From mboxrd@z Thu Jan 1 00:00:00 1970 From: Claude Code Review Bot To: dri-devel-reviews@example.com Subject: Claude review: Re: [PATCH] staging: fbtft: fb_ra8875: replace udelay() with usleep_range() Date: Fri, 13 Mar 2026 14:46:09 +1000 Message-ID: In-Reply-To: <2026031151-fragrant-geometry-4b3f@gregkh> References: <2026031151-fragrant-geometry-4b3f@gregkh> <2026031151-fragrant-geometry-4b3f@gregkh> X-Mailer: Claude Code Patch Reviewer Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Patch Review **Overall: Correct and appropriate change.** The two conversions at lines 213 and 234 of `fb_ra8875.c` are in `write_reg= 8_bus8()`, which is a `.write_register` callback that calls `par->fbtftops.= write()` (i.e., `fbtft_write_spi`). SPI transfers can sleep, so this functi= on is already in process context =E2=80=94 `usleep_range()` is safe here. ```c - udelay(100); + usleep_range(100, 200); ``` The chosen range of `(100, 200)` is reasonable for a 100 =C2=B5s delay. **Minor nit (not a blocker):** The commit message says "as recommended by c= heckpatch.pl" =E2=80=94 while true, it would be slightly better to mention = the actual technical benefit (the function is in process/sleepable context = since it performs SPI transfers) rather than citing checkpatch as the prima= ry motivation. Checkpatch-driven patches that blindly convert `udelay` to `= usleep_range` without verifying context safety are a common source of bugs;= mentioning why it's safe here would strengthen the commit message. **No functional issues found.** --- Generated by Claude Code Patch Reviewer