Commit Graph

657 Commits

Author SHA1 Message Date
Jingyue Wu 48a9bdc6aa [NVPTX] make load on global readonly memory to use ldg
Summary:
[NVPTX] make load on global readonly memory to use ldg

Summary:
As describe in [1], ld.global.nc may be used to load memory by nvcc when
__restrict__ is used and compiler can detect whether read-only data cache
is safe to use.

This patch will try to check whether ldg is safe to use and use them to
replace ld.global when possible. This change can improve the performance
by 18~29% on affected kernels (ratt*_kernel and rwdot*_kernel) in 
S3D benchmark of shoc [2]. 

Patched by Xuetian Weng. 

[1] http://docs.nvidia.com/cuda/kepler-tuning-guide/#read-only-data-cache
[2] https://github.com/vetter/shoc

Test Plan: test/CodeGen/NVPTX/load-with-non-coherent-cache.ll

Reviewers: jholewinski, jingyue

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D11314

llvm-svn: 242713
2015-07-20 21:28:54 +00:00
Eli Bendersky b09cfb51ca Use inbounds GEPs for memcpy and memset lowering
Follow-up on discussion in http://reviews.llvm.org/D11220

llvm-svn: 242542
2015-07-17 16:42:33 +00:00
Eli Bendersky f871e09d2d Streamline the coding style in NVPTXLowerAggrCopies
Make the style consistent with LLVM style throughout and clang-format.

llvm-svn: 242439
2015-07-16 20:42:38 +00:00
Jingyue Wu e7981cee24 [NVPTX] enable SpeculativeExecution in NVPTX
Summary:
SpeculativeExecution enables a series straight line optimizations (such
as SLSR and NaryReassociate) on conditional code. For example,

  if (...)
    ... b * s ...
  if (...)
    ... (b + 1) * s ...

speculative execution can hoist b * s and (b + 1) * s from then-blocks,
so that we have

  ... b * s ...
  if (...)
    ...
  ... (b + 1) * s ...
  if (...)
    ...

Then, SLSR can rewrite (b + 1) * s to (b * s + s) because after
speculative execution b * s dominates (b + 1) * s.

The performance impact of this change is significant. It speeds up the
benchmarks running EigenFloatContractionKernelInternal16x16
(ba68f42fa6/unsupported/Eigen/CXX11/src/Tensor/TensorContractionCuda.h?at=default#cl-526)
by roughly 2%. Some internal benchmarks that have the above code pattern
are improved by up to 40%. No significant slowdowns are observed on
Eigen CUDA microbenchmarks.

Reviewers: jholewinski, broune, eliben

Subscribers: llvm-commits, jholewinski

Differential Revision: http://reviews.llvm.org/D11201

llvm-svn: 242437
2015-07-16 20:13:48 +00:00
Benjamin Kramer 5dfd8b6da0 [NVPTX] Don't leak dead instructions after unlinking them from the BasicBlock
llvm-svn: 242417
2015-07-16 16:51:48 +00:00
Eli Bendersky f14af16219 Correct lowering of memmove in NVPTX
This fixes https://llvm.org/bugs/show_bug.cgi?id=24056

Also a bit of refactoring along the way.

Differential Revision: http://reviews.llvm.org/D11220

llvm-svn: 242413
2015-07-16 16:27:19 +00:00
Mehdi Amini bd7287ebe5 Move most user of TargetMachine::getDataLayout to the Module one
Summary:
This change is part of a series of commits dedicated to have a single
DataLayout during compilation by using always the one owned by the
module.

This patch is quite boring overall, except for some uglyness in
ASMPrinter which has a getDataLayout function but has some clients
that use it without a Module (llmv-dsymutil, llvm-dwarfdump), so
some methods are taking a DataLayout as parameter.

Reviewers: echristo

Subscribers: yaron.keren, rafael, llvm-commits, jholewinski

Differential Revision: http://reviews.llvm.org/D11090

From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 242386
2015-07-16 06:11:10 +00:00
Mehdi Amini 5c0fa58e91 Remove DataLayout from TargetLoweringObjectFile, redirect to Module
Summary:
This change is part of a series of commits dedicated to have a single
DataLayout during compilation by using always the one owned by the
module.

Reviewers: echristo

Subscribers: yaron.keren, rafael, llvm-commits, jholewinski

Differential Revision: http://reviews.llvm.org/D11079

From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 242385
2015-07-16 06:04:17 +00:00
Mark Heffernan 4c8ca53f7e Enable partial and runtime loop unrolling for NVPTX.
Enable partial and runtime loop unrolling for NVPTX backend via
TTI::UnrollingPreferences with a small threshold. This partially unrolls
small loops which are often unrolled by the PTX to SASS compiler
and unrolling earlier can be beneficial.

llvm-svn: 242049
2015-07-13 18:33:21 +00:00
Duncan P. N. Exon Smith 754e21f244 MC: Remove MCSubtargetInfo() default constructor
Force all creators of `MCSubtargetInfo` to immediately initialize it,
merging the default constructor and the initializer into an initializing
constructor.  Besides cleaning up the code a little, this makes it clear
that the initializer is never called again later.

Out-of-tree backends need a trivial change: instead of calling:

    auto *X = new MCSubtargetInfo();
    InitXYZMCSubtargetInfo(X, ...);
    return X;

they should call:

    return createXYZMCSubtargetInfoImpl(...);

There's no real functionality change here.

llvm-svn: 241957
2015-07-10 22:43:42 +00:00
Jingyue Wu a277561922 [TTI] BasicTTIImpl assumes no vector registers
Summary:
Following the discussion on r241884, it's more reasonable to assume that a
target has no vector registers by default instead of letting every such
target overrides getNumberOfRegisters.

Therefore, this patch modifies BasicTTIImpl::getNumberOfRegisters to
return 0 when Vector is true, and partially reverts r241884 which
modifies NVPTXTTIImpl::getNumberOfRegisters.

It also fixes a performance bug in LoopVectorizer. Even if a target has
no vector registers, vectorization may still help ILP. So, we need both
checks to be false before disabling loop vectorization all together.

Reviewers: hfinkel

Subscribers: llvm-commits, jholewinski

Differential Revision: http://reviews.llvm.org/D11108

llvm-svn: 241942
2015-07-10 21:14:54 +00:00
Eli Bendersky 5c0039a014 Actually support volatile memcpys in NVPTX lowering
Differential Revision: http://reviews.llvm.org/D11091

llvm-svn: 241914
2015-07-10 15:40:33 +00:00
Jingyue Wu ad85c8c204 [NVPTX] declare no vector registers
Summary:
Without this patch, LoopVectorizer in certain cases (see loop-vectorize.ll)
produces code with complex control flow which hurts later optimizations. Since
NVPTX doesn't have vector registers in LLVM's sense
(NVPTXTTI::getRegisterBitWidth(true) == 32), we for now declare no vector
registers to effectively disable loop vectorization.

Reviewers: jholewinski

Subscribers: jingyue, llvm-commits, jholewinski

Differential Revision: http://reviews.llvm.org/D11089

llvm-svn: 241884
2015-07-10 04:31:56 +00:00
Eli Bendersky d880520bc2 Replace index-loops by range-based loops
NFC

llvm-svn: 241875
2015-07-09 23:06:03 +00:00
Mehdi Amini eaabc51e78 Re-instate the EVT parameter to getScalarShiftAmountTy() for OOT user
A documentation for this function would be nice by the way.

From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 241807
2015-07-09 15:12:23 +00:00
Mehdi Amini 157e5a6d10 Remove getDataLayout() from TargetSelectionDAGInfo (had no users)
Summary:
Remove empty subclass in the process.

This change is part of a series of commits dedicated to have a single
DataLayout during compilation by using always the one owned by the
module.

Reviewers: echristo

Subscribers: jholewinski, llvm-commits, rafael, yaron.keren, ted

Differential Revision: http://reviews.llvm.org/D11045

From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 241780
2015-07-09 02:10:08 +00:00
Mehdi Amini a749f2ad47 Remove getDataLayout() from TargetLowering
Summary:
This change is part of a series of commits dedicated to have a single
DataLayout during compilation by using always the one owned by the
module.

Reviewers: echristo

Subscribers: yaron.keren, rafael, llvm-commits, jholewinski

Differential Revision: http://reviews.llvm.org/D11042

From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 241779
2015-07-09 02:09:52 +00:00
Mehdi Amini 0cdec1e2ab Make isLegalAddressingMode() taking DataLayout as an argument
Summary:
This change is part of a series of commits dedicated to have a single
DataLayout during compilation by using always the one owned by the
module.

Reviewers: echristo

Subscribers: jholewinski, llvm-commits, rafael, yaron.keren

Differential Revision: http://reviews.llvm.org/D11040

From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 241778
2015-07-09 02:09:40 +00:00
Mehdi Amini 9639d650bb Make TargetLowering::getShiftAmountTy() taking DataLayout as an argument
Summary:
This change is part of a series of commits dedicated to have a single
DataLayout during compilation by using always the one owned by the
module.

Reviewers: echristo

Subscribers: jholewinski, llvm-commits, rafael, yaron.keren

Differential Revision: http://reviews.llvm.org/D11037

From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 241776
2015-07-09 02:09:20 +00:00
Mehdi Amini 44ede33a69 Make TargetLowering::getPointerTy() taking DataLayout as an argument
Summary:
This change is part of a series of commits dedicated to have a single
DataLayout during compilation by using always the one owned by the
module.

Reviewers: echristo

Subscribers: jholewinski, ted, yaron.keren, rafael, llvm-commits

Differential Revision: http://reviews.llvm.org/D11028

From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 241775
2015-07-09 02:09:04 +00:00
Mehdi Amini 5010ebf181 Make TargetTransformInfo keeping a reference to the Module DataLayout
DataLayout is no longer optional. It was initialized with or without
a DataLayout, and the DataLayout when supplied could have been the
one from the TargetMachine.

Summary:
This change is part of a series of commits dedicated to have a single
DataLayout during compilation by using always the one owned by the
module.

Reviewers: echristo

Subscribers: jholewinski, llvm-commits, rafael, yaron.keren

Differential Revision: http://reviews.llvm.org/D11021

From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 241774
2015-07-09 02:08:42 +00:00
Mehdi Amini 56228dabfa Redirect DataLayout from TargetMachine to Module in ComputeValueVTs()
Summary:
Avoid using the TargetMachine owned DataLayout and use the Module owned
one instead. This requires passing the DataLayout up the stack to
ComputeValueVTs().

This change is part of a series of commits dedicated to have a single
DataLayout during compilation by using always the one owned by the
module.

Reviewers: echristo

Subscribers: jholewinski, yaron.keren, rafael, llvm-commits

Differential Revision: http://reviews.llvm.org/D11019

From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 241773
2015-07-09 01:57:34 +00:00
Eli Bendersky 8e131f8cbc Cosmetic cleanups - NFC
Remove commented lines, trailing whitespace, etc.

llvm-svn: 241687
2015-07-08 16:33:21 +00:00
Daniel Sanders f423f5627c Change the last few internal StringRef triples into Triple objects.
Summary:
This concludes the patch series to eliminate StringRef forms of GNU triples
from the internals of LLVM that began in r239036.

At this point, the StringRef-form of GNU Triples should only be used in the
public API (including IR serialization) and a couple objects that directly
interact with the API (most notably the Module class). The next step is to
replace these Triple objects with the TargetTuple object that will represent
our authoratative/unambiguous internal equivalent to GNU Triples.

Reviewers: rengolin

Subscribers: llvm-commits, jholewinski, ted, rengolin

Differential Revision: http://reviews.llvm.org/D10962

llvm-svn: 241472
2015-07-06 16:56:07 +00:00
Benjamin Kramer 9bfb627a0e [TargetLowering] StringRefize asm constraint getters.
There is some functional change here because it changes target code from
atoi(3) to StringRef::getAsInteger which has error checking. For valid
constraints there should be no difference.

llvm-svn: 241411
2015-07-05 19:29:18 +00:00
Jingyue Wu a0a56601c0 [NVPTX] expand extload/truncstore for vectors of floats
Summary:
According to PTX ISA:

For convenience, ld, st, and cvt instructions permit source and destination data operands to be wider than the instruction-type size, so that narrow values may be loaded, stored, and converted using regular-width registers. For example, 8-bit or 16-bit values may be held directly in 32-bit or 64-bit registers when being loaded, stored, or converted to other types and sizes. The operand type checking rules are relaxed for bit-size and integer (signed and unsigned) instruction types; floating-point instruction types still require that the operand type-size matches exactly, unless the operand is of bit-size type.

So, the ISA does not support load with extending/store with truncatation for floating numbers. This is reflected in setting the loadext/truncstore actions to expand in the code for floating numbers, but vectors of floating numbers are not taken care of.

As a result, loading a vector of floats followed by a fp_extend may be combined by DAGCombiner to a extload, and the extload may be lowered to NVPTXISD::LoadV2 with extending information. However, NVPTXISD::LoadV2 does not perform extending, and no extending instructions are inserted. Finally, PTX instructions with mismatched types are generated, like
ld.v2.f32 {%fd3, %fd4}, [%rd2]

This patch adds the correct actions for vectors of floats, so DAGCombiner would not create loads with extending, and correct code is generated.

Patched by Gang Hu. 

Test Plan: Test case attached.

Reviewers: jingyue

Reviewed By: jingyue

Subscribers: llvm-commits, jholewinski

Differential Revision: http://reviews.llvm.org/D10876

llvm-svn: 241191
2015-07-01 21:32:42 +00:00
Jingyue Wu 77b5b385ee [NVPTX] Move NVPTXPeephole after NVPTXPrologEpilogPass
Summary:
Offset of frame index is calculated by NVPTXPrologEpilogPass. Before
that the correct offset of stack objects cannot be obtained, which
leads to wrong offset if there are more than 2 frame objects. This patch
move NVPTXPeephole after NVPTXPrologEpilogPass. Because the frame index
is already replaced by %VRFrame in NVPTXPrologEpilogPass, we check
VRFrame register instead, and try to remove the VRFrame if there
is no usage after NVPTXPeephole pass.

Patched by Xuetian Weng. 

Test Plan:
Strengthened test/CodeGen/NVPTX/local-stack-frame.ll to check the
offset calculation based on SP and SPL.

Reviewers: jholewinski, jingyue

Reviewed By: jingyue

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D10853

llvm-svn: 241185
2015-07-01 20:08:06 +00:00
Jingyue Wu b374dc651c [NVPTX] cleanups and refacotring in NVPTXFrameLowering.cpp
Summary: NFC

Test Plan: no regression

Reviewers: wengxt

Reviewed By: wengxt

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D10849

llvm-svn: 241118
2015-06-30 21:28:31 +00:00
Jingyue Wu 9fe08c4bb3 [NVPTX] Fix issue introduced in D10321
Summary:
Really check if %SP is not used in other places, instead of checking only exact
one non-dbg use.

Patched by Xuetian Weng. 

Test Plan:
@foo4 in test/CodeGen/NVPTX/local-stack-frame.ll, create a case that
SP will appear twice.

Reviewers: jholewinski, jingyue

Reviewed By: jingyue

Subscribers: llvm-commits, sfantao, jholewinski

Differential Revision: http://reviews.llvm.org/D10844

llvm-svn: 241099
2015-06-30 18:59:19 +00:00
Samuel Antao 01ee64c2ea Force relocation mode to be default, regardless of what is passed to the backend.
llvm-svn: 241081
2015-06-30 17:18:00 +00:00
Jingyue Wu 3203818bf7 [NVPTX] noop when kernel pointers are already global
Summary:
Some front ends make kernel pointers global already. In that case,
handlePointerParams does nothing.

Test Plan: more tests in lower-kernel-ptr-arg.ll

Reviewers: grosser

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D10779

llvm-svn: 240849
2015-06-26 22:35:43 +00:00
Jingyue Wu 9c71150bfb Add NVPTXPeephole pass to reduce unnecessary address cast
Summary:
This patch first change the register that holds local address for stack
frame to %SPL. Then the new NVPTXPeephole pass will try to scan the
following pattern

   %vreg0<def> = LEA_ADDRi64 <fi#0>, 4
   %vreg1<def> = cvta_to_local %vreg0

and transform it into

   %vreg1<def> = LEA_ADDRi64 %VRFrameLocal, 4

Patched by Xuetian Weng

Test Plan: test/CodeGen/NVPTX/local-stack-frame.ll

Reviewers: jholewinski, jingyue

Reviewed By: jingyue

Subscribers: eliben, jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D10549

llvm-svn: 240587
2015-06-24 20:20:16 +00:00
Rafael Espindola c233f74e6e Simplify the Mangler interface now that DataLayout is mandatory.
We only need to pass in a DataLayout when mangling a raw string, not when
constructing the mangler.

llvm-svn: 240405
2015-06-23 13:59:29 +00:00
Alexander Kornienko f00654e31b Revert r240137 (Fixed/added namespace ending comments using clang-tidy. NFC)
Apparently, the style needs to be agreed upon first.

llvm-svn: 240390
2015-06-23 09:49:53 +00:00
Alexander Kornienko 70bc5f1398 Fixed/added namespace ending comments using clang-tidy. NFC
The patch is generated using this command:

tools/clang/tools/extra/clang-tidy/tool/run-clang-tidy.py -fix \
  -checks=-*,llvm-namespace-comment -header-filter='llvm/.*|clang/.*' \
  llvm/lib/


Thanks to Eugene Kosov for the original patch!

llvm-svn: 240137
2015-06-19 15:57:42 +00:00
Jingyue Wu cd3afea451 Add NVPTXLowerAlloca pass to convert alloca'ed memory to local address
Summary:
This is done by first adding two additional instructions to convert the
alloca returned address to local and convert it back to generic. Then
replace all uses of alloca instruction with the converted generic
address. Then we can rely NVPTXFavorNonGenericAddrSpace pass to combine
the generic addresscast and the corresponding Load, Store, Bitcast, GEP
Instruction together.

Patched by Xuetian Weng (xweng@google.com). 

Test Plan: test/CodeGen/NVPTX/lower-alloca.ll

Reviewers: jholewinski, jingyue

Reviewed By: jingyue

Subscribers: meheff, broune, eliben, jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D10483

llvm-svn: 239964
2015-06-17 22:31:02 +00:00
Daniel Sanders c81f450f1a Clean up redundant copies of Triple objects. NFC
Summary:

Reviewers: rengolin

Reviewed By: rengolin

Subscribers: llvm-commits, rengolin, jholewinski

Differential Revision: http://reviews.llvm.org/D10382

llvm-svn: 239823
2015-06-16 15:44:21 +00:00
Daniel Sanders 3e5de88dac Replace string GNU Triples with llvm::Triple in TargetMachine. NFC.
Summary:
For the moment, TargetMachine::getTargetTriple() still returns a StringRef.

This continues the patch series to eliminate StringRef forms of GNU triples
from the internals of LLVM that began in r239036.

Reviewers: rengolin

Reviewed By: rengolin

Subscribers: ted, llvm-commits, rengolin, jholewinski

Differential Revision: http://reviews.llvm.org/D10362

llvm-svn: 239554
2015-06-11 19:41:26 +00:00
Ahmed Bougacha c88bf54366 [CodeGen] ArrayRef'ize cond/pred in various TII APIs. NFC.
llvm-svn: 239553
2015-06-11 19:30:37 +00:00
Pete Cooper 7cbe58d3c5 Remove MachineModuleInfo::UsedFunctions as it has no users.
It hasn't been used since r130964.

This also removes MachineModuleInfo::isUsedFunction and
MachineModuleInfo::AnalyzeModule, both of which were only
there to support UsedFunctions.

llvm-svn: 239501
2015-06-11 01:04:56 +00:00
Daniel Sanders a73f1fdb19 Replace string GNU Triples with llvm::Triple in MCSubtargetInfo and create*MCSubtargetInfo(). NFC.
Summary:
This continues the patch series to eliminate StringRef forms of GNU triples
from the internals of LLVM that began in r239036.

Reviewers: rafael

Reviewed By: rafael

Subscribers: rafael, ted, jfb, llvm-commits, rengolin, jholewinski

Differential Revision: http://reviews.llvm.org/D10311

llvm-svn: 239467
2015-06-10 12:11:26 +00:00
Jingyue Wu 75589ffcc2 [NVPTX] fix a crash bug in NVPTXFavorNonGenericAddrSpaces
Summary:
We used to assume V->RAUW only modifies the operand list of V's user.
However, if V and V's user are Constants, RAUW may replace and invalidate V's
user entirely.

This patch fixes the above issue by letting the caller replace the
operand instead of calling RAUW on Constants.

Test Plan: @nested_const_expr and @rauw in access-non-generic.ll

Reviewers: broune, jholewinski

Reviewed By: broune, jholewinski

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D10345

llvm-svn: 239435
2015-06-09 21:50:32 +00:00
Samuel Antao cd50135a29 The constant initialization for globals in NVPTX is generated as an
array of bytes. The generation of this byte arrays was expecting 
the host to be little endian, which prevents big endian hosts to be 
used in the generation of the PTX code. This patch fixes the 
problem by changing the way the bytes are extracted so that it 
works for either little and big endian.

llvm-svn: 239412
2015-06-09 16:29:34 +00:00
Daniel Sanders 329fc9b68a [nvptx] Only support the 'm' inline assembly memory constraint. NFC.
Summary:
NVPTX doesn't seem to support any additional constraints. Therefore remove
the target hook.

No functional change intended.

Reviewers: jholewinski

Reviewed By: jholewinski

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D8209

llvm-svn: 239395
2015-06-09 10:34:05 +00:00
Matt Arsenault 8b643559d4 MC: Add target hook to control symbol quoting
llvm-svn: 239370
2015-06-09 00:31:39 +00:00
Jingyue Wu 2e4d1dd0ed [NVPTX] run SROA after NVPTXFavorNonGenericAddrSpaces
Summary:
This cleans up most allocas NVPTXLowerKernelArgs emits for byval
parameters.

Test Plan: makes bug21465.ll more stronger to verify no redundant local load/store.

Reviewers: eliben, jholewinski

Reviewed By: eliben, jholewinski

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D10322

llvm-svn: 239368
2015-06-09 00:05:56 +00:00
Jingyue Wu a2f6027a31 [NVPTX] roll forward r239082
NVPTXISelDAGToDAG translates "addrspacecast to param" to
NVPTX::nvvm_ptr_gen_to_param

Added an llc test in bug21465.

llvm-svn: 239100
2015-06-04 21:28:26 +00:00
Jingyue Wu b8f38668d5 Revert r239082
llc crashed for NVPTX backend

llvm-svn: 239094
2015-06-04 21:07:08 +00:00
Jingyue Wu f3a8079b75 [NVPTX] kernel pointer arguments point to the global address space
Summary:
With this patch, NVPTXLowerKernelArgs converts a kernel pointer argument to a
pointer in the global address space. This change, along with
NVPTXFavorNonGenericAddrSpaces, allows the NVPTX backend to emit ld.global.*
and st.global.* for accessing kernel pointer arguments.

Minor changes:
1. refactor: extract function convertToPointerInAddrSpace
2. fix a bug in the test case in bug21465.ll

Test Plan: lower-kernel-ptr-arg.ll

Reviewers: eliben, meheff, jholewinski

Reviewed By: jholewinski

Subscribers: wengxt, jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D10154

llvm-svn: 239082
2015-06-04 20:19:38 +00:00
Daniel Sanders 7813ae879e Replace string GNU Triples with llvm::Triple in MCAsmInfo subclasses and create*AsmInfo(). NFC.
Summary:
This is the first of several patches to eliminate StringRef forms of GNU
triples from the internals of LLVM. After this is complete, GNU triples
will be replaced by a more authoratitive representation in the form of
an LLVM TargetTuple.

Reviewers: rengolin

Reviewed By: rengolin

Subscribers: ted, llvm-commits, rengolin, jholewinski

Differential Revision: http://reviews.llvm.org/D10236

llvm-svn: 239036
2015-06-04 13:12:25 +00:00
Benjamin Kramer db220dbf02 Push constness through LoopInfo::isLoopHeader and clean it up a bit.
NFC.

llvm-svn: 238843
2015-06-02 15:28:27 +00:00
Matt Arsenault bd7d80a4a6 Add address space argument to isLegalAddressingMode
This is important because of different addressing modes
depending on the address space for GPU targets.

This only adds the argument, and does not update
any of the uses to provide the correct address space.

llvm-svn: 238723
2015-06-01 05:31:59 +00:00
Jim Grosbach 13760bd152 MC: Clean up MCExpr naming. NFC.
llvm-svn: 238634
2015-05-30 01:25:56 +00:00
Jingyue Wu 995dde2799 [NVPTXFavorNonGenericAddrSpaces] recursively trace into GEP and BitCast
Summary:
This patch allows NVPTXFavorNonGenericAddrSpaces to remove addrspacecast
from longer chains consisting of GEPs and BitCasts. For example, it can
now optimize

  %0 = addrspacecast [10 x float] addrspace(3)* @a to [10 x float]*
  %1 = gep [10 x float]* %0, i64 0, i64 %i
  %2 = bitcast float* %1 to i32*
  %3 = load i32* %2 ; emits ld.u32

to

  %0 = gep [10 x float] addrspace(3)* @a, i64 0, i64 %i
  %1 = bitcast float addrspace(3)* %0 to i32 addrspace(3)*
  %3 = load i32 addrspace(3)* %1 ; emits ld.shared.f32

Test Plan: @ld_int_from_global_float in access-non-generic.ll

Reviewers: broune, eliben, jholewinski, meheff

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D10074

llvm-svn: 238574
2015-05-29 17:00:27 +00:00
Benjamin Kramer dba7ee90b5 Don't call utostr in Twine/raw_ostream contexts.
Creating temporary std::strings there is unnecessary.

llvm-svn: 238412
2015-05-28 11:24:24 +00:00
Jingyue Wu c2a014697a [NaryReassociate] Run EarlyCSE after NaryReassociate
Summary:
This patch made two improvements to NaryReassociate and the NVPTX pipeline

1. Run EarlyCSE/GVN after NaryReassociate to get rid of redundant common
expressions.

2. When adding an instruction to SeenExprs, maps both the SCEV before and after
reassociation to that instruction.

Test Plan: updated @reassociate_gep_nsw in nary-gep.ll

Reviewers: meheff, broune

Reviewed By: broune

Subscribers: dberlin, jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D9947

llvm-svn: 238396
2015-05-28 04:56:52 +00:00
Rafael Espindola 0709a7bd1a Move alignment from MCSectionData to MCSection.
This starts merging MCSection and MCSectionData.

There are a few issues with the current split between MCSection and
MCSectionData.

* It optimizes the the not as important case. We want the production
of .o files to be really fast, but the split puts the information used
for .o emission in a separate data structure.

* The ELF/COFF/MachO hierarchy is not represented in MCSectionData,
leading to some ad-hoc ways to represent the various flags.

* It makes it harder to remember where each item is.

The attached patch starts merging the two by moving the alignment from
MCSectionData to MCSection.

Most of the patch is actually just dropping 'const', since
MCSectionData is mutable, but MCSection was not.

llvm-svn: 237936
2015-05-21 19:20:38 +00:00
Jim Grosbach 6f482000e9 MC: Clean up method names in MCContext.
The naming was a mish-mash of old and new style. Update to be consistent
with the new. NFC.

llvm-svn: 237594
2015-05-18 18:43:14 +00:00
Pete Cooper 81902a3ae4 Remove MCAssembler.h include from MCStreamer.h and fix users of MCStreamer.h
llvm-svn: 237483
2015-05-15 22:19:42 +00:00
Pete Cooper 3de83e4098 Remove 3 includes from MCInstrDesc.h and explicitly include them where needed
llvm-svn: 237481
2015-05-15 21:58:42 +00:00
Jim Grosbach 4c98cf77d9 MC: MCCodeGenInfo naming update. NFC.
s/InitMCCodeGenInfo/initMCCodeGenInfo/

llvm-svn: 237471
2015-05-15 19:13:31 +00:00
Jim Grosbach e9119e41ef MC: Modernize MCOperand API naming. NFC.
MCOperand::Create*() methods renamed to MCOperand::create*().

llvm-svn: 237275
2015-05-13 18:37:00 +00:00
Matthias Braun d04893fa36 Change getTargetNodeName() to produce compiler warnings for missing cases, fix them
llvm-svn: 236775
2015-05-07 21:33:59 +00:00
Quentin Colombet 61b305edfd [ShrinkWrap] Add (a simplified version) of shrink-wrapping.
This patch introduces a new pass that computes the safe point to insert the
prologue and epilogue of the function.
The interest is to find safe points that are cheaper than the entry and exits
blocks.

As an example and to avoid regressions to be introduce, this patch also
implements the required bits to enable the shrink-wrapping pass for AArch64.


** Context **

Currently we insert the prologue and epilogue of the method/function in the
entry and exits blocks. Although this is correct, we can do a better job when
those are not immediately required and insert them at less frequently executed
places.
The job of the shrink-wrapping pass is to identify such places.


** Motivating example **

Let us consider the following function that perform a call only in one branch of
a if:
define i32 @f(i32 %a, i32 %b)  {
 %tmp = alloca i32, align 4
 %tmp2 = icmp slt i32 %a, %b
 br i1 %tmp2, label %true, label %false

true:
 store i32 %a, i32* %tmp, align 4
 %tmp4 = call i32 @doSomething(i32 0, i32* %tmp)
 br label %false

false:
 %tmp.0 = phi i32 [ %tmp4, %true ], [ %a, %0 ]
 ret i32 %tmp.0
}

On AArch64 this code generates (removing the cfi directives to ease
readabilities):
_f:                                     ; @f
; BB#0:
  stp x29, x30, [sp, #-16]!
  mov  x29, sp
  sub sp, sp, #16             ; =16
  cmp  w0, w1
  b.ge  LBB0_2
; BB#1:                                 ; %true
  stur  w0, [x29, #-4]
  sub x1, x29, #4             ; =4
  mov  w0, wzr
  bl  _doSomething
LBB0_2:                                 ; %false
  mov  sp, x29
  ldp x29, x30, [sp], #16
  ret

With shrink-wrapping we could generate:
_f:                                     ; @f
; BB#0:
  cmp  w0, w1
  b.ge  LBB0_2
; BB#1:                                 ; %true
  stp x29, x30, [sp, #-16]!
  mov  x29, sp
  sub sp, sp, #16             ; =16
  stur  w0, [x29, #-4]
  sub x1, x29, #4             ; =4
  mov  w0, wzr
  bl  _doSomething
  add sp, x29, #16            ; =16
  ldp x29, x30, [sp], #16
LBB0_2:                                 ; %false
  ret

Therefore, we would pay the overhead of setting up/destroying the frame only if
we actually do the call.


** Proposed Solution **

This patch introduces a new machine pass that perform the shrink-wrapping
analysis (See the comments at the beginning of ShrinkWrap.cpp for more details).
It then stores the safe save and restore point into the MachineFrameInfo
attached to the MachineFunction.
This information is then used by the PrologEpilogInserter (PEI) to place the
related code at the right place. This pass runs right before the PEI.

Unlike the original paper of Chow from PLDI’88, this implementation of
shrink-wrapping does not use expensive data-flow analysis and does not need hack
to properly avoid frequently executed point. Instead, it relies on dominance and
loop properties.

The pass is off by default and each target can opt-in by setting the
EnableShrinkWrap boolean to true in their derived class of TargetPassConfig.
This setting can also be overwritten on the command line by using
-enable-shrink-wrap.

Before you try out the pass for your target, make sure you properly fix your
emitProlog/emitEpilog/adjustForXXX method to cope with basic blocks that are not
necessarily the entry block.


** Design Decisions **

1. ShrinkWrap is its own pass right now. It could frankly be merged into PEI but
for debugging and clarity I thought it was best to have its own file.
2. Right now, we only support one save point and one restore point. At some
point we can expand this to several save point and restore point, the impacted
component would then be:
- The pass itself: New algorithm needed.
- MachineFrameInfo: Hold a list or set of Save/Restore point instead of one
  pointer.
- PEI: Should loop over the save point and restore point.
Anyhow, at least for this first iteration, I do not believe this is interesting
to support the complex cases. We should revisit that when we motivating
examples.

Differential Revision: http://reviews.llvm.org/D9210

<rdar://problem/3201744>

llvm-svn: 236507
2015-05-05 17:38:16 +00:00
Duncan P. N. Exon Smith a9308c49ef IR: Give 'DI' prefix to debug info metadata
Finish off PR23080 by renaming the debug info IR constructs from `MD*`
to `DI*`.  The last of the `DIDescriptor` classes were deleted in
r235356, and the last of the related typedefs removed in r235413, so
this has all baked for about a week.

Note: If you have out-of-tree code (like a frontend), I recommend that
you get everything compiling and tests passing with the *previous*
commit before updating to this one.  It'll be easier to keep track of
what code is using the `DIDescriptor` hierarchy and what you've already
updated, and I think you're extremely unlikely to insert bugs.  YMMV of
course.

Back to *this* commit: I did this using the rename-md-di-nodes.sh
upgrade script I've attached to PR23080 (both code and testcases) and
filtered through clang-format-diff.py.  I edited the tests for
test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns
were off-by-three.  It should work on your out-of-tree testcases (and
code, if you've followed the advice in the previous paragraph).

Some of the tests are in badly named files now (e.g.,
test/Assembler/invalid-mdcompositetype-missing-tag.ll should be
'dicompositetype'); I'll come back and move the files in a follow-up
commit.

llvm-svn: 236120
2015-04-29 16:38:44 +00:00
Eric Christopher f4bf3779d8 Fix a [-Werror,-Winconsistent-missing-override] problem in the
NVPTX overrides.

llvm-svn: 236007
2015-04-28 18:06:27 +00:00
Justin Holewinski 3d2a976197 [NVPTX] Handle addrspacecast constant expressions in aggregate initializers
We need to track if an AddrSpaceCast expression was seen when
generating an MCExpr for a ConstantExpr.  This change introduces a
custom lowerConstant method to the NVPTX asm printer that will create
NVPTXGenericMCSymbolRefExpr nodes at the appropriate places to encode
the information that a given symbol needs to be casted to a generic
address.

llvm-svn: 236000
2015-04-28 17:18:30 +00:00
Sergey Dmitrouk 842a51bad8 Reapply r235977 "[DebugInfo] Add debug locations to constant SD nodes"
[DebugInfo] Add debug locations to constant SD nodes

This adds debug location to constant nodes of Selection DAG and updates
all places that create constants to pass debug locations
(see PR13269).

Can't guarantee that all locations are correct, but in a lot of cases choice
is obvious, so most of them should be. At least all tests pass.

Tests for these changes do not cover everything, instead just check it for
SDNodes, ARM and AArch64 where it's easy to get incorrect locations on
constants.

This is not complete fix as FastISel contains workaround for wrong debug
locations, which drops locations from instructions on processing constants,
but there isn't currently a way to use debug locations from constants there
as llvm::Constant doesn't cache it (yet). Although this is a bit different
issue, not directly related to these changes.

Differential Revision: http://reviews.llvm.org/D9084

llvm-svn: 235989
2015-04-28 14:05:47 +00:00
Daniel Jasper 48e93f7181 Revert "[DebugInfo] Add debug locations to constant SD nodes"
This breaks a test:
http://bb.pgr.jp/builders/cmake-llvm-x86_64-linux/builds/23870

llvm-svn: 235987
2015-04-28 13:38:35 +00:00
Sergey Dmitrouk adb4c69d5c [DebugInfo] Add debug locations to constant SD nodes
This adds debug location to constant nodes of Selection DAG and updates
all places that create constants to pass debug locations
(see PR13269).

Can't guarantee that all locations are correct, but in a lot of cases choice
is obvious, so most of them should be. At least all tests pass.

Tests for these changes do not cover everything, instead just check it for
SDNodes, ARM and AArch64 where it's easy to get incorrect locations on
constants.

This is not complete fix as FastISel contains workaround for wrong debug
locations, which drops locations from instructions on processing constants,
but there isn't currently a way to use debug locations from constants there
as llvm::Constant doesn't cache it (yet). Although this is a bit different
issue, not directly related to these changes.

Differential Revision: http://reviews.llvm.org/D9084

llvm-svn: 235977
2015-04-28 11:56:37 +00:00
Lang Hames 9ff69c8f4d [AsmPrinter] Make AsmPrinter's OutStreamer member a unique_ptr.
AsmPrinter owns the OutStreamer, so an owning pointer makes sense here. Using a
reference for this is crufty.

llvm-svn: 235752
2015-04-24 19:11:51 +00:00
Jingyue Wu 72fca6c89b Resurrect r235688
We should skip vector types which are not SCEVable.

test/CodeGen/NVPTX/sched2.ll passes

llvm-svn: 235695
2015-04-24 04:22:39 +00:00
Jingyue Wu 62af99b0db Revert r235688
Seems breaking builds

llvm-svn: 235690
2015-04-24 03:26:11 +00:00
Jingyue Wu 312fd0242d [NVPTX] Emits "generic()" depending on the original address space
Summary:
Fixes a bug in the NVPTX codegen. The code used to miss necessary "generic()"
on aggregates of addrspacecasts.

Test Plan: addrspacecast-gvar.ll

Reviewers: eliben, jholewinski

Reviewed By: jholewinski

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D9130

llvm-svn: 235689
2015-04-24 02:57:30 +00:00
Jingyue Wu 3daace5295 [NVPTX] enable NaryReassociate in NVPTX
Summary:
We run NaryReassociate right after SLSR because SLSR enables many
opportunities for NaryReassociate. For example, in nary-slsr.ll

  foo((a + b) + c);
  foo((a + b * 2) + c);
  foo((a + b * 3) + c);   // 2 muls and 6 adds

after SLSR:

  ab = a + b;
  foo(ab + c);
  ab2 = ab + b;
  foo(ab2 + c);
  ab3 = ab2 + b;
  foo(ab3 + c);           // 6 adds

after NaryReassociate:

  abc = (a + b) + c;
  foo(abc);
  ab2c = abc + b;
  foo(ab2c);
  ab3c = ab2c + b;
  foo(ab3c);              // 4 adds

Test Plan: nary-slsr.ll

Reviewers: jholewinski, eliben

Reviewed By: eliben

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D9066

llvm-svn: 235688
2015-04-24 02:54:06 +00:00
Jingyue Wu 3286ec1484 [NVPTX] run SeparateConstOffsetFromGEP before SLSR
Summary:
We pick this order because SeparateConstOffsetFromGEP may create more
opportunities for SLSR.

Test Plan:
reassociate-geps-and-slsr.ll
no performance regression on internal benchmarks

Reviewers: meheff

Subscribers: llvm-commits, jholewinski

Differential Revision: http://reviews.llvm.org/D9230

llvm-svn: 235632
2015-04-23 20:00:04 +00:00
Jingyue Wu 66a161f05e [NVPTX] do not run DCE after SLSR and SeparateConstOffsetFromGEP
Summary:
With D9096 and D9101, there's no need to run DCE after SLSR and
SeparateConstOffsetFromGEP.

Test Plan: no regression

Reviewers: jholewinski, meheff

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D9172

llvm-svn: 235415
2015-04-21 20:47:15 +00:00
Duncan P. N. Exon Smith b273d06b63 DebugInfo: Gut DIScope, DIEnumerator and DISubrange
The only class the still has API left is `DIDescriptor` itself.

llvm-svn: 235067
2015-04-16 01:37:00 +00:00
Duncan P. N. Exon Smith 35ef22cf53 DebugInfo: Gut DICompileUnit and DIFile
Continuing gutting `DIDescriptor` subclasses; this edition,
`DICompileUnit` and `DIFile`.  In the name of PR23080.

llvm-svn: 235055
2015-04-15 23:19:27 +00:00
Rafael Espindola 5560a4cfbd Use raw_pwrite_stream in the object writer/streamer.
The ELF object writer will take advantage of that in the next commit.

llvm-svn: 234950
2015-04-14 22:14:34 +00:00
Duncan P. N. Exon Smith 537b4a8159 DebugInfo: Gut DISubprogram and DILexicalBlock*
Gut the `DIDescriptor` wrappers around `MDLocalScope` subclasses.  Note
that `DILexicalBlock` wraps `MDLexicalBlockBase`, not `MDLexicalBlock`.

llvm-svn: 234850
2015-04-14 03:40:37 +00:00
Benjamin Kramer 619c4e57ba Reduce dyn_cast<> to isa<> or cast<> where possible.
No functional change intended.

llvm-svn: 234586
2015-04-10 11:24:51 +00:00
Jingyue Wu 5da831cc31 Divergence analysis for GPU programs
Summary:
Some optimizations such as jump threading and loop unswitching can negatively
affect performance when applied to divergent branches. The divergence analysis
added in this patch conservatively estimates which branches in a GPU program
can diverge. This information can then help LLVM to run certain optimizations
selectively.

Test Plan: test/Analysis/DivergenceAnalysis/NVPTX/diverge.ll

Reviewers: resistor, hfinkel, eliben, meheff, jholewinski

Subscribers: broune, bjarke.roune, madhur13490, tstellarAMD, dberlin, echristo, jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D8576

llvm-svn: 234567
2015-04-10 05:03:50 +00:00
Duncan P. N. Exon Smith e686f1591f CodeGen: Stop using DIDescriptor::is*() and auto-casting
Same as r234255, but for lib/CodeGen and lib/Target.

llvm-svn: 234258
2015-04-06 23:27:40 +00:00
Rafael Espindola 8ca44f0b5c Implement unique sections with an unique ID.
This allows the compiler/assembly programmer to switch back to a
section. This in turn fixes the bootstrap failure on powerpc (tested
on gcc110) without changing the ppc codegen at all.

I will try to cleanup the various getELFSection overloads in a  followup patch.
Just using a default argument now would lead to ambiguities.

llvm-svn: 234099
2015-04-04 18:02:01 +00:00
David Blaikie aa41cd57e0 [opaque pointer type] More GEP IRBuilder API migrations...
llvm-svn: 234058
2015-04-03 21:33:42 +00:00
David Blaikie 93c5444fe0 [opaque pointer type] More GEP API migrations in IRBuilder uses
The plan here is to push the API changes out from the common components
(like Constant::getGetElementPtr and IRBuilder::CreateGEP related
functions) and just update callers to either pass the type if it's
obvious, or pass null.

Do this with LoadInst as well and anything else that comes up, then to
start porting specific uses to not pass null anymore - this may require
some refactoring in each case.

llvm-svn: 234042
2015-04-03 19:41:44 +00:00
David Blaikie 4a2e73b066 [opaque pointer type] API migration for GEP constant factories
Require the pointee type to be passed explicitly and assert that it is
correct. For now it's possible to pass nullptr here (and I've done so in
a few places in this patch) but eventually that will be disallowed once
all clients have been updated or removed. It'll be a long road to get
all the way there... but if you have the cahnce to update your callers
to pass the type explicitly without depending on a pointer's element
type, that would be a good thing to do soon and a necessary thing to do
eventually.

llvm-svn: 233938
2015-04-02 18:55:32 +00:00
Eric Christopher f8019408dc Replace the MCSubtargetInfo parameter with a Triple when creating
an MCInstPrinter. Update all callers and use where we wanted a Triple
previously.

llvm-svn: 233648
2015-03-31 00:10:04 +00:00
Eric Christopher 48e697f7fe Remove unused MCSubtargetInfo argument from the NVPTX MCInstPrinter ctors.
llvm-svn: 233611
2015-03-30 22:03:16 +00:00
Eric Christopher c7c5592b7e Remove unused Target argument from MCInstPrinter ctor functions.
llvm-svn: 233607
2015-03-30 21:52:21 +00:00
Justin Holewinski 0afca26e90 [NVPTX] Associate a minimum PTX version for each SM architecture
When a new SM architecture is introduced, it is only supported by the
current PTX version and later.  Make sure we are using at least the
minimum PTX version for the target architecture.

This also removes support for PTX ISA < 3.2.

llvm-svn: 233583
2015-03-30 19:30:55 +00:00
Duncan P. N. Exon Smith 9dffcd04f7 CodeGen: Use the new DebugLoc API, NFC
Update lib/CodeGen (and lib/Target) to use the new `DebugLoc` API.

llvm-svn: 233582
2015-03-30 19:14:47 +00:00
Justin Holewinski f94d5b5137 [NVPTX] Add options for PTX 4.1/4.2 and SM 3.2/3.7/5.2/5.3
llvm-svn: 233575
2015-03-30 18:12:50 +00:00
Yaron Keren 075759aadd Remove more superfluous .str() and replace std::string concatenation with Twine.
Following r233392, http://llvm.org/viewvc/llvm-project?rev=233392&view=rev.

llvm-svn: 233555
2015-03-30 15:42:36 +00:00
Akira Hatanaka fb2289cb1b Delete MCInstPrinter::AvailableFeatures.
All the ports have been fixed to read the feature bits from the subtarget passed
to the print methods. Also, delete the call to setAvailableFeatures in the
constructor of NVPTX's instprinter as the instprinter wasn't using the feature
bits anywhere.

llvm-svn: 233486
2015-03-28 21:07:24 +00:00
Akira Hatanaka b46d0234a6 [MCInstPrinter] Enable MCInstPrinter to change its behavior based on the
per-function subtarget.

Currently, code-gen passes the default or generic subtarget to the constructors
of MCInstPrinter subclasses (see LLVMTargetMachine::addPassesToEmitFile), which
enables some targets (AArch64, ARM, and X86) to change their instprinter's
behavior based on the subtarget feature bits. Since the backend can now use
different subtargets for each function, instprinter has to be changed to use the
per-function subtarget rather than the default subtarget.

This patch takes the first step towards enabling instprinter to change its
behavior based on the per-function subtarget. It adds a bit "PassSubtarget" to
AsmWriter which tells table-gen to pass a reference to MCSubtargetInfo to the
various print methods table-gen auto-generates. 

I will follow up with changes to instprinters of AArch64, ARM, and X86.

llvm-svn: 233411
2015-03-27 20:36:02 +00:00
Rafael Espindola b61beca40c Close unique sections when switching away from them.
It is not possible to switch back to unique secitons, so close them
automatically when switching away.

llvm-svn: 233380
2015-03-27 15:01:40 +00:00
Andrew Kaylor 5c73e1f85c Disabling warnings for MSVC build to enable /W4 use.
Differential Revision: http://reviews.llvm.org/D8572

llvm-svn: 233133
2015-03-24 23:37:10 +00:00
David Blaikie 156d46eda0 Opaque Pointer Types: GEP API migrations to specify the gep type explicitly
The changes to InstCombine (& SCEV) do seem a bit silly - it doesn't make
anything obviously better to have the caller access the pointers element
type (the thing I'm trying to remove) than the GEP itself, but it's a
helpful migration step. This will allow me to more obviously lock down
GEP (& Load, etc) API usage, then fix all the code that accesses pointer
element types except the places that need to be removed (most of the
InstCombines) anyway - at which point I'll need to just remove all that
code because it won't be meaningful anymore (there will be no pointer
types, so no bitcasts to combine)

SCEV looks like it'll need some restructuring - we'll have to do a bit
more work for GEP canonicalization, since it'll depend on how it's used
if we can even manage to canonicalize it to a non-ugly GEP. I guess we
can do some fun stuff like voting (do 2 out of 3 load from the GEP with
a certain type that gives a pretty GEP? Does every typed use of the GEP
use either a specific type or a generic type (i8*, etc)?)

llvm-svn: 233131
2015-03-24 23:34:31 +00:00
Benjamin Kramer 799003bf8c Re-sort includes with sort-includes.py and insert raw_ostream.h where it's used.
llvm-svn: 232998
2015-03-23 19:32:43 +00:00
Eli Bendersky 3e84019a39 Simplify boolean expressions with true and false using clang-tidy
Patch by Richard (legalize@xmission.com)

Differential Revision: http://reviews.llvm.org/D8521

llvm-svn: 232961
2015-03-23 16:26:23 +00:00
Eric Christopher 4d0f35a901 Remove the target independent TargetMachine::getSubtarget and
TargetMachine::getSubtargetImpl routines.

This keeps the target independent code free of bare subtarget
calls while the remainder of the backends are migrated, or not
if they don't wish to support per-function subtargets as would
be needed for function multiversioning or LTO of disparate
cpu subarchitecture types, e.g.

clang -msse4.2 -c foo.c -emit-llvm -o foo.bc
clang -c bar.c -emit-llvm -o bar.bc
llvm-link foo.bc bar.bc -o baz.bc
llc baz.bc

and get appropriate code for what the command lines requested.

llvm-svn: 232885
2015-03-21 04:22:23 +00:00
Eric Christopher 5c3dffc459 Simplify the query for a subtarget in the NVPTX pass manager.
llvm-svn: 232876
2015-03-21 03:13:03 +00:00
Artem Belevich 9e8a039318 Add support for __nvvm_reflect changes in libdevice in CUDA-7.0
Summary:
CUDA 7.0's libdevice uses slightly different IR to call __nvvm_reflect
and that triggers an assertion in nvvm_reflect optimization pass. This
change allows nvvm_reflect pass to deal with both old and new ways to
pass an argument to __nvvm_reflect.

Test Plan: ninja check-all

Reviewers: eliben, echristo

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D8399

llvm-svn: 232732
2015-03-19 17:05:35 +00:00
Rafael Espindola 69244c3e78 two or more, use a for.
llvm-svn: 232688
2015-03-18 23:15:49 +00:00
David Blaikie 9f380a3ca0 Fix uses of reserved identifiers starting with an underscore followed by an uppercase letter
This covers essentially all of llvm's headers and libs. One or two weird
cases I wasn't sure were worth/appropriate to fix.

llvm-svn: 232394
2015-03-16 18:06:57 +00:00
Daniel Sanders bf5b80f5f9 Make each target map all inline assembly memory constraints to InlineAsm::Constraint_m. NFC.
Summary:
This is instead of doing this in target independent code and is the last
non-functional change before targets begin to distinguish between
different memory constraints when selecting code for the ISD::INLINEASM
node.

Next, each target will individually move away from the idea that all
memory constraints behave like 'm'.

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D8173

llvm-svn: 232373
2015-03-16 13:13:41 +00:00
David Blaikie 096b1da29d [opaque pointer type] more gep API migration
llvm-svn: 232274
2015-03-14 19:53:33 +00:00
Daniel Sanders 60f1db0525 Recommit r232027 with PR22883 fixed: Add infrastructure for support of multiple memory constraints.
The operand flag word for ISD::INLINEASM nodes now contains a 15-bit
memory constraint ID when the operand kind is Kind_Mem. This constraint
ID is a numeric equivalent to the constraint code string and is converted
with a target specific hook in TargetLowering.

This patch maps all memory constraints to InlineAsm::Constraint_m so there
is no functional change at this point. It just proves that using these
previously unused bits in the encoding of the flag word doesn't break
anything.

The next patch will make each target preserve the current mapping of
everything to Constraint_m for itself while changing the target independent
implementation of the hook to return Constraint_Unknown appropriately. Each
target will then be adapted in separate patches to use appropriate
Constraint_* values.

PR22883 was caused the matching operands copying the whole of the operand flags
for the matched operand. This included the constraint id which needed to be
replaced with the operand number. This has been fixed with a conversion
function. Following on from this, matching operands also used the operand
number as the constraint id. This has been fixed by looking up the matched
operand and taking it from there. 

llvm-svn: 232165
2015-03-13 12:45:09 +00:00
Hal Finkel e78e52ba9b Revert "r232027 - Add infrastructure for support of multiple memory constraints"
This (r232027) has caused PR22883; so it seems those bits might be used by
something else after all. Reverting until we can figure out what else to do.

Original commit message:

The operand flag word for ISD::INLINEASM nodes now contains a 15-bit
memory constraint ID when the operand kind is Kind_Mem. This constraint
ID is a numeric equivalent to the constraint code string and is converted
with a target specific hook in TargetLowering.

This patch maps all memory constraints to InlineAsm::Constraint_m so there
is no functional change at this point. It just proves that using these
previously unused bits in the encoding of the flag word doesn't break anything.

The next patch will make each target preserve the current mapping of
everything to Constraint_m for itself while changing the target independent
implementation of the hook to return Constraint_Unknown appropriately. Each
target will then be adapted in separate patches to use appropriate Constraint_*
values.

llvm-svn: 232093
2015-03-12 20:09:39 +00:00
Daniel Sanders 41c072e63b Add infrastructure for support of multiple memory constraints.
Summary:
The operand flag word for ISD::INLINEASM nodes now contains a 15-bit
memory constraint ID when the operand kind is Kind_Mem. This constraint
ID is a numeric equivalent to the constraint code string and is converted
with a target specific hook in TargetLowering.

This patch maps all memory constraints to InlineAsm::Constraint_m so there
is no functional change at this point. It just proves that using these
previously unused bits in the encoding of the flag word doesn't break anything.

The next patch will make each target preserve the current mapping of
everything to Constraint_m for itself while changing the target independent
implementation of the hook to return Constraint_Unknown appropriately. Each
target will then be adapted in separate patches to use appropriate Constraint_*
values.

Reviewers: hfinkel

Reviewed By: hfinkel

Subscribers: hfinkel, jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D8171

llvm-svn: 232027
2015-03-12 11:00:48 +00:00
Jingyue Wu e8290f21b5 [NVPTXAsmPrinter] do not print .align on function headers
Summary:
PTX does not allow .align directives on function headers.

Fixes PR21551.

Test Plan: test/Codegen/NVPTX/function-align.ll

Reviewers: eliben, jholewinski

Reviewed By: eliben, jholewinski

Subscribers: llvm-commits, eliben, jpienaar, jholewinski

Differential Revision: http://reviews.llvm.org/D8274

llvm-svn: 232004
2015-03-12 01:50:30 +00:00
Mehdi Amini 93e1ea167e Move the DataLayout to the generic TargetMachine, making it mandatory.
Summary:
I don't know why every singled backend had to redeclare its own DataLayout.
There was a virtual getDataLayout() on the common base TargetMachine, the
default implementation returned nullptr. It was not clear from this that
we could assume at call site that a DataLayout will be available with
each Target.

Now getDataLayout() is no longer virtual and return a pointer to the
DataLayout member of the common base TargetMachine. I plan to turn it into
a reference in a future patch.

The only backend that didn't have a DataLayout previsouly was the CPPBackend.
It now initializes the default DataLayout. This commit is NFC for all the
other backends.

Test Plan: clang+llvm ninja check-all

Reviewers: echristo

Subscribers: jfb, jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D8243

From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 231987
2015-03-12 00:07:24 +00:00
Eric Christopher 7af9528747 Have getCalleeSavedRegs take a non-null MachineFunction all the
time. The target independent code was passing in one all the
time and targets weren't checking validity before using. Update
a few calls to pass in a MachineFunction where necessary.

llvm-svn: 231970
2015-03-11 21:41:28 +00:00
Rafael Espindola 6b9998b3eb Create symbols marking the start of a section earlier.
This lets us pass the symbol to the constructor and avoid the mutable field.

This also opens the way for outputting the symbol only when needed, instead
of outputting them at the start of the file.

llvm-svn: 231859
2015-03-10 22:00:25 +00:00
Benjamin Kramer d58792f38b Don't use LLVM_LIBRARY_VISIBILITY in cpp files.
llvm-svn: 231831
2015-03-10 20:07:44 +00:00
Benjamin Kramer 414c0964c3 NVPTX: move NVPTXAllocaHoisting into the cpp file
Also initialize without using static initialization.

llvm-svn: 231822
2015-03-10 19:20:52 +00:00
Benjamin Kramer 42a31e1f5e NVPTX: Remove copy of LLVMInitializeNVPTXAsmPrinter.
If anyone is using this for some strange reason,
LLVMInitializeNVPTXAsmPrinter does exactly the same thing and is what
other LLVM tools are calling.

llvm-svn: 231810
2015-03-10 18:19:24 +00:00
Rafael Espindola 760bf9520a Remove incredibly confusing isBaseAddressKnownZero.
When referring to a symbol in a dwarf section on ELF we should use

.long foo

instead of

.long foo - .debug_something

because ELF is unaware of the content of the sections and therefore needs
relocations. This has nothing to do with optimizing a -0.

llvm-svn: 231751
2015-03-10 04:11:52 +00:00
Rafael Espindola fcc2821882 Use a better name for compile unit labels.
They mark the start of a compile unit, so name them .Lcu_*. Using
Section->getLabelBeginName() makes it looks like they mark the start of the
section.

While at it, switch to createTempSymbol to avoid collisions with labels
created in inline assembly. Not sure if a "don't crash" test is worth it.

With this getLabelBeginName is dead, delete it.

llvm-svn: 231750
2015-03-10 03:58:36 +00:00
Mehdi Amini a28d91d81b DataLayout is mandatory, update the API to reflect it with references.
Summary:
Now that the DataLayout is a mandatory part of the module, let's start
cleaning the codebase. This patch is a first attempt at doing that.

This patch is not exactly NFC as for instance some places were passing
a nullptr instead of the DataLayout, possibly just because there was a
default value on the DataLayout argument to many functions in the API.
Even though it is not purely NFC, there is no change in the
validation.

I turned as many pointer to DataLayout to references, this helped
figuring out all the places where a nullptr could come up.

I had initially a local version of this patch broken into over 30
independant, commits but some later commit were cleaning the API and
touching part of the code modified in the previous commits, so it
seemed cleaner without the intermediate state.

Test Plan:

Reviewers: echristo

Subscribers: llvm-commits

From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 231740
2015-03-10 02:37:25 +00:00
Reid Kleckner 294fa7aa9d TableGen: Use 'enum : uint64_t' for feature flags to fix -Wmicrosoft
clang-cl would warn that this value is not representable in 'int':
  enum { FeatureX = 1ULL << 31 };
All MS enums are 'ints' unless otherwise specified, so we have to use an
explicit type.  The AMDGPU target just hit 32 features, triggering this
warning.

Now that we have C++11 strong enum types, we can also eliminate the
'const uint64_t' codepath from tablegen and just use 'enum : uint64_t'.

llvm-svn: 231697
2015-03-09 20:23:14 +00:00
Rafael Espindola 028f8b43e2 Delete dead code. NFC.
llvm-svn: 231682
2015-03-09 18:48:29 +00:00
Benjamin Kramer a52f696e6f Move unreferenced passes into the cpp file
NFC.

llvm-svn: 231661
2015-03-09 15:50:58 +00:00
David Blaikie dc3f01e9cf Simplify expressions involving boolean constants with clang-tidy
Patch by Richard (legalize at xmission dot com).

Differential Revision: http://reviews.llvm.org/D8154

llvm-svn: 231617
2015-03-09 01:57:13 +00:00
Mehdi Amini 46a43556db Make DataLayout Non-Optional in the Module
Summary:
DataLayout keeps the string used for its creation.

As a side effect it is no longer needed in the Module.
This is "almost" NFC, the string is no longer
canonicalized, you can't rely on two "equals" DataLayout
having the same string returned by getStringRepresentation().

Get rid of DataLayoutPass: the DataLayout is in the Module

The DataLayout is "per-module", let's enforce this by not
duplicating it more than necessary.
One more step toward non-optionality of the DataLayout in the
module.

Make DataLayout Non-Optional in the Module

Module->getDataLayout() will never returns nullptr anymore.

Reviewers: echristo

Subscribers: resistor, llvm-commits, jholewinski

Differential Revision: http://reviews.llvm.org/D7992

From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 231270
2015-03-04 18:43:29 +00:00
Paul Robinson c583abc888 Remove useless .debug_macinfo section setup.
llvm-svn: 231001
2015-03-02 19:52:42 +00:00
Benjamin Kramer e86b0f186b NVPTX: Remove dead code.
Fun fact: This file was never referenced since the initial checkin of
the NVPTX backend.

llvm-svn: 230957
2015-03-02 13:16:28 +00:00
Benjamin Kramer 5fbfe2ffdc Convert push_back loops into append calls.
No functionality change intended.

llvm-svn: 230849
2015-02-28 13:20:15 +00:00
Eric Christopher 11e4df73c8 getRegForInlineAsmConstraint wants to use TargetRegisterInfo for
a lookup, pass that in rather than use a naked call to getSubtargetImpl.
This involved passing down and around either a TargetMachine or
TargetRegisterInfo. Update all callers/definitions around the targets
and SelectionDAG.

llvm-svn: 230699
2015-02-26 22:38:43 +00:00
Eric Christopher 23a3a7c871 Remove an argument-less call to getSubtargetImpl from TargetLoweringBase.
This required plumbing a TargetRegisterInfo through computeRegisterProperties
and into findRepresentativeClass which uses it for register class
iteration. This required passing a subtarget into a few target specific
initializations of TargetLowering.

llvm-svn: 230583
2015-02-26 00:00:24 +00:00
Benjamin Kramer ea68a944a1 Demote vectors to arrays. No functionality change.
llvm-svn: 229861
2015-02-19 15:26:17 +00:00
Eric Christopher ca929f2469 Avoid using a self-referential initializer and fix up uses.
llvm-svn: 229790
2015-02-19 00:22:47 +00:00
Eric Christopher 02389e3886 Remove all use of is64bit off of NVPTXSubtarget and clean up code
accordingly. This changes the constructors of a number of classes
that don't need to know the subtarget's 64-bitness.

llvm-svn: 229787
2015-02-19 00:08:27 +00:00
Eric Christopher beffc4e84f Remove all use of getDrvInterface off of NVPTXSubtarget and clean
up code accordingly. Delete code that was checking for all cases
of an enum.

llvm-svn: 229786
2015-02-19 00:08:23 +00:00
Eric Christopher 6aad8b1801 Migrate the NVPTX backend asm printer to a per function subtarget.
This involved moving two non-subtarget dependent features (64-bitness
and the driver interface) to the NVPTX target machine and updating
the uses (or migrating around the subtarget use for ease of review).
Otherwise use the cached subtarget or create a default subtarget
based on the TargetMachine cpu and feature string for the module
level assembler emission.

llvm-svn: 229785
2015-02-19 00:08:14 +00:00
Duncan P. N. Exon Smith b5054333ec NVPTX: Canonicalize access to function attributes, NFC
Canonicalize access to function attributes to use the simpler API.

getAttributes().getAttribute(AttributeSet::FunctionIndex, Kind)
  => getFnAttribute(Kind)

getAttributes().hasAttribute(AttributeSet::FunctionIndex, Kind)
  => hasFnAttribute(Kind)

llvm-svn: 229260
2015-02-14 15:35:43 +00:00
Chandler Carruth 30d69c2e36 [PM] Remove the old 'PassManager.h' header file at the top level of
LLVM's include tree and the use of using declarations to hide the
'legacy' namespace for the old pass manager.

This undoes the primary modules-hostile change I made to keep
out-of-tree targets building. I sent an email inquiring about whether
this would be reasonable to do at this phase and people seemed fine with
it, so making it a reality. This should allow us to start bootstrapping
with modules to a certain extent along with making it easier to mix and
match headers in general.

The updates to any code for users of LLVM are very mechanical. Switch
from including "llvm/PassManager.h" to "llvm/IR/LegacyPassManager.h".
Qualify the types which now produce compile errors with "legacy::". The
most common ones are "PassManager", "PassManagerBase", and
"FunctionPassManager".

llvm-svn: 229094
2015-02-13 10:01:29 +00:00
Benjamin Kramer 5f6a907288 MathExtras: Bring Count(Trailing|Leading)Ones and CountPopulation in line with countTrailingZeros
Update all callers.

llvm-svn: 228930
2015-02-12 15:35:40 +00:00
Jingyue Wu d7966ff3b9 Add straight-line strength reduction to LLVM
Summary:
Straight-line strength reduction (SLSR) is implemented in GCC but not yet in
LLVM. It has proven to effectively simplify statements derived from an unrolled
loop, and can potentially benefit many other cases too. For example,

LLVM unrolls

  #pragma unroll
  foo (int i = 0; i < 3; ++i) {
    sum += foo((b + i) * s);
  }

into

  sum += foo(b * s);
  sum += foo((b + 1) * s);
  sum += foo((b + 2) * s);

However, no optimizations yet reduce the internal redundancy of the three
expressions:

  b * s
  (b + 1) * s
  (b + 2) * s

With SLSR, LLVM can optimize these three expressions into:

  t1 = b * s
  t2 = t1 + s
  t3 = t2 + s

This commit is only an initial step towards implementing a series of such
optimizations. I will implement more (see TODO in the file commentary) in the
near future. This optimization is enabled for the NVPTX backend for now.
However, I am more than happy to push it to the standard optimization pipeline
after more thorough performance tests.

Test Plan: test/StraightLineStrengthReduce/slsr.ll

Reviewers: eliben, HaoLiu, meheff, hfinkel, jholewinski, atrick

Reviewed By: jholewinski, atrick

Subscribers: karthikthecool, jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D7310

llvm-svn: 228016
2015-02-03 19:37:06 +00:00
Jingyue Wu 5bbcdaa8d9 Remove usernames from TODOs, NFC
making the style consistent with the rest

llvm-svn: 227991
2015-02-03 17:57:38 +00:00
Jingyue Wu 49a766e468 Resurrect the assertion removed by r227717
Summary: MSVC can compile "LoopID->getOperand(0) == LoopID" when LoopID is MDNode*.

Test Plan: no regression

Reviewers: mkuper

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D7327

llvm-svn: 227853
2015-02-02 20:41:11 +00:00
Chandler Carruth c956ab6603 [multiversion] Switch the TTI queries from TargetMachine to Subtarget
now that we have a correct and cached subtarget specific to the
function.

Also, finish providing a cached per-function subtarget in the core
LLVMTargetMachine -- that layer hadn't switched over yet.

The only use of the TargetMachine was to re-lookup a subtarget for
a particular function to work around the fact that TTI was immutable.
Now that it is per-function and we haved a cached subtarget, use it.

This still leaves a few interfaces with real warts on them where we were
passing Function objects through the TTI interface. I'll remove these
and clean their usage up in subsequent commits now that this isn't
necessary.

llvm-svn: 227738
2015-02-01 14:22:17 +00:00
Chandler Carruth c340ca839c [multiversion] Remove the cached TargetMachine pointer from the
intermediate TTI implementation template and instead query up to the
derived class for both the TargetMachine and the TargetLowering.

Most of the derived types had a TLI cached already and there is no need
to store a less precisely typed target machine pointer.

This will in turn make it much cleaner to look up the TLI via
a per-function subtarget instead of the generic subtarget, and it will
pave the way toward pulling the subtarget used for unroll preferences
into the same form once we are *always* using the function to look up
the correct subtarget.

llvm-svn: 227737
2015-02-01 14:01:15 +00:00
Chandler Carruth 8b04c0d26a [multiversion] Switch all of the targets over to use the
TargetIRAnalysis access path directly rather than implementing getTTI.

This even removes getTTI from the interface. It's more efficient for
each target to just register a precise callback that creates their
specific TTI.

As part of this, all of the targets which are building their subtargets
individually per-function now build their TTI instance with the function
and thus look up the correct subtarget and cache it. NVPTX, R600, and
XCore currently don't leverage this functionality, but its trivial for
them to add it now.

llvm-svn: 227735
2015-02-01 13:20:00 +00:00
Chandler Carruth ee642690ea [multiversion] Remove a false freedom to leave the TargetMachine pointer
null.

For some reason some of the original TTI code supported a null target
machine. This seems to have been legacy, and I made matters worse when
refactoring this code by spreading that pattern further through the
various targets.

The TargetMachine can't actually be null, and it doesn't make sense to
support that use case. I've now consistently removed it and removed all
of the code trying to cope with that situation. This is probably good,
as several targets *didn't* cope with it being null despite the null
default argument in their constructors. =]

llvm-svn: 227734
2015-02-01 12:38:24 +00:00
Jingyue Wu 0220df0dfd [NVPTX] Emit .pragma "nounroll" for loops marked with nounroll
Summary:
CUDA driver can unroll loops when jit-compiling PTX. To prevent CUDA
driver from unrolling a loop marked with llvm.loop.unroll.disable is not
unrolled by CUDA driver, we need to emit .pragma "nounroll" at the
header of that loop.

This patch also extracts getting unroll metadata from loop ID metadata
into a shared helper function.

Test Plan: test/CodeGen/NVPTX/nounroll.ll

Reviewers: eliben, meheff, jholewinski

Reviewed By: jholewinski

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D7041

llvm-svn: 227703
2015-02-01 02:27:45 +00:00
Chandler Carruth d8b3e9a420 [PM] Remove a bunch of stale TTI creation method declarations. I nuked
their definitions, but forgot to clean up all the declarations which are
in different files.

llvm-svn: 227698
2015-02-01 00:22:15 +00:00
Chandler Carruth 93dcdc47db [PM] Switch the TargetMachine interface from accepting a pass manager
base which it adds a single analysis pass to, to instead return the type
erased TargetTransformInfo object constructed for that TargetMachine.

This removes all of the pass variants for TTI. There is now a single TTI
*pass* in the Analysis layer. All of the Analysis <-> Target
communication is through the TTI's type erased interface itself. While
the diff is large here, it is nothing more that code motion to make
types available in a header file for use in a different source file
within each target.

I've tried to keep all the doxygen comments and file boilerplate in line
with this move, but let me know if I missed anything.

With this in place, the next step to making TTI work with the new pass
manager is to introduce a really simple new-style analysis that produces
a TTI object via a callback into this routine on the target machine.
Once we have that, we'll have the building blocks necessary to accept
a function argument as well.

llvm-svn: 227685
2015-01-31 11:17:59 +00:00
Chandler Carruth 705b185f90 [PM] Change the core design of the TTI analysis to use a polymorphic
type erased interface and a single analysis pass rather than an
extremely complex analysis group.

The end result is that the TTI analysis can contain a type erased
implementation that supports the polymorphic TTI interface. We can build
one from a target-specific implementation or from a dummy one in the IR.

I've also factored all of the code into "mix-in"-able base classes,
including CRTP base classes to facilitate calling back up to the most
specialized form when delegating horizontally across the surface. These
aren't as clean as I would like and I'm planning to work on cleaning
some of this up, but I wanted to start by putting into the right form.

There are a number of reasons for this change, and this particular
design. The first and foremost reason is that an analysis group is
complete overkill, and the chaining delegation strategy was so opaque,
confusing, and high overhead that TTI was suffering greatly for it.
Several of the TTI functions had failed to be implemented in all places
because of the chaining-based delegation making there be no checking of
this. A few other functions were implemented with incorrect delegation.
The message to me was very clear working on this -- the delegation and
analysis group structure was too confusing to be useful here.

The other reason of course is that this is *much* more natural fit for
the new pass manager. This will lay the ground work for a type-erased
per-function info object that can look up the correct subtarget and even
cache it.

Yet another benefit is that this will significantly simplify the
interaction of the pass managers and the TargetMachine. See the future
work below.

The downside of this change is that it is very, very verbose. I'm going
to work to improve that, but it is somewhat an implementation necessity
in C++ to do type erasure. =/ I discussed this design really extensively
with Eric and Hal prior to going down this path, and afterward showed
them the result. No one was really thrilled with it, but there doesn't
seem to be a substantially better alternative. Using a base class and
virtual method dispatch would make the code much shorter, but as
discussed in the update to the programmer's manual and elsewhere,
a polymorphic interface feels like the more principled approach even if
this is perhaps the least compelling example of it. ;]

Ultimately, there is still a lot more to be done here, but this was the
huge chunk that I couldn't really split things out of because this was
the interface change to TTI. I've tried to minimize all the other parts
of this. The follow up work should include at least:

1) Improving the TargetMachine interface by having it directly return
   a TTI object. Because we have a non-pass object with value semantics
   and an internal type erasure mechanism, we can narrow the interface
   of the TargetMachine to *just* do what we need: build and return
   a TTI object that we can then insert into the pass pipeline.
2) Make the TTI object be fully specialized for a particular function.
   This will include splitting off a minimal form of it which is
   sufficient for the inliner and the old pass manager.
3) Add a new pass manager analysis which produces TTI objects from the
   target machine for each function. This may actually be done as part
   of #2 in order to use the new analysis to implement #2.
4) Work on narrowing the API between TTI and the targets so that it is
   easier to understand and less verbose to type erase.
5) Work on narrowing the API between TTI and its clients so that it is
   easier to understand and less verbose to forward.
6) Try to improve the CRTP-based delegation. I feel like this code is
   just a bit messy and exacerbating the complexity of implementing
   the TTI in each target.

Many thanks to Eric and Hal for their help here. I ended up blocked on
this somewhat more abruptly than I expected, and so I appreciate getting
it sorted out very quickly.

Differential Revision: http://reviews.llvm.org/D7293

llvm-svn: 227669
2015-01-31 03:43:40 +00:00
Eric Christopher 3084aff498 Migrate a bare getSubtarget call to query the MachineFunction
for the target dependent one.

llvm-svn: 227542
2015-01-30 01:50:09 +00:00
Eric Christopher bef0a373b4 Migrate NVPTXISelLowering to take the subtarget that it's dependent
upon as an argument and store/use that in the entire function.

llvm-svn: 227541
2015-01-30 01:50:07 +00:00
Eric Christopher 9745b3aae0 Remove unused argument.
llvm-svn: 227539
2015-01-30 01:41:01 +00:00
Eric Christopher 147bba2385 Migrate NVPTXISelDAGToDAG's getSubtarget to a runOnMachineFunction
version. Update NVPTXInstrInfo accordingly.

llvm-svn: 227538
2015-01-30 01:40:59 +00:00
Chandler Carruth b81dfa6378 [LPM] Stop using the string based preservation API. It is an
abomination.

For starters, this API is incredibly slow. In order to lookup the name
of a pass it must take a memory fence to acquire a pointer to the
managed static pass registry, and then potentially acquire locks while
it consults this registry for information about what passes exist by
that name. This stops the world of LLVMs in your process no matter
how little they cared about the result.

To make this more joyful, you'll note that we are preserving many passes
which *do not exist* any more, or are not even analyses which one might
wish to have be preserved. This means we do all the work only to say
"nope" with no error to the user.

String-based APIs are a *bad idea*. String-based APIs that cannot
produce any meaningful error are an even worse idea. =/

I have a patch that simply removes this API completely, but I'm hesitant
to commit it as I don't really want to perniciously break out-of-tree
users of the old pass manager. I'd rather they just have to migrate to
the new one at some point. If others disagree and would like me to kill
it with fire, just say the word. =]

llvm-svn: 227294
2015-01-28 04:57:56 +00:00
Justin Holewinski d4d2e9bd0e [NVPTX] Generate a more optimal sequence for select of i1
Instead of creating a pattern like "(p && a) || ((!p) && b)",
just expand the i8 operands to i32 and perform the selp on them.

Fixes PR22246

llvm-svn: 227123
2015-01-26 19:52:20 +00:00
Justin Holewinski 23df659e6d [NVPTX] Handle floating-point conversion patterns that are not explicitly ordered or unordered
Fixes PR22322

llvm-svn: 227117
2015-01-26 19:11:20 +00:00
Eric Christopher 8b7706517c Move DataLayout back to the TargetMachine from TargetSubtargetInfo
derived classes.

Since global data alignment, layout, and mangling is often based on the
DataLayout, move it to the TargetMachine. This ensures that global
data is going to be layed out and mangled consistently if the subtarget
changes on a per function basis. Prior to this all targets(*) have
had subtarget dependent code moved out and onto the TargetMachine.

*One target hasn't been migrated as part of this change: R600. The
R600 port has, as a subtarget feature, the size of pointers and
this affects global data layout. I've currently hacked in a FIXME
to enable progress, but the port needs to be updated to either pass
the 64-bitness to the TargetMachine, or fix the DataLayout to
avoid subtarget dependent features.

llvm-svn: 227113
2015-01-26 19:03:15 +00:00
David Blaikie 9459832ebd std::unique_ptrify the MCStreamer argument to createAsmPrinter
llvm-svn: 226414
2015-01-18 20:29:04 +00:00
NAKAMURA Takumi a50a89a081 Update libdeps in NVPTXCodeGen, since r225944.
llvm-svn: 226055
2015-01-14 23:01:36 +00:00
Olivier Sallenave c8d13bd370 Override the TLI callback enableAggressiveFMAFusion and return true. Indeed, fmul, fmadd and fadd nodes cost the same number of cycles, so we can enable more combining heuristics to produce more fmadd nodes.
llvm-svn: 225984
2015-01-14 14:47:24 +00:00
Chandler Carruth d9903888d9 [cleanup] Re-sort all the #include lines in LLVM using
utils/sort_includes.py.

I clearly haven't done this in a while, so more changed than usual. This
even uncovered a missing include from the InstrProf library that I've
added. No functionality changed here, just mechanical cleanup of the
include order.

llvm-svn: 225974
2015-01-14 11:23:27 +00:00
Duncan P. N. Exon Smith 9f6bddd4b2 NVPTX: Use MapMetadata() instead of custom/stale/untested logic
Copy the `GVMap` over to a standard `ValueToValueMapTy` so that we can
reuse the `MapMetadata()` logic.  Unfortunately the `GVMap` can't just
be replaced, since `MapMetadata()` likes to modify the map, but at least
this will prevent NVPTX from bitrotting.

llvm-svn: 225944
2015-01-14 05:14:30 +00:00
Duncan P. N. Exon Smith f864ae2745 NVPTX: Remove bogus remap logic for global variable address spaces
The comment is incorrect, and the code mangles debug info.  Remove the
bad logic, which wasn't tested anyway.

llvm-svn: 225943
2015-01-14 05:13:18 +00:00
Ahmed Bougacha 2b6917b020 [SelectionDAG] Allow targets to specify legality of extloads' result
type (in addition to the memory type).

The *LoadExt* legalization handling used to only have one type, the
memory type.  This forced users to assume that as long as the extload
for the memory type was declared legal, and the result type was legal,
the whole extload was legal.

However, this isn't always the case.  For instance, on X86, with AVX,
this is legal:
    v4i32 load, zext from v4i8
but this isn't:
    v4i64 load, zext from v4i8
Whereas v4i64 is (arguably) legal, even without AVX2.

Note that the same thing was done a while ago for truncstores (r46140),
but I assume no one needed it yet for extloads, so here we go.

Calls to getLoadExtAction were changed to add the value type, found
manually in the surrounding code.

Calls to setLoadExtAction were mechanically changed, by wrapping the
call in a loop, to match previous behavior.  The loop iterates over
the MVT subrange corresponding to the memory type (FP vectors, etc...).
I also pulled neighboring setTruncStoreActions into some of the loops;
those shouldn't make a difference, as the additional types are illegal.
(e.g., i128->i1 truncstores on PPC.)

No functional change intended.

Differential Revision: http://reviews.llvm.org/D6532

llvm-svn: 225421
2015-01-08 00:51:32 +00:00
Ahmed Bougacha 67dd2d25a3 [CodeGen] Use MVT iterator_ranges in legality loops. NFC intended.
A few loops do trickier things than just iterating on an MVT subset,
so I'll leave them be for now.
Follow-up of r225387.

llvm-svn: 225392
2015-01-07 21:27:10 +00:00
Craig Topper d3c02f177a Replace several 'assert(false' with 'llvm_unreachable' or fold a condition into the assert.
llvm-svn: 225160
2015-01-05 10:15:49 +00:00
Jingyue Wu e4c9cf04f5 [NVPTX] Fix bugs related to isSingleValueType
Summary:
With isSingleValueType starting to treat vector types as single-value types,
code that uses this interface needs to be updated.

Test Plan:
vector-global.ll
nvcl-param-align.ll

Reviewers: jholewinski

Reviewed By: jholewinski

Subscribers: llvm-commits, meheff, eliben, jholewinski

Differential Revision: http://reviews.llvm.org/D6573

llvm-svn: 224440
2014-12-17 17:59:04 +00:00
Matt Arsenault 31a52ad48c NVPTX: Remove duplicate of AsmPrinter::lowerConstant
llvm-svn: 224355
2014-12-16 19:16:17 +00:00
Matthias Braun 7e37a5f523 [CodeGen] Add print and verify pass after each MachineFunctionPass by default
Previously print+verify passes were added in a very unsystematic way, which is
annoying when debugging as you miss intermediate steps and allows bugs to stay
unnotice when no verification is performed.

To make this change practical I added the possibility to explicitely disable
verification. I used this option on all places where no verification was
performed previously (because alot of places actually don't pass the
MachineVerifier).
In the long term these problems should be fixed properly and verification
enabled after each pass. I'll enable some more verification in subsequent
commits.

This is the 2nd attempt at this after realizing that PassManager::add() may
actually delete the pass.

llvm-svn: 224059
2014-12-11 21:26:47 +00:00
Rafael Espindola 01c73610d0 This reverts commit r224043 and r224042.
check-llvm was failing.

llvm-svn: 224045
2014-12-11 20:03:57 +00:00
Matthias Braun a7c82a9f1d [CodeGen] Add print and verify pass after each MachineFunctionPass by default
Previously print+verify passes were added in a very unsystematic way, which is
annoying when debugging as you miss intermediate steps and allows bugs to stay
unnotice when no verification is performed.

To make this change practical I added the possibility to explicitely disable
verification. I used this option on all places where no verification was
performed previously (because alot of places actually don't pass the
MachineVerifier).
In the long term these problems should be fixed properly and verification
enabled after each pass. I'll enable some more verification in subsequent
commits.

llvm-svn: 224042
2014-12-11 19:42:05 +00:00
Duncan P. N. Exon Smith 5bf8fef580 IR: Split Metadata from Value
Split `Metadata` away from the `Value` class hierarchy, as part of
PR21532.  Assembly and bitcode changes are in the wings, but this is the
bulk of the change for the IR C++ API.

I have a follow-up patch prepared for `clang`.  If this breaks other
sub-projects, I apologize in advance :(.  Help me compile it on Darwin
I'll try to fix it.  FWIW, the errors should be easy to fix, so it may
be simpler to just fix it yourself.

This breaks the build for all metadata-related code that's out-of-tree.
Rest assured the transition is mechanical and the compiler should catch
almost all of the problems.

Here's a quick guide for updating your code:

  - `Metadata` is the root of a class hierarchy with three main classes:
    `MDNode`, `MDString`, and `ValueAsMetadata`.  It is distinct from
    the `Value` class hierarchy.  It is typeless -- i.e., instances do
    *not* have a `Type`.

  - `MDNode`'s operands are all `Metadata *` (instead of `Value *`).

  - `TrackingVH<MDNode>` and `WeakVH` referring to metadata can be
    replaced with `TrackingMDNodeRef` and `TrackingMDRef`, respectively.

    If you're referring solely to resolved `MDNode`s -- post graph
    construction -- just use `MDNode*`.

  - `MDNode` (and the rest of `Metadata`) have only limited support for
    `replaceAllUsesWith()`.

    As long as an `MDNode` is pointing at a forward declaration -- the
    result of `MDNode::getTemporary()` -- it maintains a side map of its
    uses and can RAUW itself.  Once the forward declarations are fully
    resolved RAUW support is dropped on the ground.  This means that
    uniquing collisions on changing operands cause nodes to become
    "distinct".  (This already happened fairly commonly, whenever an
    operand went to null.)

    If you're constructing complex (non self-reference) `MDNode` cycles,
    you need to call `MDNode::resolveCycles()` on each node (or on a
    top-level node that somehow references all of the nodes).  Also,
    don't do that.  Metadata cycles (and the RAUW machinery needed to
    construct them) are expensive.

  - An `MDNode` can only refer to a `Constant` through a bridge called
    `ConstantAsMetadata` (one of the subclasses of `ValueAsMetadata`).

    As a side effect, accessing an operand of an `MDNode` that is known
    to be, e.g., `ConstantInt`, takes three steps: first, cast from
    `Metadata` to `ConstantAsMetadata`; second, extract the `Constant`;
    third, cast down to `ConstantInt`.

    The eventual goal is to introduce `MDInt`/`MDFloat`/etc. and have
    metadata schema owners transition away from using `Constant`s when
    the type isn't important (and they don't care about referring to
    `GlobalValue`s).

    In the meantime, I've added transitional API to the `mdconst`
    namespace that matches semantics with the old code, in order to
    avoid adding the error-prone three-step equivalent to every call
    site.  If your old code was:

        MDNode *N = foo();
        bar(isa             <ConstantInt>(N->getOperand(0)));
        baz(cast            <ConstantInt>(N->getOperand(1)));
        bak(cast_or_null    <ConstantInt>(N->getOperand(2)));
        bat(dyn_cast        <ConstantInt>(N->getOperand(3)));
        bay(dyn_cast_or_null<ConstantInt>(N->getOperand(4)));

    you can trivially match its semantics with:

        MDNode *N = foo();
        bar(mdconst::hasa               <ConstantInt>(N->getOperand(0)));
        baz(mdconst::extract            <ConstantInt>(N->getOperand(1)));
        bak(mdconst::extract_or_null    <ConstantInt>(N->getOperand(2)));
        bat(mdconst::dyn_extract        <ConstantInt>(N->getOperand(3)));
        bay(mdconst::dyn_extract_or_null<ConstantInt>(N->getOperand(4)));

    and when you transition your metadata schema to `MDInt`:

        MDNode *N = foo();
        bar(isa             <MDInt>(N->getOperand(0)));
        baz(cast            <MDInt>(N->getOperand(1)));
        bak(cast_or_null    <MDInt>(N->getOperand(2)));
        bat(dyn_cast        <MDInt>(N->getOperand(3)));
        bay(dyn_cast_or_null<MDInt>(N->getOperand(4)));

  - A `CallInst` -- specifically, intrinsic instructions -- can refer to
    metadata through a bridge called `MetadataAsValue`.  This is a
    subclass of `Value` where `getType()->isMetadataTy()`.

    `MetadataAsValue` is the *only* class that can legally refer to a
    `LocalAsMetadata`, which is a bridged form of non-`Constant` values
    like `Argument` and `Instruction`.  It can also refer to any other
    `Metadata` subclass.

(I'll break all your testcases in a follow-up commit, when I propagate
this change to assembly.)

llvm-svn: 223802
2014-12-09 18:38:53 +00:00
Jacques Pienaar 0c7dc9f7c3 Test commit.
llvm-svn: 223310
2014-12-03 23:21:02 +00:00
Duncan P. N. Exon Smith c280ff0e47 NVPTX: Delete dead code
`MDNode` does not inherit from `User`, and it never has a name.

llvm-svn: 223198
2014-12-03 04:13:23 +00:00
Jingyue Wu 5b62eb9b48 [NVPTX] Do not emit .weak symbols for NVPTX
Summary:
".weak" symbols cannot be consumed by ptxas (PR21685). This patch makes the
weak directive in MCAsmPrinter customizable, and disables emitting ".weak"
symbols for NVPTX.

Test Plan: weak-linkage.ll

Reviewers: jholewinski

Reviewed By: jholewinski

Subscribers: majnemer, jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D6455

llvm-svn: 223077
2014-12-01 21:16:17 +00:00
Craig Topper c50d64b07b Replace neverHasSideEffects=1 with hasSideEffects=0 in all .td files.
llvm-svn: 222801
2014-11-26 00:46:26 +00:00
Reid Kleckner 357600eab5 Add out of line virtual destructors to all LLVMTargetMachine subclasses
These recently all grew a unique_ptr<TargetLoweringObjectFile> member in
r221878.  When anyone calls a virtual method of a class, clang-cl
requires all virtual methods to be semantically valid. This includes the
implicit virtual destructor, which triggers instantiation of the
unique_ptr destructor, which fails because the type being deleted is
incomplete.

This is just part of the ongoing saga of PR20337, which is affecting
Blink as well. Because the MSVC ABI doesn't have key functions, we end
up referencing the vtable and implicit destructor on any virtual call
through a class. We don't actually end up emitting the dtor, so it'd be
good if we could avoid this unneeded type completion work.

llvm-svn: 222480
2014-11-20 23:37:18 +00:00
Aditya Nandakumar 3053155652 We can get the TLOF from the TargetMachine - so constructor no longer requires TargetLoweringObjectFile to be passed.
llvm-svn: 221926
2014-11-13 21:29:21 +00:00
Aditya Nandakumar a27193297f This patch changes the ownership of TLOF from TargetLoweringBase to TargetMachine so that different subtargets could share the TLOF effectively
llvm-svn: 221878
2014-11-13 09:26:31 +00:00
Jingyue Wu a41cf018b8 Fix broken doxygen annotations, NFC
llvm-svn: 221801
2014-11-12 18:25:06 +00:00
Jingyue Wu 8a12cea5f1 Disable indvar widening if arithmetics on the wider type are more expensive
Summary:
Reapply r221772. The old patch breaks the bot because the @indvar_32_bit test
was run whether NVPTX was enabled or not.

IndVarSimplify should not widen an indvar if arithmetics on the wider
indvar are more expensive than those on the narrower indvar. For
instance, although NVPTX64 treats i64 as a legal type, an ADD on i64 is
twice as expensive as that on i32, because the hardware needs to
simulate a 64-bit integer using two 32-bit integers.

Split from D6188, and based on D6195 which adds NVPTXTargetTransformInfo.

Fixes PR21148.

Test Plan:
Added @indvar_32_bit that verifies we do not widen an indvar if the arithmetics
on the wider type are more expensive. This test is run only when NVPTX is
enabled.

Reviewers: jholewinski, eliben, meheff, atrick

Reviewed By: atrick

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D6196

llvm-svn: 221799
2014-11-12 18:09:15 +00:00
Jingyue Wu a48273390c Reverts r221772 which fails tests
llvm-svn: 221773
2014-11-12 07:19:25 +00:00
Jingyue Wu 635a9b14fa Disable indvar widening if arithmetics on the wider type are more expensive
Summary:
IndVarSimplify should not widen an indvar if arithmetics on the wider
indvar are more expensive than those on the narrower indvar. For
instance, although NVPTX64 treats i64 as a legal type, an ADD on i64 is
twice as expensive as that on i32, because the hardware needs to
simulate a 64-bit integer using two 32-bit integers.

Split from D6188, and based on D6195 which adds NVPTXTargetTransformInfo.

Fixes PR21148.

Test Plan:
Added @indvar_32_bit that verifies we do not widen an indvar if the arithmetics
on the wider type are more expensive.

Reviewers: jholewinski, eliben, meheff, atrick

Reviewed By: atrick

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D6196

llvm-svn: 221772
2014-11-12 06:58:45 +00:00
Rafael Espindola 35a12a85a1 Remove a bit of dead code.
Every "real" object file implements this an ptx doesn't use it.

llvm-svn: 221746
2014-11-12 01:27:22 +00:00
Duncan P. N. Exon Smith de36e8040f Revert "IR: MDNode => Value"
Instead, we're going to separate metadata from the Value hierarchy.  See
PR21532.

This reverts commit r221375.
This reverts commit r221373.
This reverts commit r221359.
This reverts commit r221167.
This reverts commit r221027.
This reverts commit r221024.
This reverts commit r221023.
This reverts commit r220995.
This reverts commit r220994.

llvm-svn: 221711
2014-11-11 21:30:22 +00:00
Jingyue Wu dfd4eb9285 [NVPTX] Remove dead code in NVPTXTargetTransformInfo (NFC)
llvm-svn: 221668
2014-11-11 05:24:04 +00:00
Jingyue Wu 0c981bd7df [NVPTX] Add an NVPTX-specific TargetTransformInfo
Summary:
It currently only implements hasBranchDivergence, and will be extended
in later diffs.

Split from D6188.

Test Plan: make check-all

Reviewers: jholewinski

Reviewed By: jholewinski

Subscribers: llvm-commits, meheff, eliben, jholewinski

Differential Revision: http://reviews.llvm.org/D6195

llvm-svn: 221619
2014-11-10 18:38:25 +00:00
Eli Bendersky 799c564236 Clean up NVPTXLowerStructArgs.cpp. NFC
* Remove unnecessary const_casts and C-style casts
* Simplify attribute access code
* Simplify ArrayRef creation
* 80-col and clang-format

llvm-svn: 221464
2014-11-06 17:05:49 +00:00
Aaron Ballman e77ffe35bf Fixing some -Wcast-qual warnings; NFC.
llvm-svn: 221454
2014-11-06 14:32:30 +00:00
Justin Holewinski 3d140fcfd1 [NVPTX] Add NVPTXLowerStructArgs pass
This works around the limitation that PTX does not allow .param space
loads/stores with arbitrary pointers.

If a function has a by-val struct ptr arg, say foo(%struct.x *byval %d), then
add the following instructions to the first basic block :

%temp = alloca %struct.x, align 8
%tt1 = bitcast %struct.x * %d to i8 *
%tt2 = llvm.nvvm.cvt.gen.to.param %tt2
%tempd = bitcast i8 addrspace(101) * to %struct.x addrspace(101) *
%tv = load %struct.x addrspace(101) * %tempd
store %struct.x %tv, %struct.x * %temp, align 8

The above code allocates some space in the stack and copies the incoming
struct from param space to local space. Then replace all occurences of %d
by %temp.

Fixes PR21465.

llvm-svn: 221377
2014-11-05 18:19:30 +00:00
Duncan P. N. Exon Smith c5754a65e6 IR: MDNode => Value: NamedMDNode::getOperator()
Change `NamedMDNode::getOperator()` from returning `MDNode *` to
returning `Value *`.  To reduce boilerplate at some call sites, add a
`getOperatorAsMDNode()` for named metadata that's expected to only
return `MDNode` -- for now, that's everything, but debug node named
metadata (such as llvm.dbg.cu and llvm.dbg.sp) will soon change.  This
is part of PR21433.

Note that there's a follow-up patch to clang for the API change.

llvm-svn: 221375
2014-11-05 18:16:03 +00:00
Duncan P. N. Exon Smith 3872d0084c IR: MDNode => Value: Instruction::getMetadata()
Change `Instruction::getMetadata()` to return `Value` as part of
PR21433.

Update most callers to use `Instruction::getMDNode()`, which wraps the
result in a `cast_or_null<MDNode>`.

llvm-svn: 221024
2014-11-01 00:10:31 +00:00
Jingyue Wu ea51161a94 [NVPTX] aligned byte-buffers for vector return types
Summary:
Fixes PR21100 which is caused by inconsistency between the declared return type
and the expected return type at the call site. The new behavior is consistent
with nvcc and the NVPTXTargetLowering::getPrototype function.

Test Plan: test/Codegen/NVPTX/vector-return.ll

Reviewers: jholewinski

Reviewed By: jholewinski

Subscribers: llvm-commits, meheff, eliben, jholewinski

Differential Revision: http://reviews.llvm.org/D5612

llvm-svn: 220607
2014-10-25 03:46:16 +00:00
Rafael Espindola c606bfe660 Fix a bit of confusion about .set and produce more readable assembly.
Every target we support has support for assembly that looks like

a = b - c
.long a

What is special about MachO is that the above combination suppresses the
production of a relocation.

With this change we avoid producing the intermediary labels when they don't
add any value.

llvm-svn: 220256
2014-10-21 01:17:30 +00:00
Benjamin Kramer 2c99e413ba Reduce double set lookups. NFC.
llvm-svn: 219505
2014-10-10 15:32:50 +00:00
Benjamin Kramer c6cc58e703 Remove unnecessary copying or replace it with moves in a bunch of places.
NFC.

llvm-svn: 219061
2014-10-04 16:55:56 +00:00
Tilmann Scheller 383b4fff4c [NVPTX] Remove dead code.
Found by the Clang static analyzer.

llvm-svn: 218874
2014-10-02 15:12:48 +00:00
Aaron Ballman 0bb041b5f4 Reverting NFC changes from r218050. Instead, the warning was disabled for GCC in r218059, so these changes are no longer required.
llvm-svn: 218062
2014-09-18 17:34:23 +00:00