Blob Blame History Raw
From: Trond Myklebust <trond.myklebust@hammerspace.com>
Date: Thu, 15 Apr 2021 15:41:57 -0400
Subject: [PATCH] NFS: Don't discard pNFS layout segments that are marked for
 return
Git-commit: 39fd01863616964f009599e50ca5c6ea9ebf88d6
Patch-mainline: v5.13-rc1
References: git-fixes

If the pNFS layout segment is marked with the NFS_LSEG_LAYOUTRETURN
flag, then the assumption is that it has some reporting requirement
to perform through a layoutreturn (e.g. flexfiles layout stats or error
information).

Fixes: e0b7d420f72a ("pNFS: Don't discard layout segments that are marked for return")
Cc: stable@vger.kernel.org
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Acked-by: NeilBrown <neilb@suse.com>

---
 fs/nfs/pnfs.c |    5 +++++
 1 file changed, 5 insertions(+)

--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -2337,6 +2337,9 @@ pnfs_mark_matching_lsegs_return(struct p
 
 	assert_spin_locked(&lo->plh_inode->i_lock);
 
+	if (test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags))
+		tmp_list = &lo->plh_return_segs;
+
 	list_for_each_entry_safe(lseg, next, &lo->plh_segs, pls_list)
 		if (pnfs_match_lseg_recall(lseg, return_range, seq)) {
 			dprintk("%s: marking lseg %p iomode %d "
@@ -2344,6 +2347,8 @@ pnfs_mark_matching_lsegs_return(struct p
 				lseg, lseg->pls_range.iomode,
 				lseg->pls_range.offset,
 				lseg->pls_range.length);
+			if (test_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags))
+				tmp_list = &lo->plh_return_segs;
 			if (mark_lseg_invalid(lseg, tmp_list))
 				continue;
 			remaining++;