| Message ID | 20191025161717.106825.14421.stgit@awfm-01.aw.intel.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | Few more rc fixes | expand |
On Fri, Oct 25, 2019 at 03:58:17PM -0400, Dennis Dalessandro wrote: > Here we have 4 more bug fixing patches. These are all marked stable as well so > it would be good to get these in to 5.4 if possible. > > The allow for speeds patch is not a "feature" it is a bug that causes problems > if the host comes up in gen4. Is very small. The other 3 from Kaike fix TID RDMA > bugs. A few more lines of code here but all relegated to TID RDMA. > > > James Erwin (1): > IB/hfi1: Allow for all speeds higher than gen3 With v2 of this one > Kaike Wan (3): > IB/hfi1: Ensure r_tid_ack is valid before building TID RDMA ACK packet > IB/hfi1: Calculate flow weight based on QP MTU for TID RDMA > IB/hfi1: TID RDMA WRITE should not return IB_WC_RNR_RETRY_EXC_ERR Applied to for-rc, thanks Jason
Here we have 4 more bug fixing patches. These are all marked stable as well so it would be good to get these in to 5.4 if possible. The allow for speeds patch is not a "feature" it is a bug that causes problems if the host comes up in gen4. Is very small. The other 3 from Kaike fix TID RDMA bugs. A few more lines of code here but all relegated to TID RDMA. --- James Erwin (1): IB/hfi1: Allow for all speeds higher than gen3 Kaike Wan (3): IB/hfi1: Ensure r_tid_ack is valid before building TID RDMA ACK packet IB/hfi1: Calculate flow weight based on QP MTU for TID RDMA IB/hfi1: TID RDMA WRITE should not return IB_WC_RNR_RETRY_EXC_ERR drivers/infiniband/hw/hfi1/init.c | 1 - drivers/infiniband/hw/hfi1/pcie.c | 4 ++ drivers/infiniband/hw/hfi1/rc.c | 16 +++++---- drivers/infiniband/hw/hfi1/tid_rdma.c | 57 +++++++++++++++++++-------------- drivers/infiniband/hw/hfi1/tid_rdma.h | 3 +- 5 files changed, 44 insertions(+), 37 deletions(-) -- -Denny