From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 1D2A82E1757; Tue, 15 Jul 2025 13:55:16 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1752587717; cv=none; b=ONqRVlWfsNnSfreESNxXoQd17rMMTUbJUOt0jW7KtjvYJf3853wmytLjhGk6eJEma+Hcsa2NylfW5VidCLm5vJoYOUgJzlHxhx3yyJm7zPBlkOO062ZAT1Ixu9j2T8uneldPxwW8nlMsDPPwjPA8szEsB/HY2AT0CvZWGg0NXbA= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1752587717; c=relaxed/simple; bh=nGEJaNNZ4tP/c2AZNLXjI1a8ovpLriGZH+KZzeUpMwE=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=AGNoB8ofkvD3gY6uQVAv+4ypTtjoQDYw73bH6Qm3YJTepyWb4e7I+wcgJ+6ELKI0DNGI6Ob+u8P3RgGhNotTIgx7kBPgeCGnxgyIj5B2VqKaa1zT0yc6s220eYkc7W6Fv1Intoxbv724hbOiEVEMK/r+iyaTAlbx5mTQXcXXjLA= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b=cS+qpyM1; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b="cS+qpyM1" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 57D6AC4CEE3; Tue, 15 Jul 2025 13:55:16 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linuxfoundation.org; s=korg; t=1752587716; bh=nGEJaNNZ4tP/c2AZNLXjI1a8ovpLriGZH+KZzeUpMwE=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=cS+qpyM12x4AGAF9vnRmhYfHiZfL9XPuzXpt7m2seX+8JeKnWXbsETBy1hdE+Isg9 +JENUljTl28y/qKSKb7EPLrhNKmxgOb93WM9QSkm6ckjTalwUR/aJvEjD9ck17vp9e TeMMlBIkxJu1GTvMIRrA9BfdSk9ze8nuobWrBrC8= From: Greg Kroah-Hartman To: stable@vger.kernel.org Cc: Greg Kroah-Hartman , patches@lists.linux.dev, Tigran Mkrtchyan , Trond Myklebust , Anna Schumaker , Sasha Levin Subject: [PATCH 5.10 116/208] NFSv4/flexfiles: Fix handling of NFS level errors in I/O Date: Tue, 15 Jul 2025 15:13:45 +0200 Message-ID: <20250715130815.594396669@linuxfoundation.org> X-Mailer: git-send-email 2.50.1 In-Reply-To: <20250715130810.830580412@linuxfoundation.org> References: <20250715130810.830580412@linuxfoundation.org> User-Agent: quilt/0.68 X-stable: review X-Patchwork-Hint: ignore Precedence: bulk X-Mailing-List: patches@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit 5.10-stable review patch. If anyone has any objections, please let me know. ------------------ From: Trond Myklebust [ Upstream commit 38074de35b015df5623f524d6f2b49a0cd395c40 ] Allow the flexfiles error handling to recognise NFS level errors (as opposed to RPC level errors) and handle them separately. The main motivator is the NFSERR_PERM errors that get returned if the NFS client connects to the data server through a port number that is lower than 1024. In that case, the client should disconnect and retry a READ on a different data server, or it should retry a WRITE after reconnecting. Reviewed-by: Tigran Mkrtchyan Fixes: d67ae825a59d ("pnfs/flexfiles: Add the FlexFile Layout Driver") Signed-off-by: Trond Myklebust Signed-off-by: Anna Schumaker Signed-off-by: Sasha Levin --- fs/nfs/flexfilelayout/flexfilelayout.c | 119 ++++++++++++++++++------- 1 file changed, 85 insertions(+), 34 deletions(-) diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c index 46b106785eb82..f8962eaec87bc 100644 --- a/fs/nfs/flexfilelayout/flexfilelayout.c +++ b/fs/nfs/flexfilelayout/flexfilelayout.c @@ -1103,6 +1103,7 @@ static void ff_layout_reset_read(struct nfs_pgio_header *hdr) } static int ff_layout_async_handle_error_v4(struct rpc_task *task, + u32 op_status, struct nfs4_state *state, struct nfs_client *clp, struct pnfs_layout_segment *lseg, @@ -1113,34 +1114,42 @@ static int ff_layout_async_handle_error_v4(struct rpc_task *task, struct nfs4_deviceid_node *devid = FF_LAYOUT_DEVID_NODE(lseg, idx); struct nfs4_slot_table *tbl = &clp->cl_session->fc_slot_table; - switch (task->tk_status) { - case -NFS4ERR_BADSESSION: - case -NFS4ERR_BADSLOT: - case -NFS4ERR_BAD_HIGH_SLOT: - case -NFS4ERR_DEADSESSION: - case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION: - case -NFS4ERR_SEQ_FALSE_RETRY: - case -NFS4ERR_SEQ_MISORDERED: + switch (op_status) { + case NFS4_OK: + case NFS4ERR_NXIO: + break; + case NFSERR_PERM: + if (!task->tk_xprt) + break; + xprt_force_disconnect(task->tk_xprt); + goto out_retry; + case NFS4ERR_BADSESSION: + case NFS4ERR_BADSLOT: + case NFS4ERR_BAD_HIGH_SLOT: + case NFS4ERR_DEADSESSION: + case NFS4ERR_CONN_NOT_BOUND_TO_SESSION: + case NFS4ERR_SEQ_FALSE_RETRY: + case NFS4ERR_SEQ_MISORDERED: dprintk("%s ERROR %d, Reset session. Exchangeid " "flags 0x%x\n", __func__, task->tk_status, clp->cl_exchange_flags); nfs4_schedule_session_recovery(clp->cl_session, task->tk_status); - break; - case -NFS4ERR_DELAY: + goto out_retry; + case NFS4ERR_DELAY: nfs_inc_stats(lseg->pls_layout->plh_inode, NFSIOS_DELAY); fallthrough; - case -NFS4ERR_GRACE: + case NFS4ERR_GRACE: rpc_delay(task, FF_LAYOUT_POLL_RETRY_MAX); - break; - case -NFS4ERR_RETRY_UNCACHED_REP: - break; + goto out_retry; + case NFS4ERR_RETRY_UNCACHED_REP: + goto out_retry; /* Invalidate Layout errors */ - case -NFS4ERR_PNFS_NO_LAYOUT: - case -ESTALE: /* mapped NFS4ERR_STALE */ - case -EBADHANDLE: /* mapped NFS4ERR_BADHANDLE */ - case -EISDIR: /* mapped NFS4ERR_ISDIR */ - case -NFS4ERR_FHEXPIRED: - case -NFS4ERR_WRONG_TYPE: + case NFS4ERR_PNFS_NO_LAYOUT: + case NFS4ERR_STALE: + case NFS4ERR_BADHANDLE: + case NFS4ERR_ISDIR: + case NFS4ERR_FHEXPIRED: + case NFS4ERR_WRONG_TYPE: dprintk("%s Invalid layout error %d\n", __func__, task->tk_status); /* @@ -1153,6 +1162,11 @@ static int ff_layout_async_handle_error_v4(struct rpc_task *task, pnfs_destroy_layout(NFS_I(inode)); rpc_wake_up(&tbl->slot_tbl_waitq); goto reset; + default: + break; + } + + switch (task->tk_status) { /* RPC connection errors */ case -ECONNREFUSED: case -EHOSTDOWN: @@ -1166,26 +1180,56 @@ static int ff_layout_async_handle_error_v4(struct rpc_task *task, nfs4_delete_deviceid(devid->ld, devid->nfs_client, &devid->deviceid); rpc_wake_up(&tbl->slot_tbl_waitq); - fallthrough; + break; default: - if (ff_layout_avoid_mds_available_ds(lseg)) - return -NFS4ERR_RESET_TO_PNFS; -reset: - dprintk("%s Retry through MDS. Error %d\n", __func__, - task->tk_status); - return -NFS4ERR_RESET_TO_MDS; + break; } + + if (ff_layout_avoid_mds_available_ds(lseg)) + return -NFS4ERR_RESET_TO_PNFS; +reset: + dprintk("%s Retry through MDS. Error %d\n", __func__, + task->tk_status); + return -NFS4ERR_RESET_TO_MDS; + +out_retry: task->tk_status = 0; return -EAGAIN; } /* Retry all errors through either pNFS or MDS except for -EJUKEBOX */ static int ff_layout_async_handle_error_v3(struct rpc_task *task, + u32 op_status, + struct nfs_client *clp, struct pnfs_layout_segment *lseg, u32 idx) { struct nfs4_deviceid_node *devid = FF_LAYOUT_DEVID_NODE(lseg, idx); + switch (op_status) { + case NFS_OK: + case NFSERR_NXIO: + break; + case NFSERR_PERM: + if (!task->tk_xprt) + break; + xprt_force_disconnect(task->tk_xprt); + goto out_retry; + case NFSERR_ACCES: + case NFSERR_BADHANDLE: + case NFSERR_FBIG: + case NFSERR_IO: + case NFSERR_NOSPC: + case NFSERR_ROFS: + case NFSERR_STALE: + goto out_reset_to_pnfs; + case NFSERR_JUKEBOX: + nfs_inc_stats(lseg->pls_layout->plh_inode, NFSIOS_DELAY); + goto out_retry; + default: + break; + } + switch (task->tk_status) { /* File access problems. Don't mark the device as unavailable */ case -EACCES: @@ -1204,6 +1248,7 @@ static int ff_layout_async_handle_error_v3(struct rpc_task *task, nfs4_delete_deviceid(devid->ld, devid->nfs_client, &devid->deviceid); } +out_reset_to_pnfs: /* FIXME: Need to prevent infinite looping here. */ return -NFS4ERR_RESET_TO_PNFS; out_retry: @@ -1214,6 +1259,7 @@ static int ff_layout_async_handle_error_v3(struct rpc_task *task, } static int ff_layout_async_handle_error(struct rpc_task *task, + u32 op_status, struct nfs4_state *state, struct nfs_client *clp, struct pnfs_layout_segment *lseg, @@ -1232,10 +1278,11 @@ static int ff_layout_async_handle_error(struct rpc_task *task, switch (vers) { case 3: - return ff_layout_async_handle_error_v3(task, lseg, idx); - case 4: - return ff_layout_async_handle_error_v4(task, state, clp, + return ff_layout_async_handle_error_v3(task, op_status, clp, lseg, idx); + case 4: + return ff_layout_async_handle_error_v4(task, op_status, state, + clp, lseg, idx); default: /* should never happen */ WARN_ON_ONCE(1); @@ -1286,6 +1333,7 @@ static void ff_layout_io_track_ds_error(struct pnfs_layout_segment *lseg, switch (status) { case NFS4ERR_DELAY: case NFS4ERR_GRACE: + case NFS4ERR_PERM: break; case NFS4ERR_NXIO: ff_layout_mark_ds_unreachable(lseg, idx); @@ -1318,7 +1366,8 @@ static int ff_layout_read_done_cb(struct rpc_task *task, trace_ff_layout_read_error(hdr); } - err = ff_layout_async_handle_error(task, hdr->args.context->state, + err = ff_layout_async_handle_error(task, hdr->res.op_status, + hdr->args.context->state, hdr->ds_clp, hdr->lseg, hdr->pgio_mirror_idx); @@ -1483,7 +1532,8 @@ static int ff_layout_write_done_cb(struct rpc_task *task, trace_ff_layout_write_error(hdr); } - err = ff_layout_async_handle_error(task, hdr->args.context->state, + err = ff_layout_async_handle_error(task, hdr->res.op_status, + hdr->args.context->state, hdr->ds_clp, hdr->lseg, hdr->pgio_mirror_idx); @@ -1529,8 +1579,9 @@ static int ff_layout_commit_done_cb(struct rpc_task *task, trace_ff_layout_commit_error(data); } - err = ff_layout_async_handle_error(task, NULL, data->ds_clp, - data->lseg, data->ds_commit_index); + err = ff_layout_async_handle_error(task, data->res.op_status, + NULL, data->ds_clp, data->lseg, + data->ds_commit_index); trace_nfs4_pnfs_commit_ds(data, err); switch (err) { -- 2.39.5