| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
|
|
| |
The enhanced version will be used in D48893 and related patches
and an almost identical (fadd is different) version is proposed
in D28907, so adding this as a preliminary step.
llvm-svn: 336444
|
|
|
|
|
|
|
| |
This is likely to be used in D48987 and similar patches,
so adding it as an NFC preliminary step.
llvm-svn: 336442
|
|
|
|
|
|
|
|
|
| |
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
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
LoopBlockNumber is a DenseMap<BasicBlock*, int>, comparing the result of
find() will compare a pair<BasicBlock*, int>. That's of course depending
on pointer ordering which varies from run to run. Reverse iteration
doesn't find this because we're copying to a vector first.
This bug has been there since 2016 but only recently showed up on clang
selfhost with FDO and ThinLTO, which is also why I didn't manage to get
a reasonable test case for this. Add an assert that would've caught
this.
llvm-svn: 336439
|
|
|
|
|
|
|
|
| |
This adds:
- outer shareable TLB Maintenance instructions, and
- TLB range maintenance instructions.
llvm-svn: 336434
|
|
|
|
|
|
| |
Now with the asm operand definition included.
llvm-svn: 336432
|
|
|
|
| |
llvm-svn: 336428
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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
|
|
|
|
|
|
| |
It's causing build errors.
llvm-svn: 336422
|
|
|
|
|
|
|
|
| |
These instructions are added to AArch64 only.
Differential Revision: https://reviews.llvm.org/D48926
llvm-svn: 336421
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously we only iterated over functions reachable from the set of
external functions in the module. But since some of the passes under
this (notably the always-inliner and coroutine lowerer) are required for
correctness, they need to run over everything.
This just adds an extra layer of iteration over the CallGraph to keep
track of which functions we've already visited and get the next batch of
SCCs.
Should fix PR38029.
llvm-svn: 336419
|
|
|
|
|
|
|
|
| |
This adds the Armv8.4-A Trace synchronization barrier (TSB) instruction.
Differential Revision: https://reviews.llvm.org/D48918
llvm-svn: 336418
|
|
|
|
|
|
|
|
| |
The intrinsics can be implemented with a f32/f64 llvm.fma intrinsic and an insert into a zero vector.
There are a couple regressions here due to SelectionDAG not being able to pull an fneg through an extract_vector_elt. I'm not super worried about this though as InstCombine should be able to do it before we get to SelectionDAG.
llvm-svn: 336416
|
|
|
|
|
|
| |
are done"
llvm-svn: 336410
|
|
|
|
|
|
|
|
|
|
| |
unmasked 512-bit intrinsics with rounding mode.
This upgrades all of the intrinsics to use fneg instructions to convert fma into fmsub/fnmsub/fnmadd/fmsubadd. And uses a select instruction for masking.
This matches how clang uses the intrinsics these days.
llvm-svn: 336409
|
|
|
|
|
|
|
|
| |
Power 9 does not have a hardware instruction for frem but we can call fmodf128.
Differential Revision: https://reviews.llvm.org/D48552
llvm-svn: 336406
|
|
|
|
|
|
|
|
|
|
|
| |
It seems like the debugger first computes a symbol's bucket,
and then does a binary search of entries in the bucket using the
symbol's name in order to find it. If the bucket entries are not
in sorted order, this obviously won't work. After this patch a
couple of simple test cases show that we generate an exactly
identical GSI hash stream, which is very nice.
llvm-svn: 336405
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Summary: The lib paths are not correctly picked up for OpenEmbedded sysroots
(like arm-oe-linux-gnueabi). I fix this in a follow-up clang patch. But in
order to add the correct libs I need to detect if the vendor is oe. For this
reason, it is first necessary to teach llvm to detect oe vendor, which is what
this patch does.
Reviewers: chandlerc, compnerd, rengolin, javed.absar
Reviewed By: compnerd
Subscribers: kristof.beyls, dexonsmith, llvm-commits
Differential Revision: https://reviews.llvm.org/D48861
llvm-svn: 336401
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Summary:
If LOCK prefix is not the first prefix in an instruction, LLVM
disassembler silently drops the prefix.
The fix is to select a proper instruction with a builtin LOCK prefix if
one exists.
Reviewers: craig.topper
Reviewed By: craig.topper
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D49001
llvm-svn: 336400
|
|
|
|
|
|
|
| |
There were a couple of issues reported (PR38047, PR37929) - I'll reland
the patch when I figure out and fix the rootcause.
llvm-svn: 336393
|
|
|
|
|
|
|
|
|
|
|
|
| |
Summary: This was missing in D48839 (rL336145).
Reviewers: aardappel
Subscribers: dschuff, sbc100, jgravelle-google, sunfish, llvm-commits
Differential Revision: https://reviews.llvm.org/D48992
llvm-svn: 336390
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Summary:
CompileOnDemandLayer.cpp uses function in these libraries, and builds
with `-DSHARED_LIB=ON` fail without this.
Reviewers: lhames
Subscribers: mgorny, llvm-commits
Differential Revision: https://reviews.llvm.org/D48995
llvm-svn: 336389
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
a deficiency in TableGen that has been addressed in r336334.
[AArch64][SVE] Asm: Support for predicated FP rounding instructions.
This patch also adds instructions for predicated FP square-root and
reciprocal exponent.
The added instructions are:
- FRINTI Round to integral value (current FPCR rounding mode)
- FRINTX Round to integral value (current FPCR rounding mode, signalling inexact)
- FRINTA Round to integral value (to nearest, with ties away from zero)
- FRINTN Round to integral value (to nearest, with ties to even)
- FRINTZ Round to integral value (toward zero)
- FRINTM Round to integral value (toward minus Infinity)
- FRINTP Round to integral value (toward plus Infinity)
- FSQRT Floating-point square root
- FRECPX Floating-point reciprocal exponent
llvm-svn: 336387
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
writing them to a buffer and re-loading them.
Also introduces a multithreaded variant of SimpleCompiler
(MultiThreadedSimpleCompiler) for compiling IR concurrently on multiple
threads.
These changes are required to JIT IR on multiple threads correctly.
No test case yet. I will be looking at how to modify LLI / LLJIT to test
multithreaded JIT support soon.
llvm-svn: 336385
|
|
|
|
| |
llvm-svn: 336384
|
|
|
|
|
|
|
|
| |
'llvm.fma'. Add upgrade tests for all.
Still need to remove the AVX512 masked versions.
llvm-svn: 336383
|
|
|
|
|
|
| |
Differential Revision: https://reviews.llvm.org/D48954
llvm-svn: 336376
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Better NaN handling for AMDGCN fmed3.
All operands are checked for NaN now. The checks
were moved before the canonicalization to provide
a better mapping from fclamp. Changed the behaviour
of fmed3(x,y,NaN) to return max(x,y) instead of
min(x,y) in light of this. Updated tests as a result
and added some new cases to cover the fix.
Patch by Alan Baker
llvm-svn: 336375
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Avoid using allocateKernArg / AssignFn. We do not want any
of the type splitting properties of normal calling convention
lowering.
For now at least this exists alongside the IR argument lowering
pass. This is necessary to handle struct padding correctly while
some arguments are still skipped by the IR argument lowering
pass.
llvm-svn: 336373
|
|
|
|
|
|
| |
Normally InstCombine would have simplified these to SRL/AND instructions but we may still see these during SLP vectorization etc.
llvm-svn: 336371
|
|
|
|
|
|
|
|
|
|
|
| |
instructions
Map the following instructions to the proper float128 lib calls:
pow[i], exp[2], log[2|10], sin, cos, fmin, fmax
Differential Revision: https://reviews.llvm.org/D48544
llvm-svn: 336361
|
|
|
|
|
|
|
|
|
|
| |
from opcodes. NFCI.
This is an early step towards matching Instructions by attributes other than the opcode. This will be necessary for cast/call alternates which share the same opcode but have different types/intrinsicIDs etc. - which we could vectorize as long as we split them using the alternate mechanism.
Differential Revision: https://reviews.llvm.org/D48945
llvm-svn: 336344
|
|
|
|
|
|
|
|
|
|
|
|
| |
Wait states are not properly being inserted after buffer_store for v_interp instructions.
Add VALU to V_INTERP instructions so that the GCNHazardRecognizer can
check and insert the appropriate wait states when needed.
Differential Revision: https://reviews.llvm.org/D48772
Change-Id: Id540c9b074fc69b5c1de6b182276aa089c74aa64
llvm-svn: 336339
|
|
|
|
| |
llvm-svn: 336331
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Similar to PR/25526, fast-regalloc introduces spills at the end of basic
blocks. When this occurs in between an ll and sc, the stores can cause the
atomic sequence to fail.
This patch fixes the issue by introducing more pseudos to represent atomic
operations and moving their lowering to after the expansion of postRA
pseudos.
This version addresses issues with the initial implementation and covers
all atomic operations.
This resolves PR/32020.
Thanks to James Cowgill for reporting the issue!
Patch By: Simon Dardis
Differential Revision: https://reviews.llvm.org/D31287
llvm-svn: 336328
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Resolves:
Unsupported ARM Neon intrinsics in Target-specific DAG combine
function for VLDDUP
https://bugs.llvm.org/show_bug.cgi?id=38031
Related diff: D48439
Differential Revision: https://reviews.llvm.org/D48920
llvm-svn: 336325
|
|
|
|
|
|
|
| |
in TableGen, for which there is already a patch in Phabricator
(D48937) that needs to be committed first.
llvm-svn: 336324
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch also adds instructions for predicated FP square-root and
reciprocal exponent.
The added instructions are:
- FRINTI Round to integral value (current FPCR rounding mode)
- FRINTX Round to integral value (current FPCR rounding mode, signalling inexact)
- FRINTA Round to integral value (to nearest, with ties away from zero)
- FRINTN Round to integral value (to nearest, with ties to even)
- FRINTZ Round to integral value (toward zero)
- FRINTM Round to integral value (toward minus Infinity)
- FRINTP Round to integral value (toward plus Infinity)
- FSQRT Floating-point square root
- FRECPX Floating-point reciprocal exponent
llvm-svn: 336322
|
|
|
|
|
|
|
|
|
| |
We were miscompiling i8 loads, so reject them as unsupported narrow operations
for now.
Differential Revision: https://reviews.llvm.org/D48944
llvm-svn: 336319
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch implements the following varieties:
- Unpredicated signed max, e.g. smax z0.h, z1.h, #-128
- Unpredicated signed min, e.g. smin z0.h, z1.h, #-128
- Unpredicated unsigned max, e.g. umax z0.h, z1.h, #255
- Unpredicated unsigned min, e.g. umin z0.h, z1.h, #255
- Predicated signed max, e.g. smax z0.h, p0/m, z0.h, z1.h
- Predicated signed min, e.g. smin z0.h, p0/m, z0.h, z1.h
- Predicated signed abd, e.g. sabd z0.h, p0/m, z0.h, z1.h
- Predicated unsigned max, e.g. umax z0.h, p0/m, z0.h, z1.h
- Predicated unsigned min, e.g. umin z0.h, p0/m, z0.h, z1.h
- Predicated unsigned abd, e.g. uabd z0.h, p0/m, z0.h, z1.h
llvm-svn: 336317
|
|
|
|
|
|
|
|
|
|
|
|
| |
Optimize code sequences for integer conversion to fp128 when the integer is a result of:
* float->int
* float->long
* double->int
* double->long
Differential Revision: https://reviews.llvm.org/D48429
llvm-svn: 336316
|
|
|
|
|
|
| |
independent FMA and extractelement/insertelement.
llvm-svn: 336315
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The alignment specified by a constant for the field
`BumpPointerAllocator::InitialBuffer` exceeded the alignment
guaranteed by `malloc` and `new` on Windows. This change set
the alignment value to that of `long double`, which is defined
by the used platform.
It fixes https://bugs.llvm.org/show_bug.cgi?id=37944.
Differential Revision: https://reviews.llvm.org/D48889
llvm-svn: 336311
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Non-homogenous aggregates are passed in consecutive GPRs, in GPRs and in memory,
or in memory. This patch ensures that float128 members of non-homogenous
aggregates are passed via VSX registers.
This is done via custom lowering a bitcast of a build_pari(i64,i64) to float128
to a new PPCISD node, BUILD_FP128.
Differential Revision: https://reviews.llvm.org/D48308
llvm-svn: 336310
|
|
|
|
|
|
|
|
|
| |
Legalize and emit code for quad-precision floating point operation conversion of
single-precision value to quad-precision.
Differential Revision: https://reviews.llvm.org/D47569
llvm-svn: 336307
|
|
|
|
|
|
|
|
|
|
| |
This patch enable parameter passing and return by value for float128 types.
Passing aggregate/union which contain float128 members will be submitted in
subsequent patches.
Differential Revision: https://reviews.llvm.org/D47552
llvm-svn: 336306
|
|
|
|
|
|
|
|
| |
for the v1i1 mask to have come from a scalar_to_vector from GR8.
We have patterns for SELECTS that top at v1i1 and we have a pattern for (v1i1 (scalar_to_vector GR8)). The patterns being removed here do the same thing as the two other patterns combined so there is no need for them.
llvm-svn: 336305
|
|
|
|
|
|
|
|
|
|
| |
input.
Previously we could only negate the FMADD opcodes. This used to be mostly ok when we lowered FMA intrinsics during lowering. But with the move to llvm.fma from target specific intrinsics, we can combine (fneg (fma)) to (fmsub) earlier. So if we start with (fneg (fma (fneg))) we would get stuck at (fmsub (fneg)).
This patch fixes that so we can also combine things like (fmsub (fneg)).
llvm-svn: 336304
|
|
|
|
|
|
|
|
|
|
| |
in clang.
There's a regression in here due to inability to combine fneg inputs of X86ISD::FMSUB/FNMSUB/FNMADD nodes.
More removals to come, but I wanted to stop and fix the regression that showed up in this first.
llvm-svn: 336303
|
|
|
|
|
|
|
|
|
| |
Legalize and emit code for round & convert float128 to double precision and
single precision.
Differential Revision: https://reviews.llvm.org/D46997
llvm-svn: 336299
|