Updates the names (e.g. widen => extend, saturate => sat) and opcodes of all
SIMD instructions to match the finalized SIMD spec. Deliberately does not change
the public interface in wasm_simd128.h yet; that will require more care.
Depends on D98466.
Differential Revision: https://reviews.llvm.org/D98676
Removes the instruction definitions, intrinsics, and builtins for qfma/qfms,
signselect, and prefetch instructions, which were not included in the final
WebAssembly SIMD spec.
Depends on D98457.
Differential Revision: https://reviews.llvm.org/D98466
Now that the WebAssembly SIMD specification is finalized and engines are
generally up-to-date, there is no need for a separate target feature for gating
SIMD instructions that engines have not implemented. With this change,
v128.const is now enabled by default with the simd128 target feature.
Differential Revision: https://reviews.llvm.org/D98457
[amdgpu] Update med3 combine to skip i64
Fixes an assumption that a type which is not i32 will be i16. This asserts
when trying to sign/zero extend an i64 to i32.
Test case was cut down from an openmp application. Variations on it are hit by
other combines before reaching the problematic one, e.g. replacing the
immediate values with other function arguments changes the codegen path and
misses this combine.
Reviewed By: arsenm
Differential Revision: https://reviews.llvm.org/D98872
Extend this to support ComputeNumSignBits of the (used) source vector elements so that we can handle more than just the case where we're sext_in_reg from the source element signbit.
Noticed while investigating the poor codegen in D98587.
byval requires an implicit copy between the caller and callee such
that the callee may write into the stack area without it modifying the
value in the parent. Previously, this was passing through the raw
pointer value which would break if the callee wrote into it.
Most of the time, this copy can be optimized out (however we don't
have the optimization SelectionDAG does yet).
This will trigger more fallbacks for AMDGPU now, since we don't have
legalization for memcpy yet (although we should stop using byval
anyway).
LLVM test CodeGen/AArch64/machine-outliner-retaddr-sign-thunk.ll uses
a string substitution block that contains a regex matching block. This
seems like as a copy/paste from other similar test where the match also
defines a variable, hence the [[]] syntax. In this case however this is
a CHECK-NOT variable so nothing should match. No variable definition is
thus expected and the square brackets can be dropped.
Reviewed By: chill
Differential Revision: https://reviews.llvm.org/D98853
Add ISD::ABS to the existing unary instructions handling for splat detection
This is similar to D83605, but doesn't appear to need to touch any of the wasm refactoring.
Differential Revision: https://reviews.llvm.org/D98778
This patch adds support for masked scatter intrinsics on scalable vector
types. It is mostly an extension of the earlier masked gather support
introduced in D96263, since the addressing mode legalization is the
same.
Reviewed By: craig.topper
Differential Revision: https://reviews.llvm.org/D96486
This patch supports the masked gather intrinsics in RVV.
The RVV indexed load/store instructions only support the "unsigned unscaled"
addressing mode; indices are implicitly zero-extended or truncated to XLEN and
are treated as byte offsets. This ISA supports the intrinsics directly, but not
the majority of various forms of the MGATHER SDNode that LLVM combines to. Any
signed or scaled indexing is extended to the XLEN value type and scaled
accordingly. This is done during DAG combining as widening the index types to
XLEN may produce illegal vectors that require splitting, e.g.
nxv16i8->nxv16i64.
Support for scalable-vector CONCAT_VECTORS was added to avoid spilling via the
stack when lowering split legalized index operands.
Reviewed By: craig.topper
Differential Revision: https://reviews.llvm.org/D96263
Without this patch, bitcasts of fixed-length mask vectors would go
through the stack.
Reviewed By: craig.topper
Differential Revision: https://reviews.llvm.org/D98779
D93594 depend on the dominate tree and loop information. It increased
the compile time when build with -O0. However this is just to amend the
dominate tree and loop information, so that it is unnecessary to
re-analyze them again. Given the dominate tree of loop information are
absent in this pass, we can avoid amending them.
Differential Revision: https://reviews.llvm.org/D98773
Don't rewrite an add instruction with 2 SET_CC operands into a csel
instruction. The total instruction sequence uses an extra instruction and
register. Preventing this allows us to match a `(add, csel)` pattern and
rewrite this into a `cinc`.
Differential Revision: https://reviews.llvm.org/D98704
This patch changes the operand order of masked vmslt[u]
from (mask, rs1, scalar, maskedoff, vl)
to (maskedoff, rs1, scalar, mask, vl).
Reviewed By: craig.topper
Differential Revision: https://reviews.llvm.org/D98839
The offset in HVX loads/stores is only 4 bits long, so often an
extra register is needed to hold the address. Minimize the number
of such registers by "standardizing" the base addresses and reusing
preexisting base registers when replacing frame indices.
These are pseudos without any users, so DCE was killing them in the combiner.
Marking them as having side effects doesn't seem quite right since they don't.
Gives a nice 0.3% geomean size win on CTMark -Os.
Differential Revision: https://reviews.llvm.org/D98811
The previous technique relied on early-exiting the legalizer predicate
initialization, leaving an empty rule table. That causes a fallback
for most instructions, but some have legacy rules defined like G_ZEXT
which can try continue, but then crash.
We should fall back earlier, in the translator, to avoid this issue.
Differential Revision: https://reviews.llvm.org/D98730
Given a sextload i16, we can usually generate "ldrsh [rn. rm]". If we
don't naturally have a rn, rm addressing mode, we can either generate
"ldrh [rn, #0]; sxth" or "mov rm, #0; ldrsh [rn. rm]".
We currently generate the first, always creating a sxth. They are both
the same number of instructions, but if we generate the second then the
mov #0 will likely be CSE'd or pulled out of a loop, etc.
This adjusts the ISel patterns to do that, creating a mov instead of a
sxth.
Differential Revision: https://reviews.llvm.org/D98693
Previously NEON used a target specific intrinsic for frintn, given that
the FROUNDEVEN ISD node now exists, move over to that instead and add
codegen support for that node for both NEON and fixed length SVE.
Differential Revision: https://reviews.llvm.org/D98487
This patch adds an optimization path for BUILD_VECTOR nodes where the
majority of the elements are identical. These can be splatted, with the
remaining elements patched up with INSERT_VECTOR_ELTs. The threshold can
be tweaked as required - it is currently conservative. Undef elements
are disregarded when judging the dominance of a particular element. This
allows them to be covered by the splat value.
In addition, vectors of 2 elements are always optimized to a splat (for
the upper element) and an insert at element zero.
This optimization is disabled when optimizing for size.
Reviewed By: craig.topper
Differential Revision: https://reviews.llvm.org/D98700
This patch is to replace the fixed value with expression.
Keep .file section as fixed values as it might be changed. The
remaining sections will hardly be modified. So the Index values
are sequential. By using expression, we can avoid the fixed value
changes in coming patches.
This is a follow-up of patch D97117.
Reviewed By: hubert.reinterpretcast, shchenz
Differential Revision: https://reviews.llvm.org/D98620
This commit folds sxtw'd or uxtw'd offsets into gather loads where
possible with a DAGCombine optimization.
As an example, the following code:
1 #include <arm_sve.h>
2
3 svuint64_t func(svbool_t pred, const int32_t *base, svint64_t offsets) {
4 return svld1sw_gather_s64offset_u64(
5 pred, base, svextw_s64_x(pred, offsets)
6 );
7 }
would previously lower to the following assembly:
sxtw z0.d, p0/m, z0.d
ld1sw { z0.d }, p0/z, [x0, z0.d]
ret
but now lowers to:
ld1sw { z0.d }, p0/z, [x0, z0.d, sxtw]
ret
Differential Revision: https://reviews.llvm.org/D97858
The InstrEmitter can sometimes insert a copy after an IMPLICIT_DEF
before connecting it to the vector instruction. This occurs when
constrainRegClass reduces to a class with less than 4 registers.
I believe LMUL8 on masked instructions triggers this since the
result can only use the v8, v16, or v24 register group as the mask
is using v0.
Reviewed By: frasercrmck
Differential Revision: https://reviews.llvm.org/D98567
This commit implements an IR-level optimization to eliminate idempotent
SVE mul/fmul intrinsic calls. Currently, the following patterns are
captured:
fmul pg (dup_x 1.0) V => V
mul pg (dup_x 1) V => V
fmul pg V (dup_x 1.0) => V
mul pg V (dup_x 1) => V
fmul pg V (dup v pg 1.0) => V
mul pg V (dup v pg 1) => V
The result of this commit is that code such as:
1 #include <arm_sve.h>
2
3 svfloat64_t foo(svfloat64_t a) {
4 svbool_t t = svptrue_b64();
5 svfloat64_t b = svdup_f64(1.0);
6 return svmul_m(t, a, b);
7 }
will lower to a nop.
This commit does not capture all possibilities; only the simple cases
described above. There is still room for further optimisation.
Differential Revision: https://reviews.llvm.org/D98033
The default promotion uses zero extends that become shifts. We
cam use sign extend instead which is better for RISCV.
I've used two different implementations based on whether we
have minu/maxu instructions.
Differential Revision: https://reviews.llvm.org/D98683
This pass runs in any situations but we skip it when it is not O0 and the
function doesn't have optnone attribute. With -O0, the def of shape to amx
intrinsics is near the amx intrinsics code. We are not able to find a
point which post-dominate all the shape and dominate all amx intrinsics.
To decouple the dependency of the shape, we transform amx intrinsics
to scalar operation, so that compiling doesn't fail. In long term, we
should improve fast register allocation to allocate amx register.
Reviewed By: pengfei
Differential Revision: https://reviews.llvm.org/D93594
Avoid making a temporary copy of byval argument if all accesses are loads and
therefore the pointer to the parameter can not escape.
This avoids excessive global memory accesses when each kernel makes its own
copy.
Differential revision: https://reviews.llvm.org/D98469
RA can insert something like a sub1_sub2 COPY of a wide VGPR
tuple which results in the unaligned acces with v_pk_mov_b32
after the copy is expanded. This is regression after D97316.
Differential Revision: https://reviews.llvm.org/D98549
