From: "Modem, Bhanuprakash" <bhanuprakash.modem@intel.com>
To: "Joshi, Kunal1" <kunal1.joshi@intel.com>,
"igt-dev@lists.freedesktop.org" <igt-dev@lists.freedesktop.org>
Subject: Re: [PATCH i-g-t 2/2] tests/intel/kms_big_joiner: add new test to validate force bigjoiner
Date: Wed, 24 Jan 2024 14:48:24 +0530 [thread overview]
Message-ID: <969cdec6-1b12-4b3b-9b68-51aabd7e796c@intel.com> (raw)
In-Reply-To: <CY8PR11MB68893AACF5F28CD7C675237DDA7B2@CY8PR11MB6889.namprd11.prod.outlook.com>
Hi Kunal,
On 24-01-2024 01:55 pm, Joshi, Kunal1 wrote:
> Hello Bhanu,
>
> -----Original Message-----
> From: Modem, Bhanuprakash <bhanuprakash.modem@intel.com>
> Sent: Wednesday, January 24, 2024 1:25 PM
> To: Joshi, Kunal1 <kunal1.joshi@intel.com>; igt-dev@lists.freedesktop.org
> Cc: Sharma, Swati2 <swati2.sharma@intel.com>; B S, Karthik <karthik.b.s@intel.com>
> Subject: Re: [PATCH i-g-t 2/2] tests/intel/kms_big_joiner: add new test to validate force bigjoiner
>
> Hi Kunal,
>
> On 18-01-2024 09:48 pm, Kunal Joshi wrote:
>> Earlier bigjoiner was enabled only for below cases
>> a) for mode having hdisplay > 5k
>> b) if clock requirement were higher than 1 pipe could handle
>>
>> https://patchwork.freedesktop.org/series/124730/
>> With above series we can force bigjoiner without such constraints.
>>
>> Cc: Swati Sharma <swati2.sharma@intel.com>
>> Cc: Karthik B S <karthik.b.s@intel.com>
>> Cc: Bhanuprakash Modem <bhanuprakash.modem@intel.com>
>> Signed-off-by: Kunal Joshi <kunal1.joshi@intel.com>
>> ---
>> tests/intel/kms_big_joiner.c | 92 +++++++++++++++++++++++++++++++++---
>> 1 file changed, 86 insertions(+), 6 deletions(-)
>>
>> diff --git a/tests/intel/kms_big_joiner.c
>> b/tests/intel/kms_big_joiner.c index aba2adfbe..bc1c49023 100644
>> --- a/tests/intel/kms_big_joiner.c
>> +++ b/tests/intel/kms_big_joiner.c
>> @@ -46,8 +46,10 @@
>> *
>> * SUBTEST: 2x-modeset
>> * Description: Verify simultaneous modeset on 2 big joiner outputs
>> + *
>> + * SUBTEST: force-bigjoiner
>> + * Description: Verify bigjoiner on non bigjoiner outputs by force
>> */
>> -
>> IGT_TEST_DESCRIPTION("Test big joiner");
>>
>> struct bigjoiner_output {
>> @@ -67,6 +69,23 @@ typedef struct {
>>
>> static int max_dotclock;
>>
>> +static void force_big_joiner_on_all_outputs(data_t *data, bool
>> +enable) {
>> + igt_output_t *output;
>> +
>> + for_each_connected_output(&data->display, output) {
>> + /*
>> + * Force big joiner on output
>> + */
>> + igt_assert_f(igt_force_bigjoiner_enable(data->drm_fd, output->name, enable) >= 0,
>> + "Failed to %s big joiner for connector %s\n",
>> + enable ? "enable" : "disable", output->name);
>> + igt_assert_f(igt_check_force_bigjoiner_status(data->drm_fd, output->name) == enable ? 1 : 0,
>> + "Failed to force big joiner on output %s\n", output->name);
>> + }
>> + igt_display_commit2(&data->display, COMMIT_ATOMIC); }
>> +
>> static void test_invalid_modeset(data_t *data)
>> {
>> igt_output_t *output;
>> @@ -107,6 +126,53 @@ static void test_invalid_modeset(data_t *data)
>> igt_assert_lt(ret, 0);
>> }
>>
>> +static void cleanup(data_t *data, igt_output_t *output, struct igt_fb *primary_fb, igt_plane_t **primary) {
>> + for_each_connected_output(&data->display, output) {
>> + igt_output_set_pipe(output, PIPE_NONE);
>> + }
>> + force_big_joiner_on_all_outputs(data, false);
>> + free(primary_fb);
>> + free(primary);
>> +}
>> +
>> +static void force_bigjoiner(data_t *data, int simultaneous_bigjoner)
>
> Instead of writing the new logic, let's try to utilize the existing infra. All we need to do is Fix the logic (force bigjoiner) to identify the selected output as bigjoiner supported.
>
> - Bhanu
> [Kunal Joshi]
> The existing logic is to test valid big joiner outputs, this test aims to force big joiner output on
> non bigjoiner output.
>
> So would prefer to add a new test for this behaviour. Please explain if you think otherwise.
Existing logic:
- Iterate all connected outputs
- Check bigjoiner capability (We need to tweak this logic: to force the
bigjoiner & check the capability)
- Add the output to data if it is a bigjoiner capable
- Run the test using available outputs in data
By tweaking the above logic, we can re-use the existing functions to
test the force-bigjoiner.
Please let me know if you need more information.
- Bhanu
>
>> +{
>> + int no_of_outputs;
>> + drmModeModeInfo *mode;
>> + igt_output_t *output;
>> + igt_display_t *display = &data->display;
>> + enum pipe pipe;
>> + struct igt_fb *primary_fb = malloc(data->n_pipes * sizeof(struct igt_fb));
>> + igt_plane_t **primary = malloc(data->n_pipes * sizeof(igt_plane_t
>> +*));
>> +
>> + igt_display_reset(display);
>> + pipe = 0;
>> + no_of_outputs = 0;
>> +
>> + force_big_joiner_on_all_outputs(data, true);
>> + for_each_connected_output(display, output) {
>> + mode = igt_output_get_mode(output);
>> + igt_create_color_fb(data->drm_fd, mode->hdisplay, mode->vdisplay,
>> + DRM_FORMAT_XRGB8888, DRM_FORMAT_MOD_LINEAR, 1.0, 0.0, 0.0, &primary_fb[no_of_outputs]);
>> + igt_output_set_pipe(output, pipe);
>> + primary[no_of_outputs] = igt_output_get_plane_type(output, DRM_PLANE_TYPE_PRIMARY);
>> + igt_plane_set_fb(primary[no_of_outputs], &primary_fb[no_of_outputs]);
>> + igt_info("Using (pipe %s + %s) to run the subtest with mode %dx%d@%d.\n",
>> + kmstest_pipe_name(pipe), igt_output_name(output),
>> + output->override_mode.hdisplay, output->override_mode.vdisplay, output->override_mode.vrefresh);
>> + no_of_outputs++;
>> + pipe = pipe + 2;
>> + if (pipe >= data->n_pipes || no_of_outputs >= simultaneous_bigjoner)
>> + break;
>> + }
>> + if (no_of_outputs != simultaneous_bigjoner) {
>> + cleanup(data, output, primary_fb, primary);
>> + igt_skip("Can't test all outputs\n");
>> + }
>> + igt_display_commit2(display, COMMIT_ATOMIC);
>> + cleanup(data, output, primary_fb, primary); }
>> +
>> static void test_basic_modeset(data_t *data)
>> {
>> drmModeModeInfo *mode;
>> @@ -257,24 +323,27 @@ igt_main
>> j++;
>> }
>>
>> - igt_require_f(count > 0, "No output with 5k+ mode (or) clock > max-dot-clock found\n");
>> -
>> - igt_create_pattern_fb(data.drm_fd, width, height, DRM_FORMAT_XRGB8888,
>> - DRM_FORMAT_MOD_LINEAR, &data.fb);
>> }
>>
>> igt_describe("Verify the basic modeset on big joiner mode on all pipes");
>> igt_subtest_with_dynamic("basic") {
>> + igt_require_f(count > 0, "No output with 5k+ mode (or) clock > max-dot-clock found\n");
>> + igt_create_pattern_fb(data.drm_fd, width, height, DRM_FORMAT_XRGB8888,
>> + DRM_FORMAT_MOD_LINEAR, &data.fb);
>> for (i = 0; i < data.n_pipes - 1; i++) {
>> data.pipe1 = pipe_seq[i];
>> igt_dynamic_f("pipe-%s", kmstest_pipe_name(pipe_seq[i]))
>> test_basic_modeset(&data);
>> }
>> + igt_remove_fb(data.drm_fd, &data.fb);
>> }
>>
>> igt_describe("Verify if the modeset on the adjoining pipe is rejected "
>> "when the pipe is active with a big joiner modeset");
>> igt_subtest_with_dynamic("invalid-modeset") {
>> + igt_require_f(count > 0, "No output with 5k+ mode (or) clock > max-dot-clock found\n");
>> + igt_create_pattern_fb(data.drm_fd, width, height, DRM_FORMAT_XRGB8888,
>> + DRM_FORMAT_MOD_LINEAR, &data.fb);
>> data.pipe1 = pipe_seq[j - 1];
>>
>> igt_display_reset(&data.display);
>> @@ -323,11 +392,15 @@ igt_main
>> test_invalid_modeset(&data);
>> }
>> }
>> + igt_remove_fb(data.drm_fd, &data.fb);
>> }
>>
>> igt_describe("Verify simultaneous modeset on 2 big joiner outputs");
>> igt_subtest_with_dynamic("2x-modeset") {
>> igt_require_f(count > 1, "2 outputs with big joiner modes are
>> required\n");
>> + igt_create_pattern_fb(data.drm_fd, width, height, DRM_FORMAT_XRGB8888,
>> + DRM_FORMAT_MOD_LINEAR, &data.fb);
>> +
>> igt_require_f(data.n_pipes > 3, "Minumum of 4 pipes are required\n");
>> for (i = 0; (i + 2) < data.n_pipes - 1; i++) {
>> data.pipe1 = pipe_seq[i];
>> @@ -335,10 +408,17 @@ igt_main
>> igt_dynamic_f("pipe-%s-%s", kmstest_pipe_name(pipe_seq[i]), kmstest_pipe_name(pipe_seq[i + 2]))
>> test_dual_display(&data);
>> }
>> + igt_remove_fb(data.drm_fd, &data.fb);
>> + }
>> +
>> + igt_describe("Verify bigjoiner on non bigjoiner outputs by force");
>> + igt_subtest_with_dynamic("force-bigjoiner") {
>> + for (i = 1; i < valid_output+1; i++)
>> + igt_dynamic_f("%dx", i)
>> + force_bigjoiner(&data, i);
>> }
>>
>> igt_fixture {
>> - igt_remove_fb(data.drm_fd, &data.fb);
>> igt_display_fini(&data.display);
>> drm_close_driver(data.drm_fd);
>> }
> Thanks and Regards
> [Kunal Joshi]
>
next prev parent reply other threads:[~2024-01-24 9:19 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-01-18 16:18 [PATCH i-g-t 0/2] add force bigjoiner test Kunal Joshi
2024-01-18 16:18 ` [PATCH i-g-t 1/2] lib/igt_kms: add support to force big joiner Kunal Joshi
2024-01-24 7:44 ` Modem, Bhanuprakash
2024-01-18 16:18 ` [PATCH i-g-t 2/2] tests/intel/kms_big_joiner: add new test to validate force bigjoiner Kunal Joshi
2024-01-24 7:54 ` Modem, Bhanuprakash
2024-01-24 8:25 ` Joshi, Kunal1
2024-01-24 9:18 ` Modem, Bhanuprakash [this message]
2024-01-18 17:47 ` ✓ Fi.CI.BAT: success for add force bigjoiner test (rev2) Patchwork
2024-01-18 18:15 ` ✓ CI.xeBAT: " Patchwork
2024-01-18 20:24 ` ✗ Fi.CI.IGT: failure " 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=969cdec6-1b12-4b3b-9b68-51aabd7e796c@intel.com \
--to=bhanuprakash.modem@intel.com \
--cc=igt-dev@lists.freedesktop.org \
--cc=kunal1.joshi@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