LLVM backend for 6502
Go to file
Sanjay Patel 0332323ab6 [x86] Implement combineRepeatedFPDivisors
Set the transform bar at 2 divisions because the fastest current
x86 FP divider circuit is in SandyBridge / Haswell at 10 cycle
latency (best case) relative to a 5 cycle multiplier. 
So that's the worst case for this transform (no latency win), 
but multiplies are obviously pipelined while divisions are not,
so there's still a big throughput win which we would expect to
show up in typical FP code.

These are the sequences I'm comparing:

  divss   %xmm2, %xmm0
  mulss   %xmm1, %xmm0
  divss   %xmm2, %xmm0

Becomes:

  movss   LCPI0_0(%rip), %xmm3    ## xmm3 = mem[0],zero,zero,zero
  divss   %xmm2, %xmm3
  mulss   %xmm3, %xmm0
  mulss   %xmm1, %xmm0
  mulss   %xmm3, %xmm0

[Ignore for the moment that we don't optimize the chain of 3 multiplies
into 2 independent fmuls followed by 1 dependent fmul...this is the DAG
version of: https://llvm.org/bugs/show_bug.cgi?id=21768 ...if we fix that,
then the transform becomes even more profitable on all targets.]

Differential Revision: http://reviews.llvm.org/D8941



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@235012 91177308-0d34-0410-b5e6-96231b3b80d8
2015-04-15 15:22:55 +00:00
autoconf Try to unbreak Clang build to export LLVM_ABI_BREAKING_CHECKS, fixup for r233310. 2015-04-01 11:46:15 +00:00
bindings Go bindings: make various DIBuilder arguments optional. 2015-04-08 20:18:57 +00:00
cmake [msvc] Strip /W[0-4] before appending /W4 2015-04-14 16:57:54 +00:00
docs Correct 'teh' and other typos / repeated words. 2015-04-14 20:52:58 +00:00
examples DebugInfo: Add implicit conversion from DISubprogram to DIScope 2015-04-14 16:19:44 +00:00
include Clarify that Twine::toVector *appends* the Twine to the given SmallString. 2015-04-15 11:27:32 +00:00
lib [x86] Implement combineRepeatedFPDivisors 2015-04-15 15:22:55 +00:00
projects
test [x86] Implement combineRepeatedFPDivisors 2015-04-15 15:22:55 +00:00
tools Change range-based for-loop to use const auto&. No functionality change. 2015-04-15 03:17:49 +00:00
unittests Only recalculate DFS Numbers if invalid. Invalidate DFS numbers on reset. Add unit test to verify recalculation 2015-04-14 19:49:26 +00:00
utils Change range-based for-loops to be -Wrange-loop-analysis clean. 2015-04-15 01:21:15 +00:00
.arcconfig
.clang-format
.clang-tidy
.gitignore Ignore compile_commands.json only at the root of the tree. 2015-03-26 18:55:42 +00:00
CMakeLists.txt [CMake] Updates to llvm-shlib to support overriding exports list and added an option to export all symbols. 2015-04-13 21:29:46 +00:00
CODE_OWNERS.TXT Tom is also responsible for the 3.6 branch. 2015-03-16 18:15:27 +00:00
configure Try to unbreak Clang build to export LLVM_ABI_BREAKING_CHECKS, fixup for r233310. 2015-04-01 11:46:15 +00:00
CREDITS.TXT
LICENSE.TXT Update for a new year. 2015-03-12 01:25:29 +00:00
llvm.spec.in
LLVMBuild.txt
Makefile
Makefile.common
Makefile.config.in Try to unbreak Clang build to export LLVM_ABI_BREAKING_CHECKS, fixup for r233310. 2015-04-01 11:46:15 +00:00
Makefile.rules
README.txt Revert test commit at revision 233535. 2015-03-30 12:39:03 +00:00

Low Level Virtual Machine (LLVM)
================================

This directory and its subdirectories contain source code for LLVM,
a toolkit for the construction of highly optimized compilers,
optimizers, and runtime environments.

LLVM is open source software. You may freely distribute it under the terms of
the license agreement found in LICENSE.txt.

Please see the documentation provided in docs/ for further
assistance with LLVM, and in particular docs/GettingStarted.rst for getting
started with LLVM and docs/README.txt for an overview of LLVM's
documentation setup.

If you're writing a package for LLVM, see docs/Packaging.rst for our
suggestions.