[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAC=S1ngpXtsLUF=A-MxVnmWY7YgwvxHKSe0NuN_n=G+7-kpG7Q@mail.gmail.com>
Date: Wed, 28 Aug 2024 17:41:00 +0800
From: Fei Shao <fshao@...omium.org>
To: jason-jh.lin@...iatek.com, Chun-Kuang Hu <chunkuang.hu@...nel.org>
Cc: AngeloGioacchino Del Regno <angelogioacchino.delregno@...labora.com>,
Philipp Zabel <p.zabel@...gutronix.de>, David Airlie <airlied@...il.com>,
Daniel Vetter <daniel@...ll.ch>, Singo Chang <singo.chang@...iatek.com>,
Nancy Lin <nancy.lin@...iatek.com>, Project_Global_Chrome_Upstream_Group@...iatek.com,
dri-devel@...ts.freedesktop.org, linux-mediatek@...ts.infradead.org,
linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH v3 2/2] drm/mediatek: change config_lock from spin_lock to spin_lock_irqsave
Hi Jason,
On Tue, Aug 27, 2024 at 10:58 PM Jason-JH.Lin via B4 Relay
<devnull+jason-jh.lin.mediatek.com@...nel.org> wrote:
>
> From: "Jason-JH.Lin" <jason-jh.lin@...iatek.com>
>
> Operations within spin_locks are limited to fast memory access and
> confirmation of minimum lock duration.
>
> Although using spin_lock with config_lock seems to ensure shorter hold
> times, it is safer to use spin_lock_irqsave due to potential deadlocks
> and nested interrupt scenarios in interrupt contexts.
>
> So change config_lock from spin_lock to spin_lock_irqsave.
>
> Fixes: 7f82d9c43879 ("drm/mediatek: Clear pending flag when cmdq packet is done")
The appropriate fix tag should be:
Fixes: 1bbb2be61bbb ("drm/mediatek: Fix missing configuration flags in
mtk_crtc_ddp_config()")
I don't know if it's worth resending though.
I guess it's up to CK if he's willing to replace it when picking up the patch.
But the change itself is valid, so after fixing that:
Reviewed-by: Fei Shao <fshao@...omium.org>
Regards,
Fei
> Signed-off-by: Jason-JH.Lin <jason-jh.lin@...iatek.com>
Powered by blists - more mailing lists