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] [thread-next>] [day] [month] [year] [list]
Date: Tue, 18 Jun 2024 21:55:06 +0300
From: Dmitry Baryshkov <dmitry.baryshkov@...aro.org>
To: Sibi Sankar <quic_sibis@...cinc.com>
Cc: andersson@...nel.org, konrad.dybcio@...aro.org, djakov@...nel.org, 
	robh+dt@...nel.org, krzysztof.kozlowski+dt@...aro.org, 
	srinivas.kandagatla@...aro.org, linux-kernel@...r.kernel.org, linux-arm-msm@...r.kernel.org, 
	devicetree@...r.kernel.org, linux-pm@...r.kernel.org, quic_rgottimu@...cinc.com, 
	quic_kshivnan@...cinc.com, conor+dt@...nel.org, abel.vesa@...aro.org
Subject: Re: [PATCH V2 2/3] soc: qcom: icc-bwmon: Allow for interrupts to be
 shared across instances

On Tue, Jun 18, 2024 at 09:13:05PM GMT, Sibi Sankar wrote:
> The multiple BWMONv4 instances available on the X1E80100 SoC use the
> same interrupt number. Mark them are shared to allow for re-use across
> instances. Handle the ensuing race introduced by relying on bwmon_disable
> to disable the interrupt and coupled with explicit request/free irqs.
> 
> Signed-off-by: Sibi Sankar <quic_sibis@...cinc.com>
> ---
> 
> v2:
> * Use explicit request/free irq and add comments regarding the race
>   introduced when adding the IRQF_SHARED flag. [Krzysztof/Dmitry]
> 
>  drivers/soc/qcom/icc-bwmon.c | 14 +++++++++++---
>  1 file changed, 11 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/soc/qcom/icc-bwmon.c b/drivers/soc/qcom/icc-bwmon.c
> index fb323b3364db..4a4e28b41509 100644
> --- a/drivers/soc/qcom/icc-bwmon.c
> +++ b/drivers/soc/qcom/icc-bwmon.c
> @@ -781,9 +781,10 @@ static int bwmon_probe(struct platform_device *pdev)
>  	bwmon->dev = dev;
>  
>  	bwmon_disable(bwmon);
> -	ret = devm_request_threaded_irq(dev, bwmon->irq, bwmon_intr,
> -					bwmon_intr_thread,
> -					IRQF_ONESHOT, dev_name(dev), bwmon);
> +
> +	/* SoCs with multiple cpu-bwmon instances can end up using a shared interrupt line */

... using devm_ here might result in the IRQ handler being executed
after bwmon_disable in bwmon_remove()

> +	ret = request_threaded_irq(bwmon->irq, bwmon_intr, bwmon_intr_thread,
> +				   IRQF_ONESHOT | IRQF_SHARED, dev_name(dev), bwmon);
>  	if (ret)
>  		return dev_err_probe(dev, ret, "failed to request IRQ\n");
>  
> @@ -798,6 +799,13 @@ static void bwmon_remove(struct platform_device *pdev)
>  	struct icc_bwmon *bwmon = platform_get_drvdata(pdev);
>  
>  	bwmon_disable(bwmon);
> +
> +	/*
> +	 * Handle the race introduced, when dealing with multiple bwmon instances
> +	 * using a shared interrupt line, by relying on bwmon_disable to disable
> +	 * the interrupt and followed by an explicit free.
> +	 */

This sounds more like a part of the commit message. The comment before
request_threaded_irq() should be enough.

> +	free_irq(bwmon->irq, bwmon);
>  }
>  
>  static const struct icc_bwmon_data msm8998_bwmon_data = {
> -- 
> 2.34.1
> 

-- 
With best wishes
Dmitry

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ