1. 25 May, 2019 1 commit
  2. 17 May, 2019 1 commit
  3. 16 May, 2019 15 commits
  4. 13 May, 2019 2 commits
    • Paulo Zanoni's avatar
      egl-query-surface: really print the width in the error message · ac955893
      Paulo Zanoni authored
      This looks like a copy/paste error.
      Signed-off-by: Paulo Zanoni's avatarPaulo Zanoni <paulo.r.zanoni@intel.com>
    • Paulo Zanoni's avatar
      egl: ask the X11 WM to keep windows the size we expect · b0ba8444
      Paulo Zanoni authored
      Despite the fact that the windows are very small, my Window Manager
      (Cinnamon) decides to not create them with the original size (it does
      118x136 instead of 119x137). This makes egl-query-suface
      --attr=EGL_{WIDTH,HEIGHT} fail on my machine.  Simply using
      XSetWMNormalHints seems to fix the problem for me.
      More stubborn WMs may still still decide to not comply with the
      request, so an alternative could be to query the window size through
      X11 and compare that to what EGL gives us, but I would prefer to try
      the less invasive approach first.
       - We don't need the alloc/free functions (Eric Engestrom).
       - Also set the obsolete width/height flags since they don't work and
         really old WMs might be looking at them.
       - Don't create a separate function since  now we only have a single
         function call.
      Tested-by: Eric Engestrom's avatarEric Engestrom <eric.engestrom@intel.com>
      Reviewed-by: Tapani Pälli <tapani.palli@intel.com> (v1)
      Signed-off-by: Paulo Zanoni's avatarPaulo Zanoni <paulo.r.zanoni@intel.com>
  5. 09 May, 2019 1 commit
  6. 08 May, 2019 1 commit
    • Pierre-Eric Pelloux-Prayer's avatar
      Avoid undefined behavior in glsl-routing test · 17bccd6b
      Pierre-Eric Pelloux-Prayer authored
      This test reads attribute in the FS (e.g: gl_SecondaryColor) without writing
      them in the VS.
      It shouldn't affect the output since in this case the eq() function returns 0.0,
      so it's equivalent to: r += 0.0 * gl_SecondaryColor.
      The problems are:
        - if gl_SecondaryColor is unwritten, its value is allowed to be NaN, in which
      case it will affect the output color and test will fail even if the driver is valid
        - the GLSL 1.20 specs says (7.6 Varying variables):
      "The values in gl_Color  and  gl_SecondaryColor  will be derived automatically by
      the system from gl_FrontColor, gl_BackColor, gl_FrontSecondaryColor, and
      gl_BackSecondaryColor  based on which face is visible."
      And the paragraph before says:
      "A particular [varying variable] should be written to if any functionality in a
      corresponding fragment shader or fixed pipeline uses it or state derived from it.
      Otherwise, behavior is undefined."
      This commit replaces the 'r += eq(...) * gl_SecondaryColor' lines with
      'if (eq(...)) r += gl_SecondaryColor' ones which aren't affected by the above
      Fixes the test on my AMD Vega10 system.
      Reviewed-by: Eric Anholt's avatarEric Anholt <eric@anholt.net>
  7. 07 May, 2019 1 commit
  8. 06 May, 2019 1 commit
  9. 02 May, 2019 2 commits
  10. 30 Apr, 2019 4 commits
  11. 24 Apr, 2019 1 commit
  12. 23 Apr, 2019 5 commits
  13. 22 Apr, 2019 1 commit
  14. 20 Apr, 2019 1 commit
  15. 18 Apr, 2019 1 commit
    • Andres Gomez's avatar
      arb_separate_shader_objects: add gl_PerVertex tests · c8aa5747
      Andres Gomez authored
      From the ARB_separate_shader_objects spec v.25:
        " To use any built-in input or output in the gl_PerVertex and
          gl_PerFragment blocks in separable program objects, shader code
          must redeclare those blocks prior to use.  A separable program
          will fail to link if:
            * it contains multiple shaders of a single type with different
              redeclarations of these built-in input and output blocks; or
            * any shader uses a built-in block member not found in the
              redeclaration of that block."
      Additionally, added the "SEPARABLE PROGRAM ENABLED" directive to the
      Signed-off-by: Andres Gomez's avatarAndres Gomez <agomez@igalia.com>
      Acked-by: 's avatarMarek Olšák <marek.olsak@amd.com>
      Reviewed-by: 's avatarIan Romanick <ian.d.romanick@intel.com>
  16. 17 Apr, 2019 2 commits