| Commit message (Collapse) | Author | Age | Files | Lines |
| ... | |
| |
|
|
|
|
|
|
| |
This helps because LoopAccessReport is passed around as a const
reference and we derive the basic block passed as the Value parameter
from the instruction in LoopAccessReport.
llvm-svn: 276191
|
| |
|
|
| |
llvm-svn: 276190
|
| |
|
|
|
|
|
|
| |
After r276153 the pass applies to both kernels and regular functions.
Differential Revision: https://reviews.llvm.org/D22583
llvm-svn: 276189
|
| |
|
|
|
|
|
| |
This adds an (incomplete, inefficient) framework for deciding what to do with
some operation on a given type.
llvm-svn: 276184
|
| |
|
|
|
|
|
|
|
|
|
| |
At -O0, cmpxchg survives AtomicExpand: it's mostly straightforward
to select it in fast-isel, and let the pseudo be expanded later.
extractvalues on the result are the tricky part: the generic logic
only works for legal types (and it would be painful to make it
support illegal types), so we can only support i32/i64 cmpxchg.
llvm-svn: 276183
|
| |
|
|
| |
llvm-svn: 276182
|
| |
|
|
|
|
|
|
|
| |
We hoisted PHIs without respecting their special insertion point in the
block, leading to verfier errors.
This fixes PR28626.
llvm-svn: 276181
|
| |
|
|
|
|
|
|
|
| |
We can replace the return values with undef if we replaced all
the call uses with a constant/undef.
Differential Revision: https://reviews.llvm.org/D22336
llvm-svn: 276174
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
moving loads.
Summary:
Previously we wouldn't move loads/stores across instructions that had
side-effects, where that was defined as may-write or may-throw. But
this is not sufficiently restrictive: Stores can't safely be moved
across instructions that may load.
This patch also adds a DEBUG check that all instructions in our chain
are either loads or stores.
Reviewers: asbirlea
Subscribers: llvm-commits, jholewinski, arsenm, mzolotukhin
Differential Revision: https://reviews.llvm.org/D22547
llvm-svn: 276171
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Summary:
Previously if we had a chain that contained a side-effecting
instruction, we wouldn't vectorize it at all. Now we'll vectorize
everything that comes before the side-effecting instruction.
Reviewers: asbirlea
Subscribers: arsenm, jholewinski, llvm-commits, mzolotukhin
Differential Revision: https://reviews.llvm.org/D22536
llvm-svn: 276170
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A seemingly common use for the walker's getClobberingMemoryAccess
function is:
```
MemoryAccess *getClobber(MemorySSAWalker *W, MemoryUseOrDef *MUD) {
const Instruction *I = MUD->getMemoryInst();
return W->getClobberingMemoryAccess(I);
}
```
Which is kind of redundant, since walkers will ultimately query MSSA to
find out which MemoryAccess `I` maps to (...which is always `MUD`).
So, this patch adds an overload of getClobberingMemoryAccess that
accepts MemoryAccesses directly. As a result, the Instruction overload
of getClobberingMemoryAccess becomes a lightweight wrapper around our
new overload.
Additionally, this patch un`virtual`izes the Instruction overload of
getClobberingMemoryAccess, since there doesn't seem to be a walker that
benefits from that being virtual, and I can't think of how else one
would implement it. Happy to make it virtual again if we would benefit
from doing so.
llvm-svn: 276169
|
| |
|
|
|
|
|
|
| |
This should be all the low-level instruction selection needs to determine how
to implement an operation, with the remaining context taken from the opcode
(e.g. G_ADD vs G_FADD) or other flags not based on type (e.g. fast-math).
llvm-svn: 276158
|
| |
|
|
| |
llvm-svn: 276156
|
| |
|
|
|
|
|
|
| |
Fixes a crash in llvm_unreachable when a function has array return type.
Differential Revision: https://reviews.llvm.org/D22524
llvm-svn: 276154
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Avoid unnecessary spills of byval arguments of device functions to
local space on SASS level and subsequent pointer conversion to generic
address space that follows. Instead, make a local copy in IR, provide
a way to access arguments directly, and let LLVM optimize the copy away
when possible.
Differential Review: https://reviews.llvm.org/D21421
llvm-svn: 276153
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Summary:
Mirroring most cleanup changed from compiler-rt/lib/builtins/cpu_model.
x86 methods are still returning a bool.
Reviewers: llvm-commits, echristo, craig.topper, sanjoy
Subscribers: mehdi_amini
Differential Revision: https://reviews.llvm.org/D22480
llvm-svn: 276149
|
| |
|
|
|
|
|
|
| |
As noted in https://reviews.llvm.org/D22537 , we can use this functionality in
visitSelectInstWithICmp() and InstSimplify, but currently we have duplicated
code.
llvm-svn: 276140
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In D12090, the ExprValueMap was added to reuse existing value during SCEV expansion.
However, const folding and sext/zext distribution can make the reuse still difficult.
A simplified case is: suppose we know S1 expands to V1 in ExprValueMap, and
S1 = S2 + C_a
S3 = S2 + C_b
where C_a and C_b are different SCEVConstants. Then we'd like to expand S3 as
V1 - C_a + C_b instead of expanding S2 literally. It is helpful when S2 is a
complex SCEV expr and S2 has no entry in ExprValueMap, which is usually caused
by the fact that S3 is generated from S1 after const folding.
In order to do that, we represent ExprValueMap as a mapping from SCEV to
ValueOffsetPair. We will save both S1->{V1, 0} and S2->{V1, C_a} into the
ExprValueMap when we create SCEV for V1. When S3 is expanded, it will first
expand S2 to V1 - C_a because of S2->{V1, C_a} in the map, then expand S3 to
V1 - C_a + C_b.
Differential Revision: https://reviews.llvm.org/D21313
llvm-svn: 276136
|
| |
|
|
| |
llvm-svn: 276135
|
| |
|
|
|
|
| |
Differential Revision: https://reviews.llvm.org/D22526
llvm-svn: 276119
|
| |
|
|
|
|
|
|
| |
Makes InstCombine infloop when compiling v8.
This reverts commit r275989 and r276105.
llvm-svn: 276106
|
| |
|
|
|
|
|
|
| |
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
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Retry r275776 (no changes, we suspect the issue was with another commit).
The current logic for handling inline asm operands in DAGToDAGISel interprets
the operands by looking for constants, which should represent the flags
describing the kind of operand we're dealing with (immediate, memory, register
def etc). The operands representing actual data are skipped only if they are
non-const, with the exception of immediate operands which are skipped explicitly
when a flag describing an immediate is found.
The oversight is that memory operands may be const too (e.g. for device drivers
reading a fixed address), so we should explicitly skip the operand following a
flag describing a memory operand. If we don't, we risk interpreting that
constant as a flag, which is definitely not intended.
Fixes PR26038
Differential Revision: https://reviews.llvm.org/D22103
llvm-svn: 276101
|
| |
|
|
|
|
| |
pattern.
llvm-svn: 276088
|
| |
|
|
|
|
| |
without relying on file ordering.
llvm-svn: 276087
|
| |
|
|
|
|
| |
VEXTRACT(F/I)128/VINSERT(I/F)128. NFC
llvm-svn: 276086
|
| |
|
|
|
|
| |
instructions with less repeated code. NFC
llvm-svn: 276085
|
| |
|
|
|
|
|
|
|
| |
Inference of the 'returned' attribute was fixed in r276008, lets try
turning the backend support back on.
This reverts commit r275677.
llvm-svn: 276081
|
| |
|
|
|
|
|
| |
The new OptimizationRemarkEmitter analysis pass is hooked up to both new
and old PM passes.
llvm-svn: 276080
|
| |
|
|
|
|
|
| |
This reverts commit r276064, and thus reapplies r275891 and r275883 with
a fix for PR28608.
llvm-svn: 276077
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Summary:
getVectorizablePrefix previously didn't work properly in the face of
aliasing loads/stores. It unwittingly assumed that the loads/stores
appeared in the BB in address order. If they didn't, it would do the
wrong thing.
Reviewers: asbirlea, tstellarAMD
Subscribers: arsenm, llvm-commits, mzolotukhin
Differential Revision: https://reviews.llvm.org/D22535
llvm-svn: 276072
|
| |
|
|
|
|
|
|
|
| |
Reverting this commit for now as it seems to be causing failures on
test-suite tests on the clang-ppc64le-linux-lnt bot.
This reverts commit r276044.
llvm-svn: 276068
|
| |
|
|
|
|
|
|
|
| |
Add a check that the layout predecessor of a block is an actual CFG
predecssor of the block as well. No current code fails this check, but
upcoming patches can trigger this, and it makes sense to separate it
out.
llvm-svn: 276066
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Revert "[LoopSimplify] Update LCSSA after separating nested loops."
This reverts commit r275891.
Revert "[LCSSA] Post-process PHI-nodes created by SSAUpdate when constructing LCSSA form."
This reverts commit r275883.
llvm-svn: 276064
|
| |
|
|
|
|
|
|
|
| |
We just set PreserveLCSSA to always true since we don't have an
analogous method `mustPreserveAnalysisID(LCSSA)`.
Also port LoopInfo verifier pass to test LoopUnrollPass.
llvm-svn: 276063
|
| |
|
|
|
|
|
|
| |
canTailDuplicate accepts two blocks and returns true if the first can be
duplicated into the second successfully. Use this function to
encapsulate the heuristic.
llvm-svn: 276062
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Summary: substr doesn't modify the string, so this line has no effect.
Reviewers: majnemer
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D22540
llvm-svn: 276057
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Summary:
Previously, the insertion point for stores was the last instruction in
Chain *before calling getVectorizablePrefixEndIdx*. Thus if
getVectorizablePrefixEndIdx didn't return Chain.size(), we still would
insert at the last instruction in Chain.
This patch changes our internal API a bit in an attempt to make it less
prone to this sort of error. As a result, we end up recalculating the
Chain's boundary instructions, but I think worrying about the speed hit
of this is a premature optimization right now.
Reviewers: asbirlea, tstellarAMD
Subscribers: mzolotukhin, arsenm, llvm-commits
Differential Revision: https://reviews.llvm.org/D22534
llvm-svn: 276056
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Summary:
The DEBUG message was hard to read because two Values were being printed
on the same line with only the delimiter "aliases". This change makes
us print each Value on its own line.
Reviewers: asbirlea
Subscribers: llvm-commits, arsenm, mzolotukhin
Differential Revision: https://reviews.llvm.org/D22533
llvm-svn: 276055
|
| |
|
|
|
|
|
|
|
|
| |
Reviewers: asbirlea
Subscribers: mzolotukhin, llvm-commits, arsenm
Differential Revision: https://reviews.llvm.org/D22532
llvm-svn: 276054
|
| |
|
|
| |
llvm-svn: 276052
|
| |
|
|
|
|
|
|
|
|
|
| |
If 2.5 ulp is acceptable, denormals are not required, and
isn't a reciprocal which will already be handled, replace
with a faster fdiv.
Simplify the lowering tests by using per function
subtarget features.
llvm-svn: 276051
|
| |
|
|
| |
llvm-svn: 276050
|
| |
|
|
|
|
| |
Differential Revision: http://reviews.llvm.org/D22545
llvm-svn: 276048
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Summary: Make MemorySSA::dominates/locallydominates constant time
Reviewers: george.burgess.iv, gberry
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D22527
llvm-svn: 276046
|
| |
|
|
|
|
|
|
| |
Patch by Andrew Paprocki!
Differential Revision: https://reviews.llvm.org/D18359
llvm-svn: 276045
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is a variant of scavengeRegister() that works for
enterBasicBlockEnd()/backward(). The benefit of the backward mode is
that it is not affected by incomplete kill flags.
This patch also changes
PrologEpilogInserter::doScavengeFrameVirtualRegs() to use the register
scavenger in backwards mode.
Differential Revision: http://reviews.llvm.org/D21885
llvm-svn: 276044
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This adds two pieces:
- RegisterScavenger:::enterBasicBlockEnd() which behaves similar to
enterBasicBlock() but starts tracking at the end of the basic block.
- A RegisterScavenger::backward() method. It is subtly different
from the existing unprocess() method which only considers uses with
the kill flag set: If a value is dead at the end of a basic block with
a last use inside the basic block, unprocess() will fail to mark it as
live. However we cannot change/fix this behaviour because unprocess()
needs to perform the exact reverse operation of forward().
Differential Revision: http://reviews.llvm.org/D21873
llvm-svn: 276043
|
| |
|
|
|
|
|
|
| |
Add check for legal data types when expanding into a Newton series.
Differential Revision: https://reviews.llvm.org/D22267
llvm-svn: 276041
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
the source type
The pattern may look more obviously like a sext if written as:
define i32 @g(i16 %x) {
%zext = zext i16 %x to i32
%xor = xor i32 %zext, 32768
%add = add i32 %xor, -32768
ret i32 %add
}
We already have that fold in visitAdd().
Differential Revision: https://reviews.llvm.org/D22477
llvm-svn: 276035
|