Re: [PATCH 3/5] cifsd: add file operations

From: Matthew Wilcox
Date: Mon Mar 22 2021 - 09:16:16 EST

On Mon, Mar 22, 2021 at 06:03:21PM +0900, Sergey Senozhatsky wrote:
> On (21/03/22 08:15), Matthew Wilcox wrote:
> >
> > What's the scenario for which your allocator performs better than slub
> >
> IIRC request and reply buffers can be up to 4M in size. So this stuff
> just allocates a number of fat buffers and keeps them around so that
> it doesn't have to vmalloc(4M) for every request and every response.

That makes a lot more sense; I was thrown off by the kvmalloc, which
is usually used for allocations that might be smaller than PAGE_SIZE.

So what this patch is really saying is that vmalloc() should include
some caching, so it can defer freeing until there's memory pressure
or it's built up a large (percpu) backlog of freed areas.

Vlad, have you thought about this?