From mboxrd@z Thu Jan  1 00:00:00 1970
From: Claude Code Review Bot <claude-review@example.com>
To: dri-devel-reviews@example.com
Subject: Claude review: Re: [PATCH v2] staging: fbtft: fb_ra8875: replace udelays with fsleep
Date: Fri, 27 Feb 2026 12:54:37 +1000
Message-ID: <review-patch2-aZ_6QvJPZplh6xtd@smile.fi.intel.com>
In-Reply-To: <aZ_6QvJPZplh6xtd@smile.fi.intel.com>
References: <2026022515-booth-silent-f3bd@gregkh> <aZ_6QvJPZplh6xtd@smile.fi.intel.com>
X-Mailer: Claude Code Patch Reviewer
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0

Patch Review

**Status**: Looks good.

The v2 patch correctly uses `fsleep(100)` which is the kernel's recommended=
 flexible sleep API. For a 100us delay, `fsleep()` will internally choose `=
usleep_range()` (since 100us falls in the 10us=E2=80=9320ms range where `us=
leep_range` is preferred), so the end result is essentially the same as v1 =
but without the caller needing to pick the range bounds.

```c
-	udelay(100);
+	fsleep(100);
```

Both replacements are in `write_reg8_bus8()`, which is clearly a sleepable =
context =E2=80=94 it calls `par->fbtftops.write()` (an SPI transfer functio=
n) both before and after the delays.

**Minor observations:**
- The commit message and changelog are well-written and explain the rationa=
le clearly.
- The `v2:` changelog below the `---` line properly documents what changed =
between versions.

**Reviewed-by worthy**: Yes, this is a straightforward and correct cleanup.=
 No issues found.

---
Generated by Claude Code Patch Reviewer