Conversation
rppt
pushed a commit
that referenced
this pull request
Jun 2, 2026
rmnet_dellink() removes the endpoint from the hash table with hlist_del_init_rcu() and then immediately frees it with kfree(). However, RCU readers on the receive path (rmnet_rx_handler -> __rmnet_map_ingress_handler) may still hold a reference to the endpoint and dereference ep->egress_dev after the memory has been freed. The endpoint is a kmalloc-32 object, and the stale read at offset 8 corresponds to the egress_dev pointer. BUG: unable to handle page fault for address: ffffffffde942eef Oops: 0002 [#1] SMP NOPTI CPU: 1 UID: 0 PID: 137 Comm: poc_write Not tainted 7.0.0+ #4 PREEMPTLAZY RIP: 0010:rmnet_vnd_rx_fixup (rmnet_vnd.c:27) Call Trace: <TASK> __rmnet_map_ingress_handler (rmnet_handlers.c:48 rmnet_handlers.c:101) rmnet_rx_handler (rmnet_handlers.c:129 rmnet_handlers.c:235) __netif_receive_skb_core.constprop.0 (net/core/dev.c:6096) __netif_receive_skb_one_core (net/core/dev.c:6208) netif_receive_skb (net/core/dev.c:6467) tun_get_user (drivers/net/tun.c:1955) tun_chr_write_iter (drivers/net/tun.c:2003) vfs_write (fs/read_write.c:688) ksys_write (fs/read_write.c:740) </TASK> Add an rcu_head field to struct rmnet_endpoint and replace kfree() with kfree_rcu() so the endpoint memory remains valid through the RCU grace period. Also remove the rmnet_vnd_dellink() call and inline only the nr_rmnet_devs decrement, since rmnet_vnd_dellink() would set ep->egress_dev to NULL during the grace period, creating a data race with lockless readers. Fixes: ceed73a ("drivers: net: ethernet: qualcomm: rmnet: Initial implementation") Reported-by: Xiang Mei <xmei5@asu.edu> Signed-off-by: Weiming Shi <bestswngs@gmail.com> Link: https://patch.msgid.link/20260514122511.3083479-2-bestswngs@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
rppt
pushed a commit
that referenced
this pull request
Jun 14, 2026
Since the start of the git history, brport_store() always acquired the
bridge lock. Back then this decision made sense: The bridge lock
protects the STP state of the bridge and its ports and at that time the
function was only used by two STP related attributes (cost and
priority).
Nowadays, brport_store() processes a lot more attributes and most of
them do not need the bridge lock:
* Bridge flags: Only require RTNL. Read locklessly by the data path.
Annotations can be added in net-next.
* FDB port flushing: Only requires the FDB lock.
* Multicast attributes: Only require the multicast lock.
* Group forward mask: Only requires RTNL. Read locklessly by the data
path. Annotations can be added in net-next.
* Backup port: Only requires RTNL. Read locklessly by the data path.
This is a problem as the bridge calls dev_set_promiscuity() when certain
bridge port flags change and this function can sleep since the commit
cited below, resulting in a splat such as [1].
Fix this by reducing the scope of the bridge lock and only take it when
processing the two STP related attributes that require it. Remove the
now stale comment from br_switchdev_set_port_flag(). The
SWITCHDEV_F_DEFER flag can be removed in net-next.
[1]
BUG: sleeping function called from invalid context at net/core/dev_addr_lists.c:1262
in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 372, name: bash
preempt_count: 201, expected: 0
RCU nest depth: 0, expected: 0
5 locks held by bash/372:
#0: ffff88810c51c3f0 (sb_writers#7){.+.+}-{0:0}, at: ksys_write (fs/read_write.c:740)
#1: ffff888115ce9480 (&of->mutex){+.+.}-{4:4}, at: kernfs_fop_write_iter (fs/kernfs/file.c:343)
#2: ffff88810b9fd330 (kn->active#37){.+.+}-{0:0}, at: kernfs_fop_write_iter (fs/kernfs/file.c:80 fs/kernfs/file.c:344)
#3: ffffffffa59473a0 (rtnl_mutex){+.+.}-{4:4}, at: brport_store (net/bridge/br_sysfs_if.c:326)
#4: ffff8881099d2d58 (&br->lock){+...}-{3:3}, at: brport_store (./include/linux/spinlock.h:348 net/bridge/br_sysfs_if.c:345)
Preemption disabled at:
0x0
Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
Call Trace:
<TASK>
dump_stack_lvl (lib/dump_stack.c:94 lib/dump_stack.c:120)
__might_resched.cold (kernel/sched/core.c:9163)
netif_rx_mode_run (net/core/dev_addr_lists.c:1262)
netif_rx_mode_sync (net/core/dev_addr_lists.c:1428)
dev_set_promiscuity (net/core/dev_api.c:289)
br_manage_promisc (net/bridge/br_if.c:135 net/bridge/br_if.c:172)
br_port_flags_change (net/bridge/br_if.c:242 net/bridge/br_if.c:747)
store_learning (net/bridge/br_sysfs_if.c:79 net/bridge/br_sysfs_if.c:235)
brport_store (net/bridge/br_sysfs_if.c:346)
kernfs_fop_write_iter (fs/kernfs/file.c:352)
new_sync_write (fs/read_write.c:595)
vfs_write (fs/read_write.c:688)
ksys_write (fs/read_write.c:740)
do_syscall_64 (arch/x86/entry/syscall_64.c:63 arch/x86/entry/syscall_64.c:94)
entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:121)
Fixes: 78cd408 ("net: add missing instance lock to dev_set_promiscuity")
Reviewed-by: Nikolay Aleksandrov <nikolay@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Link: https://patch.msgid.link/20260526064818.272516-3-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
rppt
pushed a commit
that referenced
this pull request
Jun 14, 2026
…kernel/git/kvmarm/kvmarm into HEAD KVM/arm64 fixes for 7.1, take #4 - Restore CONFIG_PKVM_DISABLE_STAGE2_ON_PANIC to its former glory by making sure the config symbol is correctly spelled out in the code - Don't reset the AArch32 view of the PMU counters to zero when the guest is writing to them - Fix an assorted collection of memory leaks in the newly added tracing code - Fix the capping of ZCR_EL2 which could be used in an unsanitised way by an L2 guest
rppt
pushed a commit
that referenced
this pull request
Jun 14, 2026
Writing to the netdevsim debugfs file "netdevsim/netdevsimN/fib/nexthop_bucket_activity" enters nsim_nexthop_bucket_activity_write(), which looks up a nexthop in data->nexthop_ht under rtnl_lock(). If a network namespace teardown, devlink reload or device deletion runs concurrently, nsim_fib_destroy() frees that rhashtable (and the surrounding nsim_fib_data) while the write is still in flight, leading to a slab-use-after-free: BUG: KASAN: slab-use-after-free in nsim_nexthop_bucket_activity_write+0xb9e/0xdf0 Read of size 4 at addr ff1100001a379808 by task syz.0.11967/27894 CPU: 0 UID: 0 PID: 27894 Comm: syz.0.11967 Not tainted 7.1.0-rc4-gf6f1bfc1980a #4 Call Trace: nsim_nexthop_bucket_activity_write+0xb9e/0xdf0 full_proxy_write+0x135/0x1a0 vfs_write+0x2e2/0x1040 ksys_write+0x146/0x270 __x64_sys_write+0x76/0xb0 do_syscall_64+0xb9/0x5b0 entry_SYSCALL_64_after_hwframe+0x74/0x7c Allocated by task 15957: rhashtable_init_noprof+0x3ec/0x860 nsim_fib_create+0x371/0xca0 nsim_drv_probe+0xd60/0x15c0 ... new_device_store+0x425/0x7f0 Freed by task 24: rhashtable_free_and_destroy+0x10d/0x620 nsim_fib_destroy+0xc9/0x1c0 nsim_dev_reload_destroy+0x1e7/0x530 nsim_dev_reload_down+0x6b/0xd0 devlink_reload+0x1b5/0x770 devlink_pernet_pre_exit+0x25d/0x3a0 ops_undo_list+0x1b7/0xb90 cleanup_net+0x47f/0x8a0 The buggy address belongs to the object at ff1100001a379800 which belongs to the cache kmalloc-1k of size 1024 The freed 1k object is the bucket table of data->nexthop_ht. Shortly after, the dangling table is dereferenced again and the machine also takes a GPF in __rht_bucket_nested() from the same call site. The root cause is a lifetime mismatch: the debugfs files reference nsim_fib_data (the writer dereferences data->nexthop_ht), but the interface is not bracketed around the lifetime of that data. nsim_fib_destroy() freed both rhashtables and only removed the debugfs directory afterwards, and nsim_fib_create() created the debugfs files before the rhashtables were initialized and, on the error path, freed them before removing the files. debugfs keeps the file itself alive across a ->write() via debugfs_file_get()/debugfs_file_put() (fs/debugfs/file.c), but it does not keep data->nexthop_ht alive, so the in-flight writer dereferenced freed memory. rtnl_lock() in the writer does not help, because the teardown path does not take rtnl around rhashtable_free_and_destroy(). Fix it by bracketing the debugfs interface around the data it exposes, keeping nsim_fib_create() and nsim_fib_destroy() symmetric: - In nsim_fib_destroy(), tear down the debugfs files before the data structures they reference. debugfs_remove_recursive() drops the initial active-user reference and then waits for every in-flight ->write() to drop its reference before returning, and rejects new opens (__debugfs_file_removed(), fs/debugfs/inode.c). Once it returns, no debugfs accessor can reach the FIB data, so the rhashtables and nsim_fib_data can be destroyed safely. This also covers the bool knobs in the same directory, which store pointers into the same nsim_fib_data, and the final kfree(data). - In nsim_fib_create(), create the debugfs files after the rhashtables and notifiers are set up. This closes the same race on the error-unwind path, where a concurrent writer could otherwise observe a half-constructed instance or a table that the unwind has already freed. (With only the destroy-side change, a writer racing the create window instead dereferences an uninitialized data->nexthop_ht.) This is reproducible by racing, in a loop, writes to /sys/kernel/debug/netdevsim/netdevsimN/fib/nexthop_bucket_activity against a teardown of the same netdevsim instance -- a devlink reload ("devlink dev reload netdevsim/netdevsimN"), destroying the network namespace it lives in, or "echo N > /sys/bus/netdevsim/del_device". It was found with syzkaller; a syzkaller reproducer is available. A standalone C reproducer does not trigger it reliably because the race needs the netns-teardown/reload path. Cc: <stable+noautosel@kernel.org> # netdevsim is a test harness, it's never loaded on production systems Signed-off-by: Zijing Yin <yzjaurora@gmail.com> Reviewed-by: Ido Schimmel <idosch@nvidia.com> Link: https://patch.msgid.link/20260529135718.1804031-1-yzjaurora@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
rppt
pushed a commit
that referenced
this pull request
Jun 14, 2026
…ernel/git/ath/ath Jeff Johnson says: ================== ath.git patches for v7.2 (PR #4) An assortment of cleanups and minor bug fixes across wcn36xx, ath9k, ath10k, ath11k, and ath12k. ================== Signed-off-by: Johannes Berg <johannes.berg@intel.com>
rppt
pushed a commit
that referenced
this pull request
Jun 14, 2026
strset_add_str_mem() might reallocate the strset data buffer in order to accommodate the provided string 's'. However, if 's' points to a string already present in the buffer, it becomes dangling after the realloc. This leads to a use-after-free when attempting to memcpy() the string into the new buffer. One scenario that triggers this problematic path is when resolve_btfids attempts to patch kfunc prototypes using existing BTF parameter names: | resolve_btfids: function bpf_list_push_back_impl already exists in BTF | Segmentation fault (core dumped) Compiling resolve_btfids with fsanitize=address generates a detailed report of the UAF: | ================================================================= | ERROR: AddressSanitizer: heap-use-after-free on address 0x7f4c4a500bd4 | ==1507892==ERROR: AddressSanitizer: heap-use-after-free on address 0x7f4c4a500bd4 at pc 0x55d25155a2a8 bp 0x7ffcef879060 sp 0x7ffcef878818 | READ of size 5 at 0x7f4c4a500bd4 thread T0 | #0 0x55d25155a2a7 in memcpy (tools/bpf/resolve_btfids/resolve_btfids+0xcf2a7) | #1 0x55d2515d708e in strset__add_str tools/lib/bpf/strset.c:162:2 | #2 0x55d2515c730b in btf__add_str tools/lib/bpf/btf.c:2109:8 | #3 0x55d2515c9020 in btf__add_func_param tools/lib/bpf/btf.c:3108:14 | #4 0x55d25159f0b5 in process_kfunc_with_implicit_args tools/bpf/resolve_btfids/main.c:1196:9 | #5 0x55d25159e004 in btf2btf tools/bpf/resolve_btfids/main.c:1229:9 | #6 0x55d25159cee7 in main tools/bpf/resolve_btfids/main.c:1535:6 | #7 0x7f4c78e29f76 in __libc_start_call_main csu/../sysdeps/nptl/libc_start_call_main.h:58:16 | #8 0x7f4c78e2a026 in __libc_start_main csu/../csu/libc-start.c:360:3 | #9 0x55d2514bb860 in _start (tools/bpf/resolve_btfids/resolve_btfids+0x30860) | | 0x7f4c4a500bd4 is located 13268 bytes inside of 2829000-byte region [0x7f4c4a4fd800,0x7f4c4a7b02c8) | freed by thread T0 here: | #0 0x55d25155b700 in realloc (tools/bpf/resolve_btfids/resolve_btfids+0xd0700) | #1 0x55d2515c426c in libbpf_reallocarray tools/lib/bpf/./libbpf_internal.h:220:9 | #2 0x55d2515c426c in libbpf_add_mem tools/lib/bpf/btf.c:224:13 | | previously allocated by thread T0 here: | #0 0x55d25155b2e3 in malloc (tools/bpf/resolve_btfids/resolve_btfids+0xd02e3) | #1 0x55d2515d6e7d in strset__new tools/lib/bpf/strset.c:58:20 While resolve_btfids could be refactored to avoid this call path, let's instead fix this issue at the source in strset__add_str() and avoid similar scenarios. Let's check if set->strs_data was reallocated and whether 's' points to an internal string within the old strset buffer. In such case, 's' is reconstructed to point to the new buffer. While already here, also fix strset__find_str() which suffers from the same problem by factoring out the common operations into a new helper function strset_str_append(). Fixes: 90d76d3 ("libbpf: Extract internal set-of-strings datastructure APIs") Suggested-by: Andrii Nakryiko <andrii@kernel.org> Suggested-by: Mykyta Yatsenko <yatsenko@meta.com> Signed-off-by: Carlos Llamas <cmllamas@google.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20260523162722.2718940-1-cmllamas@google.com
rppt
pushed a commit
that referenced
this pull request
Jun 14, 2026
A deadlock occurs in the audit subsystem when duplicating executable-related rules. When a file is moved (e.g., via do_renameat2()), the VFS layer locks the parent directory (I_MUTEX_PARENT), which synchronously triggers an fsnotify_move event. If an existing executable audit rule matches the file being moved, the audit subsystem catches this event and calls audit_dupe_exe() to duplicate the watch and update the rule. Then, audit_alloc_mark() would call kern_path_parent() to resolve the path, leading to a blind attempt to acquire the exact same I_MUTEX_PARENT lock already held by the task, resulting in the following recursive locking deadlock: ============================================ WARNING: possible recursive locking detected 6.12.0-55.27.1.el10_0.x86_64+debug #1 Not tainted -------------------------------------------- mv/5099 is trying to acquire lock: ffff888132845358 (&inode->i_sb->s_type->i_mutex_dir_key/1){+.+.}-{3:3}, at: __kern_path_locked+0x10a/0x2f0 but task is already holding lock: ffff888132846b58 (&inode->i_sb->s_type->i_mutex_dir_key/1){+.+.}-{3:3}, at: lock_two_directories+0x13f/0x2b0 other info that might help us debug this: Possible unsafe locking scenario: CPU0 ---- lock(&inode->i_sb->s_type->i_mutex_dir_key/1); lock(&inode->i_sb->s_type->i_mutex_dir_key/1); *** DEADLOCK *** May be due to missing lock nesting notation 6 locks held by mv/5099: #0: ffff888112a9c440 (sb_writers#13) at: do_renameat2+0x34c/0xbc0 #1: ffff888112a9c790 (&type->s_vfs_rename_key#3) at: do_renameat2+0x415/0xbc0 #2: ffff888132846b58 (&inode->i_sb->s_type->i_mutex_dir_key/1) at: lock_two_directories+0x13f/0x2b0 #3: ffff888132845358 (&inode->i_sb->s_type->i_mutex_dir_key/5) at: lock_two_directories+0x175/0x2b0 #4: ffffffffb3a1fb10 (&fsnotify_mark_srcu) at: fsnotify+0x454/0x28a0 #5: ffffffffaf886230 (audit_filter_mutex) at: audit_update_watch+0x36/0x11e0 stack backtrace: Call Trace: <TASK> dump_stack_lvl+0x6f/0xb0 print_deadlock_bug.cold+0xbd/0xca validate_chain+0x83a/0xf00 __lock_acquire+0xcac/0x1d20 lock_acquire.part.0+0x11b/0x360 down_write_nested+0x9f/0x230 __kern_path_locked+0x10a/0x2f0 kern_path_locked+0x26/0x40 audit_alloc_mark+0xfb/0x4f0 audit_dupe_exe+0x6c/0xe0 audit_dupe_rule+0x6c2/0xc00 audit_update_watch+0x4cc/0x11e0 audit_watch_handle_event+0x12c/0x1b0 send_to_group+0x5d0/0x8b0 fsnotify+0x615/0x28a0 fsnotify_move+0x1d8/0x630 vfs_rename+0xdcd/0x1df0 do_renameat2+0x9d4/0xbc0 __x64_sys_renameat+0x192/0x260 do_syscall_64+0x92/0x180 entry_SYSCALL_64_after_hwframe+0x76/0x7e RIP: 0033:0x7f0491fe8c4e Code: 0f 1f 40 00 48 8b 15 c1 e1 16 00 f7 d8 64 89 02 b8 ff ff ff ff c3 66 0f 1f 44 00 00 f3 0f 1e fa 49 89 ca b8 08 01 00 00 0f 05 <48> 3d 00 f0 ff ff 77 0a c3 66 0f 1f 84 00 00 00 00 00 48 8b 15 89 RSP: 002b:00007ffc7210bf38 EFLAGS: 00000246 ORIG_RAX: 0000000000000108 RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f0491fe8c4e RDX: 0000000000000003 RSI: 00007ffc7210e6c8 RDI: 00000000ffffff9c RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000001 R10: 00005575eb2dae2a R11: 0000000000000246 R12: 00005575eb2dae2a R13: 00007ffc7210e6c8 R14: 0000000000000003 R15: 00000000ffffff9c </TASK> The aforementioned deadlock can be consistently reproduced by running the script below: audit-dupe-exe-deadlock.sh -------------------------- #!/bin/bash auditctl -D mkdir -p /tmp/foo touch /tmp/file auditctl -a always,exit -F exe=/tmp/file -F path=/tmp/file -S all -k dr mv /tmp/file /tmp/foo/file rm -Rf /tmp/foo This patch fixes the issue by introducing struct audit_watch_ctx to pass the fsnotify event context down to audit_alloc_mark(). By utilizing the already-resolved directory inode provided by the event, we bypass the kern_path_parent() path resolution entirely, safely avoiding the recursive lock. Furthermore, it explicitly allows duplicate fsnotify marks (allow_dups = 1) during the rename update, allowing the new rule's mark to safely coexist with the old rule's mark until the old rule is freed. P.S.: This issue was identified and reproduced during a comprehensive code coverage analysis of the audit subsystem. The full report is available at the link below: https://people.redhat.com/rrobaina/audit-code-coverage-analysis.pdf P.P.S: With the permission of both Ricardo and Nathan, I've squashed a fixup patch from Nathan that addresses a compile time error when CONFIG_AUDITSYSCALL=n. Cc: stable@kernel.org Fixes: 34d99af ("audit: implement audit by executable") Acked-by: Waiman Long <longman@redhat.com> Acked-by: Richard Guy Briggs <rgb@redhat.com> Signed-off-by: Nathan Chancellor <nathan@kernel.org> Signed-off-by: Ricardo Robaina <rrobaina@redhat.com> [PM: my link metadata into the msg, apply fix from NC] Signed-off-by: Paul Moore <paul@paul-moore.com>
rppt
pushed a commit
that referenced
this pull request
Jun 14, 2026
…unlock race When the FUTEX_ROBUST_UNLOCK mechanism is used for unlocking (PI-)futexes, then the unlock sequence in user space looks like this: 1) robust_list_set_op_pending(mutex); 2) robust_list_remove(mutex); lval = gettid(); 3) if (atomic_try_cmpxchg(&mutex->lock, lval, 0)) 4) robust_list_clear_op_pending(); else 5) sys_futex(OP | FUTEX_ROBUST_UNLOCK, ....); That still leaves a minimal race window between #3 and #4 where the mutex could be acquired by some other task, which observes that it is the last user and: 1) unmaps the mutex memory 2) maps a different file, which ends up covering the same address When then the original task exits before reaching #5 then the kernel robust list handling observes the pending op entry and tries to fix up user space. In case that the newly mapped data contains the TID of the exiting thread at the address of the mutex/futex the kernel will set the owner died bit in that memory and therefore corrupt unrelated data. On X86 this boils down to this simplified assembly sequence: mov %esi,%eax // Load TID into EAX xor %ecx,%ecx // Set ECX to 0 #3 lock cmpxchg %ecx,(%rdi) // Try the TID -> 0 transition .Lstart: jnz .Lend #4 movq %rcx,(%rdx) // Clear list_op_pending .Lend: If the cmpxchg() succeeds and the task is interrupted before it can clear list_op_pending in the robust list head (#4) and the task crashes in a signal handler or gets killed then it ends up in do_exit() and subsequently in the robust list handling, which then might run into the unmap/map issue described above. This is only relevant when user space was interrupted and a signal is pending. The fix-up has to be done before signal delivery is attempted because: 1) The signal might be fatal so get_signal() ends up in do_exit() 2) The signal handler might crash or the task is killed before returning from the handler. At that point the instruction pointer in pt_regs is not longer the instruction pointer of the initially interrupted unlock sequence. The right place to handle this is in __exit_to_user_mode_loop() before invoking arch_do_signal_or_restart() as this covers obviously both scenarios. As this is only relevant when the task was interrupted in user space, this is tied to RSEQ and the generic entry code as RSEQ keeps track of user space interrupts unconditionally even if the task does not have a RSEQ region installed. That makes the decision very lightweight: if (current->rseq.user_irq && within(regs, csr->unlock_ip_range)) futex_fixup_robust_unlock(regs, csr); futex_fixup_robust_unlock() then invokes a architecture specific function to return the pending op pointer or NULL. The function evaluates the register content to decide whether the pending ops pointer in the robust list head needs to be cleared. Assuming the above unlock sequence, then on x86 this decision is the trivial evaluation of the zero flag: return regs->eflags & X86_EFLAGS_ZF ? regs->dx : NULL; Other architectures might need to do more complex evaluations due to LLSC, but the approach is valid in general. The size of the pointer is determined from the matching range struct, which covers both 32-bit and 64-bit builds including COMPAT. The unlock sequence is going to be placed in the VDSO so that the kernel can keep everything synchronized, especially the register usage. The resulting code sequence for user space is: if (__vdso_futex_robust_list$SZ_try_unlock(lock, tid, &pending_op) != tid) err = sys_futex($OP | FUTEX_ROBUST_UNLOCK,....); Both the VDSO unlock and the kernel side unlock ensure that the pending_op pointer is always cleared when the lock becomes unlocked. Signed-off-by: Thomas Gleixner <tglx@kernel.org> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: André Almeida <andrealmeid@igalia.com> Link: https://patch.msgid.link/20260602090535.773669210@kernel.org
rppt
pushed a commit
that referenced
this pull request
Jun 14, 2026
When the FUTEX_ROBUST_UNLOCK mechanism is used for unlocking (PI-)futexes, then the unlock sequence in userspace looks like this: 1) robust_list_set_op_pending(mutex); 2) robust_list_remove(mutex); lval = gettid(); 3) if (atomic_try_cmpxchg(&mutex->lock, lval, 0)) 4) robust_list_clear_op_pending(); else 5) sys_futex(OP,...FUTEX_ROBUST_UNLOCK); That still leaves a minimal race window between #3 and #4 where the mutex could be acquired by some other task which observes that it is the last user and: 1) unmaps the mutex memory 2) maps a different file, which ends up covering the same address When then the original task exits before reaching #5 then the kernel robust list handling observes the pending op entry and tries to fix up user space. In case that the newly mapped data contains the TID of the exiting thread at the address of the mutex/futex the kernel will set the owner died bit in that memory and therefore corrupt unrelated data. Provide a VDSO function which exposes the critical section window in the VDSO symbol table. The resulting addresses are updated in the task's mm when the VDSO is (re)map()'ed. The core code detects when a task was interrupted within the critical section and is about to deliver a signal. It then invokes an architecture specific function which determines whether the pending op pointer has to be cleared or not. The unlock assembly sequence on 64-bit is: mov %esi,%eax // Load TID into EAX xor %ecx,%ecx // Set ECX to 0 lock cmpxchg %ecx,(%rdi) // Try the TID -> 0 transition .Lstart: jnz .Lend movq %rcx,(%rdx) // Clear list_op_pending .Lend: ret So the decision can be simply based on the ZF state in regs->flags. The pending op pointer is always in DX independent of the build mode (32/64-bit) to make the pending op pointer retrieval uniform. The size of the pointer is stored in the matching criticial section range struct and the core code retrieves it from there. So the pointer retrieval function does not have to care. It is bit-size independent: return regs->flags & X86_EFLAGS_ZF ? regs->dx : NULL; There are two entry points to handle the different robust list pending op pointer size: __vdso_futex_robust_list64_try_unlock() __vdso_futex_robust_list32_try_unlock() The 32-bit VDSO provides only __vdso_futex_robust_list32_try_unlock(). The 64-bit VDSO provides always __vdso_futex_robust_list64_try_unlock() and when COMPAT is enabled also the list32 variant, which is required to support multi-size robust list pointers used by gaming emulators. The unlock function is inspired by an idea from Mathieu Desnoyers. Signed-off-by: Thomas Gleixner <tglx@kernel.org> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: André Almeida <andrealmeid@igalia.com> Acked-by: Uros Bizjak <ubizjak@gmail.com> Link: https://lore.kernel.org/20260311185409.1988269-1-mathieu.desnoyers@efficios.com Link: https://patch.msgid.link/20260602090535.883796247@kernel.org
rppt
pushed a commit
that referenced
this pull request
Jun 14, 2026
tl;dr: Use stop_machine() and a state machine based on the "MULTI_STOP" pattern to implement core TDX module update logic. Long version: TDX module updates require careful synchronization with other TDX operations. The requirements are (#1/#2 reflect current behavior that must be preserved): 1. SEAMCALLs need to be callable from both process and IRQ contexts. 2. SEAMCALLs need to be able to run concurrently across CPUs 3. During updates, only update-related SEAMCALLs are permitted; all other SEAMCALLs shouldn't be called. 4. During updates, all online CPUs must participate in the update work. No single lock primitive satisfies all requirements. For instance, rwlock_t handles #1/#2 but fails #4: CPUs spinning with IRQs disabled cannot be directed to perform update work. Use stop_machine() as it is the only well-understood mechanism that can meet all requirements. And TDX module updates consist of several steps (See Intel Trust Domain Extensions (Intel TDX) Module Base Architecture Specification, Chapter "TD-Preserving TDX module Update"). Ordering requirements between steps mandate lockstep synchronization across all CPUs. multi_cpu_stop() provides a good example of executing a multi-step task in lockstep across CPUs, but it does not synchronize the individual steps inside the callback itself. Implement a similar state machine as the skeleton for TDX module updates. Each state represents one step in the update flow, and the state advances only after all CPUs acknowledge completion of the current step. This acknowledgment mechanism provides the required lockstep execution. The update flow is intentionally simpler than multi_cpu_stop() in two ways: a) use a spinlock to protect the control data instead of atomic_t and explicit memory barriers. b) omit touch_nmi_watchdog() and rcu_momentary_eqs(), which exist there for debugging and are not strictly needed for this update flow Potential alternative to stop_machine() ======================================= An alternative approach is to lock all KVM entry points and kick all vCPUs. Here, KVM entry points refer to KVM VM/vCPU ioctl entry points, implemented in KVM common code (virt/kvm). Adding a locking mechanism there would affect all architectures KVM supports. And to lock only TDX vCPUs, new logic would be needed to identify TDX vCPUs, which the KVM common code currently lacks. This would add significant complexity and maintenance overhead to KVM for this TDX-specific use case, so don't take this approach. [ dhansen: normal changelog/style munging ] Signed-off-by: Chao Gao <chao.gao@intel.com> Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com> Reviewed-by: Xu Yilun <yilun.xu@linux.intel.com> Reviewed-by: Tony Lindgren <tony.lindgren@linux.intel.com> Reviewed-by: Kai Huang <kai.huang@intel.com> Reviewed-by: Kiryl Shutsemau (Meta) <kas@kernel.org> Reviewed-by: Rick Edgecombe <rick.p.edgecombe@intel.com> Link: https://patch.msgid.link/20260520133909.409394-15-chao.gao@intel.com
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.