- Linux-stable-mirror - lists.linaro.org

[PATCH] net-sysfs: fix error handling in netdev_register_kobject()

by Ma Ke

Once device_add() failed, we should call put_device() to decrement reference count for cleanup. Or it could cause memory leak. As comment of device_add() says, 'if device_add() succeeds, you should call device_del() when you want to get rid of it. If device_add() has not succeeded, use only put_device() to drop the reference count'. Found by code review. Cc: stable(a)vger.kernel.org Fixes: 8ed633b9baf9 ("Revert "net-sysfs: Fix memory leak in netdev_register_kobject"") Signed-off-by: Ma Ke <make24(a)iscas.ac.cn> --- net/core/net-sysfs.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 07cb99b114bd..f443eacc9237 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -2169,6 +2169,7 @@ int netdev_register_kobject(struct net_device *ndev) error = device_add(dev); if (error) + put_device(dev); return error; error = register_queue_kobjects(ndev); -- 2.25.1

4 months

3
3
0 0

[PATCH] mtd: fix error handling in uif_init()

by Ma Ke

Once cdev_device_add() failed, we should call put_device() to decrement reference count for cleanup. Or it could cause memory leak. Found by code review. Cc: stable(a)vger.kernel.org Fixes: 493cfaeaa0c9 ("mtd: utilize new cdev_device_add helper function") Signed-off-by: Ma Ke <make24(a)iscas.ac.cn> --- drivers/mtd/ubi/build.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index ef6a22f372f9..ca4c54cf9fd4 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -486,6 +486,7 @@ static int uif_init(struct ubi_device *ubi) kill_volumes(ubi); cdev_device_del(&ubi->cdev, &ubi->dev); out_unreg: + put_device(&ubi->dev); unregister_chrdev_region(ubi->cdev.dev, ubi->vtbl_slots + 1); ubi_err(ubi, "cannot initialize UBI %s, error %d", ubi->ubi_name, err); -- 2.25.1

4 months

2
1
0 0

[PATCH] [POWERPC] ps3: fix error handling in ps3_system_bus_device_register()

by Ma Ke

Once device_register() failed, we should call put_device() to decrement reference count for cleanup. Or it could cause memory leak. As comment of device_register() says, 'NOTE: _Never_ directly free @dev after calling this function, even if it returned an error! Always use put_device() to give up the reference initialized in this function instead.' Found by code review. Cc: stable(a)vger.kernel.org Fixes: a3d4d6435b56 ("[POWERPC] ps3: add ps3 platform system bus support") Signed-off-by: Ma Ke <make24(a)iscas.ac.cn> --- arch/powerpc/platforms/ps3/system-bus.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/powerpc/platforms/ps3/system-bus.c b/arch/powerpc/platforms/ps3/system-bus.c index afbaabf182d0..c477d0ee523a 100644 --- a/arch/powerpc/platforms/ps3/system-bus.c +++ b/arch/powerpc/platforms/ps3/system-bus.c @@ -769,6 +769,9 @@ int ps3_system_bus_device_register(struct ps3_system_bus_device *dev) pr_debug("%s:%d add %s\n", __func__, __LINE__, dev_name(&dev->core)); result = device_register(&dev->core); + if (result) + put_device(&dev->core); + return result; } -- 2.25.1

4 months

2
1
0 0

[PATCH 6.6.y] zram: fix NULL pointer in comp_algorithm_show()

by jianqi.ren.cn＠windriver.com

From: Liu Shixin <liushixin2(a)huawei.com> [ Upstream commit f364cdeb38938f9d03061682b8ff3779dd1730e5 ] LTP reported a NULL pointer dereference as followed: CPU: 7 UID: 0 PID: 5995 Comm: cat Kdump: loaded Not tainted 6.12.0-rc6+ #3 Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015 pstate: 40400005 (nZcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : __pi_strcmp+0x24/0x140 lr : zcomp_available_show+0x60/0x100 [zram] sp : ffff800088b93b90 x29: ffff800088b93b90 x28: 0000000000000001 x27: 0000000000400cc0 x26: 0000000000000ffe x25: ffff80007b3e2388 x24: 0000000000000000 x23: ffff80007b3e2390 x22: ffff0004041a9000 x21: ffff80007b3e2900 x20: 0000000000000000 x19: 0000000000000000 x18: 0000000000000000 x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000 x14: 0000000000000000 x13: 0000000000000000 x12: 0000000000000000 x11: 0000000000000000 x10: ffff80007b3e2900 x9 : ffff80007b3cb280 x8 : 0101010101010101 x7 : 0000000000000000 x6 : 0000000000000000 x5 : 0000000000000040 x4 : 0000000000000000 x3 : 00656c722d6f7a6c x2 : 0000000000000000 x1 : ffff80007b3e2900 x0 : 0000000000000000 Call trace: __pi_strcmp+0x24/0x140 comp_algorithm_show+0x40/0x70 [zram] dev_attr_show+0x28/0x80 sysfs_kf_seq_show+0x90/0x140 kernfs_seq_show+0x34/0x48 seq_read_iter+0x1d4/0x4e8 kernfs_fop_read_iter+0x40/0x58 new_sync_read+0x9c/0x168 vfs_read+0x1a8/0x1f8 ksys_read+0x74/0x108 __arm64_sys_read+0x24/0x38 invoke_syscall+0x50/0x120 el0_svc_common.constprop.0+0xc8/0xf0 do_el0_svc+0x24/0x38 el0_svc+0x38/0x138 el0t_64_sync_handler+0xc0/0xc8 el0t_64_sync+0x188/0x190 The zram->comp_algs[ZRAM_PRIMARY_COMP] can be NULL in zram_add() if comp_algorithm_set() has not been called. User can access the zram device by sysfs after device_add_disk(), so there is a time window to trigger the NULL pointer dereference. Move it ahead device_add_disk() to make sure when user can access the zram device, it is ready. comp_algorithm_set() is protected by zram->init_lock in other places and no such problem. Link: https://lkml.kernel.org/r/20241108100147.3776123-1-liushixin2@huawei.com Fixes: 7ac07a26dea7 ("zram: preparation for multi-zcomp support") Signed-off-by: Liu Shixin <liushixin2(a)huawei.com> Reviewed-by: Sergey Senozhatsky <senozhatsky(a)chromium.org> Cc: Jens Axboe <axboe(a)kernel.dk> Cc: Minchan Kim <minchan(a)kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> [This fix does not backport zram_comp_params_reset which was introduced after v6.6, in commit f2bac7ad187d ("zram: introduce zcomp_params structure")] Signed-off-by: Jianqi Ren <jianqi.ren.cn(a)windriver.com> Signed-off-by: He Zhe <zhe.he(a)windriver.com> --- Verified the build test --- drivers/block/zram/zram_drv.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c index b73038ad86f7..44cf0e51d7db 100644 --- a/drivers/block/zram/zram_drv.c +++ b/drivers/block/zram/zram_drv.c @@ -2247,6 +2247,8 @@ static int zram_add(void) zram->disk->private_data = zram; snprintf(zram->disk->disk_name, 16, "zram%d", device_id); + comp_algorithm_set(zram, ZRAM_PRIMARY_COMP, default_compressor); + /* Actual capacity set using sysfs (/sys/block/zram<id>/disksize */ set_capacity(zram->disk, 0); /* zram devices sort of resembles non-rotational disks */ @@ -2281,8 +2283,6 @@ static int zram_add(void) if (ret) goto out_cleanup_disk; - comp_algorithm_set(zram, ZRAM_PRIMARY_COMP, default_compressor); - zram_debugfs_register(zram); pr_info("Added device: %s\n", zram->disk->disk_name); return device_id; -- 2.25.1

4 months

2
1
0 0

[PATCH 5.15.y] fs/ntfs3: Fix shift-out-of-bounds in ntfs_fill_super

by Miguel García

From: Konstantin Komarov <almaz.alexandrovich(a)paragon-software.com> commit 91a4b1ee78cb ("fs/ntfs3: Fix shift-out-of-bounds in ntfs_fill_super") This patch is a backport and fixes an UBSAN warning about shift-out-of-bounds in ntfs_fill_super() function of the NTFS3 driver. The original code incorrectly calculated MFT record size, causing undefined behavior when performing bit shifts with values that exceed type limits. The fix has been verified by executing the syzkaller reproducer test case. After applying this patch, the system successfully handles the test case without kernel panic or UBSAN warnings. Bug: https://syzkaller.appspot.com/bug?extid=010986becd65dbf9464b Reported-by: syzbot+010986becd65dbf9464b(a)syzkaller.appspotmail.com Cc: <stable(a)vger.kernel.org> Signed-off-by: Konstantin Komarov <almaz.alexandrovich(a)paragon-software.com> Signed-off-by: Miguel Garcia Roman <miguelgarciaroman8(a)gmail.com> (cherry picked from commit 91a4b1ee78cb100b19b70f077c247f211110348f) --- fs/ntfs3/ntfs_fs.h | 2 ++ fs/ntfs3/super.c | 63 +++++++++++++++++++++++++++++++++++----------- 2 files changed, 50 insertions(+), 15 deletions(-) diff --git a/fs/ntfs3/ntfs_fs.h b/fs/ntfs3/ntfs_fs.h index 7b46926e920c..c5ef0e0cdf59 100644 --- a/fs/ntfs3/ntfs_fs.h +++ b/fs/ntfs3/ntfs_fs.h @@ -43,9 +43,11 @@ enum utf16_endian; #define MINUS_ONE_T ((size_t)(-1)) /* Biggest MFT / smallest cluster */ #define MAXIMUM_BYTES_PER_MFT 4096 +#define MAXIMUM_SHIFT_BYTES_PER_MFT 12 #define NTFS_BLOCKS_PER_MFT_RECORD (MAXIMUM_BYTES_PER_MFT / 512) #define MAXIMUM_BYTES_PER_INDEX 4096 +#define MAXIMUM_SHIFT_BYTES_PER_INDEX 12 #define NTFS_BLOCKS_PER_INODE (MAXIMUM_BYTES_PER_INDEX / 512) /* NTFS specific error code when fixup failed. */ diff --git a/fs/ntfs3/super.c b/fs/ntfs3/super.c index 78b086527331..abab388e413f 100644 --- a/fs/ntfs3/super.c +++ b/fs/ntfs3/super.c @@ -680,7 +680,7 @@ static u32 true_sectors_per_clst(const struct NTFS_BOOT *boot) * ntfs_init_from_boot - Init internal info from on-disk boot sector. */ static int ntfs_init_from_boot(struct super_block *sb, u32 sector_size, - u64 dev_size) + u64 dev_size) { struct ntfs_sb_info *sbi = sb->s_fs_info; int err; @@ -705,12 +705,12 @@ static int ntfs_init_from_boot(struct super_block *sb, u32 sector_size, /* 0x55AA is not mandaroty. Thanks Maxim Suhanov*/ /*if (0x55 != boot->boot_magic[0] || 0xAA != boot->boot_magic[1]) - * goto out; + * goto out; */ boot_sector_size = (u32)boot->bytes_per_sector[1] << 8; if (boot->bytes_per_sector[0] || boot_sector_size < SECTOR_SIZE || - !is_power_of_2(boot_sector_size)) { + !is_power_of_2(boot_sector_size)) { goto out; } @@ -733,15 +733,49 @@ static int ntfs_init_from_boot(struct super_block *sb, u32 sector_size, /* Check MFT record size. */ if ((boot->record_size < 0 && - SECTOR_SIZE > (2U << (-boot->record_size))) || - (boot->record_size >= 0 && !is_power_of_2(boot->record_size))) { + SECTOR_SIZE > (2U << (-boot->record_size))) || + (boot->record_size >= 0 && !is_power_of_2(boot->record_size))) { + goto out; + } + + /* Calculate cluster size */ + sbi->cluster_size = boot_sector_size * sct_per_clst; + sbi->cluster_bits = blksize_bits(sbi->cluster_size); + + if (boot->record_size >= 0) { + record_size = (u32)boot->record_size << sbi->cluster_bits; + } else if (-boot->record_size <= MAXIMUM_SHIFT_BYTES_PER_MFT) { + record_size = 1u << (-boot->record_size); + } else { + ntfs_err(sb, "%s: invalid record size %d.", "NTFS", + boot->record_size); + goto out; + } + + sbi->record_size = record_size; + sbi->record_bits = blksize_bits(record_size); + sbi->attr_size_tr = (5 * record_size >> 4); // ~320 bytes + + if (record_size > MAXIMUM_BYTES_PER_MFT) { + ntfs_err(sb, "Unsupported bytes per MFT record %u.", + record_size); + goto out; + } + + if (boot->index_size >= 0) { + sbi->index_size = (u32)boot->index_size << sbi->cluster_bits; + } else if (-boot->index_size <= MAXIMUM_SHIFT_BYTES_PER_INDEX) { + sbi->index_size = 1u << (-boot->index_size); + } else { + ntfs_err(sb, "%s: invalid index size %d.", "NTFS", + boot->index_size); goto out; } /* Check index record size. */ if ((boot->index_size < 0 && - SECTOR_SIZE > (2U << (-boot->index_size))) || - (boot->index_size >= 0 && !is_power_of_2(boot->index_size))) { + SECTOR_SIZE > (2U << (-boot->index_size))) || + (boot->index_size >= 0 && !is_power_of_2(boot->index_size))) { goto out; } @@ -762,9 +796,6 @@ static int ntfs_init_from_boot(struct super_block *sb, u32 sector_size, dev_size += sector_size - 1; } - sbi->cluster_size = boot_sector_size * sct_per_clst; - sbi->cluster_bits = blksize_bits(sbi->cluster_size); - sbi->mft.lbo = mlcn << sbi->cluster_bits; sbi->mft.lbo2 = mlcn2 << sbi->cluster_bits; @@ -785,9 +816,9 @@ static int ntfs_init_from_boot(struct super_block *sb, u32 sector_size, sbi->cluster_mask = sbi->cluster_size - 1; sbi->cluster_mask_inv = ~(u64)sbi->cluster_mask; sbi->record_size = record_size = boot->record_size < 0 - ? 1 << (-boot->record_size) - : (u32)boot->record_size - << sbi->cluster_bits; + ? 1 << (-boot->record_size) + : (u32)boot->record_size + << sbi->cluster_bits; if (record_size > MAXIMUM_BYTES_PER_MFT || record_size < SECTOR_SIZE) goto out; @@ -801,8 +832,8 @@ static int ntfs_init_from_boot(struct super_block *sb, u32 sector_size, ALIGN(sizeof(enum ATTR_TYPE), 8); sbi->index_size = boot->index_size < 0 - ? 1u << (-boot->index_size) - : (u32)boot->index_size << sbi->cluster_bits; + ? 1u << (-boot->index_size) + : (u32)boot->index_size << sbi->cluster_bits; sbi->volume.ser_num = le64_to_cpu(boot->serial_num); @@ -879,6 +910,8 @@ static int ntfs_init_from_boot(struct super_block *sb, u32 sector_size, return err; } + + /* * ntfs_fill_super - Try to mount. */ -- 2.34.1

4 months

4
7
0 0

FAILED: patch "[PATCH] mm: fix kernel BUG when userfaultfd_move encounters swapcache" failed to apply to 6.13-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.13-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.13.y git checkout FETCH_HEAD git cherry-pick -x c50f8e6053b0503375c2975bf47f182445aebb4c # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2025030926-tapestry-rabid-7392@gregkh' --subject-prefix 'PATCH 6.13.y' HEAD^.. Possible dependencies: thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From c50f8e6053b0503375c2975bf47f182445aebb4c Mon Sep 17 00:00:00 2001 From: Barry Song <baohua(a)kernel.org> Date: Wed, 26 Feb 2025 13:14:00 +1300 Subject: [PATCH] mm: fix kernel BUG when userfaultfd_move encounters swapcache userfaultfd_move() checks whether the PTE entry is present or a swap entry. - If the PTE entry is present, move_present_pte() handles folio migration by setting: src_folio->index = linear_page_index(dst_vma, dst_addr); - If the PTE entry is a swap entry, move_swap_pte() simply copies the PTE to the new dst_addr. This approach is incorrect because, even if the PTE is a swap entry, it can still reference a folio that remains in the swap cache. This creates a race window between steps 2 and 4. 1. add_to_swap: The folio is added to the swapcache. 2. try_to_unmap: PTEs are converted to swap entries. 3. pageout: The folio is written back. 4. Swapcache is cleared. If userfaultfd_move() occurs in the window between steps 2 and 4, after the swap PTE has been moved to the destination, accessing the destination triggers do_swap_page(), which may locate the folio in the swapcache. However, since the folio's index has not been updated to match the destination VMA, do_swap_page() will detect a mismatch. This can result in two critical issues depending on the system configuration. If KSM is disabled, both small and large folios can trigger a BUG during the add_rmap operation due to: page_pgoff(folio, page) != linear_page_index(vma, address) [ 13.336953] page: refcount:6 mapcount:1 mapping:00000000f43db19c index:0xffffaf150 pfn:0x4667c [ 13.337520] head: order:2 mapcount:1 entire_mapcount:0 nr_pages_mapped:1 pincount:0 [ 13.337716] memcg:ffff00000405f000 [ 13.337849] anon flags: 0x3fffc0000020459(locked|uptodate|dirty|owner_priv_1|head|swapbacked|node=0|zone=0|lastcpupid=0xffff) [ 13.338630] raw: 03fffc0000020459 ffff80008507b538 ffff80008507b538 ffff000006260361 [ 13.338831] raw: 0000000ffffaf150 0000000000004000 0000000600000000 ffff00000405f000 [ 13.339031] head: 03fffc0000020459 ffff80008507b538 ffff80008507b538 ffff000006260361 [ 13.339204] head: 0000000ffffaf150 0000000000004000 0000000600000000 ffff00000405f000 [ 13.339375] head: 03fffc0000000202 fffffdffc0199f01 ffffffff00000000 0000000000000001 [ 13.339546] head: 0000000000000004 0000000000000000 00000000ffffffff 0000000000000000 [ 13.339736] page dumped because: VM_BUG_ON_PAGE(page_pgoff(folio, page) != linear_page_index(vma, address)) [ 13.340190] ------------[ cut here ]------------ [ 13.340316] kernel BUG at mm/rmap.c:1380! [ 13.340683] Internal error: Oops - BUG: 00000000f2000800 [#1] PREEMPT SMP [ 13.340969] Modules linked in: [ 13.341257] CPU: 1 UID: 0 PID: 107 Comm: a.out Not tainted 6.14.0-rc3-gcf42737e247a-dirty #299 [ 13.341470] Hardware name: linux,dummy-virt (DT) [ 13.341671] pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 13.341815] pc : __page_check_anon_rmap+0xa0/0xb0 [ 13.341920] lr : __page_check_anon_rmap+0xa0/0xb0 [ 13.342018] sp : ffff80008752bb20 [ 13.342093] x29: ffff80008752bb20 x28: fffffdffc0199f00 x27: 0000000000000001 [ 13.342404] x26: 0000000000000000 x25: 0000000000000001 x24: 0000000000000001 [ 13.342575] x23: 0000ffffaf0d0000 x22: 0000ffffaf0d0000 x21: fffffdffc0199f00 [ 13.342731] x20: fffffdffc0199f00 x19: ffff000006210700 x18: 00000000ffffffff [ 13.342881] x17: 6c203d2120296567 x16: 6170202c6f696c6f x15: 662866666f67705f [ 13.343033] x14: 6567617028454741 x13: 2929737365726464 x12: ffff800083728ab0 [ 13.343183] x11: ffff800082996bf8 x10: 0000000000000fd7 x9 : ffff80008011bc40 [ 13.343351] x8 : 0000000000017fe8 x7 : 00000000fffff000 x6 : ffff8000829eebf8 [ 13.343498] x5 : c0000000fffff000 x4 : 0000000000000000 x3 : 0000000000000000 [ 13.343645] x2 : 0000000000000000 x1 : ffff0000062db980 x0 : 000000000000005f [ 13.343876] Call trace: [ 13.344045] __page_check_anon_rmap+0xa0/0xb0 (P) [ 13.344234] folio_add_anon_rmap_ptes+0x22c/0x320 [ 13.344333] do_swap_page+0x1060/0x1400 [ 13.344417] __handle_mm_fault+0x61c/0xbc8 [ 13.344504] handle_mm_fault+0xd8/0x2e8 [ 13.344586] do_page_fault+0x20c/0x770 [ 13.344673] do_translation_fault+0xb4/0xf0 [ 13.344759] do_mem_abort+0x48/0xa0 [ 13.344842] el0_da+0x58/0x130 [ 13.344914] el0t_64_sync_handler+0xc4/0x138 [ 13.345002] el0t_64_sync+0x1ac/0x1b0 [ 13.345208] Code: aa1503e0 f000f801 910f6021 97ff5779 (d4210000) [ 13.345504] ---[ end trace 0000000000000000 ]--- [ 13.345715] note: a.out[107] exited with irqs disabled [ 13.345954] note: a.out[107] exited with preempt_count 2 If KSM is enabled, Peter Xu also discovered that do_swap_page() may trigger an unexpected CoW operation for small folios because ksm_might_need_to_copy() allocates a new folio when the folio index does not match linear_page_index(vma, addr). This patch also checks the swapcache when handling swap entries. If a match is found in the swapcache, it processes it similarly to a present PTE. However, there are some differences. For example, the folio is no longer exclusive because folio_try_share_anon_rmap_pte() is performed during unmapping. Furthermore, in the case of swapcache, the folio has already been unmapped, eliminating the risk of concurrent rmap walks and removing the need to acquire src_folio's anon_vma or lock. Note that for large folios, in the swapcache handling path, we directly return -EBUSY since split_folio() will return -EBUSY regardless if the folio is under writeback or unmapped. This is not an urgent issue, so a follow-up patch may address it separately. [v-songbaohua(a)oppo.com: minor cleanup according to Peter Xu] Link: https://lkml.kernel.org/r/20250226024411.47092-1-21cnbao@gmail.com Link: https://lkml.kernel.org/r/20250226001400.9129-1-21cnbao@gmail.com Fixes: adef440691ba ("userfaultfd: UFFDIO_MOVE uABI") Signed-off-by: Barry Song <v-songbaohua(a)oppo.com> Acked-by: Peter Xu <peterx(a)redhat.com> Reviewed-by: Suren Baghdasaryan <surenb(a)google.com> Cc: Andrea Arcangeli <aarcange(a)redhat.com> Cc: Al Viro <viro(a)zeniv.linux.org.uk> Cc: Axel Rasmussen <axelrasmussen(a)google.com> Cc: Brian Geffon <bgeffon(a)google.com> Cc: Christian Brauner <brauner(a)kernel.org> Cc: David Hildenbrand <david(a)redhat.com> Cc: Hugh Dickins <hughd(a)google.com> Cc: Jann Horn <jannh(a)google.com> Cc: Kalesh Singh <kaleshsingh(a)google.com> Cc: Liam R. Howlett <Liam.Howlett(a)oracle.com> Cc: Lokesh Gidra <lokeshgidra(a)google.com> Cc: Matthew Wilcox (Oracle) <willy(a)infradead.org> Cc: Michal Hocko <mhocko(a)suse.com> Cc: Mike Rapoport (IBM) <rppt(a)kernel.org> Cc: Nicolas Geoffray <ngeoffray(a)google.com> Cc: Ryan Roberts <ryan.roberts(a)arm.com> Cc: Shuah Khan <shuah(a)kernel.org> Cc: ZhangPeng <zhangpeng362(a)huawei.com> Cc: Tangquan Zheng <zhengtangquan(a)oppo.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> diff --git a/mm/userfaultfd.c b/mm/userfaultfd.c index af3dfc3633db..c45b672e10d1 100644 --- a/mm/userfaultfd.c +++ b/mm/userfaultfd.c @@ -18,6 +18,7 @@ #include <asm/tlbflush.h> #include <asm/tlb.h> #include "internal.h" +#include "swap.h" static __always_inline bool validate_dst_vma(struct vm_area_struct *dst_vma, unsigned long dst_end) @@ -1076,16 +1077,14 @@ static int move_present_pte(struct mm_struct *mm, return err; } -static int move_swap_pte(struct mm_struct *mm, +static int move_swap_pte(struct mm_struct *mm, struct vm_area_struct *dst_vma, unsigned long dst_addr, unsigned long src_addr, pte_t *dst_pte, pte_t *src_pte, pte_t orig_dst_pte, pte_t orig_src_pte, pmd_t *dst_pmd, pmd_t dst_pmdval, - spinlock_t *dst_ptl, spinlock_t *src_ptl) + spinlock_t *dst_ptl, spinlock_t *src_ptl, + struct folio *src_folio) { - if (!pte_swp_exclusive(orig_src_pte)) - return -EBUSY; - double_pt_lock(dst_ptl, src_ptl); if (!is_pte_pages_stable(dst_pte, src_pte, orig_dst_pte, orig_src_pte, @@ -1094,6 +1093,16 @@ static int move_swap_pte(struct mm_struct *mm, return -EAGAIN; } + /* + * The src_folio resides in the swapcache, requiring an update to its + * index and mapping to align with the dst_vma, where a swap-in may + * occur and hit the swapcache after moving the PTE. + */ + if (src_folio) { + folio_move_anon_rmap(src_folio, dst_vma); + src_folio->index = linear_page_index(dst_vma, dst_addr); + } + orig_src_pte = ptep_get_and_clear(mm, src_addr, src_pte); set_pte_at(mm, dst_addr, dst_pte, orig_src_pte); double_pt_unlock(dst_ptl, src_ptl); @@ -1141,6 +1150,7 @@ static int move_pages_pte(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmd, __u64 mode) { swp_entry_t entry; + struct swap_info_struct *si = NULL; pte_t orig_src_pte, orig_dst_pte; pte_t src_folio_pte; spinlock_t *src_ptl, *dst_ptl; @@ -1322,6 +1332,8 @@ static int move_pages_pte(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmd, orig_dst_pte, orig_src_pte, dst_pmd, dst_pmdval, dst_ptl, src_ptl, src_folio); } else { + struct folio *folio = NULL; + entry = pte_to_swp_entry(orig_src_pte); if (non_swap_entry(entry)) { if (is_migration_entry(entry)) { @@ -1335,9 +1347,53 @@ static int move_pages_pte(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmd, goto out; } - err = move_swap_pte(mm, dst_addr, src_addr, dst_pte, src_pte, - orig_dst_pte, orig_src_pte, dst_pmd, - dst_pmdval, dst_ptl, src_ptl); + if (!pte_swp_exclusive(orig_src_pte)) { + err = -EBUSY; + goto out; + } + + si = get_swap_device(entry); + if (unlikely(!si)) { + err = -EAGAIN; + goto out; + } + /* + * Verify the existence of the swapcache. If present, the folio's + * index and mapping must be updated even when the PTE is a swap + * entry. The anon_vma lock is not taken during this process since + * the folio has already been unmapped, and the swap entry is + * exclusive, preventing rmap walks. + * + * For large folios, return -EBUSY immediately, as split_folio() + * also returns -EBUSY when attempting to split unmapped large + * folios in the swapcache. This issue needs to be resolved + * separately to allow proper handling. + */ + if (!src_folio) + folio = filemap_get_folio(swap_address_space(entry), + swap_cache_index(entry)); + if (!IS_ERR_OR_NULL(folio)) { + if (folio_test_large(folio)) { + err = -EBUSY; + folio_put(folio); + goto out; + } + src_folio = folio; + src_folio_pte = orig_src_pte; + if (!folio_trylock(src_folio)) { + pte_unmap(&orig_src_pte); + pte_unmap(&orig_dst_pte); + src_pte = dst_pte = NULL; + put_swap_device(si); + si = NULL; + /* now we can block and wait */ + folio_lock(src_folio); + goto retry; + } + } + err = move_swap_pte(mm, dst_vma, dst_addr, src_addr, dst_pte, src_pte, + orig_dst_pte, orig_src_pte, dst_pmd, dst_pmdval, + dst_ptl, src_ptl, src_folio); } out: @@ -1354,6 +1410,8 @@ static int move_pages_pte(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmd, if (src_pte) pte_unmap(src_pte); mmu_notifier_invalidate_range_end(&range); + if (si) + put_swap_device(si); return err; }

4 months

3
2
0 0

[PATCH 6.1.y 1/1] hrtimer: Use and report correct timerslack values for realtime tasks

by Felix Moessbauer

commit ed4fb6d7ef68111bb539283561953e5c6e9a6e38 upstream. The timerslack_ns setting is used to specify how much the hardware timers should be delayed, to potentially dispatch multiple timers in a single interrupt. This is a performance optimization. Timers of realtime tasks (having a realtime scheduling policy) should not be delayed. This logic was inconsitently applied to the hrtimers, leading to delays of realtime tasks which used timed waits for events (e.g. condition variables). Due to the downstream override of the slack for rt tasks, the procfs reported incorrect (non-zero) timerslack_ns values. This is changed by setting the timer_slack_ns task attribute to 0 for all tasks with a rt policy. By that, downstream users do not need to specially handle rt tasks (w.r.t. the slack), and the procfs entry shows the correct value of "0". Setting non-zero slack values (either via procfs or PR_SET_TIMERSLACK) on tasks with a rt policy is ignored, as stated in "man 2 PR_SET_TIMERSLACK": Timer slack is not applied to threads that are scheduled under a real-time scheduling policy (see sched_setscheduler(2)). The special handling of timerslack on rt tasks in downstream users is removed as well. Signed-off-by: Felix Moessbauer <felix.moessbauer(a)siemens.com> Signed-off-by: Thomas Gleixner <tglx(a)linutronix.de> Link: https://lore.kernel.org/all/20240814121032.368444-2-felix.moessbauer@siemen… --- fs/proc/base.c | 9 +++++---- fs/select.c | 11 ++++------- kernel/sched/core.c | 8 ++++++++ kernel/sys.c | 2 ++ kernel/time/hrtimer.c | 18 +++--------------- 5 files changed, 22 insertions(+), 26 deletions(-) diff --git a/fs/proc/base.c b/fs/proc/base.c index ecc45389ea793..82e4a8805bae6 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -2633,10 +2633,11 @@ static ssize_t timerslack_ns_write(struct file *file, const char __user *buf, } task_lock(p); - if (slack_ns == 0) - p->timer_slack_ns = p->default_timer_slack_ns; - else - p->timer_slack_ns = slack_ns; + if (task_is_realtime(p)) + slack_ns = 0; + else if (slack_ns == 0) + slack_ns = p->default_timer_slack_ns; + p->timer_slack_ns = slack_ns; task_unlock(p); out: diff --git a/fs/select.c b/fs/select.c index 3f730b8581f65..e66b6189845ea 100644 --- a/fs/select.c +++ b/fs/select.c @@ -77,19 +77,16 @@ u64 select_estimate_accuracy(struct timespec64 *tv) { u64 ret; struct timespec64 now; + u64 slack = current->timer_slack_ns; - /* - * Realtime tasks get a slack of 0 for obvious reasons. - */ - - if (rt_task(current)) + if (slack == 0) return 0; ktime_get_ts64(&now); now = timespec64_sub(*tv, now); ret = __estimate_accuracy(&now); - if (ret < current->timer_slack_ns) - return current->timer_slack_ns; + if (ret < slack) + return slack; return ret; } diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 0a483fd9f5de5..9be8a509b5f3f 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -7380,6 +7380,14 @@ static void __setscheduler_params(struct task_struct *p, else if (fair_policy(policy)) p->static_prio = NICE_TO_PRIO(attr->sched_nice); + /* rt-policy tasks do not have a timerslack */ + if (task_is_realtime(p)) { + p->timer_slack_ns = 0; + } else if (p->timer_slack_ns == 0) { + /* when switching back to non-rt policy, restore timerslack */ + p->timer_slack_ns = p->default_timer_slack_ns; + } + /* * __sched_setscheduler() ensures attr->sched_priority == 0 when * !rt_policy. Always setting this ensures that things like diff --git a/kernel/sys.c b/kernel/sys.c index d06eda1387b69..06a9a87a8d3e0 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -2477,6 +2477,8 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, error = current->timer_slack_ns; break; case PR_SET_TIMERSLACK: + if (task_is_realtime(current)) + break; if (arg2 <= 0) current->timer_slack_ns = current->default_timer_slack_ns; diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index 8db65e2db14c7..f6d799646dd9c 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -2090,14 +2090,9 @@ long hrtimer_nanosleep(ktime_t rqtp, const enum hrtimer_mode mode, struct restart_block *restart; struct hrtimer_sleeper t; int ret = 0; - u64 slack; - - slack = current->timer_slack_ns; - if (dl_task(current) || rt_task(current)) - slack = 0; hrtimer_init_sleeper_on_stack(&t, clockid, mode); - hrtimer_set_expires_range_ns(&t.timer, rqtp, slack); + hrtimer_set_expires_range_ns(&t.timer, rqtp, current->timer_slack_ns); ret = do_nanosleep(&t, mode); if (ret != -ERESTART_RESTARTBLOCK) goto out; @@ -2278,7 +2273,7 @@ void __init hrtimers_init(void) /** * schedule_hrtimeout_range_clock - sleep until timeout * @expires: timeout value (ktime_t) - * @delta: slack in expires timeout (ktime_t) for SCHED_OTHER tasks + * @delta: slack in expires timeout (ktime_t) * @mode: timer mode * @clock_id: timer clock to be used */ @@ -2305,13 +2300,6 @@ schedule_hrtimeout_range_clock(ktime_t *expires, u64 delta, return -EINTR; } - /* - * Override any slack passed by the user if under - * rt contraints. - */ - if (rt_task(current)) - delta = 0; - hrtimer_init_sleeper_on_stack(&t, clock_id, mode); hrtimer_set_expires_range_ns(&t.timer, *expires, delta); hrtimer_sleeper_start_expires(&t, mode); @@ -2331,7 +2319,7 @@ EXPORT_SYMBOL_GPL(schedule_hrtimeout_range_clock); /** * schedule_hrtimeout_range - sleep until timeout * @expires: timeout value (ktime_t) - * @delta: slack in expires timeout (ktime_t) for SCHED_OTHER tasks + * @delta: slack in expires timeout (ktime_t) * @mode: timer mode * * Make the current task sleep until the given expiry time has -- 2.47.2

4 months

2
1
0 0

[PATCH 5.4 0/4] sctp: sysctl: fix argument passed to container_of

by Magali Lemes

Patches "sctp: sysctl: cookie_hmac_alg: avoid using current->nsproxy" and "sctp: sysctl: auth_enable: avoid using current->nsproxy" have been mixed up when backported to 5.4. The `member` argument passed to `container_of` has been swapped in both proc_sctp_do_auth() and proc_sctp_do_hmac_alg(). For instance, accessing /proc/sys/net/sctp/cookie_hmac_alg can now cause a kernel oops. Fix this by reverting the wrong backports and re-applying them correctly. Magali Lemes (2): Revert "sctp: sysctl: cookie_hmac_alg: avoid using current->nsproxy" Revert "sctp: sysctl: auth_enable: avoid using current->nsproxy" Matthieu Baerts (NGI0) (2): sctp: sysctl: cookie_hmac_alg: avoid using current->nsproxy sctp: sysctl: auth_enable: avoid using current->nsproxy net/sctp/sysctl.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) -- 2.48.1

4 months

2
8
0 0

[PATCH stable 5.15] lib/buildid: Handle memfd_secret() files in build_id_parse()

by Chen Linxuan

Backport of a similar change from commit 5ac9b4e935df ("lib/buildid: Handle memfd_secret() files in build_id_parse()") to address an issue where accessing secret memfd contents through build_id_parse() would trigger faults. Original report and repro can be found in [0]. [0] https://lore.kernel.org/bpf/ZwyG8Uro%2FSyTXAni@ly-workstation/ This repro will cause BUG: unable to handle kernel paging request in build_id_parse in 5.15/6.1/6.6. Some other discussions can be found in [1]. [1] https://lore.kernel.org/bpf/20241104175256.2327164-1-jolsa@kernel.org/T/#u Cc: stable(a)vger.kernel.org Fixes: 88a16a130933 ("perf: Add build id data in mmap2 event") Signed-off-by: Chen Linxuan <chenlinxuan(a)deepin.org> --- lib/buildid.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/lib/buildid.c b/lib/buildid.c index 9fc46366597e..9db35305f257 100644 --- a/lib/buildid.c +++ b/lib/buildid.c @@ -5,6 +5,7 @@ #include <linux/elf.h> #include <linux/kernel.h> #include <linux/pagemap.h> +#include <linux/secretmem.h> #define BUILD_ID 3 @@ -157,6 +158,12 @@ int build_id_parse(struct vm_area_struct *vma, unsigned char *build_id, if (!vma->vm_file) return -EINVAL; +#ifdef CONFIG_SECRETMEM + /* reject secretmem folios created with memfd_secret() */ + if (vma->vm_file->f_mapping->a_ops == &secretmem_aops) + return -EFAULT; +#endif + page = find_get_page(vma->vm_file->f_mapping, 0); if (!page) return -EFAULT; /* page not mapped */ -- 2.48.1

4 months

2
1
0 0

[PATCH 6.6.y 1/1] hrtimer: Use and report correct timerslack values for realtime tasks

by Felix Moessbauer

commit ed4fb6d7ef68111bb539283561953e5c6e9a6e38 upstream. The timerslack_ns setting is used to specify how much the hardware timers should be delayed, to potentially dispatch multiple timers in a single interrupt. This is a performance optimization. Timers of realtime tasks (having a realtime scheduling policy) should not be delayed. This logic was inconsitently applied to the hrtimers, leading to delays of realtime tasks which used timed waits for events (e.g. condition variables). Due to the downstream override of the slack for rt tasks, the procfs reported incorrect (non-zero) timerslack_ns values. This is changed by setting the timer_slack_ns task attribute to 0 for all tasks with a rt policy. By that, downstream users do not need to specially handle rt tasks (w.r.t. the slack), and the procfs entry shows the correct value of "0". Setting non-zero slack values (either via procfs or PR_SET_TIMERSLACK) on tasks with a rt policy is ignored, as stated in "man 2 PR_SET_TIMERSLACK": Timer slack is not applied to threads that are scheduled under a real-time scheduling policy (see sched_setscheduler(2)). The special handling of timerslack on rt tasks in downstream users is removed as well. Signed-off-by: Felix Moessbauer <felix.moessbauer(a)siemens.com> Signed-off-by: Thomas Gleixner <tglx(a)linutronix.de> Link: https://lore.kernel.org/all/20240814121032.368444-2-felix.moessbauer@siemen… --- fs/proc/base.c | 9 +++++---- fs/select.c | 11 ++++------- kernel/sched/core.c | 8 ++++++++ kernel/sys.c | 2 ++ kernel/time/hrtimer.c | 18 +++--------------- 5 files changed, 22 insertions(+), 26 deletions(-) diff --git a/fs/proc/base.c b/fs/proc/base.c index 699f085d4de7d..91fe20b7657c0 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -2633,10 +2633,11 @@ static ssize_t timerslack_ns_write(struct file *file, const char __user *buf, } task_lock(p); - if (slack_ns == 0) - p->timer_slack_ns = p->default_timer_slack_ns; - else - p->timer_slack_ns = slack_ns; + if (task_is_realtime(p)) + slack_ns = 0; + else if (slack_ns == 0) + slack_ns = p->default_timer_slack_ns; + p->timer_slack_ns = slack_ns; task_unlock(p); out: diff --git a/fs/select.c b/fs/select.c index 3f730b8581f65..e66b6189845ea 100644 --- a/fs/select.c +++ b/fs/select.c @@ -77,19 +77,16 @@ u64 select_estimate_accuracy(struct timespec64 *tv) { u64 ret; struct timespec64 now; + u64 slack = current->timer_slack_ns; - /* - * Realtime tasks get a slack of 0 for obvious reasons. - */ - - if (rt_task(current)) + if (slack == 0) return 0; ktime_get_ts64(&now); now = timespec64_sub(*tv, now); ret = __estimate_accuracy(&now); - if (ret < current->timer_slack_ns) - return current->timer_slack_ns; + if (ret < slack) + return slack; return ret; } diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 784a4f8409453..3d6dc03e4a7d3 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -7530,6 +7530,14 @@ static void __setscheduler_params(struct task_struct *p, else if (fair_policy(policy)) p->static_prio = NICE_TO_PRIO(attr->sched_nice); + /* rt-policy tasks do not have a timerslack */ + if (task_is_realtime(p)) { + p->timer_slack_ns = 0; + } else if (p->timer_slack_ns == 0) { + /* when switching back to non-rt policy, restore timerslack */ + p->timer_slack_ns = p->default_timer_slack_ns; + } + /* * __sched_setscheduler() ensures attr->sched_priority == 0 when * !rt_policy. Always setting this ensures that things like diff --git a/kernel/sys.c b/kernel/sys.c index 44b5759903332..355de0b65c235 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -2535,6 +2535,8 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, error = current->timer_slack_ns; break; case PR_SET_TIMERSLACK: + if (task_is_realtime(current)) + break; if (arg2 <= 0) current->timer_slack_ns = current->default_timer_slack_ns; diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index e99b1305e1a5f..5db6912b8f6e1 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -2093,14 +2093,9 @@ long hrtimer_nanosleep(ktime_t rqtp, const enum hrtimer_mode mode, struct restart_block *restart; struct hrtimer_sleeper t; int ret = 0; - u64 slack; - - slack = current->timer_slack_ns; - if (rt_task(current)) - slack = 0; hrtimer_init_sleeper_on_stack(&t, clockid, mode); - hrtimer_set_expires_range_ns(&t.timer, rqtp, slack); + hrtimer_set_expires_range_ns(&t.timer, rqtp, current->timer_slack_ns); ret = do_nanosleep(&t, mode); if (ret != -ERESTART_RESTARTBLOCK) goto out; @@ -2281,7 +2276,7 @@ void __init hrtimers_init(void) /** * schedule_hrtimeout_range_clock - sleep until timeout * @expires: timeout value (ktime_t) - * @delta: slack in expires timeout (ktime_t) for SCHED_OTHER tasks + * @delta: slack in expires timeout (ktime_t) * @mode: timer mode * @clock_id: timer clock to be used */ @@ -2308,13 +2303,6 @@ schedule_hrtimeout_range_clock(ktime_t *expires, u64 delta, return -EINTR; } - /* - * Override any slack passed by the user if under - * rt contraints. - */ - if (rt_task(current)) - delta = 0; - hrtimer_init_sleeper_on_stack(&t, clock_id, mode); hrtimer_set_expires_range_ns(&t.timer, *expires, delta); hrtimer_sleeper_start_expires(&t, mode); @@ -2334,7 +2322,7 @@ EXPORT_SYMBOL_GPL(schedule_hrtimeout_range_clock); /** * schedule_hrtimeout_range - sleep until timeout * @expires: timeout value (ktime_t) - * @delta: slack in expires timeout (ktime_t) for SCHED_OTHER tasks + * @delta: slack in expires timeout (ktime_t) * @mode: timer mode * * Make the current task sleep until the given expiry time has -- 2.47.2

4 months

2
1
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror