When using AAPCS-compliant frame chains with PACBTI return address
signing, there ware a number of bugs in the generation of the frame
pointer and function prologues. The most obvious was that we sometimes
would modify r11 before pushing it to the stack, so it wasn't preserved
as required by the PCS. We also sometimes did not push R11 and LR
adjacent to one another on the stack, or used R11 as a frame pointer
without pointing it at the saved value of R11, both of which are
required to have an AAPCS compliant frame chain.
The original work of this patch was done by James Westwood, reviewed as
#82801 and #81249, with some tidy-ups done by Mark Murray and myself.
Before b7b28e770c46, AreSupportedUsers will skip
MemTransferInst, it may cause unexpected assertion.
https://godbolt.org/z/z5d691fj1
In b7b28e770c46, we start to allow MemTransferInst,
we should allow it in adjustByValArgAlignment too.
The original goal of this pass was to focus on vector operations with
VLMAX. However, users often utilize only part of the result, and such
usage may come from the vectorizer.
We found that relaxing this constraint can capture more optimization
opportunities, such as non-power-of-2 code generation and vector
operation sequences with different VLs.
---------
Co-authored-by: Kito Cheng <kito.cheng@sifive.com>
This change is part of this proposal:
https://discourse.llvm.org/t/rfc-all-the-math-intrinsics/78294
Based on example PR #96222 and fix PR #101268, with some differences due
to 2-arg intrinsic and intermediate refactor (RuntimeLibCalls.cpp).
- Add llvm.experimental.constrained.atan2 - Intrinsics.td,
ConstrainedOps.def, LangRef.rst
- Add to ISDOpcodes.h and TargetSelectionDAG.td, connect to intrinsic in
BasicTTIImpl.h, and LibFunc_ in SelectionDAGBuilder.cpp
- Update LegalizeDAG.cpp, LegalizeFloatTypes.cpp, LegalizeVectorOps.cpp,
and LegalizeVectorTypes.cpp
- Update isKnownNeverNaN in SelectionDAG.cpp
- Update SelectionDAGDumper.cpp
- Update libcalls - RuntimeLibcalls.def, RuntimeLibcalls.cpp
- TargetLoweringBase.cpp - Expand for vectors, promote f16
- X86ISelLowering.cpp - Expand f80, promote f32 to f64 for MSVC
Part 4 for Implement the atan2 HLSL Function #70096.
Similar to #111287, this moves the UseOutlineAtomics legalization rules to a
boolean predicate as opposed to needing the be nested functions.
There appeared to be a pair of redundant customIfs for s128 sizes (assuming
only scalars are supported).
fp conversion V_CVT_F_F/V_CVT_F_U instructions true16 format were
previously implemented using fake16 profile.
With the MC support inplace, correct and support these instructions in
true16/fake16 format in CodeGen
Modify VOP3 profile and pesudo, and add encoding info for VOP3 True16
including DPP and DPP8 in true16 and fake16 format.
This patch applies true16/fake16 changes and asm/dasm changes to
V_ADD_NC_U16
V_ADD_NC_I16
V_SUB_NC_U16
V_SUB_NC_I16
The patch adds support for lround when the output type of the rounding
is i32.
The support for a rounding result of type i64 existed before this patch.
Under AArch64 it is common and will become more common to have operation
legalization rules dependant on a feature of the architecture. For
example HasFP16 or the newer CSSC integer min/max instructions, among
many others. With the current legalization rules this either means
adding a custom predicate based on the feature as in
`legalIf([=](const LegalityQuery &Query) { return HasFP16 && ...; }` or
splitting the legalization rules into pieces that place rules optionally
into them base on the features available.
This patch proposes an alternative where the existing routines like
legalFor(..) are provided a boolean predicate, which if false skips
adding the rule. It makes the rules cleaner and will hopefully allow
them to scale better as we add more features.
The SVE predicates for loads/stores I have changed to just be always
available. Scalable vectors without SVE have never been supported, but
it could also add a condition.
Fix a check for extending loads in DAGCombiner,
where if the result type has more bits than the
loaded type it should count as an extending load.
All backends apart from AArch64 ignore this
ExtTy argument to shouldReduceLoadWidth, so this
change currently only impacts AArch64.
Combine is needed to clear redundant ANDs with 1 that will be
created by reg-bank-select to clean-up high bits in register.
Fix replaceRegWith from CombinerHelper:
If copy had to be inserted, first create copy then delete MI.
If MI is deleted first insert point is not valid.
This aligns with GCC. LoongArch kernel developers requested that this
option generate some corresponding relations in a section, including the
addresses of the jump instruction(jr) and the `MachineJumpTableEntry`.
Reviewed By: heiher
Pull Request: https://github.com/llvm/llvm-project/pull/102411
- create a clang built-in in Builtins.td
- add semantic checking in SemaHLSL.cpp
- link the WaveReadLaneAt api in hlsl_intrinsics.h
- add lowering to spirv backend op GroupNonUniformShuffle
with Scope = 2 (Group) in SPIRVInstructionSelector.cpp
- add WaveReadLaneAt intrinsic to IntrinsicsDirectX.td and mapping
to DXIL.td
- add tests for HLSL intrinsic lowering to spirv intrinsic in
WaveReadLaneAt.hlsl
- add tests for sema checks in WaveReadLaneAt-errors.hlsl
- add spir-v backend tests in WaveReadLaneAt.ll
- add test to show scalar dxil lowering functionality
- note that this doesn't include support for the scalarizer to
handle WaveReadLaneAt will be added in a future pr
This is the first part #70104
The aim is to have the same set of promotions on fixed-length bf16
vectors as on fixed-length f16 vectors, and then deduplicate them
similarly to what was done for scalable vectors.
It looks like fneg/fabs/fcopysign end up getting expanded because fsub
is now legal, and the default operation action must be expand.
If `cvta.param` is used in regular functions, it may produce an
invalid pointer. It's unclear if it's a bug in ptxas or we're not using `cvta.param` correctly, but,
regardless of the underlying reason, the instruction has to be disabled for non-kernels, at least for now.
This PR fixes implementation of `SPIRVInstrInfo::analyzeBranch()` and
adds implementations of `SPIRVInstrInfo::removeBranch()` and
`SPIRVInstrInfo::insertBranch()` to support Branch Folding and If
Conversion optimization.
The attached test case failed before this PR due to report_fatal_error()
firing on missing implementation of `SPIRVInstrInfo::removeBranch()`.
The new test case is not able to pass spirv-val check at the moment due
to the issue described in
https://github.com/llvm/llvm-project/issues/110652 , this is not related
to this PR.
This PR also updates instructions definition in tablegen to set
isBranch=1 for relevant instructions.
A reduction instruction always has a passthru operand, so the scalar
operand should always be vs1 which is at index 3.
Even though the destination operand is also scalar, I think the passthru
will need to preserve all elements so I haven't included it.
This patch adds icmp+select patterns for integer min/max matchers in
SDPatternMatch, similar to those in IR PatternMatch.
Reapply #111774.
Closes#108218.
The iterator passed to `fixupCalleeSaveRestoreStackOffset` may be
incorrect when it tries to skip over the instructions that get the
current value of 'vg', when there is a 'rdsvl' instruction straight
after the prologue. That's because it doesn't check that the instruction
is still a 'frame-setup' instruction.
LDV could reorder reinserted fragment and non-fragment debug values for
the same variable (compared to the input order), potentially resulting
in stale values being presented.
For example, before:
DBG_VALUE 1001, $noreg, !13, !DIExpression(DW_OP_LLVM_fragment, 0, 16)
DBG_VALUE 1002, $noreg, !13, !DIExpression(DW_OP_LLVM_fragment, 16, 16)
DBG_VALUE %0, $noreg, !13, !DIExpression()
After (without this patch):
DBG_VALUE %stack.0, 0, !13, !DIExpression()
DBG_VALUE 1002, $noreg, !13, !DIExpression(DW_OP_LLVM_fragment, 16, 16)
DBG_VALUE 1001, $noreg, !13, !DIExpression(DW_OP_LLVM_fragment, 0, 16)
It would also reorder DBG_VALUEs for different variables. Although that
does not matter for the debug information output, it resulted in some
noise in before/after pass diffs.
This should hopefully align so that instruction referencing and
DBG_VALUE emit debug instructions in the same order (see the
sdag-salvage-add.ll change).
