[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CABb+yY3P_iFnKamY=7z5rnNW+BQW4-vAbeS=S33BpqCKR1Rhuw@mail.gmail.com>
Date: Mon, 1 Feb 2021 00:28:29 -0600
From: Jassi Brar <jassisinghbrar@...il.com>
To: haidong yao <yaohaidong369@...il.com>
Cc: natechancellor@...il.com,
Nick Desaulniers <ndesaulniers@...gle.com>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
clang-built-linux@...glegroups.com,
Orson Zhai <orsonzhai@...il.com>,
Chunyan Zhang <zhang.lyra@...il.com>
Subject: Re: [PATCH] add chan->cl check in mbox_chan_received_data()
On Thu, Jan 7, 2021 at 5:53 AM haidong yao <yaohaidong369@...il.com> wrote:
>
> Hi Jassi Brar
>
> Thank you very much for your reply.
>
> Look at the function sprd_mbox_outbox_isr .
>
> Chan is !NULL.
>
> chan->cl is NULL when the client driver not loaded, the controller
> driver don't know the client driver loaded successfully, so, I do not
> use mbox_free_channel.
>
> Here,How do you know chan->cl is ok?
>
The channel is supposed to get/send data _only_ if it is being used by a client.
Which you can mark in .startup() and .shutdown().
Checking for chan->cl will make your symptoms disappear but that is
not the right fix for the issue.
The right fix is to EITHER not cause Rx/Tx interrupt on a channel not
being used, OR not send it to upper layers.
thanks.
Powered by blists - more mailing lists