This patch updates how splat loads handled and is an extension of D106555.
Particularly, for v2i64/v4f32/v4i32 types, they are updated to handle only
non-extending loads. For v8i16/v16i8 types, they are updated to handle extending
loads only if the memory VT is the same vector element VT type.
A test case has been added to illustrate a scenario where a PPCISD::LD_SPLAT
node should not be produced. In this test, it depicts the following f64
extending load used in a v2f64 build vector, but the extending load is actually
used in more places other than the build vector (such as in t12 and t16).
```
Type-legalized selection DAG: %bb.0 'test:entry'
SelectionDAG has 20 nodes:
t0: ch = EntryToken
t4: i64,ch = CopyFromReg t0, Register:i64 %1
t6: i64,ch = CopyFromReg t0, Register:i64 %2
t11: f64,ch = load<(load (s64) from %ir.b, !tbaa !7)> t0, t4, undef:i64
t16: f64 = fadd t31, t37
t34: ch = store<(store (s64) into %ir.c, !tbaa !7)> t31:1, t16, t6, undef:i64
t36: ch = TokenFactor t34, t37:1
t27: v2f64 = BUILD_VECTOR t37, t37
t22: ch,glue = CopyToReg t36, Register:v2f64 $v2, t27
t12: f64 = fadd t11, t37
t28: ch = store<(store (s64) into %ir.b, !tbaa !7)> t11:1, t12, t4, undef:i64
t31: f64,ch = load<(load (s64) from %ir.c, !tbaa !7)> t28, t6, undef:i64
t2: i64,ch = CopyFromReg t0, Register:i64 %0
t37: f64,ch = load<(load (s32) from %ir.a, !tbaa !3), anyext from f32> t0, t2, undef:i64
t23: ch = PPCISD::RET_FLAG t22, Register:v2f64 $v2, t22:1
```
Differential Revision: https://reviews.llvm.org/D117803
When a comparison is extended and it would be free to extend the
arguments to that comparison, we can propagate the extend into those arguments.
This prevents extra instructions being generated to extend the result of the
comparison, which is not free to extend.
This is a resubmission of D116812 with fixes that need another review.
Differential Revision: https://reviews.llvm.org/D118139
Adds patterns of the form "(and a, (not b)) -> bic".
NOTE: With this support I'm inclined to remove AArch64ISD::BIC,
but will leave that investigation for another time.
Differential Revision: https://reviews.llvm.org/D118365
These splats -- whether BUILD_VECTOR or SPLAT_VECTOR -- are formed by
first extracting a value from a vector and splatting it to all elements
of the destination vector. These could be performed more optimally,
avoiding the drop to scalar, using RVV's vrgather, for example.
The LMULMAX check names didn't match the options we were passing to llc
(they were swapped around) and we were silently missing coverage for one
test which differs between RV32 and RV64.
Previous folds by combineSetCCMOVMSK might have converted these to CMP when changing the bitwidth, and the CMP->SUB fold might not have happened (or will happen)
On the level of the generated object files, both symbols (both
original and alias) are generally indistinguishable - both are
regular defined symbols. But previously, only the original
function had the COFF ComplexType set to IMAGE_SYM_DTYPE_FUNCTION,
while the symbol created via an alias had the type set to
IMAGE_SYM_DTYPE_NULL.
This matches what GCC does, which emits directives for setting the
COFF symbol type for this kind of alias symbol too.
This makes a difference when GNU ld.bfd exports symbols without
dllexport directives or a def file - it seems to decide between
function or data exports based on the COFF symbol type. This means
that functions created via aliases, like some C++ constructors,
are exported as data symbols (missing the thunk for calling without
dllimport).
The hasnt been an issue when doing the same with LLD, as LLD decides
between function or data export based on the flags of the section
that the symbol points at.
This should fix the root cause of
https://github.com/msys2/MINGW-packages/issues/10547.
Differential Revision: https://reviews.llvm.org/D118328
rG9103b73fe052 was assuming that we could OR/AND with the source vector, but that will fail on float/double vectors without bitcasting - it also missed the case that any_of checks might be testing less than all the source elements
This is similar to D116619, but now it handles `invoke`s. The reason we
didn't handle `invoke`s back then was we didn't support Wasm EH + Wasm
SjLj together, and the only case SjLj transformation will see `invoke`s
is when we are using Wasm EH. (In Emscripten EH, they would have been
transformed to `call`s to invoke wrappers.)
But after D117610 we support Wasm EH + Wasm SjLj together and we can
nullify `invoke`s to `setjmp` when there is no other longjmpable calls
within the function. Actually this is very unlikely to happen in
practice, because we treat destructors as longjmpable and also treat
`__cxa_end_catch` as longjmpable even if it is not.
Reviewed By: dschuff
Differential Revision: https://reviews.llvm.org/D118408
Wasm SjLj converts longjmpable calls into `invoke`s that unwind to
`%catch.longjmp.dispatch` BB, from where we check if the thrown
exception is a `longjmp`. But in case a call already has a `funclet`
attribute, i.e., it is within a catch scope, we have to unwind to its
unwind destination first to preserve the scoping structure. That will
eventually unwind to `%catch.longjmp.dispatch`, because all
`catchswitch` and `cleanupret` that unwind to caller are redirected to
`%catch.dispatch.longjmp` during Wasm SjLj transformation.
But the prevous code assumed `cleanuppad`'s parent pad was always an
instruction, and didn't handle when a `cleanuppad`'s parent is `none`.
This CL handles this case, and makes the `while` loop more intuitive by
removing `FromPad` condition and explicitly inserting `break`s.
Reviewed By: dschuff
Differential Revision: https://reviews.llvm.org/D118407
Wasm EH, used with either of Emscripten SjLj or Wasm SjLj, does not
allow `setjmp` calls to be placed within a `catch` clause, because we
don't support jumping into a `catch` block. Emscripten EH does not have
this restriction. But I think this restriction wouldn't prevent most of
use cases. This CL errors out with a clear messsage for this case.
Reviewed By: dschuff, kripken
Differential Revision: https://reviews.llvm.org/D118286
When we create an invoke wrapper call, if the original call instruction
has a `noreturn` attribute, we shouldn't copy it, because we expect
invoke wrapper calls to return. This generated incorrect `free` call
before an invoke wrapper call that calls `__cxa_throw`, because
`__cxa_throw` has `noreturn` attribute.
Reviewed By: dschuff
Differential Revision: https://reviews.llvm.org/D118274
In D117610 we treated `__cxa_end_catch` longjmpable even though it was
not to make unwind destination relationships correct. But we only need
to do this in Wasm SjLj, and doing this in Emscripten SjLj does not make
the code incorrect but add unnecessary invokes. This CL treats
`__cxa_end_catch` longjmpable only in Wasm SjLj.
Reviewed By: dschuff
Differential Revision: https://reviews.llvm.org/D117943
According to riscv-v-spec-1.0, widening signed(vs2)-unsigned integer multiply
vwmulsu.vv vd, vs2, vs1, vm # vector-vector
vwmulsu.vx vd, vs2, rs1, vm # vector-scalar
It is worth noting that signed op is only for vs2.
For vwmulsu.vv, we can swap two ops, and don't care which is sign extension,
but for vwmulsu.vx signExt can not be a vector extended from scalar (rs1).
I specifically added two functions ending with _swap in the test case.
Reviewed By: craig.topper
Differential Revision: https://reviews.llvm.org/D118215
Return false from ShouldShrinkFPConstant(), so that these constants are stored
in their full size on the constant pool, even if they could have been shrunk
and used with an extending load.
This is better since LD is faster than LDE, and it also enables reg/mem opcodes.
Review: Ulrich Weigand
Differential Revision: https://reviews.llvm.org/D117927
By reordering the objects on the stack frame after looking at the users, a
better utilization of displacement operands will result. This means less
needed Load Address instructions for the accessing of these objects.
This is important for very large functions where otherwise small changes
could cause a lot more/less accesses go out of range.
Note: this is not yet enabled for SystemZXPLINKFrameLowering, but should be.
Review: Ulrich Weigand
Differential Revision: https://reviews.llvm.org/D115690
Since it's introduction, the qrdmlah has been represented as a qrdmulh
and a sadd_sat. This doesn't produce the same result for all input
values though. This patch fixes that by introducing a qrdmlah (and
qrdmlsh) intrinsic specifically for the vqrdmlah and sqrdmlah
instructions. The old test cases will now produce a qrdmulh and sqadd,
as expected.
Fixes#53120 and #50905 and #51761.
Differential Revision: https://reviews.llvm.org/D117592
This adds the following changes:
* Fold: vselect(<all active predicate>, x, y) => x
* Extend isAllActivePredicate to take vscale_range into account, e.g.
isAllActivePredicate(vl16) for nxv16i1 and vscale == 1 => true.
isAllActivePredicate(vl32) for nxv16i1 and vscale == 2 => true.
Differential Revision: https://reviews.llvm.org/D118147
InstCombine performs this more generally with SimplifyUsingDistributiveLaws, but we don't need anything that complex here - this is mainly to fix up cases where logic ops get created late on during lowering, often in conjunction with sext/zext ops for type legalization.
https://alive2.llvm.org/ce/z/gGpY5v
Swizzled accesses are not merged, but there is no particular reason not
to merge two instructions if any of the intervening instructions happens
to be a swizzled access.
This moves the check for swizzled accesses out of checkAndPrepareMerge
into collectMergeableInsts where I think it makes more sense.
Differential Revision: https://reviews.llvm.org/D118267
We already perform some basic folds (add/sub with zero etc.) on scalar types, this patch adds some basic support for constant splats as well in a few cases (we can add more with future test coverage).
In the cases I've enabled, we can handle buildvector implicit truncation as we're not creating new constant nodes from the vector types - we're just returning existing nodes. This allows us to get a number of extra cases in the aarch64 tests.
I haven't enabled support for undefs in buildvector splats, as we're often checking for zero/allones patterns that return the original constant and we shouldn't be returning undef elements in some of these cases - we can enable this later if we're OK with creating new constants.
Differential Revision: https://reviews.llvm.org/D118264
The ISel patterns for PFALSE helps recognise the instructions as being
free of side-effects, which helps MachineCSE remove redundant
PFALSE instructions.
Reviewed By: dmgreen
Differential Revision: https://reviews.llvm.org/D118054
This patch adds support for expanding VP_MERGE through a sequence of
vector operations producing a full-length mask setting up the elements
past EVL/pivot to be false, combining this with the original mask, and
culminating in a full-length vector select.
This expansion should work for any data type, though the only use for
RVV is for boolean vectors, which themselves rely on an expansion for
the VSELECT.
Reviewed By: craig.topper
Differential Revision: https://reviews.llvm.org/D118058
CSKY arch has multiple FPU instruction versions such as FPU, FPUv2 and FPUv3 to implement floating operations.
For now, we just only support FPUv2 and FPUv3.
It includes the encoding, asm parsing of instructions and codegen of DAG nodes.
