qubes-linux-kernel/patches.xen/xen-netfront-detach-crash.patch
Marek Marczykowski-Górecki fdbdf930ba
Apply patch for xen-netfront crash on network detach
It still isn't clear this fix doesn't introduce some memory leak, but
even if so, it's much better to have memory leak in rare situation, than
crash preventing netvm change operation. Details in this thread:
http://xen.markmail.org/thread/pw5edbtqienjx4q5
Unfortunately no one responded there about possible memory leak...

Fixes QubesOS/qubes-issues#975
2015-10-31 15:53:31 +01:00

27 lines
1.1 KiB
Diff

When it get to free_page(queue->grant_tx_page[i]), the use counter on this page
is already 0, which cause a crash. Not sure if this is the proper fix
(according to git log this may introduce some memory leak), but at least it
prevent the crash.
Details in this thread:
http://xen.markmail.org/thread/pw5edbtqienjx4q5
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c
index f821a97..a5efbb0 100644
--- a/drivers/net/xen-netfront.c
+++ b/drivers/net/xen-netfront.c
@@ -1065,9 +1069,10 @@ static void xennet_release_tx_bufs(struct netfront_queue *queue)
skb = queue->tx_skbs[i].skb;
get_page(queue->grant_tx_page[i]);
- gnttab_end_foreign_access(queue->grant_tx_ref[i],
- GNTMAP_readonly,
- (unsigned long)page_address(queue->grant_tx_page[i]));
+ gnttab_end_foreign_access_ref(
+ queue->grant_tx_ref[i], GNTMAP_readonly);
+ gnttab_release_grant_reference(
+ &queue->gref_tx_head, queue->grant_tx_ref[i]);
queue->grant_tx_page[i] = NULL;
queue->grant_tx_ref[i] = GRANT_INVALID_REF;
add_id_to_freelist(&queue->tx_skb_freelist, queue->tx_skbs, i);