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/amdgpu: use ACK polling for page-write completion Date: Thu, 04 Jun 2026 14:17:38 +1000 Message-ID: In-Reply-To: <20260601093226.1255621-1-kunal.devanandzodape@amd.com> References: <20260601093226.1255621-1-kunal.devanandzodape@amd.com> X-Mailer: Claude Code Patch Reviewer Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit MIME-Version: 1.0 Overall Series Review Subject: drm/amdgpu: use ACK polling for page-write completion Author: Kunal Zodape Patches: 6 Reviewed: 2026-06-04T14:17:38.468071 --- This mbox contains two versions (v1 and v2) of the same single patch, replacing a fixed 10 ms `msleep()` after each EEPROM page write with ACK polling. The optimization is well-motivated: EEPROM internal write cycles are typically ~5 ms but the driver was always waiting the worst-case 10 ms. ACK polling is the standard technique described in EEPROM datasheets (AT24CM02 Section 7.4) and was even marked as a TODO in the existing code. **v1** used an open-coded `ktime` + `do/while` loop. **v2** addressed review feedback (credited to Jani Nikula) by switching to `read_poll_timeout()`, which is the right kernel idiom. v2 is a clean, correct patch. The approach, error handling, and use of `read_poll_timeout()` are all sound. I have one minor concern about variable reuse worth calling out, but no blocking issues. --- --- Generated by Claude Code Patch Reviewer