llvm-6502/test
Bob Wilson 84c5eed15b This patch combines several changes from Evan Cheng for rdar://8659675.
Making use of VFP / NEON floating point multiply-accumulate / subtraction is
difficult on current ARM implementations for a few reasons.
1. Even though a single vmla has latency that is one cycle shorter than a pair
   of vmul + vadd, a RAW hazard during the first (4? on Cortex-a8) can cause
   additional pipeline stall. So it's frequently better to single codegen
   vmul + vadd.
2. A vmla folowed by a vmul, vmadd, or vsub causes the second fp instruction to
   stall for 4 cycles. We need to schedule them apart.
3. A vmla followed vmla is a special case. Obvious issuing back to back RAW
   vmla + vmla is very bad. But this isn't ideal either:
     vmul
     vadd
     vmla
   Instead, we want to expand the second vmla:
     vmla
     vmul
     vadd
   Even with the 4 cycle vmul stall, the second sequence is still 2 cycles
   faster.

Up to now, isel simply avoid codegen'ing fp vmla / vmls. This works well enough
but it isn't the optimial solution. This patch attempts to make it possible to
use vmla / vmls in cases where it is profitable.

A. Add missing isel predicates which cause vmla to be codegen'ed.
B. Make sure the fmul in (fadd (fmul)) has a single use. We don't want to
   compute a fmul and a fmla.
C. Add additional isel checks for vmla, avoid cases where vmla is feeding into
   fp instructions (except for the #3 exceptional case).
D. Add ARM hazard recognizer to model the vmla / vmls hazards.
E. Add a special pre-regalloc case to expand vmla / vmls when it's likely the
   vmla / vmls will trigger one of the special hazards.

Enable these fp vmlx codegen changes for Cortex-A9.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129775 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-19 18:11:57 +00:00
..
Analysis PR9634: Don't unconditionally tell the AliasSetTracker that the PreheaderLoad 2011-04-07 01:35:06 +00:00
Archive
Assembler Reapply r129401 with patch for clang. 2011-04-13 00:36:11 +00:00
Bindings/Ocaml
Bitcode Add intrinsics @llvm.arm.neon.vmulls and @llvm.arm.neon.vmullu.* back. Frontends 2011-03-29 23:06:19 +00:00
BugPoint
CodeGen This patch combines several changes from Evan Cheng for rdar://8659675. 2011-04-19 18:11:57 +00:00
DebugInfo Fix a ton of comment typos found by codespell. Patch by 2011-04-15 05:18:47 +00:00
ExecutionEngine
Feature Revert r129235 pending a vetting of the EH rewrite. 2011-04-10 23:18:04 +00:00
FrontendAda Will not compile without the spec! 2011-03-31 10:03:32 +00:00
FrontendC Fix a ton of comment typos found by codespell. Patch by 2011-04-15 05:18:47 +00:00
FrontendC++ Remove test to check line numbers. There are other numerous tests in our test harness to check line number information. 2011-04-18 22:27:20 +00:00
FrontendFortran
FrontendObjC Testcase for r127301. 2011-03-09 01:05:00 +00:00
FrontendObjC++
Integer
lib
Linker Correctly merge available_externally and regular definitions when they have 2011-02-01 05:33:52 +00:00
LLVMC
MC Thumb2 BFC was insufficiently encoded. 2011-04-15 22:52:15 +00:00
Object Revert "Object: Renable the tests now that none of the build bots complain about aliasing." 2011-01-21 06:27:04 +00:00
Other
Scripts
TableGen Fix a ton of comment typos found by codespell. Patch by 2011-04-15 05:18:47 +00:00
Transforms Add test cases for Jay's r129641 and fix a 32-bit-centric testcase in a file with a 64-bit datalayout. 2011-04-16 14:31:50 +00:00
Unit
Verifier
CMakeLists.txt Fix compiler command line used by lit.py when working with NMake 2011-04-12 22:10:38 +00:00
lit.cfg test/lit.cfg: Add PATHEXT to 'substitution', to recognize tools on Windows hosts. Thanks to Danil Malyshev! 2011-02-24 12:34:34 +00:00
lit.site.cfg.in CMake: Add the new option LLVM_LIT_TOOLS_DIR. It can specify "Path to GnuWin32 tools". 2011-02-09 04:18:58 +00:00
Makefile Reapply r127073(partially): Introduce $(ECHOPATH) to print DOSish path string on MSYS bash for alternative of $(ECHO). 2011-03-08 12:25:10 +00:00
Makefile.tests
site.exp.in
TestRunner.sh