On 9/10/21 9:32 AM, Al Viro wrote: > On Fri, Sep 10, 2021 at 09:08:02AM -0600, Jens Axboe wrote: > >>> You actually can cut it down even more - nr_segs + iov remains constant >>> all along, so you could get away with just 3 words here... I would be >> >> Mmm, the iov pointer remains constant? Maybe I'm missing your point, but >> the various advance functions are quite happy to increment iter->iov or >> iter->bvec, so we need to restore them. From a quick look, looks like >> iter->nr_segs is modified for advancing too. >> >> What am I missing? > > i->iov + i->nr_segs does not change - the places incrementing the former > will decrement the latter by the same amount. So it's enough to store > either of those - the other one can be recovered by subtracting the > saved value from the current i->iov + i->nr_segs. Ahh, clever. Yes that should work just fine. Let me test that and send out a proposal. Thanks Al. -- Jens Axboe
