They should mostly be fine, as long the shader compiles. The reported
error should be enough to diagnose something going wrong in case of
failure.

This should help avoiding temporary regressions when new properties are
introduced in mesa.git before they are either handled or filtered out by
virgl, as was the case with commit fbe6e92899.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

If the shader failed to be finished, it should be removed from the
hashtable if it was already inserted. Use the goto error path in this
case to handle shader destroy and prevent potential later lookup of
invalid shader from the hashtable.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

Some piglit tests read/write outside of fix-sized arrays. Return an
error in this case. Note: further investigation needed to avoid
such access in the first place.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

This just moves a check outside the loop, as this is a bit of
a hotpath.

We take a reference to this at startup, don't keep looking it up
drop tracking res_handle as well.

This is just a cleanup, to avoid seeing the calculation in a few
places.

If we lookup the same sampler view then ignore this.

This catches about 10% of the sampler view updates in xonotic

If we have a lot of draws we end up traversing the linked list
looking for the same shader on each draw call even if the
shader is the same one we used the last time.

This removes a chunk of CPU overhead in the draw path.

This seems to do better in xonotic traces, we at least don't
traverse as much of the list to pick up the shaders.

I think we should be using a hash table here really.

This with a change in the mesa side should fix the remaining
render condition test fails

They heavily rely on egl/linux currently

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>

Found thanks to IWYU.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>

On some systems, unistd is not implicitely included

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>

To fix warnings on !eventfd systems

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>

This resource validation check was incorrect, fix so that
xonotic can run in the guest again.

this fixes clearbuffer-mixed-format

We weren't converting integer sampler results correctly,
so bad things ensued, this uses the info in the sampler
view to recreate the proper return values and do the
correct casting.

missing case for alignment 6, 2 case is correct.

this fixes a bunch of RGB16F fails.

Need to use the correct index into the array.

We need to ignore the writemask for frag depth writes,
this fixes a bunch of depth draw/copy pixels tests.

Add fragment gl_ViewportIndex support.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>

Fix using undeclared gl_Layer from fragment shader.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>

disconnect using the proper functions, the API in theory
works but mesa has bugs so lets avoid it for now.

This fixes 1d texture depth tests.

We are seeing shaders with 0 and 2 inputs, but no 1, so we need
to handle gaps properly.

This fixes some regressions in drawpixels after some mesa changes
on the guest.

TXQ may query levels and size.

Fixes all piglit texturesize tests and others.

[airlied: use EMIT_BUF_WITH_RET].

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

Commit 53286c6f

 introduced a regression, there is a missing break in the switch..

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

We use the library unconditionally, however not check it in
configure phase. Let's produce an error there and not at
compilation phase.
And while at it, drop -ldl from LDFLAGS too. I suspect it's
coming from the same source because in the gbm pkg-config file
private libs are to be linked with dl. But not only we are not a
gbm private library, we don't use dl*() anywhere.

Signed-off-by: Michal Privoznik <mprivozn@redhat.com>