linux-kernel - Re: [PATCH] nvmet-rdma: use spin_lock_bh() on rsp_wr_wait

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: <709e6baa-8ac5-4f2c-984b-0d7b06deb7d0@grimberg.me>
Date:   Tue, 26 Sep 2023 22:55:14 +0300
From:   Sagi Grimberg <sagi@...mberg.me>
To:     Chengfeng Ye <dg573847474@...il.com>, hch@....de, kch@...dia.com
Cc:     linux-nvme@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] nvmet-rdma: use spin_lock_bh() on rsp_wr_wait_lock



On 9/26/23 20:22, Chengfeng Ye wrote:
> It seems to me that read_cqe.done could be executed under softirq
> context,

It doesn't though.

>   as done callbacks always do, and it acquires rsp_wr_wait_lock
> along the following call chain.

The CQ polling context is IB_POLL_WORKQUEUE, which does not
run in softirq.