We can use the same input for both operands to get a free compare with zero.
We already use this trick in a couple places where we explicitly create PTESTM with the same input twice. This generalizes it.
I'm hoping to remove the ISD opcodes and move this to isel patterns like we do for scalar cmp/test.
llvm-svn: 323605
Legalization is still biased to turn LT compares in to GT by swapping operands to avoid needing extra isel patterns to commute.
I'm hoping to remove TESTM/TESTNM next and this should simplify that by making EQ/NE more similar.
llvm-svn: 323604
If broadcasting from another shuffle, attempt to simplify it.
We can probably generalize this a lot more (embedding in combineX86ShufflesRecursively), but BROADCAST is one of the more troublesome as it accepts inputs of different sizes to the result.
llvm-svn: 323602
Previously we had to materialize all 1s in a register using vpternlog or pcmpeq and then xor with that. By using vpternlog directly we can do it in one operation.
This is implemented using isel patterns, but we should maybe consider creating a generalized vpternlog combiner.
llvm-svn: 323572
A correctly aligned address may happen to be separated into a variable
part and a constant part, where the constant part does not match the
alignment needed in a load/store that uses this address. Such a constant
cannot be used as an immediate offset in an indexed instruction.
When lowering a global address, make sure that if there is an offset
folded into the global, the offset is valid for all uses in load/store
instructions.
llvm-svn: 323562
X86ISelLowering.cpp:34130:5: error: return type 'llvm::SDValue' must
match previous return type 'const llvm::SDValue' when lambda expression
has unspecified explicit return type
llvm-svn: 323557
Previously some targets printed their own message at the start of Select to indicate what they were selecting. For the targets that didn't, it means there was no print of the root node before any custom handling in the target executed. So if the target did something custom and never called SelectNodeCommon, no print would be made. For the targets that did print a message in Select, if they didn't custom handle a node SelectNodeCommon would reprint the root node before walking the isel table.
It seems better to just print the message before the call to Select so all targets behave the same. And then remove the root node printing from SelectNodeCommon and just leave a message that says we're starting the table search.
There were also some oddities in blank line behavior. Usually due to a \n after a call to SelectionDAGNode::dump which already inserted a new line.
llvm-svn: 323551
We currently coalesce v4i32 extracts from all 4 elements to 2 v2i64 extracts + shifts/sign-extends.
This seems to have been added back in the days when we tended to spill vectors and reload scalars, or ended up with repeated shuffles moving everything down to 0'th index. I don't think either of these are likely these days as we have better EXTRACT_VECTOR_ELT and VECTOR_SHUFFLE handling, and the existing code tends to make it very difficult for various vector and load combines.
Differential Revision: https://reviews.llvm.org/D42308
llvm-svn: 323541
- using qualified pointer addrspace in intrinsics class to avoid .f32 mangling
- changed too common atomic mangling to ds
- added missing intrinsics to AMDGPUTTIImpl::getTgtMemIntrinsic
Reviewed by: b-sumner
Differential Revision: https://reviews.llvm.org/D42383
llvm-svn: 323516
load instruction
The function `Thumb1InstrInfo::loadRegFromStackSlot` accepts only the `tGPR`
register class. The function serves to emit a `tLDRspi` instruction and
certainly any subset of the `tGPR` register class is a valid destination of the
load.
Differential revision: https://reviews.llvm.org/D42535
llvm-svn: 323514
This is the groundwork for Armv8.2-A FP16 code generation .
Clang passes and returns _Float16 values as floats, together with the required
bitconverts and truncs etc. to implement correct AAPCS behaviour, see D42318.
We will implement half-precision argument passing/returning lowering in the ARM
backend soon, but for now this means that this:
_Float16 sub(_Float16 a, _Float16 b) {
return a + b;
}
gets lowered to this:
define float @sub(float %a.coerce, float %b.coerce) {
entry:
%0 = bitcast float %a.coerce to i32
%tmp.0.extract.trunc = trunc i32 %0 to i16
%1 = bitcast i16 %tmp.0.extract.trunc to half
<SNIP>
%add = fadd half %1, %3
<SNIP>
}
When FullFP16 is *not* supported, we don't make f16 a legal type, and we get
legalization for "free", i.e. nothing changes and everything works as before.
And also f16 argument passing/returning is handled.
When FullFP16 is supported, we do make f16 a legal type, and have 2 places that
we need to patch up: f16 argument passing and returning, which involves minor
tweaks to avoid unnecessary code generation for some bitcasts.
As a "demonstrator" that this works for the different FP16, FullFP16, softfp
modes, etc., I've added match rules to the VSUB instruction description showing
that we can codegen this instruction from IR, but more importantly, also to
some conversion instructions. These conversions were causing issue before in
the FP16 and FullFP16 cases.
I've also added match rules to the VLDRH and VSTRH desriptions, so that we can
actually compile the entire half-precision sub code example above. This showed
that these loads and stores had the wrong addressing mode specified: AddrMode5
instead of AddrMode5FP16, which turned out not be implemented at all, so that
has also been added.
This is the minimal patch that shows all the different moving parts. In patch
2/3 I will add some efficient lowering of bitcasts, and in 2/3 I will add the
remaining Armv8.2-A FP16 instruction descriptions.
Thanks to Sam Parker and Oliver Stannard for their help and reviews!
Differential Revision: https://reviews.llvm.org/D38315
llvm-svn: 323512
Type legalization would prevent any i64 operands to the build_vector from existing before we get here. The coverage bots show this code as uncovered.
llvm-svn: 323506
The original autoupgrade for kunpck intrinsics used a bitcasted scalar shift, or, and. This combine would turn this into a concat_vectors. Now the kunpck intrinsics are autoupgraded to a vector shuffle that will become a concat_vectors.
llvm-svn: 323504
This listed all legal 128-bit integer types individually, but since we already know we have a legal type and its integer, we can just check is128BitVector.
llvm-svn: 323502
When pass creates a MOV instruction for
lea (%base,%index,1), %dst => mov %base,%dst; add %index,%dst
modification it should clean the killed flag for base
if base is equal to index.
Otherwise verifier complains about usage of killed register in add instruction.
Reviewers: lsaba, zvi, zansari, aaboud
Reviewed By: lsaba
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D42522
llvm-svn: 323497
This patch enables aggressive FMA by default on T99, and provides a -mllvm
option to enable the same on other AArch64 micro-arch's (-mllvm
-aarch64-enable-aggressive-fma).
Test case demonstrating the effects on T99 is included.
Patch by: steleman (Stefan Teleman)
Differential Revision: https://reviews.llvm.org/D40696
llvm-svn: 323474
The asm parser puts the lock prefix in the MCInst flags so we need to check that in addition to TSFlags. This matches what the ATT printer does.
llvm-svn: 323469
The regular expressions and the imul names caused some instructions to be matched by multiple regexs creating unpredictable results.
This changes them all to use explicit instrs instead.
While doing this I also found that some instructions in Skylake were missing load latency so I fixed that too.
llvm-svn: 323406
The IMUL instruction names mixed with the prefix matching of the instregex lead to some strange matches. The worst being that several memory instructions are using the register form latency.
I don't know what the right answer is, so I've left TODOs and will try to work with the AMD folks to get this cleaned up.
llvm-svn: 323405
MMX instrutions all start with MMX_ so the 64 isn't needed for disambigutation.
SSE/AVX1 instructions are assumed 128-bit so we don't need to say 128.
AVX2 instructions should use a Y to indicate 256-bits.
llvm-svn: 323402
These were treated as optional suffixes, but the regular expressions are already prefix matches so this is unnecessary. It breaks the binary search optimization in tablegen due to the top level question mark.
llvm-svn: 323401
The code in EmitFunctionEntryCode needs to know the maximum stack
alignment, but it runs very early in the selection process (before
lowering). The final stack alignment may change during lowering, so
the code needs to be moved to where the alignment is known.
llvm-svn: 323374
The tablegen imported patterns for sext(load(a)) don't check for single uses
of the load or delete the original after matching. As a result two loads are
left in the generated code. This particular issue will be fixed by adding
support for a G_SEXTLOAD opcode in future.
There are however other potential issues around this that wouldn't be fixed by
a G_SEXTLOAD, so until we have a proper solution we don't try to handle volatile
loads at all in the AArch64 selector.
Fixes/works around PR36018.
llvm-svn: 323371
As discussed in D41484, PMADDWD for 'zero extended' vXi32 is nearly always a better option than PMULLD:
On SNB it will result in code that isn't any faster, but not any slower so we may as well keep it.
On KNL it only has half the throughput, so I've disabled it on there - ideally there'd be a better way than this.
Differential Revision: https://reviews.llvm.org/D42258
llvm-svn: 323367
Summary:
Move reserveRegisterTuples into AMDGPURegisterInfo and use it in
R600RegisterInfo::getReservedRegs and
R600InstrInfo::reserveIndirectRegisters to ensure that all super
registers of reserved registers are also marked as reserved.
Before this change, under certain circumstances, the registers %t1_x and
%t1_xyzw would be marked as reserved, but %t1_xy and %t1_xyz would not
be, leading to the register allocator sometimes assigning a register to
%t1_xy, which is invalid since %t1_x is reserved.
Reviewers: arsenm, tstellar, MatzeB, qcolombet
Subscribers: kzhuravl, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D42448
llvm-svn: 323356
Summary: For long shifts, the inlined version takes about 20 instructions on Thumb1. To avoid the code bloat, expand to __aeabi_ calls if target is Thumb1.
Reviewers: samparker
Reviewed By: samparker
Subscribers: samparker, aemerson, javed.absar, kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D42401
llvm-svn: 323354
The regexs are treated as a prefix match already so the checking for optional text at the end provides no value. Instead it prevents the binary search optimization in tablegen from kicking in due to the top level question mark.
llvm-svn: 323351
Summary:
Loads/stores of some NEON vector types are promoted to other vector
types with different lane sizes but same vector size. This is not a
problem in little-endian but, when in big-endian, it requires
additional byte reversals required to preserve the lane ordering
while keeping the right endianness of the data inside each lane.
For example:
%1 = load <4 x half>, <4 x half>* %p
results in the following assembly:
ld1 { v0.2s }, [x1]
rev32 v0.4h, v0.4h
This patch changes the promotion of these loads/stores so that the
actual vector load/store (LD1/ST1) takes care of the endianness
correctly and there is no need for further byte reversals. The
previous code now results in the following assembly:
ld1 { v0.4h }, [x1]
Reviewers: olista01, SjoerdMeijer, efriedma
Reviewed By: efriedma
Subscribers: aemerson, rengolin, javed.absar, llvm-commits, kristof.beyls
Differential Revision: https://reviews.llvm.org/D42235
llvm-svn: 323325
This matches what MSVC does for alloca() function calls on ARM.
Even if MSVC doesn't support VLAs at the language level, it does
support the alloca function.
On the clang level, both the _alloca() (when emulating MSVC, which is
what the alloca() function expands to) and __builtin_alloca() builtin
functions, and VLAs, map to the same LLVM IR "alloca" function - so
within LLVM they're not distinguishable from each other.
Differential Revision: https://reviews.llvm.org/D42292
llvm-svn: 323308
I think these instructions used to be named differently and the regular expression reflected that. I guess we must have correct itinerary information that made this not matter for the scheduler test?
llvm-svn: 323305
There are a couple tricky things with this patch.
I had to add an override of isVectorLoadExtDesirable to stop DAG combine from combining sign_extend with loads after legalization since we legalize sextload using a load+sign_extend. Overriding this hook actually prevents a lot sextloads from being created in the first place.
I also had to add isel patterns because DAG combine blindly combines sign_extend+truncate to a smaller sign_extend which defeats what legalization was trying to do.
Differential Revision: https://reviews.llvm.org/D42407
llvm-svn: 323301
This is already a simplification, and should help with avoiding a plt
reference when calling an intrinsic with -fno-plt.
With this change we return false for null GVs, so the caller only
needs to check the new metadata to decide if it should use foo@plt or
*foo@got.
llvm-svn: 323297
Remove FeatureSlowMisaligned128Store from cyclone flags.
This flag causes splitting of 16 byte wide stores into 2 stored of 8
bytes. This was useful on older apple CPUs which were slow for 16byte
stores that were not aligned on 16byte. As the compiler often cannot
predict the actual alignment, the splitting was choosen.
This has been a topic for a lot of debate as the splitting also
decreases performance for some benchmarks. Measuring the effects on
newer apple chips (rdar://35525421) shows that it harms more cases than
it helps. So it is time to retire this workaround.
llvm-svn: 323289
Summary:
Fix an issue that's similar to what D41411 fixed:
float(__int128(float_var)) shouldn't be optimized to xscvdpsxds +
xscvsxdsp, as they mean (float)(int64_t)float_var.
Reviewers: jtony, hfinkel, echristo
Subscribers: sanjoy, nemanjai, hiraditya, llvm-commits, kbarton
Differential Revision: https://reviews.llvm.org/D42400
llvm-svn: 323270
I don't know if the unused classes were intended to be used and that the VEX version is really different than the legacy SSE version. Agner's tables don't show any differences. I'm just cleaning up assuming the current behavior is correct.
llvm-svn: 323263
All other intrinsic instructions put the _Int on the end. This make these instructions consistent and gets the prefix instregexs in the scheduler models to pick them up.
llvm-svn: 323261
Minor refactor to make it possible for LowerBUILD_VECTORAsVariablePermute to be used with a wider variety of shuffles op and types.
I'd have liked to add v4i32/v4f32 support as well but we don't see v4i32 index extractions at the moment (which is why I created D42308)
After this I intend to begin adding scaling support for PSHUFB (v8i16, v4i32, v2i64)) and VPERMPS (v4f64, v4i64).
Differential Revision: https://reviews.llvm.org/D42431
llvm-svn: 323260
In addition to that, make sure that there are no boolean vector types that
are associated with multiple register classes. Specifically, remove v32i1
and v64i1 from integer register classes. These types will correspond to
results of vector comparisons, and as such should belong to the vector
predicate class. Having them in scalar registers as well makes legalization
ambiguous.
llvm-svn: 323229
The grow_memory and current_memory instructions are expected to be
officially renamed to mem.grow and mem.size. Introduce new intrinsics
with the new names. These new names aren't yet official, so for now,
use them at your own risk.
Also, take this opportunity to add arguments for the currently unused
immediate field in those instructions.
llvm-svn: 323222
The existing code was already doing something very similar to subvector insertion so this allows us to remove the nearly duplicate code.
This patch is a little larger than it should be due to differences between the DQI handling between the two today.
llvm-svn: 323212
Some nodes produce multiple values so when obtaining the type of an ISD::OR we
need to make sure we ask for the correct one. Hopefully that's all of them.
llvm-svn: 323205
Summary:
For the most part its better to keep v32i1 as a mask type of a narrower width than trying to promote it to a ymm register.
I had to add some overrides to the methods that get the types for the calling convention so that we still use v32i8 for argument/return purposes.
There are still some regressions in here. I definitely saw some around shuffles. I think we probably should move vXi1 shuffle from lowering to a DAG combine where I think the extend and truncate we have to emit would be better combined.
I think we also need a DAG combine to remove trunc from (extract_vector_elt (trunc))
Overall this removes something like 13000 CHECK lines from lit tests.
Reviewers: zvi, RKSimon, delena, spatel
Reviewed By: RKSimon
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D42031
llvm-svn: 323201
I'm not sure there's any way to generate these folding cases especially the movzx ones since even the register form is never emitted by codegen.
I'm just adding them to remove the difference with the autogenerated version of the folding table.
llvm-svn: 323200
As detailed in rL317463, PSHUFB (like most variable shuffle instructions) uses Op[0] for the source vector and Op[1] for the shuffle index vector, VPERMV works in reverse which is probably where the confusion comes from.
Differential Revision: https://reviews.llvm.org/D42380
llvm-svn: 323190
- Alter abs for micromips to have both AFGR64 and FGR64
variants, same as sqrt
- Remove sqrt and abs from MicroMips32r6InstrInfo.td,
use micromips FGR64 variants
- Restrict non-micromips abs/sqrt with NotInMicroMips
predicate
Differential revision: https://reviews.llvm.org/D41439
llvm-svn: 323184
Summary:
If we can match as a zero extend there's no need to flip the order to get an encoding benefit. As movzx is 3 bytes with independent source/dest registers. The shortest 'and' we could make is also 3 bytes unless we get lucky in the register allocator and its on AL/AX/EAX which have a 2 byte encoding.
This patch was more impressive before r322957 went in. It removed some of the same Ands that got deleted by that patch.
Reviewers: spatel, RKSimon
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D42313
llvm-svn: 323175
Some of the NOREX instructions are used in 32-bit mode making this printing confusing. It also doesn't provide a lot of value since you can see the h-register being used by the instruction.
llvm-svn: 323174
Add missing patterns for inserting v1i1 into a zero vector. Use insert_subvector to zero upper bits before inserting an element into a vXi1 vector. Replace kshift based isel pattern with insert_subvector based pattern now that code that caused the pattern has been fixed to emit insert_subvector.
llvm-svn: 323173
Summary:
First, we need to explain the core of the vulnerability. Note that this
is a very incomplete description, please see the Project Zero blog post
for details:
https://googleprojectzero.blogspot.com/2018/01/reading-privileged-memory-with-side.html
The basis for branch target injection is to direct speculative execution
of the processor to some "gadget" of executable code by poisoning the
prediction of indirect branches with the address of that gadget. The
gadget in turn contains an operation that provides a side channel for
reading data. Most commonly, this will look like a load of secret data
followed by a branch on the loaded value and then a load of some
predictable cache line. The attacker then uses timing of the processors
cache to determine which direction the branch took *in the speculative
execution*, and in turn what one bit of the loaded value was. Due to the
nature of these timing side channels and the branch predictor on Intel
processors, this allows an attacker to leak data only accessible to
a privileged domain (like the kernel) back into an unprivileged domain.
The goal is simple: avoid generating code which contains an indirect
branch that could have its prediction poisoned by an attacker. In many
cases, the compiler can simply use directed conditional branches and
a small search tree. LLVM already has support for lowering switches in
this way and the first step of this patch is to disable jump-table
lowering of switches and introduce a pass to rewrite explicit indirectbr
sequences into a switch over integers.
However, there is no fully general alternative to indirect calls. We
introduce a new construct we call a "retpoline" to implement indirect
calls in a non-speculatable way. It can be thought of loosely as
a trampoline for indirect calls which uses the RET instruction on x86.
Further, we arrange for a specific call->ret sequence which ensures the
processor predicts the return to go to a controlled, known location. The
retpoline then "smashes" the return address pushed onto the stack by the
call with the desired target of the original indirect call. The result
is a predicted return to the next instruction after a call (which can be
used to trap speculative execution within an infinite loop) and an
actual indirect branch to an arbitrary address.
On 64-bit x86 ABIs, this is especially easily done in the compiler by
using a guaranteed scratch register to pass the target into this device.
For 32-bit ABIs there isn't a guaranteed scratch register and so several
different retpoline variants are introduced to use a scratch register if
one is available in the calling convention and to otherwise use direct
stack push/pop sequences to pass the target address.
This "retpoline" mitigation is fully described in the following blog
post: https://support.google.com/faqs/answer/7625886
We also support a target feature that disables emission of the retpoline
thunk by the compiler to allow for custom thunks if users want them.
These are particularly useful in environments like kernels that
routinely do hot-patching on boot and want to hot-patch their thunk to
different code sequences. They can write this custom thunk and use
`-mretpoline-external-thunk` *in addition* to `-mretpoline`. In this
case, on x86-64 thu thunk names must be:
```
__llvm_external_retpoline_r11
```
or on 32-bit:
```
__llvm_external_retpoline_eax
__llvm_external_retpoline_ecx
__llvm_external_retpoline_edx
__llvm_external_retpoline_push
```
And the target of the retpoline is passed in the named register, or in
the case of the `push` suffix on the top of the stack via a `pushl`
instruction.
There is one other important source of indirect branches in x86 ELF
binaries: the PLT. These patches also include support for LLD to
generate PLT entries that perform a retpoline-style indirection.
The only other indirect branches remaining that we are aware of are from
precompiled runtimes (such as crt0.o and similar). The ones we have
found are not really attackable, and so we have not focused on them
here, but eventually these runtimes should also be replicated for
retpoline-ed configurations for completeness.
For kernels or other freestanding or fully static executables, the
compiler switch `-mretpoline` is sufficient to fully mitigate this
particular attack. For dynamic executables, you must compile *all*
libraries with `-mretpoline` and additionally link the dynamic
executable and all shared libraries with LLD and pass `-z retpolineplt`
(or use similar functionality from some other linker). We strongly
recommend also using `-z now` as non-lazy binding allows the
retpoline-mitigated PLT to be substantially smaller.
When manually apply similar transformations to `-mretpoline` to the
Linux kernel we observed very small performance hits to applications
running typical workloads, and relatively minor hits (approximately 2%)
even for extremely syscall-heavy applications. This is largely due to
the small number of indirect branches that occur in performance
sensitive paths of the kernel.
When using these patches on statically linked applications, especially
C++ applications, you should expect to see a much more dramatic
performance hit. For microbenchmarks that are switch, indirect-, or
virtual-call heavy we have seen overheads ranging from 10% to 50%.
However, real-world workloads exhibit substantially lower performance
impact. Notably, techniques such as PGO and ThinLTO dramatically reduce
the impact of hot indirect calls (by speculatively promoting them to
direct calls) and allow optimized search trees to be used to lower
switches. If you need to deploy these techniques in C++ applications, we
*strongly* recommend that you ensure all hot call targets are statically
linked (avoiding PLT indirection) and use both PGO and ThinLTO. Well
tuned servers using all of these techniques saw 5% - 10% overhead from
the use of retpoline.
We will add detailed documentation covering these components in
subsequent patches, but wanted to make the core functionality available
as soon as possible. Happy for more code review, but we'd really like to
get these patches landed and backported ASAP for obvious reasons. We're
planning to backport this to both 6.0 and 5.0 release streams and get
a 5.0 release with just this cherry picked ASAP for distros and vendors.
This patch is the work of a number of people over the past month: Eric, Reid,
Rui, and myself. I'm mailing it out as a single commit due to the time
sensitive nature of landing this and the need to backport it. Huge thanks to
everyone who helped out here, and everyone at Intel who helped out in
discussions about how to craft this. Also, credit goes to Paul Turner (at
Google, but not an LLVM contributor) for much of the underlying retpoline
design.
Reviewers: echristo, rnk, ruiu, craig.topper, DavidKreitzer
Subscribers: sanjoy, emaste, mcrosier, mgorny, mehdi_amini, hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D41723
llvm-svn: 323155
- Change inserted add ( V_ADD_{I|U}32_e32 ) to _e64 version ( V_ADD_{I|U}32_e64 ) so that the add uses a vreg for the carry; this prevents inserted v_add from killing VCC; the _e64 version doesn't accept a literal in its encoding, so we need to introduce a mov instr as well to get the imm into a register.
- Change pass name to "SI Load Store Optimizer"; this removes the '/', which complicates scripts.
Differential Revision: https://reviews.llvm.org/D42124
llvm-svn: 323153
As noted in another review, this loop is confusing. This commit cleans it up
somewhat.
Differential Revision: https://reviews.llvm.org/D42312
llvm-svn: 323136
Dsp and dspr2 require MIPS revision 2, while msa requires revision 5. Adding
warnings for cases when these flags are used with earlier revision.
Patch by Milos Stojanovic.
Differential Revision: https://reviews.llvm.org/D40490
llvm-svn: 323131
The change in r322988 caused a failure in the bootstrap build bot.
The problem was that directly gluing a BR_CCMASK node to a
compare-and-swap could lead to issues if other nodes were
chained in between. There is then no way to create a topological
sort that respects both the chain sequence and the glue property.
Fixed for now by rejecting the optimization in this case. As a
future enhancement, we may be able to handle additional cases
by swapping chain links around.
llvm-svn: 323129
Improves the code generation for v4f16 FCMP instructions when FullFP16 is not supported.
Generating FCTVL(s) rather than a longer series of FCVTs.
Differential Revision: https://reviews.llvm.org/D41772
llvm-svn: 323118
Primarily, this allows us to use the aggressive extraction mechanisms in combineExtractWithShuffle earlier and make use of UNDEF elements that may be lost during lowering.
Reapplied after rL322279 was reverted at rL322335 due to PR35918, underlying issue was fixed at rL322644.
llvm-svn: 323104
Summary:
This patch adds support for parsing/printing of named or unnamed
patterns that are used in SVE's PTRUE instruction, amongst others.
The pattern can be specified as a named pattern to initialize the predicate
vector or it can be specified as an immediate in the range 0-31.
Reviewers: fhahn, rengolin, evandro, mcrosier, t.p.northover
Reviewed By: fhahn
Subscribers: aemerson, javed.absar, tschuett, kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D41818
llvm-svn: 323098
1. ReachingDefsAnalysis - Allows to identify for each instruction what is the “closest” reaching def of a certain register. Used by BreakFalseDeps (for clearance calculation) and ExecutionDomainFix (for arbitrating conflicting domains).
2. ExecutionDomainFix - Changes the variant of the instructions in order to minimize domain crossings.
3. BreakFalseDeps - Breaks false dependencies.
4. LoopTraversal - Creatws a traversal order of the basic blocks that is optimal for loops (introduced in revision L293571). Both ExecutionDomainFix and ReachingDefsAnalysis use this to determine the order they will traverse the basic blocks.
This also included the following changes to ExcecutionDepsFix original logic:
1. BreakFalseDeps and ReachingDefsAnalysis logic no longer restricted by a register class.
2. ReachingDefsAnalysis tracks liveness of reg units instead of reg indices into a given reg class.
Additional changes in affected files:
1. X86 and ARM targets now inherit from ExecutionDomainFix instead of ExecutionDepsFix. BreakFalseDeps also was added to the passes they activate.
2. Comments and references to ExecutionDepsFix replaced with ExecutionDomainFix and BreakFalseDeps, as appropriate.
Additional refactoring changes will follow.
This commit is (almost) NFC.
The only functional change is that now BreakFalseDeps will break dependency for all register classes.
Since no additional instructions were added to the list of instructions that have false dependencies, there is no actual change yet.
In a future commit several instructions (and tests) will be added.
This is the first of multiple patches that fix bugzilla https://bugs.llvm.org/show_bug.cgi?id=33869
Most of the patches are intended at refactoring the existent code.
Additional relevant reviews:
https://reviews.llvm.org/D40331https://reviews.llvm.org/D40332https://reviews.llvm.org/D40333https://reviews.llvm.org/D40334
Differential Revision: https://reviews.llvm.org/D40330
Change-Id: Icaeb75e014eff96a8f721377783f9a3e6c679275
llvm-svn: 323087
Summary:
This patch adds an implementation of targetShrinkDemandedConstant that tries to keep shrinkdemandedbits from removing bits that would otherwise have been recognized as a movzx.
We still need a follow patch to stop moving ands across srl if the and could be represented as a movzx before the shift but not after. I think this should help with some of the cases that D42088 ended up removing during isel.
Reviewers: spatel, RKSimon
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D42265
llvm-svn: 323048
This change applies to places where we would turn 128/256-bit code into 512-bit in order to get a wider element type through sext/zext. Any 512-bit types that already existed in the IR/DAG will be left that way.
The width preference has no effect on codegen behavior when the target does not have AVX512 enabled. So AVX/AVX2 codegen cannot be limited via this mechanism yet.
If the preference is lower than 256 we may still use a 256 bit type to do the operation. Constraining to 128 bits makes it much more difficult to support some operations. For many of these cases we need to change element width while keeping element count constant which is easiest done by switching between 256 and 128 bit.
The preference is only obeyed when AVX512 and VLX are available. This means the preference is not obeyed for KNL, but is obeyed for SKX, Cannonlake, and Icelake. For KNL, the only way to do masked operation is on 512-bit registers so we would have to completely disable masking to obey the preference. We would also lose support for gather, scatter, ctlz, vXi64 multiplies, etc. This may change in the future, but this simplifies the initial implementation.
Differential Revision: https://reviews.llvm.org/D41895
llvm-svn: 323016
This will cause the vectorizers to do some limiting of the vector widths they create. This is not a strict limit. There are reasons I know of that the loop vectorizer will generate larger vectors for.
I've written this in such a way that the interface will only return a properly supported width(0/128/256/512) even if the attribute says something funny like 384 or 10.
This has been split from D41895 with the remainder in a follow up commit.
llvm-svn: 323015
On current machines we have load-on-condition instructions that can be
used to directly implement the SETCC semantics. If we have those, it is
always preferable to use them instead of generating the IPM sequence.
llvm-svn: 322989
In order to implement a test whether a compare-and-swap succeeded, the
SystemZ back-end currently emits a rather inefficient sequence of first
converting the CC result into an integer, and then testing that integer
against zero. This commit changes the back-end to simply directly test
the CC value set by the compare-and-swap instruction.
llvm-svn: 322988
The SystemZ back-end uses a sequence of IPM followed by arithmetic
operations to implement the SETCC primitive. This is currently done
early during SelectionDAG. This patch moves generating those sequences
to much later in SelectionDAG (during PreprocessISelDAG).
This doesn't change much in generated code by itself, but it allows
further enhancements that will be checked-in as follow-on commits.
llvm-svn: 322987
Fix a performance regression caused by r322737.
While trying to make it easier to replace compares with existing adds and
subtracts, I accidentally stopped it from doing so in some cases. This should
fix that. I'm also fixing another potential bug in that commit.
Differential Revision: https://reviews.llvm.org/D42263
llvm-svn: 322972
RuntimeLibcallSignatures previously manually initialized all the libcall
names into an array and searched it linearly for the first match to lookup
the corresponding index.
r322802 switched that to initializing a map keyed by the libcall name.
Neither of these approaches works correctly because some libcall numbers use
the same name on different platforms (e.g. the "l" suffixed functions
use f80 or f128 or ppcf128).
This change fixes that by ensuring that each name only goes into the map
once. It also adds tests.
Differential Revision: https://reviews.llvm.org/D42271
llvm-svn: 322971
Sign-extension opcodes have been split into a separate proposal from
the main threads proposal, so switch them to their own target
feature. See:
https://github.com/WebAssembly/sign-extension-ops
llvm-svn: 322966
Try to reverse the constant-shrinking that happens in SimplifyDemandedBits()
for 'and' masks when it results in a smaller sign-extended immediate.
We are also able to detect dead 'and' ops here (the mask is all ones). In
that case, we replace and return without selecting the 'and'.
Other targets might want to share some of this logic by enabling this under a
target hook, but I didn't see diffs for simple cases with PowerPC or AArch64,
so they may already have some specialized logic for this kind of thing or have
different needs.
This should solve PR35907:
https://bugs.llvm.org/show_bug.cgi?id=35907
Differential Revision: https://reviews.llvm.org/D42088
llvm-svn: 322957
This avoids playing games with pseudo pass IDs and avoids using an
unreliable MRI::isSSA() check to determine whether register allocation
has happened.
Note that this renames:
- MachineLICMID -> EarlyMachineLICM
- PostRAMachineLICMID -> MachineLICMID
to be consistent with the EarlyTailDuplicate/TailDuplicate naming.
llvm-svn: 322927
Re-commit of r322200: The testcase shouldn't hit machineverifiers
anymore with r322917 in place.
Large callframes (calls with several hundreds or thousands or
parameters) could lead to situations in which the emergency spillslot is
out of range to be addressed relative to the stack pointer.
This commit forces the use of a frame pointer in the presence of large
callframes.
This commit does several things:
- Compute max callframe size at the end of instruction selection.
- Add mirFileLoaded target callback. Use it to compute the max callframe size
after loading a .mir file when the size wasn't specified in the file.
- Let TargetFrameLowering::hasFP() return true if there exists a
callframe > 255 bytes.
- Always place the emergency spillslot close to FP if we have a frame
pointer.
- Note that `useFPForScavengingIndex()` would previously return false
when a base pointer was available leading to the emergency spillslot
getting allocated late (that's the whole effect of this callback).
Which made no sense to me so I took this case out: Even though the
emergency spillslot is technically not referenced by FP in this case
we still want it allocated early.
Differential Revision: https://reviews.llvm.org/D40876
llvm-svn: 322919
Do not create CALLSEQ_START/CALLSEQ_END when there is no callframe to
setup and the callframe size is 0.
- Fixes an invalid callframe nesting for byval arguments, which would
look like this before this patch (as in `big-byval.ll`):
...
ADJCALLSTACKDOWN 32768, 0, ... # Setup for extfunc
...
ADJCALLSTACKDOWN 0, 0, ... # setup for memcpy
...
BL &memcpy ...
ADJCALLSTACKUP 0, 0, ... # destroy for memcpy
...
BL &extfunc
ADJCALLSTACKUP 32768, 0, ... # destroy for extfunc
- Saves us two instructions in the common case of zero-sized stackframes.
- Remove an unnecessary scheduling barrier (hence the small unittest
changes).
Differential Revision: https://reviews.llvm.org/D42006
llvm-svn: 322917
This adds a new instrinsic to support the rdpid instruction. The implementation is a bit weird because the intrinsic is defined as always returning 32-bits, but the assembler support thinks the instruction produces a 64-bit register in 64-bit mode. But really it zeros the upper 32 bits. So I had to add separate patterns where 64-bit mode uses an extract_subreg.
Differential Revision: https://reviews.llvm.org/D42205
llvm-svn: 322910
Summary:
This patch implements d16 support for image load, image store and image sample intrinsics.
Reviewers:
Matt, Brian.
Differential Revision:
https://reviews.llvm.org/D3991
llvm-svn: 322903
Remove the tight coupling between llvm/CodeGenRuntimeLibcalls.def and
the table of supported singatures for wasm. This will allow adding new libcalls
without changing wasm's signature table.
Also, some cleanup:
Use ManagedStatics instead of const tables to avoid memory/binary bloat.
Use a StringMap instead of a linear search for name lookup.
Differential Revision: https://reviews.llvm.org/D35592
llvm-svn: 322802
Every known PE COFF target emits /EXPORT: linker flags into a .drective
section. The AsmPrinter should handle this.
While we're at it, use global_values() and emit each export flag with
its own .ascii directive. This should make the .s file output more
readable.
llvm-svn: 322788
Summary:
This patch adds a new target option in order to control GlobalISel.
This will allow the users to enable/disable GlobalISel prior to the
backend by calling `TargetMachine::setGlobalISel(bool Enable)`.
No test case as there is already a test to check GlobalISel
command line options.
See: CodeGen/AArch64/GlobalISel/gisel-commandline-option.ll.
Reviewers: qcolombet, aemerson, ab, dsanders
Reviewed By: qcolombet
Subscribers: rovka, javed.absar, kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D42137
llvm-svn: 322773
Trying to link
__attribute__((weak, visibility("hidden"))) extern int foo;
int *main(void) {
return &foo;
}
on OS X fails with
ld: 32-bit RIP relative reference out of range (-4294971318 max is +/-2GB): from _main (0x100000FAB) to _foo@0x00001000 (0x00000000) in '_main' from test.o for architecture x86_64
The problem being that 0 cannot be computed as a fixed difference from
%rip. Exactly the same issue exists on ELF and we can use the same
solution.
llvm-svn: 322739
This extends my previous patches to also optimize overflow-checked multiplies during SelectionDAG.
Differential revision: https://reviews.llvm.org/D40922
llvm-svn: 322738
The ARM backend contains code that tries to optimize compares by replacing them with an existing instruction that sets the flags the same way. This allows it to replace a "cmp" with a "adds", generalizing the code that replaces "cmp" with "sub". It also heuristically disables sinking of instructions that could potentially be used to replace compares (currently only if they're next to each other).
Differential revision: https://reviews.llvm.org/D38378
llvm-svn: 322737
Most are just replaced with instrs lists, but a few regexps have been further generalized to match more instructions with a single pattern.
llvm-svn: 322734
If we are splatting pairs of 32-bit elements, we can use a 64-bit broadcast to get the job done.
We could probably could probably do this with other sizes too, for example four 16-bit elements. Or we could broadcast pairs of 16-bit elements using a 32-bit element broadcast. But I've left that as a future improvement.
I've also restricted this to AVX2 only because we can only broadcast loads under AVX.
Differential Revision: https://reviews.llvm.org/D42086
llvm-svn: 322730
We legalize selects of masks with scalar conditions using a bitcast to an integer type. But if we are in 32-bit mode we can't convert v64i1 to i64. So instead split the v64i1 to v32i1 and concat it back together. Each half will then be legalized by bitcasting to i32 which is fine.
The test case is a little indirect. If we have the v64i1 select in IR it will get legalized by legalize vector ops which has a run of type legalization after it. That type legalization run is able to fix this i64 bitcast. So in order to avoid that we need a build_vector of a splat which legalize vector ops will ignore. Legalize DAG will then turn that into a select via LowerBUILD_VECTORvXi1. And the select will get legalized. In this case there is no type legalizer run to cleanup the bitcast.
This fixes pr35972.
llvm-svn: 322724
candidates with coldcc attribute.
This patch adds support for the coldcc calling convention for Power.
This changes the set of non-volatile registers. It includes a pass to stress
test the implementation by marking all static directly called functions with
the coldcc attribute through the option -enable-coldcc-stress-test. It also
includes an option, -ppc-enable-coldcc, to add the coldcc attribute to
functions which are cold at all call sites based on BlockFrequencyInfo when
the containing function does not call any non cold functions.
Differential Revision: https://reviews.llvm.org/D38413
llvm-svn: 322721
BRCTH is capable of a long branch which needs to be recognized during branch
relaxation. This is done by checking for ExtraRelaxSize == 0.
Review: Ulrich Weigand
llvm-svn: 322688
This assert typically happens if an unstructured CFG is passed
to the pass. This can happen if the pass is run independently
without the structurizer.
llvm-svn: 322685
Summary:
Loading a vector of 4 half-precision FP sometimes results in an LD1
of 2 single-precision FP + a reversal. This results in an incorrect
byte swap due to the conversion from little endian to big endian.
In order to generate the correct byte swap, it is easier to
generate the correct LD1 of 4 half-precision FP, thus avoiding the
subsequent reversal.
Reviewers: craig.topper, jmolloy, olista01
Reviewed By: olista01
Subscribers: efriedma, samparker, SjoerdMeijer, rogfer01, aemerson, rengolin, javed.absar, kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D41863
llvm-svn: 322663
When the compressed instruction set is enabled, the 16-bit c.nop can be
generated if necessary.
Differential Revision: https://reviews.llvm.org/D41221
Patch by Shiva Chen.
llvm-svn: 322658
Mark G_FPEXT and G_FPTRUNC as legal or libcall, depending on hardware
support, but only for conversions between float and double.
Also add the necessary boilerplate so that the LegalizerHelper can
introduce the required libcalls. This also works only for float and
double, but isn't too difficult to extend when the need arises.
llvm-svn: 322651
The match* functions have the annoying behavior of modifying its inputs.
Save and restore the inputs, just in case the early out for AVX512 is
hit. This is still not great and its only a matter of time this kind of
bug happens again, but I couldn't come up with a better pattern without
rewriting significant chunks of this code. Fixes PR35977.
llvm-svn: 322644
When "xer" is specified as clobbered register in inline assembler, clang can accept it, but llvm simply ignore it when lowered to machine instructions. It may cause problems later in scheduler.
This patch adds a new register XER aliased to CARRY, and adds it to register class CARRYRC. Now PPCTargetLowering::getRegForInlineAsmConstraint can return correct register number for inline asm constraint "{xer}", and scheduler behave correctly.
Differential Revision: https://reviews.llvm.org/D41967
llvm-svn: 322591
Summary:
This patch adds CustomRenderer which renders the matched
operands to the specified instruction.
Targets can enable the matching of SDNodeXForm by adding
a definition that inherits from GICustomOperandRenderer and
GISDNodeXFormEquiv as follows.
def gi_imm8 : GICustomOperandRenderer<"renderImm8”>,
GISDNodeXFormEquiv<imm8_xform>;
Custom renderer functions should be of the form:
void render(MachineInstrBuilder &MIB, const MachineInstr &I);
Reviewers: dsanders, ab, rovka
Reviewed By: dsanders
Subscribers: kristof.beyls, javed.absar, llvm-commits, mgrang, qcolombet
Differential Revision: https://reviews.llvm.org/D42012
llvm-svn: 322582
These pseudos are not supposed to be visible to user.
This patch reduced the auto-generated instruction matcher. For example,
the following words are removed from keyword list of LLVM BPF assembler.
- MCK__35_, // '#'
- MCK__COLON_, // ':'
- MCK__63_, // '?'
- MCK_ADJCALLSTACKDOWN, // 'ADJCALLSTACKDOWN'
- MCK_ADJCALLSTACKUP, // 'ADJCALLSTACKUP'
- MCK_PSEUDO, // 'PSEUDO'
- MCK_Select, // 'Select'
Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Jiong Wang <jiong.wang@netronome.com>
llvm-svn: 322535
As commented on the existing code:
// The Reg operand should be a virtual register, which is defined
// outside the current basic block. DAG combiner has done a pretty
// good job in removing truncating inside a single basic block.
However, when the Reg operand comes from bpf_load_[byte | half | word]
intrinsics, the generic optimizer doesn't understand their results are
zero extended, so these single basic block elimination opportunities were
missed.
Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Jiong Wang <jiong.wang@netronome.com>
llvm-svn: 322534
Prior to this we had a separate instruction and register class that excluded eax to prevent matching the instruction that would encode with 0x90.
This patch changes this to just use an InstAlias to force xchgl %eax, %eax to use XCHG32rr instruction in 64-bit mode. This gets rid of the separate instruction and register class.
llvm-svn: 322532
As mentioned on PR35869, (and came up recently on D41517) we don't create a MMX zero register via the PXOR but instead perform a spill to stack from a XMM zero register.
This patch adds support for direct MMX zero vector creation and should make it easier to add better constant vector creation in the future as well.
Differential Revision: https://reviews.llvm.org/D41908
llvm-svn: 322525
Add support for custom execution domain fixing and implement support for BLENDPD/BLENDPS/PBLENDD/PBLENDW.
Differential Revision: https://reviews.llvm.org/D42042
llvm-svn: 322524
In some cases we do not copy implicit defs from pseudo to real
VOP instructions. It has no visible impact at the moment thus no
tests are affected or added.
Differential Revision: https://reviews.llvm.org/D41783
llvm-svn: 322496
Since a load and test instruction treat its operands as signed, it can only
replace a logical compare for EQ/NE uses.
Review: Ulrich Weigand
https://bugs.llvm.org/show_bug.cgi?id=35662
llvm-svn: 322488
Summary:
For example, VSQRTSDZr and VSQRTSSZr were missing the predicate.
Also fix braces indentation and braces for consistency.
Reviewers: craig.topper, RKSimon
Suscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41983
llvm-svn: 322478
We have to take special care to avoid the cases where the result of the truncate would be padded with zero elements.
Ideally we'd just use ISD::TRUNCATE for these cases instead.
llvm-svn: 322454
Extend vXi1 conditions of vXi8/vXi16 selects even before type legalization gets a chance to split wide vectors. Previously we would only extend 128 and 256 bit vectors. But if we start with a 512 bit vector or wider that needs to be split we wouldn't extend until after the split had taken place. By extending early we improve the results of type legalization.
Don't widen condition of 128/256 bit vXi16/vXi8 selects when we have BWI but not VLX. We can still use a mask register by widening the select to 512-bits instead. This is similar to what we do for compares already.
llvm-svn: 322450
In addition to the existing match as part of a loop-reduction, add a
straightforward pattern match for DAG-contained patterns.
Reviewers: RKSimon, craig.topper
Subscribers: llvm-commits
Reviewed By: RKSimon
Differential Revision: https://reviews.llvm.org/D41811
llvm-svn: 322446
This avoids having the result type stick around until lowering where we have to extend the setcc and insert a truncate. If we get the types converted early we can do more to optimize it.
llvm-svn: 322432
*Mostly* NFC. Still updating the test though just for completeness.
This moves the hasAddressTaken check to MachineOutliner.cpp and replaces it
with a per-basic block test rather than a per-function test. The old test was
too conservative and was preventing functions in C programs from being
outlined even though they were safe to outline.
This was mostly a problem in C sources.
llvm-svn: 322425
Summary:
A recent change
321556: AMDGPU: Remove mayLoad/hasSideEffects from MIMG stores
can allow the machine instruction scheduler to move an image store past
an image load using the same descriptor.
V2: Fixed by marking image ops as mayAlias and isAliased. This may be
overly conservative, and we may need to revisit.
V3: Reverted test change done on 321556.
Reviewers: arsenm, nhaehnle, dstuttard
Subscribers: llvm-commits, t-tye, yaxunl, wdng, kzhuravl
Differential Revision: https://reviews.llvm.org/D41969
llvm-svn: 322419
Fix typos in the default scheduling resources when using the post indexed
addressing modes.
Differential revision: https://reviews.llvm.org/D40511
llvm-svn: 322392
We can probably take this a step further since the only
user of the isUsed flag is AsmParser it should probably
be doing this explicitly. For now this is a step in the
right direction though.
Differential Revision: https://reviews.llvm.org/D41971
llvm-svn: 322386
Part of the fix for https://bugs.llvm.org/show_bug.cgi?id=35812.
This patch ensures that the compare operand for the atomic compare and swap
is properly zero-extended to 32 bits if applicable.
A follow-up commit will fix the extension for the SETCC node generated when
expanding an ATOMIC_CMP_SWAP_WITH_SUCCESS. That will complete the bug fix.
Differential Revision: https://reviews.llvm.org/D41856
llvm-svn: 322372
For hard float with VFP4, it is legal. Otherwise, we use libcalls.
This needs a bit of support in the LegalizerHelper for soft float
because we didn't handle G_FMA libcalls yet. The support is trivial, as
the only difference between G_FMA and other libcalls that we already
handle is that it has 3 input operands rather than just 2.
llvm-svn: 322366
This patch teaches the Arm back-end to generate the SMMULR, SMMLAR and SMMLSR
instructions from equivalent IR patterns.
Differential Revision: https://reviews.llvm.org/D41775
llvm-svn: 322361
Without a register with a size being mentioned the instruction is ambiguous in at&t syntax. With Intel syntax the memory operation caries a size that can be used to disambiguate.
llvm-svn: 322356
While the suffix isn't required to disambiguate the instructions, it is required in order to parse the instructions when the suffix is specified in order to match the GNU assembler.
llvm-svn: 322354
Summary:
Very basic stack instrumentation using tagged pointers.
Tag for N'th alloca in a function is built as XOR of:
* base tag for the function, which is just some bits of SP (poor
man's random)
* small constant which is a function of N.
Allocas are aligned to 16 bytes. On every ReturnInst allocas are
re-tagged to catch use-after-return.
This implementation has a bunch of issues that will be taken care of
later:
1. lifetime intrinsics referring to tagged pointers are not
recognized in SDAG. This effectively disables stack coloring.
2. Generated code is quite inefficient. There is one extra
instruction at each memory access that adds the base tag to the
untagged alloca address. It would be better to keep tagged SP in a
callee-saved register and address allocas as an offset of that XOR
retag, but that needs better coordination between hwasan
instrumentation pass and prologue/epilogue insertion.
3. Lifetime instrinsics are ignored and use-after-scope is not
implemented. This would be harder to do than in ASan, because we
need to use a differently tagged pointer depending on which
lifetime.start / lifetime.end the current instruction is dominated
/ post-dominated.
Reviewers: kcc, alekseyshl
Subscribers: srhines, kubamracek, javed.absar, hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D41602
llvm-svn: 322324
The PeepholeOptimizer would fail for vregs without a definition. If this
was caused by an undef operand abort to keep the code simple (so we
don't need to add logic everywhere to replicate the undef flag).
Differential Revision: https://reviews.llvm.org/D40763
llvm-svn: 322319
While updating clang tests for having clang set dso_local I noticed
that:
- There are *a lot* of tests to update.
- Many of the updates are redundant.
They are redundant because a GV is "obviously dso_local". This patch
starts formalizing that a bit by requiring that internal and private
GVs be dso_local too. Since they all are, we don't have to print
dso_local to the textual representation, making it a bit more compact
and easier to read.
llvm-svn: 322317
Summary: This is a preparatory step for D41811: refactoring code for breaking vector operands of binary operation to legal-types.
Reviewers: RKSimon, craig.topper, spatel
Reviewed By: RKSimon
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41925
llvm-svn: 322296
In practice, this patch has no effect on scheduling.
There is no test case as there already exists a comprehensive test case for
LSE Atomics.
Patch by Stefan Teleman
Differential Revision: https://reviews.llvm.org/D40694
llvm-svn: 322291
Primarily, this allows us to use the aggressive extraction mechanisms in combineExtractWithShuffle earlier and make use of UNDEF elements that may be lost during lowering.
llvm-svn: 322279
Summary:
As RKSimon suggested in pr35820, in the case that Src is smaller in
bit-size than Indices, need to widen Src to avoid type mismatch.
Fixes pr35820
Reviewers: RKSimon, craig.topper
Reviewed By: RKSimon
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41865
llvm-svn: 322272
Although the register scavenger can often find a spare register, an emergency
spill slot is needed to guarantee success. Reserve this slot in cases where
the function is known to have a large stack (meaning the scavenger may be
needed when forming stack addresses).
llvm-svn: 322269
If the index is v2i64 we can use the scatter instruction that has v4i32/v4f32 data register, v2i64 index, and v2i1 mask. Similar was already done for gather.
Implement custom widening for v2i32 data to remove the code that reverses type legalization during lowering.
llvm-svn: 322254
Revert for now as the testcase is hitting a pre-existing verifier error
that manifest as a failure when expensive checks are enabled (or
-verify-machineinstrs) is used.
This reverts commit r322200.
llvm-svn: 322231
After D41349, we can no get a MCSubtargetInfo into the MCAsmBackend constructor. This allows us to get NOPL from a subtarget feature rather than a CPU name blacklist.
Differential Revision: https://reviews.llvm.org/D41721
llvm-svn: 322227
Branch relaxation is needed to support branch displacements that overflow the
instruction's immediate field.
Differential Revision: https://reviews.llvm.org/D40830
llvm-svn: 322224
This is a prerequisite for the branch relaxation pass, and allows a number of
optimisation passes (e.g. BranchFolding and MachineBlockPlacement) to work.
Differential Revision: https://reviews.llvm.org/D40808
llvm-svn: 322222
Includes support for expanding va_copy. Also adds support for using 'aligned'
registers when necessary for vararg calls, and ensure the frame pointer always
points to the bottom of the vararg spill region. This is necessary to ensure
that the saved return address and stack pointer are always available at fixed
known offsets of the frame pointer.
Differential Revision: https://reviews.llvm.org/D40805
llvm-svn: 322215
Currently we infer the scale at isel time by analyzing whether the base is a constant 0 or not. If it is we assume scale is 1, else we take it from the element size of the pass thru or stored value. This seems a little weird and I think it makes more sense to make it explicit in the DAG rather than doing tricky things in the backend.
Most of this patch is just making sure we copy the scale around everywhere.
Differential Revision: https://reviews.llvm.org/D40055
llvm-svn: 322210
ADRP instructions weren't being outlined because they're PC-relative and thus
fail the LR checks. This patch adds a special case for ADRPs to
getOutliningType to make sure that ADRPs can be outlined and updates the MIR
test.
llvm-svn: 322207
Large callframes (calls with several hundreds or thousands or
parameters) could lead to situations in which the emergency spillslot is
out of range to be addressed relative to the stack pointer.
This commit forces the use of a frame pointer in the presence of large
callframes.
This commit does several things:
- Compute max callframe size at the end of instruction selection.
- Add mirFileLoaded target callback. Use it to compute the max callframe size
after loading a .mir file when the size wasn't specified in the file.
- Let TargetFrameLowering::hasFP() return true if there exists a
callframe > 255 bytes.
- Always place the emergency spillslot close to FP if we have a frame
pointer.
- Note that `useFPForScavengingIndex()` would previously return false
when a base pointer was available leading to the emergency spillslot
getting allocated late (that's the whole effect of this callback).
Which made no sense to me so I took this case out: Even though the
emergency spillslot is technically not referenced by FP in this case
we still want it allocated early.
Differential Revision: https://reviews.llvm.org/D40876
llvm-svn: 322200
For hard float, it is legal.
For soft float, we need to lower to 0 - x first, and then we can use the
libcall for G_FSUB. This is undoing some of the canonicalization
performed by the IRTranslator (which introduces G_FNEG when it sees a
0 - x). Ideally, that canonicalization would be performed by a
pre-legalizer pass that would allow targets to opt out of this behaviour
rather than dance around it in the legalizer.
llvm-svn: 322168
Summary:
This extends TableGen's AsmMatcherEmitter with code that generates
a table with tied-operand constraints. The constraints are checked
when parsing the instruction. If an operand is not equal to its tied operand,
the assembler will give an error.
Patch [2/3] in a series to add operand constraint checks for SVE's predicated ADD/SUB.
Reviewers: olista01, rengolin, mcrosier, fhahn, craig.topper, evandro, echristo
Reviewed By: fhahn
Subscribers: javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D41446
llvm-svn: 322166
Since a load and test instruction treat its operands as signed, it can only
replace a logical compare for EQ/NE uses.
Review: Ulrich Weigand
https://bugs.llvm.org/show_bug.cgi?id=35662
llvm-svn: 322161
This patch makes the following changes to the schedule of instructions in the
prologue and epilogue.
The stack pointer update is moved down in the prologue so that the callee saves
do not have to wait for the update to happen.
Saving the lr is moved down in the prologue to hide the latency of the mflr.
The stack pointer is moved up in the epilogue so that restoring of the lr can
happen sooner.
The mtlr is moved up in the epilogue so that it is away form the blr at the end
of the epilogue. The latency of the mtlr can now be hidden by the loads of the
callee saved registers.
This commit is almost identical to this one: r322036 except that two warnings
that broke build bots have been fixed.
The revision number is D41737 as before.
llvm-svn: 322124
Summary:
I had a case where multiple nested uniform ifs resulted in code that did
v_cmp comparisons, combining the results with s_and_b64, s_or_b64 and
s_xor_b64 and using the resulting mask in s_cbranch_vccnz, without first
ensuring that bits for inactive lanes were clear.
There was already code for inserting an "s_and_b64 vcc, exec, vcc" to
clear bits for inactive lanes in the case that the branch is instruction
selected as s_cbranch_scc1 and is then changed to s_cbranch_vccnz in
SIFixSGPRCopies. I have added the same code into SILowerControlFlow for
the case that the branch is instruction selected as s_cbranch_vccnz.
This de-optimizes the code in some cases where the s_and is not needed,
because vcc is the result of a v_cmp, or multiple v_cmp instructions
combined by s_and/s_or. We should add a pass to re-optimize those cases.
Reviewers: arsenm, kzhuravl
Subscribers: wdng, yaxunl, t-tye, llvm-commits, dstuttard, timcorringham, nhaehnle
Differential Revision: https://reviews.llvm.org/D41292
llvm-svn: 322119
Summary:
If the vector type is transformed to non-vector single type, the compile
may crash trying to get vector information about non-vector type.
Reviewers: RKSimon, spatel, mkuper, hfinkel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41862
llvm-svn: 322106
Normally target independent DAG combine would do this combine based on getSetCCResultType, but with VLX getSetCCResultType returns a vXi1 type preventing the DAG combining from kicking in.
But doing this combine can allow us to remove the explicit sign extend that would otherwise be emitted.
This patch adds a target specific DAG combine to combine the sext+setcc when the result type is the same size as the input to the setcc. I've restricted this to FP compares and things that can be represented with PCMPEQ and PCMPGT since we don't have full integer compare support on the older ISAs.
Differential Revision: https://reviews.llvm.org/D41850
llvm-svn: 322101
In -debug output we print "pred:" whenever a MachineOperand is a
predicate operand in the instruction descriptor, and "opt:" whenever a
MachineOperand is an optional def in the instruction descriptor.
Differential Revision: https://reviews.llvm.org/D41870
llvm-svn: 322096
Fixed issue that was found on sanitizer-x86_64-linux-fast.
I changed the result type of 'Parser.getTok().getString().lower()'
in AArch64AsmParser::tryParseSVEPredicateVector() from 'StringRef' to
'auto', since StringRef::lower() returns a std::string.
llvm-svn: 322092
Summary:
Parsing of the '/m' (merging) or '/z' (zeroing) suffix of a predicate operand.
Patch [2/3] in a series to add predicated ADD/SUB instructions for SVE.
Reviewers: rengolin, mcrosier, evandro, fhahn, echristo, MatzeB, t.p.northover
Reviewed By: fhahn
Subscribers: t.p.northover, MatzeB, aemerson, javed.absar, tschuett, llvm-commits, kristof.beyls
Differential Revision: https://reviews.llvm.org/D41442
llvm-svn: 322070
Summary:
This commit enables some of the arithmetic instructions for Nios2 ISA (for both
R1 and R2 revisions), implements facilities required to emit those instructions
and provides LIT tests for added instructions.
Reviewed By: hfinkel
Differential Revision: https://reviews.llvm.org/D41236
Author: belickim <mateusz.belicki@intel.com>
llvm-svn: 322069
CET (Control-Flow Enforcement Technology) introduces a new mechanism called IBT (Indirect Branch Tracking).
According to IBT, each Indirect branch should land on dedicated ENDBR instruction (End Branch).
The new pass adds ENDBR instructions for every indirect jmp/call (including jumps using jump tables / switches).
For more information, please see the following:
https://software.intel.com/sites/default/files/managed/4d/2a/control-flow-enforcement-technology-preview.pdf
Differential Revision: https://reviews.llvm.org/D40482
Change-Id: Icb754489faf483a95248f96982a4e8b1009eb709
llvm-svn: 322062
The code that checks the immediate wasn't masking to the lower 3-bits like the code in X86InstrInfo.cpp that's used by the peephole pass does.
llvm-svn: 322060
The CTRLoop pass performs checks on the argument of certain libcalls/intrinsics,
and assumes the arguments must be of a simple type. This isn't always the case
though. For example if we unroll and vectorize a loop we may end up with vectors
larger then the largest legal type, along with intrinsics that operate on those
wider types. This happened in the ffmpeg build, where we unrolled a loop and
ended up with a sqrt intrinsic that operated on V16f64, triggering an assertion.
Differential Revision: https://reviews.llvm.org/D41758
llvm-svn: 322055
I had to drop fast-isel-abort from a test because we can't fast isel some of the mask stuff. When we used intrinsics we implicitly fell back to SelectionDAG for the intrinsic call without triggering the abort error. But with native IR that doesn't happen the same way.
llvm-svn: 322050
The pattern was this
def : Pat<(i32 (zext (i8 (bitconvert (v8i1 VK8:$src))))),
(MOVZX32rr8 (EXTRACT_SUBREG (i32 (COPY_TO_REGCLASS VK8:$src, GR32)), sub_8bit))>, Requires<[NoDQI]>;
but if you just let (i32 (zext X)) match byte itself you'll get MOVZX32rr8. And if you let (i8 (bitconvert (v8i1 VK8:$src))) match by itself you'll get (EXTRACT_SUBREG (i32 (COPY_TO_REGCLASS VK8:$src, GR32)), sub_8bit).
So we can just let isel do the two patterns naturally.
llvm-svn: 322049
This commit does two things. Firstly, it adds a collection of flags which can
be passed along to the target to encode information about the MBB that an
instruction lives in to the outliner.
Second, it adds some of those flags to the AArch64 outliner in order to add
more stack instructions to the list of legal instructions that are handled
by the outliner. The two flags added check if
- There are calls in the MachineBasicBlock containing the instruction
- The link register is available in the entire block
If the link register is available and there are no calls, then a stack
instruction can always be outlined without fixups, regardless of what it is,
since in this case, the outliner will never modify the stack to create a
call or outlined frame.
The motivation for doing this was checking which instructions are most often
missed by the outliner. Instructions like, say
%sp<def> = ADDXri %sp, 32, 0; flags: FrameDestroy
are very common, but cannot be outlined in the case that the outliner might
modify the stack. This commit allows us to outline instructions like this.
llvm-svn: 322048
This patch makes the following changes to the schedule of instructions in the
prologue and epilogue.
The stack pointer update is moved down in the prologue so that the callee saves
do not have to wait for the update to happen.
Saving the lr is moved down in the prologue to hide the latency of the mflr.
The stack pointer is moved up in the epilogue so that restoring of the lr can
happen sooner.
The mtlr is moved up in the epilogue so that it is away form the blr at the end
of the epilogue. The latency of the mtlr can now be hidden by the loads of the
callee saved registers.
Differential Revision: https://reviews.llvm.org/D41737
llvm-svn: 322036
This patch allows `r7` to be used, regardless of its use as a frame pointer, as
a temporary register when popping `lr`, and also falls back to using a high
temporary register if, for some reason, we weren't able to find a suitable low
one.
Differential revision: https://reviews.llvm.org/D40961
Fixes https://bugs.llvm.org/show_bug.cgi?id=35481
llvm-svn: 321989
(Target)FrameLowering::determineCalleeSaves can be called multiple
times. I don't think it should have side-effects as creating stack
objects and setting global MachineFunctionInfo state as it is doing
today (in other back-ends as well).
This moves the creation of stack objects from determineCalleeSaves to
assignCalleeSavedSpillSlots.
Differential Revision: https://reviews.llvm.org/D41703
llvm-svn: 321987
I had removed the qualifiers around the autogenerated folding table so I could compare with the manual table, but didn't intend to commit the change.
llvm-svn: 321971
Summary:
There are few oddities that occur due to v1i1, v8i1, v16i1 being legal without v2i1 and v4i1 being legal when we don't have VLX. Particularly during legalization of v2i32/v4i32/v2i64/v4i64 masked gather/scatter/load/store. We end up promoting the mask argument to these during type legalization and then have to widen the promoted type to v8iX/v16iX and truncate it to get the element size back down to v8i1/v16i1 to use a 512-bit operation. Since need to fill the upper bits of the mask we have to fill with 0s at the promoted type.
It would be better if we could just have the v2i1/v4i1 types as legal so they don't undergo any promotion. Then we can just widen with 0s directly in a k register. There are no real v4i1/v2i1 instructions anyway. Everything is done on a larger register anyway.
This also fixes an issue that we couldn't implement a masked vextractf32x4 from zmm to xmm properly.
We now have to support widening more compares to 512-bit to get a mask result out so new tablegen patterns got added.
I had to hack the legalizer for widening the operand of a setcc a bit so it didn't try create a setcc returning v4i32, extract from it, then try to promote it using a sign extend to v2i1. Now we create the setcc with v4i1 if the original setcc's result type is v2i1. Then extract that and don't sign extend it at all.
There's definitely room for improvement with some follow up patches.
Reviewers: RKSimon, zvi, guyblank
Reviewed By: RKSimon
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D41560
llvm-svn: 321967
Summary:
I believe legalization is really expecting that ReplaceNodeResults will return something with the same type as the thing that's being legalized. Ultimately, it uses the output to replace the uses in the DAG so the type should match to make that work.
There are two relevant cases here. When crbits are enabled, then i1 is a legal type and getSetCCResultType should return i1. In this case, the truncate will be between i1 and i1 and should be removed (SelectionDAG::getNode does this). Otherwise, getSetCCResultType will be i32 and the legalizer will promote the truncate to be i32 -> i32 which will be similarly removed.
With this fixed we can remove some code from PromoteIntRes_SETCC that seemed to only exist to deal with the intrinsic being replaced with a larger type without changing the other operand. With the truncate being used for connectivity this doesn't happen anymore.
Reviewers: hfinkel
Reviewed By: hfinkel
Subscribers: nemanjai, llvm-commits, kbarton
Differential Revision: https://reviews.llvm.org/D41654
llvm-svn: 321959
The instructions that load 64-bits or an xmm register should be TB_NO_REVERSE to avoid the load being widened during unfold. The instructions that load 128-bits need to ensure 128-bit alignment.
llvm-svn: 321956
The memory form of the xmm->xmm version only writes 64-bits. If we use it in the folding tables and its get used for a stack spill, only half the slot will be written. Then a reload may read all 128-bits which will pull in garbage. But without the spill the upper bits of the register would have been zero. By not folding we would preserve the zeros.
llvm-svn: 321950
We don't do fine grained feature control like this on features prior to AVX512.
We do still have checks in place in the assembly parser itself that prevents %zmm references or %xmm16-31 from being parsed without at least -mattr=avx512f. Same for rounding control and mask operands. That will prevent the table matcher from matching for any instructions that need those features and that's probably good enough.
llvm-svn: 321947
This is the last step needed to fix PR33325:
https://bugs.llvm.org/show_bug.cgi?id=33325
We're trading branch and compares for loads and logic ops.
This makes the code smaller and hopefully faster in most cases.
The 24-byte test shows an interesting construct: we load the trailing scalar
elements into vector registers and generate the same pcmpeq+movmsk code that
we expected for a pair of full vector elements (see the 32- and 64-byte tests).
Differential Revision: https://reviews.llvm.org/D41714
llvm-svn: 321934
Without this we allow "vmovd %rax, %xmm0", but not "vmovd %rax, %xmm16"
This exists due to continue a silly bug where really old versions of the GNU assembler required movd instead of movq on these instructions. This compatibility hack then crept forward to avx version too, but we didn't propagate it to avx512.
llvm-svn: 321903
This behavior existed to work with an old version of the gnu assembler on MacOS that only accepted this form. Newer versions of GNU assembler and the current LLVM derived version of the assembler on MacOS support movq as well.
llvm-svn: 321898
Currently the assembler would accept, e.g. `ldr r0, [s0, #12]` and similar.
This patch add checks that only general-purpose registers are used in address
operands, shifted registers, and shift amounts.
Differential revision: https://reviews.llvm.org/D39910
llvm-svn: 321866
Instead of using, for example, `dup v0.4s, wzr`, which transfers between
register files, use the more efficient `movi v0.4s, #0` instead.
Differential revision: https://reviews.llvm.org/D41515
llvm-svn: 321824
Wide Thumb2 instructions should be emitted into the object file as pairs of
16-bit words of the appropriate endianness, not one 32-bit word.
Differential revision: https://reviews.llvm.org/D41185
llvm-svn: 321799
Select G_PHI to PHI and manually constrain the result register. This is
very similar to how COPY is handled, so extract and reuse some of that
code.
llvm-svn: 321797
We used to handle G_CONSTANT with pointer type by forcing the type of
the result register to s32 and then letting TableGen handle it.
Unfortunately, setting the type only works for generic virtual
registers, that haven't yet been constrained to a register class (e.g.
those used only by a COPY later on). If the result register has already
been constrained as a use of a previously selected instruction, then
setting the type will assert.
It would be nice to be able to teach TableGen to select pointer
constants the same as integer constants, but since it's such an edge
case (at the moment the only pointer constant that we're generally
interested in is 0, and that is mostly used for comparisons and selects,
which are also not supported by TableGen) it's probably not worth the
effort right now. Instead, handle pointer constants with some trivial
handwritten code.
llvm-svn: 321793
This custom inserter was added in r124272 at which time it added about bunch of Defs for Win64. In r150708, those defs were removed leaving only the "return BB". So I think this means the custom inserter is a NOP these days.
This patch removes the remaining code and stops tagging the instructions for custom insertion
Differential Revision: https://reviews.llvm.org/D41671
llvm-svn: 321747
Currently we use SIGN_EXTEND in lowerMasksToReg as part of calling convention setup, but we don't require a specific value for the upper bits.
This patch changes it to ANY_EXTEND which will be lowered as SIGN_EXTEND if it ends up sticking around.
llvm-svn: 321746
After D41349, we can now directly access MCSubtargetInfo from
createARM*AsmBackend. This patch makes use of this, avoiding the need to
create a fresh MCSubtargetInfo (which was previously always done with a blank
CPU and feature string). Given the total size of the change remains pretty
tiny and we're removing the old explicit destructor, I changed the STI field
to a reference rather than a pointer.
Differential Revision: https://reviews.llvm.org/D41693
llvm-svn: 321707
Summary:
Add a register class for SVE predicate operands that can only be p0-p7 (as opposed to p0-p15)
Patch [1/3] in a series to add predicated ADD/SUB instructions for SVE.
Reviewers: rengolin, mcrosier, evandro, fhahn, echristo, olista01, SjoerdMeijer, javed.absar
Reviewed By: fhahn
Subscribers: aemerson, javed.absar, tschuett, kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D41441
llvm-svn: 321699
Currently it's not possible to access MCSubtargetInfo from a TgtMCAsmBackend.
D20830 threaded an MCSubtargetInfo reference through
MCAsmBackend::relaxInstruction, but this isn't the only function that would
benefit from access. This patch removes the Triple and CPUString arguments
from createMCAsmBackend and replaces them with MCSubtargetInfo.
This patch just changes the interface without making any intentional
functional changes. Once in, several cleanups are possible:
* Get rid of the awkward MCSubtargetInfo handling in ARMAsmBackend
* Support 16-bit instructions when valid in MipsAsmBackend::writeNopData
* Get rid of the CPU string parsing in X86AsmBackend and just use a SubtargetFeature for HasNopl
* Emit 16-bit nops in RISCVAsmBackend::writeNopData if the compressed instruction set extension is enabled (see D41221)
This change initially exposed PR35686, which has since been resolved in r321026.
Differential Revision: https://reviews.llvm.org/D41349
llvm-svn: 321692
This is an extension of D31156 with the goal that we'll allow memcmp() == 0 expansion
for x86 to use 2 pairs of loads per block.
The memcmp expansion pass (formerly part of CGP) will generate this kind of pattern
with oversized integer compares, so we want to transform these into x86-specific vector
nodes before legalization splits things into scalar chunks.
See PR33325 for more details:
https://bugs.llvm.org/show_bug.cgi?id=33325
Differential Revision: https://reviews.llvm.org/D41618
llvm-svn: 321656
Tests updated to explicitly use fast-isel at -O0 instead of implicitly.
This change also allows an explicit -fast-isel option to override an
implicitly enabled global-isel. Otherwise -fast-isel would have no effect at -O0.
Differential Revision: https://reviews.llvm.org/D41362
llvm-svn: 321655
Summary:
isReg() in AArch64AsmParser.cpp is a bit of a misnomer, and would be better named 'isScalarReg()' instead.
Patch [1/3] in a series to add operand constraint checks for SVE's predicated ADD/SUB.
Reviewers: rengolin, mcrosier, evandro, fhahn, echristo
Reviewed By: fhahn
Subscribers: aemerson, javed.absar, llvm-commits, kristof.beyls
Differential Revision: https://reviews.llvm.org/D41445
llvm-svn: 321646
Currently the promotion for these ignores the normal getTypeToPromoteTo and instead just tries to double the element width. This is because the default behavior of getTypeToPromote to just adds 1 to the SimpleVT, which has the affect of increasing the element count while keeping the scalar size the same.
If multiple steps are required to get to a legal operation type, int_to_fp will be promoted multiple times. And fp_to_int will keep trying wider types in a loop until it finds one that works.
getTypeToPromoteTo does have the ability to query a promotion map to get the type and not do the increasing behavior. It seems better to just let the target specify the promotion type in the map explicitly instead of letting the legalizer iterate via widening.
FWIW, it's worth I think for any other vector operations that need to be promoted, we have to specify the type explicitly because the default behavior of getTypeToPromote isn't useful for vectors. The other types of promotion already require either the element count is constant or the total vector width is constant, but neither happens by incrementing the SimpleVT enum.
Differential Revision: https://reviews.llvm.org/D40664
llvm-svn: 321629
We can use zmm move with zero masking for this. We already had patterns for using a masked move, but we didn't check for the zero masking case separately.
llvm-svn: 321612
The CONCAT_VECTORS will be lowered to INSERT_SUBVECTOR later. In the modified cases this seems to be enough to trick a later DAG combine into running in a different order than allows the ANDs to be removed.
I'll admit this is a bit of a hack that happens to work, but using CONCAT_VECTORS is more consistent with other legalization code anyway.
llvm-svn: 321611
Don't combine buildvector(binop(),binop(),binop(),binop()) -> binop(buildvector(), buildvector()) if its a splat - keep the binop scalar and just splat the result to avoid large vector constants.
llvm-svn: 321607
We end up using an i8 load via an isel pattern from v8i1 anyway. This just makes it more explicit. This seems to improve codgen in some cases and I'd like to kill off some of the load patterns.
llvm-svn: 321598
If the callee and caller use different calling convensions, we cannot apply TCO if the callee requires arguments on stack; e.g. C calling convention and Fast CC use the same registers for parameter passing, but the stack offset is not necessarily same.
This patch also recommit r319218 "[PowerPC] Allow tail calls of fastcc functions from C CallingConv functions." by @sfertile since the problem reported in r320106 should be fixed.
Differential Revision: https://reviews.llvm.org/D40893
llvm-svn: 321579
Initially, if the `c` constraint applied to the wrong data type that
causes LLVM to assert. This commit replaces the assert by an error
message.
llvm-svn: 321565
Currently all images are lowered to have a single
image PseudoSourceValue. Image stores happen to have
overly strict mayLoad/mayStore/hasSideEffects flags
set on them, so this happens to work. When these
are fixed to be correct, the scheduler breaks
this because the identical PSVs are assumed to
be the same address. These need to be unique
to the image resource value.
llvm-svn: 321555
As noted in PR34686, we are relying on a PSHUFD+PSHUFLW+PSHUFHW shuffle chain for most general vXi16 unary shuffles.
This patch checks for simpler PSHUFLW+PSHUFD and PSHUFHW+PSHUFD cases beforehand, building on some existing code that just handled splat shuffles.
By doing so we also prevent premature use of PSHUFB shuffles which can be slower and require the creation/loading of constant shuffle masks.
We now have the 'fast-variable-shuffle' option for hardware that prefers combining 2 or more shuffles to VPSHUFB etc.
Differential Revision: https://reviews.llvm.org/D38318
llvm-svn: 321553
Revision 320791 introduced a pass that transforms reg+reg instructions to
reg+imm if they're fed by "load immediate". However, it didn't
handle out-of-range shifts correctly as reported in PR35688.
This patch fixes that and therefore the PR.
Furthermore, there was undefined behaviour in the patch where the RHS of an
initialization expression was 32 bits and constant `1` was shifted left 32
bits. This was fixed by ensuring the RHS is 64 bits just like the LHS.
Differential Revision: https://reviews.llvm.org/D41369
llvm-svn: 321551
Previously we used an extend from v8i1 to v8i32/v8i64. Then extracted to the final width. But if we have VLX we should extract first. This way we don't end up with an overly large extend.
This allows us to use vcmpeq to make all ones for the sign extend when DQI isn't available. Otherwise we get a VPTERNLOG.
If we make v2i1/v4i1 legal like proposed in D41560, we could always do this and rely on the lowering of the extend to widen when necessary.
llvm-svn: 321538
-Use MinAlign instead of std::min.
-Use SelectionDAG::getMemBasePlusOffset.
-Apply offset to the pointer info for the second load/store created.
llvm-svn: 321536
The exception handler thunk needs to reference the LSDA of the parent
function, which won't be emitted if it's available_externally.
Fixes PR35736. ThinLTO ends up producing available_externally functions
that use _CxxFrameHandler3.
llvm-svn: 321532
If there are 17 or more leading zeros to the v4i32 elements, then we can use PMADD for the integer multiply when PMULLD is unavailable or slow.
The 17 bits need to be zero as the PMADDWD performs a v8i16 signed-mul-extend + pairwise-add - the upper 16 so we're adding a zero pair and the 17th bit so we don't incorrectly sign extend.
Differential Revision: https://reviews.llvm.org/D41484
llvm-svn: 321516
My original implementation ran as a DAG combine post type legalization, but it turns out we don't run that DAG combine step if type legalization didn't change anything. Attempts to make the combine run before type legalization as well hit other issues.
So just do it in LowerMUL where we can catch more cases.
llvm-svn: 321496
r319980 added new patterns to the machine combiner for transforming (fsub (fmul
x y) z) into (fmla (fneg z) x y). That is, fsub's where the first source
operand is an fmul are transformed. We previously only matched the case where
the second source operand of an fsub was an fmul, transforming (fsub z (fmul x
y)) into (fmls z x y). Now, if we have an fsub where both source operands are
fmuls, both of the above patterns are applicable.
However, the order in which we add the patterns to the list of candidates
determines the transformation that takes place, since only the first pattern
that matches will be used. This patch changes the order these two patterns are
added to the list of candidates such that we prefer the case where the second
source operand is an fmul (the fmls case), rather than the other one (the
fmla/fneg case). When both source operands are fmuls, this ordering results in
fewer instructions.
Differential Revision: https://reviews.llvm.org/D41587
llvm-svn: 321491
Returning SDValue() means nothing changed, SDValue(N,0) means there was a change but the worklist management was taken care of.
I don't know if this has a real effect other than making sure the combine counter in the DAG combiner gets updated, but it is the correct thing to do.
llvm-svn: 321463
Normally we catch this during lowering, but vXi64 mul is considered legal when we have AVX512DQ.
This DAG combine allows us to avoid PMULLQ with AVX512DQ if we can prove its unnecessary. PMULLQ is 3 uops that take 4 cycles each. While pmuldq/pmuludq is only one 4 cycle uop.
llvm-svn: 321437
Match regular x87 memory fold instructions with load/sideeffects tags, to prevent the schedulers from re-ordering them across the fnstcw/fldcw sequences for truncating stores while they are still pseudo during the stack conversion pass.
llvm-svn: 321424
Previously we extended v2i1 to v2f64 and then tried to use cvtuqq2pd/cvtqq2pd, but that only works with avx512dq. So we ended up scalarizing it. Now we widen to v4i1 first and extend to v4i32.
llvm-svn: 321420
Immediately after it is created we check if its equal to another EVT. Then we inconsistently use one or the other variables in the code below.
Instead do the equality check directly on the getValueType result and remove the variable. Use the origina VT variable throughout the remaining code.
llvm-svn: 321406
getOperand returns an SDValue that contains the node and the result number. There is no guarantee that the result number if 0. By using the -> operator we are calling SDNode::getValueType rather than SDValue::getValueType. This requires supplying a result number and we shouldn't assume it was 0.
I don't have a test case. Just noticed while cleaning up some other code and saw that it occurred in other places.
llvm-svn: 321397
Re-land r321234. It had to be reverted because it broke the shared
library build. The shared library build broke because there was a
missing LLVMBuild dependency from lib/Passes (which calls
TargetMachine::getTargetIRAnalysis) to lib/Target. As far as I can
tell, this problem was always there but was somehow masked
before (perhaps because TargetMachine::getTargetIRAnalysis was a
virtual function).
Original commit message:
This makes the TargetMachine interface a bit simpler. We still need
the std::function in TargetIRAnalysis to avoid having to add a
dependency from Analysis to Target.
See discussion:
http://lists.llvm.org/pipermail/llvm-dev/2017-December/119749.html
I avoided adding all of the backend owners to this review since the
change is simple, but let me know if you feel differently about this.
Reviewers: echristo, MatzeB, hfinkel
Reviewed By: hfinkel
Subscribers: jholewinski, jfb, arsenm, dschuff, mcrosier, sdardis, nemanjai, nhaehnle, javed.absar, sbc100, jgravelle-google, aheejin, kbarton, llvm-commits
Differential Revision: https://reviews.llvm.org/D41464
llvm-svn: 321375
Despite what the comment said there isn't better codegen for 512-bit vectors. The 128/256/512 bit implementation jus stores to memory and loads an element. There's no advantage to doing that with a larger size. In fact in many cases it causes a stack realignment and generates worse code.
llvm-svn: 321369
Pointer constants are pretty rare, since we usually represent them as
integer constants and then cast to pointer. One notable exception is the
null pointer constant, which is represented directly as a G_CONSTANT 0
with pointer type. Mark it as legal and make sure it is selected like
any other integer constant.
llvm-svn: 321354
Previously prefetch was only considered legal if sse was enabled, but it should be supported with 3dnow as well.
The prfchw flag now imply at least some form of prefetch without the write hint is available, either the sse or 3dnow version. This is true even if 3dnow and sse are explicitly disabled.
Similarly prefetchwt1 feature implies availability of prefetchw and the the prefetcht0/1/2/nta instructions. This way we can support _MM_HINT_ET0 using prefetchw and _MM_HINT_ET1 with prefetchwt1. And its assumed that if we have levels for the write hint we would have levels for the non-write hint, thus why we enable the sse prefetch instructions.
I believe this behavior is consistent with gcc. I've updated the prefetch.ll to test all of these combinations.
llvm-svn: 321335
The penalty is currently getting applied in a bunch of places where it
doesn't make sense, like bitcasts (which are free) and calls (which
were getting the call penalty applied twice). Instead, just apply the
penalty to binary operators and floating-point casts.
While I'm here, also fix getFPOpCost() to do the right thing in more
cases, so we don't have to dig into function attributes.
Differential Revision: https://reviews.llvm.org/D41522
llvm-svn: 321332
This should only affect what we do for v8i16. Previously we went to v8i64, but if we have VLX we only need v8i32. This prevents an unnecessary zmm usage.
llvm-svn: 321303
We should have equally good shuffle options for v8i32 with VLX. This was spotted during my attempts to remove 512-bit vectors from SKX.
We still use 512-bits for v16i1, v32i1, and v64i1. I'm less sure we can handle those well with narrower vectors. i32 and i64 element sizes get the best shuffle support.
llvm-svn: 321291
The build failure was caused by an assertion in pre-legalization DAGCombine:
Combining: t6: ppcf128 = uint_to_fp t5
... into: t20: f32 = PPCISD::FCFIDUS t19
which is clearly wrong since ppcf128 are definitely different type with f32 and
we cannot change the node value type when do DAGCombine. The fix is don't
handle ppc_fp128 or i1 conversions in PPCTargetLowering::combineFPToIntToFP and
leave it to downstream to legalize it and expand it to small legal types.
Differential Revision: https://reviews.llvm.org/D41411
llvm-svn: 321276
Summary:
This makes the TargetMachine interface a bit simpler. We still need
the std::function in TargetIRAnalysis to avoid having to add a
dependency from Analysis to Target.
See discussion:
http://lists.llvm.org/pipermail/llvm-dev/2017-December/119749.html
I avoided adding all of the backend owners to this review since the
change is simple, but let me know if you feel differently about this.
Reviewers: echristo, MatzeB, hfinkel
Reviewed By: hfinkel
Subscribers: jholewinski, jfb, arsenm, dschuff, mcrosier, sdardis, nemanjai, nhaehnle, javed.absar, sbc100, jgravelle-google, aheejin, kbarton, llvm-commits
Differential Revision: https://reviews.llvm.org/D41464
llvm-svn: 321234
The AArch64 backend contains code to optimize {s,u}{add,sub}.with.overflow during SelectionDAG. This commit ports that code to the ARM backend.
Differential revision: https://reviews.llvm.org/D35635
llvm-svn: 321224
Add support for 'objdump -print-imm-hex' for imm64, operand imm
and branch target. If user programs encode immediate values
as hex numbers, such an option will make it easy to correlate
asm insns with source code. This option also makes it easy
to correlate imm values with insn encoding.
There is one changed behavior in this patch. In old way, we
print the 64bit imm as u64:
O << (uint64_t)Op.getImm();
and the new way is:
O << formatImm(Op.getImm());
The formatImm is defined in llvm/MC/MCInstPrinter.h as
format_object<int64_t> formatImm(int64_t Value)
So the new way to print 64bit imm is i64 type.
If a 64bit value has the highest bit set, the old way
will print the value as a positive value and the
new way will print as a negative value. The new way
is consistent with x86_64.
For the code (see the test program):
...
if (a == 0xABCDABCDabcdabcdULL)
...
x86_64 objdump, with and without -print-imm-hex, looks like:
48 b8 cd ab cd ab cd ab cd ab movabsq $-6067004223159161907, %rax
48 b8 cd ab cd ab cd ab cd ab movabsq $-0x5432543254325433, %rax
Signed-off-by: Yonghong Song <yhs@fb.com>
llvm-svn: 321215
Multiple Closure objects can be created and stored for a single function. It's not a good idea to devote so many fields of it to storing pointers and references to global data structures of the pass. The closure class should only store the things needed to represent the closure itself.
This patch refactors many of the methods of Closure to belong to the pass object and to pass around a reference to the current Closure. The Closure class gains a few simple methods to add instructions and edges, and to return iterators to edges and instructions
Differential Revision: https://reviews.llvm.org/D41327
llvm-svn: 321213
Gather/scatter can implicitly sign extend from i32->i64 on indices. So if we know the sign bit of the input to a zext is 0 we can use the implicit extension.
llvm-svn: 321209
The function createTailCallBranchInstr assumes that the iterator MBBI is valid.
However, only one use of MBBI is guarded in the function.
Fix this by adding an assert.
Differential Revision: https://reviews.llvm.org/D41358
llvm-svn: 321205
This patch turns shuffles of fadd/fsub with fmul into fmsubadd.
Patch by Dmitry Venikov
Differential Revision: https://reviews.llvm.org/D40335
llvm-svn: 321200
We get an assertion in RegBankSelect for code along the lines of
my_32_bit_int = my_64_bit_int, which tends to translate into a 64-bit
load, followed by a G_TRUNC, followed by a 32-bit store. This appears in
a couple of places in the test-suite.
At the moment, the legalizer doesn't distinguish between integer and
floating point scalars, so a 64-bit load will be marked as legal for
targets with VFP, and so will the rest of the sequence, leading to a
slightly bizarre G_TRUNC reaching RegBankSelect.
Since the current support for 64-bit integers is rather immature, this
patch works around the issue by explicitly handling this case in
RegBankSelect and InstructionSelect. In the future, we may want to
revisit this decision and make sure 64-bit integer loads are narrowed
before reaching RegBankSelect.
llvm-svn: 321165
Summary:
Implement lower of unsigned saturation on an interval [0, k] where k + 1 is a power of two using USAT instruction in a similar way to how [~k, k] is lowered using SSAT on ARM models that supports it.
Patch by Marten Svanfeldt
Reviewers: t.p.northover, pbarrio, eastig, SjoerdMeijer, javed.absar, fhahn
Reviewed By: fhahn
Subscribers: fhahn, aemerson, javed.absar, llvm-commits, kristof.beyls
Differential Revision: https://reviews.llvm.org/D41348
llvm-svn: 321164
This patch resubmits the SVE ZIP1/ZIP2 patch series consisting of
of r320992, r320986, r320973, and r320970 by reverting
https://reviews.llvm.org/rL321024.
The issue that caused r321024 has been addressed in https://reviews.llvm.org/rL321158,
so this patch-series should be safe to resubmit.
llvm-svn: 321163
Implement the 'Current Cache Size' register that has been introduced
as part of the Armv8.3 architecture. I originally missed this, and
(hopefully) should be the final patch for assembler support.
Differential Revision: https://reviews.llvm.org/D41396
llvm-svn: 321155
The gather instruction will implicitly sign extend to the pointer width, we don't need to further extend it. This can prevent unnecessary splitting in some cases.
There's still an issue that lowering on non-VLX can introduce another sign extend that doesn't get combined with shifts from a lowered sign_extend_inreg.
llvm-svn: 321152
Not sure how to test this cause I think the worst that happens is that we don't revisit the node a second time to look for additional combines. We used UpdateNodeOperands so the updating the DAG work was already done.
llvm-svn: 321148
This patch fixes a bug in the redundant compare elimination reported in https://reviews.llvm.org/rL320786 and re-enables the optimization.
The redundant compare elimination assumes that we can replace signed comparison with unsigned comparison for the equality check. But due to the difference in the sign extension behavior we cannot change the opcode if the comparison is against an immediate and the most significant bit of the immediate is one.
Differential Revision: https://reviews.llvm.org/D41385
llvm-svn: 321147
These optimizations depend on the ExplicitLocals pass to lower TEE
instructions, which is disabled in the ELF ABI, so disable them too.
llvm-svn: 321131
We try to prevent shuffle combining to value types that would stop the folding of masked operations, but by just returning early, we were failing to try different shuffle types.
The TODOs are all still relevant here to improve codegen but we're lacking test examples.
llvm-svn: 321085
As mentioned in D38318 and D40865, modern Intel processors prefer to combine multiple shuffles to a variable shuffle mask (PSHUFB/VPERMPS etc.) instead of having multiple stage 'fixed' shuffles which put more pressure on Port 5 (at the expense of extra shuffle mask loads).
This patch provides a FeatureFastVariableShuffle target flag for Haswell+ CPUs that prefers combining 2 or more fixed shuffles to a single variable shuffle (default is 3 shuffles).
The long term aim is to drive more of this from schedule data (probably via the MC) but we're not close to being ready for that yet.
Differential Revision: https://reviews.llvm.org/D41323
llvm-svn: 321074
Extension to D39729 which performed this for vXi16, with the same bit flipping to handle SMAX/SMIN/UMAX cases, vXi8 UMIN horizontal reductions can be performed.
This makes use of the fact that by performing a pair-wise i8 SHUFFLE/UMIN before PHMINPOSUW, we both get the UMIN of each pair but also zero-extend the upper bits ready for v8i16.
Differential Revision: https://reviews.llvm.org/D41294
llvm-svn: 321070
This instruction is encoded as zero, so we have handle that case when checking
for unimplemented opcodes when producing the encoding for an instruction.
llvm-svn: 321066
BWI supports shifting by word amounts. Even if VLX isn't support we can still widen to v32i16 and extract the lower half. For SKX its preferrable to not use 512-bit vector if we can.
llvm-svn: 321059
Previously, we were checking for MVTs with sizes betwen 8 and 64 which only includes i8, i16, i32, and i64 today. But I don't think we should assume that and should list the types that are legal for x86. I also don't think we need i64 since type legalization is guaranteed to split those up.
llvm-svn: 321058
My reading of the SDM says that all bits of the shift amount are used. If the value of the element is larger than the number of bits the result the shift result is zero. So I think we need to zero_extend here to avoid garbage in the upper bits.
In reality we lower any_extend as zero_extend so in most cases it would be hard to hit this.
llvm-svn: 321055
Note:
- X86ISelLowering: setLibcallName(SINCOS) was superfluous as
InitLibcalls() already does it.
- ARMISelLowering: Setting libcallnames for sincos/sincosf seemed
superfluous as in the darwin case it wouldn't be used while for all
other cases InitLibcalls already does it.
llvm-svn: 321036
Empty string should be equivalent to "generic" which doesn't allow NOPL. Force tests to use specificy 'pentiumpro' to guarantee NOPL.
Fixes PR35686
llvm-svn: 321026
This reverts changes r320992, r320986, r320973, and r320970.
r320970 by itself breaks the test case, and the rest depend on it.
Test case will land soon.
llvm-svn: 321024
LR was undefined entering outlined functions that contain calls. This made the
machine verifier unhappy when expensive checks were enabled. This fixes that.
llvm-svn: 321014
Hiroshi Inoue