This revision was automatically updated to reflect the committed changes.
Closed by commit rL257559: [Bugfix] Fix ICE on constexpr vector splat.
(authored by gbiv).
Changed prior to commit:
http://reviews.llvm.org/D14877?vs=44187&id=44701#toc
Repository:
rL LLVM
http://reviews.llvm.org/D148
rsmith accepted this revision.
This revision is now accepted and ready to land.
Comment at: lib/CodeGen/CGExprScalar.cpp:816-817
@@ +815,4 @@
+// the same as the vector's element type (sans qualifiers)
+assert(DstType->castAs()->getElementType().getTypePtr() ==
+
george.burgess.iv marked 2 inline comments as done.
george.burgess.iv added a comment.
Ping :)
http://reviews.llvm.org/D14877
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
george.burgess.iv added a comment.
> Maybe we could remove CK_BooleanToSignedFloating and model that conversion as
> a sequence of CK_BooleanToSignedIntegral followed by CK_IntegralToFloating? I
> don't imagine it's a common operation, so the simpler AST representation is
> probably worth more
george.burgess.iv updated this revision to Diff 44187.
george.burgess.iv marked an inline comment as done.
george.burgess.iv added a comment.
Addressed all feedback, and added tests for `bool` -> `[n x i128]` case.
http://reviews.llvm.org/D14877
Files:
include/clang/AST/ASTContext.h
include
rsmith added a comment.
Maybe we could remove `CK_BooleanToSignedFloating` and model that conversion as
a sequence of `CK_BooleanToSignedIntegral` followed by `CK_IntegralToFloating`?
I don't imagine it's a common operation, so the simpler AST representation is
probably worth more than the mino
george.burgess.iv added a comment.
Ping :)
http://reviews.llvm.org/D14877
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
george.burgess.iv added a comment.
Ping :)
http://reviews.llvm.org/D14877
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
george.burgess.iv updated this revision to Diff 42686.
george.burgess.iv added a comment.
TIL you can use OpenCL vectors in C++ files. Fixed a bug related to not knowing
this + added tests for the constexpr cases.
Also, is there a better way to do the `bool` -> `APFloat` conversion in
ExprConst
george.burgess.iv added inline comments.
Comment at: lib/Sema/SemaExpr.cpp:5576
@@ +5575,3 @@
+return ExprError();
+ return ImpCastExprToType(CastExprRes.get(), DestElemTy, CK);
+}
rsmith wrote:
> Looking at `ScalarExprEmitter::VisitCastExpr`, it seems like w
george.burgess.iv updated this revision to Diff 42683.
george.burgess.iv marked an inline comment as done.
george.burgess.iv added a comment.
> Changes to ExprConstant and CGExprConstant appear to be pure cleanups; please
> check those in as a separate change
r255314, thanks.
> When this is do
rsmith added a comment.
Changes to ExprConstant and CGExprConstant appear to be pure cleanups; please
check those in as a separate change.
When this is done, you should also be able to remove the `IgnoreImpCasts` and
`EmitScalarConversion` calls in the `CK_VectorSplat` handling in
`ScalarExprE
george.burgess.iv added a comment.
Ping :)
http://reviews.llvm.org/D14877
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
george.burgess.iv updated this revision to Diff 40895.
george.burgess.iv added a comment.
We now add implicit casts to splatted literals, as Richard suggested, instead
of trying to handle this as a special case in `ExprConstant`.
http://reviews.llvm.org/D14877
Files:
include/clang/Sema/Sema.
george.burgess.iv added inline comments.
Comment at: lib/CodeGen/CGExprConstant.cpp:1362-1363
@@ -1360,3 +1361,4 @@
+Inits[I] = llvm::ConstantFP::get(VMContext, Elt.getFloat());
else
-Inits.push_back(llvm::ConstantFP::get(VMContext, Elt.getFloat()));
+
majnemer added a subscriber: majnemer.
Comment at: lib/CodeGen/CGExprConstant.cpp:1362-1363
@@ -1360,3 +1361,4 @@
+Inits[I] = llvm::ConstantFP::get(VMContext, Elt.getFloat());
else
-Inits.push_back(llvm::ConstantFP::get(VMContext, Elt.getFloat()));
+
It would seem cleaner to build an ImplicitCastExpr node in Sema between the
operand and the splat node.
On Nov 20, 2015 11:04 AM, "George Burgess IV"
wrote:
> george.burgess.iv created this revision.
> george.burgess.iv added a reviewer: rsmith.
> george.burgess.iv added a subscriber: cfe-commits
george.burgess.iv created this revision.
george.burgess.iv added a reviewer: rsmith.
george.burgess.iv added a subscriber: cfe-commits.
When evaluating constexpr vector splats, we weren't doing appropriate type
conversions on the literal we were splatting, causing assertion failures in
cases lik
18 matches
Mail list logo
