- Kernel - mailweb.openeuler.org

[PATCH OLK-6.6] sched/soft_domain: Increasing NR_MAX_CLUSTER to 32
by Zhang Qiao 25 Feb '26

25 Feb '26

hulk inclusion category: bugfix bugzilla: https://gitcode.com/openeuler/kernel/issues/8540 -------------------------------- Bumped the maximum number of clusters per node from 16 to 32 to resolve an out-of-bounds/array overflow issue. Fixes: 645a1ba256ef ("sched: topology: Build soft domain for LLC") Signed-off-by: Zhang Qiao <zhangqiao22(a)huawei.com> --- kernel/sched/soft_domain.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/kernel/sched/soft_domain.c b/kernel/sched/soft_domain.c index 3c52680ca82f..7b569fdb2c6e 100644 --- a/kernel/sched/soft_domain.c +++ b/kernel/sched/soft_domain.c @@ -17,6 +17,8 @@ #include <linux/sort.h> +#define NR_MAX_CLUSTER 32 + static DEFINE_STATIC_KEY_TRUE(__soft_domain_switch); static int __init soft_domain_switch_setup(char *str) @@ -50,6 +52,7 @@ static int build_soft_sub_domain(int nid, struct cpumask *cpus) const struct cpumask *span = cpumask_of_node(nid); struct soft_domain *sf_d = NULL; int i; + int cls_cnt = 0; sf_d = kzalloc_node(sizeof(struct soft_domain) + cpumask_size(), GFP_KERNEL, nid); @@ -63,6 +66,12 @@ static int build_soft_sub_domain(int nid, struct cpumask *cpus) for_each_cpu_and(i, span, cpus) { struct soft_subdomain *sub_d = NULL; + cls_cnt++; + if (cls_cnt > NR_MAX_CLUSTER) { + pr_info("clsuter number > %d, unsupport soft domain.\n", NR_MAX_CLUSTER); + return -EINVAL; + } + sub_d = kzalloc_node(sizeof(struct soft_subdomain) + cpumask_size(), GFP_KERNEL, nid); if (!sub_d) { @@ -138,8 +147,6 @@ void build_soft_domain(void) static DEFINE_MUTEX(soft_domain_mutex); -#define NR_MAX_CLUSTER 16 - struct domain_node { struct soft_subdomain *sud_d; unsigned int attached; -- 2.18.0.huawei.25

1 0

[PATCH openEuler-1.0-LTS] squashfs: fix memory leak in squashfs_fill_super
by Long Li 25 Feb '26

25 Feb '26

From: Phillip Lougher <phillip(a)squashfs.org.uk> stable inclusion from stable-v5.10.240 commit 9bdb2e7f9b11a1c9bb7096d5726fe5a738fa5d73 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/9383 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=… -------------------------------- commit b64700d41bdc4e9f82f1346c15a3678ebb91a89c upstream. If sb_min_blocksize returns 0, squashfs_fill_super exits without freeing allocated memory (sb->s_fs_info). Fix this by moving the call to sb_min_blocksize to before memory is allocated. Link: https://lkml.kernel.org/r/20250811223740.110392-1-phillip@squashfs.org.uk Fixes: 734aa85390ea ("Squashfs: check return result of sb_min_blocksize") Signed-off-by: Phillip Lougher <phillip(a)squashfs.org.uk> Reported-by: Scott GUO <scottzhguo(a)tencent.com> Closes: https://lore.kernel.org/all/20250811061921.3807353-1-scott_gzh@163.com Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Conflicts: fs/squashfs/super.c [Context conflicts] Signed-off-by: Long Li <leo.lilong(a)huawei.com> --- fs/squashfs/super.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/fs/squashfs/super.c b/fs/squashfs/super.c index 41756ca57ff2..119c0d63dafc 100644 --- a/fs/squashfs/super.c +++ b/fs/squashfs/super.c @@ -85,10 +85,15 @@ static int squashfs_fill_super(struct super_block *sb, void *data, int silent) unsigned short flags; unsigned int fragments; u64 lookup_table_start, xattr_id_table_start, next_table; - int err; + int err, devblksize = sb_min_blocksize(sb, SQUASHFS_DEVBLK_SIZE); TRACE("Entered squashfs_fill_superblock\n"); + if (!devblksize) { + ERROR("squashfs: unable to set blocksize\n"); + return -EINVAL; + } + sb->s_fs_info = kzalloc(sizeof(*msblk), GFP_KERNEL); if (sb->s_fs_info == NULL) { ERROR("Failed to allocate squashfs_sb_info\n"); @@ -96,12 +101,7 @@ static int squashfs_fill_super(struct super_block *sb, void *data, int silent) } msblk = sb->s_fs_info; - msblk->devblksize = sb_min_blocksize(sb, SQUASHFS_DEVBLK_SIZE); - if (!msblk->devblksize) { - ERROR("squashfs: unable to set blocksize\n"); - return -EINVAL; - } - + msblk->devblksize = devblksize; msblk->devblksize_log2 = ffz(~msblk->devblksize); mutex_init(&msblk->meta_index_mutex); -- 2.39.2

2 1

[PATCH OLK-6.6] efivarfs: fix error propagation in efivar_entry_get()
by Yongjian Sun 25 Feb '26

25 Feb '26

From: Kohei Enju <kohei(a)enjuk.jp> mainline inclusion from mainline-v6.19-rc8 commit 4b22ec1685ce1fc0d862dcda3225d852fb107995 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/13706 CVE: CVE-2026-23156 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- efivar_entry_get() always returns success even if the underlying __efivar_entry_get() fails, masking errors. This may result in uninitialized heap memory being copied to userspace in the efivarfs_file_read() path. Fix it by returning the error from __efivar_entry_get(). Fixes: 2d82e6227ea1 ("efi: vars: Move efivar caching layer into efivarfs") Cc: <stable(a)vger.kernel.org> # v6.1+ Signed-off-by: Kohei Enju <kohei(a)enjuk.jp> Signed-off-by: Ard Biesheuvel <ardb(a)kernel.org> Signed-off-by: Yongjian Sun <sunyongjian1(a)huawei.com> --- fs/efivarfs/vars.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/efivarfs/vars.c b/fs/efivarfs/vars.c index 13bc60698955..b10aa5afd7f7 100644 --- a/fs/efivarfs/vars.c +++ b/fs/efivarfs/vars.c @@ -609,7 +609,7 @@ int efivar_entry_get(struct efivar_entry *entry, u32 *attributes, err = __efivar_entry_get(entry, attributes, size, data); efivar_unlock(); - return 0; + return err; } /** -- 2.39.2

2 1

[PATCH OLK-6.6] io_uring/io-wq: check IO_WQ_BIT_EXIT inside work run loop
by Yongjian Sun 25 Feb '26

25 Feb '26

From: Jens Axboe <axboe(a)kernel.dk> mainline inclusion from mainline-v6.19-rc7 commit 10dc959398175736e495f71c771f8641e1ca1907 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/13663 CVE: CVE-2026-23113 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Currently this is checked before running the pending work. Normally this is quite fine, as work items either end up blocking (which will create a new worker for other items), or they complete fairly quickly. But syzbot reports an issue where io-wq takes seemingly forever to exit, and with a bit of debugging, this turns out to be because it queues a bunch of big (2GB - 4096b) reads with a /dev/msr* file. Since this file type doesn't support ->read_iter(), loop_rw_iter() ends up handling them. Each read returns 16MB of data read, which takes 20 (!!) seconds. With a bunch of these pending, processing the whole chain can take a long time. Easily longer than the syzbot uninterruptible sleep timeout of 140 seconds. This then triggers a complaint off the io-wq exit path: INFO: task syz.4.135:6326 blocked for more than 143 seconds. Not tainted syzkaller #0 Blocked by coredump. "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. task:syz.4.135 state:D stack:26824 pid:6326 tgid:6324 ppid:5957 task_flags:0x400548 flags:0x00080000 Call Trace: <TASK> context_switch kernel/sched/core.c:5256 [inline] __schedule+0x1139/0x6150 kernel/sched/core.c:6863 __schedule_loop kernel/sched/core.c:6945 [inline] schedule+0xe7/0x3a0 kernel/sched/core.c:6960 schedule_timeout+0x257/0x290 kernel/time/sleep_timeout.c:75 do_wait_for_common kernel/sched/completion.c:100 [inline] __wait_for_common+0x2fc/0x4e0 kernel/sched/completion.c:121 io_wq_exit_workers io_uring/io-wq.c:1328 [inline] io_wq_put_and_exit+0x271/0x8a0 io_uring/io-wq.c:1356 io_uring_clean_tctx+0x10d/0x190 io_uring/tctx.c:203 io_uring_cancel_generic+0x69c/0x9a0 io_uring/cancel.c:651 io_uring_files_cancel include/linux/io_uring.h:19 [inline] do_exit+0x2ce/0x2bd0 kernel/exit.c:911 do_group_exit+0xd3/0x2a0 kernel/exit.c:1112 get_signal+0x2671/0x26d0 kernel/signal.c:3034 arch_do_signal_or_restart+0x8f/0x7e0 arch/x86/kernel/signal.c:337 __exit_to_user_mode_loop kernel/entry/common.c:41 [inline] exit_to_user_mode_loop+0x8c/0x540 kernel/entry/common.c:75 __exit_to_user_mode_prepare include/linux/irq-entry-common.h:226 [inline] syscall_exit_to_user_mode_prepare include/linux/irq-entry-common.h:256 [inline] syscall_exit_to_user_mode_work include/linux/entry-common.h:159 [inline] syscall_exit_to_user_mode include/linux/entry-common.h:194 [inline] do_syscall_64+0x4ee/0xf80 arch/x86/entry/syscall_64.c:100 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7fa02738f749 RSP: 002b:00007fa0281ae0e8 EFLAGS: 00000246 ORIG_RAX: 00000000000000ca RAX: fffffffffffffe00 RBX: 00007fa0275e6098 RCX: 00007fa02738f749 RDX: 0000000000000000 RSI: 0000000000000080 RDI: 00007fa0275e6098 RBP: 00007fa0275e6090 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000 R13: 00007fa0275e6128 R14: 00007fff14e4fcb0 R15: 00007fff14e4fd98 There's really nothing wrong here, outside of processing these reads will take a LONG time. However, we can speed up the exit by checking the IO_WQ_BIT_EXIT inside the io_worker_handle_work() loop, as syzbot will exit the ring after queueing up all of these reads. Then once the first item is processed, io-wq will simply cancel the rest. That should avoid syzbot running into this complaint again. Cc: stable(a)vger.kernel.org Link: https://lore.kernel.org/all/68a2decc.050a0220.e29e5.0099.GAE@google.com/ Reported-by: syzbot+4eb282331cab6d5b6588(a)syzkaller.appspotmail.com Signed-off-by: Jens Axboe <axboe(a)kernel.dk> Signed-off-by: Yongjian Sun <sunyongjian1(a)huawei.com> --- io_uring/io-wq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/io_uring/io-wq.c b/io_uring/io-wq.c index 1c4ef4e4eb52..c848a5018d12 100644 --- a/io_uring/io-wq.c +++ b/io_uring/io-wq.c @@ -554,9 +554,9 @@ static void io_worker_handle_work(struct io_wq_acct *acct, __releases(&acct->lock) { struct io_wq *wq = worker->wq; - bool do_kill = test_bit(IO_WQ_BIT_EXIT, &wq->state); do { + bool do_kill = test_bit(IO_WQ_BIT_EXIT, &wq->state); struct io_wq_work *work; /* -- 2.39.2

2 1

[PATCH OLK-6.6] dmaengine: mmp_pdma: Fix race condition in mmp_pdma_residue()
by Zhang Yuwei 25 Feb '26

25 Feb '26

From: Guodong Xu <guodong(a)riscstar.com> mainline inclusion from mainline-v6.19-rc6 commit a143545855bc2c6e1330f6f57ae375ac44af00a7 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/13727 CVE: CVE-2025-71221 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Add proper locking in mmp_pdma_residue() to prevent use-after-free when accessing descriptor list and descriptor contents. The race occurs when multiple threads call tx_status() while the tasklet on another CPU is freeing completed descriptors: CPU 0 CPU 1 ----- ----- mmp_pdma_tx_status() mmp_pdma_residue() -> NO LOCK held list_for_each_entry(sw, ..) DMA interrupt dma_do_tasklet() -> spin_lock(&desc_lock) list_move(sw->node, ...) spin_unlock(&desc_lock) | dma_pool_free(sw) <- FREED! -> access sw->desc <- UAF! This issue can be reproduced when running dmatest on the same channel with multiple threads (threads_per_chan > 1). Fix by protecting the chain_running list iteration and descriptor access with the chan->desc_lock spinlock. Signed-off-by: Juan Li <lijuan(a)linux.spacemit.com> Signed-off-by: Guodong Xu <guodong(a)riscstar.com> Link: https://patch.msgid.link/20251216-mmp-pdma-race-v1-1-976a224bb622@riscstar.… Signed-off-by: Vinod Koul <vkoul(a)kernel.org> Conflicts: drivers/dma/mmp_pdma.c [commit 35e40bf761fcb not merged] Signed-off-by: Zhang Yuwei <zhangyuwei20(a)huawei.com> --- drivers/dma/mmp_pdma.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/dma/mmp_pdma.c b/drivers/dma/mmp_pdma.c index ebdfdcbb4f7a..3a035a31bec5 100644 --- a/drivers/dma/mmp_pdma.c +++ b/drivers/dma/mmp_pdma.c @@ -763,6 +763,7 @@ static unsigned int mmp_pdma_residue(struct mmp_pdma_chan *chan, dma_cookie_t cookie) { struct mmp_pdma_desc_sw *sw; + unsigned long flags; u32 curr, residue = 0; bool passed = false; bool cyclic = chan->cyclic_first != NULL; @@ -779,6 +780,8 @@ static unsigned int mmp_pdma_residue(struct mmp_pdma_chan *chan, else curr = readl(chan->phy->base + DSADR(chan->phy->idx)); + spin_lock_irqsave(&chan->desc_lock, flags); + list_for_each_entry(sw, &chan->chain_running, node) { u32 start, end, len; @@ -822,6 +825,7 @@ static unsigned int mmp_pdma_residue(struct mmp_pdma_chan *chan, continue; if (sw->async_tx.cookie == cookie) { + spin_unlock_irqrestore(&chan->desc_lock, flags); return residue; } else { residue = 0; @@ -829,6 +833,8 @@ static unsigned int mmp_pdma_residue(struct mmp_pdma_chan *chan, } } + spin_unlock_irqrestore(&chan->desc_lock, flags); + /* We should only get here in case of cyclic transactions */ return residue; } -- 2.22.0

2 1

[PATCH OLK-5.10] dmaengine: mmp_pdma: Fix race condition in mmp_pdma_residue()
by Zhang Yuwei 25 Feb '26

25 Feb '26

From: Guodong Xu <guodong(a)riscstar.com> mainline inclusion from mainline-v6.19-rc6 commit a143545855bc2c6e1330f6f57ae375ac44af00a7 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/13727 CVE: CVE-2025-71221 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Add proper locking in mmp_pdma_residue() to prevent use-after-free when accessing descriptor list and descriptor contents. The race occurs when multiple threads call tx_status() while the tasklet on another CPU is freeing completed descriptors: CPU 0 CPU 1 ----- ----- mmp_pdma_tx_status() mmp_pdma_residue() -> NO LOCK held list_for_each_entry(sw, ..) DMA interrupt dma_do_tasklet() -> spin_lock(&desc_lock) list_move(sw->node, ...) spin_unlock(&desc_lock) | dma_pool_free(sw) <- FREED! -> access sw->desc <- UAF! This issue can be reproduced when running dmatest on the same channel with multiple threads (threads_per_chan > 1). Fix by protecting the chain_running list iteration and descriptor access with the chan->desc_lock spinlock. Signed-off-by: Juan Li <lijuan(a)linux.spacemit.com> Signed-off-by: Guodong Xu <guodong(a)riscstar.com> Link: https://patch.msgid.link/20251216-mmp-pdma-race-v1-1-976a224bb622@riscstar.… Signed-off-by: Vinod Koul <vkoul(a)kernel.org> Conflicts: drivers/dma/mmp_pdma.c [commit 35e40bf761fcb not merged] Signed-off-by: Zhang Yuwei <zhangyuwei20(a)huawei.com> --- drivers/dma/mmp_pdma.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/dma/mmp_pdma.c b/drivers/dma/mmp_pdma.c index 4eb63f1ad224..c56cb954b392 100644 --- a/drivers/dma/mmp_pdma.c +++ b/drivers/dma/mmp_pdma.c @@ -764,6 +764,7 @@ static unsigned int mmp_pdma_residue(struct mmp_pdma_chan *chan, dma_cookie_t cookie) { struct mmp_pdma_desc_sw *sw; + unsigned long flags; u32 curr, residue = 0; bool passed = false; bool cyclic = chan->cyclic_first != NULL; @@ -780,6 +781,8 @@ static unsigned int mmp_pdma_residue(struct mmp_pdma_chan *chan, else curr = readl(chan->phy->base + DSADR(chan->phy->idx)); + spin_lock_irqsave(&chan->desc_lock, flags); + list_for_each_entry(sw, &chan->chain_running, node) { u32 start, end, len; @@ -823,6 +826,7 @@ static unsigned int mmp_pdma_residue(struct mmp_pdma_chan *chan, continue; if (sw->async_tx.cookie == cookie) { + spin_unlock_irqrestore(&chan->desc_lock, flags); return residue; } else { residue = 0; @@ -830,6 +834,8 @@ static unsigned int mmp_pdma_residue(struct mmp_pdma_chan *chan, } } + spin_unlock_irqrestore(&chan->desc_lock, flags); + /* We should only get here in case of cyclic transactions */ return residue; } -- 2.22.0

2 1

[PATCH] Prefer steal task within LLC and fallback to native newidle_balance on failure
by Chen Jinghuang 25 Feb '26

25 Feb '26

--- kernel/sched/fair.c | 15 +++++++++++++-- kernel/sched/features.h | 1 + 2 files changed, 14 insertions(+), 2 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index e60f19cb0fee..70ba6ee3c058 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -10343,9 +10343,15 @@ done: __maybe_unused; */ rq_idle_stamp_update(rq); - new_tasks = newidle_balance(rq, rf); - if (new_tasks == 0) + if (sched_feat(STEALPOC)) { new_tasks = try_steal(rq, rf); + if (new_tasks == 0) + new_tasks = newidle_balance(rq, rf); + } else { + new_tasks = newidle_balance(rq, rf); + if (new_tasks == 0) + new_tasks = try_steal(rq, rf); + } schedstat_end_time(rq, time); if (new_tasks) @@ -14313,6 +14319,11 @@ static int newidle_balance(struct rq *this_rq, struct rq_flags *rf) int continue_balancing = 1; u64 domain_cost; + if (sched_feat(STEALPOC)) { + if (sd->flags & SD_SHARE_PKG_RESOURCES) + continue; + } + update_next_balance(sd, &next_balance); if (this_rq->avg_idle < curr_cost + sd->max_newidle_lb_cost) diff --git a/kernel/sched/features.h b/kernel/sched/features.h index 1f665fbf0137..c74329fd86b9 100644 --- a/kernel/sched/features.h +++ b/kernel/sched/features.h @@ -63,6 +63,7 @@ SCHED_FEAT(SIS_UTIL, true) * Improves CPU utilization. */ SCHED_FEAT(STEAL, false) +SCHED_FEAT(STEALPOC, false) #endif #ifdef CONFIG_SCHED_PARAL -- 2.34.1

1 0

[PATCH OLK-6.6 0/2] CVE-2026-23094
by Liu Kai 25 Feb '26

25 Feb '26

CVE-2026-23094 Chenghai Huang (1): uacce: fix isolate sysfs check condition Liu Kai (1): Revert "uacce: fix isolate sysfs check condition" drivers/misc/uacce/uacce.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) -- 2.34.1

2 3

[PATCH OLK-5.10] migrate: correct lock ordering for hugetlb file folios
by Jinjiang Tu 24 Feb '26

24 Feb '26

From: "Matthew Wilcox (Oracle)" <willy(a)infradead.org> stable inclusion from stable-v5.10.249 commit e7396d23f9d5739f56cf9ab430c3a169f5508394 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/13630 CVE: CVE-2026-23097 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit b7880cb166ab62c2409046b2347261abf701530e upstream. Syzbot has found a deadlock (analyzed by Lance Yang): 1) Task (5749): Holds folio_lock, then tries to acquire i_mmap_rwsem(read lock). 2) Task (5754): Holds i_mmap_rwsem(write lock), then tries to acquire folio_lock. migrate_pages() -> migrate_hugetlbs() -> unmap_and_move_huge_page() <- Takes folio_lock! -> remove_migration_ptes() -> __rmap_walk_file() -> i_mmap_lock_read() <- Waits for i_mmap_rwsem(read lock)! hugetlbfs_fallocate() -> hugetlbfs_punch_hole() <- Takes i_mmap_rwsem(write lock)! -> hugetlbfs_zero_partial_page() -> filemap_lock_hugetlb_folio() -> filemap_lock_folio() -> __filemap_get_folio <- Waits for folio_lock! The migration path is the one taking locks in the wrong order according to the documentation at the top of mm/rmap.c. So expand the scope of the existing i_mmap_lock to cover the calls to remove_migration_ptes() too. This is (mostly) how it used to be after commit c0d0381ade79. That was removed by 336bf30eb765 for both file & anon hugetlb pages when it should only have been removed for anon hugetlb pages. Link: https://lkml.kernel.org/r/20260109041345.3863089-2-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) <willy(a)infradead.org> Fixes: 336bf30eb765 ("hugetlbfs: fix anon huge page migration race") Reported-by: syzbot+2d9c96466c978346b55f(a)syzkaller.appspotmail.com Link: https://lore.kernel.org/all/68e9715a.050a0220.1186a4.000d.GAE@google.com Debugged-by: Lance Yang <lance.yang(a)linux.dev> Acked-by: David Hildenbrand (Red Hat) <david(a)kernel.org> Acked-by: Zi Yan <ziy(a)nvidia.com> Cc: Alistair Popple <apopple(a)nvidia.com> Cc: Byungchul Park <byungchul(a)sk.com> Cc: Gregory Price <gourry(a)gourry.net> Cc: Jann Horn <jannh(a)google.com> Cc: Joshua Hahn <joshua.hahnjy(a)gmail.com> Cc: Liam Howlett <liam.howlett(a)oracle.com> Cc: Lorenzo Stoakes <lorenzo.stoakes(a)oracle.com> Cc: Matthew Brost <matthew.brost(a)intel.com> Cc: Rakie Kim <rakie.kim(a)sk.com> Cc: Rik van Riel <riel(a)surriel.com> Cc: Vlastimil Babka <vbabka(a)suse.cz> Cc: Ying Huang <ying.huang(a)linux.alibaba.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Jinjiang Tu <tujinjiang(a)huawei.com> --- mm/migrate.c | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) diff --git a/mm/migrate.c b/mm/migrate.c index cf8c05ea821e..7dc6846af90e 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -1342,6 +1342,7 @@ static int unmap_and_move_huge_page(new_page_t get_new_page, struct page *new_hpage; struct anon_vma *anon_vma = NULL; struct address_space *mapping = NULL; + enum ttu_flags ttu = TTU_MIGRATION|TTU_IGNORE_MLOCK; /* * Migratability of hugepages depends on architectures and their size. @@ -1389,9 +1390,6 @@ static int unmap_and_move_huge_page(new_page_t get_new_page, goto put_anon; if (page_mapped(hpage)) { - bool mapping_locked = false; - enum ttu_flags ttu = TTU_MIGRATION|TTU_IGNORE_MLOCK; - if (!PageAnon(hpage)) { /* * In shared mappings, try_to_unmap could potentially @@ -1403,15 +1401,11 @@ static int unmap_and_move_huge_page(new_page_t get_new_page, if (unlikely(!mapping)) goto unlock_put_anon; - mapping_locked = true; ttu |= TTU_RMAP_LOCKED; } try_to_unmap(hpage, ttu); page_was_mapped = 1; - - if (mapping_locked) - i_mmap_unlock_write(mapping); } if (!page_mapped(hpage)) @@ -1419,7 +1413,11 @@ static int unmap_and_move_huge_page(new_page_t get_new_page, if (page_was_mapped) remove_migration_ptes(hpage, - rc == MIGRATEPAGE_SUCCESS ? new_hpage : hpage, false); + rc == MIGRATEPAGE_SUCCESS ? new_hpage : hpage, + (ttu & TTU_RMAP_LOCKED) ? true : false); + + if (ttu & TTU_RMAP_LOCKED) + i_mmap_unlock_write(mapping); unlock_put_anon: unlock_page(new_hpage); -- 2.43.0

2 1

[PATCH OLK-6.6] migrate: correct lock ordering for hugetlb file folios
by Jinjiang Tu 24 Feb '26

24 Feb '26

From: "Matthew Wilcox (Oracle)" <willy(a)infradead.org> stable inclusion from stable-v6.6.122 commit 526394af4e8ade89cacd1a9ce2b97712712fcc34 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/13630 CVE: CVE-2026-23097 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit b7880cb166ab62c2409046b2347261abf701530e upstream. Syzbot has found a deadlock (analyzed by Lance Yang): 1) Task (5749): Holds folio_lock, then tries to acquire i_mmap_rwsem(read lock). 2) Task (5754): Holds i_mmap_rwsem(write lock), then tries to acquire folio_lock. migrate_pages() -> migrate_hugetlbs() -> unmap_and_move_huge_page() <- Takes folio_lock! -> remove_migration_ptes() -> __rmap_walk_file() -> i_mmap_lock_read() <- Waits for i_mmap_rwsem(read lock)! hugetlbfs_fallocate() -> hugetlbfs_punch_hole() <- Takes i_mmap_rwsem(write lock)! -> hugetlbfs_zero_partial_page() -> filemap_lock_hugetlb_folio() -> filemap_lock_folio() -> __filemap_get_folio <- Waits for folio_lock! The migration path is the one taking locks in the wrong order according to the documentation at the top of mm/rmap.c. So expand the scope of the existing i_mmap_lock to cover the calls to remove_migration_ptes() too. This is (mostly) how it used to be after commit c0d0381ade79. That was removed by 336bf30eb765 for both file & anon hugetlb pages when it should only have been removed for anon hugetlb pages. Link: https://lkml.kernel.org/r/20260109041345.3863089-2-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) <willy(a)infradead.org> Fixes: 336bf30eb765 ("hugetlbfs: fix anon huge page migration race") Reported-by: syzbot+2d9c96466c978346b55f(a)syzkaller.appspotmail.com Link: https://lore.kernel.org/all/68e9715a.050a0220.1186a4.000d.GAE@google.com Debugged-by: Lance Yang <lance.yang(a)linux.dev> Acked-by: David Hildenbrand (Red Hat) <david(a)kernel.org> Acked-by: Zi Yan <ziy(a)nvidia.com> Cc: Alistair Popple <apopple(a)nvidia.com> Cc: Byungchul Park <byungchul(a)sk.com> Cc: Gregory Price <gourry(a)gourry.net> Cc: Jann Horn <jannh(a)google.com> Cc: Joshua Hahn <joshua.hahnjy(a)gmail.com> Cc: Liam Howlett <liam.howlett(a)oracle.com> Cc: Lorenzo Stoakes <lorenzo.stoakes(a)oracle.com> Cc: Matthew Brost <matthew.brost(a)intel.com> Cc: Rakie Kim <rakie.kim(a)sk.com> Cc: Rik van Riel <riel(a)surriel.com> Cc: Vlastimil Babka <vbabka(a)suse.cz> Cc: Ying Huang <ying.huang(a)linux.alibaba.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Conflicts: mm/migrate.c [Context conflicts due to 3f6171ebb378 ("mm: try skipping pinned page before unmapping in migration").] Signed-off-by: Jinjiang Tu <tujinjiang(a)huawei.com> --- mm/migrate.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/mm/migrate.c b/mm/migrate.c index 216f45c9570f..2ca451efaaad 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -1438,6 +1438,7 @@ static int unmap_and_move_huge_page(new_folio_t get_new_folio, int page_was_mapped = 0; struct anon_vma *anon_vma = NULL; struct address_space *mapping = NULL; + enum ttu_flags ttu = 0; if (folio_ref_count(src) == 1) { /* page was freed from under us. So we are done. */ @@ -1479,8 +1480,6 @@ static int unmap_and_move_huge_page(new_folio_t get_new_folio, goto put_anon; if (folio_mapped(src)) { - enum ttu_flags ttu = 0; - /* * Racy check folio is pinned, try to avoid unnecessary page * table unmapping and restoring, because the subsequent @@ -1505,9 +1504,6 @@ static int unmap_and_move_huge_page(new_folio_t get_new_folio, try_to_migrate(src, ttu); page_was_mapped = 1; - - if (ttu & TTU_RMAP_LOCKED) - i_mmap_unlock_write(mapping); } if (!folio_mapped(src)) @@ -1515,7 +1511,11 @@ static int unmap_and_move_huge_page(new_folio_t get_new_folio, if (page_was_mapped) remove_migration_ptes(src, - rc == MIGRATEPAGE_SUCCESS ? dst : src, false); + rc == MIGRATEPAGE_SUCCESS ? dst : src, + ttu ? true : false); + + if (ttu & TTU_RMAP_LOCKED) + i_mmap_unlock_write(mapping); unlock_put_anon: folio_unlock(dst); -- 2.43.0

2 1