1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-19 19:12:56 +02:00
llvm-mirror/utils
Andrea Di Biagio 23f0c22225 [tblgen][llvm-mca] Add the ability to describe move elimination candidates via tablegen.
This patch adds the ability to identify instructions that are "move elimination
candidates". It also allows scheduling models to describe processor register
files that allow move elimination.

A move elimination candidate is an instruction that can be eliminated at
register renaming stage.
Each subtarget can specify which instructions are move elimination candidates
with the help of tablegen class "IsOptimizableRegisterMove" (see
llvm/Target/TargetInstrPredicate.td).

For example, on X86, BtVer2 allows both GPR and MMX/SSE moves to be eliminated.
The definition of 'IsOptimizableRegisterMove' for BtVer2 looks like this:

```
def : IsOptimizableRegisterMove<[
  InstructionEquivalenceClass<[
    // GPR variants.
    MOV32rr, MOV64rr,

    // MMX variants.
    MMX_MOVQ64rr,

    // SSE variants.
    MOVAPSrr, MOVUPSrr,
    MOVAPDrr, MOVUPDrr,
    MOVDQArr, MOVDQUrr,

    // AVX variants.
    VMOVAPSrr, VMOVUPSrr,
    VMOVAPDrr, VMOVUPDrr,
    VMOVDQArr, VMOVDQUrr
  ], CheckNot<CheckSameRegOperand<0, 1>> >
]>;
```

Definitions of IsOptimizableRegisterMove from processor models of a same
Target are processed by the SubtargetEmitter to auto-generate a target-specific
override for each of the following predicate methods:

```
bool TargetSubtargetInfo::isOptimizableRegisterMove(const MachineInstr *MI)
const;
bool MCInstrAnalysis::isOptimizableRegisterMove(const MCInst &MI, unsigned
CPUID) const;
```

By default, those methods return false (i.e. conservatively assume that there
are no move elimination candidates).

Tablegen class RegisterFile has been extended with the following information:
 - The set of register classes that allow move elimination.
 - Maxium number of moves that can be eliminated every cycle.
 - Whether move elimination is restricted to moves from registers that are
   known to be zero.

This patch is structured in three part:

A first part (which is mostly boilerplate) adds the new
'isOptimizableRegisterMove' target hooks, and extends existing register file
descriptors in MC by introducing new fields to describe properties related to
move elimination.

A second part, uses the new tablegen constructs to describe move elimination in
the BtVer2 scheduling model.

A third part, teaches llm-mca how to query the new 'isOptimizableRegisterMove'
hook to mark instructions that are candidates for move elimination. It also
teaches class RegisterFile how to describe constraints on move elimination at
PRF granularity.

llvm-mca tests for btver2 show differences before/after this patch.

Differential Revision: https://reviews.llvm.org/D53134

llvm-svn: 344334
2018-10-12 11:23:04 +00:00
..
benchmark Add benchmark and benchmark_main to the Utils folder in IDEs. 2018-09-21 22:55:57 +00:00
bugpoint
count
crosstool
docker [docker] Fix LLVM_EXTERNAL_PROJECTS cmake variable value 2018-07-24 18:34:13 +00:00
emacs [emacs] Indent statement continuation to match clang-format 2018-08-02 08:50:41 +00:00
FileCheck Refactor FileCheck to make it usable as an API 2018-08-07 21:58:49 +00:00
fpcmp
gdb-scripts [gdb] Fix SmallVector pretty printer after r337514 2018-07-23 21:33:51 +00:00
git
git-svn [git-llvm] Fix some issues surrouding EOL conversion on Windows. 2018-10-09 23:42:28 +00:00
jedit
kate Add (very partial) Kate syntax highlighting definition for TableGen 2018-07-18 18:35:27 +00:00
KillTheDoctor Remove @brief commands from doxygen comments, too. 2018-05-01 16:10:38 +00:00
lint
lit make lit builtins a package 2018-09-28 17:55:18 +00:00
llvm-build Remove llvm-build's --configure-target-def-file. 2018-04-20 17:21:10 +00:00
llvm-lit [lit] Actually do normalize the case of files in the config map. 2017-09-21 21:27:11 +00:00
LLVMVisualizers Fix typo in the MSVC Visualizer for SmallVector class 2018-08-06 16:47:24 +00:00
Misc
not Refactor ExecuteAndWait to take StringRefs. 2018-06-12 17:43:52 +00:00
PerfectShuffle
release merge-request.sh: Add 7.0 metabug 2018-09-28 02:30:42 +00:00
Reviewing Make email options of find_interesting_reviews more flexible. 2018-06-29 07:16:27 +00:00
sanitizers UBSan blacklist workaround for bot timeouts 2018-06-25 21:28:35 +00:00
TableGen [tblgen][llvm-mca] Add the ability to describe move elimination candidates via tablegen. 2018-10-12 11:23:04 +00:00
Target/ARM
testgen
textmate
unittest [unittests] Do not use llvm::sort in googlemock 2018-09-20 04:27:32 +00:00
UpdateTestChecks Re-land r335297 "[X86] Implement more of x86-64 large and medium PIC code models" 2018-07-23 21:14:35 +00:00
valgrind
vim vim: rename singlethread to syncscope 2018-03-22 16:39:54 +00:00
vscode
yaml-bench Remove \brief commands from doxygen comments. 2018-05-01 15:54:18 +00:00
abtest.py utils/abtest: Refactor and add bisection method 2018-09-07 17:08:44 +00:00
bisect
bisect-skip-count
bugpoint_gisel_reducer.py Add a utility to reduce GlobalISel tests 2018-01-23 19:47:10 +00:00
check-each-file
clang-parse-diagnostics-file
codegen-diff
countloc.sh
create_ladder_graph.py
demangle_tree.py Add a utility script to stress test the demangler. 2018-08-30 20:53:48 +00:00
DSAclean.py
DSAextract.py
extract_symbols.py Fix some user facing typos 2018-03-17 17:30:08 +00:00
findmisopt
findoptdiff
findsym.pl
GenLibDeps.pl
GetRepositoryPath
GetSourceVersion
getsrcs.sh
indirect_calls.py Remove a stale comment cut and pasted from another file. 2018-04-03 17:07:05 +00:00
lldbDataFormatters.py [utils] Fix the llvm::Optional data formatter 2018-07-23 21:59:06 +00:00
llvm-compilers-check
llvm-gisel-cov.py [globalisel][tablegen] Generate rule coverage and use it to identify untested rules 2017-11-16 00:46:35 +00:00
llvm-native-gxx
llvm.grm
LLVMBuild.txt
llvmdo
llvmgrep
prepare-code-coverage-artifact.py [Coverage] Apply filtered paths to summary 2018-10-11 04:00:51 +00:00
schedcover.py [MachineScheduler] Add itinerary to schedcover.py. Make default work in the command line filter 2018-03-27 04:26:39 +00:00
shuffle_fuzz.py
shuffle_select_fuzz_tester.py Adding a shufflevector and select LLVM IR instructions fuzz tool 2017-10-31 11:39:31 +00:00
sort_includes.py
unicode-case-fold.py Resubmit r325107 (case folding DJB hash) 2018-02-21 22:36:31 +00:00
update_analyze_test_checks.py [NFC] chmod +x utils/update_analyze_test_checks.py 2018-06-15 12:41:50 +00:00
update_cc_test_checks.py Fix LLVM IR check lines in utils/update_cc_test_checks.py 2018-03-14 17:47:07 +00:00
update_llc_test_checks.py Re-land r335297 "[X86] Implement more of x86-64 large and medium PIC code models" 2018-07-23 21:14:35 +00:00
update_mca_test_checks.py [utils] Ensure that update_mca_test_checks.py writes prefixes in alphabetical order 2018-10-04 14:42:19 +00:00
update_mir_test_checks.py update_mir_test_checks: Fix handling of IR input after r326284 2018-03-12 18:06:58 +00:00
update_test_checks.py [UpdateTestChecks] Remove unnecessary return from add_ir_checks 2018-04-05 09:30:42 +00:00
UpdateCMakeLists.pl
wciia.py