When using the ARM AAPCS, HFAs (Homogeneous Floating-point Aggregates) must
be passed in a block of consecutive floating-point registers, or on the stack.
This means that unused floating-point registers cannot be back-filled with
part of an HFA, however this can currently happen. This patch, along with the
corresponding clang patch (http://reviews.llvm.org/D3083) prevents this.
llvm-svn: 208413
Summary:
Adds MIPS32r6/MIPS64r6 and checks the compatibility requirements for these
processors.
I've also included comments to describe removed and re-encoded instructions,
along with placeholder def's for the new instructions but there are no
functional changes to codegen at this point.
Reviewers: jkolek, vmedic
Reviewed By: vmedic
Differential Revision: http://reviews.llvm.org/D3622
llvm-svn: 208399
Handle lowering of global addresses for PIC mode compilation on Windows. Always
use the movw/movt load to load the address as Windows on ARM requires ARMv7+ and
is a pure Thumb environment.
llvm-svn: 208385
Summary:
Also ran clang-format on the function. The code added is the last else
if block.
Reviewers: nadav, craig.topper, delena
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D3518
llvm-svn: 208372
This patch teaches the backend how to combine packed SSE2/AVX2 arithmetic shift
intrinsics.
The rules are:
- Always fold a packed arithmetic shift by zero to its first operand;
- Convert a packed arithmetic shift intrinsic dag node into a ISD::SRA only if
the shift count is known to be smaller than the vector element size.
This patch also teaches to function 'getTargetVShiftByConstNode' how fold
target specific vector shifts by zero.
Added two new tests to verify that the DAGCombiner is able to fold
sequences of SSE2/AVX2 packed arithmetic shift calls.
llvm-svn: 208342
When building on Windows, the default target is Windows. Windows on ARM does
not support ARM mode compilation, resulting in test failures. Simply specify a
triple to ensure that we are testing the correct behaviour.
llvm-svn: 208340
Summary:
Vectors built with zeros and elements in the same order as another
(source) vector are optimized to be built using a single insertps
instruction.
Also optimize when we move one element in a vector to a different place
in that vector while zeroing out some of the other elements.
Further optimizations are possible, described in TODO comments.
I will be implementing at least some of them in the near future.
Added some tests for different cases where this optimization triggers.
Reviewers: nadav, delena, craig.topper
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D3521
llvm-svn: 208271
Prior to r208252, the FMA 231 family was marked as isCommutable. However the
memory variants of this family are not commutable. Therefore, we did not
implemented the findCommutedOpIndices for those variants and missed that
the default implementation (more or less: commute indices 1 and 2) was
firing behind our back.
As a result, as demonstrated in the test case before the fix, we were
transforming a = b * c + a into a = a * c + b.
I.e., before r208252 we were generating for this test case:
vmovaps %xmm0, %xmm1
vmoss (%rsi), %xmm0
vfmadd231ss (%rdi), %xmm1, %xmm0
Instead of:
vmoss (%rsi), %xmm1
vfmadd231ss (%rdi), %xmm1, %xmm0
<rdar://problem/16800495>
llvm-svn: 208260
this patch disables the dead register elimination pass and the load/store pair
optimization pass at -O0. The ILP optimizations don't require the optimization
level to be checked because the call to addILPOpts is predicated with the
necessary check. The AdvSIMDScalar pass is disabled by default at all
optimization levels. This patch leaves that pass disabled by default.
Also, move command-line options into ARM64TargetMachine.cpp and add a few
additional flags to aid in debugging. This fixes an issue with the
-debug-pass=Structure flag where passes were printed, but not actually run
(i.e., AdvSIMDScalar pass).
llvm-svn: 208223
When performing a scalar comparison that feeds into a vector select,
it's actually better to do the comparison on the vector side: the
scalar route would be "CMP -> CSEL -> DUP", the vector is "CM -> DUP"
since the vector comparisons are all mask based.
llvm-svn: 208210
The AAPCS states that values passed in registers must have a value as though
they had been loaded with "LDR". LDR is equivalent to "LD1.64 vX.1D" - that is,
loading scalars to vector registers and loading 1-element vectors is equivalent.
The logic implemented here is to ensure that at all call boundaries and during
formal argument lowering all vectors are treated as their bitwidth-based floating
point scalar counterpart, which is always one of f64 or f128 (v2i32 -> f64,
v4i32 -> f128 etc). A BITCAST is inserted so that the appropriate REV will be
generated during code generation.
llvm-svn: 208198
Because we've canonicalised on using LD1/ST1, every time we do a bitcast
between vector types we must do an equivalent lane reversal.
Consider a simple memory load followed by a bitconvert then a store.
v0 = load v2i32
v1 = BITCAST v2i32 v0 to v4i16
store v4i16 v2
In big endian mode every memory access has an implicit byte swap. LDR and
STR do a 64-bit byte swap, whereas LD1/ST1 do a byte swap per lane - that
is, they treat the vector as a sequence of elements to be byte-swapped.
The two pairs of instructions are fundamentally incompatible. We've decided
to use LD1/ST1 only to simplify compiler implementation.
LD1/ST1 perform the equivalent of a sequence of LDR/STR + REV. This makes
the original code sequence: v0 = load v2i32
v1 = REV v2i32 (implicit)
v2 = BITCAST v2i32 v1 to v4i16
v3 = REV v4i16 v2 (implicit)
store v4i16 v3
But this is now broken - the value stored is different to the value loaded
due to lane reordering. To fix this, on every BITCAST we must perform two
other REVs:
v0 = load v2i32
v1 = REV v2i32 (implicit)
v2 = REV v2i32
v3 = BITCAST v2i32 v2 to v4i16
v4 = REV v4i16
v5 = REV v4i16 v4 (implicit)
store v4i16 v5
This means an extra two instructions, but actually in most cases the two REV
instructions can be combined into one. For example:
(REV64_2s (REV64_4h X)) === (REV32_4h X)
There is also no 128-bit REV instruction. This must be synthesized with an
EXT instruction.
Most bitconverts require some sort of conversion. The only exceptions are:
a) Identity conversions - vNfX <-> vNiX
b) Single-lane-to-scalar - v1fX <-> fX or v1iX <-> iX
Even though there are hundreds of changed lines, I have a fairly high confidence
that they are somewhat correct. The changes to add two REV instructions per
bitcast were pretty mechanical, and once I'd done that I threw the resulting
.td at a script I wrote which combined the two REVs together (and added
an EXT instruction, for f128) based on an instruction description I gave it.
This was much less prone to error than doing it all manually, plus my brain
would not just have melted but would have vapourised.
llvm-svn: 208194
This completes the port of r204814 (cpirker "AArch64_BE function argument
passing for ARM ABI") from AArch64 to ARM64, and fixes a bunch of issues
found during later development along the way. The biggest of these was
that the alignment fixup logic wasn't replicated into all the places it
should have been.
llvm-svn: 208192
The ARM::BLX instruction is an ARM mode instruction. The Windows on ARM target
is limited to Thumb instructions. Correctly use the thumb mode tBLXr
instruction. This would manifest as an errant write into the object file as the
instruction is 4-bytes in length rather than 2. The result would be a corrupted
object file that would eventually result in an executable that would crash at
runtime.
llvm-svn: 208152
remove it from the list of unspilled registers. Otherwise the following
attempt to keep the stack aligned by picking an extra GPR register to
spill will not work as it picks up r11.
llvm-svn: 208129
Before this patch, the backend always emitted a store+load sequence to
bitconvert from f64 to i64 the input operand of a ISD::BITCAST dag node that
performed a bitconvert from type MVT::f64 to type MVT::v2i32. The resulting
i64 node was then used to build a v2i32 vector.
With this patch, the backend now produces a cheaper SCALAR_TO_VECTOR from
MVT::f64 to MVT::v2f64. That SCALAR_TO_VECTOR is then followed by a "free"
bitcast to type MVT::v4i32. The elements of the resulting
v4i32 are then extracted to build a v2i32 vector (which is illegal and
therefore promoted to MVT::v2i64).
This is in general cheaper than emitting a stack store+load sequence
to bitconvert the operand from type f64 to type i64.
llvm-svn: 208107
This patch implements the infrastructure to use named register constructs in
programs that need access to specific registers (bare metal, kernels, etc).
So far, only the stack pointer is supported as a technology preview, but as it
is, the intrinsic can already support all non-allocatable registers from any
architecture.
llvm-svn: 208104
An alias has the address of what it points to, so it also has the same
alignment.
This allows a few optimizations to see past aliases for free.
llvm-svn: 208103
The Win64 docs are very clear that anything larger than 8 bytes is
passed by reference, and GCC MinGW64 honors that for __modti3 and
friends.
Patch by Jameson Nash!
llvm-svn: 208029
Summary:
Also ran clang-format on the function. The code added is the last else
if block.
Reviewers: nadav, craig.topper
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D3518
llvm-svn: 207992
Windows on ARM does not conform to AEABI. However, memset would be emitted
using the AEABI signature, resulting in inverted parameters. Handle this
special case appropriately.
llvm-svn: 207943
Both MinGW and cygwin (i686) construct export directives without the global
leader prefix. This is mostly due to the fact that they use GNU ld which does
not correctly handle the export directive. This apparently has been been broken
for a while. However, this was recently reported as being broken by
mingwandroid and diorcety of the msys2 project.
Remove the global leader prefix if targeting MinGW or cygwin, otherwise, retain
the global leader prefix. Add an explicit test for cygwin's behaviour of export
directives.
llvm-svn: 207926
While post-indexed LD1/ST1 instructions do exist for vector loads,
this patch makes use of the more flexible addressing-modes in LDR/STR
instructions.
llvm-svn: 207838
Given the following C code llvm currently generates suboptimal code for
x86-64:
__m128 bss4( const __m128 *ptr, size_t i, size_t j )
{
float f = ptr[i][j];
return (__m128) { f, f, f, f };
}
=================================================
define <4 x float> @_Z4bss4PKDv4_fmm(<4 x float>* nocapture readonly %ptr, i64 %i, i64 %j) #0 {
%a1 = getelementptr inbounds <4 x float>* %ptr, i64 %i
%a2 = load <4 x float>* %a1, align 16, !tbaa !1
%a3 = trunc i64 %j to i32
%a4 = extractelement <4 x float> %a2, i32 %a3
%a5 = insertelement <4 x float> undef, float %a4, i32 0
%a6 = insertelement <4 x float> %a5, float %a4, i32 1
%a7 = insertelement <4 x float> %a6, float %a4, i32 2
%a8 = insertelement <4 x float> %a7, float %a4, i32 3
ret <4 x float> %a8
}
=================================================
shlq $4, %rsi
addq %rdi, %rsi
movslq %edx, %rax
vbroadcastss (%rsi,%rax,4), %xmm0
retq
=================================================
The movslq is uneeded, but is present because of the trunc to i32 and then
sext back to i64 that the backend adds for vbroadcastss.
We can't remove it because it changes the meaning. The IR that clang
generates is already suboptimal. What clang really should emit is:
%a4 = extractelement <4 x float> %a2, i64 %j
This patch makes that legal. A separate patch will teach clang to do it.
Differential Revision: http://reviews.llvm.org/D3519
llvm-svn: 207801
Oliver Stannard