Commit Graph

281 Commits

Author SHA1 Message Date
Matthew Simpson 235e479984 Reapply "[SLP] Initialize VectorizedValue when gathering"
The test case included in r279125 exposed existing undefined behavior in the
SLP vectorizer that it did not introduce. This patch reapplies the original
patch, but modifies the test case to avoid hitting the undefined behavior. This
allows us to close PR28330 while keeping the UBSan bot happy. The undefined
behavior the original test uncovered will be addressed in a follow-on patch.

Reference: https://llvm.org/bugs/show_bug.cgi?id=28330
llvm-svn: 279370
2016-08-20 14:49:02 +00:00
Vitaly Buka cc7db13bf0 Revert "[SLP] Initialize VectorizedValue when gathering" to fix ubsan bot.
This reverts commit r279125.

https://reviews.llvm.org/D23410

llvm-svn: 279363
2016-08-20 07:09:39 +00:00
Matthew Simpson 11db6b6b8c [SLP] Initialize VectorizedValue when gathering
We abort building vectorizable trees in some cases (e.g., if the maximum
recursion depth is reached, if the region size is too large, etc.). If this
happens for a reduction, we can be left with a root entry that needs to be
gathered. For these cases, we need make sure we actually set VectorizedValue to
the resulting vector.

This patch ensures we properly set VectorizedValue, and it also ensures the
insertelement sequence generated for the gathers is inserted at the correct
location.

Reference: https://llvm.org/bugs/show_bug.cgi?id=28330
Differential Revison: https://reviews.llvm.org/D23410

llvm-svn: 279125
2016-08-18 19:50:32 +00:00
Simon Pilgrim 5d5ca9c0cb [X86][SSE] Add initial costs for vector CTTZ/CTLZ
llvm-svn: 277716
2016-08-04 10:51:41 +00:00
Simon Pilgrim 9e201eac32 [SLPVectorizer][X86] Added vXi8/vXi16 sitofp/uitofp tests
Dropped useless 2i32-2f32 test

llvm-svn: 277281
2016-07-30 21:01:34 +00:00
Simon Pilgrim f5134a2867 [SLPVectorizer][X86] Added SITOFP/UITOFP vectorization tests
llvm-svn: 277275
2016-07-30 18:43:30 +00:00
Michael Kuperstein 38e7298093 [SLPVectorizer] Vectorize reverse-order loads in horizontal reductions
When vectorizing a tree rooted at a store bundle, we currently try to sort the
stores before building the tree, so that the stores can be vectorized. For other
trees, the order of the root bundle - which determines the order of all other
bundles - is arbitrary. That is bad, since if a leaf bundle of consecutive loads
happens to appear in the wrong order, we will not vectorize it.

This is partially mitigated when the root is a binary operator, by trying to
build a "reversed" tree when that's considered profitable. This patch extends the
workaround we have for binops to trees rooted in a horizontal reduction.

This fixes PR28474.

Differential Revision: https://reviews.llvm.org/D22554

llvm-svn: 276477
2016-07-22 21:28:48 +00:00
Simon Pilgrim 1b4f511aaa [X86][SSE] Add cost model values for CTPOP of vectors
This patch adds costs for the vectorized implementations of CTPOP, the default values were seriously underestimating the cost of these and was encouraging vectorization on targets where serialized use of POPCNT would be much better.

Differential Revision: https://reviews.llvm.org/D22456

llvm-svn: 276104
2016-07-20 10:41:28 +00:00
Simon Pilgrim 1b2ab113fb [SLPVectorizer][X86] Added sqrt vectorization tests
llvm-svn: 275788
2016-07-18 13:20:54 +00:00
Simon Pilgrim 4ca42e232d [SLPVectorizer][X86] Added fma vectorization tests
llvm-svn: 274889
2016-07-08 17:19:13 +00:00
Elena Demikhovsky 971fbfda1e Vector GEP test: renamed + some comments
Differential revision: http://reviews.llvm.org/D21957

llvm-svn: 274611
2016-07-06 08:11:23 +00:00
Elena Demikhovsky 6f2ec8104a Fixed crash of SLP Vectorizer on KNL
The bug is connected to vector GEPs.
https://llvm.org/bugs/show_bug.cgi?id=28313

llvm-svn: 273919
2016-06-27 20:07:00 +00:00
Simon Pilgrim bc35f9f702 [SLPVectorizer][X86] Added ceil/floor/nearbyint/rint/trunc vectorization tests
llvm-svn: 273420
2016-06-22 14:07:46 +00:00
Simon Pilgrim 356e823b51 [X86][SSE] Add cost model for BSWAP of vectors
The BSWAP of vector types is quite efficiently implemented using vector shuffles on SSE/AVX targets, we should reflect the typical cost of this to encourage vectorization.

Differential Revision: http://reviews.llvm.org/D21521

llvm-svn: 273217
2016-06-20 23:08:21 +00:00
Sean Silva e0a9e66040 [PM] Port SLPVectorizer to the new PM
This uses the "runImpl" approach to share code with the old PM.

Porting to the new PM meant abandoning the anonymous namespace enclosing
most of SLPVectorizer.cpp which is a bit of a bummer (but not a big deal
compared to having to pull the pass class into a header which the new PM
requires since it calls the constructor directly).

llvm-svn: 272766
2016-06-15 08:43:40 +00:00
Simon Pilgrim 3fc09f7be6 [CostModel][X86][SSE] Updated costs for vector BITREVERSE ops on SSSE3+ targets
To account for the fast PSHUFB implementation now available

llvm-svn: 272484
2016-06-11 19:23:02 +00:00
Michael Zolotukhin 987ab631fa [SLPVectorizer] Handle GEP with differing constant index types
Summary:
This fixes PR27617.

Bug description: The SLPVectorizer asserts on encountering GEPs with different index types, such as i8 and i64.

The patch includes a simple relaxation of the assert to allow constants being of different types, along with a regression test that will provoke the unrelaxed assert.

Reviewers: nadav, mzolotukhin

Subscribers: JesperAntonsson, llvm-commits, mzolotukhin

Differential Revision: http://reviews.llvm.org/D20685

Patch by Jesper Antonsson!

llvm-svn: 272206
2016-06-08 21:55:16 +00:00
Simon Pilgrim ba319ded5e [Analysis] Enabled BITREVERSE as a vectorizable intrinsic
Allows XOP to vectorize BITREVERSE - other targets will follow as their costmodels improve.

llvm-svn: 271803
2016-06-04 20:21:07 +00:00
Guozhi Wei b994f4cdbc [SLP] Pass in correct alignment when query memory access cost
This patch fixes bug https://llvm.org/bugs/show_bug.cgi?id=27897.

When query memory access cost, current SLP always passes in alignment value of 1 (unaligned), so it gets a very high cost of scalar memory access, and wrongly vectorize memory loads in the test case.

It can be fixed by simply giving correct alignment.

llvm-svn: 271333
2016-05-31 20:41:19 +00:00
Simon Pilgrim 45964c3742 [SLPVectorizer][X86] Regenerated SEXT/ZEXT cast vectorization tests
Added 256-bit vector test as well

llvm-svn: 268811
2016-05-06 22:22:18 +00:00
Simon Pilgrim 2def0a878a [SLPVectorizer][X86] Added BSWAP/BITREVERSE vectorization tests
llvm-svn: 268803
2016-05-06 21:41:55 +00:00
Simon Pilgrim a2220ea456 [SLPVectorizer][X86] Added CTPOP/CTLZ/CTTZ vectorization tests
llvm-svn: 268800
2016-05-06 21:33:01 +00:00
David Majnemer 13d5526392 [SLPVectorizer] Add operand bundles to vectorized functions
SLPVectorizing a call site should result in further propagation of its
bundles.

llvm-svn: 268004
2016-04-29 07:09:51 +00:00
Arch D. Robison 0e61034018 [SLPVectorizer] Extend SLP Vectorizer to deal with aggregates.
The refactoring portion part was done as r267748.

http://reviews.llvm.org/D14185

llvm-svn: 267899
2016-04-28 16:11:45 +00:00
Matthew Simpson e5dfb08fcb [TTI] Add hook for vector extract with extension
This change adds a new hook for estimating the cost of vector extracts followed
by zero- and sign-extensions. The motivating example for this change is the
SMOV and UMOV instructions on AArch64. These instructions move data from vector
to general purpose registers while performing the corresponding extension
(sign-extend for SMOV and zero-extend for UMOV) at the same time. For these
operations, TargetTransformInfo can assume the extensions are free and only
report the cost of the vector extract. The SLP vectorizer has been updated to
make use of the new hook.

Differential Revision: http://reviews.llvm.org/D18523

llvm-svn: 267725
2016-04-27 15:20:21 +00:00
Adrian Prantl 75819aedf6 [PR27284] Reverse the ownership between DICompileUnit and DISubprogram.
Currently each Function points to a DISubprogram and DISubprogram has a
scope field. For member functions the scope is a DICompositeType. DIScopes
point to the DICompileUnit to facilitate type uniquing.

Distinct DISubprograms (with isDefinition: true) are not part of the type
hierarchy and cannot be uniqued. This change removes the subprograms
list from DICompileUnit and instead adds a pointer to the owning compile
unit to distinct DISubprograms. This would make it easy for ThinLTO to
strip unneeded DISubprograms and their transitively referenced debug info.

Motivation
----------

Materializing DISubprograms is currently the most expensive operation when
doing a ThinLTO build of clang.

We want the DISubprogram to be stored in a separate Bitcode block (or the
same block as the function body) so we can avoid having to expensively
deserialize all DISubprograms together with the global metadata. If a
function has been inlined into another subprogram we need to store a
reference the block containing the inlined subprogram.

Attached to https://llvm.org/bugs/show_bug.cgi?id=27284 is a python script
that updates LLVM IR testcases to the new format.

http://reviews.llvm.org/D19034
<rdar://problem/25256815>

llvm-svn: 266446
2016-04-15 15:57:41 +00:00
David Majnemer 12fd50410d [SLPVectorizer] Vectorizing the libm sqrt to llvm's sqrt intrinsic requires nnan
To quote the langref "Unlike sqrt in libm, however, llvm.sqrt has
undefined behavior for negative numbers other than -0.0 (which allows
for better optimization, because there is no need to worry about errno
being set). llvm.sqrt(-0.0) is defined to return -0.0 like IEEE sqrt."

This means that it's unsafe to replace sqrt with llvm.sqrt unless the
call is annotated with nnan.

Thanks to Hal Finkel for pointing this out!

llvm-svn: 265521
2016-04-06 07:04:53 +00:00
David Majnemer 25d03dbcde [SLPVectorizer] Vectorize libcalls of sqrt
We didn't realize that we could transform the libcall into a vectorized
intrinsic.

llvm-svn: 265493
2016-04-06 00:14:59 +00:00
David Majnemer 6f1f85f0e1 [SLPVectorizer] Don't insert an extractelement before a catchswitch
A catchswitch cannot be preceded by another instruction in the same
basic block (other than a PHI node).

Instead, insert the extract element right after the materialization of
the vectorized value.  This isn't optimal but is a reasonable compromise
given the constraints of WinEH.

This fixes PR27163.

llvm-svn: 265157
2016-04-01 17:28:15 +00:00
Adrian Prantl b8089516a5 testcase gardening: update the emissionKind enum to the new syntax. (NFC)
llvm-svn: 265081
2016-04-01 00:16:49 +00:00
Adrian Prantl b939a25707 Move the DebugEmissionKind enum from DIBuilder into DICompileUnit.
This mostly cosmetic patch moves the DebugEmissionKind enum from DIBuilder
into DICompileUnit. DIBuilder is not the right place for this enum to live
in — a metadata consumer should not have to include DIBuilder.h.
I also added a Verifier check that checks that the emission kind of a
DICompileUnit is actually legal.

http://reviews.llvm.org/D18612
<rdar://problem/25427165>

llvm-svn: 265077
2016-03-31 23:56:58 +00:00
Paul Robinson 51fa0a87c3 Fix tests that used CHECK-NEXT-NOT and CHECK-DAG-NOT.
FileCheck actually doesn't support combo suffixes.

Differential Revision: http://reviews.llvm.org/D17588

llvm-svn: 262054
2016-02-26 19:40:34 +00:00
Matthew Simpson 92821cb4a8 Reapply commit r259357 with a fix for PR26629
Commit r259357 was reverted because it caused PR26629. We were assuming all
roots of a vectorizable tree could be truncated to the same width, which is not
the case in general. This commit reapplies the patch along with a fix and a new
test case to ensure we don't regress because of this issue again. This should
fix PR26629.

llvm-svn: 261212
2016-02-18 14:14:40 +00:00
David Majnemer f48bcb2bd9 Revert "Reapply commit r258404 with fix."
This reverts commit r259357, it caused PR26629.

llvm-svn: 261137
2016-02-17 19:02:36 +00:00
Matthew Simpson 45dee06177 Add test case missing from r259357 (NFC)
llvm-svn: 259385
2016-02-01 19:09:24 +00:00
Matthew Simpson c578d67407 Reapply commit r258404 with fix.
The previous patch caused PR26364. The fix is to ensure that we don't enter a
cycle when iterating over use-def chains.

llvm-svn: 259357
2016-02-01 13:38:29 +00:00
David Majnemer b2416bd2a7 Revert "Reapply commit r258404 with fix"
This reverts commit r258929, it caused PR26364.

llvm-svn: 259148
2016-01-29 02:43:22 +00:00
Matthew Simpson b95861d35e Reapply commit r258404 with fix
This patch is the second attempt to reapply commit r258404. There was bug in
the initial patch and subsequent fix (mentioned below).

The initial patch caused an assertion because we were computing smaller type
sizes for instructions that cannot be demoted. The fix first determines the
instructions that will be demoted, and then applies the smaller type size to
only those instructions.

This should fix PR26239 and PR26307.

llvm-svn: 258929
2016-01-27 13:43:27 +00:00
Matthew Simpson 61d5a18469 Revert "Reapply commit r258404 with fix"
This commit exposes a crash in computeKnownBits on the Chromium buildbots.
Reverting to investigate.

Reference: https://llvm.org/bugs/show_bug.cgi?id=26307
llvm-svn: 258812
2016-01-26 15:45:49 +00:00
Matthew Simpson cfe5e2c846 Reapply commit r25804 with fix
We were hitting an assertion because we were computing smaller type sizes for
instructions that cannot be demoted. The fix first determines the instructions
that will be demoted, and then applies the smaller type size to only those
instructions.

This should fix PR26239.

llvm-svn: 258705
2016-01-25 19:24:29 +00:00
Matthew Simpson 486bace5cc Revert "[SLP] Truncate expressions to minimum required bit width"
This reverts commit r258404.

llvm-svn: 258408
2016-01-21 17:17:20 +00:00
Matthew Simpson cb17d72170 [SLP] Truncate expressions to minimum required bit width
This change attempts to produce vectorized integer expressions in bit widths
that are narrower than their scalar counterparts. The need for demotion arises
especially on architectures in which the small integer types (e.g., i8 and i16)
are not legal for scalar operations but can still be used in vectors. Like
similar work done within the loop vectorizer, we rely on InstCombine to perform
the actual type-shrinking. We use the DemandedBits analysis and
ComputeNumSignBits from ValueTracking to determine the minimum required bit
width of an expression.

Differential revision: http://reviews.llvm.org/D15815

llvm-svn: 258404
2016-01-21 16:31:55 +00:00
Matthew Simpson 57fe1b10db Reapply r257800 with fix
The fix uniques the bundle of getelementptr indices we are about to vectorize
since it's possible for the same index to be used by multiple instructions.
The original commit message is below.

[SLP] Vectorize the index computations of getelementptr instructions.

This patch seeds the SLP vectorizer with getelementptr indices. The primary
motivation in doing so is to vectorize gather-like idioms beginning with
consecutive loads (e.g., g[a[0] - b[0]] + g[a[1] - b[1]] + ...). While these
cases could be vectorized with a top-down phase, seeding the existing bottom-up
phase with the index computations avoids the complexity, compile-time, and
phase ordering issues associated with a full top-down pass. Only bundles of
single-index getelementptrs with non-constant differences are considered for
vectorization.

llvm-svn: 257918
2016-01-15 18:51:51 +00:00
Matthew Simpson 9258e013a2 Revert "[SLP] Vectorize the index computations of getelementptr instructions."
This reverts commit r257800.

llvm-svn: 257888
2016-01-15 13:10:46 +00:00
Keno Fischer 81e2e9ef86 Reapply r257105 "[Verifier] Check that debug values have proper size"
I originally reapplied this in 257550, but had to revert again due to bot
breakage. The only change in this version is to allow either the TypeSize
or the TypeAllocSize of the variable to be the one represented in debug info
(hopefully in the future we can figure out how to encode the difference).
Additionally, several bot failures following r257550, were due to
optimizer bugs now fixed in r257787 and r257795.

r257550 commit message was:

```
The follow extra changes were made to test cases:

Manually making the variable be the actual type instead of a pointer
to avoid pointer-size differences in generic code:

    LLVM :: DebugInfo/Generic/2010-03-24-MemberFn.ll
    LLVM :: DebugInfo/Generic/2010-04-06-NestedFnDbgInfo.ll
    LLVM :: DebugInfo/Generic/2010-05-03-DisableFramePtr.ll
    LLVM :: DebugInfo/Generic/varargs.ll

Delete sizing information from debug info for the same reason
(but the presence of the pointer was important to the test case):

    LLVM :: DebugInfo/Generic/restrict.ll
    LLVM :: DebugInfo/Generic/tu-composite.ll
    LLVM :: Linker/type-unique-type-array-a.ll
    LLVM :: Linker/type-unique-simple2.ll

Fixing an incorrect DW_OP_deref

    LLVM :: DebugInfo/Generic/2010-05-03-OriginDIE.ll

Fixing a missing DW_OP_deref

    LLVM :: DebugInfo/Generic/incorrect-variable-debugloc.ll

Additionally, clang should no longer complain during bootstrap should no
longer happen after r257534.

The original commit message was:
``
Summary:
Teach the Verifier to make sure that the storage size given to llvm.dbg.declare
or the value size given to llvm.dbg.value agree with what is declared in
DebugInfo. This is implicitly assumed in a number of passes (e.g. in SROA).
Additionally this catches a number of common mistakes, such as passing a
pointer when a value was intended or vice versa.

One complication comes from stack coloring which modifies the original IR when
it merges allocas in order to make sure that if AA falls back to the IR it gets
the correct result. However, given this new invariant, indiscriminately
replacing one alloca by a different (differently sized one) is no longer valid.
Fix this by just undefing out any use of the alloca in a dbg.declare in this
case.

Additionally, I had to fix a number of test cases. Of particular note:
- I regenerated dbg-changes-codegen-branch-folding.ll from the given source as
  it was affected by the bug fixed in r256077
- two-cus-from-same-file.ll was changed to avoid having a variable-typed debug
  variable as that would depend on the target, even though this test is
  supposed to be generic
- I had to manually declared size/align for reference type. See also the
  discussion for D14275/r253186.
- fpstack-debuginstr-kill.ll required changing `double` to `long double`
- most others were just a question of adding OP_deref
``

```

llvm-svn: 257850
2016-01-15 00:46:17 +00:00
Matthew Simpson 791fd160c3 [SLP] Vectorize the index computations of getelementptr instructions.
This patch seeds the SLP vectorizer with getelementptr indices. The primary
motivation in doing so is to vectorize gather-like idioms beginning with
consecutive loads (e.g., g[a[0] - b[0]] + g[a[1] - b[1]] + ...). While these
cases could be vectorized with a top-down phase, seeding the existing bottom-up
phase with the index computations avoids the complexity, compile-time, and
phase ordering issues associated with a full top-down pass. Only bundles of
single-index getelementptrs with non-constant differences are considered for
vectorization.

Differential Revision: http://reviews.llvm.org/D14829

llvm-svn: 257800
2016-01-14 20:46:27 +00:00
Keno Fischer 78e5c9e6e2 Re-Revert r257105 (Verifier debug info changes)
While I investigate some new buildbot failures. This was originally reapplied
as r257550 and r257558.

llvm-svn: 257563
2016-01-13 02:31:14 +00:00
Keno Fischer 25916079ff Reapply r257105 "[Verifier] Check that debug values have proper size"
The follow extra changes were made to test cases:

Manually making the variable be the actual type instead of a pointer
to avoid pointer-size differences in generic code:

    LLVM :: DebugInfo/Generic/2010-03-24-MemberFn.ll
    LLVM :: DebugInfo/Generic/2010-04-06-NestedFnDbgInfo.ll
    LLVM :: DebugInfo/Generic/2010-05-03-DisableFramePtr.ll
    LLVM :: DebugInfo/Generic/varargs.ll

Delete sizing information from debug info for the same reason
(but the presence of the pointer was important to the test case):

    LLVM :: DebugInfo/Generic/restrict.ll
    LLVM :: DebugInfo/Generic/tu-composite.ll
    LLVM :: Linker/type-unique-type-array-a.ll
    LLVM :: Linker/type-unique-simple2.ll

Fixing an incorrect DW_OP_deref

    LLVM :: DebugInfo/Generic/2010-05-03-OriginDIE.ll

Fixing a missing DW_OP_deref

    LLVM :: DebugInfo/Generic/incorrect-variable-debugloc.ll

Additionally, clang should no longer complain during bootstrap should no
longer happen after r257534.

The original commit message was:
```
Summary:
Teach the Verifier to make sure that the storage size given to llvm.dbg.declare
or the value size given to llvm.dbg.value agree with what is declared in
DebugInfo. This is implicitly assumed in a number of passes (e.g. in SROA).
Additionally this catches a number of common mistakes, such as passing a
pointer when a value was intended or vice versa.

One complication comes from stack coloring which modifies the original IR when
it merges allocas in order to make sure that if AA falls back to the IR it gets
the correct result. However, given this new invariant, indiscriminately
replacing one alloca by a different (differently sized one) is no longer valid.
Fix this by just undefing out any use of the alloca in a dbg.declare in this
case.

Additionally, I had to fix a number of test cases. Of particular note:
- I regenerated dbg-changes-codegen-branch-folding.ll from the given source as
  it was affected by the bug fixed in r256077
- two-cus-from-same-file.ll was changed to avoid having a variable-typed debug
  variable as that would depend on the target, even though this test is
  supposed to be generic
- I had to manually declared size/align for reference type. See also the
  discussion for D14275/r253186.
- fpstack-debuginstr-kill.ll required changing `double` to `long double`
- most others were just a question of adding OP_deref
```

llvm-svn: 257550
2016-01-13 00:31:44 +00:00
Keno Fischer ea33a25816 Temporarily revert r257105 "[Verifier] Check that debug values have proper size"
Looks like there's a case where clang generates debug info that triggers
the new verifier check. Reverting while investigating.

llvm-svn: 257107
2016-01-07 22:39:11 +00:00
Keno Fischer b3326be6ad [Verifier] Check that debug values have proper size
Summary:
Teach the Verifier to make sure that the storage size given to llvm.dbg.declare
or the value size given to llvm.dbg.value agree with what is declared in
DebugInfo. This is implicitly assumed in a number of passes (e.g. in SROA).
Additionally this catches a number of common mistakes, such as passing a
pointer when a value was intended or vice versa.

One complication comes from stack coloring which modifies the original IR when
it merges allocas in order to make sure that if AA falls back to the IR it gets
the correct result. However, given this new invariant, indiscriminately
replacing one alloca by a different (differently sized one) is no longer valid.
Fix this by just undefing out any use of the alloca in a dbg.declare in this
case.

Additionally, I had to fix a number of test cases. Of particular note:
- I regenerated dbg-changes-codegen-branch-folding.ll from the given source as
  it was affected by the bug fixed in r256077
- two-cus-from-same-file.ll was changed to avoid having a variable-typed debug
  variable as that would depend on the target, even though this test is
  supposed to be generic
- I had to manually declared size/align for reference type. See also the
  discussion for D14275/r253186.
- fpstack-debuginstr-kill.ll required changing `double` to `long double`
- most others were just a question of adding OP_deref

Reviewers: aprantl
Differential Revision: http://reviews.llvm.org/D14276

llvm-svn: 257105
2016-01-07 22:18:37 +00:00
Charlie Turner b69b92855d [NFC] Update horizontal reduction test cases.
These testcases no longer need to specify -slp-vectorize-hor, since it was
enabled by default in r252733.

llvm-svn: 255783
2015-12-16 17:22:24 +00:00
Mehdi Amini b0e3192a48 Fix SLPVectorizer commutativity reordering
The SLPVectorizer had a very crude way of trying to benefit
from associativity: it tried to optimize for splat/broadcast
or in order to have the same operator on the same side.
This is benefitial to the cost model and allows more vectorization
to occur.
This patch improve the logic and make the detection optimal (locally,
we don't look at the full tree but only at the immediate children).

Should fix https://llvm.org/bugs/show_bug.cgi?id=25247

Reviewers: mzolotukhin

Differential Revision: http://reviews.llvm.org/D13996

From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 252337
2015-11-06 20:17:51 +00:00
Peter Collingbourne d4bff30370 DI: Reverse direction of subprogram -> function edge.
Previously, subprograms contained a metadata reference to the function they
described. Because most clients need to get or set a subprogram for a given
function rather than the other way around, this created unneeded inefficiency.

For example, many passes needed to call the function llvm::makeSubprogramMap()
to build a mapping from functions to subprograms, and the IR linker needed to
fix up function references in a way that caused quadratic complexity in the IR
linking phase of LTO.

This change reverses the direction of the edge by storing the subprogram as
function-level metadata and removing DISubprogram's function field.

Since this is an IR change, a bitcode upgrade has been provided.

Fixes PR23367. An upgrade script for textual IR for out-of-tree clients is
attached to the PR.

Differential Revision: http://reviews.llvm.org/D14265

llvm-svn: 252219
2015-11-05 22:03:56 +00:00
Charlie Turner ab3215fa11 [SLP] Be more aggressive about reduction width selection.
Summary:
This change could be way off-piste, I'm looking for any feedback on whether it's an acceptable approach.

It never seems to be a problem to gobble up as many reduction values as can be found, and then to attempt to reduce the resulting tree. Some of the workloads I'm looking at have been aggressively unrolled by hand, and by selecting reduction widths that are not constrained by a vector register size, it becomes possible to profitably vectorize. My test case shows such an unrolling which SLP was not vectorizing (on neither ARM nor X86) before this patch, but with it does vectorize.

I measure no significant compile time impact of this change when combined with D13949 and D14063. There are also no significant performance regressions on ARM/AArch64 in SPEC or LNT.

The more principled approach I thought of was to generate several candidate tree's and use the cost model to pick the cheapest one. That seemed like quite a big design change (the algorithms seem very much one-shot), and would likely be a costly thing for compile time. This seemed to do the job at very little cost, but I'm worried I've misunderstood something!

Reviewers: nadav, jmolloy

Subscribers: mssimpso, llvm-commits, aemerson

Differential Revision: http://reviews.llvm.org/D14116

llvm-svn: 251428
2015-10-27 17:59:03 +00:00
Charlie Turner cd6e8cf8c2 [SLP] Try a bit harder to find reduction PHIs
Summary:
Currently, when the SLP vectorizer considers whether a phi is part of a reduction, it dismisses phi's whose incoming blocks are not the same as the block containing the phi. For the patterns I'm looking at, extending this rule to allow phis whose incoming block is a containing loop latch allows me to vectorize certain workloads.

There is no significant compile-time impact, and combined with D13949, no performance improvement measured in ARM/AArch64 in any of SPEC2000, SPEC2006 or LNT.

Reviewers: jmolloy, mcrosier, nadav

Subscribers: mssimpso, nadav, aemerson, llvm-commits

Differential Revision: http://reviews.llvm.org/D14063

llvm-svn: 251425
2015-10-27 17:54:16 +00:00
Charlie Turner 74c387feb7 [SLP] Treat SelectInsts as reduction values.
Summary:
Certain workloads, in particular sum-of-absdiff loops, can be vectorized using SLP if it can treat select instructions as reduction values.

The test case is a bit awkward. The AArch64 cost model needs some tuning to not be so pessimistic about selects. I've had to tweak the SLP threshold here.

Reviewers: jmolloy, mzolotukhin, spatel, nadav

Subscribers: nadav, mssimpso, aemerson, llvm-commits

Differential Revision: http://reviews.llvm.org/D13949

llvm-svn: 251424
2015-10-27 17:49:11 +00:00
Michael Zolotukhin fc783e91e0 [SLP] Don't vectorize loads of non-packed types (like i1, i2).
Summary:
Given an array of i2 elements, 4 consecutive scalar loads will be lowered to
i8-sized loads and thus will access 4 consecutive bytes in memory. If we
vectorize these loads into a single <4 x i2> load, it'll access only 1 byte in
memory. Hence, we should prohibit vectorization in such cases.

PS: Initial patch was proposed by Arnold.

Reviewers: aschwaighofer, nadav, hfinkel

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D13277

llvm-svn: 248943
2015-09-30 21:05:43 +00:00
Erik Eckstein 91c49810f2 SLPVectorizer: add a test to check if the minimum region size works.
This is an addition to rL248917.

llvm-svn: 248923
2015-09-30 17:28:19 +00:00
Erik Eckstein 848c1aa452 SLPVectorizer: limit the scheduling region size per basic block.
Usually large blocks are not a problem. But if a large block (> 10k instructions)
contains many (potential) chains of vector instructions, and those chains are
spread over a wide range of instructions, then scheduling becomes a compile time problem.
This change introduces a limit for the accumulate scheduling region size of a block.
For real-world functions this limit will never be exceeded (it's about 10x larger than
the maximum value seen in the test-suite and external test suite).

llvm-svn: 248917
2015-09-30 17:00:44 +00:00
Duncan P. N. Exon Smith 814b8e91c7 DI: Require subprogram definitions to be distinct
As a follow-up to r246098, require `DISubprogram` definitions
(`isDefinition: true`) to be 'distinct'.  Specifically, add an assembler
check, a verifier check, and bitcode upgrading logic to combat testcase
bitrot after the `DIBuilder` change.

While working on the testcases, I realized that
test/Linker/subprogram-linkonce-weak-odr.ll isn't relevant anymore.  Its
purpose was to check for a corner case in PR22792 where two subprogram
definitions match exactly and share the same metadata node.  The new
verifier check, requiring that subprogram definitions are 'distinct',
precludes that possibility.

I updated almost all the IR with the following script:

    git grep -l -E -e '= !DISubprogram\(.* isDefinition: true' |
    grep -v test/Bitcode |
    xargs sed -i '' -e 's/= \(!DISubprogram(.*, isDefinition: true\)/= distinct \1/'

Likely some variant of would work for out-of-tree testcases.

llvm-svn: 246327
2015-08-28 20:26:49 +00:00
Michael Zolotukhin 6002295c6a [SLP] Add one more test case for propagating 'nontemporal' attributes.
llvm-svn: 245644
2015-08-21 00:08:39 +00:00
Michael Zolotukhin 51b00e6d82 [SLP] Propagate 'nontemporal' attribute into vectorized instructions.
llvm-svn: 245633
2015-08-20 22:28:15 +00:00
Silviu Baranga b322aa6f53 [CostModel][AArch64] Increase cost of vector insert element and add missing cast costs
Summary:
Increase the estimated costs for insert/extract element operations on
AArch64. This is motivated by results from benchmarking interleaved
accesses.

Add missing costs for zext/sext/trunc instructions and some integer to
floating point conversions. These costs were previously calculated
by scalarizing these operation and were affected by the cost increase of
the insert/extract element operations.

Reviewers: rengolin

Subscribers: mcrosier, aemerson, rengolin, llvm-commits

Differential Revision: http://reviews.llvm.org/D11939

llvm-svn: 245226
2015-08-17 16:05:09 +00:00
Duncan P. N. Exon Smith 55ca964e94 DI: Disallow uniquable DICompileUnits
Since r241097, `DIBuilder` has only created distinct `DICompileUnit`s.
The backend is liable to start relying on that (if it hasn't already),
so make uniquable `DICompileUnit`s illegal and automatically upgrade old
bitcode.  This is a nice cleanup, since we can remove an unnecessary
`DenseSet` (and the associated uniquing info) from `LLVMContextImpl`.

Almost all the testcases were updated with this script:

    git grep -e '= !DICompileUnit' -l -- test |
    grep -v test/Bitcode |
    xargs sed -i '' -e 's,= !DICompileUnit,= distinct !DICompileUnit,'

I imagine something similar should work for out-of-tree testcases.

llvm-svn: 243885
2015-08-03 17:26:41 +00:00
Duncan P. N. Exon Smith ed013cd221 DI: Remove DW_TAG_arg_variable and DW_TAG_auto_variable
Remove the fake `DW_TAG_auto_variable` and `DW_TAG_arg_variable` tags,
using `DW_TAG_variable` in their place Stop exposing the `tag:` field at
all in the assembly format for `DILocalVariable`.

Most of the testcase updates were generated by the following sed script:

    find test/ -name "*.ll" -o -name "*.mir" |
    xargs grep -l 'DILocalVariable' |
    xargs sed -i '' \
      -e 's/tag: DW_TAG_arg_variable, //' \
      -e 's/tag: DW_TAG_auto_variable, //'

There were only a handful of tests in `test/Assembly` that I needed to
update by hand.

(Note: a follow-up could change `DILocalVariable::DILocalVariable()` to
set the tag to `DW_TAG_formal_parameter` instead of `DW_TAG_variable`
(as appropriate), instead of having that logic magically in the backend
in `DbgVariable`.  I've added a FIXME to that effect.)

llvm-svn: 243774
2015-07-31 18:58:39 +00:00
Wei Mi d6f7252e2e [SLP vectorizer]: Choose the best consecutive candidate to pair with a store instruction.
The patch changes the SLPVectorizer::vectorizeStores to choose the immediate
succeeding or preceding candidate for a store instruction when it has multiple
consecutive candidates. In this way it has better chance to find more slp
vectorization opportunities.

Differential Revision: http://reviews.llvm.org/D10445

llvm-svn: 243666
2015-07-30 17:40:39 +00:00
Sanjay Patel 1319446195 [SLPVectorizer] Try different vectorization factors for store chains
...and set max vector register size based on target 

This patch is based on discussion on the llvmdev mailing list:
http://lists.cs.uiuc.edu/pipermail/llvmdev/2015-July/087405.html

and also solves:
https://llvm.org/bugs/show_bug.cgi?id=17170

Several FIXME/TODO items are noted in comments as potential improvements.

Differential Revision: http://reviews.llvm.org/D10950

llvm-svn: 241760
2015-07-08 23:40:55 +00:00
Sanjay Patel 1a6a58caf5 change CHECK to CHECK-LABEL for more precision
llvm-svn: 241422
2015-07-05 23:19:16 +00:00
Sanjay Patel 787e12aec1 remove unnecessary test specifications
llvm-svn: 241419
2015-07-05 22:37:51 +00:00
Sanjay Patel 8ee056c5e7 minimize test case and remove unnecessary opt passes
llvm-svn: 241418
2015-07-05 22:30:12 +00:00
Michael Zolotukhin 4d8ffa082c [SLP] Vectorize for all-constant entries.
Differential Revision: http://reviews.llvm.org/D10531

llvm-svn: 240144
2015-06-19 17:40:15 +00:00
Matt Arsenault 5eb5eb59fc AMDGPU: Fix some places missed in rename
llvm-svn: 240143
2015-06-19 17:39:03 +00:00
David Majnemer 7fddeccb8b Move the personality function from LandingPadInst to Function
The personality routine currently lives in the LandingPadInst.

This isn't desirable because:
- All LandingPadInsts in the same function must have the same
  personality routine.  This means that each LandingPadInst beyond the
  first has an operand which produces no additional information.

- There is ongoing work to introduce EH IR constructs other than
  LandingPadInst.  Moving the personality routine off of any one
  particular Instruction and onto the parent function seems a lot better
  than have N different places a personality function can sneak onto an
  exceptional function.

Differential Revision: http://reviews.llvm.org/D10429

llvm-svn: 239940
2015-06-17 20:52:32 +00:00
Akira Hatanaka 3058d0f080 Let llc and opt override "-target-cpu" and "-target-features" via command line
options.

This commit fixes a bug in llc and opt where "-mcpu" and "-mattr" wouldn't
override function attributes "-target-cpu" and "-target-features" in the IR.

Differential Revision: http://reviews.llvm.org/D9537

llvm-svn: 236677
2015-05-06 23:54:14 +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
David Blaikie 23af64846f [opaque pointer type] Add textual IR support for explicit type parameter to the call instruction
See r230786 and r230794 for similar changes to gep and load
respectively.

Call is a bit different because it often doesn't have a single explicit
type - usually the type is deduced from the arguments, and just the
return type is explicit. In those cases there's no need to change the
IR.

When that's not the case, the IR usually contains the pointer type of
the first operand - but since typed pointers are going away, that
representation is insufficient so I'm just stripping the "pointerness"
of the explicit type away.

This does make the IR a bit weird - it /sort of/ reads like the type of
the first operand: "call void () %x(" but %x is actually of type "void
()*" and will eventually be just of type "ptr". But this seems not too
bad and I don't think it would benefit from repeating the type
("void (), void () * %x(" and then eventually "void (), ptr %x(") as has
been done with gep and load.

This also has a side benefit: since the explicit type is no longer a
pointer, there's no ambiguity between an explicit type and a function
that returns a function pointer. Previously this case needed an explicit
type (eg: a function returning a void() function was written as
"call void () () * @x(" rather than "call void () * @x(" because of the
ambiguity between a function returning a pointer to a void() function
and a function returning void).

No ambiguity means even function pointer return types can just be
written alone, without writing the whole function's type.

This leaves /only/ the varargs case where the explicit type is required.

Given the special type syntax in call instructions, the regex-fu used
for migration was a bit more involved in its own unique way (as every
one of these is) so here it is. Use it in conjunction with the apply.sh
script and associated find/xargs commands I've provided in rr230786 to
migrate your out of tree tests. Do let me know if any of this doesn't
cover your cases & we can iterate on a more general script/regexes to
help others with out of tree tests.

About 9 test cases couldn't be automatically migrated - half of those
were functions returning function pointers, where I just had to manually
delete the function argument types now that we didn't need an explicit
function type there. The other half were typedefs of function types used
in calls - just had to manually drop the * from those.

import fileinput
import sys
import re

pat = re.compile(r'((?:=|:|^|\s)call\s(?:[^@]*?))(\s*$|\s*(?:(?:\[\[[a-zA-Z0-9_]+\]\]|[@%](?:(")?[\\\?@a-zA-Z0-9_.]*?(?(3)"|)|{{.*}}))(?:\(|$)|undef|inttoptr|bitcast|null|asm).*$)')
addrspace_end = re.compile(r"addrspace\(\d+\)\s*\*$")
func_end = re.compile("(?:void.*|\)\s*)\*$")

def conv(match, line):
  if not match or re.search(addrspace_end, match.group(1)) or not re.search(func_end, match.group(1)):
    return line
  return line[:match.start()] + match.group(1)[:match.group(1).rfind('*')].rstrip() + match.group(2) + line[match.end():]

for line in sys.stdin:
  sys.stdout.write(conv(re.search(pat, line), line))

llvm-svn: 235145
2015-04-16 23:24:18 +00:00
Duncan P. N. Exon Smith 988a7f8b79 DebugInfo: Fix bad debug info for compile units and types
Fix debug info in these tests, which started failing with a WIP patch to
verify compile units and types.  The problems look like they were all
caused by bitrot.  They fell into these categories:

  - Using `!{i32 0}` instead of `!{}`.
  - Using `!{null}` instead of `!{}`.
  - Using `!MDExpression()` instead of `!{}`.
  - Using `!8` instead of `!{!8}`.
  - `file:` references that pointed at `MDCompileUnit`s instead of the
    same `MDFile` as the compile unit.
  - `file:` references that were numerically off-by-one or (off-by-ten).

llvm-svn: 233415
2015-03-27 20:46:33 +00:00
Duncan P. N. Exon Smith 166121ad0b Verifier: Check debug info intrinsic arguments
Verify that debug info intrinsic arguments are valid.  (These checks
will not recurse through the full debug info graph, so they don't need
to be cordoned of in `DebugInfoVerifier`.)

With those checks in place, changing the `DbgIntrinsicInst` accessors to
downcast to `MDLocalVariable` and `MDExpression` is natural (added isa
specializations in `Metadata.h` to support this).

Added tests to `test/Verifier` for the new -verify checks, and fixed the
debug info in all the in-tree tests.

If you have out-of-tree testcases that have started to fail to -verify,
hopefully the verify checks are helpful.  The most likely problem is
that the expression argument is `!{}` (instead of `!MDExpression()`).

llvm-svn: 232296
2015-03-15 01:21:30 +00:00
David Blaikie f72d05bc7b [opaque pointer type] Add textual IR support for explicit type parameter to gep operator
Similar to gep (r230786) and load (r230794) changes.

Similar migration script can be used to update test cases, which
successfully migrated all of LLVM and Polly, but about 4 test cases
needed manually changes in Clang.

(this script will read the contents of stdin and massage it into stdout
- wrap it in the 'apply.sh' script shown in previous commits + xargs to
apply it over a large set of test cases)

import fileinput
import sys
import re

rep = re.compile(r"(getelementptr(?:\s+inbounds)?\s*\()((<\d*\s+x\s+)?([^@]*?)(|\s*addrspace\(\d+\))\s*\*(?(3)>)\s*)(?=$|%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|zeroinitializer|<|\[\[[a-zA-Z]|\{\{)", re.MULTILINE | re.DOTALL)

def conv(match):
  line = match.group(1)
  line += match.group(4)
  line += ", "
  line += match.group(2)
  return line

line = sys.stdin.read()
off = 0
for match in re.finditer(rep, line):
  sys.stdout.write(line[off:match.start()])
  sys.stdout.write(conv(match))
  off = match.end()
sys.stdout.write(line[off:])

llvm-svn: 232184
2015-03-13 18:20:45 +00:00
Duncan P. N. Exon Smith e274180f0e DebugInfo: Move new hierarchy into place
Move the specialized metadata nodes for the new debug info hierarchy
into place, finishing off PR22464.  I've done bootstraps (and all that)
and I'm confident this commit is NFC as far as DWARF output is
concerned.  Let me know if I'm wrong :).

The code changes are fairly mechanical:

  - Bumped the "Debug Info Version".
  - `DIBuilder` now creates the appropriate subclass of `MDNode`.
  - Subclasses of DIDescriptor now expect to hold their "MD"
    counterparts (e.g., `DIBasicType` expects `MDBasicType`).
  - Deleted a ton of dead code in `AsmWriter.cpp` and `DebugInfo.cpp`
    for printing comments.
  - Big update to LangRef to describe the nodes in the new hierarchy.
    Feel free to make it better.

Testcase changes are enormous.  There's an accompanying clang commit on
its way.

If you have out-of-tree debug info testcases, I just broke your build.

  - `upgrade-specialized-nodes.sh` is attached to PR22564.  I used it to
    update all the IR testcases.
  - Unfortunately I failed to find way to script the updates to CHECK
    lines, so I updated all of these by hand.  This was fairly painful,
    since the old CHECKs are difficult to reason about.  That's one of
    the benefits of the new hierarchy.

This work isn't quite finished, BTW.  The `DIDescriptor` subclasses are
almost empty wrappers, but not quite: they still have loose casting
checks (see the `RETURN_FROM_RAW()` macro).  Once they're completely
gutted, I'll rename the "MD" classes to "DI" and kill the wrappers.  I
also expect to make a few schema changes now that it's easier to reason
about everything.

llvm-svn: 231082
2015-03-03 17:24:31 +00:00
David Blaikie a79ac14fa6 [opaque pointer type] Add textual IR support for explicit type parameter to load instruction
Essentially the same as the GEP change in r230786.

A similar migration script can be used to update test cases, though a few more
test case improvements/changes were required this time around: (r229269-r229278)

import fileinput
import sys
import re

pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)")

for line in sys.stdin:
  sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line))

Reviewers: rafael, dexonsmith, grosser

Differential Revision: http://reviews.llvm.org/D7649

llvm-svn: 230794
2015-02-27 21:17:42 +00:00
David Blaikie 79e6c74981 [opaque pointer type] Add textual IR support for explicit type parameter to getelementptr instruction
One of several parallel first steps to remove the target type of pointers,
replacing them with a single opaque pointer type.

This adds an explicit type parameter to the gep instruction so that when the
first parameter becomes an opaque pointer type, the type to gep through is
still available to the instructions.

* This doesn't modify gep operators, only instructions (operators will be
  handled separately)

* Textual IR changes only. Bitcode (including upgrade) and changing the
  in-memory representation will be in separate changes.

* geps of vectors are transformed as:
    getelementptr <4 x float*> %x, ...
  ->getelementptr float, <4 x float*> %x, ...
  Then, once the opaque pointer type is introduced, this will ultimately look
  like:
    getelementptr float, <4 x ptr> %x
  with the unambiguous interpretation that it is a vector of pointers to float.

* address spaces remain on the pointer, not the type:
    getelementptr float addrspace(1)* %x
  ->getelementptr float, float addrspace(1)* %x
  Then, eventually:
    getelementptr float, ptr addrspace(1) %x

Importantly, the massive amount of test case churn has been automated by
same crappy python code. I had to manually update a few test cases that
wouldn't fit the script's model (r228970,r229196,r229197,r229198). The
python script just massages stdin and writes the result to stdout, I
then wrapped that in a shell script to handle replacing files, then
using the usual find+xargs to migrate all the files.

update.py:
import fileinput
import sys
import re

ibrep = re.compile(r"(^.*?[^%\w]getelementptr inbounds )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")
normrep = re.compile(       r"(^.*?[^%\w]getelementptr )(((?:<\d* x )?)(.*?)(| addrspace\(\d\)) *\*(|>)(?:$| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$))")

def conv(match, line):
  if not match:
    return line
  line = match.groups()[0]
  if len(match.groups()[5]) == 0:
    line += match.groups()[2]
  line += match.groups()[3]
  line += ", "
  line += match.groups()[1]
  line += "\n"
  return line

for line in sys.stdin:
  if line.find("getelementptr ") == line.find("getelementptr inbounds"):
    if line.find("getelementptr inbounds") != line.find("getelementptr inbounds ("):
      line = conv(re.match(ibrep, line), line)
  elif line.find("getelementptr ") != line.find("getelementptr ("):
    line = conv(re.match(normrep, line), line)
  sys.stdout.write(line)

apply.sh:
for name in "$@"
do
  python3 `dirname "$0"`/update.py < "$name" > "$name.tmp" && mv "$name.tmp" "$name"
  rm -f "$name.tmp"
done

The actual commands:
From llvm/src:
find test/ -name *.ll | xargs ./apply.sh
From llvm/src/tools/clang:
find test/ -name *.mm -o -name *.m -o -name *.cpp -o -name *.c | xargs -I '{}' ../../apply.sh "{}"
From llvm/src/tools/polly:
find test/ -name *.ll | xargs ./apply.sh

After that, check-all (with llvm, clang, clang-tools-extra, lld,
compiler-rt, and polly all checked out).

The extra 'rm' in the apply.sh script is due to a few files in clang's test
suite using interesting unicode stuff that my python script was throwing
exceptions on. None of those files needed to be migrated, so it seemed
sufficient to ignore those cases.

Reviewers: rafael, dexonsmith, grosser

Differential Revision: http://reviews.llvm.org/D7636

llvm-svn: 230786
2015-02-27 19:29:02 +00:00
Chandler Carruth 63aaa98d94 [slp] Fix a nasty bug in the SLP vectorizer that Joerg pointed out.
Apparently some code finally started to tickle this after my
canonicalization changes to instcombine.

The bug stems from trying to form a vector type out of scalars that
aren't compatible at all. In this example, from x86_mmx values. The code
in the vectorizer that checks for reasonable types whas checking for
aggregates or vectors, but there are lots of other types that should
just never reach the vectorizer.

Debugging this was made more confusing by the lie in an assert in
VectorType::get() -- it isn't that the types are *primitive*. The types
must be integer, pointer, or floating point types. No other types are
allowed.

I've improved the assert and added a helper to the vectorizer to handle
the element type validity checks. It now re-uses the VectorType static
function and then further excludes weird target-specific types that we
probably shouldn't be touching here (x86_fp80 and ppc_fp128). Neither of
these are really reachable anyways (neither 80-bit nor 128-bit things
will get vectorized) but it seems better to just eagerly exclude such
nonesense.

I've added a test case, but while it definitely covers two of the paths
through this code there may be more paths that would benefit from test
coverage. I'm not familiar enough with the SLP vectorizer to synthesize
test cases for all of these, but was able to update the code itself by
inspection.

llvm-svn: 228899
2015-02-12 02:30:56 +00:00
Erik Eckstein 7330e358f6 Fix: SLPVectorizer crashes with assertion when vectorizing a cmp instruction.
The commit r225977 uncovered this bug. The problem was that the vectorizer tried to
read the second operand of an already deleted instruction.
The bug didn't show up before r225977 because the freed memory still contained a non-null pointer.
With r225977 deletion of instructions is delayed and the read operand pointer is always null.

llvm-svn: 227800
2015-02-02 12:45:34 +00:00
Erik Eckstein 98df6da740 SLPVectorizer: fix wrong scheduling of atomic load/stores.
This fixes PR22306.

llvm-svn: 227077
2015-01-26 09:07:04 +00:00
Karthik Bhat 0b0f4660fa Fix Operandreorder logic in SLPVectorizer to generate longer vectorizable chain.
This patch fixes 2 issues in reorderInputsAccordingToOpcode
1) AllSameOpcodeLeft and AllSameOpcodeRight was being calculated incorrectly resulting in code not being vectorized in few cases.
2) Adds logic to reorder operands if we get longer chain of consecutive loads enabling vectorization. Handled the same for cases were we have AltOpcode.
Thanks Michael for inputs and review.
Review: http://reviews.llvm.org/D6677

llvm-svn: 226547
2015-01-20 06:11:00 +00:00
Duncan P. N. Exon Smith 9885469922 IR: Move MDLocation into place
This commit moves `MDLocation`, finishing off PR21433.  There's an
accompanying clang commit for frontend testcases.  I'll attach the
testcase upgrade script I used to PR21433 to help out-of-tree
frontends/backends.

This changes the schema for `DebugLoc` and `DILocation` from:

    !{i32 3, i32 7, !7, !8}

to:

    !MDLocation(line: 3, column: 7, scope: !7, inlinedAt: !8)

Note that empty fields (line/column: 0 and inlinedAt: null) don't get
printed by the assembly writer.

llvm-svn: 226048
2015-01-14 22:27:36 +00:00
Suyog Sarda 43fae93da8 Revert 224119 "This patch recognizes (+ (+ v0, v1) (+ v2, v3)), reorders them for bundling into vector of loads,
and vectorizes it." 

This was re-ordering floating point data types resulting in mismatch in output.

llvm-svn: 224424
2014-12-17 10:34:27 +00:00
Duncan P. N. Exon Smith be7ea19b58 IR: Make metadata typeless in assembly
Now that `Metadata` is typeless, reflect that in the assembly.  These
are the matching assembly changes for the metadata/value split in
r223802.

  - Only use the `metadata` type when referencing metadata from a call
    intrinsic -- i.e., only when it's used as a `Value`.

  - Stop pretending that `ValueAsMetadata` is wrapped in an `MDNode`
    when referencing it from call intrinsics.

So, assembly like this:

    define @foo(i32 %v) {
      call void @llvm.foo(metadata !{i32 %v}, metadata !0)
      call void @llvm.foo(metadata !{i32 7}, metadata !0)
      call void @llvm.foo(metadata !1, metadata !0)
      call void @llvm.foo(metadata !3, metadata !0)
      call void @llvm.foo(metadata !{metadata !3}, metadata !0)
      ret void, !bar !2
    }
    !0 = metadata !{metadata !2}
    !1 = metadata !{i32* @global}
    !2 = metadata !{metadata !3}
    !3 = metadata !{}

turns into this:

    define @foo(i32 %v) {
      call void @llvm.foo(metadata i32 %v, metadata !0)
      call void @llvm.foo(metadata i32 7, metadata !0)
      call void @llvm.foo(metadata i32* @global, metadata !0)
      call void @llvm.foo(metadata !3, metadata !0)
      call void @llvm.foo(metadata !{!3}, metadata !0)
      ret void, !bar !2
    }
    !0 = !{!2}
    !1 = !{i32* @global}
    !2 = !{!3}
    !3 = !{}

I wrote an upgrade script that handled almost all of the tests in llvm
and many of the tests in cfe (even handling many `CHECK` lines).  I've
attached it (or will attach it in a moment if you're speedy) to PR21532
to help everyone update their out-of-tree testcases.

This is part of PR21532.

llvm-svn: 224257
2014-12-15 19:07:53 +00:00
Suyog Sarda 2b27fc78a2 Typo Correction in Test Case. NFC.
llvm-svn: 224244
2014-12-15 12:19:46 +00:00
Suyog Sarda 384095e65c This patch recognizes (+ (+ v0, v1) (+ v2, v3)), reorders them for bundling into vector of loads,
and vectorizes it. 
 
 Test case :
 
       float hadd(float* a) {
           return (a[0] + a[1]) + (a[2] + a[3]);
        }
 
 
 AArch64 assembly before patch :
 
        ldp	s0, s1, [x0]
 	ldp	s2, s3, [x0, #8]
 	fadd	s0, s0, s1
 	fadd	s1, s2, s3
 	fadd	s0, s0, s1
 	ret
 
 AArch64 assembly after patch :
 
        ldp	d0, d1, [x0]
 	fadd	v0.2s, v0.2s, v1.2s
 	faddp	s0, v0.2s
 	ret

Reviewed Link : http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20141208/248531.html

llvm-svn: 224119
2014-12-12 12:53:44 +00:00
Suyog Sarda aba97f4aba Vectorize a reduction chain feeding into a 'return' statement.
e.x 
return (a[0]+b[0]) + (a[1]+b[1])

Differential Revision: http://reviews.llvm.org/D6227

llvm-svn: 222364
2014-11-19 16:07:38 +00:00
Hal Finkel 3b7fc86677 [SLPVectorize] Basic ephemeral-value awareness
The SLP vectorizer should not vectorize ephemeral values. These are used to
express information to the optimizer, and vectorizing them does not lead to
faster code (because the ephemeral values are dropped prior to code generation,
vectorized or not), and obscures the information the instructions are
attempting to communicate (the logic that interprets the arguments to
@llvm.assume generically does not understand vectorized conditions).

Also, uses by ephemeral values are free (because they, and the necessary
extractelement instructions, will be dropped prior to code generation).

llvm-svn: 219816
2014-10-15 17:35:01 +00:00
Duncan P. N. Exon Smith 176b691d32 Revert "Revert "DI: Fold constant arguments into a single MDString""
This reverts commit r218918, effectively reapplying r218914 after fixing
an Ocaml bindings test and an Asan crash.  The root cause of the latter
was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a
PR to investigate who requires the loose check (and why).

Original commit message follows.

--

This patch addresses the first stage of PR17891 by folding constant
arguments together into a single MDString.  Integers are stringified and
a `\0` character is used as a separator.

Part of PR17891.

Note: I've attached my testcases upgrade scripts to the PR.  If I've
just broken your out-of-tree testcases, they might help.

llvm-svn: 219010
2014-10-03 20:01:09 +00:00
Duncan P. N. Exon Smith 786cd049fc Revert "DI: Fold constant arguments into a single MDString"
This reverts commit r218914 while I investigate some bots.

llvm-svn: 218918
2014-10-02 22:15:31 +00:00
Duncan P. N. Exon Smith 571f97bd90 DI: Fold constant arguments into a single MDString
This patch addresses the first stage of PR17891 by folding constant
arguments together into a single MDString.  Integers are stringified and
a `\0` character is used as a separator.

Part of PR17891.

Note: I've attached my testcases upgrade scripts to the PR.  If I've
just broken your out-of-tree testcases, they might help.

llvm-svn: 218914
2014-10-02 21:56:57 +00:00
Adrian Prantl 87b7eb9d0f Move the complex address expression out of DIVariable and into an extra
argument of the llvm.dbg.declare/llvm.dbg.value intrinsics.

Previously, DIVariable was a variable-length field that has an optional
reference to a Metadata array consisting of a variable number of
complex address expressions. In the case of OpPiece expressions this is
wasting a lot of storage in IR, because when an aggregate type is, e.g.,
SROA'd into all of its n individual members, the IR will contain n copies
of the DIVariable, all alike, only differing in the complex address
reference at the end.

By making the complex address into an extra argument of the
dbg.value/dbg.declare intrinsics, all of the pieces can reference the
same variable and the complex address expressions can be uniqued across
the CU, too.
Down the road, this will allow us to move other flags, such as
"indirection" out of the DIVariable, too.

The new intrinsics look like this:
declare void @llvm.dbg.declare(metadata %storage, metadata %var, metadata %expr)
declare void @llvm.dbg.value(metadata %storage, i64 %offset, metadata %var, metadata %expr)

This patch adds a new LLVM-local tag to DIExpressions, so we can detect
and pretty-print DIExpression metadata nodes.

What this patch doesn't do:

This patch does not touch the "Indirect" field in DIVariable; but moving
that into the expression would be a natural next step.

http://reviews.llvm.org/D4919
rdar://problem/17994491

Thanks to dblaikie and dexonsmith for reviewing this patch!

Note: I accidentally committed a bogus older version of this patch previously.
llvm-svn: 218787
2014-10-01 18:55:02 +00:00
Adrian Prantl b458dc2eee Revert r218778 while investigating buldbot breakage.
"Move the complex address expression out of DIVariable and into an extra"

llvm-svn: 218782
2014-10-01 18:10:54 +00:00
Adrian Prantl 25a7174e7a Move the complex address expression out of DIVariable and into an extra
argument of the llvm.dbg.declare/llvm.dbg.value intrinsics.

Previously, DIVariable was a variable-length field that has an optional
reference to a Metadata array consisting of a variable number of
complex address expressions. In the case of OpPiece expressions this is
wasting a lot of storage in IR, because when an aggregate type is, e.g.,
SROA'd into all of its n individual members, the IR will contain n copies
of the DIVariable, all alike, only differing in the complex address
reference at the end.

By making the complex address into an extra argument of the
dbg.value/dbg.declare intrinsics, all of the pieces can reference the
same variable and the complex address expressions can be uniqued across
the CU, too.
Down the road, this will allow us to move other flags, such as
"indirection" out of the DIVariable, too.

The new intrinsics look like this:
declare void @llvm.dbg.declare(metadata %storage, metadata %var, metadata %expr)
declare void @llvm.dbg.value(metadata %storage, i64 %offset, metadata %var, metadata %expr)

This patch adds a new LLVM-local tag to DIExpressions, so we can detect
and pretty-print DIExpression metadata nodes.

What this patch doesn't do:

This patch does not touch the "Indirect" field in DIVariable; but moving
that into the expression would be a natural next step.

http://reviews.llvm.org/D4919
rdar://problem/17994491

Thanks to dblaikie and dexonsmith for reviewing this patch!

llvm-svn: 218778
2014-10-01 17:55:39 +00:00
Chad Rosier 70d54ac848 [AArch64] Improve cost model to handle sdiv by a pow-of-two.
This patch improves the target-specific cost model to better handle signed
division by a power of two. The immediate result is that this enables the SLP
vectorizer to do a better job.

http://reviews.llvm.org/D5469
PR20714

llvm-svn: 218607
2014-09-29 13:59:31 +00:00