linux-kernel - Re: [PATCH] iomap: add allocation cache for iomap

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <aV6HawIBiF-fQo31@infradead.org>
Date: Wed, 7 Jan 2026 08:18:51 -0800
From: Christoph Hellwig <hch@...radead.org>
To: guzebing <guzebing1612@...il.com>
Cc: Christoph Hellwig <hch@...radead.org>, brauner@...nel.org,
	djwong@...nel.org, linux-xfs@...r.kernel.org,
	linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org,
	guzebing <guzebing@...edance.com>,
	Fengnan Chang <changfengnan@...edance.com>
Subject: Re: [PATCH] iomap: add allocation cache for iomap_dio

On Mon, Jan 05, 2026 at 05:27:54PM +0800, guzebing wrote:
> Yes, I try to use a dedicated kmem cache to allocate cache for iomap-dio
> structure. However, when system memory is sufficient, kmalloc and kmem cache
> deliver identical performance.

Thanks for benchmarking this.

> > Also any chance you could factor this into common code?
> > 
> For a mempool, we first allocate with kmalloc or kmem cache and finally fall
> back to a reserved cache—this is for reliability. It’s not a great fit for
> our high‑performance scenario.
> 
> Additionally, the current need for frequent allocation/free (hundreds of
> thousands to millions of times per second) may be more suitable for the bio
> or dio structures; beyond those, I’m not sure whether similar scenarios
> exist.
> 
> If we were to extract a generic implementation solely for this, would it
> yield significant benefits? Do you have any good suggestions?


Factoring mean the percpu cache.  But given that it's been so long that
I looked at the code all the defails of this have been paged out from
my brain.  I can take another look when you resend it.