The sqrt case is faster and we already do this for the case where
the exponent is 0.25. This adds the 0.75 case which is also not
sensitive to signed zeros.
Patch by Whitney Tsang (Whitney)
Differential revision: https://reviews.llvm.org/D57434
llvm-svn: 353557
Replace OR(SHL,SRL) pattern with ISD::FSHR (legalization expands this later if necessary) - this helps with the scale == 0 'undefined' drop-through case that was discussed on D55720.
llvm-svn: 353546
Make behavior of G_LOAD in widenScalar same as for G_ZEXTLOAD and
G_SEXTLOAD. That is perform widenScalarDst to size given by the target
and avoid additional checks in common code. Targets can reorder or add
additional rules in LegalizeRuleSet for the opcode to achieve desired
behavior.
Select extending load that does not have specified type of extension
into zero extending load.
Select truncating store that stores number of bytes indicated by size
in MachineMemoperand.
Differential Revision: https://reviews.llvm.org/D57454
llvm-svn: 353520
This is part of https://bugs.llvm.org/show_bug.cgi?id=40442.
Vector legalization is implemented for the add/sub overflow opcodes.
UMULO/SMULO are also handled as far as legalization is concerned, but
they don't support vector expansion yet (so no tests for them).
The vector result widening implementation is suboptimal, because it
could result in a legalization loop.
Differential Revision: https://reviews.llvm.org/D57639
llvm-svn: 353464
Move the (add (umax X, C), -C) --> (usubsat X, C) X86 combine into generic DAGCombiner
First of a number of saturated arithmetic folds that can be moved out of X86-specific code for PR40111.
Differential Revision: https://reviews.llvm.org/D57754
llvm-svn: 353457
This is pretty much directly ported from SelectionDAG. Doesn't include
the shift by non-constant but known bits version, since there isn't a
globalisel version of computeKnownBits yet.
This shows a disadvantage of targets not specifically which type
should be used for the shift amount. If type 0 is legalized before
type 1, the operations on the shift amount type use the wider type
(which are also less likely to legalize). This can be avoided by
targets specifying legalization actions on type 1 earlier than for
type 0.
llvm-svn: 353455
I noticed that we are missing this canonicalization in IR:
rL352515
...and then realized that we don't get this right in SDAG either,
so this has to be fixed first regardless of what we choose to do in IR.
The existing fold was limited to scalars and using the wrong predicate
to guard the transform. We have a boolean contents TLI query that can
be used to decide which direction to fold.
This may eventually lead back to the problems/question in:
https://bugs.llvm.org/show_bug.cgi?id=40486
...but it makes no difference to that yet.
Differential Revision: https://reviews.llvm.org/D57401
llvm-svn: 353433
Introduce a new function which handles instructions with multiple type
indices, but have the same number of vector elements.
Also legalize v2s16 shifts when applicable.
llvm-svn: 353432
Summary: This code tries to handle the case where IBB is an EHPad, but there's an earlier check that uses PBB->hasEHPadSuccessor(). Where PBB is a predecessor of IBB. The hasEHPadSuccessor function would have visited IBB and seen that it was an EHPad and returned false. This would prevent us from reaching this code with IBB as an EHPad.
Looks like this code was originally added in rL37427 (ancient) and made dead in rL143001.
Reviewers: rnk, void, efriedma
Reviewed By: rnk
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D57358
llvm-svn: 353375
There was a lot of repeated code wrt unary math intrinsics in
translateKnownIntrinsic. This factors out the repeated MIRBuilder code into
two functions: translateSimpleUnaryIntrinsic and getSimpleUnaryIntrinsicOpcode.
This simplifies adding simple unary intrinsics, since after this, all you have
to do is add the mapping to SimpleUnaryIntrinsicOpcodes.
Differential Revision: https://reviews.llvm.org/D57774
llvm-svn: 353316
Summary:
According to
https://docs.nvidia.com/cuda/archive/10.0/ptx-writers-guide-to-interoperability/index.html#cuda-specific-dwarf,
the compiler should emit the DW_AT_address_class attribute for all
variable and parameter. It means, that DW_AT_address_class attribute
should be used in the non-standard way to support compatibility with the
cuda-gdb debugger.
Clang is able to generate the information about the variable address
class. This information is emitted as the expression sequence
`DW_OP_constu <DWARF Address Space> DW_OP_swap DW_OP_xderef`. The patch
tries to find all such expressions and transform them into
`DW_AT_address_class <DWARF Address Space>` if target is NVPTX and the debugger is gdb.
If the expression is not found, then default values are used. For the
local variables <DWARF Address Space> is set to ADDR_local_space(6), for
the globals <DWARF Address Space> is set to ADDR_global_space(5). The
values are taken from the table in the same section 5.2. CUDA-Specific
DWARF Definitions.
Reviewers: echristo, probinson
Subscribers: jholewinski, aprantl, llvm-commits
Differential Revision: https://reviews.llvm.org/D57157
llvm-svn: 353203
This patch improves code generation for some AArch64 ACLE intrinsics. It adds
support to CGP to duplicate and sink operands to their user, if they can be
folded into a target instruction, like zexts and sub into usubl. It adds a
TargetLowering hook shouldSinkOperands, which looks at the operands of
instructions to see if sinking is profitable.
I decided to add a new target hook, as for the sinking to be profitable,
at least on AArch64, we have to look at multiple operands of an
instruction, instead of looking at the users of a zext for example.
The sinking is done in CGP, because it works around an instruction
selection limitation. If instruction selection is not limited to a
single basic block, this patch should not be needed any longer.
Alternatively this could be done in the LoopSink pass, which tries to
undo LICM for instructions in blocks that are not executed frequently.
Note that we do not force the operands to sink to have a single user,
because we duplicate them before sinking. Therefore this is only
desirable if they really can be done for free. Additionally we could
consider the impact on live ranges later on.
This should fix https://bugs.llvm.org/show_bug.cgi?id=40025.
As for performance, we have internal code that uses intrinsics and can
be speed up by 10% by this change.
Reviewers: SjoerdMeijer, t.p.northover, samparker, efriedma, RKSimon, spatel
Reviewed By: samparker
Differential Revision: https://reviews.llvm.org/D57377
llvm-svn: 353152
The fewerElementsVectors implementation for load/stores
handles the scalar reduction case just as well, so drop
the redundant code in narrowScalar. This also introduces
support for narrowing irregular size breakdowns for
scalars.
llvm-svn: 353125
Summary:
If the index isn't constant, this transform inserts a multiply and an add on the index to calculating the base pointer for a scalar load. But we still create a memory operand with an offset of 0 and the size of the scalar access. But the access is really to an unknown offset within the original access size.
This can cause the machine scheduler to incorrectly calculate dependencies between this load and other accesses. In the case we saw, there was a 32 byte vector store that was split into two 16 byte stores, one with offset 0 and one with offset 16. The size of the memory operand for both was 16. The scheduler correctly detected the alias with the offset 0 store, but not the offset 16 store.
This patch discards the pointer info so we don't incorrectly detect aliasing. I wasn't sure if we could keep using the original offset and size without risking some other transform on the load changing the size.
I tried to reduce a test case, but there's still a lot of memory operations needed to get the scheduler to do the bad reordering. So it looked pretty fragile to maintain.
Reviewers: efriedma
Reviewed By: efriedma
Subscribers: arphaman, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D57616
llvm-svn: 353124
Don't handle vector conditions.
I think this can be merged in the future with
fewerElementsVectorSelect, although this becomes slightly tricky with
a vector condition.
llvm-svn: 353122
Try to use the underlying source registers.
This enables legalization in more cases where some irregular
operations are widened and others narrowed.
This seems to make the test_combines_2 AArch64 test worse, since the
MERGE_VALUES has multiple uses. Since this should be required for
legalization, a hasOneUse check is probably inappropriate (or maybe
should only be used if the merge is legal?).
llvm-svn: 353121
A number of of tests were using imm operands, not cimm. Since CSE
relies on the exact ConstantInt* pointer used, and implicit
conversions are generally evil, also enforce the bitsize of the types.
llvm-svn: 353113
Aliases of functions are now marked as function symbols even if
they are bitcast to some other other non-function type.
This is important for WebAssembly where object and function
symbols can't alias each other.
Fixes PR38866
Differential Revision: https://reviews.llvm.org/D57538
llvm-svn: 353109
The LiveDebugValues pass recognizes spills but not restores, which can
cause large gaps in location information for some variables, depending
on control flow. This patch make LiveDebugValues recognize restores and
generate appropriate DBG_VALUE instructions.
This patch was posted previously with r352642 and reverted in r352666 due
to buildbot errors. A missing return statement was the cause for the
failures.
Reviewers: aprantl, NicolaPrica
Differential Revision: https://reviews.llvm.org/D57271
llvm-svn: 353089
This fixes two problems with CSE done in buildConstant. First, this
would hit an assert when used with a vector result type. Solve this by
allowing CSE on the vector elements, but not on the result vector for
now.
Second, this was also performing the CSE based on the input
ConstantInt pointer. The underlying buildConstant could potentially
convert the constant depending on the result type, giving in a
different ConstantInt*. Stop allowing the APInt and ConstantInt forms
from automatically casting to the result type to avoid any similar
problems in the future.
llvm-svn: 353077
This reverts commit 8bbd570fd5205a04d88d2e5513a6e4adbd028039.
Apparently adding ffloor breaks AMDGPU somehow, so I need to back this out
while I look into it.
llvm-svn: 353064
Add an intrinsic that takes 2 unsigned integers with the scale of them
provided as the third argument and performs fixed point multiplication on
them.
This is a part of implementing fixed point arithmetic in clang where some of
the more complex operations will be implemented as intrinsics.
Differential Revision: https://reviews.llvm.org/D55625
llvm-svn: 353059
This is no-functional-change-intended although there could
be intermediate variations caused by a difference in the
debug info produced by setting that from the builder's
insertion point.
I'm updating the IR test file associated with this code just
to show that the naming differences from using the builder
are visible.
The motivation for adding a helper function is that we are
likely to extend this code to deal with other overflow ops.
llvm-svn: 353056
