We now emit a move of -1 before the cmov and do the addition after the cmov just like the case with an extra addition.
This may be slightly worse for code size, but is more consistent with other compilers. And we might be able to hoist the mov -1 outside of loops.
llvm-svn: 338613
Summary:
D25878, which added support for !absolute_symbol for normal X86 ISel,
did not add support for materializing references to absolute symbols for
X86 FastISel. This causes build failures because FastISel generates
PC-relative relocations for absolute symbols. Fall back to normal ISel
for references to !absolute_symbol GVs. Fix for PR38200.
Reviewers: pcc, craig.topper
Reviewed By: pcc
Subscribers: hiraditya, llvm-commits, kcc
Differential Revision: https://reviews.llvm.org/D50116
llvm-svn: 338599
There is nothing x86-specific about this code, so it'd be nice to make this available for other targets to use in the future (and get it out of X86ISelLowering!).
Differential Revision: https://reviews.llvm.org/D50083
llvm-svn: 338586
Summary:
Add _L to _LZ image intrinsic table mapping to table gen.
In ISelLowering check if image intrinsic has lod and if it's equal
to zero, if so remove lod and change opcode to equivalent mapped _LZ.
Change-Id: Ie24cd7e788e2195d846c7bd256151178cbb9ec71
Subscribers: arsenm, mehdi_amini, kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, steven_wu, dexonsmith, llvm-commits
Differential Revision: https://reviews.llvm.org/D49483
llvm-svn: 338523
The DAG combiner logic to simplify AND masks in shift counts is invalid.
While it is true that the SystemZ shift instructions ignore all but the
low 6 bits of the shift count, it is still invalid to simplify the AND
masks while the DAG still uses the standard shift operators (which are
*not* defined to match the SystemZ instruction behavior).
Instead, this patch performs equivalent operations during instruction
selection. For completely removing the AND, this now happens via
additional DAG match patterns implemented by a multi-alternative
PatFrags. For simplifying a 32-bit AND to a 16-bit AND, the existing DAG
patterns were already mostly OK, they just needed an output XForm to
actually truncate the immediate value.
Unfortunately, the latter change also exposed a bug in TableGen: it
seems XForms are currently only handled correctly for direct operands of
the outermost operation node. This patch also fixes that bug by simply
recurring through the whole pattern. This should be NFC for all other
targets.
Differential Revision: https://reviews.llvm.org/D50096
llvm-svn: 338521
Summary:
This patch improves Inliner to provide causes/reasons for negative inline decisions.
1. It adds one new message field to InlineCost to report causes for Always and Never instances. All Never and Always instantiations must provide a simple message.
2. Several functions that used to return the inlining results as boolean are changed to return InlineResult which carries the cause for negative decision.
3. Changed remark priniting and debug output messages to provide the additional messages and related inline cost.
4. Adjusted tests for changed printing.
Patch by: yrouban (Yevgeny Rouban)
Reviewers: craig.topper, sammccall, sgraenitz, NutshellySima, shchenz, chandlerc, apilipenko, javed.absar, tejohnson, dblaikie, sanjoy, eraman, xbolva00
Reviewed By: tejohnson, xbolva00
Subscribers: xbolva00, llvm-commits, arsenm, mehdi_amini, eraman, haicheng, steven_wu, dexonsmith
Differential Revision: https://reviews.llvm.org/D49412
llvm-svn: 338494
It's not strictly required by the transform of the cmov and the add, but it makes sure we restrict it to the cases we know we want to match.
While there canonicalize the operand order of the cmov to simplify the matching and emitting code.
llvm-svn: 338492
EFLAGS copy lowering.
If you have a branch of LLVM, you may want to cherrypick this. It is
extremely unlikely to hit this case empirically, but it will likely
manifest as an "impossible" branch being taken somewhere, and will be
... very hard to debug.
Hitting this requires complex conditions living across complex control
flow combined with some interesting memory (non-stack) initialized with
the results of a comparison. Also, because you have to arrange for an
EFLAGS copy to be in *just* the right place, almost anything you do to
the code will hide the bug. I was unable to reduce anything remotely
resembling a "good" test case from the place where I hit it, and so
instead I have constructed synthetic MIR testing that directly exercises
the bug in question (as well as the good behavior for completeness).
The issue is that we would mistakenly assume any SETcc with a valid
condition and an initial operand that was a register and a virtual
register at that to be a register *defining* SETcc...
It isn't though....
This would in turn cause us to test some other bizarre register,
typically the base pointer of some memory. Now, testing this register
and using that to branch on doesn't make any sense. It even fails the
machine verifier (if you are running it) due to the wrong register
class. But it will make it through LLVM, assemble, and it *looks*
fine... But wow do you get a very unsual and surprising branch taken in
your actual code.
The fix is to actually check what kind of SETcc instruction we're
dealing with. Because there are a bunch of them, I just test the
may-store bit in the instruction. I've also added an assert for sanity
that ensure we are, in fact, *defining* the register operand. =D
llvm-svn: 338481
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
When lowering calling conventions, prefer to decompose vectors
into the constitute register types. This avoids artifical constraints
to satisfy a wide super-register.
This improves code quality because now optimizations don't need to
deal with the super-register constraint. For example the immediate
folding code doesn't deal with 4 component reg_sequences, so by
breaking the register down earlier the existing immediate folding
code is able to work.
This also avoids the need for the shader input processing code
to manually split vector types.
llvm-svn: 338416
Don't declare them as X86SchedWritePair when the folded class will never be used.
Note: MOVBE (load/store endian conversion) instructions tend to have a very different behaviour to BSWAP.
llvm-svn: 338412
As was done for vector rotations, we can efficiently use ISD::MULHU for vXi8/vXi16 ISD::SRL lowering.
Shift-by-zero cases are still problematic (mainly on v32i8 due to extra AND/ANDN/OR or VPBLENDVB blend masks but v8i16/v16i16 aren't great either if PBLENDW fails) so I've limited this first patch to known non-zero cases if we can't easily use PBLENDW.
Differential Revision: https://reviews.llvm.org/D49562
llvm-svn: 338407
Summary:
Similar to D49636, but for PMADDUBSW. This instruction has the additional complexity that the addition of the two products saturates to 16-bits rather than wrapping around. And one operand is treated as signed and the other as unsigned.
A C example that triggers this pattern
```
static const int N = 128;
int8_t A[2*N];
uint8_t B[2*N];
int16_t C[N];
void foo() {
for (int i = 0; i != N; ++i)
C[i] = MIN(MAX((int16_t)A[2*i]*(int16_t)B[2*i] + (int16_t)A[2*i+1]*(int16_t)B[2*i+1], -32768), 32767);
}
```
Reviewers: RKSimon, spatel, zvi
Reviewed By: RKSimon, zvi
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D49829
llvm-svn: 338402
This commit fixes two issues with the liveness information after the
call:
1) The code always spills RCX and RDX if InProlog == true, which results
in an use of undefined phys reg.
2) FinalReg, JoinReg, RoundedReg, SizeReg are not added as live-ins to
the basic blocks that use them, therefore they are seen undefined.
https://llvm.org/PR38376
Differential Revision: https://reviews.llvm.org/D50020
llvm-svn: 338400
Summary:
This patch improves Inliner to provide causes/reasons for negative inline decisions.
1. It adds one new message field to InlineCost to report causes for Always and Never instances. All Never and Always instantiations must provide a simple message.
2. Several functions that used to return the inlining results as boolean are changed to return InlineResult which carries the cause for negative decision.
3. Changed remark priniting and debug output messages to provide the additional messages and related inline cost.
4. Adjusted tests for changed printing.
Patch by: yrouban (Yevgeny Rouban)
Reviewers: craig.topper, sammccall, sgraenitz, NutshellySima, shchenz, chandlerc, apilipenko, javed.absar, tejohnson, dblaikie, sanjoy, eraman, xbolva00
Reviewed By: tejohnson, xbolva00
Subscribers: xbolva00, llvm-commits, arsenm, mehdi_amini, eraman, haicheng, steven_wu, dexonsmith
Differential Revision: https://reviews.llvm.org/D49412
llvm-svn: 338387
We could choose a free 0 for this, but this
matches the behavior for fmul undef, 1.0. Also,
the NaN use is more useful for folding use operations
although if it's not eliminated it is more expensive
in terms of code size.
llvm-svn: 338376
This patch teaches llvm-mca how to identify dependency breaking instructions on
btver2.
An example of dependency breaking instructions is the zero-idiom XOR (example:
`XOR %eax, %eax`), which always generates zero regardless of the actual value of
the input register operands.
Dependency breaking instructions don't have to wait on their input register
operands before executing. This is because the computation is not dependent on
the inputs.
Not all dependency breaking idioms are also zero-latency instructions. For
example, `CMPEQ %xmm1, %xmm1` is independent on
the value of XMM1, and it generates a vector of all-ones.
That instruction is not eliminated at register renaming stage, and its opcode is
issued to a pipeline for execution. So, the latency is not zero.
This patch adds a new method named isDependencyBreaking() to the MCInstrAnalysis
interface. That method takes as input an instruction (i.e. MCInst) and a
MCSubtargetInfo.
The default implementation of isDependencyBreaking() conservatively returns
false for all instructions. Targets may override the default behavior for
specific CPUs, and return a value which better matches the subtarget behavior.
In future, we should teach to Tablegen how to automatically generate the body of
isDependencyBreaking from scheduling predicate definitions. This would allow us
to expose the knowledge about dependency breaking instructions to the machine
schedulers (and, potentially, other codegen passes).
Differential Revision: https://reviews.llvm.org/D49310
llvm-svn: 338372
Since z13, the max group size will be 2 if any μop has more than 3 register
sources.
This has been ignored sofar in the SystemZHazardRecognizer, but is now
handled by recognizing those instructions and adjusting the tracking of
decoding and the cost heuristic for grouping.
Review: Ulrich Weigand
https://reviews.llvm.org/D49847
llvm-svn: 338368
isFNEG was duplicating much of what was done by getTargetConstantBitsFromNode in its own calls to getTargetConstantFromNode.
Noticed while reviewing D48467.
llvm-svn: 338358
Contrary to ELF, we don't add any markers that distinguish data generated
with .short/.long from normal instructions, so the .inst directive only
adds compatibility with assembly that uses it.
Differential Revision: https://reviews.llvm.org/D49936
llvm-svn: 338356
Contrary to ELF, we don't add any markers that distinguish data generated
with .long from normal instructions, so the .inst directive only adds
compatibility with assembly that uses it.
Differential Revision: https://reviews.llvm.org/D49935
llvm-svn: 338355
In one place we checked X86Subtarget.slowLEA() to decide if the pass should run. But to decide what the pass should we only check isSLM. This resulted in Goldmont going down the Bonnell path.
llvm-svn: 338342
Also refactors some existing code to materialize addresses for the large code
model so it can be shared between G_GLOBAL_VALUE and G_BLOCK_ADDR.
This implements PR36390.
Differential Revision: https://reviews.llvm.org/D49903
llvm-svn: 338337
The vector contains the SDNodes that these functions create. The number of nodes is always a small number so we should use SmallVector to avoid a heap allocation.
llvm-svn: 338329
This teaches the outliner to save LR to a register rather than the stack when
possible. This allows us to avoid bumping the stack in outlined functions in
some cases. By doing this, in a later patch, we can teach the outliner to do
something like this:
f1:
...
bl OUTLINED_FUNCTION
...
f2:
...
move LR's contents to a register
bl OUTLINED_FUNCTION
move the register's contents back
instead of falling back to saving LR in both cases.
llvm-svn: 338278
This patch enables instructions that are destructive on their
destination- and first source operand, to be prefixed with a
MOVPRFX instruction.
This patch also adds a variety of tests:
- positive tests for all instructions and forms that accept a
movprfx for either or both predicated and unpredicated forms.
- negative tests for all instructions and forms that do not accept
an unpredicated or predicated movprfx.
- negative tests for the diagnostics that get emitted when a MOVPRFX
instruction is used incorrectly.
This is patch [2/2] in a series to add MOVPRFX instructions:
- Patch [1/2]: https://reviews.llvm.org/D49592
- Patch [2/2]: https://reviews.llvm.org/D49593
Reviewers: rengolin, SjoerdMeijer, samparker, fhahn, javed.absar
Reviewed By: SjoerdMeijer
Differential Revision: https://reviews.llvm.org/D49593
llvm-svn: 338261
This patch adds predicated and unpredicated MOVPRFX instructions, which
can be prepended to SVE instructions that are destructive on their first
source operand, to make them a constructive operation, e.g.
add z1.s, p0/m, z1.s, z2.s <=> z1 = z1 + z2
can be made constructive:
movprfx z0, z1
add z0.s, p0/m, z0.s, z2.s <=> z0 = z1 + z2
The predicated MOVPRFX instruction can additionally be used to zero
inactive elements, e.g.
movprfx z0.s, p0/z, z1.s
add z0.s, p0/m, z0.s, z2.s
Not all instructions can be prefixed with the MOVPRFX instruction
which is why this patch also adds a mechanism to validate prefixed
instructions. The exact rules when a MOVPRFX applies is detailed in
the SVE supplement of the Architectural Reference Manual.
This is patch [1/2] in a series to add MOVPRFX instructions:
- Patch [1/2]: https://reviews.llvm.org/D49592
- Patch [2/2]: https://reviews.llvm.org/D49593
Reviewers: rengolin, SjoerdMeijer, samparker, fhahn, javed.absar
Reviewed By: SjoerdMeijer
Differential Revision: https://reviews.llvm.org/D49592
llvm-svn: 338258
The machine verifier asserts with:
Assertion failed: (isMBB() && "Wrong MachineOperand accessor"), function getMBB, file ../include/llvm/CodeGen/MachineOperand.h, line 542.
It calls analyzeBranch which tries to call getMBB if the opcode is
JMP_1, but in this case we do:
JMP_1 @OUTLINED_FUNCTION
I believe we have to use TAILJMPd64 instead of JMP_1 since JMP_1 is used
with brtarget8.
Differential Revision: https://reviews.llvm.org/D49299
llvm-svn: 338237
Summary:
These instructions interact with hardware blocks outside the shader core,
and they can have "scalar" side effects even when EXEC = 0. We don't
want these scalar side effects to occur when all lanes want to skip
these instructions, so always add the execz skip branch instruction
for basic blocks that contain them.
Also ensure that we skip scalar stores / atomics, though we don't
code-gen those yet.
Reviewers: arsenm, rampitec
Subscribers: kzhuravl, wdng, yaxunl, dstuttard, tpr, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D48431
Change-Id: Ieaeb58352e2789ffd64745603c14970c60819d44
llvm-svn: 338235
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
The WHILE instructions generate a predicate that is true while the
comparison of the first scalar operand (incremented for each predicate
element) with the second scalar operand is true and false thereafter.
WHILELE While incrementing signed scalar less than or equal to scalar
WHILELO While incrementing unsigned scalar lower than scalar
WHILELS While incrementing unsigned scalar lower than or same as scalar
WHILELT While incrementing signed scalar less than scalar
e.g.
whilele p0.s, x0, x1
generates predicate p0 (for 32bit elements) by incrementing
(signed) x0 and comparing that vector to splat(x1).
llvm-svn: 338211
The instructions added in this patch permit active elements within
a vector to be processed sequentially without unpacking the vector.
PFIRST Set the first active element to true.
PNEXT Find next active element in predicate.
CTERMEQ Compare and terminate loop when equal.
CTERMNE Compare and terminate loop when not equal.
llvm-svn: 338210
X86 normally requires immediates to be a signed 32-bit value which would exclude i64 0x80000000. But for add/sub we can negate the constant and use the opposite instruction.
llvm-svn: 338204
This patch adds PFALSE (unconditionally sets all elements of
the predicate to false) and PTEST (set the status flags for the
predicate).
llvm-svn: 338198
SelectionDAGBuilder widens v3i32/v3f32 arguments to
to v4i32/v4f32 which consume an additional register.
In addition to wasting argument space, this produces extra
instructions since now it appears the 4th vector component has
a meaningful value to most combines.
llvm-svn: 338197
This patch adds support for instructions that partition a predicate
based on data-dependent termination conditions in a loop.
BRKA Break after the first true condition
BRKAS Break after the first true condition, setting condition flags
BRKB Break before the first true condition
BRKBS Break before the first true condition, setting condition flags
BRKPA Break after the first true condition, propagating from the
previous partition
BRKPAS Break after the first true condition, propagating from the
previous partition, setting condition flags
BRKPB Break before the first true condition, propagating from the
previous partition
BRKPBS Break before the first true condition, propagating from the
previous partition, setting condition flags
BRKN Propagate break to next partition
BKRNS Propagate break to next partition, setting condition flags
llvm-svn: 338196
Summary:
Moved Explicit Locals pass to last.
Made that pass obligatory.
Made it convert from register to stack based instructions, and removed the registers.
Fixes to related code that was expecting register based instructions.
Added the correct testing flag to all tests, depending on what the
format they were expecting so far.
Translated one test to stack format as example: reg-stackify-stack.ll
tested:
llvm-lit -v `find test -name WebAssembly`
unittests/MC/*
Reviewers: dschuff, sunfish
Subscribers: sbc100, jgravelle-google, eraman, aheejin, llvm-commits
Differential Revision: https://reviews.llvm.org/D49160
llvm-svn: 338164
Fixed the ASAN failure from before in r338148, so recommiting.
This patch enables the MachineOutliner by default in AArch64 under -Oz.
The MachineOutliner offers around a 4.5% improvement on the current -Oz code
size improvements.
We have done work into improving the debuggability of outlined code, so that
users of -Oz won't be surprised by the optimization. We have also been executing
the LLVM test suite and common external tests such as the SPEC suites
continuously with no issue. The outliner has a low compile-time overhead of
roughly 1%. At this point, the outliner would be a really good addition to the
-Oz pass pipeline!
llvm-svn: 338160
There was a missing check for if a candidate list was entirely deleted. This
adds that check.
This fixes an asan failure caused by running test/CodeGen/AArch64/addsub_ext.ll
with the MachineOutliner enabled.
llvm-svn: 338148
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
Errors like the following are reported by:
https://urldefense.proofpoint.com/v2/url?u=http-3A__lab.llvm.org-3A8011_builders_llvm-2Dclang-2Dx86-5F64-2Dexpensive-2Dchecks-2Dwin_builds_11261&d=DwIBAg&c=5VD0RTtNlTh3ycd41b3MUw&r=DA8e1B5r073vIqRrFz7MRA&m=929oWPCf7Bf2qQnir4GBtowB8ZAlIRWsAdTfRkDaK-g&s=9k-wbEUVpUm474hhzsmAO29VXVvbxJPWD9RTgCD71fQ&e=
*** Bad machine code: Explicit definition marked as use ***
- function: cal_align1
- basic block: %bb.0 entry (0x47edd98)
- instruction: LDB $r3, $r2, 0
- operand 0: $r3
This is because RegState info was missing for ScratchReg inside
expandMEMCPY. This caused incomplete register usage information to
MachineInstr verifier which then would complain as there could be potential
code-gen issue if the complained MachineInstr is used in place where
register usage information matters even though the memcpy expanding is not
in such case as it happens at the last stage of IR optimization pipeline.
We should always specify those register usage information which compiler
couldn't deduct automatically whenever we add a hardware register manually.
Reported-by: Builder llvm-clang-x86_64-expensive-checks-win Build #11261
Signed-off-by: Jiong Wang <jiong.wang@netronome.com>
Reviewed-by: Yonghong Song <yhs@fb.com>
llvm-svn: 338134
This patch enables the MachineOutliner by default in AArch64 under -Oz.
The MachineOutliner offers around a 4.5% improvement on the current -Oz code
size improvements.
We have done work into improving the debuggability of outlined code, so that
users of -Oz won't be surprised by the optimization. We have also been executing
the LLVM test suite and common external tests such as the SPEC suites
continuously with no issue. The outliner has a low compile-time overhead of
roughly 1%. At this point, the outliner would be a really good addition to the
-Oz pass pipeline!
llvm-svn: 338133
R600 can't handle immediates for BFE, these will be eliminated later.
Fixes powr/pow regressions n r600 since r334817
Differential Revision: https://reviews.llvm.org/D49641
llvm-svn: 338127
This patch adds support for various integer reduction operations:
SADDV signed add reduction to scalar
UADDV unsigned add reduction to scalar
SMAXV signed maximum reduction to scalar
SMINV signed minimum reduction to scalar
UMAXV unsigned maximum reduction to scalar
UMINV unsigned minimum reduction to scalar
ANDV logical AND reduction to scalar
ORV logical OR reduction to scalar
EORV logical EOR reduction to scalar
The reduction is predicated, e.g.
smaxv s0, p0, z1.s
performs a signed maximum reduction on active elements in z1,
and stores the (signed max value) result in s0.
llvm-svn: 338126
This patch adds support for various floating-point
reduction operations:
FADDA strictly-ordered add reduction, accumulating in scalar
FADDV recursive add reduction to scalar
FMAXV recursive max reduction to scalar
FMINV recursive min reduction to scalar
FMAXNMV recursive max number reduction to scalar
FMINNMV recursive min number reduction to scalar
The reduction is predicated, e.g.
fadda d0, p0, d0, z1.d
performs the add-reduction in strict order on active elements
in z1, accumulating into d0.
faddv d0, p0, z1.d
performs the add-reduction (not in strict order)
on active elements in z1, storing the result in d0.
llvm-svn: 338123
This patch adds support for transcendental acceleration
instructions 'FEXPA' (exponential accelerator) and 'FTSSEL'
(trigonometric select coefficient).
llvm-svn: 338121
Not sure why they were being explicitly excluded, but I believe all the math inside the if works. I changed the absolute value to be uint64_t instead of int64_t so INT64_MIN+1 wouldn't be signed wrap.
llvm-svn: 338101
Summary:
This is the pattern you get from the loop vectorizer for something like this
int16_t A[1024];
int16_t B[1024];
int32_t C[512];
void pmaddwd() {
for (int i = 0; i != 512; ++i)
C[i] = (A[2*i]*B[2*i]) + (A[2*i+1]*B[2*i+1]);
}
In this case we will have (add (mul (build_vector), (build_vector)), (mul (build_vector), (build_vector))). This is different than the pattern we currently match which has the build_vectors between an add and a single multiply. I'm not sure what C code would get you that pattern.
Reviewers: RKSimon, spatel, zvi
Reviewed By: zvi
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D49636
llvm-svn: 338097
If this happens the operands aren't updated and the existing node is returned. Make sure we pass this existing node up to the DAG combiner so that a proper replacement happens. Otherwise we get stuck in an infinite loop with an unoptimized node.
llvm-svn: 338090
Scale the offset of VGPR spills by the wave size when it cannot fit in the
12-bit offset immediate field and so is added to the soffset SGPR. This
accounts for hardware swizzling of scratch memory.
Differential Revision: https://reviews.llvm.org/D49448
llvm-svn: 338060
- Save/restore only registers that are used.
This includes Callee saved registers and Caller saved registers
(arguments and temporaries) for integer and FP registers.
- If there is a call in the interrupt handler, save/restore all
Caller saved registers (arguments and temporaries) and all FP registers.
- Emit special return instructions depending on "interrupt"
attribute type.
Based on initial patch by Zhaoshi Zheng.
Reviewers: asb
Reviewed By: asb
Subscribers: rkruppe, the_o, MartinMosbeck, brucehoult, rbar, johnrusso, simoncook, sabuasal, niosHD, kito-cheng, shiva0217, zzheng, edward-jones, mgrang, rogfer01, llvm-commits
Differential Revision: https://reviews.llvm.org/D48411
llvm-svn: 338047
Summary:
NVPTX target dos not use register-based frame information. Instead it
relies on the artificial local_depot that is used instead of the frame
and the data for variables must be emitted relatively to this
local_depot.
Reviewers: tra, jlebar, echristo
Subscribers: jholewinski, aprantl, JDevlieghere, llvm-commits
Differential Revision: https://reviews.llvm.org/D45963
llvm-svn: 338039
- Some of the v8.3 pointer authentication instruction inhabit the Hint space
- These instructions can be assembled to hint instructions which act as NOP instructions prior to v8.3
- This patch permits using the hint instructions for all v8a targets
- Also, correct the RETA{A,B} instructions to match the instruction attributes of RET (set isTerminator and isBarrier)
Differential Revision: https://reviews.llvm.org/D49786
llvm-svn: 338029
Override getTypeForExtReturn so that functions returning
an i32 typed value have it sign extended on MIPS64.
Also provide patterns to get rid of unneeded sign extensions
for arithmetic instructions which implicitly sign extend
their results.
Differential Revision: https://reviews.llvm.org/D48374
llvm-svn: 338019
a helper function with a nice overview comment. NFC.
This is a preperatory refactoring to implementing another component of
mitigation here that was descibed in the design document but hadn't been
implemented yet.
llvm-svn: 338016
This adds MC support for the crypto instructions that were made optional
extensions in Armv8.2-A (AArch64 only).
Differential Revision: https://reviews.llvm.org/D49370
llvm-svn: 338010
I'm not sure if this was trying to avoid optimizing the new nodes further or what. Or maybe to prevent a cycle if something tried to reform the multiply? But I don't think its a reliable way to do that. If the user of the expanded multiply is visited by the DAGCombiner after this conversion happens, the DAGCombiner will check its operands, see that they haven't been visited by the DAGCombiner before and it will then add the first node to the worklist. This process will repeat until all the new nodes are visited.
So this seems like an unreliable prevention at best. So this patch just returns the new nodes like any other combine. If this starts causing problems we can try to add target specific nodes or something to more directly prevent optimizations.
Now that we handle the combine normally, we can combine any negates the mul expansion creates into their users since those will be visited now.
llvm-svn: 338007
These calls were making sure some newly created nodes were added to worklist, but the DAGCombiner has internal support for ensuring it has visited all nodes. Any time it visits a node it ensures the operands have been queued to be visited as well. This means if we only need to return the last new node. The DAGCombiner will take care of adding its inputs thus walking backwards through all the new nodes.
llvm-svn: 337996
- Avoid duplication of regmask size calculation.
- Simplify allocateRegisterMask() call.
- Rename allocateRegisterMask() to allocateRegMask() to be consistent
with naming in MachineOperand.
llvm-svn: 337986
Some BPF JIT backends would want to optimize memcpy in their own
architecture specific way.
However, at the moment, there is no way for JIT backends to see memcpy
semantics in a reliable way. This is due to LLVM BPF backend is expanding
memcpy into load/store sequences and could possibly schedule them apart from
each other further. So, BPF JIT backends inside kernel can't reliably
recognize memcpy semantics by peephole BPF sequence.
This patch introduce new intrinsic expand infrastructure to memcpy.
To get stable in-order load/store sequence from memcpy, we first lower
memcpy into BPF::MEMCPY node which then expanded into in-order load/store
sequences in expandPostRAPseudo pass which will happen after instruction
scheduling. By this way, kernel JIT backends could reliably recognize
memcpy through scanning BPF sequence.
This new memcpy expand infrastructure is gated by a new option:
-bpf-expand-memcpy-in-order
Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Jiong Wang <jiong.wang@netronome.com>
Signed-off-by: Yonghong Song <yhs@fb.com>
llvm-svn: 337977
In SVN r334523, the first half of comdat constant pool handling was
hoisted from X86WindowsTargetObjectFile (which despite the name only
was used for msvc targets) into the arch independent
TargetLoweringObjectFileCOFF, but the other half of the handling was
left behind in X86AsmPrinter::GetCPISymbol.
With only half of the handling in place, inconsistent comdat
sections/symbols are created, causing issues with both GNU binutils
(avoided for X86 in SVN r335918) and with the MS linker, which
would complain like this:
fatal error LNK1143: invalid or corrupt file: no symbol for COMDAT section 0x4
Differential Revision: https://reviews.llvm.org/D49644
llvm-svn: 337950
Saves materializing the immediate for the "ands".
Corresponding patterns exist for lsrs+lsls, but that seems less common
in practice.
Now implemented as a DAGCombine.
Differential Revision: https://reviews.llvm.org/D49585
llvm-svn: 337945
For example v = <2 x i1> is represented as bbbbaaaa in a predicate register,
where b = v[1], a = v[0]. Extracting v[1] is equivalent to extracting bit 4
from the predicate register.
llvm-svn: 337934
Add support for lowering pointer arguments.
Changing type from pointer to integer is already done in
MipsTargetLowering::getRegisterTypeForCallingConv.
Patch by Petar Avramovic.
Differential Revision: https://reviews.llvm.org/D49419
llvm-svn: 337912
NFC changes to make scheduler TableGen files more readable, by using loops
instead of a lot of similar defs with just e.g. a latency value that changes.
https://reviews.llvm.org/D49598
Review: Ulrich Weigand, Javed Abshar
llvm-svn: 337909
code.
This consolidates all our hardening calls, and simplifies the code
a bit. It seems much more clear to handle all of these together.
No functionality changed here.
llvm-svn: 337895
This function actually does two things: it traces the predicate state
through each of the basic blocks in the function (as that isn't directly
handled by the SSA updater) *and* it hardens everything necessary in the
block as it goes. These need to be done together so that we have the
currently active predicate state to use at each point of the hardening.
However, this also made obvious that the flag to disable actual
hardening of loads was flawed -- it also disabled tracing the predicate
state across function calls within the body of each block. So this patch
sinks this debugging flag test to correctly guard just the hardening of
loads.
Unless load hardening was disabled, no functionality should change with
tis patch.
llvm-svn: 337894
The target independent AsmParser doesn't recognise .hword, .word, .dword
which are required for Mips. Currently MipsAsmParser recognises these
through dispatch to MipsAsmParser::parseDataDirective. This contains
equivalent logic to AsmParser::parseDirectiveValue. This patch allows
reuse of AsmParser::parseDirectiveValue by making use of
addAliasForDirective to support .hword, .word and .dword.
Original patch provided by Alex Bradbury at D47001 was modified to fix
handling of microMIPS symbols. The `AsmParser::parseDirectiveValue`
calls either `EmitIntValue` or `EmitValue`. In this patch we override
`EmitIntValue` in the `MipsELFStreamer` to clear a pending set of
microMIPS symbols.
Differential revision: https://reviews.llvm.org/D49539
llvm-svn: 337893
against v1.2 BCBS attacks directly.
Attacks using spectre v1.2 (a subset of BCBS) are described in the paper
here:
https://people.csail.mit.edu/vlk/spectre11.pdf
The core idea is to speculatively store over the address in a vtable,
jumptable, or other target of indirect control flow that will be
subsequently loaded. Speculative execution after such a store can
forward the stored value to subsequent loads, and if called or jumped
to, the speculative execution will be steered to this potentially
attacker controlled address.
Up until now, this could be mitigated by enableing retpolines. However,
that is a relatively expensive technique to mitigate this particular
flavor. Especially because in most cases SLH will have already mitigated
this. To fully mitigate this with SLH, we need to do two core things:
1) Unfold loads from calls and jumps, allowing the loads to be post-load
hardened.
2) Force hardening of incoming registers even if we didn't end up
needing to harden the load itself.
The reason we need to do these two things is because hardening calls and
jumps from this particular variant is importantly different from
hardening against leak of secret data. Because the "bad" data here isn't
a secret, but in fact speculatively stored by the attacker, it may be
loaded from any address, regardless of whether it is read-only memory,
mapped memory, or a "hardened" address. The only 100% effective way to
harden these instructions is to harden the their operand itself. But to
the extent possible, we'd like to take advantage of all the other
hardening going on, we just need a fallback in case none of that
happened to cover the particular input to the control transfer
instruction.
For users of SLH, currently they are paing 2% to 6% performance overhead
for retpolines, but this mechanism is expected to be substantially
cheaper. However, it is worth reminding folks that this does not
mitigate all of the things retpolines do -- most notably, variant #2 is
not in *any way* mitigated by this technique. So users of SLH may still
want to enable retpolines, and the implementation is carefuly designed to
gracefully leverage retpolines to avoid the need for further hardening
here when they are enabled.
Differential Revision: https://reviews.llvm.org/D49663
llvm-svn: 337878
We generated a subtract for the power of 2 minus one then negated the result. The negate can be optimized away by swapping the subtract operands, but DAG combine doesn't know how to do that and we don't add any of the new nodes to the worklist anyway.
This patch makes use explicitly emit the swapped subtract.
llvm-svn: 337858
Use a left shift and 2 subtracts like we do for 30. Move this out from behind the slow lea check since it doesn't even use an LEA.
Use this for multiply by 14 as well.
llvm-svn: 337856
Just some gardening here.
Similar to how we moved call information into Candidates, this moves outlined
frame information into OutlinedFunction. This allows us to remove
TargetCostInfo entirely.
Anywhere where we returned a TargetCostInfo struct, we now return an
OutlinedFunction. This establishes OutlinedFunctions as more of a general
repeated sequence, and Candidates as occurrences of those repeated sequences.
llvm-svn: 337848
When building with LTO, builtin functions that are defined but whose calls have not been inserted yet, get internalized. The Global Dead Code Elimination phase in the new LTO implementation then removes these function definitions. Later optimizations add calls to those functions, and the linker then dies complaining that there are no definitions. This CL fixes the new LTO implementation to check if a function is builtin, and if so, to not internalize (and later DCE) the function. As part of this fix I needed to move the RuntimeLibcalls.{def,h} files from the CodeGen subidrectory to the IR subdirectory. I have updated all the files that accessed those two files to access their new location.
Fixes PR34169
Patch by Caroline Tice!
Differential Revision: https://reviews.llvm.org/D49434
llvm-svn: 337847
Summary:
Enabling this fully exposes a latent bug in the instruction folding: we
never update the register constraints for the register operands when
fusing a load into another operation. The fused form could, in theory,
have different register constraints on its operands. And in fact,
TCRETURNm* needs its memory operands to use tailcall compatible
registers.
I've updated the folding code to re-constrain all the registers after
they are mapped onto their new instruction.
However, we still can't enable folding in the general case from
TCRETURNr* to TCRETURNm* because doing so may require more registers to
be available during the tail call. If the call itself uses all but one
register, and the folded load would require both a base and index
register, there will not be enough registers to allocate the tail call.
It would be better, IMO, to teach the register allocator to *unfold*
TCRETURNm* when it runs out of registers (or specifically check the
number of registers available during the TCRETURNr*) but I'm not going
to try and solve that for now. Instead, I've just blocked the forward
folding from r -> m, leaving LLVM free to unfold from m -> r as that
doesn't introduce new register pressure constraints.
The down side is that I don't have anything that will directly exercise
this. Instead, I will be immediately using this it my SLH patch. =/
Still worse, without allowing the TCRETURNr* -> TCRETURNm* fold, I don't
have any tests that demonstrate the failure to update the memory operand
register constraints. This patch still seems correct, but I'm nervous
about the degree of testing due to this.
Suggestions?
Reviewers: craig.topper
Subscribers: sanjoy, mcrosier, hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D49717
llvm-svn: 337845
Before this, TCI contained all the call information for each Candidate.
This moves that information onto the Candidates. As a result, each Candidate
can now supply how it ought to be called. Thus, Candidates will be able to,
say, call the same function in cheaper ways when possible. This also removes
that information from TCI, since it's no longer used there.
A follow-up patch for the AArch64 outliner will demonstrate this.
llvm-svn: 337840
For the final DTPREL addition, rather than a lui/daddiu/daddu triple,
LLVM was erronously emitting a daddiu/daddiu pair, treating the %dtprel_hi
as if it were a %dtprel_lo, since Mips::Hi expands unshifted for Sym64.
Instead, use a new TlsHi node and, although unnecessary due to the exact
structure of the nodes emitted, use TlsHi for local exec too to prevent
future bugs. Also garbage-collect the unused TprelLo and TlsGd nodes,
and TprelHi since its functionality is provided by the new common TlsHi node.
Patch by James Clarke.
Differential revision: https://reviews.llvm.org/D49259
llvm-svn: 337827
helper and restructure the post-load hardening to use this.
This isn't as trivial as I would have liked because the post-load
hardening used a trick that only works for it where it swapped in
a temporary register to the load rather than replacing anything.
However, there is a simple way to do this without that trick that allows
this to easily reuse a friendly API for hardening a value in a register.
That API will in turn be usable in subsequent patcehs.
This also techincally changes the position at which we insert the subreg
extraction for the predicate state, but that never resulted in an actual
instruction and so tests don't change at all.
llvm-svn: 337825
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
Summary:
We were marking G_EXTRACT operations unsupported if the output type
was larger than the input type. I don't see how this could ever actually
happen, so I dropped the constraint. Doing this makes it possible to
reuse the same legality code for G_INSERT.
Reviewers: arsenm
Reviewed By: arsenm
Subscribers: kzhuravl, wdng, nhaehnle, yaxunl, rovka, kristof.beyls, dstuttard, tpr, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D49600
llvm-svn: 337794
This code was really nasty, had several bugs in it originally, and
wasn't carrying its weight. While on Zen we have all 4 ports available
for SHRX, on all of the Intel parts with Agner's tables, SHRX can only
execute on 2 ports, giving it 1/2 the throughput of OR.
Worse, all too often this pattern required two SHRX instructions in
a chain, hurting the critical path by a lot.
Even if we end up needing to safe/restore EFLAGS, that is no longer so
bad. We pay for a uop to save the flag, but we very likely get fusion
when it is used by forming a test/jCC pair or something similar. In
practice, I don't expect the SHRX to be a significant savings here, so
I'd like to avoid the complex code required. We can always resurrect
this if/when someone has a specific performance issue addressed by it.
llvm-svn: 337781
This matches the structure used on X86 and ARM. This requires
a little bit of duplication of the parts that are equal in both
AArch64 COFF variants though.
Before SVN r335286, these classes didn't add anything that MCAsmInfoCOFF
didn't, but now they do.
This makes AArch64 match X86 in how comdat is used for float constants
for MinGW.
Differential Revision: https://reviews.llvm.org/D49637
llvm-svn: 337755
Don't try to generate large PIC code for non-ELF targets. Neither COFF
nor MachO have relocations for large position independent code, and
users have been using "large PIC" code models to JIT 64-bit code for a
while now. With this change, if they are generating ELF code, their
JITed code will truly be PIC, but if they target MachO or COFF, it will
contain 64-bit immediates that directly reference external symbols. For
a JIT, that's perfectly fine.
llvm-svn: 337740
Summary:
OpChain has subclasses, so add a virtual destructor.
This fixes an issue when deleting subclasses of OpChain (see MatchSMLAD() specifically) in r337701.
Reviewers: javed.absar
Subscribers: llvm-commits, SjoerdMeijer, samparker
Differential Revision: https://reviews.llvm.org/D49681
llvm-svn: 337713
In preparing to allow ARMParallelDSP pass to parallelise more than
smlads, I've restructed some elements:
- The ParallelMAC struct has been renamed to BinOpChain.
- The BinOpChain struct holds two value lists: LHS and RHS, as well
as inheriting from the OpChain base class.
- The OpChain struct holds all the values of the represented chain
and has had the memory locations functionality inserted into it.
- ParallelMACList becomes OpChainList and it now holds pointers
instead of objects.
Differential Revision: https://reviews.llvm.org/D49020
llvm-svn: 337701
Two minor issues: The new MCD SchedWrite name does not contain "Unit" like
all the others, so a check is needed. Also, print "LSU" instead of "LS".
Review: Ulrich Weigand
llvm-svn: 337700
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
Summary:
Pretty mechanical follow-up for D49196.
As microarchitecture.pdf notes, "20 AMD Ryzen pipeline",
"20.8 Register renaming and out-of-order schedulers":
The integer register file has 168 physical registers of 64 bits each.
The floating point register file has 160 registers of 128 bits each.
"20.14 Partial register access":
The processor always keeps the different parts of an integer register together.
...
An instruction that writes to part of a register will therefore have a false dependence
on any previous write to the same register or any part of it.
Reviewers: andreadb, courbet, RKSimon, craig.topper, GGanesh
Reviewed By: GGanesh
Subscribers: gbedwell, llvm-commits
Differential Revision: https://reviews.llvm.org/D49393
llvm-svn: 337676
a call, and then again as a return.
Also added a comment to try and explain better why we would be doing
what we're doing when hardening the (non-call) returns.
llvm-svn: 337673
This provides an overview of the algorithm used to harden specific
loads. It also brings this our terminology further in line with
hardening rather than checking.
Differential Revision: https://reviews.llvm.org/D49583
llvm-svn: 337667
This seems to be a net improvement. There's still an issue under avx512f where we have a 512-bit vpaddd, but not vpmaddwd so we end up doing two 256-bit vpmaddwds and inserting the results before a 512-bit vpaddd. It might be better to do two 512-bits paddds with zeros in the upper half. Same number of instructions, but breaks a dependency.
llvm-svn: 337656
This is a follow-up to the rL335185. Those commit adds some WrapperPat
patterns for microMIPS target. But declaration of the WrapperPat class
is under the NotInMicroMips predicate and microMIPS patterns cannot be
selected because predicate (Subtarget->inMicroMipsMode()) &&
(!Subtarget->inMicroMipsMode()) is always false.
This change move out the WrapperPat class declaration from the
NotInMicroMips predicate and enables microMIPS WrapperPat patterns.
Differential revision: https://reviews.llvm.org/D49533
llvm-svn: 337646
Ideally our ISD node types going into the isel table would have types consistent with their instruction domain. This prevents us having to duplicate patterns with different types for the same instruction.
Unfortunately, it seems our shuffle combining is currently relying on this a little remove some bitcasts. This seems to enable some switching between shufps and shufd. Hopefully there's some way we can address this in the combining.
Differential Revision: https://reviews.llvm.org/D49280
llvm-svn: 337590
CombineTo is most useful when you need to replace multiple results, avoid the worklist management, or you need to something else after the combine, etc. Otherwise you should be able to just return the new node and let DAGCombiner go through its usual worklist code.
All of the places changed in this patch look to be standard cases where we should be able to use the more stand behavior of just returning the new node.
Differential Revision: https://reviews.llvm.org/D49569
llvm-svn: 337589
We can safely use getConstant here as we're still lowering, which allows constant folding to kick in and simplify the vector shift codegen.
Noticed while working on D49562.
llvm-svn: 337578
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
This is an early step towards using SimplifyDemandedVectorElts for target shuffle combining - this merely moves the existing X86ISD::VBROADCAST simplification code to use the SimplifyDemandedVectorElts mechanism.
Adds X86TargetLowering::SimplifyDemandedVectorEltsForTargetNode to handle X86ISD::VBROADCAST - in time we can support all target shuffles (and other ops) here.
llvm-svn: 337547
As a consequence of recent discussions
(http://lists.llvm.org/pipermail/llvm-dev/2018-May/123164.html), this patch
changes the SystemZ SchedModels so that the IssueWidth is 6, which is the
decoder capacity, and NumMicroOps become the number of decoder slots needed
per instruction.
In addition, the SchedWrite latencies now match the MachineInstructions
def-operand indexes, and ReadAdvances have been added on instructions with
one register operand and one memory operand.
Review: Ulrich Weigand
https://reviews.llvm.org/D47008
llvm-svn: 337538
This patch adds the following instructions:
RBIT reverse bits within each active elemnt (predicated), e.g.
rbit z0.d, p0/m, z1.d
for 8, 16, 32 and 64 bit elements.
REV reverse order of elements in data/predicate vector
(unpredicated), e.g.
rev z0.d, z1.d
rev p0.d, p1.d
for 8, 16, 32 and 64 bit elements.
REVB reverse order of bytes within each active element, e.g.
revb z0.d, p0/m, z1.d
for 16, 32 and 64 bit elements.
REVH reverse order of 16-bit half-words within each active
element, e.g.
revh z0.d, p0/m, z1.d
for 32 and 64 bit elements.
REVW reverse order of 32-bit words within each active element,
e.g.
revw z0.d, p0/m, z1.d
for 64 bit elements.
llvm-svn: 337534
Summary:
lifetime2.C violates DR1696, which prevents reference members from being
initialized to temporaries, whose lifetime would end at the end of ctor.
Reviewers: sbc100
Subscribers: dschuff, sunfish, llvm-commits
Differential Revision: https://reviews.llvm.org/D49577
llvm-svn: 337512
remove dead declaration of a call instruction handling helper.
This moves to the 'harden' terminology that I've been trying to settle
on for returns. It also adds a really detailed comment explaining what
all we're trying to accomplish with return instructions and why.
Hopefully this makes it much more clear what exactly is being
"hardened".
Differential Revision: https://reviews.llvm.org/D49571
llvm-svn: 337510
We have a number of cases where we fail to reduce vector op widths, performing the op in a larger vector and then extracting a subvector. This is often because by default it would create illegal types.
This peephole patch attempts to handle a few common cases detailed in PR36761, which typically involved extension+conversion to vX2f64 types.
Differential Revision: https://reviews.llvm.org/D49556
llvm-svn: 337500
Returning SDValue() means nothing was changed. Returning the result of CombineTo returns the first argument of CombineTo. This is specially detected by DAGCombiner as meaning that something changed, but worklist management was already taken care of.
I think the only real effect of this change is that we now properly update the Statistic the counts the number of combines performed. That's the only thing between the check for null and the check for N in the DAGCombiner.
llvm-svn: 337491
As we already return true from needsAggressiveScheduling() for the most recent
hardware it would be cleaner to just return true for all PowerPC hardware.
Differential Revision: https://reviews.llvm.org/D48663
llvm-svn: 337488
This patch fixes the latency/throughput of LEA instructions in the BtVer2
scheduling model.
On Jaguar, A 3-operands LEA has a latency of 2cy, and a reciprocal throughput of
1. That is because it uses one cycle of SAGU followed by 1cy of ALU1. An LEA
with a "Scale" operand is also slow, and it has the same latency profile as the
3-operands LEA. An LEA16r has a latency of 3cy, and a throughput of 0.5 (i.e.
RThrouhgput of 2.0).
This patch adds a new TIIPredicate named IsThreeOperandsLEAFn to X86Schedule.td.
The tablegen backend (for instruction-info) expands that definition into this
(file X86GenInstrInfo.inc):
```
static bool isThreeOperandsLEA(const MachineInstr &MI) {
return (
(
MI.getOpcode() == X86::LEA32r
|| MI.getOpcode() == X86::LEA64r
|| MI.getOpcode() == X86::LEA64_32r
|| MI.getOpcode() == X86::LEA16r
)
&& MI.getOperand(1).isReg()
&& MI.getOperand(1).getReg() != 0
&& MI.getOperand(3).isReg()
&& MI.getOperand(3).getReg() != 0
&& (
(
MI.getOperand(4).isImm()
&& MI.getOperand(4).getImm() != 0
)
|| (MI.getOperand(4).isGlobal())
)
);
}
```
A similar method is generated in the X86_MC namespace, and included into
X86MCTargetDesc.cpp (the declaration lives in X86MCTargetDesc.h).
Back to the BtVer2 scheduling model:
A new scheduling predicate named JSlowLEAPredicate now checks if either the
instruction is a three-operands LEA, or it is an LEA with a Scale value
different than 1.
A variant scheduling class uses that new predicate to correctly select the
appropriate latency profile.
Differential Revision: https://reviews.llvm.org/D49436
llvm-svn: 337469
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
changes that are intertwined here:
1) Extracting the tracing of predicate state through the CFG to its own
function.
2) Creating a struct to manage the predicate state used throughout the
pass.
Doing #1 necessitates and motivates the particular approach for #2 as
now the predicate management is spread across different functions
focused on different aspects of it. A number of simplifications then
fell out as a direct consequence.
I went with an Optional to make it more natural to construct the
MachineSSAUpdater object.
This is probably the single largest outstanding refactoring step I have.
Things get a bit more surgical from here. My current goal, beyond
generally making this maintainable long-term, is to implement several
improvements to how we do interprocedural tracking of predicate state.
But I don't want to do that until the predicate state management and
tracing is in reasonably clear state.
Differential Revision: https://reviews.llvm.org/D49427
llvm-svn: 337446
Summary:
The use of exception handling instructions should only be enabled with
`-mattr=+exception-handling` option.
Reviewers: jgravelle-google
Subscribers: dschuff, sbc100, sunfish, llvm-commits
Differential Revision: https://reviews.llvm.org/D49391
llvm-svn: 337425
As discussed on PR38197, this canonicalizes MOVS*(N0, OP(N0, N1)) --> MOVS*(N0, SCALAR_TO_VECTOR(OP(N0[0], N1[0])))
This returns the scalar-fp codegen lost by rL336971.
Additionally it handles the OP(N1, N0)) case for commutable (FADD/FMUL) ops.
Differential Revision: https://reviews.llvm.org/D49474
llvm-svn: 337419
This is a follow-up to the rL337171. This patch fixes regression
introduced by the r337171 and enables MipsTruncIntFP pattern.
Differential revision: https://reviews.llvm.org/D49469
llvm-svn: 337392
When rL336971 removed the scalar-fp isel patterns, we lost the need for this canonicalization - commutation/folding can handle everything else.
llvm-svn: 337387
ARMSubtarget had a copy/pasted block to determine whether the target was
hard-float, but it just delegated to triple features anyway so it's better at
the TargetMachine level.
llvm-svn: 337384
This patch adds support for the following unpredicated
floating-point instructions:
FADD Floating point add
FSUB Floating point subtract
FMUL Floating point multiplication
FTSMUL Floating point trigonometric starting value
FRECPS Floating point reciprocal step
FRSQRTS Floating point reciprocal square root step
The instructions have the following assembly format:
fadd z0.h, z1.h, z2.h
and have variants for 16, 32 and 64-bit FP elements.
llvm-svn: 337383
This reverts commit 55222c9183c6e07f53a54c4061677734f54feac1.
I missed that this patch has a dependency on https://reviews.llvm.org/D49219
that has not been approved yet.
llvm-svn: 337373
The signed/unsigned DOT instructions perform a dot-product on
quadtuplets from two source vectors and accumulate the result in
the destination register. The instructions come in two forms:
Vector form, e.g.
sdot z0.s, z1.b, z2.b - signed dot product on four 8-bit quad-tuplets,
accumulating results in 32-bit elements.
udot z0.d, z1.h, z2.h - unsigned dot product on four 16-bit quad-tuplets,
accumulating results in 64-bit elements.
Indexed form, e.g.
sdot z0.s, z1.b, z2.b[3] - signed dot product on four 8-bit quad-tuplets
with specified quadtuplet from second
source vector, accumulating results in 32-bit
elements.
udot z0.d, z1.h, z2.h[1] - dot product on four 16-bit quad-tuplets
with specified quadtuplet from second
source vector, accumulating results in 64-bit
elements.
llvm-svn: 337372
Summary: This is how it appears to be handled in GCC and it prevents a
"Unknown mismatch" error in the SelectionDAGBuilder.
Reviewers: venkatra, jyknight, jrtc27
Reviewed By: jyknight, jrtc27
Subscribers: eraman, fedor.sergeev, jrtc27, llvm-commits
Differential Revision: https://reviews.llvm.org/D49218
llvm-svn: 337370
This patch adds the following predicated instructions:
UDIV Unsigned divide active elements
UDIVR Unsigned divide active elements, reverse form.
SDIV Signed divide active elements
SDIVR Signed divide active elements, reverse form.
e.g.
udiv z0.s, p0/m, z0.s, z1.s
(unsigned divide active elements in z0 by z1, store result in z0)
sdivr z0.s, p0/m, z0.s, z1.s
(signed divide active elements in z1 by z0, store result in z0)
llvm-svn: 337369
This patch adds the following instructions:
MUL - multiply vectors, e.g.
mul z0.h, p0/m, z0.h, z1.h
- multiply with immediate, e.g.
mul z0.h, z0.h, #127
SMULH - signed multiply returning high half, e.g.
smulh z0.h, p0/m, z0.h, z1.h
UMULH - unsigned multiply returning high half, e.g.
umulh z0.h, p0/m, z0.h, z1.h
llvm-svn: 337358
* Delete a no-longer-used override, and mark the other
getRegisterTypeForCallingConv() as override.
* SPE only supports i32, not i64, as the internal type, so simply remove
the type check, so that DestReg and Opc are provably always set.
GCC 6.4 did not warn about either of the above.
llvm-svn: 337350
The X86ISD::MOVLHPS/MOVHLPS should now only be emitted in SSE1 only. This means that the v2i64/v2f64 types would be illegal thus we don't need these patterns.
llvm-svn: 337349
I'm trying to restrict the MOVLHPS/MOVHLPS ISD nodes to SSE1 only. With SSE2 we can use unpcks. I believe this will allow some patterns to be cleaned up to require fewer bitcasts.
I've put in an odd isel hack to still select MOVHLPS instruction from the unpckh node to avoid changing tests and because movhlps is a shorter encoding. Ideally we'd do execution domain switching on this, but the operands are in the wrong order and are tied. We might be able to try a commute in the domain switching using custom code.
We already support domain switching for UNPCKLPD and MOVLHPS.
llvm-svn: 337348
Summary:
The Signal Processing Engine (SPE) is found on NXP/Freescale e500v1,
e500v2, and several e200 cores. This adds support targeting the e500v2,
as this is more common than the e500v1, and is in SoCs still on the
market.
This patch is very intrusive because the SPE is binary incompatible with
the traditional FPU. After discussing with others, the cleanest
solution was to make both SPE and FPU features on top of a base PowerPC
subset, so all FPU instructions are now wrapped with HasFPU predicates.
Supported by this are:
* Code generation following the SPE ABI at the LLVM IR level (calling
conventions)
* Single- and Double-precision math at the level supported by the APU.
Still to do:
* Vector operations
* SPE intrinsics
As this changes the Callee-saved register list order, one test, which
tests the precise generated code, was updated to account for the new
register order.
Reviewed by: nemanjai
Differential Revision: https://reviews.llvm.org/D44830
llvm-svn: 337347
This is the lead-up to having SPE codegen. Add the rest of the
instructions, along with MC tests.
Differential Revision: https://reviews.llvm.org/D44829
llvm-svn: 337346
Summary:
The only thing he suggested that I've skipped here is the double-wide
multiply instructions. Multiply is an area I'm nervous about there being
some hidden data-dependent behavior, and it doesn't seem important for
any benchmarks I have, so skipping it and sticking with the minimal
multiply support that matches what I know is widely used in existing
crypto libraries. We can always add double-wide multiply when we have
clarity from vendors about its behavior and guarantees.
I've tried to at least cover the fundamentals here with tests, although
I've not tried to cover every width or permutation. I can add more tests
where folks think it would be helpful.
Reviewers: craig.topper
Subscribers: sanjoy, mcrosier, hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D49413
llvm-svn: 337308
Previously we were assuming whole program compilation. Now that
separate compilation is a thing we need to update this pass.
Firstly, it can no longer assert on the existence of malloc and free.
This functions might not be in the current translation unit. If we
need them then we will generate not imports for them.
Secondly the global helper function we create should be marked as
weak since we will be generating a separate copy in each translation
unit.
Finally the names of the symbols used must be unique and fixed since
they need to agree across translation units.
Differential Revision: https://reviews.llvm.org/D49263
llvm-svn: 337301
Previously we passed 'null_frag' into the instruction definition. The multiclass is shared with MOVHPD which doesn't use null_frag. It turns out by passing X86Movsd it produces patterns equivalent to some standalone patterns.
llvm-svn: 337299
The Mips FastISel back-end does not extend i1 values while lowering icmp.
Ensure that we bail into DAG ISel when handling this case.
Patch by Dragan Mladjenovic.
Differential Revision: https://reviews.llvm.org/D49290
llvm-svn: 337288
This patch completes support for the following floating point
instructions that take FP immediates:
FADD* (addition)
FSUB (subtract)
FSUBR (subtract reverse form)
FMUL* (multiplication)
FMAX* (maximum)
FMAXNM (maximum number)
FMIN (maximum)
FMINNM (maximum number)
All operations are predicated and take a FP immediate operand,
e.g.
fadd z0.h, p0/m, z0.h, #0.5
fmin z0.s, p0/m, z0.s, #1.0
^___________^ (tied)
* Instructions added in a previous patch.
llvm-svn: 337272
rL333307 was introduced to remove automatic target triple
normalization when calling sys::getDefaultTargetTriple(), arguing
that users of the latter already called Triple::normalize()
if necessary. However, users of the C API currently have no way of
doing target triple normalization.
This patch introduces an LLVMNormalizeTargetTriple function to
the C API which wraps Triple::normalize() and can be used on
the result of LLVMGetDefaultTargetTriple to achieve the same effect.
Differential Revision: https://reviews.llvm.org/D49414
Reviewed By: whitequark
llvm-svn: 337263
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
The SPLICE instruction splices two vectors into one vector using a
predicate. It copies the active elements from the first vector, and
then fills the remaining elements with the low-numbered elements from
the second vector.
The instruction has the following form, e.g.
splice z0.b, p0, z0.b, z1.b
for 8-bit elements. It also supports 16, 32 and
64-bit elements.
llvm-svn: 337253
This patch adds an instruction that allows extracting
a vector from a pair of vectors, given an immediate index
that describes the element position to extract from.
The instruction has the following assembly:
ext z0.b, z0.b, z1.b, #imm
where #imm is an immediate between 0 and 255.
llvm-svn: 337251
The ta instruction will always trap, regardless of the value
of the integer condition codes. TRAPri is marked as using icc,
so we cannot use a pattern for TRAPri to implement ta 1, as
verify-machineinstrs can complain that icc is not defined.
Instead we implement ta 1 the same way as ta 5.
llvm-svn: 337236
This amounts to pretty ridiculous number of patterns. Ideally we'd canonicalize the X86ISD::VRNDSCALE earlier to reuse those patterns. I briefly looked into doing that, but some strict FP operations could still get converted to rint and nearbyint during isel. It's probably still worthwhile to look into. This patch is meant as a starting point to work from.
llvm-svn: 337234
This allows us to use 231 form to fold an insertelement on the add input to the fma. There is technically no software intrinsic that can use this until AVX512F, but it can be manually built up from other intrinsics.
llvm-svn: 337223
This support was partial and temporary. Now that we have
wasm object file support its no longer needed.
Differential Revision: https://reviews.llvm.org/D48744
llvm-svn: 337222
invariant instructions to be both more correct and much more powerful.
While testing, I continued to find issues with sinking post-load
hardening. Unfortunately, it was amazingly hard to create any useful
tests of this because we were mostly sinking across copies and other
loading instructions. The fact that we couldn't sink past normal
arithmetic was really a big oversight.
So first, I've ported roughly the same set of instructions from the data
invariant loads to also have their non-loading varieties understood to
be data invariant. I've also added a few instructions that came up so
often it again made testing complicated: inc, dec, and lea.
With this, I was able to shake out a few nasty bugs in the validity
checking. We need to restrict to hardening single-def instructions with
defined registers that match a particular form: GPRs that don't have
a NOREX constraint directly attached to their register class.
The (tiny!) test case included catches all of the issues I was seeing
(once we can sink the hardening at all) except for the NOREX issue. The
only test I have there is horrible. It is large, inexplicable, and
doesn't even produce an error unless you try to emit encodings. I can
keep looking for a way to test it, but I'm out of ideas really.
Thanks to Ben for giving me at least a sanity-check review. I'll follow
up with Craig to go over this more thoroughly post-commit, but without
it SLH crashes everywhere so landing it for now.
Differential Revision: https://reviews.llvm.org/D49378
llvm-svn: 337177
Instead, the pattern is tagged with the correct predicate when
it is declared. Some patterns have been duplicated as necessary.
Patch by Simon Dardis.
Differential revision: https://reviews.llvm.org/D48365
llvm-svn: 337171
Add code for selection of G_LOAD, G_STORE, G_GEP, G_FRAMEINDEX and
G_CONSTANT. Support loads and stores of i32 values.
Patch by Petar Avramovic.
Differential Revision: https://reviews.llvm.org/D48957
llvm-svn: 337168
Summary:
[[ https://bugs.llvm.org/show_bug.cgi?id=38149 | PR38149 ]]
As discussed in https://reviews.llvm.org/D49179#1158957 and later,
the IR for 'check for [no] signed truncation' pattern can be improved:
https://rise4fun.com/Alive/gBf
^ that pattern will be produced by Implicit Integer Truncation sanitizer,
https://reviews.llvm.org/D48958https://bugs.llvm.org/show_bug.cgi?id=21530
in signed case, therefore it is probably a good idea to improve it.
But the IR-optimal patter does not lower efficiently, so we want to undo it..
This handles the simple pattern.
There is a second pattern with predicate and constants inverted.
NOTE: we do not check uses here. we always do the transform.
Reviewers: spatel, craig.topper, RKSimon, javed.absar
Reviewed By: spatel
Subscribers: kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D49266
llvm-svn: 337166
Summary: These are the names used in libgcc.
Reviewers: venkatra, jyknight, ekedaigle
Reviewed By: jyknight
Subscribers: joerg, fedor.sergeev, jrtc27, llvm-commits
Differential Revision: https://reviews.llvm.org/D48915
llvm-svn: 337164
Summary: Software trap number one is the trap used for breakpoints
in the Sparc ABI.
Reviewers: jyknight, venkatra
Reviewed By: jyknight
Subscribers: fedor.sergeev, jrtc27, llvm-commits
Differential Revision: https://reviews.llvm.org/D48637
llvm-svn: 337163
Found cases that hit the assert I added. This patch factors the validity
checking into a nice helper routine and calls it when deciding to harden
post-load, and asserts it when doing so later.
I've added tests for the various ways of loading a floating point type,
as well as loading all vector permutations. Even though many of these go
to identical instructions, it seems good to somewhat comprehensively
test them.
I'm confident there will be more fixes needed here, I'll try to add
tests each time as I get this predicate adjusted.
llvm-svn: 337160
Re-apply "[AMDGPU][Waitcnt] fix "comparison of integers of different signs" build error""
( fe0a456510131f268e388c4a18a92f575c0db183 ), which was inadvertantly reverted via
2b2ee080f0164485562593b1b87291a48cea4a9a .
llvm-svn: 337156
Memory legalizer, waitcnt, and shrink passes can perturb the instructions,
which means that the post-RA hazard recognizer pass should run after them.
Otherwise, one of those passes may invalidate the work done by the hazard
recognizer. Note that this has adverse side-effect that any consecutive
S_NOP 0's, emitted by the hazard recognizer, will not be shrunk into a
single S_NOP <N>. This should be addressed in a follow-on patch.
Differential Revision: https://reviews.llvm.org/D49288
llvm-svn: 337154
This unfortunately requires a bunch of bitcasts to be added added to SUBREG_TO_REG, COPY_TO_REGCLASS, and instructions in output patterns. Otherwise tablegen seems to default to picking f128 and then we fail when something tries to get the register class for f128 which isn't always valid.
The test changes are because we were previously mixing fr128 and vr128 due to contrainRegClass finding FR128 first and passes like live range shrinking weren't handling that well.
llvm-svn: 337147
indices used by AVX2 and AVX-512 gather instructions.
The index vector is hardened by broadcasting the predicate state
into a vector register and then or-ing. We don't even have to worry
about EFLAGS here.
I've added a test for all of the gather intrinsics to make sure that we
don't miss one. A particularly interesting creation is the gather
prefetch, which needs to be marked as potentially "loading" to get the
correct behavior. It's a memory access in many ways, and is actually
relevant for SLH. Based on discussion with Craig in review, I've moved
it to be `mayLoad` and `mayStore` rather than generic side effects. This
matches how we model other prefetch instructions.
Many thanks to Craig for the review here.
Differential Revision: https://reviews.llvm.org/D49336
llvm-svn: 337144
AVX512F only has integer domain logic instructions. AVX512DQ added FP domain logic instructions.
Execution domain fixing runs before EVEX->VEX. So if we have AVX512F and not AVX512DQ we fail to do execution domain switching of the logic operations. This leads to mismatches in execution domain and more test differences.
This patch adds custom domain fixing that switches EVEX integer logic operations to VEX fp logic operations if XMM16-31 are not used.
llvm-svn: 337137
128-bit ops implicitly zero the upper bits. This should address the comment about domain crossing for the integer version without AVX2 since we can use a 128-bit VBLENDW without AVX2.
The only bad thing I see here is that we failed to reuse an vxorps in some of the tests, but I think that's already known issue.
llvm-svn: 337134
registers.
The goal of this patch is to improve the throughput analysis in llvm-mca for the
case where instructions perform partial register writes.
On x86, partial register writes are quite difficult to model, mainly because
different processors tend to implement different register merging schemes in
hardware.
When the code contains partial register writes, the IPC (instructions per
cycles) estimated by llvm-mca tends to diverge quite significantly from the
observed IPC (using perf).
Modern AMD processors (at least, from Bulldozer onwards) don't rename partial
registers. Quoting Agner Fog's microarchitecture.pdf:
" The processor always keeps the different parts of an integer register together.
For example, AL and AH are not treated as independent by the out-of-order
execution mechanism. An instruction that writes to part of a register will
therefore have a false dependence on any previous write to the same register or
any part of it."
This patch is a first important step towards improving the analysis of partial
register updates. It changes the semantic of RegisterFile descriptors in
tablegen, and teaches llvm-mca how to identify false dependences in the presence
of partial register writes (for more details: see the new code comments in
include/Target/TargetSchedule.h - class RegisterFile).
This patch doesn't address the case where a write to a part of a register is
followed by a read from the whole register. On Intel chips, high8 registers
(AH/BH/CH/DH)) can be stored in separate physical registers. However, a later
(dirty) read of the full register (example: AX/EAX) triggers a merge uOp, which
adds extra latency (and potentially affects the pipe usage).
This is a very interesting article on the subject with a very informative answer
from Peter Cordes:
https://stackoverflow.com/questions/45660139/how-exactly-do-partial-registers-on-haswell-skylake-perform-writing-al-seems-to
In future, the definition of RegisterFile can be extended with extra information
that may be used to identify delays caused by merge opcodes triggered by a dirty
read of a partial write.
Differential Revision: https://reviews.llvm.org/D49196
llvm-svn: 337123
no conditions.
This is only valid to do if we're hardening calls and rets with LFENCE
which results in an LFENCE guarding the entire entry block for us.
llvm-svn: 337089
The code tried to find the immediate by using getNumOperands() on the MachineInstr, but there might be implicit-defs after the immediate that get counted.
Instead use getNumOperands() from the instruction description which will only count the operands that are defined in the td file.
llvm-svn: 337088
AVX512 doesn't have an immediate controlled blend instruction. But blend throughput is still better than movss/sd on SKX.
This commit changes AVX512 to use the AVX blend instructions instead of MOVSS/MOVSD. This constrains the register allocation since it won't be able to use XMM16-31, but hopefully the increased throughput and reduced port 5 pressure makes up for that.
llvm-svn: 337083
This reverts commit r337021.
WARNING: MemorySanitizer: use-of-uninitialized-value
#0 0x1415cd65 in void write_signed<long>(llvm::raw_ostream&, long, unsigned long, llvm::IntegerStyle) /code/llvm-project/llvm/lib/Support/NativeFormatting.cpp:95:7
#1 0x1415c900 in llvm::write_integer(llvm::raw_ostream&, long, unsigned long, llvm::IntegerStyle) /code/llvm-project/llvm/lib/Support/NativeFormatting.cpp:121:3
#2 0x1472357f in llvm::raw_ostream::operator<<(long) /code/llvm-project/llvm/lib/Support/raw_ostream.cpp:117:3
#3 0x13bb9d4 in llvm::raw_ostream::operator<<(int) /code/llvm-project/llvm/include/llvm/Support/raw_ostream.h:210:18
#4 0x3c2bc18 in void printField<unsigned int, &(amd_kernel_code_s::amd_kernel_code_version_major)>(llvm::StringRef, amd_kernel_code_s const&, llvm::raw_ostream&) /code/llvm-project/llvm/lib/Target/AMDGPU/Utils/AMDKernelCodeTUtils.cpp:78:23
#5 0x3c250ba in llvm::printAmdKernelCodeField(amd_kernel_code_s const&, int, llvm::raw_ostream&) /code/llvm-project/llvm/lib/Target/AMDGPU/Utils/AMDKernelCodeTUtils.cpp:104:5
#6 0x3c27ca3 in llvm::dumpAmdKernelCode(amd_kernel_code_s const*, llvm::raw_ostream&, char const*) /code/llvm-project/llvm/lib/Target/AMDGPU/Utils/AMDKernelCodeTUtils.cpp:113:5
#7 0x3a46e6c in llvm::AMDGPUTargetAsmStreamer::EmitAMDKernelCodeT(amd_kernel_code_s const&) /code/llvm-project/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUTargetStreamer.cpp:161:3
#8 0xd371e4 in llvm::AMDGPUAsmPrinter::EmitFunctionBodyStart() /code/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp:204:26
[...]
Uninitialized value was created by an allocation of 'KernelCode' in the stack frame of function '_ZN4llvm16AMDGPUAsmPrinter21EmitFunctionBodyStartEv'
#0 0xd36650 in llvm::AMDGPUAsmPrinter::EmitFunctionBodyStart() /code/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp:192
llvm-svn: 337079
If an HVX vector register is to be coalesced into a vector pair, make
sure that the vector pair will not have a function call in its live range,
unless it already had one. All HVX vector registers are volatile, so
any vector register live across a function call will have to be spilled.
If a vector needs to be spilled, and it's coalesced into a vector pair
then the whole pair will need to be spilled (even if only a part of it is
live), taking extra stack space.
llvm-svn: 337073
Ryzen has something like an 18 cycle latency on these based on Agner's data. AMD's own xls is blank. So it seems like there might be something tricky here.
Agner's data for Intel CPUs indicates these are a single uop there.
Probably safest to remove them. We never generate them without an intrinsic so this should be ok.
Differential Revision: https://reviews.llvm.org/D49315
llvm-svn: 337067
-Drop the intrinsic versions of conversion instructions. These should be handled when we do vectors. They shouldn't show up in scalar code.
-Add the float<->double conversions which were missing.
-Add the AVX512 and AVX version of the conversion instructions including the unsigned integer conversions unique to AVX512
Differential Revision: https://reviews.llvm.org/D49313
llvm-svn: 337066
-Move BSF/BSR to the same group as TZCNT/LZCNT/POPCNT.
-Split some of the bit manipulation instructions away from TZCNT/LZCNT/POPCNT. These are things like 'x & (x - 1)' which are composed of a few simple arithmetic operations. These aren't nearly as complicated/surprising as counting bits.
-Move BEXTR/BZHI into their own group. They aren't like a simple arithmethic op or the bit manipulation instructions. They're more like a shift+and.
Differential Revision: https://reviews.llvm.org/D49312
llvm-svn: 337065
These were supposed to be integer types since we are selecting integer instructions.
Found while preparing to remove these patterns for another patch.
llvm-svn: 337057
This needs to refer to arguments by their original argument
index, not the argument split index which depends on what
the type splitting decides to do.
Also avoid increment PSInputNum for each split piece.
llvm-svn: 337022
This was completely broken if there was ever a struct argument, as
this information is thrown away during the argument analysis.
The offsets as passed in to LowerFormalArguments are not useful,
as they partially depend on the legalized result register type,
and they don't consider the alignment in the first place.
Ignore the Ins array, and instead figure out from the raw IR type
what we need to do. This seems to fix the padding computation
if the DAG lowering is forced (and stops breaking arguments
following padded arguments if the arguments were only partially
lowered in the IR)
llvm-svn: 337021
Follow up of rL336913: fix base class description. Thanks to Ahmed Bougacha
for pointing this out.
Differential Revision: https://reviews.llvm.org/D49284
llvm-svn: 337009
Revision r322373 fixed a bug in how we materialize constants when the CR-field
needs to be set.
However the fix is overly conservative. It will only do the transform if
AND-ing the input with the new constant produces the same new constant.
This is of course correct, but not necessarily required.
If there are no futher uses of the constant, the constant can be changed.
If there are no uses of the GPR result, the final result of the materialization
isn't important other than it needs to compare to zero correctly (lt, gt, eq).
Differential revision: https://reviews.llvm.org/D42109
llvm-svn: 337008
This patch adds support for AArch64 to cfi-verify.
This required three changes to cfi-verify. First, it generalizes checking if an instruction is a trap by adding a new isTrap flag to TableGen (and defining it for x86 and AArch64). Second, the code that ensures that the operand register is not clobbered between the CFI check and the indirect call needs to allow a single dereference (in x86 this happens as part of the jump instruction). Third, we needed to ensure that return instructions are not counted as indirect branches. Technically, returns are indirect branches and can be covered by CFI, but LLVM's forward-edge CFI does not protect them, and x86 does not consider them, so we keep that behavior.
In addition, we had to improve AArch64's code to evaluate the branch target of a MCInst to handle calls where the destination is not the first operand (which it often is not).
Differential Revision: https://reviews.llvm.org/D48836
llvm-svn: 337007
A TableGen instruction record usually contains a DAG pattern that will
describe the SelectionDAG operation that can be implemented by this
instruction. However, there will be cases where several different DAG
patterns can all be implemented by the same instruction. The way to
represent this today is to write additional patterns in the Pattern
(or usually Pat) class that map those extra DAG patterns to the
instruction. This usually also works fine.
However, I've noticed cases where the current setup seems to require
quite a bit of extra (and duplicated) text in the target .td files.
For example, in the SystemZ back-end, there are quite a number of
instructions that can implement an "add-with-overflow" operation.
The same instructions also need to be used to implement just plain
addition (simply ignoring the extra overflow output). The current
solution requires creating extra Pat pattern for every instruction,
duplicating the information about which particular add operands
map best to which particular instruction.
This patch enhances TableGen to support a new PatFrags class, which
can be used to encapsulate multiple alternative patterns that may
all match to the same instruction. It operates the same way as the
existing PatFrag class, except that it accepts a list of DAG patterns
to match instead of just a single one. As an example, we can now define
a PatFrags to match either an "add-with-overflow" or a regular add
operation:
def z_sadd : PatFrags<(ops node:$src1, node:$src2),
[(z_saddo node:$src1, node:$src2),
(add node:$src1, node:$src2)]>;
and then use this in the add instruction pattern:
defm AR : BinaryRRAndK<"ar", 0x1A, 0xB9F8, z_sadd, GR32, GR32>;
These SystemZ target changes are implemented here as well.
Note that PatFrag is now defined as a subclass of PatFrags, which
means that some users of internals of PatFrag need to be updated.
(E.g. instead of using PatFrag.Fragment you now need to use
!head(PatFrag.Fragments).)
The implementation is based on the following main ideas:
- InlinePatternFragments may now replace each original pattern
with several result patterns, not just one.
- parseInstructionPattern delays calling InlinePatternFragments
and InferAllTypes. Instead, it extracts a single DAG match
pattern from the main instruction pattern.
- Processing of the DAG match pattern part of the main instruction
pattern now shares most code with processing match patterns from
the Pattern class.
- Direct use of main instruction patterns in InferFromPattern and
EmitResultInstructionAsOperand is removed; everything now operates
solely on DAG match patterns.
Reviewed by: hfinkel
Differential Revision: https://reviews.llvm.org/D48545
llvm-svn: 336999
Spectre variant #1 for x86.
There is a lengthy, detailed RFC thread on llvm-dev which discusses the
high level issues. High level discussion is probably best there.
I've split the design document out of this patch and will land it
separately once I update it to reflect the latest edits and updates to
the Google doc used in the RFC thread.
This patch is really just an initial step. It isn't quite ready for
prime time and is only exposed via debugging flags. It has two major
limitations currently:
1) It only supports x86-64, and only certain ABIs. Many assumptions are
currently hard-coded and need to be factored out of the code here.
2) It doesn't include any options for more fine-grained control, either
of which control flow edges are significant or which loads are
important to be hardened.
3) The code is still quite rough and the testing lighter than I'd like.
However, this is enough for people to begin using. I have had numerous
requests from people to be able to experiment with this patch to
understand the trade-offs it presents and how to use it. We would also
like to encourage work to similar effect in other toolchains.
The ARM folks are actively developing a system based on this for
AArch64. We hope to merge this with their efforts when both are far
enough along. But we also don't want to block making this available on
that effort.
Many thanks to the *numerous* people who helped along the way here. For
this patch in particular, both Eric and Craig did a ton of review to
even have confidence in it as an early, rough cut at this functionality.
Differential Revision: https://reviews.llvm.org/D44824
llvm-svn: 336990
flow patterns including forks, merges, and even cyles.
This tries to cover a reasonably comprehensive set of patterns that
still don't require PHIs or PHI placement. The coverage was inspired by
the amazing variety of patterns produced when copy EFLAGS and restoring
it to implement Speculative Load Hardening. Without this patch, we
simply cannot make such complex and invasive changes to x86 instruction
sequences due to EFLAGS.
I've added "just" one test, but this test covers many different
complexities and corner cases of this approach. It is actually more
comprehensive, as far as I can tell, than anything that I have
encountered in the wild on SLH.
Because the test is so complex, I've tried to give somewhat thorough
comments and an ASCII-art diagram of the control flows to make it a bit
easier to read and maintain long-term.
Differential Revision: https://reviews.llvm.org/D49220
llvm-svn: 336985
This patch adds support for the following unpack instructions:
- PUNPKLO, PUNPKHI Unpack elements from low/high half and
place into elements of twice their size.
e.g. punpklo p0.h, p0.b
- UUNPKLO, UUNPKHI Unpack elements from low/high half and
SUNPKLO, SUNPKHI place into elements of twice their size
after zero- or sign-extending the values.
e.g. uunpklo z0.h, z0.b
llvm-svn: 336982
Previously we iseled to blend, commuted to another blend, and then commuted back to movss/movsd or blend depending on optsize. Now we do it directly.
llvm-svn: 336976
This is not an optimization we should be doing in isel. This is more suitable for a DAG combine.
My main concern is a future time when we support more FPENV. Changing a packed op to a scalar op could cause us to miss some exceptions that should have occured if we had done a packed op. A DAG combine would be better able to manage this.
llvm-svn: 336971
This re-applies r336929 with a fix to accomodate for the Mips target
scheduling multiple SelectionDAG instances into the pass pipeline.
PrologEpilogInserter and StackColoring depend on the StackProtector analysis
being alive from the point it is run until PEI, which requires that they are all
scheduled in the same FunctionPassManager. Inserting a (machine) ModulePass
between StackProtector and PEI results in these passes being in separate
FunctionPassManagers and the StackProtector is not available for PEI.
PEI and StackColoring don't use much information from the StackProtector pass,
so transfering the required information to MachineFrameInfo is cleaner than
keeping the StackProtector pass around. This commit moves the SSP layout
information to MFI instead of keeping it in the pass.
This patch set (D37580, D37581, D37582, D37583, D37584, D37585, D37586, D37587)
is a first draft of the pagerando implementation described in
http://lists.llvm.org/pipermail/llvm-dev/2017-June/113794.html.
Patch by Stephen Crane <sjc@immunant.com>
Differential Revision: https://reviews.llvm.org/D49256
llvm-svn: 336964
We were accidentally connecting it to result 0 instead of result 1. This was caught by the machine verifier that noticed the flags were dead, but we were using them somehow. I'm still not clear what actually happened downstream.
llvm-svn: 336925
canWidenShuffleElements can do a better job if given a mask with ZeroableElements info. Apparently, ZeroableElements was being only used to identify AllZero candidates, but possibly we could plug it into more shuffle matchers.
Original Patch by Zvi Rackover @zvi
Differential Revision: https://reviews.llvm.org/D42044
llvm-svn: 336903
Noticed while updating D42044, lowerV2X128VectorShuffle can improve the shuffle mask with the zeroable data to create a target shuffle mask to recognise more 'zero upper 128' patterns.
NOTE: lowerV4X128VectorShuffle could benefit as well but the code needs refactoring first to discriminate between SM_SentinelUndef and SM_SentinelZero for negative shuffle indices.
Differential Revision: https://reviews.llvm.org/D49092
llvm-svn: 336900
Mark standard encoded instructions and pseudo "standard encoded"
as not being in MIPS16e by default.
Patch by Simon Dardis.
Differential revision: https://reviews.llvm.org/D48379
llvm-svn: 336893
We now use llvm.fma.f32/f64 or llvm.x86.fmadd.f32/f64 intrinsics that use scalar types rather than vector types. So we don't these special ISD nodes that operate on the lowest element of a vector.
llvm-svn: 336883
there for a long time.
The boolean tracking whether we saw a kill of the flags was supposed to
be per-block we are scanning and instead was outside that loop and never
cleared. It requires a quite contrived test case to hit this as you have
to have multiple levels of successors and interleave them with kills.
I've included such a test case here.
This is another bug found testing SLH and extracted to its own focused
patch.
llvm-svn: 336876
multiple successors where some of the uses end up killing the EFLAGS
register.
There was a bug where rather than skipping to the next basic block
queued up with uses once we saw a kill, we stopped processing the blocks
entirely. =/
Test case produces completely nonsensical code w/o this tiny fix.
This was found testing Speculative Load Hardening and split out of that
work.
Differential Revision: https://reviews.llvm.org/D49211
llvm-svn: 336874
This converts them to what clang is now using for codegen. Unfortunately, there seem to be a few kinks to work out still. I'll try to address with follow up patches.
llvm-svn: 336871
This is marginally helpful for removing redundant extensions, and the
code is easier to read, so it seems like an all-around win. In the new
test i8-phi-ext.ll, we used to emit an AssertSext i8; now we emit an
AssertZext i2, which allows the extension of the return value to be
eliminated.
Differential Revision: https://reviews.llvm.org/D49004
llvm-svn: 336868
SITargetLowering queries SIInstrInfo in its constructor, so SIInstrInfo
must be initialized first. This fixes msan buildbot failures and was
introduced by r336851.
llvm-svn: 336861
Summary:
This is a follow-up to r335942.
- Merge SISubtarget into AMDGPUSubtarget and rename to GCNSubtarget
- Rename AMDGPUCommonSubtarget to AMDGPUSubtarget
- Merge R600Subtarget::Generation and GCNSubtarget::Generation into
AMDGPUSubtarget::Generation.
Reviewers: arsenm, jvesely
Subscribers: kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D49037
llvm-svn: 336851
We can instead block the load folding isProfitableToFold. Then isel will emit a register->register move for the zeroing part and a separate load. The PostProcessISelDAG should be able to remove the register->register move.
This saves us patterns and fixes the fact that we only had unaligned load patterns. The test changes show places where we should have been using an aligned load.
llvm-svn: 336828
Before revision 336728, the "mayLoad" flag for instruction (V)MOVLPSrm was
inferred directly from the "default" pattern associated with the instruction
definition.
r336728 removed special node X86Movlps, and all the patterns associated to it.
Now instruction (V)MOVLPSrm doesn't have a pattern associated to it, and the
'mayLoad/hasSideEffects' flags are left unset.
When the instruction info is emitted by tablegen, method
CodeGenDAGPatterns::InferInstructionFlags() sees that (V)MOVLPSrm doesn't have a
pattern, and flags are undefined. So, it conservatively sets the
"hasSideEffects" flag for it.
As a consequence, we were losing the 'mayLoad' flag, and we were gaining a
'hasSideEffect' flag in its place.
This patch fixes the issue (originally reported by Michael Holmen).
The mca tests show the differences in the instruction info flags. Instructions
that were affected by this problem were: MOVLPSrm/VMOVLPSrm/VMOVLPSZ128rm.
Differential Revision: https://reviews.llvm.org/D49182
llvm-svn: 336818
This mostly brings the P5600 scheduler model to a mostly complete
status. There are a number of instructions which trigger the
`error:'MipsP5600Model' lacks information for` error. These are certain
codegen only instructions relating to MIPS64 which can be addressed by
using the correct predicates for them. That will be done in a full-up
patch.
Patch by Simon Dardis.
Differential revision: https://reviews.llvm.org/D45245
llvm-svn: 336802
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 compact instruction shuffles active elements of vector
into lowest numbered elements and sets remaining elements
to zero.
e.g.
compact z0.s, p0, z1.s
llvm-svn: 336789
The LASTB and LASTA instructions extract the last active element,
or element after the last active, from the source vector.
The added variants are:
Scalar:
last(a|b) w0, p0, z0.b
last(a|b) w0, p0, z0.h
last(a|b) w0, p0, z0.s
last(a|b) x0, p0, z0.d
SIMD & FP Scalar:
last(a|b) b0, p0, z0.b
last(a|b) h0, p0, z0.h
last(a|b) s0, p0, z0.s
last(a|b) d0, p0, z0.d
The CLASTB and CLASTA conditionally extract the last or element after
the last active element from the source vector.
The added variants are:
Scalar:
clast(a|b) w0, p0, w0, z0.b
clast(a|b) w0, p0, w0, z0.h
clast(a|b) w0, p0, w0, z0.s
clast(a|b) x0, p0, x0, z0.d
SIMD & FP Scalar:
clast(a|b) b0, p0, b0, z0.b
clast(a|b) h0, p0, h0, z0.h
clast(a|b) s0, p0, s0, z0.s
clast(a|b) d0, p0, d0, z0.d
Vector:
clast(a|b) z0.b, p0, z0.b, z1.b
clast(a|b) z0.h, p0, z0.h, z1.h
clast(a|b) z0.s, p0, z0.s, z1.s
clast(a|b) z0.d, p0, z0.d, z1.d
Please refer to the architecture specification for more details on
the semantics of the added instructions.
llvm-svn: 336783
Summary:
These changes cover the PR#31399.
Now the ffs(x) function is lowered to (x != 0) ? llvm.cttz(x) + 1 : 0
and it corresponds to the following llvm code:
%cnt = tail call i32 @llvm.cttz.i32(i32 %v, i1 true)
%tobool = icmp eq i32 %v, 0
%.op = add nuw nsw i32 %cnt, 1
%add = select i1 %tobool, i32 0, i32 %.op
and x86 asm code:
bsfl %edi, %ecx
addl $1, %ecx
testl %edi, %edi
movl $0, %eax
cmovnel %ecx, %eax
In this case the 'test' instruction can't be eliminated because
the 'add' instruction modifies the EFLAGS, namely, ZF flag
that is set by the 'bsf' instruction when 'x' is zero.
We now produce the following code:
bsfl %edi, %ecx
movl $-1, %eax
cmovnel %ecx, %eax
addl $1, %eax
Patch by Ivan Kulagin
Reviewers: davide, craig.topper, spatel, RKSimon
Reviewed By: craig.topper
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D48765
llvm-svn: 336768
These patterns looked for a MOVSS/SD followed by a scalar_to_vector. Or a scalar_to_vector followed by a load.
In both cases we emitted a MOVSS/SD for the MOVSS/SD part, a REG_CLASS for the scalar_to_vector, and a MOVSS/SD for the load.
But we have patterns that do each of those 3 things individually so there's no reason to build large patterns.
Most of the test changes are just reorderings. The one test that had a meaningful change is pr30430.ll and it appears to be a regression. But its doing -O0 so I think it missed a lot of opportunities and was just getting lucky before.
llvm-svn: 336762
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
Some added 20 and some added 15. Its unclear when to use which value and whether they are required at all.
This patch removes them all. If we start finding real world issues we may need to add them back with proper tests.
llvm-svn: 336735
Isel currently emits movss/movsd a lot of the time and an accidental double commute turns it into a blend.
Ideally we'd select blend directly in isel under optspeed and not rely on the double commute to create blend.
llvm-svn: 336731
These ISD nodes try to select the MOVLPS and MOVLPD instructions which are special load only instructions. They load data and merge it into the lower 64-bits of an XMM register. They are logically equivalent to our MOVSD node plus a load.
There was only one place in X86ISelLowering that used MOVLPD and no places that selected MOVLPS. The one place that selected MOVLPD had to choose between it and MOVSD based on whether there was a load. But lowering is too early to tell if the load can really be folded. So in isel we have patterns that use MOVSD for MOVLPD if we can't find a load.
We also had patterns that select the MOVLPD instruction for a MOVSD if we can find a load, but didn't choose the MOVLPD ISD opcode for some reason.
So it seems better to just standardize on MOVSD ISD opcode and manage MOVSD vs MOVLPD instruction with isel patterns.
llvm-svn: 336728
I believe isProfitableToFold will stop the load folding that this was intended to overcome.
Given an (xor load, -1), isProfitableToFold will see that the immediate can be folded with the xor using a one byte immediate since it can be sign extended. It doesn't know about NOT, but the one byte immediate check is enough to stop the fold.
llvm-svn: 336712
This patch adds support for the following instructions:
CLS (Count Leading Sign bits)
CLZ (Count Leading Zeros)
CNT (Count non-zero bits)
CNOT (Logically invert boolean condition in vector)
NOT (Bitwise invert vector)
FABS (Floating-point absolute value)
FNEG (Floating-point negate)
All operations are predicated and unary, e.g.
clz z0.s, p0/m, z1.s
- CLS, CLZ, CNT, CNOT and NOT have variants for 8, 16, 32
and 64 bit elements.
- FABS and FNEG have variants for 16, 32 and 64 bit elements.
llvm-svn: 336677
An explicit untied use is not sufficient to maintain liveness of a
register redefined in a predicated instruction. For example
%1 = COPY %0
...
%1 = A2_paddif %2, %1, 1
could become
$r1 = COPY $r0
...
$r1 = A2_paddif $p0, $r1, 1
and later
$r1 = COPY $r0 ;; this is not really dead!
...
$r1 = A2_paddif $p0, $r0, 1
llvm-svn: 336662
Now that rL336250 has landed, we should prefer 2 immediate shifts + a shuffle blend over performing a multiply. Despite the increase in instructions, this is quicker (especially for slow v4i32 multiplies), avoid loads and constant pool usage. It does mean however that we increase register pressure. The code size will go up a little but by less than what we save on the constant pool data.
This patch also adds support for v16i16 to the BLEND(SHIFT(v,c1),SHIFT(v,c2)) combine, and also prevents blending on pre-SSE41 shifts if it would introduce extra blend masks/constant pool usage.
Differential Revision: https://reviews.llvm.org/D48936
llvm-svn: 336642
We're missing the EVEX equivalents of these patterns and seem to get along fine.
I think we end up with X86vzload for the obvious IR cases that would produce this DAG.
llvm-svn: 336638
Summary:
This adds support for binary atomic read-modify-write instructions:
add, sub, and, or, xor, and xchg.
This does not yet support translations of some of LLVM IR atomicrmw
instructions (nand, max, min, umax, and umin) that do not have a direct
counterpart in wasm instructions.
Reviewers: dschuff
Subscribers: sbc100, jgravelle-google, sunfish, llvm-commits
Differential Revision: https://reviews.llvm.org/D49088
llvm-svn: 336615
Added __float128 support for a number of rounding operations:
trunc
rint
nearbyint
round
floor
ceil
Differential Revision: https://reviews.llvm.org/D48415
llvm-svn: 336601
Summary:
- Changed variable/function names to be more consistent
- Improved comments in test files
- Added more tests
- Fixed a few typos
- Misc. cosmetic changes
Reviewers: dschuff
Subscribers: sbc100, jgravelle-google, sunfish, llvm-commits
Differential Revision: https://reviews.llvm.org/D49087
llvm-svn: 336598
Build error on Android; reported by and fix provided by (thanks) by Mauro Rossi <issor.oruam@gmail.com>
Fixes the following building error:
external/llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cpp:1903:61:
error: comparison of integers of different signs:
'typename iterator_traits<__wrap_iter<MachineBasicBlock **> >::difference_type'
(aka 'int') and 'unsigned int' [-Werror,-Wsign-compare]
BlockWaitcntProcessedSet.end(), &MBB) < Count)) {
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ^ ~~~~~
1 error generated.
Differential Revision: https://reviews.llvm.org/D49089
llvm-svn: 336588
These won't work for the forseeable future. These aren't allowed
from OpenCL, but IPO optimizations can make them appear.
Also directly set the attributes on functions, regardless
of the linkage rather than cloning functions like before.
llvm-svn: 336587
Summary:
This adds a reverse transform for the instcombine canonicalizations
that were added in D47980, D47981.
As discussed later, that was worse at least for the code size,
and potentially for the performance, too.
https://rise4fun.com/Alive/Zmpl
Reviewers: craig.topper, RKSimon, spatel
Reviewed By: spatel
Subscribers: reames, llvm-commits
Differential Revision: https://reviews.llvm.org/D48768
llvm-svn: 336585
These patterns mapped (v2f64 (X86vzmovl (v2f64 (scalar_to_vector FR64:$src)))) to a MOVSD and an zeroing XOR. But the complexity of a pattern for (v2f64 (X86vzmovl (v2f64))) that selects MOVQ is artificially and hides this MOVSD pattern.
Weirder still, the SSE version of the pattern was explicitly blocked on SSE41, but yet we had copied it to AVX and AVX512.
llvm-svn: 336556
This patch adds support for the following instructions:
CNTB CNTH - Determine the number of active elements implied by
CNTW CNTD the named predicate constant, multiplied by an
immediate, e.g.
cnth x0, vl8, #16
CNTP - Count active predicate elements, e.g.
cntp x0, p0, p1.b
counts the number of active elements in p1, predicated
by p0, and stores the result in x0.
llvm-svn: 336552
This patch completes support for shifts, which include:
- LSL - Logical Shift Left
- LSLR - Logical Shift Left, Reversed form
- LSR - Logical Shift Right
- LSRR - Logical Shift Right, Reversed form
- ASR - Arithmetic Shift Right
- ASRR - Arithmetic Shift Right, Reversed form
- ASRD - Arithmetic Shift Right for Divide
In the following variants:
- Predicated shift by immediate - ASR, LSL, LSR, ASRD
e.g.
asr z0.h, p0/m, z0.h, #1
(active lanes of z0 shifted by #1)
- Unpredicated shift by immediate - ASR, LSL*, LSR*
e.g.
asr z0.h, z1.h, #1
(all lanes of z1 shifted by #1, stored in z0)
- Predicated shift by vector - ASR, LSL*, LSR*
e.g.
asr z0.h, p0/m, z0.h, z1.h
(active lanes of z0 shifted by z1, stored in z0)
- Predicated shift by vector, reversed form - ASRR, LSLR, LSRR
e.g.
lslr z0.h, p0/m, z0.h, z1.h
(active lanes of z1 shifted by z0, stored in z0)
- Predicated shift left/right by wide vector - ASR, LSL, LSR
e.g.
lsl z0.h, p0/m, z0.h, z1.d
(active lanes of z0 shifted by wide elements of vector z1)
- Unpredicated shift left/right by wide vector - ASR, LSL, LSR
e.g.
lsl z0.h, z1.h, z2.d
(all lanes of z1 shifted by wide elements of z2, stored in z0)
*Variants added in previous patches.
llvm-svn: 336547
Support for SVE's TBL instruction for programmable table
lookup/permute using vector of element indices, e.g.
tbl z0.d, { z1.d }, z2.d
stores elements from z1, indexed by elements from z2, into z0.
llvm-svn: 336544
This patch adds support for:
UZP1 Concatenate even elements from two vectors
UZP2 Concatenate odd elements from two vectors
TRN1 Interleave even elements from two vectors
TRN2 Interleave odd elements from two vectors
With variants for both data and predicate vectors, e.g.
uzp1 z0.b, z1.b, z2.b
trn2 p0.s, p1.s, p2.s
llvm-svn: 336531
This replaces some asserts in lowerV2F64VectorShuffle with the similar asserts from lowerVIF64VectorShuffle which are more readable. The original asserts mentioned a blend, but there's no guarantee that it is a blend.
Also remove an if that the asserts prove is always true. Mask[0] is always less than 2 and Mask[1] is always at least 2. Therefore (Mask[0] >= 2) + (Mask[1] >= 2) == 1 must wlays be true.
llvm-svn: 336517
It only existed on SSE and AVX version. AVX512 version didn't have it.
I checked the generated table and this didn't seem necessary to creat a match preference.
llvm-svn: 336516
Summary:
{F6603964}
While there is still some discrepancies within that new group,
it is clearly separate from the other shifts.
And Agner's tables agree, these double shifts are clearly
different from the normal shifts/rotates.
I'm guessing `FeatureSlowSHLD` is related.
Indeed, a basic sched pair is *not* the /best/ match.
But keeping it in the WriteShift is /clearly/ not ideal either.
This can and likely will be fine-tuned later.
This is purely mechanical change, it does not change any numbers,
as the [lack of the change of] mca tests show.
Reviewers: craig.topper, RKSimon, andreadb
Reviewed By: craig.topper
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D49015
llvm-svn: 336515
Pre-AVX512 (which can perform a quick extend/shift/truncate), extending to 2 v8i16 for the PMULLW and then truncating is more performant than relying on the generic PBLENDVB vXi8 shift path and uses a similar amount of mask constant pool data.
Differential Revision: https://reviews.llvm.org/D48963
llvm-svn: 336513
Summary:
Motivation: {F6597954}
This only does the mechanical splitting, does not actually change
any numbers, as the tests added in previous revision show.
Reviewers: craig.topper, RKSimon, courbet
Reviewed By: craig.topper
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D48998
llvm-svn: 336511
Craig Topper