| Message ID | 20200504130917.175613.43231.stgit@awfm-01.aw.intel.com (mailing list archive) |
|---|---|
| State | Mainlined |
| Commit | fa8dac3968635dec8518a13ac78d662f2aa88e4d |
| Delegated to: | Jason Gunthorpe |
| Headers | show |
| Series | [for-rc] IB/hfi1: Fix another case where pq is left on waitlist | expand |
On Mon, May 04, 2020 at 09:09:17AM -0400, Mike Marciniszyn wrote: > The commit noted below fixed a case where a pq is left on > the sdma wait list. > > It however missed another case. > > user_sdma_send_pkts() has two calls from hfi1_user_sdma_process_request(). > > If the first one fails as indicated by -EBUSY, the pq will be placed on > the waitlist as by design. > > If the second call then succeeds, the pq is still on the waitlist > setting up a race with the interrupt handler if a subsequent request uses > a different SDMA engine > > Fix by deleting the first call. > > The use of pcount and the intent to send a short burst of packets followed > by the larger balance of packets was never correctly implemented, because > the two calls always send pcount packets no matter what. A subsequent > patch will correct that issue. > > Fixes: 9a293d1e21a6 ("IB/hfi1: Ensure pq is not left on waitlist") > Cc: <stable@vger.kernel.org> > Reviewed-by: Kaike Wan <kaike.wan@intel.com> > Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com> > --- > drivers/infiniband/hw/hfi1/user_sdma.c | 4 ---- > 1 file changed, 4 deletions(-) Applied to for-rc, thanks Jason
diff --git a/drivers/infiniband/hw/hfi1/user_sdma.c b/drivers/infiniband/hw/hfi1/user_sdma.c index 13e4203..a92346e 100644 --- a/drivers/infiniband/hw/hfi1/user_sdma.c +++ b/drivers/infiniband/hw/hfi1/user_sdma.c @@ -589,10 +589,6 @@ int hfi1_user_sdma_process_request(struct hfi1_filedata *fd, set_comp_state(pq, cq, info.comp_idx, QUEUED, 0); pq->state = SDMA_PKT_Q_ACTIVE; - /* Send the first N packets in the request to buy us some time */ - ret = user_sdma_send_pkts(req, pcount); - if (unlikely(ret < 0 && ret != -EBUSY)) - goto free_req; /* * This is a somewhat blocking send implementation.
The commit noted below fixed a case where a pq is left on the sdma wait list. It however missed another case. user_sdma_send_pkts() has two calls from hfi1_user_sdma_process_request(). If the first one fails as indicated by -EBUSY, the pq will be placed on the waitlist as by design. If the second call then succeeds, the pq is still on the waitlist setting up a race with the interrupt handler if a subsequent request uses a different SDMA engine Fix by deleting the first call. The use of pcount and the intent to send a short burst of packets followed by the larger balance of packets was never correctly implemented, because the two calls always send pcount packets no matter what. A subsequent patch will correct that issue. Fixes: 9a293d1e21a6 ("IB/hfi1: Ensure pq is not left on waitlist") Cc: <stable@vger.kernel.org> Reviewed-by: Kaike Wan <kaike.wan@intel.com> Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com> --- drivers/infiniband/hw/hfi1/user_sdma.c | 4 ---- 1 file changed, 4 deletions(-)