[Piglit] [PATCH] shader_runner: report PIGLIT_FAIL if linking unexpectedly fails

[Dylan Baker]

Quoting Brian Paul (2016-10-18 17:18:36)
> On 10/18/2016 05:43 PM, Dylan Baker wrote:
> > Quoting Brian Paul (2016-10-18 16:28:53)
> >> We were previously reporting 'pass' if linking failed.
> >> ---
> >>   tests/shaders/shader_runner.c | 1 +
> >>   1 file changed, 1 insertion(+)
> >>
> >> diff --git a/tests/shaders/shader_runner.c b/tests/shaders/shader_runner.c
> >> index b0bde2c..94865b7 100644
> >> --- a/tests/shaders/shader_runner.c
> >> +++ b/tests/shaders/shader_runner.c
> >> @@ -3529,6 +3529,7 @@ piglit_display(void)
> >>
> >>          if (!link_ok && !link_error_expected) {
> >>                  program_must_be_in_use();
> >> +               full_result = PIGLIT_FAIL;
> >>          }
> >>
> >>          piglit_present_results();
> >> --
> >> 1.9.1
> >>
> >> _______________________________________________
> >> Piglit mailing list
> >> Piglit at lists.freedesktop.org
> >> https://lists.freedesktop.org/mailman/listinfo/piglit
> >
> > And, after looking some more, there's a lot of cases where we need to handle
> > this.
> 
> Yeah, I saw that and wasn't sure what was intended.  My take was 
> regardless of what program_must_be_in_use() returns, the fact that if 
> link_ok==false and link_error_expected==false it means the test failed.
> 
> Changing it to full_result = program_must_be_in_use() works too.  I'll 
> update the patch.
> 
> -Brian
> 

That's probably correct. I think either just setting "full_results =
PIGLIT_FAIL" or "full_result = program_must_be_in_use()" is correct, but just
calling the function for no reason then setting it to PIGLIT_FAIL seems a little
strange.

With either of those changes you can add my rb,
Reviewed-by: Dylan Baker <dylan at pnwbakers.com>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 455 bytes
Desc: signature
URL: <https://lists.freedesktop.org/archives/piglit/attachments/20161019/d84b392c/attachment.sig>