup requests validations and\n classifications\" from SeongJae Park cleans up some warts which\n SeongJae observed during his earlier madvise work.\n\n - The series \"mm/hwpoison: Fix regressions in memory failure handling\"\n from Shuai Xue addresses two quite serious regressions which Shuai\n has observed in the memory-failure implementation.\n\n - The series \"mm: reliable huge page allocator\" from Johannes Weiner\n makes huge page allocations cheaper and more reliable by reducing\n fragmentation.\n\n - The series \"Minor memcg cleanups \u0026 prep for memdescs\" from Matthew\n Wilcox is preparatory work for the future implementation of memdescs.\n\n - The series \"track memory used by balloon drivers\" from Nico Pache\n introduces a way to track memory used by our various balloon drivers.\n\n - The series \"mm/damon: introduce DAMOS filter type for active pages\"\n from Nhat Pham permits users to filter for active/inactive pages,\n separately for file and anon pages.\n\n - The series \"Adding Proactive Memory Reclaim Statistics\" from Hao Jia\n separates the proactive reclaim statistics from the direct reclaim\n statistics.\n\n - The series \"mm/vmscan: don\u0027t try to reclaim hwpoison folio\" from\n Jinjiang Tu fixes our handling of hwpoisoned pages within the reclaim\n code.\n\n* tag \u0027mm-stable-2025-03-30-16-52\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (431 commits)\n mm/page_alloc: remove unnecessary __maybe_unused in order_to_pindex()\n x86/mm: restore early initialization of high_memory for 32-bits\n mm/vmscan: don\u0027t try to reclaim hwpoison folio\n mm/hwpoison: introduce folio_contain_hwpoisoned_page() helper\n cgroup: docs: add pswpin and pswpout items in cgroup v2 doc\n mm: vmscan: split proactive reclaim statistics from direct reclaim statistics\n selftests/mm: speed up split_huge_page_test\n selftests/mm: uffd-unit-tests support for hugepages \u003e 2M\n docs/mm/damon/design: document active DAMOS filter type\n mm/damon: implement a new DAMOS filter type for active pages\n fs/dax: don\u0027t disassociate zero page entries\n MM documentation: add \"Unaccepted\" meminfo entry\n selftests/mm: add commentary about 9pfs bugs\n fork: use __vmalloc_node() for stack allocation\n docs/mm: Physical Memory: Populate the \"Zones\" section\n xen: balloon: update the NR_BALLOON_PAGES state\n hv_balloon: update the NR_BALLOON_PAGES state\n balloon_compaction: update the NR_BALLOON_PAGES state\n meminfo: add a per node counter for balloon drivers\n mm: remove references to folio in __memcg_kmem_uncharge_page()\n ...\n" }, { "commit": "7c977393b8277ed319e92e4b598b26598c9d30c0", "tree": "1e8b713aca884137b5ef62370b13ab31784680bb", "parents": [ "6309a5c43b0dc629851f25b2e5ef8beff61d08e5" ], "author": { "name": "Josh Poimboeuf", "email": "jpoimboe@kernel.org", "time": "Mon Mar 31 21:26:43 2025 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Tue Apr 01 10:10:10 2025 +0200" }, "message": "objtool/loongarch: Add unwind hints in prepare_frametrace()\n\nIf \u0027regs\u0027 points to a local stack variable, prepare_frametrace() stores\nall registers to the stack. This confuses objtool as it expects them to\nbe restored from the stack later.\n\nThe stores don\u0027t affect stack tracing, so use unwind hints to hide them\nfrom objtool.\n\nFixes the following warnings:\n\n arch/loongarch/kernel/traps.o: warning: objtool: show_stack+0xe0: stack state mismatch: reg1[22]\u003d-1+0 reg2[22]\u003d-2-160\n arch/loongarch/kernel/traps.o: warning: objtool: show_stack+0xe0: stack state mismatch: reg1[23]\u003d-1+0 reg2[23]\u003d-2-152\n\nFixes: cb8a2ef0848c (\"LoongArch: Add ORC stack unwinder support\")\nReported-by: kernel test robot \u003clkp@intel.com\u003e\nTested-by: Tiezhu Yang \u003cyangtiezhu@loongson.cn\u003e\nSigned-off-by: Josh Poimboeuf \u003cjpoimboe@kernel.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nLink: https://lore.kernel.org/r/270cadd8040dda74db2307f23497bb68e65db98d.1743481539.git.jpoimboe@kernel.org\nCloses: https://lore.kernel.org/oe-kbuild-all/202503280703.OARM8SrY-lkp@intel.com/\n" }, { "commit": "6309a5c43b0dc629851f25b2e5ef8beff61d08e5", "tree": "441c6e9c036a63107864846e65c47957c9b27da3", "parents": [ "9ac50f7311dc8b39e355582f14c1e82da47a8196" ], "author": { "name": "Josh Poimboeuf", "email": "jpoimboe@kernel.org", "time": "Mon Mar 31 21:26:46 2025 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Tue Apr 01 09:12:17 2025 +0200" }, "message": "rcu-tasks: Always inline rcu_irq_work_resched()\n\nThanks to CONFIG_DEBUG_SECTION_MISMATCH, empty functions can be\ngenerated out of line. rcu_irq_work_resched() can be called from\nnoinstr code, so make sure it\u0027s always inlined.\n\nFixes: 564506495ca9 (\"rcu/context-tracking: Move deferred nocb resched to context tracking\")\nReported-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\nSigned-off-by: Josh Poimboeuf \u003cjpoimboe@kernel.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\nCc: Paul E. McKenney \u003cpaulmck@kernel.org\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nLink: https://lore.kernel.org/r/e84f15f013c07e4c410d972e75620c53b62c1b3e.1743481539.git.jpoimboe@kernel.org\nCloses: https://lore.kernel.org/d1eca076-fdde-484a-b33e-70e0d167c36d@infradead.org\n" }, { "commit": "9ac50f7311dc8b39e355582f14c1e82da47a8196", "tree": "4a88a7badde52411ea54ae4bf70c5286ef161e69", "parents": [ "09f37f2d7b21ff35b8b533f9ab8cfad2fe8f72f6" ], "author": { "name": "Josh Poimboeuf", "email": "jpoimboe@kernel.org", "time": "Mon Mar 31 21:26:45 2025 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Tue Apr 01 09:12:05 2025 +0200" }, "message": "context_tracking: Always inline ct_{nmi,irq}_{enter,exit}()\n\nThanks to CONFIG_DEBUG_SECTION_MISMATCH, empty functions can be\ngenerated out of line. These can be called from noinstr code, so make\nsure they\u0027re always inlined.\n\nFixes the following warnings:\n\n vmlinux.o: warning: objtool: irqentry_nmi_enter+0xa2: call to ct_nmi_enter() leaves .noinstr.text section\n vmlinux.o: warning: objtool: irqentry_nmi_exit+0x16: call to ct_nmi_exit() leaves .noinstr.text section\n vmlinux.o: warning: objtool: irqentry_exit+0x78: call to ct_irq_exit() leaves .noinstr.text section\n\nFixes: 6f0e6c1598b1 (\"context_tracking: Take IRQ eqs entrypoints over RCU\")\nReported-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\nSigned-off-by: Josh Poimboeuf \u003cjpoimboe@kernel.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: Frederic Weisbecker \u003cfrederic@kernel.org\u003e\nCc: Paul E. McKenney \u003cpaulmck@kernel.org\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nLink: https://lore.kernel.org/r/8509bce3f536bcd4ae7af3a2cf6930d48c5e631a.1743481539.git.jpoimboe@kernel.org\nCloses: https://lore.kernel.org/d1eca076-fdde-484a-b33e-70e0d167c36d@infradead.org\n" }, { "commit": "09f37f2d7b21ff35b8b533f9ab8cfad2fe8f72f6", "tree": "3820bfda36c06a50bab2b64abe24129d9673235c", "parents": [ "e77956e4e5c11218e60a1fe8cdbccd02476f2e56" ], "author": { "name": "Josh Poimboeuf", "email": "jpoimboe@kernel.org", "time": "Mon Mar 31 21:26:44 2025 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Tue Apr 01 09:07:13 2025 +0200" }, "message": "sched/smt: Always inline sched_smt_active()\n\nsched_smt_active() can be called from noinstr code, so it should always\nbe inlined. The CONFIG_SCHED_SMT version already has __always_inline.\nDo the same for its !CONFIG_SCHED_SMT counterpart.\n\nFixes the following warning:\n\n vmlinux.o: error: objtool: intel_idle_ibrs+0x13: call to sched_smt_active() leaves .noinstr.text section\n\nFixes: 321a874a7ef8 (\"sched/smt: Expose sched_smt_present static key\")\nReported-by: kernel test robot \u003clkp@intel.com\u003e\nSigned-off-by: Josh Poimboeuf \u003cjpoimboe@kernel.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nLink: https://lore.kernel.org/r/1d03907b0a247cf7fb5c1d518de378864f603060.1743481539.git.jpoimboe@kernel.org\nCloses: https://lore.kernel.org/r/202503311434.lyw2Tveh-lkp@intel.com/\n" }, { "commit": "e77956e4e5c11218e60a1fe8cdbccd02476f2e56", "tree": "b55f6d86d24459e3a12edc8d86fb82bf4bcd5460", "parents": [ "3e7be635937d19b91bab70695328214a3d789d51" ], "author": { "name": "David Laight", "email": "david.laight.linux@gmail.com", "time": "Mon Mar 31 21:26:42 2025 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Tue Apr 01 09:07:13 2025 +0200" }, "message": "objtool: Fix verbose disassembly if CROSS_COMPILE isn\u0027t set\n\nIn verbose mode, when printing the disassembly of affected functions, if\nCROSS_COMPILE isn\u0027t set, the objdump command string gets prefixed with\n\"(null)\".\n\nSomehow this worked before. Maybe some versions of glibc return an\nempty string instead of NULL. Fix it regardless.\n\n[ jpoimboe: Rewrite commit log. ]\n\nFixes: ca653464dd097 (\"objtool: Add verbose option for disassembling affected functions\")\nSigned-off-by: David Laight \u003cdavid.laight.linux@gmail.com\u003e\nSigned-off-by: Josh Poimboeuf \u003cjpoimboe@kernel.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nLink: https://lore.kernel.org/r/20250215142321.14081-1-david.laight.linux@gmail.com\nLink: https://lore.kernel.org/r/b931a4786bc0127aa4c94e8b35ed617dcbd3d3da.1743481539.git.jpoimboe@kernel.org\n" }, { "commit": "3e7be635937d19b91bab70695328214a3d789d51", "tree": "a3894c0d69b1d9c3355ffd5f1ed1254f11c15f89", "parents": [ "0b10177114d1e434af850b377cf5e6620dd1d525" ], "author": { "name": "Josh Poimboeuf", "email": "jpoimboe@kernel.org", "time": "Mon Mar 31 21:26:41 2025 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Tue Apr 01 09:07:13 2025 +0200" }, "message": "objtool: Change \"warning:\" to \"error: \" for fatal errors\n\nThis is similar to GCC\u0027s behavior and makes it more obvious why the\nbuild failed.\n\nSigned-off-by: Josh Poimboeuf \u003cjpoimboe@kernel.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nLink: https://lore.kernel.org/r/0ea76f4b0e7a370711ed9f75fd0792bb5979c2bf.1743481539.git.jpoimboe@kernel.org\n" }, { "commit": "0b10177114d1e434af850b377cf5e6620dd1d525", "tree": "6dff9c15ed60f58df2257ad00e967eeeea08836e", "parents": [ "c5610071a69d1c94c70e681874298b4fc6942098" ], "author": { "name": "Josh Poimboeuf", "email": "jpoimboe@kernel.org", "time": "Mon Mar 31 21:26:40 2025 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Tue Apr 01 09:07:13 2025 +0200" }, "message": "objtool: Always fail on fatal errors\n\nObjtool writes several object annotations which are used to enable\ncritical kernel runtime functionalities like static calls and\nretpoline/rethunk patching.\n\nIn the rare case where it fails to read or write an object, the\nannotations don\u0027t get written, causing runtime code patching to fail and\ncode to become corrupted.\n\nDue to the catastrophic nature of such warnings, convert them to errors\nwhich fail the build regardless of CONFIG_OBJTOOL_WERROR.\n\nReported-by: Chaitanya Kumar Borah \u003cchaitanya.kumar.borah@intel.com\u003e\nSigned-off-by: Josh Poimboeuf \u003cjpoimboe@kernel.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nLink: https://lore.kernel.org/r/7d35684ca61eac56eb2424f300ca43c5d257b170.1743481539.git.jpoimboe@kernel.org\nCloses: https://lore.kernel.org/SJ1PR11MB61295789E25C2F5197EFF2F6B9A72@SJ1PR11MB6129.namprd11.prod.outlook.com\n" }, { "commit": "c5610071a69d1c94c70e681874298b4fc6942098", "tree": "675a1f869c2ae14e89b49b4ad0fa58a50d1a9582", "parents": [ "188d90f817e13b66e03e110eb6f82e8f5f0d654b" ], "author": { "name": "Josh Poimboeuf", "email": "jpoimboe@kernel.org", "time": "Mon Mar 31 21:26:39 2025 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Tue Apr 01 09:07:13 2025 +0200" }, "message": "Revert \"objtool: Increase per-function WARN_FUNC() rate limit\"\n\nThis reverts commit 0a7fb6f07e3ad497d31ae9a2082d2cacab43d54a.\n\nThe \"skipping duplicate warnings\" warning is technically not an actual\nwarning, which can cause confusion. This feature isn\u0027t all that useful\nanyway. It\u0027s exceedingly rare for a function to have more than one\nunrelated warning.\n\nSuggested-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nSigned-off-by: Josh Poimboeuf \u003cjpoimboe@kernel.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nLink: https://lore.kernel.org/r/e5abe5e858acf1a9207a5dfa0f37d17ac9dca872.1743481539.git.jpoimboe@kernel.org\n" }, { "commit": "188d90f817e13b66e03e110eb6f82e8f5f0d654b", "tree": "5f998b866d59498804a6cf47f04db552145b4d75", "parents": [ "0d7597749f5a3ac67851d3836635d084df15fb66" ], "author": { "name": "Josh Poimboeuf", "email": "jpoimboe@kernel.org", "time": "Mon Mar 31 21:26:38 2025 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Tue Apr 01 09:07:12 2025 +0200" }, "message": "objtool: Append \"()\" to function name in \"unexpected end of section\" warning\n\nAppend with \"()\" to clarify it\u0027s a function.\n\nBefore:\n\n vmlinux.o: warning: objtool: cdns_mrvl_xspi_setup_clock: unexpected end of section .text.cdns_mrvl_xspi_setup_clock\n\nAfter:\n\n vmlinux.o: warning: objtool: cdns_mrvl_xspi_setup_clock(): unexpected end of section .text.cdns_mrvl_xspi_setup_clock\n\nFixes: c5995abe1547 (\"objtool: Improve error handling\")\nSigned-off-by: Josh Poimboeuf \u003cjpoimboe@kernel.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nLink: https://lore.kernel.org/r/692e1e0d0b15a71bd35c6b4b87f3c75cd5a57358.1743481539.git.jpoimboe@kernel.org\n" }, { "commit": "0d7597749f5a3ac67851d3836635d084df15fb66", "tree": "ca813decef0bd4fe2abf10a4dc15fd648a44bd26", "parents": [ "55c78035a1a8dfb05f1472018ce2a651701adb7d" ], "author": { "name": "Josh Poimboeuf", "email": "jpoimboe@kernel.org", "time": "Mon Mar 31 21:26:37 2025 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Tue Apr 01 09:07:12 2025 +0200" }, "message": "objtool: Ignore end-of-section jumps for KCOV/GCOV\n\nWhen KCOV or GCOV is enabled, dead code can be left behind, in which\ncase objtool silences unreachable and undefined behavior (fallthrough)\nwarnings.\n\nFallthrough warnings, and their variant \"end of section\" warnings, were\nsilenced with the following commit:\n\n 6b023c784204 (\"objtool: Silence more KCOV warnings\")\n\nAnother variant of a fallthrough warning is a jump to the end of a\nfunction. If that function happens to be at the end of a section, the\njump destination doesn\u0027t actually exist.\n\nNormally that would be a fatal objtool error, but for KCOV/GCOV it\u0027s\njust another undefined behavior fallthrough. Silence it like the\nothers.\n\nFixes the following warning:\n\n drivers/iommu/dma-iommu.o: warning: objtool: iommu_dma_sw_msi+0x92: can\u0027t find jump dest instruction at .text+0x54d5\n\nFixes: 6b023c784204 (\"objtool: Silence more KCOV warnings\")\nReported-by: Randy Dunlap \u003crdunlap@infradead.org\u003e\nSigned-off-by: Josh Poimboeuf \u003cjpoimboe@kernel.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nLink: https://lore.kernel.org/r/08fbe7d7e1e20612206f1df253077b94f178d93e.1743481539.git.jpoimboe@kernel.org\nCloses: https://lore.kernel.org/314f8809-cd59-479b-97d7-49356bf1c8d1@infradead.org/\n" }, { "commit": "55c78035a1a8dfb05f1472018ce2a651701adb7d", "tree": "946da81018e871dc599304b661239fd07d268a42", "parents": [ "ae958b12940bcd4ffa32c44684e4f2878bc5e140" ], "author": { "name": "Josh Poimboeuf", "email": "jpoimboe@kernel.org", "time": "Mon Mar 31 21:26:36 2025 -0700" }, "committer": { "name": "Ingo Molnar", "email": "mingo@kernel.org", "time": "Tue Apr 01 09:07:12 2025 +0200" }, "message": "objtool: Silence more KCOV warnings, part 2\n\nSimilar to GCOV, KCOV can leave behind dead code and undefined behavior.\nWarnings related to those should be ignored.\n\nThe previous commit:\n\n 6b023c784204 (\"objtool: Silence more KCOV warnings\")\n\n... only did so for CONFIG_CGOV_KERNEL. Also do it for CONFIG_KCOV, but\nfor real this time.\n\nFixes the following warning:\n\n vmlinux.o: warning: objtool: synaptics_report_mt_data: unexpected end of section .text.synaptics_report_mt_data\n\nFixes: 6b023c784204 (\"objtool: Silence more KCOV warnings\")\nReported-by: kernel test robot \u003clkp@intel.com\u003e\nSigned-off-by: Josh Poimboeuf \u003cjpoimboe@kernel.org\u003e\nSigned-off-by: Ingo Molnar \u003cmingo@kernel.org\u003e\nCc: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\nLink: https://lore.kernel.org/r/a44ba16e194bcbc52c1cef3d3cd9051a62622723.1743481539.git.jpoimboe@kernel.org\nCloses: https://lore.kernel.org/oe-kbuild-all/202503282236.UhfRsF3B-lkp@intel.com/\n" }, { "commit": "08733088b566b58283f0f12fb73f5db6a9a9de30", "tree": "179b90c8371e8f32673cd5a84f65ad36244d3c3a", "parents": [ "98fb679d19a17aec624d53b016953a3fcd272e8d", "c1f4534b213d7be41b5d8b815a42d201a8f2978f" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 18:39:59 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 18:39:59 2025 -0700" }, "message": "Merge tag \u0027rust-fixes-6.15-merge\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ojeda/linux\n\nPull Rust fix from Miguel Ojeda:\n \"Fix \u0027generate_rust_analyzer.py\u0027 due to typo during merge\"\n\nMea culpa, mea maxima culpa.\n\n* tag \u0027rust-fixes-6.15-merge\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/ojeda/linux:\n scripts: generate_rust_analyzer: fix pin-init name in kernel deps\n" }, { "commit": "98fb679d19a17aec624d53b016953a3fcd272e8d", "tree": "78cea0d1b5368dd6991869aae78fba81c0aebfd2", "parents": [ "4080cf02f11e337c5031013f77e0ba1a475985ee", "650f5353dcc9b6e690a1c763754fa1e98d217bfc" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 18:33:51 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 18:33:51 2025 -0700" }, "message": "Merge tag \u0027bcachefs-2025-03-31\u0027 of git://evilpiepirate.org/bcachefs\n\nPull more bcachefs updates from Kent Overstreet:\n \"All bugfixes and logging improvements\"\n\n* tag \u0027bcachefs-2025-03-31\u0027 of git://evilpiepirate.org/bcachefs: (35 commits)\n bcachefs: fix bch2_write_point_to_text() units\n bcachefs: Log original key being moved in data updates\n bcachefs: BCH_JSET_ENTRY_log_bkey\n bcachefs: Reorder error messages that include journal debug\n bcachefs: Don\u0027t use designated initializers for disk_accounting_pos\n bcachefs: Silence errors after emergency shutdown\n bcachefs: fix units in rebalance_status\n bcachefs: bch2_ioctl_subvolume_destroy() fixes\n bcachefs: Clear fs_path_parent on subvolume unlink\n bcachefs: Change btree_insert_node() assertion to error\n bcachefs: Better printing of inconsistency errors\n bcachefs: bch2_count_fsck_err()\n bcachefs: Better helpers for inconsistency errors\n bcachefs: Consistent indentation of multiline fsck errors\n bcachefs: Add an \"ignore unknown\" option to bch2_parse_mount_opts()\n bcachefs: bch2_time_stats_init_no_pcpu()\n bcachefs: Fix bch2_fs_get_tree() error path\n bcachefs: fix logging in journal_entry_err_msg()\n bcachefs: add missing newline in bch2_trans_updates_to_text()\n bcachefs: print_string_as_lines: fix extra newline\n ...\n" }, { "commit": "4080cf02f11e337c5031013f77e0ba1a475985ee", "tree": "e71b6bd5b81ada4d473720a147e1c9ee5db60eeb", "parents": [ "172f7c91674fb3e55a7a00cfcba76719000811df", "6afdc60ec30b0a9390d11b7cebed79c857ce82aa" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 17:53:44 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 17:53:44 2025 -0700" }, "message": "Merge tag \u0027fs_for_v6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs\n\nPull ext2, udf, and isofs updates from Jan Kara:\n\n - conversion of ext2 to the new mount API\n\n - small folio conversion work for ext2\n\n - a fix of an unexpected return value in udf in inode_getblk()\n\n - a fix of handling of corrupted directory in isofs\n\n* tag \u0027fs_for_v6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:\n udf: Fix inode_getblk() return value\n ext2: Make ext2_params_spec static\n ext2: create ext2_msg_fc for use during parsing\n ext2: convert to the new mount API\n ext2: Remove reference to bh-\u003eb_page\n isofs: fix KMSAN uninit-value bug in do_isofs_readdir()\n" }, { "commit": "172f7c91674fb3e55a7a00cfcba76719000811df", "tree": "60decc8c777e2cb1b3437b9e6fb7551e5d164dc1", "parents": [ "f64a72bc767f6e9ddb18fdacaeb99708c4810ada", "c73e680d1f84059e1b1ea82a537f6ccc1c563eb4" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 17:49:35 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 17:49:35 2025 -0700" }, "message": "Merge tag \u0027exfat-for-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/linkinjeon/exfat\n\nPull exfat updates from Namjae Jeon:\n\n - Fix random stack corruption and incorrect error returns in\n exfat_get_block()\n\n - Optimize exfat_get_block() by improving checking corner cases\n\n - Fix an endless loop by self-linked chain in exfat_find_last_cluster\n\n - Remove dead EXFAT_CLUSTERS_UNTRACKED codes\n\n - Add missing shutdown check\n\n - Improve the delete performance with discard mount option\n\n* tag \u0027exfat-for-6.15-rc1\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/linkinjeon/exfat:\n exfat: call bh_read in get_block only when necessary\n exfat: fix potential wrong error return from get_block\n exfat: fix missing shutdown check\n exfat: fix the infinite loop in exfat_find_last_cluster()\n exfat: fix random stack corruption after get_block\n exfat: remove count used cluster from exfat_statfs()\n exfat: support batch discard of clusters when freeing clusters\n" }, { "commit": "f64a72bc767f6e9ddb18fdacaeb99708c4810ada", "tree": "52484ef4c427d4466a00053a839a44b316b5492d", "parents": [ "8b175e2e189673643bf5b996335f0430faddf953", "ddb7ea36ba7129c2ed107e2186591128618864e1" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 17:42:26 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 17:42:26 2025 -0700" }, "message": "Merge tag \u0027v6.15rc-part1-ksmbd-server-fixes\u0027 of git://git.samba.org/ksmbd\n\nPull smb server updates from Steve French:\n\n - Two fixes for bounds checks of open contexts\n\n - Two multichannel fixes, including one for important UAF\n\n - Oplock/lease break fix for potential ksmbd connection refcount leak\n\n - Security fix to free crypto data more securely\n\n - Fix to enable allowing Kerberos authentication by default\n\n - Two RDMA/smbdirect fixes\n\n - Minor cleanup\n\n* tag \u0027v6.15rc-part1-ksmbd-server-fixes\u0027 of git://git.samba.org/ksmbd:\n ksmbd: fix r_count dec/increment mismatch\n ksmbd: fix multichannel connection failure\n ksmbd: fix use-after-free in ksmbd_sessions_deregister()\n ksmbd: use ib_device_get_netdev() instead of calling ops.get_netdev\n ksmbd: use aead_request_free to match aead_request_alloc\n Revert \"ksmbd: fix missing RDMA-capable flag for IPoIB device in ksmbd_rdma_capable_netdev()\"\n ksmbd: add bounds check for create lease context\n ksmbd: add bounds check for durable handle context\n ksmbd: make SMB_SERVER_KERBEROS5 enable by default\n ksmbd: Use str_read_write() and str_true_false() helpers\n" }, { "commit": "8b175e2e189673643bf5b996335f0430faddf953", "tree": "03965d41eaf365fd6ab8ab2be2da4c461cad9e7f", "parents": [ "b6dde1e5275ed82e4c89844e95a03f95ca48be13", "e14b64247438e5026b2fce8ffd7cdf80a87e2bfa" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 17:38:34 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 17:38:34 2025 -0700" }, "message": "Merge tag \u00276.15-rc-part1-smb3-client-fixes\u0027 of git://git.samba.org/sfrench/cifs-2.6\n\nPull smb client updates from Steve French:\n\n - Fix for network namespace refcount leak\n\n - Multichannel fix and minor multichannel debug message cleanup\n\n - Fix potential null ptr reference in SMB3 close\n\n - Fix for special file handling when reparse points not supported by\n server\n\n - Two ACL fixes one for stricter ACE validation, one for incorrect\n perms requested\n\n - Three RFC1001 fixes: one for SMB3 mounts on port 139, one for better\n default hostname, and one for better session response processing\n\n - Minor update to email address for MAINTAINERS file\n\n - Allow disabling Unicode for access to old SMB1 servers\n\n - Three minor cleanups\n\n* tag \u00276.15-rc-part1-smb3-client-fixes\u0027 of git://git.samba.org/sfrench/cifs-2.6:\n cifs: Add new mount option -o nounicode to disable SMB1 UNICODE mode\n cifs: Set default Netbios RFC1001 server name to hostname in UNC\n smb: client: Fix netns refcount imbalance causing leaks and use-after-free\n cifs: add validation check for the fields in smb_aces\n CIFS: Propagate min offload along with other parameters from primary to secondary channels.\n cifs: Improve establishing SMB connection with NetBIOS session\n cifs: Fix establishing NetBIOS session for SMB2+ connection\n cifs: Fix getting DACL-only xattr system.cifs_acl and system.smb3_acl\n cifs: Check if server supports reparse points before using them\n MAINTAINERS: reorder preferred email for Steve French\n cifs: avoid NULL pointer dereference in dbg call\n smb: client: Remove redundant check in smb2_is_path_accessible()\n smb: client: Remove redundant check in cifs_oplock_break()\n smb: mark the new channel addition log as informational log with cifs_info\n smb: minor cleanup to remove unused function declaration\n" }, { "commit": "b6dde1e5275ed82e4c89844e95a03f95ca48be13", "tree": "0c85fe04158d2a308d8e187ced84875ad6c381b1", "parents": [ "1e7857b28020ba57ca7fdafae7ac855ba326c697", "26a80762153ba0dc98258b5e6d2e9741178c5114" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 17:28:17 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 17:28:17 2025 -0700" }, "message": "Merge tag \u0027nfsd-6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux\n\nPull nfsd updates from Chuck Lever:\n \"Neil Brown contributed more scalability improvements to NFSD\u0027s open\n file cache, and Jeff Layton contributed a menagerie of repairs to\n NFSD\u0027s NFSv4 callback / backchannel implementation.\n\n Mike Snitzer contributed a change to NFS re-export support that\n disables support for file locking on a re-exported NFSv4 mount. This\n is because NFSv4 state recovery is currently difficult if not\n impossible for re-exported NFS mounts. The change aims to prevent data\n integrity exposures after the re-export server crashes.\n\n Work continues on the evolving NFSD netlink administrative API.\n\n Many thanks to the contributors, reviewers, testers, and bug reporters\n who participated during the v6.15 development cycle\"\n\n* tag \u0027nfsd-6.15\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux: (45 commits)\n NFSD: Add a Kconfig setting to enable delegated timestamps\n sysctl: Fixes nsm_local_state bounds\n nfsd: use a long for the count in nfsd4_state_shrinker_count()\n nfsd: remove obsolete comment from nfs4_alloc_stid\n nfsd: remove unneeded forward declaration of nfsd4_mark_cb_fault()\n nfsd: reorganize struct nfs4_delegation for better packing\n nfsd: handle errors from rpc_call_async()\n nfsd: move cb_need_restart flag into cb_flags\n nfsd: replace CB_GETATTR_BUSY with NFSD4_CALLBACK_RUNNING\n nfsd: eliminate cl_ra_cblist and NFSD4_CLIENT_CB_RECALL_ANY\n nfsd: prevent callback tasks running concurrently\n nfsd: disallow file locking and delegations for NFSv4 reexport\n nfsd: filecache: drop the list_lru lock during lock gc scans\n nfsd: filecache: don\u0027t repeatedly add/remove files on the lru list\n nfsd: filecache: introduce NFSD_FILE_RECENT\n nfsd: filecache: use list_lru_walk_node() in nfsd_file_gc()\n nfsd: filecache: use nfsd_file_dispose_list() in nfsd_file_close_inode_sync()\n NFSD: Re-organize nfsd_file_gc_worker()\n nfsd: filecache: remove race handling.\n fs: nfs: acl: Avoid -Wflex-array-member-not-at-end warning\n ...\n" }, { "commit": "f278b6d5bb465c7fd66f3d103812947e55b376ed", "tree": "80f0f7103f809d9a485053e203a62e15e09b4c40", "parents": [ "dd07df9ff3d148aee87fcbab99ff14f0727752f4" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Mon Mar 31 07:59:46 2025 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:53:54 2025 -0700" }, "message": "Revert \"tcp: avoid atomic operations on sk-\u003esk_rmem_alloc\"\n\nThis reverts commit 0de2a5c4b824da2205658ebebb99a55c43cdf60f.\n\nI forgot that a TCP socket could receive messages in its error queue.\n\nsock_queue_err_skb() can be called without socket lock being held,\nand changes sk-\u003esk_rmem_alloc.\n\nThe fact that skbs in error queue are limited by sk-\u003esk_rcvbuf\nmeans that error messages can be dropped if socket receive\nqueues are full, which is an orthogonal issue.\n\nIn future kernels, we could use a separate sk-\u003esk_error_mem_alloc\ncounter specifically for the error queue.\n\nFixes: 0de2a5c4b824 (\"tcp: avoid atomic operations on sk-\u003esk_rmem_alloc\")\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nLink: https://patch.msgid.link/20250331075946.31960-1-edumazet@google.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "dd07df9ff3d148aee87fcbab99ff14f0727752f4", "tree": "1fee36d0ae7dd14fce4ce37e63c845dc6e2db806", "parents": [ "9e3267cf02c240065fddfbe1a58cdb99d0b00531" ], "author": { "name": "Stanislav Fomichev", "email": "sdf@fomichev.me", "time": "Fri Mar 28 10:42:16 2025 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:53:22 2025 -0700" }, "message": "bnxt_en: bring back rtnl lock in bnxt_shutdown\n\nTaehee reports missing rtnl from bnxt_shutdown path:\n\ninetdev_event (./include/linux/inetdevice.h:256 net/ipv4/devinet.c:1585)\nnotifier_call_chain (kernel/notifier.c:85)\n__dev_close_many (net/core/dev.c:1732 (discriminator 3))\nkernel/locking/mutex.c:713 kernel/locking/mutex.c:732)\ndev_close_many (net/core/dev.c:1786)\nnetif_close (./include/linux/list.h:124 ./include/linux/list.h:215\nbnxt_shutdown (drivers/net/ethernet/broadcom/bnxt/bnxt.c:16707) bnxt_en\npci_device_shutdown (drivers/pci/pci-driver.c:511)\ndevice_shutdown (drivers/base/core.c:4820)\nkernel_restart (kernel/reboot.c:271 kernel/reboot.c:285)\n\nBring back the rtnl lock.\n\nLink: https://lore.kernel.org/netdev/CAMArcTV4P8PFsc6O2tSgzRno050DzafgqkLA2b7t\u003dFv_SY\u003dbrw@mail.gmail.com/\nFixes: 004b5008016a (\"eth: bnxt: remove most dependencies on RTNL\")\nReported-by: Taehee Yoo \u003cap420073@gmail.com\u003e\nSigned-off-by: Stanislav Fomichev \u003csdf@fomichev.me\u003e\nTested-by: Taehee Yoo \u003cap420073@gmail.com\u003e\nTested-by: Breno Leitao \u003cleitao@debian.org\u003e\nLink: https://patch.msgid.link/20250328174216.3513079-1-sdf@fomichev.me\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "9e3267cf02c240065fddfbe1a58cdb99d0b00531", "tree": "809b8cdcf17aa9d4a1a8d6f913a82cde075098a8", "parents": [ "0a541eaff5d6e5a8b40360476ac20defdce5c4b1" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Fri Mar 28 09:47:42 2025 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:53:05 2025 -0700" }, "message": "eth: gve: add missing netdev locks on reset and shutdown paths\n\nAll the misc entry points end up calling into either gve_open()\nor gve_close(), they take rtnl_lock today but since the recent\ninstance locking changes should also take the instance lock.\n\nFound by code inspection and untested.\n\nFixes: cae03e5bdd9e (\"net: hold netdev instance lock during queue operations\")\nAcked-by: Stanislav Fomichev \u003csdf@fomichev.me\u003e\nReviewed-by: Harshitha Ramamurthy \u003chramamurthy@google.com\u003e\nLink: https://patch.msgid.link/20250328164742.1268069-1-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "0a541eaff5d6e5a8b40360476ac20defdce5c4b1", "tree": "80f07126d195069574ea072ab4db7e1c093dd650", "parents": [ "323d6db6dc7decb06f2545efb9496259ddacd4f4", "b44a4c28228fc50b0af05b5d15b44c2172f112a0" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:52:42 2025 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:52:43 2025 -0700" }, "message": "Merge branch \u0027mptcp-misc-fixes-for-6-15-rc0\u0027\n\nMatthieu Baerts says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nmptcp: misc. fixes for 6.15-rc0\n\nHere are 4 unrelated patches:\n\n- Patch 1: fix a NULL pointer when two SYN-ACK for the same request are\n handled in parallel. A fix for up to v5.9.\n\n- Patch 2: selftests: fix check for the wrong FD. A fix for up to v5.17.\n\n- Patch 3: selftests: close all FDs in case of error. A fix for up to\n v5.17.\n\n- Patch 4: selftests: ignore a new generated file. A fix for 6.15-rc0.\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://patch.msgid.link/20250328-net-mptcp-misc-fixes-6-15-v1-0-34161a482a7f@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "b44a4c28228fc50b0af05b5d15b44c2172f112a0", "tree": "80f07126d195069574ea072ab4db7e1c093dd650", "parents": [ "c183165f87a486d5879f782c05a23c179c3794ab" ], "author": { "name": "Matthieu Baerts (NGI0)", "email": "matttbe@kernel.org", "time": "Fri Mar 28 15:27:19 2025 +0100" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:52:39 2025 -0700" }, "message": "selftests: mptcp: ignore mptcp_diag binary\n\nA new binary is now generated by the MPTCP selftests: mptcp_diag.\n\nLike the other binaries from this directory, there is no need to track\nthis in Git, it should then be ignored.\n\nFixes: 00f5e338cf7e (\"selftests: mptcp: Add a tool to get specific msk_info\")\nReviewed-by: Mat Martineau \u003cmartineau@kernel.org\u003e\nSigned-off-by: Matthieu Baerts (NGI0) \u003cmatttbe@kernel.org\u003e\nLink: https://patch.msgid.link/20250328-net-mptcp-misc-fixes-6-15-v1-4-34161a482a7f@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "c183165f87a486d5879f782c05a23c179c3794ab", "tree": "3f683c2c40eb8dcb07eb92a48ec6ae27d4358f11", "parents": [ "7335d4ac812917c16e04958775826d12d481c92d" ], "author": { "name": "Geliang Tang", "email": "tanggeliang@kylinos.cn", "time": "Fri Mar 28 15:27:18 2025 +0100" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:52:39 2025 -0700" }, "message": "selftests: mptcp: close fd_in before returning in main_loop\n\nThe file descriptor \u0027fd_in\u0027 is opened when cfg_input is configured, but\nnot closed in main_loop(), this patch fixes it.\n\nFixes: 05be5e273c84 (\"selftests: mptcp: add disconnect tests\")\nCc: stable@vger.kernel.org\nCo-developed-by: Cong Liu \u003cliucong2@kylinos.cn\u003e\nSigned-off-by: Cong Liu \u003cliucong2@kylinos.cn\u003e\nSigned-off-by: Geliang Tang \u003ctanggeliang@kylinos.cn\u003e\nReviewed-by: Matthieu Baerts (NGI0) \u003cmatttbe@kernel.org\u003e\nSigned-off-by: Matthieu Baerts (NGI0) \u003cmatttbe@kernel.org\u003e\nLink: https://patch.msgid.link/20250328-net-mptcp-misc-fixes-6-15-v1-3-34161a482a7f@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "7335d4ac812917c16e04958775826d12d481c92d", "tree": "bf2b89f8dcc42e133afe8e4aed0a9d89d4a67427", "parents": [ "443041deb5ef6a1289a99ed95015ec7442f141dc" ], "author": { "name": "Cong Liu", "email": "liucong2@kylinos.cn", "time": "Fri Mar 28 15:27:17 2025 +0100" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:52:39 2025 -0700" }, "message": "selftests: mptcp: fix incorrect fd checks in main_loop\n\nFix a bug where the code was checking the wrong file descriptors\nwhen opening the input files. The code was checking \u0027fd\u0027 instead\nof \u0027fd_in\u0027, which could lead to incorrect error handling.\n\nFixes: 05be5e273c84 (\"selftests: mptcp: add disconnect tests\")\nCc: stable@vger.kernel.org\nFixes: ca7ae8916043 (\"selftests: mptcp: mptfo Initiator/Listener\")\nCo-developed-by: Geliang Tang \u003cgeliang@kernel.org\u003e\nSigned-off-by: Geliang Tang \u003cgeliang@kernel.org\u003e\nSigned-off-by: Cong Liu \u003cliucong2@kylinos.cn\u003e\nReviewed-by: Matthieu Baerts (NGI0) \u003cmatttbe@kernel.org\u003e\nSigned-off-by: Matthieu Baerts (NGI0) \u003cmatttbe@kernel.org\u003e\nLink: https://patch.msgid.link/20250328-net-mptcp-misc-fixes-6-15-v1-2-34161a482a7f@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "443041deb5ef6a1289a99ed95015ec7442f141dc", "tree": "d71bff0c5a4c46632b46b121f2198aba0939b494", "parents": [ "323d6db6dc7decb06f2545efb9496259ddacd4f4" ], "author": { "name": "Gang Yan", "email": "yangang@kylinos.cn", "time": "Fri Mar 28 15:27:16 2025 +0100" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:52:39 2025 -0700" }, "message": "mptcp: fix NULL pointer in can_accept_new_subflow\n\nWhen testing valkey benchmark tool with MPTCP, the kernel panics in\n\u0027mptcp_can_accept_new_subflow\u0027 because subflow_req-\u003emsk is NULL.\n\nCall trace:\n\n mptcp_can_accept_new_subflow (./net/mptcp/subflow.c:63 (discriminator 4)) (P)\n subflow_syn_recv_sock (./net/mptcp/subflow.c:854)\n tcp_check_req (./net/ipv4/tcp_minisocks.c:863)\n tcp_v4_rcv (./net/ipv4/tcp_ipv4.c:2268)\n ip_protocol_deliver_rcu (./net/ipv4/ip_input.c:207)\n ip_local_deliver_finish (./net/ipv4/ip_input.c:234)\n ip_local_deliver (./net/ipv4/ip_input.c:254)\n ip_rcv_finish (./net/ipv4/ip_input.c:449)\n ...\n\nAccording to the debug log, the same req received two SYN-ACK in a very\nshort time, very likely because the client retransmits the syn ack due\nto multiple reasons.\n\nEven if the packets are transmitted with a relevant time interval, they\ncan be processed by the server on different CPUs concurrently). The\n\u0027subflow_req-\u003emsk\u0027 ownership is transferred to the subflow the first,\nand there will be a risk of a null pointer dereference here.\n\nThis patch fixes this issue by moving the \u0027subflow_req-\u003emsk\u0027 under the\n`own_req \u003d\u003d true` conditional.\n\nNote that the !msk check in subflow_hmac_valid() can be dropped, because\nthe same check already exists under the own_req mpj branch where the\ncode has been moved to.\n\nFixes: 9466a1ccebbe (\"mptcp: enable JOIN requests even if cookies are in use\")\nCc: stable@vger.kernel.org\nSuggested-by: Paolo Abeni \u003cpabeni@redhat.com\u003e\nSigned-off-by: Gang Yan \u003cyangang@kylinos.cn\u003e\nReviewed-by: Matthieu Baerts (NGI0) \u003cmatttbe@kernel.org\u003e\nSigned-off-by: Matthieu Baerts (NGI0) \u003cmatttbe@kernel.org\u003e\nLink: https://patch.msgid.link/20250328-net-mptcp-misc-fixes-6-15-v1-1-34161a482a7f@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "323d6db6dc7decb06f2545efb9496259ddacd4f4", "tree": "9789c69949f172e932c6fd9e2fb3d7473d99deeb", "parents": [ "0fdba88a211508984eb5df62008c29688692b134" ], "author": { "name": "Geetha sowjanya", "email": "gakula@marvell.com", "time": "Thu Mar 27 15:10:54 2025 +0530" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:52:07 2025 -0700" }, "message": "octeontx2-af: Free NIX_AF_INT_VEC_GEN irq\n\nDue to the incorrect initial vector number in\nrvu_nix_unregister_interrupts(), NIX_AF_INT_VEC_GEN is not\ngeeting free. Fix the vector number to include NIX_AF_INT_VEC_GEN\nirq.\n\nFixes: 5ed66306eab6 (\"octeontx2-af: Add devlink health reporters for NIX\")\nSigned-off-by: Geetha sowjanya \u003cgakula@marvell.com\u003e\nReviewed-by: Simon Horman \u003chorms@kernel.org\u003e\nLink: https://patch.msgid.link/20250327094054.2312-1-gakula@marvell.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "0fdba88a211508984eb5df62008c29688692b134", "tree": "0314b128685fa58ea98ef69880107caf6b526baf", "parents": [ "42f342387841891bbbd15e25d33eb510a0cf7a9a" ], "author": { "name": "Geetha sowjanya", "email": "gakula@marvell.com", "time": "Thu Mar 27 14:44:41 2025 +0530" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:52:02 2025 -0700" }, "message": "octeontx2-af: Fix mbox INTR handler when num VFs \u003e 64\n\nWhen number of RVU VFs \u003e 64, the vfs value passed to \"rvu_queue_work\"\nfunction is incorrect. Due to which mbox workqueue entries for\nVFs 0 to 63 never gets added to workqueue.\n\nFixes: 9bdc47a6e328 (\"octeontx2-af: Mbox communication support btw AF and it\u0027s VFs\")\nSigned-off-by: Geetha sowjanya \u003cgakula@marvell.com\u003e\nReviewed-by: Simon Horman \u003chorms@kernel.org\u003e\nLink: https://patch.msgid.link/20250327091441.1284-1-gakula@marvell.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "42f342387841891bbbd15e25d33eb510a0cf7a9a", "tree": "7e2f5ac3ca3a21fb6e478630c0453a46b9ac7c95", "parents": [ "5c6052802dfbae291ce9032636fb2f7bd0c705e3" ], "author": { "name": "Taehee Yoo", "email": "ap420073@gmail.com", "time": "Fri Mar 28 06:22:37 2025 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:44:49 2025 -0700" }, "message": "net: fix use-after-free in the netdev_nl_sock_priv_destroy()\n\nIn the netdev_nl_sock_priv_destroy(), an instance lock is acquired\nbefore calling net_devmem_unbind_dmabuf(), then releasing an instance\nlock(netdev_unlock(binding-\u003edev)).\nHowever, a binding is freed in the net_devmem_unbind_dmabuf().\nSo using a binding after net_devmem_unbind_dmabuf() occurs UAF.\nTo fix this UAF, it needs to use temporary variable.\n\nFixes: ba6f418fbf64 (\"net: bubble up taking netdev instance lock to callers of net_devmem_unbind_dmabuf()\")\nSigned-off-by: Taehee Yoo \u003cap420073@gmail.com\u003e\nReviewed-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\nReviewed-by: Mina Almasry \u003calmasrymina@google.com\u003e\nReviewed-by: Xuan Zhuo \u003cxuanzhuo@linux.alibaba.com\u003e\nReviewed-by: Simon Horman \u003chorms@kernel.org\u003e\nLink: https://patch.msgid.link/20250328062237.3746875-1-ap420073@gmail.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "5c6052802dfbae291ce9032636fb2f7bd0c705e3", "tree": "cd0ee3348b7d167cc4ee768740349798136479e3", "parents": [ "7220e8f4d4eec0b2f682eef45e2d36c092738413", "88dec030dfcd72fa4322181eb64db06c514f33b1" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:44:28 2025 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:44:29 2025 -0700" }, "message": "Merge branch \u0027selftests-drv-net-replace-the-rpath-helper-with-path-objects\u0027\n\nJakub Kicinski says:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nselftests: drv-net: replace the rpath helper with Path objects\n\nTrying to change the env.rpath() helper during the development\ncycle was causing a lot of conflicts between net and net-next.\nLet\u0027s get it converted now that the trees are converged.\n\nv2: https://lore.kernel.org/20250306171158.1836674-1-kuba@kernel.org\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\n\nLink: https://patch.msgid.link/20250327222315.1098596-1-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "88dec030dfcd72fa4322181eb64db06c514f33b1", "tree": "cd0ee3348b7d167cc4ee768740349798136479e3", "parents": [ "c231e12ecd45fcb34ff3b52d6557d614ba49b699" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Thu Mar 27 15:23:15 2025 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:44:26 2025 -0700" }, "message": "selftests: net: use Path helpers in ping\n\nNow that net and net-next have converged we can use the Path\nhelpers in the ping test without conflicts.\n\nReviewed-by: Willem de Bruijn \u003cwillemb@google.com\u003e\nLink: https://patch.msgid.link/20250327222315.1098596-4-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "c231e12ecd45fcb34ff3b52d6557d614ba49b699", "tree": "858c4a2a64c172ea9d8e488dfeb0740ed1e2498e", "parents": [ "e514d77334a63f1dcb9a3b47d5aee8f51d66cb1d" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Thu Mar 27 15:23:14 2025 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:44:25 2025 -0700" }, "message": "selftests: net: use the dummy bpf from net/lib\n\nCommit 29b036be1b0b (\"selftests: drv-net: test XDP, HDS auto and\nthe ioctl path\") added an sample XDP_PASS prog in net/lib, so\nthat we can reuse it in various sub-directories. Delete the old\nsample and use the one from the lib in existing tests.\n\nAcked-by: Stanislav Fomichev \u003csdf@fomichev.me\u003e\nReviewed-by: Willem de Bruijn \u003cwillemb@google.com\u003e\nLink: https://patch.msgid.link/20250327222315.1098596-3-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "e514d77334a63f1dcb9a3b47d5aee8f51d66cb1d", "tree": "4cba27bfdfddac445d386ca459df3c039788252b", "parents": [ "7220e8f4d4eec0b2f682eef45e2d36c092738413" ], "author": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Thu Mar 27 15:23:13 2025 -0700" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:44:25 2025 -0700" }, "message": "selftests: drv-net: replace the rpath helper with Path objects\n\nThe single letter + \"path\" helpers do not have many fans (see Link).\nUse a Path object with a better name. test_dir is the replacement\nfor rpath(), net_lib_dir is a new path of the $ksft/net/lib directory.\n\nThe Path() class overloads the \"/\" operator and can be cast to string\nautomatically, so to get a path to a file tests can do:\n\n path \u003d env.test_dir / \"binary\"\n\nLink: https://lore.kernel.org/CA+FuTSemTNVZ5MxXkq8T9P\u003dDYm\u003dnSXcJnL7CJBPZNAT_9UFisQ@mail.gmail.com\nReviewed-by: Willem de Bruijn \u003cwillemb@google.com\u003e\nLink: https://patch.msgid.link/20250327222315.1098596-2-kuba@kernel.org\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "7220e8f4d4eec0b2f682eef45e2d36c092738413", "tree": "156d9ee5ba924b78d750cb9f6dffa0c82d332422", "parents": [ "4f1eaabb4b66a1f7473f584e14e15b2ac19dfaf3" ], "author": { "name": "Eric Dumazet", "email": "edumazet@google.com", "time": "Thu Mar 27 14:44:39 2025 +0000" }, "committer": { "name": "Jakub Kicinski", "email": "kuba@kernel.org", "time": "Mon Mar 31 16:43:55 2025 -0700" }, "message": "net: lapbether: use netdev_lockdep_set_classes() helper\n\ndrivers/net/wan/lapbether.c uses stacked devices.\nLike similar drivers, it must use netdev_lockdep_set_classes()\nto avoid LOCKDEP splats.\n\nThis is similar to commit 9bfc9d65a1dc (\"hamradio:\nuse netdev_lockdep_set_classes() helper\")\n\nFixes: 7e4d784f5810 (\"net: hold netdev instance lock during rtnetlink operations\")\nReported-by: syzbot+377b71db585c9c705f8e@syzkaller.appspotmail.com\nCloses: https://lore.kernel.org/lkml/67cd611c.050a0220.14db68.0073.GAE@google.com/T/#u\nSigned-off-by: Eric Dumazet \u003cedumazet@google.com\u003e\nAcked-by: Stanislav Fomichev \u003csdf@fomichev.me\u003e\nLink: https://patch.msgid.link/20250327144439.2463509-1-edumazet@google.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\n" }, { "commit": "1e7857b28020ba57ca7fdafae7ac855ba326c697", "tree": "fcd9756e7e36b116acdd77c158e982c628c96bd6", "parents": [ "46d29f23a7213d3ce3925725057344a28f7de2b0" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 14:19:55 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 14:19:55 2025 -0700" }, "message": "x86: don\u0027t re-generate cpufeaturemasks.h so eagerly\n\nIt turns out the code to generate the x86 cpufeaturemasks.h header was\nway too aggressive, and would re-generate it whenever the timestamp on\nthe kernel config file changed.\n\nNow, the regular \u0027make *config\u0027 tools are fairly careful to not rewrite\nthe kernel config file unless the contents change, but other usecases\naren\u0027t that careful.\n\nMichael Kelley reports that \u0027make-kpkg\u0027 ends up doing \"make syncconfig\"\nmultiple times in prepping to build, and will modify the config file in\nthe process (and then modify it back, but by then the timestamps have\nchanged).\n\nJakub Kicinski reports that the netdev CI does something similar in how\nit generates the config file in multiple steps.\n\nIn both cases, the config file timestamp updates then cause the\ncpufeaturemasks.h file to be regenerated, and that in turn then causes\nlots of unnecessary rebuilds due to all the normal dependencies.\n\nFix it by using our \u0027filechk\u0027 infrastructure in the Makefile to generate\nthe header file. That will only write a new version of the file if the\ncontents of the file have actually changed.\n\nFixes: 841326332bcb (\"x86/cpufeatures: Generate the \u003casm/cpufeaturemasks.h\u003e header based on build config\")\nReported-by: Michael Kelley \u003cmhklinux@outlook.com\u003e\nReported-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\nLink: https://lore.kernel.org/all/SN6PR02MB415756D1829740F6E8AC11D1D4D82@SN6PR02MB4157.namprd02.prod.outlook.com/\nLink: https://lore.kernel.org/all/20250328162311.08134fa6@kernel.org/\nCc: Peter Anvin \u003chpa@zytor.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n" }, { "commit": "46d29f23a7213d3ce3925725057344a28f7de2b0", "tree": "b8da87706b76c1fa84239c7c74f4eeb2cf4974ea", "parents": [ "609706855d90bcab6080ba2cd030b9af322a1f0c", "028a58ec154257e618c27fb0eba8d9e30379bc3d" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 13:37:22 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 13:37:22 2025 -0700" }, "message": "Merge tag \u0027trace-ringbuffer-v6.15-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace\n\nPull ring-buffer updates from Steven Rostedt:\n\n - Restructure the persistent memory to have a \"scratch\" area\n\n Instead of hard coding the KASLR offset in the persistent memory by\n the ring buffer, push that work up to the callers of the persistent\n memory as they are the ones that need this information. The offsets\n and such is not important to the ring buffer logic and it should not\n be part of that.\n\n A scratch pad is now created when the caller allocates a ring buffer\n from persistent memory by stating how much memory it needs to save.\n\n - Allow where modules are loaded to be saved in the new scratch pad\n\n Save the addresses of modules when they are loaded into the\n persistent memory scratch pad.\n\n - A new module_for_each_mod() helper function was created\n\n With the acknowledgement of the module maintainers a new module\n helper function was created to iterate over all the currently loaded\n modules. This has a callback to be called for each module. This is\n needed for when tracing is started in the persistent buffer and the\n currently loaded modules need to be saved in the scratch area.\n\n - Expose the last boot information where the kernel and modules were\n loaded\n\n The last_boot_info file is updated to print out the addresses of\n where the kernel \"_text\" location was loaded from a previous boot, as\n well as where the modules are loaded. If the buffer is recording the\n current boot, it only prints \"# Current\" so that it does not expose\n the KASLR offset of the currently running kernel.\n\n - Allow the persistent ring buffer to be released (freed)\n\n To have this in production environments, where the kernel command\n line can not be changed easily, the ring buffer needs to be freed\n when it is not going to be used. The memory for the buffer will\n always be allocated at boot up, but if the system isn\u0027t going to\n enable tracing, the memory needs to be freed. Allow it to be freed\n and added back to the kernel memory pool.\n\n - Allow stack traces to print the function names in the persistent\n buffer\n\n Now that the modules are saved in the persistent ring buffer, if the\n same modules are loaded, the printing of the function names will\n examine the saved modules. If the module is found in the scratch area\n and is also loaded, then it will do the offset shift and use kallsyms\n to display the function name. If the address is not found, it simply\n displays the address from the previous boot in hex.\n\n* tag \u0027trace-ringbuffer-v6.15-2\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:\n tracing: Use _text and the kernel offset in last_boot_info\n tracing: Show last module text symbols in the stacktrace\n ring-buffer: Remove the unused variable bmeta\n tracing: Skip update_last_data() if cleared and remove active check for save_mod()\n tracing: Initialize scratch_size to zero to prevent UB\n tracing: Fix a compilation error without CONFIG_MODULES\n tracing: Freeable reserved ring buffer\n mm/memblock: Add reserved memory release function\n tracing: Update modules to persistent instances when loaded\n tracing: Show module names and addresses of last boot\n tracing: Have persistent trace instances save module addresses\n module: Add module_for_each_mod() function\n tracing: Have persistent trace instances save KASLR offset\n ring-buffer: Add ring_buffer_meta_scratch()\n ring-buffer: Add buffer meta data for persistent ring buffer\n ring-buffer: Use kaslr address instead of text delta\n ring-buffer: Fix bytes_dropped calculation issue\n" }, { "commit": "609706855d90bcab6080ba2cd030b9af322a1f0c", "tree": "14f3e806f0bda5c07347e71706993b88f30c8902", "parents": [ "802f0d58d52e8e34e08718479475ccdff0caffa0", "4bb5d82b66002b770f8917d68ab4fbefcb7f5f9b" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 09:56:08 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 09:56:08 2025 -0700" }, "message": "Merge tag \u0027trace-latency-v6.15-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace\n\nPull tracing documentation fix from Steven Rostedt:\n \"Documentation fix for runtime verifier\n\n The runtime verifier documents that were created were not referenced\n in the indices, which caused warning when building the documentation\n tree. Those documents are now added to the rv indices\"\n\n* tag \u0027trace-latency-v6.15-3\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:\n Documentation/rv: Add sched pages to the indices\n" }, { "commit": "802f0d58d52e8e34e08718479475ccdff0caffa0", "tree": "305f3be98d12b0c6881a6c59eb92e795e6088e51", "parents": [ "4e82c87058f45e79eeaa4d5bcc3b38dd3dce7209", "35d13f841a3d8159ef20d5e32a9ed3faa27875bc" ], "author": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 08:52:33 2025 -0700" }, "committer": { "name": "Linus Torvalds", "email": "torvalds@linux-foundation.org", "time": "Mon Mar 31 08:52:33 2025 -0700" }, "message": "Merge tag \u0027perf-tools-for-v6.15-2025-03-27\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools\n\nPull perf tools updates from Namhyung Kim:\n \"perf record:\n\n - Introduce latency profiling using scheduler information.\n\n The latency profiling is to show impacts on wall-time rather than\n cpu-time. By tracking context switches, it can weight samples and\n find which part of the code contributed more to the execution\n latency.\n\n The value (period) of the sample is weighted by dividing it by the\n number of parallel execution at the moment. The parallelism is\n tracked in perf report with sched-switch records. This will reduce\n the portion that are run in parallel and in turn increase the\n portion of serial executions.\n\n For now, it\u0027s limited to profile processes, IOW system-wide\n profiling is not supported. You can add --latency option to enable\n this.\n\n $ perf record --latency -- make -C tools/perf\n\n I\u0027ve run the above command for perf build which adds -j option to\n make with the number of CPUs in the system internally. Normally\n it\u0027d show something like below:\n\n $ perf report -F overhead,comm\n ...\n #\n # Overhead Command\n # ........ ...............\n #\n 78.97% cc1\n 6.54% python3\n 4.21% shellcheck\n 3.28% ld\n 1.80% as\n 1.37% cc1plus\n 0.80% sh\n 0.62% clang\n 0.56% gcc\n 0.44% perl\n 0.39% make\n \t ...\n\n The cc1 takes around 80% of the overhead as it\u0027s the actual\n compiler. However it runs in parallel so its contribution to\n latency may be less than that. Now, perf report will show both\n overhead and latency (if --latency was given at record time) like\n below:\n\n $ perf report -s comm\n ...\n #\n # Overhead Latency Command\n # ........ ........ ...............\n #\n 78.97% 48.66% cc1\n 6.54% 25.68% python3\n 4.21% 0.39% shellcheck\n 3.28% 13.70% ld\n 1.80% 2.56% as\n 1.37% 3.08% cc1plus\n 0.80% 0.98% sh\n 0.62% 0.61% clang\n 0.56% 0.33% gcc\n 0.44% 1.71% perl\n 0.39% 0.83% make\n \t ...\n\n You can see latency of cc1 goes down to around 50% and python3 and\n ld contribute a lot more than their overhead. You can use --latency\n option in perf report to get the same result but ordered by\n latency.\n\n $ perf report --latency -s comm\n\n perf report:\n\n - As a side effect of the latency profiling work, it adds a new\n output field \u0027latency\u0027 and a sort key \u0027parallelism\u0027. The below is a\n result from my system with 64 CPUs. The build was well-parallelized\n but contained some serial portions.\n\n $ perf report -s parallelism\n ...\n #\n # Overhead Latency Parallelism\n # ........ ........ ...........\n #\n 16.95% 1.54% 62\n 13.38% 1.24% 61\n 12.50% 70.47% 1\n 11.81% 1.06% 63\n 7.59% 0.71% 60\n 4.33% 12.20% 2\n 3.41% 0.33% 59\n 2.05% 0.18% 64\n 1.75% 1.09% 9\n 1.64% 1.85% 5\n ...\n\n - Support Feodra mini-debuginfo which is a LZMA compressed symbol\n table inside \".gnu_debugdata\" ELF section.\n\n perf annotate:\n\n - Add --code-with-type option to enable data-type profiling with the\n usual annotate output.\n\n Instead of focusing on data structure, it shows code annotation\n together with data type it accesses in case the instruction refers\n to a memory location (and it was able to resolve the target data\n type). Currently it only works with --stdio.\n\n $ perf annotate --stdio --code-with-type\n ...\n Percent | Source code \u0026 Disassembly of vmlinux for cpu/mem-loads,ldlat\u003d30/pp (18 samples, percent: local period)\n ----------------------------------------------------------------------------------------------------------------------\n : 0 0xffffffff81050610 \u003c__fdget\u003e:\n 0.00 : ffffffff81050610: callq 0xffffffff81c01b80 \u003c__fentry__\u003e # data-type: (stack operation)\n 0.00 : ffffffff81050615: pushq %rbp # data-type: (stack operation)\n 0.00 : ffffffff81050616: movq %rsp, %rbp\n 0.00 : ffffffff81050619: pushq %r15 # data-type: (stack operation)\n 0.00 : ffffffff8105061b: pushq %r14 # data-type: (stack operation)\n 0.00 : ffffffff8105061d: pushq %rbx # data-type: (stack operation)\n 0.00 : ffffffff8105061e: subq $0x10, %rsp\n 0.00 : ffffffff81050622: movl %edi, %ebx\n 0.00 : ffffffff81050624: movq %gs:0x7efc4814(%rip), %rax # 0x14e40 \u003ccurrent_task\u003e # data-type: struct task_struct* +0\n 0.00 : ffffffff8105062c: movq 0x8d0(%rax), %r14 # data-type: struct task_struct +0x8d0 (files)\n 0.00 : ffffffff81050633: movl (%r14), %eax # data-type: struct files_struct +0 (count.counter)\n 0.00 : ffffffff81050636: cmpl $0x1, %eax\n 0.00 : ffffffff81050639: je 0xffffffff810506a9 \u003c__fdget+0x99\u003e\n 0.00 : ffffffff8105063b: movq 0x20(%r14), %rcx # data-type: struct files_struct +0x20 (fdt)\n 0.00 : ffffffff8105063f: movl (%rcx), %eax # data-type: struct fdtable +0 (max_fds)\n 0.00 : ffffffff81050641: cmpl %ebx, %eax\n 0.00 : ffffffff81050643: jbe 0xffffffff810506ef \u003c__fdget+0xdf\u003e\n 0.00 : ffffffff81050649: movl %ebx, %r15d\n 5.56 : ffffffff8105064c: movq 0x8(%rcx), %rdx # data-type: struct fdtable +0x8 (fd)\n \t...\n\n The \"# data-type:\" part was added with this change. The first few\n entries are not very interesting. But later you can it accesses a\n couple of fields in the task_struct, files_struct and fdtable.\n\n perf trace:\n\n - Support syscall tracing for different ABI. For example it can trace\n system calls for 32-bit applications on 64-bit kernel\n transparently.\n\n - Add --summary-mode\u003dtotal option to show global syscall summary. The\n default is \u0027thread\u0027 to show per-thread syscall summary.\n\n Python support:\n\n - Add more interfaces to \u0027perf\u0027 module to parse events, and config,\n enable or disable the event list properly so that it can implement\n basic functionalities purely in Python. There is an example code\n for these new interfaces in python/tracepoint.py.\n\n - Add mypy and pylint support to enable build time checking. Fix some\n code based on the findings from these tools.\n\n Internals:\n\n - Introduce io_dir__readdir() API to make directory traveral (usually\n for proc or sysfs) efficient with less memory footprint.\n\n JSON vendor events:\n\n - Add events and metrics for ARM Neoverse N3 and V3\n\n - Update events and metrics on various Intel CPUs\n\n - Add/update events for a number of SiFive processors\"\n\n* tag \u0027perf-tools-for-v6.15-2025-03-27\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools: (229 commits)\n perf bpf-filter: Fix a parsing error with comma\n perf report: Fix a memory leak for perf_env on AMD\n perf trace: Fix wrong size to bpf_map__update_elem call\n perf tools: annotate asm_pure_loop.S\n perf python: Fix setup.py mypy errors\n perf test: Address attr.py mypy error\n perf build: Add pylint build tests\n perf build: Add mypy build tests\n perf build: Rename TEST_LOGS to SHELL_TEST_LOGS\n tools/build: Don\u0027t pass test log files to linker\n perf bench sched pipe: fix enforced blocking reads in worker_thread\n perf tools: Fix is_compat_mode build break in ppc64\n perf build: filter all combinations of -flto for libperl\n perf vendor events arm64 AmpereOneX: Fix frontend_bound calculation\n perf vendor events arm64: AmpereOne/AmpereOneX: Mark LD_RETIRED impacted by errata\n perf trace: Fix evlist memory leak\n perf trace: Fix BTF memory leak\n perf trace: Make syscall table stable\n perf syscalltbl: Mask off ABI type for MIPS system calls\n perf build: Remove Makefile.syscalls\n ...\n" }, { "commit": "ebca08fef88febdb0a898cefa7c99b9e25b3a984", "tree": "9582e46d0616f002cce6297679158cc5ad930ddb", "parents": [ "064009ebc171dc11aaa86c0bdce939640e1d033d" ], "author": { "name": "Gergo Koteles", "email": "soyer@irl.hu", "time": "Fri Mar 28 22:08:56 2025 +0100" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Mon Mar 31 17:02:14 2025 +0200" }, "message": "ACPI: video: Handle fetching EDID as ACPI_TYPE_PACKAGE\n\nThe _DDC method should return a buffer, or an integer in case of an error.\nBut some Lenovo laptops incorrectly return EDID as buffer in ACPI package.\n\nCalling _DDC generates this ACPI Warning:\nACPI Warning: \\_SB.PCI0.GP17.VGA.LCD._DDC: Return type mismatch - \\\nfound Package, expected Integer/Buffer (20240827/nspredef-254)\n\nUse the first element of the package to get the EDID buffer.\n\nThe DSDT:\n\nName (AUOP, Package (0x01)\n{\n\tBuffer (0x80)\n\t{\n\t...\n\t}\n})\n\n...\n\nMethod (_DDC, 1, NotSerialized) // _DDC: Display Data Current\n{\n\tIf ((PAID \u003d\u003d AUID))\n {\n\t\tReturn (AUOP) /* \\_SB_.PCI0.GP17.VGA_.LCD_.AUOP */\n\t}\n\tElseIf ((PAID \u003d\u003d IVID))\n\t{\n\t\tReturn (IVOP) /* \\_SB_.PCI0.GP17.VGA_.LCD_.IVOP */\n\t}\n\tElseIf ((PAID \u003d\u003d BOID))\n\t{\n\t\tReturn (BOEP) /* \\_SB_.PCI0.GP17.VGA_.LCD_.BOEP */\n\t}\n\tElseIf ((PAID \u003d\u003d SAID))\n\t{\n\t\tReturn (SUNG) /* \\_SB_.PCI0.GP17.VGA_.LCD_.SUNG */\n\t}\n\n\tReturn (Zero)\n}\n\nLink: https://uefi.org/htmlspecs/ACPI_Spec_6_4_html/Apx_B_Video_Extensions/output-device-specific-methods.html#ddc-return-the-edid-for-this-device\nCc: stable@vger.kernel.org\nFixes: c6a837088bed (\"drm/amd/display: Fetch the EDID from _DDC if available for eDP\")\nCloses: https://gitlab.freedesktop.org/drm/amd/-/issues/4085\nSigned-off-by: Gergo Koteles \u003csoyer@irl.hu\u003e\nReviewed-by: Hans de Goede \u003chdegoede@redhat.com\u003e\nReviewed-by: Mario Limonciello \u003cmario.limonciello@amd.com\u003e\nLink: https://patch.msgid.link/61c3df83ab73aba0bc7a941a443cd7faf4cf7fb0.1743195250.git.soyer@irl.hu\nSigned-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n" }, { "commit": "2d066800a4276340a97acc75c148892eb6f8781a", "tree": "5f2e3f82ea17440b1587dde8cd75f6e71ae65022", "parents": [ "1dfe2a220e9cd85861a853b00d8620222b960c1f" ], "author": { "name": "Joanne Koong", "email": "joannelkoong@gmail.com", "time": "Mon Mar 17 17:41:52 2025 -0700" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@redhat.com", "time": "Mon Mar 31 14:59:27 2025 +0200" }, "message": "fuse: remove unneeded atomic set in uring creation\n\nWhen the ring is allocated, it is kzalloc-ed. ring-\u003equeue_refs will\nalready be initialized to 0 by default. It does not need to be\natomically set to 0.\n\nSigned-off-by: Joanne Koong \u003cjoannelkoong@gmail.com\u003e\nReviewed-by: Bernd Schubert \u003cbschubert@ddn.com\u003e\nSigned-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n" }, { "commit": "1dfe2a220e9cd85861a853b00d8620222b960c1f", "tree": "f9bb92e91d7710034478dcbd33affb2f2588b4db", "parents": [ "27992ef80770d61a57f6c3a551735b08cefdffa3" ], "author": { "name": "Joanne Koong", "email": "joannelkoong@gmail.com", "time": "Mon Mar 17 17:30:28 2025 -0700" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@redhat.com", "time": "Mon Mar 31 14:59:27 2025 +0200" }, "message": "fuse: fix uring race condition for null dereference of fc\n\nThere is a race condition leading to a kernel crash from a null\ndereference when attemping to access fc-\u003elock in\nfuse_uring_create_queue(). fc may be NULL in the case where another\nthread is creating the uring in fuse_uring_create() and has set\nfc-\u003ering but has not yet set ring-\u003efc when fuse_uring_create_queue()\nreads ring-\u003efc. There is another race condition as well where in\nfuse_uring_register(), ring-\u003enr_queues may still be 0 and not yet set\nto the new value when we compare qid against it.\n\nThis fix sets fc-\u003ering only after ring-\u003efc and ring-\u003enr_queues have been\nset, which guarantees now that ring-\u003efc is a proper pointer when any\nqueues are created and ring-\u003enr_queues reflects the right number of\nqueues if ring is not NULL. We must use smp_store_release() and\nsmp_load_acquire() semantics to ensure the ordering will remain correct\nwhere fc-\u003ering is assigned only after ring-\u003efc and ring-\u003enr_queues have\nbeen assigned.\n\nSigned-off-by: Joanne Koong \u003cjoannelkoong@gmail.com\u003e\nFixes: 24fe962c86f5 (\"fuse: {io-uring} Handle SQEs - register commands\")\nReviewed-by: Bernd Schubert \u003cbschubert@ddn.com\u003e\nSigned-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n" }, { "commit": "27992ef80770d61a57f6c3a551735b08cefdffa3", "tree": "50e66236fc0e87e387dc44ba08114a0c266e0bf8", "parents": [ "2412085da370836945c2daa61c5cee38dd979e0d" ], "author": { "name": "Bernd Schubert", "email": "bschubert@ddn.com", "time": "Mon Dec 16 22:14:07 2024 +0100" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@redhat.com", "time": "Mon Mar 31 14:59:27 2025 +0200" }, "message": "fuse: Increase FUSE_NAME_MAX to PATH_MAX\n\nOur file system has a translation capability for S3-to-posix.\nThe current value of 1kiB is enough to cover S3 keys, but\ndoes not allow encoding of %xx escape characters.\nThe limit is increased to (PATH_MAX - 1), as we need\n3 x 1024 and that is close to PATH_MAX (4kB) already.\n-1 is used as the terminating null is not included in the\nlength calculation.\n\nTesting large file names was hard with libfuse/example file systems,\nso I created a new memfs that does not have a 255 file name length\nlimitation.\nhttps://github.com/libfuse/libfuse/pull/1077\n\nThe connection is initialized with FUSE_NAME_LOW_MAX, which\nis set to the previous value of FUSE_NAME_MAX of 1024. With\nFUSE_MIN_READ_BUFFER of 8192 that is enough for two file names\n+ fuse headers.\nWhen FUSE_INIT reply sets max_pages to a value \u003e 1 we know\nthat fuse daemon supports request buffers of at least 2 pages\n(+ header) and can therefore hold 2 x PATH_MAX file names - operations\nlike rename or link that need two file names are no issue then.\n\nSigned-off-by: Bernd Schubert \u003cbschubert@ddn.com\u003e\nSigned-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n" }, { "commit": "2412085da370836945c2daa61c5cee38dd979e0d", "tree": "e70e6d953cf8757d3935c17ce047d3032335167e", "parents": [ "9b17cb59a7db983a967c3658fe9a2f250f588bbd" ], "author": { "name": "Bernd Schubert", "email": "bschubert@ddn.com", "time": "Mon Dec 16 22:14:06 2024 +0100" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@redhat.com", "time": "Mon Mar 31 14:59:27 2025 +0200" }, "message": "fuse: Allocate only namelen buf memory in fuse_notify_\n\nfuse_notify_inval_entry and fuse_notify_delete were using fixed allocations\nof FUSE_NAME_MAX to hold the file name. Often that large buffers are not\nneeded as file names might be smaller, so this uses the actual file name\nsize to do the allocation.\n\nSigned-off-by: Bernd Schubert \u003cbschubert@ddn.com\u003e\nReviewed-by: Jingbo Xu \u003cjefflexu@linux.alibaba.com\u003e\nSigned-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n" }, { "commit": "9b17cb59a7db983a967c3658fe9a2f250f588bbd", "tree": "b894295fe2f370c7fe28451f4b826c91f08f18bb", "parents": [ "0f6439f61a6e2ddc92b98362c6d1afc210f56a90" ], "author": { "name": "Joanne Koong", "email": "joannelkoong@gmail.com", "time": "Wed Jan 22 13:55:28 2025 -0800" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@redhat.com", "time": "Mon Mar 31 14:59:27 2025 +0200" }, "message": "fuse: add default_request_timeout and max_request_timeout sysctls\n\nIntroduce two new sysctls, \"default_request_timeout\" and\n\"max_request_timeout\". These control how long (in seconds) a server can\ntake to reply to a request. If the server does not reply by the timeout,\nthen the connection will be aborted. The upper bound on these sysctl\nvalues is 65535.\n\n\"default_request_timeout\" sets the default timeout if no timeout is\nspecified by the fuse server on mount. 0 (default) indicates no default\ntimeout should be enforced. If the server did specify a timeout, then\ndefault_request_timeout will be ignored.\n\n\"max_request_timeout\" sets the max amount of time the server may take to\nreply to a request. 0 (default) indicates no maximum timeout. If\nmax_request_timeout is set and the fuse server attempts to set a\ntimeout greater than max_request_timeout, the system will use\nmax_request_timeout as the timeout. Similarly, if default_request_timeout\nis greater than max_request_timeout, the system will use\nmax_request_timeout as the timeout. If the server does not request a\ntimeout and default_request_timeout is set to 0 but max_request_timeout\nis set, then the timeout will be max_request_timeout.\n\nPlease note that these timeouts are not 100% precise. The request may\ntake roughly an extra FUSE_TIMEOUT_TIMER_FREQ seconds beyond the set max\ntimeout due to how it\u0027s internally implemented.\n\n$ sysctl -a | grep fuse.default_request_timeout\nfs.fuse.default_request_timeout \u003d 0\n\n$ echo 65536 | sudo tee /proc/sys/fs/fuse/default_request_timeout\ntee: /proc/sys/fs/fuse/default_request_timeout: Invalid argument\n\n$ echo 65535 | sudo tee /proc/sys/fs/fuse/default_request_timeout\n65535\n\n$ sysctl -a | grep fuse.default_request_timeout\nfs.fuse.default_request_timeout \u003d 65535\n\n$ echo 0 | sudo tee /proc/sys/fs/fuse/default_request_timeout\n0\n\n$ sysctl -a | grep fuse.default_request_timeout\nfs.fuse.default_request_timeout \u003d 0\n\n[Luis Henriques: Limit the timeout to the range [FUSE_TIMEOUT_TIMER_FREQ,\nfuse_max_req_timeout]]\n\nSigned-off-by: Joanne Koong \u003cjoannelkoong@gmail.com\u003e\nReviewed-by: Bernd Schubert \u003cbschubert@ddn.com\u003e\nReviewed-by: Josef Bacik \u003cjosef@toxicpanda.com\u003e\nReviewed-by: Sergey Senozhatsky \u003csenozhatsky@chromium.org\u003e\nReviewed-by: Luis Henriques \u003cluis@igalia.com\u003e\nReviewed-by: Jeff Layton \u003cjlayton@kernel.org\u003e\nSigned-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n" }, { "commit": "0f6439f61a6e2ddc92b98362c6d1afc210f56a90", "tree": "afefdaeeee1cfddeff0017232c15c40752a76ef5", "parents": [ "eef36cf6a7016cb5353d4b0a9dbdfbd52c4bd973" ], "author": { "name": "Joanne Koong", "email": "joannelkoong@gmail.com", "time": "Wed Jan 22 13:55:27 2025 -0800" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@redhat.com", "time": "Mon Mar 31 14:59:25 2025 +0200" }, "message": "fuse: add kernel-enforced timeout option for requests\n\nThere are situations where fuse servers can become unresponsive or\nstuck, for example if the server is deadlocked. Currently, there\u0027s no\ngood way to detect if a server is stuck and needs to be killed manually.\n\nThis commit adds an option for enforcing a timeout (in seconds) for\nrequests where if the timeout elapses without the server responding to\nthe request, the connection will be automatically aborted.\n\nPlease note that these timeouts are not 100% precise. For example, the\nrequest may take roughly an extra FUSE_TIMEOUT_TIMER_FREQ seconds beyond\nthe requested timeout due to internal implementation, in order to\nmitigate overhead.\n\n[SzM: Bump the API version number]\n\nSigned-off-by: Joanne Koong \u003cjoannelkoong@gmail.com\u003e\nReviewed-by: Jeff Layton \u003cjlayton@kernel.org\u003e\nSigned-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n" }, { "commit": "eef36cf6a7016cb5353d4b0a9dbdfbd52c4bd973", "tree": "754ca2303a400fee1ba602d3d91f1bdf82a1f150", "parents": [ "8344213571b2ac8caf013cfd3b37bc3467c3a893" ], "author": { "name": "Miklos Szeredi", "email": "mszeredi@redhat.com", "time": "Fri Feb 14 11:00:53 2025 +0100" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@redhat.com", "time": "Mon Mar 31 14:55:36 2025 +0200" }, "message": "fuse: optmize missing FUSE_LINK support\n\nIf filesystem doesn\u0027t support FUSE_LINK (i.e. returns -ENOSYS), then\nremember this and next time return immediately, without incurring the\noverhead of a round trip to the server.\n\nSigned-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n" }, { "commit": "8344213571b2ac8caf013cfd3b37bc3467c3a893", "tree": "18ddb52846f42cad4fbf183cde8a953f42a9ac28", "parents": [ "841c7b812c038661e4f659d1b9c9a366c6d24b71" ], "author": { "name": "Matt Johnston", "email": "matt@codeconstruct.com.au", "time": "Fri Feb 14 09:17:53 2025 +0800" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@redhat.com", "time": "Mon Mar 31 14:55:36 2025 +0200" }, "message": "fuse: Return EPERM rather than ENOSYS from link()\n\nlink() is documented to return EPERM when a filesystem doesn\u0027t support\nthe operation, return that instead.\n\nLink: https://github.com/libfuse/libfuse/issues/925\nSigned-off-by: Matt Johnston \u003cmatt@codeconstruct.com.au\u003e\nSigned-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n" }, { "commit": "841c7b812c038661e4f659d1b9c9a366c6d24b71", "tree": "b481dd0f0dcf79abc871a82634730bf11478430d", "parents": [ "09098e62e4be8f0755e58d6078aaf27cbd9a3a8d" ], "author": { "name": "Luis Henriques", "email": "luis@igalia.com", "time": "Fri Feb 07 13:35:02 2025 +0000" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@redhat.com", "time": "Mon Mar 31 14:55:36 2025 +0200" }, "message": "fuse: removed unused function fuse_uring_create() from header\n\nFunction fuse_uring_create() is used only from dev_uring.c and does not\nneed to be exposed in the header file. Furthermore, it has the wrong\nsignature.\n\nWhile there, also remove the \u0027struct fuse_ring\u0027 forward declaration.\n\nSigned-off-by: Luis Henriques \u003cluis@igalia.com\u003e\nReviewed-by: Bernd Schubert \u003cbschubert@ddn.com\u003e\nSigned-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n" }, { "commit": "09098e62e4be8f0755e58d6078aaf27cbd9a3a8d", "tree": "d1c03f07301687a382b7efb94de2160e1f9cf41f", "parents": [ "4701f33a10702d5fc577c32434eb62adde0a1ae1" ], "author": { "name": "Bernd Schubert", "email": "bschubert@ddn.com", "time": "Tue Mar 25 18:29:31 2025 +0100" }, "committer": { "name": "Miklos Szeredi", "email": "mszeredi@redhat.com", "time": "Mon Mar 31 14:53:02 2025 +0200" }, "message": "fuse: {io-uring} Fix a possible req cancellation race\n\ntask-A (application) might be in request_wait_answer and\ntry to remove the request when it has FR_PENDING set.\n\ntask-B (a fuse-server io-uring task) might handle this\nrequest with FUSE_IO_URING_CMD_COMMIT_AND_FETCH, when\nfetching the next request and accessed the req from\nthe pending list in fuse_uring_ent_assign_req().\nThat code path was not protected by fiq-\u003elock and so\nmight race with task-A.\n\nFor scaling reasons we better don\u0027t use fiq-\u003elock, but\nadd a handler to remove canceled requests from the queue.\n\nThis also removes usage of fiq-\u003elock from\nfuse_uring_add_req_to_ring_ent() altogether, as it was\nthere just to protect against this race and incomplete.\n\nAlso added is a comment why FR_PENDING is not cleared.\n\nFixes: c090c8abae4b (\"fuse: Add io-uring sqe commit and fetch support\")\nCc: \u003cstable@vger.kernel.org\u003e # v6.14\nReported-by: Joanne Koong \u003cjoannelkoong@gmail.com\u003e\nCloses: https://lore.kernel.org/all/CAJnrk1ZgHNb78dz-yfNTpxmW7wtT88A\u003dm-zF0ZoLXKLUHRjNTw@mail.gmail.com/\nSigned-off-by: Bernd Schubert \u003cbschubert@ddn.com\u003e\nReviewed-by: Joanne Koong \u003cjoannelkoong@gmail.com\u003e\nSigned-off-by: Miklos Szeredi \u003cmszeredi@redhat.com\u003e\n" }, { "commit": "4f1eaabb4b66a1f7473f584e14e15b2ac19dfaf3", "tree": "4acb5710a30ddf11a585654a0291389e9ff95022", "parents": [ "2ea396448f26d0d7d66224cb56500a6789c7ed07" ], "author": { "name": "Jim Liu", "email": "jim.t90615@gmail.com", "time": "Thu Mar 27 14:29:42 2025 +0800" }, "committer": { "name": "David S. Miller", "email": "davem@davemloft.net", "time": "Mon Mar 31 13:47:41 2025 +0100" }, "message": "net: phy: broadcom: Correct BCM5221 PHY model detection\n\nCorrect detect condition is applied to the entire 5221 family of PHYs.\n\nFixes: 3abbd0699b67 (\"net: phy: broadcom: add support for BCM5221 phy\")\nSigned-off-by: Jim Liu \u003cjim.t90615@gmail.com\u003e\nReviewed-by: Michal Swiatkowski \u003cmichal.swiatkowski@linux.intel.com\u003e\nReviewed-by: Florian Fainelli \u003cflorian.fainelli@broadcom.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n" }, { "commit": "9e9b893404d43894d69a18dd2fc8fcf1c36abb7e", "tree": "376f9595c13f1ab370224e625e1120eaed954934", "parents": [ "21e0ff5b10ec1b61fda435d42db4ba80d0cdfded" ], "author": { "name": "Giovanni Gherdovich", "email": "ggherdovich@suse.cz", "time": "Fri Mar 28 15:30:39 2025 +0100" }, "committer": { "name": "Rafael J. Wysocki", "email": "rafael.j.wysocki@intel.com", "time": "Mon Mar 31 13:59:04 2025 +0200" }, "message": "ACPI: processor: idle: Return an error if both P_LVL{2,3} idle states are invalid\n\nPrior to commit 496121c02127 (\"ACPI: processor: idle: Allow probing on\nplatforms with one ACPI C-state\"), the acpi_idle driver wouldn\u0027t load on\nsystems without a valid C-State at least as deep as C2.\n\nThe behavior was desirable for guests on hypervisors such as VMWare\nESXi, which by default don\u0027t have the _CST ACPI method, and set the C2\nand C3 latencies to 101 and 1001 microseconds respectively via the FADT,\nto signify they\u0027re unsupported.\n\nSince the above change though, these virtualized deployments end up\nloading acpi_idle, and thus entering the default C1 C-State set by\nacpi_processor_get_power_info_default(); this is undesirable for a\nsystem that\u0027s communicating to the OS it doesn\u0027t want C-States (missing\n_CST, and invalid C2/C3 in FADT).\n\nMake acpi_processor_get_power_info_fadt() return -ENODEV in that case,\nso that acpi_processor_get_cstate_info() exits early and doesn\u0027t set\npr-\u003eflags.power \u003d 1.\n\nFixes: 496121c02127 (\"ACPI: processor: idle: Allow probing on platforms with one ACPI C-state\")\nSigned-off-by: Giovanni Gherdovich \u003cggherdovich@suse.cz\u003e\nReviewed-by: Zhang Rui \u003crui.zhang@intel.com\u003e\nLink: https://patch.msgid.link/20250328143040.9348-1-ggherdovich@suse.cz\n[ rjw: Changelog edits ]\nSigned-off-by: Rafael J. Wysocki \u003crafael.j.wysocki@intel.com\u003e\n" }, { "commit": "bd496a44f041da9ef3afe14d1d6193d460424e91", "tree": "cf8fb0b5e587f854d69dc99f104750e8d51316fc", "parents": [ "465e5486aa5e1cdedc910bc3487ca92c5e6d51c4" ], "author": { "name": "Manjunatha Venkatesh", "email": "manjunatha.venkatesh@nxp.com", "time": "Wed Mar 26 18:00:46 2025 +0530" }, "committer": { "name": "Alexandre Belloni", "email": "alexandre.belloni@bootlin.com", "time": "Mon Mar 31 11:44:00 2025 +0200" }, "message": "i3c: Add NULL pointer check in i3c_master_queue_ibi()\n\nThe I3C master driver may receive an IBI from a target device that has not\nbeen probed yet. In such cases, the master calls `i3c_master_queue_ibi()`\nto queue an IBI work task, leading to \"Unable to handle kernel read from\nunreadable memory\" and resulting in a kernel panic.\n\nTypical IBI handling flow:\n1. The I3C master scans target devices and probes their respective drivers.\n2. The target device driver calls `i3c_device_request_ibi()` to enable IBI\n and assigns `dev-\u003eibi \u003d ibi`.\n3. The I3C master receives an IBI from the target device and calls\n `i3c_master_queue_ibi()` to queue the target device driver’s IBI\n handler task.\n\nHowever, since target device events are asynchronous to the I3C probe\nsequence, step 3 may occur before step 2, causing `dev-\u003eibi` to be `NULL`,\nleading to a kernel panic.\n\nAdd a NULL pointer check in `i3c_master_queue_ibi()` to prevent accessing\nan uninitialized `dev-\u003eibi`, ensuring stability.\n\nFixes: 3a379bbcea0af (\"i3c: Add core I3C infrastructure\")\nCc: stable@vger.kernel.org\nLink: https://lore.kernel.org/lkml/Z9gjGYudiYyl3bSe@lizhi-Precision-Tower-5810/\nSigned-off-by: Manjunatha Venkatesh \u003cmanjunatha.venkatesh@nxp.com\u003e\nReviewed-by: Frank Li \u003cFrank.Li@nxp.com\u003e\nLink: https://lore.kernel.org/r/20250326123047.2797946-1-manjunatha.venkatesh@nxp.com\nSigned-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n" }, { "commit": "465e5486aa5e1cdedc910bc3487ca92c5e6d51c4", "tree": "a913ae3ec2cd3d53e3513ea4628da40b1ebe5a00", "parents": [ "e8d2d287e26d9bd9114cf258a123a6b70812442e" ], "author": { "name": "Andy Shevchenko", "email": "andriy.shevchenko@linux.intel.com", "time": "Fri Mar 21 21:30:06 2025 +0200" }, "committer": { "name": "Alexandre Belloni", "email": "alexandre.belloni@bootlin.com", "time": "Mon Mar 31 11:41:48 2025 +0200" }, "message": "i3c: master: Drop duplicate check before calling OF APIs\n\nOF APIs are usually NULL-aware and returns an error in case when\ndevice node is not present or supported. We already have a check\nfor the returned value, no need to check for the parameter.\n\nSigned-off-by: Andy Shevchenko \u003candriy.shevchenko@linux.intel.com\u003e\nReviewed-by: Frank Li \u003cFrank.Li@nxp.com\u003e\nLink: https://lore.kernel.org/r/20250321193044.457649-1-andriy.shevchenko@linux.intel.com\nSigned-off-by: Alexandre Belloni \u003calexandre.belloni@bootlin.com\u003e\n" } ], "next": "c1f4534b213d7be41b5d8b815a42d201a8f2978f" }