From: Alexander Duyck <alexander.du...@gmail.com> Date: Sun, 27 Sep 2015 15:58:56 -0700
> On 09/26/2015 10:36 PM, David Miller wrote: >> From: Alexander Duyck <adu...@mirantis.com> >> Date: Tue, 22 Sep 2015 14:56:08 -0700 >> >>> Rather than carry around a value of budget that is 0 or less we can >>> instead >>> just loop through and pass 0 to each napi->poll call. If any driver >>> returns a value for work done that is non-zero then we can report that >>> driver and continue rather than allowing a bad actor to make the >>> budget >>> value negative and pass that negative value to napi->poll. >> Unfortunately we have drivers that won't do any TX work if the budget >> is zero. > > Well that is what we are doing right now. The fact is the call starts > out with a budget of 0, and it is somewhat hidden from the call since > the budget is assigned a value of 0 in netpoll_poll_dev. That is one > of the things I was wanting do address because that is clear as mud > from looking at poll_one_napi. Based on the code you would assume > budget starts out as a non-zero value and it doesn't. I see, thanks for explaining. -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
