fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2026-05-07 22:18:13 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 4
37857 commits 66 branches 934 tags 1 GiB
Commit graph

37857 commits

This branch
This branch
All branches
Author SHA1 Message Date
Eric Anholt
66d4c65ee2 glsl2: Make the dead code handler make its own talloc context. 
This way, we don't need to pass in a parse state, and the context
doesn't grow with the number of passes through optimization.
 2010-07-27 11:46:05 -07:00
Eric Anholt
8bbdf6e7cd mesa: Allow large temporary indices coming into the temporary reg allocator. 
This gets glsl-vs-raytrace, glsl-fs-raytrace running on the new
compiler.
 2010-07-27 11:45:27 -07:00
Eric Anholt
85cd64ee17 glsl2: Talloc type names. 
Otherwise, we end up losing structure names after compile time, and
dumping IR often ends up reporting some other mysterious string.
 2010-07-27 11:14:59 -07:00
Jerome Glisse
2b3b76a4a0 r600g: always emit literal after emiting an alu instruction 
Make sure we always fill in the literal after alu instruction.

Signed-off-by: Jerome Glisse <jglisse@redhat.com>
 2010-07-27 13:50:11 -04:00
Eric Anholt
9a770ee49f ir_to_mesa: Set the swizzle on constant struct src regs. 
MESA_GLSL=nopt now produces believable output for glsl-fs-raytrace.
 2010-07-27 10:41:01 -07:00
Eric Anholt
fdbaca931a ir_to_mesa: Fix stray "break" that broke functions of >1 argument. 2010-07-27 10:08:38 -07:00
Eric Anholt
832aad989e glsl2: Add optimization pass for algebraic simplifications. 
This cleans up the assembly output of almost all the non-logic tests
glsl-algebraic-*.  glsl-algebraic-pow-two needs love (basically,
flattening to a temporary and squaring it).
 2010-07-27 09:43:52 -07:00
Eric Anholt
5533c6e380 ir_validate: Check the types of expression operations. 2010-07-27 09:43:52 -07:00
José Fonseca
4fbffb7d90 llvmpipe: Use lp_build_select_bitwise() where appropriate. 
Fixes fdo 29269.
 2010-07-27 13:09:47 +01:00
José Fonseca
2a45972fb2 gallivm: Add lp_build_select_bitwise() alternative to lp_build_select_bitwise 
When

  (mask & a) | (~mask & b)

is meant instead of

   mask ? a : b
 2010-07-27 13:09:46 +01:00
Eric Anholt
6a1401eb88 glsl2: Fix missing visit_continue return in ir_validate. 2010-07-27 00:18:57 -07:00
Brian Paul
fef5ae949f cell: comment-out unused fields, functions 2010-07-26 20:50:17 -06:00
Brian Paul
dd0cf2e1f5 cell: make functions static 2010-07-26 20:49:54 -06:00
Brian Paul
0315c00f58 cell: fix segfault when freeing samplers 2010-07-26 20:49:35 -06:00
Brian Paul
2a8021667d cell: comment-out unneeded padding field 2010-07-26 20:49:09 -06:00
Brian Paul
d2c714627c cell: added const qualifier 2010-07-26 20:48:46 -06:00
Brian Paul
b17ee335e3 util: fix unused function warning on non-x86 2010-07-26 20:48:29 -06:00
Eric Anholt
576d01ad8c ir_to_mesa: Fix up handling of void function returns. 
void functions have a type of glsl_type::void_type, not a null type.
 2010-07-26 19:41:39 -07:00
Eric Anholt
c91809e1e4 ir_to_mesa: Actually allocate the right size for constant matrix temps. 2010-07-26 19:41:23 -07:00
Eric Anholt
5b6890a388 ir_to_mesa: Add support for structure constants. 
Fixes:
TPPStreamCompiler::assignOperands
 2010-07-26 19:40:47 -07:00
Eric Anholt
fbaca31352 glsl2: Also steal the constant components of aggregate-typed ir_constants. 2010-07-26 19:30:19 -07:00
Kenneth Graunke
eb2cc4f1b1 glsl2: Steal ir_variable's constant_value field. 
Fixes a link-time crash in glsl-vs-cross-3.
 2010-07-26 19:09:10 -07:00
Eric Anholt
e2130d4e39 glsl2: Remove old .gitignore file from when we were out of tree. 2010-07-26 17:54:35 -07:00
Eric Anholt
afe125e0a1 Merge remote branch 'origin/master' into glsl2 
This pulls in multiple i965 driver fixes which will help ensure better
testing coverage during development, and also gets past the conflicts
of the src/mesa/shader -> src/mesa/program move.

Conflicts:
	src/mesa/Makefile
	src/mesa/main/shaderapi.c
	src/mesa/main/shaderobj.h
 2010-07-26 17:53:27 -07:00
Eric Anholt
d64343f1ae ir_to_mesa: Flag when we use the address reg. 
Hardware backends will get angry otherwise.
 2010-07-26 17:45:37 -07:00
Kristian Høgsberg
5403ca79b2 glx: Remove function prototypes no longer necessary 2010-07-26 17:37:59 -04:00
Eric Anholt
2621100458 i965: Fix reversed naming of the operations in compute-to-mrf optimization. 
Also fix up comments, so that the difference between the two passes is
clarified.
 2010-07-26 13:08:25 -07:00
Eric Anholt
b3ea15f12b i965: Clean up a few magic numbers to use brw_defines.h defs. 2010-07-26 13:08:25 -07:00
Eric Anholt
ca0f4e2c10 i965: Use MIN2, MAX2 instead of rolling our own. 2010-07-26 13:08:25 -07:00
Eric Anholt
9e31adfa46 i965: Fold the "is arithmetic" bit of 965 opcodes into the opcode list. 2010-07-26 13:08:25 -07:00
Eric Anholt
b9c84515a5 i965: Remove some duped register size/count definitions 2010-07-26 13:08:25 -07:00
Eric Anholt
22f839292f i965: Move the GRF-to-MRF optimizations to brw_optimize.c. 2010-07-26 13:08:25 -07:00
Benjamin Segovia
a64def5f2a i965: Improve (i.e. remove) some grf-to-mrf unnecessary moves 
Several routines directly analyze the grf-to-mrf moves from the Gen
binary code. When it is possible, the mov is removed and the message
register is directly written in the arithmetic instruction

Also redundant mrf-to-grf moves are removed (frequently for example,
when sampling many textures with the same uv)

Code was tested with piglit, warsow and nexuiz on an Ironlake
machine. No regression was found there

Note that the optimizations are *deactivated* on Gen4 and Gen6 since I
did test them properly yet. No reason there are bugs but who knows

The optimizations are currently done in branch free programs *only*.
Considering branches is more complicated and there are actually two
paths: one for branch free programs and one for programs with branches

Also some other optimizations should be done during the emission
itself but considering that some code is shader between vertex shaders
(AOS) and pixel shaders (SOA) and that we may have branches or not, it
is pretty hard to both factorize the code and have one good set of
strategies
 2010-07-26 13:08:25 -07:00
Kristian Høgsberg
e4aa9440d9 glx: Drop debug fprintf that snug in with the previous commit 2010-07-26 16:06:05 -04:00
Kristian Høgsberg
2235b1c72d glx: Enable copy subbuffer patch when GLX_DIRECT_RENDERING is #defined 
Depending on __DRI_COPY_SUB_BUFFER doesn't work when we no longer include
dri_interface.h.

https://bugs.freedesktop.org/show_bug.cgi?id=29264
 2010-07-26 15:52:45 -04:00
nobled
c88fc26ac9 st/egl: Fix debug line 
Acked-by: Jakob Bornecrantz <jakob@vmware.com>
 2010-07-26 12:27:21 -07:00
nobled
3cef6c42bc util: fix CPU detection on OS X 
s/PIPE_OS_DARWIN/PIPE_OS_APPLE, since there is no PIPE_OS_DARWIN.

Acked-by: Vinson Lee <vlee@vmware.com>
 2010-07-26 12:27:01 -07:00
nobled
b1ef3e0863 st/xorg: fix use-after-free 
Acked-by: Jakob Bornecrantz <jakob@vmware.com>
 2010-07-26 12:26:58 -07:00
Jakob Bornecrantz
0697d41fce i965g: Enable llvm in dri driver if built 2010-07-26 12:26:53 -07:00
nobled
4830237660 i915g: Fix llvm build 
Acked-by: Jakob Bornecrantz <jakob@vmware.com>
 2010-07-26 12:25:05 -07:00
Kristian Høgsberg
bc34aa6128 glx: Drop duplicate psc field in dri context struct 
Same problem as fixed for drisw in 4d58b5b482.
 2010-07-26 15:18:56 -04:00
Eric Anholt
941b893032 i965: Allow VS MOVs to use immediate constants. 
Clarifies program assembly, and with a little tweak to always use
constant_map, we could cut down on constant buffer payload.
 2010-07-26 10:11:31 -07:00
Stephan Schmid
0bebdc230f r600g: implememt the LIT instruction 2010-07-26 10:34:05 -04:00
Marek Olšák
a3a42e4696 util: fix another mutex leak in mempool 
By fixing one, I introduced another. Crap.
 2010-07-26 14:56:48 +02:00
Marek Olšák
d26fb69169 util: fix mutex leaks in mempool 2010-07-26 14:53:27 +02:00
Dave Airlie
8c26dc2bfe r300g: fix macro substitution problem 
isn't a problem yet, but have issues in hiz branch.

Signed-off-by: Dave Airlie <airlied@redhat.com>
 2010-07-26 12:02:14 +10:00
Marek Olšák
6f2936c654 r300g: implement D24X8 texture sampling for r3xx-r4xx 
Because the hw can't sample it, I reinterpret the format as G16R16 and
sample the G component. This gives 16 bits of precision, which should be
enough for depth texturing (surprisingly, the sampled values are exactly
the same as in D16 textures).

This also enables EXT_packed_depth_stencil on those old chipsets, finally.
 2010-07-25 23:40:51 +02:00
Marek Olšák
451a0ddb19 r300g: make sure a texture is large enough for the CBZB clear 
The number of macrotiles in the Y direction must be even, otherwise memory
corruption may happen (e.g. broken fonts). Basically, if we get a buffer
in resource_from_handle, we can determine from the buffer size whether it's
safe to use the CBZB clear or not.
 2010-07-25 10:33:09 +02:00
Marek Olšák
49330fc5ac r300g: do not use TXPITCH_EN if the width is POT and the height is NPOT 2010-07-25 10:25:21 +02:00
Marek Olšák
c92d232061 r300g: do not use TXPITCH_EN for power-of-two textures from the DDX 
We were using TXPITCH_EN for textures from the DDX since ever, for nothing.
 2010-07-25 10:25:21 +02:00