3 files changed, 47 insertions, 39 deletions

diff --git a/fs/nfs/filelayout/filelayout.c b/fs/nfs/filelayout/filelayout.c
index 9c96e3e5ed35..5c4551117c58 100644
--- a/fs/nfs/filelayout/filelayout.c
+++ b/fs/nfs/filelayout/filelayout.c
@@ -181,6 +181,8 @@ static int filelayout_async_handle_error(struct rpc_task *task,
 	case -EIO:
 	case -ETIMEDOUT:
 	case -EPIPE:
+	case -EPROTO:
+	case -ENODEV:
 		dprintk("%s DS connection error %d\n", __func__,
 			task->tk_status);
 		nfs4_mark_deviceid_unavailable(devid);
@@ -486,7 +488,7 @@ filelayout_read_pagelist(struct nfs_pgio_header *hdr)
 	/* Perform an asynchronous read to ds */
 	nfs_initiate_pgio(ds_clnt, hdr, hdr->cred,
 			  NFS_PROTO(hdr->inode), &filelayout_read_call_ops,
-			  0, RPC_TASK_SOFTCONN);
+			  0, RPC_TASK_SOFTCONN, NULL);
 	return PNFS_ATTEMPTED;
 }
 
@@ -528,7 +530,7 @@ filelayout_write_pagelist(struct nfs_pgio_header *hdr, int sync)
 	/* Perform an asynchronous write */
 	nfs_initiate_pgio(ds_clnt, hdr, hdr->cred,
 			  NFS_PROTO(hdr->inode), &filelayout_write_call_ops,
-			  sync, RPC_TASK_SOFTCONN);
+			  sync, RPC_TASK_SOFTCONN, NULL);
 	return PNFS_ATTEMPTED;
 }
 
@@ -603,14 +605,6 @@ filelayout_check_layout(struct pnfs_layout_hdr *lo,
 
 	dprintk("--> %s\n", __func__);
 
-	/* FIXME: remove this check when layout segment support is added */
-	if (lgr->range.offset != 0 ||
-	    lgr->range.length != NFS4_MAX_UINT64) {
-		dprintk("%s Only whole file layouts supported. Use MDS i/o\n",
-			__func__);
-		goto out;
-	}
-
 	if (fl->pattern_offset > lgr->range.offset) {
 		dprintk("%s pattern_offset %lld too large\n",
 				__func__, fl->pattern_offset);
@@ -652,19 +646,19 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo,
 {
 	struct xdr_stream stream;
 	struct xdr_buf buf;
-	struct page *scratch;
+	struct folio *scratch;
 	__be32 *p;
 	uint32_t nfl_util;
 	int i;
 
 	dprintk("%s: set_layout_map Begin\n", __func__);
 
-	scratch = alloc_page(gfp_flags);
+	scratch = folio_alloc(gfp_flags, 0);
 	if (!scratch)
 		return -ENOMEM;
 
 	xdr_init_decode_pages(&stream, &buf, lgr->layoutp->pages, lgr->layoutp->len);
-	xdr_set_scratch_page(&stream, scratch);
+	xdr_set_scratch_folio(&stream, scratch);
 
 	/* 20 = ufl_util (4), first_stripe_index (4), pattern_offset (8),
 	 * num_fh (4) */
@@ -730,11 +724,11 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo,
 			fl->fh_array[i]->size);
 	}
 
-	__free_page(scratch);
+	folio_put(scratch);
 	return 0;
 
 out_err:
-	__free_page(scratch);
+	folio_put(scratch);
 	return -EIO;
 }
 
@@ -781,6 +775,12 @@ filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid,
 	return &fl->generic_hdr;
 }
 
+static bool
+filelayout_lseg_is_striped(const struct nfs4_filelayout_segment *flseg)
+{
+	return flseg->num_fh > 1;
+}
+
 /*
  * filelayout_pg_test(). Called by nfs_can_coalesce_requests()
  *
@@ -801,6 +801,8 @@ filelayout_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev,
 	size = pnfs_generic_pg_test(pgio, prev, req);
 	if (!size)
 		return 0;
+	else if (!filelayout_lseg_is_striped(FILELAYOUT_LSEG(pgio->pg_lseg)))
+		return size;
 
 	/* see if req and prev are in the same stripe */
 	if (prev) {
@@ -839,7 +841,12 @@ fl_pnfs_update_layout(struct inode *ino,
 
 	lseg = pnfs_update_layout(ino, ctx, pos, count, iomode, strict_iomode,
 				  gfp_flags);
-	if (IS_ERR_OR_NULL(lseg))
+	if (IS_ERR(lseg)) {
+		/* Fall back to MDS on recoverable errors */
+		if (!nfs_error_is_fatal_on_server(PTR_ERR(lseg)))
+			lseg = NULL;
+		goto out;
+	} else if (!lseg)
 		goto out;
 
 	lo = NFS_I(ino)->layout;
@@ -847,6 +854,8 @@ fl_pnfs_update_layout(struct inode *ino,
 
 	status = filelayout_check_deviceid(lo, fl, gfp_flags);
 	if (status) {
+		pnfs_error_mark_layout_for_return(ino, lseg);
+		pnfs_set_lo_fail(lseg);
 		pnfs_put_lseg(lseg);
 		lseg = NULL;
 	}
@@ -858,15 +867,15 @@ static void
 filelayout_pg_init_read(struct nfs_pageio_descriptor *pgio,
 			struct nfs_page *req)
 {
-	pnfs_generic_pg_check_layout(pgio);
+	pnfs_generic_pg_check_layout(pgio, req);
 	if (!pgio->pg_lseg) {
 		pgio->pg_lseg = fl_pnfs_update_layout(pgio->pg_inode,
 						      nfs_req_openctx(req),
-						      0,
-						      NFS4_MAX_UINT64,
+						      req_offset(req),
+						      req->wb_bytes,
 						      IOMODE_READ,
 						      false,
-						      GFP_KERNEL);
+						      nfs_io_gfp_mask());
 		if (IS_ERR(pgio->pg_lseg)) {
 			pgio->pg_error = PTR_ERR(pgio->pg_lseg);
 			pgio->pg_lseg = NULL;
@@ -882,15 +891,15 @@ static void
 filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio,
 			 struct nfs_page *req)
 {
-	pnfs_generic_pg_check_layout(pgio);
+	pnfs_generic_pg_check_layout(pgio, req);
 	if (!pgio->pg_lseg) {
 		pgio->pg_lseg = fl_pnfs_update_layout(pgio->pg_inode,
 						      nfs_req_openctx(req),
-						      0,
-						      NFS4_MAX_UINT64,
+						      req_offset(req),
+						      req->wb_bytes,
 						      IOMODE_RW,
 						      false,
-						      GFP_NOFS);
+						      nfs_io_gfp_mask());
 		if (IS_ERR(pgio->pg_lseg)) {
 			pgio->pg_error = PTR_ERR(pgio->pg_lseg);
 			pgio->pg_lseg = NULL;
@@ -1002,7 +1011,7 @@ static int filelayout_initiate_commit(struct nfs_commit_data *data, int how)
 		data->args.fh = fh;
 	return nfs_initiate_commit(ds_clnt, data, NFS_PROTO(data->inode),
 				   &filelayout_commit_call_ops, how,
-				   RPC_TASK_SOFTCONN);
+				   RPC_TASK_SOFTCONN, NULL);
 out_err:
 	pnfs_generic_prepare_to_resend_writes(data);
 	pnfs_generic_commit_release(data);
@@ -1075,7 +1084,7 @@ filelayout_setup_ds_info(struct pnfs_ds_commit_info *fl_cinfo,
 	unsigned int size = (fl->stripe_type == STRIPE_SPARSE) ?
 		fl->dsaddr->ds_num : fl->dsaddr->stripe_count;
 
-	new = pnfs_alloc_commit_array(size, GFP_NOIO);
+	new = pnfs_alloc_commit_array(size, nfs_io_gfp_mask());
 	if (new) {
 		spin_lock(&inode->i_lock);
 		array = pnfs_add_commit_array(fl_cinfo, new, lseg);
@@ -1101,7 +1110,6 @@ static const struct pnfs_commit_ops filelayout_commit_ops = {
 	.clear_request_commit	= pnfs_generic_clear_request_commit,
 	.scan_commit_lists	= pnfs_generic_scan_commit_lists,
 	.recover_commit_reqs	= pnfs_generic_recover_commit_reqs,
-	.search_commit_reqs	= pnfs_generic_search_commit_reqs,
 	.commit_pagelist	= filelayout_commit_pagelist,
 };
 
diff --git a/fs/nfs/filelayout/filelayout.h b/fs/nfs/filelayout/filelayout.h
index 79323b5dab0c..c7bb5da93307 100644
--- a/fs/nfs/filelayout/filelayout.h
+++ b/fs/nfs/filelayout/filelayout.h
@@ -51,7 +51,7 @@ struct nfs4_file_layout_dsaddr {
 	u32				stripe_count;
 	u8				*stripe_indices;
 	u32				ds_num;
-	struct nfs4_pnfs_ds		*ds_list[1];
+	struct nfs4_pnfs_ds		*ds_list[] __counted_by(ds_num);
 };
 
 struct nfs4_filelayout_segment {
diff --git a/fs/nfs/filelayout/filelayoutdev.c b/fs/nfs/filelayout/filelayoutdev.c
index 86c3f7e69ec4..df79aeb68db4 100644
--- a/fs/nfs/filelayout/filelayoutdev.c
+++ b/fs/nfs/filelayout/filelayoutdev.c
@@ -35,6 +35,7 @@
 #include "../internal.h"
 #include "../nfs4session.h"
 #include "filelayout.h"
+#include "../nfs4trace.h"
 
 #define NFSDBG_FACILITY		NFSDBG_PNFS_LD
 
@@ -72,17 +73,18 @@ nfs4_fl_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev,
 	struct nfs4_file_layout_dsaddr *dsaddr = NULL;
 	struct xdr_stream stream;
 	struct xdr_buf buf;
-	struct page *scratch;
+	struct folio *scratch;
 	struct list_head dsaddrs;
 	struct nfs4_pnfs_ds_addr *da;
+	struct net *net = server->nfs_client->cl_net;
 
 	/* set up xdr stream */
-	scratch = alloc_page(gfp_flags);
+	scratch = folio_alloc(gfp_flags, 0);
 	if (!scratch)
 		goto out_err;
 
 	xdr_init_decode_pages(&stream, &buf, pdev->pages, pdev->pglen);
-	xdr_set_scratch_page(&stream, scratch);
+	xdr_set_scratch_folio(&stream, scratch);
 
 	/* Get the stripe count (number of stripe index) */
 	p = xdr_inline_decode(&stream, 4);
@@ -136,9 +138,7 @@ nfs4_fl_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev,
 		goto out_err_free_stripe_indices;
 	}
 
-	dsaddr = kzalloc(sizeof(*dsaddr) +
-			(sizeof(struct nfs4_pnfs_ds *) * (num - 1)),
-			gfp_flags);
+	dsaddr = kzalloc(struct_size(dsaddr, ds_list, num), gfp_flags);
 	if (!dsaddr)
 		goto out_err_free_stripe_indices;
 
@@ -160,8 +160,7 @@ nfs4_fl_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev,
 
 		mp_count = be32_to_cpup(p); /* multipath count */
 		for (j = 0; j < mp_count; j++) {
-			da = nfs4_decode_mp_ds_addr(server->nfs_client->cl_net,
-						    &stream, gfp_flags);
+			da = nfs4_decode_mp_ds_addr(net, &stream, gfp_flags);
 			if (da)
 				list_add_tail(&da->da_node, &dsaddrs);
 		}
@@ -171,9 +170,10 @@ nfs4_fl_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev,
 			goto out_err_free_deviceid;
 		}
 
-		dsaddr->ds_list[i] = nfs4_pnfs_ds_add(&dsaddrs, gfp_flags);
+		dsaddr->ds_list[i] = nfs4_pnfs_ds_add(net, &dsaddrs, gfp_flags);
 		if (!dsaddr->ds_list[i])
 			goto out_err_drain_dsaddrs;
+		trace_fl_getdevinfo(server, &pdev->dev_id, dsaddr->ds_list[i]->ds_remotestr);
 
 		/* If DS was already in cache, free ds addrs */
 		while (!list_empty(&dsaddrs)) {
@@ -186,7 +186,7 @@ nfs4_fl_alloc_deviceid_node(struct nfs_server *server, struct pnfs_device *pdev,
 		}
 	}
 
-	__free_page(scratch);
+	folio_put(scratch);
 	return dsaddr;
 
 out_err_drain_dsaddrs:
@@ -204,7 +204,7 @@ out_err_free_deviceid:
 out_err_free_stripe_indices:
 	kfree(stripe_indices);
 out_err_free_scratch:
-	__free_page(scratch);
+	folio_put(scratch);
 out_err:
 	dprintk("%s ERROR: returning NULL\n", __func__);
 	return NULL;