Commit graph

16375 commits

Author SHA1 Message Date
Tom Stellard
50ff2dc0a4 radeon/llvm: Add special nodes for SALU operations on VCC
The VCC register is tricky because the SALU views it as 64-bit, but the
VALU views it as 1-bit.  In order to deal with this we've added some
special bitcast and binary operations to help convert from the 64-bit
SALU view to the 1-bit VALU view and vice versa.
2012-07-27 17:08:08 +00:00
Tom Stellard
c424975572 radeon/llvm: Add i1 registers for SI. 2012-07-27 17:08:08 +00:00
Tom Stellard
bdda1cb914 radeon/llvm: Fix CCReg definitions on SI 2012-07-27 17:08:08 +00:00
Tom Stellard
ae9be358f2 radeonsi: Enable PIPE_SHADER_CAP_INTEGERS 2012-07-27 17:08:08 +00:00
Tom Stellard
022b54359a radeonsi: Add support for loading integers from constant memory 2012-07-27 17:08:07 +00:00
Tom Stellard
ad95bcb31f radeon/llvm: Add bitconvert patterns for SI 2012-07-27 17:08:07 +00:00
Tom Stellard
4cab682184 radeon/llvm: Add custom lowering for SELECT_CC nodes on SI 2012-07-27 17:08:07 +00:00
Tom Stellard
ba76684292 radeon/llvm: Move conditional pattern leafs to common tablegen file 2012-07-27 17:08:07 +00:00
Tom Stellard
d36455ba2c radeon/llvm: Implement getSetCCResultType for SI 2012-07-27 17:08:07 +00:00
Tom Stellard
e8825ce6e1 radeon/llvm: Custom lower BR_CC for SI 2012-07-27 17:08:07 +00:00
Tom Stellard
87272e9e25 radeon/llvm: Move lowering of BR_CC node to R600ISelLowering
SI will handle BR_CC different from R600, so we need to move it
out of the shared instruction selector.
2012-07-27 17:08:07 +00:00
Tom Stellard
92823fb72a radeon/llvm: Move lowering of SETCC node to R600ISelLowering
SI will handle SETCC different from R600, so we need to move it
out of the shared instruction selector.
2012-07-27 17:08:06 +00:00
Tom Stellard
46d12c99a2 radeon/llvm: Use correct node type when lowering SETCC 2012-07-27 17:08:06 +00:00
Tom Stellard
47d1b0a809 radeon/llvm: Move LowerSELECT_CC into R600ISelLowering
SI will handle SELECT_CC different from R600, so we need to move it out
of the shared instruction selector.
2012-07-27 17:08:06 +00:00
Eric Anholt
11ff18fcf5 automake: Remove OPT_FLAGS.
If you want to change your compiler arguments, just set CFLAGS/CXXFLAGS.
Having Mesa have this separate variable is a great way to have your arguments
not thoroughly propagated to all compiler invocations.

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2012-07-26 17:30:06 -07:00
Eric Anholt
87a1c4f233 automake: Remove ARCH_FLAGS.
In all current uses, it was appended to CFLAGS, which already had -m32.  If
you want to do some other flag supplied to compiler invocations, there's
CFLAGS/CXXFLAGS.

Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
2012-07-26 17:30:06 -07:00
Brian Paul
f37f1a7209 svga: initialize svga_compile_key to zeros to be safe 2012-07-26 16:00:31 -06:00
Brian Paul
dafa77201f svga: fix invalid memory reference in needs_to_create_zero()
The emit->key.fkey info is only valid if we're generating a fragment shader.
We should not look at it if we're generating a vertex shader.

When generating a vertex shader, the value of emit->key.fkey.num_textures was
garbage and the loop over num_textures would read invalid data.  At best
this would cause us to emit an unused constant.  At worse, we could segfault.
Just by dumb luck, fkey.num_textures was usually a smallish integer.

NOTE: This is a candidate for the 8.0 branch.

Reviewed-by: José Fonseca <jfonseca@vmware.com>
2012-07-26 16:00:31 -06:00
Vinson Lee
4f109ca4e8 scons: Fix build with clang.
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Brian Paul <brianp@vmware.com>
2012-07-25 17:04:30 -07:00
Olivier Galibert
fa76d04aea softpipe: fix copy/paste error in tex sample code
Fixes https://bugs.freedesktop.org/show_bug.cgi?id=52369

Reviewed-by: Brian Paul <brianp@vmware.com>
2012-07-25 07:47:19 -06:00
Anthony G. Basile
f35e380dd2 Fix compile time errors when building against uclibc
Mesa misses a few checks when compiling on a uclibc system
which cause it to fall back on glibc-ism.  This patch
addresses those issues.

Acked-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
2012-07-24 13:00:47 -07:00
Jerome Glisse
1ffac44e83 r600g: enable streamout only on 2.14 or latter kernel
The kernel streamout support was supposed to get into 3.3 along
the tiling change and thus use the same kernel version bump of
2.13 to report userspace that streamout register were supported.

This is not what happen. So as streamout kernel support did not
bump the kernel driver version, rely on kernel 2.14 version bump
to know if streamout is enabled or not. Which means you need at
least 3.4 kernel.

Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2012-07-24 15:08:31 -04:00
Christian König
de3335dba8 radeonsi: remove old state handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:30 +02:00
Christian König
9b213c871a radeonsi: move everything else into the new handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:30 +02:00
Christian König
53d47889e6 radeonsi: move format handling into si_state.c
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:30 +02:00
Christian König
73dd906ba0 radeonsi: move remaining sampler state into si_state.c
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:30 +02:00
Christian König
ca9cf611b6 radeonsi: move draw state into new handling
Split it out into si_state_draw.c

Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:30 +02:00
Christian König
0d6b0b512a radeonsi: move constants to new state handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:30 +02:00
Christian König
baf2039756 radeonsi: move sampler states into new handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:30 +02:00
Christian König
3c09f11e5c radeonsi: move shaders to new handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:30 +02:00
Christian König
bd2a5cf328 radeonsi: move spi into new handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:30 +02:00
Christian König
840f05da6b radeonsi: move init state to new handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:30 +02:00
Christian König
e4e6f954ae radeonsi: move draw_info to new state handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:30 +02:00
Christian König
76660dfcce radeonsi: move CB_TARGET_MASK into fb/blend state
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:30 +02:00
Christian König
e6937211da radeonsi: move stencil_ref to new handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:30 +02:00
Christian König
b41b3eb989 radeonsi: move dsa state to new handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:29 +02:00
Christian König
bd18a316e1 radeonsi: move infeered fb/rs state to new handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:29 +02:00
Christian König
f67fae0e43 radeonsi: move rasterizer state into new handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:29 +02:00
Christian König
835098a529 radeonsi: move framebuffer to new handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:29 +02:00
Christian König
7e011d92c9 radeonsi: move viewport to new handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:29 +02:00
Christian König
43f414f7b7 radeonsi: move scissor state to new state handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:29 +02:00
Christian König
9cbbe0d4e6 radeonsi: move clip state to new handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:29 +02:00
Christian König
0a091a4824 radeonsi: move blend color to new state handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:29 +02:00
Christian König
63636ae52a radeonsi: move blender to new state handling
Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:29 +02:00
Christian König
bf7302a6e1 radeonsi: rework state handling v2
Add a complete new state handling for SI.

v2: fix spelling error

Signed-off-by: Christian König <deathsimple@vodafone.de>
2012-07-24 12:29:29 +02:00
José Fonseca
c30bf68946 gallivm: Prefer the standard JIT engine whenever possible.
Testing shows that the standard JIT engine retrofited with AVX support is quite
stable and as capable to handle AVX instructions as MC-JIT is.

And the old JIT is much more memory efficient, as we don't need to
allocate one engine instance per shader, as we do for MC-JIT due to its
incompleteness.

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
2012-07-23 17:46:38 +01:00
Jerome Glisse
cb149bf9e1 r600g: don't emit forbidden reg with old kernel on evergreen
Fix https://bugs.freedesktop.org/show_bug.cgi?id=52313

Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2012-07-23 11:42:36 -04:00
Jerome Glisse
b7b5a77ec0 r600g: don't emit forbidden register on old kernel
Fix https://bugs.freedesktop.org/show_bug.cgi?id=52313

Signed-off-by: Jerome Glisse <jglisse@redhat.com>
2012-07-23 11:28:25 -04:00
Vincent Lejeune
bc4b4c605c radeon/llvm: Fix a bug with IF LOGICALNZ with int operand
Signed-off-by: Tom Stellard <thomas.stellard@amd.com>
2012-07-23 15:04:36 +00:00
Tom Stellard
044de40cb0 pipe_loader: Try to connect with the X server before probing pciids v2
When X is running it is neccesary for pipe_loader to authenticate with
DRM, in order to be able to use the device.

This makes it possible to run OpenCL programs while X is running.

v2:
  - Fix C++ style comments
  - Drop Xlib-xcb dependency
  - Close the X connection when done
  - Split auth code into separate function

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
2012-07-23 13:25:36 +00:00