This lets us avoid allocing new buffers for renderbuffers, finalized miptrees,
and PBO-uploaded textures when there's an unreferenced but still active one
cached, while also avoiding CPU waits for batchbuffers and CPU-uploaded
textures. The size of BOs allocated for a desktop running current GL
cairogears on i915 is cut in half with this.
Note that this means we require libdrm 2.4.5.
(cherry picked from commit 40dd024be6)
We were asking for something illegal (write_domain != 0 && read_domains !=
write_domain) because at the time of writing the region surfaces were used
for texturing occasionally as well, and we weren't really clear on the model
GEM was going to use.
This reliably triggered a kernel bug with domain handling, resulting in
oglconform mustpass.c failure. Of course, it only became visible after
01bc4d441f cleaned up some gratuitous flushing.
(cherry picked from commit 078e8a61b2)
It is possible that an object whose vertices all are outside of a
view plane is passed to clip thread due to the RHW workaround. This
object should be rejected by clip thread. Fix bug #19879
(cherry picked from commit 68915fd6fa)
The default for EmitCondCodes got flipped when gallium-0.2 was merged.
This fixes GLSL if/else/endif regressions.
Drivers that use GLSL should always explicitly set the flag to be safe.
(cherry picked from commit a9e753c84c)
Tries to use either GLX_MESA_swap_control or GLX_SGI_video_sync to
detect whether the display is synchronized to the vertical blank. If
it detects this, a message will be printed. HOPEFULLY this will
prevent some of the bug reports such as "glxgears only gets 59.7fps.
What's wrong with my driver?"
(cherry picked from commit 58b9cd411f)
The wrap_texture() function doesn't set the renderbuffer PutRowRGB() method,
which is used to implement DrawPixels(). This fix adds an implementation
of this method.
(cherry picked from commit 523febe12e)
On Mobile chipsets, we often enable PipeB instead of PipeA, but the test
in here was insufficient, falling back to pipe A if the area
intersection returned zero. Therefore, in the case where a window went
off to the top of the left of the screen, it would freeze, waiting on
the wrong vblank.
Fix this mess by checking the sarea for a crtc being zero sized, and in
that case always default to the other one.
(cherry picked from commit 0b5266ff64)
intelMakeCurrent is called before intelWindowMoved (in fact, it calls
it), so calculation of the correct vblank crtc has not happened yet.
Fix this by making a function that fixes up a set of vblank flags and
call if from both functions.
(cherry picked from commit 39b4061bb9)
When doing line stipple, the stipple count resets on each line segment,
unless the primitive is a GL_LINE_LOOP or a GL_LINE_STRIP.
The existing code correctly identifies the need for a software fallback
to handle conformant line stipple on GL_LINE_LOOP primitives, but
neglects to make the same assessment on GL_LINE_STRIP primitives.
This fixes it so they match.
(cherry picked from commit 73658ff04f)
Back-face stencil operations didn't work correctly because this value was
zero. It needs to be 1 or 2. The only place it's set otherwise is in
glEnable/Disable(GL_STENCIL_TEST_TWO_SIDE_EXT).
(cherry picked from commit 2a968113a9)
Track separate back-face stencil state for OpenGL 2.0 /
GL_ATI_separate_stencil and GL_EXT_stencil_two_side. This allows all
three to be enabled in a driver. One set of state is set via the 2.0
or ATI functions and is used when STENCIL_TEST_TWO_SIDE_EXT is
disabled. The other is set by StencilFunc and StencilOp when the
active stencil face is set to BACK. The GL_EXT_stencil_two_side spec has
more details.
http://opengl.org/registry/specs/EXT/stencil_two_side.txt
(cherry picked from commit dde7cb9628)
