kernel/linux.git/fs/nfs, branch v5.15.213

NFSv4/flexfiles: honor FF_FLAGS_NO_IO_THRU_MDS in pg_get_mirror_count_write

2026-07-24T13:51:38+00:00

[ Upstream commit 1d62e659c0bf11649cf48e002c2a55d148f2610a ] The FF_FLAGS_NO_IO_THRU_MDS flag lives on each lseg, so any fallback decision made when there is no current lseg (e.g. between LAYOUTRETURN and the next LAYOUTGET) cannot run the per-lseg check. Introduce a sticky hdr-level ditto for FF_FLAGS_NO_IO_THRU_MDS in struct nfs4_flexfile_layout::flags (NFS4_FF_HDR_NO_IO_THRU_MDS bit), set whenever ff_layout_alloc_lseg() parses an lseg with the flag. The bit is never cleared for the lifetime of the layout hdr; the server is assumed to be consistent in its no-fallback policy per file. kzalloc() in ff_layout_alloc_layout_hdr() zero-initializes the field. Use the new ff_layout_hdr_no_fallback_to_mds() helper to gate ff_layout_pg_get_mirror_count_write(): when pnfs_update_layout() returns NULL (e.g. NFS_LAYOUT_BULK_RECALL, pnfs_layout_io_test_failed, pnfs_layoutgets_blocked) the existing code unconditionally calls nfs_pageio_reset_write_mds(). This is a source of unwanted WRITE to MDS. Fix it by checking NFS4_FF_HDR_NO_IO_THRU_MDS bit, and if set surface -EAGAIN instead; the writepage-side caller (nfs_do_writepage() for buffered, nfs_direct_write_reschedule() for O_DIRECT) then redirties the request so writeback retries via pNFS. Fixes: 260074cd8413 ("pNFS/flexfiles: Add support for FF_FLAGS_NO_IO_THRU_MDS") Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Mike Snitzer Signed-off-by: Anna Schumaker Signed-off-by: Sasha Levin

NFSv4/flexfiles: honor FF_FLAGS_NO_IO_THRU_MDS on fatal DS connect errors

2026-07-24T13:51:38+00:00

[ Upstream commit 7a375cafc14ed151508f908ea5681caf0a9cc1d6 ] Commit f06bedfa62d5 ("pNFS/flexfiles: don't attempt pnfs on fatal DS errors") teaches ff_layout_{read,write}_pagelist() to return PNFS_NOT_ATTEMPTED when nfs4_ff_layout_prepare_ds() fails with a nfs_error_is_fatal() errno (e.g. -ETIMEDOUT from a SOFTCONN connect deadline, -ENOMEM, -ERESTARTSYS), so that the client gives up instead of spinning. pnfs_do_{read,write}() then dispatches the I/O through pnfs_{read,write}_through_mds() → nfs_pageio_reset_{read,write}_mds(). That fallback is unconditional and silently violates FF_FLAGS_NO_IO_THRU_MDS: when the layout segment carries the flag (typically single-mirror appliance layouts where MDS I/O is explicitly forbidden), the out_failed: path's \`&& !ds_fatal_error\` clause overrides the flag's short-circuit through ff_layout_avoid_mds_available_ds() and routes the I/O to the MDS file handle anyway. This is reachable in practice during a data-server restart: SOFTCONN exhaustion produces -ETIMEDOUT, which is fatal per nfs_error_is_fatal(), which triggers PNFS_NOT_ATTEMPTED, which silently goes to MDS. Preserve the upstream "don't spin on fatal errors" intent for layouts that permit MDS fallback. For layouts with FF_FLAGS_NO_IO_THRU_MDS set, mark the layout for return and request PNFS_TRY_AGAIN instead; if the server cannot supply a usable layout the failure now surfaces cleanly via pnfs_update_layout(), rather than via silent MDS I/O that contradicts the flag. Fixes: f06bedfa62d5 ("pNFS/flexfiles: don't attempt pnfs on fatal DS errors") Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Mike Snitzer Signed-off-by: Anna Schumaker Signed-off-by: Sasha Levin

NFSv4/pnfs: defer return_range callbacks until after inode unlock

2026-07-24T13:51:38+00:00

[ Upstream commit 77b160b2d863d37f36b6c38e80a7d259ce939e69 ] Sometimes unmounting an NFS filesystem mounted with pNFS SCSI layouts triggers the following warning: BUG: scheduling while atomic: umount.nfs4/... __schedule_bug+0xbd/0x100 schedule_debug.constprop.0+0x19f/0x220 __schedule+0x10d/0x10a0 schedule+0x74/0x190 schedule_timeout+0xf5/0x220 io_schedule_timeout+0xd5/0x160 __wait_for_common+0x186/0x4b0 blk_execute_rq+0x2ef/0x3a0 scsi_execute_cmd+0x1ff/0x700 sd_pr_out_command.isra.0+0x242/0x380 [sd_mod] bl_unregister_scsi.constprop.0+0x109/0x3c0 [blocklayoutdriver] bl_unregister_dev+0x175/0x1c0 [blocklayoutdriver] bl_free_device+0x1f/0x1b0 [blocklayoutdriver] bl_free_deviceid_node+0x12/0x30 [blocklayoutdriver] nfs4_put_deviceid_node+0x171/0x360 [nfsv4] ext_tree_remove+0x11c/0x1d0 [blocklayoutdriver] _pnfs_return_layout+0x416/0x900 [nfsv4] nfs4_evict_inode+0x108/0x130 [nfsv4] evict+0x316/0x750 dispose_list+0xf1/0x1a0 evict_inodes+0x33f/0x440 generic_shutdown_super+0xc9/0x4e0 kill_anon_super+0x3a/0x90 nfs_kill_super+0x44/0x60 [nfs] deactivate_locked_super+0xb8/0x1b0 cleanup_mnt+0x25a/0x380 task_work_run+0x13e/0x210 exit_to_user_mode_loop+0x169/0x400 do_syscall_64+0x467/0x1550 entry_SYSCALL_64_after_hwframe+0x76/0x7e The warning occurs because the block layout driver unregisters the SCSI device while the inode lock is still held. Device unregistration issues a SCSI PR command, which may sleep, resulting in a "scheduling while atomic" warning. During layout return, ext_tree_remove() invokes the layout driver's return_range callback while holding the inode lock. For block layouts, this callback eventually calls bl_unregister_scsi(), which may block in scsi_execute_cmd() while issuing PR commands to the device. Fix this by deferring the return_range callbacks until after the inode lock has been released. The layout header reference count is incremented before invoking return_range(), ensuring that the layout header remains valid while the layout driver removes extents from the extent tree. Fixes: c88953d87f5c8 ("pnfs: add return_range method") Signed-off-by: Dai Ngo Signed-off-by: Anna Schumaker Signed-off-by: Sasha Levin

pNFS/filelayout: fix cheking if a layout is striped

2026-07-24T13:51:38+00:00

[ Upstream commit 91668417d4e925c98cae4a55b1b9860380ddbf16 ] A layout can still be striped with num_fh = 1 as it is perfectly possible that both MDS and DSs can handle the same filehandle. Hence check according to stripe_count > 1, which is the correct check to begin with. We should not be called with flseg->dsaddr = NULL, but if for some reason we do, return our best guess with is flseg->num_fh > 1. Fixes: a6b9d2fa0024 ("pNFS/filelayout: Fix coalescing test for single DS") Signed-off-by: Sagi Grimberg Signed-off-by: Anna Schumaker Signed-off-by: Sasha Levin

NFSv4: include MAY_WRITE in open permission mask for O_TRUNC

2026-07-24T13:51:25+00:00

commit 5140f099ecd8a2f2808b7f7b720ee1bad8468974 upstream. POSIX requires write permission to truncate a file, so an open() that specifies O_TRUNC must be authorized for write access regardless of the O_ACCMODE access mode. nfs_open_permission_mask() builds the access mask passed to nfs_may_open(), which is the local authorization gate for OPENs the client serves itself from a cached write delegation via the can_open_delegated() path in nfs4_try_open_cached(). The mask is derived from O_ACCMODE alone, so an open(O_RDONLY | O_TRUNC) against a file the caller cannot write requests only MAY_READ and passes the local check. The OPEN is then satisfied locally and the truncation is issued to the server as a SETATTR(size=0) over the delegation stateid, which the server accepts under standard write-delegation semantics. POSIX requires that this open fail with EACCES. Include MAY_WRITE in the mask whenever O_TRUNC is set so the local check matches the access the server would have enforced. Suggested-by: Trond Myklebust Fixes: af22f94ae02a ("NFSv4: Simplify _nfs4_do_access()") Cc: stable@vger.kernel.org Signed-off-by: Benjamin Coddington Signed-off-by: Anna Schumaker Signed-off-by: Greg Kroah-Hartman

NFSv4/flexfiles: reject zero filehandle version count

2026-07-24T13:51:17+00:00

[ Upstream commit 2c6bb3c40bc24f6aa8dfbe6fe98c3ad6389203f2 ] ff_layout_alloc_lseg() decodes the filehandle-version array count from the flexfiles layout body. The value is used as the count for kzalloc_objs(), and the current code only rejects NULL. A zero count yields ZERO_SIZE_PTR, which can be stored in dss_info->fh_versions even though later flexfiles paths assume that at least one filehandle version exists. Reject fh_count == 0 before the allocation, matching the existing zero version_count validation in the flexfiles GETDEVICEINFO parser. A QEMU/KASAN run with a malformed flexfiles layout hit: KASAN: null-ptr-deref in range [0x0000000000000010-0x0000000000000017] RIP: 0010:ff_layout_encode_ff_layoutupdate.isra.0+0x15f/0x750 ff_layout_encode_layoutreturn+0x683/0x970 nfs4_xdr_enc_layoutreturn+0x278/0x3a0 Kernel panic - not syncing: Fatal exception The patched kernel rejects the malformed layout without KASAN/oops/panic, and a valid fh_count=1 regression still opens, reads, and unmounts cleanly. Cc: stable@vger.kernel.org Fixes: d67ae825a59d ("pnfs/flexfiles: Add the FlexFile Layout Driver") Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Michael Bommarito Signed-off-by: Anna Schumaker Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

NFSv4/pNFS: reject zero-length r_addr in nfs4_decode_mp_ds_addr

2026-07-04T11:39:40+00:00

commit 41fe0f7b84f0cb822ae10ab08592996a592b2a25 upstream. nfs4_decode_mp_ds_addr() decodes the r_netid and r_addr opaques of a netaddr4 from a GETDEVICEINFO multipath-DS body, then immediately calls strrchr(buf, '.') to locate the port separator. Both decodes use xdr_stream_decode_string_dup(), and the current code checks only "nlen < 0" / "rlen < 0" before dereferencing the returned string. When the on-wire opaque has length zero, xdr_stream_decode_opaque_inline() returns 0 and xdr_stream_decode_string_dup() falls through to its "*str = NULL; return ret" tail, leaving buf NULL with a return value of 0. The "< 0" check does not catch this, and the next line is strrchr(NULL, '.'), a kernel NULL pointer dereference reachable from any pNFS-flexfile client mounted against a malicious or compromised metadata server. Reject the zero-length cases explicitly so the decoder fails with -EBADMSG (treated as a malformed GETDEVICEINFO body) instead of panicking the client. Cc: stable@vger.kernel.org Fixes: 6b7f3cf96364 ("nfs41: pull decode_ds_addr from file layout to generic pnfs") Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Michael Bommarito Signed-off-by: Anna Schumaker Signed-off-by: Greg Kroah-Hartman

pNFS: Fix use-after-free in pnfs_update_layout()

2026-07-04T11:39:40+00:00

commit 13e198a90ca4050f4bee8a3f23680389a6563ccc upstream. When hitting the NFS_LAYOUT_RETURN branch in pnfs_update_layout(), the code calls pnfs_prepare_to_retry_layoutget(lo). If it succeeds, pnfs_put_layout_hdr(lo) is called before trace_pnfs_update_layout(), which still references 'lo'. This results in a use-after-free when the tracepoint accesses lo's fields. Fix this by moving the tracepoint call before pnfs_put_layout_hdr(lo). Fixes: 2c8d5fc37fe2 ("pNFS: Stricter ordering of layoutget and layoutreturn") Cc: stable@vger.kernel.org Signed-off-by: Wentao Liang Signed-off-by: Anna Schumaker Signed-off-by: Greg Kroah-Hartman

nfs/blocklayout: Fix compilation error (`make W=1`) in bl_write_pagelist()

2026-06-01T15:35:36+00:00

[ Upstream commit f83c8dda456ce4863f346aa26d88efa276eda35d ] Clang compiler is not happy about set but unused variable (when dprintk() is no-op): .../blocklayout/blocklayout.c:384:9: error: variable 'count' set but not used [-Werror,-Wunused-but-set-variable] Remove a leftover from the previous cleanup. Fixes: 3a6fd1f004fc ("pnfs/blocklayout: remove read-modify-write handling in bl_write_pagelist") Acked-by: Anna Schumaker Reviewed-by: Jeff Layton Signed-off-by: Andy Shevchenko Signed-off-by: Chuck Lever Signed-off-by: Sasha Levin

pNFS: fix a missing wake up while waiting on NFS_LAYOUT_DRAIN

2026-03-04T12:19:44+00:00

[ Upstream commit 5248d8474e594d156bee1ed10339cc16e207a28b ] It is possible to have a task get stuck on waiting on the NFS_LAYOUT_DRAIN in the following scenario 1. cpu a: waiter test NFS_LAYOUT_DRAIN (1) and plh_outstanding (1) 2. cpu b: atomic_dec_and_test() -> clear bit -> wake up 3. cpu c: sets NFS_LAYOUT_DRAIN again 4. cpu a: calls wait_on_bit() sleeps forever. To expand on this we have say 2 outstanding pnfs write IO that get ESTALE which causes both to call pnfs_destroy_layout() and set the NFS_LAYOUT_DRAIN bit but the 1st one doesn't call the pnfs_put_layout_hdr() yet (as that would prevent the 2nd ESTALE write from trying to call pnfs_destroy_layout()). If the 1st ESTALE write is the one that initially sets the NFS_LAYOUT_DRAIN so that new IO on this file initiates new LAYOUTGET. Another new write would find NFS_LAYOUT_DRAIN set and phl_outstanding>0 (step 1) and would wait_on_bit(). LAYOUTGET completes doing step 2. Now, the 2nd of ESTALE writes is calling pnfs_destory_layout() and set the NFS_LAYOUT_DRAIN bit (step 3). Finally, the waiting write wakes up to check the bit and goes back to sleep. The problem revolves around the fact that if NFS_LAYOUT_INVALID_STID was already set, it should not do the work of pnfs_mark_layout_stateid_invalid(), thus NFS_LAYOUT_DRAIN will not be set more than once for an invalid layout. Suggested-by: Trond Myklebust Fixes: 880265c77ac4 ("pNFS: Avoid a live lock condition in pnfs_update_layout()") Signed-off-by: Olga Kornievskaia Signed-off-by: Anna Schumaker Signed-off-by: Sasha Levin