@@ -4912,3 +4922,279 @@ void AutoType::Profile(llvm::FoldingSetNodeID &ID,
const ASTContext &Context) {
Profile(ID, Context, getDeducedType(), getKeyword(), isDependentType(),
getTypeConstraintConcept(), getTypeConstraintArguments());
}
+
+FunctionEffect::~Function
@@ -3922,6 +3922,42 @@ bool Sema::MergeFunctionDecl(FunctionDecl *New,
NamedDecl *&OldD, Scope *S,
return true;
}
+ const auto OldFX = Old->getFunctionEffects();
+ const auto NewFX = New->getFunctionEffects();
+ if (OldFX != NewFX) {
+const auto Diffs = FunctionE
@@ -2380,6 +2382,1239 @@ class UnsafeBufferUsageReporter : public
UnsafeBufferUsageHandler {
};
} // namespace
+//
=
+
+// Temporary debugging option
+#define FX_ANALYZER_VERIFY_DECL_LIST 1
+
+namesp
@@ -3922,6 +3922,42 @@ bool Sema::MergeFunctionDecl(FunctionDecl *New,
NamedDecl *&OldD, Scope *S,
return true;
}
+ const auto OldFX = Old->getFunctionEffects();
+ const auto NewFX = New->getFunctionEffects();
+ if (OldFX != NewFX) {
+const auto Diffs = FunctionE
https://github.com/dougsonos edited
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -3922,6 +3922,42 @@ bool Sema::MergeFunctionDecl(FunctionDecl *New,
NamedDecl *&OldD, Scope *S,
return true;
}
+ const auto OldFX = Old->getFunctionEffects();
+ const auto NewFX = New->getFunctionEffects();
+ if (OldFX != NewFX) {
+const auto Diffs = FunctionE
https://github.com/dougsonos edited
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -3922,6 +3922,42 @@ bool Sema::MergeFunctionDecl(FunctionDecl *New,
NamedDecl *&OldD, Scope *S,
return true;
}
+ const auto OldFX = Old->getFunctionEffects();
+ const auto NewFX = New->getFunctionEffects();
+ if (OldFX != NewFX) {
+const auto Diffs = FunctionE
@@ -3922,6 +3922,42 @@ bool Sema::MergeFunctionDecl(FunctionDecl *New,
NamedDecl *&OldD, Scope *S,
return true;
}
+ const auto OldFX = Old->getFunctionEffects();
+ const auto NewFX = New->getFunctionEffects();
+ if (OldFX != NewFX) {
+const auto Diffs = FunctionE
@@ -3922,6 +3922,42 @@ bool Sema::MergeFunctionDecl(FunctionDecl *New,
NamedDecl *&OldD, Scope *S,
return true;
}
+ const auto OldFX = Old->getFunctionEffects();
+ const auto NewFX = New->getFunctionEffects();
+ if (OldFX != NewFX) {
+const auto Diffs = FunctionE
https://github.com/dougsonos edited
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/dougsonos edited
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -3922,6 +3922,42 @@ bool Sema::MergeFunctionDecl(FunctionDecl *New,
NamedDecl *&OldD, Scope *S,
return true;
}
+ const auto OldFX = Old->getFunctionEffects();
+ const auto NewFX = New->getFunctionEffects();
+ if (OldFX != NewFX) {
+const auto Diffs = FunctionE
https://github.com/dougsonos edited
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/dougsonos edited
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
dougsonos wrote:
> I would maybe try going with that then for now (and maybe add a comment about
> that too); I’m not sure my function pointer example is really the same
> situation, but I remember finding an example that was analogous but for
> `noreturn` _somwhere_ in the test cases, but I d
Sirraide wrote:
> Just tried. In C mode, it works to do nothing about effect changes in
> isFunctionConversion (!).
I would maybe try going with that then for now (and maybe add a comment about
that too); I’m not sure my function pointer example is really the same
situation, but I remember fi
dougsonos wrote:
> > have you tried maybe simply not setting `Changed` to `true` (perhaps only
> > in C mode)?
Just tried. In C mode, it works to do nothing about effect changes in
isFunctionConversion (!).
https://github.com/llvm/llvm-project/pull/84983
__
Sirraide wrote:
> Adapted from `test/Sema/initialize-noreturn.c`
Seeing as that test does check that we’re issuing an error for this, and seeing
that this seems to be handled in `isFunctionConversion()`, it really just seems
like that function is supposed to error if there is a conversion (i.e
Sirraide wrote:
>
> I was testing with `noreturn` earlier. It's hard to compare, because (at
> least in C23):
>
> ```
> ../clang/test/Sema/attr-nolock-wip.cpp:15:19: error: 'noreturn' attribute
> cannot be applied to types
>15 | void noret(int) [[noreturn]];
> | ^
dougsonos wrote:
> > have you tried maybe simply not setting `Changed` to `true` (perhaps only
> > in C mode)?
>
> It’s worth pointing out that C also warns (and C23 straight-up _errors_) if
> you replace `nolock` w/ `noreturn` iirc, so either that’s also a bug or
> that’s intended; I think I
Sirraide wrote:
> swapping the two didn’t seem to change anything
In one specific place where this is called in C mode, that is. From what I
could see, the ordering seems to be correct for other calls to this function.
https://github.com/llvm/llvm-project/pull/84983
___
Sirraide wrote:
> have you tried maybe simply not setting `Changed` to `true` (perhaps only in
> C mode)?
It’s worth pointing out that C also warns (and C23 straight-up *errors*) if you
replace `nolock` w/ `noreturn` iirc, so either that’s also a bug or that’s
intended; I think I was going to
Sirraide wrote:
> the naming of "From" and "To" seems backwards.
I noticed that as well; as I recall, swapping the two didn’t seem to change
anything. I might open a pr or issue about this to either fix this or at least
get some clarification as to whether that was really intended. I got a bit
Sirraide wrote:
> I've probably been staring at this way too long, but here's what's going on.
> My test is:
My bad, I remember looking into this yesterday or the day before; from what I
could tell, it seems I may have forgotten to share what I found here: I think
the problem is that the code
dougsonos wrote:
> > * Conversions between functions / function pointers with and without the
> > attribute work in C++ but not in C and I'm a bit lost (there's a bunch of
> > debug logging around this e.g. Sema::IsFunctionConversion and
> > Sema::ImpCastExprToType)
I've probably been staring
@@ -3922,6 +3922,42 @@ bool Sema::MergeFunctionDecl(FunctionDecl *New,
NamedDecl *&OldD, Scope *S,
return true;
}
+ const auto OldFX = Old->getFunctionEffects();
+ const auto NewFX = New->getFunctionEffects();
+ if (OldFX != NewFX) {
+const auto Diffs = FunctionE
@@ -3922,6 +3922,42 @@ bool Sema::MergeFunctionDecl(FunctionDecl *New,
NamedDecl *&OldD, Scope *S,
return true;
}
+ const auto OldFX = Old->getFunctionEffects();
+ const auto NewFX = New->getFunctionEffects();
+ if (OldFX != NewFX) {
+const auto Diffs = FunctionE
https://github.com/dougsonos edited
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -10778,6 +10778,101 @@ def warn_imp_cast_drops_unaligned : Warning<
"implicit cast from type %0 to type %1 drops __unaligned qualifier">,
InGroup>;
+def warn_func_effect_allocates : Warning<
+ "'%0' function '%1' must not allocate or deallocate memory">,
+ InGroup;
+
@@ -3922,6 +3922,42 @@ bool Sema::MergeFunctionDecl(FunctionDecl *New,
NamedDecl *&OldD, Scope *S,
return true;
}
+ const auto OldFX = Old->getFunctionEffects();
+ const auto NewFX = New->getFunctionEffects();
+ if (OldFX != NewFX) {
+const auto Diffs = FunctionE
@@ -10778,6 +10778,101 @@ def warn_imp_cast_drops_unaligned : Warning<
"implicit cast from type %0 to type %1 drops __unaligned qualifier">,
InGroup>;
+def warn_func_effect_allocates : Warning<
+ "'%0' function '%1' must not allocate or deallocate memory">,
+ InGroup;
+
@@ -10778,6 +10778,101 @@ def warn_imp_cast_drops_unaligned : Warning<
"implicit cast from type %0 to type %1 drops __unaligned qualifier">,
InGroup>;
+def warn_func_effect_allocates : Warning<
+ "'%0' function '%1' must not allocate or deallocate memory">,
+ InGroup;
+
@@ -4912,3 +4922,279 @@ void AutoType::Profile(llvm::FoldingSetNodeID &ID,
const ASTContext &Context) {
Profile(ID, Context, getDeducedType(), getKeyword(), isDependentType(),
getTypeConstraintConcept(), getTypeConstraintArguments());
}
+
+FunctionEffect::~Function
@@ -4181,6 +4185,127 @@ class FunctionNoProtoType : public FunctionType, public
llvm::FoldingSetNode {
}
};
+class FunctionEffect;
+class FunctionEffectSet;
+
+// It is the user's responsibility to keep this in set form: elements are
+// ordered and unique.
+// We could hid
https://github.com/shafik commented:
Quick drive by comment
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -4181,6 +4185,127 @@ class FunctionNoProtoType : public FunctionType, public
llvm::FoldingSetNode {
}
};
+class FunctionEffect;
+class FunctionEffectSet;
+
+// It is the user's responsibility to keep this in set form: elements are
+// ordered and unique.
+// We could hid
https://github.com/shafik edited https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
dougsonos wrote:
> > * the bug where `AttributedType` sugar gets lost on lambdas (when the
> > "inferred" return type gets converted to a concrete one) happens here and
> > the `nolock(false)` attribute is lost from `h`.
>
> Opened an issue for this (#85120) because it really does seem like a
@@ -11100,6 +11136,48 @@ Attr
*Sema::getImplicitCodeSegOrSectionAttrForFunction(const FunctionDecl *FD,
return nullptr;
}
+// Should only be called when getFunctionEffects() returns a non-empty set.
+// Decl should be a FunctionDecl or BlockDecl.
+void Sema::CheckAddCallabl
@@ -11100,6 +11136,48 @@ Attr
*Sema::getImplicitCodeSegOrSectionAttrForFunction(const FunctionDecl *FD,
return nullptr;
}
+// Should only be called when getFunctionEffects() returns a non-empty set.
+// Decl should be a FunctionDecl or BlockDecl.
+void Sema::CheckAddCallabl
Sirraide wrote:
> * the bug where `AttributedType` sugar gets lost on lambdas (when the
> "inferred" return type gets converted to a concrete one) happens here and the
> `nolock(false)` attribute is lost from `h`.
Opened an issue for this (#85120) because it really does seem like a bug to me.
dougsonos wrote:
> Since you mention it is attached to the type, is it mangled then differently.
> e.g.:
> Does the above f calls to 2 different functions?
This example is problematic because in this construct, `g` and `h` degenerate
into function pointers, and inference doesn't work across fu
pinskia wrote:
One question since the attribute is applied to types is there a way to get the
nolock/noalloc from type?.
e.g.
```
template [[nolock(T)]] void f(T a) { a(); }
```
Will the above work or is there no way to implement that currently?
Since you mention it is attached to the type, is
@@ -4912,3 +4922,279 @@ void AutoType::Profile(llvm::FoldingSetNodeID &ID,
const ASTContext &Context) {
Profile(ID, Context, getDeducedType(), getKeyword(), isDependentType(),
getTypeConstraintConcept(), getTypeConstraintArguments());
}
+
+FunctionEffect::~Function
@@ -0,0 +1,84 @@
+// RUN: %clang_cc1 -fsyntax-only -fblocks -verify %s
dougsonos wrote:
OK, maybe "syntax" then...
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
@@ -11100,6 +11136,48 @@ Attr
*Sema::getImplicitCodeSegOrSectionAttrForFunction(const FunctionDecl *FD,
return nullptr;
}
+// Should only be called when getFunctionEffects() returns a non-empty set.
+// Decl should be a FunctionDecl or BlockDecl.
+void Sema::CheckAddCallabl
https://github.com/dougsonos edited
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/dougsonos edited
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/dougsonos edited
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/dougsonos edited
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -4912,3 +4922,279 @@ void AutoType::Profile(llvm::FoldingSetNodeID &ID,
const ASTContext &Context) {
Profile(ID, Context, getDeducedType(), getKeyword(), isDependentType(),
getTypeConstraintConcept(), getTypeConstraintArguments());
}
+
+FunctionEffect::~Function
@@ -4912,3 +4922,279 @@ void AutoType::Profile(llvm::FoldingSetNodeID &ID,
const ASTContext &Context) {
Profile(ID, Context, getDeducedType(), getKeyword(), isDependentType(),
getTypeConstraintConcept(), getTypeConstraintArguments());
}
+
+FunctionEffect::~Function
@@ -0,0 +1,84 @@
+// RUN: %clang_cc1 -fsyntax-only -fblocks -verify %s
Sirraide wrote:
It’s usually not required to separate code that errors and code that is
supposed to compile without errors, but if it is required for something like
this, then sure, go ahead
@@ -4912,3 +4922,279 @@ void AutoType::Profile(llvm::FoldingSetNodeID &ID,
const ASTContext &Context) {
Profile(ID, Context, getDeducedType(), getKeyword(), isDependentType(),
getTypeConstraintConcept(), getTypeConstraintArguments());
}
+
+FunctionEffect::~Function
@@ -4912,3 +4922,279 @@ void AutoType::Profile(llvm::FoldingSetNodeID &ID,
const ASTContext &Context) {
Profile(ID, Context, getDeducedType(), getKeyword(), isDependentType(),
getTypeConstraintConcept(), getTypeConstraintArguments());
}
+
+FunctionEffect::~Function
@@ -0,0 +1,84 @@
+// RUN: %clang_cc1 -fsyntax-only -fblocks -verify %s
dougsonos wrote:
Hm, the standard diagnostic for diagnosing incompatible attributes produces an
error, and the last-pass constraint analysis is skipped when there are errors.
I'm going to ke
@@ -4912,3 +4922,279 @@ void AutoType::Profile(llvm::FoldingSetNodeID &ID,
const ASTContext &Context) {
Profile(ID, Context, getDeducedType(), getKeyword(), isDependentType(),
getTypeConstraintConcept(), getTypeConstraintArguments());
}
+
+FunctionEffect::~Function
@@ -10778,6 +10778,101 @@ def warn_imp_cast_drops_unaligned : Warning<
"implicit cast from type %0 to type %1 drops __unaligned qualifier">,
InGroup>;
+def warn_func_effect_allocates : Warning<
+ "'%0' function '%1' must not allocate or deallocate memory">,
+ InGroup;
+
@@ -4912,3 +4922,279 @@ void AutoType::Profile(llvm::FoldingSetNodeID &ID,
const ASTContext &Context) {
Profile(ID, Context, getDeducedType(), getKeyword(), isDependentType(),
getTypeConstraintConcept(), getTypeConstraintArguments());
}
+
+FunctionEffect::~Function
https://github.com/dougsonos deleted
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -395,6 +395,33 @@ bool Sema::checkStringLiteralArgumentAttr(const ParsedAttr
&AL, unsigned ArgNum,
return checkStringLiteralArgumentAttr(AL, ArgExpr, Str, ArgLocation);
}
+/// Check if the argument \p ArgNum of \p Attr is a compile-time constant
+/// integer (boolean) ex
https://github.com/dougsonos edited
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -10778,6 +10778,101 @@ def warn_imp_cast_drops_unaligned : Warning<
"implicit cast from type %0 to type %1 drops __unaligned qualifier">,
InGroup>;
+def warn_func_effect_allocates : Warning<
+ "'%0' function '%1' must not allocate or deallocate memory">,
+ InGroup;
+
@@ -10778,6 +10778,101 @@ def warn_imp_cast_drops_unaligned : Warning<
"implicit cast from type %0 to type %1 drops __unaligned qualifier">,
InGroup>;
+def warn_func_effect_allocates : Warning<
+ "'%0' function '%1' must not allocate or deallocate memory">,
+ InGroup;
+
@@ -4912,3 +4922,279 @@ void AutoType::Profile(llvm::FoldingSetNodeID &ID,
const ASTContext &Context) {
Profile(ID, Context, getDeducedType(), getKeyword(), isDependentType(),
getTypeConstraintConcept(), getTypeConstraintArguments());
}
+
+FunctionEffect::~Function
@@ -0,0 +1,26 @@
+// RUN: %clang_cc1 -fsyntax-only -fblocks -verify %s
+// R UN: %clang_cc1 -fsyntax-only -fblocks -verify -x c -std=c23 %s
+
+#if !__has_attribute(clang_nolock)
+#error "the 'nolock' attribute is not available"
+#endif
+
+void unannotated(void);
+void nolock(void)
@@ -0,0 +1,84 @@
+// RUN: %clang_cc1 -fsyntax-only -fblocks -verify %s
dougsonos wrote:
Thanks, yes, I will merge them (other than the one that is checking the AST
dump).
https://github.com/llvm/llvm-project/pull/84983
__
@@ -0,0 +1,26 @@
+// RUN: %clang_cc1 -fsyntax-only -fblocks -verify %s
+// R UN: %clang_cc1 -fsyntax-only -fblocks -verify -x c -std=c23 %s
+
+#if !__has_attribute(clang_nolock)
+#error "the 'nolock' attribute is not available"
+#endif
+
+void unannotated(void);
+void nolock(void)
https://github.com/Sirraide edited
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -395,6 +395,33 @@ bool Sema::checkStringLiteralArgumentAttr(const ParsedAttr
&AL, unsigned ArgNum,
return checkStringLiteralArgumentAttr(AL, ArgExpr, Str, ArgLocation);
}
+/// Check if the argument \p ArgNum of \p Attr is a compile-time constant
+/// integer (boolean) ex
@@ -339,6 +349,11 @@ namespace {
bool didParseNoDeref() const { return parsedNoDeref; }
+void setParsedNolock(unsigned char v) { parsedNolock = v; }
+unsigned char getParsedNolock() const { return parsedNolock; }
+void setParsedNoalloc(unsigned char v) { parse
@@ -395,6 +395,33 @@ bool Sema::checkStringLiteralArgumentAttr(const ParsedAttr
&AL, unsigned ArgNum,
return checkStringLiteralArgumentAttr(AL, ArgExpr, Str, ArgLocation);
}
+/// Check if the argument \p ArgNum of \p Attr is a compile-time constant
+/// integer (boolean) ex
@@ -4912,3 +4922,279 @@ void AutoType::Profile(llvm::FoldingSetNodeID &ID,
const ASTContext &Context) {
Profile(ID, Context, getDeducedType(), getKeyword(), isDependentType(),
getTypeConstraintConcept(), getTypeConstraintArguments());
}
+
+FunctionEffect::~Function
@@ -2380,6 +2382,1239 @@ class UnsafeBufferUsageReporter : public
UnsafeBufferUsageHandler {
};
} // namespace
+//
=
+
+// Temporary debugging option
+#define FX_ANALYZER_VERIFY_DECL_LIST 1
+
+namesp
@@ -4912,3 +4922,279 @@ void AutoType::Profile(llvm::FoldingSetNodeID &ID,
const ASTContext &Context) {
Profile(ID, Context, getDeducedType(), getKeyword(), isDependentType(),
getTypeConstraintConcept(), getTypeConstraintArguments());
}
+
+FunctionEffect::~Function
@@ -4912,3 +4922,279 @@ void AutoType::Profile(llvm::FoldingSetNodeID &ID,
const ASTContext &Context) {
Profile(ID, Context, getDeducedType(), getKeyword(), isDependentType(),
getTypeConstraintConcept(), getTypeConstraintArguments());
}
+
+FunctionEffect::~Function
@@ -4912,3 +4922,279 @@ void AutoType::Profile(llvm::FoldingSetNodeID &ID,
const ASTContext &Context) {
Profile(ID, Context, getDeducedType(), getKeyword(), isDependentType(),
getTypeConstraintConcept(), getTypeConstraintArguments());
}
+
+FunctionEffect::~Function
@@ -4912,3 +4922,279 @@ void AutoType::Profile(llvm::FoldingSetNodeID &ID,
const ASTContext &Context) {
Profile(ID, Context, getDeducedType(), getKeyword(), isDependentType(),
getTypeConstraintConcept(), getTypeConstraintArguments());
}
+
+FunctionEffect::~Function
@@ -4912,3 +4922,279 @@ void AutoType::Profile(llvm::FoldingSetNodeID &ID,
const ASTContext &Context) {
Profile(ID, Context, getDeducedType(), getKeyword(), isDependentType(),
getTypeConstraintConcept(), getTypeConstraintArguments());
}
+
+FunctionEffect::~Function
@@ -3666,11 +3673,14 @@ void FunctionProtoType::Profile(llvm::FoldingSetNodeID
&ID, QualType Result,
// Finally we have a trailing return type flag (bool)
// combined with AArch64 SME Attributes, to save space:
// int
+ // Then add the FunctionEffects
//
// T
@@ -10778,6 +10778,101 @@ def warn_imp_cast_drops_unaligned : Warning<
"implicit cast from type %0 to type %1 drops __unaligned qualifier">,
InGroup>;
+def warn_func_effect_allocates : Warning<
+ "'%0' function '%1' must not allocate or deallocate memory">,
+ InGroup;
+
https://github.com/dougsonos edited
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/dougsonos edited
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -10778,6 +10778,101 @@ def warn_imp_cast_drops_unaligned : Warning<
"implicit cast from type %0 to type %1 drops __unaligned qualifier">,
InGroup>;
+def warn_func_effect_allocates : Warning<
+ "'%0' function '%1' must not allocate or deallocate memory">,
+ InGroup;
+
@@ -3666,11 +3673,14 @@ void FunctionProtoType::Profile(llvm::FoldingSetNodeID
&ID, QualType Result,
// Finally we have a trailing return type flag (bool)
// combined with AArch64 SME Attributes, to save space:
// int
+ // Then add the FunctionEffects
//
// T
@@ -10778,6 +10778,101 @@ def warn_imp_cast_drops_unaligned : Warning<
"implicit cast from type %0 to type %1 drops __unaligned qualifier">,
InGroup>;
+def warn_func_effect_allocates : Warning<
+ "'%0' function '%1' must not allocate or deallocate memory">,
+ InGroup;
+
@@ -1402,6 +1402,29 @@ def CXX11NoReturn : InheritableAttr {
let Documentation = [CXX11NoReturnDocs];
}
+def NoLock : DeclOrTypeAttr {
+ let Spellings = [CXX11<"clang", "nolock">,
+ C23<"clang", "nolock">,
+ GNU<"clang_nolock">];
+
+ //
@@ -10778,6 +10778,101 @@ def warn_imp_cast_drops_unaligned : Warning<
"implicit cast from type %0 to type %1 drops __unaligned qualifier">,
InGroup>;
+def warn_func_effect_allocates : Warning<
+ "'%0' function '%1' must not allocate or deallocate memory">,
+ InGroup;
+
https://github.com/dougsonos edited
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -1402,6 +1402,29 @@ def CXX11NoReturn : InheritableAttr {
let Documentation = [CXX11NoReturnDocs];
}
+def NoLock : DeclOrTypeAttr {
+ let Spellings = [CXX11<"clang", "nolock">,
+ C23<"clang", "nolock">,
+ GNU<"clang_nolock">];
+
+ //
https://github.com/dougsonos reopened
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
dougsonos wrote:
> > * Does the FunctionEffect / FunctionEffectSet abstraction make sense
> > (Type.h)? The idea is that an abstract effect system for functions could
> > easily support things like "TCB with types" or adding "nowait" to the
> > "nolock/noalloc" group of effects.
>
> Hmm, this
https://github.com/dougsonos closed
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Sirraide wrote:
> * Conversions between functions / function pointers with and without the
> attribute work in C++ but not in C and I'm a bit lost (there's a bunch of
> debug logging around this e.g. Sema::IsFunctionConversion and
> Sema::ImpCastExprToType)
Er, I’ll see if I can find some tim
Sirraide wrote:
> - Does the FunctionEffect / FunctionEffectSet abstraction make sense
> (Type.h)? The idea is that an abstract effect system for functions could
> easily support things like "TCB with types" or adding "nowait" to the
> "nolock/noalloc" group of effects.
Hmm, this is a difficu
Sirraide wrote:
> * FunctionEffect/FunctionEffectSet need to be serialized as part of
> FunctionProtoType
That’s in `TypeProperties.td` from what I recall. You might be able to pack
this into an integer or something similar perhaps.
https://github.com/llvm/llvm-project/pull/84983
pinskia wrote:
Does it make sense to have some C testcases too? Likewise some testcases
testing the __attribute__ style attribute?
I would say more testcases the better really.
https://github.com/llvm/llvm-project/pull/84983
___
cfe-commits mailing li
@@ -1402,6 +1402,29 @@ def CXX11NoReturn : InheritableAttr {
let Documentation = [CXX11NoReturnDocs];
}
+def NoLock : DeclOrTypeAttr {
+ let Spellings = [CXX11<"clang", "nolock">,
+ C23<"clang", "nolock">,
+ GNU<"clang_nolock">];
+
+ //
@@ -18321,6 +18321,28 @@ bool Sema::CheckOverridingFunctionAttributes(const
CXXMethodDecl *New,
return true;
}
+ // Virtual overrides: check for matching effects.
Sirraide wrote:
Just a remark: Despite what I said wrt function declaration merging, I a
1 - 100 of 122 matches
Mail list logo
