Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit e2c63e09 authored by Trond Myklebust's avatar Trond Myklebust
Browse files

Merge branch 'flexfiles' 

* flexfiles: (53 commits)
  pnfs: lookup new lseg at lseg boundary
  nfs41: .init_read and .init_write can be called with valid pg_lseg
  pnfs: Update documentation on the Layout Drivers
  pnfs/flexfiles: Add the FlexFile Layout Driver
  nfs: count DIO good bytes correctly with mirroring
  nfs41: wait for LAYOUTRETURN before retrying LAYOUTGET
  nfs: add a helper to set NFS_ODIRECT_RESCHED_WRITES to direct writes
  nfs41: add NFS_LAYOUT_RETRY_LAYOUTGET to layout header flags
  nfs/flexfiles: send layoutreturn before freeing lseg
  nfs41: introduce NFS_LAYOUT_RETURN_BEFORE_CLOSE
  nfs41: allow async version layoutreturn
  nfs41: add range to layoutreturn args
  pnfs: allow LD to ask to resend read through pnfs
  nfs: add nfs_pgio_current_mirror helper
  nfs: only reset desc->pg_mirror_idx when mirroring is supported
  nfs41: add a debug warning if we destroy an unempty layout
  pnfs: fail comparison when bucket verifier not set
  nfs: mirroring support for direct io
  nfs: add mirroring support to pgio layer
  pnfs: pass ds_commit_idx through the commit path
  ...

Conflicts:
	fs/nfs/pnfs.c
	fs/nfs/pnfs.h
parents cc3ea893 7c13789e

Documentation/filesystems/nfs/pnfs.txt

+7 −6
Original line number Original line Diff line number Diff line
@@ -57,15 +57,16 @@ bit is set, preventing any new lsegs from being added. 
layout drivers

layout drivers

--------------

--------------





PNFS utilizes what is called layout drivers. The STD defines 3 basic

PNFS utilizes what is called layout drivers. The STD defines 4 basic

layout types: "files" "objects" and "blocks". For each of these types

layout types: "files", "objects", "blocks", and "flexfiles". For each

there is a layout-driver with a common function-vectors table which

of these types there is a layout-driver with a common function-vectors

are called by the nfs-client pnfs-core to implement the different layout

table which are called by the nfs-client pnfs-core to implement the

types.

different layout types.





Files-layout-driver code is in: fs/nfs/nfs4filelayout.c && nfs4filelayoutdev.c

Files-layout-driver code is in: fs/nfs/filelayout/.. directory

Objects-layout-deriver code is in: fs/nfs/objlayout/.. directory

Objects-layout-deriver code is in: fs/nfs/objlayout/.. directory

Blocks-layout-deriver code is in: fs/nfs/blocklayout/.. directory

Blocks-layout-deriver code is in: fs/nfs/blocklayout/.. directory

Flexfiles-layout-driver code is in: fs/nfs/flexfilelayout/.. directory





objects-layout setup

objects-layout setup

--------------------

--------------------

fs/nfs/Kconfig

+5 −0
Original line number Original line Diff line number Diff line
@@ -128,6 +128,11 @@ config PNFS_OBJLAYOUT 
	depends on NFS_V4_1 && SCSI_OSD_ULD

	depends on NFS_V4_1 && SCSI_OSD_ULD

	default NFS_V4

	default NFS_V4





config PNFS_FLEXFILE_LAYOUT

	tristate

	depends on NFS_V4_1 && NFS_V3

	default m



config NFS_V4_1_IMPLEMENTATION_ID_DOMAIN

config NFS_V4_1_IMPLEMENTATION_ID_DOMAIN

	string "NFSv4.1 Implementation ID Domain"

	string "NFSv4.1 Implementation ID Domain"

	depends on NFS_V4_1

	depends on NFS_V4_1

fs/nfs/Makefile

+2 −1
Original line number Original line Diff line number Diff line
@@ -27,9 +27,10 @@ nfsv4-y := nfs4proc.o nfs4xdr.o nfs4state.o nfs4renewd.o nfs4super.o nfs4file.o 
	  dns_resolve.o nfs4trace.o

	  dns_resolve.o nfs4trace.o

nfsv4-$(CONFIG_NFS_USE_LEGACY_DNS) += cache_lib.o

nfsv4-$(CONFIG_NFS_USE_LEGACY_DNS) += cache_lib.o

nfsv4-$(CONFIG_SYSCTL)	+= nfs4sysctl.o

nfsv4-$(CONFIG_SYSCTL)	+= nfs4sysctl.o

nfsv4-$(CONFIG_NFS_V4_1)	+= pnfs.o pnfs_dev.o

nfsv4-$(CONFIG_NFS_V4_1)	+= pnfs.o pnfs_dev.o pnfs_nfs.o

nfsv4-$(CONFIG_NFS_V4_2)	+= nfs42proc.o

nfsv4-$(CONFIG_NFS_V4_2)	+= nfs42proc.o





obj-$(CONFIG_PNFS_FILE_LAYOUT) += filelayout/

obj-$(CONFIG_PNFS_FILE_LAYOUT) += filelayout/

obj-$(CONFIG_PNFS_OBJLAYOUT) += objlayout/

obj-$(CONFIG_PNFS_OBJLAYOUT) += objlayout/

obj-$(CONFIG_PNFS_BLOCK) += blocklayout/

obj-$(CONFIG_PNFS_BLOCK) += blocklayout/

obj-$(CONFIG_PNFS_FLEXFILE_LAYOUT) += flexfilelayout/

fs/nfs/blocklayout/blocklayout.c

+2 −0
Original line number Original line Diff line number Diff line
@@ -860,12 +860,14 @@ static const struct nfs_pageio_ops bl_pg_read_ops = { 
	.pg_init = bl_pg_init_read,

	.pg_init = bl_pg_init_read,

	.pg_test = bl_pg_test_read,

	.pg_test = bl_pg_test_read,

	.pg_doio = pnfs_generic_pg_readpages,

	.pg_doio = pnfs_generic_pg_readpages,

	.pg_cleanup = pnfs_generic_pg_cleanup,

};

};





static const struct nfs_pageio_ops bl_pg_write_ops = {

static const struct nfs_pageio_ops bl_pg_write_ops = {

	.pg_init = bl_pg_init_write,

	.pg_init = bl_pg_init_write,

	.pg_test = bl_pg_test_write,

	.pg_test = bl_pg_test_write,

	.pg_doio = pnfs_generic_pg_writepages,

	.pg_doio = pnfs_generic_pg_writepages,

	.pg_cleanup = pnfs_generic_pg_cleanup,

};

};





static struct pnfs_layoutdriver_type blocklayout_type = {

static struct pnfs_layoutdriver_type blocklayout_type = {

fs/nfs/direct.c

+95 −17
Original line number Original line Diff line number Diff line
@@ -66,6 +66,10 @@ static struct kmem_cache *nfs_direct_cachep; 
/*

/*

 * This represents a set of asynchronous requests that we're waiting on

 * This represents a set of asynchronous requests that we're waiting on

 */

 */

struct nfs_direct_mirror {

	ssize_t count;

};



struct nfs_direct_req {

struct nfs_direct_req {

	struct kref		kref;		/* release manager */

	struct kref		kref;		/* release manager */
@@ -78,8 +82,13 @@ struct nfs_direct_req { 
	/* completion state */

	/* completion state */

	atomic_t		io_count;	/* i/os we're waiting for */

	atomic_t		io_count;	/* i/os we're waiting for */

	spinlock_t		lock;		/* protect completion state */

	spinlock_t		lock;		/* protect completion state */



	struct nfs_direct_mirror mirrors[NFS_PAGEIO_DESCRIPTOR_MIRROR_MAX];

	int			mirror_count;



	ssize_t			count,		/* bytes actually processed */

	ssize_t			count,		/* bytes actually processed */

				bytes_left,	/* bytes left to be sent */

				bytes_left,	/* bytes left to be sent */

				io_start,	/* start of IO */

				error;		/* any reported error */

				error;		/* any reported error */

	struct completion	completion;	/* wait for i/o completion */

	struct completion	completion;	/* wait for i/o completion */
@@ -108,26 +117,56 @@ static inline int put_dreq(struct nfs_direct_req *dreq) 
	return atomic_dec_and_test(&dreq->io_count);

	return atomic_dec_and_test(&dreq->io_count);

}

}





void nfs_direct_set_resched_writes(struct nfs_direct_req *dreq)

{

	dreq->flags = NFS_ODIRECT_RESCHED_WRITES;

}

EXPORT_SYMBOL_GPL(nfs_direct_set_resched_writes);



static void

nfs_direct_good_bytes(struct nfs_direct_req *dreq, struct nfs_pgio_header *hdr)

{

	int i;

	ssize_t count;



	WARN_ON_ONCE(hdr->pgio_mirror_idx >= dreq->mirror_count);



	count = dreq->mirrors[hdr->pgio_mirror_idx].count;

	if (count + dreq->io_start < hdr->io_start + hdr->good_bytes) {

		count = hdr->io_start + hdr->good_bytes - dreq->io_start;

		dreq->mirrors[hdr->pgio_mirror_idx].count = count;

	}



	/* update the dreq->count by finding the minimum agreed count from all

	 * mirrors */

	count = dreq->mirrors[0].count;



	for (i = 1; i < dreq->mirror_count; i++)

		count = min(count, dreq->mirrors[i].count);



	dreq->count = count;

}



/*

/*

 * nfs_direct_select_verf - select the right verifier

 * nfs_direct_select_verf - select the right verifier

 * @dreq - direct request possibly spanning multiple servers

 * @dreq - direct request possibly spanning multiple servers

 * @ds_clp - nfs_client of data server or NULL if MDS / non-pnfs

 * @ds_clp - nfs_client of data server or NULL if MDS / non-pnfs

 * @ds_idx - index of data server in data server list, only valid if ds_clp set

 * @commit_idx - commit bucket index for the DS

 *

 *

 * returns the correct verifier to use given the role of the server

 * returns the correct verifier to use given the role of the server

 */

 */

static struct nfs_writeverf *

static struct nfs_writeverf *

nfs_direct_select_verf(struct nfs_direct_req *dreq,

nfs_direct_select_verf(struct nfs_direct_req *dreq,

		       struct nfs_client *ds_clp,

		       struct nfs_client *ds_clp,

		       int ds_idx)

		       int commit_idx)

{

{

	struct nfs_writeverf *verfp = &dreq->verf;

	struct nfs_writeverf *verfp = &dreq->verf;





#ifdef CONFIG_NFS_V4_1

#ifdef CONFIG_NFS_V4_1

	if (ds_clp) {

	if (ds_clp) {

		/* pNFS is in use, use the DS verf */

		/* pNFS is in use, use the DS verf */

		if (ds_idx >= 0 && ds_idx < dreq->ds_cinfo.nbuckets)

		if (commit_idx >= 0 && commit_idx < dreq->ds_cinfo.nbuckets)

			verfp = &dreq->ds_cinfo.buckets[ds_idx].direct_verf;

			verfp = &dreq->ds_cinfo.buckets[commit_idx].direct_verf;

		else

		else

			WARN_ON_ONCE(1);

			WARN_ON_ONCE(1);

	}

	}
@@ -148,8 +187,7 @@ static void nfs_direct_set_hdr_verf(struct nfs_direct_req *dreq, 
{

{

	struct nfs_writeverf *verfp;

	struct nfs_writeverf *verfp;





	verfp = nfs_direct_select_verf(dreq, hdr->ds_clp,

	verfp = nfs_direct_select_verf(dreq, hdr->ds_clp, hdr->ds_commit_idx);

				      hdr->ds_idx);

	WARN_ON_ONCE(verfp->committed >= 0);

	WARN_ON_ONCE(verfp->committed >= 0);

	memcpy(verfp, &hdr->verf, sizeof(struct nfs_writeverf));

	memcpy(verfp, &hdr->verf, sizeof(struct nfs_writeverf));

	WARN_ON_ONCE(verfp->committed < 0);

	WARN_ON_ONCE(verfp->committed < 0);
@@ -169,8 +207,7 @@ static int nfs_direct_set_or_cmp_hdr_verf(struct nfs_direct_req *dreq, 
{

{

	struct nfs_writeverf *verfp;

	struct nfs_writeverf *verfp;





	verfp = nfs_direct_select_verf(dreq, hdr->ds_clp,

	verfp = nfs_direct_select_verf(dreq, hdr->ds_clp, hdr->ds_commit_idx);

					 hdr->ds_idx);

	if (verfp->committed < 0) {

	if (verfp->committed < 0) {

		nfs_direct_set_hdr_verf(dreq, hdr);

		nfs_direct_set_hdr_verf(dreq, hdr);

		return 0;

		return 0;
@@ -193,7 +230,11 @@ static int nfs_direct_cmp_commit_data_verf(struct nfs_direct_req *dreq, 




	verfp = nfs_direct_select_verf(dreq, data->ds_clp,

	verfp = nfs_direct_select_verf(dreq, data->ds_clp,

					 data->ds_commit_index);

					 data->ds_commit_index);

	WARN_ON_ONCE(verfp->committed < 0);



	/* verifier not set so always fail */

	if (verfp->committed < 0)

		return 1;



	return memcmp(verfp, &data->verf, sizeof(struct nfs_writeverf));

	return memcmp(verfp, &data->verf, sizeof(struct nfs_writeverf));

}

}
@@ -249,6 +290,18 @@ void nfs_init_cinfo_from_dreq(struct nfs_commit_info *cinfo, 
	cinfo->completion_ops = &nfs_direct_commit_completion_ops;

	cinfo->completion_ops = &nfs_direct_commit_completion_ops;

}

}





static inline void nfs_direct_setup_mirroring(struct nfs_direct_req *dreq,

					     struct nfs_pageio_descriptor *pgio,

					     struct nfs_page *req)

{

	int mirror_count = 1;



	if (pgio->pg_ops->pg_get_mirror_count)

		mirror_count = pgio->pg_ops->pg_get_mirror_count(pgio, req);



	dreq->mirror_count = mirror_count;

}



static inline struct nfs_direct_req *nfs_direct_req_alloc(void)

static inline struct nfs_direct_req *nfs_direct_req_alloc(void)

{

{

	struct nfs_direct_req *dreq;

	struct nfs_direct_req *dreq;
@@ -263,6 +316,7 @@ static inline struct nfs_direct_req *nfs_direct_req_alloc(void) 
	INIT_LIST_HEAD(&dreq->mds_cinfo.list);

	INIT_LIST_HEAD(&dreq->mds_cinfo.list);

	dreq->verf.committed = NFS_INVALID_STABLE_HOW;	/* not set yet */

	dreq->verf.committed = NFS_INVALID_STABLE_HOW;	/* not set yet */

	INIT_WORK(&dreq->work, nfs_direct_write_schedule_work);

	INIT_WORK(&dreq->work, nfs_direct_write_schedule_work);

	dreq->mirror_count = 1;

	spin_lock_init(&dreq->lock);

	spin_lock_init(&dreq->lock);





	return dreq;

	return dreq;
@@ -369,7 +423,8 @@ static void nfs_direct_read_completion(struct nfs_pgio_header *hdr) 
	if (test_bit(NFS_IOHDR_ERROR, &hdr->flags) && (hdr->good_bytes == 0))

	if (test_bit(NFS_IOHDR_ERROR, &hdr->flags) && (hdr->good_bytes == 0))

		dreq->error = hdr->error;

		dreq->error = hdr->error;

	else

	else

		dreq->count += hdr->good_bytes;

		nfs_direct_good_bytes(dreq, hdr);



	spin_unlock(&dreq->lock);

	spin_unlock(&dreq->lock);





	while (!list_empty(&hdr->pages)) {

	while (!list_empty(&hdr->pages)) {
@@ -547,6 +602,7 @@ ssize_t nfs_file_direct_read(struct kiocb *iocb, struct iov_iter *iter, 




	dreq->inode = inode;

	dreq->inode = inode;

	dreq->bytes_left = count;

	dreq->bytes_left = count;

	dreq->io_start = pos;

	dreq->ctx = get_nfs_open_context(nfs_file_open_context(iocb->ki_filp));

	dreq->ctx = get_nfs_open_context(nfs_file_open_context(iocb->ki_filp));

	l_ctx = nfs_get_lock_context(dreq->ctx);

	l_ctx = nfs_get_lock_context(dreq->ctx);

	if (IS_ERR(l_ctx)) {

	if (IS_ERR(l_ctx)) {
@@ -579,6 +635,20 @@ out: 
	return result;

	return result;

}

}





static void

nfs_direct_write_scan_commit_list(struct inode *inode,

				  struct list_head *list,

				  struct nfs_commit_info *cinfo)

{

	spin_lock(cinfo->lock);

#ifdef CONFIG_NFS_V4_1

	if (cinfo->ds != NULL && cinfo->ds->nwritten != 0)

		NFS_SERVER(inode)->pnfs_curr_ld->recover_commit_reqs(list, cinfo);

#endif

	nfs_scan_commit_list(&cinfo->mds->list, list, cinfo, 0);

	spin_unlock(cinfo->lock);

}



static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)

static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)

{

{

	struct nfs_pageio_descriptor desc;

	struct nfs_pageio_descriptor desc;
@@ -586,20 +656,23 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq) 
	LIST_HEAD(reqs);

	LIST_HEAD(reqs);

	struct nfs_commit_info cinfo;

	struct nfs_commit_info cinfo;

	LIST_HEAD(failed);

	LIST_HEAD(failed);

	int i;





	nfs_init_cinfo_from_dreq(&cinfo, dreq);

	nfs_init_cinfo_from_dreq(&cinfo, dreq);

	pnfs_recover_commit_reqs(dreq->inode, &reqs, &cinfo);

	nfs_direct_write_scan_commit_list(dreq->inode, &reqs, &cinfo);

	spin_lock(cinfo.lock);

	nfs_scan_commit_list(&cinfo.mds->list, &reqs, &cinfo, 0);

	spin_unlock(cinfo.lock);





	dreq->count = 0;

	dreq->count = 0;

	for (i = 0; i < dreq->mirror_count; i++)

		dreq->mirrors[i].count = 0;

	get_dreq(dreq);

	get_dreq(dreq);





	nfs_pageio_init_write(&desc, dreq->inode, FLUSH_STABLE, false,

	nfs_pageio_init_write(&desc, dreq->inode, FLUSH_STABLE, false,

			      &nfs_direct_write_completion_ops);

			      &nfs_direct_write_completion_ops);

	desc.pg_dreq = dreq;

	desc.pg_dreq = dreq;





	req = nfs_list_entry(reqs.next);

	nfs_direct_setup_mirroring(dreq, &desc, req);



	list_for_each_entry_safe(req, tmp, &reqs, wb_list) {

	list_for_each_entry_safe(req, tmp, &reqs, wb_list) {

		if (!nfs_pageio_add_request(&desc, req)) {

		if (!nfs_pageio_add_request(&desc, req)) {

			nfs_list_remove_request(req);

			nfs_list_remove_request(req);
@@ -646,7 +719,7 @@ static void nfs_direct_commit_complete(struct nfs_commit_data *data) 
		nfs_list_remove_request(req);

		nfs_list_remove_request(req);

		if (dreq->flags == NFS_ODIRECT_RESCHED_WRITES) {

		if (dreq->flags == NFS_ODIRECT_RESCHED_WRITES) {

			/* Note the rewrite will go through mds */

			/* Note the rewrite will go through mds */

			nfs_mark_request_commit(req, NULL, &cinfo);

			nfs_mark_request_commit(req, NULL, &cinfo, 0);

		} else

		} else

			nfs_release_request(req);

			nfs_release_request(req);

		nfs_unlock_and_release_request(req);

		nfs_unlock_and_release_request(req);
@@ -721,7 +794,7 @@ static void nfs_direct_write_completion(struct nfs_pgio_header *hdr) 
		dreq->error = hdr->error;

		dreq->error = hdr->error;

	}

	}

	if (dreq->error == 0) {

	if (dreq->error == 0) {

		dreq->count += hdr->good_bytes;

		nfs_direct_good_bytes(dreq, hdr);

		if (nfs_write_need_commit(hdr)) {

		if (nfs_write_need_commit(hdr)) {

			if (dreq->flags == NFS_ODIRECT_RESCHED_WRITES)

			if (dreq->flags == NFS_ODIRECT_RESCHED_WRITES)

				request_commit = true;

				request_commit = true;
@@ -745,7 +818,8 @@ static void nfs_direct_write_completion(struct nfs_pgio_header *hdr) 
		nfs_list_remove_request(req);

		nfs_list_remove_request(req);

		if (request_commit) {

		if (request_commit) {

			kref_get(&req->wb_kref);

			kref_get(&req->wb_kref);

			nfs_mark_request_commit(req, hdr->lseg, &cinfo);

			nfs_mark_request_commit(req, hdr->lseg, &cinfo,

				hdr->ds_commit_idx);

		}

		}

		nfs_unlock_and_release_request(req);

		nfs_unlock_and_release_request(req);

	}

	}
@@ -826,6 +900,9 @@ static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq, 
				result = PTR_ERR(req);

				result = PTR_ERR(req);

				break;

				break;

			}

			}



			nfs_direct_setup_mirroring(dreq, &desc, req);



			nfs_lock_request(req);

			nfs_lock_request(req);

			req->wb_index = pos >> PAGE_SHIFT;

			req->wb_index = pos >> PAGE_SHIFT;

			req->wb_offset = pos & ~PAGE_MASK;

			req->wb_offset = pos & ~PAGE_MASK;
@@ -934,6 +1011,7 @@ ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter, 




	dreq->inode = inode;

	dreq->inode = inode;

	dreq->bytes_left = count;

	dreq->bytes_left = count;

	dreq->io_start = pos;

	dreq->ctx = get_nfs_open_context(nfs_file_open_context(iocb->ki_filp));

	dreq->ctx = get_nfs_open_context(nfs_file_open_context(iocb->ki_filp));

	l_ctx = nfs_get_lock_context(dreq->ctx);

	l_ctx = nfs_get_lock_context(dreq->ctx);

	if (IS_ERR(l_ctx)) {

	if (IS_ERR(l_ctx)) {