[Clang][OpenCL] Compiler crash on __builtin_assume_aligned in OpenCL

[ritter-x2a]

Using the return value of __builtin_assume_aligned in OpenCL hits an assertion in clang. I don't have a strong opinion on whether the builtin should be supported in OpenCL since it's not part of the Khronos spec, but it shouldn't hit an assertion.
Observed with a RelWithDebInfo trunk build, on Ubuntu 22.04.

Reproducer:

void f(__global int *g) {
  __global int *ag = __builtin_assume_aligned(g, 16);
}

When compiling this via clang -c test.cl, clang first reports unexpected diagnostics (I don't see how bools are involved) and then hits an assertion:

test.cl:2:17: error: incompatible integer to pointer conversion initializing '__global int *__private' with an expression of type 'bool' [-Wint-conversion]
    2 |   __global int *ag = __builtin_assume_aligned(g, 16);
      |                 ^                             ~~~~~~
clang: /home/faritter/playground/llvm/llvm-project/llvm/lib/IR/Instructions.cpp:2974: static llvm::CastInst* llvm::CastInst::Create(llvm::Instruction::CastOps, llvm::Value*, llvm::Type*, const llvm::Twine&, llvm::InsertPosition): Assertion `castIsValid(op, S, Ty) && "Invalid cast!"' failed.
PLEASE submit a bug report to https://github.com/llvm/llvm-project/issues/ and include the crash backtrace, preprocessed source, and associated run script.
Stack dump:
0.	Program arguments: ./build/bin/clang -c test.cl
1.	<eof> parser at end of file
2.	test.cl:1:6: LLVM IR generation of declaration 'f'
3.	test.cl:1:6: Generating code for declaration 'f'
 #0 0x000056e464acc7ff llvm::sys::PrintStackTrace(llvm::raw_ostream&, int) /home/faritter/playground/llvm/llvm-project/llvm/lib/Support/Unix/Signals.inc:802:3
[...]
#13 0x000056e4636a9852 llvm::IRBuilderBase::CreateCast(llvm::Instruction::CastOps, llvm::Value*, llvm::Type*, llvm::Twine const&, llvm::MDNode*, llvm::FMFSource) /home/faritter/playground/llvm/llvm-project/llvm/include/llvm/IR/IRBuilder.h:2193:0
#14 0x000056e4636a9852 llvm::IRBuilderBase::CreateIntCast(llvm::Value*, llvm::Type*, bool, llvm::Twine const&) /home/faritter/playground/llvm/llvm-project/llvm/include/llvm/IR/IRBuilder.h:2231:0
#15 0x000056e464e2f677 (anonymous namespace)::ScalarExprEmitter::VisitCastExpr(clang::CastExpr*) /home/faritter/playground/llvm/llvm-project/clang/lib/CodeGen/CGExprScalar.cpp:2574:44
#16 0x000056e464e2c104 Visit /home/faritter/playground/llvm/llvm-project/clang/lib/CodeGen/CGExprScalar.cpp:449:3
#17 0x000056e464e2c104 clang::CodeGen::CodeGenFunction::EmitScalarExpr(clang::Expr const*, bool) /home/faritter/playground/llvm/llvm-project/clang/lib/CodeGen/CGExprScalar.cpp:5591:13
[...]

The full backtrace, preprocessed source, and run script are attached:

backtrace.txt

test-ff43fa.cl.txt

test-ff43fa.sh.txt

[llvmbot]

@llvm/issue-subscribers-opencl

Author: Fabian Ritter (ritter-x2a)

Using the return value of `__builtin_assume_aligned` in OpenCL hits an assertion in clang. I don't have a strong opinion on whether the builtin should be supported in OpenCL since it's not part of the Khronos spec, but it shouldn't hit an assertion. Observed with a RelWithDebInfo trunk build, on Ubuntu 22.04.

Reproducer:

void f(__global int *g) {
  __global int *ag = __builtin_assume_aligned(g, 16);
}

When compiling this via clang -c test.cl, clang first reports unexpected diagnostics (I don't see how bools are involved) and then hits an assertion:

test.cl:2:17: error: incompatible integer to pointer conversion initializing '__global int *__private' with an expression of type 'bool' [-Wint-conversion]
    2 |   __global int *ag = __builtin_assume_aligned(g, 16);
      |                 ^                             ~~~~~~
clang: /home/faritter/playground/llvm/llvm-project/llvm/lib/IR/Instructions.cpp:2974: static llvm::CastInst* llvm::CastInst::Create(llvm::Instruction::CastOps, llvm::Value*, llvm::Type*, const llvm::Twine&, llvm::InsertPosition): Assertion `castIsValid(op, S, Ty) && "Invalid cast!"' failed.
PLEASE submit a bug report to https://github.com/llvm/llvm-project/issues/ and include the crash backtrace, preprocessed source, and associated run script.
Stack dump:
0.	Program arguments: ./build/bin/clang -c test.cl
1.	<eof> parser at end of file
2.	test.cl:1:6: LLVM IR generation of declaration 'f'
3.	test.cl:1:6: Generating code for declaration 'f'
 #<!-- -->0 0x000056e464acc7ff llvm::sys::PrintStackTrace(llvm::raw_ostream&amp;, int) /home/faritter/playground/llvm/llvm-project/llvm/lib/Support/Unix/Signals.inc:802:3
[...]
#<!-- -->13 0x000056e4636a9852 llvm::IRBuilderBase::CreateCast(llvm::Instruction::CastOps, llvm::Value*, llvm::Type*, llvm::Twine const&amp;, llvm::MDNode*, llvm::FMFSource) /home/faritter/playground/llvm/llvm-project/llvm/include/llvm/IR/IRBuilder.h:2193:0
#<!-- -->14 0x000056e4636a9852 llvm::IRBuilderBase::CreateIntCast(llvm::Value*, llvm::Type*, bool, llvm::Twine const&amp;) /home/faritter/playground/llvm/llvm-project/llvm/include/llvm/IR/IRBuilder.h:2231:0
#<!-- -->15 0x000056e464e2f677 (anonymous namespace)::ScalarExprEmitter::VisitCastExpr(clang::CastExpr*) /home/faritter/playground/llvm/llvm-project/clang/lib/CodeGen/CGExprScalar.cpp:2574:44
#<!-- -->16 0x000056e464e2c104 Visit /home/faritter/playground/llvm/llvm-project/clang/lib/CodeGen/CGExprScalar.cpp:449:3
#<!-- -->17 0x000056e464e2c104 clang::CodeGen::CodeGenFunction::EmitScalarExpr(clang::Expr const*, bool) /home/faritter/playground/llvm/llvm-project/clang/lib/CodeGen/CGExprScalar.cpp:5591:13
[...]

The full backtrace, preprocessed source, and run script are attached:

backtrace.txt

test-ff43fa.cl.txt

test-ff43fa.sh.txt

[llvmbot]

@llvm/issue-subscribers-clang-codegen

Author: Fabian Ritter (ritter-x2a)

Using the return value of `__builtin_assume_aligned` in OpenCL hits an assertion in clang. I don't have a strong opinion on whether the builtin should be supported in OpenCL since it's not part of the Khronos spec, but it shouldn't hit an assertion. Observed with a RelWithDebInfo trunk build, on Ubuntu 22.04.

Reproducer:

void f(__global int *g) {
  __global int *ag = __builtin_assume_aligned(g, 16);
}

When compiling this via clang -c test.cl, clang first reports unexpected diagnostics (I don't see how bools are involved) and then hits an assertion:

test.cl:2:17: error: incompatible integer to pointer conversion initializing '__global int *__private' with an expression of type 'bool' [-Wint-conversion]
    2 |   __global int *ag = __builtin_assume_aligned(g, 16);
      |                 ^                             ~~~~~~
clang: /home/faritter/playground/llvm/llvm-project/llvm/lib/IR/Instructions.cpp:2974: static llvm::CastInst* llvm::CastInst::Create(llvm::Instruction::CastOps, llvm::Value*, llvm::Type*, const llvm::Twine&, llvm::InsertPosition): Assertion `castIsValid(op, S, Ty) && "Invalid cast!"' failed.
PLEASE submit a bug report to https://github.com/llvm/llvm-project/issues/ and include the crash backtrace, preprocessed source, and associated run script.
Stack dump:
0.	Program arguments: ./build/bin/clang -c test.cl
1.	<eof> parser at end of file
2.	test.cl:1:6: LLVM IR generation of declaration 'f'
3.	test.cl:1:6: Generating code for declaration 'f'
 #<!-- -->0 0x000056e464acc7ff llvm::sys::PrintStackTrace(llvm::raw_ostream&amp;, int) /home/faritter/playground/llvm/llvm-project/llvm/lib/Support/Unix/Signals.inc:802:3
[...]
#<!-- -->13 0x000056e4636a9852 llvm::IRBuilderBase::CreateCast(llvm::Instruction::CastOps, llvm::Value*, llvm::Type*, llvm::Twine const&amp;, llvm::MDNode*, llvm::FMFSource) /home/faritter/playground/llvm/llvm-project/llvm/include/llvm/IR/IRBuilder.h:2193:0
#<!-- -->14 0x000056e4636a9852 llvm::IRBuilderBase::CreateIntCast(llvm::Value*, llvm::Type*, bool, llvm::Twine const&amp;) /home/faritter/playground/llvm/llvm-project/llvm/include/llvm/IR/IRBuilder.h:2231:0
#<!-- -->15 0x000056e464e2f677 (anonymous namespace)::ScalarExprEmitter::VisitCastExpr(clang::CastExpr*) /home/faritter/playground/llvm/llvm-project/clang/lib/CodeGen/CGExprScalar.cpp:2574:44
#<!-- -->16 0x000056e464e2c104 Visit /home/faritter/playground/llvm/llvm-project/clang/lib/CodeGen/CGExprScalar.cpp:449:3
#<!-- -->17 0x000056e464e2c104 clang::CodeGen::CodeGenFunction::EmitScalarExpr(clang::Expr const*, bool) /home/faritter/playground/llvm/llvm-project/clang/lib/CodeGen/CGExprScalar.cpp:5591:13
[...]

The full backtrace, preprocessed source, and run script are attached:

backtrace.txt

test-ff43fa.cl.txt

test-ff43fa.sh.txt

[svenvh]

Something seems to be wrong in the AST already:

|         `-CallExpr 0x5588757c1078 <col:47, col:52> 'bool'                                                                                                                                                        
|           |-ImplicitCastExpr 0x5588757c1060 <<invalid sloc>> 'void *(*)(const __global void *, unsigned long, ...)' <FunctionToPointerDecay>                                                                     
|           | `-DeclRefExpr 0x5588757c1008 <<invalid sloc>> 'void *(const __global void *, unsigned long, ...)' Function 0x5588757c0dd8 '__builtin_assume_aligned' 'void *(const __global void *, unsigned long, ...)'

whereas one would expect it to look more like the following:

|         `-CallExpr 0x56528cc36878 <col:13, col:43> 'void *'                                                                                                                                                      
|           |-ImplicitCastExpr 0x56528cc36860 <col:13> 'void *(*)(const void *, unsigned long, ...)' <BuiltinFnToFnPtr>                                                                                            
|           | `-DeclRefExpr 0x56528cc367b8 <col:13> '<builtin fn type>' Function 0x56528cc36588 '__builtin_assume_aligned' 'void *(const void *, unsigned long, ...)'

[svenvh]

I think we're hitting a known limitation of rewriteBuiltinFunctionDecl here, which still has a

/// TODO: Handle pointer return types.