Many tests had this flag removed because of the G_BITCAST emission
issue. Now that the PR is merged, we can re-enable this additional
check.
2 tests (basic_int_types) just have the TODO removed because they are
not useful for SPIR-V as-is: SPIR-V requires reg2mem/mem2reg to run,
which removes all the body. Integers are used in other spirv tests, and
seems like testing for spirv32/64 and relying on others for the logical
target coverage should be fine.
Signed-off-by: Nathan Gauër <brioche@google.com>
This PR implements instruction selection for G_BITCAST on an earlier
stage to avoid MachineVerifier complains on subtle semantics difference
between G_BITCAST and OpBitcast.
We do instruction selections for OpBitcast after IR Translation instead
of calling MIB.buildBitcast() generating the general op code G_BITCAST,
because when MachineVerifier validates G_BITCAST we see a check of a
kind: 'if Source Type is equal to Destination Type then report error
"bitcast must change the type"'. This doesn't take into account the
notion of a typed pointer that is important for SPIR-V where a user may
and should use bitcast between pointers with different pointee types
(https://registry.khronos.org/SPIR-V/specs/unified1/SPIRV.html#OpBitcast).
It's important for correct lowering in SPIR-V, because interpretation of
the data type is not left to instructions that utilize the pointer, but
encoded by the pointer declaration, and the SPIRV target can and must
handle the declaration and use of pointers that specify the type of data
they point to.
It's not feasible to improve validation of G_BITCAST using just
information provided by low level types of source and destination.
Therefore we don't produce G_BITCAST as the general op code with
semantics different from OpBitcast, but rather lower to OpBitcast
immediately.
See discussion in https://github.com/llvm/llvm-project/pull/110270 for
even more context.
G_BITCAST emission in the SPIR-V backend is not accepted by the
verifier. DIsabling verifier for impacted tests until
https://github.com/llvm/llvm-project/pull/114216 is merged.
Signed-off-by: Nathan Gauër <brioche@google.com>
The "topological" sorting was behaving incorrectly in some cases:
the exit of a loop could have a lower rank than a node in the loop.
This causes issues when structurizing some patterns, and also codegen
issues as we could generate BBs in the incorrect order in regard to the
SPIR-V spec.
Fixing this ordering alone broke other parts of the structurizer, which
by luck worked. Had to fix those.
Added more test cases, especially to test basic patterns.
I also needed to tweak/disable some tests for 2 reasons:
- SPIR-V now required reg2mem/mem2reg to run. Meaning dead stores
are optimized away. Some tests require tweaks to avoid having the
whole function removed.
- Mem2Reg will generate variable & load/stores. This generates
G_BITCAST in several cases. And there is currently something wrong
we do with G_BITCAST which causes MIR verifier to complain.
Until this is resolved, I disabled -verify-machineinstrs flag on
those tests.
---------
Signed-off-by: Nathan Gauër <brioche@google.com>
Line ending policies were changed in the parent, dccebddb3b80. To make
it easier to resolve downstream merge conflicts after line-ending
policies are adjusted this is a separate whitespace-only commit. If you
have merge conflicts as a result, you can simply `git add --renormalize
-u && git merge --continue` or `git add --renormalize -u && git rebase
--continue` - depending on your workflow.
- 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
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.
- add degrees builtin
- link degrees api in hlsl_intrinsics.h
- add degrees intrinsic to IntrinsicsDirectX.td
- add degrees intrinsic to IntrinsicsSPIRV.td
- add lowering from clang builtin to dx/spv intrinsics in CGBuiltin.cpp
- add semantic checks to SemaHLSL.cpp
- add expansion of directx intrinsic to llvm fmul for DirectX in
DXILIntrinsicExpansion.cpp
- add mapping to spir-v intrinsic in SPIRVInstructionSelector.cpp
- add test coverage:
- degrees.hlsl -> check hlsl lowering to dx/spv degrees intrinsics
- degrees-errors.hlsl/half-float-only-errors -> check semantic warnings
- hlsl-intrinsics/degrees.ll -> check lowering of spir-v degrees
intrinsic to SPIR-V backend
- DirectX/degrees.ll -> check expansion and scalarization of directx
degrees intrinsic to fmul
Resolves#99104
Implement the intrinsic `llvm.spv.handle.fromBinding`, which returns the
handle for a global resource. This involves creating a global variable
that matches the return-type, set, and binding in the call, and
returning the handle to that resource.
This commit implements the scalar version. It does not handle arrays of
resources yet. It also does not handle storage buffers yet. We do not
have the type for the storage buffers designed yet.
Part of #81036
Reverts llvm/llvm-project#110800
`llvm\test\CodeGen\DirectX\radians.ll` is failing after this change.
@adam-yang please send a new PR with the issue resolved once you've had
time to investigate.
The SPIRV backend has a special type named `spirv.Image`. This type is
meant to correspond to the OpTypeImage instruction in SPIR-V, but there
is one difference. The access qualifier operand in OpTypeImage is
optional. On top of that, the access qualifiers are only valid for
kernels, and not for shaders.
We want to reuse this type when generating shader from HLSL, but we
can't use the access qualifier. This commit make the access qualifer
optional in the target extension type.
The same is done for `spirv.SampledImage`.
Contributes to #81036
There are llvm intrinsics which we do not expect to actually represent
code after lowering or which are not implemented yet but can be found in
a customer's LLVM IR input. We do not want translation to crash when
these llvm intrinsics are found, and this PR fixes the issue with
translation crash for some known cases, aligned with Khronos Translator.
This PR reworks implementation of OpSpecConstantOp with ptr-cast
operation (PtrCastToGeneric, GenericCastToPtr). Previous implementation
didn't take into account a lot of use cases, including multiple
inclusion of pointers, reference to a pointer from OpName, etc. A
reproducer is attached as a new test case.
This PR also fixes wrong type inference for IR patterns which generate
new virtual registers without SPIRV type. Previous implementation
assumed always that result has the same address space as a source that
is not the fact, and, for example, led to impossibility to emit a
ptr-cast operation in the reproducer, because wrong type inference
rendered source and destination with the same address space, eliminating
translation of G_ADDRSPACE_CAST.
This PR improves type inference and fixes inconsistency between
previously deduced element type of a pointer and function's return type.
It fixes https://github.com/llvm/llvm-project/issues/109401 by ensuring
that OpPhi is consistent with respect to operand types.
Windows have different separators for paths than Unix based OS. One of
the tests in debug-compilation-unit.ll didn't have Win supported '\\'
variant which broken test suite on that OS.
Two main goals of this PR are:
* to support "Arithmetic with Overflow" intrinsics, including the
special case when those intrinsics are being generated by the
CodeGenPrepare pass during translations with optimization;
* to redirect intrinsics with aggregate return type to be lowered via
GlobalISel operations instead of SPIRV-specific unfolding/lowering (see
https://github.com/llvm/llvm-project/pull/95012).
There is a new test case
`llvm/test/CodeGen/SPIRV/passes/translate-aggregate-uaddo.ll` that
describes and checks the general logics of the translation.
This PR continues a series of PRs aimed to identify and fix flaws in
code emission, to improve pass rates for the mode with expensive checks
set on (see https://github.com/llvm/llvm-project/pull/101732,
https://github.com/llvm/llvm-project/pull/104104,
https://github.com/llvm/llvm-project/pull/106966), having in mind the
ultimate goal of proceeding towards the non-experimental status of
SPIR-V Backend.
The reproducers are:
1) consider `llc -O3 -mtriple=spirv64-unknown-unknown ...` with:
```
define spir_func i32 @foo(i32 %a, ptr addrspace(4) %p) {
entry:
br label %l1
l1:
%e = phi i32 [ %a, %entry ], [ %i, %body ]
%i = add nsw i32 %e, 1
%fl = icmp eq i32 %i, 0
br i1 %fl, label %exit, label %body
body:
store i8 42, ptr addrspace(4) %p
br label %l1
exit:
ret i32 %i
}
```
2) consider `llc -O0 -mtriple=spirv64-unknown-unknown ...` with:
```
define spir_func i32 @foo(i32 %a, ptr addrspace(4) %p) {
entry:
br label %l1
l1: ; preds = %body, %entry
%e = phi i32 [ %a, %entry ], [ %math, %body ]
%0 = call { i32, i1 } @llvm.uadd.with.overflow.i32(i32 %e, i32 1)
%math = extractvalue { i32, i1 } %0, 0
%ov = extractvalue { i32, i1 } %0, 1
br i1 %ov, label %exit, label %body
body: ; preds = %l1
store i8 42, ptr addrspace(4) %p, align 1
br label %l1
exit: ; preds = %l1
ret i32 %math
}
```
This change adds support for correctly lowering the `__scoped` Clang
builtins, and corresponding scoped LLVM instructions. These were
previously unconditionally lowered to Device scope, which is possibly incorrect.
Furthermore, the default / implicit scope is changed from Device (an
OpenCL assumption) to AllSvmDevices (aka System), since the SPIR-V BE is not
OpenCL specific / can ingest IR coming from other language front-ends. OpenCL
defaulting to Device scope is now reflected in the front-end handling of atomic
ops, which seems preferable.
AsmPrinter always creates a symbol for the end of function if valid
debug info is present. However, this breaks SPIR-V target's output,
because SPIR-V specification allows label instructions only inside a
block, not after the function body (see
https://registry.khronos.org/SPIR-V/specs/unified1/SPIRV.html#OpLabel).
This PR proposes to disable emission of label instructions after the
function body if the target is SPIR-V.
This PR is a fix of the
https://github.com/llvm/llvm-project/issues/102732 issue.
This commit adds an initial SPIR-V structurizer.
It leverages the previously merged passes, and the convergence region
analysis to determine the correct merge and continue blocks for SPIR-V.
The first part does a branch cleanup (simplifying switches, and
legalizing them), then merge instructions are added to cycles,
convergent and later divergent blocks.
Then comes the important part: splitting critical edges, and making sure
the divergent construct boundaries don't cross.
- we split blocks with multiple headers into 2 blocks.
- we split blocks that are a merge blocks for 2 or more constructs:
SPIR-V spec disallow a merge block to be shared by 2
loop/switch/condition construct.
- we split merge & continue blocks: SPIR-V spec disallow a basic block
to be both a continue block, and a merge block.
- we remove superfluous headers: when a header doesn't bring more info
than the parent on the divergence state, it must be removed.
This PR leverages the merged SPIR-V simulator for testing, as long as
spirv-val. For now, most DXC structurization tests are passing. The
unsupported ones are either caused by unsupported features like switches
on boolean types, or switches in region exits, because the MergeExit
pass doesn't support those yet (there is a FIXME).
This PR is quite large, and the addition not trivial, so I tried to keep
it simple. E.G: as soon as the CFG changes, I recompute the dominator
trees and other structures instead of updating them.
---------
Signed-off-by: Nathan Gauër <brioche@google.com>
Add new elementwise popcount builtin to support HLSL function
'countbits'.
elementwise popcount only accepts integer types.
Add hlsl intrinsic 'countbits'
Closes#99094
The commit introduces support for fundamental DI instruction. Metadata
handlers required for this instruction is stored inside debug records
(https://llvm.org/docs/SourceLevelDebugging.html) parts of the module
which rises the necessity of it's traversal.
When running SPIR-V Backend with optimization levels higher than 0, we
observe GEP Operator's as a new factor, massively used to convey the
semantics of the original LLVM IR. Previously, an issue related to GEP
Operator was mentioned and fixed on the consumer side of toolchains
(see, for example, Khronos Trandslator Issue
https://github.com/KhronosGroup/SPIRV-LLVM-Translator/issues/2486 and PR
https://github.com/KhronosGroup/SPIRV-LLVM-Translator/pull/2487).
However, there is a case when GenCode creates G_PTR_ADD to convey the
original semantics under optimization levels higher than 0 where it's
SPIR-V Backend that fails to translate source LLVM IR correctly.
Consider the following reproducer:
```
%struct = type { i32, [257 x i8], [257 x i8], [129 x i8], i32, i64, i64, i64, i64, i64, i64 }
@Mem = linkonce_odr dso_local addrspace(1) global %struct zeroinitializer, align 8
define weak dso_local spir_func void @__devicelib_assert_fail(ptr addrspace(4) noundef %expr, i32 noundef %line, i1 %fl) {
entry:
%cmp = icmp eq i32 %line, 0
br i1 %cmp, label %lbl, label %exit
lbl:
store i32 %line, ptr addrspace(1) getelementptr inbounds (i8, ptr addrspace(1) @Mem, i64 648), align 8
br i1 %fl, label %lbl, label %exit
exit:
ret void
}
```
converted to the following machine instructions by SPIR-V Backend:
```
%4:type(s64) = OpTypeInt 32, 0
%22:type(s64) = OpTypePointer 5, %4:type(s64)
%2:type(s64) = OpTypeInt 8, 0
%28:type(s64) = OpTypePointer 5, %2:type(s64)
%10:pid(p1) = G_GLOBAL_VALUE @Mem
%36:type(s64) = OpTypeStruct %4:type(s64), %32:type(s64), %32:type(s64), %34:type(s64), %4:type(s64), %35:type(s64), %35:type(s64), %35:type(s64), %35:type(s64), %35:type(s64), %35:type(s64)
%37:iid(s32) = G_INTRINSIC_W_SIDE_EFFECTS intrinsic(@llvm.spv.const.composite)
%8:iid(s32) = ASSIGN_TYPE %37:iid(s32), %36:type(s64)
G_INTRINSIC_W_SIDE_EFFECTS intrinsic(@llvm.spv.init.global), %10:pid(p1), %8:iid(s32)
%29:pid(p1) = nuw G_PTR_ADD %10:pid, %16:iid(s64)
%15:pid(p1) = nuw ASSIGN_TYPE %29:pid(p1), %28:type(s64)
%27:pid(p2) = G_BITCAST %15:pid(p1)
%17:pid(p2) = ASSIGN_TYPE %27:pid(p2), %22:type(s64)
G_STORE %1:iid(s32), %17:pid(p2) :: (store (s32) into %ir.3, align 8, addrspace 1)
```
On the next stage of instruction selection this `G_PTR_ADD`-related
pattern would be interpreted as an initialization of a global variable
and converted to an invalid constant GEP pattern that, in its turn,
would fail to be verified by LLVM during back translation from SPIR-V to
LLVM IR.
This PR introduces a fix for the problem by adding one more case of
`G_PTR_ADD` translation, when we use a non-const GEP to convey the
meaning. The reproducer is attached as a new test case.
This commits add the WaveIsFirstLane() hlsl intrinsinc. This intrinsic
uses the convergence intrinsincs for the SPIR-V backend. On the DXIL
side, I'm not sure what the strategy is for convergence, so I
implemented that like in DXC: a normal builtin function.
Signed-off-by: Nathan Gauër <brioche@google.com>
This PR is to ensure that OpExtInst instructions generated by
NonSemantic_Shader_DebugInfo_100 are not mixed up with other OpExtInst
instructions.
Original implementation
(https://github.com/llvm/llvm-project/pull/97558) has introduced an
issue by moving OpExtInst instruction with the 3rd operand equal to
DebugSource (value 35) or DebugCompilationUnit (value 1) even if
OpExtInst is not generated by NonSemantic_Shader_DebugInfo_100
implementation code.
The reproducer is attached as a new test case. The code of the test case
reproduces the issue, because "lgamma" has the same code (35) inside
OpenCL_std as DebugSource inside NonSemantic_Shader_DebugInfo_100.
This PR improves correctness of emitted MIR between passes for branching
instructions and thus increase number of passing tests when expensive
checks are on. Specifically, we address here such issues with machine
verifier as:
* fix switch generation: generate correct successors and undo the
"address taken" status to reflect that a successor doesn't actually
correspond to an IR-level basic block;
* fix incorrect definition of OpBranch and OpBranchConditional in
TableGen (SPIRVInstrInfo.td) to set isBarrier status properly and set a
correct type of virtual registers;
* fix a case when Phi refers to a type definition that goes after the
Phi instruction, so that the virtual register definition of the type
doesn't dominate all uses.
This PR decrease number of failing tests under expensive checks from 56
to 50.
OpenCL's vload_half builtin expects two arguments, but the current
TableGen definition expects three.
This change fixes the mismatch and adds a test to check this.
I don't know if this is the right thing to do or if the emitter should
be normalizing path separators, but this gets this test to pass on
windows where `sys::path::append` will use "\" instead of "/".
This is another follow up to #97558, as #105889 managed to fix the
build, but not all of the tests.
This commit introduces emission of DebugSource, DebugCompileUnit from
NonSemantic.Shader.DebugInfo.100 and required OpString with filename.
NonSemantic.Shader.DebugInfo.100 is divided, following DWARF into two
main concepts – emitting DIE and Line.
In DWARF .debug_abbriev and .debug_info sections are responsible for
emitting tree with information (DEIs) about e.g. types, compilation
unit. Corresponding to that in NonSemantic.Shader.DebugInfo.100 have
instructions like DebugSource, DebugCompileUnit etc. which preforms same
role in SPIR-V file. The difference is in fact that in SPIR-V there are
no sections but logical layout which forces order of the instruction
emission.
The NonSemantic.Shader.DebugInfo.100 requires for this type of global
information to be emitted after OpTypeXXX and OpConstantXXX
instructions.
One of the goals was to minimize changes and interaction with
SPIRVModuleAnalysis as possible which current commit achieves by
emitting it’s instructions directly into MachineFunction.
The possibility of duplicates are mitigated by guard inside pass which
emits the global information only once in one function.
By that method duplicates don’t have chance to be emitted.
From that point, adding new debug global instructions should be
straightforward.
This adds the SPIRV fdot, sdot, and udot intrinsics and allows them to
be created at codegen depending on the target architecture. This
required moving some of the DXIL-specific choices to DXIL instruction
expansion out of codegen and providing it with at a more generic fdot
intrinsic as well.
Removed some stale comments that gave the obsolete impression that type
conversions should be expected to match overloads.
The SPIRV intrinsic handling involves generating multiply and add
operations for integers and the existing OpDot operation for floating
point.
New tests for generating SPIRV float and integer dot intrinsics are
added as well as expanding HLSL tests to include SPIRV generation
Used new dot product intrinsic generation to implement normalize() in SPIRV
Incidentally changed existing dot intrinsic definitions to use
DefaultAttrsIntrinsic to match the newly added inrinsics
Fixes#88056
This PR continues https://github.com/llvm/llvm-project/pull/101732
changes in virtual register processing aimed to improve correctness of
emitted MIR between passes from the perspective of MachineVerifier.
Namely, the following changes are introduced:
* register classes (lib/Target/SPIRV/SPIRVRegisterInfo.td) and
instruction patterns (lib/Target/SPIRV/SPIRVInstrInfo.td) are corrected
and simplified (by removing unnecessary sophisticated options) -- e.g.,
this PR gets rid of duplicating 32/64 bits patterns, removes ANYID
register class and simplifies definition of the rest of register
classes,
* hardcoded LLT scalar types in passes before instruction selection are
corrected -- the goal is to have correct bit width before instruction
selection, and use 64 bits registers for pattern matching in the
instruction selection pass; 32-bit registers remain where they are
described in such terms by SPIR-V specification (like, for example,
creation of virtual registers for scope/mem semantics operands),
* rework virtual register type/class assignment for calls/builtins
lowering,
* a series of minor changes to fix validity of emitted code between
passes:
- ensure that that bitcast changes the type,
- fix the pattern for instruction selection for OpExtInst,
- simplify inline asm operands usage,
- account for arbitrary integer sizes / update legalizer rules;
* add '-verify-machineinstrs' to existed test cases.
See also https://github.com/llvm/llvm-project/issues/88129 that this PR
may resolve.
This PR fixes a great number of issues reported by MachineVerifier and,
as a result, reduces a number of failed test cases for the mode with
expensive checks set on from ~200 to ~57.
The SPIR-V spec required basic blocks to respect some kind of ordering
(A block dominating another cannot be after in the binary layout).
---------
Signed-off-by: Nathan Gauër <brioche@google.com>
Implement support for HLSL intrinsic saturate.
Implement DXIL codegen for the intrinsic saturate by lowering it to DXIL
Op dx.saturate.
Implement SPIRV codegen by transforming saturate(x) to clamp(x, 0.0f,
1.0f).
Add tests for DXIL and SPIRV CodeGen.
This PR addresses a TODO in
lib/Target/SPIRV/SPIRVInstructionSelector.cpp by adding implementation
of the non-const G_BUILD_VECTOR, and fix emission of the
OpGroupBroadcast instruction for the case when the `..._group_broadcast`
builtin has more than one `local_id` argument and `OpGroupBroadcast`
requires a newly constructed vector with 2 or 3 components instead of
originally passed series of `local_id` arguments.
This PR may resolve https://github.com/llvm/llvm-project/issues/97310 if
the reason for the reported fail is an incorrectly generated
OpGroupBroadcast instruction that was definitely a case.
Existing test is hardened and a new test is added to cover this special
case of the OpGroupBroadcast instruction emission.
