Igt-dev Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
To: Alex Hung <alex.hung@amd.com>,
	igt-dev@lists.freedesktop.org, kamil.konieczny@linux.intel.com
Cc: hersenxs.wu@amd.com
Subject: Re: [igt-dev] [PATCH v4 2/2] tests/amdgpu: Add test for ODM combine
Date: Mon, 21 Aug 2023 16:18:42 -0400	[thread overview]
Message-ID: <97b7e6a3-bf81-4a17-b224-1351d8dedf53@amd.com> (raw)
In-Reply-To: <581a47bb-0e4c-7408-d107-5ad0e5da8c9e@amd.com>

Thanks for the review Alex.

Kamil,

There seems to be a false positive failure for this patchset: 
https://lists.freedesktop.org/archives/igt-dev/2023-August/060105.html

Do you have any further review suggestions or is it okay to merge ?


On 2023-08-21 11:39, Alex Hung wrote:
> Reviewed-by: Alex Hung <alex.hung@amd.com>
> 
> On 2023-08-21 08:07, Aurabindo Pillai wrote:
>> Add a test that checks whether ODM Combine mode is triggered when high
>> refresh rate timing is committed on hardware it supports. This test
>> requires a DSC capable monitor.
>>
>> Changes in V2: Added check to skip test on hardware that does not have
>> ODM combine segments debugfs endpoint exposed
>>
>> Changes in V3: Removed data->w,h since its not used; Removed license
>> text and use SPDX identifier instead.
>>
>> Changes in V4: Checkpatch.pl style fixes
>>
>> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
>> ---
>>   tests/amdgpu/amd_odm.c   | 189 +++++++++++++++++++++++++++++++++++++++
>>   tests/amdgpu/meson.build |   1 +
>>   2 files changed, 190 insertions(+)
>>   create mode 100644 tests/amdgpu/amd_odm.c
>>
>> diff --git a/tests/amdgpu/amd_odm.c b/tests/amdgpu/amd_odm.c
>> new file mode 100644
>> index 000000000..2781ad1c2
>> --- /dev/null
>> +++ b/tests/amdgpu/amd_odm.c
>> @@ -0,0 +1,189 @@
>> +// SPDX-License-Identifier: MIT
>> +/*
>> + * Copyrights 2023 Advanced Micro Devices, Inc.
>> + */
>> +
>> +#include <fcntl.h>
>> +#include <stdio.h>
>> +#include <xf86drmMode.h>
>> +
>> +#include "igt.h"
>> +#include "igt_amd.h"
>> +#include "igt_edid.h"
>> +
>> +IGT_TEST_DESCRIPTION("Test whether ODM Combine mode is triggered when 
>> timings with high refresh "
>> +             "rate is committed");
>> +
>> +enum odmc_mode {
>> +    ODMC_2_TO_1,
>> +    ODMC_4_TO_1,
>> +};
>> +
>> +/* Common test data. */
>> +struct data {
>> +    igt_display_t display;
>> +    igt_plane_t *primary;
>> +    igt_output_t *output;
>> +    igt_pipe_t *pipe;
>> +    drmModeModeInfoPtr mode;
>> +    enum pipe pipe_id;
>> +    int fd;
>> +};
>> +
>> +static const drmModeModeInfo test_mode[] = {
>> +    { 1278720,
>> +    3840, 3952, 3984, 4000, 0,
>> +    2160, 2210, 2215, 2220, 0,
>> +    30,
>> +    DRM_MODE_FLAG_NHSYNC,
>> +    0x40,
>> +    "4k144\0",
>> +    }, /* from HP Omen 27c */
>> +
>> +};
>> +
>> +#define TEST_MODE_IDX_ODMC_2_TO_1 0
>> +
>> +static void test_init(struct data *data)
>> +{
>> +    igt_display_t *display = &data->display;
>> +
>> +    /* It doesn't matter which pipe we choose on amdpgu. */
>> +    data->pipe_id = PIPE_A;
>> +    data->pipe = &data->display.pipes[data->pipe_id];
>> +
>> +    igt_display_reset(display);
>> +
>> +    /* find a connected non-HDMI output */
>> +    data->output = NULL;
>> +    for (int i = 0; i < data->display.n_outputs; ++i) {
>> +        drmModeConnector *connector = 
>> data->display.outputs[i].config.connector;
>> +
>> +        if (connector->connection == DRM_MODE_CONNECTED)
>> +            data->output = &data->display.outputs[i];
>> +    }
>> +    igt_require_f(data->output, "Requires a connected output\n");
>> +
>> +    data->mode = igt_output_get_mode(data->output);
>> +    igt_assert(data->mode);
>> +
>> +    igt_skip_on_f(!igt_amd_output_has_odm_combine_segments(data->fd, 
>> data->output->name),
>> +              "ASIC does not support reading ODM combine segments\n");
>> +
>> +    igt_skip_on_f(!is_dp_dsc_supported(data->fd, data->output->name),
>> +              "The monitor must be DSC capable\n");
>> +
>> +    igt_skip_on_f(data->output->config.connector->connector_type == 
>> DRM_MODE_CONNECTOR_HDMIA ||
>> +              data->output->config.connector->connector_type == 
>> DRM_MODE_CONNECTOR_HDMIB,
>> +              "ODM Combine isn't supported on HDMI 1.x\n");
>> +
>> +    data->primary = igt_pipe_get_plane_type(data->pipe, 
>> DRM_PLANE_TYPE_PRIMARY);
>> +    igt_output_set_pipe(data->output, data->pipe_id);
>> +
>> +    igt_display_reset(display);
>> +}
>> +
>> +static void test_fini(struct data *data)
>> +{
>> +    igt_display_reset(&data->display);
>> +    igt_display_commit_atomic(&data->display, 
>> DRM_MODE_ATOMIC_ALLOW_MODESET, 0);
>> +}
>> +
>> +/* Forces a mode for a connector. */
>> +static void force_output_mode(struct data *d, igt_output_t *output,
>> +                  const drmModeModeInfo *mode)
>> +{
>> +    /* This allows us to create a virtual sink. */
>> +    if (!igt_output_is_connected(output)) {
>> +        kmstest_force_edid(d->fd, output->config.connector,
>> +                   igt_kms_get_4k_edid());
>> +
>> +        kmstest_force_connector(d->fd, output->config.connector,
>> +                    FORCE_CONNECTOR_DIGITAL);
>> +    }
>> +
>> +    igt_output_override_mode(output, mode);
>> +}
>> +
>> +static void run_test_odmc(struct data *data, enum odmc_mode m, const 
>> drmModeModeInfo *mode)
>> +{
>> +    igt_display_t *display = &data->display;
>> +    struct igt_fb buffer;
>> +    char buf[256];
>> +    int ret, seg, fd;
>> +    int i = 0;
>> +
>> +    test_init(data);
>> +
>> +    force_output_mode(data, data->output, mode);
>> +
>> +    igt_create_color_fb(display->drm_fd, mode->hdisplay,
>> +                mode->vdisplay, DRM_FORMAT_XRGB8888,
>> +                DRM_FORMAT_MOD_LINEAR, 1.f, 0.f, 0.f,
>> +                &buffer);
>> +
>> +    igt_output_set_pipe(data->output, i);
>> +
>> +    igt_plane_set_fb(data->primary, &buffer);
>> +
>> +    ret = igt_display_try_commit_atomic(display,
>> +                        DRM_MODE_ATOMIC_ALLOW_MODESET |
>> +                        DRM_MODE_ATOMIC_TEST_ONLY,
>> +                        NULL);
>> +    igt_skip_on_f(ret != 0, "Unsupported mode\n");
>> +
>> +    igt_display_commit_atomic(display, DRM_MODE_ATOMIC_ALLOW_MODESET, 
>> NULL);
>> +
>> +    fd = igt_debugfs_connector_dir(data->fd, data->output->name, 
>> O_RDONLY);
>> +    igt_assert(fd >= 0);
>> +
>> +    ret = igt_debugfs_simple_read(fd, "odm_combine_segments", buf, 
>> sizeof(buf));
>> +    close(fd);
>> +    igt_require(ret > 0);
>> +
>> +    seg = strtol(buf, NULL, 0);
>> +
>> +    switch (m) {
>> +    case ODMC_2_TO_1:
>> +        igt_assert_f(seg == 2,
>> +                 "ODM Combine uses %d segments for connector %s, 
>> expected 2\n",
>> +                 seg, data->output->name);
>> +        break;
>> +    case ODMC_4_TO_1:
>> +        igt_assert_f(seg == 4,
>> +                 "ODM Combine uses %d segments for connector %s, 
>> expected 4\n",
>> +                 seg, data->output->name);
>> +        break;
>> +    }
>> +
>> +    igt_remove_fb(display->drm_fd, &buffer);
>> +
>> +    test_fini(data);
>> +}
>> +
>> +igt_main
>> +{
>> +    struct data data;
>> +
>> +    memset(&data, 0, sizeof(data));
>> +
>> +    igt_fixture
>> +    {
>> +        data.fd = drm_open_driver_master(DRIVER_ANY);
>> +
>> +        kmstest_set_vt_graphics_mode();
>> +
>> +        igt_display_require(&data.display, data.fd);
>> +        igt_require(&data.display.is_atomic);
>> +        igt_display_require_output(&data.display);
>> +
>> +    }
>> +
>> +    igt_subtest_f("odm-combine-2-to-1-%s", 
>> test_mode[TEST_MODE_IDX_ODMC_2_TO_1].name)
>> +        run_test_odmc(&data, ODMC_2_TO_1, 
>> &test_mode[TEST_MODE_IDX_ODMC_2_TO_1]);
>> +
>> +    igt_fixture
>> +    {
>> +        igt_display_fini(&data.display);
>> +    }
>> +}
>> diff --git a/tests/amdgpu/meson.build b/tests/amdgpu/meson.build
>> index 6c6166167..6032a38e8 100644
>> --- a/tests/amdgpu/meson.build
>> +++ b/tests/amdgpu/meson.build
>> @@ -35,6 +35,7 @@ if libdrm_amdgpu.found()
>>                 'amd_vm',
>>                 'amd_vrr_range',
>>                 'amd_mall',
>> +              'amd_odm',
>>               ]
>>       amdgpu_deps += libdrm_amdgpu
>>   endif

-- 
--

Thanks & Regards,
Jay

  reply	other threads:[~2023-08-21 20:18 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-21 14:07 [igt-dev] [PATCH v4 1/2] lib/amdgpu: Add check for ODM combine segments debugfs endpoint Aurabindo Pillai
2023-08-21 14:07 ` [igt-dev] [PATCH v4 2/2] tests/amdgpu: Add test for ODM combine Aurabindo Pillai
2023-08-21 15:39   ` Alex Hung
2023-08-21 20:18     ` Aurabindo Pillai [this message]
2023-08-22 19:45       ` Kamil Konieczny
2023-08-22 20:18         ` Aurabindo Pillai
2023-08-23  8:02           ` Kamil Konieczny
2023-08-23 13:38             ` Aurabindo Pillai
2023-08-21 15:38 ` [igt-dev] [PATCH v4 1/2] lib/amdgpu: Add check for ODM combine segments debugfs endpoint Alex Hung
2023-08-21 17:58 ` [igt-dev] ✗ GitLab.Pipeline: warning for series starting with [v4,1/2] " Patchwork
2023-08-21 18:23 ` [igt-dev] ○ CI.xeBAT: info " Patchwork
2023-08-21 18:40 ` [igt-dev] ✗ Fi.CI.BAT: failure " Patchwork
2023-08-22 18:17 ` [igt-dev] ✗ GitLab.Pipeline: warning for series starting with [v4,1/2] lib/amdgpu: Add check for ODM combine segments debugfs endpoint (rev4) Patchwork
2023-08-22 18:46 ` [igt-dev] ○ CI.xeBAT: info " Patchwork
2023-08-22 18:55 ` [igt-dev] ✓ Fi.CI.BAT: success " Patchwork
2023-08-22 23:29 ` [igt-dev] ✗ Fi.CI.IGT: failure " Patchwork
2023-08-23  8:05   ` Kamil Konieczny
2023-08-23 10:26     ` Yedireswarapu, SaiX Nandan
2023-08-23 10:07 ` [igt-dev] ✓ Fi.CI.IGT: success " 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=97b7e6a3-bf81-4a17-b224-1351d8dedf53@amd.com \
    --to=aurabindo.pillai@amd.com \
    --cc=alex.hung@amd.com \
    --cc=hersenxs.wu@amd.com \
    --cc=igt-dev@lists.freedesktop.org \
    --cc=kamil.konieczny@linux.intel.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox