[igt-dev] [PATCH i-g-t 1/4] xe/xe_sysfs: Updated tests to use xe_sysfs_gt_open

Upadhyay, Tejas tejas.upadhyay at intel.com
Wed Aug 2 10:00:04 UTC 2023



> -----Original Message-----
> From: Dandamudi, Priyanka <priyanka.dandamudi at intel.com>
> Sent: Wednesday, August 2, 2023 3:24 PM
> To: Upadhyay, Tejas <tejas.upadhyay at intel.com>; Kumar, Janga Rahul
> <janga.rahul.kumar at intel.com>; igt-dev at lists.freedesktop.org
> Subject: RE: [PATCH i-g-t 1/4] xe/xe_sysfs: Updated tests to use
> xe_sysfs_gt_open
> 
> 
> 
> > -----Original Message-----
> > From: Upadhyay, Tejas <tejas.upadhyay at intel.com>
> > Sent: 02 August 2023 03:11 PM
> > To: Dandamudi, Priyanka <priyanka.dandamudi at intel.com>; Kumar, Janga
> > Rahul <janga.rahul.kumar at intel.com>; igt-dev at lists.freedesktop.org
> > Subject: RE: [PATCH i-g-t 1/4] xe/xe_sysfs: Updated tests to use
> > xe_sysfs_gt_open
> >
> >
> >
> > > -----Original Message-----
> > > From: Dandamudi, Priyanka <priyanka.dandamudi at intel.com>
> > > Sent: Wednesday, August 2, 2023 2:09 PM
> > > To: Kumar, Janga Rahul <janga.rahul.kumar at intel.com>; igt-
> > > dev at lists.freedesktop.org; Upadhyay, Tejas
> > > <tejas.upadhyay at intel.com>; Dandamudi, Priyanka
> > > <priyanka.dandamudi at intel.com>
> > > Subject: [PATCH i-g-t 1/4] xe/xe_sysfs: Updated tests to use
> > > xe_sysfs_gt_open
> > >
> > > From: Priyanka Dandamudi <priyanka.dandamudi at intel.com>
> > >
> > > Tests have been updated to use xe_sysfs_gt_open function.
> > >
> > > v2: Updated code to check when root user try to set value beyond
> > > local min/max scheduler property.
> > >
> > > Cc: Tejas Upadhyay <tejas.upadhyay at intel.com>
> > > Signed-off-by: Priyanka Dandamudi <priyanka.dandamudi at intel.com>
> > > ---
> > >  tests/xe/xe_sysfs_defaults.c  | 10 ++++++----
> > > tests/xe/xe_sysfs_scheduler.c
> > > | 12 +++++++-----
> > >  2 files changed, 13 insertions(+), 9 deletions(-)
> > >
> > > diff --git a/tests/xe/xe_sysfs_defaults.c
> > > b/tests/xe/xe_sysfs_defaults.c index 5c9de43e9..ff958161b 100644
> > > --- a/tests/xe/xe_sysfs_defaults.c
> > > +++ b/tests/xe/xe_sysfs_defaults.c
> > > @@ -65,25 +65,27 @@ igt_main
> > >
> > >  		sys_fd = igt_sysfs_open(xe);
> > >  		igt_require(sys_fd != -1);
> > > +		close(sys_fd);
> > >  	}
> > >
> > >  	igt_subtest_with_dynamic("engine-defaults") {
> > >  		xe_for_each_gt(xe, gt) {
> > >  			int engines_fd = -1;
> > > -			char buf[100];
> > > +			int gt_fd = -1;
> > >
> > > -			sprintf(buf, "device/gt%d/engines", gt);
> > > -			engines_fd = openat(sys_fd, buf, O_RDONLY);
> > > +			gt_fd = xe_sysfs_gt_open(xe, gt);
> > > +			igt_require(gt_fd != -1);
> > > +			engines_fd = openat(gt_fd, "engines", O_RDONLY);
> > >  			igt_require(engines_fd != -1);
> > >
> > >  			igt_sysfs_engines(xe, engines_fd, NULL,
> > test_defaults);
> > >
> > >  			close(engines_fd);
> > > +			 close(gt_fd);
> > >  		}
> > >  	}
> > >
> > >  	igt_fixture {
> > > -		close(sys_fd);
> > >  		xe_device_put(xe);
> > >  		close(xe);
> > >  	}
> > > diff --git a/tests/xe/xe_sysfs_scheduler.c
> > > b/tests/xe/xe_sysfs_scheduler.c index 1da0f541a..f199d6be8 100644
> > > --- a/tests/xe/xe_sysfs_scheduler.c
> > > +++ b/tests/xe/xe_sysfs_scheduler.c
> > > @@ -94,7 +94,7 @@ static void test_min_max(int xe, int engine, const
> > > char
> > > **property)
> > >
> > >  	igt_sysfs_printf(engine, property[0], "%d", default_min);
> > >  	igt_sysfs_scanf(engine, property[0], "%u", &set);
> > > -	igt_assert_eq(set, default_min);
> > > +	igt_assert_neq(set, default_min);
> >
> > Also what is default_min here, is it min/max under .defaults/? To test
> > min/max, I think it should sysfs min/max that we should check against.
> 
> default_min here is under min/max under .defaults. This is sort of a negative
> test when we try to test beyond sysfs min/max and within default min/max it
> should fail. This I changed to negative, as the requirement has been changed.
> 

Ok, how about checking negative test also keeping sysfs min/max in equation. As requirement is all limited in sysfs min/max range. Default has no relevant here.

Am I right?

Thanks,
Tejas  

> Regards,
> Priyanka
> 
> >
> > Tejas
> > >
> > >  	igt_sysfs_printf(engine, property[0], "%d", min);
> > >  	igt_sysfs_scanf(engine, property[0], "%u", &set); @@ -184,6 +184,7
> > > @@ igt_main
> > >
> > >  		sys_fd = igt_sysfs_open(xe);
> > >  		igt_require(sys_fd != -1);
> > > +		close(sys_fd);
> > >  	}
> > >
> > >  	for (int i = 0; i < count; i++) {
> > > @@ -191,20 +192,21 @@ igt_main
> > >  			igt_subtest_with_dynamic_f("%s-%s", property[i][0],
> > > t->name) {
> > >  				xe_for_each_gt(xe, gt) {
> > >  					int engines_fd = -1;
> > > -					char buf[100];
> > > +					int gt_fd = -1;
> > >
> > > -					sprintf(buf, "device/gt%d/engines",
> > > gt);
> > > -					engines_fd = openat(sys_fd, buf,
> > > O_RDONLY);
> > > +					gt_fd = xe_sysfs_gt_open(xe, gt);
> > > +					igt_require(gt_fd != -1);
> > > +					engines_fd = openat(gt_fd, "engines",
> > > O_RDONLY);
> > >  					igt_require(engines_fd != -1);
> > >
> > >  					igt_sysfs_engines(xe, engines_fd,
> > property[i], t->fn);
> > >  					close(engines_fd);
> > > +					close(gt_fd);
> > >  				}
> > >  			}
> > >  		}
> > >  	}
> > >  	igt_fixture {
> > > -		close(sys_fd);
> > >  		xe_device_put(xe);
> > >  		close(xe);
> > >  	}
> > > --
> > > 2.25.1



More information about the igt-dev mailing list