- Kernel - mailweb.openeuler.org

[PATCH OLK-5.10 0/2] mainline patch submission for OLK-5.10
by Chen Yuxi 03 Sep '26

03 Sep '26

Zide Chen (2): perf/x86/intel/uncore: Keep PCI PMUs working when MMIO/MSR setup fails perf/x86/intel/uncore: Fix uncore_box ref/unref ordering arch/x86/events/intel/uncore.c | 48 ++++++++++++++-------------------- 1 file changed, 20 insertions(+), 28 deletions(-) -- 2.34.1

2 3

[PATCH OLK-6.6 0/2] mainline patch submission for OLK-6.6
by Chen Yuxi 03 Sep '26

03 Sep '26

Zide Chen (2): perf/x86/intel/uncore: Keep PCI PMUs working when MMIO/MSR setup fails perf/x86/intel/uncore: Fix uncore_box ref/unref ordering arch/x86/events/intel/uncore.c | 48 ++++++++++++++-------------------- 1 file changed, 20 insertions(+), 28 deletions(-) -- 2.34.1

2 3

[PATCH OLK-5.10 V1] sched/fair: Fix overflow in update_tg_cfs_runnable()
by Yao Yiqi 03 Sep '26

03 Sep '26

From: "Chen, Yu C" <yu.c.chen(a)intel.com> mainline inclusion from mainline-v7.3-rc1 commit 4f166adb5cb0525d9e32d45729fd8f28c80acbee category: bugfix bugzilla: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- A divide-by-zero crash is observed when running hackbench: [14697.488452] CPU: 112 UID: 0 PID: 124791 Comm: hackbench Not tainted 7.1.0-rc2+ [14697.492627] RIP: 0010:propagate_entity_load_avg+0x35f/0x3e0 [14697.506799] <TASK> [14697.507411] __dequeue_task+0x2b4/0xc70 [14697.508677] dequeue_task_fair+0x36/0x370 [14697.509047] dequeue_task+0x101/0x2f0 [14697.509426] __schedule+0x1b1/0x1a00 [14697.510868] anon_pipe_read+0x3da/0x450 [14697.511400] vfs_read+0x361/0x390 [14697.512053] __x64_sys_read+0x19/0x30 The divide-by-zero happens here: if (scale_load_down(gcfs_rq->load.weight)) { load_sum = div_u64(gcfs_rq->avg.load_sum, scale_load_down(gcfs_rq->load.weight)); } gcfs_rq->load.weight is an insane large value and is truncated to the lower 32 bits by div_u64, which happen to be 0. Using AI for investigation, the cause is a u32 overflow in update_tg_cfs_runnable(), and flat pickup became a victim when using tg_tasks(): u32 new_sum, divider; ... new_sum = se->avg.runnable_avg * divider; <-- boom The following sequence shows how this triggers the crash: propagate_entity_load_avg() update_tg_cfs_runnable() # u32 overflow corrupts runnable_sum __update_load_avg_cfs_rq() ___update_load_avg() # computes insane runnable_avg update_tg_load_avg() # propagates to tg->runnable_avg update_cfs_group() calc_concur_shares() tg_tasks() # long-to-int truncation, negative nr reweight_entity() # corrupted se->load.weight update_load_add() # corrupted cfs_rq->load.weight propagate_entity_load_avg() update_tg_cfs_load() div_u64() # divide-by-zero Fix by widening new_sum from u32 to u64 (no need to force tg_tasks() to return unsigned long after this fix) Fixes: 95246d1ec80b ("sched/pelt: Relax the sync of runnable_sum with runnable_avg") Assisted-by: Claude:claude-opus-4.6 Signed-off-by: Chen Yu <yu.c.chen(a)intel.com> Signed-off-by: Peter Zijlstra (Intel) <peterz(a)infradead.org> Link: https://patch.msgid.link/a22eea2b-4c4a-4623-9a44-d7b18c0c91c8@intel.com Signed-off-by: Yao Yiqi <yaoyiqi3(a)huawei.com> --- kernel/sched/fair.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index bcba09a9dffd..c40ee813cb15 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -3692,7 +3692,8 @@ static inline void update_tg_cfs_runnable(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq *gcfs_rq) { long delta_sum, delta_avg = gcfs_rq->avg.runnable_avg - se->avg.runnable_avg; - u32 new_sum, divider; + u64 new_sum; + u32 divider; /* Nothing to update */ if (!delta_avg) @@ -3706,7 +3707,7 @@ update_tg_cfs_runnable(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cf /* Set new sched_entity's runnable */ se->avg.runnable_avg = gcfs_rq->avg.runnable_avg; - new_sum = se->avg.runnable_avg * divider; + new_sum = (u64)se->avg.runnable_avg * divider; delta_sum = (long)new_sum - (long)se->avg.runnable_sum; se->avg.runnable_sum = new_sum; -- 2.34.1

2 1

[PATCH OLK-6.6 V1] sched/fair: Fix overflow in update_tg_cfs_runnable()
by Yao Yiqi 03 Sep '26

03 Sep '26

From: "Chen, Yu C" <yu.c.chen(a)intel.com> mainline inclusion from mainline-v7.3-rc1 commit 4f166adb5cb0525d9e32d45729fd8f28c80acbee category: bugfix bugzilla: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- A divide-by-zero crash is observed when running hackbench: [14697.488452] CPU: 112 UID: 0 PID: 124791 Comm: hackbench Not tainted 7.1.0-rc2+ [14697.492627] RIP: 0010:propagate_entity_load_avg+0x35f/0x3e0 [14697.506799] <TASK> [14697.507411] __dequeue_task+0x2b4/0xc70 [14697.508677] dequeue_task_fair+0x36/0x370 [14697.509047] dequeue_task+0x101/0x2f0 [14697.509426] __schedule+0x1b1/0x1a00 [14697.510868] anon_pipe_read+0x3da/0x450 [14697.511400] vfs_read+0x361/0x390 [14697.512053] __x64_sys_read+0x19/0x30 The divide-by-zero happens here: if (scale_load_down(gcfs_rq->load.weight)) { load_sum = div_u64(gcfs_rq->avg.load_sum, scale_load_down(gcfs_rq->load.weight)); } gcfs_rq->load.weight is an insane large value and is truncated to the lower 32 bits by div_u64, which happen to be 0. Using AI for investigation, the cause is a u32 overflow in update_tg_cfs_runnable(), and flat pickup became a victim when using tg_tasks(): u32 new_sum, divider; ... new_sum = se->avg.runnable_avg * divider; <-- boom The following sequence shows how this triggers the crash: propagate_entity_load_avg() update_tg_cfs_runnable() # u32 overflow corrupts runnable_sum __update_load_avg_cfs_rq() ___update_load_avg() # computes insane runnable_avg update_tg_load_avg() # propagates to tg->runnable_avg update_cfs_group() calc_concur_shares() tg_tasks() # long-to-int truncation, negative nr reweight_entity() # corrupted se->load.weight update_load_add() # corrupted cfs_rq->load.weight propagate_entity_load_avg() update_tg_cfs_load() div_u64() # divide-by-zero Fix by widening new_sum from u32 to u64 (no need to force tg_tasks() to return unsigned long after this fix) Fixes: 95246d1ec80b ("sched/pelt: Relax the sync of runnable_sum with runnable_avg") Assisted-by: Claude:claude-opus-4.6 Signed-off-by: Chen Yu <yu.c.chen(a)intel.com> Signed-off-by: Peter Zijlstra (Intel) <peterz(a)infradead.org> Link: https://patch.msgid.link/a22eea2b-4c4a-4623-9a44-d7b18c0c91c8@intel.com Signed-off-by: Yao Yiqi <yaoyiqi3(a)huawei.com> --- kernel/sched/fair.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 05b92febb57a..2df39c1003d7 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -4640,7 +4640,8 @@ static inline void update_tg_cfs_runnable(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq *gcfs_rq) { long delta_sum, delta_avg = gcfs_rq->avg.runnable_avg - se->avg.runnable_avg; - u32 new_sum, divider; + u64 new_sum; + u32 divider; /* Nothing to update */ if (!delta_avg) @@ -4654,7 +4655,7 @@ update_tg_cfs_runnable(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cf /* Set new sched_entity's runnable */ se->avg.runnable_avg = gcfs_rq->avg.runnable_avg; - new_sum = se->avg.runnable_avg * divider; + new_sum = (u64)se->avg.runnable_avg * divider; delta_sum = (long)new_sum - (long)se->avg.runnable_sum; se->avg.runnable_sum = new_sum; -- 2.34.1

2 1

[PATCH openEuler-1.0-LTS] ipv4: validate IPV4_DEVCONF attributes properly
by JiangJieHua 03 Sep '26

03 Sep '26

hulk inclusion category: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/9803 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- syzbot reported a UBSAN undefined behavior issue in vif_delete(): UBSAN: Undefined behaviour in net/ipv4/ipmr.c:720:43 signed integer overflow: -2147483648 - 1 cannot be represented in type 'int' When vif_delete() is called with vifi = -2147483648 (INT_MIN), the operation "vifi - 1" causes a signed integer overflow, which is undefined behavior in C. This occurs because the function does not validate the vifi range before using it as an array index. Fix by adding a proper range check at the beginning of vif_delete() to ensure vifi falls within [0, maxvif) before any arithmetic operation is performed. This prevents the overflow and eliminates the UBSAN warning. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: JiangJieHua <jiangjiehua1(a)huawei.com> --- net/ipv4/devinet.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 64f0fa0be370e..acf60f0a01a4a 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1796,6 +1796,9 @@ static int inet_validate_link_af(const struct net_device *dev, if (cfgid <= 0 || cfgid > IPV4_DEVCONF_MAX) return -EINVAL; + + if (cfgid == IPV4_DEVCONF_MC_FORWARDING) + return -EINVAL; } } -- 2.33.8

2 1

[PATCH openEuler-1.0-LTS] ipv4: validate IPV4_DEVCONF attributes properly
by JiangJieHua 03 Sep '26

03 Sep '26

hulk inclusion category: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/9803 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- syzbot reported a UBSAN undefined behavior issue in vif_delete(): UBSAN: Undefined behaviour in net/ipv4/ipmr.c:720:43 signed integer overflow: -2147483648 - 1 cannot be represented in type 'int' When vif_delete() is called with vifi = -2147483648 (INT_MIN), the operation "vifi - 1" causes a signed integer overflow, which is undefined behavior in C. This occurs because the function does not validate the vifi range before using it as an array index. Fix by adding a proper range check at the beginning of vif_delete() to ensure vifi falls within [0, maxvif) before any arithmetic operation is performed. This prevents the overflow and eliminates the UBSAN warning. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: JiangJieHua <jiangjiehua1(a)huawei.com> --- net/ipv4/devinet.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 64f0fa0be370e..acf60f0a01a4a 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1796,6 +1796,9 @@ static int inet_validate_link_af(const struct net_device *dev, if (cfgid <= 0 || cfgid > IPV4_DEVCONF_MAX) return -EINVAL; + + if (cfgid == IPV4_DEVCONF_MC_FORWARDING) + return -EINVAL; } } -- 2.33.8

2 1

[PATCH openEuler-1.0-LTS] ipv4: validate IPV4_DEVCONF attributes properly
by JiangJieHua 03 Sep '26

03 Sep '26

hulk inclusion category: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/9803 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- syzbot reported a UBSAN undefined behavior issue in vif_delete(): UBSAN: Undefined behaviour in net/ipv4/ipmr.c:720:43 signed integer overflow: -2147483648 - 1 cannot be represented in type 'int' When vif_delete() is called with vifi = -2147483648 (INT_MIN), the operation "vifi - 1" causes a signed integer overflow, which is undefined behavior in C. This occurs because the function does not validate the vifi range before using it as an array index. Fix by adding a proper range check at the beginning of vif_delete() to ensure vifi falls within [0, maxvif) before any arithmetic operation is performed. This prevents the overflow and eliminates the UBSAN warning. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: JiangJieHua <jiangjiehua1(a)huawei.com> Therefore, this self-developed adaptation adopts an equivalent manual validation approach: add checks for key attributes in inet_validate_link_af(), reject read-only attributes (e.g., mc_forwarding) by returning -EINVAL, preventing them from being modified via netlink. This approach aligns with the upstream fix's objective, does not introduce additional kernel ABI changes, and has no impact on existing netlink users. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: JiangJieHua <jiangjiehua1(a)huawei.com> --- net/ipv4/devinet.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 64f0fa0be370e..acf60f0a01a4a 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1796,6 +1796,9 @@ static int inet_validate_link_af(const struct net_device *dev, if (cfgid <= 0 || cfgid > IPV4_DEVCONF_MAX) return -EINVAL; + + if (cfgid == IPV4_DEVCONF_MC_FORWARDING) + return -EINVAL; } } -- 2.33.8

2 1

[PATCH OLK-6.6 0/4] smb: client: backport fixes for CVE-2026-72031
by Lu Chentao 03 Sep '26

03 Sep '26

Backport four upstream smb client fixes required to address CVE-2026-72031 on OLK-6.6. Lu Chentao (4): [OLK-6.6] cifs: prevent readdir from changing file size due to stale directory metadata [OLK-6.6] cifs: fix cifsFileInfo leak on kmalloc failure in deferred close drain paths [OLK-6.6] cifs: consolidate time_last_write stamp into _cifsFileInfo_put() [OLK-6.6] cifs: fix time_last_write stamp placement in setattr/truncate paths fs/smb/client/cifsfs.c | 1 + fs/smb/client/cifsglob.h | 1 + fs/smb/client/file.c | 65 ++++++++++++++++++++++++++++++++++++---- fs/smb/client/inode.c | 25 +++++++++++++++- fs/smb/client/misc.c | 27 +++++++++++++---- 5 files changed, 106 insertions(+), 13 deletions(-) -- 2.52.0

2 5

[PATCH openEuler-1.0-LTS] ext4: validate readdir offset before accessing dirent
by Yao Kai 03 Sep '26

03 Sep '26

hulk inclusion category: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/9553 CVE: NA ------------------------ A corrupted directory can trigger the following KASAN report when ext4_readdir() resumes from an invalid position: BUG: KASAN: use-after-free in __ext4_check_dir_entry+0x5ef/0x820 Read of size 2 at addr ffff88810a646000 by task repro_linear/509 Call Trace: <TASK> dump_stack_lvl+0x53/0x70 print_report+0xd0/0x630 kasan_report+0xce/0x100 __ext4_check_dir_entry+0x5ef/0x820 ext4_readdir+0xcde/0x2b70 iterate_dir+0x1a1/0x520 __x64_sys_getdents64+0x12b/0x220 do_syscall_64+0xf9/0x540 entry_SYSCALL_64_after_hwframe+0x77/0x7f </TASK> KASAN reports use-after-free because the out-of-bounds access lands in an adjacent freed page. The directory buffer itself is still referenced. ext4_dir_llseek() invalidates the directory cookie so that ext4_readdir() rescans directory entries from the start of the block. The rescan checks only the lower bound of rec_len before advancing. A corrupted rec_len can therefore place the offset where the block has insufficient space for a complete directory entry. The rescan itself may dereference that truncated entry, or the main loop may pass it to __ext4_check_dir_entry(). The latter reads de->rec_len before validating the range. For example: block offset 0 4092 4096 |---- de1.rec_len = 4092 -----|----| de2.inode | de2.rec_len ^ OOB, reported as UAF de2 starts at offset 4092 in this 4 KiB block. Its four-byte inode fits in the block, but its rec_len starts at offset 4096 and crosses the boundary. The minimum safe length is inode-dependent. Encrypted and casefolded directory entries need eight additional hash bytes, while a valid metadata checksum tail is only 12 bytes. Cache the metadata checksum feature state and derive the minimum directory entry length from the on-disk format. Use it to bound both the rescan and the offset passed to the main loop. Report an offset in a truncated block tail and skip the remainder of the block, while continuing to accept an offset exactly at the block boundary. Reported-by: syzbot+5322c5c260eb44d209ed(a)syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=5322c5c260eb44d209ed Fixes: ac27a0ec112a ("[PATCH] ext4: initial copy of files from ext3") Signed-off-by: Yao Kai <yaokai34(a)huawei.com> Reviewed-by: Zhihao Cheng <chengzhihao1(a)huawei.com> Reviewed-by: Jan Kara <jack(a)suse.cz> Reviewed-by: Zhang Yi <yi.zhang(a)huawei.com> --- fs/ext4/dir.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c index 5def57966807..9c5ff45bfd6b 100644 --- a/fs/ext4/dir.c +++ b/fs/ext4/dir.c @@ -216,7 +216,9 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx) * dirent right now. Scan from the start of the block * to make sure. */ if (!inode_eq_iversion(inode, file->f_version)) { - for (i = 0; i < sb->s_blocksize && i < offset; ) { + for (i = 0; + i <= sb->s_blocksize - EXT4_DIR_REC_LEN(1) && + i < offset;) { de = (struct ext4_dir_entry_2 *) (bh->b_data + i); /* It's too expensive to do a full @@ -237,6 +239,16 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx) file->f_version = inode_query_iversion(inode); } + if (unlikely(offset < sb->s_blocksize && + offset > sb->s_blocksize - EXT4_DIR_REC_LEN(1))) { + EXT4_ERROR_FILE(file, bh->b_blocknr, + "bad entry in directory: %s - offset=%u, size=%lu", + "directory entry too close to block end", + offset, sb->s_blocksize); + ctx->pos = round_up(ctx->pos, sb->s_blocksize); + goto next_block; + } + while (ctx->pos < inode->i_size && offset < sb->s_blocksize) { de = (struct ext4_dir_entry_2 *) (bh->b_data + offset); @@ -282,6 +294,7 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx) ctx->pos += ext4_rec_len_from_disk(de->rec_len, sb->s_blocksize); } +next_block: if ((ctx->pos < inode->i_size) && !dir_relax_shared(inode)) goto done; brelse(bh); -- 2.43.0

2 1

[PATCH openEuler-1.0-LTS] perf sched: Fix register_pid() overflow, strcpy, and BUG_ON
by Gu Bowen 03 Sep '26

03 Sep '26

From: Arnaldo Carvalho de Melo <acme(a)redhat.com> mainline inclusion from mainline-v7.2-rc1 commit 5949d339f5ec98752d56dcd4e36f619a59d513a5 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/18115 CVE: CVE-2026-80671 Reference: https://web.git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/comm… -------------------------------- register_pid() has several issues when processing untrusted perf.data: 1. Integer overflow: (pid + 1) * sizeof(struct task_desc *) can wrap to a small value on 32-bit systems when pid is large (e.g. 0x40000000), causing realloc to return a tiny buffer followed by out-of-bounds writes in the initialization loop. 2. Heap buffer overflow: strcpy(task->comm, comm) copies the untrusted comm string into a fixed 20-byte COMM_LEN buffer with no length check. 3. BUG_ON on allocation failure: perf.data is untrusted input, so allocation failures should be handled gracefully rather than killing the process. 4. Realloc of sched->tasks assigned directly back, leaking the old pointer on failure; nr_tasks incremented before the realloc, leaving corrupted state on failure. Cap pid at PID_MAX_LIMIT (4194304, matching the kernel's maximum on 64-bit), replace strcpy with strlcpy, guard against NULL comm, replace BUG_ON with NULL returns using safe realloc patterns, and add NULL checks in callers that dereference the result. Fixes: ec156764d424 ("perf sched: Import schedbench.c") Reported-by: sashiko-bot <sashiko-bot(a)kernel.org> Cc: Ingo Molnar <mingo(a)elte.hu> Assisted-by: Claude:claude-opus-4.6 Signed-off-by: Arnaldo Carvalho de Melo <acme(a)redhat.com> Conflicts: tools/perf/builtin-sched.c [Context conflicts.] Signed-off-by: Gu Bowen <gubowen5(a)huawei.com> --- tools/perf/builtin-sched.c | 40 ++++++++++++++++++++++++++++---------- 1 file changed, 30 insertions(+), 10 deletions(-) diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c index 8927952a6d26..ff664763602b 100644 --- a/tools/perf/builtin-sched.c +++ b/tools/perf/builtin-sched.c @@ -43,6 +43,7 @@ #define COMM_LEN 20 #define SYM_LEN 129 #define MAX_PID 1024000 +#define PID_MAX_LIMIT 4194304 /* kernel limit on 64-bit */ struct sched_atom; @@ -435,17 +436,28 @@ static void add_sched_event_sleep(struct perf_sched *sched, struct task_desc *ta static struct task_desc *register_pid(struct perf_sched *sched, unsigned long pid, const char *comm) { - struct task_desc *task; + struct task_desc *task, **tasks_p; static int pid_max; + /* perf.data is untrusted — cap pid to prevent overflow in size calculations */ + if (pid >= PID_MAX_LIMIT) { + pr_err("pid %lu exceeds limit %d, skipping\n", pid, PID_MAX_LIMIT); + return NULL; + } + if (sched->pid_to_task == NULL) { if (sysctl__read_int("kernel/pid_max", &pid_max) < 0) pid_max = MAX_PID; - BUG_ON((sched->pid_to_task = calloc(pid_max, sizeof(struct task_desc *))) == NULL); + sched->pid_to_task = calloc(pid_max, sizeof(struct task_desc *)); + if (sched->pid_to_task == NULL) + return NULL; } if (pid >= (unsigned long)pid_max) { - BUG_ON((sched->pid_to_task = realloc(sched->pid_to_task, (pid + 1) * - sizeof(struct task_desc *))) == NULL); + void *p = realloc(sched->pid_to_task, (pid + 1) * sizeof(struct task_desc *)); + + if (p == NULL) + return NULL; + sched->pid_to_task = p; while (pid >= (unsigned long)pid_max) sched->pid_to_task[pid_max++] = NULL; } @@ -456,9 +468,11 @@ static struct task_desc *register_pid(struct perf_sched *sched, return task; task = zalloc(sizeof(*task)); + if (task == NULL) + return NULL; task->pid = pid; - task->nr = sched->nr_tasks; - strcpy(task->comm, comm); + if (comm) + strlcpy(task->comm, comm, sizeof(task->comm)); /* * every task starts in sleeping state - this gets ignored * if there's no wakeup pointing to this sleep state: @@ -466,10 +480,12 @@ static struct task_desc *register_pid(struct perf_sched *sched, add_sched_event_sleep(sched, task, 0, 0); sched->pid_to_task[pid] = task; - sched->nr_tasks++; - sched->tasks = realloc(sched->tasks, sched->nr_tasks * sizeof(struct task_desc *)); - BUG_ON(!sched->tasks); - sched->tasks[task->nr] = task; + tasks_p = realloc(sched->tasks, (sched->nr_tasks + 1) * sizeof(struct task_desc *)); + if (!tasks_p) + return NULL; + sched->tasks = tasks_p; + sched->tasks[sched->nr_tasks] = task; + task->nr = sched->nr_tasks++; if (verbose > 0) printf("registered task #%ld, PID %ld (%s)\n", sched->nr_tasks, pid, comm); @@ -813,6 +829,8 @@ replay_wakeup_event(struct perf_sched *sched, waker = register_pid(sched, sample->tid, "<unknown>"); wakee = register_pid(sched, pid, comm); + if (waker == NULL || wakee == NULL) + return -1; add_sched_event_wakeup(sched, waker, sample->time, wakee); return 0; @@ -855,6 +873,8 @@ static int replay_switch_event(struct perf_sched *sched, prev = register_pid(sched, prev_pid, prev_comm); next = register_pid(sched, next_pid, next_comm); + if (prev == NULL || next == NULL) + return -1; sched->cpu_last_switched[cpu] = timestamp; -- 2.43.0

2 1

[PATCH OLK-5.10] openvswitch: fix GSO userspace truncation underflow
by Yi Yang 03 Sep '26

03 Sep '26

From: Kyle Zeng <kylebot(a)openai.com> stable inclusion from stable-v5.10.265 commit 50a6a85f3d6b1d22d8436848606cdef5d2c490b4 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/16932 CVE: CVE-2026-68123 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 4032f8ed10fcb84d41c508dfb04be96589f78dfe ] OVS_ACTION_ATTR_TRUNC currently stores a delta from the original skb length in OVS_CB(skb)->cutlen. When a later userspace action segments a GSO skb, queue_gso_packets() reuses that delta for each smaller segment. A segment can then reach queue_userspace_packet() with cutlen greater than skb->len, underflowing the length passed to skb_zerocopy(). Store the maximum preserved length instead and bound each consumer against the current skb length. Use U32_MAX as the no-truncation sentinel so the value remains valid if skb geometry changes before a consumer handles it. Fixes: f2a4d086ed4c ("openvswitch: Add packet truncation support.") Cc: stable(a)vger.kernel.org Assisted-by: Codex:gpt-5.5 Signed-off-by: Kyle Zeng <kylebot(a)openai.com> Reviewed-by: Ilya Maximets <i.maximets(a)ovn.org> Reviewed-by: Aaron Conole <aconole(a)redhat.com> Link: https://patch.msgid.link/20260707221635.27489-1-kylebot@openai.com Signed-off-by: Paolo Abeni <pabeni(a)redhat.com> [5.10.y supports neither OVS_ACTION_ATTR_PSAMPLE nor OVS drop reasons] Signed-off-by: Ilya Maximets <i.maximets(a)ovn.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Conflicts: net/openvswitch/actions.c [The two parameters of the max function are of inconsistent types.] Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- net/openvswitch/actions.c | 15 +++++---------- net/openvswitch/datapath.c | 25 ++++++++++++++----------- net/openvswitch/datapath.h | 2 +- net/openvswitch/vport.c | 2 +- 4 files changed, 21 insertions(+), 23 deletions(-) diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index a02ea493c269..cdf7696d2a20 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -918,12 +918,8 @@ static void do_output(struct datapath *dp, struct sk_buff *skb, int out_port, u16 mru = OVS_CB(skb)->mru; u32 cutlen = OVS_CB(skb)->cutlen; - if (unlikely(cutlen > 0)) { - if (skb->len - cutlen > ovs_mac_header_len(key)) - pskb_trim(skb, skb->len - cutlen); - else - pskb_trim(skb, ovs_mac_header_len(key)); - } + if (unlikely(cutlen < skb->len)) + pskb_trim(skb, max(cutlen, (u32)ovs_mac_header_len(key))); if (likely(!mru || (skb->len <= mru + vport->dev->hard_header_len))) { @@ -1299,22 +1295,21 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb, clone = skb_clone(skb, GFP_ATOMIC); if (clone) do_output(dp, clone, port, key); - OVS_CB(skb)->cutlen = 0; + OVS_CB(skb)->cutlen = U32_MAX; break; } case OVS_ACTION_ATTR_TRUNC: { struct ovs_action_trunc *trunc = nla_data(a); - if (skb->len > trunc->max_len) - OVS_CB(skb)->cutlen = skb->len - trunc->max_len; + OVS_CB(skb)->cutlen = trunc->max_len; break; } case OVS_ACTION_ATTR_USERSPACE: output_userspace(dp, skb, key, a, attr, len, OVS_CB(skb)->cutlen); - OVS_CB(skb)->cutlen = 0; + OVS_CB(skb)->cutlen = U32_MAX; break; case OVS_ACTION_ATTR_HASH: diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index 1c69aa986633..529eb9d6acab 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -240,7 +240,7 @@ void ovs_dp_process_packet(struct sk_buff *skb, struct sw_flow_key *key) upcall.cmd = OVS_PACKET_CMD_MISS; upcall.portid = ovs_vport_find_upcall_portid(p, skb); upcall.mru = OVS_CB(skb)->mru; - error = ovs_dp_upcall(dp, skb, key, &upcall, 0); + error = ovs_dp_upcall(dp, skb, key, &upcall, U32_MAX); switch (error) { case 0: case -EAGAIN: @@ -400,7 +400,8 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, struct sk_buff *nskb = NULL; struct sk_buff *user_skb = NULL; /* to be queued to userspace */ struct nlattr *nla; - size_t len; + size_t msg_size; + size_t skb_len; unsigned int hlen; int err, dp_ifindex; u64 hash; @@ -421,7 +422,8 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, skb = nskb; } - if (nla_attr_size(skb->len) > USHRT_MAX) { + skb_len = min(skb->len, cutlen); + if (nla_attr_size(skb_len) > USHRT_MAX) { err = -EFBIG; goto out; } @@ -436,13 +438,13 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, * padding logic. Only perform zerocopy if padding is not required. */ if (dp->user_features & OVS_DP_F_UNALIGNED) - hlen = skb_zerocopy_headlen(skb); + hlen = min(skb_zerocopy_headlen(skb), cutlen); else - hlen = skb->len; + hlen = skb_len; - len = upcall_msg_size(upcall_info, hlen - cutlen, - OVS_CB(skb)->acts_origlen); - user_skb = genlmsg_new(len, GFP_ATOMIC); + msg_size = upcall_msg_size(upcall_info, hlen, + OVS_CB(skb)->acts_origlen); + user_skb = genlmsg_new(msg_size, GFP_ATOMIC); if (!user_skb) { err = -ENOMEM; goto out; @@ -503,7 +505,7 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, } /* Add OVS_PACKET_ATTR_LEN when packet is truncated */ - if (cutlen > 0 && + if (skb_len < skb->len && nla_put_u32(user_skb, OVS_PACKET_ATTR_LEN, skb->len)) { err = -ENOBUFS; goto out; @@ -528,9 +530,9 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, err = -ENOBUFS; goto out; } - nla->nla_len = nla_attr_size(skb->len - cutlen); + nla->nla_len = nla_attr_size(skb_len); - err = skb_zerocopy(user_skb, skb, skb->len - cutlen, hlen); + err = skb_zerocopy(user_skb, skb, skb_len, hlen); if (err) goto out; @@ -587,6 +589,7 @@ static int ovs_packet_cmd_execute(struct sk_buff *skb, struct genl_info *info) packet->ignore_df = 1; } OVS_CB(packet)->mru = mru; + OVS_CB(packet)->cutlen = U32_MAX; if (a[OVS_PACKET_ATTR_HASH]) { hash = nla_get_u64(a[OVS_PACKET_ATTR_HASH]); diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h index 38f7d3e66ca6..62e7ad32f837 100644 --- a/net/openvswitch/datapath.h +++ b/net/openvswitch/datapath.h @@ -96,7 +96,7 @@ struct datapath { * @mru: The maximum received fragement size; 0 if the packet is not * fragmented. * @acts_origlen: The netlink size of the flow actions applied to this skb. - * @cutlen: The number of bytes from the packet end to be removed. + * @cutlen: The number of bytes in the packet to preserve on output. */ struct ovs_skb_cb { struct vport *input_vport; diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c index da733b92ae8a..ced707ff6dd9 100644 --- a/net/openvswitch/vport.c +++ b/net/openvswitch/vport.c @@ -436,7 +436,7 @@ int ovs_vport_receive(struct vport *vport, struct sk_buff *skb, OVS_CB(skb)->input_vport = vport; OVS_CB(skb)->mru = 0; - OVS_CB(skb)->cutlen = 0; + OVS_CB(skb)->cutlen = U32_MAX; if (unlikely(dev_net(skb->dev) != ovs_dp_get_net(vport->dp))) { u32 mark; -- 2.25.1

2 1

[PATCH OLK-6.6] net/mlx5: Fix MCIA register buffer overflow on 32 dword reads
by Yi Yang 03 Sep '26

03 Sep '26

From: Gal Pressman <gal(a)nvidia.com> mainline inclusion from mainline-v7.2-rc5 commit 11c057d23465c7a5817a7284c896d19d54c0b616 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/17144 CVE: CVE-2026-68293 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- The MCIA register can return up to 32 dwords (128 bytes) when the device advertises the mcia_32dwords capability, but struct mlx5_ifc_mcia_reg_bits only defines dword_0..11, leaving room for just 12 dwords (48 bytes) of data. mlx5_query_mcia() clamps the read size to mlx5_mcia_max_bytes() and then memcpy()s that many bytes out of the register, potentially reading past the end of the 'out' buffer. On kernels built with FORTIFY_SOURCE this is caught as a buffer overflow while reading the module EEPROM via ethtool: detected buffer overflow in memcpy kernel BUG at lib/string_helpers.c:1048! RIP: 0010:fortify_panic+0x13/0x20 Call Trace: mlx5_query_mcia.isra.0+0x200/0x210 [mlx5_core] mlx5_query_module_eeprom_by_page+0x4a/0xa0 [mlx5_core] mlx5e_get_module_eeprom_by_page+0xbb/0x120 [mlx5_core] eeprom_prepare_data+0xf3/0x170 ethnl_default_doit+0xf1/0x3b0 Extend the mcia_reg layout to 32 dwords. Fixes: 271907ee2f29 ("net/mlx5: Query the maximum MCIA register read size from firmware") Signed-off-by: Gal Pressman <gal(a)nvidia.com> Reviewed-by: Alex Lazar <alazar(a)nvidia.com> Signed-off-by: Tariq Toukan <tariqt(a)nvidia.com> Link: https://patch.msgid.link/20260717072338.1240582-1-tariqt@nvidia.com Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Conflicts: drivers/net/ethernet/mellanox/mlx5/core/port.c [Commit 2e4c44b12f4d ("net/mlx5: Refactor EEPROM query error handling to return status separately") was not merged. Context conflicts.] Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- drivers/net/ethernet/mellanox/mlx5/core/port.c | 4 ++-- include/linux/mlx5/mlx5_ifc.h | 13 +------------ 2 files changed, 3 insertions(+), 14 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c index a5622b44385e..ec739b2f0ada 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c @@ -316,7 +316,7 @@ static int mlx5_query_module_id(struct mlx5_core_dev *dev, int module_num, status); return -EIO; } - ptr = MLX5_ADDR_OF(mcia_reg, out, dword_0); + ptr = MLX5_ADDR_OF(mcia_reg, out, dwords); *module_id = ptr[0]; @@ -401,7 +401,7 @@ static int mlx5_query_mcia(struct mlx5_core_dev *dev, return -EIO; } - ptr = MLX5_ADDR_OF(mcia_reg, out, dword_0); + ptr = MLX5_ADDR_OF(mcia_reg, out, dwords); memcpy(data, ptr, size); return size; diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index c59fd31719a1..573461c9dced 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -11320,18 +11320,7 @@ struct mlx5_ifc_mcia_reg_bits { u8 reserved_at_60[0x20]; - u8 dword_0[0x20]; - u8 dword_1[0x20]; - u8 dword_2[0x20]; - u8 dword_3[0x20]; - u8 dword_4[0x20]; - u8 dword_5[0x20]; - u8 dword_6[0x20]; - u8 dword_7[0x20]; - u8 dword_8[0x20]; - u8 dword_9[0x20]; - u8 dword_10[0x20]; - u8 dword_11[0x20]; + u8 dwords[0x400]; }; struct mlx5_ifc_dcbx_param_bits { -- 2.25.1

2 1

[PATCH OLK-6.6] RDMA/nldev: Fix locking when accessing mr->pd
by Chen Jinghuang 03 Sep '26

03 Sep '26

From: Jason Gunthorpe <jgg(a)nvidia.com> mainline inclusion from mainline-v7.2-rc1 commit 50d5c02ab8e62325548bd3a6e6b758a9dcd6e7c3 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/17658 CVE: CVE-2026-74334 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Sashiko points out that, due to rereg_mr, the PD is actually variable and all the touches in nldev are racy. Use mr->device instead of mr->pd->device. Getting the PD restrack ID is more tricky. To avoid disturbing all the happy paths, add an rdma_restrack_sync() operation which is sort of like flush_workqueue() or synchronize_irq(): after it returns, all the old nldev touches to the mr are gone and everything sees the new PD. This makes it safe to reach into the PD pointer. Fixes: da5c85078215 ("RDMA/nldev: add driver-specific resource tracking") Link: https://patch.msgid.link/r/4-v1-29ebd2c229b5+fd5-ib_mr_pd_jgg@nvidia.com Signed-off-by: Jason Gunthorpe <jgg(a)nvidia.com> Conflicts: drivers/infiniband/core/restrack.c drivers/infiniband/core/uverbs_cmd.c include/rdma/ib_verbs.h [In restrack.c, rdma_restrack_count() in target tree lacks show_details parameter and RESTRACK_DD mark logic; only added xa_is_zero() check required by rdma_restrack_sync(). Removed RESTRACK_DD xa_get_mark() check in rdma_restrack_sync() as target tree does not have driver-specific resource tracking (RESTRACK_DD mark). In uverbs_cmd.c, kept downstream #include "uverbs_profile.h" alongside new #include "restrack.h".] Signed-off-by: Chen Jinghuang <chenjinghuang2(a)huawei.com> --- drivers/infiniband/core/nldev.c | 15 +++++---- drivers/infiniband/core/restrack.c | 50 +++++++++++++++++++++++++++- drivers/infiniband/core/restrack.h | 1 + drivers/infiniband/core/uverbs_cmd.c | 10 ++++-- include/rdma/ib_verbs.h | 5 +++ 5 files changed, 72 insertions(+), 9 deletions(-) diff --git a/drivers/infiniband/core/nldev.c b/drivers/infiniband/core/nldev.c index 9d6627bde61a..6c4008a1e18f 100644 --- a/drivers/infiniband/core/nldev.c +++ b/drivers/infiniband/core/nldev.c @@ -644,7 +644,7 @@ static int fill_res_mr_entry(struct sk_buff *msg, bool has_cap_net_admin, struct rdma_restrack_entry *res, uint32_t port) { struct ib_mr *mr = container_of(res, struct ib_mr, res); - struct ib_device *dev = mr->pd->device; + struct ib_device *dev = mr->device; if (has_cap_net_admin) { if (nla_put_u32(msg, RDMA_NLDEV_ATTR_RES_RKEY, mr->rkey)) @@ -660,9 +660,12 @@ static int fill_res_mr_entry(struct sk_buff *msg, bool has_cap_net_admin, if (nla_put_u32(msg, RDMA_NLDEV_ATTR_RES_MRN, res->id)) return -EMSGSIZE; - if (!rdma_is_kernel_res(res) && - nla_put_u32(msg, RDMA_NLDEV_ATTR_RES_PDN, mr->pd->res.id)) - return -EMSGSIZE; + if (!rdma_is_kernel_res(res)) { + struct ib_pd *pd = READ_ONCE(mr->pd); + + if (nla_put_u32(msg, RDMA_NLDEV_ATTR_RES_PDN, pd->res.id)) + return -EMSGSIZE; + } if (fill_res_name_pid(msg, res)) return -EMSGSIZE; @@ -676,7 +679,7 @@ static int fill_res_mr_raw_entry(struct sk_buff *msg, bool has_cap_net_admin, struct rdma_restrack_entry *res, uint32_t port) { struct ib_mr *mr = container_of(res, struct ib_mr, res); - struct ib_device *dev = mr->pd->device; + struct ib_device *dev = mr->device; if (!dev->ops.fill_res_mr_entry_raw) return -EINVAL; @@ -966,7 +969,7 @@ static int fill_stat_mr_entry(struct sk_buff *msg, bool has_cap_net_admin, struct rdma_restrack_entry *res, uint32_t port) { struct ib_mr *mr = container_of(res, struct ib_mr, res); - struct ib_device *dev = mr->pd->device; + struct ib_device *dev = mr->device; if (nla_put_u32(msg, RDMA_NLDEV_ATTR_RES_MRN, res->id)) goto err; diff --git a/drivers/infiniband/core/restrack.c b/drivers/infiniband/core/restrack.c index 438ed3588175..7ffb7701516f 100644 --- a/drivers/infiniband/core/restrack.c +++ b/drivers/infiniband/core/restrack.c @@ -68,8 +68,11 @@ int rdma_restrack_count(struct ib_device *dev, enum rdma_restrack_type type) u32 cnt = 0; xa_lock(&rt->xa); - xas_for_each(&xas, e, U32_MAX) + xas_for_each(&xas, e, U32_MAX) { + if (xa_is_zero(e)) + continue; cnt++; + } xa_unlock(&rt->xa); return cnt; } @@ -266,6 +269,51 @@ int rdma_restrack_put(struct rdma_restrack_entry *res) } EXPORT_SYMBOL(rdma_restrack_put); +/** + * rdma_restrack_sync() - Fence concurrent netlink dumps on an entry + * @res: resource entry + * + * After this returns any concurrent netlink dump threads will see the current + * value of the object. This is useful if the object has to be changed and there + * is not locking to protect the nl side. Eg for mr->pd. This effectively + * destroys the object from a kref/xarray perspective and then immediately + * restores it. The kref is acting like a lock to barrier concurrent nl threads. + * Callers must ensure rdma_restrack_del() is not concurrently called. + */ +void rdma_restrack_sync(struct rdma_restrack_entry *res) +{ + struct rdma_restrack_entry *old; + struct rdma_restrack_root *rt; + struct task_struct *task; + struct ib_device *dev; + + if (!res->valid || res->no_track) + return; + + dev = res_to_dev(res); + if (WARN_ON(!dev)) + return; + + rt = &dev->res[res->type]; + + old = xa_cmpxchg(&rt->xa, res->id, res, XA_ZERO_ENTRY, GFP_KERNEL); + if (WARN_ON(old != res)) + return; + + task = res->task; + if (task) + get_task_struct(task); + rdma_restrack_put(res); + wait_for_completion(&res->comp); + reinit_completion(&res->comp); + if (task) + res->task = task; + kref_init(&res->kref); + + xa_cmpxchg(&rt->xa, res->id, XA_ZERO_ENTRY, res, GFP_KERNEL); +} +EXPORT_SYMBOL(rdma_restrack_sync); + /** * rdma_restrack_del() - delete object from the reource tracking database * @res: resource entry diff --git a/drivers/infiniband/core/restrack.h b/drivers/infiniband/core/restrack.h index 6a04fc41f738..75b8d1005a98 100644 --- a/drivers/infiniband/core/restrack.h +++ b/drivers/infiniband/core/restrack.h @@ -27,6 +27,7 @@ int rdma_restrack_init(struct ib_device *dev); void rdma_restrack_clean(struct ib_device *dev); void rdma_restrack_add(struct rdma_restrack_entry *res); void rdma_restrack_del(struct rdma_restrack_entry *res); +void rdma_restrack_sync(struct rdma_restrack_entry *res); void rdma_restrack_new(struct rdma_restrack_entry *res, enum rdma_restrack_type type); void rdma_restrack_set_name(struct rdma_restrack_entry *res, diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 2e4265ba35b7..f442cca3c05c 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -46,6 +46,7 @@ #include "uverbs.h" #include "core_priv.h" +#include "restrack.h" /* * Copy a response to userspace. If the provided 'resp' is larger than the @@ -813,6 +814,10 @@ static int ib_uverbs_rereg_mr(struct uverbs_attr_bundle *attrs) ret = PTR_ERR(new_pd); goto put_uobjs; } + if (new_pd == orig_pd) { + uobj_put_obj_read(new_pd); + cmd.flags &= ~IB_MR_REREG_PD; + } } else { new_pd = mr->pd; } @@ -858,9 +863,10 @@ static int ib_uverbs_rereg_mr(struct uverbs_attr_bundle *attrs) mr = new_mr; } else { if (cmd.flags & IB_MR_REREG_PD) { - atomic_dec(&orig_pd->usecnt); - mr->pd = new_pd; atomic_inc(&new_pd->usecnt); + WRITE_ONCE(mr->pd, new_pd); + rdma_restrack_sync(&mr->res); + atomic_dec(&orig_pd->usecnt); } if (cmd.flags & IB_MR_REREG_TRANS) { mr->iova = cmd.hca_va; diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 3eea95e51928..05b101693cd5 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -1833,6 +1833,11 @@ struct ib_dm { struct ib_mr { struct ib_device *device; + /* + * Due to IB_MR_REREG_PD pd is not a fixed pointer and can change. For a + * user MR, this value should only be read from a system call that holds + * the uobject lock, or the driver should disable in-place REREG_PD. + */ struct ib_pd *pd; u32 lkey; u32 rkey; -- 2.34.1

2 1

[PATCH OLK-6.6] net/mlx5: Fix MCIA register buffer overflow on 32 dword reads
by Yi Yang 03 Sep '26

03 Sep '26

From: Gal Pressman <gal(a)nvidia.com> mainline inclusion from mainline-v5.13-rc1 commit 11c057d23465c7a5817a7284c896d19d54c0b616 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/17144 CVE: CVE-2026-68293 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- The MCIA register can return up to 32 dwords (128 bytes) when the device advertises the mcia_32dwords capability, but struct mlx5_ifc_mcia_reg_bits only defines dword_0..11, leaving room for just 12 dwords (48 bytes) of data. mlx5_query_mcia() clamps the read size to mlx5_mcia_max_bytes() and then memcpy()s that many bytes out of the register, potentially reading past the end of the 'out' buffer. On kernels built with FORTIFY_SOURCE this is caught as a buffer overflow while reading the module EEPROM via ethtool: detected buffer overflow in memcpy kernel BUG at lib/string_helpers.c:1048! RIP: 0010:fortify_panic+0x13/0x20 Call Trace: mlx5_query_mcia.isra.0+0x200/0x210 [mlx5_core] mlx5_query_module_eeprom_by_page+0x4a/0xa0 [mlx5_core] mlx5e_get_module_eeprom_by_page+0xbb/0x120 [mlx5_core] eeprom_prepare_data+0xf3/0x170 ethnl_default_doit+0xf1/0x3b0 Extend the mcia_reg layout to 32 dwords. Fixes: 271907ee2f29 ("net/mlx5: Query the maximum MCIA register read size from firmware") Signed-off-by: Gal Pressman <gal(a)nvidia.com> Reviewed-by: Alex Lazar <alazar(a)nvidia.com> Signed-off-by: Tariq Toukan <tariqt(a)nvidia.com> Link: https://patch.msgid.link/20260717072338.1240582-1-tariqt@nvidia.com Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Conflicts: drivers/net/ethernet/mellanox/mlx5/core/port.c [Commit 2e4c44b12f4d ("net/mlx5: Refactor EEPROM query error handling to return status separately") was not merged. Context conflicts.] Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- drivers/net/ethernet/mellanox/mlx5/core/port.c | 4 ++-- include/linux/mlx5/mlx5_ifc.h | 13 +------------ 2 files changed, 3 insertions(+), 14 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c index a5622b44385e..ec739b2f0ada 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c @@ -316,7 +316,7 @@ static int mlx5_query_module_id(struct mlx5_core_dev *dev, int module_num, status); return -EIO; } - ptr = MLX5_ADDR_OF(mcia_reg, out, dword_0); + ptr = MLX5_ADDR_OF(mcia_reg, out, dwords); *module_id = ptr[0]; @@ -401,7 +401,7 @@ static int mlx5_query_mcia(struct mlx5_core_dev *dev, return -EIO; } - ptr = MLX5_ADDR_OF(mcia_reg, out, dword_0); + ptr = MLX5_ADDR_OF(mcia_reg, out, dwords); memcpy(data, ptr, size); return size; diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index c59fd31719a1..573461c9dced 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -11320,18 +11320,7 @@ struct mlx5_ifc_mcia_reg_bits { u8 reserved_at_60[0x20]; - u8 dword_0[0x20]; - u8 dword_1[0x20]; - u8 dword_2[0x20]; - u8 dword_3[0x20]; - u8 dword_4[0x20]; - u8 dword_5[0x20]; - u8 dword_6[0x20]; - u8 dword_7[0x20]; - u8 dword_8[0x20]; - u8 dword_9[0x20]; - u8 dword_10[0x20]; - u8 dword_11[0x20]; + u8 dwords[0x400]; }; struct mlx5_ifc_dcbx_param_bits { -- 2.25.1

2 1

[PATCH openEuler-1.0-LTS] ipv4: validate IPV4_DEVCONF attributes properly
by JiangJieHua 03 Sep '26

03 Sep '26

Offering: HULK hulk inclusion categroy: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/9803 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- As the IPV4_DEVCONF netlink attributes are not being validated, it is possible to use netlink to set read-only values like mc_forwarding. In addition, valid ranges are not being validated neither but that is less relevant as they aren't in sysctl. The upstream patch introduces a complete NLA policy table to validate all IPV4_DEVCONF attributes. However, 4.19 kernels do not support macros such as NLA_REJECT and NLA_POLICY_RANGE. Directly backporting the upstream patch would cause build failures and introduce significant mechanism changes. Therefore, this self-developed adaptation adopts an equivalent manual validation approach: add checks for key attributes in inet_validate_link_af(), reject read-only attributes (e.g., mc_forwarding) by returning -EINVAL, preventing them from being modified via netlink. This approach aligns with the upstream fix's objective, does not introduce additional kernel ABI changes, and has no impact on existing netlink users. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: JiangJieHua <jiangjiehua1(a)huawei.com> --- net/ipv4/devinet.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 64f0fa0be370e..acf60f0a01a4a 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1796,6 +1796,9 @@ static int inet_validate_link_af(const struct net_device *dev, if (cfgid <= 0 || cfgid > IPV4_DEVCONF_MAX) return -EINVAL; + + if (cfgid == IPV4_DEVCONF_MC_FORWARDING) + return -EINVAL; } } -- 2.33.8

2 1

[PATCH openEuler-1.0-LTS] RDMA/mlx5: Fix undefined shift of user RQ WQE size
by Chen Jinghuang 03 Sep '26

03 Sep '26

From: Maher Sanalla <msanalla(a)nvidia.com> mainline inclusion from mainline-v7.2-rc1 commit d881d60223aac8fdc12b227d89c76e131e92a9cd category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/18115 CVE: CVE-2026-74297 Reference: https://web.git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/comm… -------------------------------- set_rq_size() computes the RQ WQE size as "1 << rq_wqe_shift" based on the user-provided rq_wqe_shift, which is only checked to be greater than 32, so shifts of 32 are still accepted. A shift of 31 also overflows a signed integer, leading to undefined behavior. Use check_shl_overflow() to compute the RQ WQE size and reject any invalid values. Fixes: e126ba97dba9 ("mlx5: Add driver for Mellanox Connect-IB adapters") Link: https://patch.msgid.link/r/20260611-maher-sec-fixes-v1-1-cd8eb2542869@nvidi… Signed-off-by: Maher Sanalla <msanalla(a)nvidia.com> Signed-off-by: Edward Srouji <edwards(a)nvidia.com> Signed-off-by: Jason Gunthorpe <jgg(a)nvidia.com> Conflicts: drivers/infiniband/hw/mlx5/qp.c [a trival context conflict] Signed-off-by: Chen Jinghuang <chenjinghuang2(a)huawei.com> --- drivers/infiniband/hw/mlx5/qp.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index ef0f710587ad..5d4625e7e0b4 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -260,12 +260,12 @@ static int set_rq_size(struct mlx5_ib_dev *dev, struct ib_qp_cap *cap, } else { if (ucmd) { qp->rq.wqe_cnt = ucmd->rq_wqe_count; - if (ucmd->rq_wqe_shift > BITS_PER_BYTE * sizeof(ucmd->rq_wqe_shift)) - return -EINVAL; qp->rq.wqe_shift = ucmd->rq_wqe_shift; - if ((1 << qp->rq.wqe_shift) / sizeof(struct mlx5_wqe_data_seg) < qp->wq_sig) + if (check_shl_overflow(1, qp->rq.wqe_shift, &wqe_size)) return -EINVAL; - qp->rq.max_gs = (1 << qp->rq.wqe_shift) / sizeof(struct mlx5_wqe_data_seg) - qp->wq_sig; + if (wqe_size / sizeof(struct mlx5_wqe_data_seg) < qp->wq_sig) + return -EINVAL; + qp->rq.max_gs = wqe_size / sizeof(struct mlx5_wqe_data_seg) - qp->wq_sig; qp->rq.max_post = qp->rq.wqe_cnt; } else { wqe_size = qp->wq_sig ? sizeof(struct mlx5_wqe_signature_seg) : 0; -- 2.34.1

2 1

[PATCH OLK-6.6] tcp: clear sock_ops cb flags before force-closing a child socket
by JiangJieHua 03 Sep '26

03 Sep '26

From: Sechang Lim <rhkrqnwk98(a)gmail.com> mainline inclusion from mainline-v7.2-rc1 commit 990348e5bb457697c2f1f7f7b65154a3334d9d2b category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/17954 CVE: CVE-2026-74268 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- A child socket inherits the listener's bpf_sock_ops_cb_flags via sk_clone_lock(). If its setup fails in tcp_v4_syn_recv_sock() / tcp_v6_syn_recv_sock(), the child is freed through put_and_exit, where inet_csk_prepare_forced_close() drops the socket lock and tcp_done() runs without it. If BPF_SOCK_OPS_STATE_CB_FLAG was inherited, tcp_done() -> tcp_set_state() calls tcp_call_bpf(), which expects the lock and trips sock_owned_by_me(): WARNING: include/net/sock.h:1799 at tcp_set_state+0x433/0x550 RIP: 0010:tcp_set_state+0x433/0x550 include/net/sock.h:1799 Call Trace: <IRQ> tcp_done+0xba/0x250 net/ipv4/tcp.c:5095 tcp_v4_syn_recv_sock+0x850/0xa50 net/ipv4/tcp_ipv4.c:1787 tcp_check_req+0xf30/0x1360 net/ipv4/tcp_minisocks.c:926 tcp_v4_rcv+0x1047/0x1b50 net/ipv4/tcp_ipv4.c:2164 </IRQ> The child is freed before it is ever established, so it should run no sock_ops callback. Clear its cb flags in inet_csk_prepare_for_destroy_sock(), the common point for the IPv4, IPv6 and chtls forced-close paths and for the MPTCP ->syn_recv_sock() failure path (dispose_child), which reaches tcp_done() on a child that was never established too. Suggested-by: Jiayuan Chen <jiayuan.chen(a)linux.dev> Fixes: d44874910a26 ("bpf: Add BPF_SOCK_OPS_STATE_CB") Signed-off-by: Sechang Lim <rhkrqnwk98(a)gmail.com> Reviewed-by: Jiayuan Chen <jiayuan.chen(a)linux.dev> Reviewed-by: Kuniyuki Iwashima <kuniyu(a)google.com> Reviewed-by: Eric Dumazet <edumazet(a)google.com> Link: https://patch.msgid.link/20260611092923.1895982-1-rhkrqnwk98@gmail.com Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Conflicts: include/net/inet_connection_sock.h net/ipv4/inet_connection_sock.c [In OLK-6.6, inet_csk_prepare_for_destroy_sock() remains a static inline function in include/net/inet_connection_sock.h, while upstream already moved it into inet_connection_sock.c via commit 7051b54fb5aa (due to header dependency). This backport cherry-picks the inline-to-non-inline conversion (changing the definition in inet_connection_sock.h to a declaration), retains OLK-6.6's this_cpu_inc(*sk->sk_prot->orphan_count) implementation without pulling in the orphan_count refactoring, and adds a sk->sk_protocol == IPPROTO_TCP guard to isolate the DCCP path.] Signed-off-by: JiangJieHua <jiangjiehua1(a)huawei.com> --- include/net/inet_connection_sock.h | 8 +------- include/net/tcp.h | 9 +++++++++ net/ipv4/inet_connection_sock.c | 9 +++++++++ 3 files changed, 19 insertions(+), 7 deletions(-) diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index 327ae14ca82b2..c16f600e5c5c3 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -303,14 +303,8 @@ reqsk_timeout(struct request_sock *req, unsigned long max_timeout) return (unsigned long)min_t(u64, timeout, max_timeout); } -static inline void inet_csk_prepare_for_destroy_sock(struct sock *sk) -{ - /* The below has to be done to allow calling inet_csk_destroy_sock */ - sock_set_flag(sk, SOCK_DEAD); - this_cpu_inc(*sk->sk_prot->orphan_count); -} - void inet_csk_destroy_sock(struct sock *sk); +void inet_csk_prepare_for_destroy_sock(struct sock *sk); void inet_csk_prepare_forced_close(struct sock *sk); /* diff --git a/include/net/tcp.h b/include/net/tcp.h index 8674996bd39c6..e1a345d5a40b1 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -2515,6 +2515,11 @@ static inline int tcp_call_bpf_3arg(struct sock *sk, int op, u32 arg1, u32 arg2, return tcp_call_bpf(sk, op, 3, args); } +static inline void tcp_clear_sock_ops_cb_flags(struct sock *sk) +{ + tcp_sk(sk)->bpf_sock_ops_cb_flags = 0; +} + #else static inline int tcp_call_bpf(struct sock *sk, int op, u32 nargs, u32 *args) { @@ -2532,6 +2537,10 @@ static inline int tcp_call_bpf_3arg(struct sock *sk, int op, u32 arg1, u32 arg2, return -EPERM; } +static inline void tcp_clear_sock_ops_cb_flags(struct sock *sk) +{ +} + #endif static inline u32 tcp_timeout_init(struct sock *sk) diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 9f1d17cd5eebd..bca256306c9ee 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -1240,6 +1240,15 @@ void inet_csk_destroy_sock(struct sock *sk) } EXPORT_SYMBOL(inet_csk_destroy_sock); +void inet_csk_prepare_for_destroy_sock(struct sock *sk) +{ + /* The below has to be done to allow calling inet_csk_destroy_sock */ + if (sk->sk_protocol == IPPROTO_TCP) + tcp_clear_sock_ops_cb_flags(sk); + sock_set_flag(sk, SOCK_DEAD); + this_cpu_inc(*sk->sk_prot->orphan_count); +} + /* This function allows to force a closure of a socket after the call to * tcp/dccp_create_openreq_child(). */ -- 2.33.8

2 1

[PATCH OLK-6.6] tcp: clear sock_ops cb flags before force-closing a child socket
by JiangJieHua 03 Sep '26

03 Sep '26

From: Sechang Lim <rhkrqnwk98(a)gmail.com> mainline inclusion from mainline-v7.2-rc1 commit 990348e5bb457697c2f1f7f7b65154a3334d9d2b category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/17954 CVE: CVE-2026-74268 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- A child socket inherits the listener's bpf_sock_ops_cb_flags via sk_clone_lock(). If its setup fails in tcp_v4_syn_recv_sock() / tcp_v6_syn_recv_sock(), the child is freed through put_and_exit, where inet_csk_prepare_forced_close() drops the socket lock and tcp_done() runs without it. If BPF_SOCK_OPS_STATE_CB_FLAG was inherited, tcp_done() -> tcp_set_state() calls tcp_call_bpf(), which expects the lock and trips sock_owned_by_me(): WARNING: include/net/sock.h:1799 at tcp_set_state+0x433/0x550 RIP: 0010:tcp_set_state+0x433/0x550 include/net/sock.h:1799 Call Trace: <IRQ> tcp_done+0xba/0x250 net/ipv4/tcp.c:5095 tcp_v4_syn_recv_sock+0x850/0xa50 net/ipv4/tcp_ipv4.c:1787 tcp_check_req+0xf30/0x1360 net/ipv4/tcp_minisocks.c:926 tcp_v4_rcv+0x1047/0x1b50 net/ipv4/tcp_ipv4.c:2164 </IRQ> The child is freed before it is ever established, so it should run no sock_ops callback. Clear its cb flags in inet_csk_prepare_for_destroy_sock(), the common point for the IPv4, IPv6 and chtls forced-close paths and for the MPTCP ->syn_recv_sock() failure path (dispose_child), which reaches tcp_done() on a child that was never established too. Suggested-by: Jiayuan Chen <jiayuan.chen(a)linux.dev> Fixes: d44874910a26 ("bpf: Add BPF_SOCK_OPS_STATE_CB") Signed-off-by: Sechang Lim <rhkrqnwk98(a)gmail.com> Reviewed-by: Jiayuan Chen <jiayuan.chen(a)linux.dev> Reviewed-by: Kuniyuki Iwashima <kuniyu(a)google.com> Reviewed-by: Eric Dumazet <edumazet(a)google.com> Link: https://patch.msgid.link/20260611092923.1895982-1-rhkrqnwk98@gmail.com Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Conflicts: include/net/inet_connection_sock.h net/ipv4/inet_connection_sock.c [In OLK-6.6, inet_csk_prepare_for_destroy_sock() remains a static inline function in include/net/inet_connection_sock.h, while upstream already moved it into inet_connection_sock.c via commit 7051b54fb5aa (due to header dependency). This backport cherry-picks the inline-to-non-inline conversion (changing the definition in inet_connection_sock.h to a declaration), retains OLK-6.6's this_cpu_inc(*sk->sk_prot->orphan_count) implementation without pulling in the orphan_count refactoring, and adds a sk->sk_protocol == IPPROTO_TCP guard to isolate the DCCP path.] Signed-off-by: JiangJieHua <jiangjiehua1(a)huawei.com> --- include/net/inet_connection_sock.h | 8 +------- include/net/tcp.h | 9 +++++++++ net/ipv4/inet_connection_sock.c | 9 +++++++++ 3 files changed, 19 insertions(+), 7 deletions(-) diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index 327ae14ca82b2..c16f600e5c5c3 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -303,14 +303,8 @@ reqsk_timeout(struct request_sock *req, unsigned long max_timeout) return (unsigned long)min_t(u64, timeout, max_timeout); } -static inline void inet_csk_prepare_for_destroy_sock(struct sock *sk) -{ - /* The below has to be done to allow calling inet_csk_destroy_sock */ - sock_set_flag(sk, SOCK_DEAD); - this_cpu_inc(*sk->sk_prot->orphan_count); -} - void inet_csk_destroy_sock(struct sock *sk); +void inet_csk_prepare_for_destroy_sock(struct sock *sk); void inet_csk_prepare_forced_close(struct sock *sk); /* diff --git a/include/net/tcp.h b/include/net/tcp.h index 8674996bd39c6..e1a345d5a40b1 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -2515,6 +2515,11 @@ static inline int tcp_call_bpf_3arg(struct sock *sk, int op, u32 arg1, u32 arg2, return tcp_call_bpf(sk, op, 3, args); } +static inline void tcp_clear_sock_ops_cb_flags(struct sock *sk) +{ + tcp_sk(sk)->bpf_sock_ops_cb_flags = 0; +} + #else static inline int tcp_call_bpf(struct sock *sk, int op, u32 nargs, u32 *args) { @@ -2532,6 +2537,10 @@ static inline int tcp_call_bpf_3arg(struct sock *sk, int op, u32 arg1, u32 arg2, return -EPERM; } +static inline void tcp_clear_sock_ops_cb_flags(struct sock *sk) +{ +} + #endif static inline u32 tcp_timeout_init(struct sock *sk) diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 9f1d17cd5eebd..bca256306c9ee 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -1240,6 +1240,15 @@ void inet_csk_destroy_sock(struct sock *sk) } EXPORT_SYMBOL(inet_csk_destroy_sock); +void inet_csk_prepare_for_destroy_sock(struct sock *sk) +{ + /* The below has to be done to allow calling inet_csk_destroy_sock */ + if (sk->sk_protocol == IPPROTO_TCP) + tcp_clear_sock_ops_cb_flags(sk); + sock_set_flag(sk, SOCK_DEAD); + this_cpu_inc(*sk->sk_prot->orphan_count); +} + /* This function allows to force a closure of a socket after the call to * tcp/dccp_create_openreq_child(). */ -- 2.33.8

2 1

[PATCH OLK-6.6 00/33] arm_mpam: Introduce Narrow-PARTID feature
by Zeng Heng 03 Sep '26

03 Sep '26

Background ========== On x86, the resctrl allows creating up to num_rmids monitoring groups under parent control group. However, ARM64 MPAM is currently limited by the PMG (Performance Monitoring Group) count, which is typically much smaller than the theoretical RMID limit. This creates a significant scalability gap: users expecting fine-grained per-process or per-thread monitoring quickly exhaust the PMG space, even when plenty of reqPARTIDs remain available. The Narrow-PARTID feature, defined in the ARM MPAM architecture, addresses this by associating reqPARTIDs with intPARTIDs through a programmable many-to-one mapping. This allows the kernel to present more logical monitoring contexts. Design Overview =============== The implementation extends the RMID encoding to carry reqPARTID information: RMID = reqPARTID * NUM_PMG + PMG In this patchset, a monitoring group is uniquely identified by the combination of reqPARTID and PMG. The closid is represented by intPARTID, which is exactly the original PARTID. For systems with homogeneous MSCs (all supporting Narrow-PARTID), the driver exposes the full reqPARTID range directly. For heterogeneous systems where some MSCs lack Narrow-PARTID support, the driver utilizes PARTIDs beyond the intPARTID range as reqPARTIDs to expand monitoring capacity. The sole exception is when MBA MSCs lack Narrow-PARTID support, their percentage-based control mechanism prevents the use of PARTIDs as reqPARTIDs. Capacity Improvements ===================== -------------------------------------------------------------------------- The maximum | Sub-monitoring groups | System-wide number of | under a control group | monitoring groups -------------------------------------------------------------------------- Without | | reqPARTID | PMG | intPARTID * PMG -------------------------------------------------------------------------- reqPARTID | | static allocation | (reqPARTID // intPARTID) * PMG | reqPARTID * PMG -------------------------------------------------------------------------- reqPARTID | | dynamic allocation | (reqPARTID − intPARTID + 1) * PMG | reqPARTID * PMG -------------------------------------------------------------------------- Under MPAM, the number of reqPARTID is always greater than or equal to intPARTID. Dave Martin (1): arm_mpam: Set INTERNAL as needed when setting MSC controls James Morse (1): debugfs: Add helpers for creating cpumask entries in debugfs Yushan Wang (1): cputype: add cputype for HIP13 Zeng Heng (30): arm_mpam: Add intPARTID and reqPARTID support for narrow PARTID feature arm_mpam: Refactor rmid to reqPARTID/PMG mapping arm_mpam: Propagate control group config to sub-monitoring groups fs/resctrl: Add rmid_entry state helpers arm_mpam: Implement dynamic reqPARTID allocation for monitoring groups fs/resctrl: Wire up rmid expansion and reclaim functions arm64/mpam: Add mpam_sync_config() for dynamic rmid expansion fs/resctrl: Prevent rmid parsing errors by flushing limbo on umount arm_mpam: Disable reqPARTID expansion when Narrow-PARTID is unavailable arm_mpam: Add MBWU counter scaling support fs/resctrl: Fix incorrect closid propagation during monitor group reparenting arm64/mpam: Fix wd boundary check and GENMASK range in percent-mbw conversion arm64/mpam: Add HISI_HIP13 support fs/resctrl: Introduce per-resource MBM enable check fs/resctrl: Add arch-specific MBM overflow setup helpers arm64/mpam: Add selective L2 MBM overflow checking arm64/mpam: Add MBM overflow handling for HIP13 arm64/mpam: Allow zero-bit CPBM for L2 cache arm64/mpam: Adapt SMMU IOMMU interface for Narrow PARTID support arm64/mpam: Add MBOPT (Memory Bandwidth Optimal) feature resctrl: Remove unused mbm_core field from rdt_domain fs/resctrl: Introduce resctrl_arch_mbm_update() for arch-specific MBM event handling arm64/mpam: Add MBM_TOTAL_EVENT configurability check arm64/mpam: Add debugfs entries to show the MSC/RIS the driver discovered iommu/arm-smmu-v3: Add debug logging for MPAM STE configuration Revert "fs/resctrl: Re-allocate rmid for the monitor when migrating across control groups" fs/resctrl: Fix iommu_group and kobject reference leaks arm_mpam: Fix NULL pointer dereference after freeing comp->cfg arm_mpam: Wait for SRCU grace period before destroying RIS arm_mpam: Fix percpu memory leak in __setup_ppi() error path arch/arm64/include/asm/cputype.h | 2 + arch/arm64/include/asm/mpam.h | 17 +- drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 3 + drivers/platform/mpam/mpam_devices.c | 271 ++++++++- drivers/platform/mpam/mpam_internal.h | 21 +- drivers/platform/mpam/mpam_resctrl.c | 617 ++++++++++++++++---- fs/debugfs/file.c | 64 ++ fs/resctrl/internal.h | 1 - fs/resctrl/monitor.c | 123 ++-- fs/resctrl/rdtgroup.c | 132 +++-- include/linux/arm_mpam.h | 20 +- include/linux/debugfs.h | 6 + include/linux/resctrl.h | 35 +- include/linux/resctrl_types.h | 2 + tools/arch/arm64/include/asm/cputype.h | 2 + 15 files changed, 1035 insertions(+), 281 deletions(-) -- 2.43.0

2 34

[PATCH openEuler-1.0-LTS] ipv4: validate IPV4_DEVCONF attributes properly
by JiangJieHua 03 Sep '26

03 Sep '26

Offering: HULK HULK inclusion categroy: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/9803 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- As the IPV4_DEVCONF netlink attributes are not being validated, it is possible to use netlink to set read-only values like mc_forwarding. In addition, valid ranges are not being validated neither but that is less relevant as they aren't in sysctl. The upstream patch introduces a complete NLA policy table to validate all IPV4_DEVCONF attributes. However, 4.19 kernels do not support macros such as NLA_REJECT and NLA_POLICY_RANGE. Directly backporting the upstream patch would cause build failures and introduce significant mechanism changes. Therefore, this self-developed adaptation adopts an equivalent manual validation approach: add checks for key attributes in inet_validate_link_af(), reject read-only attributes (e.g., mc_forwarding) by returning -EINVAL, preventing them from being modified via netlink. This approach aligns with the upstream fix's objective, does not introduce additional kernel ABI changes, and has no impact on existing netlink users. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: JiangJieHua <jiangjiehua1(a)huawei.com> --- net/ipv4/devinet.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 64f0fa0be370e..acf60f0a01a4a 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1796,6 +1796,9 @@ static int inet_validate_link_af(const struct net_device *dev, if (cfgid <= 0 || cfgid > IPV4_DEVCONF_MAX) return -EINVAL; + + if (cfgid == IPV4_DEVCONF_MC_FORWARDING) + return -EINVAL; } } -- 2.33.8

2 1

[PATCH OLK-5.10] openvswitch: fix GSO userspace truncation underflow
by Yi Yang 03 Sep '26

03 Sep '26

From: Kyle Zeng <kylebot(a)openai.com> stable inclusion from stable-v5.10.265 commit 50a6a85f3d6b1d22d8436848606cdef5d2c490b4 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/16932 CVE: CVE-2026-68123 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 4032f8ed10fcb84d41c508dfb04be96589f78dfe ] OVS_ACTION_ATTR_TRUNC currently stores a delta from the original skb length in OVS_CB(skb)->cutlen. When a later userspace action segments a GSO skb, queue_gso_packets() reuses that delta for each smaller segment. A segment can then reach queue_userspace_packet() with cutlen greater than skb->len, underflowing the length passed to skb_zerocopy(). Store the maximum preserved length instead and bound each consumer against the current skb length. Use U32_MAX as the no-truncation sentinel so the value remains valid if skb geometry changes before a consumer handles it. Fixes: f2a4d086ed4c ("openvswitch: Add packet truncation support.") Cc: stable(a)vger.kernel.org Assisted-by: Codex:gpt-5.5 Signed-off-by: Kyle Zeng <kylebot(a)openai.com> Reviewed-by: Ilya Maximets <i.maximets(a)ovn.org> Reviewed-by: Aaron Conole <aconole(a)redhat.com> Link: https://patch.msgid.link/20260707221635.27489-1-kylebot@openai.com Signed-off-by: Paolo Abeni <pabeni(a)redhat.com> [5.10.y supports neither OVS_ACTION_ATTR_PSAMPLE nor OVS drop reasons] Signed-off-by: Ilya Maximets <i.maximets(a)ovn.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- net/openvswitch/actions.c | 15 +++++---------- net/openvswitch/datapath.c | 25 ++++++++++++++----------- net/openvswitch/datapath.h | 2 +- net/openvswitch/vport.c | 2 +- 4 files changed, 21 insertions(+), 23 deletions(-) diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index a02ea493c269..1e087b7f6dc6 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -918,12 +918,8 @@ static void do_output(struct datapath *dp, struct sk_buff *skb, int out_port, u16 mru = OVS_CB(skb)->mru; u32 cutlen = OVS_CB(skb)->cutlen; - if (unlikely(cutlen > 0)) { - if (skb->len - cutlen > ovs_mac_header_len(key)) - pskb_trim(skb, skb->len - cutlen); - else - pskb_trim(skb, ovs_mac_header_len(key)); - } + if (unlikely(cutlen < skb->len)) + pskb_trim(skb, max(cutlen, ovs_mac_header_len(key))); if (likely(!mru || (skb->len <= mru + vport->dev->hard_header_len))) { @@ -1299,22 +1295,21 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb, clone = skb_clone(skb, GFP_ATOMIC); if (clone) do_output(dp, clone, port, key); - OVS_CB(skb)->cutlen = 0; + OVS_CB(skb)->cutlen = U32_MAX; break; } case OVS_ACTION_ATTR_TRUNC: { struct ovs_action_trunc *trunc = nla_data(a); - if (skb->len > trunc->max_len) - OVS_CB(skb)->cutlen = skb->len - trunc->max_len; + OVS_CB(skb)->cutlen = trunc->max_len; break; } case OVS_ACTION_ATTR_USERSPACE: output_userspace(dp, skb, key, a, attr, len, OVS_CB(skb)->cutlen); - OVS_CB(skb)->cutlen = 0; + OVS_CB(skb)->cutlen = U32_MAX; break; case OVS_ACTION_ATTR_HASH: diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index 1c69aa986633..529eb9d6acab 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -240,7 +240,7 @@ void ovs_dp_process_packet(struct sk_buff *skb, struct sw_flow_key *key) upcall.cmd = OVS_PACKET_CMD_MISS; upcall.portid = ovs_vport_find_upcall_portid(p, skb); upcall.mru = OVS_CB(skb)->mru; - error = ovs_dp_upcall(dp, skb, key, &upcall, 0); + error = ovs_dp_upcall(dp, skb, key, &upcall, U32_MAX); switch (error) { case 0: case -EAGAIN: @@ -400,7 +400,8 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, struct sk_buff *nskb = NULL; struct sk_buff *user_skb = NULL; /* to be queued to userspace */ struct nlattr *nla; - size_t len; + size_t msg_size; + size_t skb_len; unsigned int hlen; int err, dp_ifindex; u64 hash; @@ -421,7 +422,8 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, skb = nskb; } - if (nla_attr_size(skb->len) > USHRT_MAX) { + skb_len = min(skb->len, cutlen); + if (nla_attr_size(skb_len) > USHRT_MAX) { err = -EFBIG; goto out; } @@ -436,13 +438,13 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, * padding logic. Only perform zerocopy if padding is not required. */ if (dp->user_features & OVS_DP_F_UNALIGNED) - hlen = skb_zerocopy_headlen(skb); + hlen = min(skb_zerocopy_headlen(skb), cutlen); else - hlen = skb->len; + hlen = skb_len; - len = upcall_msg_size(upcall_info, hlen - cutlen, - OVS_CB(skb)->acts_origlen); - user_skb = genlmsg_new(len, GFP_ATOMIC); + msg_size = upcall_msg_size(upcall_info, hlen, + OVS_CB(skb)->acts_origlen); + user_skb = genlmsg_new(msg_size, GFP_ATOMIC); if (!user_skb) { err = -ENOMEM; goto out; @@ -503,7 +505,7 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, } /* Add OVS_PACKET_ATTR_LEN when packet is truncated */ - if (cutlen > 0 && + if (skb_len < skb->len && nla_put_u32(user_skb, OVS_PACKET_ATTR_LEN, skb->len)) { err = -ENOBUFS; goto out; @@ -528,9 +530,9 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, err = -ENOBUFS; goto out; } - nla->nla_len = nla_attr_size(skb->len - cutlen); + nla->nla_len = nla_attr_size(skb_len); - err = skb_zerocopy(user_skb, skb, skb->len - cutlen, hlen); + err = skb_zerocopy(user_skb, skb, skb_len, hlen); if (err) goto out; @@ -587,6 +589,7 @@ static int ovs_packet_cmd_execute(struct sk_buff *skb, struct genl_info *info) packet->ignore_df = 1; } OVS_CB(packet)->mru = mru; + OVS_CB(packet)->cutlen = U32_MAX; if (a[OVS_PACKET_ATTR_HASH]) { hash = nla_get_u64(a[OVS_PACKET_ATTR_HASH]); diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h index 38f7d3e66ca6..62e7ad32f837 100644 --- a/net/openvswitch/datapath.h +++ b/net/openvswitch/datapath.h @@ -96,7 +96,7 @@ struct datapath { * @mru: The maximum received fragement size; 0 if the packet is not * fragmented. * @acts_origlen: The netlink size of the flow actions applied to this skb. - * @cutlen: The number of bytes from the packet end to be removed. + * @cutlen: The number of bytes in the packet to preserve on output. */ struct ovs_skb_cb { struct vport *input_vport; diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c index da733b92ae8a..ced707ff6dd9 100644 --- a/net/openvswitch/vport.c +++ b/net/openvswitch/vport.c @@ -436,7 +436,7 @@ int ovs_vport_receive(struct vport *vport, struct sk_buff *skb, OVS_CB(skb)->input_vport = vport; OVS_CB(skb)->mru = 0; - OVS_CB(skb)->cutlen = 0; + OVS_CB(skb)->cutlen = U32_MAX; if (unlikely(dev_net(skb->dev) != ovs_dp_get_net(vport->dp))) { u32 mark; -- 2.25.1

2 1

openEuler Kernel SIG双周例会
by openEuler conference 03 Sep '26

03 Sep '26

您好！ Kernel 邀请您参加 2026-09-04 14:00 召开的WeLink会议(自动录制) 会议主题：openEuler Kernel SIG双周例会会议发起人： LiaoTao_Wave 会议内容： 1. 进展update 2. 议题征集中（新增议题可直接填报至会议纪要看板）会议链接：https://meeting.huaweicloud.com:36443/#/j/989418007 会议纪要&签到链接：https://etherpad.openeuler.org/p/Kernel-meetings 更多资讯尽在：https://www.openeuler.org/zh/ Hello! Kernel invites you to attend the WeLink conference(auto recording) will be held at 2026-09-04 14:00 The subject of the conference is openEuler Kernel SIG双周例会 The sponsor of the conference is LiaoTao_Wave Summary: 1. 进展update 2. 议题征集中（新增议题可直接填报至会议纪要看板） You can join the meeting at https://meeting.huaweicloud.com:36443/#/j/989418007 Add topics at https://etherpad.openeuler.org/p/Kernel-meetings More information: https://www.openeuler.org/en/

1 0

[PATCH OLK-5.10] net/sched: cls_api: Always acquire rtnl_lock when destroying locked classifiers
by superdcc97＠163.com 02 Sep '26

02 Sep '26

From: Jamal Hadi Salim <jhs(a)mojatatu.com> mainline inclusion from mainline-v7.2-rc7 commit a347304b2ca1a5377d5bd2d8a72e4b4f12afe648 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/18291 CVE: CVE-2026-74700 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Another challenge with unlocked filters. There is a short window in tc_new_tfilter where a tcf_proto can be found and briefly referenced by a totally unrelated, unlocked classifier's request and cause a race. Feng created a poc which created this race with two threads, one creating a u32 filter and other a flower filter in the same chain/prio: 1. Both threads enter tc_new_tfilter, both find the chain empty, both drop filter_chain_lock 2. u32 finishes tcf_proto_create("u32") first, calls tcf_chain_tp_insert_unique() -> inserts u32_tp into the chain 3. flower finishes tcf_proto_create("flower") later, calls tcf_chain_tp_insert_unique() -> tcf_chain_tp_find() now sees u32_tp already there, takes a reference on it, destroys flower's own tp_new and returns u32_tp to the caller. Flower then hits the kind mismatch check (because it requested for kind "flower" but tp->ops->kind is "u32") and goes through the errout path which calls tcf_proto_put() on u32_tp. If the u32 thread has already gone through its own errout (its change() call failed on the PoC's empty options) and dropped its create and insert refs, flower's put is the last one and drops u32_tp's refcnt to zero. At this point tp->ops->destroy() runs in a context that never took rtnl_lock. When that happens, it might cause a UAF like the following (illustrated by the PoC): [ +0.000710] BUG: KASAN: slab-use-after-free in u32_init (net/sched/cls_u32.c:393) [ +0.000281] Read of size 8 at addr ffff888120022f00 by task poc_feng_xue/524 Call Trace: u32_init (net/sched/cls_u32.c:393) tc_new_tfilter (net/sched/cls_api.c:2378) Allocated by task 526: u32_init (net/sched/cls_u32.c:378) tc_new_tfilter (net/sched/cls_api.c:2378) Freed by task 522: kfree u32_destroy (net/sched/cls_u32.c:662) tcf_proto_destroy (net/sched/cls_api.c:446) tcf_proto_put (net/sched/cls_api.c:459) tc_new_tfilter (net/sched/cls_api.c:2459) Fix this by having tcf_proto_destroy() take rtnl_lock around tp->ops->destroy() for locked classifiers whenever rtnl is not held. To explain why I used a temp variable "not_lockless" I'd like to point to a semi-related note on rtnl_held vs TCF_PROTO_OPS_DOIT_UNLOCKED (adding here for future cleanup if deemed necessary): The rtnl_held parameter and the TCF_PROTO_OPS_DOIT_UNLOCKED flag are redundant sources of truth for whether rtnl_lock is held. Among the nine classifier destroy(..rtnl_held..) callbacks, only flower consults the rtnl_held parameter which it propagates to tc_setup_cb_destroy() and tc_setup_cb_call(). The other eight (u32, flow, bpf, cgroup, route, basic, fw, mall) ignore it entirely;-> those that call tc_setup_cb_destroy() (u32, bpf, mall) hardcode true always instead of forwarding the parameter. A future cleanup should remove the rtnl_held parameter from the destroy callback signature entirely and have callers rely solely on their knowledge whether they are running in an unlocked context. Fixes: 12db03b65c2b ("net: sched: extend proto ops to support unlocked classifiers") Reported-by: Feng Xue <feng.xue(a)outlook.com> Tested-by: Victor Nogueira <victor(a)mojatatu.com> Signed-off-by: Jamal Hadi Salim <jhs(a)mojatatu.com> Link: https://patch.msgid.link/20260801125632.360365-1-jhs@mojatatu.com Signed-off-by: Paolo Abeni <pabeni(a)redhat.com> Conflicts: net/sched/cls_api.c [commit a12c76a03386 ("net: sched: refine software bypass handling in tc_run") is not backport, which lead to conflicts] Signed-off-by: Dong Chenchen <dongchenchen2(a)huawei.com> --- net/sched/cls_api.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index b6dd697a3d5f..a263cbf969be 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -292,7 +292,22 @@ static void tcf_chain_put(struct tcf_chain *chain); static void tcf_proto_destroy(struct tcf_proto *tp, bool rtnl_held, bool sig_destroy, struct netlink_ext_ack *extack) { - tp->ops->destroy(tp, rtnl_held, extack); + /* A locked classifier's destroy callback (e.g. u32_destroy) uses + * rtnl_dereference() and mutates shared structures (e.g. the + * tc_u_common hash list) that are only safe under rtnl_lock. When an + * unlocked classifier's request (e.g. flower on ingress) loses the + * tcf_chain_tp_insert_unique() race and ends up dropping the last + * reference on a locked classifier's proto, destroy() would run + * without rtnl held. Take it here in that case. + */ + bool not_lockless = !rtnl_held && + !(tp->ops->flags & TCF_PROTO_OPS_DOIT_UNLOCKED); + + if (not_lockless) + rtnl_lock(); + tp->ops->destroy(tp, rtnl_held || not_lockless, extack); + if (not_lockless) + rtnl_unlock(); if (sig_destroy) tcf_proto_signal_destroyed(tp->chain, tp); tcf_chain_put(tp->chain); -- 2.43.0

2 1

[PATCH OLK-6.6 0/6] net/sched: fix UAF when destroying locked classifiers (CVE-2026-74700)
by superdcc97＠163.com 02 Sep '26

02 Sep '26

From: Dong Chenchen <dongchenchen2(a)huawei.com> This series fixes CVE-2026-74700, a slab-use-after-free (UAF) in the tc_new_tfilter() path of net/sched/cls_api.c. A short race window in tc_new_tfilter() allows a tcf_proto to be found and briefly referenced by a totally unrelated, unlocked classifier's request. When an unlocked classifier's request (e.g. flower on ingress) loses the tcf_chain_tp_insert_unique() race and ends up dropping the last reference on a locked classifier's proto, the destroy() callback runs without rtnl_lock held. That can corrupt shared classifier structures (e.g. the tc_u_common hash list) and causes a use-after-free. The final patch fixes this by having tcf_proto_destroy() take rtnl_lock around the destroy() callback for locked classifiers whenever rtnl is not already held. The earlier patches are the prerequisite backports that reworked the filter counting / software-bypass bookkeeping (skip_sw counter, filter counter, make skip_sw actually skip software, UAF fix in fl_dump_key, and refine software bypass handling). They are Stable-dep-of prerequisites for the main fix and are required for it to apply correctly on OLK-6.6. Asbjørn Sloth Tønnesen (3): net: sched: cls_api: add skip_sw counter net: sched: cls_api: add filter counter net: sched: make skip_sw actually skip software Jamal Hadi Salim (1): net/sched: cls_api: Always acquire rtnl_lock when destroying locked classifiers Jianbo Liu (1): net: sched: cls_api: fix slab-use-after-free in fl_dump_key Xin Long (1): net: sched: refine software bypass handling in tc_run include/net/pkt_cls.h | 18 +++++++++++++++ include/net/sch_generic.h | 3 +++ net/core/dev.c | 13 +++++++++++ net/sched/cls_api.c | 47 ++++++++++++++++++++++++++++++++++++++- net/sched/cls_bpf.c | 2 ++ net/sched/cls_flower.c | 2 ++ net/sched/cls_matchall.c | 2 ++ net/sched/cls_u32.c | 4 ++++ 8 files changed, 90 insertions(+), 1 deletion(-) -- 2.43.0

2 7

[PATCH OLK-5.10] netfilter: xt_connmark: reject invalid shift parameters
by superdcc97＠163.com 02 Sep '26

02 Sep '26

From: Dong Chenchen <dongchenchen2(a)huawei.com> mainline inclusion from mainline-v7.2-rc3 commit 1b47026fb4b35bac850ad6e8a4ad7fc018e09ebc category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/17361 CVE: CVE-2026-72347 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Revision 2 of the CONNMARK target accepts user-controlled shift parameters and applies them to 32-bit mark values in connmark_tg_shift(). A shift_bits value of 32 or more triggers an undefined-shift bug when the rule is evaluated. Invalid shift_dir values are also accepted and silently fall back to the left-shift path. Reject invalid revision-2 shift parameters in connmark_tg_check() so malformed rules fail at installation time, before they can reach the packet path. Fixes: 472a73e00757 ("netfilter: xt_conntrack: Support bit-shifting for CONNMARK & MARK targets.") Reported-by: Yuan Tan <yuantan098(a)gmail.com> Reported-by: Zhengchuan Liang <zcliangcn(a)gmail.com> Reported-by: Xin Liu <dstsmallbird(a)foxmail.com> Assisted-by: Codex:GPT-5.4 Signed-off-by: Wyatt Feng <bronzed_45_vested(a)icloud.com> Reviewed-by: Ren Wei <enjou1224z(a)gmail.com> Reviewed-by: Phil Sutter <phil(a)nwl.cc> Signed-off-by: Florian Westphal <fw(a)strlen.de> Conflicts: net/netfilter/xt_connmark.c [commit 0bfcb7b71e735 registers CONNMARK rev2 separately for NFPROTO_IPV4 and NFPROTO_IPV6, which is not merged] Signed-off-by: Dong Chenchen <dongchenchen2(a)huawei.com> --- net/netfilter/xt_connmark.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/net/netfilter/xt_connmark.c b/net/netfilter/xt_connmark.c index ad3c033db64e..3004c5229938 100644 --- a/net/netfilter/xt_connmark.c +++ b/net/netfilter/xt_connmark.c @@ -112,6 +112,16 @@ static int connmark_tg_check(const struct xt_tgchk_param *par) return ret; } +static int connmark_tg_check_v2(const struct xt_tgchk_param *par) +{ + const struct xt_connmark_tginfo2 *info = par->targinfo; + + if (info->shift_dir > D_SHIFT_RIGHT || info->shift_bits >= 32) + return -EINVAL; + + return connmark_tg_check(par); +} + static void connmark_tg_destroy(const struct xt_tgdtor_param *par) { nf_ct_netns_put(par->net, par->family); @@ -162,7 +172,7 @@ static struct xt_target connmark_tg_reg[] __read_mostly = { .name = "CONNMARK", .revision = 2, .family = NFPROTO_UNSPEC, - .checkentry = connmark_tg_check, + .checkentry = connmark_tg_check_v2, .target = connmark_tg_v2, .targetsize = sizeof(struct xt_connmark_tginfo2), .destroy = connmark_tg_destroy, -- 2.43.0

2 1

[PATCH OLK-5.10] openvswitch: fix GSO userspace truncation underflow
by Yi Yang 02 Sep '26

02 Sep '26

From: Kyle Zeng <kylebot(a)openai.com> stable inclusion from stable-v5.10.265 commit 50a6a85f3d6b1d22d8436848606cdef5d2c490b4 category: bugfix bugzilla: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 4032f8ed10fcb84d41c508dfb04be96589f78dfe ] OVS_ACTION_ATTR_TRUNC currently stores a delta from the original skb length in OVS_CB(skb)->cutlen. When a later userspace action segments a GSO skb, queue_gso_packets() reuses that delta for each smaller segment. A segment can then reach queue_userspace_packet() with cutlen greater than skb->len, underflowing the length passed to skb_zerocopy(). Store the maximum preserved length instead and bound each consumer against the current skb length. Use U32_MAX as the no-truncation sentinel so the value remains valid if skb geometry changes before a consumer handles it. Fixes: f2a4d086ed4c ("openvswitch: Add packet truncation support.") Cc: stable(a)vger.kernel.org Assisted-by: Codex:gpt-5.5 Signed-off-by: Kyle Zeng <kylebot(a)openai.com> Reviewed-by: Ilya Maximets <i.maximets(a)ovn.org> Reviewed-by: Aaron Conole <aconole(a)redhat.com> Link: https://patch.msgid.link/20260707221635.27489-1-kylebot@openai.com Signed-off-by: Paolo Abeni <pabeni(a)redhat.com> [5.10.y supports neither OVS_ACTION_ATTR_PSAMPLE nor OVS drop reasons] Signed-off-by: Ilya Maximets <i.maximets(a)ovn.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- net/openvswitch/actions.c | 15 +++++---------- net/openvswitch/datapath.c | 25 ++++++++++++++----------- net/openvswitch/datapath.h | 2 +- net/openvswitch/vport.c | 2 +- 4 files changed, 21 insertions(+), 23 deletions(-) diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index a02ea493c269..1e087b7f6dc6 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -918,12 +918,8 @@ static void do_output(struct datapath *dp, struct sk_buff *skb, int out_port, u16 mru = OVS_CB(skb)->mru; u32 cutlen = OVS_CB(skb)->cutlen; - if (unlikely(cutlen > 0)) { - if (skb->len - cutlen > ovs_mac_header_len(key)) - pskb_trim(skb, skb->len - cutlen); - else - pskb_trim(skb, ovs_mac_header_len(key)); - } + if (unlikely(cutlen < skb->len)) + pskb_trim(skb, max(cutlen, ovs_mac_header_len(key))); if (likely(!mru || (skb->len <= mru + vport->dev->hard_header_len))) { @@ -1299,22 +1295,21 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb, clone = skb_clone(skb, GFP_ATOMIC); if (clone) do_output(dp, clone, port, key); - OVS_CB(skb)->cutlen = 0; + OVS_CB(skb)->cutlen = U32_MAX; break; } case OVS_ACTION_ATTR_TRUNC: { struct ovs_action_trunc *trunc = nla_data(a); - if (skb->len > trunc->max_len) - OVS_CB(skb)->cutlen = skb->len - trunc->max_len; + OVS_CB(skb)->cutlen = trunc->max_len; break; } case OVS_ACTION_ATTR_USERSPACE: output_userspace(dp, skb, key, a, attr, len, OVS_CB(skb)->cutlen); - OVS_CB(skb)->cutlen = 0; + OVS_CB(skb)->cutlen = U32_MAX; break; case OVS_ACTION_ATTR_HASH: diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index 1c69aa986633..529eb9d6acab 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -240,7 +240,7 @@ void ovs_dp_process_packet(struct sk_buff *skb, struct sw_flow_key *key) upcall.cmd = OVS_PACKET_CMD_MISS; upcall.portid = ovs_vport_find_upcall_portid(p, skb); upcall.mru = OVS_CB(skb)->mru; - error = ovs_dp_upcall(dp, skb, key, &upcall, 0); + error = ovs_dp_upcall(dp, skb, key, &upcall, U32_MAX); switch (error) { case 0: case -EAGAIN: @@ -400,7 +400,8 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, struct sk_buff *nskb = NULL; struct sk_buff *user_skb = NULL; /* to be queued to userspace */ struct nlattr *nla; - size_t len; + size_t msg_size; + size_t skb_len; unsigned int hlen; int err, dp_ifindex; u64 hash; @@ -421,7 +422,8 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, skb = nskb; } - if (nla_attr_size(skb->len) > USHRT_MAX) { + skb_len = min(skb->len, cutlen); + if (nla_attr_size(skb_len) > USHRT_MAX) { err = -EFBIG; goto out; } @@ -436,13 +438,13 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, * padding logic. Only perform zerocopy if padding is not required. */ if (dp->user_features & OVS_DP_F_UNALIGNED) - hlen = skb_zerocopy_headlen(skb); + hlen = min(skb_zerocopy_headlen(skb), cutlen); else - hlen = skb->len; + hlen = skb_len; - len = upcall_msg_size(upcall_info, hlen - cutlen, - OVS_CB(skb)->acts_origlen); - user_skb = genlmsg_new(len, GFP_ATOMIC); + msg_size = upcall_msg_size(upcall_info, hlen, + OVS_CB(skb)->acts_origlen); + user_skb = genlmsg_new(msg_size, GFP_ATOMIC); if (!user_skb) { err = -ENOMEM; goto out; @@ -503,7 +505,7 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, } /* Add OVS_PACKET_ATTR_LEN when packet is truncated */ - if (cutlen > 0 && + if (skb_len < skb->len && nla_put_u32(user_skb, OVS_PACKET_ATTR_LEN, skb->len)) { err = -ENOBUFS; goto out; @@ -528,9 +530,9 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, err = -ENOBUFS; goto out; } - nla->nla_len = nla_attr_size(skb->len - cutlen); + nla->nla_len = nla_attr_size(skb_len); - err = skb_zerocopy(user_skb, skb, skb->len - cutlen, hlen); + err = skb_zerocopy(user_skb, skb, skb_len, hlen); if (err) goto out; @@ -587,6 +589,7 @@ static int ovs_packet_cmd_execute(struct sk_buff *skb, struct genl_info *info) packet->ignore_df = 1; } OVS_CB(packet)->mru = mru; + OVS_CB(packet)->cutlen = U32_MAX; if (a[OVS_PACKET_ATTR_HASH]) { hash = nla_get_u64(a[OVS_PACKET_ATTR_HASH]); diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h index 38f7d3e66ca6..62e7ad32f837 100644 --- a/net/openvswitch/datapath.h +++ b/net/openvswitch/datapath.h @@ -96,7 +96,7 @@ struct datapath { * @mru: The maximum received fragement size; 0 if the packet is not * fragmented. * @acts_origlen: The netlink size of the flow actions applied to this skb. - * @cutlen: The number of bytes from the packet end to be removed. + * @cutlen: The number of bytes in the packet to preserve on output. */ struct ovs_skb_cb { struct vport *input_vport; diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c index da733b92ae8a..ced707ff6dd9 100644 --- a/net/openvswitch/vport.c +++ b/net/openvswitch/vport.c @@ -436,7 +436,7 @@ int ovs_vport_receive(struct vport *vport, struct sk_buff *skb, OVS_CB(skb)->input_vport = vport; OVS_CB(skb)->mru = 0; - OVS_CB(skb)->cutlen = 0; + OVS_CB(skb)->cutlen = U32_MAX; if (unlikely(dev_net(skb->dev) != ovs_dp_get_net(vport->dp))) { u32 mark; -- 2.25.1

2 1

[PATCH OLK-5.10 0/2] CVE-2026-72390: Fix race and UAF in sch_teql
by superdcc97＠163.com 02 Sep '26

02 Sep '26

From: Dong Chenchen <dongchenchen2(a)huawei.com> This series backports the CVE-2026-72390 fix for the OLK-5.10 kernel. Patch 1 introduces a per-master slaves_lock spinlock and RCU accessors around the master->slaves list to prevent concurrent mutation and use-after-free in the TEQL qdisc. Patch 2 is a follow-up that switches the rcu/spin primitives in teql_master_xmit() from the _bh variants to plain variants, since ndo_start_xmit is already called with BH disabled. Both patches preserve the original author sign-offs and add the adapter signature. Jamal Hadi Salim (2): net/sched: sch_teql: Introduce slaves_lock to avoid race condition and UAF net/sched: sch_teql: move rcu_read_lock()/spin_lock() from _bh variants net/sched/sch_teql.c | 128 +++++++++++++++++++++++++++++-------------- 1 file changed, 88 insertions(+), 40 deletions(-) -- 2.43.0

2 3

[PATCH OLK-5.10] net/sched: cls_u32: validate offshift to prevent shift-out-of-bounds
by superdcc97＠163.com 02 Sep '26

02 Sep '26

From: "Cen Zhang (Microsoft)" <blbllhy(a)gmail.com> mainline inclusion from mainline-v7.2-rc6 commit aef96eead2860cbfa371e4471d4f04412213b958 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/17793 CVE: CVE-2026-74544 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- u32_change() copies the user-provided tc_u32_sel.offshift (unsigned char, 0-255) into the kernel knode object without bounds validation. When a packet later hits u32_classify() with TC_U32_VAROFFSET set, it evaluates `ntohs(offmask & *data) >> offshift` where the left operand is a 16-bit value promoted to a 32-bit int. Any offshift >= 32 is undefined behavior per C11 6.5.7p3, triggerable by an unprivileged user via user/network namespaces. UBSAN: shift-out-of-bounds in net/sched/cls_u32.c:236:43 shift exponent 32 is too large for 32-bit type int Fix this by rejecting offshift >= 16 during filter creation in u32_change(). Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: AutonomousCodeSecurity(a)microsoft.com Link: https://lore.kernel.org/all/20260720034514.23053-1-blbllhy@gmail.com Signed-off-by: Cen Zhang (Microsoft) <blbllhy(a)gmail.com> Acked-by: Jamal Hadi Salim <jhs(a)mojatatu.com> Tested-by: Jamal Hadi Salim <jhs(a)mojatatu.com> Tested-by: Victor Nogueira <victor(a)mojatatu.com> Link: https://patch.msgid.link/20260723044955.89471-1-blbllhy@gmail.com Signed-off-by: Paolo Abeni <pabeni(a)redhat.com> Conflicts: net/sched/cls_u32.c [commit 69050f8d6d07 is not backport, which lead to context conflict] Signed-off-by: Dong Chenchen <dongchenchen2(a)huawei.com> --- net/sched/cls_u32.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c index e501390ccd75..0db59ea65e48 100644 --- a/net/sched/cls_u32.c +++ b/net/sched/cls_u32.c @@ -1067,6 +1067,13 @@ static int u32_change(struct net *net, struct sk_buff *in_skb, goto erridr; } + if (s->offshift >= 16) { + NL_SET_ERR_MSG_MOD(extack, + "offshift must be less than 16"); + err = -EINVAL; + goto erridr; + } + n = kzalloc(struct_size(n, sel.keys, s->nkeys), GFP_KERNEL); if (n == NULL) { err = -ENOBUFS; -- 2.43.0

2 1

[PATCH OLK-5.10] netfilter: nf_tables: make nft_object rhltable per table
by superdcc97＠163.com 02 Sep '26

02 Sep '26

From: Pablo Neira Ayuso <pablo(a)netfilter.org> mainline inclusion from mainline-v7.2-rc6 commit f4f699790590bd0896c48a71e9232a65198f92f0 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/17332 CVE: CVE-2026-74565 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- The nft_object rhltable is global, this allows for accessing objects that are being dismangled from lookup path by other existing netns. Given the nft_obj_destroy() releases the object inmediately, this might lead to use-after-free of these objects that are being released. Make the existing rhltable per table to address this issue to deal with with the nft_rcv_nl_event() path too. Update nft_obj_lookup() to take the table as non-const, otherwise, compiler complains when passing the objname_ht to rhltable_lookup(). Fixes: 4d44175aa5bb ("netfilter: nf_tables: handle nft_object lookups via rhltable") Suggested-by: Florian Westphal <fw(a)strlen.de> Signed-off-by: Pablo Neira Ayuso <pablo(a)netfilter.org> Conflicts: include/net/netfilter/nf_tables.h net/netfilter/nf_tables_api.c [commit f4f699790590bd0896c48a71e9232a65198f92f0 ("netfilter: nf_tables: make nft_object rhltable per table") is not backport, which lead to conflicts] [Structural conflict, no upstream prerequisite found in Fixes..CVE range: all flagged deviations are 5.10 branch code shape, not missing logic. classification: Class B forced-minimal adaptation; target control flow preserved; only CVE-related per-object-hashtable changes applied.] [Deviations: (1) 5.10 nf_tables_table_destroy() takes struct nft_ctx *ctx instead of struct nft_table *table, so objname_ht is destroyed via ctx->table->objname_ht, equivalent to upstream table->objname_ht; (2) 5.10 commit loop has no loop-local table variable (upstream has table = trans->table), so nft_obj_del() is called with trans->ctx.table, which is the same table the object was inserted into; (3) 5.10 merged nf_tables_getobj() (upstream split into nf_tables_getobj_single), the const removal on table applies there; (4) 5.10 module_init uses positional err1..err6 labels (upstream uses named labels); removing the global rhltable_init block requires renumbering so that nft_offload_init failure unwinds through err4 (unregister_netdevice_notifier) and nfnetlink_subsys_register failure through err5 (nft_offload_exit), preserving the exact upstream cleanup order and removing only the global rhltable_destroy for the object-name hashtable.] [Refcount/lifetime safety: object name-hashtable insertion and removal are paired at the table level (rhltable_insert in nf_tables_newobj vs rhltable_remove in nft_obj_del); the per-table hashtable is initialized in nf_tables_newtable and torn down in nf_tables_table_destroy / err_trans path; no dev_hold/refcount/lock operations were added, removed, or reordered relative to upstream; the CVE fix (isolating object lookup from concurrently-released objects across netns) is preserved verbatim.] Signed-off-by: Dong Chenchen <dongchenchen2(a)huawei.com> --- include/net/netfilter/nf_tables.h | 4 +++- net/netfilter/nf_tables_api.c | 40 +++++++++++++++---------------- 2 files changed, 22 insertions(+), 22 deletions(-) diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 9168a6a8934c..fb9de90d00e7 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -1103,6 +1103,7 @@ static inline void nft_use_inc_restore(u32 *use) * @sets: sets in the table * @objects: stateful objects in the table * @flowtables: flow tables in the table + * @objname_ht: hashtable for objects lookup by name * @hgenerator: handle generator state * @handle: table handle * @use: number of chain references to this table @@ -1118,6 +1119,7 @@ struct nft_table { struct list_head sets; struct list_head objects; struct list_head flowtables; + struct rhltable objname_ht; u64 hgenerator; u64 handle; u32 use; @@ -1190,7 +1192,7 @@ static inline void *nft_obj_data(const struct nft_object *obj) #define nft_expr_obj(expr) *((struct nft_object **)nft_expr_priv(expr)) struct nft_object *nft_obj_lookup(const struct net *net, - const struct nft_table *table, + struct nft_table *table, const struct nlattr *nla, u32 objtype, u8 genmask); diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 2beef989cced..fd498aecdd13 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -44,8 +44,6 @@ enum { NFT_VALIDATE_DO, }; -static struct rhltable nft_objname_ht; - static u32 nft_chain_hash(const void *data, u32 len, u32 seed); static u32 nft_chain_hash_obj(const void *data, u32 len, u32 seed); static int nft_chain_hash_cmp(struct rhashtable_compare_arg *, const void *); @@ -1300,6 +1298,10 @@ static int nf_tables_newtable(struct net *net, struct sock *nlsk, if (err) goto err_chain_ht; + err = rhltable_init(&table->objname_ht, &nft_objname_ht_params); + if (err < 0) + goto err_obj_ht; + INIT_LIST_HEAD(&table->chains); INIT_LIST_HEAD(&table->sets); INIT_LIST_HEAD(&table->objects); @@ -1316,6 +1318,8 @@ static int nf_tables_newtable(struct net *net, struct sock *nlsk, list_add_tail_rcu(&table->list, &nft_net->tables); return 0; err_trans: + rhltable_destroy(&table->objname_ht); +err_obj_ht: rhltable_destroy(&table->chains_ht); err_chain_ht: kfree(table->udata); @@ -1474,6 +1478,7 @@ static void nf_tables_table_destroy(struct nft_ctx *ctx) return; rhltable_destroy(&ctx->table->chains_ht); + rhltable_destroy(&ctx->table->objname_ht); kfree(ctx->table->name); kfree(ctx->table->udata); kfree(ctx->table); @@ -6150,7 +6155,7 @@ void nft_unregister_obj(struct nft_object_type *obj_type) EXPORT_SYMBOL_GPL(nft_unregister_obj); struct nft_object *nft_obj_lookup(const struct net *net, - const struct nft_table *table, + struct nft_table *table, const struct nlattr *nla, u32 objtype, u8 genmask) { @@ -6166,7 +6171,7 @@ struct nft_object *nft_obj_lookup(const struct net *net, !lockdep_commit_lock_is_held(net)); rcu_read_lock(); - list = rhltable_lookup(&nft_objname_ht, &k, nft_objname_ht_params); + list = rhltable_lookup(&table->objname_ht, &k, nft_objname_ht_params); if (!list) goto out; @@ -6436,7 +6441,7 @@ static int nf_tables_newobj(struct net *net, struct sock *nlsk, if (err < 0) goto err_trans; - err = rhltable_insert(&nft_objname_ht, &obj->rhlhead, + err = rhltable_insert(&table->objname_ht, &obj->rhlhead, nft_objname_ht_params); if (err < 0) goto err_obj_ht; @@ -6622,7 +6627,7 @@ static int nf_tables_getobj(struct net *net, struct sock *nlsk, const struct nfgenmsg *nfmsg = nlmsg_data(nlh); u8 genmask = nft_genmask_cur(net); int family = nfmsg->nfgen_family; - const struct nft_table *table; + struct nft_table *table; struct nft_object *obj; struct sk_buff *skb2; bool reset = false; @@ -8159,9 +8164,9 @@ static void nf_tables_commit_chain(struct net *net, struct nft_chain *chain) nf_tables_commit_chain_free_rules_old(g0); } -static void nft_obj_del(struct nft_object *obj) +static void nft_obj_del(struct nft_table *table, struct nft_object *obj) { - rhltable_remove(&nft_objname_ht, &obj->rhlhead, nft_objname_ht_params); + rhltable_remove(&table->objname_ht, &obj->rhlhead, nft_objname_ht_params); list_del_rcu(&obj->list); } @@ -8741,7 +8746,7 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) } break; case NFT_MSG_DELOBJ: - nft_obj_del(nft_trans_obj(trans)); + nft_obj_del(trans->ctx.table, nft_trans_obj(trans)); nf_tables_obj_notify(&trans->ctx, nft_trans_obj(trans), NFT_MSG_DELOBJ); break; @@ -8988,7 +8993,7 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action) nft_trans_destroy(trans); } else { nft_use_dec_restore(&trans->ctx.table->use); - nft_obj_del(nft_trans_obj(trans)); + nft_obj_del(trans->ctx.table, nft_trans_obj(trans)); } break; case NFT_MSG_DELOBJ: @@ -9633,7 +9638,7 @@ static void __nft_release_table(struct net *net, struct nft_table *table) nft_set_destroy(&ctx, set); } list_for_each_entry_safe(obj, ne, &table->objects, list) { - nft_obj_del(obj); + nft_obj_del(table, obj); nft_use_dec(&table->use); nft_obj_destroy(&ctx, obj); } @@ -9734,26 +9739,20 @@ static int __init nf_tables_module_init(void) if (err < 0) goto err3; - err = rhltable_init(&nft_objname_ht, &nft_objname_ht_params); - if (err < 0) - goto err4; - err = nft_offload_init(); if (err < 0) - goto err5; + goto err4; /* must be last */ err = nfnetlink_subsys_register(&nf_tables_subsys); if (err < 0) - goto err6; + goto err5; nft_chain_route_init(); return err; -err6: - nft_offload_exit(); err5: - rhltable_destroy(&nft_objname_ht); + nft_offload_exit(); err4: unregister_netdevice_notifier(&nf_tables_flowtable_notifier); err3: @@ -9777,7 +9776,6 @@ static void __exit nf_tables_module_exit(void) cancel_work_sync(&trans_gc_work); cancel_work_sync(&trans_destroy_work); rcu_barrier(); - rhltable_destroy(&nft_objname_ht); nf_tables_core_module_exit(); } -- 2.43.0

2 1

[PATCH OLK-5.10] xfrm: fix stale skb->prev after async crypto steals a GSO segment
by superdcc97＠163.com 02 Sep '26

02 Sep '26

From: Petr Wozniak <petr.wozniak(a)gmail.com> mainline inclusion from mainline-v7.2-rc4 commit 3f4c3919baf0944ad96580467c302bc6c7758b00 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/17032 CVE: CVE-2026-68426 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- skb_gso_segment() leaves the segment list head with ->prev pointing at the last segment, an invariant validate_xmit_skb_list() relies on when it sets its tail pointer (tail = skb->prev). When validate_xmit_xfrm() walks a GSO list and some segments are stolen by async crypto (->xmit() returns -EINPROGRESS), those segments are unlinked from the list but the head ->prev is never updated. If the last segment is the one stolen, the returned head still has ->prev pointing at it, even though it is now owned by the crypto engine and may be freed. validate_xmit_skb_list() later does tail->next = skb, writing through that stale pointer -- a use-after-free. Repoint skb->prev at the last retained segment before returning. Fixes: f53c723902d1 ("net: Add asynchronous callbacks for xfrm on layer 2.") Signed-off-by: Petr Wozniak <petr.wozniak(a)gmail.com> Signed-off-by: Steffen Klassert <steffen.klassert(a)secunet.com> Conflicts: net/xfrm/xfrm_device.c [mainline commit 6860b467f569 ("xfrm: propagate -EINPROGRESS from validate_xmit_xfrm()") is not backport, which lead to conflicts: this is a Class B non-Stable-dep-of prerequisite (pure control-flow change turning NULL return into ERR_PTR(-EINPROGRESS), no dev_hold/refcount/ lock/UAF involvement). Applied forced minimal adaptation per P3: kept target's "return skb;" (5.10 callers already handle NULL via "if (!skb) continue;"), applied only the CVE change "if (skb) skb->prev = pskb;" to repoint the GSO head ->prev at the last retained segment so validate_xmit_skb_list() cannot chain onto a segment stolen by async crypto (use-after-free). Refcount/lifetime analysis: added code takes no reference and acquires no lock; ownership of stolen skbs remains with the crypto engine via the xfrm_dev_resume() path, unchanged] Signed-off-by: Dong Chenchen <dongchenchen2(a)huawei.com> --- net/xfrm/xfrm_device.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/net/xfrm/xfrm_device.c b/net/xfrm/xfrm_device.c index 0e7424b8a5fa..062caeb893a5 100644 --- a/net/xfrm/xfrm_device.c +++ b/net/xfrm/xfrm_device.c @@ -215,6 +215,14 @@ struct sk_buff *validate_xmit_xfrm(struct sk_buff *skb, netdev_features_t featur pskb = skb2; } + /* skb_gso_segment() set skb->prev to the last segment, but async + * crypto may have stolen it above without updating ->prev. Repoint + * it at the last retained segment so validate_xmit_skb_list() does + * not chain onto a segment now owned by the crypto engine. + */ + if (skb) + skb->prev = pskb; + return skb; } EXPORT_SYMBOL_GPL(validate_xmit_xfrm); -- 2.43.0

2 2

[PATCH OLK-6.6] tcp: clear sock_ops cb flags before force-closing a child socket
by JiangJieHua 02 Sep '26

02 Sep '26

From: Sechang Lim <rhkrqnwk98(a)gmail.com> mainline inclusion from mainline-v7.2-rc1 commit 990348e5bb457697c2f1f7f7b65154a3334d9d2b category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/17954 CVE: CVE-2026-74268 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- A child socket inherits the listener's bpf_sock_ops_cb_flags via sk_clone_lock(). If its setup fails in tcp_v4_syn_recv_sock() / tcp_v6_syn_recv_sock(), the child is freed through put_and_exit, where inet_csk_prepare_forced_close() drops the socket lock and tcp_done() runs without it. If BPF_SOCK_OPS_STATE_CB_FLAG was inherited, tcp_done() -> tcp_set_state() calls tcp_call_bpf(), which expects the lock and trips sock_owned_by_me(): WARNING: include/net/sock.h:1799 at tcp_set_state+0x433/0x550 RIP: 0010:tcp_set_state+0x433/0x550 include/net/sock.h:1799 Call Trace: <IRQ> tcp_done+0xba/0x250 net/ipv4/tcp.c:5095 tcp_v4_syn_recv_sock+0x850/0xa50 net/ipv4/tcp_ipv4.c:1787 tcp_check_req+0xf30/0x1360 net/ipv4/tcp_minisocks.c:926 tcp_v4_rcv+0x1047/0x1b50 net/ipv4/tcp_ipv4.c:2164 </IRQ> The child is freed before it is ever established, so it should run no sock_ops callback. Clear its cb flags in inet_csk_prepare_for_destroy_sock(), the common point for the IPv4, IPv6 and chtls forced-close paths and for the MPTCP ->syn_recv_sock() failure path (dispose_child), which reaches tcp_done() on a child that was never established too. Suggested-by: Jiayuan Chen <jiayuan.chen(a)linux.dev> Fixes: d44874910a26 ("bpf: Add BPF_SOCK_OPS_STATE_CB") Signed-off-by: Sechang Lim <rhkrqnwk98(a)gmail.com> Reviewed-by: Jiayuan Chen <jiayuan.chen(a)linux.dev> Reviewed-by: Kuniyuki Iwashima <kuniyu(a)google.com> Reviewed-by: Eric Dumazet <edumazet(a)google.com> Link: https://patch.msgid.link/20260611092923.1895982-1-rhkrqnwk98@gmail.com Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Conflicts: include/net/inet_connection_sock.h net/ipv4/inet_connection_sock.c [In OLK-6.6, inet_csk_prepare_for_destroy_sock() remains a static inline function in include/net/inet_connection_sock.h, while upstream already moved it into inet_connection_sock.c via commit 7051b54fb5aa (due to header dependency). This backport cherry-picks the inline-to-non-inline conversion (changing the definition in inet_connection_sock.h to a declaration), retains OLK-6.6's this_cpu_inc(*sk->sk_prot->orphan_count) implementation without pulling in the orphan_count refactoring, and adds a sk->sk_protocol == IPPROTO_TCP guard to isolate the DCCP path.] Signed-off-by: JiangJieHua <jiangjiehua1(a)huawei.com> --- include/net/inet_connection_sock.h | 8 +------- include/net/tcp.h | 9 +++++++++ net/ipv4/inet_connection_sock.c | 9 +++++++++ 3 files changed, 19 insertions(+), 7 deletions(-) diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index 327ae14ca82b2..c16f600e5c5c3 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -303,14 +303,8 @@ reqsk_timeout(struct request_sock *req, unsigned long max_timeout) return (unsigned long)min_t(u64, timeout, max_timeout); } -static inline void inet_csk_prepare_for_destroy_sock(struct sock *sk) -{ - /* The below has to be done to allow calling inet_csk_destroy_sock */ - sock_set_flag(sk, SOCK_DEAD); - this_cpu_inc(*sk->sk_prot->orphan_count); -} - void inet_csk_destroy_sock(struct sock *sk); +void inet_csk_prepare_for_destroy_sock(struct sock *sk); void inet_csk_prepare_forced_close(struct sock *sk); /* diff --git a/include/net/tcp.h b/include/net/tcp.h index 8674996bd39c6..e1a345d5a40b1 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -2515,6 +2515,11 @@ static inline int tcp_call_bpf_3arg(struct sock *sk, int op, u32 arg1, u32 arg2, return tcp_call_bpf(sk, op, 3, args); } +static inline void tcp_clear_sock_ops_cb_flags(struct sock *sk) +{ + tcp_sk(sk)->bpf_sock_ops_cb_flags = 0; +} + #else static inline int tcp_call_bpf(struct sock *sk, int op, u32 nargs, u32 *args) { @@ -2532,6 +2537,10 @@ static inline int tcp_call_bpf_3arg(struct sock *sk, int op, u32 arg1, u32 arg2, return -EPERM; } +static inline void tcp_clear_sock_ops_cb_flags(struct sock *sk) +{ +} + #endif static inline u32 tcp_timeout_init(struct sock *sk) diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 9f1d17cd5eebd..bca256306c9ee 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -1240,6 +1240,15 @@ void inet_csk_destroy_sock(struct sock *sk) } EXPORT_SYMBOL(inet_csk_destroy_sock); +void inet_csk_prepare_for_destroy_sock(struct sock *sk) +{ + /* The below has to be done to allow calling inet_csk_destroy_sock */ + if (sk->sk_protocol == IPPROTO_TCP) + tcp_clear_sock_ops_cb_flags(sk); + sock_set_flag(sk, SOCK_DEAD); + this_cpu_inc(*sk->sk_prot->orphan_count); +} + /* This function allows to force a closure of a socket after the call to * tcp/dccp_create_openreq_child(). */ -- 2.33.8

2 1

[PATCH OLK-6.6] tcp: clear sock_ops cb flags before force-closing a child socket
by JiangJieHua 02 Sep '26

02 Sep '26

From: Sechang Lim <rhkrqnwk98(a)gmail.com> mainline inclusion from mainline-v7.2-rc1 commit 990348e5bb457697c2f1f7f7b65154a3334d9d2b category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/17954 CVE: CVE-2026-74268 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- A child socket inherits the listener's bpf_sock_ops_cb_flags via sk_clone_lock(). If its setup fails in tcp_v4_syn_recv_sock() / tcp_v6_syn_recv_sock(), the child is freed through put_and_exit, where inet_csk_prepare_forced_close() drops the socket lock and tcp_done() runs without it. If BPF_SOCK_OPS_STATE_CB_FLAG was inherited, tcp_done() -> tcp_set_state() calls tcp_call_bpf(), which expects the lock and trips sock_owned_by_me(): WARNING: include/net/sock.h:1799 at tcp_set_state+0x433/0x550 RIP: 0010:tcp_set_state+0x433/0x550 include/net/sock.h:1799 Call Trace: <IRQ> tcp_done+0xba/0x250 net/ipv4/tcp.c:5095 tcp_v4_syn_recv_sock+0x850/0xa50 net/ipv4/tcp_ipv4.c:1787 tcp_check_req+0xf30/0x1360 net/ipv4/tcp_minisocks.c:926 tcp_v4_rcv+0x1047/0x1b50 net/ipv4/tcp_ipv4.c:2164 </IRQ> The child is freed before it is ever established, so it should run no sock_ops callback. Clear its cb flags in inet_csk_prepare_for_destroy_sock(), the common point for the IPv4, IPv6 and chtls forced-close paths and for the MPTCP ->syn_recv_sock() failure path (dispose_child), which reaches tcp_done() on a child that was never established too. Suggested-by: Jiayuan Chen <jiayuan.chen(a)linux.dev> Fixes: d44874910a26 ("bpf: Add BPF_SOCK_OPS_STATE_CB") Signed-off-by: Sechang Lim <rhkrqnwk98(a)gmail.com> Reviewed-by: Jiayuan Chen <jiayuan.chen(a)linux.dev> Reviewed-by: Kuniyuki Iwashima <kuniyu(a)google.com> Reviewed-by: Eric Dumazet <edumazet(a)google.com> Link: https://patch.msgid.link/20260611092923.1895982-1-rhkrqnwk98@gmail.com Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Conflicts: net/ipv4/inet_connection_sock.h net/ipv4/inet_connection_sock.c [In OLK-6.6, inet_csk_prepare_for_destroy_sock() remains a static inline function in include/net/inet_connection_sock.h, while upstream already moved it into inet_connection_sock.c via commit 7051b54fb5aa (due to header dependency). This backport cherry-picks the inline-to-non-inline conversion (changing the definition in inet_connection_sock.h to a declaration), retains OLK-6.6's this_cpu_inc(*sk->sk_prot->orphan_count) implementation without pulling in the orphan_count refactoring, and adds a sk->sk_protocol == IPPROTO_TCP guard to isolate the DCCP path.] Signed-off-by: JiangJieHua <jiangjiehua1(a)huawei.com> --- include/net/inet_connection_sock.h | 8 +------- include/net/tcp.h | 9 +++++++++ net/ipv4/inet_connection_sock.c | 9 +++++++++ 3 files changed, 19 insertions(+), 7 deletions(-) diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index 327ae14ca82b2..c16f600e5c5c3 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -303,14 +303,8 @@ reqsk_timeout(struct request_sock *req, unsigned long max_timeout) return (unsigned long)min_t(u64, timeout, max_timeout); } -static inline void inet_csk_prepare_for_destroy_sock(struct sock *sk) -{ - /* The below has to be done to allow calling inet_csk_destroy_sock */ - sock_set_flag(sk, SOCK_DEAD); - this_cpu_inc(*sk->sk_prot->orphan_count); -} - void inet_csk_destroy_sock(struct sock *sk); +void inet_csk_prepare_for_destroy_sock(struct sock *sk); void inet_csk_prepare_forced_close(struct sock *sk); /* diff --git a/include/net/tcp.h b/include/net/tcp.h index 8674996bd39c6..e1a345d5a40b1 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -2515,6 +2515,11 @@ static inline int tcp_call_bpf_3arg(struct sock *sk, int op, u32 arg1, u32 arg2, return tcp_call_bpf(sk, op, 3, args); } +static inline void tcp_clear_sock_ops_cb_flags(struct sock *sk) +{ + tcp_sk(sk)->bpf_sock_ops_cb_flags = 0; +} + #else static inline int tcp_call_bpf(struct sock *sk, int op, u32 nargs, u32 *args) { @@ -2532,6 +2537,10 @@ static inline int tcp_call_bpf_3arg(struct sock *sk, int op, u32 arg1, u32 arg2, return -EPERM; } +static inline void tcp_clear_sock_ops_cb_flags(struct sock *sk) +{ +} + #endif static inline u32 tcp_timeout_init(struct sock *sk) diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 9f1d17cd5eebd..bca256306c9ee 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -1240,6 +1240,15 @@ void inet_csk_destroy_sock(struct sock *sk) } EXPORT_SYMBOL(inet_csk_destroy_sock); +void inet_csk_prepare_for_destroy_sock(struct sock *sk) +{ + /* The below has to be done to allow calling inet_csk_destroy_sock */ + if (sk->sk_protocol == IPPROTO_TCP) + tcp_clear_sock_ops_cb_flags(sk); + sock_set_flag(sk, SOCK_DEAD); + this_cpu_inc(*sk->sk_prot->orphan_count); +} + /* This function allows to force a closure of a socket after the call to * tcp/dccp_create_openreq_child(). */ -- 2.33.8

2 1

[PATCH OLK-5.10] bpf: Fix UAF in sock clone early bailouts
by superdcc97＠163.com 02 Sep '26

02 Sep '26

From: Matt Bobrowski <mattbobrowski(a)google.com> mainline inclusion from mainline-v7.2-rc4 commit 7cbd0c4cebe4c9f678d15e6b9ba975e1155a107f category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/17028 CVE: CVE-2026-68399 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Similar to recent commit 9b51a6155d14 ("bpf,fork: wipe ->bpf_storage before bailouts that access it"), sk_clone() performs an initial shallow copy of the socket field ->sk_bpf_storage via sock_copy() for the cloned socket newsk. If sk_clone() bails out early (e.g. if sk_filter_charge() fails) prior to calling bpf_sk_storage_clone(), newsk->sk_bpf_storage still points to the parent socket's BPF local storage. When newsk is subsequently freed via sk_free(), the deallocation path (__sk_destruct() -> bpf_sk_storage_free()) destroys the parent socket's BPF local storage, leading to a use-after-free (UAF) on the parent socket. Fix this by resetting newsk->sk_bpf_storage to NULL immediately after sock_copy() in sk_clone(), and remove the now redundant initialization from bpf_sk_storage_clone(). Fixes: 6ac99e8f23d4 ("bpf: Introduce bpf sk local storage") Fixes: f12dd75959b0 ("bpf: net: Set sk_bpf_storage back to NULL for cloned sk") Signed-off-by: Matt Bobrowski <mattbobrowski(a)google.com> Signed-off-by: Daniel Borkmann <daniel(a)iogearbox.net> Reviewed-by: Kuniyuki Iwashima <kuniyu(a)google.com> Acked-by: Daniel Borkmann <daniel(a)iogearbox.net> Link: https://lore.kernel.org/bpf/20260709025316.999913-1-mattbobrowski@google.com Conflicts: net/core/bpf_sk_storage.c net/core/sock.c [commit 0db4941d9dae1 uses rcu_read_lock_dont_migrate() which does not exist on 5.10 (uses rcu_read_lock()), so the hunk was adapted manually] Signed-off-by: Dong Chenchen <dongchenchen2(a)huawei.com> --- net/core/bpf_sk_storage.c | 2 -- net/core/sock.c | 3 +++ 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c index 3fad2f5b920e..7923d1a52fe4 100644 --- a/net/core/bpf_sk_storage.c +++ b/net/core/bpf_sk_storage.c @@ -190,8 +190,6 @@ int bpf_sk_storage_clone(const struct sock *sk, struct sock *newsk) struct bpf_local_storage_elem *selem; int ret = 0; - RCU_INIT_POINTER(newsk->sk_bpf_storage, NULL); - rcu_read_lock(); sk_storage = rcu_dereference(sk->sk_bpf_storage); diff --git a/net/core/sock.c b/net/core/sock.c index 8904b0d87751..e1380c3ebf17 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1939,6 +1939,9 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority) sock_copy(newsk, sk); newsk->sk_prot_creator = prot; +#ifdef CONFIG_BPF_SYSCALL + RCU_INIT_POINTER(newsk->sk_bpf_storage, NULL); +#endif /* SANITY */ if (likely(newsk->sk_net_refcnt)) { -- 2.43.0

2 1

[PATCH OLK-6.6 V1 0/3] CVE-2026-80721
by Yao Yiqi 02 Sep '26

02 Sep '26

Pauli Virtanen (3): Bluetooth: ISO: avoid deadlocks in iso_sock_timeout Bluetooth: ISO: ensure no dangling hcon references in iso_conn Bluetooth: ISO: fix refcounting of iso_conn net/bluetooth/iso.c | 107 ++++++++++++++++++++++++++------------------ 1 file changed, 64 insertions(+), 43 deletions(-) -- 2.34.1

2 4

[PATCH OLK-6.6] net/sched: serialize qdisc_rtab_list against concurrent get/put
by superdcc97＠163.com 01 Sep '26

01 Sep '26

From: Aldo Ariel Panzardo <qwe.aldo(a)gmail.com> stable inclusion from stable-v6.6.153 commit 4131dd0b6f67acddd616ed7c244e1d3eedd46e7b category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/16947 CVE: CVE-2026-68138 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit f43ee0c0730d6191629b5ee1ceae27b1ebfdc047 ] qdisc_get_rtab() and qdisc_put_rtab() mutate the process-global singly linked list qdisc_rtab_list and a plain non-atomic 'int refcnt' with no lock. This was only safe because every caller historically held the RTNL mutex, which serialized all rate-table lookups, inserts and frees. That invariant no longer holds. cls_flower sets TCF_PROTO_OPS_DOIT_UNLOCKED, so tc_new_tfilter() keeps rtnl_held == false for it and sets TCA_ACT_FLAGS_NO_RTNL. That flag propagates through tcf_exts_validate_ex() -> tcf_action_init() -> tcf_action_init_1() -> tcf_police_init(), which calls qdisc_get_rtab()/qdisc_put_rtab() with the RTNL mutex NOT held. Two RTM_NEWTFILTER requests on different CPUs, each adding a flower filter with a police action carrying the same rate, then race on qdisc_rtab_list and on the non-atomic refcnt, leading to a use-after-free / double-free of the kmalloc-2k struct qdisc_rate_table. qdisc_rtab_list is a single global (not per-netns), so the corrupted object is shared system-wide. BUG: KASAN: slab-use-after-free in qdisc_put_rtab+0x12f/0x160 qdisc_put_rtab+0x12f/0x160 tcf_police_init+0xda9/0x1590 tcf_action_init_1+0x460/0x6b0 tcf_action_init+0x439/0xa40 tcf_exts_validate_ex+0x42d/0x550 fl_change+0xddd/0x7da0 tc_new_tfilter+0xaa7/0x2420 rtnetlink_rcv_msg+0x95e/0xe90 which belongs to the cache kmalloc-2k of size 2048 Protect qdisc_rtab_list and the refcount with a dedicated spinlock. The (sleeping, GFP_KERNEL) allocation in qdisc_get_rtab() is performed before taking the lock; if a concurrent inserter added an identical table in the meantime the freshly allocated one is freed under the lock, so no duplicate is leaked. qdisc_put_rtab() now decrements the refcount and unlinks under the same lock. Fixes: 470502de5bdb ("net: sched: unlock rules update API") Suggested-by: Eric Dumazet <edumazet(a)google.com> Signed-off-by: Aldo Ariel Panzardo <qwe.aldo(a)gmail.com> Cc: stable(a)vger.kernel.org Acked-by: Jamal Hadi Salim <jhs(a)mojatatu.com> Reviewed-by: Eric Dumazet <edumazet(a)google.com> Link: https://patch.msgid.link/20260715114114.446841-1-qwe.aldo@gmail.com Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Dong Chenchen <dongchenchen2(a)huawei.com> --- net/sched/sch_api.c | 24 +++++++++++++++++++----- 1 file changed, 19 insertions(+), 5 deletions(-) diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index ff7bad762496..3f5db8d378a8 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -411,33 +411,39 @@ static __u8 __detect_linklayer(struct tc_ratespec *r, __u32 *rtab) } return TC_LINKLAYER_ETHERNET; } static struct qdisc_rate_table *qdisc_rtab_list; +static DEFINE_SPINLOCK(qdisc_rtab_lock); struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r, struct nlattr *tab, struct netlink_ext_ack *extack) { - struct qdisc_rate_table *rtab; + struct qdisc_rate_table *rtab, *new_rtab; if (tab == NULL || r->rate == 0 || r->cell_log == 0 || r->cell_log >= 32 || nla_len(tab) != TC_RTAB_SIZE) { NL_SET_ERR_MSG(extack, "Invalid rate table parameters for searching"); return NULL; } + new_rtab = kmalloc(sizeof(*new_rtab), GFP_KERNEL); + + spin_lock(&qdisc_rtab_lock); for (rtab = qdisc_rtab_list; rtab; rtab = rtab->next) { if (!memcmp(&rtab->rate, r, sizeof(struct tc_ratespec)) && !memcmp(&rtab->data, nla_data(tab), 1024)) { rtab->refcnt++; + spin_unlock(&qdisc_rtab_lock); + kfree(new_rtab); return rtab; } } - rtab = kmalloc(sizeof(*rtab), GFP_KERNEL); + rtab = new_rtab; if (rtab) { rtab->rate = *r; rtab->refcnt = 1; memcpy(rtab->data, nla_data(tab), 1024); if (r->linklayer == TC_LINKLAYER_UNAWARE) @@ -445,30 +451,38 @@ struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r, rtab->next = qdisc_rtab_list; qdisc_rtab_list = rtab; } else { NL_SET_ERR_MSG(extack, "Failed to allocate new qdisc rate table"); } + spin_unlock(&qdisc_rtab_lock); return rtab; } EXPORT_SYMBOL(qdisc_get_rtab); void qdisc_put_rtab(struct qdisc_rate_table *tab) { struct qdisc_rate_table *rtab, **rtabp; - if (!tab || --tab->refcnt) + if (!tab) return; + spin_lock(&qdisc_rtab_lock); + if (--tab->refcnt) { + spin_unlock(&qdisc_rtab_lock); + return; + } + for (rtabp = &qdisc_rtab_list; (rtab = *rtabp) != NULL; rtabp = &rtab->next) { if (rtab == tab) { *rtabp = rtab->next; - kfree(rtab); - return; + break; } } + spin_unlock(&qdisc_rtab_lock); + kfree(tab); } EXPORT_SYMBOL(qdisc_put_rtab); static LIST_HEAD(qdisc_stab_list); -- 2.43.0

2 1

[PATCH OLK-6.6] workqueue: Reduce expensive locks for unbound workqueue
by Yao Kai 01 Sep '26

01 Sep '26

From: Wangyang Guo <wangyang.guo(a)intel.com> mainline inclusion from mainline-v6.13-rc1 commit 85f0d8e39affb7b88401b1e0542230a7af985b96 category: perf bugzilla: https://atomgit.com/openeuler/kernel/issues/9846 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- For unbound workqueue, pwqs usually map to just a few pools. Most of the time, pwqs will be linked sequentially to wq->pwqs list by cpu index. Usually, consecutive CPUs have the same workqueue attribute (e.g. belong to the same NUMA node). This makes pwqs with the same pool cluster together in the pwq list. Only do lock/unlock if the pool has changed in flush_workqueue_prep_pwqs(). This reduces the number of expensive lock operations. The performance data shows this change boosts FIO by 65x in some cases when multiple concurrent threads write to xfs mount points with fsync. FIO Benchmark Details - FIO version: v3.35 - FIO Options: ioengine=libaio,iodepth=64,norandommap=1,rw=write, size=128M,bs=4k,fsync=1 - FIO Job Configs: 64 jobs in total writing to 4 mount points (ramdisks formatted as xfs file system). - Kernel Codebase: v6.12-rc5 - Test Platform: Xeon 8380 (2 sockets) Reviewed-by: Tim Chen <tim.c.chen(a)linux.intel.com> Signed-off-by: Wangyang Guo <wangyang.guo(a)intel.com> Reviewed-by: Lai Jiangshan <jiangshanlai(a)gmail.com> Signed-off-by: Tejun Heo <tj(a)kernel.org> Signed-off-by: Yao Kai <yaokai34(a)huawei.com> --- kernel/workqueue.c | 22 ++++++++++++++++++---- 1 file changed, 18 insertions(+), 4 deletions(-) diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 6810faac117a..666764c79df3 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -3243,16 +3243,28 @@ static bool flush_workqueue_prep_pwqs(struct workqueue_struct *wq, { bool wait = false; struct pool_workqueue *pwq; + struct worker_pool *current_pool = NULL; if (flush_color >= 0) { WARN_ON_ONCE(atomic_read(&wq->nr_pwqs_to_flush)); atomic_set(&wq->nr_pwqs_to_flush, 1); } + /* + * For unbound workqueue, pwqs will map to only a few pools. + * Most of the time, pwqs within the same pool will be linked + * sequentially to wq->pwqs by cpu index. So in the majority + * of pwq iters, the pool is the same, only doing lock/unlock + * if the pool has changed. This can largely reduce expensive + * lock operations. + */ for_each_pwq(pwq, wq) { - struct worker_pool *pool = pwq->pool; - - raw_spin_lock_irq(&pool->lock); + if (current_pool != pwq->pool) { + if (likely(current_pool)) + raw_spin_unlock_irq(&current_pool->lock); + current_pool = pwq->pool; + raw_spin_lock_irq(&current_pool->lock); + } if (flush_color >= 0) { WARN_ON_ONCE(pwq->flush_color != -1); @@ -3269,9 +3281,11 @@ static bool flush_workqueue_prep_pwqs(struct workqueue_struct *wq, pwq->work_color = work_color; } - raw_spin_unlock_irq(&pool->lock); } + if (current_pool) + raw_spin_unlock_irq(&current_pool->lock); + if (flush_color >= 0 && atomic_dec_and_test(&wq->nr_pwqs_to_flush)) complete(&wq->first_flusher->done); -- 2.43.0

2 1

[PATCH openEuler-1.0-LTS] ixgbevf: fix use-after-free in VEPA multicast source pruning
by Luo Gengkun 01 Sep '26

01 Sep '26

From: Michael Bommarito <michael.bommarito(a)gmail.com> mainline inclusion from mainline-v7.1-rc5 commit 5d49b568c188dc77199d8d2b959c91da8cc27cf1 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/16356 CVE: CVE-2026-64113 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… ---------------------------------------------------------------------- ixgbevf_clean_rx_irq() prunes frames whose source MAC matches the VF's own address (VEPA multicast workaround) by freeing the skb and continuing to the next descriptor: dev_kfree_skb_irq(skb); continue; The skb pointer is declared outside the while loop and persists across iterations. Because the continue skips the "skb = NULL" reset at the bottom of the loop, the next iteration enters the "else if (skb)" path and calls ixgbevf_add_rx_frag() on the freed skb, dereferencing skb_shinfo(skb)->nr_frags - a use-after-free in NAPI softirq context. The sibling driver iavf already handles this correctly by nulling the pointer before continuing. Apply the same pattern here. I do not have ixgbevf hardware; the bug was found by static analysis (scan_drop_continue_loops.py + semgrep drop_continue_in_loop, multi-tool corroboration with the highest score in the scan). The UAF was confirmed under KASAN by loading a test module that reproduces the exact code pattern (alloc skb, kfree_skb, then read skb_shinfo(skb)->nr_frags): BUG: KASAN: slab-use-after-free in ixgbevf_uaf_test_init+0x100/0x1000 Read of size 8 at addr 000000006163ae78 by task insmod/30 freed 208-byte region [000000006163adc0, 000000006163ae90) QEMU emulates igb (82576) but not ixgbe (82599), and the igbvf VF driver does not include the VEPA source pruning path, so a full end-to-end reproduction with emulated hardware was not possible. Fixes: bad17234ba70 ("ixgbevf: Change receive model to use double buffered page based receives") Cc: stable(a)vger.kernel.org Signed-off-by: Michael Bommarito <michael.bommarito(a)gmail.com> Reviewed-by: Simon Horman <horms(a)kernel.org> Tested-by: Rafal Romanowski <rafal.romanowski(a)intel.com> Signed-off-by: Tony Nguyen <anthony.l.nguyen(a)intel.com> Link: https://patch.msgid.link/20260515182419.1597859-8-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Signed-off-by: Luo Gengkun <luogengkun2(a)huawei.com> --- drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c index 450c0a95d36d..f6e924db3286 100644 --- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c +++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c @@ -1204,6 +1204,7 @@ static int ixgbevf_clean_rx_irq(struct ixgbevf_q_vector *q_vector, ether_addr_equal(rx_ring->netdev->dev_addr, eth_hdr(skb)->h_source)) { dev_kfree_skb_irq(skb); + skb = NULL; continue; } -- 2.34.1

2 1

[PATCH 0/3] CVE-2026-80721
by Yao Yiqi 01 Sep '26

01 Sep '26

Pauli Virtanen (3): Bluetooth: ISO: avoid deadlocks in iso_sock_timeout Bluetooth: ISO: ensure no dangling hcon references in iso_conn Bluetooth: ISO: fix refcounting of iso_conn net/bluetooth/iso.c | 107 ++++++++++++++++++++++++++------------------ 1 file changed, 64 insertions(+), 43 deletions(-) -- 2.34.1

1 3

[PATCH OLK-6.6 V1 1/3] Bluetooth: ISO: avoid deadlocks in iso_sock_timeout
by Yao Yiqi 01 Sep '26

01 Sep '26

From: Pauli Virtanen <pav(a)iki.fi> stable inclusion from stable-v6.18.44 commit 82e982f54f962f72646868ddbb2c3bd9ea178568 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/18470 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 200fa1629c57a3ca2b03d3ca63fd3a9bfd910c43 ] iso_sock_timeout() takes lock_sock, so sync disabling the timer while holding that lock may deadlock. iso_sock_timeout() may also run concurrently with iso_conn_del(), which leads to UAF [Task 1] [Task hdev->workqueue] iso_sock_timeout iso_conn_del iso_conn_hold_unless_zero iso_chan_del `------------> iso_conn_put caller frees hcon iso_conn_put iso_conn_free conn->hcon->iso_data = NULL; /* UAF */ Fix the deadlock by removing the disable from the lock_sock sections. Move the timer from iso_conn to iso_pinfo to decouple it from iso_conn which may need to be freed in lock_sock section. Convert some of the clear_timer to disable_timer. Fixes: dc26097bdb86 ("Bluetooth: ISO: Use kref to track lifetime of iso_conn") Signed-off-by: Pauli Virtanen <pav(a)iki.fi> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz(a)intel.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Conflicts: net/bluetooth/iso.c [Context conflicts] Signed-off-by: Yao Yiqi <yaoyiqi3(a)huawei.com> --- net/bluetooth/iso.c | 60 ++++++++++++++++++++++----------------------- 1 file changed, 29 insertions(+), 31 deletions(-) diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index 56aab26ced0e..97649e371f28 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -36,8 +36,6 @@ struct iso_conn { spinlock_t lock; struct sock *sk; - struct delayed_work timeout_work; - struct sk_buff *rx_skb; __u32 rx_len; __u16 tx_sn; @@ -81,6 +79,7 @@ struct iso_pinfo { __u8 base_len; __u8 base[BASE_MAX_LENGTH]; struct iso_conn *conn; + struct delayed_work timeout_work; }; static struct bt_iso_qos default_qos; @@ -115,9 +114,6 @@ static void iso_conn_free(struct kref *ref) hci_conn_drop(conn->hcon); } - /* Ensure no more work items will run since hci_conn has been dropped */ - disable_delayed_work_sync(&conn->timeout_work); - kfree_skb(conn->rx_skb); kfree(conn); @@ -158,48 +154,45 @@ static struct sock *iso_sock_hold(struct iso_conn *conn) static void iso_sock_timeout(struct work_struct *work) { - struct iso_conn *conn = container_of(work, struct iso_conn, - timeout_work.work); - struct sock *sk; - - conn = iso_conn_hold_unless_zero(conn); - if (!conn) - return; - - iso_conn_lock(conn); - sk = iso_sock_hold(conn); - iso_conn_unlock(conn); - iso_conn_put(conn); - - if (!sk) - return; + struct iso_pinfo *pi = container_of(work, struct iso_pinfo, + timeout_work.work); + struct sock *sk = &pi->bt.sk; BT_DBG("sock %p state %d", sk, sk->sk_state); lock_sock(sk); - sk->sk_err = ETIMEDOUT; - sk->sk_state_change(sk); + if (!sock_flag(sk, SOCK_ZAPPED)) { + sk->sk_err = ETIMEDOUT; + sk->sk_state_change(sk); + } release_sock(sk); - sock_put(sk); } static void iso_sock_set_timer(struct sock *sk, long timeout) { + lockdep_assert(lockdep_sock_is_held(sk)); + + cancel_delayed_work(&iso_pi(sk)->timeout_work); + if (!iso_pi(sk)->conn) return; BT_DBG("sock %p state %d timeout %ld", sk, sk->sk_state, timeout); - cancel_delayed_work(&iso_pi(sk)->conn->timeout_work); - schedule_delayed_work(&iso_pi(sk)->conn->timeout_work, timeout); + schedule_delayed_work(&iso_pi(sk)->timeout_work, timeout); } static void iso_sock_clear_timer(struct sock *sk) { - if (!iso_pi(sk)->conn) - return; + BT_DBG("sock %p state %d", sk, sk->sk_state); + cancel_delayed_work(&iso_pi(sk)->timeout_work); +} + +static void iso_sock_disable_timer(struct sock *sk) +{ + lockdep_assert(!lockdep_sock_is_held(sk)); BT_DBG("sock %p state %d", sk, sk->sk_state); - cancel_delayed_work(&iso_pi(sk)->conn->timeout_work); + disable_delayed_work_sync(&iso_pi(sk)->timeout_work); } /* ---- ISO connections ---- */ @@ -224,7 +217,6 @@ static struct iso_conn *iso_conn_add(struct hci_conn *hcon) kref_init(&conn->ref); spin_lock_init(&conn->lock); - INIT_DELAYED_WORK(&conn->timeout_work, iso_sock_timeout); hcon->iso_data = conn; conn->hcon = hcon; @@ -300,6 +292,8 @@ static void iso_conn_del(struct hci_conn *hcon, int err) return; } + iso_sock_disable_timer(sk); + lock_sock(sk); /* While a PA sync hcon is in the process of closing, @@ -319,7 +313,6 @@ static void iso_conn_del(struct hci_conn *hcon, int err) } } - iso_sock_clear_timer(sk); iso_chan_del(sk, err); release_sock(sk); iso_sock_kill(sk); @@ -800,6 +793,8 @@ static void iso_sock_cleanup_listen(struct sock *parent) */ static void iso_sock_kill(struct sock *sk) { + iso_sock_disable_timer(sk); + lock_sock(sk); if (!sock_flag(sk, SOCK_ZAPPED) || sk->sk_socket || @@ -895,8 +890,9 @@ static void __iso_sock_close(struct sock *sk) /* Must be called on unlocked socket. */ static void iso_sock_close(struct sock *sk) { + iso_sock_disable_timer(sk); + lock_sock(sk); - iso_sock_clear_timer(sk); __iso_sock_close(sk); release_sock(sk); } @@ -964,6 +960,8 @@ static struct sock *iso_sock_alloc(struct net *net, struct socket *sock, iso_pi(sk)->qos = default_qos; + INIT_DELAYED_WORK(&iso_pi(sk)->timeout_work, iso_sock_timeout); + bt_sock_link(&iso_sk_list, sk); return sk; } -- 2.34.1

2 3

[PATCH OLK-6.6 0/3] CVE-2026-74653
by Cai Xinchen 01 Sep '26

01 Sep '26

PATCH 1: Add serial8250_handle_irq_locked for PATCH 3 PATCH 2: The bugfix for PATCH 1 PATCH 3: CVE-2026-74653 Ilpo Järvinen (1): serial: 8250: Add serial8250_handle_irq_locked() Jacques Nilo (1): serial: 8250: dispatch SysRq character in serial8250_handle_irq() Ryan Wilbur (1): serial: 8250_of: clear stuck empty-FIFO RX-timeout on LPC32xx drivers/tty/serial/8250/8250_of.c | 44 +++++++++++++++++++++++++++++ drivers/tty/serial/8250/8250_port.c | 28 +++++++++++++----- include/linux/serial_8250.h | 1 + 3 files changed, 66 insertions(+), 7 deletions(-) -- 2.18.0.huawei.25

2 4

[PATCH OLK-6.6 0/3] *** fix CVE-2026-72454 ***
by Lin Ruifeng 01 Sep '26

01 Sep '26

*** fix CVE-2026-72454 *** Adrian Hunter (3): i3c: mipi-i3c-hci: Consolidate spinlocks i3c: mipi-i3c-hci: Fix race between DMA ring dequeue and interrupt handler i3c: mipi-i3c-hci: Fix race in i3c_hci_addr_to_dev() drivers/i3c/master/mipi-i3c-hci/core.c | 41 ++++++++++++++++++++++++-- drivers/i3c/master/mipi-i3c-hci/dma.c | 24 +++++++-------- drivers/i3c/master/mipi-i3c-hci/hci.h | 2 ++ drivers/i3c/master/mipi-i3c-hci/ibi.h | 13 +------- drivers/i3c/master/mipi-i3c-hci/pio.c | 23 +++++++-------- 5 files changed, 64 insertions(+), 39 deletions(-) -- 2.34.1

2 4

[PATCH OLK-6.6] entry: Fix seccomp bypass after ptrace with TSYNC
by Jinjie Ruan 01 Sep '26

01 Sep '26

mainline inclusion from mainline-v7.3-rc1 commit 4a3591287fb7f808e209b4974ed337f609a2006b category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/18562 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- Sashiko review pointed out the following issue. If a thread is stopped in syscall_trace_enter() for ptrace, another thread can install a seccomp filter with SECCOMP_FILTER_FLAG_TSYNC (e.g., via seccomp_attach_filter()). This will successfully set SYSCALL_WORK_SECCOMP on the stopped thread, but syscall_trace_enter() evaluates a cached 'work' variable sampled on entry. Consequently, the subsequent check for SYSCALL_WORK_SECCOMP misses the newly assigned flag, and the filter is silently bypassed. This race condition could allow an unprivileged process to execute a prohibited system call (e.g., execve) that the newly installed filter was intended to block, especially since the tracer might have modified the system call number during the ptrace stop. Fix this by re-reading the syscall_work flags after ptrace handling, so that any new SYSCALL_WORK_SECCOMP flag set by another thread via TSYNC during the ptrace stop is observed before the subsequent seccomp check. Fixes: 142781e108b1 ("entry: Provide generic syscall entry functionality") Signed-off-by: Jinjie Ruan <ruanjinjie(a)huawei.com> Signed-off-by: Thomas Gleixner <tglx(a)kernel.org> Cc: stable(a)vger.kernel.org Link: https://lore.kernel.org/all/20260629132914.1135C1F000E9@smtp.kernel.org/ Link: https://patch.msgid.link/20260713025712.416366-1-ruanjinjie@huawei.com Conflicts: include/linux/entry-common.h [Context and rework conflict. ] Signed-off-by: Jinjie Ruan <ruanjinjie(a)huawei.com> --- kernel/entry/common.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/kernel/entry/common.c b/kernel/entry/common.c index 90843cc38588..fb61ca3ab93c 100644 --- a/kernel/entry/common.c +++ b/kernel/entry/common.c @@ -45,6 +45,9 @@ long syscall_trace_enter(struct pt_regs *regs, long syscall, ret = ptrace_report_syscall_entry(regs); if (ret || (work & SYSCALL_WORK_SYSCALL_EMU)) return -1L; + + /* ptrace might have changed work flags */ + work = READ_ONCE(current_thread_info()->syscall_work); } /* Do seccomp after ptrace, to catch any tracer changes. */ -- 2.34.1

2 1

[PATCH OLK-5.10] [Backport] driver core: use READ_ONCE() for dev->driver in dev_has_sync_state()
by Lin Ruifeng 01 Sep '26

01 Sep '26

From: Danilo Krummrich <dakr(a)kernel.org> stable inclusion from stable-v5.10.261 commit 89789e4c141904506163dcb91c7289a074573931 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/18448 CVE: CVE-2026-80677 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit e9506871a8ea304cde48ff4a57226df2aadddae3 ] dev_has_sync_state() reads dev->driver twice without holding device_lock() -- once for the NULL check and once to dereference ->sync_state. Some callers only hold device_links_write_lock, which doesn't prevent a concurrent unbind from clearing dev->driver via device_unbind_cleanup(). Fix it by reading dev->driver exactly once with READ_ONCE(), pairing with the WRITE_ONCE() in device_set_driver(). Link: https://lore.kernel.org/driver-core/DHW8QPU1VU1F.3P6PH69HLFBYC@kernel.org/ Fixes: ac338acf514e ("driver core: Add dev_has_sync_state()") Reviewed-by: Rafael J. Wysocki (Intel) <rafael(a)kernel.org> Acked-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Reviewed-by: Saravana Kannan <saravanak(a)kernel.org> Link: https://patch.msgid.link/20260418162221.1121873-1-dakr@kernel.org Signed-off-by: Danilo Krummrich <dakr(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Lin Ruifeng <linruifeng4(a)huawei.com> --- include/linux/device.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/include/linux/device.h b/include/linux/device.h index 65dac0c81e47..863d943715cc 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -894,9 +894,12 @@ static inline struct device_node *dev_of_node(struct device *dev) static inline bool dev_has_sync_state(struct device *dev) { + struct device_driver *drv; + if (!dev) return false; - if (dev->driver && dev->driver->sync_state) + drv = READ_ONCE(dev->driver); + if (drv && drv->sync_state) return true; if (dev->bus && dev->bus->sync_state) return true; -- 2.34.1

2 1

[PATCH OLK-6.6] dmaengine: idxd: fix double free of wq, engine, and group structs
by Lin Ruifeng 01 Sep '26

01 Sep '26

From: Yuho Choi <dbgh9129(a)gmail.com> stable inclusion from stable-v7.1.8 commit c93a9f652b7373ac8ee5bfcc18a9af76959b6c6c category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/18548 CVE: CVE-2026-80698 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit ec2d428b2e32dd157de8f86a86dd85c5b2c8f45c ] The release callbacks for wq, engine, and group devices (idxd_conf_wq_release, idxd_conf_engine_release, idxd_conf_group_release) each call kfree() on the enclosing struct. The setup error paths and cleanup functions also call kfree() explicitly after put_device(), producing a double free whenever put_device() drops the reference count to zero and fires the release. In the setup functions, device_initialize() is called before device_add(), so the reference count is exactly 1 at the error sites. put_device() unconditionally fires the release, which frees the struct; the subsequent explicit kfree() then operates on freed memory. For idxd_setup_wqs(), the wq release callback also owns opcap_bmap and wqcfg. The error unwind additionally freed those fields explicitly before calling put_device(), causing further double frees on both. Remove the redundant explicit kfree() calls from all setup error paths and cleanup functions for wq, engine, and group structs, delegating sole ownership of those allocations to the release callbacks. Fixes: 7c5dd23e57c1 ("dmaengine: idxd: fix wq conf_dev 'struct device' lifetime") Fixes: 75b911309060 ("dmaengine: idxd: fix engine conf_dev lifetime") Fixes: defe49f96012 ("dmaengine: idxd: fix group conf_dev lifetime") Signed-off-by: Yuho Choi <dbgh9129(a)gmail.com> Acked-by: Vinicius Costa Gomes <vinicius.gomes(a)intel.com> Reviewed-by: Frank Li <Frank.Li(a)nxp.com> Link: https://patch.msgid.link/20260415205452.67155-1-dbgh9129@gmail.com Signed-off-by: Vinod Koul <vkoul(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Lin Ruifeng <linruifeng4(a)huawei.com> --- drivers/dma/idxd/init.c | 36 +++++------------------------------- 1 file changed, 5 insertions(+), 31 deletions(-) diff --git a/drivers/dma/idxd/init.c b/drivers/dma/idxd/init.c index 624a47b64bb4..eaa0523a41ed 100644 --- a/drivers/dma/idxd/init.c +++ b/drivers/dma/idxd/init.c @@ -150,18 +150,12 @@ static void idxd_cleanup_interrupts(struct idxd_device *idxd) static void idxd_clean_wqs(struct idxd_device *idxd) { - struct idxd_wq *wq; struct device *conf_dev; int i; for (i = 0; i < idxd->max_wqs; i++) { - wq = idxd->wqs[i]; - if (idxd->hw.wq_cap.op_config) - bitmap_free(wq->opcap_bmap); - kfree(wq->wqcfg); - conf_dev = wq_confdev(wq); + conf_dev = wq_confdev(idxd->wqs[i]); put_device(conf_dev); - kfree(wq); } bitmap_free(idxd->wq_enable_map); kfree(idxd->wqs); @@ -203,7 +197,6 @@ static int idxd_setup_wqs(struct idxd_device *idxd) rc = dev_set_name(conf_dev, "wq%d.%d", idxd->id, wq->id); if (rc < 0) { put_device(conf_dev); - kfree(wq); goto err_unwind; } @@ -217,7 +210,6 @@ static int idxd_setup_wqs(struct idxd_device *idxd) wq->wqcfg = kzalloc_node(idxd->wqcfg_size, GFP_KERNEL, dev_to_node(dev)); if (!wq->wqcfg) { put_device(conf_dev); - kfree(wq); rc = -ENOMEM; goto err_unwind; } @@ -225,9 +217,7 @@ static int idxd_setup_wqs(struct idxd_device *idxd) if (idxd->hw.wq_cap.op_config) { wq->opcap_bmap = bitmap_zalloc(IDXD_MAX_OPCAP_BITS, GFP_KERNEL); if (!wq->opcap_bmap) { - kfree(wq->wqcfg); put_device(conf_dev); - kfree(wq); rc = -ENOMEM; goto err_unwind; } @@ -242,13 +232,8 @@ static int idxd_setup_wqs(struct idxd_device *idxd) err_unwind: while (--i >= 0) { - wq = idxd->wqs[i]; - if (idxd->hw.wq_cap.op_config) - bitmap_free(wq->opcap_bmap); - kfree(wq->wqcfg); - conf_dev = wq_confdev(wq); + conf_dev = wq_confdev(idxd->wqs[i]); put_device(conf_dev); - kfree(wq); } bitmap_free(idxd->wq_enable_map); @@ -260,15 +245,12 @@ static int idxd_setup_wqs(struct idxd_device *idxd) static void idxd_clean_engines(struct idxd_device *idxd) { - struct idxd_engine *engine; struct device *conf_dev; int i; for (i = 0; i < idxd->max_engines; i++) { - engine = idxd->engines[i]; - conf_dev = engine_confdev(engine); + conf_dev = engine_confdev(idxd->engines[i]); put_device(conf_dev); - kfree(engine); } kfree(idxd->engines); } @@ -303,7 +285,6 @@ static int idxd_setup_engines(struct idxd_device *idxd) rc = dev_set_name(conf_dev, "engine%d.%d", idxd->id, engine->id); if (rc < 0) { put_device(conf_dev); - kfree(engine); goto err; } @@ -314,10 +295,8 @@ static int idxd_setup_engines(struct idxd_device *idxd) err: while (--i >= 0) { - engine = idxd->engines[i]; - conf_dev = engine_confdev(engine); + conf_dev = engine_confdev(idxd->engines[i]); put_device(conf_dev); - kfree(engine); } kfree(idxd->engines); @@ -326,13 +305,10 @@ static int idxd_setup_engines(struct idxd_device *idxd) static void idxd_clean_groups(struct idxd_device *idxd) { - struct idxd_group *group; int i; for (i = 0; i < idxd->max_groups; i++) { - group = idxd->groups[i]; - put_device(group_confdev(group)); - kfree(group); + put_device(group_confdev(idxd->groups[i])); } kfree(idxd->groups); } @@ -367,7 +343,6 @@ static int idxd_setup_groups(struct idxd_device *idxd) rc = dev_set_name(conf_dev, "group%d.%d", idxd->id, group->id); if (rc < 0) { put_device(conf_dev); - kfree(group); goto err; } @@ -392,7 +367,6 @@ static int idxd_setup_groups(struct idxd_device *idxd) while (--i >= 0) { group = idxd->groups[i]; put_device(group_confdev(group)); - kfree(group); } kfree(idxd->groups); -- 2.34.1

2 1

[PATCH OLK-5.10] entry: Fix seccomp bypass after ptrace with TSYNC
by Jinjie Ruan 01 Sep '26

01 Sep '26

mainline inclusion from mainline-v7.3-rc1 commit 4a3591287fb7f808e209b4974ed337f609a2006b category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/18562 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- Sashiko review pointed out the following issue. If a thread is stopped in syscall_trace_enter() for ptrace, another thread can install a seccomp filter with SECCOMP_FILTER_FLAG_TSYNC (e.g., via seccomp_attach_filter()). This will successfully set SYSCALL_WORK_SECCOMP on the stopped thread, but syscall_trace_enter() evaluates a cached 'work' variable sampled on entry. Consequently, the subsequent check for SYSCALL_WORK_SECCOMP misses the newly assigned flag, and the filter is silently bypassed. This race condition could allow an unprivileged process to execute a prohibited system call (e.g., execve) that the newly installed filter was intended to block, especially since the tracer might have modified the system call number during the ptrace stop. Fix this by re-reading the syscall_work flags after ptrace handling, so that any new SYSCALL_WORK_SECCOMP flag set by another thread via TSYNC during the ptrace stop is observed before the subsequent seccomp check. Fixes: 142781e108b1 ("entry: Provide generic syscall entry functionality") Signed-off-by: Jinjie Ruan <ruanjinjie(a)huawei.com> Signed-off-by: Thomas Gleixner <tglx(a)kernel.org> Cc: stable(a)vger.kernel.org Link: https://lore.kernel.org/all/20260629132914.1135C1F000E9@smtp.kernel.org/ Link: https://patch.msgid.link/20260713025712.416366-1-ruanjinjie@huawei.com Conflicts: include/linux/entry-common.h [Context and rework conflict. ] Signed-off-by: Jinjie Ruan <ruanjinjie(a)huawei.com> --- kernel/entry/common.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/kernel/entry/common.c b/kernel/entry/common.c index b3690d7ea3cb..5706a8f2e90c 100644 --- a/kernel/entry/common.c +++ b/kernel/entry/common.c @@ -51,6 +51,9 @@ static long syscall_trace_enter(struct pt_regs *regs, long syscall, ret = arch_syscall_enter_tracehook(regs); if (ret || (ti_work & _TIF_SYSCALL_EMU)) return -1L; + + /* ptrace might have changed work flags */ + ti_work = READ_ONCE(current_thread_info()->flags); } /* Do seccomp after ptrace, to catch any tracer changes. */ -- 2.34.1

2 1

[PATCH OLK-6.6] KVM: arm64: vgic-v3: Restrict ICH_AP1Rn_EL2 accesses to 64-bit only with vNMI
by Jinqian Yang 01 Sep '26

01 Sep '26

ICH_AP1Rn_EL2 is 64-bit only when vNMI is in use, with the NMI priority in ICH_AP1R0_EL2[63:32]; otherwise the upper bits are RES0. Gate the save/restore access width on vgic_v3_vcpu_has_vnmi() which checks both vgic.has_nmi and ID_AA64PFR1_EL1.NMI, so that the NMI priority is preserved across context switch only when vNMI is actually enabled for the guest. Also drop the early assignment of kvm->arch.pfr1_nmi in kvm_arch_init_vm(). Without it vNMI is no longer enabled by default at VM creation on hosts with FEAT_NMI; instead the VMM must explicitly set ID_AA64PFR1_EL1.NMI via SET_ONE_REG to enable vNMI for the guest. Signed-off-by: Jinqian Yang <yangjinqian1(a)huawei.com> --- arch/arm64/include/asm/kvm_hyp.h | 1 + arch/arm64/kvm/arm.c | 3 -- arch/arm64/kvm/hyp/vgic-v3-sr.c | 55 +++++++++++++++++++++++++++----- 3 files changed, 48 insertions(+), 11 deletions(-) diff --git a/arch/arm64/include/asm/kvm_hyp.h b/arch/arm64/include/asm/kvm_hyp.h index 127ad280e68b..c4121bf0aadc 100644 --- a/arch/arm64/include/asm/kvm_hyp.h +++ b/arch/arm64/include/asm/kvm_hyp.h @@ -82,6 +82,7 @@ void __vgic_v3_activate_traps(struct vgic_v3_cpu_if *cpu_if); void __vgic_v3_deactivate_traps(struct vgic_v3_cpu_if *cpu_if); void __vgic_v3_save_aprs(struct vgic_v3_cpu_if *cpu_if); void __vgic_v3_restore_aprs(struct vgic_v3_cpu_if *cpu_if); +bool vgic_v3_vcpu_has_vnmi(struct kvm_vcpu *vcpu); int __vgic_v3_perform_cpuif_access(struct kvm_vcpu *vcpu); #ifdef __KVM_NVHE_HYPERVISOR__ diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c index 2b560eba3e0b..9a4be08730da 100644 --- a/arch/arm64/kvm/arm.c +++ b/arch/arm64/kvm/arm.c @@ -381,9 +381,6 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) /* The maximum number of VCPUs is limited by the host's GIC model */ kvm->max_vcpus = kvm_arm_default_max_vcpus(); - if (cpus_have_const_cap(ARM64_HAS_NMI) && !static_branch_unlikely(&vgic_v3_cpuif_trap)) - kvm->arch.pfr1_nmi = ID_AA64PFR1_EL1_NMI_IMP; - kvm_arm_init_hypercalls(kvm); if (!kvm_is_realm(kvm)) diff --git a/arch/arm64/kvm/hyp/vgic-v3-sr.c b/arch/arm64/kvm/hyp/vgic-v3-sr.c index 7a9e132f68db..ab406332762e 100644 --- a/arch/arm64/kvm/hyp/vgic-v3-sr.c +++ b/arch/arm64/kvm/hyp/vgic-v3-sr.c @@ -335,10 +335,40 @@ void __vgic_v3_deactivate_traps(struct vgic_v3_cpu_if *cpu_if) write_gicreg(0, ICH_HCR_EL2); } +static struct kvm_vcpu *vgic_v3_cpu_if_to_vcpu(struct vgic_v3_cpu_if *cpu_if) +{ + struct vgic_cpu *vgic_cpu = container_of(cpu_if, struct vgic_cpu, vgic_v3); + + return container_of(vgic_cpu, struct kvm_vcpu, arch.vgic_cpu); +} + +/* + * vNMI is only usable if the guest has selected NMI support + * (vgic.has_nmi) and the vcpu has FEAT_NMI (pfr1_nmi). + * + * This drives the width of the ICH_AP1Rn_EL2 accesses: with vNMI the + * registers are 64bit (the NMI priority lives in ICH_AP1R0_EL2[63]), + * without it only the low 32 bits are valid. + */ +bool vgic_v3_vcpu_has_vnmi(struct kvm_vcpu *vcpu) +{ + struct kvm *kvm = kern_hyp_va(vcpu->kvm); + + return kvm->arch.vgic.has_nmi && + kvm->arch.pfr1_nmi == ID_AA64PFR1_EL1_NMI_IMP; +} + void __vgic_v3_save_aprs(struct vgic_v3_cpu_if *cpu_if) { u64 val; u32 nr_pre_bits; + /* + * ICH_AP1Rn_EL2 are only 64bit when the vcpu has vNMI: the NMI + * priority is encoded in ICH_AP1R0_EL2[63]. Without vNMI, the + * upper bits are RES0 and only the low 32 bits must be + * preserved. + */ + bool vnmi = vgic_v3_vcpu_has_vnmi(vgic_v3_cpu_if_to_vcpu(cpu_if)); val = read_gicreg(ICH_VTR_EL2); nr_pre_bits = vtr_to_nr_pre_bits(val); @@ -357,14 +387,18 @@ void __vgic_v3_save_aprs(struct vgic_v3_cpu_if *cpu_if) switch (nr_pre_bits) { case 7: - cpu_if->vgic_ap1r[3] = __vgic_v3_read_ap1rn(3); - cpu_if->vgic_ap1r[2] = __vgic_v3_read_ap1rn(2); + cpu_if->vgic_ap1r[3] = vnmi ? __vgic_v3_read_ap1rn(3) + : (u32)__vgic_v3_read_ap1rn(3); + cpu_if->vgic_ap1r[2] = vnmi ? __vgic_v3_read_ap1rn(2) + : (u32)__vgic_v3_read_ap1rn(2); fallthrough; case 6: - cpu_if->vgic_ap1r[1] = __vgic_v3_read_ap1rn(1); + cpu_if->vgic_ap1r[1] = vnmi ? __vgic_v3_read_ap1rn(1) + : (u32)__vgic_v3_read_ap1rn(1); fallthrough; default: - cpu_if->vgic_ap1r[0] = __vgic_v3_read_ap1rn(0); + cpu_if->vgic_ap1r[0] = vnmi ? __vgic_v3_read_ap1rn(0) + : (u32)__vgic_v3_read_ap1rn(0); } } @@ -372,6 +406,7 @@ void __vgic_v3_restore_aprs(struct vgic_v3_cpu_if *cpu_if) { u64 val; u32 nr_pre_bits; + bool vnmi = vgic_v3_vcpu_has_vnmi(vgic_v3_cpu_if_to_vcpu(cpu_if)); val = read_gicreg(ICH_VTR_EL2); nr_pre_bits = vtr_to_nr_pre_bits(val); @@ -390,14 +425,18 @@ void __vgic_v3_restore_aprs(struct vgic_v3_cpu_if *cpu_if) switch (nr_pre_bits) { case 7: - __vgic_v3_write_ap1rn(cpu_if->vgic_ap1r[3], 3); - __vgic_v3_write_ap1rn(cpu_if->vgic_ap1r[2], 2); + __vgic_v3_write_ap1rn(vnmi ? cpu_if->vgic_ap1r[3] + : (u32)cpu_if->vgic_ap1r[3], 3); + __vgic_v3_write_ap1rn(vnmi ? cpu_if->vgic_ap1r[2] + : (u32)cpu_if->vgic_ap1r[2], 2); fallthrough; case 6: - __vgic_v3_write_ap1rn(cpu_if->vgic_ap1r[1], 1); + __vgic_v3_write_ap1rn(vnmi ? cpu_if->vgic_ap1r[1] + : (u32)cpu_if->vgic_ap1r[1], 1); fallthrough; default: - __vgic_v3_write_ap1rn(cpu_if->vgic_ap1r[0], 0); + __vgic_v3_write_ap1rn(vnmi ? cpu_if->vgic_ap1r[0] + : (u32)cpu_if->vgic_ap1r[0], 0); } } -- 2.33.0

2 1

[PATCH OLK-6.6] vhost_iotlb: bound map allocation in add_range
by Lin Ruifeng 01 Sep '26

01 Sep '26

From: Linfeng Sun <linfeng.sun.dev(a)gamil.com> stable inclusion from stable-v7.1.9 commit ae128dd19040ee06a4f8143c7ced4d18080d7a9a category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/18250 CVE: CVE-2026-74713 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 1ed35ac7f3fe2b4396bdd29ac3a7f0ebc0829e94 ] vhost_iotlb_add_range_ctx() only retires an old entry when the table has a non-zero limit, has exactly reached that limit and has VHOST_IOTLB_FLAG_RETIRE set. Non-retiring tables can keep allocating entries after reaching their configured limit. Existing vhost devices allocate their IOTLB with max_iotlb_entries from vhost.c, which defaults to 2048 and is tunable by module parameter. Use the caller-provided limit at the allocation point instead of adding a separate default in the common IOTLB helper, and reject non-positive values in vhost paths that can report an error. Other vhost IOTLB users should not create zero-limit tables when entries can be populated from userspace or guest-controlled requests. Add caller-side max_iotlb_entries parameters for mlx5 vDPA, VDUSE and vhost-vDPA. Reject non-positive VDUSE and vhost-vDPA values, and require at least two entries for vdpa_sim and mlx5 vDPA paths that install full-range mappings, since those mappings are split into two IOTLB entries. Handle full-range mappings in the common helper by checking that the IOTLB can hold both split entries before inserting the first half. This avoids returning an error after leaving a half mapping behind. When the table is full, keep the existing retire behavior for retiring tables and return -ENOSPC for non-retiring tables. Reuse the retired map node instead of freeing it and allocating a replacement, so a stream of IOTLB updates cannot keep forcing GFP_ATOMIC allocations after the table has reached its limit. If a zero-limit IOTLB still reaches the common helper, treat it as a configuration error and return -EINVAL. I found this bug myself, though the patch was written with AI assistance. Fixes: 0bbe30668d89 ("vhost: factor out IOTLB") Assisted-by: OpenAI-Codex:GPT-5 Signed-off-by: Linfeng Sun <linfeng.sun.dev(a)gamil.com> Message-ID: <AMYAtgAiKmgYcSQT5ukl-4qq.3.1781960405943.Hmail.241270009(a)hdu.edu.cn> Signed-off-by: Michael S. Tsirkin <mst(a)redhat.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Conflicts: drivers/vdpa/mlx5/core/mr.c drivers/vdpa/vdpa_sim/vdpa_sim.c drivers/vdpa/vdpa_user/iova_domain.c drivers/vhost/iotlb.c drivers/vhost/vdpa.c drivers/vhost/vhost.c [Context Conflicts] Signed-off-by: Lin Ruifeng <linruifeng4(a)huawei.com> --- drivers/vdpa/mlx5/core/mlx5_vdpa.h | 2 ++ drivers/vdpa/mlx5/core/mr.c | 3 ++ drivers/vdpa/mlx5/core/resources.c | 11 ++++++- drivers/vdpa/vdpa_sim/vdpa_sim.c | 4 ++- drivers/vdpa/vdpa_user/iova_domain.c | 11 ++++++- drivers/vhost/iotlb.c | 47 +++++++++++++++++++--------- drivers/vhost/vdpa.c | 11 +++++-- drivers/vhost/vhost.c | 8 +++++ 8 files changed, 78 insertions(+), 19 deletions(-) diff --git a/drivers/vdpa/mlx5/core/mlx5_vdpa.h b/drivers/vdpa/mlx5/core/mlx5_vdpa.h index ca56242972b3..869219529ff1 100644 --- a/drivers/vdpa/mlx5/core/mlx5_vdpa.h +++ b/drivers/vdpa/mlx5/core/mlx5_vdpa.h @@ -11,6 +11,8 @@ #define MLX5V_ETH_HARD_MTU (ETH_HLEN + VLAN_HLEN + ETH_FCS_LEN) +extern int mlx5_vdpa_max_iotlb_entries; + struct mlx5_vdpa_direct_mr { u64 start; u64 end; diff --git a/drivers/vdpa/mlx5/core/mr.c b/drivers/vdpa/mlx5/core/mr.c index 165730656934..b7c1d532ccbc 100644 --- a/drivers/vdpa/mlx5/core/mr.c +++ b/drivers/vdpa/mlx5/core/mr.c @@ -577,6 +577,9 @@ static int _mlx5_vdpa_create_mr(struct mlx5_vdpa_dev *mvdev, { int err; + if (mlx5_vdpa_max_iotlb_entries < 2) + return -EINVAL; + err = _mlx5_vdpa_create_dvq_mr(mvdev, iotlb, asid); if (err) return err; diff --git a/drivers/vdpa/mlx5/core/resources.c b/drivers/vdpa/mlx5/core/resources.c index d5a59c9035fb..00b66438ab53 100644 --- a/drivers/vdpa/mlx5/core/resources.c +++ b/drivers/vdpa/mlx5/core/resources.c @@ -3,8 +3,14 @@ #include <linux/iova.h> #include <linux/mlx5/driver.h> +#include <linux/moduleparam.h> #include "mlx5_vdpa.h" +int mlx5_vdpa_max_iotlb_entries = 2048; +module_param_named(max_iotlb_entries, mlx5_vdpa_max_iotlb_entries, int, 0444); +MODULE_PARM_DESC(max_iotlb_entries, + "Maximum number of iotlb entries. (default: 2048)"); + static int alloc_pd(struct mlx5_vdpa_dev *dev, u32 *pdn, u16 uid) { struct mlx5_core_dev *mdev = dev->mdev; @@ -229,7 +235,10 @@ int mlx5_vdpa_destroy_mkey(struct mlx5_vdpa_dev *mvdev, u32 mkey) static int init_ctrl_vq(struct mlx5_vdpa_dev *mvdev) { - mvdev->cvq.iotlb = vhost_iotlb_alloc(0, 0); + if (mlx5_vdpa_max_iotlb_entries < 2) + return -EINVAL; + + mvdev->cvq.iotlb = vhost_iotlb_alloc(mlx5_vdpa_max_iotlb_entries, 0); if (!mvdev->cvq.iotlb) return -ENOMEM; diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.c b/drivers/vdpa/vdpa_sim/vdpa_sim.c index af59d90ba9a8..36b7c12e2a73 100644 --- a/drivers/vdpa/vdpa_sim/vdpa_sim.c +++ b/drivers/vdpa/vdpa_sim/vdpa_sim.c @@ -34,7 +34,7 @@ MODULE_PARM_DESC(batch_mapping, "Batched mapping 1 -Enable; 0 - Disable"); static int max_iotlb_entries = 2048; module_param(max_iotlb_entries, int, 0444); MODULE_PARM_DESC(max_iotlb_entries, - "Maximum number of iotlb entries for each address space. 0 means unlimited. (default: 2048)"); + "Maximum number of iotlb entries for each address space. (default: 2048)"); static bool use_va = true; module_param(use_va, bool, 0444); @@ -199,6 +199,8 @@ struct vdpasim *vdpasim_create(struct vdpasim_dev_attr *dev_attr, if (!dev_attr->alloc_size) return ERR_PTR(-EINVAL); + if (max_iotlb_entries < 2) + return ERR_PTR(-EINVAL); if (config->mask & BIT_ULL(VDPA_ATTR_DEV_FEATURES)) { if (config->device_features & diff --git a/drivers/vdpa/vdpa_user/iova_domain.c b/drivers/vdpa/vdpa_user/iova_domain.c index cc8d26b97187..0a7791803132 100644 --- a/drivers/vdpa/vdpa_user/iova_domain.c +++ b/drivers/vdpa/vdpa_user/iova_domain.c @@ -12,11 +12,17 @@ #include <linux/file.h> #include <linux/anon_inodes.h> #include <linux/highmem.h> +#include <linux/moduleparam.h> #include <linux/vmalloc.h> #include <linux/vdpa.h> #include "iova_domain.h" +static int max_iotlb_entries = 2048; +module_param(max_iotlb_entries, int, 0444); +MODULE_PARM_DESC(max_iotlb_entries, + "Maximum number of iotlb entries. (default: 2048)"); + static int vduse_iotlb_add_range(struct vduse_iova_domain *domain, u64 start, u64 last, u64 addr, unsigned int perm, @@ -607,12 +613,15 @@ vduse_domain_create(unsigned long iova_limit, size_t bounce_size) bounce_pfns = PAGE_ALIGN(bounce_size) >> BOUNCE_MAP_SHIFT; if (iova_limit <= bounce_size) return NULL; + + if (max_iotlb_entries <= 0) + return NULL; domain = kzalloc(sizeof(*domain), GFP_KERNEL); if (!domain) return NULL; - domain->iotlb = vhost_iotlb_alloc(0, 0); + domain->iotlb = vhost_iotlb_alloc(max_iotlb_entries, 0); if (!domain->iotlb) goto err_iotlb; diff --git a/drivers/vhost/iotlb.c b/drivers/vhost/iotlb.c index ea61330a3431..a6228dd24931 100644 --- a/drivers/vhost/iotlb.c +++ b/drivers/vhost/iotlb.c @@ -20,6 +20,14 @@ INTERVAL_TREE_DEFINE(struct vhost_iotlb_map, rb, __u64, __subtree_last, START, LAST, static inline, vhost_iotlb_itree); +static void vhost_iotlb_map_unlink(struct vhost_iotlb *iotlb, + struct vhost_iotlb_map *map) +{ + vhost_iotlb_itree_remove(map, &iotlb->root); + list_del(&map->link); + iotlb->nmaps--; +} + /** * vhost_iotlb_map_free - remove a map node and free it * @iotlb: the IOTLB @@ -28,10 +36,8 @@ INTERVAL_TREE_DEFINE(struct vhost_iotlb_map, void vhost_iotlb_map_free(struct vhost_iotlb *iotlb, struct vhost_iotlb_map *map) { - vhost_iotlb_itree_remove(map, &iotlb->root); - list_del(&map->link); + vhost_iotlb_map_unlink(iotlb, map); kfree(map); - iotlb->nmaps--; } EXPORT_SYMBOL_GPL(vhost_iotlb_map_free); @@ -57,14 +63,25 @@ int vhost_iotlb_add_range_ctx(struct vhost_iotlb *iotlb, if (last < start) return -EFAULT; + if (!iotlb->limit) + return -EINVAL; + /* If the range being mapped is [0, ULONG_MAX], split it into two entries * otherwise its size would overflow u64. */ if (start == 0 && last == ULONG_MAX) { u64 mid = last / 2; - int err = vhost_iotlb_add_range_ctx(iotlb, start, mid, addr, - perm, opaque); + int err; + + if (iotlb->limit < 2) + return -ENOSPC; + if (!(iotlb->flags & VHOST_IOTLB_FLAG_RETIRE) && + iotlb->nmaps > iotlb->limit - 2) + return -ENOSPC; + + err = vhost_iotlb_add_range_ctx(iotlb, start, mid, addr, + perm, opaque); if (err) return err; @@ -72,17 +89,19 @@ int vhost_iotlb_add_range_ctx(struct vhost_iotlb *iotlb, start = mid + 1; } - if (iotlb->limit && - iotlb->nmaps == iotlb->limit && - iotlb->flags & VHOST_IOTLB_FLAG_RETIRE) { - map = list_first_entry(&iotlb->list, typeof(*map), link); - vhost_iotlb_map_free(iotlb, map); + if (iotlb->nmaps >= iotlb->limit) { + if (iotlb->flags & VHOST_IOTLB_FLAG_RETIRE) { + map = list_first_entry(&iotlb->list, typeof(*map), link); + vhost_iotlb_map_unlink(iotlb, map); + } else { + return -ENOSPC; + } + } else { + map = kmalloc(sizeof(*map), GFP_ATOMIC); + if (!map) + return -ENOMEM; } - map = kmalloc(sizeof(*map), GFP_ATOMIC); - if (!map) - return -ENOMEM; - map->start = start; map->size = last - start + 1; map->last = last; diff --git a/drivers/vhost/vdpa.c b/drivers/vhost/vdpa.c index 0583374602f8..72c71b2fb198 100644 --- a/drivers/vhost/vdpa.c +++ b/drivers/vhost/vdpa.c @@ -37,6 +37,11 @@ enum { #define VHOST_VDPA_DEV_MAX (1U << MINORBITS) +static int max_iotlb_entries = 2048; +module_param(max_iotlb_entries, int, 0444); +MODULE_PARM_DESC(max_iotlb_entries, + "Maximum number of iotlb entries. (default: 2048)"); + #define VHOST_VDPA_IOTLB_BUCKETS 16 struct vhost_vdpa_as { @@ -114,12 +119,14 @@ static struct vhost_vdpa_as *vhost_vdpa_alloc_as(struct vhost_vdpa *v, u32 asid) if (asid >= v->vdpa->nas) return NULL; + if (max_iotlb_entries <= 0) + return NULL; as = kmalloc(sizeof(*as), GFP_KERNEL); if (!as) return NULL; - vhost_iotlb_init(&as->iotlb, 0, 0); + vhost_iotlb_init(&as->iotlb, max_iotlb_entries, 0); as->id = asid; hlist_add_head(&as->hash_link, head); @@ -1833,7 +1840,7 @@ static int vhost_vdpa_probe(struct vdpa_device *vdpa) goto err; } - vhost_iotlb_init(&v->resv_iotlb, 0, 0); + vhost_iotlb_init(&v->resv_iotlb, max_iotlb_entries, 0); r = dev_set_name(&v->dev, "vhost-vdpa-%u", minor); if (r) diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index a58be2369fdc..2b410ddffac2 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -977,6 +977,9 @@ EXPORT_SYMBOL_GPL(vhost_dev_set_owner); static struct vhost_iotlb *iotlb_alloc(void) { + if (max_iotlb_entries <= 0) + return NULL; + return vhost_iotlb_alloc(max_iotlb_entries, VHOST_IOTLB_FLAG_RETIRE); } @@ -1817,6 +1820,8 @@ static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m) return -EOPNOTSUPP; if (mem.nregions > max_mem_regions) return -E2BIG; + if (max_iotlb_entries <= 0) + return -EINVAL; newmem = kvzalloc(struct_size(newmem, regions, mem.nregions), GFP_KERNEL); if (!newmem) @@ -2111,6 +2116,9 @@ int vhost_init_device_iotlb(struct vhost_dev *d) struct vhost_iotlb *niotlb, *oiotlb; int i; + if (max_iotlb_entries <= 0) + return -EINVAL; + niotlb = iotlb_alloc(); if (!niotlb) return -ENOMEM; -- 2.34.1

2 1

[PATCH OLK-6.6] bpf: Simplify sanitize_err() signature
by Pu Lehui 01 Sep '26

01 Sep '26

From: Eduard Zingerman <eddyz87(a)gmail.com> mainline inclusion from mainline-v7.2-rc7 commit a15970d916b39acc7c60a0a99c27a6e378690aa9 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/18233 CVE: CVE-2026-74720 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- The sanitize_err() function is called when: - ptr += scalar - scalar += ptr - scalar += scalar ALU operations are processed. This commit drops offset and pointer registers parameters from its signature to simplify the follow-up changes for 'scalar += ptr' case. regs[src].type is safe to access, as it is not mutated by the callers. Signed-off-by: Yiyang Chen <chenyy23(a)mails.tsinghua.edu.cn> Acked-by: Shung-Hsi Yu <shung-hsi.yu(a)suse.com> Link: https://patch.msgid.link/20260729-c3-035-public-bpf-v4-v4-1-8ee297e2346b@ma… Signed-off-by: Eduard Zingerman <eddyz87(a)gmail.com> Conflicts: kernel/bpf/verifier.c [ctx conflicts] Signed-off-by: Pu Lehui <pulehui(a)huawei.com> --- kernel/bpf/verifier.c | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index ee8c00981cea..b59ee8344873 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -12725,23 +12725,21 @@ static void sanitize_mark_insn_seen(struct bpf_verifier_env *env) env->insn_aux_data[env->insn_idx].seen = env->pass_cnt; } -static int sanitize_err(struct bpf_verifier_env *env, - const struct bpf_insn *insn, int reason, - const struct bpf_reg_state *off_reg, - const struct bpf_reg_state *dst_reg) +static int sanitize_err(struct bpf_verifier_env *env, const struct bpf_insn *insn, int reason) { static const char *err = "pointer arithmetic with it prohibited for !root"; const char *op = BPF_OP(insn->code) == BPF_ADD ? "add" : "sub"; u32 dst = insn->dst_reg, src = insn->src_reg; + struct bpf_reg_state *regs = cur_regs(env); switch (reason) { case REASON_BOUNDS: verbose(env, "R%d has unknown scalar with mixed signed bounds, %s\n", - off_reg == dst_reg ? dst : src, err); + regs[src].type == SCALAR_VALUE ? src : dst, err); break; case REASON_TYPE: verbose(env, "R%d has pointer with unsupported alu operation, %s\n", - off_reg == dst_reg ? src : dst, err); + regs[src].type == SCALAR_VALUE ? dst : src, err); break; case REASON_PATHS: verbose(env, "R%d tried to %s from different maps, paths or scalars, %s\n", @@ -12923,7 +12921,7 @@ static int adjust_ptr_min_max_vals(struct bpf_verifier_env *env, ret = sanitize_ptr_alu(env, insn, ptr_reg, off_reg, dst_reg, &info, false); if (ret < 0) - return sanitize_err(env, insn, ret, off_reg, dst_reg); + return sanitize_err(env, insn, ret); } switch (opcode) { @@ -13087,7 +13085,7 @@ static int adjust_ptr_min_max_vals(struct bpf_verifier_env *env, ret = sanitize_ptr_alu(env, insn, dst_reg, off_reg, dst_reg, &info, true); if (ret < 0) - return sanitize_err(env, insn, ret, off_reg, dst_reg); + return sanitize_err(env, insn, ret); } return 0; @@ -13719,7 +13717,7 @@ static int adjust_scalar_min_max_vals(struct bpf_verifier_env *env, if (sanitize_needed(opcode)) { ret = sanitize_val_alu(env, insn); if (ret < 0) - return sanitize_err(env, insn, ret, NULL, NULL); + return sanitize_err(env, insn, ret); } /* Calculate sign/unsigned bounds and tnum for alu32 and alu64 bit ops. -- 2.34.1

2 1

[PATCH OLK-5.10] vhost/vdpa: validate virtqueue index in mmap and fault paths
by Fanhua Li 01 Sep '26

01 Sep '26

From: Qihang Tang <q.h.hack.winter(a)gmail.com> stable inclusion from stable-v5.10.261 commit 0f310bac6db9bd3bb1655707d692d9d2a86eeb17 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/18146 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 929e4f044621c8cc30b612fb74e1410bef09e41b ] vhost_vdpa_mmap() and vhost_vdpa_fault() use vma->vm_pgoff as a virtqueue index for get_vq_notification(), but they do not validate that the index is smaller than v->nvqs. The ioctl path already performs both a bounds check and array_index_nospec(), but the mmap/fault path only checks that the index fits in u16. This allows an out-of-range queue index to reach driver-specific get_vq_notification() callbacks. Fix this by extracting a unified vhost_vdpa_get_vq_notification() helper that validates the queue index against v->nvqs and applies array_index_nospec() before calling the driver callback. Both the mmap and fault paths use this helper, and the bounds checking is consolidated into a single location. From source inspection, the most defensible impact is out-of-bounds access in the callback path, potentially leading to invalid PFN remaps and crash/DoS. Fixes: ddd89d0a059d ("vhost_vdpa: support doorbell mapping via mmap") Acked-by: Eugenio Pérez <eperezma(a)redhat.com> Acked-by: Michael S. Tsirkin <mst(a)redhat.com> Signed-off-by: Qihang Tang <q.h.hack.winter(a)gmail.com> Signed-off-by: Michael S. Tsirkin <mst(a)redhat.com> Message-ID: <20260508075821.92656-1-q.h.hack.winter(a)gmail.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Conflicts: drivers/vhost/vdpa.c [Fanhua Li: context conflict] Signed-off-by: Fanhua Li <lifanhua5(a)huawei.com> --- drivers/vhost/vdpa.c | 29 ++++++++++++++++++++++------- 1 file changed, 22 insertions(+), 7 deletions(-) diff --git a/drivers/vhost/vdpa.c b/drivers/vhost/vdpa.c index 88c9d7665cedb..56d99c3d0a389 100644 --- a/drivers/vhost/vdpa.c +++ b/drivers/vhost/vdpa.c @@ -1567,16 +1567,32 @@ static int vhost_vdpa_release(struct inode *inode, struct file *filep) } #ifdef CONFIG_MMU -static vm_fault_t vhost_vdpa_fault(struct vm_fault *vmf) +static int +vhost_vdpa_get_vq_notification(struct vhost_vdpa *v, unsigned long index, + struct vdpa_notification_area *notify) { - struct vhost_vdpa *v = vmf->vma->vm_file->private_data; struct vdpa_device *vdpa = v->vdpa; const struct vdpa_config_ops *ops = vdpa->config; + + if (index > 65535 || index >= v->nvqs) + return -EINVAL; + + index = array_index_nospec(index, v->nvqs); + + *notify = ops->get_vq_notification(vdpa, index); + + return 0; +} + +static vm_fault_t vhost_vdpa_fault(struct vm_fault *vmf) +{ + struct vhost_vdpa *v = vmf->vma->vm_file->private_data; struct vdpa_notification_area notify; struct vm_area_struct *vma = vmf->vma; - u16 index = vma->vm_pgoff; + unsigned long index = vma->vm_pgoff; - notify = ops->get_vq_notification(vdpa, index); + if (vhost_vdpa_get_vq_notification(v, index, &notify)) + return VM_FAULT_SIGBUS; vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); if (remap_pfn_range(vma, vmf->address & PAGE_MASK, @@ -1605,8 +1621,6 @@ static int vhost_vdpa_mmap(struct file *file, struct vm_area_struct *vma) return -EINVAL; if (vma->vm_flags & VM_READ) return -EINVAL; - if (index > 65535) - return -EINVAL; if (!ops->get_vq_notification) return -ENOTSUPP; @@ -1614,7 +1628,8 @@ static int vhost_vdpa_mmap(struct file *file, struct vm_area_struct *vma) * support the doorbell which sits on the page boundary and * does not share the page with other registers. */ - notify = ops->get_vq_notification(vdpa, index); + if (vhost_vdpa_get_vq_notification(v, index, &notify)) + return -EINVAL; if (notify.addr & (PAGE_SIZE - 1)) return -EINVAL; if (vma->vm_end - vma->vm_start != notify.size) -- 2.43.0

2 1

[PATCH OLK-6.6 1/1] KVM: arm64: vgic-v3: Restrict ICH_AP1Rn_EL2 accesses to 64-bit only with vNMI
by Jinqian Yang 31 Aug '26

31 Aug '26

ICH_AP1Rn_EL2 is 64-bit only when vNMI is in use, with the NMI priority in ICH_AP1R0_EL2[63:32]; otherwise the upper bits are RES0. Gate the save/restore access width on vgic_v3_vcpu_has_vnmi() which checks both vgic.has_nmi and ID_AA64PFR1_EL1.NMI, so that the NMI priority is preserved across context switch only when vNMI is actually enabled for the guest. Also drop the early assignment of kvm->arch.pfr1_nmi in kvm_arch_init_vm(). Without it vNMI is no longer enabled by default at VM creation on hosts with FEAT_NMI; instead the VMM must explicitly set ID_AA64PFR1_EL1.NMI via SET_ONE_REG to enable vNMI for the guest. Signed-off-by: Jinqian Yang <yangjinqian1(a)huawei.com> --- arch/arm64/include/asm/kvm_hyp.h | 1 + arch/arm64/kvm/arm.c | 3 -- arch/arm64/kvm/hyp/vgic-v3-sr.c | 55 +++++++++++++++++++++++++++----- 3 files changed, 48 insertions(+), 11 deletions(-) diff --git a/arch/arm64/include/asm/kvm_hyp.h b/arch/arm64/include/asm/kvm_hyp.h index 127ad280e68b..c4121bf0aadc 100644 --- a/arch/arm64/include/asm/kvm_hyp.h +++ b/arch/arm64/include/asm/kvm_hyp.h @@ -82,6 +82,7 @@ void __vgic_v3_activate_traps(struct vgic_v3_cpu_if *cpu_if); void __vgic_v3_deactivate_traps(struct vgic_v3_cpu_if *cpu_if); void __vgic_v3_save_aprs(struct vgic_v3_cpu_if *cpu_if); void __vgic_v3_restore_aprs(struct vgic_v3_cpu_if *cpu_if); +bool vgic_v3_vcpu_has_vnmi(struct kvm_vcpu *vcpu); int __vgic_v3_perform_cpuif_access(struct kvm_vcpu *vcpu); #ifdef __KVM_NVHE_HYPERVISOR__ diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c index 2b560eba3e0b..9a4be08730da 100644 --- a/arch/arm64/kvm/arm.c +++ b/arch/arm64/kvm/arm.c @@ -381,9 +381,6 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) /* The maximum number of VCPUs is limited by the host's GIC model */ kvm->max_vcpus = kvm_arm_default_max_vcpus(); - if (cpus_have_const_cap(ARM64_HAS_NMI) && !static_branch_unlikely(&vgic_v3_cpuif_trap)) - kvm->arch.pfr1_nmi = ID_AA64PFR1_EL1_NMI_IMP; - kvm_arm_init_hypercalls(kvm); if (!kvm_is_realm(kvm)) diff --git a/arch/arm64/kvm/hyp/vgic-v3-sr.c b/arch/arm64/kvm/hyp/vgic-v3-sr.c index 7a9e132f68db..ab406332762e 100644 --- a/arch/arm64/kvm/hyp/vgic-v3-sr.c +++ b/arch/arm64/kvm/hyp/vgic-v3-sr.c @@ -335,10 +335,40 @@ void __vgic_v3_deactivate_traps(struct vgic_v3_cpu_if *cpu_if) write_gicreg(0, ICH_HCR_EL2); } +static struct kvm_vcpu *vgic_v3_cpu_if_to_vcpu(struct vgic_v3_cpu_if *cpu_if) +{ + struct vgic_cpu *vgic_cpu = container_of(cpu_if, struct vgic_cpu, vgic_v3); + + return container_of(vgic_cpu, struct kvm_vcpu, arch.vgic_cpu); +} + +/* + * vNMI is only usable if the guest has selected NMI support + * (vgic.has_nmi) and the vcpu has FEAT_NMI (pfr1_nmi). + * + * This drives the width of the ICH_AP1Rn_EL2 accesses: with vNMI the + * registers are 64bit (the NMI priority lives in ICH_AP1R0_EL2[63]), + * without it only the low 32 bits are valid. + */ +bool vgic_v3_vcpu_has_vnmi(struct kvm_vcpu *vcpu) +{ + struct kvm *kvm = kern_hyp_va(vcpu->kvm); + + return kvm->arch.vgic.has_nmi && + kvm->arch.pfr1_nmi == ID_AA64PFR1_EL1_NMI_IMP; +} + void __vgic_v3_save_aprs(struct vgic_v3_cpu_if *cpu_if) { u64 val; u32 nr_pre_bits; + /* + * ICH_AP1Rn_EL2 are only 64bit when the vcpu has vNMI: the NMI + * priority is encoded in ICH_AP1R0_EL2[63]. Without vNMI, the + * upper bits are RES0 and only the low 32 bits must be + * preserved. + */ + bool vnmi = vgic_v3_vcpu_has_vnmi(vgic_v3_cpu_if_to_vcpu(cpu_if)); val = read_gicreg(ICH_VTR_EL2); nr_pre_bits = vtr_to_nr_pre_bits(val); @@ -357,14 +387,18 @@ void __vgic_v3_save_aprs(struct vgic_v3_cpu_if *cpu_if) switch (nr_pre_bits) { case 7: - cpu_if->vgic_ap1r[3] = __vgic_v3_read_ap1rn(3); - cpu_if->vgic_ap1r[2] = __vgic_v3_read_ap1rn(2); + cpu_if->vgic_ap1r[3] = vnmi ? __vgic_v3_read_ap1rn(3) + : (u32)__vgic_v3_read_ap1rn(3); + cpu_if->vgic_ap1r[2] = vnmi ? __vgic_v3_read_ap1rn(2) + : (u32)__vgic_v3_read_ap1rn(2); fallthrough; case 6: - cpu_if->vgic_ap1r[1] = __vgic_v3_read_ap1rn(1); + cpu_if->vgic_ap1r[1] = vnmi ? __vgic_v3_read_ap1rn(1) + : (u32)__vgic_v3_read_ap1rn(1); fallthrough; default: - cpu_if->vgic_ap1r[0] = __vgic_v3_read_ap1rn(0); + cpu_if->vgic_ap1r[0] = vnmi ? __vgic_v3_read_ap1rn(0) + : (u32)__vgic_v3_read_ap1rn(0); } } @@ -372,6 +406,7 @@ void __vgic_v3_restore_aprs(struct vgic_v3_cpu_if *cpu_if) { u64 val; u32 nr_pre_bits; + bool vnmi = vgic_v3_vcpu_has_vnmi(vgic_v3_cpu_if_to_vcpu(cpu_if)); val = read_gicreg(ICH_VTR_EL2); nr_pre_bits = vtr_to_nr_pre_bits(val); @@ -390,14 +425,18 @@ void __vgic_v3_restore_aprs(struct vgic_v3_cpu_if *cpu_if) switch (nr_pre_bits) { case 7: - __vgic_v3_write_ap1rn(cpu_if->vgic_ap1r[3], 3); - __vgic_v3_write_ap1rn(cpu_if->vgic_ap1r[2], 2); + __vgic_v3_write_ap1rn(vnmi ? cpu_if->vgic_ap1r[3] + : (u32)cpu_if->vgic_ap1r[3], 3); + __vgic_v3_write_ap1rn(vnmi ? cpu_if->vgic_ap1r[2] + : (u32)cpu_if->vgic_ap1r[2], 2); fallthrough; case 6: - __vgic_v3_write_ap1rn(cpu_if->vgic_ap1r[1], 1); + __vgic_v3_write_ap1rn(vnmi ? cpu_if->vgic_ap1r[1] + : (u32)cpu_if->vgic_ap1r[1], 1); fallthrough; default: - __vgic_v3_write_ap1rn(cpu_if->vgic_ap1r[0], 0); + __vgic_v3_write_ap1rn(vnmi ? cpu_if->vgic_ap1r[0] + : (u32)cpu_if->vgic_ap1r[0], 0); } } -- 2.33.0

2 1