| 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
| ||
|
Message-ID: <7605f778-6b20-47e4-bd65-7a0d85fff736@kernel.org>
Date: Thu, 7 Aug 2025 22:51:27 +0200
From: Hans Verkuil <hverkuil+cisco@...nel.org>
To: Laurent Pinchart <laurent.pinchart@...asonboard.com>
Cc: Jacopo Mondi <jacopo.mondi@...asonboard.com>,
Mauro Carvalho Chehab <mchehab@...nel.org>, Devarsh Thakkar
<devarsht@...com>, Benoit Parrot <bparrot@...com>,
Hans Verkuil <hverkuil@...nel.org>, Mike Isely <isely@...ox.com>,
Hans de Goede <hansg@...nel.org>,
Parthiban Veerasooran <parthiban.veerasooran@...rochip.com>,
Christian Gromm <christian.gromm@...rochip.com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>, Alex Shi
<alexs@...nel.org>, Yanteng Si <si.yanteng@...ux.dev>,
Dongliang Mu <dzm91@...t.edu.cn>, Jonathan Corbet <corbet@....net>,
Tomasz Figa <tfiga@...omium.org>, Marek Szyprowski
<m.szyprowski@...sung.com>, Andy Walls <awalls@...metrocast.net>,
Michael Tretter <m.tretter@...gutronix.de>,
Pengutronix Kernel Team <kernel@...gutronix.de>,
Bin Liu <bin.liu@...iatek.com>, Matthias Brugger <matthias.bgg@...il.com>,
AngeloGioacchino Del Regno <angelogioacchino.delregno@...labora.com>,
Dmitry Osipenko <digetx@...il.com>, Thierry Reding
<thierry.reding@...il.com>, Jonathan Hunter <jonathanh@...dia.com>,
Mirela Rabulea <mirela.rabulea@....com>, Shawn Guo <shawnguo@...nel.org>,
Sascha Hauer <s.hauer@...gutronix.de>, Fabio Estevam <festevam@...il.com>,
Kieran Bingham <kieran.bingham+renesas@...asonboard.com>,
Michal Simek <michal.simek@....com>, Ming Qian <ming.qian@....com>,
Zhou Peng <eagle.zhou@....com>, Xavier Roumegue
<xavier.roumegue@....nxp.com>, Philipp Zabel <p.zabel@...gutronix.de>,
Vikash Garodia <quic_vgarodia@...cinc.com>,
Dikshita Agarwal <quic_dikshita@...cinc.com>,
Abhinav Kumar <abhinav.kumar@...ux.dev>,
Bryan O'Donoghue <bryan.odonoghue@...aro.org>,
Sylwester Nawrocki <sylvester.nawrocki@...il.com>,
Jernej Skrabec <jernej.skrabec@...il.com>, Chen-Yu Tsai <wens@...e.org>,
Samuel Holland <samuel@...lland.org>,
Daniel Almeida <daniel.almeida@...labora.com>,
Neil Armstrong <neil.armstrong@...aro.org>,
Kevin Hilman <khilman@...libre.com>, Jerome Brunet <jbrunet@...libre.com>,
Martin Blumenstingl <martin.blumenstingl@...glemail.com>,
Nas Chung <nas.chung@...psnmedia.com>,
Jackson Lee <jackson.lee@...psnmedia.com>,
Minghsiu Tsai <minghsiu.tsai@...iatek.com>,
Houlong Wei <houlong.wei@...iatek.com>,
Andrew-CT Chen <andrew-ct.chen@...iatek.com>,
Tiffany Lin <tiffany.lin@...iatek.com>,
Yunfei Dong <yunfei.dong@...iatek.com>,
Geert Uytterhoeven <geert+renesas@...der.be>,
Magnus Damm <magnus.damm@...il.com>,
Mikhail Ulyanov <mikhail.ulyanov@...entembedded.com>,
Jacob Chen <jacob-chen@...wrt.com>,
Ezequiel Garcia <ezequiel@...guardiasur.com.ar>,
Heiko Stuebner <heiko@...ech.de>,
Detlev Casanova <detlev.casanova@...labora.com>,
Krzysztof Kozlowski <krzk@...nel.org>, Alim Akhtar
<alim.akhtar@...sung.com>, Sylwester Nawrocki <s.nawrocki@...sung.com>,
Łukasz Stelmach <l.stelmach@...sung.com>,
Andrzej Pietrasiewicz <andrzejtp2010@...il.com>,
Jacek Anaszewski <jacek.anaszewski@...il.com>,
Andrzej Hajda <andrzej.hajda@...el.com>,
Fabien Dessenne <fabien.dessenne@...s.st.com>,
Hugues Fruchet <hugues.fruchet@...s.st.com>,
Jean-Christophe Trotin <jean-christophe.trotin@...s.st.com>,
Maxime Coquelin <mcoquelin.stm32@...il.com>,
Alexandre Torgue <alexandre.torgue@...s.st.com>,
Nicolas Dufresne <nicolas.dufresne@...labora.com>,
Benjamin Gaignard <benjamin.gaignard@...labora.com>,
Steve Longerbeam <slongerbeam@...il.com>, Maxime Ripard
<mripard@...nel.org>, Paul Kocialkowski <paulk@...-base.io>,
Niklas Söderlund <niklas.soderlund@...natech.se>,
Robert Foss <rfoss@...nel.org>, Todor Tomov <todor.too@...il.com>,
Vladimir Zapolskiy <vladimir.zapolskiy@...aro.org>,
Corentin Labbe <clabbe@...libre.com>,
Sakari Ailus <sakari.ailus@...ux.intel.com>,
Bingbu Cao <bingbu.cao@...el.com>, Tianshu Qiu <tian.shu.qiu@...el.com>,
Stanislaw Gruszka <stanislaw.gruszka@...ux.intel.com>,
linux-media@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-staging@...ts.linux.dev, linux-doc@...r.kernel.org,
linux-arm-kernel@...ts.infradead.org, linux-mediatek@...ts.infradead.org,
linux-tegra@...r.kernel.org, imx@...ts.linux.dev,
linux-renesas-soc@...r.kernel.org, linux-arm-msm@...r.kernel.org,
linux-samsung-soc@...r.kernel.org, linux-sunxi@...ts.linux.dev,
linux-usb@...r.kernel.org, linux-amlogic@...ts.infradead.org,
linux-rockchip@...ts.infradead.org,
linux-stm32@...md-mailman.stormreply.com, mjpeg-users@...ts.sourceforge.net
Subject: Re: [PATCH 27/65] media: Reset file->private_data to NULL in
v4l2_fh_del()
On 07/08/2025 22:25, Laurent Pinchart wrote:
> On Thu, Aug 07, 2025 at 08:00:06PM +0300, Laurent Pinchart wrote:
>> On Thu, Aug 07, 2025 at 11:50:07AM +0300, Laurent Pinchart wrote:
>>> On Wed, Aug 06, 2025 at 02:45:14PM +0200, Hans Verkuil wrote:
>>>> On 02/08/2025 11:22, Jacopo Mondi wrote:
>>>>> From: Laurent Pinchart <laurent.pinchart@...asonboard.com>
>>>>>
>>>>> Multiple drivers that use v4l2_fh and call v4l2_fh_del() manually reset
>>>>> the file->private_data pointer to NULL in their video device .release()
>>>>> file operation handler. Move the code to the v4l2_fh_del() function to
>>>>> avoid direct access to file->private_data in drivers. This requires
>>>>> adding a file pointer argument to the function.
>>>>>
>>>>> Changes to drivers have been generated with the following coccinelle
>>>>> semantic patch:
>>>>>
>>>>> @@
>>>>> expression fh;
>>>>> identifier filp;
>>>>> identifier release;
>>>>> type ret;
>>>>> @@
>>>>> ret release(..., struct file *filp, ...)
>>>>> {
>>>>> <...
>>>>> - filp->private_data = NULL;
>>>>> ...
>>>>> - v4l2_fh_del(fh);
>>>>> + v4l2_fh_del(fh, filp);
>>>>> ...>
>>>>> }
>>>>>
>>>>> @@
>>>>> expression fh;
>>>>> identifier filp;
>>>>> identifier release;
>>>>> type ret;
>>>>> @@
>>>>> ret release(..., struct file *filp, ...)
>>>>> {
>>>>> <...
>>>>> - v4l2_fh_del(fh);
>>>>> + v4l2_fh_del(fh, filp);
>>>>> ...
>>>>> - filp->private_data = NULL;
>>>>> ...>
>>>>> }
>>>>>
>>>>> @@
>>>>> expression fh;
>>>>> identifier filp;
>>>>> identifier release;
>>>>> type ret;
>>>>> @@
>>>>> ret release(..., struct file *filp, ...)
>>>>> {
>>>>> <...
>>>>> - v4l2_fh_del(fh);
>>>>> + v4l2_fh_del(fh, filp);
>>>>> ...>
>>>>> }
>>>>>
>>>>> Manual changes have been applied to Documentation/ to update the usage
>>>>> patterns, to drivers/media/v4l2-core/v4l2-fh.c to update the
>>>>> v4l2_fh_del() prototype and reset file->private_data, and to
>>>>> include/media/v4l2-fh.h to update the v4l2_fh_del() function prototype
>>>>> and its documentation.
>>>>>
>>>>> Additionally, white space issues have been fixed manually in
>>>>> drivers/usb/gadget/function/uvc_v4l2.c
>>>>>
>>>>> Signed-off-by: Laurent Pinchart <laurent.pinchart@...asonboard.com>
>>>>> Signed-off-by: Jacopo Mondi <jacopo.mondi@...asonboard.com>
>>>>> ---
>>>>> Documentation/driver-api/media/v4l2-fh.rst | 4 ++--
>>>>> Documentation/translations/zh_CN/video4linux/v4l2-framework.txt | 4 ++--
>>>>> drivers/media/pci/cx18/cx18-fileops.c | 4 ++--
>>>>> drivers/media/pci/ivtv/ivtv-fileops.c | 4 ++--
>>>>> drivers/media/pci/saa7164/saa7164-encoder.c | 2 +-
>>>>> drivers/media/pci/saa7164/saa7164-vbi.c | 2 +-
>>>>> drivers/media/platform/allegro-dvt/allegro-core.c | 2 +-
>>>>> drivers/media/platform/amlogic/meson-ge2d/ge2d.c | 2 +-
>>>>> drivers/media/platform/amphion/vpu_v4l2.c | 4 ++--
>>>>> drivers/media/platform/chips-media/coda/coda-common.c | 4 ++--
>>>>> drivers/media/platform/chips-media/wave5/wave5-helper.c | 2 +-
>>>>> drivers/media/platform/imagination/e5010-jpeg-enc.c | 4 ++--
>>>>> drivers/media/platform/m2m-deinterlace.c | 2 +-
>>>>> drivers/media/platform/mediatek/jpeg/mtk_jpeg_core.c | 4 ++--
>>>>> drivers/media/platform/mediatek/mdp/mtk_mdp_m2m.c | 4 ++--
>>>>> drivers/media/platform/mediatek/mdp3/mtk-mdp3-m2m.c | 4 ++--
>>>>> .../media/platform/mediatek/vcodec/decoder/mtk_vcodec_dec_drv.c | 4 ++--
>>>>> .../media/platform/mediatek/vcodec/encoder/mtk_vcodec_enc_drv.c | 4 ++--
>>>>> drivers/media/platform/nvidia/tegra-vde/v4l2.c | 2 +-
>>>>> drivers/media/platform/nxp/dw100/dw100.c | 2 +-
>>>>> drivers/media/platform/nxp/imx-jpeg/mxc-jpeg.c | 4 ++--
>>>>> drivers/media/platform/nxp/imx-pxp.c | 2 +-
>>>>> drivers/media/platform/nxp/imx8-isi/imx8-isi-m2m.c | 2 +-
>>>>> drivers/media/platform/nxp/mx2_emmaprp.c | 2 +-
>>>>> drivers/media/platform/qcom/iris/iris_vidc.c | 3 +--
>>>>> drivers/media/platform/qcom/venus/core.c | 2 +-
>>>>> drivers/media/platform/renesas/rcar_fdp1.c | 2 +-
>>>>> drivers/media/platform/renesas/rcar_jpu.c | 4 ++--
>>>>> drivers/media/platform/renesas/vsp1/vsp1_video.c | 2 +-
>>>>> drivers/media/platform/rockchip/rga/rga.c | 2 +-
>>>>> drivers/media/platform/rockchip/rkvdec/rkvdec.c | 2 +-
>>>>> drivers/media/platform/samsung/exynos-gsc/gsc-m2m.c | 4 ++--
>>>>> drivers/media/platform/samsung/exynos4-is/fimc-m2m.c | 4 ++--
>>>>> drivers/media/platform/samsung/s5p-g2d/g2d.c | 2 +-
>>>>> drivers/media/platform/samsung/s5p-jpeg/jpeg-core.c | 4 ++--
>>>>> drivers/media/platform/samsung/s5p-mfc/s5p_mfc.c | 4 ++--
>>>>> drivers/media/platform/st/sti/bdisp/bdisp-v4l2.c | 4 ++--
>>>>> drivers/media/platform/st/sti/delta/delta-v4l2.c | 4 ++--
>>>>> drivers/media/platform/st/sti/hva/hva-v4l2.c | 4 ++--
>>>>> drivers/media/platform/st/stm32/dma2d/dma2d.c | 2 +-
>>>>> drivers/media/platform/sunxi/sun8i-di/sun8i-di.c | 2 +-
>>>>> drivers/media/platform/sunxi/sun8i-rotate/sun8i_rotate.c | 2 +-
>>>>> drivers/media/platform/ti/omap3isp/ispvideo.c | 5 ++---
>>>>> drivers/media/platform/ti/vpe/vpe.c | 2 +-
>>>>> drivers/media/platform/verisilicon/hantro_drv.c | 4 ++--
>>>>> drivers/media/test-drivers/vicodec/vicodec-core.c | 2 +-
>>>>> drivers/media/test-drivers/vim2m.c | 2 +-
>>>>> drivers/media/test-drivers/visl/visl-core.c | 2 +-
>>>>> drivers/media/usb/pvrusb2/pvrusb2-v4l2.c | 3 +--
>>>>> drivers/media/v4l2-core/v4l2-fh.c | 7 ++++---
>>>>> drivers/media/v4l2-core/v4l2-subdev.c | 5 ++---
>>>>> drivers/staging/media/imx/imx-media-csc-scaler.c | 4 ++--
>>>>> drivers/staging/media/meson/vdec/vdec.c | 2 +-
>>>>> drivers/staging/media/sunxi/cedrus/cedrus.c | 2 +-
>>>>> drivers/staging/most/video/video.c | 4 ++--
>>>>> drivers/usb/gadget/function/uvc_v4l2.c | 3 +--
>>>>> include/media/v4l2-fh.h | 5 ++++-
>>>>> 57 files changed, 89 insertions(+), 90 deletions(-)
>>>>>
>>>>
>>>> <snip>
>>>>
>>>>> diff --git a/drivers/media/v4l2-core/v4l2-fh.c b/drivers/media/v4l2-core/v4l2-fh.c
>>>>> index b59b1084d8cdf1b62da12879e21dbe56c2109648..df3ba9d4674bd25626cfcddc2d0cb28c233e3cc3 100644
>>>>> --- a/drivers/media/v4l2-core/v4l2-fh.c
>>>>> +++ b/drivers/media/v4l2-core/v4l2-fh.c
>>>>> @@ -67,7 +67,7 @@ int v4l2_fh_open(struct file *filp)
>>>>> }
>>>>> EXPORT_SYMBOL_GPL(v4l2_fh_open);
>>>>>
>>>>> -void v4l2_fh_del(struct v4l2_fh *fh)
>>>>> +void v4l2_fh_del(struct v4l2_fh *fh, struct file *filp)
>>>>
>>>> Instead of adding a second argument, perhaps it is better to
>>>> just provide the filp pointer. After all, you can get the v4l2_fh
>>>> from filp->private_data.
>>>>
>>>> It simplifies the code a bit.
>>>
>>> That's an interesting idea. I'll give it a try.
>>
>> We end up with code like (e.g. in v4l2_fh_release(), with similar
>> constructs in lots of drivers)
>>
>> if (fh) {
>> v4l2_fh_del(filp);
>> v4l2_fh_exit(fh);
>> kfree(fh);
>> }
>>
>> compared to
>>
>> if (fh) {
>> v4l2_fh_del(fh, filp);
>> v4l2_fh_exit(fh);
>> kfree(fh);
>> }
>>
>> with the existing patch. I find the fact that v4l2_fh_del() takes a
>> different pointer than v4l2_fh_exit() a bit disturbing. If you think
>> it's better I'll drop the fh argument in v2.
>
> I gave it a try, and looking at the function prototype, its
> documentation, the imbalance with v4l2_fh_add(), and the code in the
> callers, I think keeping both arguments would look cleaner. Please tell
> me if you feel strongly about this, I can still submit a patch to drop
> the argument. It can very easily be scripted with coccinelle and doesn't
> conflict with the rest of the series, so it could also be done later.
Looking at all the drivers that call v4l2_fh_del/exit I always see v4l2_fh_del()
directly followed by v4l2_fh_exit(). I think it would make a lot of sense to just
combine the two as a single function: v4l2_fh_del_exit(filp).
That simplifies the code and solves the imbalance.
Regards,
Hans
>
>>>>> {
>>>>> unsigned long flags;
>>>>>
>>>>> @@ -75,6 +75,8 @@ void v4l2_fh_del(struct v4l2_fh *fh)
>>>>> list_del_init(&fh->list);
>>>>> spin_unlock_irqrestore(&fh->vdev->fh_lock, flags);
>>>>> v4l2_prio_close(fh->vdev->prio, fh->prio);
>>>>> +
>>>>> + filp->private_data = NULL;
>>>>> }
>>>>> EXPORT_SYMBOL_GPL(v4l2_fh_del);
>>>>>
>>>>> @@ -94,10 +96,9 @@ int v4l2_fh_release(struct file *filp)
>>>>> struct v4l2_fh *fh = file_to_v4l2_fh(filp);
>>>>>
>>>>> if (fh) {
>>>>> - v4l2_fh_del(fh);
>>>>> + v4l2_fh_del(fh, filp);
>>>>> v4l2_fh_exit(fh);
>>>>> kfree(fh);
>>>>> - filp->private_data = NULL;
>>>>> }
>>>>> return 0;
>>>>> }
>>>>
>>>> <snip>
>>>>
>>>>> diff --git a/include/media/v4l2-fh.h b/include/media/v4l2-fh.h
>>>>> index d8fcf49f10e09452b73499f4a9bd1285bc2835a5..5e4c761635120608e0b588e0b0daf63e69588d38 100644
>>>>> --- a/include/media/v4l2-fh.h
>>>>> +++ b/include/media/v4l2-fh.h
>>>>> @@ -114,12 +114,15 @@ int v4l2_fh_open(struct file *filp);
>>>>> * v4l2_fh_del - Remove file handle from the list of file handles.
>>>>> *
>>>>> * @fh: pointer to &struct v4l2_fh
>>>>> + * @filp: pointer to &struct file associated with @fh
>>>>> + *
>>>>> + * The function resets filp->private_data to NULL.
>>>>> *
>>>>> * .. note::
>>>>> * Must be called in v4l2_file_operations->release\(\) handler if the driver
>>>>> * uses &struct v4l2_fh.
>>>>> */
>>>>> -void v4l2_fh_del(struct v4l2_fh *fh);
>>>>> +void v4l2_fh_del(struct v4l2_fh *fh, struct file *filp);
>>>>>
>>>>> /**
>>>>> * v4l2_fh_exit - Release resources related to a file handle.
>
Powered by blists - more mailing lists