[Intel-wired-lan] [PATCH net-next] ice: Allow operation with reduced device MSI-X

G, GurucharanX gurucharanx.g at intel.com
Mon Sep 5 08:23:42 UTC 2022



> -----Original Message-----
> From: Intel-wired-lan <intel-wired-lan-bounces at osuosl.org> On Behalf Of
> Tony Nguyen
> Sent: Tuesday, August 23, 2022 12:27 AM
> To: intel-wired-lan at lists.osuosl.org
> Subject: [Intel-wired-lan] [PATCH net-next] ice: Allow operation with
> reduced device MSI-X
> 
> The driver currently takes an all or nothing approach for device MSI-X
> vectors. Meaning if it does not get its full allocation, it will fail and not load.
> There is no reason it can't work with a reduced number of MSI-X vectors.
> Take a similar approach as commit 741106f7bd8d ("ice: Improve MSI-X
> fallback logic") and, instead, adjust the MSI-X request to make use of what is
> available.
> 
> Signed-off-by: Tony Nguyen <anthony.l.nguyen at intel.com>
> ---
>  drivers/net/ethernet/intel/ice/ice_main.c | 187 ++++++++++++----------
>  1 file changed, 102 insertions(+), 85 deletions(-)
> 

Based on the below test results, marking this as tested
1.Verify the traffic between the SUT and TG with variable ping and variable payload of TCP/UDP
2.Verified ITR count
3.Kept regression for 5-6 hours with basic payload.
4.Driver reload test tried 10-15 times but couldn't find any issue on dmesg logs
5.Transferred the packets on multiple port
6.Reduce the no of online CPU cores. Then reload the driver multiple times and make the cpu cores online.
7.Split the Tacoma rapids single port into 4 ports through devlink split command and observe any changes. The devlink split command wasn't supporting to split the NIC into 8 ports.

Tested-by: Gurucharan <gurucharanx.g at intel.com> (A Contingent worker at Intel)




More information about the Intel-wired-lan mailing list