linux-kernel - Re: [PATCH] drm/vc4: fix ref count leak in vc4_dsi_encoder

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>] [day] [month] [year] [list]

Message-ID: <a81cacf1-88e5-18d1-9d01-8e8d32f6f0a7@web.de>
Date:   Sun, 14 Jun 2020 14:16:50 +0200
From:   Markus Elfring <Markus.Elfring@....de>
To:     Navid Emamdoost <navid.emamdoost@...il.com>,
        dri-devel@...ts.freedesktop.org
Cc:     Navid Emamdoost <emamd001@....edu>, Kangjie Lu <kjlu@....edu>,
        Stephen McCamant <smccaman@....edu>,
        Qiushi Wu <wu000273@....edu>,
        Andy Shevchenko <andriy.shevchenko@...ux.intel.com>,
        Daniel Vetter <daniel@...ll.ch>,
        David Airlie <airlied@...ux.ie>, Eric Anholt <eric@...olt.net>,
        kernel-janitors@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] drm/vc4: fix ref count leak in vc4_dsi_encoder_enable

> in vc4_dsi_encoder_enable, the call to pm_runtime_get_sync increments
> the counter even in case of failure, leading to incorrect
> ref count. In case of failure, decrement the ref count before returning.

* Can the term “reference count” become relevant also for this commit message
  besides other possible adjustments?

* Can it be nicer to combine proposed updates for this software module
  as a patch series (with a cover letter)?

* Would you like to add the tag “Fixes”?


…
> +++ b/drivers/gpu/drm/vc4/vc4_dsi.c
…
> @@ -1088,6 +1088,8 @@  static void vc4_dsi_encoder_enable(struct drm_encoder *encoder)
>  		dev_info(&dsi->pdev->dev, "DSI regs after:\n");
>  		drm_print_regset32(&p, &dsi->regset);
>  	}
> +out:
> +	pm_runtime_put(dev);
>  }
…

* Perhaps use the label “put_runtime” instead?

* Do you propose to perform an additional function call always
  (and not only according to failure cases)?

* How do you think about calling the function “pm_runtime_put_noidle”?

Regards,
Markus