diff options
| author | Trond Myklebust <trond.myklebust@hammerspace.com> | 2022-03-21 12:34:19 -0400 |
|---|---|---|
| committer | Trond Myklebust <trond.myklebust@hammerspace.com> | 2022-03-22 15:52:56 -0400 |
| commit | 515dcdcd48736576c6f5c197814da6f81c60a21e (patch) | |
| tree | 1e60066cbf650cf06a95386681744d867490156b /fs/nfs/internal.h | |
| parent | b2648015d4521de21ed3c9f48f718e023860b8c1 (diff) | |
NFS: nfsiod should not block forever in mempool_alloc()
The concern is that since nfsiod is sometimes required to kick off a
commit, it can get locked up waiting forever in mempool_alloc() instead
of failing gracefully and leaving the commit until later.
Try to allocate from the slab first, with GFP_KERNEL | __GFP_NORETRY,
then fall back to a non-blocking attempt to allocate from the memory
pool.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Diffstat (limited to 'fs/nfs/internal.h')
| -rw-r--r-- | fs/nfs/internal.h | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 194840a97e3a..57b0497105c8 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -587,6 +587,13 @@ nfs_write_match_verf(const struct nfs_writeverf *verf, !nfs_write_verifier_cmp(&req->wb_verf, &verf->verifier); } +static inline gfp_t nfs_io_gfp_mask(void) +{ + if (current->flags & PF_WQ_WORKER) + return GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN; + return GFP_KERNEL; +} + /* unlink.c */ extern struct rpc_task * nfs_async_rename(struct inode *old_dir, struct inode *new_dir, |