From: "Ville Syrjälä" <ville.syrjala@linux.intel.com>
To: Arthur Grillo <arthurgrillo@riseup.net>
Cc: igt-dev@lists.freedesktop.org,
Petri Latvala <adrinael@adrinael.net>,
Arkadiusz Hiler <arek@hiler.eu>,
Kamil Konieczny <kamil.konieczny@linux.intel.com>,
Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>,
Bhanuprakash Modem <bhanuprakash.modem@intel.com>,
Ashutosh Dixit <ashutosh.dixit@intel.com>,
Pekka Paalanen <pekka.paalanen@collabora.com>,
Louis Chauvet <louis.chauvet@bootlin.com>
Subject: Re: [PATCH i-g-t v6 12/12] lib/igt_fb: Add non sub-sampled YUV formats
Date: Mon, 28 Oct 2024 18:07:49 +0200 [thread overview]
Message-ID: <Zx-21REAlrgC3DHs@intel.com> (raw)
In-Reply-To: <20240912-kms_fb_stress-dev-v6-12-71e9a65397e9@riseup.net>
On Thu, Sep 12, 2024 at 09:39:36PM -0300, Arthur Grillo wrote:
> This adds support for the formats:
> - NV24
> - NV42
> - YU24
> - YU42
I would split this into two patches: 2 plane vs. 3 plane
>
> Signed-off-by: Arthur Grillo <arthurgrillo@riseup.net>
> ---
> lib/igt_fb.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 46 insertions(+)
>
> diff --git a/lib/igt_fb.c b/lib/igt_fb.c
> index 7f87d40ed3d6..b1301d08d9f8 100644
> --- a/lib/igt_fb.c
> +++ b/lib/igt_fb.c
> @@ -257,6 +257,16 @@ static const struct format_desc_struct {
> .num_planes = 2, .plane_bpp = { 8, 16, },
> .hsub = 2, .vsub = 2,
> },
> + { .name = "NV24", .depth = -1, .drm_id = DRM_FORMAT_NV24,
> + .cairo_id = CAIRO_FORMAT_RGB24, .convert = true,
> + .num_planes = 2, .plane_bpp = { 8, 16, },
> + .hsub = 1, .vsub = 1,
> + },
> + { .name = "NV42", .depth = -1, .drm_id = DRM_FORMAT_NV42,
> + .cairo_id = CAIRO_FORMAT_RGB24, .convert = true,
> + .num_planes = 2, .plane_bpp = { 8, 16, },
> + .hsub = 1, .vsub = 1,
> + },
> { .name = "NV61", .depth = -1, .drm_id = DRM_FORMAT_NV61,
> .cairo_id = CAIRO_FORMAT_RGB24, .convert = true,
> .num_planes = 2, .plane_bpp = { 8, 16, },
> @@ -292,6 +302,11 @@ static const struct format_desc_struct {
> .num_planes = 3, .plane_bpp = { 8, 8, 8, },
> .hsub = 2, .vsub = 1,
> },
> + { .name = "YU24", .depth = -1, .drm_id = DRM_FORMAT_YUV444,
> + .cairo_id = CAIRO_FORMAT_RGB24, .convert = true,
> + .num_planes = 3, .plane_bpp = { 8, 8, 8, },
> + .hsub = 1, .vsub = 1,
> + },
> { .name = "YV12", .depth = -1, .drm_id = DRM_FORMAT_YVU420,
> .cairo_id = CAIRO_FORMAT_RGB24, .convert = true,
> .num_planes = 3, .plane_bpp = { 8, 8, 8, },
> @@ -302,6 +317,11 @@ static const struct format_desc_struct {
> .num_planes = 3, .plane_bpp = { 8, 8, 8, },
> .hsub = 2, .vsub = 1,
> },
> + { .name = "YV24", .depth = -1, .drm_id = DRM_FORMAT_YVU444,
> + .cairo_id = CAIRO_FORMAT_RGB24, .convert = true,
> + .num_planes = 3, .plane_bpp = { 8, 8, 8, },
> + .hsub = 1, .vsub = 1,
> + },
> { .name = "Y410", .depth = -1, .drm_id = DRM_FORMAT_Y410,
> .cairo_id = CAIRO_FORMAT_RGBA128F, .convert = true,
> .num_planes = 1, .plane_bpp = { 32, },
> @@ -1166,8 +1186,10 @@ static void clear_yuv_buffer(struct igt_fb *fb)
> break;
> case DRM_FORMAT_YUV420:
> case DRM_FORMAT_YUV422:
> + case DRM_FORMAT_YUV444:
> case DRM_FORMAT_YVU420:
> case DRM_FORMAT_YVU422:
> + case DRM_FORMAT_YVU444:
Where's the corresponding stuff for the 2 plane formats?
I was also pondering about using igt_reduce_format() to
make the switch a bit less bloated, but I suppose that
won't work for eg. YUYV vs. UYVY. Maybe it could be used
in some other places though.
> igt_assert(ARRAY_SIZE(plane_size) == 3);
> memset(ptr + fb->offsets[0],
> full_range ? 0x00 : 0x10,
> @@ -3495,6 +3517,8 @@ static void get_yuv_parameters(struct igt_fb *fb, struct yuv_parameters *params)
> case DRM_FORMAT_NV12:
> case DRM_FORMAT_NV16:
> case DRM_FORMAT_NV21:
> + case DRM_FORMAT_NV24:
> + case DRM_FORMAT_NV42:
> case DRM_FORMAT_NV61:
> case DRM_FORMAT_P010:
> case DRM_FORMAT_P012:
> @@ -3505,8 +3529,10 @@ static void get_yuv_parameters(struct igt_fb *fb, struct yuv_parameters *params)
>
> case DRM_FORMAT_YUV420:
> case DRM_FORMAT_YUV422:
> + case DRM_FORMAT_YUV444:
> case DRM_FORMAT_YVU420:
> case DRM_FORMAT_YVU422:
> + case DRM_FORMAT_YVU444:
> params->ay_inc = 1;
> params->uv_inc = 1;
> break;
> @@ -3536,11 +3562,15 @@ static void get_yuv_parameters(struct igt_fb *fb, struct yuv_parameters *params)
> case DRM_FORMAT_NV12:
> case DRM_FORMAT_NV16:
> case DRM_FORMAT_NV21:
> + case DRM_FORMAT_NV24:
> + case DRM_FORMAT_NV42:
> case DRM_FORMAT_NV61:
> case DRM_FORMAT_YUV420:
> case DRM_FORMAT_YUV422:
> + case DRM_FORMAT_YUV444:
> case DRM_FORMAT_YVU420:
> case DRM_FORMAT_YVU422:
> + case DRM_FORMAT_YVU444:
> case DRM_FORMAT_P010:
> case DRM_FORMAT_P012:
> case DRM_FORMAT_P016:
> @@ -3568,12 +3598,14 @@ static void get_yuv_parameters(struct igt_fb *fb, struct yuv_parameters *params)
> switch (fb->drm_format) {
> case DRM_FORMAT_NV12:
> case DRM_FORMAT_NV16:
> + case DRM_FORMAT_NV24:
> params->y_offset = fb->offsets[0];
> params->u_offset = fb->offsets[1];
> params->v_offset = fb->offsets[1] + 1;
> break;
>
> case DRM_FORMAT_NV21:
> + case DRM_FORMAT_NV42:
> case DRM_FORMAT_NV61:
> params->y_offset = fb->offsets[0];
> params->u_offset = fb->offsets[1] + 1;
> @@ -3582,6 +3614,7 @@ static void get_yuv_parameters(struct igt_fb *fb, struct yuv_parameters *params)
>
> case DRM_FORMAT_YUV420:
> case DRM_FORMAT_YUV422:
> + case DRM_FORMAT_YUV444:
> params->y_offset = fb->offsets[0];
> params->u_offset = fb->offsets[1];
> params->v_offset = fb->offsets[2];
> @@ -3589,6 +3622,7 @@ static void get_yuv_parameters(struct igt_fb *fb, struct yuv_parameters *params)
>
> case DRM_FORMAT_YVU420:
> case DRM_FORMAT_YVU422:
> + case DRM_FORMAT_YVU444:
> params->y_offset = fb->offsets[0];
> params->u_offset = fb->offsets[2];
> params->v_offset = fb->offsets[1];
> @@ -4322,14 +4356,18 @@ static void fb_convert(struct fb_convert *cvt)
> case DRM_FORMAT_NV12:
> case DRM_FORMAT_NV16:
> case DRM_FORMAT_NV21:
> + case DRM_FORMAT_NV24:
> + case DRM_FORMAT_NV42:
> case DRM_FORMAT_NV61:
> case DRM_FORMAT_UYVY:
> case DRM_FORMAT_VYUY:
> case DRM_FORMAT_YUV420:
> case DRM_FORMAT_YUV422:
> + case DRM_FORMAT_YUV444:
> case DRM_FORMAT_YUYV:
> case DRM_FORMAT_YVU420:
> case DRM_FORMAT_YVU422:
> + case DRM_FORMAT_YVU444:
> case DRM_FORMAT_YVYU:
> convert_yuv_to_rgb24(cvt);
> return;
> @@ -4340,14 +4378,18 @@ static void fb_convert(struct fb_convert *cvt)
> case DRM_FORMAT_NV12:
> case DRM_FORMAT_NV16:
> case DRM_FORMAT_NV21:
> + case DRM_FORMAT_NV24:
> + case DRM_FORMAT_NV42:
> case DRM_FORMAT_NV61:
> case DRM_FORMAT_UYVY:
> case DRM_FORMAT_VYUY:
> case DRM_FORMAT_YUV420:
> case DRM_FORMAT_YUV422:
> + case DRM_FORMAT_YUV444:
> case DRM_FORMAT_YUYV:
> case DRM_FORMAT_YVU420:
> case DRM_FORMAT_YVU422:
> + case DRM_FORMAT_YVU444:
> case DRM_FORMAT_YVYU:
> convert_rgb24_to_yuv(cvt);
> return;
> @@ -4921,11 +4963,15 @@ bool igt_format_is_yuv(uint32_t drm_format)
> case DRM_FORMAT_NV12:
> case DRM_FORMAT_NV16:
> case DRM_FORMAT_NV21:
> + case DRM_FORMAT_NV24:
> + case DRM_FORMAT_NV42:
> case DRM_FORMAT_NV61:
> case DRM_FORMAT_YUV420:
> case DRM_FORMAT_YUV422:
> + case DRM_FORMAT_YUV444:
> case DRM_FORMAT_YVU420:
> case DRM_FORMAT_YVU422:
> + case DRM_FORMAT_YVU444:
> case DRM_FORMAT_P010:
> case DRM_FORMAT_P012:
> case DRM_FORMAT_P016:
>
> --
> 2.46.0
--
Ville Syrjälä
Intel
next prev parent reply other threads:[~2024-10-28 16:07 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-09-13 0:39 [PATCH i-g-t v6 00/12] benchmarks/kms_fb_stress: benchmark formats Arthur Grillo
2024-09-13 0:39 ` [PATCH i-g-t v6 01/12] benchmarks/kms_fb_stress: Set DRM_CLIENT_CAP_WRITEBACK_CONNECTORS before igt_require() Arthur Grillo
2024-10-22 16:41 ` Kamil Konieczny
2024-09-13 0:39 ` [PATCH i-g-t v6 02/12] benchmarks/kms_fb_stress: Assert that we have a supported pipe Arthur Grillo
2024-09-13 0:39 ` [PATCH i-g-t v6 03/12] benchmarks/kms_fb_stress: Separate the benchmark in a subtest Arthur Grillo
2024-09-13 0:39 ` [PATCH i-g-t v6 04/12] benchmarks/kms_fb_stress: Free resources on the stress procedure Arthur Grillo
2024-09-13 0:39 ` [PATCH i-g-t v6 05/12] benchmarks/kms_fb_stress: Move FB creation to a separate function Arthur Grillo
2024-09-13 0:39 ` [PATCH i-g-t v6 06/12] benchmarks/kms_fb_stress: Don't paint the FB's if the format is not supported by Pixman Arthur Grillo
2024-09-13 0:39 ` [PATCH i-g-t v6 07/12] benchmarks/kms_fb_stress: Set writeback rectangle size outside the loop Arthur Grillo
2024-09-13 0:39 ` [PATCH i-g-t v6 08/12] benchmarks/kms_fb_stress: Add enable switches for every plane Arthur Grillo
2024-09-13 0:39 ` [PATCH i-g-t v6 09/12] benchmarks/kms_fb_stress: Log the KMS structure Arthur Grillo
2024-09-13 0:39 ` [PATCH i-g-t v6 10/12] benchmarks/kms_fb_stress: Create a benchmark for primary plane formats Arthur Grillo
2024-09-13 0:39 ` [PATCH i-g-t v6 11/12] benchmarks/kms_fb_stress: Create a benchmark for writeback formats Arthur Grillo
2024-09-13 0:39 ` [PATCH i-g-t v6 12/12] lib/igt_fb: Add non sub-sampled YUV formats Arthur Grillo
2024-10-22 15:06 ` Kamil Konieczny
2024-10-28 15:34 ` Kamil Konieczny
2024-10-28 16:07 ` Ville Syrjälä [this message]
2024-09-13 1:26 ` ✓ CI.xeBAT: success for benchmarks/kms_fb_stress: benchmark formats (rev3) Patchwork
2024-09-13 1:46 ` ✗ Fi.CI.BAT: failure " Patchwork
2024-09-13 12:03 ` Arthur Grillo
2024-09-13 18:36 ` ✗ CI.xeFULL: " Patchwork
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=Zx-21REAlrgC3DHs@intel.com \
--to=ville.syrjala@linux.intel.com \
--cc=adrinael@adrinael.net \
--cc=arek@hiler.eu \
--cc=arthurgrillo@riseup.net \
--cc=ashutosh.dixit@intel.com \
--cc=bhanuprakash.modem@intel.com \
--cc=igt-dev@lists.freedesktop.org \
--cc=juhapekka.heikkila@gmail.com \
--cc=kamil.konieczny@linux.intel.com \
--cc=louis.chauvet@bootlin.com \
--cc=pekka.paalanen@collabora.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.