提交 b0ac694f 编写于 作者: S Stefano Stabellini

xen/disk: don't leak stack data via response ring

Rather than constructing a local structure instance on the stack, fill
the fields directly on the shared ring, just like other (Linux)
backends do. Build on the fact that all response structure flavors are
actually identical (aside from alignment and padding at the end).

This is XSA-216.

Reported by: Anthony Perard <anthony.perard@citrix.com>
Signed-off-by: NJan Beulich <jbeulich@suse.com>
Signed-off-by: NStefano Stabellini <sstabellini@kernel.org>
Acked-by: NAnthony PERARD <anthony.perard@citrix.com>
上级 577caa26
......@@ -769,31 +769,30 @@ static int blk_send_response_one(struct ioreq *ioreq)
struct XenBlkDev *blkdev = ioreq->blkdev;
int send_notify = 0;
int have_requests = 0;
blkif_response_t resp;
void *dst;
resp.id = ioreq->req.id;
resp.operation = ioreq->req.operation;
resp.status = ioreq->status;
blkif_response_t *resp;
/* Place on the response ring for the relevant domain. */
switch (blkdev->protocol) {
case BLKIF_PROTOCOL_NATIVE:
dst = RING_GET_RESPONSE(&blkdev->rings.native, blkdev->rings.native.rsp_prod_pvt);
resp = (blkif_response_t *) RING_GET_RESPONSE(&blkdev->rings.native,
blkdev->rings.native.rsp_prod_pvt);
break;
case BLKIF_PROTOCOL_X86_32:
dst = RING_GET_RESPONSE(&blkdev->rings.x86_32_part,
blkdev->rings.x86_32_part.rsp_prod_pvt);
resp = (blkif_response_t *) RING_GET_RESPONSE(&blkdev->rings.x86_32_part,
blkdev->rings.x86_32_part.rsp_prod_pvt);
break;
case BLKIF_PROTOCOL_X86_64:
dst = RING_GET_RESPONSE(&blkdev->rings.x86_64_part,
blkdev->rings.x86_64_part.rsp_prod_pvt);
resp = (blkif_response_t *) RING_GET_RESPONSE(&blkdev->rings.x86_64_part,
blkdev->rings.x86_64_part.rsp_prod_pvt);
break;
default:
dst = NULL;
return 0;
}
memcpy(dst, &resp, sizeof(resp));
resp->id = ioreq->req.id;
resp->operation = ioreq->req.operation;
resp->status = ioreq->status;
blkdev->rings.common.rsp_prod_pvt++;
RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(&blkdev->rings.common, send_notify);
......
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