From mboxrd@z Thu Jan 1 00:00:00 1970 From: Claude Code Review Bot To: dri-devel-reviews@example.com Subject: Claude review: drm/msm/hdmi: switch to of_drm_get_bridge_by_endpoint() Date: Sat, 16 May 2026 15:02:22 +1000 Message-ID: In-Reply-To: <20260511-drm-bridge-alloc-getput-panel_or_bridge-v6-3-f61c9e498b3f@bootlin.com> References: <20260511-drm-bridge-alloc-getput-panel_or_bridge-v6-0-f61c9e498b3f@bootlin.com> <20260511-drm-bridge-alloc-getput-panel_or_bridge-v6-3-f61c9e498b3f@bootlin.com> X-Mailer: Claude Code Patch Reviewer Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit MIME-Version: 1.0 Patch Review This is the most involved conversion because the MSM HDMI driver's bridge is in a separate `hdmi_bridge` struct (allocated in `hdmi_bridge.c`), not embedded in the `hdmi` struct. So unlike later patches, this one can't use the auto-put `bridge.next_bridge` field and must manage the refcount explicitly. The conversion correctly: - Handles the `-ENODEV` case by setting `hdmi->next_bridge = NULL` (preserving old behavior) - Changes all early-return paths after the bridge lookup to `goto err_put_bridge` - Adds `drm_bridge_put(hdmi->next_bridge)` in both `err_put_bridge` and `msm_hdmi_dev_remove()` The `drm_bridge_put()` in the remove function is safe even if `next_bridge` is NULL (patch 1 ensures this). **Minor nit:** There's a double blank line introduced at the `devm_clk_bulk_get` error path: ```c if (ret) - return ret; + goto err_put_bridge; + hdmi->extp_clk = devm_clk_get_optional(dev, "extp"); ``` No functional issues. This patch needs review/ack from MSM maintainers. --- Generated by Claude Code Patch Reviewer