summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)AuthorFilesLines
2009-06-17st/mesa: rework viewport/scissor initialization codegallium-mesa-7.4Brian Paul2-22/+3
This is a hand-pick of some of the new viewport code committed to mesa_7_5_branch. See commit 3f856c6b6b7fa95ef97a8712876de88d7d57932e for reference. This is just the state_tracker part. The core Mesa part came from merging in mesa_7_4_branch.
2009-06-17mesa: rework viewport/scissor initialization codeBrian Paul3-12/+34
The first time a context is bound to a drawable, the viewport and scissor bounds are initialized to the buffer's size. This is actually a bit tricky. A new _mesa_check_init_viewport() function is called in several places to check if the viewport has been initialized. We also use a new ctx->ViewportInitialized flag instead of the overloaded ctx->FirstTimeCurrent flag. Hand-picked from mesa_7_5_branch, commit 3f856c6b6b7fa95ef97a8712876de88d7d57932e
2009-06-17mesa: added null ptr check in Fake_glXCreatePixmap()Brian Paul1-1/+1
Fixes segfault in progs/xdemos/glxgears_pixmap.c (cherry picked from master, commit d18c57aaeac37cde0cb551191ecd3c3a56a0ffba)
2009-06-17GLX: attempt to fix glean makeCurrent test cases.Brian Paul2-2/+6
Two parts to this: One we don't keep pointers to possibly freed memory anymore once we unbind the drawables from the context. Brian I need to figure out what the comment you made there, can we get a glean/piglit test so we can fix it properly? If the new gc is the same as the oldGC, we call the unbind even though we just bound it in that function. doh. (cherry picked from master, commit 77506dac8e81e9548a7e9680ce367175fe5747af)
2009-06-17mesa: fix incorrect viewport clamping in _mesa_set_viewport()Brian Paul1-2/+2
A 0 by 0 viewport size is legal. Don't clamp against lower bound of one. The error checking earlier in the function prevents negative values.
2009-06-17mesa: fix REMAINDER() macroBrian Paul1-1/+1
The results were incorrect for some negative values of A. See bug 21872. (cherry picked from mesa_7_5_branch, commit ed7f4b42307bff4633689d6781cd3643f10041e5)
2009-06-17intel: Release fb backing regions in intelDestroyBuffer()Brian Paul1-1/+25
Fixes memory leak when destroying framebuffers. (cherry picked from mesa_7_5_branch, commit d027e8feff7d38cccadc6aaccc0454b21ce4dca0)
2009-06-17glsl: fix warnings, update comments, s/TRUE/GL_TRUE/Brian Paul1-3/+4
(cherry picked from master, commit 7fdd64ab29576e607434fb8c82ddfa61e8ea6aa8)
2009-06-17glsl: Handle continuation characters in preprocessor.Brian Paul1-2/+63
(cherry picked from master, commit cc22620e4b11425997f3bc1fc70f4c88cec22d2e)
2009-06-17glsl: Fix symbol replacement handling in preprocessor.Brian Paul1-13/+15
(cherry picked from master, commit d9617deb008b75f4a605a30408aeb1948139c33e)
2009-06-17glsl: Expand nested preprocessor macros.Brian Paul1-1/+14
(cherry picked from master, commit ef8caec29ae73bb2bbeb48f0578d839ef29348cd)
2009-06-17glsl: Fix preprocessor define argument parsing for dead sections.Brian Paul1-4/+4
(cherry picked from master, commit 19a54d9f1055c366fd77026dd67007a8d5921f58)
2009-06-17glsl: preprocessor debug code (disabled)Brian Paul1-0/+5
(cherry picked from master, commit 854151ba62e4b1cb6a755f6c131ee2e0f680f39b)
2009-06-17osmesa: Allow building standalone in all three channel widthsDan Nicholson1-27/+9
autoconf had been designating the 8 bit libOSMesa as the default standalone osmesa, but the Makefile expected it to be linked to libGL. Fix up the osmesa Makefile so that it allows any of the combinations of standalone and channel width to be built. Fixes bug #21980. (cherry picked from commit 7441dcd90b01df8351026af8bbb50e11bb86071a)
2009-06-17mesa: fix loop over generic attribs in update_arrays()Brian Paul1-1/+1
(cherry picked from commit 1045481dd96dec6e37f4b623b1dbae8af381de75)
2009-06-17vbo: fix crash in vbo_exec_bind_arrays()Brian Paul1-0/+1
When a vertex shader uses generic vertex attribute 0, but not gl_Vertex, we need to set attribute[16] to point to attribute[0]. We were setting the attribute size, but not the pointer. Fixes crash in glsl/multitex.c when using the VertCoord attribute instead of gl_Vertex. (cherry picked from commit c3538969e1ae3e626a618934aa8f35a7a22ddb39)
2009-06-17mesa: allow GL_BITMAP type in _mesa_image_image_stride()Brian Paul1-13/+23
It's possible to hand a GL_COLOR_INDEX/GL_BITMAP image to glTexImage3D() which gets converted to RGBA via the glPixelMap tables. This fixes a failure with piglit/fdo10370 with Gallium. (cherry picked from commit 995456f9305593005f8466520314ee087f3d422a)
2009-06-17mesa: bump version to 7.4.3Brian Paul1-3/+3
2009-06-17vbo: fix incorrect loop limit in bind_array_obj()Brian Paul1-1/+4
The generic_array[] is 16 elements in size, but the loop was doing 32 iterations. The out of bounds array write was clobbering the following inputs[] array but as luck would have it, that didn't matter. (cherry picked from commit 8da09e6924ca22ba7951d5a7673dfab2a711a997)
2009-06-17mesa: add missing glGet*() case for GL_VERTEX_ARRAY_BINDING_APPLEBrian Paul2-1/+16
(cherry picked from commit 70c4b81e88e18e354e8dfaf47e5455e463b207d8)
2009-06-17mesa: remove texUnit->_Current = NULL assignmentBrian Paul1-1/+0
A follow-on to commit 46a2aad2cfec0ae54bc1a3ba5aacf6f59065d64b which fixes a texture object memory leak. See bug 21756.
2009-06-17mesa: allow depth/stencil textures to be attached to GL_STENCIL_ATTACHMENTBrian Paul1-3/+13
See sourceforge bug #2793846. (cherry picked from commit 042d9a513213b1fa356c0d80abc62b9327e0bcc2)
2009-06-17mesa: assign trb->Base.StencilBits in update_wrapper().Brian Paul1-0/+1
When we render to a depth/stencil texture there are stencil bits. (cherry picked from commit c99a60c40d4ece363d37a5af895124f08a645c6b)
2009-06-17mesa: fix texture object reference counting in texture_override()Brian Paul1-1/+1
Fixes bug 21756: r200: Mesa-7.4.2 release regression: "Quit on exit".
2009-06-17Initialize psp->waitX/waitGL for swrast_dri.so.Aidan Thornton1-0/+2
Fixes http://bugs.freedesktop.org/show_bug.cgi?id=21053 . (cherry picked from commit f2445dfd85b42aafe0634e17b2929b4122ff3f03)
2009-06-17mesa: bump version to 7.4.2Brian Paul1-3/+4
2009-06-17i965: fix render to FBO/texture orientation bugsBrian Paul1-6/+7
Anytime we're not rendering to the default/window FBO, need to invert rendering, not just when rendering to a texture. Otherwise, if a FBO consists of a mix of textures and renderbuffers the up/down orientation was inconsistant. Fixes shadowtex.c bad rendering. (cherry picked from commit ba367f68ccacf255f78ac0c8dd066e64bbb1e5c2)
2009-06-17mesa: Fixed a texture memory leakBrian Paul1-0/+3
The current texture for any particular texture unit is given an additional reference in update_texture_state(); but if the context is closed before that texture can be released (which is quite frequent in normal use, unless a program unbinds and deletes the texture and renders without it to force a call to update_texture_state(), the memory is lost. This affects general Mesa; but the i965 is particularly affected because it allocates a considerable amount of additional memory for each allocated texture. (cherry picked from master, commit c230767d6956b63a2b101acb48f98823bb5dd31a)
2009-06-17Flag GLSL programs as depending on constant state when they use uniforms.Eric Anholt1-0/+5
This is not a cherry-pick, as it got fixed in master with the _NEW_PROGRAM_CONSTANTS change. Fixes bad rendering with various GLSL programs idr and I have been writing.
2009-06-17glx: replace Xmalloc() calls with Xcalloc()Brian Paul2-2/+2
Fixes a bug where psp->WaitX was uninitialized. Reported by Chris Clayton. (cherry picked from commit dd4c142e90a0cba5b445990bb522ce9199d7f565)
2009-06-17mesa: fix state validation bug for glCopyTex[Sub]Image()Brian Paul1-5/+16
We need to make sure the framebuffer state is up to date to make sure we read pixels from the right buffer when doing a texture image copy. (cherry-picked from master, commit 63f01309801c5a900d8d7f5ccd63413e33ff9bff)
2009-06-17r300: Increase reference count of texture objects referenced by current state.Michel Dänzer4-9/+11
Fixes a use-after-free reported in http://bugs.freedesktop.org/show_bug.cgi?id=20539, so this possibly fixes that bug. It has been confirmed to fix http://bugs.freedesktop.org/show_bug.cgi?id=17895 . (cherry picked from commit c28707b50701b1cf8727be29d61e2d939c6ee58f)
2009-06-17mesa: Fix buffer overflow when parsing generic vertex attributes.Brian Paul1-3/+9
(cherry picked from master, commit fa92756400ccfbb3f0201df634feb45ab4f98352)
2009-06-17R300: add quadpipe overridesAlex Deucher1-4/+13
RV410 SE chips only have 1 quadpipe. Also, handle other R300 chip with quadpipe override
2009-06-17swrast: add missing break in clamp_rect_coord_linear()Brian Paul1-0/+1
See bug 21461. (cherry picked from master, commit dcf571aff9de1a4298c4d2c4148d84cdc4daf02e)
2009-06-15add patches from bug 21997Alan Hourihane6-29/+346
2009-06-01st/mesa: fix incorrect sprite origin when drawing to FBO/textureBrian Paul1-1/+2
Need to take the draw buffer's up/down orientation into consideration when setting the sprite_coord_mode field. Fixes inverted sprites when drawing into an FBO. (cherry picked from commit 0e8a5a84742adf6e99236f246c77325fad174204)
2009-06-01draw: reset extra_vp_outputs.slot to zero in widepoint_flush()Brian Paul1-0/+1
Fixes a crash when clearing the window with a quad after drawing large points. We were asking the draw module how many vertex shader outputs there were and got 3 instead of 2. This led to creating vertices with too many attributes and trying to read invalid memory. We reset extra_vp_outputs.slot to zero in the aaline/aapoint stage's flush functions already. This omission was just an oversight in the wide_point stage. (cherry picked from commit 427554211b5b8ab2c6afcd5bd574ac97fb3457da)
2009-05-25xlib: Check GALLIUM_TRACE env var instead of XMESA_TRACE.José Fonseca2-2/+5
This avoids a xlib winsys segfault when XMESA_TRACE was set, but GALLIUM_TRACE was not.
2009-05-25xlib: Don't override DEFINES.José Fonseca1-1/+1
Specially, overriding -DUSE_XSHM will cause the xlib winsys to segfault.
2009-05-25trace: Defer the cast to after the check for enabled trace.José Fonseca1-7/+10
Prevents segfault when trace is disabled.
2009-05-24trace: Make calls dumping threadsafe.José Fonseca1-0/+7
2009-05-24trace: Make call no global.José Fonseca1-3/+4
2009-05-24python/retrace: Fix flush dirty logic.José Fonseca1-1/+1
2009-05-24python/retrace: Try to cope with failures creating textures.José Fonseca1-0/+6
2009-05-24python/retrace: Use colors on windows console.José Fonseca1-0/+70
2009-05-24python/retrace: Allow to specify the range of calls of interest.José Fonseca1-0/+7
2009-05-24xlib/trace: Fixup xlib traceJakob Bornecrantz1-5/+11
2009-05-24trace: Simplify cast wrappers.José Fonseca5-50/+27
2009-05-24xlib: Fix scons conditional compilation logic.José Fonseca2-2/+2