Commit Graph

114104 Commits

Author SHA1 Message Date
Peter Collingbourne
0bf03cb473 LowerBitSets: Introduce global layout builder.
The builder is based on a layout algorithm that tries to keep members of
small bit sets together. The new layout compresses Chromium's bit sets to
around 15% of their original size.

Differential Revision: http://reviews.llvm.org/D7796

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230394 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 23:17:02 +00:00
Philip Reames
7a1721e7c3 Improve the getting started instructions in the GC docs
This is still gcroot vs gc.statepoint agnostic.  I'm just trying to clarify the general documentation at this point.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230393 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 23:12:27 +00:00
David Majnemer
af91a06d0f PrologEpilogInserter: Clean up math in calculateFrameObjectOffsets
There is no need to open-code the alignment calculation, we have a
handy RoundUpToAlignment function which "Does The Right Thing (TM)".

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230392 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 23:08:13 +00:00
Sanjay Patel
7b59672e39 remove function names from comments; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230391 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 22:43:06 +00:00
Simon Pilgrim
41cda40157 Reapplied D7816 & rL230177 & rL230278 - with an additional fix toensure that the smallest build vector input scalar type is always used. Additional (crash) test cases already committed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230388 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 22:08:56 +00:00
Simon Pilgrim
0115755020 Added test case for PR22678 (check CONCAT_VECTORS DAG combiner pass doesn't introduce illegal types)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230386 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 21:46:23 +00:00
Justin Bogner
564b4a8e3e InstrProf: Test for appropriate linkage of the profiling structures
This test checks that the symbols instrprof creates have appropriate
linkage. The tests already exist in clang in a slightly different form
from before we sunk profile generation into an LLVM pass, but that's
an awkward place for them now. I'll remove/simplify the clang versions
shortly.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230383 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 21:42:42 +00:00
Andrew Kaylor
8f475e9d77 Fixing eol-style
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230378 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 20:49:35 +00:00
Hans Wennborg
2f7ccecf4e ReleaseNotes: add notes about VS2014 Update 4 requirement
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230363 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 20:36:40 +00:00
Arch D. Robison
16310941b4 Fix typo: qual -> equal
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230361 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 20:11:49 +00:00
Philip Reames
989ffa936e Starting to cleanup the garbage collection documentation
In this change:
- Put the getting started section first
- Create a dedicated section to document the built in collector strategies
- Move discuss of ShadowStack into new section
- Add placeholders for erlang, ocaml, and statepoint-example collectors

There will be many more changes following.  I plan on full integrating the documentation for gc.statepoint and gc.root.  I want to make it much clearer on how to get started and what users should expect in terms of effort.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230359 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 19:44:46 +00:00
Eric Christopher
7c611d59cc Revert:
Author: Simon Pilgrim <llvm-dev@redking.me.uk>
Date:   Mon Feb 23 23:04:28 2015 +0000

    Fix based on post-commit comment on D7816 & rL230177 - BUILD_VECTOR operand truncation was using the the BV's output scalar type instead of the input type.

and

Author: Simon Pilgrim <llvm-dev@redking.me.uk>
Date:   Sun Feb 22 18:17:28 2015 +0000

    [DagCombiner] Generalized BuildVector Vector Concatenation

    The CONCAT_VECTORS combiner pass can transform the concat of two BUILD_VECTOR nodes into a single BUILD_VECTOR node.

    This patch generalises this to support any number of BUILD_VECTOR nodes, and also permits UNDEF nodes to be included as well.

    This was noticed as AVX vec128 -> vec256 canonicalization sometimes creates a CONCAT_VECTOR with a real vec128 lower and an vec128 UNDEF upper.

    Differential Revision: http://reviews.llvm.org/D7816

as the root cause of PR22678 which is causing an assertion inside the DAG combiner.

I'll follow up to the main thread as well.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230358 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 19:11:00 +00:00
Eric Christopher
f8c57a105e Rename UpdateRegAllocHint to match style guidelines.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230357 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 19:10:57 +00:00
Matthias Braun
dd1a6e074d AArch64: Relax assert about large shift sizes.
The reason why these large shift sizes happen is because OpaqueConstants
currently inhibit alot of DAG combining, but that has to be addressed in
another commit (like the proposal in D6946).

Differential Revision: http://reviews.llvm.org/D6940

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230355 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 18:52:04 +00:00
Matthias Braun
8b1add3c13 DAGCombiner: Move variable definitions closer to use; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230354 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 18:52:01 +00:00
Matthias Braun
5f146de042 DAGCombiner: Move variable declaration closer to definiion; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230353 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 18:51:59 +00:00
Tom Stellard
ba150ed636 R600/SI: Remove isel mubuf legalization
We legalize mubuf instructions post-instruction selection, so this
code is no longer needed.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230352 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 17:59:19 +00:00
Tim Northover
5530ac99e6 ARM: treat [N x i32] and [N x i64] as AAPCS composite types
The logic is almost there already, with our special homogeneous aggregate
handling. Tweaking it like this allows front-ends to emit AAPCS compliant code
without ever having to count registers or add discarded padding arguments.

Only arrays of i32 and i64 are needed to model AAPCS rules, but I decided to
apply the logic to all integer arrays for more consistency.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230348 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 17:22:34 +00:00
Tobias Grosser
bc8f84c030 Revert "Raising minimum required CMake version to 2.8.12.2."
This reverts commit r230062.

Debian stable (wheezy) ships still with cmake 2.8.9.

The commit broke my LLVM/Polly buildbot, to my knowledge our only Linux+cmake
buildbot.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230343 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 16:39:46 +00:00
Sanjay Patel
a90eb87f7e simplify control flow; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230342 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 16:26:02 +00:00
Hans Wennborg
b499b73e30 Revert r230280: "Bugfix: SCEVExpander incorrectly marks increment operations as no-wrap"
This caused PR22674, failing this assert:

Instructions.h:2281: llvm::Value* llvm::PHINode::getOperand(unsigned int) const: Assertion `i_nocapture < OperandTraits<PHINode>::operands(this) && "getOperand() out of range!"' failed.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230341 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 16:19:29 +00:00
Michael Kuperstein
09d756a7e0 [x32] Mark RBX as reserved when EBX is the base pointer.
This should have gone into r230334.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230339 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 16:13:16 +00:00
Sanjay Patel
269510242b fix typo in comment; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230338 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 16:11:05 +00:00
Michael Kuperstein
2379e8a2ee [x32] x32 should use ebx as the base pointer.
This fixes the original issue in PR22655, but not the secondary one.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230334 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 15:27:13 +00:00
Hal Finkel
cccfe54933 [SDAG] Handle LowerOperation returning its input consistently
For almost all node types, if the target requested custom lowering, and
LowerOperation returned its input, we'd treat the original node as legal. This
did not work, however, for many loads and stores, because they follow
slightly different code paths, and we did not account for the possibility of
LowerOperation returning its input at those call sites.

I think that we now handle this consistently everywhere. At the call sites in
LegalizeDAG, we used to assert in this case, so there's no functional change
for any existing code there. For the call sites in LegalizeVectorOps, this
really only affects whether or not we set Changed = true, but I think makes the
semantics clearer.

No test case here, but it will be covered by an upcoming PowerPC commit adding
QPX support.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230332 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 12:59:47 +00:00
Toma Tabacu
41ecf2b99e [mips] Reformat some TableGen definitions. NFC.
Summary: Separated some instruction and pseudo-instruction definitions from InstAlias definitions, added banner for pseudo-instructions and removed a redundant whitespace from a pseudo-instruction definition. No functional change.

Reviewers: dsanders

Reviewed By: dsanders

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D7552

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230327 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 11:52:19 +00:00
Kuba Brecka
81dce4c02c Fix alloca_instruments_all_paddings.cc test to work under higher -O levels (llvm part)
When AddressSanitizer only a single dynamic alloca and no static allocas, due to an early exit from FunctionStackPoisoner::poisonStack we forget to unpoison the dynamic alloca.  This patch fixes that.

Reviewed at http://reviews.llvm.org/D7810



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230316 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 09:47:05 +00:00
Craig Topper
c3b9d471f6 [X86] Remove the AbsMem32 type from the assembly parser. Only really need the 16-bit version which will automatically get prioritized over AbsMem.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230313 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 08:02:13 +00:00
Reed Kotler
aecbb87ee8 Beginning of alloca implementation for Mips fast-isel
Summary: Begin to add various address modes; including alloca.

Test Plan: Make sure there are no regressions in test-suite at O0/02 in mips32r1/r2

Reviewers: dsanders

Reviewed By: dsanders

Subscribers: echristo, rfuhler, llvm-commits

Differential Revision: http://reviews.llvm.org/D6426

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230300 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 02:36:45 +00:00
Bob Wilson
b664a60ecf Fix handling of negative offsets for AddrModeT2_i8s4 in rewriteT2FrameIndex.
This is a follow up to r230233 to fix something that I noticed by
inspection. The AddrModeT2_i8s4 addressing mode does not support
negative offsets. I spent a good chunk of the day trying to come up with
a testcase for this but was not successful. This addressing mode is used
to spill and restore GPRPair registers in Thumb2 code and that does not
happen often. We also make very limited used of negative offsets when
lowering frame indexes. I am going ahead with the change anyway, because
I am pretty confident that it is correct. I also added a missing assertion
to check that the low bits of the scaled offset are zero.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230297 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 01:37:31 +00:00
Sanjoy Das
edf1700d33 Fix bug 22641
The bug was a result of getPreStartForExtend interpreting nsw/nuw
flags on an add recurrence more strongly than is legal.  {S,+,X}<nsw>
implies S+X is nsw only if the backedge of the loop is taken at least
once.

NOTE: I had accidentally committed an unrelated change with the commit
message of this change in r230275 (r230275 was reverted in r230279).
This is the correct change for this commit message.

Differential Revision: http://reviews.llvm.org/D7808



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230291 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 01:02:42 +00:00
Manman Ren
bd26e6a006 [LTO API] add lto_codegen_set_module to set the destination module.
When debugging LTO issues with ld64, we use -save-temps to save the merged
optimized bitcode file, then invoke ld64 again on the single bitcode file to
speed up debugging code generation passes and ld64 stuff after code generation.

llvm linking a single bitcode file via lto_codegen_add_module will generate a
different bitcode file from the single input. With the newly-added
lto_codegen_set_module, we can make sure the destination module is the same as
the input.

lto_codegen_set_module will transfer the ownship of the module to code
generator.

rdar://19024554


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230290 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 00:45:56 +00:00
Adam Nemet
989a0200c2 [LoopAccesses] LAA::getInfo to use const reference for stride parameter
And other required const-correctness fixes to make this work.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230289 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 00:41:59 +00:00
David Majnemer
fbdee9f0c0 X86: Only use 'lea' in Win64 epilogues if a frame pointer exists
We can only use 'add' in epilogues, 'lea' is not permitted unless we've
established a frame pointer in the prologue.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230286 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 00:11:32 +00:00
Sanjoy Das
f922d9cfe4 New instcombine rule: max(~a,~b) -> ~min(a, b)
This case is interesting because ScalarEvolutionExpander lowers min(a,
b) as ~max(~a,~b).  I think the profitability heuristics can be made
more clever/aggressive, but this is a start.

Differential Revision: http://reviews.llvm.org/D7821



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230285 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-24 00:08:41 +00:00
Sanjoy Das
8d16a81c33 Bugfix: SCEVExpander incorrectly marks increment operations as no-wrap
When emitting the increment operation, SCEVExpander marks the
operation as nuw or nsw based on the flags on the preincrement SCEV.
This is incorrect because, for instance, it is possible that {-6,+,1}
is <nuw> while {-6,+,1}+1 = {-5,+,1} is not.

This change teaches SCEV to mark the increment as nuw/nsw only if it
can explicitly prove that the increment operation won't overflow.

Apart from the attached test case, another (more realistic) manifestation
of the bug can be seen in Transforms/IndVarSimplify/pr20680.ll.

NOTE: this change was landed with an incorrect commit message in
rL230275 and was reverted for that reason in rL230279.  This commit
message is the correct one.

Differential Revision: http://reviews.llvm.org/D7778



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230280 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-23 23:22:58 +00:00
Sanjoy Das
69048edf8a Revert 230275.
230275 got committed with an incorrect commit message due to a mixup
on my side.  Will re-land in a few moments with the correct commit
message.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230279 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-23 23:13:22 +00:00
Simon Pilgrim
a631cd85ca Fix based on post-commit comment on D7816 & rL230177 - BUILD_VECTOR operand truncation was using the the BV's output scalar type instead of the input type.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230278 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-23 23:04:28 +00:00
Andrea Di Biagio
770e106ed6 [X86] Teach how to custom lower double-to-half conversions under fast-math.
This patch teaches the backend how to expand a double-half conversion into
a double-float conversion immediately followed by a float-half conversion.
We do this only under fast-math, and if float-half conversions are legal
for the target.

Added test CodeGen/X86/fastmath-float-half-conversion.ll

Differential Revision: http://reviews.llvm.org/D7832


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230276 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-23 22:59:02 +00:00
Sanjoy Das
7ebbc8de2f Fix bug 22641
The bug was a result of getPreStartForExtend interpreting nsw/nuw
flags on an add recurrence more strongly than is legal.  {S,+,X}<nsw>
implies S+X is nsw only if the backedge of the loop is taken at least
once.

Differential Revision: http://reviews.llvm.org/D7808



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230275 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-23 22:55:13 +00:00
Rafael Espindola
38ce6c7c1c Fix invalid cast.
Fixes PR22525.

Patch by Ben Longbons with testcase by me.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230271 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-23 21:51:06 +00:00
David Majnemer
ad6622575c X86: Use a smaller 'mov' instruction for stack probe calls
Prologue emission, in some cases, requires calls to a stack probe helper
function.  The amount of stack to probe is passed as a register
argument in the Win64 ABI but the instruction sequence used is
pessimistic: it assumes that the number of bytes to probe is greater
than 4 GB.

Instead, select a more appropriate opcode depending on the number of
bytes we are going to probe.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230270 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-23 21:50:30 +00:00
David Majnemer
d71e4c6218 X86: Use 'mov' instead of 'lea' in Win64 SEH prologues when possible
'mov' and 'lea' are equivalent when the displacement applied with 'lea'
is zero.  However, 'mov' should encode smaller.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230269 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-23 21:50:27 +00:00
David Majnemer
16ae406776 X86: Explain why we cannot use a 'mov' in a Win64 epilogue
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230268 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-23 21:50:25 +00:00
David Majnemer
10c4458d7d X86: Consistently use 'epilogue' instead of 'epilog'
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230267 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-23 21:50:18 +00:00
Sanjay Patel
4aef46411f add newline for easier reading; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230265 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-23 21:32:09 +00:00
Bruno Cardoso Lopes
6bf5b2b094 [AsmPrinter] Access pointers to globals via pcrel GOT entries
Front-ends could use global unnamed_addr to hold pointers to other
symbols, like @gotequivalent below:

@foo = global i32 42
@gotequivalent = private unnamed_addr constant i32* @foo

@delta = global i32 trunc (i64 sub (i64 ptrtoint (i32** @gotequivalent to i64),
                                    i64 ptrtoint (i32* @delta to i64))
                           to i32)

The global @delta holds a data "PC"-relative offset to @gotequivalent,
an unnamed pointer to @foo. The darwin/x86-64 assembly output for this follows:

 .globl  _foo
_foo:
 .long   42

 .globl  _gotequivalent
_gotequivalent:
 .quad   _foo

 .globl  _delta
_delta:
 .long   _gotequivalent-_delta

Since unnamed_addr indicates that the address is not significant, only
the content, we can optimize the case above by replacing pc-relative
accesses to "GOT equivalent" globals, by a PC relative access to the GOT
entry of the final symbol instead. Therefore, "delta" can contain a pc
relative relocation to foo's GOT entry and we avoid the emission of
"gotequivalent", yielding the assembly code below:

 .globl  _foo
_foo:
 .long   42

 .globl  _delta
_delta:
 .long   _foo@GOTPCREL+4

There are a couple of advantages of doing this: (1) Front-ends that need
to emit a great deal of data to store pointers to external symbols could
save space by not emitting such "got equivalent" globals and (2) IR
constructs combined with this opt opens a way to represent GOT pcrel
relocations by using the LLVM IR, which is something we previously had
no way to express.

Differential Revision: http://reviews.llvm.org/D6922

rdar://problem/18534217

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230264 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-23 21:26:18 +00:00
Justin Bogner
90c4e8b70c InstrProf: Teach llvm-cov to show the max count instead of the last
When multiple regions start on the same line, llvm-cov was just
showing the count of the last one as the line count. This can be
confusing and misleading for things like one-liner loops, where the
count at the end isn't very interesting, or even "if" statements with
an opening brace at the end of the line.

Instead, use the maximum of all of the region start counts.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230263 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-23 21:21:34 +00:00
Andrew Kaylor
5b7c499806 Removing unused private field.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230259 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-23 21:03:30 +00:00
Bruno Cardoso Lopes
a7db376a63 [X86][MMX] Fix test to reflect current codegen
This test failed in several buildbots, a bit unclear how that happen
since this was the previous behavior before r230248.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@230258 91177308-0d34-0410-b5e6-96231b3b80d8
2015-02-23 20:57:46 +00:00