Received-SPF: pass (google.com: domain of linux-kernel+bounces-181087-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
From: Louis Chauvet <louis.chauvet@bootlin.com>
Date: Thu, 16 May 2024 15:05:02 +0200
Subject: [PATCH v8 17/17] drm/vkms: Add support for DRM_FORMAT_R*
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
Message-Id: <20240516-yuv-v8-17-cf8d6f86430e@bootlin.com>
References: <20240516-yuv-v8-0-cf8d6f86430e@bootlin.com>
In-Reply-To: <20240516-yuv-v8-0-cf8d6f86430e@bootlin.com>
To: Rodrigo Siqueira <rodrigosiqueiramelo@gmail.com>, 
 Melissa Wen <melissa.srw@gmail.com>, 
 =?utf-8?q?Ma=C3=ADra_Canal?= <mairacanal@riseup.net>, 
 Haneen Mohammed <hamohammed.sa@gmail.com>, Daniel Vetter <daniel@ffwll.ch>, 
 Maarten Lankhorst <maarten.lankhorst@linux.intel.com>, 
 Maxime Ripard <mripard@kernel.org>, Thomas Zimmermann <tzimmermann@suse.de>, 
 David Airlie <airlied@gmail.com>, rdunlap@infradead.org, 
 arthurgrillo@riseup.net, Jonathan Corbet <corbet@lwn.net>, 
 pekka.paalanen@haloniitty.fi
Cc: dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org, 
 jeremie.dautheribes@bootlin.com, miquel.raynal@bootlin.com, 
 thomas.petazzoni@bootlin.com, seanpaul@google.com, marcheu@google.com, 
 nicolejadeyee@google.com, Louis Chauvet <louis.chauvet@bootlin.com>, 
 Pekka Paalanen <pekka.paalanen@collabora.com>

This add the support for:
- R1/R2/R4/R8

R1 format was tested with [1] and [2].

[1]: https://lore.kernel.org/r/20240313-new_rotation-v2-0-6230fd5cae59@bootlin.com
[2]: https://lore.kernel.org/igt-dev/20240306-b4-kms_tests-v1-0-8fe451efd2ac@bootlin.com/

Reviewed-by: Pekka Paalanen <pekka.paalanen@collabora.com>
Signed-off-by: Louis Chauvet <louis.chauvet@bootlin.com>
---
 drivers/gpu/drm/vkms/vkms_formats.c | 110 +++++++++++++++++++++++++++++++++++-
 drivers/gpu/drm/vkms/vkms_plane.c   |   4 ++
 2 files changed, 113 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/vkms/vkms_formats.c b/drivers/gpu/drm/vkms/vkms_formats.c
index dac874980e3d..88edf773b11b 100644
--- a/drivers/gpu/drm/vkms/vkms_formats.c
+++ b/drivers/gpu/drm/vkms/vkms_formats.c
@@ -241,6 +241,16 @@ static struct pixel_argb_u16 argb_u16_from_RGB565(const u16 *pixel)
 	return out_pixel;
 }
 
+static struct pixel_argb_u16 argb_u16_from_gray8(u16 gray)
+{
+	return argb_u16_from_u8888(255, gray, gray, gray);
+}
+
+static struct pixel_argb_u16 argb_u16_from_gray16(u16 gray)
+{
+	return argb_u16_from_u16161616(255, gray, gray, gray);
+}
+
 VISIBLE_IF_KUNIT struct pixel_argb_u16 argb_u16_from_yuv888(u8 y, u8 channel_1, u8 channel_2,
 							   const struct conversion_matrix *matrix)
 {
@@ -278,7 +288,7 @@ EXPORT_SYMBOL_IF_KUNIT(argb_u16_from_yuv888);
  * The following functions are read_line function for each pixel format supported by VKMS.
  *
  * They read a line starting at the point @x_start,@y_start following the @direction. The result
- * is stored in @out_pixel and in the format ARGB16161616.
+ * is stored in @out_pixel and in a 64 bits format, see struct pixel_argb_u16.
  *
  * These functions are very repetitive, but the innermost pixel loops must be kept inside these
  * functions for performance reasons. Some benchmarking was done in [1] where having the innermost
@@ -306,6 +316,96 @@ static void magenta_to_argb_u16(const struct vkms_plane_state *plane, int x_star
 	}
 }
 
+static void Rx_read_line(const struct vkms_plane_state *plane, int x_start,
+			 int y_start, enum pixel_read_direction direction, int count,
+			 struct pixel_argb_u16 out_pixel[])
+{
+	struct pixel_argb_u16 *end = out_pixel + count;
+	int bits_per_pixel = drm_format_info_bpp(plane->frame_info->fb->format, 0);
+	u8 *src_pixels;
+	int rem_x, rem_y;
+
+	WARN_ONCE(drm_format_info_block_height(plane->frame_info->fb->format, 0) != 1,
+		  "%s() only support formats with block_h == 1", __func__);
+
+	packed_pixels_addr(plane->frame_info, x_start, y_start, 0, &src_pixels, &rem_x, &rem_y);
+	int bit_offset = (8 - bits_per_pixel) - rem_x * bits_per_pixel;
+	int step = get_block_step_bytes(plane->frame_info->fb, direction, 0);
+	int mask = (0x1 << bits_per_pixel) - 1;
+	int lum_per_level = 0xFFFF / mask;
+
+	if (direction == READ_LEFT_TO_RIGHT || direction == READ_RIGHT_TO_LEFT) {
+		int restart_bit_offset;
+		int step_bit_offset;
+
+		if (direction == READ_LEFT_TO_RIGHT) {
+			restart_bit_offset = 8 - bits_per_pixel;
+			step_bit_offset = -bits_per_pixel;
+		} else {
+			restart_bit_offset = 0;
+			step_bit_offset = bits_per_pixel;
+		}
+
+		while (out_pixel < end) {
+			u8 val = ((*src_pixels) >> bit_offset) & mask;
+
+			*out_pixel = argb_u16_from_gray16((int)val * lum_per_level);
+
+			bit_offset += step_bit_offset;
+			if (bit_offset < 0 || 8 <= bit_offset) {
+				bit_offset = restart_bit_offset;
+				src_pixels += step;
+			}
+			out_pixel += 1;
+		}
+	} else if (direction == READ_TOP_TO_BOTTOM || direction == READ_BOTTOM_TO_TOP) {
+		while (out_pixel < end) {
+			u8 val = (*src_pixels >> bit_offset) & mask;
+			*out_pixel = argb_u16_from_gray16((int)val * lum_per_level);
+			src_pixels += step;
+			out_pixel += 1;
+		}
+	}
+}
+
+static void R1_read_line(const struct vkms_plane_state *plane, int x_start,
+			 int y_start, enum pixel_read_direction direction, int count,
+			 struct pixel_argb_u16 out_pixel[])
+{
+	Rx_read_line(plane, x_start, y_start, direction, count, out_pixel);
+}
+
+static void R2_read_line(const struct vkms_plane_state *plane, int x_start,
+			 int y_start, enum pixel_read_direction direction, int count,
+			 struct pixel_argb_u16 out_pixel[])
+{
+	Rx_read_line(plane, x_start, y_start, direction, count, out_pixel);
+}
+
+static void R4_read_line(const struct vkms_plane_state *plane, int x_start,
+			 int y_start, enum pixel_read_direction direction, int count,
+			 struct pixel_argb_u16 out_pixel[])
+{
+	Rx_read_line(plane, x_start, y_start, direction, count, out_pixel);
+}
+
+static void R8_read_line(const struct vkms_plane_state *plane, int x_start,
+			 int y_start, enum pixel_read_direction direction, int count,
+			 struct pixel_argb_u16 out_pixel[])
+{
+	struct pixel_argb_u16 *end = out_pixel + count;
+	u8 *src_pixels;
+	int step = get_block_step_bytes(plane->frame_info->fb, direction, 0);
+
+	packed_pixels_addr_1x1(plane->frame_info, x_start, y_start, 0, &src_pixels);
+
+	while (out_pixel < end) {
+		*out_pixel = argb_u16_from_gray8(*src_pixels);
+		src_pixels += step;
+		out_pixel += 1;
+	}
+}
+
 static void ARGB8888_read_line(const struct vkms_plane_state *plane, int x_start, int y_start,
 			       enum pixel_read_direction direction, int count,
 			       struct pixel_argb_u16 out_pixel[])
@@ -628,6 +728,14 @@ pixel_read_line_t get_pixel_read_line_function(u32 format)
 	case DRM_FORMAT_YVU422:
 	case DRM_FORMAT_YVU444:
 		return &planar_yuv_read_line;
+	case DRM_FORMAT_R1:
+		return &R1_read_line;
+	case DRM_FORMAT_R2:
+		return &R2_read_line;
+	case DRM_FORMAT_R4:
+		return &R4_read_line;
+	case DRM_FORMAT_R8:
+		return &R8_read_line;
 	default:
 		/*
 		 * This is a bug in vkms_plane_atomic_check(). All the supported
diff --git a/drivers/gpu/drm/vkms/vkms_plane.c b/drivers/gpu/drm/vkms/vkms_plane.c
index 8f764a108b00..67f891e7ac58 100644
--- a/drivers/gpu/drm/vkms/vkms_plane.c
+++ b/drivers/gpu/drm/vkms/vkms_plane.c
@@ -30,6 +30,10 @@ static const u32 vkms_formats[] = {
 	DRM_FORMAT_YVU420,
 	DRM_FORMAT_YVU422,
 	DRM_FORMAT_YVU444,
+	DRM_FORMAT_R1,
+	DRM_FORMAT_R2,
+	DRM_FORMAT_R4,
+	DRM_FORMAT_R8,
 };
 
 static struct drm_plane_state *

-- 
2.43.2