[Piglit] [PATCH] arb_program_interface_query: corrected AoA's index variable expectation

[Andres Gomez]

On Sat, 2019-02-16 at 10:05 +1100, Timothy Arceri wrote:
> NAK.
> 
> The problem is this will end up making the test fail on the Nvidia blob.
> Technically neither is incorrect, but the test does show Mesa's failure 
> to detect the unused element. Again this is not technically a failure of 
> the spec as it's dependent on the implementations ability to detect 
> active array elements. However I'd rather leave this than work around 
> our substandard detection of inactive elements.

I wonder if it's really worthy, then, to test something that is
implementation dependent.

> See [1] for more information, and note I also rejected the Mesa solution 
> proposed by Andrii in the bug report because it was too much code for 
> something that didn't actually remove the unused components but just hid 
> them from the resource list.
> 
> I think if we actually want to fix this properly then we could do it by 
> making a NIR linker for GLSL.
> 
> [1] https://bugs.freedesktop.org/show_bug.cgi?id=92822

OK. Thanks for the thorough explanation! I'll drop this patch.

> 
> On 9/2/19 3:59 am, Andres Gomez wrote:
> > Naming conventions, from the GL_ARB_program_interface_query extension:
> > 
> >   "   * For an active variable declared as an array of an aggregate
> >         data type (structures or arrays), a separate entry will be
> >         generated for each active array element, unless noted
> >         immediately below.  The name of each entry is formed by
> >         concatenating the name of the array, the "[" character, an
> >         integer identifying the element number, and the "]" character.
> >         These enumeration rules are applied recursively, treating each
> >         enumerated array element as a separate active variable."
> > 
> > Cc: Timothy Arceri <tarceri at itsqueeze.com>
> > Cc: Martin Peres <martin.peres at linux.intel.com>
> > Signed-off-by: Andres Gomez <agomez at igalia.com>
> > ---
> >   .../spec/arb_program_interface_query/getprogramresourceindex.c  | 2 +-
> >   1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/tests/spec/arb_program_interface_query/getprogramresourceindex.c b/tests/spec/arb_program_interface_query/getprogramresourceindex.c
> > index 16b38e2d5..2afc9eeb9 100755
> > --- a/tests/spec/arb_program_interface_query/getprogramresourceindex.c
> > +++ b/tests/spec/arb_program_interface_query/getprogramresourceindex.c
> > @@ -167,7 +167,7 @@ static const struct subtest_index_t index_subtests[] = {
> >   	{   vs_aofa,              GL_PROGRAM_INPUT,          "vs_input2", false, -1, GL_NO_ERROR },
> >   	{   vs_aofa,              GL_PROGRAM_INPUT,       "vs_input2[0]",  true, -1, GL_NO_ERROR },
> >   	{   vs_aofa,              GL_PROGRAM_INPUT,    "vs_input2[0][0]",  true, -1, GL_NO_ERROR },
> > -	{   vs_aofa,              GL_PROGRAM_INPUT,    "vs_input2[1][0]", false, -1, GL_NO_ERROR },
> > +	{   vs_aofa,              GL_PROGRAM_INPUT,    "vs_input2[1][0]",  true, -1, GL_NO_ERROR },
> >   	{   vs_aofa,              GL_PROGRAM_INPUT,    "vs_input2[0][1]", false, -1, GL_NO_ERROR },
> >   	{    vs_sub,          GL_VERTEX_SUBROUTINE,                "vss",  true, -1, GL_NO_ERROR },
> >   	{    vs_sub,          GL_VERTEX_SUBROUTINE,               "vss2",  true, -1, GL_NO_ERROR },
> > 
-- 
Br,

Andres