eugenis added a comment.
second attempt in r247494
Repository:
rL LLVM
http://reviews.llvm.org/D12087
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Thanks. I just reproduced it on release w/o asserts build and fixed locally.
The other problem is fixed as well, I'll try re-landing now. I'll keep
an eye on the bot later today and will revert again if necessary.
On Fri, Sep 11, 2015 at 6:02 PM, H.J. Lu wrote:
> On Fri, Sep 11, 2015 at 4:45 PM,
On Fri, Sep 11, 2015 at 4:45 PM, Evgenii Stepanov wrote:
> Does it say that there is no entry basic block? I.e. the output
> apparently looks like
>
> define void @h() #1 {
> store void ()* @f1, void ()** @p, align 8
>
> Could you confirm it? Never seen this behavior.
>
> I'm going to revert the
eugenis added inline comments.
Comment at: lib/CodeGen/CodeGenModule.cpp:543
@@ +542,3 @@
+ if (Fn->isUsedByMetadata())
+llvm::ValueAsMetadata::handleRAUW(Fn, StubFn);
+}
As the comment says.
W/o this, the debug info for .alwaysinline instructions is attached
eugenis updated this revision to Diff 34610.
eugenis added a comment.
Fixed the debug info problem.
Repository:
rL LLVM
http://reviews.llvm.org/D12087
Files:
lib/CodeGen/CGCXX.cpp
lib/CodeGen/CGClass.cpp
lib/CodeGen/CGOpenMPRuntime.cpp
lib/CodeGen/CodeGenModule.cpp
lib/CodeGen/Code
Does it say that there is no entry basic block? I.e. the output
apparently looks like
define void @h() #1 {
store void ()* @f1, void ()** @p, align 8
Could you confirm it? Never seen this behavior.
I'm going to revert the change due to this and also one broken gdb
test (something wrong with de
On Fri, Sep 11, 2015 at 1:31 PM, Evgeniy Stepanov via cfe-commits
wrote:
> eugenis closed this revision.
> eugenis added a comment.
>
> r247465, thanks for the review!
>
On Fedora 22/x86-64, I got
FAIL: Clang :: CodeGen/always_inline.c (3049 of 26132)
TEST 'Clang :: CodeGen/
eugenis closed this revision.
eugenis added a comment.
r247465, thanks for the review!
Repository:
rL LLVM
http://reviews.llvm.org/D12087
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-c
eugenis updated this revision to Diff 34578.
eugenis added a comment.
rebase, fix a merge conflict
Repository:
rL LLVM
http://reviews.llvm.org/D12087
Files:
lib/CodeGen/CGCXX.cpp
lib/CodeGen/CGClass.cpp
lib/CodeGen/CGOpenMPRuntime.cpp
lib/CodeGen/CodeGenModule.cpp
lib/CodeGen/CodeG
eugenis added a comment.
I'm going to commit this tomorrow unless someone speaks up.
Repository:
rL LLVM
http://reviews.llvm.org/D12087
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-com
eugenis added inline comments.
Comment at: lib/CodeGen/CodeGenModule.h:505
@@ +504,3 @@
+
+ llvm::SmallVector AlwaysInlineFunctions;
+
Right. Sorry I missed the comment.
I've switched this to a vector.
Repository:
rL LLVM
http://reviews.llvm.org/D12087
__
eugenis updated this revision to Diff 34095.
Repository:
rL LLVM
http://reviews.llvm.org/D12087
Files:
lib/CodeGen/CGCXX.cpp
lib/CodeGen/CGClass.cpp
lib/CodeGen/CGOpenMPRuntime.cpp
lib/CodeGen/CodeGenModule.cpp
lib/CodeGen/CodeGenModule.h
lib/CodeGen/ItaniumCXXABI.cpp
test/CodeGe
rsmith added inline comments.
Comment at: lib/CodeGen/CodeGenModule.h:505
@@ +504,3 @@
+
+ llvm::SetVector AlwaysInlineFunctions;
+
Did you try making this a vector? It'd be nice to avoid the set overhead here
if we can.
Repository:
rL LLVM
http://reviews.l
eugenis added a comment.
In http://reviews.llvm.org/D12087#239243, @yaron.keren wrote:
> In CGCXX.cpp, may be fixable after this commit:
>
> // FIXME: An extern template instantiation will create functions with
> // linkage "AvailableExternally". In libc++, some classes also define
> // mem
eugenis updated this revision to Diff 33997.
Repository:
rL LLVM
http://reviews.llvm.org/D12087
Files:
lib/CodeGen/CGCXX.cpp
lib/CodeGen/CGClass.cpp
lib/CodeGen/CGOpenMPRuntime.cpp
lib/CodeGen/CodeGenModule.cpp
lib/CodeGen/CodeGenModule.h
lib/CodeGen/ItaniumCXXABI.cpp
test/CodeGe
yaron.keren added a comment.
In CGCXX.cpp, may be fixable after this commit:
// FIXME: An extern template instantiation will create functions with
// linkage "AvailableExternally". In libc++, some classes also define
// members with attribute "AlwaysInline" and expect no reference to
// b
eugenis updated this revision to Diff 33868.
eugenis marked 6 inline comments as done.
Repository:
rL LLVM
http://reviews.llvm.org/D12087
Files:
lib/CodeGen/CGCXX.cpp
lib/CodeGen/CGClass.cpp
lib/CodeGen/CGOpenMPRuntime.cpp
lib/CodeGen/CodeGenModule.cpp
lib/CodeGen/CodeGenModule.h
l
rsmith added inline comments.
Comment at: lib/CodeGen/CodeGenModule.cpp:485
@@ +484,3 @@
+ std::string Name = Fn->getName();
+ std::string InlineName = Name + ".inlinefunction";
+ Fn->setName(InlineName);
I have a slight preference for ".alwaysinline" over ".in
eugenis marked 2 inline comments as done.
eugenis added a comment.
Repository:
rL LLVM
http://reviews.llvm.org/D12087
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
eugenis updated this revision to Diff 33766.
Repository:
rL LLVM
http://reviews.llvm.org/D12087
Files:
lib/CodeGen/CGCXX.cpp
lib/CodeGen/CGClass.cpp
lib/CodeGen/CGOpenMPRuntime.cpp
lib/CodeGen/CodeGenModule.cpp
lib/CodeGen/CodeGenModule.h
lib/CodeGen/ItaniumCXXABI.cpp
test/CodeGe
rnk accepted this revision.
rnk added a reviewer: rnk.
rnk added a comment.
This revision is now accepted and ready to land.
lgtm
Comment at: lib/CodeGen/CodeGenModule.cpp:463
@@ +462,3 @@
+
+// if (U.getUser() == GV) continue;
+
Commented out code?
===
eugenis added inline comments.
Comment at: lib/CodeGen/CodeGenModule.cpp:469-470
@@ +468,4 @@
+ llvm::LLVMContext &Ctx = getModule().getContext();
+ llvm::Function *StubFn =
+ llvm::Function::Create(FT, Fn->getLinkage(), Name, &getModule());
+ assert(StubFn->getName() == N
eugenis set the repository for this revision to rL LLVM.
eugenis updated this revision to Diff 33746.
eugenis marked an inline comment as done.
Repository:
rL LLVM
http://reviews.llvm.org/D12087
Files:
lib/CodeGen/CGCXX.cpp
lib/CodeGen/CGClass.cpp
lib/CodeGen/CGOpenMPRuntime.cpp
lib/Co
rnk added a subscriber: rnk.
Comment at: lib/CodeGen/CodeGenModule.cpp:469-470
@@ +468,4 @@
+ llvm::LLVMContext &Ctx = getModule().getContext();
+ llvm::Function *StubFn =
+ llvm::Function::Create(FT, Fn->getLinkage(), Name, &getModule());
+ assert(StubFn->getName() == Nam
On Mon, Aug 17, 2015 at 1:59 PM, David Blaikie wrote:
>
>
> On Mon, Aug 17, 2015 at 11:07 AM, Evgeniy Stepanov via cfe-commits
> wrote:
>>
>> eugenis created this revision.
>> eugenis added reviewers: chandlerc, rsmith.
>> eugenis added a subscriber: cfe-commits.
>> eugenis set the repository for
On Mon, Aug 17, 2015 at 1:59 PM, David Blaikie wrote:
>
>
> On Mon, Aug 17, 2015 at 11:07 AM, Evgeniy Stepanov via cfe-commits <
> cfe-commits@lists.llvm.org> wrote:
>
>> eugenis created this revision.
>> eugenis added reviewers: chandlerc, rsmith.
>> eugenis added a subscriber: cfe-commits.
>> e
eugenis added inline comments.
Comment at: lib/CodeGen/CodeGenModule.cpp:513
@@ +512,3 @@
+void CodeGenModule::RewriteAlwaysInlineFunctions() {
+ for (llvm::Function *Fn : AlwaysInlineFunctions)
+RewriteAlwaysInlineFunction(Fn);
Done. Should I make it a "Smal
eugenis updated this revision to Diff 32341.
http://reviews.llvm.org/D12087
Files:
lib/CodeGen/CGCXX.cpp
lib/CodeGen/CGClass.cpp
lib/CodeGen/CGOpenMPRuntime.cpp
lib/CodeGen/CodeGenModule.cpp
lib/CodeGen/CodeGenModule.h
lib/CodeGen/ItaniumCXXABI.cpp
test/CodeGen/2008-05-19-AlwaysInli
On Mon, Aug 17, 2015 at 11:07 AM, Evgeniy Stepanov via cfe-commits <
cfe-commits@lists.llvm.org> wrote:
> eugenis created this revision.
> eugenis added reviewers: chandlerc, rsmith.
> eugenis added a subscriber: cfe-commits.
> eugenis set the repository for this revision to rL LLVM.
>
> Currently
rsmith added inline comments.
Comment at: lib/CodeGen/CodeGenModule.cpp:513
@@ +512,3 @@
+void CodeGenModule::RewriteAlwaysInlineFunctions() {
+ for (llvm::Function *Fn : AlwaysInlineFunctions)
+RewriteAlwaysInlineFunction(Fn);
Does it matter that this traver
eugenis marked an inline comment as done.
Comment at: lib/CodeGen/CodeGenModule.cpp:447-448
@@ +446,4 @@
+if (C && !isa(C)) {
+ C->handleOperandChange(GV, IndirectReplacement, &U);
+ continue;
+}
Good catch.
Comment at: lib/Code
eugenis updated this revision to Diff 32338.
http://reviews.llvm.org/D12087
Files:
lib/CodeGen/CGCXX.cpp
lib/CodeGen/CGClass.cpp
lib/CodeGen/CGOpenMPRuntime.cpp
lib/CodeGen/CodeGenModule.cpp
lib/CodeGen/CodeGenModule.h
lib/CodeGen/ItaniumCXXABI.cpp
test/CodeGen/2008-05-19-AlwaysInli
probinson added a subscriber: probinson.
Comment at: lib/CodeGen/CodeGenModule.cpp:496
@@ +495,3 @@
+ llvm::ReturnInst::Create(Ctx, CI, BB);
+ }
+
What does the debug info look like for the wrapper?
http://reviews.llvm.org/D12087
___
eugenis removed rL LLVM as the repository for this revision.
eugenis updated this revision to Diff 32330.
http://reviews.llvm.org/D12087
Files:
lib/CodeGen/CGCXX.cpp
lib/CodeGen/CodeGenModule.cpp
lib/CodeGen/CodeGenModule.h
lib/CodeGen/ItaniumCXXABI.cpp
test/CodeGen/2008-05-19-AlwaysInl
chandlerc added a comment.
FYI, we should send an RFC to llvm-dev about the design change of always_inline
and make sure folks generally like the IR-level direction as well. We can point
at this review as an example.
I'm happy to write that up and send it if that's useful?
Repository:
rL LL
rsmith added inline comments.
Comment at: lib/CodeGen/CodeGenModule.cpp:447-448
@@ +446,4 @@
+ C->handleOperandChange(GV, IndirectReplacement, &U);
+} else if (!isa(U.getUser()) &&
+ !isa(U.getUser())) {
+ U.set(IndirectReplacement);
Do
eugenis created this revision.
eugenis added reviewers: chandlerc, rsmith.
eugenis added a subscriber: cfe-commits.
eugenis set the repository for this revision to rL LLVM.
Currently always_inline definitions are emitted as (in most cases) an
available_externally llvm function with an alwaysinlin
37 matches
Mail list logo
