From: Nickolai Zeldovich <nickolai@csail.mit.edu>
To: Stephen Boyd <sboyd@kernel.org>,
Brian Masney <bmasney+clk@redhat.com>,
Jerome Brunet <jbrunet+clk@baylibre.com>
Cc: linux-clk@vger.kernel.org, linux-kernel@vger.kernel.org,
Nickolai Zeldovich <nickolai@csail.mit.edu>
Subject: [PATCH v2] clk: Free the parent map only in __clk_release()
Date: Fri, 23 Oct 2026 23:39:04 -0400 [thread overview]
Message-ID: <20261024033904.830766-1-nickolai@csail.mit.edu> (raw)
In-Reply-To: <20261016164050.790594-1-nickolai@csail.mit.edu>
Since commit 9d05ae531c2c ("clk: Initialize struct clk_core kref
earlier"), every failure in __clk_register() frees the parent map
through __clk_release(). clk_core_populate_parent_map() still cleans up
after itself on failure, so an allocation failure while registering a
clock with parents either dereferences a NULL map (core->parents is NULL
while core->num_parents is set) or frees the parent names and the map a
second time.
Leave the cleanup to __clk_release(): return the error without freeing
anything, and have clk_core_free_parent_map() check for a missing map
instead of num_parents == 0. The map is zero-allocated and
kfree_const(NULL) is a no-op, so a partially filled map is freed
correctly.
Fixes: 9d05ae531c2c ("clk: Initialize struct clk_core kref earlier")
Cc: stable@vger.kernel.org
Reviewed-by: Jerome Brunet <jbrunet@baylibre.com>
Signed-off-by: Nickolai Zeldovich <nickolai@csail.mit.edu>
---
Notes:
Changes in v2:
- Shortened the commit message (Jerome, Brian).
- Added Cc: stable (Brian) and Jerome's Reviewed-by.
- No code change; rebased onto current master.
v1: https://lore.kernel.org/r/20261016164050.790594-1-nickolai@csail.mit.edu
Developed with the assistance of an LLM.
Reproduced on riscv64 under QEMU by rebinding a fixed-factor clock
provider through its sysfs bind file with failslab aimed at the parent
map allocation: without the patch the kernel oopses with a NULL pointer
dereference in __clk_release(); with it the bind fails with -ENOMEM and
"failed to register clk" is logged. The second failure path (a parent
name copy failing) was reproduced only in a userspace transcription of
the four functions under AddressSanitizer. The clk KUnit suites pass on
x86_64 and riscv64.
drivers/clk/clk.c | 11 ++---------
1 file changed, 2 insertions(+), 9 deletions(-)
diff --git a/drivers/clk/clk.c b/drivers/clk/clk.c
index fef87167a60b5..c32c3e9d62bdb 100644
--- a/drivers/clk/clk.c
+++ b/drivers/clk/clk.c
@@ -4353,15 +4353,8 @@ static int clk_core_populate_parent_map(struct clk_core *core,
WARN(1, "Must specify parents if num_parents > 0\n");
}
- if (ret) {
- do {
- kfree_const(parents[i].name);
- kfree_const(parents[i].fw_name);
- } while (--i >= 0);
- kfree(parents);
-
+ if (ret)
return ret;
- }
}
return 0;
@@ -4371,7 +4364,7 @@ static void clk_core_free_parent_map(struct clk_core *core)
{
int i = core->num_parents;
- if (!core->num_parents)
+ if (!core->parents)
return;
while (--i >= 0) {
base-commit: 4da9dedf3fc59137fd86d664693776aa38cb97f5
--
2.53.0
prev parent reply other threads:[~2026-10-24 3:39 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-10-16 16:40 [PATCH v1] " Nickolai Zeldovich
2026-10-19 8:59 ` Jerome Brunet
2026-10-19 22:16 ` Brian Masney
2026-10-24 3:39 ` Nickolai Zeldovich [this message]
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=20261024033904.830766-1-nickolai@csail.mit.edu \
--to=nickolai@csail.mit.edu \
--cc=bmasney+clk@redhat.com \
--cc=jbrunet+clk@baylibre.com \
--cc=linux-clk@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=sboyd@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
all inboxes | Powered by JetHome®