2026年9月19日(土) 3:24 Nowlin, Alexander <[email protected]>: > > On 9/16/2026 6:13 PM, Takeru Hayasaka wrote: > > > Hi all, > > > > > > Gentle ping on this series, posted about a 2 month ago: > > > https://patchwork.ozlabs.org/project/intel-wired-lan/list/?series=5137 > > > 64 > > > > > > Please let me know if anything is needed from my side. > > > > Hi Takeru, > > > > Our validation came across an issue while testing this and we're still > > working on getting the scope of the > problem. The display (ethtool -n) is > > not showing the expected fields; we're working to see if it's just the > > > display or whether it is not programming it correctly as well. > > Hi Takeru, > > We have finished the validation that Tony had mentioned in the previous > response. > > Based on our testing the underlying functionality of the patch is working, as > we confirmed that traffic is routed to multiple queues as expected. > Therefore, the only issue with the patch is the ethtool -n display not > listing the expected fields; without this patch applied the GTP fields are > displayed as expected. > > Thank you, > Alexander > > > Thanks, > > Tony > > > > > Thanks, > > > Takeru >
Hi Alexander, Tony, Thanks for taking the time to validate this, and sorry for the incomplete patch. I can reproduce the display issue here. It's the lookup in ice_get_rss_cfg(): the patch adds IPV_OTHER to the stored hdrs, but ethtool's hdrs never carry that bit, so the exact match fails and nothing is shown. Programming is unaffected, as you observed. I'll send a v2 that fixes the lookup. Thanks, Takeru
