On Fri, 2017-01-20 at 16:57 -0500, Laurence Oberman wrote: > Hello Bart, if this the match to what you shared with me earlier I have > tested the IB/SRP part of this series. > > This is the IB/RDMA/SRP stack, testing both ib_srp and ib_srpt. > > 1. Tested direct and buffered read I/O with 4MB I/O and smaller I/O sizes > direct to mpath devices > 2. Tested read and write I/O with 4MB and smaller I/O to XFS file systems. > 3. Removed a SRP paths one by one and let then reconnect with no issues > > All my tests were SRP using connectX-4 (mlx5) at EDR (100G) > > For: Optimize DMA mapping tree, branch for next from latest copy on: > https://github.com/bvanassche/linux > Tested-by: Laurence Oberman <loberman@xxxxxxxxxx> Hello Laurence, The code I posted is identical to the code I asked you to test. Thanks for the help with testing! I tested the following ULP and HBA drivers myself with this patch series: ib_srp, ib_srpt, ib_mlx4 and rxe. Bart.-- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
