linux-kernel - Re: [PATCH v1 2/3] remoteproc: qcom: q6v5: Do not report crash if SSR is disabled

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20221228162240.y2eaki6tzhkprxg3@builder.lan>
Date:   Wed, 28 Dec 2022 10:22:40 -0600
From:   Bjorn Andersson <andersson@...nel.org>
To:     Gokul krishna Krishnakumar <quic_gokukris@...cinc.com>
Cc:     Andy Gross <agross@...nel.org>,
        Konrad Dybcio <konrad.dybcio@...ainline.org>,
        Mathieu Poirier <mathieu.poirier@...aro.org>,
        linux-arm-msm@...r.kernel.org, linux-remoteproc@...r.kernel.org,
        linux-kernel@...r.kernel.org, Trilok Soni <quic_tsoni@...cinc.com>,
        Satya Durga Srinivasu Prabhala <quic_satyap@...cinc.com>,
        Rajendra Nayak <quic_rjendra@...cinc.com>,
        Elliot Berman <quic_eberman@...cinc.com>,
        Guru Das Srinagesh <quic_gurus@...cinc.com>
Subject: Re: [PATCH v1 2/3] remoteproc: qcom: q6v5: Do not report crash if
 SSR is disabled

On Mon, Sep 19, 2022 at 09:00:39AM -0700, Gokul krishna Krishnakumar wrote:
> In case recovery is disabled, do not report the rproc crash
> to the framework. If recovery is enabled after we start the
> crash handler we may end up in a weird state by informing
> clients of a crash twice, resulting in undefined behaviour.
> 

Afaict rproc_report_crash() schedules work which does nothing useful if
!rproc->recovery_disabled.

Can you please help me understand the issue you're seeing, and
preferably spell out what the resulting weird state is.

Thanks,
Bjorn

> Change-Id: If0d9bf5aa2c6f9e25adcefaca14b2de60fcb1a7a
> Signed-off-by: Gokul krishna Krishnakumar <quic_gokukris@...cinc.com>
> ---
>  drivers/remoteproc/qcom_q6v5.c | 10 ++++++----
>  1 file changed, 6 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/remoteproc/qcom_q6v5.c b/drivers/remoteproc/qcom_q6v5.c
> index 89f5384..1b9e1e1 100644
> --- a/drivers/remoteproc/qcom_q6v5.c
> +++ b/drivers/remoteproc/qcom_q6v5.c
> @@ -103,6 +103,8 @@ static void qcom_q6v5_crash_handler_work(struct work_struct *work)
>  
>  	mutex_lock(&rproc->lock);
>  
> +	rproc->state = RPROC_CRASHED;
> +
>  	list_for_each_entry_reverse(subdev, &rproc->subdevs, node) {
>  		if (subdev->stop)
>  			subdev->stop(subdev, true);
> @@ -139,8 +141,8 @@ static irqreturn_t q6v5_wdog_interrupt(int irq, void *data)
>  	q6v5->running = false;
>  	if (q6v5->rproc->recovery_disabled)
>  		schedule_work(&q6v5->crash_handler);
> -
> -	rproc_report_crash(q6v5->rproc, RPROC_WATCHDOG);
> +	else
> +		rproc_report_crash(q6v5->rproc, RPROC_WATCHDOG);
>  
>  	return IRQ_HANDLED;
>  }
> @@ -163,8 +165,8 @@ static irqreturn_t q6v5_fatal_interrupt(int irq, void *data)
>  	q6v5->running = false;
>  	if (q6v5->rproc->recovery_disabled)
>  		schedule_work(&q6v5->crash_handler);
> -
> -	rproc_report_crash(q6v5->rproc, RPROC_FATAL_ERROR);
> +	else
> +		rproc_report_crash(q6v5->rproc, RPROC_FATAL_ERROR);
>  
>  	return IRQ_HANDLED;
>  }
> -- 
> 2.7.4
>