[igt-dev] [PATCH i-g-t] tests/kms_cdclk: Minor fix

Shankar, Uma uma.shankar at intel.com
Wed Jul 7 05:08:38 UTC 2021



> > -----Original Message-----
> > From: Sharma, Swati2 <swati2.sharma at intel.com>
> > Sent: Wednesday, July 7, 2021 2:58 AM
> > To: igt-dev at lists.freedesktop.org
> > Cc: Sharma, Swati2 <swati2.sharma at intel.com>; Shankar, Uma
> > <uma.shankar at intel.com>
> > Subject: [igt-dev][PATCH i-g-t] tests/kms_cdclk: Minor fix

Missed to call out earlier, Please rename the patch header with what is the
minor fix.  Something like: "Fix cdclock failure condition" or something similar.

> > Driver still works in a such way that we never reduce CDCLK, but
> > always only increase it. That was done in order to avoid continuous
> > CDCLK switching,
> >
> > So, in case if new CDCLK is 307200 and there is no change in CDCLK
> > beacuse of above stated driver optimization; pass the test else check lt condition.
> 
> As discussed on offline chats, enhance this test to lower modes for all connectors in
> order to lower cd clock and get a proper DVFS behavior, along with the above
> mentioned kernel change.
> 
> But for now, this change helps avoid getting false failure reports, so:
> Reviewed-by: Uma Shankar <uma.shakar at intel.com>
> 
> >
> > Cc: Uma Shankar <uma.shankar at intel.com>
> > Signed-off-by: Swati Sharma <swati2.sharma at intel.com>
> > ---
> >  tests/kms_cdclk.c | 4 +++-
> >  1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/tests/kms_cdclk.c b/tests/kms_cdclk.c index
> > d2a9fa5f..818cbfd5 100644
> > --- a/tests/kms_cdclk.c
> > +++ b/tests/kms_cdclk.c
> > @@ -31,6 +31,7 @@ IGT_TEST_DESCRIPTION("Test cdclk features : crawling");
> >  #define HDISPLAY_4K     3840
> >  #define VDISPLAY_4K     2160
> >  #define VREFRESH	60
> > +#define MAX_CDCLK_4K    307200
> >
> >  /* Test flags */
> >  enum {
> > @@ -228,7 +229,8 @@ static void test_mode_transition(data_t *data,
> > enum pipe pipe, igt_output_t *out
> >  	igt_info("CD clock frequency %d -> %d\n", cdclk_ref, cdclk_new);
> >
> >  	/* cdclk should bump */
> > -	igt_assert_lt(cdclk_ref, cdclk_new);
> > +	if (cdclk_new != MAX_CDCLK_4K)
> > +		igt_assert_lt(cdclk_ref, cdclk_new);
> >
> >  	/* cleanup */
> >  	do_cleanup_display(display);
> > --
> > 2.25.1



More information about the igt-dev mailing list