mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "A. Sverdlin" <alexander.sverdlin@siemens.com>
To: linux-gpio@vger.kernel.org
Cc: Alexander Sverdlin <alexander.sverdlin@siemens.com>,
	Tony Lindgren <tony@atomide.com>,
	Haojian Zhuang <haojian.zhuang@linaro.org>,
	Linus Walleij <linusw@kernel.org>,
	Lokesh Vutla <lokeshvutla@ti.com>, Nishanth Menon <nm@ti.com>,
	Rob Herring <robh@kernel.org>,
	linux-arm-kernel@lists.infradead.org, linux-omap@vger.kernel.org,
	linux-kernel@vger.kernel.org, stable@vger.kernel.org
Subject: [PATCH 1/2] pinctrl: single: don't leak group/function tables on consumer re-probe
Date: Thu,  1 Oct 2026 10:52:10 +0200	[thread overview]
Message-ID: <20261001085217.1778645-2-alexander.sverdlin@siemens.com> (raw)
In-Reply-To: <20261001085217.1778645-1-alexander.sverdlin@siemens.com>

From: Alexander Sverdlin <alexander.sverdlin@siemens.com>

For every mapping pcs_parse_{one,bits}_in_pinctrl_entry() builds a
pcs_function, a value table, a pin array and a pingroup-name array and
registers them via pcs_add_function()/pinctrl_generic_add_group(). The
core deduplicates groups and functions by name and keeps the first
instance for the controller's lifetime, so the tables built for any later
mapping of the same node - e.g. each consumer re-probe - are dropped by
the core and leaked.

Free the redundant copies once the function turns out to be a duplicate;
the group is deduplicated together with it. Don't remove the deduplicated
instance from the core: its selector is a plain counter (num_groups /
num_functions), so removing any but the last entry corrupts the selector
space.

kmemleak stays silent because the leaked tables remain reachable from the
controller's devres list.

Fixes: 8b8b091bf07f ("pinctrl: Add one-register-per-pin type device tree based pinctrl driver")
Cc: stable@vger.kernel.org
Signed-off-by: Alexander Sverdlin <alexander.sverdlin@siemens.com>
---
 drivers/pinctrl/pinctrl-single.c | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/drivers/pinctrl/pinctrl-single.c b/drivers/pinctrl/pinctrl-single.c
index e0eb7240a9859..e75e32723100d 100644
--- a/drivers/pinctrl/pinctrl-single.c
+++ b/drivers/pinctrl/pinctrl-single.c
@@ -1097,6 +1097,19 @@ static int pcs_parse_one_pinctrl_entry(struct pcs_device *pcs,
 	} else {
 		*num_maps = 1;
 	}
+
+	/*
+	 * The core deduplicates groups/functions by name and keeps the first
+	 * for the controller's lifetime; drop our now-redundant copies when the
+	 * node is mapped again (consumer re-probe). The group is deduplicated
+	 * together with the function, so testing the latter is enough.
+	 */
+	if (pinmux_generic_get_function(pcs->pctl, fsel)->data != function) {
+		devm_kfree(pcs->dev, pins);
+		devm_kfree(pcs->dev, vals);
+		devm_kfree(pcs->dev, function);
+		devm_kfree(pcs->dev, pgnames);
+	}
 	mutex_unlock(&pcs->mutex);
 
 	return 0;
@@ -1235,6 +1248,14 @@ static int pcs_parse_bits_in_pinctrl_entry(struct pcs_device *pcs,
 	(*map)->data.mux.function = np->name;
 
 	*num_maps = 1;
+
+	/* See pcs_parse_one_pinctrl_entry() for the deduplication rationale. */
+	if (pinmux_generic_get_function(pcs->pctl, fsel)->data != function) {
+		devm_kfree(pcs->dev, pins);
+		devm_kfree(pcs->dev, vals);
+		devm_kfree(pcs->dev, function);
+		devm_kfree(pcs->dev, pgnames);
+	}
 	mutex_unlock(&pcs->mutex);
 
 	return 0;
-- 
2.55.0


  reply	other threads:[~2026-10-01  8:52 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-01  8:52 [PATCH 0/2] pinctrl: fix group/function table leaks " A. Sverdlin
2026-10-01  8:52 ` A. Sverdlin [this message]
2026-10-01  8:52 ` [PATCH 2/2] pinctrl: ti-iodelay: fix memory " A. Sverdlin
2026-11-04 12:30 ` [PATCH 0/2] pinctrl: fix group/function table " Linus Walleij
2026-11-04 12:59   ` Sverdlin, Alexander

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=20261001085217.1778645-2-alexander.sverdlin@siemens.com \
    --to=alexander.sverdlin@siemens.com \
    --cc=haojian.zhuang@linaro.org \
    --cc=linusw@kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-gpio@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-omap@vger.kernel.org \
    --cc=lokeshvutla@ti.com \
    --cc=nm@ti.com \
    --cc=robh@kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=tony@atomide.com \
    /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

all inboxes | Powered by JetHome®