From mboxrd@z Thu Jan 1 00:00:00 1970 From: Claude Code Review Bot To: dri-devel-reviews@example.com Subject: Claude review: Initial support Cadence MHDP8501(HDMI/DP) for i.MX8MQ Date: Mon, 25 May 2026 22:48:47 +1000 Message-ID: In-Reply-To: <20260519-dcss-hdmi-upstreaming-v23-0-5615524a9c63@oss.nxp.com> References: <20260519-dcss-hdmi-upstreaming-v23-0-5615524a9c63@oss.nxp.com> X-Mailer: Claude Code Patch Reviewer Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Overall Series Review Subject: Initial support Cadence MHDP8501(HDMI/DP) for i.MX8MQ Author: Laurentiu Palcu Patches: 9 Reviewed: 2026-05-25T22:48:47.938216 --- This is v23 of a patch series adding initial Cadence MHDP8501 HDMI/DP bridg= e and HDP-TX PHY support for the i.MX8MQ SoC. The series has been through e= xtensive review over 23 revisions and is well-structured: 1. Extract shared mailbox helpers into `drivers/soc/cadence/` 2. Refactor existing mhdp8546 to use the new helpers 3-4. PHY DT binding + driver 5-6. Bridge DT binding + driver 7-8. DTS enablement The overall architecture is sound: the mailbox layer is properly shared, th= e PHY child-node approach avoids duplicating MMIO mappings, and the bridge = driver uses modern DRM atomic APIs. The series clearly benefited from exten= sive review =E2=80=94 most common pitfalls have been addressed. However, se= veral issues remain across the patches. **Key concerns:** - Potential race between IRQ handling and device removal in the mhdp8501 br= idge driver - The `phy_powered` flag lacks sufficient synchronization in certain paths - The HDMI I2C adapter is extremely limited (only SCDC slave) =E2=80=94 thi= s should be clearly documented in the code - Copyright dates still say 2023-2024 despite being submitted in 2026 - The mhdp8546 patch adds `struct cdns_mhdp_base base` to `cdns_mhdp_device= ` while keeping the existing `regs`/`sapb_regs` fields, creating data dupli= cation --- --- Generated by Claude Code Patch Reviewer