2009-08-10 12:35:16 +01:00
|
|
|
LLVMPIPE -- a fork of softpipe that employs LLVM for code generation.
|
|
|
|
|
|
|
|
|
2009-08-18 17:36:13 +01:00
|
|
|
Status
|
|
|
|
======
|
|
|
|
|
|
|
|
Done so far is:
|
|
|
|
|
|
|
|
- TGSI -> LLVM fragment shader translation
|
|
|
|
- same level of support as the TGSI SSE2 exec machine
|
|
|
|
- texture sampling via an intrinsic call
|
|
|
|
- done in SoA
|
|
|
|
- input interpolation also code generated
|
|
|
|
|
|
|
|
- blend -> LLVM (including logic ops)
|
|
|
|
- SoA and AoS, but only the former used
|
|
|
|
|
|
|
|
- code is generic
|
|
|
|
- intermediates can be vectors of floats, ubytes, fixed point, etc, and of
|
|
|
|
any width and length
|
|
|
|
- not all operations are implemented for these types yet though
|
|
|
|
|
|
|
|
Most mesa/progs/demos/* work. Speed is on par with Keith's softpipe-opt branch,
|
|
|
|
which includes hand written fast implementations for common cases.
|
|
|
|
|
|
|
|
To do (probably by this order):
|
|
|
|
- code generate the rest of the fragment pipeline, namely the
|
|
|
|
depth/alpha/stencil state
|
|
|
|
- concatenate the fragment pipeline (shader + depth/stencil/alpha + blend) in a
|
|
|
|
single function
|
|
|
|
- code generate texture sampling
|
|
|
|
- translate TGSI control flow instructions
|
|
|
|
- code generate the triangle setup and rasterization
|
|
|
|
|
|
|
|
|
2009-08-10 12:35:16 +01:00
|
|
|
Requirements
|
|
|
|
============
|
|
|
|
|
|
|
|
- Linux
|
|
|
|
|
2009-08-16 11:50:17 +01:00
|
|
|
- udis86, http://udis86.sourceforge.net/
|
|
|
|
|
|
|
|
git clone git://udis86.git.sourceforge.net/gitroot/udis86
|
|
|
|
cd udis86
|
|
|
|
./configure --with-pic
|
|
|
|
make
|
|
|
|
sudo make install
|
|
|
|
|
2009-08-10 12:35:16 +01:00
|
|
|
- LLVM. On Debian based distributions do:
|
|
|
|
|
|
|
|
aptitude install llvm-dev
|
2009-08-10 15:43:04 +01:00
|
|
|
|
2009-08-16 11:50:17 +01:00
|
|
|
There is a typo in one of the llvm-dev 2.5 headers, that causes compilation
|
2009-08-10 15:43:04 +01:00
|
|
|
errors in the debug build:
|
|
|
|
|
|
|
|
--- /usr/include/llvm-c/Core.h.orig 2009-08-10 15:38:54.000000000 +0100
|
|
|
|
+++ /usr/include/llvm-c/Core.h 2009-08-10 15:38:25.000000000 +0100
|
|
|
|
@@ -831,7 +831,7 @@
|
|
|
|
template<typename T>
|
|
|
|
inline T **unwrap(LLVMValueRef *Vals, unsigned Length) {
|
|
|
|
#if DEBUG
|
|
|
|
- for (LLVMValueRef *I = Vals, E = Vals + Length; I != E; ++I)
|
|
|
|
+ for (LLVMValueRef *I = Vals, *E = Vals + Length; I != E; ++I)
|
|
|
|
cast<T>(*I);
|
|
|
|
#endif
|
|
|
|
return reinterpret_cast<T**>(Vals);
|
2009-08-10 12:35:16 +01:00
|
|
|
|
|
|
|
- A x86 or amd64 processor with support for sse2, sse3, and sse4.1 SIMD
|
|
|
|
instructions. This is necessary because we emit several SSE intrinsics for
|
|
|
|
convenience. See /proc/cpuinfo to know what your CPU supports.
|
|
|
|
|
|
|
|
- scons (although it should be straightforward to fix the Makefiles as well)
|
|
|
|
|
|
|
|
|
|
|
|
Building
|
|
|
|
========
|
|
|
|
|
|
|
|
To build everything invoke scons as:
|
|
|
|
|
|
|
|
scons debug=yes statetrackers=mesa drivers=llvmpipe winsys=xlib dri=false -k
|
|
|
|
|
|
|
|
|
|
|
|
Using
|
|
|
|
=====
|
|
|
|
|
|
|
|
Building will create a drop-in alternative for libGL.so. To use it set the
|
|
|
|
environment variables:
|
|
|
|
|
2009-08-10 16:12:51 +01:00
|
|
|
export LD_LIBRARY_PATH=$PWD/build/linux-x86-debug/lib:$LD_LIBRARY_PATH
|
|
|
|
export LD_LIBRARY_PATH=$PWD/build/linux-x86_64-debug/lib:$LD_LIBRARY_PATH
|
2009-08-10 12:35:16 +01:00
|
|
|
|
|
|
|
|
|
|
|
Unit testing
|
|
|
|
============
|
|
|
|
|
|
|
|
Building will also create several unit tests in
|
|
|
|
build/linux-???-debug/gallium/drivers/llvmpipe:
|
|
|
|
|
|
|
|
- lp_test_blend: blending
|
|
|
|
- lp_test_conv: SIMD vector conversion
|
|
|
|
- lp_test_format: pixel unpacking/packing
|
|
|
|
|
|
|
|
|