From: sashiko-bot@kernel.org
To: "Ayush Singh" <ayush@beagleboard.org>
Cc: robh@kernel.org, conor+dt@kernel.org, devicetree@vger.kernel.org,
ojeda@kernel.org
Subject: Re: [PATCH 1/7] greybus: connection: Export gb_connection_get() and gb_connection_put()
Date: Thu, 20 Aug 2026 09:41:42 +0000 [thread overview]
Message-ID: <20260820094142.BBCEF1F000E9@smtp.kernel.org> (raw)
In-Reply-To: <20260820-gb-uart-transport-v1-1-282da14ab7b7@beagleboard.org>
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] Exporting `gb_connection_get/put` allows connections to outlive their parent objects, causing a Use-After-Free in tracepoints.
--
--- Patch [1]: [PATCH 1/7] greybus: connection: Export gb_connection_get() and gb_connection_put() ---
commit d7d6f3dcc2daeb414806a0f2b9dbabd440f8ec1e
Author: Ayush Singh <ayush@beagleboard.org>
greybus: connection: Export gb_connection_get() and gb_connection_put()
This commit exports gb_connection_get() and gb_connection_put() to allow
modular drivers to manage connection references. It changes them to non-static
and adds EXPORT_SYMBOL_GPL declarations.
> diff --git a/drivers/greybus/connection.c b/drivers/greybus/connection.c
> index bd04485decb3d..f8fda22bc4667 100644
> --- a/drivers/greybus/connection.c
> +++ b/drivers/greybus/connection.c
[ ... ]
> -static void gb_connection_get(struct gb_connection *connection)
> +void gb_connection_get(struct gb_connection *connection)
> {
> kref_get(&connection->kref);
>
> trace_gb_connection_get(connection);
> }
> +EXPORT_SYMBOL_GPL(gb_connection_get);
>
> -static void gb_connection_put(struct gb_connection *connection)
> +void gb_connection_put(struct gb_connection *connection)
> {
> trace_gb_connection_put(connection);
>
> kref_put(&connection->kref, gb_connection_kref_release);
> }
> +EXPORT_SYMBOL_GPL(gb_connection_put);
[Severity: High]
Does exporting this function introduce a regression where a connection can
outlive its parent host device or bundle, potentially causing a use-after-free
in the tracepoints?
If an external module caches a connection reference and drops it asynchronously
after the Greybus core has torn down and freed the gb_bundle and
gb_host_device, calling gb_connection_put() will trigger
trace_gb_connection_put() and potentially trace_gb_connection_release().
These trace events rely on a shared TP_fast_assign block in
drivers/greybus/greybus_trace.h that unconditionally dereferences the parent
objects:
drivers/greybus/greybus_trace.h:gb_connection() {
TP_fast_assign(
__entry->hd_bus_id = connection->hd->bus_id;
__entry->bundle_id = connection->bundle ?
connection->bundle->id : BUNDLE_ID_NONE;
...
}
If the parent hd or bundle has already been freed, will these tracepoints
dereference freed memory?
--
Sashiko AI review · https://sashiko.dev/#/patchset/20260820-gb-uart-transport-v1-0-282da14ab7b7@beagleboard.org?part=1
next prev parent reply other threads:[~2026-08-20 9:41 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-08-20 9:25 [PATCH 0/7] Add Greybus Sotfsvc and UART Node drivers Ayush Singh
2026-08-20 9:25 ` [PATCH 1/7] greybus: connection: Export gb_connection_get() and gb_connection_put() Ayush Singh
2026-08-20 9:41 ` sashiko-bot [this message]
2026-08-20 9:25 ` [PATCH 2/7] greybus: connection: Add gb_connection_hd_find_by_intf() Ayush Singh
2026-08-20 9:32 ` sashiko-bot
2026-08-20 9:25 ` [PATCH 3/7] rust: crc_ccitt: add CRC-CCITT abstraction Ayush Singh
2026-08-20 9:33 ` sashiko-bot
2026-08-20 9:25 ` [PATCH 4/7] rust: kernel: Add greybus abstractions Ayush Singh
2026-08-20 9:42 ` sashiko-bot
2026-08-20 9:25 ` [PATCH 5/7] drivers: greybus: Add software SVC implementation Ayush Singh
2026-08-20 9:40 ` sashiko-bot
2026-08-20 9:25 ` [PATCH 6/7] greybus: Add Rust UART node driver Ayush Singh
2026-08-20 9:38 ` sashiko-bot
2026-08-20 9:25 ` [PATCH 7/7] dt-bindings: beagle: Add BeagleConnect Freedom Ayush Singh
2026-08-20 9:31 ` sashiko-bot
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=20260820094142.BBCEF1F000E9@smtp.kernel.org \
--to=sashiko-bot@kernel.org \
--cc=ayush@beagleboard.org \
--cc=conor+dt@kernel.org \
--cc=devicetree@vger.kernel.org \
--cc=ojeda@kernel.org \
--cc=robh@kernel.org \
--cc=sashiko-reviews@lists.linux.dev \
/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