Comment 2 for bug 1702768

------- Comment From <email address hidden> 2017-07-10 15:54 EDT-------
(In reply to comment #5)
> I built a test kernel with a pick of commit 377aa6b0efba. The test kernel
> can be downloaded from:
>
> http://kernel.ubuntu.com/~jsalisbury/lp1702768/
>
> Can you test this kernel and see if it resolves this bug?

I tried this kernel and it is ok.
[ 128.224664] (0004:01:00.0): E-Switch: E-Switch enable SRIOV: nvfs(1) mode (1)
[ 128.234634] (0004:01:00.0): E-Switch: SRIOV enabled: active vports(2)
[ 128.234818] mlx5_core 0004:01:00.0: VF BAR0: [mem 0x240000000000-0x2401ffffffff 64bit pref] shifted to [mem 0x240000000000-0x2401ffffffff 64bit pref] (Disabling 1 VFs shifted by 0)
[ 128.234836] pci 0004:01: 0.2: [PE# 00] VF 0004:01:00.2 associated with PE#0
[ 128.235086] pci 0004:01: 0.2: [PE# 00] Setting up 32-bit TCE table at 0..80000000
[ 128.238861] pci 0004:01: 0.2: [PE# 00] Setting up window#0 0..7fffffff pg=1000
[ 128.238972] pci 0004:01: 0.2: [PE# 00] Enabling 64-bit DMA bypass
[ 128.344614] pci 0004:01:00.2: [15b3:1014] type 00 class 0x020000
[ 128.344942] pci 0004:01:00.2: Max Payload Size set to 512 (was 128, max 512)
[ 128.345403] iommu: Adding device 0004:01:00.2 to group 6
[ 128.345871] mlx5_core 0004:01:00.2: enabling device (0000 -> 0002)
[ 128.345907] mlx5_core 0004:01:00.2: Using 64-bit DMA iommu bypass
[ 128.346076] mlx5_core 0004:01:00.2: firmware version: 12.20.1010
[ 128.902589] mlx5_core 0004:01:00.2: MLX5E: StrdRq(0) RqSz(1024) StrdSz(1) RxCqeCmprss(0)
[ 128.903017] mlx5_core 0004:01:00.2: Assigned random MAC address 2a:fc:b7:49:03:1b
[ 129.007113] mlx5_core 0004:01:00.2 enP4p1s0f2: renamed from eth0
[ 129.015731] mlx5_ib: Mellanox Connect-IB Infiniband driver v2.2-1 (Feb 2014)

uname -a
4.10.0-26-generic #30~lp1702768 SMP Mon Jul 10 18:37:50 UTC 2017 ppc64le ppc64le ppc64le GNU/Linux