| Message ID | 20230824075500.1735790-1-junfeng.guo@intel.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | Introduce the Parser Library | expand |
On Thu, 24 Aug 2023 15:54:45 +0800 Junfeng Guo wrote: > Current software architecture for flow filtering offloading limited > the capability of Intel Ethernet 800 Series Dynamic Device > Personalization (DDP) Package. The flow filtering offloading in the > driver is enabled based on the naming parsers, each flow pattern is > represented by a protocol header stack. And there are multiple layers > (e.g., virtchnl) to maintain their own enum/macro/structure > to represent a protocol header (IP, TCP, UDP ...), thus the extra > parsers to verify if a pattern is supported by hardware or not as > well as the extra converters that to translate represents between > different layers. Every time a new protocol/field is requested to be > supported, the corresponding logic for the parsers and the converters > needs to be modified accordingly. Thus, huge & redundant efforts are > required to support the increasing flow filtering offloading features, > especially for the tunnel types flow filtering. You keep breaking the posting guidelines :( I already complained to people at Intel about you. The only way to push back that I can think of is to start handing out posting suspensions for all @intel.com addresses on netdev. Please don't make us stoop that low.
From: Jakub Kicinski <kuba@kernel.org> Date: Thu, 24 Aug 2023 08:20:39 -0700 > On Thu, 24 Aug 2023 15:54:45 +0800 Junfeng Guo wrote: >> Current software architecture for flow filtering offloading limited >> the capability of Intel Ethernet 800 Series Dynamic Device >> Personalization (DDP) Package. The flow filtering offloading in the >> driver is enabled based on the naming parsers, each flow pattern is >> represented by a protocol header stack. And there are multiple layers >> (e.g., virtchnl) to maintain their own enum/macro/structure >> to represent a protocol header (IP, TCP, UDP ...), thus the extra >> parsers to verify if a pattern is supported by hardware or not as >> well as the extra converters that to translate represents between >> different layers. Every time a new protocol/field is requested to be >> supported, the corresponding logic for the parsers and the converters >> needs to be modified accordingly. Thus, huge & redundant efforts are >> required to support the increasing flow filtering offloading features, >> especially for the tunnel types flow filtering. > > You keep breaking the posting guidelines :( > I already complained to people at Intel about you. > > The only way to push back that I can think of is to start handing out > posting suspensions for all @intel.com addresses on netdev. Please Ah, that collective responsibility :D > don't make us stoop that low. But seriously, please don't. Intel is huge and we physically can't keep an eye on every developer or patch. I personally don't even know what team the submitter is from. Spending 8 hrs a day on tracking every @intel.com submission on netdev is also not something I'd like to do at work (I mean, I'd probably like reviewing every line coming out of my org, had I 120-150 hrs a day...). I know that sounds cheap, but that's how I see it :z > > Thanks, Olek
On Fri, 25 Aug 2023 12:52:27 +0200 Alexander Lobakin wrote: > > You keep breaking the posting guidelines :( > > I already complained to people at Intel about you. > > > > The only way to push back that I can think of is to start handing out > > posting suspensions for all @intel.com addresses on netdev. Please > > Ah, that collective responsibility :D I'd call it delegating the responsibilities :) > > don't make us stoop that low. > > But seriously, please don't. Intel is huge and we physically can't keep > an eye on every developer or patch. I personally don't even know what > team the submitter is from. > Spending 8 hrs a day on tracking every @intel.com submission on > netdev is also not something I'd like to do at work (I mean, I'd > probably like reviewing every line coming out of my org, had I > 120-150 hrs a day...). I know that sounds cheap, but that's how I see > it :z Intel may be huge, but this patch is for ice specifically. And the author knows enough to put presumably-internally-mandated iwl-next tag in the subject. So how about someone steps up and points them at a manual before 4 versions are posted?