mesa: Changes to 'refs/tags/9.2-branchpoint'
Tag '9.2-branchpoint' created by Eric Anholt <eric@anholt.net> at 2014-03-28 07:00 +0000
Mesa 9.2 branchpoint
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=qLy/
-----END PGP SIGNATURE-----
Changes since 9.1-branchpoint:
Abdiel Janulgue (9):
dri2: Create image from texture
intel: expose dimensions and offsets of a miptree level in DRIImage
intel: Expose intel_miptree_create_internal as intel_miptree_create_layout.
intel: add pixel offset calculator for miptree levels
i965: Account for offsets when updating SURFACE_STATE.
intel: Create a miptree using offsets in intel_set_texture_image_region
intel: Account for mt->offset in intel_miptree_map
intel: implement create image from texture
intel: make sure to setup image dimension in image_from_planar setup
Adam Jackson (7):
glx: Build with VISIBILITY_CFLAGS in automake
gallivm: Minor comment cleanup
llvmpipe: Work without sse2 if llvm is new enough
linux: Don't emit a .note.ABI-tag section anymore (#26663)
gallivm: Fix altivec intrinsics for 8xi16 add/sub
llvmpipe: Remove x/y from cmd_bin
gallium: Fix llvmpipe on big-endian machines
Adam Sampson (1):
autotools: oprofilejit should be included in the list of LLVM components required
Adhemerval Zanella (2):
llvmpipe: remove lp_swizzled_cbuf
llvmpipe: fix vertex_header mask store in big-endian
Adrian Marius Negreanu (1):
android: fix Android.mk bug in mesa/drivers/dri/common
Alan Coopersmith (2):
integer overflow in XF86DRIOpenConnection() [CVE-2013-1993 1/2]
integer overflow in XF86DRIGetClientDriverName() [CVE-2013-1993 2/2]
Alan Hourihane (3):
Add missing GL_TEXTURE_CUBE_MAP entry in _mesa_legal_texture_dimensions
Unreference sampler object when it's currently bound to texture unit.
mesa: fix glGetInteger*(GL_SAMPLER_BINDING).
Alex Deucher (42):
radeonsi: add support for Oland chips
radeonsi: default PA_SC_RASTER_CONFIG to 0
radeonsi: add Oland pci ids
r600g: don't enable ReZ mode on evergreen
r600g: add PS_PARTIAL_FLUSH flag
r600g: r6xx deadlock workaround (v6)
r600g: fixup PS_PARTIAL_FLUSH flag handling for cayman
r600g: synchronize streamout buffers on r6xx too (v3)
r600g: add missing emit_flush for R600_CONTEXT_FLUSH_AND_INV case
r600g: enable CP DMA on 6xx
r600g: add Richland APU pci ids
r600g: Use blitter rather than DMA for 128bpp on cayman (v3)
r600g: properly set non_disp tiling mode for DMA (v2)
r600g: emit DB_SRESULTS_COMPARE_STATE0 on r6xx/r7xx
r600g: don't emit SQ_DYN_GPR_RESOURCE_LIMIT_1 on cayman
r600g: use CP DMA for buffer clears on evergreen+
r600g: add new richland pci ids
radeonsi: add new SI pci ids
r600g: don't emit surface_sync after FLUSH_AND_INV_EVENT
radeonsi: add support for hainan chips
radeonsi: update r600_get_llvm_processor_name for hainan
radeonsi: add Hainan pci ids
winsys/radeon: add env var to disable VM on Cayman/Trinity
radeonsi: rename SI chip class from TAHITI to SI
radeonsi: initial support for CIK chips
radeonsi: initial PM4 changes for CIK
radeonsi: register updates for CIK
radeonsi: fix VGT_PRIMITIVE_TYPE emit for CIK
radeonsi: emit TA_BC_BASE_ADDR_HI for border color on CIK
radeonsi: emit additional shader pgm rsrc registers for CIK
radeonsi: properly handle DB tiling setup on CIK
radeonsi: store chip class in the pm4 struct
radeonsi: update surface sync packet emit for CIK
radeonsi: PA_CL_ENHANCE is privileged on CIK
radeonsi: emit PA_SC_RASTER_CONFIG[_1] on cik
radeonsi: add llvm processor names for CIK
radeonsi: disable 2D tiling on CIK for now
radeonsi: add bonaire pci ids
radeonsi: add kabini pci ids
r600g: adjust flush flags (v3)
radeon: bump libdrm_radeon requirement for CIK support
r600g: don't use the CB/DB CP COHER logic on r6xx
Alexander Monakov (1):
Honor GLX_DONT_CARE in MATCH_MASK
Alexander von Gluck IV (4):
Mesa: Use mmap on Haiku for executable memory vs malloc
Mapi: Use mmap on Haiku for executable memory vs malloc
Gallium: Use mmap on Haiku for executable memory vs malloc
Haiku: Add Gallium winsys and target code
Andreas Boll (34):
mesa: bump version to 9.2 (devel)
gallium: handle unhandled PIPE_CAP_TEXTURE_BUFFER_OFFSET_ALIGNMENT
docs: replace CVS with git
docs: update sourcetree
docs: mention some other supported APIs
docs: use proper title for index.html
docs: update making a new mesa release info
docs: document removal of makedepend build dependency
docs: import release notes for 9.0.3, add news item
targets/xa-vmwgfx: Force c++ linker to fix undefined symbols
llvmpipe/build: add DLOPEN_LIBS and PTHREAD_LIBS to the lp_test_* targets
docs: import release notes for 9.1, add news item
docs: add news item for mesa-demos 8.1.0 release
radeon/llvm: Link against libgallium.la to fix an undefined symbol
mesa: use ieee fp on s390 and m68k
build: Enable x86 assembler on Hurd.
osmesa: fix out-of-tree build
gallium/egl: fix out-of-tree build
docs: update release notes for 9.2
configure.ac: Allow OpenGL ES1 and ES2 only with enabled OpenGL
docs: fix 9.1.2 release notes
intel: remove executable bit from C file
egl-static: use automake conditionals for defining FEATURE_{GL,ES1,ES2}
intel: use automake conditionals for defining FEATURE_{ES1,ES2}
build: remove unused API_DEFINES
scons: remove IN_DRI_DRIVER
mesa: Add a script to generate the list of fixed bugs
docs: add info about bugzilla_mesa.sh script
mesa: add usage examples to get-pick-list and shortlog scripts
scons: remove radeon build
scons: remove nouveau build
glapi: Add some missing static_dispatch="false" annotations to es_EXT.xml
build: Unify PACKAGE_VERSION on autotools, scons and Android
build: Use PACKAGE_VERSION from autoconf
Andreas Hartmetz (1):
radeonsi: Add ipo to LLVM_COMPONENTS
Andreas Oberritter (1):
ax_prog_flex.m4: change grep syntax to accept e.g. flex.real
Anthony G. Basile (1):
gallium: fix build on uclibc system
Anuj Phogat (11):
meta: Allocate texture before initializing texture coordinates
mesa: Fix FB blitting in case of zero size src or dst rect
intel: Add a null pointer check before dereferencing the pointer
i965: Fix build failure
mesa: Implement ext_framebuffer_multisample_blit_scaled extension
intel: Change the register type from UW to UD in blorp engine
intel: Add multisample scaled blitting in blorp engine
mesa: Return ZeroVec/dummyReg instead of NULL pointer
i965/blorp: Add bilinear filtering of samples for multisample scaled blits
i965: Enable ext_framebuffer_multisample_blit_scaled on intel h/w
i965/blorp: Fix clear rectangle alignment in fast color clear
Aras Pranckevicius (1):
GLSL: fix lower_jumps to report progress properly
Armin K (2):
gallivm: Fix build with LLVM 3.3
radeon: Fix build with LLVM 3.3
Arnas Milasevicius (1):
gallium: Remove draw_arrays() and draw_arrays_instanced() functions
Ben Skeggs (1):
nvc0: enable very initial support for nvf0 (GK110)
Ben Widawsky (1):
i965: Actually use the user timeout in glClientWaitSync.
Bill York (1):
docs: updated instructions for Mesa on Windows
Brian Paul (270):
xlib: fix incorrect GL_ANGLE_texture_compression_dxt enable
xlib: stop use _mesa_enable_extension(), just set the boolean flags
st/mesa: set ctx->Const.MaxSamples = 0, not 1
xlib: use _mesa_generate_mipmap() for mipmap generation, not meta
osmesa: use _mesa_generate_mipmap() for mipmap generation, not meta
mesa: don't enable GL_EXT_framebuffer_multisample for software drivers
st/mesa: only enable GL_EXT_framebuffer_multisample if GL_MAX_SAMPLES >= 2
docs: more VMware guest driver info, tips
mesa: fix comment typo: s/formaat/format/
mesa: remove ctx->Driver.Error() hook
trace: allow GALLIUM_TRACE=stdout/stderr
trace: add void to function definition
trace: measure time for each gallium call
svga: add, use SVGA3D_SURFACE_HINT_VOLUME flag
svga: check for NaN shader immediates
st/mesa: whitespace/indentation fix
mesa: don't use format chooser code for glCompressedTexImage
st/mesa: don't choose DXT formats if we can't do DXT compression
st/mesa: improve the format choosing code for DrawPixels
st/mesa: merge st_ChooseTextureFormat_renderable() into st_ChooseTextureFormat()
st/mesa: simplify the update_single_texture() function
gallium: add SQRT shader opcode
tgsi: add support for new SQRT opcode
gallivm: implement support for SQRT opcode
gallium/drivers: handle PIPE_SHADER_CAP_TGSI_SQRT_SUPPORTED query
st/mesa: emit SQRT opcode when driver supports it
svga: fix sRGB rendering
llvmpipe: remove extraneous const qualifier
util: fix incorrect Z bit masking in util_clear_depth_stencil()
st/mesa: fix format query for GL_ARB_texture_rg
trace: move struct tr_list to tr_texture.h
trace: whitespace, comment clean-ups
mesa: remove old version comment from gl.h
trace: fix incorrect trace_surface::base.context pointer
trace: add context pointer sanity checking
draw: fix non-perspective interpolation in interp()
st/mesa: implement glBitmap unpacking from a PBO, for the cache path
st/xlib: initialize the drawable size in create_xmesa_buffer()
llvmpipe: fix handling of 0 x 0 framebuffer size
st/mesa: fix trimming of GL_QUAD_STRIP
st/mesa: check for dummy programs in destroy_program_variants()
st/mesa: fix polygon offset state translation logic
draw: fix broken polygon offset stage
llvmpipe: add missing checks for polygon offset point/line modes
llvmpipe: rename polygon offset fields to something more specific
st/mesa: fix debug_printf() format string warning
svga: remove some extraneous whitespace
svga: fix comment typos
winsys/null: fix var typo templet->templat
st/mesa: remove some conditionals in update_raster_state()
st/mesa: convert ir_triop_lrp to TGSI_OPCODE_LRP
svga: always link with C++
mesa: add switch case for ir_txf_ms to silence warning
st/mesa: add switch case for ir_txf_ms to silence warning
mesa: flush current state when querying GL_EDGE_FLAG
llvmpipe: tweak CMD_BLOCK_MAX and LP_SCENE_MAX_SIZE
llvmpipe: add some scene limit sanity check assertions
llvmpipe: remove unused cmd_block_list struct
llvmpipe: fix incorrect 'j' array index in dummy texture code
vbo: fix crash found with shared display lists
llvmpipe: remove the power of two sizeof(struct cmd_block) assertion
draw: add const qualifier to silence compiler warning
docs: document another Viewperf bug
st/mesa: add PIPE_FORMAT_R16G16B16A16_UNORM renderbuffer support
st/osmesa: new OSMesa gallium state tracker
st/osmesa: add new Makefile.am
targets/osmesa: new OSMesa gallium target
target/osmesa: add new Makefile.am
configure: wire-up new OSMesa gallium state tracker and target
docs: rewrite the OSMesa info / instructions
glx: init screen = 0 to fix optimized build warning
draw: init pre_clip_pos = NULL to fix optimized build warning
st/osmesa: silence some optimized build warnings
softpipe: fix up NUM_ENTRIES confusion
driconf: reorder options, reformat comments, etc
driconf: add a miscellaneous section and always_have_depth_buffer option
st/dri: add support for the always_have_depth_buffer option
tgsi: add initializer data to fix MSVC compile error
tgsi: fix uninitialized declaration array fields
winsys/svga: improve error/debug message output
meta: minor indentation fix
vega: fix MSVC warning about missing return statement
meta: fix incorrect slice, r coordinate computation
postprocess: silence some MSVC float/int warnings
rbug: silence some MSVC signed/unsigned warnings
softpipe: silence some MSVC float/double warnings
softpipe: silence some MSVC signed/unsigned warnings
softpipe: silence some asst. MSVC type warnings in sp_tex_sample.c
scons: check for existance of 'MSVC_VERSION' in env
swrast: init vars to silence warnings
gallivm: init vars to silence warnings
gallium: undef PACKAGE_* macros to silence warnings
draw: fix some build breakage when LLVM is not used
llvmpipe: add 'f' suffix to 1.0 in fixed_to_float()
llvmpipe: put .bmp extension on dumped image files
st/mesa: add ir_lod case in GLSL->TGSI code to silence warning
drirc: set always_have_depth_buffer for Topogon
docs: add a new page documenting known application issues
mesa: remove platform checks around __builtin_ffs, __builtin_ffsll
llvmpipe: use triangle subdivision to avoid fixed-point overflow issues
draw: use pipe_transfer_unmap() to match pipe_transfer_map()
hud: flush/unmap the vertex buffer before drawing
st/mesa: rewrite comment in st_manager.c
svga: add case for PIPE_CAP_QUERY_PIPELINE_STATISTICS
gallium/hud: try L8 texture for font if I8 format isn't supported
svga: refactor occlusion query code
svga: add HUD queries for number of draw calls, number of fallbacks
util: move functions from u_resource.c to u_transfer.c
util: add new util_resource_size() function in u_resource.[ch]
svga: add new memory-used HUD query
gallium/hud: replace malloc w/ MALLOC
util: add debug_memory_check_block(), debug_memory_tag()
st/wgl: make stw_current_context() non-static
st/wgl: add HUD support
gallium/hud: initialize sampler state
gallium/hud: add GALLIUM_HUD_PERIOD env var
st/xlib: add HUD support for xlib/GLX
gallium/u_blitter: fix is_blit_generic_supported() stencil checking
softpipe: misc updates to image dumping in softpipe_flush()
st/osmesa: re-use buffers in OSMesaMakeCurrent()
docs: initial 9.2 release notes file
st/mesa: remove #include mfeatures.h
mesa: remove #ifdef FEATURE_ES2, add some comments instead
mesa: remove old comment about FEATURE_GL
st/mesa: remove #if FEATURE_GL/ES tests
mesa: remove unused DD_LINE_SMOOTH flag
mesa: remove DD_SEPARATE_SPECULAR flag
mesa: remove DD_LINE_STIPPLE flag
mesa: remove DD_POINT_SMOOTH flag
mesa: remove DD_POINT_ATTEN flag
mesa: remove DD_TRI_OFFSET flag
mesa: remove DD_TRI_STIPPLE flag
mesa: remove DD_TRI_SMOOTH flag
mesa: remove DD_TRI_UNFILLED flag
mesa: remove DD_TRI_LIGHT_TWOSIDE flag
mesa: remove gl_context::_TriangleCaps
st/mesa: clean up formatting in st_cb_msaa.c
st/wgl: fix issue with SwapBuffers of minimized windows
glapi: remove FEATURE_remap_table test (it's always defined)
mesa: remove FEATURE_remap_table from remap.[ch]
glapi: no longer emit #include "mfeatures.h" in generated files
mesa: remove #include "mfeatures.h" from numerous source files
mesa: remove mfeatures.h
mesa: generate glGetInteger/Boolean/Float/Doublev() code for all APIs
mesa: enable GL_ARB_texture_float if TEXTURE_FLOAT_ENABLED is defined
svga: whitespace, comment fixes in svga_pipe_fs/vs.c
svga: whitespace, comment fixes in svga_pipe_query.c
st/mesa: minor indentation fixes
mesa: Remove extra MapBufferRange in create_beginend_table()
mesa: use new _mesa_inside_begin_end() function
mesa: use new _mesa_inside_dlist_begin_end() function
docs: document issue with Viewperf proe-05 test 6
mesa: make _mesa_save_vtxfmt_init() static
mesa: updated read_buffer_enum_to_index() comment
llvmpipe: bump LP_MAX_THREADS to 16
llvmpipe: replace LP_MAX_THREADS with screen->num_threads in query code
mesa: fix the compressed TexSubImage size checking code
swrast: add casts for ImageSlices pointer arithmetic
docs: s/Aprile/April/
configure: remove FEATURE_GL/ES1/ES2
configure: remove IN_DRI_DRIVER
mesa: fix save_ShadeModel() logic and add new comments
mesa: adjust PRIM_x constants for geometry shaders
mesa: fix CurrentSavePrimitive <= GL_POLYGON tests
mesa: refactor _mesa_valid_prim_mode()
mesa: simplify save_Begin() error checking
mesa: remove unused PRIM_INSIDE_UNKNOWN_PRIM constant
mesa: add missing error check in _mesa_EndList()
vbo: fix parameter validation for saving dlist glDraw* functions
mesa: don't install glDraw* functions into the BeginEnd dispatch table
vbo: remove redundant vfmt->Begin = _save_Begin assignment
mesa: comments, formatting fixes in dlist code
vbo: fix error detection in vbo_save_playback_vertex_list()
vbo: fix initial value of ctx->Driver.CurrentSavePrimitive
mesa: add names of geometry shader prims in gl_enums.py
mesa: change _mesa_inside_dlist_begin_end() to handle PRIM_UNKNOWN
mesa: remove _MESA_INIT_DLIST_VTXFMT() macro
mesa: remove _MESA_INIT_ARRAYELT_VTXFMT() macro
mesa: remove _MESA_INIT_EVAL_VTXFMT() macro
vbo: add new vbo_initialize_exec_dispatch(), vbo_initialize_save_dispatch()
mesa: simplify dispatch for glDraw* functions
mesa: remove GLvertexformat::Rectf()
mesa: remove GLvertexformat::EvalMesh1(), EvalMesh2()
mesa: update GLvertexformat comments
mesa: update comments, simplify code in vtxfmt.c
vbo: create a few utility functions for merging primitives
vbo: implement primitive merging for glBegin/End sequences
vbo: update comments for vbo_save_NotifyBegin()
mesa; change ctx->Driver.SaveNeedFlush to boolean, and document it.
mesa: change ctx->Driver.NeedFlush to GLbitfield and update comment
vbo: use new no-op ArrayElement in _mesa_noop_vtxfmt_init()
mesa: whitespace, formatting fixes, etc in api_arrayelt.c
vbo: add comments, const qualifiers
st/mesa: generate GL_OUT_OF_MEMORY if we can't create the index buffer
docs: remove ^M chars from GL3.txt
svga: misc whitespace and comment fixes in svga_cmd.c
st/glx/xlib: check for null ctx pointer in glXIsDirect()
xlib: check for null ctx pointer in glXIsDirect()
st/glx/xlib: assorted whitespace, comment fixes
st/mesa: add switch cases for new IR enums to silence warnings
st/glx: add null ctx check in glXDestroyContext()
xlib: add null ctx check in glXDestroyContext()
svga: add PIPE_CAP_MAX_VIEWPORTS to switch to silence warning
draw: add cast in debug_printf() to silence warning
mesa: asst. whitespace, formatting fixes in teximage.c
mesa: fix error checking of DXT sRGB formats in _mesa_base_tex_format()
mesa: fix hodge podge indentation, update comments in texformat.c
meta: move vertex array enables for mipmap generation
mesa: handle missing read buffer in _mesa_get_color_read_format/type()
mesa: change MAX_PROGRAM_ADDRESS_REGS to 1, clamp to it in state tracker
tgsi: replace tgsi_file_names tgsi_file_names[] with tgsi_file_name() function
docs: update some environment variable info
mesa: add missing texture strings in tex_target_name()
swrast: add texfetch code for some XBGR formats
mesa: remove unused BITSET64 macros
mesa: wrap comments, code to 78 columns in multisample.c
svga: formatting fixes in svga_tgsi_insn.c
svga: move some svga/tgsi functions
svga: whitespace, comment, formatting fixes in svga_tgsi_emit.h
svga: reindent svga_tgsi.c
indices: add some comments
svga: add some comments about primitive conversion
svga: add null ptr check in svga_get_tex_sampler_view()
svga: minor cleanups, comments in svga_tgsi_insn.c
gallium/docs: more documentation for pipe_resource::array_size
gallium/hud: do not use free() for the free_query_data hook
svga: use svga_texture() helper instead of casting
svga: add array index assertion in svga_validate_sampler_view()
svga: use new svga_age_texture_view() helper
svga: fix layer/level mix-up in svga_mark_surface_dirty()
svga: use new svga_define_texture_level() helper
svga: allow 3D transfers in svga_texture_transfer_map()
util: s/int/unsigned/ to silence some MSVC warnings
util: add some casts to silence some MSVC warnings
util: int/unsigned changes to silence some MSVC warnings
vega: add some casts to silence MSVC warnings
os: add cast in os_time_sleep() to silence MSVC warning
hud: add cast to silence MSVC warning
hud: include stdio.h since we use fprintf(), fscanf(), etc
hud: add float casts to silence MSVC warnings
st/mesa: make rtt_level, face, slice unsigned to silence MSVC warnings
st/mesa: add casts to silence MSVC warnings
mesa: update glext.h to version 20130624
mesa: remove const from glDebugMessageCallbackARB() function parameter
mesa: add const qualifier to glMultiDrawElementsEXT() indices param
mesa: move declarations before code
glx: move declarations before code
glx: return True/False instead of GL_TRUE/GL_FALSE
svga: use switch statement in svga_shader_type()
svga: pass svga_compile_key by reference instead of value
mesa: add cast to silence MSVC warning
mesa: fix inconsistent function declaration, definitions
mesa: s/GLint/GLuint/ to fix MSVC warnings in mipmap.c
mesa: add casts to fix MSVC warnings in multisample.c
mesa: s/GLuint/GLint/ to silence MSVC warning in textore.c
mesa: s/unsigned/int/ to fix MSVC warning in uniforms.c
tgsi: s/unsigned/int/ to silence MSVC warning
util: add casts to silence MSVC warnings in u_blit.c
hud: silence some MSVC warnings
softpipe: silence some MSVC warnings
mesa: whitespace, formatting, 80-column wrapping
os: add os_get_process_name() function
mesa: simplify some _mesa_IsEnabled() queries
mesa: fix Address Sanitizer (ASan) issue in _mesa_add_parameter()
mesa: update glext.h to version 20130708
tgsi: use X component of the second operand in exec_scalar_binary()
tgsi: exec TGSI_OPCODE_SQRT as a scalar instruction, not vector
tgsi: fix-up KILP comments
tgsi: rename the TGSI fragment kill opcodes
nv30: fix KILL_IF breakage
Bryan Cain (3):
draw: account for separate shader objects in geometry shader code
draw: use geometry shader info in clip_init_state if appropriate
nv50: initialize kick_notify callback in nv50_create
Carl Worth (4):
i965: Avoid segfault in gen6_upload_state
glcpp: Fix post-decrement underflow in loop-control variable
glcpp: Add test case for recently fixed loop-control underflow bug.
docs: Import 9.1.5 release notes
Chad Versace (21):
intel: Remove intel_mipmap_tree::wraps_etc
i965: Fix typo in doxygen hyperlink
android: Define PACKAGE_VERSION/BUGREPORT in CFLAGS
intel: Allocate hiz in intel_renderbuffer_move_to_temp()
intel: Refactor selection of miptree tiling
i965/blorp: Align rectangle primitive for hiz ops
intel: Add field intel_mipmap_slice::has_hiz
i965/blorp: Add fields brw_blorp_mip_info::level,layer
i965: Change signature of brw_get_depthstencil_tile_masks()
intel: Replace checks for hiz_mt with intel_has*hiz()
i965: Remove brw_context::depthstencil::hiz_mt
intel/hsw: Enable hiz (v2)
egl/dri2: Fix min/max swap interval of configs
egl/android: Fix error condition for EGL_ANDROID_image_native_buffer
mesa: Expose GL_OES_texture_npot on GLES1
i965: Fix glColorPointer(GL_FIXED)
i965,i915: Return early if miptree allocation fails
i965/hsw: Change L3 MOCS of 3DSTATE_VERTEX_BUFFERS
i965/hsw: Change L3 MOCS of SURFACE_STAT
i965/hsw: Change L3 MOCS of 3DSTATE_CONSTANT_VS/PS
i965/hsw: Change L3 MOCS for depth, hiz, and stencil
Chia-I Wu (183):
i965: fix SURFACE_STATE dumping
i965/gen7: fix 3DSTATE_LINE_STIPPLE_PATTERN
i965/gen7: fix encoding of (huge) surface size for BRW_SURFACE_BUFFER
winsys/intel: new winsys for intel
ilo: new pipe driver for Intel GEN6+
ilo: add debug flags settable through ILO_DEBUG
ilo: hook up pipe_screen param and fence functions
ilo: hook up pipe screen format functions
ilo: hook up pipe screen resource functions
ilo: add command parser
ilo: hook up pipe context flush function
ilo: add functions to manage shaders
ilo: hook up pipe context state functions
ilo: hook up pipe context blit functions
ilo: hook up pipe context transfer functions
ilo: hook up pipe context query functions
ilo: add GEN6 GPE
ilo: add GEN7 GPE
ilo: add 3D pipeline for GEN6
ilo: add GEN7 support for 3D pipeline
ilo: hook up pipe context 3D functions
ilo: add support for time/occlusion/primitive queries
ilo: hook up pipe context video functions
ilo: hook up pipe context GPGPU functions
ilo: add a toy shader compiler
ilo: compile VS/GS/FS with the toy compiler
ilo: add the driver to the build system
ilo: fix indentation of ilo_gpe_gen*.h
ilo: add ilo_dev_info shared by the screen and contexts
ilo: use ilo_dev_info in GPE and 3D pipeline
ilo: use ilo_dev_info in toy compiler
ilo: move device limits to ilo_dev_info or to GPEs
ilo: support and prefer compact array spacing
ilo: use UMS layout for render targets
ilo: fix tiling/size for special-purpose resources
ilo: allow one-off flags to be specified for CP
ilo: expose register indices of OUTs in ilo_shader
ilo: add gen6_pipeline_update_max_svbi()
ilo: implement GEN7 SO GPE functions
ilo: correctly program SO states for GEN7
ilo: reset SO write offsets for new SO targets
ilo: enable SO support on GEN7
ilo: fix PIPE_CAP_MAX_STREAM_OUTPUT_BUFFERS
util/prim: fix primitive trimming for triangles with adjacency
util/prim: clean up and add comments
draw: use u_assembled_prim() instead of u_assembled_primitive()
util/prim: fix the name of the include guard
util/prim: use vertex count info in u_validate_pipe_prim()
util/prim: assorted fixes for u_decomposed_prims_for_vertices()
util/prim: add u_reduced_prims_for_vertices()
ilo: use u_reduced_prims_for_vertices()
st/egl: fix bulid errors on Android 4.2
gallium: fix type of flags in pipe_context::flush()
ilo: correctly set return types of sampler messages
android: add ilo to the build system
android: libsync is needed on Android 4.2+ for any driver
tgsi: add tgsi_util_get_texture_coord_dim()
tgsi: clean up exec_tex()
tgsi: reorder opcodes in opcode type inference
tgsi: refactor tgsi_opcode_infer_dst_type()
tgsi: refactor tgsi_opcode_infer_src_type()
tgsi: fix operand type of TGSI_OPCODE_NOT
ilo: use tgsi_util_get_texture_coord_dim()
ilo: remove our own type inference
ilo: allow bo format to differ from that requested
ilo: allow for different mapping methods
ilo: support mapping with a staging system buffer
ilo: add support for PIPE_FORMAT_ETC1_RGB8
ilo: move transfer-related functions to a new file
ilo: rename ilo_resource to ilo_texture
ilo: rework ilo_texture
ilo: simplify ilo_texture_get_slice_offset()
ilo: minor cleanups for transfers
ilo: no writeback without PIPE_TRANSFER_WRITE
ilo: refactor transfer mapping
ilo: rework transfer mapping method choosing
ilo: clean up transfer format conversion
winsys/intel: test for and expose address swizzling
ilo: add support for stencil resources on GEN7+
ilo: emit 3DSTATE_STENCIL_BUFFER on GEN7+
ilo: update headers from i965
ilo: update format mappings
ilo: harware contexts are only for the render ring
ilo: replace cp hooks by cp owner and flush callback
ilo: refactor blitter_xy_color_blt()
ilo: use BLT engine to copy between buffers
ilo: use BLT engine to copy between textures
ilo: avoid unnecessary emission of SO states
ilo: correctly set view extent in SURFACE_STATE
ilo: set more fields of 3DSTATE_DEPTH_BUFFER
tgsi: add buffer texture to tgsi_util_get_texture_coord_dim()
ilo: add support for texture buffer objects
ilo: advertise supports for pure integer formats
ilo: add support for TBOs on GEN6
ilo: add support for indirect access of CONST in FS
ilo: enable multiple constant buffers
ilo: move hardware limits to ilo_gpe.h
ilo: switch to ilo states for VF stage
ilo: switch to ilo states for SOL stage
ilo: switch to ilo states for CLIP and SF stages
ilo: switch to ilo states for WM stage
ilo: switch to ilo states for CC stage
ilo: switch to ilo states for shaders and resources
ilo: introduce viewport CSO
ilo: construct SCISSOR_RECT in set_scissor_states()
ilo: introduce sampler CSO
ilo: introduce blend CSO
ilo: simplify emit_3DSTATE_DEPTH_BUFFER()
ilo: introduce vertex element CSO
ilo: add ilo_view_surface for SURFACE_STATE
ilo: add emit_SURFACE_STATE() for sampler views
ilo: use emit_SURFACE_STATE() for constant buffers
ilo: use emit_SURFACE_STATE() for render targets
ilo: construct 3DSTATE_CLIP in create_rasterizer_state()
ilo: construct 3DSTATE_SF in create_rasterizer_state()
ilo: enable bo reuse
ilo: unmap cp bo before destroying it
ilo: clean up states upon context destroy
ilo: use slab allocator for transfers
ilo: add support for TEX2/TXB2/TXL2 in fs
ilo: advertise PIPE_CAP_CUBE_MAP_ARRAY
u_vbuf: fix index buffer leak
util: add util_resource_is_array_texture()
ilo: fix textureSize() for single-layered array textures
ilo: remove unnecessary tex_set_bo/buf_set_bo
ilo: access bo size directly
ilo: get rid of function tables in winsys
ilo: update winsys interface
winsys/intel: reorganize functions
winsys/intel: make struct intel_bo alias drm_intel_bo
ilo: fix for util_blitter_clear() changes
ilo: re-emit states that involve resources
ilo: initialize dirty flags in ilo_init_states()
ilo: bump up PIPE_CAP_GLSL_FEATURE_LEVEL to 140
ilo: mapping a resource may make some states dirty
st/mesa: fix temp texture bindings in st_CopyPixels()
ilo: silence a compiler warning
ilo: add some assertions to help debugging
ilo: fix for half-float vertex arrays
ilo: replace util_blitter by ilo_blitter
ilo: add BLT-based blitting methods to ilo_blitter
ilo: add pipe-based copy method to ilo_blitter
ilo: construct depth/stencil command in create_surface()
ilo: rework shader cache
ilo: rename cache_seqno to uploaded
ilo: replace a boolean by bool
ilo: move internal shader interface to a new header
ilo: use multiple entry points for shader creation
ilo: clean up finalize_shader_states()
ilo: add functions to query shaders
ilo: introduce ilo_shader_cso for VS
ilo: use ilo_shader_cso for GS
ilo: convert pipe_rasterizer_state to ilo_rasterizer_wm
ilo: use ilo_rasterizer_state exclusively in GPE
ilo: use ilo_shader_cso for FS
ilo: map SO registers at shader compile time
ilo: use ilo_shader_state exclusively in GPE
ilo: move SBE setup code to ilo_shader.c
ilo: speed up ilo_shader_select_kernel_routing() a bit
ilo: remove ilo_shader_internal.h from GEN7 pipeline
ilo: remove ilo_shader_internal.h from GEN6 pipeline
ilo: honor render condition in blitter
ilo: support PIPE_CAP_USER_CONSTANT_BUFFERS
ilo: make pipe_draw_info a context state
ilo: support PIPE_CAP_USER_INDEX_BUFFERS
ilo: use a bitmask for enabled constant buffers
ilo: plug a potential index buffer leak
ilo: avoid potential dangling pointer dereference
ilo: track if primitive restart has changed
ilo: use shorter names for dirty flags
ilo: clean up states that take a start_slot
ilo: clean up ilo_set_constant_buffer()
ilo: remove ilo_cbuf_state::count
ilo: clean up states that use ilo_view_surface
ilo: honor surface padding requirements
ilo: correctly initialize undefined registers in fs
ilo: reduce PIPE_CAP_MAX_TEXTURE_CUBE_LEVELS to 12
winsys/intel: build with VISIBILITY_CFLAGS
glsl/build: build builtin_compiler with VISIBILITY_CFLAGS
ilo: clean up ilo_blitter_pipe_begin()
ilo: mark some states dirty when they are really changed
ilo: move a santiy check into its assert()
ilo: skip 3DSTATE_INDEX_BUFFER when possible
Chih-Wei Huang (1):
r600g/sb: Fix Android build v2
Chris Forbes (56):
glapi: add ARB_texture_multisample
mesa: add texobj support for ARB_texture_multisample
tests: add ARB_texture_multisample enums to table
glsl: add support for ARB_texture_multisample
mesa: add new max sample count state
i965: expose new max sample counts
mesa: implement GetMultisamplefv
mesa: implement sample mask
i965: add support for sample mask on Gen6+
i965: expose sample positions
mesa: support multisample textures in framebuffer completeness check
mesa: implement TexImage*Multisample
i965: add support for multisample textures
i965: Support multisampling in surface_state for textures
i965: take the target into account for Gen7 MSAA modes
i965: add a new virtual opcode: SHADER_OPCODE_TXF_MS
i965/vs: add support for ir_txf_ms on Gen6+
i965/fs: add support for ir_txf_ms on Gen6+
i965: enable ARB_texture_multisample on Gen6+
docs: Mark some things done in GL3.txt
mesa: allow internalformat_query with multisample texture targets
mesa: helper for checking renderbuffer sample count
mesa: use _mesa_check_sample_count() for multisample textures
i965: bump MAX_DEPTH_TEXTURE_SAMPLES to 4/8
mesa: only check sample count if we actually wanted multisampling
mesa: provide default implementation of QuerySamplesForFormat
mesa: extract _mesa_is_legal_tex_storage_format helper
mesa: add support for immutable textures to teximagemultisample()
glapi: add definition of ARB_texture_storage_multisample
mesa: add enable bit for ARB_texture_storage_multisample
mesa: improve reported function name in Tex*Multisample
mesa: allow multisample texture targets in [Get]TexParameter*
i965: enable ARB_texture_storage_multisample on Gen6+
docs: mark ARB_texture_storage_multisample done
mesa: don't memcmp() off the end of a cache key.
i965: report correct sample positions
mesa: fix bogus comment about PrimitiveRestart fields
i965/fs: Don't try to use bogus interpolation modes pre-Gen6.
mesa: add accessor for effective stencil ref
mesa: Use accessor for stencil reference values in glGet
intel: Use accessor for stencil reference values
nouveau: Use accessor for stencil reference values
radeon: Use accessor for stencil reference values
st: Use accessor for stencil reference values
swrast: Use accessor for stencil reference values
mesa: Stop clamping stencil reference value at specification time
ARB_fp: accept duplicate precision options
i965: Enable guardband clipping on Gen4/5.
i965/clip: Add support for gl_ClipVertex
i965/vs: add support for emitting gl_ClipVertex
dlist: fix save_SamplerParameteri
i965: Shrink Gen5 VUE map layout to be the same as Gen4.
i965: fix alpha test for MRT
i965/vs: Gen4/5: enable front colors if back colors are written
i965: Gen4/5: use IEEE floating point mode for GLSL shaders.
i965/Gen4: Zero extra coordinates for ir_tex
Christian König (63):
radeonsi: support constants as TEX coordinates
radeonsi: remove constant index limitation v3
radeonsi: fix trivial warning
radeon/llvm: fix trivial warnings
r600g/llvm: Update CONSTANT_BUFFER address space definition
radeonsi: switch to v*i8 for resources and samplers v2
radeon/llvm: replace shader type intrinsic with function attribute
radeon/llvm: make SGPRs proper function arguments v2
radeonsi: remove SI.vs.load.buffer.index
radeonsi: rework input interpolation
radeonsi: add LLVMNoUnwindAttribute to intrinsic
radeon/llvm: enable LICM and DCE pass v2
radeon/llvm: document LLVM commit
radeon/llvm: fix LLVM dependencies
tgsi/ureg: cleanup local temporary emission v2
tgsi/ureg: implement support for array temporaries
glsl_to_tgsi: use get_temp for all allocations
glsl_to_tgsi: allocate arrays separately v2
glsl_to_tgsi: remove indirect addressing limitations
tgsi: remove TGSI_FILE_(IMMEDIATE|TEMP)_ARRAY
tgsi: add ArrayID to declarations
tgsi: use separate structure for indirect address v2
tgsi: add ArrayID documentation v2
radeon/llvm: rework input fetch and output store
radeonsi: switch to using resource destribtors for constants v2
radeonsi: implement indirect adressing of constants
radeonsi: enable indirect adressing
glsl_to_tgsi: make simplify_cmp work with arrays
glsl_to_tgsi: avoid creating arrays if driver doesn't support them
radeon/llvm: remove uneeded inclusion
radeonsi: remove wqm intrinsic
radeonsi: mark all loads as constant
radeonsi: mark most intrinsics as readnone/nounwind
radeonsi: add preloading of all constants
radeonsi: add preloading for all samplers
radeon/llvm: document LLVM commit
radeon/llvm: move system value fetching to common code
radeonsi: add instanceid support
radeonsi: add start instance support
radeonsi: add instance divisor support v3
st/vdpau: fix subtitle related bug v2
radeonsi: remove sampler writemask v3
radeon/winsys: add uvd ring support to winsys v3
radeon/uvd: add UVD implementation v5
autoconf: enable detection of vdpau and xvmc by default
r600/uvd: cleanup disabling tiling on pre EG asics
radeonsi: cleanup disabling tiling for UVD v3
radeon/uvd: stop using anonymous unions
r600/uvd: stop advertising MPEG4 on UVD 2.x chips v2
vl/compositor: cleanup background clearing
vl/buffer: use 2D_ARRAY instead of 3D textures
st/vdpau: fix background handling in the mixer
radeon/uvd: fix quant scan order for mpeg2
radeon/uvd: fix some MPEG4 artifacts
vl/buffers: fix typo in function name
vl/idct: fix for commit 7d2f2a0c890b1993532a45c8c392c28950ddc06e
radeon/uvd: enable interlaced buffers by default
vl/vdpau: fix PresentationQueueQuerySurfaceStatus
st/vdpau: invalidate the handles on destruction
st/vdpau: remove vlCreateHTAB from surface functions
st/vdpau: destroy handle table only when it's empty
r600/uvd: fix check for UVD 2.x
radeon/uvd: fall back to shader based decoding for MPEG2 on UVD 2.x v2
Christoph Bumiller (82):
mesa: implement GL_ARB_texture_buffer_range
gallium: add PIPE_CAP_TEXTURE_BUFFER_OFFSET_ALIGNMENT
st/mesa: add support for GL_ARB_texture_buffer_range
nv50,nvc0: fix/enable texture buffer objects
nv50/ir/ra: swap copyCompound args if src is compound and dst isn't
nv50/ir/ra: fix confusion with conditional RegisterSet::occupy
nv50/ir/ra: also add pre-existing MERGE,SPLIT to constraint list
nv50/ir/ra: make sure all used function inputs get assigned a reg
nv50/ir: fix comparison of system values
nv50/ir: add various new OPs that will be needed for compute
nv50/ir/opt: make optimization aware of atomics, barriers, surface ops
nv50/ir: initialize CodeEmitters' specialized target fields
nvc0/ir/emit: implement surface related ops
nvc0/ir/emit: implement move to and logic ops on predicates
nv50/ir: add support for indirect BRA,CALL
nv50/ir/opt: CALLs cannot load
nvc0/ir/target: some ops can't be predicated, e.g. CALL
nv50/ir/emit: handle OP_ATOM
nvc0/ir/emit: implement INSBF, EXTBF, PERMT and ATOM
nvc0/ir: lower atomics in s[]
nv50/ir: extend moveSources for delta < 0
nvc0/ir: add formatted surface load lib code, move to extra header
nvc0/ir: implement lowering of surface ops for nve4
nvc0/ir: don't replace load from input in COMPUTE progs with VFETCH
nv50/ir/tgsi: handle TGSI_OPCODE_LOAD,STORE
nv50/ir/tgsi: add support for atomics
nv50/ir: add support for barriers
nv50/ir: don't assert on type in Modifier.applyTo if it is 0
nv50/ir/opt: try to make use of SUCLAMP addend
nv50/ir/ssa: add a few comments regarding RenamePass
nv50/ir: print function inputs and outputs
nv50/ir: fix size adjustment for sched info for multiple functions
nvc0/ir: use large issue delay after RET, too
nvc0/ir: skip back edges when determining latest sched value
nv50/ir: implement splitting of 64 bit ops after RA
nv50/ir: add support for different sampler and resource index on nve4
nv50,nvc0: copy writable flag on surface creation
nouveau: align PIPE_BIND_SHADER,COMPUTE_RESOURCEs to 256 bytes
nvc0: add SHADER/COMPUTE_RESOURCE bind flags to format table
nvc0/ir/emit: fix emission of large address offsets
nv50/ir: add CCTL (cache control) op
nvc0/ir: try to fix CAS (CompareAndSwap)
nvc0: implement compute support for nve4
nvc0: they removed the NTID,NCTAID,GRIDID registers on nve4
nvc0: add MP trap handler for nve4
gallium/tests: check format in compute tests, make selectable
nvc0: avoid crash on updating RASTERIZE_ENABLE state
gallium: add TGSI_SEMANTIC_TEXCOORD,PCOORD v3
nvc0: fix max varying count, move CLIPVERTEX,FOG out of the way
nv50: fix 3D render target setup
nv50,nvc0: fix 3d blits, restore viewport after blit
nvc0: use NOUVEAU_GETPARAM_GRAPH_UNITS to get MP count
nvc0: enable compression when supported
nvc0: implement MP performance counters
gallium: add PIPE_CAP_QUERY_PIPELINE_STATISTICS
gallium/docs: fix definition of PIPE_QUERY_SO_STATISTICS
gallium/hud: add support for PIPE_QUERY_PIPELINE_STATISTICS
nvc0: use fence to check state of queries that don't write sequence
nvc0: use correct hw query for PRIMITIVES_GENERATED
nvc0: disable compressed storage type 0xdb for now
nvc0: add some driver statistics queries
nvc0: add some metrics to driver specific queries
nvc0: read PM counters for each warp scheduler separately
nvc0: demagic some of the NVE4_COMPUTE_UPLOAD methods
nouveau: accelerate buffer copies in resource_copy_region
st/mesa: fix bitmap,drawpix,drawtex for PIPE_CAP_TGSI_TEXCOORD
nv50,nvc0: disable DEPTH_RANGE_NEAR/FAR clipping during blit
nvc0: fix for 2d engine R source formats writing RRR1 and not R001
nvc0: honour scaled coordiantes setting for linear textures
nv50: account for pesky prefetch in size calculation of linear textures
nvc0: place staging textures in GART and map them directly
nvc0: fix 128 bit compressed storage type selection
nv50,nvc0: remove MS resolve formats hack
nv50,nvc0: add RGBX16/32_FLOAT formats
nvc0: fix 2D engine MS2 resolve
nvc0: patch up TEX cases with 5 or 6 sources on nve4
nvc0: implement multisample textures
nv50: fix 4th component of Lx_SINT/UINT formats
nv50: set BORDER_COLOR_SRGB in sampler objects
st/mesa: optionally apply texture swizzle to border color v2
nv50: add remaining RGBX formats
r600g: x/y coordinates must be divided by block dim in dma blit
Christopher James Halse Rogers (1):
i965: Fix leak in blorp CopyTexSubImage2D
Chí-Thanh Christopher Nguyễn (1):
targets/dri-i915: Force c++ linker in all cases
Courtney Goeltzenleuchter (3):
ilo: Add support for HW primitive restart.
ilo: simplify shader variant handling
ilo: convert generic depth-stencil-alpha pipe state to ilo pipe state
Daniel Martin (1):
Fix build of swrast only without libdrm
Daniel van Vugt (1):
gbm: Remember to init format on gbm_dri_bo_create.
Dave Airlie (15):
u_blitter: don't create illegal shaders for 1D/3D/RECT/CUBE MSAA
glsl: Implement ARB_texture_query_lod
glapi: fix storage multisample build errors
mesa: fix texture storage multisample prototypes harder.
st/mesa: fix levels in initial texture creation
st/mesa: fix UBO offsets.
r600g: fix two issues in compressed msaa reading code
gallium: add get_sample_position interface
st/mesa: add support for get sample position
st/mesa: add support for ARB_texture_multisample (v3)
r600g: add get_sample_position support (v3)
ralloc: don't write to memory in case of alloc fail.
i965: fix problem with constant out of bounds access (v2)
i965: fix problem with constant out of bounds access (v3)
tgsi: text parser: fix parsing of array in declaration
Dmitry Cherkassov (1):
clover: add dynamic_cast results checking down in clSetKernelArgument() code path.
Emil Velikov (13):
nv50/codegen: do not emitATOM() if the subOp is unknown
nvc0: compile nve4_cache_split_name() only in debug build
nvc0: bail out early during nve4_compute_setup()
nvc0: set ret variable if launch desc allocation failed
docs: restructure release notes into separate folder
docs: move specs to a separate folder
docs: Add some missing release notes
docs: Update 'Making new mesa release'
configure: enable vdpau and xvmc detection, with gallium
nv50: avoid crash on updating RASTERIZE_ENABLE state
glsl/ast: Silence uninitialized variable warnings in the release build
glsl: Silence unused variable warning in the release build
gallium/util: use explicily sized types for {un, }pack_rgba_{s, u}int
Eric Anholt (307):
intel: Use a CPU map of the batch on LLC-sharing architectures.
i965: Fix segfaults from 45a28a927ab7f29ff325b9326d386a39ba538c18
i965: Disable write masking when setting up texturing m0.
glx: Add a little comment about what dri2FlushFrontBuffer() does.
glx: Centralize the code for context flushing.
i965: Fix indentation of index buffer rebase code.
i965: Fix access mode of index buffer rebase.
i965: Add perf debug for a corner case.
i965: Simplify VS push constant upload code since removal of old path.
i965: Remove some stale comments about the brw_constant_buffer atom.
vbo: Print display list debug using printf() like dlist.c does.
vbo: Merge GL_QUADS drawing requests in display lists.
i965/gen4: Work around missing sRGB RGB DXT1 support.
i965: Re-enable the -RHW workaround for original gen4 chips.
i965/fs: Use a helper function for checking for flow control instructions.
i965: Remove writemask support from brw_SAMPLE().
intel: Allow blit readpixels even when the pack alignment is set.
i965/gen7: Set up all samplers even if samplers are sparsely used.
i965/fs: Do a general SEND dependency workaround for the original 965.
i965/fs: Add a bit more instruction dumping useful for upcoming work.
i965/fs: Fix copy propagation with smearing.
i965/fs: Delay setup of uniform loads until after pre-regalloc scheduling.
i965/fs: Only do CSE when the dst types match.
i965/fs: Enable CSE on uniform pull constant loads.
i965/fs: Fix broken math on values loaded from uniform buffers on gen6.
mesa: Fix setup of ctx->Point.PointSprite for GLES2.
i965: Fix the W value of deprecated pointcoords on pre-gen6.
i965: Add WARN_ONCE for depthstencil workarounds we shouldn't be hitting.
intel: Add a flag for miptree mapping to disable transcoding.
intel: Fix software copying of miptree faces for weird formats.
i965: Make sRGB-capable framebuffers by default.
intel: Use the new "ctx" local variable I just added some more.
mesa: Remove extra #define MAXSTRING duplicating MAX_DEBUG_MESSAGE_LENGTH.
mesa: Replace open-coded _mesa_lookup_enum_by_nr().
mesa: Move debug type/severity enums to mesa core.
mesa: Fix _mesa_problem() on context destroy after application debug output
mesa: Merge handling of application-provided and built-in error sources.
mesa: Add support for GL_ARB_debug_output with dynamic ID allocation.
intel: Hook up the WARN_ONCE macro to GL_ARB_debug_output.
intel: Finish renaming fallback_debug() to perf_debug().
i965: Make perf_debug() output to GL_ARB_debug_output in a debug context.
intel: Add missing perf debug for a stall on mapping a BO.
mesa: Report ARB_debug_output for both shader errors and warnings.
mesa: Remove now-unused gl_winsys_error and gl_shader_error enums.
mesa: Remove the enum for the oom-within-debug-output case.
mesa: Remove the special enum for _mesa_error debug output.
intel: Remove the struct intel_region reuse hash table.
intel: Remove num_mapped_regions tracking.
intel: Transition intel_region_map() to being a miptree operation.
intel: Fold intel_region_copy() into its one caller.
intel: Improve the test for readpixels blit path format checking.
intel: Improve the matching (more formats!) for TexImage from PBOs.
intel: Remove some unused debug flags.
i965/fs: Fix register allocation for uniform pull constants in 16-wide.
i965/fs: Add a comment about about an implementation detail.
Reply to: