From aed06d36ba4e7fe90f2d4a85835cee7c80ea72a7 Mon Sep 17 00:00:00 2001 From: "Dr. David Alan Gilbert" Date: Sat, 22 Feb 2025 01:35:30 +0000 Subject: ceph: Remove osd_client deadcode osd_req_op_extent_osd_data_pagelist() was added in 2013 as part of commit a4ce40a9a7c1 ("libceph: combine initializing and setting osd data") but never used. The last use of osd_req_op_cls_request_data_pagelist() was removed in 2017's commit ecd4a68a26a2 ("rbd: switch rbd_obj_method_sync() to ceph_osdc_call()") Remove them. Signed-off-by: Dr. David Alan Gilbert Reviewed-by: Viacheslav Dubeyko Signed-off-by: Ilya Dryomov --- include/linux/ceph/osd_client.h | 6 ------ 1 file changed, 6 deletions(-) (limited to 'include') diff --git a/include/linux/ceph/osd_client.h b/include/linux/ceph/osd_client.h index d55b30057a45..50b14a5661c7 100644 --- a/include/linux/ceph/osd_client.h +++ b/include/linux/ceph/osd_client.h @@ -490,9 +490,6 @@ extern void osd_req_op_extent_osd_data_pages(struct ceph_osd_request *, struct page **pages, u64 length, u32 alignment, bool pages_from_pool, bool own_pages); -extern void osd_req_op_extent_osd_data_pagelist(struct ceph_osd_request *, - unsigned int which, - struct ceph_pagelist *pagelist); #ifdef CONFIG_BLOCK void osd_req_op_extent_osd_data_bio(struct ceph_osd_request *osd_req, unsigned int which, @@ -509,9 +506,6 @@ void osd_req_op_extent_osd_data_bvec_pos(struct ceph_osd_request *osd_req, void osd_req_op_extent_osd_iter(struct ceph_osd_request *osd_req, unsigned int which, struct iov_iter *iter); -extern void osd_req_op_cls_request_data_pagelist(struct ceph_osd_request *, - unsigned int which, - struct ceph_pagelist *pagelist); extern void osd_req_op_cls_request_data_pages(struct ceph_osd_request *, unsigned int which, struct page **pages, u64 length, -- cgit v1.2.3 From 569330a34a31a52c904239439984a59972c11d28 Mon Sep 17 00:00:00 2001 From: Manish Pandey Date: Fri, 11 Apr 2025 17:46:30 +0530 Subject: scsi: ufs: Introduce quirk to extend PA_HIBERN8TIME for UFS devices MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Samsung UFS devices require additional time in hibern8 mode before exiting, beyond the negotiated handshaking phase between the host and device. Introduce a quirk to increase the PA_HIBERN8TIME parameter by 100 µs, a value derived from experiments, to ensure a proper hibernation process. Signed-off-by: Manish Pandey Link: https://lore.kernel.org/r/20250411121630.21330-3-quic_mapa@quicinc.com Reviewed-by: Bean Huo Reviewed-by: Manivannan Sadhasivam Signed-off-by: Martin K. Petersen --- include/ufs/ufs_quirks.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include') diff --git a/include/ufs/ufs_quirks.h b/include/ufs/ufs_quirks.h index 41ff44dfa1db..f52de5ed1b3b 100644 --- a/include/ufs/ufs_quirks.h +++ b/include/ufs/ufs_quirks.h @@ -107,4 +107,10 @@ struct ufs_dev_quirk { */ #define UFS_DEVICE_QUIRK_DELAY_AFTER_LPM (1 << 11) +/* + * Some ufs devices may need more time to be in hibern8 before exiting. + * Enable this quirk to give it an additional 100us. + */ +#define UFS_DEVICE_QUIRK_PA_HIBER8TIME (1 << 12) + #endif /* UFS_QUIRKS_H_ */ -- cgit v1.2.3 From 16c22c56d4282584742022a37d4f79a46ca6094a Mon Sep 17 00:00:00 2001 From: Daniel Jurgens Date: Tue, 4 Mar 2025 10:14:42 -0600 Subject: virtio_pci: Use self group type for cap commands Section 2.12.1.2 of v1.4 of the VirtIO spec states: The device and driver capabilities commands are currently defined for self group type. 1. VIRTIO_ADMIN_CMD_CAP_ID_LIST_QUERY 2. VIRTIO_ADMIN_CMD_DEVICE_CAP_GET 3. VIRTIO_ADMIN_CMD_DRIVER_CAP_SET Fixes: bfcad518605d ("virtio: Manage device and driver capabilities via the admin commands") Signed-off-by: Daniel Jurgens Reviewed-by: Parav Pandit Message-Id: <20250304161442.90700-1-danielj@nvidia.com> Signed-off-by: Michael S. Tsirkin --- include/uapi/linux/virtio_pci.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/uapi/linux/virtio_pci.h b/include/uapi/linux/virtio_pci.h index 8549d4571257..c691ac210ce2 100644 --- a/include/uapi/linux/virtio_pci.h +++ b/include/uapi/linux/virtio_pci.h @@ -246,6 +246,7 @@ struct virtio_pci_cfg_cap { #define VIRTIO_ADMIN_CMD_LIST_USE 0x1 /* Admin command group type. */ +#define VIRTIO_ADMIN_GROUP_TYPE_SELF 0x0 #define VIRTIO_ADMIN_GROUP_TYPE_SRIOV 0x1 /* Transitional device admin command. */ -- cgit v1.2.3 From a940e0a685575424d33324ec7f0089045249de0a Mon Sep 17 00:00:00 2001 From: Stefano Garzarella Date: Mon, 3 Mar 2025 09:52:37 +0100 Subject: vhost: fix VHOST_*_OWNER documentation VHOST_OWNER_SET and VHOST_OWNER_RESET are used in the documentation instead of VHOST_SET_OWNER and VHOST_RESET_OWNER respectively. To avoid confusion, let's use the right names in the documentation. No change to the API, only the documentation is involved. Signed-off-by: Stefano Garzarella Message-Id: <20250303085237.19990-1-sgarzare@redhat.com> Signed-off-by: Michael S. Tsirkin --- include/uapi/linux/vhost.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/uapi/linux/vhost.h b/include/uapi/linux/vhost.h index b95dd84eef2d..d4b3e2ae1314 100644 --- a/include/uapi/linux/vhost.h +++ b/include/uapi/linux/vhost.h @@ -28,10 +28,10 @@ /* Set current process as the (exclusive) owner of this file descriptor. This * must be called before any other vhost command. Further calls to - * VHOST_OWNER_SET fail until VHOST_OWNER_RESET is called. */ + * VHOST_SET_OWNER fail until VHOST_RESET_OWNER is called. */ #define VHOST_SET_OWNER _IO(VHOST_VIRTIO, 0x01) /* Give up ownership, and reset the device to default values. - * Allows subsequent call to VHOST_OWNER_SET to succeed. */ + * Allows subsequent call to VHOST_SET_OWNER to succeed. */ #define VHOST_RESET_OWNER _IO(VHOST_VIRTIO, 0x02) /* Set up/modify memory layout */ -- cgit v1.2.3 From 1ae5e4c0626d6954114d9725990ac9c498f3b1ab Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Tue, 8 Apr 2025 12:48:55 +0300 Subject: device property: Add a note to the fwnode.h Add a note to the fwnode.h that the header should not be used directly in the leaf drivers, they all should use the higher level APIs and the respective headers. The purpose of this note is to give guidance to driver writers to avoid repeating a common mistake. Signed-off-by: Andy Shevchenko Reviewed-by: Sakari Ailus Reviewed-by: Rafael J. Wysocki Reviewed-by: Zijun Hu Link: https://lore.kernel.org/r/20250408095229.1298005-1-andriy.shevchenko@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- include/linux/fwnode.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include') diff --git a/include/linux/fwnode.h b/include/linux/fwnode.h index 6fa0a268d538..097be89487bf 100644 --- a/include/linux/fwnode.h +++ b/include/linux/fwnode.h @@ -2,6 +2,11 @@ /* * fwnode.h - Firmware device node object handle type definition. * + * This header file provides low-level data types and definitions for firmware + * and device property providers. The respective API header files supplied by + * them should contain all of the requisite data types and definitions for end + * users, so including it directly should not be necessary. + * * Copyright (C) 2015, Intel Corporation * Author: Rafael J. Wysocki */ -- cgit v1.2.3 From c9b19ea63036fc537a69265acea1b18dabd1cbd3 Mon Sep 17 00:00:00 2001 From: Marek Szyprowski Date: Tue, 15 Apr 2025 09:56:59 +0200 Subject: dma-mapping: avoid potential unused data compilation warning When CONFIG_NEED_DMA_MAP_STATE is not defined, dma-mapping clients might report unused data compilation warnings for dma_unmap_*() calls arguments. Redefine macros for those calls to let compiler to notice that it is okay when the provided arguments are not used. Reported-by: Andy Shevchenko Suggested-by: Jakub Kicinski Signed-off-by: Marek Szyprowski Tested-by: Andy Shevchenko Link: https://lore.kernel.org/r/20250415075659.428549-1-m.szyprowski@samsung.com --- include/linux/dma-mapping.h | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h index b79925b1c433..85ab710ec0e7 100644 --- a/include/linux/dma-mapping.h +++ b/include/linux/dma-mapping.h @@ -629,10 +629,14 @@ static inline int dma_mmap_wc(struct device *dev, #else #define DEFINE_DMA_UNMAP_ADDR(ADDR_NAME) #define DEFINE_DMA_UNMAP_LEN(LEN_NAME) -#define dma_unmap_addr(PTR, ADDR_NAME) (0) -#define dma_unmap_addr_set(PTR, ADDR_NAME, VAL) do { } while (0) -#define dma_unmap_len(PTR, LEN_NAME) (0) -#define dma_unmap_len_set(PTR, LEN_NAME, VAL) do { } while (0) +#define dma_unmap_addr(PTR, ADDR_NAME) \ + ({ typeof(PTR) __p __maybe_unused = PTR; 0; }) +#define dma_unmap_addr_set(PTR, ADDR_NAME, VAL) \ + do { typeof(PTR) __p __maybe_unused = PTR; } while (0) +#define dma_unmap_len(PTR, LEN_NAME) \ + ({ typeof(PTR) __p __maybe_unused = PTR; 0; }) +#define dma_unmap_len_set(PTR, LEN_NAME, VAL) \ + do { typeof(PTR) __p __maybe_unused = PTR; } while (0) #endif #endif /* _LINUX_DMA_MAPPING_H */ -- cgit v1.2.3 From 50492f942c281af4a48f8028f8409d7b8f2655d9 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Micka=C3=ABl=20Sala=C3=BCn?= Date: Wed, 16 Apr 2025 17:47:11 +0200 Subject: landlock: Fix documentation for landlock_create_ruleset(2) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Move and fix the flags documentation, and improve formatting. It makes more sense and it eases maintenance to document syscall flags in landlock.h, where they are defined. This is already the case for landlock_restrict_self(2)'s flags. The flags are now rendered like the syscall's parameters and description. Cc: Günther Noack Cc: Paul Moore Link: https://lore.kernel.org/r/20250416154716.1799902-1-mic@digikod.net Signed-off-by: Mickaël Salaün --- include/uapi/linux/landlock.h | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/uapi/linux/landlock.h b/include/uapi/linux/landlock.h index d9d0cb827117..9a4b64be9869 100644 --- a/include/uapi/linux/landlock.h +++ b/include/uapi/linux/landlock.h @@ -53,12 +53,16 @@ struct landlock_ruleset_attr { __u64 scoped; }; -/* - * sys_landlock_create_ruleset() flags: +/** + * DOC: landlock_create_ruleset_flags + * + * **Flags** + * + * %LANDLOCK_CREATE_RULESET_VERSION + * Get the highest supported Landlock ABI version (starting at 1). * - * - %LANDLOCK_CREATE_RULESET_VERSION: Get the highest supported Landlock ABI - * version. - * - %LANDLOCK_CREATE_RULESET_ERRATA: Get a bitmask of fixed issues. + * %LANDLOCK_CREATE_RULESET_ERRATA + * Get a bitmask of fixed issues for the current Landlock ABI version. */ /* clang-format off */ #define LANDLOCK_CREATE_RULESET_VERSION (1U << 0) -- cgit v1.2.3 From 25b1fc1cdc8931cf26e8d169f65ad07dfd653ca2 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Micka=C3=ABl=20Sala=C3=BCn?= Date: Wed, 16 Apr 2025 17:47:12 +0200 Subject: landlock: Fix documentation for landlock_restrict_self(2) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix, deduplicate, and improve rendering of landlock_restrict_self(2)'s flags documentation. The flags are now rendered like the syscall's parameters and description. Cc: Günther Noack Cc: Paul Moore Link: https://lore.kernel.org/r/20250416154716.1799902-2-mic@digikod.net Signed-off-by: Mickaël Salaün --- include/uapi/linux/landlock.h | 61 +++++++++++++++++++++++++------------------ 1 file changed, 36 insertions(+), 25 deletions(-) (limited to 'include') diff --git a/include/uapi/linux/landlock.h b/include/uapi/linux/landlock.h index 9a4b64be9869..8b2a1dc5c70b 100644 --- a/include/uapi/linux/landlock.h +++ b/include/uapi/linux/landlock.h @@ -69,31 +69,42 @@ struct landlock_ruleset_attr { #define LANDLOCK_CREATE_RULESET_ERRATA (1U << 1) /* clang-format on */ -/* - * sys_landlock_restrict_self() flags: - * - * - %LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF: Do not create any log related to the - * enforced restrictions. This should only be set by tools launching unknown - * or untrusted programs (e.g. a sandbox tool, container runtime, system - * service manager). Because programs sandboxing themselves should fix any - * denied access, they should not set this flag to be aware of potential - * issues reported by system's logs (i.e. audit). - * - %LANDLOCK_RESTRICT_SELF_LOG_NEW_EXEC_ON: Explicitly ask to continue - * logging denied access requests even after an :manpage:`execve(2)` call. - * This flag should only be set if all the programs than can legitimately be - * executed will not try to request a denied access (which could spam audit - * logs). - * - %LANDLOCK_RESTRICT_SELF_LOG_SUBDOMAINS_OFF: Do not create any log related - * to the enforced restrictions coming from future nested domains created by - * the caller or its descendants. This should only be set according to a - * runtime configuration (i.e. not hardcoded) by programs launching other - * unknown or untrusted programs that may create their own Landlock domains - * and spam logs. The main use case is for container runtimes to enable users - * to mute buggy sandboxed programs for a specific container image. Other use - * cases include sandboxer tools and init systems. Unlike - * %LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF, - * %LANDLOCK_RESTRICT_SELF_LOG_SUBDOMAINS_OFF does not impact the requested - * restriction (if any) but only the future nested domains. +/** + * DOC: landlock_restrict_self_flags + * + * **Flags** + * + * %LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF + * Do not create any log related to the enforced restrictions. This should + * only be set by tools launching unknown or untrusted programs (e.g. a + * sandbox tool, container runtime, system service manager). Because + * programs sandboxing themselves should fix any denied access, they should + * not set this flag to be aware of potential issues reported by system's + * logs (i.e. audit). + * + * %LANDLOCK_RESTRICT_SELF_LOG_NEW_EXEC_ON + * Explicitly ask to continue logging denied access requests even after an + * :manpage:`execve(2)` call. This flag should only be set if all the + * programs than can legitimately be executed will not try to request a + * denied access (which could spam audit logs). + * + * %LANDLOCK_RESTRICT_SELF_LOG_SUBDOMAINS_OFF + * Do not create any log related to the enforced restrictions coming from + * future nested domains created by the caller or its descendants. This + * should only be set according to a runtime configuration (i.e. not + * hardcoded) by programs launching other unknown or untrusted programs that + * may create their own Landlock domains and spam logs. The main use case + * is for container runtimes to enable users to mute buggy sandboxed + * programs for a specific container image. Other use cases include + * sandboxer tools and init systems. Unlike + * ``LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF``, + * ``LANDLOCK_RESTRICT_SELF_LOG_SUBDOMAINS_OFF`` does not impact the + * requested restriction (if any) but only the future nested domains. + * + * It is allowed to only pass the + * ``LANDLOCK_RESTRICT_SELF_LOG_SUBDOMAINS_OFF`` flag with a @ruleset_fd + * value of -1. + * */ /* clang-format off */ #define LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF (1U << 0) -- cgit v1.2.3 From 47ce2af848b7301d8571f0e01a0d7c7162d51e4a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Micka=C3=ABl=20Sala=C3=BCn?= Date: Wed, 16 Apr 2025 17:47:13 +0200 Subject: landlock: Update log documentation MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix and improve documentation related to landlock_restrict_self(2)'s flags. Update the LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF documentation according to the current semantic. Cc: Günther Noack Cc: Paul Moore Link: https://lore.kernel.org/r/20250416154716.1799902-3-mic@digikod.net Signed-off-by: Mickaël Salaün --- include/uapi/linux/landlock.h | 64 +++++++++++++++++++++++++------------------ 1 file changed, 38 insertions(+), 26 deletions(-) (limited to 'include') diff --git a/include/uapi/linux/landlock.h b/include/uapi/linux/landlock.h index 8b2a1dc5c70b..f030adc462ee 100644 --- a/include/uapi/linux/landlock.h +++ b/include/uapi/linux/landlock.h @@ -74,37 +74,49 @@ struct landlock_ruleset_attr { * * **Flags** * + * By default, denied accesses originating from programs that sandbox themselves + * are logged via the audit subsystem. Such events typically indicate unexpected + * behavior, such as bugs or exploitation attempts. However, to avoid excessive + * logging, access requests denied by a domain not created by the originating + * program are not logged by default. The rationale is that programs should know + * their own behavior, but not necessarily the behavior of other programs. This + * default configuration is suitable for most programs that sandbox themselves. + * For specific use cases, the following flags allow programs to modify this + * default logging behavior. + * + * The %LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF and + * %LANDLOCK_RESTRICT_SELF_LOG_NEW_EXEC_ON flags apply to the newly created + * Landlock domain. + * * %LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF - * Do not create any log related to the enforced restrictions. This should - * only be set by tools launching unknown or untrusted programs (e.g. a - * sandbox tool, container runtime, system service manager). Because - * programs sandboxing themselves should fix any denied access, they should - * not set this flag to be aware of potential issues reported by system's - * logs (i.e. audit). + * Disables logging of denied accesses originating from the thread creating + * the Landlock domain, as well as its children, as long as they continue + * running the same executable code (i.e., without an intervening + * :manpage:`execve(2)` call). This is intended for programs that execute + * unknown code without invoking :manpage:`execve(2)`, such as script + * interpreters. Programs that only sandbox themselves should not set this + * flag, so users can be notified of unauthorized access attempts via system + * logs. * * %LANDLOCK_RESTRICT_SELF_LOG_NEW_EXEC_ON - * Explicitly ask to continue logging denied access requests even after an - * :manpage:`execve(2)` call. This flag should only be set if all the - * programs than can legitimately be executed will not try to request a - * denied access (which could spam audit logs). + * Enables logging of denied accesses after an :manpage:`execve(2)` call, + * providing visibility into unauthorized access attempts by newly executed + * programs within the created Landlock domain. This flag is recommended + * only when all potential executables in the domain are expected to comply + * with the access restrictions, as excessive audit log entries could make + * it more difficult to identify critical events. * * %LANDLOCK_RESTRICT_SELF_LOG_SUBDOMAINS_OFF - * Do not create any log related to the enforced restrictions coming from - * future nested domains created by the caller or its descendants. This - * should only be set according to a runtime configuration (i.e. not - * hardcoded) by programs launching other unknown or untrusted programs that - * may create their own Landlock domains and spam logs. The main use case - * is for container runtimes to enable users to mute buggy sandboxed - * programs for a specific container image. Other use cases include - * sandboxer tools and init systems. Unlike - * ``LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF``, - * ``LANDLOCK_RESTRICT_SELF_LOG_SUBDOMAINS_OFF`` does not impact the - * requested restriction (if any) but only the future nested domains. - * - * It is allowed to only pass the - * ``LANDLOCK_RESTRICT_SELF_LOG_SUBDOMAINS_OFF`` flag with a @ruleset_fd - * value of -1. - * + * Disables logging of denied accesses originating from nested Landlock + * domains created by the caller or its descendants. This flag should be set + * according to runtime configuration, not hardcoded, to avoid suppressing + * important security events. It is useful for container runtimes or + * sandboxing tools that may launch programs which themselves create + * Landlock domains and could otherwise generate excessive logs. Unlike + * ``LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF``, this flag only affects + * future nested domains, not the one being created. It can also be used + * with a @ruleset_fd value of -1 to mute subdomain logs without creating a + * domain. */ /* clang-format off */ #define LANDLOCK_RESTRICT_SELF_LOG_SAME_EXEC_OFF (1U << 0) -- cgit v1.2.3 From c1a79b1a583654f24b17da81ba868b0064077243 Mon Sep 17 00:00:00 2001 From: Naohiro Aota Date: Wed, 19 Mar 2025 10:49:16 +0900 Subject: block: introduce zone capacity helper {bdev,disk}_zone_capacity() takes block_device or gendisk and sector position and returns the zone capacity of the corresponding zone. With that, move disk_nr_zones() and blk_zone_plug_bio() to consolidate them in the same #ifdef block. Signed-off-by: Naohiro Aota Reviewed-by: Damien Le Moal Reviewed-by: Johannes Thumshirn Reviewed-by: Chaitanya Kulkarni Signed-off-by: David Sterba --- include/linux/blkdev.h | 67 +++++++++++++++++++++++++++++++++----------------- 1 file changed, 45 insertions(+), 22 deletions(-) (limited to 'include') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index d37751789bf5..c57babb0adb9 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -691,23 +691,6 @@ static inline bool blk_queue_is_zoned(struct request_queue *q) (q->limits.features & BLK_FEAT_ZONED); } -#ifdef CONFIG_BLK_DEV_ZONED -static inline unsigned int disk_nr_zones(struct gendisk *disk) -{ - return disk->nr_zones; -} -bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs); -#else /* CONFIG_BLK_DEV_ZONED */ -static inline unsigned int disk_nr_zones(struct gendisk *disk) -{ - return 0; -} -static inline bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs) -{ - return false; -} -#endif /* CONFIG_BLK_DEV_ZONED */ - static inline unsigned int disk_zone_no(struct gendisk *disk, sector_t sector) { if (!blk_queue_is_zoned(disk->queue)) @@ -715,11 +698,6 @@ static inline unsigned int disk_zone_no(struct gendisk *disk, sector_t sector) return sector >> ilog2(disk->queue->limits.chunk_sectors); } -static inline unsigned int bdev_nr_zones(struct block_device *bdev) -{ - return disk_nr_zones(bdev->bd_disk); -} - static inline unsigned int bdev_max_open_zones(struct block_device *bdev) { return bdev->bd_disk->queue->limits.max_open_zones; @@ -826,6 +804,51 @@ static inline u64 sb_bdev_nr_blocks(struct super_block *sb) (sb->s_blocksize_bits - SECTOR_SHIFT); } +#ifdef CONFIG_BLK_DEV_ZONED +static inline unsigned int disk_nr_zones(struct gendisk *disk) +{ + return disk->nr_zones; +} +bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs); + +/** + * disk_zone_capacity - returns the zone capacity of zone containing @sector + * @disk: disk to work with + * @sector: sector number within the querying zone + * + * Returns the zone capacity of a zone containing @sector. @sector can be any + * sector in the zone. + */ +static inline unsigned int disk_zone_capacity(struct gendisk *disk, + sector_t sector) +{ + sector_t zone_sectors = disk->queue->limits.chunk_sectors; + + if (sector + zone_sectors >= get_capacity(disk)) + return disk->last_zone_capacity; + return disk->zone_capacity; +} +static inline unsigned int bdev_zone_capacity(struct block_device *bdev, + sector_t pos) +{ + return disk_zone_capacity(bdev->bd_disk, pos); +} +#else /* CONFIG_BLK_DEV_ZONED */ +static inline unsigned int disk_nr_zones(struct gendisk *disk) +{ + return 0; +} +static inline bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs) +{ + return false; +} +#endif /* CONFIG_BLK_DEV_ZONED */ + +static inline unsigned int bdev_nr_zones(struct block_device *bdev) +{ + return disk_nr_zones(bdev->bd_disk); +} + int bdev_disk_changed(struct gendisk *disk, bool invalidate); void put_disk(struct gendisk *disk); -- cgit v1.2.3 From 183a08715af1491d381b4e22efd61578fbe05fa5 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Thu, 10 Apr 2025 03:16:26 -0400 Subject: virtgpu: don't reset on shutdown It looks like GPUs are used after shutdown is invoked. Thus, breaking virtio gpu in the shutdown callback is not a good idea - guest hangs attempting to finish console drawing, with these warnings: [ 20.504464] WARNING: CPU: 0 PID: 568 at drivers/gpu/drm/virtio/virtgpu_vq.c:358 virtio_gpu_queue_ctrl_sgs+0x236/0x290 [virtio_gpu] [ 20.505685] Modules linked in: nft_fib_inet nft_fib_ipv4 nft_fib_ipv6 nft_fib nft_reject_inet nf_reject_ipv4 nf_reject_ipv6 nft_reject nft_ct nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 rfkill ip_set nf_tables nfnetlink vfat fat intel_rapl_msr intel_rapl_common intel_uncore_frequency_common nfit libnvdimm kvm_intel kvm rapl iTCO_wdt iTCO_vendor_support virtio_gpu virtio_dma_buf pcspkr drm_shmem_helper i2c_i801 drm_kms_helper lpc_ich i2c_smbus virtio_balloon joydev drm fuse xfs libcrc32c ahci libahci crct10dif_pclmul crc32_pclmul crc32c_intel libata virtio_net ghash_clmulni_intel net_failover virtio_blk failover serio_raw dm_mirror dm_region_hash dm_log dm_mod [ 20.511847] CPU: 0 PID: 568 Comm: kworker/0:3 Kdump: loaded Tainted: G W ------- --- 5.14.0-578.6675_1757216455.el9.x86_64 #1 [ 20.513157] Hardware name: Red Hat KVM/RHEL, BIOS edk2-20241117-3.el9 11/17/2024 [ 20.513918] Workqueue: events drm_fb_helper_damage_work [drm_kms_helper] [ 20.514626] RIP: 0010:virtio_gpu_queue_ctrl_sgs+0x236/0x290 [virtio_gpu] [ 20.515332] Code: 00 00 48 85 c0 74 0c 48 8b 78 08 48 89 ee e8 51 50 00 00 65 ff 0d 42 e3 74 3f 0f 85 69 ff ff ff 0f 1f 44 00 00 e9 5f ff ff ff <0f> 0b e9 3f ff ff ff 48 83 3c 24 00 74 0e 49 8b 7f 40 48 85 ff 74 [ 20.517272] RSP: 0018:ff34f0a8c0787ad8 EFLAGS: 00010282 [ 20.517820] RAX: 00000000fffffffb RBX: 0000000000000000 RCX: 0000000000000820 [ 20.518565] RDX: 0000000000000000 RSI: ff34f0a8c0787be0 RDI: ff218bef03a26300 [ 20.519308] RBP: ff218bef03a26300 R08: 0000000000000001 R09: ff218bef07224360 [ 20.520059] R10: 0000000000008dc0 R11: 0000000000000002 R12: ff218bef02630028 [ 20.520806] R13: ff218bef0263fb48 R14: ff218bef00cb8000 R15: ff218bef07224360 [ 20.521555] FS: 0000000000000000(0000) GS:ff218bef7ba00000(0000) knlGS:0000000000000000 [ 20.522397] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 20.522996] CR2: 000055ac4f7871c0 CR3: 000000010b9f2002 CR4: 0000000000771ef0 [ 20.523740] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 20.524477] DR3: 0000000000000000 DR6: 00000000fffe07f0 DR7: 0000000000000400 [ 20.525223] PKRU: 55555554 [ 20.525515] Call Trace: [ 20.525777] [ 20.526003] ? show_trace_log_lvl+0x1c4/0x2df [ 20.526464] ? show_trace_log_lvl+0x1c4/0x2df [ 20.526925] ? virtio_gpu_queue_fenced_ctrl_buffer+0x82/0x2c0 [virtio_gpu] [ 20.527643] ? virtio_gpu_queue_ctrl_sgs+0x236/0x290 [virtio_gpu] [ 20.528282] ? __warn+0x7e/0xd0 [ 20.528621] ? virtio_gpu_queue_ctrl_sgs+0x236/0x290 [virtio_gpu] [ 20.529256] ? report_bug+0x100/0x140 [ 20.529643] ? handle_bug+0x3c/0x70 [ 20.530010] ? exc_invalid_op+0x14/0x70 [ 20.530421] ? asm_exc_invalid_op+0x16/0x20 [ 20.530862] ? virtio_gpu_queue_ctrl_sgs+0x236/0x290 [virtio_gpu] [ 20.531506] ? virtio_gpu_queue_ctrl_sgs+0x174/0x290 [virtio_gpu] [ 20.532148] virtio_gpu_queue_fenced_ctrl_buffer+0x82/0x2c0 [virtio_gpu] [ 20.532843] virtio_gpu_primary_plane_update+0x3e2/0x460 [virtio_gpu] [ 20.533520] drm_atomic_helper_commit_planes+0x108/0x320 [drm_kms_helper] [ 20.534233] drm_atomic_helper_commit_tail+0x45/0x80 [drm_kms_helper] [ 20.534914] commit_tail+0xd2/0x130 [drm_kms_helper] [ 20.535446] drm_atomic_helper_commit+0x11b/0x140 [drm_kms_helper] [ 20.536097] drm_atomic_commit+0xa4/0xe0 [drm] [ 20.536588] ? __pfx___drm_printfn_info+0x10/0x10 [drm] [ 20.537162] drm_atomic_helper_dirtyfb+0x192/0x270 [drm_kms_helper] [ 20.537823] drm_fbdev_shmem_helper_fb_dirty+0x43/0xa0 [drm_shmem_helper] [ 20.538536] drm_fb_helper_damage_work+0x87/0x160 [drm_kms_helper] [ 20.539188] process_one_work+0x194/0x380 [ 20.539612] worker_thread+0x2fe/0x410 [ 20.540007] ? __pfx_worker_thread+0x10/0x10 [ 20.540456] kthread+0xdd/0x100 [ 20.540791] ? __pfx_kthread+0x10/0x10 [ 20.541190] ret_from_fork+0x29/0x50 [ 20.541566] [ 20.541802] ---[ end trace 0000000000000000 ]--- It looks like the shutdown is called in the middle of console drawing, so we should either wait for it to finish, or let drm handle the shutdown. This patch implements this second option: Add an option for drivers to bypass the common break+reset handling. As DRM is careful to flush/synchronize outstanding buffers, it looks like GPU can just have a NOP there. Reviewed-by: Eric Auger Tested-by: Eric Auger Fixes: 8bd2fa086a04 ("virtio: break and reset virtio devices on device_shutdown()") Cc: Eric Auger Cc: Jocelyn Falempe Signed-off-by: Michael S. Tsirkin Message-Id: <8490dbeb6f79ed039e6c11d121002618972538a3.1744293540.git.mst@redhat.com> --- include/linux/virtio.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include') diff --git a/include/linux/virtio.h b/include/linux/virtio.h index 4d16c13d0df5..64cb4b04be7a 100644 --- a/include/linux/virtio.h +++ b/include/linux/virtio.h @@ -220,6 +220,8 @@ size_t virtio_max_dma_size(const struct virtio_device *vdev); * occurs. * @reset_done: optional function to call after transport specific reset * operation has finished. + * @shutdown: synchronize with the device on shutdown. If provided, replaces + * the virtio core implementation. */ struct virtio_driver { struct device_driver driver; @@ -237,6 +239,7 @@ struct virtio_driver { int (*restore)(struct virtio_device *dev); int (*reset_prepare)(struct virtio_device *dev); int (*reset_done)(struct virtio_device *dev); + void (*shutdown)(struct virtio_device *dev); }; #define drv_to_virtio(__drv) container_of_const(__drv, struct virtio_driver, driver) -- cgit v1.2.3 From dc915672f9176799e48ac23a155f48742b15ec6c Mon Sep 17 00:00:00 2001 From: Dave Jiang Date: Thu, 17 Apr 2025 17:29:33 -0700 Subject: cxl: Fix devm host device for CXL fwctl initialization Testing revealed the following error message for a CXL memdev that has Feature support: [ 56.690430] cxl mem0: Resources present before probing Attach the allocation of cxl_fwctl to the parent device of cxl_memdev. devm_add_* calls for cxl_memdev should not happen before the memdev probe function or outside the scope of the memdev driver. cxl_test missed this bug because cxl_test always arranges for the cxl_mem driver to be loaded before cxl_mock_mem runs. So the driver core always finds the devres list idle in that case. [DJ: Updated subject title and added commit log suggestion from djbw] Fixes: 858ce2f56b52 ("cxl: Add FWCTL support to CXL") Reviewed-by: Dan Williams Reviewed-by: Alison Schofield Link: https://lore.kernel.org/linux-cxl/6801aea053466_71fe2944c@dwillia2-xfh.jf.intel.com.notmuch/ Link: https://patch.msgid.link/20250418002933.406439-1-dave.jiang@intel.com Signed-off-by: Dave Jiang --- include/cxl/features.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/cxl/features.h b/include/cxl/features.h index a3bb34694c06..5f7f842765a5 100644 --- a/include/cxl/features.h +++ b/include/cxl/features.h @@ -66,7 +66,7 @@ struct cxl_memdev; #ifdef CONFIG_CXL_FEATURES inline struct cxl_features_state *to_cxlfs(struct cxl_dev_state *cxlds); int devm_cxl_setup_features(struct cxl_dev_state *cxlds); -int devm_cxl_setup_fwctl(struct cxl_memdev *cxlmd); +int devm_cxl_setup_fwctl(struct device *host, struct cxl_memdev *cxlmd); #else static inline struct cxl_features_state *to_cxlfs(struct cxl_dev_state *cxlds) { @@ -78,7 +78,8 @@ static inline int devm_cxl_setup_features(struct cxl_dev_state *cxlds) return -EOPNOTSUPP; } -static inline int devm_cxl_setup_fwctl(struct cxl_memdev *cxlmd) +static inline int devm_cxl_setup_fwctl(struct device *host, + struct cxl_memdev *cxlmd) { return -EOPNOTSUPP; } -- cgit v1.2.3 From 2814a7d3d2ff5d2cdd22936f641f758fdb971fa0 Mon Sep 17 00:00:00 2001 From: Davidlohr Bueso Date: Thu, 17 Apr 2025 18:59:16 -0700 Subject: fs/buffer: introduce sleeping flavors for pagecache lookups Add __find_get_block_nonatomic() and sb_find_get_block_nonatomic() calls for which users will be converted where safe. These versions will take the folio lock instead of the mapping's private_lock. Reviewed-by: Jan Kara Signed-off-by: Davidlohr Bueso Link: https://kdevops.org/ext4/v6.15-rc2.html # [0] Link: https://lore.kernel.org/all/aAAEvcrmREWa1SKF@bombadil.infradead.org/ # [1] Link: https://lore.kernel.org/20250418015921.132400-3-dave@stgolabs.net Tested-by: kdevops@lists.linux.dev Reviewed-by: Luis Chamberlain Signed-off-by: Christian Brauner --- include/linux/buffer_head.h | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'include') diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h index f0a4ad7839b6..c791aa9a08da 100644 --- a/include/linux/buffer_head.h +++ b/include/linux/buffer_head.h @@ -222,6 +222,8 @@ void __wait_on_buffer(struct buffer_head *); wait_queue_head_t *bh_waitq_head(struct buffer_head *bh); struct buffer_head *__find_get_block(struct block_device *bdev, sector_t block, unsigned size); +struct buffer_head *__find_get_block_nonatomic(struct block_device *bdev, + sector_t block, unsigned size); struct buffer_head *bdev_getblk(struct block_device *bdev, sector_t block, unsigned size, gfp_t gfp); void __brelse(struct buffer_head *); @@ -397,6 +399,12 @@ sb_find_get_block(struct super_block *sb, sector_t block) return __find_get_block(sb->s_bdev, block, sb->s_blocksize); } +static inline struct buffer_head * +sb_find_get_block_nonatomic(struct super_block *sb, sector_t block) +{ + return __find_get_block_nonatomic(sb->s_bdev, block, sb->s_blocksize); +} + static inline void map_bh(struct buffer_head *bh, struct super_block *sb, sector_t block) { -- cgit v1.2.3 From 2d900efff915fe24c3948d28eef9078953d87fec Mon Sep 17 00:00:00 2001 From: Davidlohr Bueso Date: Thu, 17 Apr 2025 18:59:21 -0700 Subject: mm/migrate: fix sleep in atomic for large folios and buffer heads The large folio + buffer head noref migration scenarios are being naughty and blocking while holding a spinlock. As a consequence of the pagecache lookup path taking the folio lock this serializes against migration paths, so they can wait for each other. For the private_lock atomic case, a new BH_Migrate flag is introduced which enables the lookup to bail. This allows the critical region of the private_lock on the migration path to be reduced to the way it was before ebdf4de5642fb6 ("mm: migrate: fix reference check race between __find_get_block() and migration"), that is covering the count checks. The scope is always noref migration. Reported-by: kernel test robot Reported-by: syzbot+f3c6fda1297c748a7076@syzkaller.appspotmail.com Closes: https://lore.kernel.org/oe-lkp/202503101536.27099c77-lkp@intel.com Fixes: 3c20917120ce61 ("block/bdev: enable large folio support for large logical block sizes") Reviewed-by: Jan Kara Co-developed-by: Luis Chamberlain Signed-off-by: Davidlohr Bueso Link: https://kdevops.org/ext4/v6.15-rc2.html # [0] Link: https://lore.kernel.org/all/aAAEvcrmREWa1SKF@bombadil.infradead.org/ # [1] Link: https://lore.kernel.org/20250418015921.132400-8-dave@stgolabs.net Tested-by: kdevops@lists.linux.dev # [0] [1] Reviewed-by: Luis Chamberlain Signed-off-by: Christian Brauner --- include/linux/buffer_head.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h index c791aa9a08da..0029ff880e27 100644 --- a/include/linux/buffer_head.h +++ b/include/linux/buffer_head.h @@ -34,6 +34,7 @@ enum bh_state_bits { BH_Meta, /* Buffer contains metadata */ BH_Prio, /* Buffer should be submitted with REQ_PRIO */ BH_Defer_Completion, /* Defer AIO completion to workqueue */ + BH_Migrate, /* Buffer is being migrated (norefs) */ BH_PrivateStart,/* not a state bit, but the first bit available * for private allocation by other entities -- cgit v1.2.3 From d1f7256a5a525a44ac6a81d0a8ff931317b2acbf Mon Sep 17 00:00:00 2001 From: Mateusz Guzik Date: Fri, 18 Apr 2025 14:57:56 +0200 Subject: fs: fall back to file_ref_put() for non-last reference This reduces the slowdown in face of multiple callers issuing close on what turns out to not be the last reference. Signed-off-by: Mateusz Guzik Link: https://lore.kernel.org/20250418125756.59677-1-mjguzik@gmail.com Reviewed-by: Jan Kara Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-lkp/202504171513.6d6f8a16-lkp@intel.com Signed-off-by: Christian Brauner --- include/linux/file_ref.h | 19 ++++++------------- 1 file changed, 6 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/include/linux/file_ref.h b/include/linux/file_ref.h index 7db62fbc0500..31551e4cb8f3 100644 --- a/include/linux/file_ref.h +++ b/include/linux/file_ref.h @@ -61,7 +61,6 @@ static inline void file_ref_init(file_ref_t *ref, unsigned long cnt) atomic_long_set(&ref->refcnt, cnt - 1); } -bool __file_ref_put_badval(file_ref_t *ref, unsigned long cnt); bool __file_ref_put(file_ref_t *ref, unsigned long cnt); /** @@ -178,20 +177,14 @@ static __always_inline __must_check bool file_ref_put(file_ref_t *ref) */ static __always_inline __must_check bool file_ref_put_close(file_ref_t *ref) { - long old, new; + long old; old = atomic_long_read(&ref->refcnt); - do { - if (unlikely(old < 0)) - return __file_ref_put_badval(ref, old); - - if (old == FILE_REF_ONEREF) - new = FILE_REF_DEAD; - else - new = old - 1; - } while (!atomic_long_try_cmpxchg(&ref->refcnt, &old, new)); - - return new == FILE_REF_DEAD; + if (likely(old == FILE_REF_ONEREF)) { + if (likely(atomic_long_try_cmpxchg(&ref->refcnt, &old, FILE_REF_DEAD))) + return true; + } + return file_ref_put(ref); } /** -- cgit v1.2.3 From ce6815585d460c610e9881a5d347c0a34da287e4 Mon Sep 17 00:00:00 2001 From: "Russell King (Oracle)" Date: Wed, 16 Apr 2025 22:53:19 +0100 Subject: net: phylink: mac_link_(up|down)() clarifications As a result of an email from the fbnic author, I reviewed the phylink documentation, and I have decided to clarify the wording in the mac_link_(up|down)() kernel documentation as this was written from the point of view of mvneta/mvpp2 and is misleading. The documentation talks about forcing the link - indeed, this is what is done in the mvneta and mvpp2 drivers but not at the physical layer but the MACs idea, which has the effect of only allowing or stopping packet flow at the MAC. This "link" needs to be controlled when using a PHY or fixed link to start or stop packet flow at the MAC. However, as the MAC and PCS are tightly integrated, if the MACs idea of the link is forced down, it has the side effect that there is no way to determine that the media link has come up - in this mode, the MAC must be allowed to follow its built-in PCS so we can read the link state. Frame the documentation in more generic terms, to avoid the thought that the physical media link to the partner needs in some way to be forced up or down with these calls; it does not. If that were to be done, it would be a self-fulfilling prophecy - e.g. if the media link goes down, then mac_link_down() will be called, and if the media link is then placed into a forced down state, there is no possibility that the media link will ever come up again - clearly this is a wrong interpretation. These methods are notifications to the MAC about what has happened to the media link state - either from the PHY, or a PCS, or whatever mechanism fixed-link is using. Thus, reword them to get away from talking about changing link state to avoid confusion with media link state. This is not a change of any requirements of these methods. Also, remove the obsolete references to EEE for these methods, we now have the LPI functions for configuring the EEE parameters which renders this redundant, and also makes the passing of "phy" to the mac_link_up() function obsolete. Signed-off-by: Russell King (Oracle) Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/E1u5Ah5-001GO1-7E@rmk-PC.armlinux.org.uk Signed-off-by: Jakub Kicinski --- include/linux/phylink.h | 31 ++++++++++++++++++++----------- 1 file changed, 20 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/linux/phylink.h b/include/linux/phylink.h index 1f5773ab5660..30659b615fca 100644 --- a/include/linux/phylink.h +++ b/include/linux/phylink.h @@ -361,23 +361,29 @@ int mac_finish(struct phylink_config *config, unsigned int mode, phy_interface_t iface); /** - * mac_link_down() - take the link down + * mac_link_down() - notification that the link has gone down * @config: a pointer to a &struct phylink_config. * @mode: link autonegotiation mode * @interface: link &typedef phy_interface_t mode * - * If @mode is not an in-band negotiation mode (as defined by - * phylink_autoneg_inband()), force the link down and disable any - * Energy Efficient Ethernet MAC configuration. Interface type - * selection must be done in mac_config(). + * Notifies the MAC that the link has gone down. This will not be called + * unless mac_link_up() has been previously called. + * + * The MAC should stop processing packets for transmission and reception. + * phylink will have called netif_carrier_off() to notify the networking + * stack that the link has gone down, so MAC drivers should not make this + * call. + * + * If @mode is %MLO_AN_INBAND, then this function must not prevent the + * link coming up. */ void mac_link_down(struct phylink_config *config, unsigned int mode, phy_interface_t interface); /** - * mac_link_up() - allow the link to come up + * mac_link_up() - notification that the link has come up * @config: a pointer to a &struct phylink_config. - * @phy: any attached phy + * @phy: any attached phy (deprecated - please use LPI interfaces) * @mode: link autonegotiation mode * @interface: link &typedef phy_interface_t mode * @speed: link speed @@ -385,7 +391,10 @@ void mac_link_down(struct phylink_config *config, unsigned int mode, * @tx_pause: link transmit pause enablement status * @rx_pause: link receive pause enablement status * - * Configure the MAC for an established link. + * Notifies the MAC that the link has come up, and the parameters of the + * link as seen from the MACs point of view. If mac_link_up() has been + * called previously, there will be an intervening call to mac_link_down() + * before this method will be subsequently called. * * @speed, @duplex, @tx_pause and @rx_pause indicate the finalised link * settings, and should be used to configure the MAC block appropriately @@ -397,9 +406,9 @@ void mac_link_down(struct phylink_config *config, unsigned int mode, * that the user wishes to override the pause settings, and this should * be allowed when considering the implementation of this method. * - * If in-band negotiation mode is disabled, allow the link to come up. If - * @phy is non-%NULL, configure Energy Efficient Ethernet by calling - * phy_init_eee() and perform appropriate MAC configuration for EEE. + * Once configured, the MAC may begin to process packets for transmission + * and reception. + * * Interface type selection must be done in mac_config(). */ void mac_link_up(struct phylink_config *config, struct phy_device *phy, -- cgit v1.2.3 From 82efd569a8909f2b13140c1b3de88535aea0b051 Mon Sep 17 00:00:00 2001 From: Vlastimil Babka Date: Wed, 23 Apr 2025 10:21:29 +0200 Subject: locking/local_lock: fix _Generic() matching of local_trylock_t MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Michael Larabel reported [1] a nginx performance regression in v6.15-rc3 and bisected it to commit 51339d99c013 ("locking/local_lock, mm: replace localtry_ helpers with local_trylock_t type") The problem is the _Generic() usage with a default association that masks the fact that "local_trylock_t *" association is not being selected as expected. Replacing the default with the only other expected type "local_lock_t *" reveals the underlying problem: include/linux/local_lock_internal.h:174:26: error: ‘_Generic’ selector of type ‘__seg_gs local_lock_t *’ is not compatible with any association The local_locki's are part of __percpu structures and thus the __percpu attribute is needed to associate the type properly. Add the attribute and keep the default replaced to turn any further mismatches into compile errors. The failure to recognize local_try_lock_t in __local_lock_release() means that a local_trylock[_irqsave]() operation will set tl->acquired to 1 (there's no _Generic() part in the trylock code), but then local_unlock[_irqrestore]() will not set tl->acquired back to 0, so further trylock operations will always fail on the same cpu+lock, while non-trylock operations continue to work - a lockdep_assert() is also not being executed in the _Generic() part of local_lock() code. This means consume_stock() and refill_stock() operations will fail deterministically, resulting in taking the slow paths and worse performance. Fixes: 51339d99c013 ("locking/local_lock, mm: replace localtry_ helpers with local_trylock_t type") Reported-by: Michael Larabel Closes: https://www.phoronix.com/review/linux-615-nginx-regression/2 [1] Signed-off-by: Vlastimil Babka Acked-by: Alexei Starovoitov Signed-off-by: Linus Torvalds --- include/linux/local_lock_internal.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/local_lock_internal.h b/include/linux/local_lock_internal.h index bf2bf40d7b18..8d5ac16a9b17 100644 --- a/include/linux/local_lock_internal.h +++ b/include/linux/local_lock_internal.h @@ -102,11 +102,11 @@ do { \ l = (local_lock_t *)this_cpu_ptr(lock); \ tl = (local_trylock_t *)l; \ _Generic((lock), \ - local_trylock_t *: ({ \ + __percpu local_trylock_t *: ({ \ lockdep_assert(tl->acquired == 0); \ WRITE_ONCE(tl->acquired, 1); \ }), \ - default:(void)0); \ + __percpu local_lock_t *: (void)0); \ local_lock_acquire(l); \ } while (0) @@ -171,11 +171,11 @@ do { \ tl = (local_trylock_t *)l; \ local_lock_release(l); \ _Generic((lock), \ - local_trylock_t *: ({ \ + __percpu local_trylock_t *: ({ \ lockdep_assert(tl->acquired == 1); \ WRITE_ONCE(tl->acquired, 0); \ }), \ - default:(void)0); \ + __percpu local_lock_t *: (void)0); \ } while (0) #define __local_unlock(lock) \ -- cgit v1.2.3 From e03463d247ddac66e71143468373df3d74a3a6bd Mon Sep 17 00:00:00 2001 From: "Darrick J. Wong" Date: Wed, 23 Apr 2025 12:53:57 -0700 Subject: block: hoist block size validation code to a separate function Hoist the block size validation code to bdev_validate_blocksize so that we can call it from filesystems that don't care about the bdev pagecache manipulations of set_blocksize. Signed-off-by: Darrick J. Wong Reviewed-by: Luis Chamberlain Reviewed-by: Christoph Hellwig Link: https://lore.kernel.org/r/174543795720.4139148.840349813093799165.stgit@frogsfrogsfrogs Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index e39c45bc0a97..c6b478cfb32d 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -1614,6 +1614,7 @@ static inline void bio_end_io_acct(struct bio *bio, unsigned long start_time) return bio_end_io_acct_remapped(bio, start_time, bio->bi_bdev); } +int bdev_validate_blocksize(struct block_device *bdev, int block_size); int set_blocksize(struct file *file, int size); int lookup_bdev(const char *pathname, dev_t *dev); -- cgit v1.2.3 From c63202140d4b411d27380805c4d68eb11407b7f2 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 23 Apr 2025 07:37:39 +0200 Subject: block: move blkdev_{get,put} _no_open prototypes out of blkdev.h These are only to be used by block internal code. Remove the comment as we grew more users due to reworking block device node opening. Signed-off-by: Christoph Hellwig Reviewed-by: Christian Brauner Acked-by: Tejun Heo Link: https://lore.kernel.org/r/20250423053810.1683309-2-hch@lst.de Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 4 ---- 1 file changed, 4 deletions(-) (limited to 'include') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index c6b478cfb32d..14abaeedaa88 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -1671,10 +1671,6 @@ int bd_prepare_to_claim(struct block_device *bdev, void *holder, const struct blk_holder_ops *hops); void bd_abort_claiming(struct block_device *bdev, void *holder); -/* just for blk-cgroup, don't use elsewhere */ -struct block_device *blkdev_get_no_open(dev_t dev); -void blkdev_put_no_open(struct block_device *bdev); - struct block_device *I_BDEV(struct inode *inode); struct block_device *file_bdev(struct file *bdev_file); bool disk_live(struct gendisk *disk); -- cgit v1.2.3