Messages in this thread |  | | From | Christian Schoenebeck <> | Subject | Re: [PATCH 3/3] 9p: Add mempools for RPCs | Date | Tue, 05 Jul 2022 11:59:55 +0200 |
| |
On Montag, 4. Juli 2022 16:19:46 CEST Kent Overstreet wrote: > On Mon, Jul 04, 2022 at 03:39:32PM +0200, Christian Schoenebeck wrote: > > So simple that it already had one obvious bug (at least). But as it seems > > that Dominique already supports your patch, I refrain from enumerating > > more reasons. > > So snippy.
Yeah, the tone makes the music. If you adjust yours, then I'll do, too.
> > > > However that's exactly what I was going to address with my already > > > > posted > > > > patches (relevant patches regarding this issue here being 9..12): > > > > https://lore.kernel.org/all/cover.1640870037.git.linux_oss@crudebyte.c > > > > om/ > > > > And in the cover letter (section "STILL TODO" ... "3.") I was > > > > suggesting > > > > to > > > > subsequently subdivide kmem_cache_alloc() into e.g. 4 allocation size > > > > categories? Because that's what my already posted patches do anyway. > > > > > > Yeah that sounds like you're just reimplementing kmalloc. > > > > Quite exaggerated statement. > > I'm just pointing out that kmalloc() is just a frontend around > kmem_cache_alloc() that picks the cache based on the size parameter... so... > still sounds like you are? > > Not that there's never a legitimate reason to do so, but it does raise an > eyebrow.
So you are saying this change was useless as well then? https://github.com/torvalds/linux/commit/91a76be37ff89795526c452a6799576b03bec501
Like already discussed in the other email, I omitted those cache size granularity changes for good reasons, until proofen by benchmark that they would actually help.
Best regards, Christian Schoenebeck
|  |