From 05cbbb395f193a1d15e0f749eff8abe5c9c49b62 Mon Sep 17 00:00:00 2001 From: Lars Ellenberg Date: Fri, 16 Jan 2015 17:41:55 +0100 Subject: [PATCH] drbd: Fix spurious disk-timeout (You should not use disk-timeout anyways, see the man page for why...) We add incoming requests to the tail of some ring list. On local completion, requests are removed from that list. The timer looks only at the head of that ring list, so is supposed to only see the oldest request. All protected by a spinlock. The request object is created with timestamps zeroed out. The timestamp was only filled in just before the actual submit. But to actually submit the request, we need to give up the spinlock. If you are unlucky, there is no older still pending request, the timer looks at a new request with timestamp still zero (before it even was submitted), and 0 + timeout is most likely older than "now". Better assign the timestamp right when we put the request object on said ring list. Signed-off-by: Philipp Reisner Signed-off-by: Lars Ellenberg Signed-off-by: Jens Axboe --- drivers/block/drbd/drbd_req.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/block/drbd/drbd_req.c b/drivers/block/drbd/drbd_req.c index 55fca685fca5..7660f6e749ff 100644 --- a/drivers/block/drbd/drbd_req.c +++ b/drivers/block/drbd/drbd_req.c @@ -1167,7 +1167,6 @@ drbd_submit_req_private_bio(struct drbd_request *req) * stable storage, and this is a WRITE, we may not even submit * this bio. */ if (get_ldev(device)) { - req->pre_submit_jif = jiffies; if (drbd_insert_fault(device, rw == WRITE ? DRBD_FAULT_DT_WR : rw == READ ? DRBD_FAULT_DT_RD @@ -1311,6 +1310,7 @@ static void drbd_send_and_submit(struct drbd_device *device, struct drbd_request &device->pending_master_completion[rw == WRITE]); if (req->private_bio) { /* needs to be marked within the same spinlock */ + req->pre_submit_jif = jiffies; list_add_tail(&req->req_pending_local, &device->pending_completion[rw == WRITE]); _req_mod(req, TO_BE_SUBMITTED); -- 2.20.1