In some cases Clang does not perform merging of instructions AND and TST (aka
ANDS xzr).
Example:
tst x2, x1
and x3, x2, x1
to:
ands x3, x2, x1
This patch add such merging during instruction selection: when AND is replaced
with ANDS instruction in LowerSELECT_CC, all users of AND also should be
changed for using this ANDS instruction
Short discussion on mailing list:
http://llvm.1065342.n5.nabble.com/llvm-dev-ARM-Peephole-optimization-instructions-tst-add-tp133109.html
Patch by Pavel Kosov.
Differential Revision: https://reviews.llvm.org/D71701
I added test cases that rely on the availability of the PPC target into
the general directory for the loop vectorizer. This causes failures on
bots that don't build the PPC target. Moving them to the PowerPC directory
to fix this.
A recent commit
(https://reviews.llvm.org/rG66c120f02560ef528a60924104ead66f330190f1) changed
the cost for calls to functions that have a vector version for some
vectorization factor. However, no check is performed for whether the
vectorization factor matches the current one being cost modeled. This leads to
attempts to widen call instructions to a vectorization factor for which such a
function does not exist, which in turn leads to an assertion failure.
This patch adds the check for vectorization factor (i.e. not just that the
called function has a vector version for some VF, but that it has a vector
version for this VF).
Differential revision: https://reviews.llvm.org/D74944
These tests generate two nodes that are identical except for the
metadata that indicates their exception behavior. The nodes get
CSEd by SelectionDAG, but the nofpexcept flag is being set based
only on the fpexcept.ignore node being created last.
We need to detect this case and intersect the flags similar to
what is already done with fast math flags.
Add extra diagnostics for the inline cost analysis under
-print-instruction-deltas cl option. When enabled along with
-debug-only=inline-cost it prints the IR of inline candidate
annotated with cost and threshold change per every instruction.
Reviewed By: apilipenko, davidxl, mtrofin
Differential Revision: https://reviews.llvm.org/D71501
Previously this code was called into two ways, either a FrameIndexSDNode
was passed in StackSlot. Or a load node was passed in the argument
called StackSlot. This was determined by a dyn_cast to FrameIndexSDNode.
In the case of a load, we had to go find the real pointer from
operand 0 and cast the node to MemSDNode to find the pointer info.
For the stack slot case, the code assumed that the stack slot
was perfectly aligned despite not being the creator of the slot.
This commit modifies the interface to make the caller responsible
for passing all of the required information to avoid all the
guess work and reverse engineering.
I'm not aware of any issues with the original code after an
earlier commit to fix the alignment of one of the stack objects.
This is just clean up to make the code less surprising.
This allows for diagnosing malformed LLVM IR debug info metadata such
as the one in the testcase.
<rdar://problem/59756060>
Differential Revision: https://reviews.llvm.org/D75212
Since all types <32b on gpr end up being assigned gpr32 regclasses, we can end
up with PHIs here which try to select between a gpr32 and an fpr16. Ideally RBS
shouldn't be selecting heterogenous regbanks for operands if possible, but we
still need to be able to deal with it here.
To fix this, if we have a gpr-bank operand < 32b in size and at least one other
operand is on the fpr bank, then we add cross-bank copies to homogenize the
operand banks. For simplicity the bank that we choose to settle on is whatever
bank the def operand has. For example:
%endbb:
%dst:gpr(s16) = G_PHI %in1:gpr(s16), %bb1, %in2:fpr(s16), %bb2
=>
%bb2:
...
%in2_copy:gpr(s16) = COPY %in2:fpr(s16)
...
%endbb:
%dst:gpr(s16) = G_PHI %in1:gpr(s16), %bb1, %in2_copy:gpr(s16), %bb2
Differential Revision: https://reviews.llvm.org/D75086
Summary:
Current LLVM code base does not use error handler with ErrorPolicy.
This patch removes ErrorPolicy from DWARFContext.
This patch is extracted from the D74308.
Reviewers: jhenderson, dblaikie, grimar, aprantl, JDevlieghere
Reviewed By: grimar
Subscribers: hiraditya, llvm-commits
Tags: #llvm, #debug-info
Differential Revision: https://reviews.llvm.org/D75118
Having this as it's own function helps to reduce indentation and allows use of return instead of wiring a value over the switch. A lambda would have also worked, but with slightly deeper nesting.
This patch fixes a bug that would cause dsymutil to collect
.swiftinterface files for the Swift stdlib and other SDK
modules. There is no advantage in copying these since they should be
loaded from the ones bundled with LLDB's embedded Swift compiler
instead and copying them will cause LLDB to recompile them from source
instead of loading their prebuilt cached counterparts in the SDK.
rdar://problem/57463247
Differential Revisions: https://reviews.llvm.org/D75196
Summary: ParsingInlineAsm was a misleading name. These values are only set for MS-style inline assembly.
Reviewed By: rnk
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D75198
As suggested in D75145 -
I'm not sure why, but several passes have this kind of disable/enable flag
implemented at the pass manager level. But that means we have to duplicate
the flag for both pass managers and add code to check the flag every time
the pass appears in the pipeline.
We want a debug option to see if this pass is misbehaving regardless of the
pass managers, so just add a disablement check at the single point before
any transforms run.
Differential Revision: https://reviews.llvm.org/D75204
CVP currently does not simplify cmps with instructions in the same
block, because LVI getPredicateAt() currently does not provide
much useful information for that case (D69686 would change that,
but is stuck.) However, if the instruction is a Phi node, then
LVI can compute the result of the predicate by threading it into
the predecessor blocks, which allows it simplify some conditions
that nothing else can handle. Relevant code:
6d6a4590c5/llvm/lib/Analysis/LazyValueInfo.cpp (L1904-L1927)
Differential Revision: https://reviews.llvm.org/D72169
This is a small pet peeve from me. This change makes sure the AVR backend uses
the correct private label prefix (.L) so that private labels are hidden in
avr-objdump.
Example code:
define i8 @foo(i1 %cond) {
br i1 %cond, label %then, label %else
then:
ret i8 3
else:
ret i8 5
}
When compiling this:
llc -march=avr -filetype=obj -o test.o test.ll
and then dumping it:
avr-objdump -d test.o
You would previously get an ugly temporary label:
00000000 <foo>:
0: 81 70 andi r24, 0x01 ; 1
2: 80 30 cpi r24, 0x00 ; 0
4: f9 f3 breq .-2 ; 0x4 <foo+0x4>
6: 83 e0 ldi r24, 0x03 ; 3
8: 08 95 ret
0000000a <LBB0_2>:
a: 85 e0 ldi r24, 0x05 ; 5
c: 08 95 ret
This patch fixes that, the output is now:
00000000 <foo>:
0: 81 70 andi r24, 0x01 ; 1
2: 80 30 cpi r24, 0x00 ; 0
4: 01 f0 breq .+0 ; 0x6 <foo+0x6>
6: 83 e0 ldi r24, 0x03 ; 3
8: 08 95 ret
a: 85 e0 ldi r24, 0x05 ; 5
c: 08 95 ret
Note that as you can see the breq operand is different. However it is
still the same after linking:
4: 11 f0 breq .+4
Differential Revision: https://reviews.llvm.org/D75124
Adjusting by 2 breaks DWARF output. With this fix, programs start to
compile and produce valid DWARF output.
Differential Revision: https://reviews.llvm.org/D74213
InstCombine removes pairs of start+end intrinsics that don't
have anything in between them. Currently this is done by starting
at the start intrinsic and scanning forwards. This patch changes
it to start at the end intrinsic and scan backwards.
The motivation here is as follows: When we process the start
intrinsic, we have not yet looked at the following instructions,
which may still get folded/removed. If they do, we will only be
able to remove the start/end pair on the next iteration. When we
process the end intrinsic, all the instructions before it have
already been visited, and we don't run into this problem.
Differential Revision: https://reviews.llvm.org/D75011
MSVC does not define __LONG_MAX__, so we were just getting lucky in this
conditional:
#if __LONG_MAX__ == __INT_MAX__
Undefined identifiers evaluate to zero in preprocessor conditionals, so
this became true, which happens to work for MSVC platforms.
Instead, use this pattern and let the compiler constant fold:
return sizeof(long) == sizeof(int) ? SwapByteOrder_32((uint32_t)C)
: SwapByteOrder_64((uint64_t)C);
- Mark memory access for bool vectors as disallowed in target lowering.
This will prevent combining bitcasts of bool vectors with stores.
- Replace the actual bitcasting code with a faster version.
- Handle casting of v16i1 to i16.
DevirtSCCRepeatedPass iteration. Needs ReviewPublic
This aims to fix a missed inlining case.
If there's a virtual call in the callee on an alloca (stack allocated object) in
the caller, and the callee is inlined into the caller, the post-inline cleanup
would devirtualize the virtual call, but if the next iteration of
DevirtSCCRepeatedPass doesn't happen (under the new pass manager), which is
based on a heuristic to determine whether to reiterate, we may miss inlining the
devirtualized call.
This enables inlining in clang/test/CodeGenCXX/member-function-pointer-calls.cpp.
empty-sections.test: add two tests adapted from @jhenderson's https://reviews.llvm.org/D74755#1882221
strip-non-alloc.test: improve. D74755 will change the attribution of an empty section.
They mostly test the behavior of Object.cpp:sectionWithinSegment : how we attribute sections to segments.
`ParentSegment` can affect some subtle layout decisions.
Reviewed By: jhenderson
Differential Revision: https://reviews.llvm.org/D74879
This may inhibit vector narrowing in general, but there's
already an inconsistency in the way that we deal with this
pattern as shown by the test diff.
We may want to add a dedicated function for narrowing fneg.
It's often folded into some other op, so moving it away from
other math ops may cause regressions that we would not see
for normal binops.
See D73978 for more details.
Summary:
This patch introduces a function to house the code needed to do the
DWARF64 detection dance. The function decodes the initial length field
and returns it as a pair containing the actual length, and the DWARF
encoding.
This patch does _not_ attempt to handle the problem of detecting lengths
which extend past the size of the section, or cases when reads of a
single contribution accidentally escape beyond its specified length, but
I think it's useful in its own right.
Reviewers: dblaikie, jhenderson, ikudrin
Subscribers: hiraditya, probinson, aprantl, JDevlieghere, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D74560
The patch was reverted in 69da40033 because of test failures on windows.
The problem was the unpredictable order of some of the error messages,
which I've tried to strenghten in that patch.
It turns out this is not possible to do in verbose mode because there
the data is being writted as it is being parsed. No amount of flushing
(as I've done in the non-verbose mode) will help that. Indeed, even
without any buffering the warning messages can end in the middle of a
line in non-verbose mode.
In this patch, I have reverted the changes which tested the relative
position of the warning message, except for the messages about
unsupported initial length, which are the ones I really wanted to test,
and which do come out reasonably.
The original commit message was:
This patch if motivated by D74560, specifically the subthread about what
to print upon encountering reserved initial length values.
If the debug_line prologue has an unsupported version, we skip parsing
the rest of the data. If we encounter an reserved initial length field,
we don't even parse the version. However, we still print out all members
(with value 0) in the dump function.
This patch introduces early exits in the Prologue::dump function so that
we print only the fields that were parsed successfully. In case of an
unsupported version, we skip printing all subsequent prologue fields --
because we don't even know if this version has those fields. In case of a
reserved unit length, we don't print anything -- if the very first field
of the prologue is invalid, it's hard to say if we even have a prologue
to begin with.
Note that the user will still be able to see the invalid/reserved
initial length value in the error message. I've modified (reordered)
debug_line_invalid.test to show that the error message comes straight
after the debug_line offset. I've also added some flush() calls to the
dumping code to ensure this is the case in all situations (without that,
the warnings could get out of sync if the output was not a terminal -- I
guess this is why std::iostreams have the tie() function).
Reviewers: jhenderson, ikudrin, dblaikie
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D75043
