The inline sequence is very long (about 70 bytes on Thumb1), so it's
not really a good idea to inline it, especially when optimizing for
size.
Differential Revision: https://reviews.llvm.org/D47917
llvm-svn: 340458
On Windows, movw+movt pairs with relocations are handled with a single
relocation that covers them both. Therefore we can't inject anything
between these instructions, otherwise the relocation (which in LLVM
only is treated as the movw instruction's relocation, while the movt
instruction's relocation is dropped) will end up bogus.
These instructions are bundled up until right before the constant
islands pass, making this effectively the only place that can split
them apart.
Differential Revision: https://reviews.llvm.org/D51032
llvm-svn: 340451
This avoids a potential infinite loop setting and unsetting bits in the
mask.
Reduced from a failure on the polly-aosp bot.
Differential Revision: https://reviews.llvm.org/D51066
llvm-svn: 340446
Add intrinsic isel patterns for sxtb16, sxtab16, uxtb16 and uxtab16
so that they can perform a ror.
Differential Revision: https://reviews.llvm.org/D51034
llvm-svn: 340405
During combining, ReduceLoadWdith is used to combine AND nodes that
mask loads into narrow loads. This patch allows the mask to be a
shifted constant. This results in a narrow load which is then left
shifted to compensate for the new offset.
Differential Revision: https://reviews.llvm.org/D50432
llvm-svn: 340261
There is no way in the universe, that doing a full-width division in
software will be faster than doing overflowing multiplication in
software in the first place, especially given that this same full-width
multiplication needs to be done anyway.
This patch replaces the previous implementation with a direct lowering
into an overflowing multiplication algorithm based on half-width
operations.
Correctness of the algorithm was verified by exhaustively checking the
output of this algorithm for overflowing multiplication of 16 bit
integers against an obviously correct widening multiplication. Baring
any oversights introduced by porting the algorithm to DAG, confidence in
correctness of this algorithm is extremely high.
Following table shows the change in both t = runtime and s = space. The
change is expressed as a multiplier of original, so anything under 1 is
“better” and anything above 1 is worse.
+-------+-----------+-----------+-------------+-------------+
| Arch | u64*u64 t | u64*u64 s | u128*u128 t | u128*u128 s |
+-------+-----------+-----------+-------------+-------------+
| X64 | - | - | ~0.5 | ~0.64 |
| i686 | ~0.5 | ~0.6666 | ~0.05 | ~0.9 |
| armv7 | - | ~0.75 | - | ~1.4 |
+-------+-----------+-----------+-------------+-------------+
Performance numbers have been collected by running overflowing
multiplication in a loop under `perf` on two x86_64 (one Intel Haswell,
other AMD Ryzen) based machines. Size numbers have been collected by
looking at the size of function containing an overflowing multiply in
a loop.
All in all, it can be seen that both performance and size has improved
except in the case of armv7 where code size has regressed for 128-bit
multiply. u128*u128 overflowing multiply on 32-bit platforms seem to
benefit from this change a lot, taking only 5% of the time compared to
original algorithm to calculate the same thing.
The final benefit of this change is that LLVM is now capable of lowering
the overflowing unsigned multiply for integers of any bit-width as long
as the target is capable of lowering regular multiplication for the same
bit-width. Previously, 128-bit overflowing multiply was the widest
possible.
Patch by Simonas Kazlauskas!
Differential Revision: https://reviews.llvm.org/D50310
llvm-svn: 339922
While searching through the use-def tree, ignore GetElementPtrInst
instructions because they don't need promoting and neither do their
indices. Otherwise, the wide indices prevent the transformation from
happening.
Differential Revision: https://reviews.llvm.org/D50762
llvm-svn: 339871
Originally committed in r339755 which was reverted in r339806 due to
an asan issue. The issue was caused by my assumption that operands to
a CallInst mapped to the FunctionType Params. CallInsts are now
handled by iterating over their ArgOperands instead of Operands.
Original Message:
Treat signed icmps as 'sinks', allowing them to be in the use-def
tree, enabling more promotions to be performed. As a sink, any
promoted incoming values need to be truncated before being used by
the signed icmp.
Differential Revision: https://reviews.llvm.org/D50067
llvm-svn: 339858
We only try to promote types with are smaller than 16-bits, but we
also need to check that the type is not less than 8-bits.
Differential Revision: https://reviews.llvm.org/D50769
llvm-svn: 339770
Treat signed icmps as 'sinks', allowing them to be in the use-def
tree, enabling more promotions to be performed. As a sink, any
promoted incoming values need to be truncated before being used by
the signed icmp.
Differential Revision: https://reviews.llvm.org/D50067
llvm-svn: 339755
Add pointers to the list of allowed types, but don't try to promote
them. Also fixed a bug with the promotion of undef values, so a new
value is now created instead of mutating in place. We also now only
promote if there's an instruction in the use-def chains other than
the icmp, sinks and sources.
Differential Revision: https://reviews.llvm.org/D50054
llvm-svn: 339754
LLVM normally prefers to minimize the number of bits set in an AND
immediate, but that doesn't always match the available ARM instructions.
In Thumb1 mode, prefer uxtb or uxth where possible; otherwise, prefer
a two-instruction sequence movs+ands or movs+bics.
Some potential improvements outlined in
ARMTargetLowering::targetShrinkDemandedConstant, but seems to work
pretty well already.
The ARMISelDAGToDAG fix ensures we don't generate an invalid UBFX
instruction due to a larger-than-expected mask. (It's orthogonal, in
some sense, but as far as I can tell it's either impossible or nearly
impossible to reproduce the bug without this change.)
According to my testing, this seems to consistently improve codesize by
a small amount by forming bic more often for ISD::AND with an immediate.
Differential Revision: https://reviews.llvm.org/D50030
llvm-svn: 339472
Enabling ARMCodeGenPrepare by default caused a whole load of
failures. This is due to zexts and truncs not being handled properly.
ZExts are messy so it's just easier to disable for now and truncs
are allowed only as 'sinks'. I still need to figure out why allowing
them as 'sources' causes so many failures. The other main changes are
that we are explicit in the types that we converting to, it's now
always 'TypeSize'. Type support is also now performed while checking
for valid opcodes as it unnecessarily complicated having the checks
are different stages.
I've moved the tests around too, so we have the zext and truncs in
their own file as well as the overflowing opcode tests.
Differential Revision: https://reviews.llvm.org/D50518
llvm-svn: 339432
LLVM triple normalization is handling "unknown" and empty components
differently; for example given "x86_64-unknown-linux-gnu" and
"x86_64-linux-gnu" which should be equivalent, triple normalization
returns "x86_64-unknown-linux-gnu" and "x86_64--linux-gnu". autoconf's
config.sub returns "x86_64-unknown-linux-gnu" for both
"x86_64-linux-gnu" and "x86_64-unknown-linux-gnu". This changes the
triple normalization to behave the same way, replacing empty triple
components with "unknown".
This addresses PR37129.
Differential Revision: https://reviews.llvm.org/D50219
llvm-svn: 339294
Normally, if any registers are spilled, we prefer to spill lr on Thumb1
so we can fold the "bx lr" into the "pop". However, if there are tail
calls involved, restoring lr is expensive, so skip the optimization in
that case.
The spill of r7 in the new test also isn't necessary, but that's
mostly orthogonal to this patch. (It's the same code in
ARMFrameLowering, but it's not related to tail calls.)
Differential Revision: https://reviews.llvm.org/D49459
llvm-svn: 339283
Summary:
Currently, in line with GCC, when specifying reserved registers like sp or pc on an inline asm() clobber list, we don't always preserve the original value across the statement. And in general, overwriting reserved registers can have surprising results.
For example:
```
extern int bar(int[]);
int foo(int i) {
int a[i]; // VLA
asm volatile(
"mov r7, #1"
:
:
: "r7"
);
return 1 + bar(a);
}
```
Compiled for thumb, this gives:
```
$ clang --target=arm-arm-none-eabi -march=armv7a -c test.c -o - -S -O1 -mthumb
...
foo:
.fnstart
@ %bb.0: @ %entry
.save {r4, r5, r6, r7, lr}
push {r4, r5, r6, r7, lr}
.setfp r7, sp, #12
add r7, sp, #12
.pad #4
sub sp, #4
movs r1, #7
add.w r0, r1, r0, lsl #2
bic r0, r0, #7
sub.w r0, sp, r0
mov sp, r0
@APP
mov.w r7, #1
@NO_APP
bl bar
adds r0, #1
sub.w r4, r7, #12
mov sp, r4
pop {r4, r5, r6, r7, pc}
...
```
r7 is used as the frame pointer for thumb targets, and this function needs to restore the SP from the FP because of the variable-length stack allocation a. r7 is clobbered by the inline assembly (and r7 is included in the clobber list), but LLVM does not preserve the value of the frame pointer across the assembly block.
This type of behavior is similar to GCC's and has been discussed on the bugtracker: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=11807 . No consensus seemed to have been reached on the way forward. Clang behavior has briefly been discussed on the CFE mailing (starting here: http://lists.llvm.org/pipermail/cfe-dev/2018-July/058392.html). I've opted for following Eli Friedman's advice to print warnings when there are reserved registers on the clobber list so as not to diverge from GCC behavior for now.
The patch uses MachineRegisterInfo's target-specific knowledge of reserved registers, just before we convert the inline asm string in the AsmPrinter.
If we find a reserved register, we print a warning:
```
repro.c:6:7: warning: inline asm clobber list contains reserved registers: R7 [-Winline-asm]
"mov r7, #1"
^
```
Reviewers: eli.friedman, olista01, javed.absar, efriedma
Reviewed By: efriedma
Subscribers: efriedma, eraman, kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D49727
llvm-svn: 339257
Summary: Extend fix for PR34170 to support inline assembly with multiple output operands that do not naturally go in the register class it is constrained to (eg. double in a 32-bit GPR as in the PR).
Reviewers: bogner, t.p.northover, lattner, javed.absar, efriedma
Reviewed By: efriedma
Subscribers: efriedma, tra, eraman, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D45437
llvm-svn: 339225
Clang support for the Armv8.2-A FP16 vector intrinsic was committed in
rC328277, but this was never followed up, i.e. the LLVM part is missing.
I've raised PR38404, and this is the first step to address this. I.e.,
this adds tests for the Armv8.2-A FP16 vector intrinsic, and thus shows
which intrinsics already work, and which need further work.
Differential Revision: https://reviews.llvm.org/D50142
llvm-svn: 338568
Disable ARMCodeGenPrepare by default again. It is causing verifier
failues in V8 that look like:
Duplicate integer as switch case
switch i32 %trunc, label %if.end13 [
i32 0, label %cleanup36
i32 0, label %if.then8
], !dbg !4981
i32 0
fatal error: error in backend: Broken function found, compilation aborted!
I will continue reducing the test case and send it along.
llvm-svn: 338452
This reapplies commit r338206 reverted by r338214 since the bug that
r338206 uncovered has been fixed in r338268.
Add support for inline assembly with matching input operand that do not
naturally go in the register class it is constrained to (eg. double in a
32-bit GPR). Note that regular input is already handled by existing
code.
llvm-svn: 338269
Code in `CC_ARM_AAPCS_Custom_Aggregate()` is responsible for handling
homogeneous aggregates for `CC_ARM_AAPCS_VFP`. When an aggregate ends up
fully on stack, the function tries to pack all resulting items of the
aggregate as tightly as possible according to AAPCS.
Once the first item was laid out, the alignment used for consecutive
items was the size of one item. This logic went wrong for 128-bit
vectors because their alignment is normally only 64 bits, and so could
result in inserting unexpected padding between the first and second
element.
The patch fixes the problem by updating the alignment with the item size
only if this results in reducing it.
Differential Revision: https://reviews.llvm.org/D49720
llvm-svn: 338233
Add support for inline assembly with matching input operand that do not
naturally go in the register class it is constrained to (eg. double in a
32-bit GPR). Note that regular input is already handled by existing
code.
llvm-svn: 338206
This feature enables the fusion of such operations on Cortex A57 and Cortex
A72, as recommended in their Software Optimisation Guides, sections 4.14 and
4.11, respectively.
Differential revision: https://reviews.llvm.org/D49563
llvm-svn: 338147
Add support for inline assembly with output operand that do not
naturally go in the register class it is constrained to (eg. double in a
32-bit GPR as in the PR).
llvm-svn: 337903
ARM Stage 2 builders have been suspiciously broken since the pass was
committed. Disabling to hopefully fix the bots and give me time to
debug.
llvm-svn: 337821
Arm specific codegen prepare is implemented to perform type promotion
on icmp operands, which can enable the removal of uxtb and uxth
(unsigned extend) instructions. This is possible because performing
type promotion before ISel alleviates this duty from the DAG builder
which has to perform legalisation, but has a limited view on data
ranges.
The pass visits any instruction operand of an icmp and creates a
worklist to traverse the use-def tree to determine whether the values
can simply be promoted. Our concern is values in the registers
overflowing the narrow (i8, i16) data range, so instructions marked
with nuw can be promoted easily. For add and sub instructions, we are
able to use the parallel dsp instructions to operate on scalar data
types and avoid overflowing bits. Underflowing adds and subs are also
permitted when the result is only used by an unsigned icmp.
Differential Revision: https://reviews.llvm.org/D48832
llvm-svn: 337687
Enable the optimization of operations on DPR and SPR via a feature instead
of checking the target.
Differential revision: https://reviews.llvm.org/D49463
llvm-svn: 337575
We were emitting incorrect calls to libm functions that LLVM had decided it
knew about because the default is soft-float.
Recommitted without breaking ELF this time.
llvm-svn: 337450
If we are only extracting vector elements via EXTRACT_VECTOR_ELT(s) we may be able to use SimplifyDemandedVectorElts to avoid unnecessary vector ops.
Differential Revision: https://reviews.llvm.org/D49262
llvm-svn: 337258
See https://reviews.llvm.org/D47106 for details.
Reviewed By: probinson
Differential Revision: https://reviews.llvm.org/D47171
This commit drops that patch's changes to:
llvm/test/CodeGen/NVPTX/f16x2-instructions.ll
llvm/test/CodeGen/NVPTX/param-load-store.ll
For some reason, the dos line endings there prevent me from commiting
via the monorepo. A follow-up commit (not via the monorepo) will
finish the patch.
llvm-svn: 336843
This fixes an issue that we were not properly supporting multiple reduction
stmts in a loop, and not generating SMLADs for these cases. The alias analysis
checks were done too early, making it too conservative.
Differential revision: https://reviews.llvm.org/D49125
llvm-svn: 336795
The original code attempted to do this, but the std::abs() call didn't
actually do anything due to implicit type conversions. Fix the type
conversions, and perform the correct check for negative immediates.
This probably has very little practical impact, but it's worth fixing
just to avoid confusion in the future, I think.
Differential Revision: https://reviews.llvm.org/D48907
llvm-svn: 336742
Added statistics for the number of SMLAD instructions created, and
als renamed the pass name to -arm-parallel-dsp.
Differential Revision: https://reviews.llvm.org/D48971
llvm-svn: 336441
D48278
Allow to reduce redundant shift masks.
For example:
x1 = x & 0xAB00
x2 = (x >> 8) & 0xAB
can be reduced to:
x1 = x & 0xAB00
x2 = x1 >> 8
It only allows folding when the masks and shift values are constants.
llvm-svn: 336426
We were miscompiling i8 loads, so reject them as unsupported narrow operations
for now.
Differential Revision: https://reviews.llvm.org/D48944
llvm-svn: 336319
Armv6 introduced instructions to perform 32-bit SIMD operations. The purpose of
this pass is to do some straightforward IR pattern matching to create ACLE DSP
intrinsics, which map on these 32-bit SIMD operations.
Currently, only the SMLAD instruction gets recognised. This instruction
performs two multiplications with 16-bit operands, and stores the result in an
accumulator. We will follow this up with patches to recognise SMLAD in more
cases, and also to generate other DSP instructions (like e.g. SADD16).
Patch by: Sam Parker and Sjoerd Meijer
Differential Revision: https://reviews.llvm.org/D48128
llvm-svn: 335850
This patch adds support for the q versions of the dup
(load-to-all-lanes) NEON intrinsics, such as vld2q_dup_f16() for
example.
Currently, non-q versions of the dup intrinsics are implemented
in clang by generating IR that first loads the elements of the
structure into the first lane with the lane (to-single-lane)
intrinsics, and then propagating it other lanes. There are at
least two problems with this approach. First, there are no
double-spaced to-single-lane byte-element instructions. For
example, there is no such instruction as 'vld2.8 { d0[0], d2[0]
}, [r0]'. That means we cannot rely on the to-single-lane
intrinsics and instructions to implement the q versions of the
dup intrinsics. Note that to-all-lanes instructions do support
all sizes of data items, including bytes.
The second problem with the current approach is that we need a
separate vdup instruction to propagate the structure to each
lane. So for vld4q_dup_f16() we would need four vdup instructions
in addition to the initial vld instruction.
This patch introduces dup LLVM intrinsics and reworks handling of
the currently supported (non-q) NEON dup intrinsics to expand
them into those LLVM intrinsics, thus eliminating the need for
using to-single-lane intrinsics and instructions.
Additionally, this patch adds support for u64 and s64 dup NEON
intrinsics. These are marked as Arch64-only in the ARM NEON
Reference, but it seems there are no reasons to not support them
in AArch32 mode. Please correct, if that is wrong.
That's what we generate with this patch applied:
vld2q_dup_f16:
vld2.16 {d0[], d2[]}, [r0]
vld2.16 {d1[], d3[]}, [r0]
vld3q_dup_f16:
vld3.16 {d0[], d2[], d4[]}, [r0]
vld3.16 {d1[], d3[], d5[]}, [r0]
vld4q_dup_f16:
vld4.16 {d0[], d2[], d4[], d6[]}, [r0]
vld4.16 {d1[], d3[], d5[], d7[]}, [r0]
Differential Revision: https://reviews.llvm.org/D48439
llvm-svn: 335733
Summary:
If a routine with no stack frame makes a sibling call, we need to
preserve the stack space check even if the local stack frame is empty,
since the call target could be a "no-split" function (in which case
the linker needs to be able to fix up the prolog sequence in order to
switch to a larger stack).
This fixes PR37807.
Reviewers: cherry, javed.absar
Subscribers: srhines, llvm-commits
Differential Revision: https://reviews.llvm.org/D48444
llvm-svn: 335604
With compilation fix.
Original commit message:
D39788 added a '.stack-size' section containing metadata on function stack sizes
to output ELF files behind the new -stack-size-section flag.
This change does following two things on top:
1) Imagine the case when there are -ffunction-sections flag given and there are text sections in COMDATs.
The patch adds a '.stack-size' section into corresponding COMDAT group, so that linker will be able to
eliminate them fast during resolving the COMDATs.
2) Patch sets a SHF_LINK_ORDER flag and links '.stack-size' with the corresponding .text.
With that linker will be able to do -gc-sections on dead stack sizes sections.
Differential revision: https://reviews.llvm.org/D46874
llvm-svn: 335336
D39788 added a '.stack-size' section containing metadata on function stack sizes
to output ELF files behind the new -stack-size-section flag.
This change does following two things on top:
1) Imagine the case when there are -ffunction-sections flag given and there are text sections in COMDATs.
The patch adds a '.stack-size' section into corresponding COMDAT group, so that linker will be able to
eliminate them fast during resolving the COMDATs.
2) Patch sets a SHF_LINK_ORDER flag and links '.stack-size' with the corresponding .text.
With that linker will be able to do -gc-sections on dead stack sizes sections.
Differential revision: https://reviews.llvm.org/D46874
llvm-svn: 335332
This sets target feature FeatureStrictAlign for Armv6-m and Armv8-m.baseline,
because it has no support for unaligned accesses.
It looks like we always pass target feature "+strict-align" from
Clang, so this is not a user facing problem, but querying the subtarget
(in e.g. llc) for unaligned access support is incorrect.
Differential Revision: https://reviews.llvm.org/D48437
llvm-svn: 335326
This option allows codegen (such as DAGCombine or MI scheduling) to use alias
analysis information, which can help with the codegen on in-order cpu's,
especially machine scheduling. Here I have done things the same way as AArch64,
adding a subtarget feature to enable this for specific cores, and enabled it for
the R52 where we have a schedule to make use of it.
Differential Revision: https://reviews.llvm.org/D48074
llvm-svn: 335249
The alignment parameter to getExtLoad is treated as a base alignment,
not the alignment of the load (base + offset). When we infer a better
alignment for a Ptr we need to ensure that it applies to the base to
prevent the alignment on the load from being wrong.
This fixes a bug where the alignment could then be used to incorrectly
prove noalias between a load and a store, leading to a miscompile.
Differential Revision: https://reviews.llvm.org/D48029
llvm-svn: 335210
Summary:
Two utils methods have essentially the same functionality. This is an attempt to merge them into one.
1. lib/Transforms/Utils/Local.cpp : MergeBasicBlockIntoOnlyPred
2. lib/Transforms/Utils/BasicBlockUtils.cpp : MergeBlockIntoPredecessor
Prior to the patch:
1. MergeBasicBlockIntoOnlyPred
Updates either DomTree or DeferredDominance
Moves all instructions from Pred to BB, deletes Pred
Asserts BB has single predecessor
If address was taken, replace the block address with constant 1 (?)
2. MergeBlockIntoPredecessor
Updates DomTree, LoopInfo and MemoryDependenceResults
Moves all instruction from BB to Pred, deletes BB
Returns if doesn't have a single predecessor
Returns if BB's address was taken
After the patch:
Method 2. MergeBlockIntoPredecessor is attempting to become the new default:
Updates DomTree or DeferredDominance, and LoopInfo and MemoryDependenceResults
Moves all instruction from BB to Pred, deletes BB
Returns if doesn't have a single predecessor
Returns if BB's address was taken
Uses of MergeBasicBlockIntoOnlyPred that need to be replaced:
1. lib/Transforms/Scalar/LoopSimplifyCFG.cpp
Updated in this patch. No challenges.
2. lib/CodeGen/CodeGenPrepare.cpp
Updated in this patch.
i. eliminateFallThrough is straightforward, but I added using a temporary array to avoid the iterator invalidation.
ii. eliminateMostlyEmptyBlock(s) methods also now use a temporary array for blocks
Some interesting aspects:
- Since Pred is not deleted (BB is), the entry block does not need updating.
- The entry block was being updated with the deleted block in eliminateMostlyEmptyBlock. Added assert to make obvious that BB=SinglePred.
- isMergingEmptyBlockProfitable assumes BB is the one to be deleted.
- eliminateMostlyEmptyBlock(BB) does not delete BB on one path, it deletes its unique predecessor instead.
- adding some test owner as subscribers for the interesting tests modified:
test/CodeGen/X86/avx-cmp.ll
test/CodeGen/AMDGPU/nested-loop-conditions.ll
test/CodeGen/AMDGPU/si-annotate-cf.ll
test/CodeGen/X86/hoist-spill.ll
test/CodeGen/X86/2006-11-17-IllegalMove.ll
3. lib/Transforms/Scalar/JumpThreading.cpp
Not covered in this patch. It is the only use case using the DeferredDominance.
I would defer to Brian Rzycki to make this replacement.
Reviewers: chandlerc, spatel, davide, brzycki, bkramer, javed.absar
Subscribers: qcolombet, sanjoy, nemanjai, nhaehnle, jlebar, tpr, kbarton, RKSimon, wmi, arsenm, llvm-commits
Differential Revision: https://reviews.llvm.org/D48202
llvm-svn: 335183
Thumb has more 16-bit encoding space dedicated to ADD than ORR, allowing both a
3-address encoding and a wider range of immediates. So, particularly when
optimizing for code size (but it doesn't make things worse elsewhere) it's
beneficial to select an OR operation to an ADD if we know overflow won't occur.
This is made even better by LLVM's penchant for putting operations in canonical
form by converting the other way.
llvm-svn: 335119
Summary:
Here we relax the old constraint which utilized unsafe with the TargetOption flag HonorSignDependentRoundingFPMathOption, with the assertion that unsafe is no longer needed or never was required for correctness on FDIV/FMUL.
Reviewers: spatel, hfinkel, wristow, arsenm, javed.absar
Reviewed By: spatel
Subscribers: efriedma, wdng, tpr
Differential Revision: https://reviews.llvm.org/D48057
llvm-svn: 334769
This would fail before because 1x vectors aren't legal,
so instead just use the scalar type.
Avoids regressions in a future AMDGPU commit to add
v4i16/v4f16 as legal types.
Test update is just the one test that this triggers
on in tree now. It wasn't checking anything before.
The result is completely changed since the selects
are eliminated. Not sure if it's considered better
or not.
llvm-svn: 334440
We currently support them only in AArch64. The NEON Reference,
however, says they are 'ARMv7, ARMv8' intrinsics.
Differential Revision: https://reviews.llvm.org/D47447
llvm-svn: 334361
It looks like this got left in by accident in r289794; I can't think of
any reason this check would be necessary. (Maybe it was meant to be a
check that the AND has one use? But we check that a few lines earlier.)
Differential Revision: https://reviews.llvm.org/D47921
llvm-svn: 334322
If no alignment is set, the abi/preferred alignment of structs will be
used which may be higher than required. This can lead to extra padding
and in the end an increase in data size.
Differential Revision: https://reviews.llvm.org/D47633
llvm-svn: 334099
On targets like Arm some relaxations may only be performed when certain
architectural features are available. As functions can be compiled with
differing levels of architectural support we must make a judgement on
whether we can relax based on the MCSubtargetInfo for the function. This
change passes through the MCSubtargetInfo for the function to
fixupNeedsRelaxation so that the decision on whether to relax can be made
per function. In this patch, only the ARM backend makes use of this
information. We must also pass the MCSubtargetInfo to applyFixup because
some fixups skip error checking on the assumption that relaxation has
occurred, to prevent code-generation errors applyFixup must see the same
MCSubtargetInfo as fixupNeedsRelaxation.
Differential Revision: https://reviews.llvm.org/D44928
llvm-svn: 334078
We currently support them only in AArch64. The NEON Reference,
however, says they are 'ARMv7, ARMv8' intrinsics.
Differential Revision: https://reviews.llvm.org/D47120
llvm-svn: 333825
We currently support them only in AArch64. The NEON Reference,
however, says they are 'ARMv7, ARMv8' intrinsics.
Differential Revision: https://reviews.llvm.org/D47121
llvm-svn: 333819
We've had Thumb1 support for ARMISD::SUBE for a while now, so this just
works. Reduces codesize a bit for 64-bit integer comparisons.
Differential Revision: https://reviews.llvm.org/D47387
llvm-svn: 333445
Chances are we'll be asked again after type legalization, but before that point
it's better to claim misaligned accesses aren't allowed than to assert.
llvm-svn: 332840
We need to clean up the DAG floating-point undef logic.
This process is similar to how we handled integer undef
logic in https://reviews.llvm.org/D43141.
And as we did there, I'm trying to reduce the patch by
changing tests that would probably become meaningless
once we correct FP undef folding.
llvm-svn: 332638
We need to clean up the DAG floating-point undef logic.
This process is similar to how we handled integer undef
logic in https://reviews.llvm.org/D43141.
And as we did there, I'm trying to reduce the patch by
changing tests that would probably become meaningless
once we correct FP undef folding.
Follow-up to:
https://reviews.llvm.org/rL332538
...because that change wasn't enough.
llvm-svn: 332637
We need to clean up the DAG floating-point undef logic.
This process is similar to how we handled integer undef
logic in D43141.
And as we did there, I'm trying to reduce the patch by
changing tests that would probably become meaningless
once we correct FP undef folding.
llvm-svn: 332539
We need to clean up the DAG floating-point undef logic.
This process is similar to how we handled integer undef
logic in D43141.
And as we did there, I'm trying to reduce the patch by
changing tests that would probably become meaningless
once we correct FP undef folding.
llvm-svn: 332538
We need to clean up the DAG floating-point undef logic.
This process is similar to how we handled integer undef
logic in D43141.
And as we did there, I'm trying to reduce the patch by
changing tests that would probably become meaningless
once we correct FP undef folding.
llvm-svn: 332537
We need to clean up the DAG floating-point undef logic.
This process is similar to how we handled integer undef
logic in D43141.
And as we did there, I'm trying to reduce the patch by
changing tests that would probably become meaningless
once we correct FP undef folding.
llvm-svn: 332533
We need to clean up the DAG floating-point undef logic.
This process is similar to how we handled integer undef
logic in D43141.
And as we did there, I'm trying to reduce the patch by
changing tests that would probably become meaningless
once we correct FP undef folding.
llvm-svn: 332532
Keep loads and stores together (target defines how many loads
and stores to gang up), such that it will help in pairing
and vectorization.
Differential Revision https://reviews.llvm.org/D46477
llvm-svn: 332482
We currently handle all aggregates by creating one large LLT, and letting the
legalizer deal with splitting them up. However using this approach means that
we can't support big endian code correctly.
This patch changes the way that the IRTranslator deals with aggregate values,
by splitting them up into their constituent element values. To do this, parts
of the translator need to be modified to deal with multiple VRegs for a single
Value.
A new Value to VReg mapper is introduced to help keep compile time under
control, currently there is no measurable impact on CTMark despite the extra
code being generated in some cases.
Patch is based on the original work of Tim Northover.
Differential Revision: https://reviews.llvm.org/D46018
llvm-svn: 332449
This is a simple hack based on what's proposed in D37686, but we can extend it if needed in follow-ups.
It gets us most of the FMF functionality that we want without adding any state bits to the flags. It
also intentionally leaves out non-FMF flags (nsw, etc) to minimize the patch.
It should provide a superset of the functionality from D46563 - the extra tests show propagation and
codegen diffs for fcmp, vecreduce, and FP libcalls.
The PPC log2() test shows the limits of this most basic approach - we only applied 'afn' to the last
node created for the call. AFAIK, there aren't any libcall optimizations based on the flags currently,
so that shouldn't make any difference.
Differential Revision: https://reviews.llvm.org/D46854
llvm-svn: 332358
In order to set breakpoints on labels and list source code around
labels, we need collect debug information for labels, i.e., label
name, the function label belong, line number in the file, and the
address label located. In order to keep these information in LLVM
IR and to allow backend to generate debug information correctly.
We create a new kind of metadata for labels, DILabel. The format
of DILabel is
!DILabel(scope: !1, name: "foo", file: !2, line: 3)
We hope to keep debug information as much as possible even the
code is optimized. So, we create a new kind of intrinsic for label
metadata to avoid the metadata is eliminated with basic block.
The intrinsic will keep existing if we keep it from optimized out.
The format of the intrinsic is
llvm.dbg.label(metadata !1)
It has only one argument, that is the DILabel metadata. The
intrinsic will follow the label immediately. Backend could get the
label metadata through the intrinsic's parameter.
We also create DIBuilder API for labels to be used by Frontend.
Frontend could use createLabel() to allocate DILabel objects, and use
insertLabel() to insert llvm.dbg.label intrinsic in LLVM IR.
Differential Revision: https://reviews.llvm.org/D45024
Patch by Hsiangkai Wang.
llvm-svn: 331841
As Roman Tereshin pointed out in https://reviews.llvm.org/D45541, the
-global-isel option is redundant when -run-pass is given. -global-isel sets up
the GlobalISel passes in the pass manager but -run-pass skips that entirely and
configures it's own pipeline.
llvm-svn: 331603
By default LLVM thinks very large vectors get aligned to their size when
passed across functions. Unfortunately no-one told the ARM backend so it
doesn't trigger stack realignment and so accesses can cause the usual
misalignment issues (e.g. a data abort).
This changes the ABI alignment to the stack alignment, which in practice
(and as a bonus) also coincides with the alignment "natural" vectors get.
llvm-svn: 331451
The logic for this combine is almost identical to the logic for a
(sext (sextload x)) combine.
This commit factors out the logic so it can be shared by both combines,
and corrects the SDLoc assigned in the zext version of the combine.
Prior to this patch, for the given test case, we would apply the
location associated with the udiv instruction to instructions which
perform the load.
Part of: llvm.org/PR37262
llvm-svn: 331303
Prior to this patch, for the given test case, we would apply the
location associated with the sdiv instruction to instructions which
perform the load.
Part of: llvm.org/PR37262.
Differential Revision: https://reviews.llvm.org/D46222
llvm-svn: 331302
Setting the right SDLoc on a newly-created zextload fixes a line table
bug which resulted in non-linear stepping behavior.
Several backend tests contained CHECK lines which relied on the IROrder
inherited from the wrong SDLoc. This patch breaks that dependence where
feasbile and regenerates test cases where not.
In some cases, changing a node's IROrder may alter register allocation
and spill behavior. This can affect performance. I have chosen not to
prevent this by applying a "known good" IROrder to SDLocs, as this may
hide a more general bug in the scheduler, or cause regressions on other
test inputs.
rdar://33755881, Part of: llvm.org/PR37262
Differential Revision: https://reviews.llvm.org/D45995
llvm-svn: 331300
Summary:
Currently only the memory size is supported but others can be added as
needed.
narrowScalar for G_LOAD and G_STORE now correctly update the
MachineMemOperand and will refuse to legalize atomics since those need more
careful expansions to maintain atomicity.
Reviewers: ab, aditya_nandakumar, bogner, rtereshin, aemerson, javed.absar
Reviewed By: aemerson
Subscribers: aemerson, rovka, kristof.beyls, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D45466
llvm-svn: 331071
This adds IR intrinsics for the ARM dot-product instructions introduced in
v8.2-A.
Differential revision: https://reviews.llvm.org/D46106
llvm-svn: 331032
Back when the R52 schedule was added in rL286949, there was no way
to enable machine schedules in ARM for specific cores. Since then a
target feature has been added. This enables the feature for R52,
removing the need to manually specify compiler flags.
llvm-svn: 331027
Debug var, expr and loc were only supported for non-fixed stack objects.
This patch adds the following fields to the "fixedStack:" entries, and
renames the ones from "stack:" to:
* debug-info-variable
* debug-info-expression
* debug-info-location
Differential Revision: https://reviews.llvm.org/D46032
llvm-svn: 330859
This was originally committed at rL328921 and reverted at rL329920 to
investigate failures in Chrome. This time I've added to the ReleaseNotes
to warn users of the potential of exposing UB and let me repeat that
here for more exposure:
Optimization of floating-point casts is improved. This may cause surprising
results for code that is relying on undefined behavior. Code sanitizers can
be used to detect affected patterns such as this:
int main() {
float x = 4294967296.0f;
x = (float)((int)x);
printf("junk in the ftrunc: %f\n", x);
return 0;
}
$ clang -O1 ftrunc.c -fsanitize=undefined ; ./a.out
ftrunc.c:5:15: runtime error: 4.29497e+09 is outside the range of
representable values of type 'int'
junk in the ftrunc: 0.000000
Original commit message:
fptosi / fptoui round towards zero, and that's the same behavior as ISD::FTRUNC,
so replace a pair of casts with the equivalent node. We don't have to account for
special cases (NaN, INF) because out-of-range casts are undefined.
Differential Revision: https://reviews.llvm.org/D44909
llvm-svn: 330437
This adds code generation support for the FP16 vmaxnm/vminnm scalar
instructions.
Differential Revision: https://reviews.llvm.org/D44675
llvm-svn: 330034
This change is exposing UB in source code - as was warned/predicted. :)
See D44909 for discussion. Reverting while we figure out how to fix things.
llvm-svn: 329920
This is causing compilation timeouts on code with long sequences of
local values and calls (i.e. foo(1); foo(2); foo(3); ...). It turns out
that code coverage instrumentation is a great way to create sequences
like this, which how our users ran into the issue in practice.
Intel has a tool that detects these kinds of non-linear compile time
issues, and Andy Kaylor reported it as PR37010.
The current sinking code scans the whole basic block once per local
value sink, which happens before emitting each call. In theory, local
values should only be introduced to be used by instructions between the
current flush point and the last flush point, so we should only need to
scan those instructions.
llvm-svn: 329822
This is a follow up of rL327695 to instruction select more variants of VSELGT
and VSELGE, for which it is necessary to custom lower SELECT.
More work is required in this area, which will be addressed soon:
- more variants need to be regression tested, but this depends on the next point.
- first LowerConstantFP need to be adjusted for fp16 values.
Differential Revision: https://reviews.llvm.org/D45205
llvm-svn: 329788
Recommitting r329283, third time lucky...
If the SRL node is only used by an AND, we may be able to set the
ExtVT to the width of the mask, making the AND redundant. To support
this, another check has been added in isLegalNarrowLoad which queries
whether the load is valid.
Differential Revision: https://reviews.llvm.org/D41350
llvm-svn: 329551
In our real world application, we found the following optimization is missed in DAGCombiner
(zext (and/or/xor (shl/shr (load x), cst), cst)) -> (and/or/xor (shl/shr (zextload x), (zext cst)), (zext cst))
If the user of original zext is an add, it may enable further lea optimization on x86.
This patch add a new function CombineZExtLogicopShiftLoad to do this optimization.
Differential Revision: https://reviews.llvm.org/D44402
llvm-svn: 329516
Should fix UBSan bot by also checking there's no "uwtable" attribute
before skipping. Otherwise the unwind table will be useless since its
moves expect CSRs to actually be preserved.
A noreturn nounwind function can be expected to never return in any way, and by
never returning it will also never have to restore any callee-saved registers
for its caller. This makes it possible to skip spills of those registers during
function entry, saving some stack space and time in the process. This is rather
useful for embedded targets with limited stack space.
Should fix PR9970.
Patch mostly by myeisha (pmb).
llvm-svn: 329494
A noreturn nounwind function can be expected to never return in any way, and by
never returning it will also never have to restore any callee-saved registers
for its caller. This makes it possible to skip spills of those registers during
function entry, saving some stack space and time in the process. This is rather
useful for embedded targets with limited stack space.
Should fix PR9970.
Patch by myeisha (pmb).
llvm-svn: 329287
Recommitting rL321259. Previosuly this caused an issue with PPCBE but
I didn't receieve a reproducer and didn't have the time to follow up.
If the issue appears again, please provide a reproducer so I can fix
it.
Original commit message:
If the SRL node is only used by an AND, we may be able to set the
ExtVT to the width of the mask, making the AND redundant. To support
this, another check has been added in isLegalNarrowLoad which queries
whether the load is valid.
Differential Revision: https://reviews.llvm.org/D41350
llvm-svn: 329160
fptosi / fptoui round towards zero, and that's the same behavior as ISD::FTRUNC,
so replace a pair of casts with the equivalent node. We don't have to account for
special cases (NaN, INF) because out-of-range casts are undefined.
Differential Revision: https://reviews.llvm.org/D44909
llvm-svn: 328921
Follow up patch of r328313 to support the UseVMOVSR constraint. Removed
some unneeded instructions from the test and removed some stray
comments.
Differential Revision: https://reviews.llvm.org/D44941
llvm-svn: 328691
%tmp = bitcast i32* %arg to i8*
%tmp1 = getelementptr inbounds i8, i8* %tmp, i32 0
- %tmp2 = load i8, i8* %tmp, align 1
+ %tmp2 = load i8, i8* %tmp1, align 1
This doesn't change the semantics of the tests but makes use of %tmp1 which was originally intended.
llvm-svn: 328642
On Hexagon "x = y" is a syntax used in most instructions, and is not
treated as a directive.
Differential Revision: https://reviews.llvm.org/D44256
llvm-svn: 328635
When targeting execute-only and fp-armv8, float constants in a compare
resulted in instruction selection failures. This is now fixed by using
vmov.f32 where possible, otherwise the floating point constant is
lowered into a integer constant that is moved into a floating point
register.
This patch also restores using fpcmp with immediate 0 under fp-armv8.
Change-Id: Ie87229706f4ed879a0c0cf66631b6047ed6c6443
llvm-svn: 328313
Summary:
DbgValue nodes were not transferred when integer DAG nodes were promoted. For example, if an i32 add node was promoted to an i64 add node by DAGTypeLegalizer::PromoteIntegerResult(), its DbgValue node was not transferred to the new node. The simple fix is to update SetPromotedInteger() to transfer DbgValues.
Add AArch64/dbg-value-i8.ll to test this change and fix ARM/debug-info-d16-reg.ll which had the wrong DILocalVariable nodes with arg numbers even though they are not for function parameters.
Patch by Se Jong Oh!
Reviewers: vsk, JDevlieghere, aprantl
Reviewed By: JDevlieghere
Subscribers: javed.absar, kristof.beyls, llvm-commits
Tags: #debug-info
Differential Revision: https://reviews.llvm.org/D44546
llvm-svn: 327919
This extends the use of this attribute on ARM and AArch64 from
SVN r325900 (where it was only checked for fixed stack
allocations on ARM/AArch64, but for all stack allocations on X86).
This also adds a testcase for the existing use of disabling the
fixed stack probe with the attribute on ARM and AArch64.
Differential Revision: https://reviews.llvm.org/D44291
llvm-svn: 327897
This is the groundwork for adding the Armv8.2-A FP16 vector intrinsics, which
uses v4f16 and v8f16 vector operands and return values. All the moving parts
are tested with two intrinsics, a 1-operand v8f16 and a 2-operand v4f16
intrinsic. In a follow-up patch the rest of the intrinsics and tests will be
added.
Differential Revision: https://reviews.llvm.org/D44538
llvm-svn: 327839
This implements lowering of SELECT_CC for f16s, which enables
codegen of VSEL with f16 types.
Differential Revision: https://reviews.llvm.org/D44518
llvm-svn: 327695
Previously if getSetccResultType returned an illegal type we just fell back to using the default promoted type. This appears to have been to handle the case where for vectors getSetccResultType returns the input type, but the input type itself isn't legal and will need to be promoted. Without the legality check we would never reach a legal type.
But just picking the promoted type to be the setcc type can create strange setccs where the result type is 128 bits and the operand type is 256 bits. If for example the result type was promoted to v8i16 from v8i1, but the input type was promoted from v8i23 to v8i32. We currently handle this with custom lowering code in X86.
This legality check also caused us reject the getSetccResultType when the input type needed to be widened or split. Even though that result wouldn't have caused legalization to get stuck.
This patch tries to fix this by detecting the getSetccResultType needs to be promoted. If its input type also needs to be promoted we'll try a ask for a new setcc result type based on its eventual promoted value. Otherwise we fall back to default type to promote to.
For any other illegal values we might get back from the initial call to getSetccResultType we just keep and allow it to be re-legalized later via splitting or widening or scalarizing.
llvm-svn: 327683
Summary:
Local values are constants, global addresses, and stack addresses that
can't be folded into the instruction that uses them. For example, when
storing the address of a global variable into memory, we need to
materialize that address into a register.
FastISel doesn't want to materialize any given local value more than
once, so it generates all local value materialization code at
EmitStartPt, which always dominates the current insertion point. This
allows it to maintain a map of local value registers, and it knows that
the local value area will always dominate the current insertion point.
The downside is that local value instructions are always emitted without
a source location. This is done to prevent jumpy line tables, but it
means that the local value area will be considered part of the previous
statement. Consider this C code:
call1(); // line 1
++global; // line 2
++global; // line 3
call2(&global, &local); // line 4
Today we end up with assembly and line tables like this:
.loc 1 1
callq call1
leaq global(%rip), %rdi
leaq local(%rsp), %rsi
.loc 1 2
addq $1, global(%rip)
.loc 1 3
addq $1, global(%rip)
.loc 1 4
callq call2
The LEA instructions in the local value area have no source location and
are treated as being on line 1. Stepping through the code in a debugger
and correlating it with the assembly won't make much sense, because
these materializations are only required for line 4.
This is actually problematic for the VS debugger "set next statement"
feature, which effectively assumes that there are no registers live
across statement boundaries. By sinking the local value code into the
statement and fixing up the source location, we can make that feature
work. This was filed as https://bugs.llvm.org/show_bug.cgi?id=35975 and
https://crbug.com/793819.
This change is obviously not enough to make this feature work reliably
in all cases, but I felt that it was worth doing anyway because it
usually generates smaller, more comprehensible -O0 code. I measured a
0.12% regression in code generation time with LLC on the sqlite3
amalgamation, so I think this is worth doing.
There are some special cases worth calling out in the commit message:
1. local values materialized for phis
2. local values used by no-op casts
3. dead local value code
Local values can be materialized for phis, and this does not show up as
a vreg use in MachineRegisterInfo. In this case, if there are no other
uses, this patch sinks the value to the first terminator, EH label, or
the end of the BB if nothing else exists.
Local values may also be used by no-op casts, which adds the register to
the RegFixups table. Without reversing the RegFixups map direction, we
don't have enough information to sink these instructions.
Lastly, if the local value register has no other uses, we can delete it.
This comes up when fastisel tries two instruction selection approaches
and the first materializes the value but fails and the second succeeds
without using the local value.
Reviewers: aprantl, dblaikie, qcolombet, MatzeB, vsk, echristo
Subscribers: dotdash, chandlerc, hans, sdardis, amccarth, javed.absar, zturner, llvm-commits, hiraditya
Differential Revision: https://reviews.llvm.org/D43093
llvm-svn: 327581
Get rid of the "; mem:" suffix and use the one we use in MIR: ":: (load 2)".
rdar://38163529
Differential Revision: https://reviews.llvm.org/D42377
llvm-svn: 327580
swifterror llvm values model the swifterror register as memory at the
LLVM IR level. ISel will perform adhoc mem-to-reg on them. swifterror
values are constraint in how they can be used. Spilling them to memory
is not allowed.
SjLjEHPrepare tried to lower swifterror values to memory which is
unecessary since the back-end will spill and reload the register as
neccessary (as long as clobbering calls are marked as such which is the
case here) and further leads to invalid IR because swifterror values
can't be stored to memory.
rdar://38164004
llvm-svn: 327521
Code generation of VLD3, VLD4, VST3 and VST4 with register writeback is
broken due to 2 separate bugs:
1) VLD1d64TPseudoWB_register and VLD1d64QPseudoWB_register are missing
rules to expand them to non pseudo MIR. These are selected for
ARMISD::VLD3_UPD/VLD4_UPD with v1i64 vectors in SelectVLD.
2) Selection of the right VLD/VST instruction is broken for load and
store of 3 and 4 v1i64 vectors. SelectVLD and SelectVST are called
with MIR opcode for fixed writeback (ie increment is access size)
and call getVLDSTRegisterUpdateOpcode() to select an opcode with
register writeback if base register update is of a different size.
Since getVLDSTRegisterUpdateOpcode() only knows about
VLD1/VLD2/VST1/VST2 the call is currently conditional on the number
of element in the vector.
However, VLD1/VST1 is selected by SelectVLD/SelectVST's caller for
load and stores of 3 or 4 v1i64 vectors. Therefore the opcode is not
updated which later lead to a fixed writeback instruction being
constructed with an extra operand for the register writeback.
This patch addresses the two issues as follows:
- it adds the necessary mapping from VLD1d64TPseudoWB_register and
VLD1d64QPseudoWB_register to VLD1d64Twb_register and
VLD1d64Qwb_register respectively. Like for the existing _fixed
variants, the cost of these is bumped for unaligned access.
- it changes the logic in SelectVLD and SelectVSD to call isVLDfixed
and isVSTfixed respectively to decide whether the opcode should be
updated. It also reworks the logic and comments for pushing the
writeback offset operand and r0 operand to clarify the logic:
writeback offset needs to be pushed if it's a register writeback,
r0 needs to be pushed if not and the instruction is a
VLD1/VLD2/VST1/VST2.
Reviewers: rengolin, t.p.northover, samparker
Reviewed By: samparker
Patch by Thomas Preud'homme <thomas.preudhomme@arm.com>
Differential Revision: https://reviews.llvm.org/D42970
llvm-svn: 326570
Eli Friedman