Skip to content

[clang][bytecode] Support vector-to-vector bitcasts #118230

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Dec 2, 2024
Merged

[clang][bytecode] Support vector-to-vector bitcasts #118230

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions clang/lib/AST/ByteCode/Compiler.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -448,8 +448,8 @@ bool Compiler<Emitter>::VisitCastExpr(const CastExpr *CE) {

QualType SubExprTy = SubExpr->getType();
std::optional<PrimType> FromT = classify(SubExprTy);
// Casts from integer to vectors in C.
if (FromT && CE->getType()->isVectorType())
// Casts from integer/vector to vector.
if (CE->getType()->isVectorType())
return this->emitBuiltinBitCast(CE);

std::optional<PrimType> ToT = classify(CE->getType());
Expand Down Expand Up @@ -6502,7 +6502,7 @@ bool Compiler<Emitter>::emitBuiltinBitCast(const CastExpr *E) {
// we later assume it to be one (i.e. a PT_Ptr). However,
// we call this function for other utility methods where
// a bitcast might be useful, so convert it to a PT_Ptr in that case.
if (SubExpr->isGLValue()) {
if (SubExpr->isGLValue() || FromType->isVectorType()) {
if (!this->visit(SubExpr))
return false;
} else if (std::optional<PrimType> FromT = classify(SubExpr)) {
Expand Down
19 changes: 19 additions & 0 deletions clang/test/AST/ByteCode/altivec.c
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,19 @@
// RUN: %clang_cc1 -triple=powerpc64-unknown-linux-gnu -target-feature +altivec -target-feature +vsx -fsyntax-only -verify=expected,both %s
// RUN: %clang_cc1 -triple=powerpc64le-unknown-linux-gnu -target-feature +altivec -target-feature -vsx -fsyntax-only -verify=expected,both %s
// RUN: %clang_cc1 -triple=powerpc-ibm-aix -target-feature +altivec -fsyntax-only -verify=expected,both %s
// RUN: %clang_cc1 -triple=powerpc64-ibm-aix -target-feature +altivec -target-feature -vsx -fsyntax-only -verify=expected,both %s

// RUN: %clang_cc1 -triple=powerpc64-unknown-linux-gnu -target-feature +altivec -target-feature +vsx -fsyntax-only -verify=expected,both -fexperimental-new-constant-interpreter %s
// RUN: %clang_cc1 -triple=powerpc64le-unknown-linux-gnu -target-feature +altivec -target-feature -vsx -fsyntax-only -verify=expected,both -fexperimental-new-constant-interpreter %s
// RUN: %clang_cc1 -triple=powerpc-ibm-aix -target-feature +altivec -fsyntax-only -verify=expected,both -fexperimental-new-constant-interpreter %s
// RUN: %clang_cc1 -triple=powerpc64-ibm-aix -target-feature +altivec -target-feature -vsx -fsyntax-only -verify=expected,both -fexperimental-new-constant-interpreter %s

// both-no-diagnostics.

/// From test/Parser/altivec.c
vector char v1 = (vector char)((vector int)(1, 2, 3, 4));
vector char v2 = (vector char)((vector float)(1.0f, 2.0f, 3.0f, 4.0f));
vector char v3 = (vector char)((vector int)('a', 'b', 'c', 'd'));
vector int v4 = (vector int)(1, 2, 3, 4);
vector float v5 = (vector float)(1.0f, 2.0f, 3.0f, 4.0f);
vector char v6 = (vector char)((vector int)(1+2, -2, (int)(2.0 * 3), -(5-3)));
3 changes: 1 addition & 2 deletions clang/test/AST/ByteCode/vectors.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,10 +55,9 @@ namespace Vector {
static_assert(__builtin_vectorelements(v2) == (32 / sizeof(double)), "");
}

/// FIXME: We need to support BitCasts between vector types.
namespace {
typedef float __attribute__((vector_size(16))) VI42;
constexpr VI42 A2 = A; // expected-error {{must be initialized by a constant expression}}
constexpr VI42 A2 = A;
}

namespace BoolToSignedIntegralCast{
Expand Down