| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
|
|
|
| |
Rewrote the SLP-vectorization as a whole-function vectorization pass. It is now able to vectorize chains across multiple basic blocks.
It still does not vectorize PHIs, but this should be easy to do now that we scan the entire function.
I removed the support for extracting values from trees.
We are now able to vectorize more programs, but there are some serious regressions in many workloads (such as flops-6 and mandel-2).
llvm-svn: 184647
|
|
|
|
|
|
|
|
| |
argument."
It doesn't work as I intended it to. This reverts commit r184638.
llvm-svn: 184641
|
|
|
|
|
|
| |
It has become an expensive operation. No functionality change.
llvm-svn: 184638
|
|
|
|
|
|
| |
Thanks to Bill Wendling for pointing this out!
llvm-svn: 184593
|
|
|
|
|
|
| |
PtrState.RRI private and delete the TODO.
llvm-svn: 184587
|
|
|
|
|
|
| |
several methods on PtrState.
llvm-svn: 184586
|
|
|
|
|
|
| |
PtrState.IsTrackingImpreciseRelease().
llvm-svn: 184583
|
|
|
|
|
|
| |
PtrState.{IsCFGHazardAfflicted,SetCFGHazardAfflicted}.
llvm-svn: 184582
|
|
|
|
|
|
| |
PtrState.GetReleaseMetadata() and PtrState.SetReleaseMetadata().
llvm-svn: 184534
|
|
|
|
|
|
| |
PtrState.IsTailCallRelease() and PtrState.SetTailCallRelease().
llvm-svn: 184533
|
|
|
|
|
|
|
|
|
|
|
|
| |
PtrState.IsKnownSafe and PtrState.SetKnownSafe.
This is apart of a series of patches to encapsulate PtrState.RRI and
make PtrState.RRI a private field of PtrState.
*NOTE* This is actually the second commit in the patch stream. I should
have put this note on the first such commit r184528.
llvm-svn: 184532
|
|
|
|
| |
llvm-svn: 184531
|
|
|
|
|
|
|
|
| |
RRInfo::Merge.
I also added some comments and performed minor code cleanups.
llvm-svn: 184528
|
|
|
|
|
|
| |
vector-register size.
llvm-svn: 184527
|
|
|
|
|
|
|
|
|
|
|
| |
This commit completely removes what is left of the simplify-libcalls
pass. All of the functionality has now been migrated to the instcombine
and functionattrs passes. The following C API functions are now NOPs:
1. LLVMAddSimplifyLibCallsPass
2. LLVMPassManagerBuilderSetDisableSimplifyLibCalls
llvm-svn: 184459
|
|
|
|
| |
llvm-svn: 184446
|
|
|
|
|
|
|
| |
We collect gather sequences when we vectorize basic blocks. Gather sequences are excellent
hints for vectorization of other basic blocks.
llvm-svn: 184444
|
|
|
|
|
|
| |
This change makes it easier to filter debug messages.
llvm-svn: 184440
|
|
|
|
|
|
|
|
| |
APFloat::isFiniteNonZero.
Turns out all the references were in llvm and not in clang.
llvm-svn: 184356
|
|
|
|
|
|
| |
caching it. The TLI may change between functions. No functionality change.
llvm-svn: 184352
|
|
|
|
|
|
| |
Register it with PassManager
llvm-svn: 184343
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Prior to this change, the considered addressing modes may be invalid since the
maximum and minimum offsets were not taking into account.
This was causing an assertion failure.
The added test case exercices that behavior.
<rdar://problem/14199725> Assertion failed: (CurScaleCost >= 0 && "Legal
addressing mode has an illegal cost!")
llvm-svn: 184341
|
|
|
|
|
|
| |
ExtractElementInst).
llvm-svn: 184325
|
|
|
|
|
|
|
|
| |
The type <3 x i8> is a common in graphics and we want to be able to vectorize it.
This changes accelerates bullet by 12% and 471_omnetpp by 5%.
llvm-svn: 184317
|
|
|
|
| |
llvm-svn: 184282
|
|
|
|
| |
llvm-svn: 184281
|
|
|
|
|
|
| |
roots.
llvm-svn: 184201
|
|
|
|
| |
llvm-svn: 184200
|
|
|
|
| |
llvm-svn: 184174
|
|
|
|
|
|
|
| |
vectorizing loops with memory accesses to non-zero address spaces. It
simply dropped the AS info. Fixes PR16306.
llvm-svn: 184103
|
|
|
|
| |
llvm-svn: 184089
|
|
|
|
| |
llvm-svn: 184084
|
|
|
|
| |
llvm-svn: 184044
|
|
|
|
| |
llvm-svn: 184041
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This pass was assuming that if hasAddressTaken() returns false for a
function, the function's only uses are call sites. That's not true
because there can be references by BlockAddresses too.
Fix the pass to handle this case. Fix
BlockAddress::replaceUsesOfWithOnConstant() to allow a function's type
to be changed by RAUW'ing the function with a bitcast of the recreated
function.
Patch by Mark Seaborn.
llvm-svn: 183933
|
|
|
|
|
|
| |
Should fix the dragonegg build bots.
llvm-svn: 183845
|
|
|
|
|
|
|
| |
Most clients have already been moved from Path V1 to V2. The ones using V1
now include PathV1.h explicitly.
llvm-svn: 183801
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Instead of a custom implementation of replaceAllUsesWith, we just call
replaceAllUsesWith and recreate llvm.used and llvm.compiler-used.
This change is particularity interesting because it makes llvm see
through what clang is doing with static used functions in extern "C"
contexts. With this change, running clang -O2 in
extern "C" {
__attribute__((used)) static void foo() {}
}
produces
@llvm.used = appending global [1 x i8*] [i8* bitcast (void ()* @foo to
i8*)], section "llvm.metadata"
define internal void @foo() #0 {
entry:
ret void
}
llvm-svn: 183756
|
|
|
|
|
|
|
| |
Variadic functions are particularly fragile in the face of ABI changes, so this
limits how much the pass changes them
llvm-svn: 183625
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
r183584 tries to derive some info from the code *AFTER* a call and apply
these derived info to the code *BEFORE* the call, which is not always safe
as the call in question may never return, and in this case, the derived
info is invalid.
Thank Duncan for pointing out this potential bug.
rdar://14073661
llvm-svn: 183606
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The MemCpyOpt pass is capable of optimizing:
callee(&S); copy N bytes from S to D.
into:
callee(&D);
subject to some legality constraints.
Assertion is triggered when the compiler tries to evalute "sizeof(typeof(D))",
while D is an opaque-typed, 'sret' formal argument of function being compiled.
i.e. the signature of the func being compiled is something like this:
T caller(...,%opaque* noalias nocapture sret %D, ...)
The fix is that when come across such situation, instead of calling some
utility functions to get the size of D's type (which will crash), we simply
assume D has at least N bytes as implified by the copy-instruction.
rdar://14073661
llvm-svn: 183584
|
|
|
|
|
|
|
|
| |
TopDownPathCount/BottomUpPathCount.
rdar://12480535
llvm-svn: 183489
|
|
|
|
| |
llvm-svn: 183461
|
|
|
|
|
|
| |
compiling chrome. This patch adds a new flag to enable vectorization on all levels and not only on -O3. It should go away once we make a decision.
llvm-svn: 183456
|
|
|
|
| |
llvm-svn: 183439
|
|
|
|
|
|
|
|
| |
little bit."
This reverts commit 183328. It caused pr16244 and broke the bots.
llvm-svn: 183422
|
|
|
|
| |
llvm-svn: 183363
|
|
|
|
| |
llvm-svn: 183360
|
|
|
|
| |
llvm-svn: 183328
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
IndVarSimplify is willing to move divide instructions outside of their
loop bodies if they are invariant of the loop. However, it may not be
safe to expand them if we do not know if they can trap.
Instead, check to see if it is not safe to expand the instruction and
skip the expansion.
This fixes PR16041.
Testcase by Rafael Ávila de Espíndola.
llvm-svn: 183239
|