[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <Zy0zM8Qc0-9SVVpB@kbusch-mbp>
Date: Thu, 7 Nov 2024 14:37:55 -0700
From: Keith Busch <kbusch@...nel.org>
To: Breno Leitao <leitao@...ian.org>
Cc: Jens Axboe <axboe@...nel.dk>, Christoph Hellwig <hch@....de>,
Sagi Grimberg <sagi@...mberg.me>, linux-nvme@...ts.infradead.org,
linux-kernel@...r.kernel.org, kernel-team@...a.com
Subject: Re: [PATCH] nvme/multipath: Fix RCU list traversal to use SRCU
primitive
On Tue, Nov 05, 2024 at 06:42:46AM -0800, Breno Leitao wrote:
> The code currently uses list_for_each_entry_rcu() while holding an SRCU
> lock, triggering false positive warnings with CONFIG_PROVE_RCU=y
> enabled:
>
> drivers/nvme/host/multipath.c:168 RCU-list traversed in non-reader section!!
> drivers/nvme/host/multipath.c:227 RCU-list traversed in non-reader section!!
> drivers/nvme/host/multipath.c:260 RCU-list traversed in non-reader section!!
>
> While the list is properly protected by SRCU lock, the code uses the
> wrong list traversal primitive. Replace list_for_each_entry_rcu() with
> list_for_each_entry_srcu() to correctly indicate SRCU-based protection
> and eliminate the false warning.
Thanks, applied to nvme-6.12. Sorry I missed this was posted until after
I'd sent this week's pull request, but this will make the next one.
Powered by blists - more mailing lists