Initial patch adding assembly support for Armv8.4-A.
Besides adding v8.4 as a supported architecture to the usual places, this also
adds target features for the different crypto algorithms. Armv8.4-A introduced
new crypto algorithms, made them optional, and allows different combinations:
- none of the v8.4 crypto functions are supported, which is independent of the
implementation of the Armv8.0 SHA1 and SHA2 instructions.
- the v8.4 SHA512 and SHA3 support is implemented, in this case the Armv8.0
SHA1 and SHA2 instructions must also be implemented.
- the v8.4 SM3 and SM4 support is implemented, which is independent of the
implementation of the Armv8.0 SHA1 and SHA2 instructions.
- all of the v8.4 crypto functions are supported, in this case the Armv8.0 SHA1
and SHA2 instructions must also be implemented.
The v8.4 crypto instructions are added to AArch64 only, and not AArch32,
and are made optional extensions to Armv8.2-A.
The user-facing Clang options will map on these new target features, their
naming will be compatible with GCC and added in follow-up patches.
The Armv8.4-A instruction sets can be downloaded here:
https://developer.arm.com/products/architecture/a-profile/exploration-tools
Differential Revision: https://reviews.llvm.org/D48625
llvm-svn: 335953
Targets should be able to define whether or not they support the outliner
without the outliner being added to the pass pipeline. Before this, the
outliner pass would be added, and ask the target whether or not it supports the
outliner.
After this, it's possible to query the target in TargetPassConfig, before the
outliner pass is created. This ensures that passing -enable-machine-outliner
will not modify the pass pipeline of any target that does not support it.
https://reviews.llvm.org/D48683
llvm-svn: 335887
Add NoTrapAfterNoreturn target option which skips emission of traps
behind noreturn calls even if TrapUnreachable is enabled.
Enable the feature on Mach-O to save code size; Comments suggest it is
not possible to enable it for the other users of TrapUnreachable.
rdar://41530228
DifferentialRevision: https://reviews.llvm.org/D48674
llvm-svn: 335877
Now that we have the ability to legalize based on MMO's. Add support for
legalizing based on AtomicOrdering and use it to correct the legalization
of the atomic instructions.
Also extend all() to be a variadic template as this ruleset now requires
3 and 4 argument versions.
llvm-svn: 335767
It isn't safe to outline sequences of instructions where x16/x17/nzcv live
across the sequence.
This teaches the outliner to check whether or not a specific canidate has
x16/x17/nzcv live across it and discard the candidate in the case that that is
true.
https://bugs.llvm.org/show_bug.cgi?id=37573https://reviews.llvm.org/D47655
llvm-svn: 335758
This patch adds a custom trunc store lowering for v4i8 vector types.
Since there is not v.4b register, the v4i8 is promoted to v4i16 (v.4h)
and default action for v4i8 is to extract each element and issue 4
byte stores.
A better strategy would be to extended the promoted v4i16 to v8i16
(with undef elements) and extract and store the word lane which
represents the v4i8 subvectores. The construction:
define void @foo(<4 x i16> %x, i8* nocapture %p) {
%0 = trunc <4 x i16> %x to <4 x i8>
%1 = bitcast i8* %p to <4 x i8>*
store <4 x i8> %0, <4 x i8>* %1, align 4, !tbaa !2
ret void
}
Can be optimized from:
umov w8, v0.h[3]
umov w9, v0.h[2]
umov w10, v0.h[1]
umov w11, v0.h[0]
strb w8, [x0, #3]
strb w9, [x0, #2]
strb w10, [x0, #1]
strb w11, [x0]
ret
To:
xtn v0.8b, v0.8h
str s0, [x0]
ret
The patch also adjust the memory cost for autovectorization, so the C
code:
void foo (const int *src, int width, unsigned char *dst)
{
for (int i = 0; i < width; i++)
*dst++ = *src++;
}
can be vectorized to:
.LBB0_4: // %vector.body
// =>This Inner Loop Header: Depth=1
ldr q0, [x0], #16
subs x12, x12, #4 // =4
xtn v0.4h, v0.4s
xtn v0.8b, v0.8h
st1 { v0.s }[0], [x2], #4
b.ne .LBB0_4
Instead of byte operations.
llvm-svn: 335735
AArch64 was only setting costs for SK_Transpose, which meant that many of the simpler shuffles (e.g. SK_Select and SK_PermuteSingleSrc for larger vector elements) was being severely overestimated by the default shuffle expansion.
This patch adds costs to help improve SLP performance and avoid a regression in reductions introduced by D48174.
I'm not very knowledgeable about AArch64 shuffle lowering so I've kept the extra costs to a minimum - someone who knows this code can add extra costs which should improve vectorization a lot more.
Differential Revision: https://reviews.llvm.org/D48172
llvm-svn: 335329
This reverts commit d8f57105010cc7e78026e511d5def873fc91e0e7.
Original Commit:
Author: Haicheng Wu <haicheng@codeaurora.org>
Date: Sun Feb 18 13:51:33 2018 +0000
[AArch64] Coalesce Copy Zero during instruction selection
Add special case for copy of zero to avoid a double copy.
Differential Revision: https://reviews.llvm.org/D36104
Author's intention is to remove a BB that has one mov instruction. In
order to do that, d8f571050 pessmizes MachineSinking by introducing a
copy, such that mov instruction is NOT moved to the BB. Optimization
downstream gets rid of the BB with only mov instruction. This works well
if we have only one fall through branch as there is only one "extra"
mov instruction.
If we have multiple fall throughs, we will have a lot of redundant movs.
In such a case, it's better to have this BB which has one mov instruction.
This is causing degradation in jpeg, fft and other codebases. I believe
if we want to remove a BB with only one branch instruction, we should not
pessimize Machine Sinking at all, and find some other solution.
llvm-svn: 335251
insertOutlinerPrologue was not used by any target, and prologue-esque code was
beginning to appear in insertOutlinerEpilogue. Refactor that into one function,
buildOutlinedFrame.
This just removes insertOutlinerPrologue and renames insertOutlinerEpilogue.
llvm-svn: 335076
This patch uses the DiagnosticPredicate for SVE predicate patterns
to improve their diagnostics, now giving a 'invalid operand' diagnostic
if the type is not an immediate or one of the expected pattern
labels.
Reviewers: samparker, SjoerdMeijer, javed.absar, fhahn
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D48220
llvm-svn: 334983
The variants added by this patch are:
- SQINC signed increment, e.g. sqinc x0, w0, all, mul #4
- SQDEC signed decrement, e.g. sqdec x0, w0, all, mul #4
- UQINC unsigned increment, e.g. uqinc w0, all, mul #4
- UQDEC unsigned decrement, e.g. uqdec w0, all, mul #4
This patch includes asmparser changes to parse a GPR64 as a GPR32 in
order to satisfy the constraint check:
x0 == GPR64(w0)
in:
sqinc x0, w0, all, mul #4
^___^ (must match)
Reviewers: rengolin, fhahn, SjoerdMeijer, samparker, javed.absar
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D47716
llvm-svn: 334980
This patch adds instructions for comparing elements from two vectors, e.g.
cmpgt p0.s, p0/z, z0.s, z1.s
and also adds support for comparing to a 64-bit wide element vector, e.g.
cmpgt p0.s, p0/z, z0.s, z1.d
The patch also contains aliases for certain comparisons, e.g.:
cmple p0.s, p0/z, z0.s, z1.s => cmpge p0.s, p0/z, z1.s, z0.s
cmplo p0.s, p0/z, z0.s, z1.s => cmphi p0.s, p0/z, z1.s, z0.s
cmpls p0.s, p0/z, z0.s, z1.s => cmphs p0.s, p0/z, z1.s, z0.s
cmplt p0.s, p0/z, z0.s, z1.s => cmpgt p0.s, p0/z, z1.s, z0.s
llvm-svn: 334931
Support for SVE's predicated select instructions to select elements
from either vector, both in a data-vector and a predicate-vector
variant.
llvm-svn: 334905
Increment/decrement scalar register by (scaled) element count given by
predicate pattern, e.g. 'incw x0, all, mul #4'.
Reviewers: rengolin, fhahn, SjoerdMeijer, samparker, javed.absar
Reviewed By: SjoerdMeijer
Differential Revision: https://reviews.llvm.org/D47713
llvm-svn: 334838
Some instructions require of a limited set of FP immediates as operands,
for example '#0.5 or #1.0' for SVE's FADD instruction.
This patch adds support for parsing and printing such FP immediates as
exact values (e.g. #0.499999 is not accepted for #0.5).
Reviewers: rengolin, fhahn, SjoerdMeijer, samparker, javed.absar
Reviewed By: SjoerdMeijer
Differential Revision: https://reviews.llvm.org/D47711
llvm-svn: 334826
Summary:
For targets I'm not familiar with, I've automatically made the "default to 1 for each resource" behaviour explicit in the td files.
For more obvious cases, I've ventured a fix.
Some notes:
- Exynos is especially fishy.
- AArch64SchedThunderX2T99.td had some truncated entries. If I understand correctly, the person who wrote that interpreted the ResourceCycle as a range. I made the decision to use the upper/lower bound for consistency with the 'Latency' value. I'm sure there is a better choice.
- The change to X86ScheduleBtVer2.td is an NFC, it just makes values more explicit.
Also see PR37310.
Reviewers: RKSimon, craig.topper, javed.absar
Subscribers: kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D46356
llvm-svn: 334586
Register x20 is a callee-saved register which may be used for other
purposes in certain contexts, for example to hold special variables
within the kernel. This change adds support for reserving this register
both to frontend and backend to make this register usable for these
purposes.
Differential Revision: https://reviews.llvm.org/D46552
llvm-svn: 334531
On targets like Arm some relaxations may only be performed when certain
architectural features are available. As functions can be compiled with
differing levels of architectural support we must make a judgement on
whether we can relax based on the MCSubtargetInfo for the function. This
change passes through the MCSubtargetInfo for the function to
fixupNeedsRelaxation so that the decision on whether to relax can be made
per function. In this patch, only the ARM backend makes use of this
information. We must also pass the MCSubtargetInfo to applyFixup because
some fixups skip error checking on the assumption that relaxation has
occurred, to prevent code-generation errors applyFixup must see the same
MCSubtargetInfo as fixupNeedsRelaxation.
Differential Revision: https://reviews.llvm.org/D44928
llvm-svn: 334078
This is setting up to fix bug 37573 cleanly.
This moves data structures that are technically both used in some way by the
target and the general-purpose outlining algorithm into MachineOutliner.h. In
particular, the `Candidate` class is of importance.
Before, the outliner passed the locations of `Candidates` to the target, which
would then make some decisions about the prospective outlined function. This
change allows us to just pass `Candidates` along to the target. This will allow
the target to discard `Candidates` that would be considered unsafe before cost
calculation. Thus, we will be able to remove the unsafe candidates described in
the bug without resorting to torching the entire prospective function.
Also, as a side-effect, it makes the outliner a bit cleaner.
https://bugs.llvm.org/show_bug.cgi?id=37573
llvm-svn: 333952
Summary:
The new rules are straightforward. The main rules to keep in mind
are:
1. NAME is an implicit template argument of class and multiclass,
and will be substituted by the name of the instantiating def/defm.
2. The name of a def/defm in a multiclass must contain a reference
to NAME. If such a reference is not present, it is automatically
prepended.
And for some additional subtleties, consider these:
3. defm with no name generates a unique name but has no special
behavior otherwise.
4. def with no name generates an anonymous record, whose name is
unique but undefined. In particular, the name won't contain a
reference to NAME.
Keeping rules 1&2 in mind should allow a predictable behavior of
name resolution that is simple to follow.
The old "rules" were rather surprising: sometimes (but not always),
NAME would correspond to the name of the toplevel defm. They were
also plain bonkers when you pushed them to their limits, as the old
version of the TableGen test case shows.
Having NAME correspond to the name of the toplevel defm introduces
"spooky action at a distance" and breaks composability:
refactoring the upper layers of a hierarchy of nested multiclass
instantiations can cause unexpected breakage by changing the value
of NAME at a lower level of the hierarchy. The new rules don't
suffer from this problem.
Some existing .td files have to be adjusted because they ended up
depending on the details of the old implementation.
Change-Id: I694095231565b30f563e6fd0417b41ee01a12589
Reviewers: tra, simon_tatham, craig.topper, MartinO, arsenm, javed.absar
Subscribers: wdng, llvm-commits
Differential Revision: https://reviews.llvm.org/D47430
llvm-svn: 333900
For immediates used in DUP instructions that have the range
-128 to 127, or a multiple of 256 in the range -32768 to 32512,
one could argue that when the result element size is 16bits (.h),
the value can be considered both signed and unsigned.
Reviewers: rengolin, fhahn, SjoerdMeijer, samparker, javed.absar
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D47619
llvm-svn: 333873
Print the first indexed element as a FP register, for example:
mov z0.d, z1.d[0]
Is now printed as:
mov z0.d, d1
Next to printing, this patch also adds aliases to parse 'mov z0.d, d1'.
Reviewers: rengolin, fhahn, samparker, SjoerdMeijer, javed.absar
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D47571
llvm-svn: 333872
Unpredicated copy of indexed SVE element to SVE vector,
along with MOV-aliases.
For example:
dup z0.h, z1.h[0]
duplicates the first 16-bit element from z1 to all elements in
the result vector z0.
Reviewers: rengolin, fhahn, samparker, SjoerdMeijer, javed.absar
Reviewed By: SjoerdMeijer
Differential Revision: https://reviews.llvm.org/D47570
llvm-svn: 333871
Predicated copy of floating-point immediate value to SVE vector,
along with MOV-aliases.
Reviewers: rengolin, fhahn, samparker, SjoerdMeijer, javed.absar
Reviewed By: javed.absar
Differential Revision: https://reviews.llvm.org/D47518
llvm-svn: 333869
Predicated copy of possibly shifted immediate value into SVE
vector, along with MOV-aliases.
Reviewers: rengolin, fhahn, samparker, SjoerdMeijer, javed.absar
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D47517
llvm-svn: 333868
Before we were relying on the any extend of the s1 to s32, but
for AAPCS we need to zero-extend it to at least s8.
Fixes PR36719
Differential Revision: https://reviews.llvm.org/D47425
llvm-svn: 333747
Unpredicated copy of floating-point immediate value into SVE vector,
along with MOV-aliases.
Reviewers: rengolin, fhahn, samparker, SjoerdMeijer, javed.absar
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D47482
llvm-svn: 333744
Instead of asserting when using the def_cfa directive with a register
different from fp, fallback on DWARF.
Easily triggered with:
.cfi_def_cfa x1, 32;
rdar://40249694
Differential Revision: https://reviews.llvm.org/D47593
llvm-svn: 333667
This is to make it clear what kind of bugs the LegalizerInfo::verifier
is able to catch and test its output
Reviewers: aemerson, qcolombet
Reviewed By: aemerson
Differential Revision: https://reviews.llvm.org/D46338
llvm-svn: 333597
Floating point immediate combining a negative sign and
a hexadecimal number, e.g. #-0x0 caused the compiler to crash.
Reviewers: rengolin, fhahn, samparker, SjoerdMeijer, javed.absar
Reviewed By: javed.absar
Differential Revision: https://reviews.llvm.org/D47483
llvm-svn: 333524
As suggested in https://bugs.llvm.org/show_bug.cgi?id=32384#c1, this change
makes the inlining of `memset()` and `memcpy()` more aggressive when
compiling for speed. The tuning remains the same when optimizing for size.
Patch by: Sebastian Pop <s.pop@samsung.com>
Evandro Menezes <e.menezes@samsung.com>
Differential revision: https://reviews.llvm.org/D45098
llvm-svn: 333429
This patch addresses the following variants:
- bitmask immediate, e.g. 'and z0.d, z0.d, #0x6'.
- unpredicated data vectors, e.g. 'and z0.d, z1.d, z2.d'.
- predicated data vectors, e.g. 'and z0.d, p0/m, z0.d, z1.d'.
And also several aliases, such as:
- ORN, alias of ORR.
- EON, alias of EOR.
- BIC, alias of AND (immediate variant)
- MOV, alias of ORR (if unpredicated and source register operands are the same)
Reviewers: rengolin, huntergr, fhahn, samparker, SjoerdMeijer, javed.absar
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D47363
llvm-svn: 333414
This patch adds addsub_imm8_opt_lsl_(i8|i16|i32|i64) operands
that are unsigned values in the range 0 to 255. For element widths of
16 bits or higher it may also be a signed multiple of 256 in the
range 0 to 65280.
Note: This also does some refactoring to reuse convenience function
getShiftedVal<shift>(), and now allows AArch64 scalar 'ADD #-4096' to be
accepted to be mapped to SUB #4096.
Reviewers: rengolin, fhahn, samparker, SjoerdMeijer, javed.absar
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D47310
llvm-svn: 333408
Unpredicated copy of optionally-shifted immediate to SVE vector,
along with MOV-aliases.
This patch contains parsing and printing support for
cpy_imm8_opt_lsl_(i8|i16|i32|i64). This operand allows a signed value in
the range -128 to +127. For element widths of 16 bits or higher it may
also be a signed multiple of 256 in the range -32768 to +32512.
For element-width of 8 bits a range of -128 to 255 is accepted, since a copy
of a byte can be considered either signed/unsigned.
Note: This patch renames tryParseAddSubImm() -> tryParseImmWithOptionalShift()
and moves the behaviour of trying to shift a plain immediate by an allowed
shift-value to its addImmWithOptionalShiftOperands() method, so that the
parsing itself is generic and allows immediates from multiple shifted operands.
This is done because an immediate can be divisible by both shifted operands.
Reviewers: rengolin, fhahn, samparker, SjoerdMeijer, javed.absar
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D47309
llvm-svn: 333263
The existing code has three different ways to try to lower a 64-bit
immediate to the sequence ORR+MOVK. The result is messy: it misses
some possible sequences, and the order of the checks means we sometimes
emit two MOVKs when we only need one.
Instead, just use a simple loop to try all possible two-instruction
ORR+MOVK sequences.
Differential Revision: https://reviews.llvm.org/D47176
llvm-svn: 333218
The AArch64 asm parser currently has custom parsing logic for .hword, .word,
and .xword. Rather than use this custom logic, we can just use
addAliasForDirective to enable the reuse of AsmParser::parseDirectiveValue.
Differential Revision: https://reviews.llvm.org/D47000
llvm-svn: 333077
When we're outlining a sequence that ends in a call, we can save up to
three instructions in the outlined function by turning the call into
a tail-call. I refer to this as thunk outlining because the resulting
outlined function looks like a thunk; suggestions welcome for a better
name.
In addition to making the outlined function shorter, thunk outlining
allows outlining calls which would otherwise be illegal to outline:
we don't need to save/restore LR, so we don't need to prove anything
about the stack access patterns of the callee.
To make this work effectively, I also added
MachineOutlinerInstrType::LegalTerminator to the generic MachineOutliner
code; this allows treating an arbitrary instruction as a terminator in
the suffix tree.
Differential Revision: https://reviews.llvm.org/D47173
llvm-svn: 333015
Summary:
This **appears** to be the last missing piece for the masked merge pattern handling in the backend.
This is [[ https://bugs.llvm.org/show_bug.cgi?id=37104 | PR37104 ]].
[[ https://bugs.llvm.org/show_bug.cgi?id=6773 | PR6773 ]] will introduce an IR canonicalization that is likely bad for the end assembly.
Previously, `andps`+`andnps` / `bsl` would be generated. (see `@out`)
Now, they would no longer be generated (see `@in`), and we need to make sure that they are generated.
Differential Revision: https://reviews.llvm.org/D46528
llvm-svn: 332904
With this we gain a little flexibility in how the generic object
writer is created.
Part of PR37466.
Differential Revision: https://reviews.llvm.org/D47045
llvm-svn: 332868
To make this work I needed to add an endianness field to MCAsmBackend
so that writeNopData() implementations know which endianness to use.
Part of PR37466.
Differential Revision: https://reviews.llvm.org/D47035
llvm-svn: 332857
Provide some free functions to reduce verbosity of endian-writing
a single value, and replace the endianness template parameter with
a field.
Part of PR37466.
Differential Revision: https://reviews.llvm.org/D47032
llvm-svn: 332757
The idea is that a client that wants split dwarf would create a
specific kind of object writer that creates two files, and use it to
create the streamer.
Part of PR37466.
Differential Revision: https://reviews.llvm.org/D47050
llvm-svn: 332749
Counting the number of instructions is both unintuitive and inaccurate.
On AArch64, this only affects the generated remarks and certain rare
pseudo-instructions, but it will have a bigger impact on other targets.
Differential Revision: https://reviews.llvm.org/D46921
llvm-svn: 332685
This breaks the code which saves and restores LR, so we can't outline
without doing something more complicated for stack adjustment.
Found by inspection; we get lucky in most cases because getMemOpInfo
only handles STRWpost, not any other pre/post-increment forms. But it
hits a couple of artificial testcases in the tree.
Differential Revision: https://reviews.llvm.org/D46920
llvm-svn: 332529
The cost computation assumes we do this correctly, but the actual
lowering was wrong.
Differential Revision: https://reviews.llvm.org/D46923
llvm-svn: 332514
For regular SVE vector operands, this patch introduces a more
sensible diagnostic when the vector has a wrong suffix (e.g. z0.s vs z0.b).
For example:
add z0.s, z1.s, z2.b -> invalid element width
^_____^
mismatch
For the vector-with-shift/extend (e.g. z0.s, uxtw #2) this patch takes
a slightly different approach and instead returns a 'invalid operand'
if the element size is not as expected. This is because the diagnostics
are more specificied to suggest using the right shift/extend suffix. This
is a trade-off not to introduce more operand classes and still provide
useful diagnostics for LD1 and PRF instructions.
For example:
ld1w z1.s, p0/z, [x0, z0.s] -> invalid shift/extend specified, expected 'z[0..31].s, (uxtw|sxtw)'
ld1w z1.d, p0/z, [x0, z0.s] -> invalid operand
^________________^
mismatch
For gather prefetches, both 'z0.s' and 'z0.d' would be allowed:
prfw #0, p0, [x0, z0.s] -> invalid shift/extend specified, expected 'z[0..31].s, (uxtw|sxtw) #2'
prfw #0, p0, [x0, z0.d] -> invalid shift/extend specified, expected 'z[0..31].d, (lsl|uxtw|sxtw) #2'
Without this change, the diagnostic would unnecessarily suggest a
different element size:
prfw #0, p0, [x0, z0.s] -> invalid shift/extend specified, expected 'z[0..31].d, (lsl|uxtw|sxtw) #2'
Reviewers: SjoerdMeijer, aemerson, fhahn, samparker, javed.absar
Reviewed By: SjoerdMeijer
Differential Revision: https://reviews.llvm.org/D46688
llvm-svn: 332483
Keep loads and stores together (target defines how many loads
and stores to gang up), such that it will help in pairing
and vectorization.
Differential Revision https://reviews.llvm.org/D46477
llvm-svn: 332482
We currently handle all aggregates by creating one large LLT, and letting the
legalizer deal with splitting them up. However using this approach means that
we can't support big endian code correctly.
This patch changes the way that the IRTranslator deals with aggregate values,
by splitting them up into their constituent element values. To do this, parts
of the translator need to be modified to deal with multiple VRegs for a single
Value.
A new Value to VReg mapper is introduced to help keep compile time under
control, currently there is no measurable impact on CTMark despite the extra
code being generated in some cases.
Patch is based on the original work of Tim Northover.
Differential Revision: https://reviews.llvm.org/D46018
llvm-svn: 332449
This patch re-introduces the "S" inline assembler constraint. This matches
an absolute symbolic address or a label reference. The primary use case is
asm("adrp %0, %1\n\t"
"add %0, %0, :lo12:%1" : "=r"(addr) : "S"(&var));
I say re-introduces as it seems like "S" was implemented in the original
AArch64 backend, but it looks like it wasn't carried forward to the merged
backend. The original implementation had A and L modifiers that could be
used to print ":lo12:" to the string. It looks like gcc doesn't use these
and :lo12: is expected to be written in the inline assembly string so I've
not implemented A and L. Clang already supports the S modifier.
Fixes PR37180
Differential Revision: https://reviews.llvm.org/D46745
llvm-svn: 332444
When storing the 0th lane of a vector, use a simpler and usually more
efficient scalar store instead. In this case, also using the unscaled
offset.
Differential revision: https://reviews.llvm.org/D46762
llvm-svn: 332394
When storing the 0th lane of a vector, use a simpler and usually more efficient scalar store instead.
Differential revision: https://reviews.llvm.org/D46655
llvm-svn: 332251
The DEBUG() macro is very generic so it might clash with other projects.
The renaming was done as follows:
- git grep -l 'DEBUG' | xargs sed -i 's/\bDEBUG\s\?(/LLVM_DEBUG(/g'
- git diff -U0 master | ../clang/tools/clang-format/clang-format-diff.py -i -p1 -style LLVM
- Manual change to APInt
- Manually chage DOCS as regex doesn't match it.
In the transition period the DEBUG() macro is still present and aliased
to the LLVM_DEBUG() one.
Differential Revision: https://reviews.llvm.org/D43624
llvm-svn: 332240
Summary:
performPostLD1Combine in AArch64ISelLowering looks for vector
insert_vector_elt of a loaded value which it can optimize into a single
LD1LANE instruction. The code checking for the pattern was not checking
if the lane index was a constant which could cause two problems:
- an assert when lowering the LD1LANE ISD node since it assumes an
constant operand
- an assert in isel if the lane index value depends on the
post-incremented base register
Both of these issues are avoided by simply checking that the lane index
is a constant.
Fixes bug 35822.
Reviewers: t.p.northover, javed.absar
Subscribers: rengolin, kristof.beyls, mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D46591
llvm-svn: 332103
Accessing the members of a large data structures needs a lot of GEPs which
usually have large offsets due to the size of the underlying data structure. If
the offsets are too large to fit into the r+i addressing mode, these GEPs cannot
be sunk to their users' blocks and many extra registers are needed then to carry
the values of these GEPs.
This patch tries to split a large data struct starting from %base like the
following.
Before:
BB0:
%base =
BB1:
%gep0 = gep %base, off0
%gep1 = gep %base, off1
%gep2 = gep %base, off2
BB2:
%load1 = load %gep0
%load2 = load %gep1
%load3 = load %gep2
After:
BB0:
%base =
%new_base = gep %base, off0
BB1:
%new_gep0 = %new_base
%new_gep1 = gep %new_base, off1 - off0
%new_gep2 = gep %new_base, off2 - off0
BB2:
%load1 = load i32, i32* %new_gep0
%load2 = load i32, i32* %new_gep1
%load3 = load i32, i32* %new_gep2
In the above example, the struct is split into two parts. The first part still
starts from %base and the second part starts from %new_base. After the
splitting, %new_gep1 and %new_gep2 have smaller offsets and then can be sunk to
BB2 and folded into their users.
The algorithm to split data structure is simple and very similar to the work of
merging SExts. First, it collects GEPs that have large offsets when iterating
the blocks. Second, it splits the underlying data structures and updates the
collected GEPs to use smaller offsets.
Differential Revision: https://reviews.llvm.org/D42759
llvm-svn: 332015
With custom lowering for vector MULLH{S,U}, it is now profitable to
vectorize a divide by constant loop for the custom types (v16i8, v8i16,
and v4i32). The cost if based on TargetLowering::Build{S,U}DIV which
uses a multiply by constant plus adjustment to express a divide by
constant.
Both {u,s}mull{2} are expressed as Instruction::Mul and shifts by
Instruction::AShr.
llvm-svn: 331873
Reverting this to see if the clang-cmake-aarch64-global-isel and
clang-cmake-aarch64-quick bots are failing because of this commit.
We know it wasn't r331819.
llvm-svn: 331846
Because we create a new kind of debug instruction, DBG_LABEL, we need to
check all passes which use isDebugValue() to check MachineInstr is debug
instruction or not. When expelling debug instructions, we should expel
both DBG_VALUE and DBG_LABEL. So, I create a new function,
isDebugInstr(), in MachineInstr to check whether the MachineInstr is
debug instruction or not.
This patch has no new test case. I have run regression test and there is
no difference in regression test.
Differential Revision: https://reviews.llvm.org/D45342
Patch by Hsiangkai Wang.
llvm-svn: 331844
Patch https://reviews.llvm.org/D41445 changed the behaviour of 'isReg()'
to also return 'true' if the parsed register operand is a vector
register. Code in the AsmMatcher checks if a register is a subclass of the
expected register class. However, even though both parsed registers map
to the same physical register, the 'v' register is of kind 'NeonVector',
where 'q' is of type Scalar, where isSubclass() does not distinguish
between the two cases.
The solution is to use an AsmOperand instead of the register directly,
and use the PredicateMethod to distinguish the two operands.
This fixes for example:
ldr v0, [x0] // 'v0' is an invalid operand for this instruction
ldr q0, [x0] // valid
Reviewers: aemerson, Gerolf, SjoerdMeijer, javed.absar
Reviewed By: aemerson
Differential Revision: https://reviews.llvm.org/D46310
llvm-svn: 331755
Summary:
Previously, a extending load was represented at (G_*EXT (G_LOAD x)).
This had a few drawbacks:
* G_LOAD had to be legal for all sizes you could extend from, even if
registers didn't naturally hold those sizes.
* All sizes you could extend from had to be allocatable just in case the
extend went missing (e.g. by optimization).
* At minimum, G_*EXT and G_TRUNC had to be legal for these sizes. As we
improve optimization of extends and truncates, this legality requirement
would spread without considerable care w.r.t when certain combines were
permitted.
* The SelectionDAG importer required some ugly and fragile pattern
rewriting to translate patterns into this style.
This patch changes the representation to:
* (G_[SZ]EXTLOAD x)
* (G_LOAD x) any-extends when MMO.getSize() * 8 < ResultTy.getSizeInBits()
which resolves these issues by allowing targets to work entirely in their
native register sizes, and by having a more direct translation from
SelectionDAG patterns.
Each extending load can be lowered by the legalizer into separate extends
and loads, however a target that supports s1 will need the any-extending
load to extend to at least s8 since LLVM does not represent memory accesses
smaller than 8 bit. The legalizer can widenScalar G_LOAD into an
any-extending load but sign/zero-extending loads need help from something
else like a combiner pass. A follow-up patch that adds combiner helpers for
for this will follow.
The new representation requires that the MMO correctly reflect the memory
access so this has been corrected in a couple tests. I've also moved the
extending loads to their own tests since they are (mostly) separate opcodes
now. Additionally, the re-write appears to have invalidated two tests from
select-with-no-legality-check.mir since the matcher table no longer contains
loads that result in s1's and they aren't legal in AArch64 anymore.
Depends on D45540
Reviewers: ab, aditya_nandakumar, bogner, rtereshin, volkan, rovka, javed.absar
Reviewed By: rtereshin
Subscribers: javed.absar, llvm-commits, kristof.beyls
Differential Revision: https://reviews.llvm.org/D45541
llvm-svn: 331601
Inspired by r331508, I did a grep and found these.
Mostly just change from dyn_cast to cast. Some cases also showed a dyn_cast result being converted to bool, so those I changed to isa.
llvm-svn: 331577
Summary: Adding support for Fast flags in the SDNode to leverage fast math sub flag usage.
Reviewers: spatel, arsenm, jbhateja, hfinkel, escha, qcolombet, echristo, wristow, javed.absar
Reviewed By: spatel
Subscribers: llvm-commits, rampitec, nhaehnle, tstellar, FarhanaAleen, nemanjai, javed.absar, jbhateja, hfinkel, wdng
Differential Revision: https://reviews.llvm.org/D45710
llvm-svn: 331547
This patch adds a custom lowering for ISD::MULH{S,U} used on divide by
constant optimization (DAGCombiner::BuildSDIV and DAGCombiner::BuildUDIV).
New patterns for smull and umull are added, so AArch64ISD::{S,U}MULL
can be correctly lowered to smull2 and umull2.
Reviewed By: SjoerdMeijer
Differential Revision: https://reviews.llvm.org/D46009
llvm-svn: 331522
We've been running doxygen with the autobrief option for a couple of
years now. This makes the \brief markers into our comments
redundant. Since they are a visual distraction and we don't want to
encourage more \brief markers in new code either, this patch removes
them all.
Patch produced by
for i in $(git grep -l '\\brief'); do perl -pi -e 's/\\brief //g' $i & done
Differential Revision: https://reviews.llvm.org/D46290
llvm-svn: 331272
There are two separate fixes here:
* The lowering code for non-extending loads should report UnableToLegalize instead of emitting the same instruction.
* The target should not be requesting lowering of non-extending loads.
llvm-svn: 331201
This patch extends the 'isSVEVectorRegWithShiftExtend' function to
improve diagnostics for SVE's gather load (scalar + vector) addressing
modes. Instead of always suggesting the 'unscaled' addressing mode,
the use of DiagnosticPredicate enables a more specific error message
in the context where the scaling is incorrect. For example:
ld1h z0.d, p0/z, [x0, z0.d, lsl #2]
^
shift amount should be '1'
Instead of suggesting the packed, unscaled addressing mode:
expected 'z[0..31].d, (uxtw|sxtw)'
the assembler now suggests using the proper scaling:
expected 'z[0..31].d, (lsl|uxtw|sxtw) #1'
Reviewers: fhahn, rengolin, samparker, SjoerdMeijer, javed.absar
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D46124
llvm-svn: 331162
Most of the add<operandname>Operands() functions are the same
and can be replaced by using a single 'RenderMethod' in
the AArch64InstrFormats.td file. Since many of the scaled
immediates (with different scaling/bits) are the same, most of
these can reuse the same AsmOperandClass.
Reviewers: fhahn, rengolin, samparker, SjoerdMeijer, javed.absar
Reviewed By: samparker
Differential Revision: https://reviews.llvm.org/D46122
llvm-svn: 331146
Summary:
Previously, a extending load was represented at (G_*EXT (G_LOAD x)).
This had a few drawbacks:
* G_LOAD had to be legal for all sizes you could extend from, even if
registers didn't naturally hold those sizes.
* All sizes you could extend from had to be allocatable just in case the
extend went missing (e.g. by optimization).
* At minimum, G_*EXT and G_TRUNC had to be legal for these sizes. As we
improve optimization of extends and truncates, this legality requirement
would spread without considerable care w.r.t when certain combines were
permitted.
* The SelectionDAG importer required some ugly and fragile pattern
rewriting to translate patterns into this style.
This patch begins changing the representation to:
* (G_[SZ]EXTLOAD x)
* (G_LOAD x) any-extends when MMO.getSize() * 8 < ResultTy.getSizeInBits()
which resolves these issues by allowing targets to work entirely in their
native register sizes, and by having a more direct translation from
SelectionDAG patterns.
This patch introduces the new generic instructions and new variation on
G_LOAD and adds lowering for them to convert back to the existing
representations.
Depends on D45466
Reviewers: ab, aditya_nandakumar, bogner, rtereshin, volkan, rovka, aemerson, javed.absar
Reviewed By: aemerson
Subscribers: aemerson, kristof.beyls, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D45540
llvm-svn: 331115
This commit makes it so that if you outline a def of some register, then the
call instruction created by the outliner actually reflects that the register
is defined by the call. It also makes it so that outlined functions don't
have the TracksLiveness property.
Outlined calls shouldn't break liveness assumptions that someone might make.
This also un-XFAILs the noredzone test, and updates the calls test.
llvm-svn: 331095
Summary:
Currently only the memory size is supported but others can be added as
needed.
narrowScalar for G_LOAD and G_STORE now correctly update the
MachineMemOperand and will refuse to legalize atomics since those need more
careful expansions to maintain atomicity.
Reviewers: ab, aditya_nandakumar, bogner, rtereshin, aemerson, javed.absar
Reviewed By: aemerson
Subscribers: aemerson, rovka, kristof.beyls, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D45466
llvm-svn: 331071
Put the first ldp at the end, so that the load-store optimizer can run
and merge the ldp and the add into a post-index ldp.
This didn't work in case no frame was needed and resulted in code size
regressions.
llvm-svn: 331044
This adds IR intrinsics for the AArch64 dot-product instructions introduced in
v8.2-A.
Differential revisioon: https://reviews.llvm.org/D46107
llvm-svn: 331036
The program might have unusual expectations for functions; for example,
the Linux kernel's build system warns if it finds references from .text
to .init.data.
I'm not sure this is something we actually want to make any guarantees
about (there isn't any explicit rule that would disallow outlining
in this case), but we might want to be conservative anyway.
Differential Revision: https://reviews.llvm.org/D46091
llvm-svn: 331007
Summary:
Use the FP for scavenged spill slot accesses to prevent corruption of
the callee-save region when the SP is re-aligned.
Based on problem and patch reported by @paulwalker-arm
This is an alternative to solution proposed in D45770
Reviewers: t.p.northover, paulwalker-arm, thegameg, javed.absar
Subscribers: qcolombet, mcrosier, paulwalker-arm, kristof.beyls, rengolin, javed.absar, llvm-commits
Differential Revision: https://reviews.llvm.org/D46063
llvm-svn: 330976
This patch adds a new shuffle kind useful for transposing a 2xn matrix. These
transpose shuffle masks read corresponding even- or odd-numbered vector
elements from two n-dimensional source vectors and write each result into
consecutive elements of an n-dimensional destination vector. The transpose
shuffle kind is meant to model the TRN1 and TRN2 AArch64 instructions. As such,
this patch also considers transpose shuffles in the AArch64 implementation of
getShuffleCost.
Differential Revision: https://reviews.llvm.org/D45982
llvm-svn: 330941
This patch extends the PredicateMethod of AsmOperands used in SVE's
LD1 instructions with a DiagnosticPredicate. This makes them 'context
sensitive' to the operand that has been parsed and tells the user to
use the right register (with expected shift/extend), rather than telling
the immediate is out of range when it actually parsed a register.
Patch [2/2] in a series to improve assembler diagnostics for SVE:
- Patch [1/2]: https://reviews.llvm.org/D45879
- Patch [2/2]: https://reviews.llvm.org/D45880
Reviewers: olista01, stoklund, craig.topper, mcrosier, rengolin, echristo, fhahn, SjoerdMeijer, evandro, javed.absar
Reviewed By: fhahn
Differential Revision: https://reviews.llvm.org/D45880
llvm-svn: 330934
Before, the outliner would grab ADRPs that used LR/W30. This patch fixes
that by checking for explicit uses of those registers before the special-casing
for ADRPs.
This also adds a test that ensures that those sorts of ADRPs won't be outlined.
llvm-svn: 330783
This reland includes a check to prevent the DAG combiner from folding an
offset that is smaller than the existing one. This can cause oscillations
between two possible DAGs, which was the cause of the hang and later assertion
failure observed on the lnt-ctmark-aarch64-O3-flto bot.
http://green.lab.llvm.org/green/job/lnt-ctmark-aarch64-O3-flto/2024/
Original commit message:
> This is a code size win in code that takes offseted addresses
> frequently, such as C++ constructors that typically need to compute
> an offseted address of a vtable. This reduces the size of Chromium
> for Android's .text section by 108KB.
Differential Revision: https://reviews.llvm.org/D45199
llvm-svn: 330630
First off, this is more correct than having the B. Second off, this was making
a bot upset. This fixes that.
Update the test to include -verify-machineinstrs as well to prevent stuff like
this slipping by non debug/assert builds in the future.
llvm-svn: 330459
Summary:
In some cases the shift/extend needs to be explicitly parsed together
with the register, rather than as a separate operand. This is needed
for addressing modes where the instruction as a whole dictates the
scaling/extend, rather than specific bits in the instruction.
By parsing them as a single operand, we avoid the need to pass an
extra operand in all CodeGen patterns (because all operands need to
have an associated value), and we avoid the need to update TableGen to
accept operands that have no associated bits in the instruction.
An added benefit of parsing them together is that the assembler
can give a sensible diagnostic if the scaling is not correct.
This is patch [2/4] in a series to add assembler/disassembler support for
SVE's contiguous LD1 (scalar+scalar) instructions:
- Patch [1/4]: https://reviews.llvm.org/D45687
- Patch [2/4]: https://reviews.llvm.org/D45688
- Patch [3/4]: https://reviews.llvm.org/D45689
- Patch [4/4]: https://reviews.llvm.org/D45690
Reviewers: fhahn, rengolin, javed.absar, huntergr, SjoerdMeijer, t.p.northover, echristo, evandro
Reviewed By: fhahn, SjoerdMeijer
Subscribers: kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D45688
llvm-svn: 330394
Summary:
- Renamed tryParseRegister to tryParseScalarRegister, which
now returns an OperandMatchResultTy.
- Moved matching of certain aliases into matchRegisterNameAlias.
- Changed type of most 'Reg' variables to 'unsigned'.
This is patch [1/4] in a series to add assembler/disassembler support for
SVE's contiguous LD1 (scalar+scalar) instructions:
- Patch [1/4]: https://reviews.llvm.org/D45687
- Patch [2/4]: https://reviews.llvm.org/D45688
- Patch [3/4]: https://reviews.llvm.org/D45689
- Patch [4/4]: https://reviews.llvm.org/D45690
Reviewers: fhahn, rengolin, javed.absar, huntergr, SjoerdMeijer, t.p.northover, echristo, evandro, samparker
Reviewed By: samparker
Subscribers: samparker, llvm-commits, kristof.beyls
Differential Revision: https://reviews.llvm.org/D45687
llvm-svn: 330311
Summary:
Added instructions for contiguous stores, ST1, with scalar+imm addressing
modes and corresponding tests. The patch also adds parsing of
'mul vl' as needed for the VL-scaled immediate.
This is patch [6/6] in a series to add assembler/disassembler support for
SVE's contiguous ST1 (scalar+imm) instructions.
Reviewers: fhahn, rengolin, javed.absar, huntergr, SjoerdMeijer, t.p.northover, echristo, evandro
Reviewed By: rengolin
Subscribers: tschuett, llvm-commits, kristof.beyls
Differential Revision: https://reviews.llvm.org/D45432
llvm-svn: 330014
Summary:
Added Z_(b|h|s|d) vector list RegisterOperands along with support to
add/print the vector lists.
This is patch [5/6] in a series to add assembler/disassembler support for
SVE's contiguous ST1 (scalar+imm) instructions.
Reviewers: fhahn, rengolin, javed.absar, huntergr, SjoerdMeijer, t.p.northover, echristo, evandro
Reviewed By: fhahn
Subscribers: tschuett, kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D45431
llvm-svn: 330000
This is a code size win in code that takes offseted addresses
frequently, such as C++ constructors that typically need to compute
an offseted address of a vtable. This reduces the size of Chromium
for Android's .text section by 108KB.
Differential Revision: https://reviews.llvm.org/D45199
llvm-svn: 329956
AFI->setRedZone(false) was put in the wrong place before, and so it only fired
on functions that didn't have stack frames. This moves that to the top of
emitPrologue to make sure that every function without a redzone has it set
correctly.
This also adds a function representing one of the early exit cases (GHC calling
convention) to the MachineOutliner noredzone test to ensure that we can outline
from functions like these, where we never use a redzone.
llvm-svn: 329922
Summary:
Merged 'addVectorList64Operands' and 'addVectorList128Operands' into a
generic 'addVectorListOperands', which can be easily extended to work
for SVE vectors.
This is patch [4/6] in a series to add assembler/disassembler support for
SVE's contiguous ST1 (scalar+imm) instructions.
Reviewers: fhahn, rengolin, javed.absar, huntergr, SjoerdMeijer, t.p.northover, echristo, evandro
Reviewed By: rengolin
Subscribers: kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D45430
llvm-svn: 329909
Summary:
Added 'RegisterKind' to the VectorListOp structure, so that this operand
type can be reused for SVE vector lists in a later patch. It also
refactors the 'tryParseVectorList' function so it can be used directly
in the ParserMethod of an operand. The parsing can now parse multiple
kinds of vectors and recover if there is no match.
This is patch [3/6] in a series to add assembler/disassembler support for
SVE's contiguous ST1 (scalar+imm) instructions.
Reviewers: fhahn, rengolin, javed.absar, huntergr, SjoerdMeijer, t.p.northover, echristo, evandro
Reviewed By: rengolin
Subscribers: kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D45429
llvm-svn: 329900
Summary:
Place parsing of a vector index into a separate function to reduce
duplication, since the code is duplicated in both the parsing of a
Neon vector register operand and a Neon vector list.
This is patch [2/6] in a series to add assembler/disassembler support for
SVE's contiguous ST1 (scalar+imm) instructions.
Reviewers: fhahn, rengolin, javed.absar, huntergr, SjoerdMeijer, t.p.northover, echristo, evandro
Reviewed By: rengolin
Subscribers: kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D45428
llvm-svn: 329809
In r329691, we would choose FP even if the offset wouldn't fit, just
because the offset is smaller than the one from BP. This made many
accesses through FP need to scavenge a register, which resulted in
slower and bigger code for no good reason.
This patch now always picks the offset that fits first, even if FP is
preferred.
llvm-svn: 329797
Summary:
Merged 'tryMatchVectorRegister' (specific to Neon) and
'tryParseSVERegister' into a single 'tryParseVectorRegister' function, and
created a generic 'parseVectorKind()' function that returns the #Elements
and ElementWidth of a vector suffix. This reduces the duplication of
this functionality between two the vector implementations.
This is patch [1/6] in a series to add assembler/disassembler support for
SVE's contiguous ST1 (scalar+imm) instructions.
Reviewers: fhahn, rengolin, javed.absar, huntergr, SjoerdMeijer, t.p.northover, echristo, evandro
Reviewed By: fhahn
Subscribers: tschuett, llvm-commits, kristof.beyls
Differential Revision: https://reviews.llvm.org/D45427
llvm-svn: 329782
Summary:
When inserting MOVs to avoid Falkor HWPF collisions, the non-base
register operand of load instructions (e.g. a register offset) was not
being considered live, so it could potentially have been used as a
scratch register, clobbering the actual offset value.
Reviewers: mcrosier
Subscribers: rengolin, javed.absar, kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D45502
llvm-svn: 329761
This commit fixes the bot failures that were coming up before with r329716.
The fix was to move the check for "isInSection()" inside of the if condition
and emit the error there instead of waiting to get past the unreachable statement.
This should work in debug and release builds now.
llvm-svn: 329746
Sjoerd Meijer