Linux kernel mirror (for testing)
git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel
os
linux
SUNRPC: Handle NULL entries in svc_rqst_release_pages
svc_rqst_release_pages() releases response pages between rq_respages
and rq_next_page. It currently passes the entire range to
release_pages(), which does not expect NULL entries.
A subsequent patch preserves the rq_next_page pointer in
svc_rdma_save_io_pages() so that it accurately records how many
response pages were consumed. After that change, the range
[rq_respages, rq_next_page)
can contain NULL entries where pages have already been transferred
to a send context.
Iterate through the range entry by entry, skipping NULLs, to handle
this case correctly.
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
+13
-7
+13
-7
net/sunrpc/svc.c
+13
-7
net/sunrpc/svc.c
···
990
990
* svc_rqst_release_pages - Release Reply buffer pages
991
991
* @rqstp: RPC transaction context
992
992
*
993
-
* Release response pages that might still be in flight after
994
-
* svc_send, and any spliced filesystem-owned pages.
993
+
* Release response pages in the range [rq_respages, rq_next_page).
994
+
* NULL entries in this range are skipped, allowing transports to
995
+
* transfer pages to a send context before this function runs.
995
996
*/
996
997
void svc_rqst_release_pages(struct svc_rqst *rqstp)
997
998
{
998
-
int i, count = rqstp->rq_next_page - rqstp->rq_respages;
999
+
struct page **pp;
999
1000
1000
-
if (count) {
1001
-
release_pages(rqstp->rq_respages, count);
1002
-
for (i = 0; i < count; i++)
1003
-
rqstp->rq_respages[i] = NULL;
1001
+
for (pp = rqstp->rq_respages; pp < rqstp->rq_next_page; pp++) {
1002
+
if (*pp) {
1003
+
if (!folio_batch_add(&rqstp->rq_fbatch,
1004
+
page_folio(*pp)))
1005
+
__folio_batch_release(&rqstp->rq_fbatch);
1006
+
*pp = NULL;
1007
+
}
1004
1008
}
1009
+
if (rqstp->rq_fbatch.nr)
1010
+
__folio_batch_release(&rqstp->rq_fbatch);
1005
1011
}
1006
1012
1007
1013
/**