Summary:
Fixes PR42973. Tests don't change because simd-arith.ll tests behavior
on unimplemented-simd128, which does not include any temporary
workarounds such as the one removed in this revision.
Reviewers: aheejin
Subscribers: dschuff, sbc100, jgravelle-google, hiraditya, sunfish, dmgreen, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D66166
llvm-svn: 368868
Now that we legalize by widening, the element types here won't change. Previously these were modeled as the elements being widened and then the instruction might become an AND or SHL/ASHR pair. But now they'll become something like a ZERO_EXTEND_VECTOR_INREG/SIGN_EXTEND_VECTOR_INREG.
For AVX2, when the destination type is legal its clear the cost should be 1 since we have extend instructions that can produce 256 bit vectors from less than 128 bit vectors. I'm a little less sure about AVX1 costs, but I think the ones I changed were definitely too high, but they might still be too high.
Differential Revision: https://reviews.llvm.org/D66169
llvm-svn: 368858
This patch implements global address lowering for 32/64 bit with small/large code models.
1.For 32bit large code model on AIX, there are newly added pseudo opcode LWZtocL & ADDIStocHA32, the support of which on MC layer will be
provided by future patches.
2.The default code model on AIX should be small code model.
3.Since AIX does not have medium code model, "report_fatal_error" when users specify it.
Differential Revision: https://reviews.llvm.org/D63547
llvm-svn: 368744
That change (r363670) could leave a copy from vgpr to sgpr. Fixed.
Differential Revision: https://reviews.llvm.org/D66133
Change-Id: I00c3fe6fda2e8e1e36f53195b881b1449c777ea4
llvm-svn: 368736
The MVE architecture has the idea of "beats", where a vector instruction can be
executed over several ticks of the architecture. This adds a similar system
into the Arm backend cost model, multiplying the cost of all vector
instructions by a factor.
This factor essentially becomes the expected difference between scalar code
and vector code, on average. MVE Vector instructions can also overlap so the a
true cost of them is often lower. But equally scalar instructions can in some
situations be dual issued, or have other optimisations such as unrolling or
make use of dsp instructions. The default is chosen as 2. This should not
prevent vectorisation is a most cases (as the vector instructions will still be
doing at least 4 times the work), but it will help prevent over vectorising in
cases where the benefits are less likely.
This adds things so far to the obvious places in ARMTargetTransformInfo, and
updates a few related costs like not treating float instructions as cost 2 just
because they are floats.
Differential Revision: https://reviews.llvm.org/D66005
llvm-svn: 368733
Summary:
While D65962 is pending for review, I landed D65475 that added one more
use of `unsigned`. Changed it to `Register`.
Reviewers: dsanders
Subscribers: dschuff, sbc100, jgravelle-google, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D66064
llvm-svn: 368727
Trying again with the code changes (and not just the new test).
Summary:
This patch fixes the offsets of fields in the stack frame linkage save
area for AIX.
Reviewers: sfertile, hubert.reinterpretcast, jasonliu, Xiangling_L, xingxue, ZarkoCA, daltenty
Reviewed By: hubert.reinterpretcast
Subscribers: wuzish, nemanjai, hiraditya, kbarton, MaskRay, jsji, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64424
Patch by Chris Bowler!
llvm-svn: 368721
Currently shufflemasks get emitted as any other constant, and you end
up with a bunch of virtual registers of G_CONSTANT with a
G_BUILD_VECTOR. The AArch64 selector then asserts on anything that
doesn't fit this pattern. This isn't an ideal representation, and
should avoid legalization and have fewer opportunities for a
representational error.
Rather than invent a new shuffle mask operand type, similar to what
ShuffleVectorSDNode does, just track the original IR Constant mask
operand. I don't completely like the idea of adding another link to
the IR, but MIR is already quite dependent on IR constants already,
and this will allow sharing the shuffle mask utility functions with
the IR.
llvm-svn: 368704
If the target shuffle mask is from a wider type, attempt to scale the mask so that the extraction can attempt to peek through.
Fixes the regression mentioned in rL368662
Reapplying this as rL368308 had to be reverted as part of rL368660 to revert rL368276
llvm-svn: 368663
If we don't demand all elements, then attempt to combine to a simpler shuffle.
At the moment we can only do this if Depth == 0 as combineX86ShufflesRecursively uses Depth to track whether the shuffle has really changed or not - we'll need to change this before we can properly start merging combineX86ShufflesRecursively into SimplifyDemandedVectorElts.
The insertps-combine.ll regression is because XFormVExtractWithShuffleIntoLoad can't see through shuffles of different widths - this will be fixed in a follow-up commit.
Reapplying this as rL368307 had to be reverted as part of rL368660 to revert rL368276
llvm-svn: 368662
This introduced a false positive MemorySanitizer warning about use of
uninitialized memory in a vectorized crc function in Chromium. That suggests
maybe something is not right with this transformation. See
https://crbug.com/992853#c7 for a reproducer.
This also reverts the follow-up commits r368307 and r368308 which
depended on this.
> This patch attempts to peek through vectors based on the demanded bits/elt of a particular ISD::EXTRACT_VECTOR_ELT node, allowing us to avoid dependencies on ops that have no impact on the extract.
>
> In particular this helps remove some unnecessary scalar->vector->scalar patterns.
>
> The wasm shift patterns are annoying - @tlively has indicated that the wasm vector shift codegen are to be refactored in the near-term and isn't considered a major issue.
>
> Differential Revision: https://reviews.llvm.org/D65887
llvm-svn: 368660
The legalizer would hit an assertion on PowerPC platform when truncating
a vector whose size is not power of 2. This patch is to add a check to
prevent vectors with such odd-size elements from being custom lowered.
Reviewed By: Hal Finkel
Differential Revision: https://reviews.llvm.org/D65261
llvm-svn: 368654
Currently we can't keep any state in the selector object that we get from
subtarget. As a result we have to plumb through all our variables through
multiple functions. This change makes it non-const and adds a virtual init()
method to allow further state to be captured for each target.
AArch64 makes use of this in this patch to cache a call to hasFnAttribute()
which is expensive to call, and is used on each selection of G_BRCOND.
Differential Revision: https://reviews.llvm.org/D65984
llvm-svn: 368652
r367088 made it so that funclets store XMM registers into their local
frame instead of storing them to the parent frame. However, that change
forgot to update the parent frame pointer offset for catch blocks. This
change does that.
Fixes crashes when an exception is rethrown in a catch block that saves
XMMs, as described in https://crbug.com/992860.
llvm-svn: 368631
This is infrastructural, will be needed for future work.
For some reason it was only used in MIMG_NoSampler, while
needed everywere we use MIMGBaseOpcode if we want to use
predicates.
Differential Revision: https://reviews.llvm.org/D66115
llvm-svn: 368626
We need AVX512BW to be able to truncate an i16 vector. If we don't
have that we have to extend i16->i32, then trunc, i32->i8. But we
won't be able to remove the min/max if we do that. At least not
without more special handling.
llvm-svn: 368623
If we're after type legalize, we should make sure we won't create
a store with an illegal type when we separate the AVG pattern
from the truncating store.
I don't know of a way to fail for this today. Just noticed while
I was in the vicinity.
llvm-svn: 368608
This teaches the cost model that the sext or zext of a load is going to be
free.
Differential Revision: https://reviews.llvm.org/D66006
llvm-svn: 368593
This pass is a port of the according pass from the HSAIL compiler.
It parses printf calls and setup runtime printf buffer.
After that it copies printf arguments to the buffer and fills in
module metadata for runtime.
Differential Revision: https://reviews.llvm.org/D24035
llvm-svn: 368592
A VDUP will perform a vector broadcast in a single instruction. Update the cost
model for MVE accordingly.
Code originally by David Sherwood.
Differential Revision: https://reviews.llvm.org/D63448
llvm-svn: 368589
This puts some of the calls in ARMTargetTransformInfo.cpp behind hasNeon()
checks, now that we have MVE, and updates all the tests accordingly.
Differential Revision: https://reviews.llvm.org/D63447
llvm-svn: 368587
Summary:
Ana Pazos reported a bug where we were not checking that an APInt would
fit into 64-bits before calling `getSExtValue()`. This caused asserts when
compiling large constants, such as i128s, as happens when compiling compiler-rt.
This patch adds a testcase and makes the callback less error-prone.
Reviewers: apazos, asb, luismarques
Reviewed By: luismarques
Subscribers: hiraditya, rbar, johnrusso, simoncook, sabuasal, niosHD, kito-cheng, shiva0217, jrtc27, MaskRay, zzheng, edward-jones, rogfer01, MartinMosbeck, brucehoult, the_o, rkruppe, PkmX, jocewei, psnobl, benna, Jim, s.egerton, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D66081
llvm-svn: 368572
Support -march=tigerlake for x86.
Compare with Icelake Client, It include 4 more new features ,they are
avx512vp2intersect, movdiri, movdir64b, shstk.
Patch by Xiang Zhang (xiangzhangllvm)
Differential Revision: https://reviews.llvm.org/D65840
llvm-svn: 368543
Due to the nature of the beat system in the MVE architecture, along with tail
predication and low-overhead loops, unrolling has less benefit compared to
normal loops. You can not, for example, hide the latency of a load with other
instructions as you can for scalar code. Preventing unrolling also makes the
code easier to read and reason about.
So if a loop contains vector code, don't enable the runtime unrolling. At least
for the time being.
Differential Revision: https://reviews.llvm.org/D65803
llvm-svn: 368530
With enough codegen complete, we can now correctly report the number and size
of vector registers for MVE, allowing auto vectorisation. This also allows FP
auto-vectorization for MVE without -Ofast/-ffast-math, due to support for IEEE
FP arithmetic and parity between scalar and vector FP behaviour.
Patch by David Sherwood.
Differential Revision: https://reviews.llvm.org/D63728
llvm-svn: 368529
Summary:
When exceptions are repeatedly thrown in the middle of handling another
exception, we call `__clang_call_terminate` with the exception pointer
(i32) as an argument. But in case of foreign exceptions, we don't have
the pointer, so we call the function with 0. (This requires
`__clang_call_terminate` can deal with 0 argument, which will be done
later)
But previously the 0 argument was not added as a `i32.const 0` but an
immediate by mistake, causing the `call` instruction to take not an i32
but rather an exnref, because an `exnref` is left on top of the value
stack if `br_on_exn` is not taken.
```
block i32
br_on_exn 0, __cpp_exception
;; exnref is on top of stack now
i32.const 0 ;; This was missing!
call __clang_call_terminate
unreachable
end
call __clang_call_terminate ;; This takes i32 extracted by br_on_exn
```
Reviewers: dschuff
Subscribers: sbc100, jgravelle-google, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65475
llvm-svn: 368527
The test case that changed is probably better served through
allowing combineTruncatedArithmetic to create narrow vectors. It
also appears InstCombine would have simplified this test case
to remove the zext and trunc anyway.
llvm-svn: 368522
On SSE41+ targets we always lower vector shuffles to ZERO_EXTEND_VECTOR_INREG, even if we don't need the extended bits.
This patch relaxes this so that we lower to ANY_EXTEND_VECTOR_INREG if we can, meaning that shuffle combines have a better idea of what elements need to be kept zero. This helps the multiple reduction code as we can now combine away a lot more of the pack+extend codes.
Differential Revision: https://reviews.llvm.org/D65741
llvm-svn: 368515
Summary:
This patch adds a special DAG combine for SSE1 to recognize the IR pattern InstCombine gives us for movmsk. This only does the recognition for a few cases where its obvious the input won't be scalarized resulting in building a vector just do to the movmsk. I've made it separate from our existing matching for movmsk since that's called in multiple places and I didn't spend time to see if the other callers would make sense here. Plus the restrictions and additional checks would complicate that.
This fixes the case from PR42870. Buts its probably still broken the presence of logic ops feeding the movmsk pattern which would further hide the v4f32 type.
Reviewers: spatel, RKSimon, xbolva00
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65689
llvm-svn: 368506
Summary:
On windows if the frame size exceed 4096 bytes, compiler need to
generate a call to _alloca_probe. X86CallFrameOptimization pass
changes the reserved stack size and cause of stack probe function
not be inserted. This patch fix the issue by detecting the call
frame size, if the size exceed 4096 bytes, drop X86CallFrameOptimization.
Reviewers: craig.topper, wxiao3, annita.zhang, rnk, RKSimon
Reviewed By: rnk
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65923
llvm-svn: 368503
Summary:
Targets often have instructions that can sign-extend certain cases faster
than the equivalent shift-left/arithmetic-shift-right. Such cases can be
identified by matching a shift-left/shift-right pair but there are some
issues with this in the context of combines. For example, suppose you can
sign-extend 8-bit up to 32-bit with a target extend instruction.
%1:_(s32) = G_SHL %0:_(s32), i32 24 # (I've inlined the G_CONSTANT for brevity)
%2:_(s32) = G_ASHR %1:_(s32), i32 24
%3:_(s32) = G_ASHR %2:_(s32), i32 1
would reasonably combine to:
%1:_(s32) = G_SHL %0:_(s32), i32 24
%2:_(s32) = G_ASHR %1:_(s32), i32 25
which no longer matches the special case. If your shifts and extend are
equal cost, this would break even as a pair of shifts but if your shift is
more expensive than the extend then it's cheaper as:
%2:_(s32) = G_SEXT_INREG %0:_(s32), i32 8
%3:_(s32) = G_ASHR %2:_(s32), i32 1
It's possible to match the shift-pair in ISel and emit an extend and ashr.
However, this is far from the only way to break this shift pair and make
it hard to match the extends. Another example is that with the right
known-zeros, this:
%1:_(s32) = G_SHL %0:_(s32), i32 24
%2:_(s32) = G_ASHR %1:_(s32), i32 24
%3:_(s32) = G_MUL %2:_(s32), i32 2
can become:
%1:_(s32) = G_SHL %0:_(s32), i32 24
%2:_(s32) = G_ASHR %1:_(s32), i32 23
All upstream targets have been configured to lower it to the current
G_SHL,G_ASHR pair but will likely want to make it legal in some cases to
handle their faster cases.
To follow-up: Provide a way to legalize based on the constant. At the
moment, I'm thinking that the best way to achieve this is to provide the
MI in LegalityQuery but that opens the door to breaking core principles
of the legalizer (legality is not context sensitive). That said, it's
worth noting that looking at other instructions and acting on that
information doesn't violate this principle in itself. It's only a
violation if, at the end of legalization, a pass that checks legality
without being able to see the context would say an instruction might not be
legal. That's a fairly subtle distinction so to give a concrete example,
saying %2 in:
%1 = G_CONSTANT 16
%2 = G_SEXT_INREG %0, %1
is legal is in violation of that principle if the legality of %2 depends
on %1 being constant and/or being 16. However, legalizing to either:
%2 = G_SEXT_INREG %0, 16
or:
%1 = G_CONSTANT 16
%2:_(s32) = G_SHL %0, %1
%3:_(s32) = G_ASHR %2, %1
depending on whether %1 is constant and 16 does not violate that principle
since both outputs are genuinely legal.
Reviewers: bogner, aditya_nandakumar, volkan, aemerson, paquette, arsenm
Subscribers: sdardis, jvesely, wdng, nhaehnle, rovka, kristof.beyls, javed.absar, hiraditya, jrtc27, atanasyan, Petar.Avramovic, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D61289
llvm-svn: 368487
As discussed on PR42825, if we are inverting the selection mask we can just swap the inputs and avoid the inversion.
Differential Revision: https://reviews.llvm.org/D65522
llvm-svn: 368438
Fast-isel was picking AFGR64 register class for processing call
arguments when +fp64 options was used. We simply check is option +fp64
is used and pick appropriate register.
Patch by Mirko Brkusanin.
Differential Revision: https://reviews.llvm.org/D65886
llvm-svn: 368433
All TLS access on Darwin is in the "general dynamic" form where we call
a function to resolve the address, so implementation is pretty simple.
llvm-svn: 368418
I've now needed to add an extra parameter to this call twice recently. Not only
is the signature getting extremely unwieldy, but just updating all of the
callsites and implementations is a pain. Putting the parameters in a struct
sidesteps both issues.
llvm-svn: 368408
As loads are combined and widened, we replaced their sext users
operands whereas we should have been replacing the uses of the sext.
I've added a load of tests, with only a few of them originally
causing assertion failures, the rest improve pattern coverage.
Differential Revision: https://reviews.llvm.org/D65740
llvm-svn: 368404
The only way to generate these was through promoting legalization
of narrow vectors, but we widen those types now. So we shouldn't
produce these nodes.
llvm-svn: 368396
Under this configuration we'll want to split the v8i64 or v16i32 into two vectors. The default legalization will try to truncate each of those 256-bit pieces one step to 128-bit, concatenate those, then truncate one more time from the new 256 to 128 bits.
With this patch we now truncate the two splits to 64-bits then concatenate those. We have to do this two different ways depending on whether have widening legalization enabled. Without widening legalization we have to manually construct X86ISD::VTRUNC to prevent the ISD::TRUNCATE with a narrow result being promoted to 128 bits with a larger element type than what we want followed by something like a pshufb to grab the lower half of each element to finish the job. With widening legalization we just get the right thing. When we switch to widening by default we can just delete the other code path.
Differential Revision: https://reviews.llvm.org/D65626
llvm-svn: 368349
This fixes znver1 so that it properly enables CMPXHG8B. We can
probably remove explicit CMPXCHG8B from CPUs that also have
CMPXCHG16B, but keeping this simple to allow cherry pick to 9.0.
Fixes PR42935.
llvm-svn: 368324
Summary:
The A64 assembly language does not require the '#' character to
introduce constant immediate operands. Avoid the '#' since the AArch64
asm parser does not accept '#' before the lane specifier and rejects the
following:
__asm__ ("fmla v2.4s, v0.4s, v1.s[%0]" :: "I"(0x1))
Fix a test to not expect the '#' and add a new test case with the above
asm.
Fixes: https://github.com/android-ndk/ndk/issues/1036
Reviewers: peter.smith, kristof.beyls
Subscribers: javed.absar, hiraditya, llvm-commits, srhines
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65550
llvm-svn: 368320
If the target shuffle mask is from a wider type, attempt to scale the mask so that the extraction can attempt to peek through.
Fixes the regression mentioned in rL368307
llvm-svn: 368308
If we don't demand all elements, then attempt to combine to a simpler shuffle.
At the moment we can only do this if Depth == 0 as combineX86ShufflesRecursively uses Depth to track whether the shuffle has really changed or not - we'll need to change this before we can properly start merging combineX86ShufflesRecursively into SimplifyDemandedVectorElts.
The insertps-combine.ll regression is because XFormVExtractWithShuffleIntoLoad can't see through shuffles of different widths - this will be fixed in a follow-up commit.
llvm-svn: 368307
Summary:
This patch enable assembly output of local commons for AIX using .lcomm
directives. Adds a EmitXCOFFLocalCommonSymbol to MCStreamer so we can emit the
AIX version of .lcomm assembly directives which include a csect name. Handle the
case of BSS locals in PPCAIXAsmPrinter by using EmitXCOFFLocalCommonSymbol. Adds
a test for generating .lcomm on AIX Targets.
Reviewers: cebowleratibm, hubert.reinterpretcast, Xiangling_L, jasonliu, sfertile
Reviewed By: sfertile
Subscribers: wuzish, nemanjai, hiraditya, kbarton, MaskRay, jsji, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64825
llvm-svn: 368306
This adds pre- and post- increment and decrements for MVE loads and stores. It
uses the builtin pre and post load/store detection, unlike Neon. Loads are
selected with the code in tryT2IndexedLoad, stores are selected with tablegen
patterns. The immediates have a +/-7bit range, multiplied by the size of the
element.
Differential Revision: https://reviews.llvm.org/D63840
llvm-svn: 368305
This adds some missing patterns for big endian loads/stores, allowing unaligned
loads/stores to also be selected with an extra VREV, which produces better code
than aligning through a stack. Also moves VLDR_P0 to not be LE only, and
adjusts some of the tests to show all that working.
Differential Revision: https://reviews.llvm.org/D65583
llvm-svn: 368304
Summary:
Clang will replace references to registers using ABI names in inline
assembly constraints with references to architecture names, but other
frontends do not. LLVM uses the regular assembly parser to parse inline asm,
so inline assembly strings can contain references to registers using their
ABI names.
This patch adds support for parsing constraints using either the ABI name or
the architectural register name. This means we do not need to implement the
ABI name replacement code in every single frontend, especially those like
Rust which are a very thin shim on top of LLVM IR's inline asm, and that
constraints can more closely match the assembly strings they refer to.
Reviewers: asb, simoncook
Reviewed By: simoncook
Subscribers: hiraditya, rbar, johnrusso, JDevlieghere, apazos, sabuasal, niosHD, kito-cheng, shiva0217, jrtc27, MaskRay, zzheng, edward-jones, rogfer01, MartinMosbeck, brucehoult, the_o, rkruppe, PkmX, jocewei, psnobl, benna, Jim, s.egerton, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65947
llvm-svn: 368303
Summary:
Currently the RISC-V backend does not realign the stack. This can be an issue even for the RV32I/RV64I ABIs (where the stack is 16-byte aligned), though is rare. It will be much more comment with RV32E (though the alignment requirements for common data types remain under-documented...).
This patch adds minimal support for stack realignment. It should cope with large realignments. It will error out if the stack needs realignment and variable sized objects are present.
It feels like a lot of the code like getFrameIndexReference and determineFrameLayout could be refactored somehow, as right now it feels fiddly and brittle. We also seem to allocate a lot more memory than GCC does for equivalent C code.
Reviewers: asb
Reviewed By: asb
Subscribers: wwei, jrtc27, s.egerton, MaskRay, Jim, lenary, hiraditya, rbar, johnrusso, simoncook, apazos, sabuasal, niosHD, kito-cheng, shiva0217, zzheng, edward-jones, rogfer01, MartinMosbeck, brucehoult, the_o, rkruppe, PkmX, jocewei, psnobl, benna, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62007
llvm-svn: 368300
G_JUMP_TABLE and G_BRJT appear from translation of switch statement.
Select these two instructions for MIPS32, both pic and non-pic.
Differential Revision: https://reviews.llvm.org/D65861
llvm-svn: 368274
VLDRH needs to have an alignment of at least 2, including the
widening/narrowing versions. This tightens up the ISel patterns for it and
alters allowsMisalignedMemoryAccesses so that unaligned accesses are expanded
through the stack. It also fixed some incorrect shift amounts, which seemed to
be passing a multiple not a shift.
Differential Revision: https://reviews.llvm.org/D65580
llvm-svn: 368256
If we're splitting the 512-bit vector anyway and we have zero/sign bits, then we might as well use pack instructions to concat and truncate at once.
Differential Revision: https://reviews.llvm.org/D65904
llvm-svn: 368210
The assert that caused this to be reverted should be fixed now.
Original commit message:
This patch changes our defualt legalization behavior for 16, 32, and
64 bit vectors with i8/i16/i32/i64 scalar types from promotion to
widening. For example, v8i8 will now be widened to v16i8 instead of
promoted to v8i16. This keeps the elements widths the same and pads
with undef elements. We believe this is a better legalization strategy.
But it carries some issues due to the fragmented vector ISA. For
example, i8 shifts and multiplies get widened and then later have
to be promoted/split into vXi16 vectors.
This has the potential to cause regressions so we wanted to get
it in early in the 10.0 cycle so we have plenty of time to
address them.
Next steps will be to merge tests that explicitly test the command
line option. And then we can remove the option and its associated
code.
llvm-svn: 368183
Prevent the LoadStoreOptimizer from pairing any load/store instructions with
instructions from the prologue/epilogue if the CFI information has encoded the
operations as separate instructions. This would otherwise lead to a mismatch
of the actual prologue size from the size as recorded in the Windows CFI.
Reviewers: efriedma, mstorsjo, ssijaric
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D65817
llvm-svn: 368164
Function MipsAsmParser::expandMemInst() did not properly handle
instruction `sc` with a symbol as an argument because first argument
would be counted twice. We add additional checks and handle this case
separately.
Patch by Mirko Brkusanin.
Differential Revision: https://reviews.llvm.org/D64252
llvm-svn: 368160
Currently we check whether LR is stored/loaded to/from inbetween the
loop decrement and loop end pseudo instructions. There's two problems
here:
- It relies on all load/store instructions being labelled as such in
tablegen.
- Actually any use of loop decrement is troublesome because the value
doesn't exist!
So we need to check for any read/write of LR that occurs between the
two instructions and revert if we find anything.
Differential Revision: https://reviews.llvm.org/D65792
llvm-svn: 368130
We have aliases that disambiguate memory forms of bt/btc/btr/bts
without suffixes to the 32-bit form. These aliases should have
been updated when the instructions were updated in r356413.
llvm-svn: 368127
The upper 4 bits of the immediate byte are used to encode a
register. We need to limit the explicit immediate to fit in the
remaining 4 bits.
Fixes PR42899.
llvm-svn: 368123
If we're after type legalization we should only be trying to turn
v2i64 into v2i32. So bitcast to v4i32, shuffle the even elements
together. Then use X86ISD::CVTSI2P. The alternative is to leave
the v2i64 type alone and let it scalarized. Hopefully keeping
it packed is better.
Fixes PR42905.
llvm-svn: 368091
https://reviews.llvm.org/D65698
This adds a KnownBits analysis pass for GISel. This was done as a
pass (compared to static functions) so that we can add other features
such as caching queries(within a pass and across passes) in the future.
This patch only adds the basic pass boiler plate, and implements a lazy
non caching knownbits implementation (ported from SelectionDAG). I've
also hooked up the AArch64PreLegalizerCombiner pass to use this - there
should be no compile time regression as the analysis is lazy.
llvm-svn: 368065
Summary:
This is tested by D61289 but has been pulled into a separate patch at
a reviewers request.
Reviewers: bogner, aditya_nandakumar, volkan, aemerson, paquette, arsenm, rovka
Reviewed By: arsenm
Subscribers: javed.absar, hiraditya, wdng, kristof.beyls, Petar.Avramovic, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D61321
llvm-svn: 368063
Summary:
Cleans X86.td's Barcelona entry to be more like the others,
by moving the features out of the `Proc<>`, thus potentially
making it possible to inherit from them.
Split off from D63628
Reviewers: craig.topper, RKSimon
Reviewed By: craig.topper
Subscribers: hiraditya, jfb, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65791
llvm-svn: 368061
Refactor emitFrameOffset to accept a StackOffset struct as its offset argument.
This method currently only supports byte offsets (MVT::i8) but will be extended
in a later patch to support scalable offsets (MVT::nxv1i8) as well.
Reviewers: thegameg, rovka, t.p.northover, efriedma, greened
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D61436
llvm-svn: 368049
To support spilling/filling of scalable vectors we need a more generic
representation of a stack offset than simply 'int'.
For this we introduce the StackOffset struct, which comprises multiple
offsets sized by their respective MVTs. Byte-offsets will thus be a simple
tuple such as { offset, MVT::i8 }. Adding two byte-offsets will result in a
byte offset { offsetA + offsetB, MVT::i8 }. When two offsets have different
types, we can canonicalise them to use the same MVT, as long as their
runtime sizes are guaranteed to have the same size-ratio as they would have
at compile-time.
When we have both scalable- and fixed-size objects on the stack, we can
create an offset that is:
({ offset_fixed, MVT::i8 } + { offset_scalable, MVT::nxv1i8 })
The struct also contains a getForFrameOffset() method that is specific to
AArch64 and decomposes the frame-offset to be used directly in instructions
that operate on the stack or index into the stack.
Note: This patch adds StackOffset as an AArch64-only concept, but we would
like to make this a generic concept/struct that is supported by all
interfaces that take or return stack offsets (currently as 'int'). Since
that would be a bigger change that is currently pending on D32530 landing,
we thought it makes sense to first show/prove the concept in the AArch64
target before proposing to roll this out further.
Reviewers: thegameg, rovka, t.p.northover, efriedma, greened
Reviewed By: rovka, greened
Differential Revision: https://reviews.llvm.org/D61435
llvm-svn: 368024
If we don't demand any non-undef shuffle elements then the assert will fail as all shuffle inputs would still be flagged as 'identity' safe.
Exposed by an incoming patch.
llvm-svn: 368022
Summary:
Before this patch MGATHER/MSCATTER is capable of representing all
common addressing modes, but only when illegal types are used.
This patch adds an IndexType property so more representations
are available when using legal types only.
Original modes:
vector of bases
base + vector of signed scaled offsets
New modes:
base + vector of signed unscaled offsets
base + vector of unsigned scaled offsets
base + vector of unsigned unscaled offsets
The current behaviour of addressing modes for gather/scatter remains
unchanged.
Patch by Paul Walker.
Reviewed By: craig.topper
Differential Revision: https://reviews.llvm.org/D65636
llvm-svn: 368008
Summary:
When the WebAssembly backend encounters a return type that doesn't
fit within i32, SelectionDAG performs sret demotion, adding an
additional argument to the start of the function that contains
a pointer to an sret buffer to use instead. However, this conflicts
with the emscripten sjlj lowering pass. There we translate calls like:
```
call {i32, i32} @foo()
```
into (in pseudo-llvm)
```
%addr = @foo
call {i32, i32} @__invoke_{i32,i32}(%addr)
```
i.e. we perform an indirect call through an extra function.
However, the sret transform now transforms this into
the equivalent of
```
%addr = @foo
%sret = alloca {i32, i32}
call {i32, i32} @__invoke_{i32,i32}(%sret, %addr)
```
(while simultaneously translation the implementation of @foo as well).
Unfortunately, this doesn't work out. The __invoke_ ABI expected
the function address to be the first argument, causing crashes.
There is several possible ways to fix this:
1. Implementing the sret rewrite at the IR level as well and performing
it as part of lowering to __invoke
2. Fixing the wasm backend to recognize that __invoke has a special ABI
3. A change to the binaryen/emscripten ABI to recognize this situation
This revision implements the middle option, teaching the backend to
treat __invoke_ functions specially in sret lowering. This is achieved
by
1) Introducing a new CallingConv ID for invoke functions
2) When this CallingConv ID is seen in the backend and the first argument
is marked as sret (a function pointer would never be marked as sret),
swapping the first two arguments.
Reviewed By: tlively, aheejin
Differential Revision: https://reviews.llvm.org/D65463
llvm-svn: 367935
FastISel already does this since the initial arm64 port was upstreamed, so
it seems there are no issues with doing this at -O0 for very small memcpys.
Gives a 0.2% geomean code size improvement on CTMark.
Differential Revision: https://reviews.llvm.org/D65758
llvm-svn: 367919
This patch changes our defualt legalization behavior for 16, 32, and
64 bit vectors with i8/i16/i32/i64 scalar types from promotion to
widening. For example, v8i8 will now be widened to v16i8 instead of
promoted to v8i16. This keeps the elements widths the same and pads
with undef elements. We believe this is a better legalization strategy.
But it carries some issues due to the fragmented vector ISA. For
example, i8 shifts and multiplies get widened and then later have
to be promoted/split into vXi16 vectors.
This has the potential to cause regressions so we wanted to get
it in early in the 10.0 cycle so we have plenty of time to
address them.
Next steps will be to merge tests that explicitly test the command
line option. And then we can remove the option and its associated
code.
llvm-svn: 367901
Patch D56593 by @courbet results in calls to `bcmp()` in some cases, should
the target support the it. Unless `TTI::MemCmpExpansionOptions()`
is overridden by the target.
In a proprietary benchmark we see a performance drop of about 12% on PNG
compression before this patch, though it passes all tests.
This patch mirrors X86 for AArch64 and initializes
`TTI::MemCmpExpansionOptions()` to then expand calls to `bcmp()` when
appropriate. No tuning of the parameters was performed, but, at this point,
it's enough to recover the performance drop above.
This problem also exists on ARM. Once a consensus is reached for AArch64, we
can work to fix ARM as well.
Authors:
- Evandro Menezes (@evandro) <e.menezes@samsung.com>
- Brian Rzycki (@brzycki) <b.rzycki@samsung.com>
Differential revision: https://reviews.llvm.org/D64805
llvm-svn: 367898
Summary:
The Arm Neoverse N1 Software Optimization Guide [1], Section "4.8 Branch
instruction alignment" states:
"Consider aligning subroutine entry points and branch targets to 32B
boundaries, within the bounds of the code-density requirements of the
program."
This patch sets the preferred function alignment on Neoverse N1 to 2^4=16B.
This was already the case in some of the latest Cortex-A CPUs. Benchmarking
in previous Cortex-A CPUs suggested that 16B alignment is already better
than the default. See commit d04ee305.
The reason we don't set it to 32B right now (as the optimisation guide
suggests) is that this will impact code size and perhaps the instruction
cache performance. Therefore we need benchmark numbers first.
I have also added testing for A75 and A76 that we were missing.
[1] https://developer.arm.com/docs/swog309707/latest
Reviewers: fhahn, greened, samparker, dmgreen
Reviewed By: dmgreen
Subscribers: dmgreen, javed.absar, kristof.beyls, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65654
llvm-svn: 367894
Summary:
This is a follow up to r367237. MachineFunction::getMachineMemOperand()
adds the offset parameter to the existing offset instead of resetting it.
So we need to reset the offset to the correct value after calling this
function.
Reviewers: arsenm
Subscribers: kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65557
llvm-svn: 367881
This was switching to use a format store for a non-format store for
f16 types. Also fixes i16/f16 stores on targets without legal f16.
The corresponding loads also need to be fixed.
llvm-svn: 367872
Without context we assume SGPR. Allowing VGPR constants theoretically
helps avoid a copy. This seems to not actually work now, and the
choice isn't based on the use bank.
llvm-svn: 367871
I'm not sure what complications these present, but the current
argument lowering is pretty much directly copied from the DAG
lowering, so I assume these work as they should.
No tests because I'm lazy and things are getting pretty close to the
point where the existing calling-conventions.ll can be shared with
SelectionDAG.
llvm-svn: 367870
Summary:
This patch adds initial support for the SVE calling convention such that
SVE types can be passed as arguments and return values to/from a
subroutine.
The SVE AAPCS states [1]:
z0-z7 are used to pass scalable vector arguments to a subroutine,
and to return scalable vector results from a function. If a
subroutine takes arguments in scalable vector or predicate
registers, or if it is a function that returns results in such
registers, it must ensure that the entire contents of z8-z23 are
preserved across the call. In other cases it need only preserve the
low 64 bits of z8-z15, as described in §5.1.2.
p0-p3 are used to pass scalable predicate arguments to a subroutine
and to return scalable predicate results from a function. If a
subroutine takes arguments in scalable vector or predicate
registers, or if it is a function that returns results in these
registers, it must ensure that p4-p15 are preserved across the call.
In other cases it need not preserve any scalable predicate register
contents.
SVE predicate and data registers are passed indirectly (i.e. spilled to the
stack and pass the address) if they exceed the registers used for argument
passing defined by the PCS referenced above. Until SVE stack support is merged
we can't spill SVE registers to the stack, so currently an llvm_unreachable is
used where we will eventually handle this.
[1] https://static.docs.arm.com/100986/0000/100986_0000.pdf
Reviewed By: ostannard
Differential Revision: https://reviews.llvm.org/D65448
llvm-svn: 367859
The test case is based on the example from the post-commit thread for:
https://reviews.llvm.org/rGc9171bd0a955
This replaces the x86-specific simple-type check from:
rL367766
with a check in the DAGCombiner. Adding the check isn't
strictly necessary after the fix from:
rL367768
...but it seems likely that we're heading for trouble if
we are creating weird types in this transform.
I combined the earlier legality check into the initial
clause to simplify the code.
So we should only try the trunc/sext transform at the
earliest combine stage, but we limit the transform to
simple types anyway because the TLI hook is probably
too lax about what it considers a free truncate.
llvm-svn: 367834
We process 2 elements at a time and expect the number of elements to be
even. Similar to D60690.
Reviewers: dmgreen, samparker, t.p.northover
Reviewed By: dmgreen
Differential Revision: https://reviews.llvm.org/D65400
llvm-svn: 367831
Summary:
This is patch is part of a serie to introduce an Alignment type.
See this thread for context: http://lists.llvm.org/pipermail/llvm-dev/2019-July/133851.html
See this patch for the introduction of the type: https://reviews.llvm.org/D64790
Reviewers: courbet, jfb, jakehehrlich
Reviewed By: jfb
Subscribers: wuzish, jholewinski, arsenm, dschuff, nemanjai, jvesely, nhaehnle, javed.absar, sbc100, jgravelle-google, hiraditya, aheejin, kbarton, asb, rbar, johnrusso, simoncook, apazos, sabuasal, niosHD, jrtc27, MaskRay, zzheng, edward-jones, rogfer01, MartinMosbeck, brucehoult, the_o, dexonsmith, PkmX, jocewei, jsji, s.egerton, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65514
llvm-svn: 367828
Add an explicit construction of the ArrayRef, gcc 5 and earlier don't
seem to select the ArrayRef constructor which takes a C array when the
construction is implicit.
Original commit message:
- Avoid a crash when IPRA calls ARMFrameLowering::determineCalleeSaves
with a null RegScavenger. Simply not updating the register scavenger
is fine because IPRA only cares about the SavedRegs vector, the acutal
code of the function has already been generated at this point.
- Add a new hook to TargetRegisterInfo to get the set of registers which
can be clobbered inside a call, even if the compiler can see both
sides, by linker-generated code.
Differential revision: https://reviews.llvm.org/D64908
llvm-svn: 367819
Summary:
The SimplifyDemandedVectorElts function can replace with undef
when no elements are demanded, but due to how it interacts with
TargetLoweringOpts, it can only do this when the node has
no other users.
Remove a now unneeded DAG combine from the X86 backend.
Reviewers: RKSimon, spatel
Reviewed By: RKSimon
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65713
llvm-svn: 367788
This adds big endian MVE patterns for bitcasts. They are defined in llvm as
being the same as a store of the existing type and the load into the new. This
means that they have to become a VREV between the two types, working in the
same way that NEON works in big-endian. This also adds some example tests for
bigendian, showing where code is and isn't different.
The main difference, especially from a testing perspective is that vectors are
passed as v2f64, and so are VREV into and out of call arguments, and the
parameters are passed in a v2f64 format. Same happens for inline assembly where
the register class is used, so it is VREV to a v16i8.
So some of this is probably not correct yet, but it is (mostly) self-consistent
and seems to be consistent with how llvm treats vectors. The rest we can
hopefully fix later. More details about big endian neon can be found in
https://llvm.org/docs/BigEndianNEON.html.
Differential Revision: https://reviews.llvm.org/D65581
llvm-svn: 367780
This avoids the crash from:
https://bugs.llvm.org/show_bug.cgi?id=42880
...and I think it's a proper constraint for the TLI hook.
But that example raises questions about what happens to get us
into this situation (created i29 types) and what happens later
(why does legalization die on those types), so I'm not sure if
we will resolve the bug based on this change.
llvm-svn: 367766
Summary:
The allocsize attribute refers to call parameters by index.
Thus, when we add the extra parameter in sjlj lowering, we
need to increment the referenced paramater in the allocsize
attribute to avoid angering the Verifier.
Reviewed By: aheejin
Differential Revision: https://reviews.llvm.org/D65470
llvm-svn: 367765
Fix for https://bugs.llvm.org/show_bug.cgi?id=42760. A tBR_JTr
instruction is duplicated by tail duplication, which results in
the same jumptable with the same label being emitted twice.
Fix this by marking tBR_JTr as not duplicable. The corresponding
ARM/Thumb instructions are already marked as not duplicable.
Additionally also mark tTBB_JT and tTBH_JT to be consistent with
Thumb2, even though this shouldn't be strictly necessary.
Differential Revision: https://reviews.llvm.org/D65606
llvm-svn: 367753
This is an old commit that exposed a bug in the GISel importer, which caused
non-truncating stores to be selected for truncating store patterns. Now that's
been fixed in r367737 this can go back in.
llvm-svn: 367739
This is consistent with the target independent intrinsic handling.
Not sure this really matters since we just pull the constant out
using getZExtValue later.
llvm-svn: 367736
With newly added debuginfo type
metadata for preserve_array_access_index() intrinsic,
this patch did the following two things:
(1). checking validity before adding a new access index
to the access chain.
(2). calculating access byte offset in IR phase
BPFAbstractMemberAccess instead of when BTF is emitted.
For (1), the metadata provided by all preserve_*_access_index()
intrinsics are used to check whether the to-be-added type
is a proper struct/union member or array element.
For (2), with all available metadata, calculating access byte
offset becomes easier in BPFAbstractMemberAccess IR phase.
This enables us to remove the unnecessary complexity in
BTFDebug.cpp.
New tests are added for
. user explicit casting to array/structure/union
. global variable (or its dereference) as the source of base
. multi demensional arrays
. array access given a base pointer
. cases where we won't generate relocation if we cannot find
type name.
Differential Revision: https://reviews.llvm.org/D65618
llvm-svn: 367735
These cases can come up when the extending loads combiner doesn't combine a
zext(load) to a zextload op, due to some other operation being in between, which
then gets simplified at a later stage.
Differential Revision: https://reviews.llvm.org/D65360
llvm-svn: 367723
Summary:
As part of this, define DenseMapInfo for MCRegister (and Register while I'm at it)
Depends on D65599
Reviewers: arsenm
Subscribers: MatzeB, qcolombet, jvesely, wdng, nhaehnle, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65605
llvm-svn: 367719
Add an equivalent ComplexRendererFns function for SelectNegArithImmed. This
allows us to select immediate adds of -1 by turning them into subtracts.
Update select-binop.mir to show that the pattern works.
Differential Revision: https://reviews.llvm.org/D65460
llvm-svn: 367700
This optimisation isn't generally profitable for ARM, because we can
save/restore many registers in the prologue and epilogue using the PUSH
and POP instructions, but mostly use individual LDR/STR instructions for
other spills.
Differential revision: https://reviews.llvm.org/D64910
llvm-svn: 367670
- Avoid a crash when IPRA calls ARMFrameLowering::determineCalleeSaves
with a null RegScavenger. Simply not updating the register scavenger
is fine because IPRA only cares about the SavedRegs vector, the acutal
code of the function has already been generated at this point.
- Add a new hook to TargetRegisterInfo to get the set of registers which
can be clobbered inside a call, even if the compiler can see both
sides, by linker-generated code.
Differential revision: https://reviews.llvm.org/D64908
llvm-svn: 367669
Summary:
When combining `extsw` and `sldi` in `PPCMIPeephole`, we have to check
if `extsw`'s second operand is a virtual register, otherwise we might
get miscompile.
Differential Revision: https://reviews.llvm.org/D65315
llvm-svn: 367645
Summary:
Fixes: https://bugs.llvm.org/show_bug.cgi?id=42441
Used to print:
<unknown>:0: error: Cannot represent a difference across sections
(the location was null).
Now prints:
err.s:20:3: error: Cannot represent a difference across sections
i32.const foo-bar
^
Note: I looked at adding a test for this, but I don't think it is
worth it. We're not testing error formatting in the Wasm backend :)
Reviewers: sbc100, jgravelle-google
Subscribers: dschuff, aheejin, sunfish, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65602
llvm-svn: 367619
AMDGPU sometimes has legal s16 and <2 x s16> operations, but all
registers are really 32-bit. An unmerge destination really should ben
widened to a 32-bit register. If widening a scalarizing vector with a
target size that matches the vector size, bitcast to integer and
extract the relevant bits with shifts.
I'm not sure if this is the right place for this. This could arguably
be part of widenScalar for the result. I also have a growing feeling
that we're missing a bitcast legalize action.
llvm-svn: 367604
If a type is larger than a legal type and needs to be split, we would previously allow the multiply to be decomposed even if the split multiply is legal. Since the shift + add/sub code would also need to be split, its not any better to decompose it.
This patch figures out what type the mul will eventually be legalized to and then uses that type for the query. I tried just returning false illegal types and letting them get handled after type legalization, but then we can't recognize and i64 constant splat on 32-bit targets since will be destroyed by type legalization. We could special case vectors of i64 to avoid that...
Differential Revision: https://reviews.llvm.org/D65533
llvm-svn: 367601
A TYPE_INDEX operand (as used by call_indirect) used to be represented
by the InstPrinter as a symbol (e.g. .Ltype_index0@TYPE_INDEX) which
was a bit of a mismatch with the WasmObjectWriter which expects an
unnamed symbol, to receive the signature from and then turn into a
reloc.
There was really no good way to round-trip this information. An earlier
version of this patch tried to attach the signature information using
a .functype, but that ran into trouble when the symbol was re-emitted
without a name. Removing the name was a giant hack also.
The current version changes the assembly syntax to have an inline
signature spec for TYPEINDEX operands that is always unnamed, which
is much more elegant both in syntax and in implementation (as now the
assembler is able to follow the same path as the regular backend)
Reviewers: sbc100, dschuff, aheejin, jgravelle-google, sunfish, tlively
Subscribers: arphaman, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64758
llvm-svn: 367590
If an operand of the `lw/sw` instructions is a symbol, these instructions
incorrectly lowered using not-position-independent chain of commands.
For PIC code we should use `lw/addiu` instructions with the `R_MIPS_GOT16`
and `R_MIPS_LO16` relocations respectively. Instead of that LLVM generates
position dependent code with the `R_MIPS_HI16` and `R_MIPS_LO16`
relocations.
This patch provides a fix for the bug by handling PIC case separately in
the `MipsAsmParser::expandMemInst`. The main idea is to generate a chain
of PIC instructions to load a symbol address into a register and then
load the address content.
The fix is not optimal and does not fix all PIC-related problems. This
is a task for subsequent patches.
Differential Revision: https://reviews.llvm.org/D65524
llvm-svn: 367580
This adds SimplifyMultipleUseDemandedBitsForTargetNode X86 support and uses it to allow us to peek through vector insertions to avoid dependencies on entire insertion chains.
llvm-svn: 367570
Summary:
GCC Accepts both (reg) and 0(reg) for atomic instruction memory
operands. These instructions do not allow for an offset in their
encoding, so in the latter case, the 0 is silently dropped.
Due to how we have structured the RISCVAsmParser, the easiest way to add
support for parsing this offset is to add a custom AsmOperand and
parser. This parser drops all the parens, and just keeps the register.
This commit also adds a custom printer for these operands, which matches
the GCC canonical printer, printing both `(a0)` and `0(a0)` as `(a0)`.
Reviewers: asb, lewis-revill
Reviewed By: asb
Subscribers: s.egerton, hiraditya, rbar, johnrusso, simoncook, apazos, sabuasal, niosHD, kito-cheng, shiva0217, jrtc27, MaskRay, zzheng, edward-jones, rogfer01, MartinMosbeck, brucehoult, the_o, rkruppe, jfb, PkmX, jocewei, psnobl, benna, Jim, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65205
llvm-svn: 367553
Summary:
While there is always a `Value::replaceAllUsesWith()`,
sometimes the replacement needs to be conditional.
I have only cleaned a few cases where `replaceUsesWithIf()`
could be used, to both add test coverage,
and show that it is actually useful.
Reviewers: jdoerfert, spatel, RKSimon, craig.topper
Reviewed By: jdoerfert
Subscribers: dschuff, sbc100, jgravelle-google, hiraditya, aheejin, george.burgess.iv, asbirlea, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65528
llvm-svn: 367548
The VREV64 instruction is apparently unpredictable if Qd == Qm, due to the
cross-beat nature of the instruction. This adds an earlyclobber to Qd, which
seems to be the same way we deal with this on other instructions like the
write-back on loads and stores.
Differential Revision: https://reviews.llvm.org/D65502
llvm-svn: 367544
Fix an issue where the compiler still allocates an emergency spill slot even
though it already decided to spill an extra callee-save register to use
as a scratch register.
Reviewers: gberry, thegameg, mstorsjo, t.p.northover
Reviewed By: thegameg
Differential Revision: https://reviews.llvm.org/D65504
llvm-svn: 367540
Fold load/store + G_GEP + G_CONSTANT when
immediate in G_CONSTANT fits into 16 bit signed integer.
Differential Revision: https://reviews.llvm.org/D65507
llvm-svn: 367535
In PowerPC, there is instruction to load vector in big endian element order when it's in little endian target.
So we can combine vector load + reverse into big endian load to eliminate the swap instruction.
Also combine vector reverse + store into big endian store.
Differential Revision: https://reviews.llvm.org/D65063
llvm-svn: 367516
Start migrating to a form that will be compatible with the global isel
emitter. Also should fix some overly lax checks on the memory type,
which allowed mis-selecting some illegal atomics.
llvm-svn: 367506
This is extremely specific, but saves three instructions when it's
legal. I don't think the code can be usefully generalized.
Differential Revision: https://reviews.llvm.org/D65351
llvm-svn: 367492
Thumb1 has very limited immediate modes, so turning an "and" into a
shift can save multiple instructions.
It's possible to simplify the generated code for test2 and test3 in
cmp-and-fold.ll a little more, but I'll implement that as a followup.
Differential Revision: https://reviews.llvm.org/D65175
llvm-svn: 367491
We have custom code that ignores the normal promoting type legalization on less than 128-bit vector types like v4i8 to emit pavgb, paddusb, psubusb since we don't have the equivalent instruction on a larger element type like v4i32. If this operation appears before a store, we can be left with an any_extend_vector_inreg followed by a truncstore after type legalization. When truncstore isn't legal, this will normally be decomposed into shuffles and a non-truncating store. This will then combine away the any_extend_vector_inreg and shuffle leaving just the store. On avx512, truncstore is legal so we don't decompose it and we had no combines to fix it.
This patch adds a new DAG combine to detect this case and emit either an extract_store for 64-bit stoers or a extractelement+store for 32 and 16 bit stores. This makes the avx512 codegen match the avx2 codegen for these situations. I'm restricting to only when -x86-experimental-vector-widening-legalization is false. When we're widening we're not likely to create this any_extend_inreg+truncstore combination. This means we should be able to remove this code when we flip the default. I would like to flip the default soon, but I need to investigate some performance regressions its causing in our branch that I wasn't seeing on trunk.
Differential Revision: https://reviews.llvm.org/D65538
llvm-svn: 367488
Summary:
This will make it possible to improve IPRA by taking into account
register usage in indirect calls.
NFC yet; this is just laying the groundwork to start building
up patches to take advantage of the information for improved register
allocation.
Reviewers: aditya_nandakumar, volkan, qcolombet, arsenm, rovka, aemerson, paquette
Subscribers: sdardis, wdng, javed.absar, hiraditya, jrtc27, atanasyan, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65488
llvm-svn: 367476
This feature instructs the backend to allow locally defined global variable
addresses to contain a pointer tag in bits 56-63 that will be ignored by
the hardware (i.e. TBI), but may be used by an instrumentation pass such
as HWASAN. It works by adding a MOVK instruction to the regular ADRP/ADD
sequence that sets bits 48-63 to the corresponding bits of the global, with
the linker bounds check disabled on the ADRP instruction to prevent the tag
from causing a link failure.
This implementation of the feature omits the MOVK when loading from or storing
to a global, which is sufficient for TBI. If the same approach is extended
to MTE, assuming that 0 is not configured as a catch-all tag, we will most
likely also need the MOVK in this case in order to avoid a tag mismatch.
Differential Revision: https://reviews.llvm.org/D65364
llvm-svn: 367475
This makes the field wider than MachineOperand::SubReg_TargetFlags so that
we don't end up silently truncating any higher bits. We should still catch
any bits truncated from the MachineOperand field as a consequence of the
assertion in MachineOperand::setTargetFlags().
Differential Revision: https://reviews.llvm.org/D65465
llvm-svn: 367474
Summary:
According to the Armv8.1-M manual CSEL, CSINC, CSINV and CSNEG are
"constrained unpredictable" when SP is used as the source register Rn.
The assembler should diagnose this case.
Reviewers: momchil.velikov, dmgreen, ostannard, simon_tatham, t.p.northover
Reviewed By: ostannard
Subscribers: javed.absar, kristof.beyls, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65505
llvm-svn: 367433
Before combining insert_subvector(insert_subvector(vec, sub0, c0), sub1, c1) patterns, ensure that the subvectors are all the same type. On AVX512 targets especially we might have a mixture of 128/256 subvector insertions.
llvm-svn: 367429
Summary:
This adds the 'f' inline assembly constraint, as supported by GCC. An
'f'-constrained operand is passed in a floating point register. Exactly
which kind of floating-point register (32-bit or 64-bit) is decided
based on the operand type and the available standard extensions (-f and
-d, respectively).
This patch adds support in both the clang frontend, and LLVM itself.
Reviewers: asb, lewis-revill
Reviewed By: asb
Subscribers: hiraditya, rbar, johnrusso, simoncook, apazos, sabuasal, niosHD, kito-cheng, shiva0217, jrtc27, MaskRay, zzheng, edward-jones, rogfer01, MartinMosbeck, brucehoult, the_o, rkruppe, PkmX, jocewei, psnobl, benna, Jim, s.egerton, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D65500
llvm-svn: 367403
Use a switch instead of many isa<> while checking for supported
values. Also be explicit about which cast instructions are supported;
This allows the removal of SIToFP from GenerateSignBits.
llvm-svn: 367402
Summary:
* Loads and stores in SVE2 are gather/scatter not contiguous, fixed by
renaming multiclasses to reflect this and also updated comments.
* Remove aliases from load/store multiclasses that reflect the behaviour
of the original form.
* Fix bug in scatter store implementation, vector list should be used as
input, not output.
Reviewed By: sdesmalen
Differential Revision: https://reviews.llvm.org/D65392
llvm-svn: 367398
This adds the required extension to RISC-V's getRegForInlineAsmConstraint
in order to be able to correctly distringuish between the 32 and 64-bit
floating point registers when the generic fX name appears in inlineasm
clobber contraints. It also adds a check to validate that callee saved
floating point registers are only saved in this case when a hard-float
ABI is selected.
Differential Revision: https://reviews.llvm.org/D64751
llvm-svn: 367397
Summary:
* Clarify comment with SVE2 for predicated shifts and move next to other
shift instructions.
* Clarify comments for various instructions.
* Move FCVTX instruction next to other fp conversions.
* Move FLOGB to next to other fp instructions and fix description.
* Remove "cons" from non-constructive multiclass for bitwise shift-right
and accumulate instructions.
Reviewed By: sdesmalen
Differential Revision: https://reviews.llvm.org/D65390
llvm-svn: 367396
Summary:
This patch fixes a bug in the following instructions that should have been
implemented as destructive. A destructive instruction is an instruction where
one of the source registers also acts as the destination register. Therefore,
the contents of the source register, when the instruction begins execution, are
replaced by the result of the instruction when the instruction completes
execution [1]:
* SRI/SLI
* EORBT/EORTB
* TBX
* Narrowing top instructions
* FP convert precision instructions
These changes are non-functional from the assembler/diassembler point-of-view
but are necessary for correct codegen.
[1] https://static.docs.arm.com/ddi0584/ae/DDI0584A_e_SVE_supp_armv8A.pdf
Reviewed By: sdesmalen
Differential Revision: https://reviews.llvm.org/D65389
llvm-svn: 367394
In PowerPC, there is instruction to load vector in big endian element order when it's in little endian target.
So we can combine vector load + reverse into big endian load to eliminate the swap instruction.
Also combine vector reverse + store into big endian store.
llvm-svn: 367382
We had couple places which still return 10 as a maximum
occupancy. Fixed.
Also print comment about occupancy as compiler see it.
Differential Revision: https://reviews.llvm.org/D65423
llvm-svn: 367381
AMDGPU uses some custom code predicates for testing alignments.
I'm still having trouble comprehending the behavior of predicate bits
in the PatFrag hierarchy. Any attempt to abstract these properties
unexpectdly fails to apply them.
llvm-svn: 367373
Summary:
- Use the passed `DL` directly as retrieving data layout from CS by
checking the called function is not reliable. Under indirect function
call, there is no called function.
Subscribers: jholewinski, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65468
llvm-svn: 367349
Summary:
return_call and return_call_indirect are only valid if the return
types of the callee and caller match. We were previously not enforcing
that, which was producing invalid modules.
Reviewers: aheejin
Subscribers: dschuff, sbc100, jgravelle-google, hiraditya, sunfish, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65246
llvm-svn: 367339
Empty condition strings are considerde always true. This removes a lot
of clutter from the generated matcher tables.
This shrinks the source size of AMDGPUGenDAGISel.inc from 7.3M to
6.1M.
llvm-svn: 367326
Addresses number of comment made on D64652 after commiting:
- Reorders function decls in the TargetLoweringObjectFileXCOFF class.
- Fix comment in MCSectionXCOFF to include description of external reference
csects.
- Convert several llvm_unreachables to report_fatal_error
- Convert several dyn_casts to casts as they are expected not to fail.
- Avoid copying DataLayout object.
llvm-svn: 367324
PR42819 showed an issue that we couldn't handle the case where we demanded a 'sub-sub-vector' of the SUBV_BROADCAST 'sub-vector' source.
This patch recognizes these cases and extracts the sub-sub-vector instead of trying to broadcast to a type smaller than the 'sub-vector' source.
llvm-svn: 367306
The code is now in a good enough state to pass the bunch of tests that
I have run (after fixing the bugs), so let's enable it by default.
Differential Revision: https://reviews.llvm.org/D65277
llvm-svn: 367297
Revert the hardware loop upon finding a LoopEnd that doesn't target
the loop header, instead of asserting a failure.
Differential Revision: https://reviews.llvm.org/D65268
llvm-svn: 367296
Summary:
The LoadStoreOptimizer was creating instructions with 2
MachineMemOperands, which meant they were assumed to alias with all other instructions,
because MachineInstr:mayAlias() returns true when an instruction has multiple
MachineMemOperands.
This was preventing these instructions from being merged again, and was
giving the scheduler less freedom to reorder them.
Reviewers: arsenm, nhaehnle
Reviewed By: arsenm
Subscribers: kzhuravl, jvesely, wdng, yaxunl, dstuttard, tpr, t-tye, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65036
llvm-svn: 367237
As discussed on rL367171, we have a problem where the depth recursion used in combineX86ShufflesRecursively was subtly different to computeKnownBits etc. - it starts at Depth=1 instead of Depth=0 like the others and has a different maximum recursion depth.
This NFC patch fixes the recursion depth to start at 0, so we can more easily reuse depth values in calls from combineX86ShufflesRecursively and its helper functions in computeKnownBits etc.
llvm-svn: 367232
For llvm/test/MC/RISCV/rv64i-aliases-invalid.s, UBSan reports:
lib/Target/RISCV/AsmParser/RISCVAsmParser.cpp:371:9: runtime error:
load of value 3879186881, which is not a valid value for type
'RISCVMCExpr::VariantKind'
SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior
lib/Target/RISCV/AsmParser/RISCVAsmParser.cpp:371:9 in
It turns out that evaluateConstantImm does not set `VK` and it remains
unitialized when doing comparisons in `isImmXLenLI()`.
Differential Revision: https://reviews.llvm.org/D65347
llvm-svn: 367230
Summary:
The existing isDivergent(Value) methods query whether a value is
divergent at its definition. However even if a value is uniform at its
definition, a use of it in another basic block can be divergent because
of divergent control flow between the def and the use.
This patch adds new isDivergent(Use) methods to DivergenceAnalysis,
LegacyDivergenceAnalysis and GPUDivergenceAnalysis.
This might allow D63953 or other similar workarounds to be removed.
Reviewers: alex-t, nhaehnle, arsenm, rtaylor, rampitec, simoll, jingyue
Reviewed By: nhaehnle
Subscribers: jfb, jvesely, wdng, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65141
llvm-svn: 367218
- Remove some unused typedefs.
- Rename BinOpChain struct to MulCandidate.
- Remove the size method of MulCandidate.
- Store only the first input of the ValueList provided to
MulCandidate, as it's the only value we care about. This means we
don't have to perform any ugly (and unnecessary) iterations of the
list later on.
llvm-svn: 367208
Summary:
If isel is presented with <2 x half> vectors then it will correctly select
v_pk_fma style instructions.
If isel is presented with e.g. <4 x half> vectors it will scalarize, unlike for
other instruction types (such as fadd, fmul etc.)
Added extra support to enable this. Updated one of the tests to include a test
for this (as well as extending the test to GFX9)
Subscribers: arsenm, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, tpr, t-tye, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D65325
Change-Id: I50a4577a3f8223fb53992af3b7d26121f65b71ee
llvm-svn: 367206
The pmaddwd inserts a truncate, if that truncate would end up
creating additional instructions instead of making a zext
narrower, then we shouldn't do it.
I've restricted this to only sse4.1 targets since on prior
targets the zext will be done in stages. So the truncate will
probably not create additional instructions. Might need some
more investigation of mul shrinking and the other pmaddwd
transform to be sure this is the right decision.
There might be a slight regression on AVX1 targets due to add
splitting. Hard to say for sure. Maybe we need to look into
using the vector reduction flag to use 2 narrow loads and a
blend instead of extracting and inserting.
llvm-svn: 367198
This reverts r340478 and r340631 and replaces them with a simpler
method of just letting DAG combine revisit the nodes to handle
the other operand.
llvm-svn: 367195
This adds the patterns required to transform xor P0, -1 to a VPNOT. The
instruction operands have to change a little for this, adding an in and an out
VCCR reg and using a custom DecodeMVEVPNOT for the decode.
Differential Revision: https://reviews.llvm.org/D65133
llvm-svn: 367192
These are some better patterns for converting between predicates and floating
points. Much like the extends, we select "1"/"-1" or "0" depending on the
predicate value. Or we perform a compare against 0 to convert to a predicate.
Differential Revision: https://reviews.llvm.org/D65103
llvm-svn: 367191
Recommit rL367100 which was reverted at rL367141. Until PR42777 is fixed, we no longer get the benefits of peeking through bitcasts but it does still remove a GetDemandedBits user and gives us the equivalent combines.
llvm-svn: 367172
Add partial instruction selection for intrinsics like this:
```
declare i32 @llvm.aarch64.stlxr(i64, i32*)
```
(This only handles the case where a G_ZEXT is feeding the intrinsic.)
Also make sure that the added store instruction actually has the memory op from
the original G_STORE.
Update select-stlxr-intrin.mir and arm64-ldxr-stxr.ll.
Differential Revision: https://reviews.llvm.org/D65355
llvm-svn: 367163
Adds machine operand lowering for MCSymbolSDNodes to the PowerPC
backend. This is needed to produce call instructions in assembly for AIX
because the callee operand is a MCSymbolSDNode. The test is XFAIL'ed for
asserts due to a (valid) assertion in PEI that the AIX ABI isn't supported yet.
Differential Revision: https://reviews.llvm.org/D63738
llvm-svn: 367133
Summary:
The bitperm feature flag is now prefixed with SVE2, as it is for all other SVE2
extensions
Patch by Maciej Gabka.
Reviewers: sdesmalen, rovka, chill, SjoerdMeijer, rengolin
Reviewed By: SjoerdMeijer, rengolin
Differential Revision: https://reviews.llvm.org/D65327
llvm-svn: 367124
In preperation for AIX support in FrameLowering: replace a number of literal
'8' that represent the stack offset of the condition register save area with
a member in PPCFrameLowering.
Patch by Chris Bowler.
llvm-svn: 367111
Void return used to have unsigned with value 0 for virtual register
but with addition of Register class and changes to arguments to lowerCall
this is no longer valid.
Check for void return by inspecting the Ty field in OrigRet.
Differential Revision: https://reviews.llvm.org/D65321
llvm-svn: 367107
Add llvm.amdgcn.softwqm intrinsic which behaves like llvm.amdgcn.wqm
only if there is other WQM computation in the shader.
Reviewers: nhaehnle, tpr
Reviewed By: nhaehnle
Subscribers: arsenm, kzhuravl, jvesely, wdng, yaxunl, dstuttard, t-tye, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64935
llvm-svn: 367097
Embedded Trace Extension and Trace Buffer Extension are optional
future architecture extensions.
(cf. https://developer.arm.com/architectures/cpu-architecture/a-profile/exploration-tools)
Their system registers are documented here:
https://developer.arm.com/docs/ddi0601/a
ETE shares register names with ETM. One exception is the ETE
TRCEXTINSELR0 register, which has the same encoding as the ETM
TRCEXTINSELR register (but different semantics). This patch treats
them as aliases: the assembler will accept both names, emitting
identical encoding, and the disassembler will keep disassembling
to TRCEXRINSELR.
Differential Revision: https://reviews.llvm.org/D63707
llvm-svn: 367093
Both WhileLoopStart and LoopEnd may get turned into a cmp and br pair,
so add an implicit def to these pseudo instructions in case that WLS
and LE aren't generated.
Differential Revision: https://reviews.llvm.org/D65275
llvm-svn: 367089
Summary:
This is an alternate approach to D57970.
Currently funclets reuse the same stack slots that are used in the
parent function for saving callee-saved xmm registers. If the parent
function modifies a callee-saved xmm register before an excpetion is
thrown, the catch handler will overwrite the original saved value.
This patch allocates space in funclets stack for saving callee-saved xmm
registers and uses RSP instead RBP to access memory.
Reviewers: andrew.w.kaylor, LuoYuanke, annita.zhang, craig.topper,
RKSimon
Subscribers: rnk, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63396
Signed-off-by: pengfei <pengfei.wang@intel.com>
llvm-svn: 367088
handleAssignments gives up pretty easily on structs, and i8 values for
some reason. The other case that doesn't work is when an implicit sret
needs to be inserted if the return size exceeds the number of return
registers.
llvm-svn: 367082
Currently, the CO-RE offset relocation does not work
if any struct/union member or array element is a typedef.
For example,
typedef const int arr_t[7];
struct input {
arr_t a;
};
func(...) {
struct input *in = ...;
... __builtin_preserve_access_index(&in->a[1]) ...
}
The BPF backend calculated default offset is 0 while
4 is the correct answer. Similar issues exist for struct/union
typedef's.
When getting struct/union member or array element type,
we should trace down to the type by skipping typedef
and qualifiers const/volatile as this is what clang did
to generate getelementptr instructions.
(const/volatile member type qualifiers are already
ignored by clang.)
This patch fixed this issue, for each access index,
skipping typedef and const/volatile/restrict BTF types.
Signed-off-by: Yonghong Song <yhs@fb.com>
Differential Revision: https://reviews.llvm.org/D65259
llvm-svn: 367062
Currently, we expect the CO-RE offset relocation records
a string encoding the original getelementptr access index,
so kernel bpf loader can decode it correctly.
For example,
struct s { int a; int b; };
struct t { int c; int d; };
#define _(x) (__builtin_preserve_access_index(x))
int get_value(const void *addr1, const void *addr2);
int test(struct s *arg1, struct t *arg2) {
return get_value(_(&arg1->b), _(&arg2->d));
}
We expect two offset relocations:
reloc 1: type s, access index 0, 1
reloc 2: type t, access index 0, 1
Two globals are created to retain access indexes for the
above two relocations with global variable names.
The first global has a name "0:1:". Unfortunately,
the second global has the name "0:1:.1" as the llvm
internals automatically add suffix ".1" to a global
with the same name. Later on, the BPF peels the last
character and record "0:1" and "0:1:." in the
relocation table.
This is not desirable. BPF backend could use the global
variable suffix knowledge to generate correct access str.
This patch rather took an approach not relying on
that knowledge. It generates "s:0:1:" and "t:0:1:" to
avoid global variable suffixes and later on generate
correct index access string "0:1" for both records.
Signed-off-by: Yonghong Song <yhs@fb.com>
Differential Revision: https://reviews.llvm.org/D65258
llvm-svn: 367030
Summary:
- As LCSSA is turned on just before isel, it may create PHI of the flow,
which is consumed by pseudo structurized CFG instructions. When that
PHIs are eliminated in O0, COPY may be placed wrongly as the these
pseudo structurized CFG instructions are considering prologue of MBB.
- Run extra `unreachable-mbb-elimination` at the end of isel to clean up
PHIs.
Reviewers: arsenm, rampitec
Subscribers: kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64353
llvm-svn: 367023
... for the vector forms of `{SQ,UQ,}{INC,DEC}P` instructions. Also continue
supporting the exsting behaviour of not requiring an explicit size
specifier. The preferred disasembly is *with* the specifier.
This is implemented by redefining intruction forms to require vector predicates
with explicit size and adding aliases, which allow a predicate with no size.
Differential Revision: https://reviews.llvm.org/D65145
llvm-svn: 367019
Summary:
In PostRA phase, we often have to find out the most recent definition
of a register. This patch adds getDefMIPostRA so that other methods
can use it rather than implementing it repeatedly.
Differential Revision: https://reviews.llvm.org/D65131
llvm-svn: 366990
Summary:
Add a new method which tries to compute the target address referenced by an operand.
This patch supports x86_64 RIP-relative addressing for now.
It is necessary to print referenced symbol names in llvm-objdump.
Reviewers: andreadb, MaskRay, grosbach, jgalenson, craig.topper
Reviewed By: MaskRay, craig.topper
Subscribers: bcain, rupprecht, jhenderson, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63847
llvm-svn: 366987
Before, we weren't able to select things like this for G_GEP:
add x0, x8, #8
And instead we'd materialize the 8.
This teaches GISel to do that. It gives some considerable code size savings
on 252.eon-- about 4%!
Differential Revision: https://reviews.llvm.org/D65248
llvm-svn: 366959
Throughout the legalizerinfo we currently make the assumption that the target
has neon and FP target features available. Fixing it will require a refactor of
the whole thing, so until then make sure we fall back.
Works around PR42734
Differential Revision: https://reviews.llvm.org/D65244
llvm-svn: 366957
Summary:
This was originally reported in D62818.
https://rise4fun.com/Alive/oPH
InstCombine does the opposite fold, in hope that `C l>>/<< Y` expression
will be hoisted out of a loop if `Y` is invariant and `X` is not.
But as it is seen from the diffs here, if it didn't get hoisted,
the produced assembly is almost universally worse.
Much like with my recent "hoist add/sub by/from const" patches,
we should get almost universal win if we hoist constant,
there is almost always an "and/test by imm" instruction,
but "shift of imm" not so much, so we may avoid having to
materialize the immediate, and thus need one less register.
And since we now shift not by constant, but by something else,
the live-range of that something else may reduce.
Special care needs to be applied not to disturb x86 `BT` / hexagon `tstbit`
instruction pattern. And to not get into endless combine loop.
Reviewers: RKSimon, efriedma, t.p.northover, craig.topper, spatel, arsenm
Reviewed By: spatel
Subscribers: hiraditya, MaskRay, wuzish, xbolva00, nikic, nemanjai, jvesely, wdng, nhaehnle, javed.absar, tpr, kristof.beyls, jsji, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62871
llvm-svn: 366955
If we have a G_MUL, and either the LHS or the RHS of that mul is the legal
shift value for a load addressing mode, we can fold it into the load.
This gives some code size savings on some SPEC tests. The best are around 2%
on 300.twolf and 3% on 254.gap.
Differential Revision: https://reviews.llvm.org/D65173
llvm-svn: 366954
This introduces a new family of combiner helper routines that re-use the
target specific cost model from SelectionDAG, and generate inline implementations
of the memcpy family of intrinsics.
The combines are only enabled at optimization levels higher than -O0, and give
very substantial performance improvements.
Differential Revision: https://reviews.llvm.org/D65167
llvm-svn: 366951
To support prefetch mode 3 we need to pad current
cacheline and fill 3 cachelines after. Current padding
is only sufficient for mode 2.
Differential Revision: https://reviews.llvm.org/D65236
llvm-svn: 366938
This removes the VCEQ/VCNE/VCGE/VCEQZ/etc nodes, just using two called VCMP and
VCMPZ with an extra operand as the condition code. I believe this will make
some combines simpler, allowing us to just look at these codes and not the
operands. It also helps fill in a missing VCGTUZ MVE selection without adding
extra nodes for it.
Differential Revision: https://reviews.llvm.org/D65072
llvm-svn: 366934
This patch adds support for recognizing cases where a larger vector type is being used to reduce just the elements in the lower subvector:
e.g. <8 x i32> reduction pattern in a <16 x i32> vector:
<4,5,6,7,u,u,u,u,u,u,u,u,u,u,u,u>
<2,3,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
<1,u,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
matchBinOpReduction returns the lower extracted subvector in such cases, assuming isExtractSubvectorCheap accepts the extraction.
I've only enabled it for X86 reduction sums so far. I intend to enable it for the bitop/minmax cases in future patches, and eventually I think its worth turning it on all the time. This is mainly just a case of ensuring calls to matchBinOpReduction don't make assumptions on the vector width based on the original vector extraction.
Fixes the x86 partial reduction sum cases in PR33758 and PR42023.
Differential Revision: https://reviews.llvm.org/D65047
llvm-svn: 366933
The prevents us from trying to convert an i1 predicate vector to a float, or
vice-versa. Better patterns are possible, which will follow in a subsequent
commit. For now we just expand them.
Differential Revision: https://reviews.llvm.org/D65066
llvm-svn: 366931
Fix an off-by-one error which made us not look at the last element of the
zero vector. This caused a miscompile in 188.ammp.
Differential Revision: https://reviews.llvm.org/D65168
llvm-svn: 366930
MVE VCMP instructions can use a general purpose register as the second operand.
This adds the combines for it, selecting from a compare of a vdup.
Differential Revision: https://reviews.llvm.org/D65061
llvm-svn: 366924
This adds a DeMorgan combine for OR's of compares to turn them into AND's,
helping prevent them from going into and out of gpr registers. It also fills in
the VCLE and VCLT nodes that MVE can select, allowing it to invert more
compares.
Differential Revision: https://reviews.llvm.org/D65059
llvm-svn: 366920
Add a number of folds to convert and(vcmp, vcmp) into a single VPT block, where
the second vcmp becomes predicated on the first.
The VCMP; VPST; VCMP will eventually be converted to VPT; VCMP in the
VPTBlockPass.
Differential Revision: https://reviews.llvm.org/D65058
llvm-svn: 366910
Much like integers, this adds MVE floating point compares and select. It
requires a lot more buildvector/shuffle code because we may need to expand the
compares without mve.fp, and requires support for and/or because of the way we
lower llvm condition codes.
Some original code by David Sherwood
Differential Revision: https://reviews.llvm.org/D65054
llvm-svn: 366909
This adds some basic, "worst case" handling for MVE predicate Or/And/Xor. It
does this by going into and out of GPRs, doing the operation on scalars.
Code by David Sherwood.
Differential Revision: https://reviews.llvm.org/D65053
llvm-svn: 366907
This change make sure that llvm does not emit an invalid IT block
by putting the constant pool in the middle of an IT block.
We have code to try to avoid putting a constant island in the middle of an
IT block, but it only works if we see an IT between the one currently
referencing CPE and possible insertion point. If the first instruction
we look at is the VLDRD after the IT , we never see the IT and does not
realize that the instruction doing the load could be in an IT block itself.
Differential Revision: https://reviews.llvm.org/D64621
Change-Id: I24cecb37cded75e8992870bd997f6226853bd920
llvm-svn: 366905
This adds support code for building and shuffling i1 predicate registers. It
generally uses two basic principles, either converting the predicate into an
scalar (through a PREDICATE_CAST) and doing scalar operations on it there, or
by converting the register to an full vector register and back.
Some of the code here is a not super efficient but will hopefully cover most
cases of moving i1 vectors around and can be improved in subsequent patches.
Some code by David Sherwood.
Differential Revision: https://reviews.llvm.org/D65052
llvm-svn: 366890
This adds the very basics for MVE vector predication, adding integer VCMP and
VSEL instruction support. This is done through predicate registers (MVT::v16i1,
MVT::v8i1, MVT::v4i1), but otherwise using same mechanics as NEON to custom
lower setcc's through ARMISD::VCXX nodes (VCEQ, VCGT, VCEQZ, etc).
An extra VCNE was added, as this can be handled sensibly by MVE's expanded
number of VCMP condition codes. (There are also VCLE and VCLT which are added
later).
VPSEL is also added here, simply selecting on the vselect.
Original code by David Sherwood.
Differential Revision: https://reviews.llvm.org/D65051
llvm-svn: 366885
While combining two loads into a single load, we often need to
reorder the pointer operands for the new load. This reordering was
broken in the cases where there was a chain of values that built up
the pointer.
Differential Revision: https://reviews.llvm.org/D65193
llvm-svn: 366881
We need to be able to load and store s128 for memcpy inlining, where we want to
generate Q register mem ops. Making these legal also requires that we add some
support in other instructions. Regbankselect should also know about these since
they have no GPR register class that can hold them, so need special handling to
live on the FPR bank.
Differential Revision: https://reviews.llvm.org/D65166
llvm-svn: 366857
Currently PowerPC backend emits code like this:
r3 = li 0
std r3, 264(r1)
r3 = li 0
std r3, 272(r1)
This patch fixes that and other cases where a register already contains a value that is loaded so we will get:
r3 = li 0
std r3, 264(r1)
std r3, 272(r1)
Differential Revision: https://reviews.llvm.org/D64220
llvm-svn: 366840
LegalizeDAG tries to legal the DAG by legalizing nodes before
their operands.
If we create a new node, we end up legalizing it after its operands.
This prevents some of the optimizations that can be done when the
operand is a build_vector since the build_vector will have been
legalized to something else.
Differential Revision: https://reviews.llvm.org/D65132
llvm-svn: 366835
When we select the XRO variants of loads, we can pull in very specific shifts
(of the size of an element). E.g.
```
ldr x1, [x2, x3, lsl #3]
```
This teaches GISel to handle these when they're coming from shifts
specifically.
This adds a new addressing mode function, `selectAddrModeShiftedExtendXReg`
which recognizes this pattern.
This also packs this up with `selectAddrModeRegisterOffset` into
`selectAddrModeXRO`. This is intended to be equivalent to `selectAddrModeXRO`
in AArch64ISelDAGtoDAG.
Also update load-addressing-modes to show that all of the cases here work.
Differential Revision: https://reviews.llvm.org/D65119
llvm-svn: 366819
While lowering test.set.loop.iterations, it wasn't checked how the
brcond was using the result and so the wls could branch to the loop
preheader instead of not entering it. The same was true for
loop.decrement.reg.
So brcond and br_cc and now lowered manually when using the hwloop
intrinsics. During this we now check whether the result has been
negated and whether we're using SETEQ or SETNE and 0 or 1. We can
then figure out which basic block the WLS and LE should be targeting.
Differential Revision: https://reviews.llvm.org/D64616
llvm-svn: 366809
Replace float load/store pair with integer load/store pair when it's only used in load/store,
because float load/store instructions cost more cycles then integer load/store.
A typical scenario is when there is a call with more than 13 float arguments passing, we need pass them by stack.
So we need a load/store pair to do such memory operation if the variable is global variable.
Differential Revision: https://reviews.llvm.org/D64195
llvm-svn: 366775
The xform has no real valuewhen it's using out of a complex pattern
output. The complex pattern was already creating TargetConstants with
i16, so this was just unnecessary machinery.
This allows global isel to import the simple cases once the complex
pattern is implemented.
llvm-svn: 366743
The build_vector will become a constant pool load. By using the
desired type initially, it ensures we don't generate a bitcast
of the constant pool load which will need to be folded with
the load.
While experimenting with another patch, I noticed that when the
load type and the constant pool type don't match, then
SimplifyDemandedBits can't handle it. While we should probably
fix that, this was a simple way to fix the issue I saw.
llvm-svn: 366732
Summary:
Since we are planning to add ADDIStocHA for 32bit in later patch, we decided
to change 64bit one first to follow naming convention with 8 behind opcode.
Patch by: Xiangling_L
Differential Revision: https://reviews.llvm.org/D64814
llvm-svn: 366731
Stubs out a TargetLoweringObjectFileXCOFF class, implementing only
SelectSectionForGlobal for common symbols. Also adds an override of
EmitGlobalVariable in PPCAIXAsmPrinter which adds a number of defensive errors
and adds support for emitting common globals.
llvm-svn: 366727
ARMLowOverheadLoops would assert a failure if it did not find all the
pseudo instructions that comprise the hardware loop. Instead of doing
this, iterate through all the instructions of the function and revert
any remaining pseudo instructions that haven't been converted.
Differential Revision: https://reviews.llvm.org/D65080
llvm-svn: 366691
We need to ensure that the number of T's is correct when adding multiple
instructions into the same VPT block.
Differential revision: https://reviews.llvm.org/D65049
llvm-svn: 366684
This patch enables us to find the source loads for each element, splitting them into a Load and ByteOffset, and attempts to recognise consecutive loads that are in fact from the same source load.
A helper function, findEltLoadSrc, recurses to find a LoadSDNode and determines the element's byte offset within it. When attempting to match consecutive loads, byte offsetted loads then attempt to matched against a previous load that has already been confirmed to be a consecutive match.
Next step towards PR16739 - after this we just need to account for shuffling/repeated elements to create a vector load + shuffle.
Fixed out of bounds load assert identified in rL366501
Differential Revision: https://reviews.llvm.org/D64551
llvm-svn: 366681
ARM has code to recognise uses of the "returned" function parameter
attribute which guarantee that the value passed to the function in r0
will be returned in r0 unmodified. IPRA replaces the regmask on call
instructions, so needs to be told about this to avoid reverting the
optimisation.
Differential revision: https://reviews.llvm.org/D64986
llvm-svn: 366669
Summary:
In the atomic optimizer, save doing a bunch of work and generating a
bunch of dead IR in the fairly common case where the result of an
atomic op (i.e. the value that was in memory before the atomic op was
performed) is not used. NFC.
Reviewers: arsenm, dstuttard, tpr
Subscribers: kzhuravl, jvesely, wdng, nhaehnle, yaxunl, t-tye, hiraditya, jfb, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64981
llvm-svn: 366667
As detailed on PR42674, we can reduce a vXi8 down until we have the final <8 x i8>, and then use PSADBW with zero, to sum those values. We then extract the bottom i8, discarding any overflow from the upper bits of the i16 result.
llvm-svn: 366636
Sometimes, you can end up with cross-bank copies between same-sized GPRs and
FPRs, which feed into G_STOREs. When these copies feed only into stores, they
aren't necessary; we can just store using the original register bank.
This provides some minor code size savings for some floating point SPEC
benchmarks. (Around 0.2% for 453.povray and 450.soplex)
This issue doesn't seem to show up due to regbankselect or anything similar. So,
this patch introduces an early select function, `contractCrossBankCopyIntoStore`
which performs the contraction when possible. The selector then continues
normally and selects the correct store opcode, eliminating needless copies
along the way.
Differential Revision: https://reviews.llvm.org/D65024
llvm-svn: 366625
Summary:
Add immutable WASM global `__tls_align` which stores the alignment
requirements of the TLS segment.
Add `__builtin_wasm_tls_align()` intrinsic to get this alignment in Clang.
The expected usage has now changed to:
__wasm_init_tls(memalign(__builtin_wasm_tls_align(),
__builtin_wasm_tls_size()));
Reviewers: tlively, aheejin, sbc100, sunfish, alexcrichton
Reviewed By: tlively
Subscribers: dschuff, jgravelle-google, hiraditya, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D65028
llvm-svn: 366624
This change reverts most of the previous register name generation.
The real problem is that RegisterTuple does not generate asm names.
Added optional operand to RegisterTuple. This way we can simplify
register name access and dramatically reduce the size of static
tables for the backend.
Differential Revision: https://reviews.llvm.org/D64967
llvm-svn: 366598
This should now handle everything except structs passed as multiple
registers.
I think most of the packing logic should be handled by
handleAssignments, but I'm unclear on what the contract is for
multiple registers. This is copying how x86 handles this.
This does change the behavior of the test_sgpr_alignment0 amdgpu_vs
test. I don't think shader arguments should try to follow the
alignment, and registers need to be repacked. I also don't think it
matters, since I think the pointers are packed to the beginning of the
argument list anyway.
llvm-svn: 366582
This is the more natural lowering, and presents more opportunities to
reduce 64-bit ops to 32-bit.
This should also help avoid issues graphics shaders have had with
64-bit values, and simplify argument lowering in globalisel.
llvm-svn: 366578
Summary:
For split-stack, if the nested argument (i.e. R10) is not used, no need to save/restore it in the prologue.
Reviewers: thanm
Reviewed By: thanm
Subscribers: mstorsjo, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64673
llvm-svn: 366569
We'd like to remove this whole function, because these are properties of
functions, not the target as a whole. These two are easy to remove
because they are only used for emitting ARM build attributes, which
expects them to represent the defaults for the whole module, not just
the last function generated.
This is needed to get correct build attributes when using IPRA on ARM,
because IPRA causes resetTargetOptions to get called before
ARMAsmPrinter::emitAttributes.
Differential revision: https://reviews.llvm.org/D64929
llvm-svn: 366562
Summary:
According to the new Armv8-M specification
https://static.docs.arm.com/ddi0553/bh/DDI0553B_h_armv8m_arm.pdf the
instructions SQRSHRL and UQRSHLL now have an additional immediate
operand <saturate>. The new assembly syntax is:
SQRSHRL<c> RdaLo, RdaHi, #<saturate>, Rm
UQRSHLL<c> RdaLo, RdaHi, #<saturate>, Rm
where <saturate> can be either 64 (the existing behavior) or 48, in
that case the result is saturated to 48 bits.
The new operand is encoded as follows:
#64 Encoded as sat = 0
#48 Encoded as sat = 1
sat is bit 7 of the instruction bit pattern.
This patch adds a new assembler operand class MveSaturateOperand which
implements parsing and encoding. Decoding is implemented in
DecodeMVEOverlappingLongShift.
Reviewers: ostannard, simon_tatham, t.p.northover, samparker, dmgreen, SjoerdMeijer
Reviewed By: simon_tatham
Subscribers: javed.absar, kristof.beyls, hiraditya, pbarrio, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64810
llvm-svn: 366555
It is necessary to generate fixups in .debug_frame or .eh_frame as
relaxation is enabled due to the address delta may be changed after
relaxation.
There is an opcode with 6-bits data in debug frame encoding. So, we
also need 6-bits fixup types.
Differential Revision: https://reviews.llvm.org/D58335
llvm-svn: 366524
I plan on adding memcpy optimizations in the GlobalISel pipeline, but we can't
do that unless we delay lowering to actual function calls. This patch changes
the translator to generate G_INTRINSIC_W_SIDE_EFFECTS for these functions, and
then have each target specify that using the new custom legalizer for intrinsics
hook that they want it expanded it a libcall.
Differential Revision: https://reviews.llvm.org/D64895
llvm-svn: 366516
Add support for folding G_GEPs into loads of the form
```
ldr reg, [base, off]
```
when possible. This can save an add before the load. Currently, this is only
supported for loads of 64 bits into 64 bit registers.
Add a new addressing mode function, `selectAddrModeRegisterOffset` which
performs this folding when it is profitable.
Also add a test for addressing modes for G_LOAD.
Differential Revision: https://reviews.llvm.org/D64944
llvm-svn: 366503
Summary:
Add `__builtin_wasm_tls_base` so that LeakSanitizer can find the thread-local
block and scan through it for memory leaks.
Reviewers: tlively, aheejin, sbc100
Subscribers: dschuff, jgravelle-google, hiraditya, sunfish, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D64900
llvm-svn: 366475
There doesn't seem to be a practical reason for these instructions to have
different restrictions on the types of relocations that they may be used
with, notwithstanding the language in the ELF AArch64 spec that implies that
specific relocations are meant to be used with specific instructions.
For example, we currently forbid the first instruction in the following
sequence, despite it currently being used by clang to generate a global
reference under -mcmodel=large:
movz x0, #:abs_g0_nc:foo
movk x0, #:abs_g1_nc:foo
movk x0, #:abs_g2_nc:foo
movk x0, #:abs_g3:foo
Therefore, allow MOVK/MOVN/MOVZ to accept the union of the set of relocations
that they currently accept individually.
Differential Revision: https://reviews.llvm.org/D64466
llvm-svn: 366461
It is necessary to generate fixups in .debug_frame or .eh_frame as
relaxation is enabled due to the address delta may be changed after
relaxation.
There is an opcode with 6-bits data in debug frame encoding. So, we
also need 6-bits fixup types.
Differential Revision: https://reviews.llvm.org/D58335
llvm-svn: 366442
This patch enables us to find the source loads for each element, splitting them into a Load and ByteOffset, and attempts to recognise consecutive loads that are in fact from the same source load.
A helper function, findEltLoadSrc, recurses to find a LoadSDNode and determines the element's byte offset within it. When attempting to match consecutive loads, byte offsetted loads then attempt to matched against a previous load that has already been confirmed to be a consecutive match.
Next step towards PR16739 - after this we just need to account for shuffling/repeated elements to create a vector load + shuffle.
Differential Revision: https://reviews.llvm.org/D64551
llvm-svn: 366441
LEA doesn't affect flags, so use it more liberally to replace an ADD when
we know that the ADD operands affect flags.
In the motivating example from PR40483:
https://bugs.llvm.org/show_bug.cgi?id=40483
...this lets us avoid duplicating a math op just to avoid flag conflict.
As mentioned in the TODO comments, this heuristic can be extended to
fire more often if that leads to more improvements.
Differential Revision: https://reviews.llvm.org/D64707
llvm-svn: 366431
Summary:
PerformVMOVRRDCombine ommits adding a offset
of 4 to the PointerInfo, when converting a
f64 = load[M]
to
{i32, i32} = {load[M], load[M + 4]}
Which would allow the machine scheduller
to break dependencies with the second load.
- pr42638
Reviewers: eli.friedman, dmgreen, ostannard
Reviewed By: ostannard
Subscribers: ostannard, javed.absar, kristof.beyls, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64870
llvm-svn: 366423
I'm not convinced the code this calls is properly vetted for
vXi1 vectors. Experimental vector widening legalization testing
for D55251 is now hitting an assertion failure inside
EltsFromConsecutiveLoads. This is occurring from a v2i1 load
having a store size different than its VT size. Hopefully
this commit will keep such issues from happening.
llvm-svn: 366405
Implement IR intrinsics for stack tagging. Generated code is very
unoptimized for now.
Two special intrinsics, llvm.aarch64.irg.sp and llvm.aarch64.tagp are
used to implement a tagged stack frame pointer in a virtual register.
Differential Revision: https://reviews.llvm.org/D64172
llvm-svn: 366360
Summary:
Since the target has no significant advantage of vectorization,
vector instructions bous threshold bonus should be optional.
amdgpu-inline-arg-alloca-cost parameter default value and the target
InliningThresholdMultiplier value tuned then respectively.
Reviewers: arsenm, rampitec
Subscribers: kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, eraman, hiraditya, haicheng, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64642
llvm-svn: 366348
The original behavior was to always emit the offsets to each call site in the
call site table as uleb128 values, however on some architectures (eg RISCV)
these uleb128 offsets into the code cannot always be resolved until link time
(because relaxation will invalidate any calculated offsets), and there are no
appropriate relocations for uleb128 values. As a consequence it needs to be
possible to specify an alternative.
This also switches RISCV to use DW_EH_PE_udata4 for call side encodings in
.gcc_except_table
Differential Revision: https://reviews.llvm.org/D63415
Patch by Edward Jones.
llvm-svn: 366329
Summary:
Missed in the original commit, use the correct callee-saved register
list for spilling, instead of the standard SVR432 list. This avoids
needlessly spilling the SPE non-volatile registers when they're not used.
As part of this, also add where missing, and sort, the spill opcode
checks for SPE and SPE4 register classes.
Reviewers: nemanjai, hfinkel, joerg
Subscribers: kbarton, jsji, llvm-commits
Differential Revision: https://reviews.llvm.org/D56703
llvm-svn: 366319
Summary:
Pointed out in a comment for D49754, register spilling will currently
spill SPE registers at almost any offset. However, the instructions
`evstdd` and `evldd` require a) 8-byte alignment, and b) a limit of 256
(unsigned) bytes from the base register, as the offset must fix into a
5-bit offset, which ranges from 0-31 (indexed in double-words).
The update to the register spill test is taken partially from the test
case shown in D49754.
Additionally, pointed out by Kei Thomsen, globals will currently use
evldd/evstdd, though the offset isn't known at compile time, so may
exceed the 8-bit (unsigned) offset permitted. This fixes that as well,
by forcing it to always use evlddx/evstddx when accessing globals.
Part of the patch contributed by Kei Thomsen.
Reviewers: nemanjai, hfinkel, joerg
Subscribers: kbarton, jsji, llvm-commits
Differential Revision: https://reviews.llvm.org/D54409
llvm-svn: 366318
Add narrowScalar to half of original size for G_ICMP.
ClampScalar G_ICMP's operands 2 and 3 to to s32.
Select G_ICMP for pointers for MIPS32. Pointer compare is same
as for integers, it is enough to declare them as legal type.
Differential Revision: https://reviews.llvm.org/D64856
llvm-svn: 366317
Migrate CallLowering::lowerReturnVal to use the same infrastructure as
lowerCall/FormalArguments and remove the now obsolete code path from
splitToValueTypes.
Forgot to push this earlier.
llvm-svn: 366308
This directive forces to use the alternate register for context pointer.
For example, this code:
.cplocal $4
jal foo
expands to:
ld $25, %call16(foo)($4)
jalr $25
Differential Revision: https://reviews.llvm.org/D64743
llvm-svn: 366300
As well as other LLVM targets we do not handle "offsettable"
memory addresses in any special way. In other words, the "o" constraint
is an exact equivalent of the "m" one. But some existing code require
the "o" constraint support.
This fixes PR42589.
Differential Revision: https://reviews.llvm.org/D64792
llvm-svn: 366299
Summary:
Currently, on Emscripten, dynamic linking is not supported with threads.
This means that if thread-local storage is used, it must be used in a
statically-linked executable. Hence, local-exec is the only possible model.
This diff compiles all TLS variables to use local-exec on Emscripten as a
temporary measure until dynamic linking is supported with threads.
The goal for this is to allow C++ types with constructors to be thread-local.
Currently, when `clang` compiles a `thread_local` variable with a constructor,
it generates `__tls_guard` variable:
@__tls_guard = internal thread_local global i8 0, align 1
As no TLS model is specified, this is treated as general-dynamic, which we do
not support (and cannot support without implementing dynamic linking support
with threads in Emscripten). As a result, any C++ constructor in `thread_local`
variables would not compile.
By compiling all `thread_local` as local-exec, `__tls_guard` will compile and
we can support C++ constructors with TLS without implementing dynamic linking
with threads.
Depends on D64537
Reviewers: tlively, aheejin, sbc100
Reviewed By: aheejin
Subscribers: dschuff, jgravelle-google, hiraditya, sunfish, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64776
llvm-svn: 366275
Summary:
Thread local variables are placed inside a `.tdata` segment. Their symbols are
offsets from the start of the segment. The address of a thread local variable
is computed as `__tls_base` + the offset from the start of the segment.
`.tdata` segment is a passive segment and `memory.init` is used once per thread
to initialize the thread local storage.
`__tls_base` is a wasm global. Since each thread has its own wasm instance,
it is effectively thread local. Currently, `__tls_base` must be initialized
at thread startup, and so cannot be used with dynamic libraries.
`__tls_base` is to be initialized with a new linker-synthesized function,
`__wasm_init_tls`, which takes as an argument a block of memory to use as the
storage for thread locals. It then initializes the block of memory and sets
`__tls_base`. As `__wasm_init_tls` will handle the memory initialization,
the memory does not have to be zeroed.
To help allocating memory for thread-local storage, a new compiler intrinsic
is introduced: `__builtin_wasm_tls_size()`. This instrinsic function returns
the size of the thread-local storage for the current function.
The expected usage is to run something like the following upon thread startup:
__wasm_init_tls(malloc(__builtin_wasm_tls_size()));
Reviewers: tlively, aheejin, kripken, sbc100
Subscribers: dschuff, jgravelle-google, hiraditya, sunfish, jfb, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D64537
llvm-svn: 366272
This is part of what is requested by PR42023:
https://bugs.llvm.org/show_bug.cgi?id=42023
There's an extension needed for FP add, but exactly how we would specify
that using flags is not clear to me, so I left that as a TODO.
We're still missing patterns for partial reductions when the input vector
is 256-bit or 512-bit, but I think that's a failure of vector narrowing.
If we can reduce the widths, then this matching should work on those tests.
Differential Revision: https://reviews.llvm.org/D64760
llvm-svn: 366268
Summary:
This is exposed by our internal testing.
The reduced testcase will assert with "Impossible reg-to-reg copy"
We can't use COPY to do 32-bit to 64-bit conversion.
Reviewers: kbarton, hfinkel, nemanjai
Reviewed By: hfinkel
Subscribers: hiraditya, MaskRay, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64499
llvm-svn: 366255
I think this manages to not break the DAG handling with the divergent
predicates because the stadalone divergent patterns end up with a
higher priority than the pattern on the instruction definition.
The 16-bit versions don't work yet.
llvm-svn: 366254
When it is AReg_1024 this results in unnecessary copying into
AGPRs of a 32 element vectors even though they are not intended
for an mfma instruction.
Differential Revision: https://reviews.llvm.org/D64815
llvm-svn: 366252
Now that the patterns use the new PatFrag address space support, the
only blocker to importing most load patterns is the addressing mode
complex patterns.
llvm-svn: 366237
Summary:
Extend the atomic optimizer to handle signed and unsigned max and min
operations, as well as add and subtract.
Reviewers: arsenm, sheredom, critson, rampitec
Subscribers: kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, jfb, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64328
llvm-svn: 366235
Apparently the check for legal instructions during instruction
select does not happen without an asserts build, so these would
successfully select in release, and fail in debug.
Make s16 and/or/xor legal. These can just be selected directly
to the 32-bit operation, as is already done in SelectionDAG, so just
make them legal.
llvm-svn: 366210
The jcvt intrinsic defined in ACLE [1] is available when ARM_FEATURE_JCVT is defined.
This change introduces the AArch64 intrinsic, wires it up to the instruction and a new clang builtin function.
The __ARM_FEATURE_JCVT macro is now defined when an Armv8.3-A or higher target is used.
I've implemented the target detection logic in Clang so that this feature is enabled for architectures from armv8.3-a onwards (so -march=armv8.4-a also enables this, for example).
make check-all didn't show any new failures.
[1] https://developer.arm.com/docs/101028/latest/data-processing-intrinsics
Differential Revision: https://reviews.llvm.org/D64495
llvm-svn: 366197
The canonical GNU form of JALR resembles a load/store instruction rather
than placing the immediate offset as a separate argument, so match this
behaviour. Also add parser-only aliases for the three-operand form, and
add other shorter aliases also emitted by GNU tools.
Differential Revision: https://reviews.llvm.org/D55277
Patch by James Clarke.
llvm-svn: 366179
RISCVAsmBackend::shouldInsertExtraNopBytesForCodeAlign() assumed that the
align specified would be greater than or equal to the minimum nop length, but
that is not always the case - for example if a user specifies ".align 0" in
assembly.
Differential Revision: https://reviews.llvm.org/D63274
Patch by Edward Jones.
llvm-svn: 366176
The bool result of shouldInsertExtraNopBytesForCodeAlign() is not checked but
the returned nop count is unconditionally read even though it could be
uninitialized.
Differential Revision: https://reviews.llvm.org/D63285
Patch by Edward Jones.
llvm-svn: 366175
Since PseudoCALL defines AsmString, it can be generated from assembly,
and so code-gen patterns should be defined separately to be consistent
with the style of the RISCV backend. Other pseudo-instructions exist
that have code-gen patterns defined directly, but these instructions are
purely for code-gen and cannot be written in assembly.
Differential Revision: https://reviews.llvm.org/D64012
Patch by James Clarke.
llvm-svn: 366174
Previously, this function didn't check the IsPCRel argument. But doing so is a
useful check for errors, and also seemingly necessary for FK_Data_4 (which we
produce a R_RISCV_32_PCREL relocation for if IsPCRel).
Other than R_RISCV_32_PCREL, this should be NFC. Future exception handling
related patches will include tests that capture this behaviour.
llvm-svn: 366172
Use the MemoryVT field. This will be necessary for tablegen to
automatically handle patterns for GlobalISel.
Doesn't handle the d16 lo/hi patterns. Those are a special case since
it involvess the custom node type.
llvm-svn: 366168
We mostly avoid sub with immediate but there are a couple cases that can create them. One is the add 128, %rax -> sub -128, %rax trick in isel. The other is when a SUB immediate gets created for a compare where both the flags and the subtract value is used. If we are unable to linearize the SelectionDAG to satisfy the flag user and the sub result user from the same instruction, we will clone the sub immediate for the two uses. The one that produces flags will eventually become a compare. The other will have its flag output dead, and could then be considered for LEA creation.
I added additional test cases to add.ll to show the the sub -128 trick gets converted to LEA and a case where we don't need to convert it.
This showed up in the current codegen for PR42571.
Differential Revision: https://reviews.llvm.org/D64574
llvm-svn: 366151
Summary:
This adds missing utility methods and copy instruction handling for
`exnref` type and also adds tests.
`tee` instruction tests are missing because `isTee` is currently only
used in ExplicitLocals pass and testing that pass in mir requires
serialization of stackified registers in mir files, which is a bit
nontrivial because `MachineFunctionInfo` only has info of vreg numbers
(which are large integers) but not the mir's register numbers. But this
change is quite trivial anyway.
Reviewers: tlively
Subscribers: dschuff, sbc100, jgravelle-google, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64705
llvm-svn: 366149
Summary:
We agreed to rename `except_ref` to `exnref` for consistency with other
reference types in
https://github.com/WebAssembly/exception-handling/issues/79. This also
renames WebAssemblyInstrExceptRef.td to WebAssemblyInstrRef.td in order
to use the file for other reference types in future.
Reviewers: dschuff
Subscribers: sbc100, jgravelle-google, hiraditya, sunfish, jfb, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64703
llvm-svn: 366145
Summary:
These are emitted as identifiers by the InstPrinter, so we should
parse them as such. These could potentially clash with symbols of
the same name, but that is out of our (the WebAssembly backend) control.
Reviewers: dschuff
Subscribers: sbc100, jgravelle-google, aheejin, sunfish, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64770
llvm-svn: 366139
Summary:
Enable hoisting and merging m0 defs that are initialized with the same
immediate value. Fixes bug where removed instructions are not considered
to interfere with other inits, and make sure to not hoist inits before block
prologues.
Reviewers: rampitec, arsenm
Reviewed By: rampitec
Subscribers: kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64766
llvm-svn: 366135
We already do this for the flat and DS instructions, although it is
certainly uglier and more verbose.
This will allow using separate pattern definitions for extload and
zextload. Currently we get away with using a single PatFrag with
custom predicate code to check if the extension type is a zextload or
anyextload. The generic mechanism the global isel emitter understands
treats these as mutually exclusive. I was considering making the
pattern emitter accept zextload or sextload extensions for anyextload
patterns, but in global isel, the different extending loads have
distinct opcodes, and there is currently no mechanism for an opcode
matcher to try multiple (and there probably is very little need for
one beyond this case).
llvm-svn: 366132
If a 1-bit value is in a 32-bit VGPR, the scalar opcodes set SCC to
whether the result is 0. If the inputs are SCC, these can be copied to
a 32-bit SGPR to produce an SCC result.
llvm-svn: 366125
This is a hack until I come up with a better way of dealing with the
pseudo-register banks used for boolean values. If the use instruction
constrains the register, the selector for the def instruction won't
see that the bank was VCC. A 1-bit SReg_32 is could ambiguously have
been SCCRegBank or VCCRegBank in wave32.
This is necessary to successfully select branches with and and/or/xor
condition.
llvm-svn: 366120
The extra test change is correct, although how it arrives there is a
bug that needs work. With wave32, the test for isVCC ambiguously
reports true for an SCC or VCC source. A new allocatable pseudo
register class for SCC may be necesssary.
llvm-svn: 366119
We need to make sure that we are sensibly dealing with vectors of types v2i64
and v2f64, even if most of the time we cannot generate native operations for
them. This mostly adds a lot of testing, plus fixes up a couple of the issues
found. And, or and xor can be legal for v2i64, and shifts combining needs a
slight fixup.
Differential Revision: https://reviews.llvm.org/D64316
llvm-svn: 366106
inttofp (trunc (extelt X, 0)) --> inttofp (extelt (bitcast X), 0)
We have pseudo-vectorization of scalar int to FP casts, so this tries to
make that more likely by replacing a truncate with a bitcast. I didn't see
any test diffs starting from 'uitofp', so I left that as a TODO. We can't
only match the shorter trunc+extract pattern because there's an opposing
transform somewhere, so we infinite loop. Waiting to try this during
lowering is another possibility.
A motivating case is shown in PR39975 and included in the test diffs here:
https://bugs.llvm.org/show_bug.cgi?id=39975
Differential Revision: https://reviews.llvm.org/D64710
llvm-svn: 366098
I think we only turn out of range shiftss to undef when
all elements are out of range or the shift amount is a splat out
of range. I'm not sure which, I didn't check.
During lowering we can split a shift where some elements
are out of range into multiple shifts. This can create a
new shift with a splat shift amount that is out of range.
This patch returns undef for this case.
Fixes PR42615.
Differential Revision: https://reviews.llvm.org/D64699
llvm-svn: 366096
Insert these during codegenprepare.
This works around a DAG issue where generic combines eliminate the and
asserting the high bits are zero, which then exposes an unknown read
source to the mul combine. It doesn't worth the hassle of trying to
insert an AssertZext or something to try to deal with it.
llvm-svn: 366094
This adds basic lowering for MVE shifts. There are many shifts in MVE, but the
instructions handled here are:
VSHL (imm)
VSHRu (imm)
VSHRs (imm)
VSHL (vector)
VSHL (register)
MVE, like NEON before it, doesn't have shift right by a vector (or register).
We instead have to negate the amount and shift in the opposite direction. This
means we have to convert any SHR's into a form of SHL (that is still signed or
unsigned) with a negated condition and selecting from there. MVE still does
have shifting by an immediate for SHL, ASR and LSR.
This adds lowering for these and for register forms, which work well for shift
lefts but may require an extra fold of neg(vdup(x)) -> vdup(neg(x)) to potentially
work optimally for right shifts.
Differential Revision: https://reviews.llvm.org/D64212
llvm-svn: 366056
This just moves the shift instruction definitions further down the
ARMInstrMVE.td file, to make positioning patterns slightly more natural.
llvm-svn: 366054
This adjusts the way that we lower NEON shifts to use a DAG target node, not
via a neon intrinsic. This is useful for handling MVE shifts operations in the
same the way. It also renames some of the immediate shift nodes for
consistency, and moves some of the processing of immediate shifts into
LowerShift allowing it to capture more cases.
Differential Revision: https://reviews.llvm.org/D64426
llvm-svn: 366051
Summary:
SSE1 only supports v4f32. But does have instructions like movlps/movhps that load/store 64-bits of memory.
This patch breaks the connection between the node VT of the vzext_load/vextract_store patterns and the memory VT. Enabling a v4f32 node with a 64-bit memory VT. I've used i64 as the memory VT here. I've written the PatFrag predicate to just check the store size not the specific VT. I think the VT will only matter for CSE purposes. We could use v2f32, but if we want to start using these operations in more places a simple integer type might make the most sense.
I'd like to maybe use this same thing for SSE2 and later as well, but that will need more work to be supported by EltsFromConsecutiveLoads to avoid regressing lit tests. I'd maybe also like to combine bitcasts with these load/stores nodes now that the types are disconnected. And I'd also like to consider canonicalizing (scalar_to_vector + load) to vzext_load.
If you want I can split the mechanical tablegen stuff where I added the 32/64 off from the sse1 change.
Reviewers: spatel, RKSimon
Reviewed By: RKSimon
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64528
llvm-svn: 366034
The vmovlb instructions can be uses to sign or zero extend vector registers
between types. This adds some patterns for them and relevant testing. The
VBICIMM generation is also put behind a hasNEON check (as is already done for
VORRIMM).
Code originally by David Sherwood.
Differential Revision: https://reviews.llvm.org/D64069
llvm-svn: 366008
This selects integer VNEG instructions, which can be especially useful with shifts.
Differential Revision: https://reviews.llvm.org/D64204
llvm-svn: 366006
This simply makes the MVE integer min and max instructions legal and adds the
relevant patterns for them.
Differential Revision: https://reviews.llvm.org/D64026
llvm-svn: 366004
This adds support for the floor/ceil/trunc/... series of instructions,
converting to various forms of VRINT. They use the same suffixes as their
floating point counterparts. There is not VTINTR, so nearbyint is expanded.
Also added a copysign test, to show it is expanded.
Differential Revision: https://reviews.llvm.org/D63985
llvm-svn: 366003
This adds the patterns for minnm and maxnm from the fminnum and fmaxnum nodes,
similar to scalar types.
Original patch by Simon Tatham
Differential Revision: https://reviews.llvm.org/D63870
llvm-svn: 366002
These should really use v32f32, but were defined as v32i32
due to the lack of the v32f32 type.
Differential Revision: https://reviews.llvm.org/D64667
llvm-svn: 365972
Summary:
Most of these functions can work for MachineInstr and MCInst
equally now.
Reviewers: dschuff
Subscribers: MatzeB, sbc100, jgravelle-google, aheejin, sunfish, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64643
llvm-svn: 365965
Split AArch64FrameLowering::resolveFrameIndexReference in two parts
* Finding frame offset for the index.
* Finding base register and offset to that register.
The second part will be used to implement a virtual frame pointer in
armv8.5 MTE stack instrumentation lowering.
Reviewers: pcc, vitalybuka, hctim, ostannard
Subscribers: javed.absar, kristof.beyls, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64171
llvm-svn: 365958
Before 2018, mesa used to use byval interchangably with inreg, which
didn't really make sense. Fix tests still using it to avoid breaking
in a future commit.
llvm-svn: 365953
This fixes https://bugs.llvm.org/show_bug.cgi?id=42606 by extending
D64213. Instead of only checking if the carry comes from a matching
operation, we now check the full chain of carries. Otherwise we might
custom lower the outermost addcarry, but then generically legalize
an inner addcarry.
Differential Revision: https://reviews.llvm.org/D64658
llvm-svn: 365949
This patch series adds support for the next-generation arch13
CPU architecture to the SystemZ backend.
This includes:
- Basic support for the new processor and its features.
- Assembler/disassembler support for new instructions.
- CodeGen for new instructions, including new LLVM intrinsics.
- Scheduler description for the new processor.
- Detection of arch13 as host processor.
Note: No currently available Z system supports the arch13
architecture. Once new systems become available, the
official system name will be added as supported -march name.
llvm-svn: 365932
We can use the C flag from NEG to detect that the input was zero.
Really we could probably use the Z flag too. But C matches what
we'd do for usubo 0, X.
Haven't found a test case for this due to the usubo formation
in CGP. But I verified if I comment out the CGP code this
transformation catches some of the same cases.
llvm-svn: 365929
Summary:
r363675 changed the exec modification helper function, now called
execMayBeModifiedBeforeUse, so that if no UseMI is specified it checks
all instructions in the basic block, even beyond the last use. That
meant that the DPP combiner no longer worked in any basic block that
ended with a control flow instruction, and in particular it didn't work
on code sequences generated by the atomic optimizer.
Fix it by reinstating the old behaviour but in a new helper function
execMayBeModifiedBeforeAnyUse, and limiting the number of instructions
scanned.
Reviewers: arsenm, vpykhtin
Subscribers: kzhuravl, nemanjai, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, kbarton, MaskRay, jfb, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64393
llvm-svn: 365910
Summary:
D64497 allowed abs/neg source modifiers on v_cndmask_b32 but it doesn't
make any sense to apply them to f16 operands; they would interpret the
bits of the value as an f32, giving nonsensical results. This patch
restricts them to f32 operands.
Reviewers: arsenm, hakzsam
Subscribers: kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64636
llvm-svn: 365904
Summary:
Useful for jumps, such as `j .`.
I am not sure who should review this. Do not hesitate to change the reviewers if needed.
Reviewers: asb, jrtc27, lenary
Reviewed By: lenary
Subscribers: MaskRay, lenary, hiraditya, rbar, johnrusso, simoncook, apazos, sabuasal, niosHD, kito-cheng, shiva0217, zzheng, edward-jones, rogfer01, MartinMosbeck, brucehoult, the_o, rkruppe, PkmX, jocewei, psnobl, benna, Jim, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D63669
Patch by John LLVM (JohnLLVM)
llvm-svn: 365881
There is not match for the `MipsJmpLink texternalsym` and `MipsJmpLink
tglobaladdr` patterns for microMIPS R6. As a result LLVM incorrectly
selects the `JALRC16` compact 2-byte instruction which takes a target
instruction address from a register only and assign `R_MIPS_32` relocation
for this instruction. This relocation completely overwrites `JALRC16`
and nearby instructions.
This patch adds missed matching patterns, selects `BALC` instruction and
assign a correct `R_MICROMIPS_PC26_S1` relocation.
Differential Revision: https://reviews.llvm.org/D64552
llvm-svn: 365870
Follow up to D58597, where it was noted that the commuted ISD::SUB variant
was having problems with lack of combines.
See also D63958 where we untangled setcc/sub pairs.
Differential Revision: https://reviews.llvm.org/D58875
llvm-svn: 365791
Summary:
As of binutils 2.32, ld has a bogus TLS relaxation error when the GD/LD
code sequence using R_X86_64_GOTPCREL (instead of R_X86_64_GOTPCRELX) is
attempted to be relaxed to IE/LE (binutils PR24784). gold and lld are good.
In gcc/config/i386/i386.md, there is a configure-time check of as/ld
support and the GOT relaxation will not be used if as/ld doesn't support
it:
if (flag_plt || !HAVE_AS_IX86_TLS_GET_ADDR_GOT)
return "call\t%P2";
return "call\t{*%p2@GOT(%1)|[DWORD PTR %p2@GOT[%1]]}";
In clang, -DENABLE_X86_RELAX_RELOCATIONS=OFF is the default. The ld.bfd
bogus error can be reproduced with:
thread_local int a;
int main() { return a; }
clang -fno-plt -fpic a.cc -fuse-ld=bfd
GOTPCRELX gained relative good support in 2016, which is considered
relatively new. It is even difficult to conditionally default to
-DENABLE_X86_RELAX_RELOCATIONS=ON due to cross compilation reasons. So
work around the ld.bfd bug by only using GOT when GOTPCRELX is enabled.
Reviewers: dalias, hjl.tools, nikic, rnk
Reviewed By: nikic
Subscribers: llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64304
llvm-svn: 365752
This patch addresses a couple of problems:
1) The maximum supported offset of LE is -4094.
2) The offset of WLS also needs to be checked, this uses a
maximum positive offset of 4094.
The use of BasicBlockUtils has been changed because the block offsets
weren't being initialised, but the isBBInRange checks both positive
and negative offsets.
ARMISelLowering has been tweaked because the test case presented
another pattern that we weren't supporting.
llvm-svn: 365749
The VQDMLAH.U8, VQDMLAH.U16 and VQDMLAH.U32 instructions don't
actually exist: the Armv8.1-M architecture spec only lists signed
forms of that instruction. The unsigned ones were added in error: they
existed in an early draft of the spec, but they were removed before
the public version, and we missed that particular spec change.
Also affects the variant forms VQDMLASH, VQRDMLAH and VQRDMLASH.
Reviewers: miyuki
Subscribers: javed.absar, kristof.beyls, hiraditya, dmgreen, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64502
llvm-svn: 365747
Skip copies between virtual registers during search for UseDefs
and DefUses.
Since each operand has one def search for UseDefs is straightforward.
But since operand can have many uses, we have to check all uses of
each copy we traverse during search for DefUses.
Differential Revision: https://reviews.llvm.org/D64486
llvm-svn: 365744
When one of the uses/defs of ambiguous instruction is also ambiguous
visit it recursively and search its uses/defs for instruction with
only one mapping available.
When all instruction in a chain are ambiguous arbitrary mapping can
be selected. For s64 operands in ambiguous chain fprb is selected since
it results in less instructions then having to narrow scalar s64 to s32.
For s32 both gprb and fprb result in same number of instructions and
gprb is selected like a general purpose option.
At the moment we always avoid cross register bank copies.
TODO: Implement a model for costs calculations of different mappings
on same instruction and cross bank copies. Allow cross bank copies
when appropriate according to cost model.
Differential Revision: https://reviews.llvm.org/D64485
llvm-svn: 365743
Two functional changes have been made here:
- Now search up from any add instruction to find the chains of
operations that we may turn into a smlad. This allows the
generation of a smlad which doesn't accumulate into a phi.
- The search function has been corrected to stop it falsely searching
up through an invalid path.
The bulk of the changes have been making the Reduction struct a class
and making it more C++y with getters and setters.
Differential Revision: https://reviews.llvm.org/D61780
llvm-svn: 365740
Summary:
Wasm does not currently support `llvm.clear_cache` intrinsic, and this
prints a proper error message instead of segfault.
Reviewers: dschuff, sbc100, sunfish
Subscribers: jgravelle-google, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64322
llvm-svn: 365731
We use the functions that convert to three address to do the
conversion, but changing an 8 or 16 bit will cause it to create
a virtual register. This can't be done after register allocation
where this pass runs.
I've switched the pass completely to a white list of instructions
that can be converted to LEA instead of a blacklist that was
incorrect. This will avoid surprises if we enhance the three
address conversion function to include additional instructions
in the future.
Fixes PR42565.
llvm-svn: 365720
Unfortunately subo formation in CGP prevents obvious ways of
testing this.
But we already have BLSI in here and the flag behavior is
well understood.
Might become more useful if we improve PR42571.
llvm-svn: 365702
Summary: llvm/IR/GlobalValue.h can't be included in MC, that creates a circular dependency between MC and IR libraries. This circular dependency is causing an issue for build system that enforce layering.
Author: Xiangling_L
Reviewers: sfertile, jasonliu, hubert.reinterpretcast, gribozavr
Reviewed By: gribozavr
Subscribers: wuzish, nemanjai, hiraditya, kbarton, MaskRay, jsji, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64445
llvm-svn: 365701
Since we have distinct types for pointers and scalars, G_INTTOPTRs can sometimes
obstruct attempts to find constant source values. These usually come about when
try to do some kind of null pointer check. Teaching getConstantVRegValWithLookThrough
about this operation allows the CBZ/CBNZ optimization to catch more cases.
This change also improves the case where we can't find a constant source at all.
Previously we would emit a cmp, cset and tbnz for that. Now we try to just emit
a cmp and conditional branch, saving an instruction.
The cumulative code size improvement of this change plus D64354 is 5.5% geomean
on arm64 CTMark -O0.
Differential Revision: https://reviews.llvm.org/D64377
llvm-svn: 365690
Some minor cleanup.
This function in Utils does the same thing as `findMIFromReg`. It also looks
through copies, which `findMIFromReg` didn't.
Delete `findMIFromReg` and use `getOpcodeDef` instead. This only happens in
`tryOptVectorDup` right now.
Update opt-shuffle-splat to show that we can look through the copies now, too.
Differential Revision: https://reviews.llvm.org/D64520
llvm-svn: 365684
There are a few places where we walk over copies throughout
AArch64InstructionSelector.cpp. In Utils, there's a function that does exactly
this which we can use instead.
Note that the utility function works with the case where we run into a COPY
from a physical register. We've run into bugs with this a couple times, so using
it should defend us from similar future bugs.
Also update opt-fold-compare.mir to show that we still handle physical registers
properly.
Differential Revision: https://reviews.llvm.org/D64513
llvm-svn: 365683
Rework the TTI cache and software prefetching APIs to prepare for the
introduction of a general system model. Changes include:
- Marking existing interfaces const and/or override as appropriate
- Adding comments
- Adding BasicTTIImpl interfaces that delegate to a subtarget
implementation
- Adding a default "no information" subtarget implementation
Only a handful of targets use these interfaces currently: AArch64,
Hexagon, PPC and SystemZ. AArch64 already has a custom subtarget
implementation, so its custom TTI implementation is migrated to use
the new facilities in BasicTTIImpl to invoke its custom subtarget
implementation. The custom TTI implementations continue to exist for
the other targets with this change. They are not moved over to
subtarget-based implementations.
The end goal is to have the default subtarget implementation defer to
the system model defined by the target. With this change, the default
subtarget implementation essentially returns "no information" for
these interfaces. None of the existing users of TTI will hit that
implementation because they define their own custom TTI
implementations and won't use the BasicTTIImpl implementations.
Once system models are in place for the targets that use these
interfaces, their custom TTI implementations can be removed.
Differential Revision: https://reviews.llvm.org/D63614
llvm-svn: 365676
This renames the type so it doesn't sound like its based off the load size - as we're moving towards supporting combining loads of different sizes.
llvm-svn: 365655
Summary:
D59191 added support for these modifiers in the assembler and
disassembler. This patch just teaches instruction selection that it can
use them.
Reviewers: arsenm, tstellar
Subscribers: kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64497
llvm-svn: 365640
Cache the LoadSDNode nodes so we can easily map to/from the element index instead of packing them together - this will be useful for future patches for PR16739 etc.
llvm-svn: 365620
This patch checks to see if the vector element loads are based off a dereferenceable pointer that covers the entire vector width, in which case we don't need to have element loads at both extremes of the vector width - just the start (base pointer) of it.
Another step towards partial vector loads......
Differential Revision: https://reviews.llvm.org/D64205
llvm-svn: 365614
Summary:
Use the same predicates as VSTMDB/VLDMIA since VPUSH/VPOP alias to
these.
Patch by Momchil Velikov.
Reviewers: ostannard, simon_tatham, SjoerdMeijer, samparker, t.p.northover, dmgreen
Reviewed By: dmgreen
Subscribers: javed.absar, kristof.beyls, hiraditya, dmgreen, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D64413
llvm-svn: 365604
In SelectionDAG AMDGPU treated these as legal, but this was mostly
because the bitcasts required for FP types were painful. Theoretically
the bitpattern should eventually match to bfi, so don't bother trying
to get the patterns to import.
llvm-svn: 365583
This should prevent doing this on pre-sse4.1 targets or for 256
bit vectors without avx2.
I don't know of a failure from this. Op legalization will probably
take care of, but seemed better to be safe.
llvm-svn: 365577
A short granule is a granule of size between 1 and `TG-1` bytes. The size
of a short granule is stored at the location in shadow memory where the
granule's tag is normally stored, while the granule's actual tag is stored
in the last byte of the granule. This means that in order to verify that a
pointer tag matches a memory tag, HWASAN must check for two possibilities:
* the pointer tag is equal to the memory tag in shadow memory, or
* the shadow memory tag is actually a short granule size, the value being loaded
is in bounds of the granule and the pointer tag is equal to the last byte of
the granule.
Pointer tags between 1 to `TG-1` are possible and are as likely as any other
tag. This means that these tags in memory have two interpretations: the full
tag interpretation (where the pointer tag is between 1 and `TG-1` and the
last byte of the granule is ordinary data) and the short tag interpretation
(where the pointer tag is stored in the granule).
When HWASAN detects an error near a memory tag between 1 and `TG-1`, it
will show both the memory tag and the last byte of the granule. Currently,
it is up to the user to disambiguate the two possibilities.
Because this functionality obsoletes the right aligned heap feature of
the HWASAN memory allocator (and because we can no longer easily test
it), the feature is removed.
Also update the documentation to cover both short granule tags and
outlined checks.
Differential Revision: https://reviews.llvm.org/D63908
llvm-svn: 365551
Basically the problem is that X86 doesn't set the Fast flag from
allowsMemoryAccess on certain CPUs due to slow unaligned memory
subtarget features. This prevents bitcasts from being folded into
loads and stores. But all vector loads and stores of the same width
are the same cost on X86.
This patch merges the allowsMemoryAccess call into isLoadBitCastBeneficial to allow X86 to skip it.
Differential Revision: https://reviews.llvm.org/D64295
llvm-svn: 365549
Thomas Lively