* [PATCH v2] pNFS: deadlock in pnfs_send_layoutreturn @ 2026-04-08 12:25 Ben Roberts 2026-04-10 14:30 ` Benjamin Coddington 0 siblings, 1 reply; 6+ messages in thread From: Ben Roberts @ 2026-04-08 12:25 UTC (permalink / raw) To: Trond Myklebust, Anna Schumaker; +Cc: linux-nfs, linux-kernel, Ben Roberts On a HPC cluster running 5.14.0-611.9.1.el9.x86_64, regular deadlocks were seen within pnfs_send_layoutreturn leading to userspace processes stuck in uninterruptible sleep, ultimately requiring reboots to clear. This was occurring frequently, sometimes multiple times per day on specific hosts with heavy load. Claude code was tasked with hunting down any potential deadlocks within pnfs_send_layoutreturn, and identified the following condition. This patch has been running in production on top of the EL9 kernel for over three months without any reoccurrence of the deadlock. The pnfs_send_layoutreturn() function can deadlock when memory allocation fails. The issue occurs in the error path where pnfs_put_layout_hdr() is called, which may trigger pnfs_layoutreturn_before_put_layout_hdr(), potentially causing a recursive call back to pnfs_send_layoutreturn(). Call chain that triggers the deadlock: 1. pnfs_send_layoutreturn() - kzalloc() fails 2. Error path calls pnfs_put_layout_hdr(lo) 3. pnfs_put_layout_hdr() calls pnfs_layoutreturn_before_put_layout_hdr() 4. If NFS_LAYOUT_RETURN_REQUESTED is still set, attempts another layoutreturn, creating recursion/deadlock The fix ensures that NFS_LAYOUT_RETURN_REQUESTED is cleared in the allocation failure path before calling pnfs_put_layout_hdr(). This prevents pnfs_layoutreturn_before_put_layout_hdr() from attempting another layout return, breaking the recursion cycle. v2 fixes a syntax error introduced while composing the original mail. Signed-off-by: Ben Roberts <ben.roberts@gsacapital.com> Assisted-by: Claude:claude-sonnet-4-5 --- fs/nfs/pnfs.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index bc13d1e69449..47bda53b2b3a 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -1361,6 +1361,7 @@ pnfs_send_layoutreturn(struct pnfs_layout_hdr *lo, if (unlikely(lrp == NULL)) { status = -ENOMEM; spin_lock(&ino->i_lock); + pnfs_clear_layoutreturn_info(lo); pnfs_clear_layoutreturn_waitbit(lo); spin_unlock(&ino->i_lock); put_cred(cred); -- 2.43.0 For details of how GSA uses your personal information, please see our Privacy Notice here: https://www.gsacapital.com/privacy-notice This email and any files transmitted with it contain confidential and proprietary information and is solely for the use of the intended recipient. If you are not the intended recipient please return the email to the sender and delete it from your computer and you must not use, disclose, distribute, copy, print or rely on this email or its contents. This communication is for informational purposes only. It is not intended as an offer or solicitation for the purchase or sale of any financial instrument or as an official confirmation of any transaction. Any comments or statements made herein do not necessarily reflect those of GSA Capital. GSA Capital Partners LLP is authorised and regulated by the Financial Conduct Authority and is registered in England and Wales at Stratton House, 5 Stratton Street, London W1J 8LA, number OC309261. GSA Capital Services Limited is registered in England and Wales at the same address, number 5320529. ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2] pNFS: deadlock in pnfs_send_layoutreturn 2026-04-08 12:25 [PATCH v2] pNFS: deadlock in pnfs_send_layoutreturn Ben Roberts @ 2026-04-10 14:30 ` Benjamin Coddington 2026-04-17 10:17 ` Ben Roberts 0 siblings, 1 reply; 6+ messages in thread From: Benjamin Coddington @ 2026-04-10 14:30 UTC (permalink / raw) To: Ben Roberts; +Cc: Trond Myklebust, Anna Schumaker, linux-nfs, linux-kernel Hi Ben, Did you reproduce and diagnose this problem on a recent upstream kernel version? If not, you probably want to report the issue to your Red Hat support channel directly instead of sending fixes upstream because the maintainers will want to see patches and fixes against the upstream code. The code in 5.14.0-611.9.1.el9 may have different behaviors. Ben On 8 Apr 2026, at 8:25, Ben Roberts wrote: > On a HPC cluster running 5.14.0-611.9.1.el9.x86_64, regular deadlocks were > seen within pnfs_send_layoutreturn leading to userspace processes stuck in > uninterruptible sleep, ultimately requiring reboots to clear. This was > occurring frequently, sometimes multiple times per day on specific hosts > with heavy load. Claude code was tasked with hunting down any potential > deadlocks within pnfs_send_layoutreturn, and identified the following > condition. This patch has been running in production on top of the EL9 > kernel for over three months without any reoccurrence of the deadlock. > > The pnfs_send_layoutreturn() function can deadlock when memory > allocation fails. The issue occurs in the error path where > pnfs_put_layout_hdr() is called, which may trigger > pnfs_layoutreturn_before_put_layout_hdr(), potentially causing > a recursive call back to pnfs_send_layoutreturn(). > > Call chain that triggers the deadlock: > 1. pnfs_send_layoutreturn() - kzalloc() fails > 2. Error path calls pnfs_put_layout_hdr(lo) > 3. pnfs_put_layout_hdr() calls pnfs_layoutreturn_before_put_layout_hdr() > 4. If NFS_LAYOUT_RETURN_REQUESTED is still set, attempts another > layoutreturn, creating recursion/deadlock > > The fix ensures that NFS_LAYOUT_RETURN_REQUESTED is cleared in the > allocation failure path before calling pnfs_put_layout_hdr(). This > prevents pnfs_layoutreturn_before_put_layout_hdr() from attempting > another layout return, breaking the recursion cycle. > > v2 fixes a syntax error introduced while composing the original mail. > > Signed-off-by: Ben Roberts <ben.roberts@gsacapital.com> > Assisted-by: Claude:claude-sonnet-4-5 > --- > fs/nfs/pnfs.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c > index bc13d1e69449..47bda53b2b3a 100644 > --- a/fs/nfs/pnfs.c > +++ b/fs/nfs/pnfs.c > @@ -1361,6 +1361,7 @@ pnfs_send_layoutreturn(struct pnfs_layout_hdr *lo, > if (unlikely(lrp == NULL)) { > status = -ENOMEM; > spin_lock(&ino->i_lock); > + pnfs_clear_layoutreturn_info(lo); > pnfs_clear_layoutreturn_waitbit(lo); > spin_unlock(&ino->i_lock); > put_cred(cred); > -- > 2.43.0 > > For details of how GSA uses your personal information, please see our Privacy Notice here: https://www.gsacapital.com/privacy-notice > > This email and any files transmitted with it contain confidential and proprietary information and is solely for the use of the intended recipient. > If you are not the intended recipient please return the email to the sender and delete it from your computer and you must not use, disclose, distribute, copy, print or rely on this email or its contents. > This communication is for informational purposes only. > It is not intended as an offer or solicitation for the purchase or sale of any financial instrument or as an official confirmation of any transaction. > Any comments or statements made herein do not necessarily reflect those of GSA Capital. > GSA Capital Partners LLP is authorised and regulated by the Financial Conduct Authority and is registered in England and Wales at Stratton House, 5 Stratton Street, London W1J 8LA, number OC309261. > GSA Capital Services Limited is registered in England and Wales at the same address, number 5320529. ^ permalink raw reply [flat|nested] 6+ messages in thread
* RE: [PATCH v2] pNFS: deadlock in pnfs_send_layoutreturn 2026-04-10 14:30 ` Benjamin Coddington @ 2026-04-17 10:17 ` Ben Roberts 2026-04-17 16:19 ` Benjamin Coddington 0 siblings, 1 reply; 6+ messages in thread From: Ben Roberts @ 2026-04-17 10:17 UTC (permalink / raw) To: Benjamin Coddington Cc: Trond Myklebust, Anna Schumaker, linux-nfs, linux-kernel Hi Ben, > Did you reproduce and diagnose this problem on a recent upstream kernel > version? I'm not able to switch the production systems to a more recent kernel at this time, and don't have a reliable way to reproduce the issue in the wild without risking production systems back on an unpatched kernel. The best evidence I have that this patch is needed, is that we were seeing this deadlock occur repeatedly under high loads and memory pressure last year before applying this patch locally. It was rolled out to all production systems in early Jan and we have not seen a single reoccurrence since. The relevant code paths look similar between the modified EL9 kernel and the current git HEAD. I've spent all this week trying to devise a precise reproduction (with a lot of help from an LLM since I'm not that familar with kernel development) on both 7.0.0-rc7 (9a9c8ce300cd) and 5.14.0-611.9.1 kernels to definitively prove this patch is needed, but without success. The initial analysis suggested the deadlock might be triggered from a single process via a recursive call. This theory has been ruled out; all calling paths triggered by a single process are guarded in such a way that a recursive call does not happen. Revised theory for why this patch helps is that the deadlock is subject to an inter-process race, where one process suffers a memory allocation failure, causing a second process to become a victim of bad state, leading to an unserviceable RPC request. The timing appears to be very sensitive and hard to reproduce on demand. The LLM-assisted analysis follows, and to my non-expert eyes seems compelling. Thanks, Ben --- When pnfs_send_layoutreturn() fails to allocate memory for the layoutreturn structure, it returns -ENOMEM after calling pnfs_clear_layoutreturn_waitbit to wake waiting processes. However, it fails to clear the NFS_LAYOUT_RETURN_REQUESTED flag via pnfs_clear_layoutreturn_info(). This creates a race condition where processes woken by pnfs_clear_layoutreturn_waitbit will see NFS_LAYOUT_RETURN_REQUESTED still set and may attempt another layoutreturn on the corrupted layout state, leading to hung tasks in rpc_wait_bit_killable(). The bug manifests during memory pressure when: 1. Process A calls pnfs_send_layoutreturn(), kzalloc fails 2. Error path calls pnfs_clear_layoutreturn_waitbit() (wakes waiters) 3. Error path does NOT call pnfs_clear_layoutreturn_info() (flag remains set) 4. Process B wakes from wait_on_bit() at _pnfs_return_layout:1484 5. Process B calls pnfs_put_layout_hdr(), refcount reaches zero 6. pnfs_layoutreturn_before_put_layout_hdr() checks flag at line 1400 7. Flag is still set, so Process B calls pnfs_send_layoutreturn() (async) 8. RPC operates on inconsistent/dying layout state 9. Process B hangs indefinitely in rpc_wait_bit_killable() The fix ensures that when allocation fails in the error path, both the waitbit AND the layoutreturn info are cleared, so waiting processes wake to consistent state and do not attempt operations on the corrupted layout. The Deadlock Mechanism Thread A (allocation failure): // Line 1360-1368 in pnfs_send_layoutreturn() if (unlikely(lrp == NULL)) { status = -ENOMEM; spin_lock(&ino->i_lock); pnfs_clear_layoutreturn_waitbit(lo); // Wakes Thread B spin_unlock(&ino->i_lock); // BUG: NFS_LAYOUT_RETURN_REQUESTED still set! put_cred(cred); pnfs_put_layout_hdr(lo); // Decrements refcount goto out; } Thread B (wakes up): // In _pnfs_return_layout() around line 1458 if (test_bit(NFS_LAYOUT_RETURN_LOCK, &lo->plh_flags)) { spin_unlock(&ino->i_lock); if (wait_on_bit(...)) // Thread B was blocked HERE goto out_put_layout_hdr; // Now wakes up and goes here spin_lock(&ino->i_lock); } // ... continues ... wait_on_bit(...); // Line 1484 - passes through (bit already clear) out_put_layout_hdr: pnfs_free_lseg_list(&tmp_list); pnfs_put_layout_hdr(lo); // Line 1487 - THE CRITICAL CALL Inside Thread B's pnfs_put_layout_hdr() call (line 306): void pnfs_put_layout_hdr(struct pnfs_layout_hdr *lo) { inode = lo->plh_inode; pnfs_layoutreturn_before_put_layout_hdr(lo); // 313 Called FIRST if (refcount_dec_and_lock(&lo->plh_refcount, &inode->i_lock)) { // If refcount reaches 0, layout is FREED here pnfs_detach_layout_hdr(lo); pnfs_free_layout_hdr(lo); // Layout memory FREED! } } Inside pnfs_layoutreturn_before_put_layout_hdr() (line 1395): static void pnfs_layoutreturn_before_put_layout_hdr(struct pnfs_layout_hdr *lo) { if (!test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags)) return; // Line 1399-1400 Would early return if flag was clear // BUG: Flag is STILL SET, so continues! spin_lock(&inode->i_lock); if (pnfs_layout_need_return(lo)) { ... send = pnfs_prepare_layoutreturn(lo, &stateid, &cred, &iomode); spin_unlock(&inode->i_lock); if (send) { // Sends ANOTHER layoutreturn RPC! pnfs_send_layoutreturn(lo, &stateid, &cred, iomode, PNFS_FL_LAYOUTRETURN_ASYNC); // 1412 } } } Why It Deadlocks The key issue is timing and state corruption: 1. Thread B initiates async RPC at line 1412 that references layout lo 2. Then Thread B returns to pnfs_put_layout_hdr() line 315 3. If refcount reaches 0: Layout is DETACHED and FREED (lines 318-323) 4. But the RPC from step 1 is still in flight! The RPC is now operating on a layout structure that may be: - Partially freed - Corrupted - Have invalid stateid - Have detached inode reference When the RPC machinery tries to complete this operation, it hits inconsistent state and hangs in rpc_wait_bit_killable() waiting for an RPC response that can never complete properly because the underlying state is corrupted. --- Example stack trace for a deadlocked process: [<0>] rpc_wait_bit_killable+0xd/0x60 [sunrpc] [<0>] __rpc_execute+0x13a/0x300 [sunrpc] [<0>] rpc_execute+0xc5/0xf0 [sunrpc] [<0>] rpc_run_task+0x14d/0x1c0 [sunrpc] [<0>] nfs4_proc_layoutreturn+0x14f/0x270 [nfsv4] [<0>] pnfs_send_layoutreturn+0x119/0x190 [nfsv4] [<0>] _pnfs_return_layout+0x1b6/0x280 [nfsv4] [<0>] nfs4_evict_inode+0x6d/0x70 [nfsv4] [<0>] evict+0xcc/0x1d0 [<0>] dispose_list+0x48/0x70 [<0>] evict_inodes+0x1a0/0x1b0 [<0>] generic_shutdown_super+0x37/0x100 [<0>] kill_anon_super+0x12/0x40 [<0>] nfs_kill_super+0x22/0x40 [nfs] [<0>] deactivate_locked_super+0x2e/0xb0 [<0>] cleanup_mnt+0x100/0x160 [<0>] task_work_run+0x59/0x90 [<0>] do_exit+0x264/0x480 [<0>] do_group_exit+0x2d/0x90 [<0>] get_signal+0x839/0x860 [<0>] arch_do_signal_or_restart+0x25/0x100 [<0>] exit_to_user_mode_loop+0x9c/0x130 [<0>] exit_to_user_mode_prepare+0xb9/0x100 [<0>] syscall_exit_to_user_mode+0x12/0x40 [<0>] do_syscall_64+0x6b/0xe0 [<0>] entry_SYSCALL_64_after_hwframe+0x78/0x80 Ben Roberts For details of how GSA uses your personal information, please see our Privacy Notice here: https://www.gsacapital.com/privacy-notice This email and any files transmitted with it contain confidential and proprietary information and is solely for the use of the intended recipient. If you are not the intended recipient please return the email to the sender and delete it from your computer and you must not use, disclose, distribute, copy, print or rely on this email or its contents. This communication is for informational purposes only. It is not intended as an offer or solicitation for the purchase or sale of any financial instrument or as an official confirmation of any transaction. Any comments or statements made herein do not necessarily reflect those of GSA Capital. GSA Capital Partners LLP is authorised and regulated by the Financial Conduct Authority and is registered in England and Wales at Stratton House, 5 Stratton Street, London W1J 8LA, number OC309261. GSA Capital Services Limited is registered in England and Wales at the same address, number 5320529. ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2] pNFS: deadlock in pnfs_send_layoutreturn 2026-04-17 10:17 ` Ben Roberts @ 2026-04-17 16:19 ` Benjamin Coddington 2026-04-22 10:46 ` Ben Roberts 0 siblings, 1 reply; 6+ messages in thread From: Benjamin Coddington @ 2026-04-17 16:19 UTC (permalink / raw) To: Ben Roberts Cc: Benjamin Coddington, Trond Myklebust, Anna Schumaker, linux-nfs, linux-kernel On 17 Apr 2026, at 3:17, Ben Roberts wrote: > Hi Ben, > >> Did you reproduce and diagnose this problem on a recent upstream kernel >> version? > > I'm not able to switch the production systems to a more recent kernel at > this time, and don't have a reliable way to reproduce the issue in the > wild without risking production systems back on an unpatched kernel. The > best evidence I have that this patch is needed, is that we were seeing > this deadlock occur repeatedly under high loads and memory pressure last > year before applying this patch locally. It was rolled out to all > production systems in early Jan and we have not seen a single reoccurrence > since. The relevant code paths look similar between the modified EL9 > kernel and the current git HEAD. > > I've spent all this week trying to devise a precise reproduction (with a > lot of help from an LLM since I'm not that familar with kernel > development) on both 7.0.0-rc7 (9a9c8ce300cd) and 5.14.0-611.9.1 kernels > to definitively prove this patch is needed, but without success. The > initial analysis suggested the deadlock might be triggered from a single > process via a recursive call. This theory has been ruled out; all calling > paths triggered by a single process are guarded in such a way that a > recursive call does not happen. Revised theory for why this patch helps > is that the deadlock is subject to an inter-process race, where one > process suffers a memory allocation failure, causing a second process to > become a victim of bad state, leading to an unserviceable RPC request. > The timing appears to be very sensitive and hard to reproduce on demand. > > The LLM-assisted analysis follows, and to my non-expert eyes seems > compelling. > > Thanks, > Ben The kzalloc failure is definitely a rarely-used/tested path, so its possible there's an issue there no one has seen yet, but from what I can see it looks like every call to pnfs_send_layoutreturn() first calls pnfs_prepare_layoutreturn(), which already clears NFS_LAYOUT_RETURN_REQUESTED. I don't see how you can end up with another proccess seeing the flag. There's at least one body of work in this area that your systems don't yet have: https://lore.kernel.org/linux-nfs/20240613050055.854323-1-trond.myklebust@hammerspace.com/ Again, I strongly recommend engaging RH support here. Ben ^ permalink raw reply [flat|nested] 6+ messages in thread
* RE: [PATCH v2] pNFS: deadlock in pnfs_send_layoutreturn 2026-04-17 16:19 ` Benjamin Coddington @ 2026-04-22 10:46 ` Ben Roberts 2026-04-22 12:52 ` Trond Myklebust 0 siblings, 1 reply; 6+ messages in thread From: Ben Roberts @ 2026-04-22 10:46 UTC (permalink / raw) To: Benjamin Coddington Cc: Trond Myklebust, Anna Schumaker, linux-nfs, linux-kernel Hi Ben, > The kzalloc failure is definitely a rarely-used/tested path, so its possible > there's an issue there no one has seen yet, but from what I can see it looks > like every call to pnfs_send_layoutreturn() first calls > pnfs_prepare_layoutreturn(), which already clears > NFS_LAYOUT_RETURN_REQUESTED. I don't see how you can end up with another > proccess seeing the flag. Understood, thank you for the feedback here. I withdraw the patch request and will take this back to the drawing board. Likely removing the patch from local systems and looking to capture more useful diagnostics if/when the problem reoccurs. > There's at least one body of work in this area that your systems > don't yet have: > https://lore.kernel.org/linux-nfs/20240613050055.854323-1-trond.myklebust@hammerspace.com/ For what it's worth, I checked this patch set and the changes are almost completely present in the EL source. Patch 2 is only partially applied. Ben Roberts For details of how GSA uses your personal information, please see our Privacy Notice here: https://www.gsacapital.com/privacy-notice This email and any files transmitted with it contain confidential and proprietary information and is solely for the use of the intended recipient. If you are not the intended recipient please return the email to the sender and delete it from your computer and you must not use, disclose, distribute, copy, print or rely on this email or its contents. This communication is for informational purposes only. It is not intended as an offer or solicitation for the purchase or sale of any financial instrument or as an official confirmation of any transaction. Any comments or statements made herein do not necessarily reflect those of GSA Capital. GSA Capital Partners LLP is authorised and regulated by the Financial Conduct Authority and is registered in England and Wales at Stratton House, 5 Stratton Street, London W1J 8LA, number OC309261. GSA Capital Services Limited is registered in England and Wales at the same address, number 5320529. ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2] pNFS: deadlock in pnfs_send_layoutreturn 2026-04-22 10:46 ` Ben Roberts @ 2026-04-22 12:52 ` Trond Myklebust 0 siblings, 0 replies; 6+ messages in thread From: Trond Myklebust @ 2026-04-22 12:52 UTC (permalink / raw) To: Ben Roberts, Benjamin Coddington; +Cc: Anna Schumaker, linux-nfs, linux-kernel On Wed, 2026-04-22 at 10:46 +0000, Ben Roberts wrote: > Hi Ben, > > > The kzalloc failure is definitely a rarely-used/tested path, so its > > possible > > there's an issue there no one has seen yet, but from what I can see > > it looks > > like every call to pnfs_send_layoutreturn() first calls > > pnfs_prepare_layoutreturn(), which already clears > > NFS_LAYOUT_RETURN_REQUESTED. I don't see how you can end up with > > another > > proccess seeing the flag. > I had first applied this patch, however when looking more closely, it seems that it will also be defeated by the by the loop in pnfs_clear_layoutreturn_info() that checks if there are any layouts that need returning, and resets NFS_LAYOUT_RETURN_REQUESTED. > Understood, thank you for the feedback here. I withdraw the patch > request and > will take this back to the drawing board. Likely removing the patch > from local > systems and looking to capture more useful diagnostics if/when the > problem > reoccurs. > > > There's at least one body of work in this area that your systems > > don't yet have: > > https://lore.kernel.org/linux-nfs/20240613050055.854323-1-trond.myklebust@hammerspace.com/ > > For what it's worth, I checked this patch set and the changes are > almost > completely present in the EL source. Patch 2 is only partially > applied. > > Ben Roberts ^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2026-04-22 12:52 UTC | newest] Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2026-04-08 12:25 [PATCH v2] pNFS: deadlock in pnfs_send_layoutreturn Ben Roberts 2026-04-10 14:30 ` Benjamin Coddington 2026-04-17 10:17 ` Ben Roberts 2026-04-17 16:19 ` Benjamin Coddington 2026-04-22 10:46 ` Ben Roberts 2026-04-22 12:52 ` Trond Myklebust
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®