2024-04-25 20:24:53

by Anna Schumaker

[permalink] [raw]
Subject: [PATCH] NFS: Fix READ_PLUS when server doesn't support OP_READ_PLUS

From: Anna Schumaker <[email protected]>

Olga showed me a case where the client was sending multiple READ_PLUS
calls to the server in parallel, and the server replied
NFS4ERR_OPNOTSUPP to each. The client would fall back to READ for the
first reply, but fail to retry the other calls.

I fix this by removing the test for NFS_CAP_READ_PLUS in
nfs4_read_plus_not_supported(). This allows us to reschedule any
READ_PLUS call that has a NFS4ERR_OPNOTSUPP return value, even after the
capability has been cleared.

Reported-by: Olga Kornievskaia <[email protected]>
Fixes: c567552612ec ("NFS: Add READ_PLUS data segment support")
Cc: [email protected] # v5.10+
Signed-off-by: Anna Schumaker <[email protected]>
---
fs/nfs/nfs4proc.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index ea390db94b62..c93c12063b3a 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -5456,7 +5456,7 @@ static bool nfs4_read_plus_not_supported(struct rpc_task *task,
struct rpc_message *msg = &task->tk_msg;

if (msg->rpc_proc == &nfs4_procedures[NFSPROC4_CLNT_READ_PLUS] &&
- server->caps & NFS_CAP_READ_PLUS && task->tk_status == -ENOTSUPP) {
+ task->tk_status == -ENOTSUPP) {
server->caps &= ~NFS_CAP_READ_PLUS;
msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ];
rpc_restart_call_prepare(task);
--
2.44.0



2024-04-26 11:43:56

by Benjamin Coddington

[permalink] [raw]
Subject: Re: [PATCH] NFS: Fix READ_PLUS when server doesn't support OP_READ_PLUS

On 25 Apr 2024, at 16:24, Anna Schumaker wrote:

> From: Anna Schumaker <[email protected]>
>
> Olga showed me a case where the client was sending multiple READ_PLUS
> calls to the server in parallel, and the server replied
> NFS4ERR_OPNOTSUPP to each. The client would fall back to READ for the
> first reply, but fail to retry the other calls.
>
> I fix this by removing the test for NFS_CAP_READ_PLUS in
> nfs4_read_plus_not_supported(). This allows us to reschedule any
> READ_PLUS call that has a NFS4ERR_OPNOTSUPP return value, even after the
> capability has been cleared.
>
> Reported-by: Olga Kornievskaia <[email protected]>
> Fixes: c567552612ec ("NFS: Add READ_PLUS data segment support")
> Cc: [email protected] # v5.10+
> Signed-off-by: Anna Schumaker <[email protected]>

Reviewed-by: Benjamin Coddington <[email protected]>

Ben

> ---
> fs/nfs/nfs4proc.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
> index ea390db94b62..c93c12063b3a 100644
> --- a/fs/nfs/nfs4proc.c
> +++ b/fs/nfs/nfs4proc.c
> @@ -5456,7 +5456,7 @@ static bool nfs4_read_plus_not_supported(struct rpc_task *task,
> struct rpc_message *msg = &task->tk_msg;
>
> if (msg->rpc_proc == &nfs4_procedures[NFSPROC4_CLNT_READ_PLUS] &&
> - server->caps & NFS_CAP_READ_PLUS && task->tk_status == -ENOTSUPP) {
> + task->tk_status == -ENOTSUPP) {
> server->caps &= ~NFS_CAP_READ_PLUS;
> msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ];
> rpc_restart_call_prepare(task);
> --
> 2.44.0