linux-kernel - Re: [PATCH v3] blk-cgroup: Replace u64 sync with spinlock for iostat update

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: <Zphvj1kXK2xcKj7J@slm.duckdns.org>
Date: Wed, 17 Jul 2024 15:27:43 -1000
From: "tj@...nel.org" <tj@...nel.org>
To: Boy Wu (吳勃誼) <Boy.Wu@...iatek.com>
Cc: "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"boris@....io" <boris@....io>,
	"linux-block@...r.kernel.org" <linux-block@...r.kernel.org>,
	"linux-mediatek@...ts.infradead.org" <linux-mediatek@...ts.infradead.org>,
	"cgroups@...r.kernel.org" <cgroups@...r.kernel.org>,
	"axboe@...nel.dk" <axboe@...nel.dk>,
	Iverlin Wang (王苳霖) <Iverlin.Wang@...iatek.com>,
	"josef@...icpanda.com" <josef@...icpanda.com>,
	"linux-arm-kernel@...ts.infradead.org" <linux-arm-kernel@...ts.infradead.org>,
	"matthias.bgg@...il.com" <matthias.bgg@...il.com>,
	"angelogioacchino.delregno@...labora.com" <angelogioacchino.delregno@...labora.com>
Subject: Re: [PATCH v3] blk-cgroup: Replace u64 sync with spinlock for iostat
 update

Hello,

On Thu, Jul 18, 2024 at 01:21:32AM +0000, Boy Wu (吳勃誼) wrote:
...
> I think this will work, but as I mentioned before, this issue is only
> on 32 bit SMP systems. Replacing u64 sync with spinlock will increase
> overhead on 64 bit systems, because u64 sync does nothing on 64 bit
> systems. However, if it is acceptable, we can proceed with this
> solution.

We can encapsulate the spinlock in some helpers and make them conditional on
32bit. However, the u64_sync -> spinlock conversions in the suggested patch
are all on cold paths, so I doubt it'd be all that noticeable especially
given that the hottest path of them all is already grabbing blkg_stat_lock,
but either way is fine by me.

Thanks.

-- 
tejun