[PATCH i-g-t 3/4] tests/kms_content_protection: Add retry logic for mst usecase
Kandpal, Suraj
suraj.kandpal at intel.com
Mon Aug 19 05:32:09 UTC 2024
> -----Original Message-----
> From: B, Jeevan <jeevan.b at intel.com>
> Sent: Monday, August 19, 2024 10:47 AM
> To: Kandpal, Suraj <suraj.kandpal at intel.com>; igt-dev at lists.freedesktop.org
> Cc: Samala, Pranay <pranay.samala at intel.com>; Nautiyal, Ankit K
> <ankit.k.nautiyal at intel.com>; Kandpal, Suraj <suraj.kandpal at intel.com>
> Subject: RE: [PATCH i-g-t 3/4] tests/kms_content_protection: Add retry logic
> for mst usecase
>
> > -----Original Message-----
> > From: igt-dev <igt-dev-bounces at lists.freedesktop.org> On Behalf Of
> > Suraj Kandpal
> > Sent: Tuesday, August 13, 2024 9:50 AM
> > To: igt-dev at lists.freedesktop.org
> > Cc: Samala, Pranay <pranay.samala at intel.com>; Nautiyal, Ankit K
> > <ankit.k.nautiyal at intel.com>; Kandpal, Suraj <suraj.kandpal at intel.com>
> > Subject: [PATCH i-g-t 3/4] tests/kms_content_protection: Add retry
> > logic for mst usecase
> >
> > Add mst retry logic where it retries 3 times before failing the test.
> > After every retry mst setup needs to be disabled so we need to add a
> > function which disables mst in one commit.
> >
> > Signed-off-by: Suraj Kandpal <suraj.kandpal at intel.com>
> > ---
> > tests/kms_content_protection.c | 84
> > +++++++++++++++++++++++++++++-----
> > 1 file changed, 73 insertions(+), 11 deletions(-)
> >
> > diff --git a/tests/kms_content_protection.c
> > b/tests/kms_content_protection.c index db6dc17b1..82a56167b 100644
> > --- a/tests/kms_content_protection.c
> > +++ b/tests/kms_content_protection.c
> > @@ -278,6 +278,35 @@ static bool test_cp_enable(igt_output_t *output,
> > enum igt_commit_style s,
> > return ret;
> > }
> >
> > +static void test_mst_cp_disable(igt_output_t *hdcp_mst_output[], enum
> > igt_commit_style s,
> Update enum from s to some proper name instead of a char.
> > + int valid_outputs)
> > +{
> > + igt_display_t *display = &data.display;
> > + igt_plane_t *primary;
> > + bool ret;
> > + int count;
> > + u64 val;
> > +
> > + for (count = 0; count < valid_outputs; count++) {
> > + primary =
> igt_output_get_plane_type(hdcp_mst_output[count],
> > DRM_PLANE_TYPE_PRIMARY);
> > + igt_plane_set_fb(primary, &data.red);
> > + igt_output_set_prop_value(hdcp_mst_output[count],
> > IGT_CONNECTOR_CONTENT_PROTECTION,
> > + CP_UNDESIRED);
> > + }
> > +
> > + igt_display_commit2(display, s);
> > +
> > + ret = wait_for_prop_value(hdcp_mst_output[0], CP_UNDESIRED,
> > + KERNEL_DISABLE_TIME_ALLOWED_MSEC);
> > + for (count = 1; count < valid_outputs; count++) {
> > + val = igt_output_get_prop(hdcp_mst_output[count],
> > +
> > IGT_CONNECTOR_CONTENT_PROTECTION);
> > + ret &= (val == CP_UNDESIRED);
> > + }
> > +
> > + igt_assert_f(ret, "Content Protection not cleared\n"); }
> Suggest to update string to "Content Protection not cleared on all outputs"
Sure will do.
> > +
> > static void test_cp_disable(igt_output_t *output, enum igt_commit_style s)
> {
> > igt_display_t *display = &data.display; @@ -630,6 +659,49 @@ static
> > void test_cp_lic_on_mst(igt_output_t *mst_outputs[], int valid_outputs, b
> > }
> > }
> >
> > +static bool
> > +test_mst_cp_enable_with_retry(igt_output_t *hdcp_mst_output[], int
> Here we are trying to enable mst with cp, "test" should be removed.
I think test here seems appropriate since not only is enabling MST we also have assertions to
check if CP has been enabled or not also, this follows a similar style of naming found in this
test when retrying for a single connector.
Also a suggestion maybe adding a new line before and after the review make the inline comments
more clearer to see I almost missed the two comments I address in the reply
Regards,
Suraj Kandpal
> > valid_outputs,
> > + int retries, int content_type) {
> > + igt_display_t *display = &data.display;
> > + int retry_orig = retries, count;
> > + bool ret;
> > + u64 val;
> > +
> > + do {
> > + if (retry_orig != retries)
> > + test_mst_cp_disable(hdcp_mst_output,
> > COMMIT_ATOMIC, valid_outputs);
> > +
> > + for (count = 0; count < valid_outputs; count++) {
> > + igt_output_set_prop_value(hdcp_mst_output[count],
> > +
> > IGT_CONNECTOR_CONTENT_PROTECTION, CP_DESIRED);
> > +
> > + if (hdcp_mst_output[count]-
> > >props[IGT_CONNECTOR_HDCP_CONTENT_TYPE])
> > +
> > igt_output_set_prop_value(hdcp_mst_output[count],
> > +
> > IGT_CONNECTOR_HDCP_CONTENT_TYPE,
> > + content_type);
> > + }
> > +
> > + igt_display_commit2(display, COMMIT_ATOMIC);
> > +
> > + ret = wait_for_prop_value(hdcp_mst_output[0], CP_ENABLED,
> > +
> > KERNEL_AUTH_TIME_ALLOWED_MSEC);
> > + for (count = 1; count < valid_outputs; count++) {
> > + val = igt_output_get_prop(hdcp_mst_output[count],
> > +
> > IGT_CONNECTOR_CONTENT_PROTECTION);
> > + ret &= (val == CP_ENABLED);
> > + }
> > +
> > + retries -= 1;
> > + if (!ret || retries)
> > + igt_debug("Retry %d/3\n", 3 - retries);
> > + } while (retries && !ret);
> > +
> > + igt_assert_f(ret, "Content Protection not enabled on MST output\n");
> > +
> > + return ret;
> > +}
> > +
> > static void
> > test_content_protection_mst(int content_type) { @@ -684,19 +756,9 @@
> > test_content_protection_mst(int content_type)
> > igt_require_f(ret == 0, "Commit failure during MST
> modeset\n");
> > }
> >
> > - for (count = 0; count < valid_outputs; count++) {
> > - igt_output_set_prop_value(hdcp_mst_output[count],
> > IGT_CONNECTOR_CONTENT_PROTECTION, CP_DESIRED);
> > -
> > - if (output->props[IGT_CONNECTOR_HDCP_CONTENT_TYPE])
> > - igt_output_set_prop_value(hdcp_mst_output[count],
> > IGT_CONNECTOR_HDCP_CONTENT_TYPE, content_type);
> > - }
> > -
> > igt_display_commit2(display, COMMIT_ATOMIC);
> >
> > - for (count = 0; count < valid_outputs; count++) {
> > - ret = wait_for_prop_value(hdcp_mst_output[count],
> > CP_ENABLED, KERNEL_AUTH_TIME_ALLOWED_MSEC);
> > - igt_assert_f(ret, "Content Protection not enabled on %s\n",
> > hdcp_mst_output[count]->name);
> > - }
> > + ret = test_mst_cp_enable_with_retry(hdcp_mst_output,
> valid_outputs,
> > 2,
> > +content_type);
> >
> > if (data.cp_tests & CP_LIC)
> > test_cp_lic_on_mst(hdcp_mst_output, valid_outputs, 0);
> > --
> > 2.43.2
More information about the igt-dev
mailing list