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: drm/panel: Add Novatek/Tianma NT37700F panel
Date: Tue, 17 Feb 2026 15:51:44 +1000
Message-ID: <review-patch3-20260217000854.131242-4-mailingradian@gmail.com>
In-Reply-To: <20260217000854.131242-4-mailingradian@gmail.com>
References: <20260217000854.131242-1-mailingradian@gmail.com> <20260217000854.131242-4-mailingradian@gmail.com>
X-Mailer: Claude Code Patch Reviewer
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
MIME-Version: 1.0

Patch Review

The panel driver follows established patterns. A few observations:

> +static void nt37700f_tianma_reset(struct nt37700f_tianma *ctx)
> +{
> +	gpiod_set_value_cansleep(ctx->reset_gpio, 1);
> +	usleep_range(1000, 2000);
> +	gpiod_set_value_cansleep(ctx->reset_gpio, 0);
> +	usleep_range(10000, 11000);
> +}

The reset GPIO is requested with `GPIOD_OUT_HIGH` in probe, meaning the panel starts in reset. The reset sequence asserts (1), waits, deasserts (0), waits. This is consistent with the `unprepare` path which sets the GPIO to 1 (asserting reset). Fine.

> +	dsi->mode_flags = MIPI_DSI_MODE_VIDEO_BURST |
> +			  MIPI_DSI_CLOCK_NON_CONTINUOUS | MIPI_DSI_MODE_LPM;

The mode flags combine video burst mode with LPM for command sequences. This is a standard pattern seen in other panel drivers (e.g., `panel-samsung-sofef00.c`).

> +static int nt37700f_tianma_bl_update_status(struct backlight_device *bl)
> +{
> +	struct mipi_dsi_device *dsi = bl_get_data(bl);
> +	u16 brightness = backlight_get_brightness(bl);
> +	int ret;
> +
> +	dsi->mode_flags &= ~MIPI_DSI_MODE_LPM;
> +
> +	ret = mipi_dsi_dcs_set_display_brightness_large(dsi, brightness);
> +	if (ret < 0)
> +		return ret;
> +
> +	dsi->mode_flags |= MIPI_DSI_MODE_LPM;

Temporarily clearing and restoring `MIPI_DSI_MODE_LPM` to send brightness commands in HS mode is not thread-safe -- if two callers hit this concurrently, the flag manipulation could race. That said, this is a common pattern found in many generated panel drivers and the backlight framework serializes calls through a mutex, so in practice this is fine.

> + * Copyright (c) 2024, The Linux Foundation. All rights reserved.

Minor: the copyright year says 2024, though the patch is being submitted in 2026. This is cosmetic and doesn't affect functionality. Also, attributing a generated driver to "The Linux Foundation" may not be accurate -- typically the author of the driver would be listed.

Overall, the driver is straightforward and correct. No functional bugs found.

---
Generated by Claude Code Patch Reviewer