| Message ID | 20190614163146.44927.95985.stgit@awfm-01.aw.intel.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | IB/hfi1: Fix a stuck qp problem | expand |
On Fri, 2019-06-14 at 12:32 -0400, Dennis Dalessandro wrote: > This series fixes a problem we encountered with a QP getting stuck. > It is a bit > bigger than we probably want for RC so I'm fine applying this to for- > next > instead especially this late in the game for 5.2. > --- > > Mike Marciniszyn (7): > IB/hfi1: Avoid hardlockup with flushlist_lock > IB/hfi1: Silence txreq allocation warnings > IB/hfi1: Create inline to get extended headers > IB/hfi1: Use aborts to trigger RC throttling > IB/hfi1: Wakeup QPs orphaned on wait list after flush > IB/hfi1: Handle wakeup of orphaned QPs for pio > IB/hfi1: Handle port down properly in pio > Hi Denny, These all look reasonably sane. Their big for a late rc, but they fix scale issues, not your typical one line typo bug, and those often are bigger fixes. I'm comfortable arguing for these. So, assuming my build test that's underway completes, applied to for-rc. Thanks.
On Mon, 2019-06-17 at 21:19 -0400, Doug Ledford wrote: > On Fri, 2019-06-14 at 12:32 -0400, Dennis Dalessandro wrote: > > This series fixes a problem we encountered with a QP getting stuck. > > It is a bit > > bigger than we probably want for RC so I'm fine applying this to > > for- > > next > > instead especially this late in the game for 5.2. > > --- > > > > Mike Marciniszyn (7): > > IB/hfi1: Avoid hardlockup with flushlist_lock > > IB/hfi1: Silence txreq allocation warnings > > IB/hfi1: Create inline to get extended headers > > IB/hfi1: Use aborts to trigger RC throttling > > IB/hfi1: Wakeup QPs orphaned on wait list after flush > > IB/hfi1: Handle wakeup of orphaned QPs for pio > > IB/hfi1: Handle port down properly in pio > > > > Hi Denny, > > These all look reasonably sane. Their They're*. I really have better grammar than that. > big for a late rc, but they fix > scale issues, not your typical one line typo bug, and those often are > bigger fixes. I'm comfortable arguing for these. So, assuming my > build test that's underway completes, applied to for-rc. Thanks. >
This series fixes a problem we encountered with a QP getting stuck. It is a bit bigger than we probably want for RC so I'm fine applying this to for-next instead especially this late in the game for 5.2. --- Mike Marciniszyn (7): IB/hfi1: Avoid hardlockup with flushlist_lock IB/hfi1: Silence txreq allocation warnings IB/hfi1: Create inline to get extended headers IB/hfi1: Use aborts to trigger RC throttling IB/hfi1: Wakeup QPs orphaned on wait list after flush IB/hfi1: Handle wakeup of orphaned QPs for pio IB/hfi1: Handle port down properly in pio drivers/infiniband/hw/hfi1/hfi.h | 31 ++++++++++++++++++ drivers/infiniband/hw/hfi1/pio.c | 21 +++++++++++- drivers/infiniband/hw/hfi1/rc.c | 53 ++++++++++++++++++------------ drivers/infiniband/hw/hfi1/sdma.c | 26 +++++++++++---- drivers/infiniband/hw/hfi1/ud.c | 4 +- drivers/infiniband/hw/hfi1/verbs.c | 14 +++++--- drivers/infiniband/hw/hfi1/verbs.h | 1 + drivers/infiniband/hw/hfi1/verbs_txreq.c | 2 + drivers/infiniband/hw/hfi1/verbs_txreq.h | 3 +- 9 files changed, 116 insertions(+), 39 deletions(-) -- -Denny