path: root/docs/relnotes/8.0.3.rst

Mesa 8.0.3 Release Notes / May 18, 2012
=======================================

Mesa 8.0.3 is a bug fix release which fixes bugs found since the 8.0.2
release.

Mesa 8.0.3 implements the OpenGL 3.0 API, but the version reported by
glGetString(GL_VERSION) depends on the particular driver being used.
Some drivers don't support all the features required in OpenGL 3.0.

See the :doc:`Compiling/Installing page <../install>` for
prerequisites for DRI hardware acceleration.

MD5 checksums
-------------

::

   b7210a6d6e4584bd852ab29294ce717e  MesaLib-8.0.3.tar.gz
   cc5ee15e306b8c15da6a478923797171  MesaLib-8.0.3.tar.bz2
   32773634a0f7e70a680505a81426eccf  MesaLib-8.0.3.zip

New features
------------

None.

Bug fixes
---------

This list is likely incomplete.

-  `Bug 28138 <https://bugs.freedesktop.org/show_bug.cgi?id=28138>`__ -
   [G45] Regnum Online, sparkling in in-game rendering
-  `Bug 30102 <https://bugs.freedesktop.org/show_bug.cgi?id=30102>`__ -
   glean depthStencil test fails BadLength with indirect non-swrast
   rendering
-  `Bug 40361 <https://bugs.freedesktop.org/show_bug.cgi?id=40361>`__ -
   Glitches on X3100 after upgrade to 7.11
-  `Bug 41152 <https://bugs.freedesktop.org/show_bug.cgi?id=41152>`__ -
   [glsl] Shader backend in Regnum Online does not work
-  `Bug 41216 <https://bugs.freedesktop.org/show_bug.cgi?id=41216>`__ -
   [bisected pineview]oglc filtercubemin(basic.sizedRGBA) fails
-  `Bug 41372 <https://bugs.freedesktop.org/show_bug.cgi?id=41372>`__ -
   i830_state.c PBO crash
-  `Bug 41495 <https://bugs.freedesktop.org/show_bug.cgi?id=41495>`__ -
   i830: intel_get_vb_max / intel_batchbuffer_space mismatch.
-  `Bug 44701 <https://bugs.freedesktop.org/show_bug.cgi?id=44701>`__ -
   Regnum online textures flickering
-  `Bug 44961 <https://bugs.freedesktop.org/show_bug.cgi?id=44961>`__ -
   [bisected i965] oglc sRGB(Mipmap.1D_textures) regressed
-  `Bug 44970 <https://bugs.freedesktop.org/show_bug.cgi?id=44970>`__ -
   [i965]oglc max_values(negative.textureSize.textureCube) segfaults
-  `Bug 45214 <https://bugs.freedesktop.org/show_bug.cgi?id=45214>`__ -
   Textures disappearing or missing in RegnumOnline OpenGL game
-  `Bug 45558 <https://bugs.freedesktop.org/show_bug.cgi?id=45558>`__ -
   cannot render on a drawable of size equal the max framebuffer size
-  `Bug 45921 <https://bugs.freedesktop.org/show_bug.cgi?id=45921>`__ -
   [r300g, bisected] Multiple piglit regressions after glsl_to_tgsi
   changes
-  `Bug 46303 <https://bugs.freedesktop.org/show_bug.cgi?id=46303>`__ -
   [SNB] segfault in intel_miptree_release()
-  `Bug 46739 <https://bugs.freedesktop.org/show_bug.cgi?id=46739>`__ -
   [snb-m-gt2+] compiz crashed with SIGSEGV in intel_miptree_release()
-  `Bug 46834 <https://bugs.freedesktop.org/show_bug.cgi?id=46834>`__ -
   small performance when playing flightgear (swrast fallback through
   GTT mapping)
-  `Bug 47126 <https://bugs.freedesktop.org/show_bug.cgi?id=47126>`__ -
   tests/fbo/fbo-array.c:109: create_array_fbo: Assertion \`glGetError()
   == 0' failed.
-  `Bug 48218 <https://bugs.freedesktop.org/show_bug.cgi?id=48218>`__ -
   brw_fs_schedule_instructions.cpp segfault due to accessing not
   allocated last_mrf_write[16]
-  `Bug 48545 <https://bugs.freedesktop.org/show_bug.cgi?id=48545>`__ -
   LLVMpipe glReadPixels Firefox hits the slow path (WebGL rendering)
-  `Bug 49124 <https://bugs.freedesktop.org/show_bug.cgi?id=49124>`__ -
   swrast/s_texfetch.c:1156: set_fetch_functions: Assertion
   \`texImage->FetchTexel' failed.

Changes
-------

The full set of changes can be viewed by using the following GIT
command:

::

     git log mesa-8.0.2..mesa-8.0.3

Alban Browaeys (1):

-  dri/i915: Fix off-by-one in i830 clip region size.

Alex Deucher (2):

-  r200: fix fog coordinate emit
-  radeon: fix fog coordinate emit

Alexander von Gluck (4):

-  llvmpipe: fix symbol conflict on Haiku
-  svga: fix typedef conflicts on Haiku
-  mesa: Don't use newlocale on Haiku
-  glsl: Don't use newlocale on Haiku

Anuj Phogat (4):

-  mesa: fix issues with texture border and array textures
-  mesa: Fix valid texture target test in
   \_mesa_GetTexLevelParameteriv()
-  mesa: Fix the cause of piglit test fbo-array failure
-  intel: Fix a case when mapping large texture fails

Brian Paul (17):

-  mesa: add a couple fast-paths to fast_read_rgba_pixels_memcpy()
-  mesa/gdi: remove wmesa_set_renderbuffer_funcs() function
-  mesa/gdi: remove clear_color() function
-  mesa: bump version to 8.0.2 in configs/default
-  swrast: include s_fragprog.h to silence warnings
-  mesa: remove LSB-first pixel packing check in glReadPixels
-  mesa: fix error in \_mesa_format_matches_format_and_type() for RGB888
-  mesa: add BGR888 code in \_mesa_format_matches_format_and_type()
-  vbo: fix node_attrsz[] usage in vbo_bind_vertex_list()
-  mesa: add missing texture integer test in glTexSubImage()
-  mesa: add missing return after \_mesa_error() in update_array()
-  glsl: propagate MaxUnrollIterations to the optimizer's loop unroller
-  st/mesa: set MaxUnrollIterations = 255
-  st/mesa: no-op glCopyPixels if source region is out of bounds
-  mesa: do more teximage error checking for generic compressed formats
-  mesa: fix/add error check in \_mesa_ColorMaterial()
-  mesa: fix glMaterial / dlist bug

Chad Versace (3):

-  glsl: Fix Android build
-  main: Fix memory leak in \_mesa_make_extension_string()
-  intel: Disable ARB_framebuffer_object in ES contexts

Chris Wilson (1):

-  i830: Compute initial number of vertices from remaining batch space

Dave Airlie (4):

-  mesa/format_unpack: add LUMINANCE 8/16 UINT/INT
-  glx/drisw: avoid segfaults when we fail to get visual
-  drisw: fix image stride calculation for 16-bit.
-  intel: fix TFP at 16-bpp

Dylan Noblesmith (7):

-  intel: fix null dereference processing HiZ buffer
-  util: fix undefined behavior
-  util: fix uninitialized table
-  egl: fix uninitialized values
-  st/vega: fix uninitialized values
-  egl-static: fix printf warning
-  i965: fix typo

Eric Anholt (19):

-  i965/fs: Jump from discard statements to the end of the program when
   done.
-  intel: Fix rendering from textures after RenderTexture().
-  mesa: Fix handling of glCopyBufferSubData() for src == dst.
-  i965/fs: Move GL_CLAMP handling to coordinate setup.
-  i965/fs: Implement GL_CLAMP behavior on texture rectangles on gen6+.
-  mesa: Fix push/pop of multisample coverage invert.
-  mesa: Include the multisample enables under GL_MULTISAMPLE_BIT attrib
   as well.
-  mesa: Fix display list handling for GL_ARB_draw_instanced.
-  mesa: Fix display lists for draw_elements_base_vertex with
   draw_instanced.
-  mesa: Add missing error check for first < 0 in
   glDrawArraysInstanced().
-  i915: Fix piglit fbo-nodepth-test on i830.
-  intel: Return success when asked to allocate a 0-width/height
   renderbuffer.
-  mesa: Throw error on glGetActiveUniform inside Begin/End.
-  i965/vs: Fix up swizzle for dereference_array of matrices.
-  glsl: Fix indentation of switch code.
-  glsl: Let the constructor figure out the types of switch-related
   expressions.
-  glsl: Reject non-scalar switch expressions.
-  glsl: Fix assertion failure on handling switch on uint expressions.
-  mesa: Check for framebuffer completeness before looking at the rb.

Eugeni Dodonov (1):

-  intel: add PCI IDs for Ivy Bridge GT2 server variant

Han Shen(沈涵) (1):

-  bin/mklib: remove '-m32' for arm linux

Ian Romanick (1):

-  mesa: Bump version number to 8.0.3

Jakob Bornecrantz (1):

-  docs: Add 8.0.2 md5sums

Jeremy Huddleston (7):

-  darwin: Eliminate a pthread mutex leak
-  darwin: Fix an error message
-  darwin: Make reported errors more user-friendly
-  darwin: Use ASL for logging
-  darwin: Unlock our mutex before destroying it
-  darwin: Eliminate a possible race condition while destroying a
   surface
-  darwin: Address a build failure on Leopard and earlier OS versions

Jon TURNEY (1):

-  Have \__glImageSize handle format GL_DEPTH_STENCIL_NV the same way as
   the server does

Jonas Maebe (2):

-  glapi: Correct size of allocated \_glapi_table struct
-  apple: Fix a use after free

Jordan Justen (1):

-  mesa: Add primitive restart support to glArrayElement

Kenneth Graunke (12):

-  i965: Actually upload sampler state pointers for the VS unit on Gen6.
-  i965/fs: Fix FB writes that tried to use the non-existent m16
   register.
-  vbo: Remove pedantic warning about 'end' beind out of bounds.
-  vbo: Ignore invalid element ranges which are outside VBO bounds.
-  vbo: Rework checking of 'end' against \_MaxElement.
-  vbo: Eliminate short-circuiting in invalid-start case.
-  i965: Fix GPU hangs in the dummy fragment shader.
-  i965: Make the dummy fragment shader work in SIMD16 mode.
-  drirc: Add force_glsl_extensions_warn workaround for Unigine Heaven.
-  i965: Avoid explicit accumulator operands in SIMD16 mode on Gen7.
-  intel: Remove pointless software fallback for glBitmap on Gen6.
-  glsl: Fix broken constant expression handling for <, <=, >, and >=.

Kurt Roeckx (2):

-  i915: Compute maximum number of verts using the actual batchbuffer
   size.
-  i915: Fix i830 polygon stipple from PBOs.

Marek Olšák (5):

-  r300g/swtcl: don't print an error when getting ClipVertex
-  r300g/swtcl: don't enter u_vbuf_mgr
-  r300g/swtcl: don't expose shader subroutine support
-  r300g/swtcl: fix polygon offset
-  r300g/swtcl: fix crash when back color is present in vertex shader

Mathias Fröhlich (1):

-  glsl: Avoid excessive loop unrolling.

Matt Turner (1):

-  Remove -ffast-math from default CFLAGS

Paul Berry (1):

-  i915: Initialize swrast_texture_image structure fields.

Roland Scheidegger (1):

-  mesa: check_index_bounds off-by-one fix

Tom Stellard (2):

-  r300/compiler: Clear loop registers in vertex shaders w/o loops
-  r300/compiler: Copy all instruction attributes during local transfoms

Vinson Lee (5):

-  mesa: Fix memory leak in \_mesa_get_uniform_location.
-  linker: Fix memory leak in count_uniform_size::visit_field.
-  swrast: Fix memory leaks in blit_linear.
-  ir_to_mesa: Fix uninitialized member in add_uniform_to_shader.
-  mesa: Fix memory leak in generate_mipmap_compressed.

Yuanhan Liu (2):

-  i915: set SPRITE_POINT_ENABLE bit correctly
-  i965: fix wrong cube/3D texture layout