From: Cong Nguyen <congnt264@gmail.com>
To: Thierry Reding <thierry.reding@kernel.org>
Cc: Jonathan Hunter <jonathanh@nvidia.com>,
Sowjanya Komatineni <skomatineni@nvidia.com>,
Luca Ceresoli <luca.ceresoli@bootlin.com>,
Mauro Carvalho Chehab <mchehab@kernel.org>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
Svyatoslav Ryhel <clamor95@gmail.com>,
linux-media@vger.kernel.org, linux-tegra@vger.kernel.org,
linux-staging@lists.linux.dev, linux-kernel@vger.kernel.org,
Cong Nguyen <congnt264@gmail.com>
Subject: [PATCH] staging: media: tegra-video: fix syncpoint leak in tegra20 channel init
Date: Sun, 26 Jul 2026 17:04:08 +0700 [thread overview]
Message-ID: <20260726100408.3262831-1-congnt264@gmail.com> (raw)
tegra20_channel_host1x_syncpt_init() requests two host1x syncpoints. It
stored the first one (mw ack) into chan->mw_ack_sp[0] and only then
requested the second one (frame start). If the second request failed the
function returned without releasing the first syncpoint, leaking it: the
caller tegra_channel_init() bails out with a bare "return ret" on a
syncpt init failure and does not run its free_syncpts error path (that
path only covers failures after syncpoint init has fully succeeded), and
the channel is subsequently torn down via the kfree()-only path, so
nothing ever releases the orphaned syncpoint.
Release the already-acquired syncpoint on the error path, and only
publish the syncpoints into the channel once both have been successfully
requested, so chan->mw_ack_sp[0] never holds a released handle. This
matches the handling in tegra210_channel_host1x_syncpt_init().
Fixes: 0e2c4117c351 ("staging: media: tegra-video: add CSI support for Tegra20 and Tegra30")
Signed-off-by: Cong Nguyen <congnt264@gmail.com>
---
drivers/staging/media/tegra-video/tegra20.c | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/drivers/staging/media/tegra-video/tegra20.c b/drivers/staging/media/tegra-video/tegra20.c
index e513e6ccb776..f8acb6834a0b 100644
--- a/drivers/staging/media/tegra-video/tegra20.c
+++ b/drivers/staging/media/tegra-video/tegra20.c
@@ -429,12 +429,13 @@ static int tegra20_channel_host1x_syncpt_init(struct tegra_vi_channel *chan)
if (!out_sp)
return dev_err_probe(vi->dev, -EBUSY, "failed to request mw ack syncpoint\n");
- chan->mw_ack_sp[0] = out_sp;
-
fs_sp = host1x_syncpt_request(&vi->client, HOST1X_SYNCPT_CLIENT_MANAGED);
- if (!fs_sp)
+ if (!fs_sp) {
+ host1x_syncpt_put(out_sp);
return dev_err_probe(vi->dev, -EBUSY, "failed to request frame start syncpoint\n");
+ }
+ chan->mw_ack_sp[0] = out_sp;
chan->frame_start_sp[0] = fs_sp;
return 0;
--
2.25.1
next reply other threads:[~2026-07-26 10:04 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-26 10:04 Cong Nguyen [this message]
2026-07-27 5:13 ` [PATCH] staging: media: tegra-video: fix syncpoint leak in tegra20 channel init Mikko Perttunen
2026-07-27 8:10 ` Svyatoslav Ryhel
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20260726100408.3262831-1-congnt264@gmail.com \
--to=congnt264@gmail.com \
--cc=clamor95@gmail.com \
--cc=gregkh@linuxfoundation.org \
--cc=jonathanh@nvidia.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=linux-staging@lists.linux.dev \
--cc=linux-tegra@vger.kernel.org \
--cc=luca.ceresoli@bootlin.com \
--cc=mchehab@kernel.org \
--cc=skomatineni@nvidia.com \
--cc=thierry.reding@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox