提交 35848f68 编写于 作者: J Jason Wang 提交者: Greg Kroah-Hartman

drivers: hv: switch to use mb() instead of smp_mb()

Even if guest were compiled without SMP support, it could not assume that host
wasn't. So switch to use mb() instead of smp_mb() to force memory barriers for
UP guest.
Signed-off-by: NJason Wang <jasowang@redhat.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: NK. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
上级 36dd6831
......@@ -32,7 +32,7 @@
void hv_begin_read(struct hv_ring_buffer_info *rbi)
{
rbi->ring_buffer->interrupt_mask = 1;
smp_mb();
mb();
}
u32 hv_end_read(struct hv_ring_buffer_info *rbi)
......@@ -41,7 +41,7 @@ u32 hv_end_read(struct hv_ring_buffer_info *rbi)
u32 write;
rbi->ring_buffer->interrupt_mask = 0;
smp_mb();
mb();
/*
* Now check to see if the ring buffer is still empty.
......@@ -71,7 +71,7 @@ u32 hv_end_read(struct hv_ring_buffer_info *rbi)
static bool hv_need_to_signal(u32 old_write, struct hv_ring_buffer_info *rbi)
{
smp_mb();
mb();
if (rbi->ring_buffer->interrupt_mask)
return false;
......@@ -442,7 +442,7 @@ int hv_ringbuffer_write(struct hv_ring_buffer_info *outring_info,
sizeof(u64));
/* Issue a full memory barrier before updating the write index */
smp_mb();
mb();
/* Now, update the write location */
hv_set_next_write_location(outring_info, next_write_location);
......@@ -549,7 +549,7 @@ int hv_ringbuffer_read(struct hv_ring_buffer_info *inring_info, void *buffer,
/* Make sure all reads are done before we update the read index since */
/* the writer may start writing to the read area once the read index */
/*is updated */
smp_mb();
mb();
/* Update the read index */
hv_set_next_read_location(inring_info, next_read_location);
......
......@@ -434,7 +434,7 @@ static void vmbus_on_msg_dpc(unsigned long data)
* will not deliver any more messages since there is
* no empty slot
*/
smp_mb();
mb();
if (msg->header.message_flags.msg_pending) {
/*
......
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