[igt-dev] [PATCH i-g-t] tests/kms_multipipe_modeset: Add test to validate max pipe configuration
Karthik B S
karthik.b.s at intel.com
Thu Apr 23 09:50:34 UTC 2020
Thanks a lot for the review.
On 4/22/2020 1:47 PM, Petri Latvala wrote:
> On Mon, Apr 13, 2020 at 03:51:44PM +0530, Karthik B S wrote:
>> Added test to validate max pipe configuration for a platform.
>> In the test, the reference CRC is collected first by doing modeset
>> on all the outputs individually.Then a simultaneous modeset is
>> done on all the outputs and the CRC is collected. This is compared
>> with the reference CRC.
>>
>> If the number of outputs connected is less than the number of pipes
>> supported by the platform, then the test skips.
>>
>> This test is added to verify if the max pipe configuration for a
>> given platform is working fine. Even though there are other tests
>> which test multipipe configuration, they test some other functionalities
>> as well together with multipipe. This is a stand alone test that
>> intends to only verify simultaneous modeset at the max pipe configuration.
>>
>> Signed-off-by: Karthik B S <karthik.b.s at intel.com>
>> ---
>> tests/Makefile.sources | 1 +
>> tests/kms_multipipe_modeset.c | 180 ++++++++++++++++++++++++++++++++++
>> tests/meson.build | 1 +
>> 3 files changed, 182 insertions(+)
>> create mode 100644 tests/kms_multipipe_modeset.c
>>
>> diff --git a/tests/Makefile.sources b/tests/Makefile.sources
>> index 4e44c98c..903c5671 100644
>> --- a/tests/Makefile.sources
>> +++ b/tests/Makefile.sources
>> @@ -60,6 +60,7 @@ TESTS_progs = \
>> kms_lease \
>> kms_legacy_colorkey \
>> kms_mmap_write_crc \
>> + kms_multipipe_modeset \
>> kms_panel_fitting \
>> kms_pipe_b_c_ivb \
>> kms_pipe_crc_basic \
>> diff --git a/tests/kms_multipipe_modeset.c b/tests/kms_multipipe_modeset.c
>> new file mode 100644
>> index 00000000..11328165
>> --- /dev/null
>> +++ b/tests/kms_multipipe_modeset.c
>> @@ -0,0 +1,180 @@
>> +/*
>> + * Copyright © 2020 Intel Corporation
>> + *
>> + * Permission is hereby granted, free of charge, to any person obtaining a
>> + * copy of this software and associated documentation files (the "Software"),
>> + * to deal in the Software without restriction, including without limitation
>> + * the rights to use, copy, modify, merge, publish, distribute, sublicense,
>> + * and/or sell copies of the Software, and to permit persons to whom the
>> + * Software is furnished to do so, subject to the following conditions:
>> + *
>> + * The above copyright notice and this permission notice (including the next
>> + * paragraph) shall be included in all copies or substantial portions of the
>> + * Software.
>> + *
>> + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
>> + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
>> + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
>> + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
>> + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
>> + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
>> + * IN THE SOFTWARE.
>> + *
>> + * Author:
>> + * Karthik B S <karthik.b.s at intel.com>
>> + */
>> +
>> +#include "igt.h"
>> +
>> +IGT_TEST_DESCRIPTION("Test simultaneous modeset on all the supported pipes");
>> +
>> +typedef struct {
>> + int drm_fd;
>> + igt_display_t display;
>> + struct igt_fb fb;
>> +} data_t;
>> +
>> +static void run_test(data_t *data, int valid_outputs)
>> +{
>> + igt_output_t *output;
>> + igt_pipe_crc_t *pipe_crcs[IGT_MAX_PIPES] = { 0 };
>> + igt_crc_t ref_crcs[IGT_MAX_PIPES], new_crcs[IGT_MAX_PIPES];
>> + igt_display_t *display = &data->display;
>> + int width = 0, height = 0, i, count = 0;
>> + int output_to_pipe[valid_outputs];
>> + igt_pipe_t *pipe;
>> + igt_plane_t *plane;
>> + drmModeModeInfo *mode;
>> +
>> + for_each_connected_output(display, output) {
>> + mode = igt_output_get_mode(output);
>> + igt_assert(mode);
>> +
>> + igt_output_set_pipe(output, PIPE_NONE);
>> +
>> + width = max(width, mode->hdisplay);
>> + height = max(height, mode->vdisplay);
>> + }
>> +
>> + igt_create_pattern_fb(data->drm_fd, width, height, DRM_FORMAT_XRGB8888,
>> + LOCAL_DRM_FORMAT_MOD_NONE, &data->fb);
>> +
>> + /* Make pipe-ouput mapping which will be used unchanged across the test */
>
> Typo, pipe-output
>
Noted. I'll fix this.
>
>> + for_each_pipe(display, i) {
>> + count = 0;
>> + for_each_connected_output(display, output) {
>> + if (igt_pipe_connector_valid(i, output) &&
>> + output->pending_pipe == PIPE_NONE) {
>> + igt_output_set_pipe(output, i);
>> + output_to_pipe[count] = i;
>> + break;
>> + }
>> + count++;
>> + }
>> + }
>> +
>> + /* Collect reference CRC by Committing individually on all outputs*/
>> + for_each_pipe(display, i) {
>> + pipe = &display->pipes[i];
>> + plane = igt_pipe_get_plane_type(pipe, DRM_PLANE_TYPE_PRIMARY);
>> +
>> + mode = NULL;
>> +
>> + if (is_i915_device(display->drm_fd))
>> + pipe_crcs[i] = igt_pipe_crc_new(display->drm_fd, i,
>> + INTEL_PIPE_CRC_SOURCE_AUTO);
>> +
>> + count = 0;
>> + for_each_connected_output(display, output) {
>> + if (output_to_pipe[count] == pipe->pipe) {
>> + igt_output_set_pipe(output, i);
>> + mode = igt_output_get_mode(output);
>> + igt_assert(mode);
>> + } else
>> + igt_output_set_pipe(output, PIPE_NONE);
>> + count++;
>> + }
>> +
>> + igt_plane_set_fb(plane, &data->fb);
>> + igt_fb_set_size(&data->fb, plane, mode->hdisplay, mode->vdisplay);
>> + igt_plane_set_size(plane, mode->hdisplay, mode->vdisplay);
>> +
>> + igt_display_commit2(display, COMMIT_ATOMIC);
>> + igt_pipe_crc_collect_crc(pipe_crcs[i], &ref_crcs[i]);
>
> This will crash on non-i915. igt_pipe_crc_collect_crc needs a non-NULL pipe_crc.
>
Could you please suggest how I can handle this?
Does it mean I'll have to make the test Intel specific?
>
>> + }
>> +
>> + /* Simultaneously commit on all outputs */
>> + for_each_pipe(display, i) {
>> + pipe = &display->pipes[i];
>> + plane = igt_pipe_get_plane_type(pipe, DRM_PLANE_TYPE_PRIMARY);
>> +
>> + mode = NULL;
>> +
>> + count = 0;
>> + for_each_connected_output(display, output) {
>> + if (output_to_pipe[count] == pipe->pipe) {
>> + igt_output_set_pipe(output, i);
>> + mode = igt_output_get_mode(output);
>> + igt_assert(mode);
>> + break;
>> + }
>> + count++;
>> + }
>> +
>> + igt_plane_set_fb(plane, &data->fb);
>> + igt_fb_set_size(&data->fb, plane, mode->hdisplay, mode->vdisplay);
>> + igt_plane_set_size(plane, mode->hdisplay, mode->vdisplay);
>> + }
>> +
>> + igt_display_commit2(display, COMMIT_ATOMIC);
>> +
>> + /* CRC Verification */
>> + for (count = 0; count < valid_outputs; count++) {
>> + igt_pipe_crc_collect_crc(pipe_crcs[count], &new_crcs[count]);
>> + igt_assert_crc_equal(&ref_crcs[count], &new_crcs[count]);
>> + }
>> +
>> + igt_plane_set_fb(plane, NULL);
>> + igt_remove_fb(data->drm_fd, &data->fb);
>> +}
>> +
>> +static void test_multipipe(data_t *data)
>> +{
>> + igt_output_t *output;
>> + int valid_outputs = 0, num_pipes;
>> +
>> + num_pipes = igt_display_get_n_pipes(&data->display);
>> + for_each_connected_output(&data->display, output)
>> + valid_outputs++;
>> +
>> + igt_require_f(valid_outputs == num_pipes,
>> + "Number of connected outputs not equal to the "
>> + "number of pipes supported\n");
>
> For debugging aid, please print the numbers as well.
>
Sure I'll update this.
Thanks,
Karthik
>
>
>
More information about the igt-dev
mailing list