aboutsummaryrefslogtreecommitdiff
path: root/fs/nfs/objlayout
diff options
context:
space:
mode:
Diffstat (limited to 'fs/nfs/objlayout')
-rw-r--r--fs/nfs/objlayout/objio_osd.c114
-rw-r--r--fs/nfs/objlayout/objlayout.c39
-rw-r--r--fs/nfs/objlayout/objlayout.h10
3 files changed, 117 insertions, 46 deletions
diff --git a/fs/nfs/objlayout/objio_osd.c b/fs/nfs/objlayout/objio_osd.c
index b47277baeba..611320753db 100644
--- a/fs/nfs/objlayout/objio_osd.c
+++ b/fs/nfs/objlayout/objio_osd.c
@@ -41,6 +41,7 @@
#include <scsi/osd_ore.h>
#include "objlayout.h"
+#include "../internal.h"
#define NFSDBG_FACILITY NFSDBG_PNFS_LD
@@ -233,7 +234,7 @@ static int __alloc_objio_seg(unsigned numdevs, gfp_t gfp_flags,
lseg = kzalloc(lseg_size, gfp_flags);
if (unlikely(!lseg)) {
- dprintk("%s: Faild allocation numdevs=%d size=%zd\n", __func__,
+ dprintk("%s: Failed allocation numdevs=%d size=%zd\n", __func__,
numdevs, lseg_size);
return -ENOMEM;
}
@@ -368,7 +369,7 @@ void objio_free_result(struct objlayout_io_res *oir)
kfree(objios);
}
-enum pnfs_osd_errno osd_pri_2_pnfs_err(enum osd_err_priority oep)
+static enum pnfs_osd_errno osd_pri_2_pnfs_err(enum osd_err_priority oep)
{
switch (oep) {
case OSD_ERR_PRI_NO_ERROR:
@@ -438,7 +439,7 @@ static void _read_done(struct ore_io_state *ios, void *private)
objlayout_read_done(&objios->oir, status, objios->sync);
}
-int objio_read_pagelist(struct nfs_read_data *rdata)
+int objio_read_pagelist(struct nfs_pgio_data *rdata)
{
struct nfs_pgio_header *hdr = rdata->header;
struct objio_state *objios;
@@ -454,7 +455,10 @@ int objio_read_pagelist(struct nfs_read_data *rdata)
objios->ios->done = _read_done;
dprintk("%s: offset=0x%llx length=0x%x\n", __func__,
rdata->args.offset, rdata->args.count);
- return ore_read(objios->ios);
+ ret = ore_read(objios->ios);
+ if (unlikely(ret))
+ objio_free_result(&objios->oir);
+ return ret;
}
/*
@@ -483,11 +487,19 @@ static void _write_done(struct ore_io_state *ios, void *private)
static struct page *__r4w_get_page(void *priv, u64 offset, bool *uptodate)
{
struct objio_state *objios = priv;
- struct nfs_write_data *wdata = objios->oir.rpcdata;
+ struct nfs_pgio_data *wdata = objios->oir.rpcdata;
struct address_space *mapping = wdata->header->inode->i_mapping;
pgoff_t index = offset / PAGE_SIZE;
- struct page *page = find_get_page(mapping, index);
+ struct page *page;
+ loff_t i_size = i_size_read(wdata->header->inode);
+
+ if (offset >= i_size) {
+ *uptodate = true;
+ dprintk("%s: g_zero_page index=0x%lx\n", __func__, index);
+ return ZERO_PAGE(0);
+ }
+ page = find_get_page(mapping, index);
if (!page) {
page = find_or_create_page(mapping, index, GFP_NOFS);
if (unlikely(!page)) {
@@ -507,8 +519,10 @@ static struct page *__r4w_get_page(void *priv, u64 offset, bool *uptodate)
static void __r4w_put_page(void *priv, struct page *page)
{
- dprintk("%s: index=0x%lx\n", __func__, page->index);
- page_cache_release(page);
+ dprintk("%s: index=0x%lx\n", __func__,
+ (page == ZERO_PAGE(0)) ? -1UL : page->index);
+ if (ZERO_PAGE(0) != page)
+ page_cache_release(page);
return;
}
@@ -517,7 +531,7 @@ static const struct _ore_r4w_op _r4w_op = {
.put_page = &__r4w_put_page,
};
-int objio_write_pagelist(struct nfs_write_data *wdata, int how)
+int objio_write_pagelist(struct nfs_pgio_data *wdata, int how)
{
struct nfs_pgio_header *hdr = wdata->header;
struct objio_state *objios;
@@ -539,8 +553,10 @@ int objio_write_pagelist(struct nfs_write_data *wdata, int how)
dprintk("%s: offset=0x%llx length=0x%x\n", __func__,
wdata->args.offset, wdata->args.count);
ret = ore_write(objios->ios);
- if (unlikely(ret))
+ if (unlikely(ret)) {
+ objio_free_result(&objios->oir);
return ret;
+ }
if (objios->sync)
_write_done(objios->ios, objios);
@@ -548,24 +564,87 @@ int objio_write_pagelist(struct nfs_write_data *wdata, int how)
return 0;
}
-static bool objio_pg_test(struct nfs_pageio_descriptor *pgio,
+/*
+ * Return 0 if @req cannot be coalesced into @pgio, otherwise return the number
+ * of bytes (maximum @req->wb_bytes) that can be coalesced.
+ */
+static size_t objio_pg_test(struct nfs_pageio_descriptor *pgio,
struct nfs_page *prev, struct nfs_page *req)
{
- if (!pnfs_generic_pg_test(pgio, prev, req))
- return false;
+ unsigned int size;
+
+ size = pnfs_generic_pg_test(pgio, prev, req);
+
+ if (!size || pgio->pg_count + req->wb_bytes >
+ (unsigned long)pgio->pg_layout_private)
+ return 0;
+
+ return min(size, req->wb_bytes);
+}
+
+static void objio_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req)
+{
+ pnfs_generic_pg_init_read(pgio, req);
+ if (unlikely(pgio->pg_lseg == NULL))
+ return; /* Not pNFS */
- return pgio->pg_count + req->wb_bytes <=
- OBJIO_LSEG(pgio->pg_lseg)->layout.max_io_length;
+ pgio->pg_layout_private = (void *)
+ OBJIO_LSEG(pgio->pg_lseg)->layout.max_io_length;
+}
+
+static bool aligned_on_raid_stripe(u64 offset, struct ore_layout *layout,
+ unsigned long *stripe_end)
+{
+ u32 stripe_off;
+ unsigned stripe_size;
+
+ if (layout->raid_algorithm == PNFS_OSD_RAID_0)
+ return true;
+
+ stripe_size = layout->stripe_unit *
+ (layout->group_width - layout->parity);
+
+ div_u64_rem(offset, stripe_size, &stripe_off);
+ if (!stripe_off)
+ return true;
+
+ *stripe_end = stripe_size - stripe_off;
+ return false;
+}
+
+static void objio_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req)
+{
+ unsigned long stripe_end = 0;
+ u64 wb_size;
+
+ if (pgio->pg_dreq == NULL)
+ wb_size = i_size_read(pgio->pg_inode) - req_offset(req);
+ else
+ wb_size = nfs_dreq_bytes_left(pgio->pg_dreq);
+
+ pnfs_generic_pg_init_write(pgio, req, wb_size);
+ if (unlikely(pgio->pg_lseg == NULL))
+ return; /* Not pNFS */
+
+ if (req->wb_offset ||
+ !aligned_on_raid_stripe(req->wb_index * PAGE_SIZE,
+ &OBJIO_LSEG(pgio->pg_lseg)->layout,
+ &stripe_end)) {
+ pgio->pg_layout_private = (void *)stripe_end;
+ } else {
+ pgio->pg_layout_private = (void *)
+ OBJIO_LSEG(pgio->pg_lseg)->layout.max_io_length;
+ }
}
static const struct nfs_pageio_ops objio_pg_read_ops = {
- .pg_init = pnfs_generic_pg_init_read,
+ .pg_init = objio_init_read,
.pg_test = objio_pg_test,
.pg_doio = pnfs_generic_pg_readpages,
};
static const struct nfs_pageio_ops objio_pg_write_ops = {
- .pg_init = pnfs_generic_pg_init_write,
+ .pg_init = objio_init_write,
.pg_test = objio_pg_test,
.pg_doio = pnfs_generic_pg_writepages,
};
@@ -576,6 +655,7 @@ static struct pnfs_layoutdriver_type objlayout_type = {
.flags = PNFS_LAYOUTRET_ON_SETATTR |
PNFS_LAYOUTRET_ON_ERROR,
+ .owner = THIS_MODULE,
.alloc_layout_hdr = objlayout_alloc_layout_hdr,
.free_layout_hdr = objlayout_free_layout_hdr,
diff --git a/fs/nfs/objlayout/objlayout.c b/fs/nfs/objlayout/objlayout.c
index 87461354530..765d3f54e98 100644
--- a/fs/nfs/objlayout/objlayout.c
+++ b/fs/nfs/objlayout/objlayout.c
@@ -53,10 +53,10 @@ objlayout_alloc_layout_hdr(struct inode *inode, gfp_t gfp_flags)
struct objlayout *objlay;
objlay = kzalloc(sizeof(struct objlayout), gfp_flags);
- if (objlay) {
- spin_lock_init(&objlay->lock);
- INIT_LIST_HEAD(&objlay->err_list);
- }
+ if (!objlay)
+ return NULL;
+ spin_lock_init(&objlay->lock);
+ INIT_LIST_HEAD(&objlay->err_list);
dprintk("%s: Return %p\n", __func__, objlay);
return &objlay->pnfs_layout;
}
@@ -148,17 +148,6 @@ end_offset(u64 start, u64 len)
return end >= start ? end : NFS4_MAX_UINT64;
}
-/* last octet in a range */
-static inline u64
-last_byte_offset(u64 start, u64 len)
-{
- u64 end;
-
- BUG_ON(!len);
- end = start + len;
- return end > start ? end - 1 : NFS4_MAX_UINT64;
-}
-
static void _fix_verify_io_params(struct pnfs_layout_segment *lseg,
struct page ***p_pages, unsigned *p_pgbase,
u64 offset, unsigned long count)
@@ -240,11 +229,11 @@ objlayout_io_set_result(struct objlayout_io_res *oir, unsigned index,
static void _rpc_read_complete(struct work_struct *work)
{
struct rpc_task *task;
- struct nfs_read_data *rdata;
+ struct nfs_pgio_data *rdata;
dprintk("%s enter\n", __func__);
task = container_of(work, struct rpc_task, u.tk_work);
- rdata = container_of(task, struct nfs_read_data, task);
+ rdata = container_of(task, struct nfs_pgio_data, task);
pnfs_ld_read_done(rdata);
}
@@ -252,7 +241,7 @@ static void _rpc_read_complete(struct work_struct *work)
void
objlayout_read_done(struct objlayout_io_res *oir, ssize_t status, bool sync)
{
- struct nfs_read_data *rdata = oir->rpcdata;
+ struct nfs_pgio_data *rdata = oir->rpcdata;
oir->status = rdata->task.tk_status = status;
if (status >= 0)
@@ -277,7 +266,7 @@ objlayout_read_done(struct objlayout_io_res *oir, ssize_t status, bool sync)
* Perform sync or async reads.
*/
enum pnfs_try_status
-objlayout_read_pagelist(struct nfs_read_data *rdata)
+objlayout_read_pagelist(struct nfs_pgio_data *rdata)
{
struct nfs_pgio_header *hdr = rdata->header;
struct inode *inode = hdr->inode;
@@ -323,11 +312,11 @@ objlayout_read_pagelist(struct nfs_read_data *rdata)
static void _rpc_write_complete(struct work_struct *work)
{
struct rpc_task *task;
- struct nfs_write_data *wdata;
+ struct nfs_pgio_data *wdata;
dprintk("%s enter\n", __func__);
task = container_of(work, struct rpc_task, u.tk_work);
- wdata = container_of(task, struct nfs_write_data, task);
+ wdata = container_of(task, struct nfs_pgio_data, task);
pnfs_ld_write_done(wdata);
}
@@ -335,7 +324,7 @@ static void _rpc_write_complete(struct work_struct *work)
void
objlayout_write_done(struct objlayout_io_res *oir, ssize_t status, bool sync)
{
- struct nfs_write_data *wdata = oir->rpcdata;
+ struct nfs_pgio_data *wdata = oir->rpcdata;
oir->status = wdata->task.tk_status = status;
if (status >= 0) {
@@ -362,7 +351,7 @@ objlayout_write_done(struct objlayout_io_res *oir, ssize_t status, bool sync)
* Perform sync or async writes.
*/
enum pnfs_try_status
-objlayout_write_pagelist(struct nfs_write_data *wdata,
+objlayout_write_pagelist(struct nfs_pgio_data *wdata,
int how)
{
struct nfs_pgio_header *hdr = wdata->header;
@@ -624,8 +613,10 @@ int objlayout_get_deviceinfo(struct pnfs_layout_hdr *pnfslay,
pd.pgbase = 0;
pd.pglen = PAGE_SIZE;
pd.mincount = 0;
+ pd.maxcount = PAGE_SIZE;
- err = nfs4_proc_getdeviceinfo(NFS_SERVER(pnfslay->plh_inode), &pd);
+ err = nfs4_proc_getdeviceinfo(NFS_SERVER(pnfslay->plh_inode), &pd,
+ pnfslay->plh_lc_cred);
dprintk("%s nfs_getdeviceinfo returned %d\n", __func__, err);
if (err)
goto err_out;
diff --git a/fs/nfs/objlayout/objlayout.h b/fs/nfs/objlayout/objlayout.h
index 880ba086be9..01e041029a6 100644
--- a/fs/nfs/objlayout/objlayout.h
+++ b/fs/nfs/objlayout/objlayout.h
@@ -114,13 +114,13 @@ extern int objio_alloc_lseg(struct pnfs_layout_segment **outp,
gfp_t gfp_flags);
extern void objio_free_lseg(struct pnfs_layout_segment *lseg);
-/* objio_free_result will free these @oir structs recieved from
+/* objio_free_result will free these @oir structs received from
* objlayout_{read,write}_done
*/
extern void objio_free_result(struct objlayout_io_res *oir);
-extern int objio_read_pagelist(struct nfs_read_data *rdata);
-extern int objio_write_pagelist(struct nfs_write_data *wdata, int how);
+extern int objio_read_pagelist(struct nfs_pgio_data *rdata);
+extern int objio_write_pagelist(struct nfs_pgio_data *wdata, int how);
/*
* callback API
@@ -168,10 +168,10 @@ extern struct pnfs_layout_segment *objlayout_alloc_lseg(
extern void objlayout_free_lseg(struct pnfs_layout_segment *);
extern enum pnfs_try_status objlayout_read_pagelist(
- struct nfs_read_data *);
+ struct nfs_pgio_data *);
extern enum pnfs_try_status objlayout_write_pagelist(
- struct nfs_write_data *,
+ struct nfs_pgio_data *,
int how);
extern void objlayout_encode_layoutcommit(
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-11-10 21:10:54 +0000