提交 799ba8d5 编写于 作者: F Fred Isaman 提交者: Trond Myklebust

NFS4.1: Add lseg to struct nfs4_fl_commit_bucket

Also create a commit_info structure to hold the bucket array and push
it up from the lseg to the layout where it really belongs.

While we are at it, fix a refcounting bug due to an (incorrect)
implicit assumption that filelayout_scan_ds_commit_list always
completely emptied the src list.

This clarifies refcounting, removes the ugly find_only_write_lseg
functions, and pushes the file layout commit code along on the path to
supporting multiple lsegs.
Signed-off-by: NFred Isaman <iisaman@netapp.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>
上级 1acbbb4e
...@@ -650,10 +650,66 @@ filelayout_free_lseg(struct pnfs_layout_segment *lseg) ...@@ -650,10 +650,66 @@ filelayout_free_lseg(struct pnfs_layout_segment *lseg)
dprintk("--> %s\n", __func__); dprintk("--> %s\n", __func__);
nfs4_fl_put_deviceid(fl->dsaddr); nfs4_fl_put_deviceid(fl->dsaddr);
kfree(fl->commit_buckets); /* This assumes a single RW lseg */
if (lseg->pls_range.iomode == IOMODE_RW) {
struct nfs4_filelayout *flo;
flo = FILELAYOUT_FROM_HDR(lseg->pls_layout);
flo->commit_info.nbuckets = 0;
kfree(flo->commit_info.buckets);
flo->commit_info.buckets = NULL;
}
_filelayout_free_lseg(fl); _filelayout_free_lseg(fl);
} }
static int
filelayout_alloc_commit_info(struct pnfs_layout_segment *lseg,
gfp_t gfp_flags)
{
struct nfs4_filelayout_segment *fl = FILELAYOUT_LSEG(lseg);
struct nfs4_filelayout *flo = FILELAYOUT_FROM_HDR(lseg->pls_layout);
struct nfs4_fl_commit_bucket *buckets;
int size;
if (fl->commit_through_mds)
return 0;
if (flo->commit_info.nbuckets != 0) {
/* This assumes there is only one IOMODE_RW lseg. What
* we really want to do is have a layout_hdr level
* dictionary of <multipath_list4, fh> keys, each
* associated with a struct list_head, populated by calls
* to filelayout_write_pagelist().
* */
return 0;
}
size = (fl->stripe_type == STRIPE_SPARSE) ?
fl->dsaddr->ds_num : fl->dsaddr->stripe_count;
buckets = kcalloc(size, sizeof(struct nfs4_fl_commit_bucket),
gfp_flags);
if (!buckets)
return -ENOMEM;
else {
int i;
spin_lock(&lseg->pls_layout->plh_inode->i_lock);
if (flo->commit_info.nbuckets != 0)
kfree(buckets);
else {
flo->commit_info.buckets = buckets;
flo->commit_info.nbuckets = size;
for (i = 0; i < size; i++) {
INIT_LIST_HEAD(&buckets[i].written);
INIT_LIST_HEAD(&buckets[i].committing);
}
}
spin_unlock(&lseg->pls_layout->plh_inode->i_lock);
return 0;
}
}
static struct pnfs_layout_segment * static struct pnfs_layout_segment *
filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid, filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid,
struct nfs4_layoutget_res *lgr, struct nfs4_layoutget_res *lgr,
...@@ -673,29 +729,6 @@ filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid, ...@@ -673,29 +729,6 @@ filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid,
_filelayout_free_lseg(fl); _filelayout_free_lseg(fl);
return NULL; return NULL;
} }
/* This assumes there is only one IOMODE_RW lseg. What
* we really want to do is have a layout_hdr level
* dictionary of <multipath_list4, fh> keys, each
* associated with a struct list_head, populated by calls
* to filelayout_write_pagelist().
* */
if ((!fl->commit_through_mds) && (lgr->range.iomode == IOMODE_RW)) {
int i;
int size = (fl->stripe_type == STRIPE_SPARSE) ?
fl->dsaddr->ds_num : fl->dsaddr->stripe_count;
fl->commit_buckets = kcalloc(size, sizeof(struct nfs4_fl_commit_bucket), gfp_flags);
if (!fl->commit_buckets) {
filelayout_free_lseg(&fl->generic_hdr);
return NULL;
}
fl->number_of_buckets = size;
for (i = 0; i < size; i++) {
INIT_LIST_HEAD(&fl->commit_buckets[i].written);
INIT_LIST_HEAD(&fl->commit_buckets[i].committing);
}
}
return &fl->generic_hdr; return &fl->generic_hdr;
} }
...@@ -747,6 +780,8 @@ static void ...@@ -747,6 +780,8 @@ static void
filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio, filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio,
struct nfs_page *req) struct nfs_page *req)
{ {
int status;
BUG_ON(pgio->pg_lseg != NULL); BUG_ON(pgio->pg_lseg != NULL);
pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
...@@ -757,7 +792,16 @@ filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio, ...@@ -757,7 +792,16 @@ filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio,
GFP_NOFS); GFP_NOFS);
/* If no lseg, fall back to write through mds */ /* If no lseg, fall back to write through mds */
if (pgio->pg_lseg == NULL) if (pgio->pg_lseg == NULL)
nfs_pageio_reset_write_mds(pgio); goto out_mds;
status = filelayout_alloc_commit_info(pgio->pg_lseg, GFP_NOFS);
if (status < 0) {
put_lseg(pgio->pg_lseg);
pgio->pg_lseg = NULL;
goto out_mds;
}
return;
out_mds:
nfs_pageio_reset_write_mds(pgio);
} }
static const struct nfs_pageio_ops filelayout_pg_read_ops = { static const struct nfs_pageio_ops filelayout_pg_read_ops = {
...@@ -793,17 +837,13 @@ filelayout_clear_request_commit(struct nfs_page *req) ...@@ -793,17 +837,13 @@ filelayout_clear_request_commit(struct nfs_page *req)
if (!test_and_clear_bit(PG_COMMIT_TO_DS, &req->wb_flags)) if (!test_and_clear_bit(PG_COMMIT_TO_DS, &req->wb_flags))
goto out; goto out;
if (list_is_singular(&req->wb_list)) { if (list_is_singular(&req->wb_list)) {
struct pnfs_layout_segment *lseg; struct nfs4_fl_commit_bucket *bucket;
/* From here we can find the bucket, but for the moment, bucket = list_first_entry(&req->wb_list,
* since there is only one relevant lseg... struct nfs4_fl_commit_bucket,
*/ written);
list_for_each_entry(lseg, &NFS_I(inode)->layout->plh_segs, pls_list) { freeme = bucket->wlseg;
if (lseg->pls_range.iomode == IOMODE_RW) { bucket->wlseg = NULL;
freeme = lseg;
break;
}
}
} }
out: out:
nfs_request_remove_commit_list(req); nfs_request_remove_commit_list(req);
...@@ -818,6 +858,7 @@ filelayout_choose_commit_list(struct nfs_page *req, ...@@ -818,6 +858,7 @@ filelayout_choose_commit_list(struct nfs_page *req,
struct nfs4_filelayout_segment *fl = FILELAYOUT_LSEG(lseg); struct nfs4_filelayout_segment *fl = FILELAYOUT_LSEG(lseg);
u32 i, j; u32 i, j;
struct list_head *list; struct list_head *list;
struct nfs4_fl_commit_bucket *buckets;
if (fl->commit_through_mds) if (fl->commit_through_mds)
return &NFS_I(req->wb_context->dentry->d_inode)->commit_list; return &NFS_I(req->wb_context->dentry->d_inode)->commit_list;
...@@ -831,15 +872,16 @@ filelayout_choose_commit_list(struct nfs_page *req, ...@@ -831,15 +872,16 @@ filelayout_choose_commit_list(struct nfs_page *req,
j = nfs4_fl_calc_j_index(lseg, j = nfs4_fl_calc_j_index(lseg,
(loff_t)req->wb_index << PAGE_CACHE_SHIFT); (loff_t)req->wb_index << PAGE_CACHE_SHIFT);
i = select_bucket_index(fl, j); i = select_bucket_index(fl, j);
list = &fl->commit_buckets[i].written; buckets = FILELAYOUT_FROM_HDR(lseg->pls_layout)->commit_info.buckets;
list = &buckets[i].written;
if (list_empty(list)) { if (list_empty(list)) {
/* Non-empty buckets hold a reference on the lseg. That ref /* Non-empty buckets hold a reference on the lseg. That ref
* is normally transferred to the COMMIT call and released * is normally transferred to the COMMIT call and released
* there. It could also be released if the last req is pulled * there. It could also be released if the last req is pulled
* off due to a rewrite, in which case it will be done in * off due to a rewrite, in which case it will be done in
* filelayout_remove_commit_req * filelayout_clear_request_commit
*/ */
get_lseg(lseg); buckets[i].wlseg = get_lseg(lseg);
} }
set_bit(PG_COMMIT_TO_DS, &req->wb_flags); set_bit(PG_COMMIT_TO_DS, &req->wb_flags);
return list; return list;
...@@ -908,32 +950,6 @@ static int filelayout_initiate_commit(struct nfs_write_data *data, int how) ...@@ -908,32 +950,6 @@ static int filelayout_initiate_commit(struct nfs_write_data *data, int how)
&filelayout_commit_call_ops, how); &filelayout_commit_call_ops, how);
} }
/*
* This is only useful while we are using whole file layouts.
*/
static struct pnfs_layout_segment *
find_only_write_lseg_locked(struct inode *inode)
{
struct pnfs_layout_segment *lseg;
list_for_each_entry(lseg, &NFS_I(inode)->layout->plh_segs, pls_list)
if (lseg->pls_range.iomode == IOMODE_RW)
return lseg;
return NULL;
}
static struct pnfs_layout_segment *find_only_write_lseg(struct inode *inode)
{
struct pnfs_layout_segment *rv;
spin_lock(&inode->i_lock);
rv = find_only_write_lseg_locked(inode);
if (rv)
get_lseg(rv);
spin_unlock(&inode->i_lock);
return rv;
}
static int static int
filelayout_scan_ds_commit_list(struct nfs4_fl_commit_bucket *bucket, int max, filelayout_scan_ds_commit_list(struct nfs4_fl_commit_bucket *bucket, int max,
spinlock_t *lock) spinlock_t *lock)
...@@ -955,6 +971,13 @@ filelayout_scan_ds_commit_list(struct nfs4_fl_commit_bucket *bucket, int max, ...@@ -955,6 +971,13 @@ filelayout_scan_ds_commit_list(struct nfs4_fl_commit_bucket *bucket, int max,
if (ret == max) if (ret == max)
break; break;
} }
if (ret) {
bucket->clseg = bucket->wlseg;
if (list_empty(src))
bucket->wlseg = NULL;
else
get_lseg(bucket->clseg);
}
return ret; return ret;
} }
...@@ -964,18 +987,14 @@ filelayout_scan_ds_commit_list(struct nfs4_fl_commit_bucket *bucket, int max, ...@@ -964,18 +987,14 @@ filelayout_scan_ds_commit_list(struct nfs4_fl_commit_bucket *bucket, int max,
static int filelayout_scan_commit_lists(struct inode *inode, int max, static int filelayout_scan_commit_lists(struct inode *inode, int max,
spinlock_t *lock) spinlock_t *lock)
{ {
struct pnfs_layout_segment *lseg; struct nfs4_fl_commit_info *fl_cinfo;
struct nfs4_filelayout_segment *fl;
int i, rv = 0, cnt; int i, rv = 0, cnt;
lseg = find_only_write_lseg_locked(inode); fl_cinfo = &FILELAYOUT_FROM_HDR(NFS_I(inode)->layout)->commit_info;
if (!lseg) if (fl_cinfo->nbuckets == 0)
goto out_done;
fl = FILELAYOUT_LSEG(lseg);
if (fl->commit_through_mds)
goto out_done; goto out_done;
for (i = 0; i < fl->number_of_buckets && max != 0; i++) { for (i = 0; i < fl_cinfo->nbuckets && max != 0; i++) {
cnt = filelayout_scan_ds_commit_list(&fl->commit_buckets[i], cnt = filelayout_scan_ds_commit_list(&fl_cinfo->buckets[i],
max, lock); max, lock);
max -= cnt; max -= cnt;
rv += cnt; rv += cnt;
...@@ -987,38 +1006,35 @@ static int filelayout_scan_commit_lists(struct inode *inode, int max, ...@@ -987,38 +1006,35 @@ static int filelayout_scan_commit_lists(struct inode *inode, int max,
static unsigned int static unsigned int
alloc_ds_commits(struct inode *inode, struct list_head *list) alloc_ds_commits(struct inode *inode, struct list_head *list)
{ {
struct pnfs_layout_segment *lseg; struct nfs4_fl_commit_info *fl_cinfo;
struct nfs4_filelayout_segment *fl; struct nfs4_fl_commit_bucket *bucket;
struct nfs_write_data *data; struct nfs_write_data *data;
int i, j; int i, j;
unsigned int nreq = 0; unsigned int nreq = 0;
/* Won't need this when non-whole file layout segments are supported fl_cinfo = &FILELAYOUT_FROM_HDR(NFS_I(inode)->layout)->commit_info;
* instead we will use a pnfs_layout_hdr structure */ bucket = fl_cinfo->buckets;
lseg = find_only_write_lseg(inode); for (i = 0; i < fl_cinfo->nbuckets; i++, bucket++) {
if (!lseg) if (list_empty(&bucket->committing))
return 0;
fl = FILELAYOUT_LSEG(lseg);
for (i = 0; i < fl->number_of_buckets; i++) {
if (list_empty(&fl->commit_buckets[i].committing))
continue; continue;
data = nfs_commitdata_alloc(); data = nfs_commitdata_alloc();
if (!data) if (!data)
break; break;
data->ds_commit_index = i; data->ds_commit_index = i;
data->lseg = lseg; data->lseg = bucket->clseg;
bucket->clseg = NULL;
list_add(&data->pages, list); list_add(&data->pages, list);
nreq++; nreq++;
} }
/* Clean up on error */ /* Clean up on error */
for (j = i; j < fl->number_of_buckets; j++) { for (j = i; j < fl_cinfo->nbuckets; j++, bucket++) {
if (list_empty(&fl->commit_buckets[i].committing)) if (list_empty(&bucket->committing))
continue; continue;
nfs_retry_commit(&fl->commit_buckets[i].committing, lseg); nfs_retry_commit(&bucket->committing, bucket->clseg);
put_lseg(lseg); /* associated with emptying bucket */ put_lseg(bucket->clseg);
bucket->clseg = NULL;
} }
put_lseg(lseg);
/* Caller will clean up entries put on list */ /* Caller will clean up entries put on list */
return nreq; return nreq;
} }
...@@ -1058,7 +1074,10 @@ filelayout_commit_pagelist(struct inode *inode, struct list_head *mds_pages, ...@@ -1058,7 +1074,10 @@ filelayout_commit_pagelist(struct inode *inode, struct list_head *mds_pages,
nfs_initiate_commit(data, NFS_CLIENT(inode), nfs_initiate_commit(data, NFS_CLIENT(inode),
data->mds_ops, how); data->mds_ops, how);
} else { } else {
nfs_init_commit(data, &FILELAYOUT_LSEG(data->lseg)->commit_buckets[data->ds_commit_index].committing, data->lseg); struct nfs4_fl_commit_info *fl_cinfo;
fl_cinfo = &FILELAYOUT_FROM_HDR(data->lseg->pls_layout)->commit_info;
nfs_init_commit(data, &fl_cinfo->buckets[data->ds_commit_index].committing, data->lseg);
filelayout_initiate_commit(data, how); filelayout_initiate_commit(data, how);
} }
} }
...@@ -1072,10 +1091,27 @@ filelayout_free_deveiceid_node(struct nfs4_deviceid_node *d) ...@@ -1072,10 +1091,27 @@ filelayout_free_deveiceid_node(struct nfs4_deviceid_node *d)
nfs4_fl_free_deviceid(container_of(d, struct nfs4_file_layout_dsaddr, id_node)); nfs4_fl_free_deviceid(container_of(d, struct nfs4_file_layout_dsaddr, id_node));
} }
static struct pnfs_layout_hdr *
filelayout_alloc_layout_hdr(struct inode *inode, gfp_t gfp_flags)
{
struct nfs4_filelayout *flo;
flo = kzalloc(sizeof(*flo), gfp_flags);
return &flo->generic_hdr;
}
static void
filelayout_free_layout_hdr(struct pnfs_layout_hdr *lo)
{
kfree(FILELAYOUT_FROM_HDR(lo));
}
static struct pnfs_layoutdriver_type filelayout_type = { static struct pnfs_layoutdriver_type filelayout_type = {
.id = LAYOUT_NFSV4_1_FILES, .id = LAYOUT_NFSV4_1_FILES,
.name = "LAYOUT_NFSV4_1_FILES", .name = "LAYOUT_NFSV4_1_FILES",
.owner = THIS_MODULE, .owner = THIS_MODULE,
.alloc_layout_hdr = filelayout_alloc_layout_hdr,
.free_layout_hdr = filelayout_free_layout_hdr,
.alloc_lseg = filelayout_alloc_lseg, .alloc_lseg = filelayout_alloc_lseg,
.free_lseg = filelayout_free_lseg, .free_lseg = filelayout_free_lseg,
.pg_read_ops = &filelayout_pg_read_ops, .pg_read_ops = &filelayout_pg_read_ops,
......
...@@ -77,6 +77,13 @@ struct nfs4_file_layout_dsaddr { ...@@ -77,6 +77,13 @@ struct nfs4_file_layout_dsaddr {
struct nfs4_fl_commit_bucket { struct nfs4_fl_commit_bucket {
struct list_head written; struct list_head written;
struct list_head committing; struct list_head committing;
struct pnfs_layout_segment *wlseg;
struct pnfs_layout_segment *clseg;
};
struct nfs4_fl_commit_info {
int nbuckets;
struct nfs4_fl_commit_bucket *buckets;
}; };
struct nfs4_filelayout_segment { struct nfs4_filelayout_segment {
...@@ -89,10 +96,19 @@ struct nfs4_filelayout_segment { ...@@ -89,10 +96,19 @@ struct nfs4_filelayout_segment {
struct nfs4_file_layout_dsaddr *dsaddr; /* Point to GETDEVINFO data */ struct nfs4_file_layout_dsaddr *dsaddr; /* Point to GETDEVINFO data */
unsigned int num_fh; unsigned int num_fh;
struct nfs_fh **fh_array; struct nfs_fh **fh_array;
struct nfs4_fl_commit_bucket *commit_buckets; /* Sort commits to ds */
int number_of_buckets;
}; };
struct nfs4_filelayout {
struct pnfs_layout_hdr generic_hdr;
struct nfs4_fl_commit_info commit_info;
};
static inline struct nfs4_filelayout *
FILELAYOUT_FROM_HDR(struct pnfs_layout_hdr *lo)
{
return container_of(lo, struct nfs4_filelayout, generic_hdr);
}
static inline struct nfs4_filelayout_segment * static inline struct nfs4_filelayout_segment *
FILELAYOUT_LSEG(struct pnfs_layout_segment *lseg) FILELAYOUT_LSEG(struct pnfs_layout_segment *lseg)
{ {
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册