[igt-dev] [PATCH i-g-t v3 2/7] tests/kms_cursor_legacy: Read crtc id for enable pipes

[Khajapasha, Mohammed]

> -----Original Message-----
> From: Hiler, Arkadiusz <arkadiusz.hiler at intel.com>
> Sent: Tuesday, June 30, 2020 5:58 PM
> To: Khajapasha, Mohammed <mohammed.khajapasha at intel.com>
> Cc: igt-dev at lists.freedesktop.org
> Subject: Re: [PATCH i-g-t v3 2/7] tests/kms_cursor_legacy: Read crtc id for
> enable pipes
> 
> On Thu, Jun 25, 2020 at 11:53:13AM +0530, Mohammed Khajapasha wrote:
> > Read the crtc ids for enable pipes only in display.
> >
> > Signed-off-by: Mohammed Khajapasha
> <mohammed.khajapasha at intel.com>
> > ---
> >  tests/kms_cursor_legacy.c | 14 +++++++++-----
> >  1 file changed, 9 insertions(+), 5 deletions(-)
> >
> > diff --git a/tests/kms_cursor_legacy.c b/tests/kms_cursor_legacy.c
> > index 344442e8..151bd31d 100644
> > --- a/tests/kms_cursor_legacy.c
> > +++ b/tests/kms_cursor_legacy.c
> > @@ -58,7 +58,7 @@ static void stress(igt_display_t *display,
> >  	uint64_t *results;
> >  	bool torture;
> >  	int n;
> > -	unsigned crtc_id[IGT_MAX_PIPES], num_crtcs;
> > +	unsigned crtc_id[IGT_MAX_PIPES] = {0}, num_crtcs;
> >
> >  	torture = false;
> >  	if (num_children < 0) {
> > @@ -84,8 +84,10 @@ static void stress(igt_display_t *display,
> >  		}
> >  	} else {
> >  		num_crtcs = 1;
> > -		arg.crtc_id = crtc_id[0] = display->pipes[pipe].crtc_id;
> > -		do_ioctl(display->drm_fd, DRM_IOCTL_MODE_CURSOR,
> &arg);
> > +		if(display->pipes[pipe].enabled) {
> > +			arg.crtc_id = crtc_id[0] = display->pipes[pipe].crtc_id;
> > +			do_ioctl(display->drm_fd,
> DRM_IOCTL_MODE_CURSOR, &arg);
> > +		}
> >  	}
> >
> >  	arg.flags = mode;
> > @@ -103,7 +105,8 @@ static void stress(igt_display_t *display,
> >  		hars_petruska_f54_1_random_perturb(child);
> >  		igt_until_timeout(timeout) {
> >  			arg.crtc_id =
> crtc_id[hars_petruska_f54_1_random_unsafe() % num_crtcs];
> > -			do_ioctl(display->drm_fd,
> DRM_IOCTL_MODE_CURSOR, &arg);
> > +			if (arg.crtc_id)
> 
> This seems to introduce assumption that a valid crtc_id != 0, is it true?

Yes, I have rechecked by running kms_cursor_legacy test and crtc id cann't be 0,
The subtest case like all-pipes-forked-move, getting failed when crtc id is 0.

--
Khajapasha
> 
> --
> Cheers,
> Arek
> 
> > +				do_ioctl(display->drm_fd,
> DRM_IOCTL_MODE_CURSOR, &arg);
> >  			count++;
> >  		}
> >
> > @@ -1390,7 +1393,8 @@ igt_main
> >  			errno = 0;
> >
> >  			igt_fixture {
> > -				igt_skip_on(n >= display.n_pipes);
> > +				igt_require_f(display.pipes[n].enabled,
> > +						"Pipe-%s not enabled\n",
> kmstest_pipe_name(n));
> >  			}
> >
> >  			igt_subtest_f("pipe-%s-single-bo",
> kmstest_pipe_name(n))
> > --
> > 2.24.1
> >