Prevents hoisting of loads above stores and sinking of stores below loads
in MergedLoadStoreMotion.cpp (rdar://15991737)

llvm-svn: 213497

Thanks to the lld-x86_64-darwin13 builder for catching this first.

llvm-svn: 213488

Prior to this change, the loop vectorizer did not make use of the alias
analysis infrastructure. Instead, it performed memory dependence analysis using
ScalarEvolution-based linear dependence checks within equivalence classes
derived from the results of ValueTracking's GetUnderlyingObjects.

Unfortunately, this meant that:
  1. The loop vectorizer had logic that essentially duplicated that in BasicAA
     for aliasing based on identified objects.
  2. The loop vectorizer could not partition the space of dependency checks
     based on information only easily available from within AA (TBAA metadata is
     currently the prime example).

This means, for example, regardless of whether -fno-strict-aliasing was
provided, the vectorizer would only vectorize this loop with a runtime
memory-overlap check:

void foo(int *a, float *b) {
  for (int i = 0; i < 1600; ++i)
    a[i] = b[i];
}

This is suboptimal because the TBAA metadata already provides the information
necessary to show that this check unnecessary. Of course, the vectorizer has a
limit on the number of such checks it will insert, so in practice, ignoring
TBAA means not vectorizing more-complicated loops that we should.

This change causes the vectorizer to use an AliasSetTracker to keep track of
the pointers in the loop. The resulting alias sets are then used to partition
the space of dependency checks, and potential runtime checks; this results in
more-efficient vectorizations.

When pointer locations are added to the AliasSetTracker, two things are done:
  1. The location size is set to UnknownSize (otherwise you'd not catch
     inter-iteration dependencies)
  2. For instructions in blocks that would need to be predicated, TBAA is
     removed (because the metadata might have a control dependency on the condition
     being speculated).

For non-predicated blocks, you can leave the TBAA metadata. This is safe
because you can't have an iteration dependency on the TBAA metadata (if you
did, and you unrolled sufficiently, you'd end up with the same pointer value
used by two accesses that TBAA says should not alias, and that would yield
undefined behavior).

llvm-svn: 213486

Summary: This patch introduces two new iterator ranges and updates existing code to use it.  No functional change intended.

Test Plan: All tests (make check-all) still pass.

Reviewers: dblaikie

Reviewed By: dblaikie

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D4481

llvm-svn: 213474

llvm-svn: 213456

IRBuilder has CreateAligned(Load|Store) functions; use them and we don't need
to make a second call to setAlignment.

No functionality change intended.

llvm-svn: 213453

There are some kinds of metadata that are safe to propagate from the scalar
instructions to the vector instructions (fpmath and tbaa currently).

Regarding TBAA, one might worry about propagating it on if-converted loads and
stores, because the metadata might have had a control dependency on the
condition, and thus actually aliased with some other non-speculated memory
access when the condition was false. However, this would be caught by the
runtime overlap checks.

llvm-svn: 213452

llvm-svn: 213448

llvm-svn: 213414

llvm-svn: 213412

Merges equivalent loads on both sides of a hammock/diamond
and hoists into into the header.
Merges equivalent stores on both sides of a hammock/diamond
and sinks it to the footer.
Can enable if conversion and tolerate better load misses
and store operand latencies.

llvm-svn: 213396

This is used to avoid instrumentation of instructions added by UBSan
in Clang frontend (see r213291). This fixes PR20085.

Reviewed in http://reviews.llvm.org/D4544.

llvm-svn: 213292

Origin is meaningless for fully initialized values. Avoid
storing origin for function arguments that are known to
be always initialized (i.e. shadow is a compile-time null
constant).

This is not about correctness, but purely an optimization.
Seems to affect compilation time of blacklisted functions
significantly.

llvm-svn: 213239

Refactor code, no functionality change, test case moved from instcombine to instsimplify.

Differential Revision: http://reviews.llvm.org/D4102
 

llvm-svn: 213231

llvm-svn: 213228

Summary:
Converting outermost zext(a) to sext(a) causes worse code when the
computation of zext(a) could be reused. For example, after converting

... = array[zext(a)]
... = array[zext(a) + 1]

to

... = array[sext(a)]
... = array[zext(a) + 1],

the program computes sext(a), which is actually unnecessary. I added one
test in split-gep-and-gvn.ll to illustrate this scenario.

Also, with r211281 and r211084, we annotate more "nuw" tags to
computation involving CUDA intrinsics such as threadIdx.x. These
annotations help with splitting GEP a lot, rendering the benefit we get
from this reverted optimization only marginal.

Test Plan: make check-all

Reviewers: eliben, meheff

Reviewed By: meheff

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D4542

llvm-svn: 213209

llvm-svn: 213189

In the original version of the patch the behaviour was like described in
the comment.  This behaviour was changed before committing it without
updating the comment.

llvm-svn: 213117

This patch modifies the existing DiagnosticInfo system to create a generic base
class that is inherited to produce diagnostic-based warnings. This is used by
the loop vectorizer to trigger a warning when vectorization is forced and
fails. Several tests have been added to verify this behavior.

Reviewed by: Arnold Schwaighofer

llvm-svn: 213110

Specifically, do not compute a union if it is statically known that one
shadow set subsumes the other.

llvm-svn: 213100

Phabricator ticket: D4246, Don't merge functions with different range metadata on call/invoke.
Thanks!

llvm-svn: 213060

Specifically, when building a union query, if we are dominated by an identical
query then use the result of that query instead.

llvm-svn: 213047

llvm-svn: 213046

llvm-svn: 213045

Don't eliminate memcpy's when the address of the pointer may itself be relevant. Fixes PR18304. Patch by David Wiberg!

llvm-svn: 212970

llvm-svn: 212963

llvm-svn: 212904

not properly handle the case where the predecessor block was the entry block to
the function.  The only in-tree client of this is JumpThreading, which worked
around the issue in its own code.  This patch moves the solution into the helper
so that JumpThreading (and other clients) do not have to replicate the same fix
everywhere.

llvm-svn: 212875

Currently ASan instrumentation pass creates a string with global name
for each instrumented global (to include global names in the error report). Global
name is already mangled at this point, and we may not be able to demangle it
at runtime (e.g. there is no __cxa_demangle on Android).

Instead, create a string with fully qualified global name in Clang, and pass it
to ASan instrumentation pass in llvm.asan.globals metadata. If there is no metadata
for some global, ASan will use the original algorithm.

This fixes https://code.google.com/p/address-sanitizer/issues/detail?id=264.

llvm-svn: 212872

No functionality change.

llvm-svn: 212850

When we sink an instruction, this can open up opportunity for the operands to be sunk - add them to the worklist

llvm-svn: 212847

This commit fixes bug http://llvm.org/bugs/show_bug.cgi?id=20103.

Thanks to Qwertyuiop for the report and the proposed fix.

llvm-svn: 212802

Partially fix PR20058: reduce compile time for loop unrolling with very high count by reducing calls to SE->forgetLoop

llvm-svn: 212782

Fix a crash in `InstCombiner::Descale()` when a multiply-by-zero gets
created as an argument to a GEP partway through an iteration, causing
-instcombine to optimize the GEP before the multiply.

rdar://problem/17615671

llvm-svn: 212742

This is the one remaining place I see where passing
isSafeToSpeculativelyExecute a DataLayout pointer might matter (at least for
loads) -- I think I got the others in r212720. Most of the other remaining
callers of isSafeToSpeculativelyExecute only use it for call sites (or
otherwise exclude loads).

llvm-svn: 212730

isSafeToSpeculativelyExecute can optionally take a DataLayout pointer. In the
past, this was mainly used to make better decisions regarding divisions known
not to trap, and so was not all that important for users concerned with "cheap"
instructions. However, now it also helps look through bitcasts for
dereferencable loads, and will also be important if/when we add a
dereferencable pointer attribute.

This is some initial work to feed a DataLayout pointer through to callers of
isSafeToSpeculativelyExecute, generally where one was already available.

llvm-svn: 212720

isDereferenceablePointer should not give up upon encountering any bitcast. If
we're casting from a pointer to a larger type to a pointer to a small type, we
can continue by examining the bitcast's operand. This missing capability
was noted in a comment in the function.

In order for this to work, isDereferenceablePointer now takes an optional
DataLayout pointer (essentially all callers already had such a pointer
available). Most code uses isDereferenceablePointer though
isSafeToSpeculativelyExecute (which already took an optional DataLayout
pointer), and to enable the LICM test case, LICM needs to actually provide its DL
pointer to isSafeToSpeculativelyExecute (which it was not doing previously).

llvm-svn: 212686

llvm-svn: 212668

Turn llvm::SpecialCaseList into a simple class that parses text files in
a specified format and knows nothing about LLVM IR. Move this class into
LLVMSupport library. Implement two users of this class:
  * DFSanABIList in DFSan instrumentation pass.
  * SanitizerBlacklist in Clang CodeGen library.
The latter will be modified to use actual source-level information from frontend
(source file names) instead of unstable LLVM IR things (LLVM Module identifier).

Remove dependency edge from ClangCodeGen/ClangDriver to LLVMTransformUtils.

No functionality change.

llvm-svn: 212643

In PR20059 ( http://llvm.org/pr20059 ), instcombine eliminates shuffles that are necessary before performing an operation that can trap (srem).

This patch calls isSafeToSpeculativelyExecute() and bails out of the optimization in SimplifyVectorOp() if needed.

Differential Revision: http://reviews.llvm.org/D4424

llvm-svn: 212629