[PATCH v3] staging: media: atomisp: fix block comment style

[PATCH v3] staging: media: atomisp: fix block comment style

[Ethan Lam]

Fixed all block comment style warnings by checkpatch.pl.
Kept original indentation in comments (requested by Andy Shevchenko)

Signed-off-by: Ethan Lam <ethanlxkernel@gmail.com>
---
 .../staging/media/atomisp/pci/atomisp_cmd.c   | 60 ++++++++++++-------
 1 file changed, 39 insertions(+), 21 deletions(-)

diff --git a/drivers/staging/media/atomisp/pci/atomisp_cmd.c b/drivers/staging/media/atomisp/pci/atomisp_cmd.c
index 3a4eb4f6d3be..c67fa4cc3f06 100644
--- a/drivers/staging/media/atomisp/pci/atomisp_cmd.c
+++ b/drivers/staging/media/atomisp/pci/atomisp_cmd.c
@@ -1380,8 +1380,10 @@ static void atomisp_update_grid_info(struct atomisp_sub_device *asd,
 	if (atomisp_css_get_grid_info(asd, pipe_id))
 		return;
 
-	/* We must free all buffers because they no longer match
-	   the grid size. */
+	/*
+	 * We must free all buffers because they no longer match
+	 * the grid size.
+	 */
 	atomisp_css_free_stat_buffers(asd);
 
 	err = atomisp_alloc_css_stat_bufs(asd, ATOMISP_INPUT_STREAM_GENERAL);
@@ -1393,8 +1395,10 @@ static void atomisp_update_grid_info(struct atomisp_sub_device *asd,
 	if (atomisp_alloc_3a_output_buf(asd)) {
 		/* Failure for 3A buffers does not influence DIS buffers */
 		if (asd->params.s3a_output_bytes != 0) {
-			/* For SOC sensor happens s3a_output_bytes == 0,
-			 * using if condition to exclude false error log */
+			/*
+			 * For SOC sensor happens s3a_output_bytes == 0,
+			 * using if condition to exclude false error log
+			 */
 			dev_err(isp->dev, "Failed to allocate memory for 3A statistics\n");
 		}
 		goto err;
@@ -1686,9 +1690,11 @@ int atomisp_3a_stat(struct atomisp_sub_device *asd, int flag,
 		return -EINVAL;
 
 	if (atomisp_compare_grid(asd, &config->grid_info) != 0) {
-		/* If the grid info in the argument differs from the current
-		   grid info, we tell the caller to reset the grid size and
-		   try again. */
+		/*
+		 * If the grid info in the argument differs from the current
+		 * grid info, we tell the caller to reset the grid size and
+		 * try again.
+		 */
 		return -EAGAIN;
 	}
 
@@ -2462,9 +2468,11 @@ int atomisp_css_cp_dvs2_coefs(struct atomisp_sub_device *asd,
 		if (sizeof(*cur) != sizeof(coefs->grid) ||
 		    memcmp(&coefs->grid, cur, sizeof(coefs->grid))) {
 			dev_err(asd->isp->dev, "dvs grid mismatch!\n");
-			/* If the grid info in the argument differs from the current
-			grid info, we tell the caller to reset the grid size and
-			try again. */
+			/*
+			 * If the grid info in the argument differs from the current
+			 * grid info, we tell the caller to reset the grid size and
+			 * try again.
+			 */
 			return -EAGAIN;
 		}
 
@@ -3028,9 +3036,11 @@ int atomisp_param(struct atomisp_sub_device *asd, int flag,
 
 		atomisp_curr_user_grid_info(asd, &config->info);
 
-		/* We always return the resolution and stride even if there is
+		/*
+		 * We always return the resolution and stride even if there is
 		 * no valid metadata. This allows the caller to get the
-		 * information needed to allocate user-space buffers. */
+		 * information needed to allocate user-space buffers.
+		 */
 		config->metadata_config.metadata_height = asd->
 			stream_env[ATOMISP_INPUT_STREAM_GENERAL].stream_info.
 			metadata_info.resolution.height;
@@ -3280,8 +3290,10 @@ atomisp_bytesperline_to_padded_width(unsigned int bytesperline,
 		return bytesperline / 2;
 	case IA_CSS_FRAME_FORMAT_RGBA888:
 		return bytesperline / 4;
-	/* The following cases could be removed, but we leave them
-	   in to document the formats that are included. */
+	/*
+	 * The following cases could be removed, but we leave them
+	 * in to document the formats that are included.
+	 */
 	case IA_CSS_FRAME_FORMAT_NV11:
 	case IA_CSS_FRAME_FORMAT_NV12:
 	case IA_CSS_FRAME_FORMAT_NV16:
@@ -3317,9 +3329,11 @@ atomisp_v4l2_framebuffer_to_css_frame(const struct v4l2_framebuffer *arg,
 	padded_width = atomisp_bytesperline_to_padded_width(
 			   arg->fmt.bytesperline, sh_format);
 
-	/* Note: the padded width on an ia_css_frame is in elements, not in
-	   bytes. The RAW frame we use here should always be a 16bit RAW
-	   frame. This is why we bytesperline/2 is equal to the padded with */
+	/*
+	 * Note: the padded width on an ia_css_frame is in elements, not in
+	 * bytes. The RAW frame we use here should always be a 16bit RAW
+	 * frame. This is why we bytesperline/2 is equal to the padded with
+	 */
 	if (ia_css_frame_allocate(&res, arg->fmt.width, arg->fmt.height,
 				       sh_format, padded_width, 0)) {
 		ret = -ENOMEM;
@@ -3928,8 +3942,10 @@ static inline int atomisp_set_sensor_mipi_to_isp(
 		    asd->stream_env[stream_id].isys_info[1].height);
 	}
 
-	/* Compatibility for sensors which provide no media bus code
-	 * in s_mbus_framefmt() nor support pad formats. */
+	/*
+	 * Compatibility for sensors which provide no media bus code
+	 * in s_mbus_framefmt() nor support pad formats.
+	 */
 	if (mipi_info && mipi_info->input_format != -1) {
 		bayer_order = mipi_info->raw_bayer_order;
 
@@ -4387,8 +4403,10 @@ int atomisp_set_fmt(struct video_device *vdev, struct v4l2_format *f)
 			ATOMISP_SUBDEV_PAD_SINK,
 			V4L2_SEL_TGT_CROP);
 
-	/* Try to enable YUV downscaling if ISP input is 10 % (either
-	 * width or height) bigger than the desired result. */
+	/*
+	 * Try to enable YUV downscaling if ISP input is 10 % (either
+	 * width or height) bigger than the desired result.
+	 */
 	if (!IS_MOFD ||
 	    isp_sink_crop.width * 9 / 10 < f->fmt.pix.width ||
 	    isp_sink_crop.height * 9 / 10 < f->fmt.pix.height ||
-- 
2.43.0

Re: [PATCH v3] staging: media: atomisp: fix block comment style

[Sakari Ailus]

Hi Ethan,

On Tue, Feb 10, 2026 at 07:35:57PM +0800, Ethan Lam wrote:
> Fixed all block comment style warnings by checkpatch.pl.
> Kept original indentation in comments (requested by Andy Shevchenko)

This can still be improved.

Please see section "Describe your changes" in
Documentation/process/submitting-patches.rst . In particular:

	Describe your changes in imperative mood, e.g. "make xyzzy do
	frotz" instead of "[This patch] makes xyzzy do frotz" or "[I]
	changed xyzzy to do frotz", as if you are giving orders to the
	codebase to change its behaviour.

No need to mention original indentation is kept as it is correct.

Is atomisp_cmd.c the only file in the atomisp driver with checkpatch.pl
comment warnings?

> 
> Signed-off-by: Ethan Lam <ethanlxkernel@gmail.com>

-- 
Regards,

Sakari Ailus