Displaying 13 results from an estimated 13 matches for "linkglobalvalueproto".
2016 Jul 29
2
[ThinLTO] assert(GS != DefinedGlobals.end()) failed in FunctionImport.cpp
...lizable
; Function Attrs: nounwind uwtable
define weak_odr void @foo(%1*) unnamed_addr #7 comdat($comdat1) align 2 personality i32 (...)* @__gxx_personality_v0 {}
is materialized to
; Function Attrs: nounwind uwtable
declare void @foo(%"type1"*) unnamed_addr #2 align 2
Inside IRLinker::linkGlobalValueProto, the materialized value is returned from getLinkedToGlobal(SGV) and assigned to DGV. However, as ForAlias is true and ShouldLink is false, DGV becomes nullptr later, and NewGV is created from copyGlobalValueProto(SGV, ShouldLink) call. Therefore, returned value from linkGlobalValueProto is differen...
2016 Jul 29
0
[ThinLTO] assert(GS != DefinedGlobals.end()) failed in FunctionImport.cpp
...) unnamed_addr #7 comdat($comdat1) align 2
> personality i32 (...)* @__gxx_personality_v0 {}
>
>
>
> is materialized to
>
>
>
> ; Function Attrs: nounwind uwtable
>
> declare void @foo(%"type1"*) unnamed_addr #2 align 2
>
>
>
> Inside IRLinker::linkGlobalValueProto, the materialized value is returned
> from getLinkedToGlobal(SGV) and assigned to DGV. However, as ForAlias is
> true and ShouldLink is false, DGV becomes nullptr later, and NewGV is
> created from copyGlobalValueProto(SGV, ShouldLink) call. Therefore,
> returned value from linkGlobalVa...
2016 Jul 29
2
[ThinLTO] assert(GS != DefinedGlobals.end()) failed in FunctionImport.cpp
...; Function Attrs: nounwind uwtable
define weak_odr void @foo(%1*) unnamed_addr #7 comdat($comdat1) align 2 personality i32 (...)* @__gxx_personality_v0 {}
is materialized to
; Function Attrs: nounwind uwtable
declare void @foo(%"type1"*) unnamed_addr #2 align 2
Inside IRLinker::linkGlobalValueProto, the materialized value is returned from getLinkedToGlobal(SGV) and assigned to DGV. However, as ForAlias is true and ShouldLink is false, DGV becomes nullptr later, and NewGV is created from copyGlobalValueProto(SGV, ShouldLink) call. Therefore, returned value from linkGlobalValueProto is differen...
2016 Jul 29
0
[ThinLTO] assert(GS != DefinedGlobals.end()) failed in FunctionImport.cpp
...stripping .llvm.{HASH}, but what I observe is that ".1" is
> still appended to the expected original name.
>
>
>
> Then where this extra ".1" comes from? It is appended when the global
> value is materialized. IRLinker::materialize function calls
> IRLinker::linkGlobalValueProto function, and inside that function if DGV is
> nullptr or ShouldLink is true then IRLinker::copyGlobalValueProto function
> is called to create a global variable in the destination module that
> corresponds to SGV. I found that newly created global variable has the
> extra ".1"...
2016 Jul 29
0
[ThinLTO] assert(GS != DefinedGlobals.end()) failed in FunctionImport.cpp
...) unnamed_addr #7 comdat($comdat1) align 2
> personality i32 (...)* @__gxx_personality_v0 {}
>
>
>
> is materialized to
>
>
>
> ; Function Attrs: nounwind uwtable
>
> declare void @foo(%"type1"*) unnamed_addr #2 align 2
>
>
>
> Inside IRLinker::linkGlobalValueProto, the materialized value is returned
> from getLinkedToGlobal(SGV) and assigned to DGV. However, as ForAlias is
> true and ShouldLink is false, DGV becomes nullptr later, and NewGV is
> created from copyGlobalValueProto(SGV, ShouldLink) call. Therefore,
> returned value from linkGlobalVa...
2016 Jul 30
2
[ThinLTO] assert(GS != DefinedGlobals.end()) failed in FunctionImport.cpp
...; Function Attrs: nounwind uwtable
define weak_odr void @foo(%1*) unnamed_addr #7 comdat($comdat1) align 2 personality i32 (...)* @__gxx_personality_v0 {}
is materialized to
; Function Attrs: nounwind uwtable
declare void @foo(%"type1"*) unnamed_addr #2 align 2
Inside IRLinker::linkGlobalValueProto, the materialized value is returned from getLinkedToGlobal(SGV) and assigned to DGV. However, as ForAlias is true and ShouldLink is false, DGV becomes nullptr later, and NewGV is created from copyGlobalValueProto(SGV, ShouldLink) call. Therefore, returned value from linkGlobalValueProto is differen...
2016 Jul 29
3
[ThinLTO] assert(GS != DefinedGlobals.end()) failed in FunctionImport.cpp
...pts to get the original name by stripping .llvm.{HASH}, but what I observe is that ".1" is still appended to the expected original name.
Then where this extra ".1" comes from? It is appended when the global value is materialized. IRLinker::materialize function calls IRLinker::linkGlobalValueProto function, and inside that function if DGV is nullptr or ShouldLink is true then IRLinker::copyGlobalValueProto function is called to create a global variable in the destination module that corresponds to SGV. I found that newly created global variable has the extra ".1" added to the name...
2016 Jul 30
1
[ThinLTO] assert(GS != DefinedGlobals.end()) failed in FunctionImport.cpp
...) unnamed_addr #7 comdat($comdat1) align 2
> personality i32 (...)* @__gxx_personality_v0 {}
>
>
>
> is materialized to
>
>
>
> ; Function Attrs: nounwind uwtable
>
> declare void @foo(%"type1"*) unnamed_addr #2 align 2
>
>
>
> Inside IRLinker::linkGlobalValueProto, the materialized value is returned
> from getLinkedToGlobal(SGV) and assigned to DGV. However, as ForAlias is
> true and ShouldLink is false, DGV becomes nullptr later, and NewGV is
> created from copyGlobalValueProto(SGV, ShouldLink) call. Therefore,
> returned value from linkGlobalVa...
2016 Aug 15
2
[ThinLTO] assert(GS != DefinedGlobals.end()) failed in FunctionImport.cpp
...;>>
>>>
>>> is materialized to
>>>
>>>
>>>
>>> ; Function Attrs: nounwind uwtable
>>>
>>> declare void @foo(%"type1"*) unnamed_addr #2 align 2
>>>
>>>
>>>
>>> Inside IRLinker::linkGlobalValueProto, the materialized value is
>>> returned from getLinkedToGlobal(SGV) and assigned to DGV. However, as
>>> ForAlias is true and ShouldLink is false, DGV becomes nullptr later, and
>>> NewGV is created from copyGlobalValueProto(SGV, ShouldLink) call.
>>> Therefore,...
2016 Aug 15
2
[ThinLTO] assert(GS != DefinedGlobals.end()) failed in FunctionImport.cpp
...) unnamed_addr #7 comdat($comdat1) align 2
> personality i32 (...)* @__gxx_personality_v0 {}
>
>
>
> is materialized to
>
>
>
> ; Function Attrs: nounwind uwtable
>
> declare void @foo(%"type1"*) unnamed_addr #2 align 2
>
>
>
> Inside IRLinker::linkGlobalValueProto, the materialized value is returned
> from getLinkedToGlobal(SGV) and assigned to DGV. However, as ForAlias is
> true and ShouldLink is false, DGV becomes nullptr later, and NewGV is
> created from copyGlobalValueProto(SGV, ShouldLink) call. Therefore,
> returned value from linkGlobalVa...
2016 Jul 28
2
[ThinLTO] assert(GS != DefinedGlobals.end()) failed in FunctionImport.cpp
...empts to get the original name by stripping .llvm.{HASH}, but what I observe is that ".1" is still appended to the expected original name.
Then where this extra ".1" comes from? It is appended when the global value is materialized. IRLinker::materialize function calls IRLinker::linkGlobalValueProto function, and inside that function if DGV is nullptr or ShouldLink is true then IRLinker::copyGlobalValueProto function is called to create a global variable in the destination module that corresponds to SGV. I found that newly created global variable has the extra ".1" added to the name...
2016 Jul 29
0
[ThinLTO] assert(GS != DefinedGlobals.end()) failed in FunctionImport.cpp
...stripping .llvm.{HASH}, but what I observe is that ".1" is
> still appended to the expected original name.
>
>
>
> Then where this extra ".1" comes from? It is appended when the global
> value is materialized. IRLinker::materialize function calls
> IRLinker::linkGlobalValueProto function, and inside that function if DGV is
> nullptr or ShouldLink is true then IRLinker::copyGlobalValueProto function
> is called to create a global variable in the destination module that
> corresponds to SGV. I found that newly created global variable has the
> extra ".1"...
2016 Jul 22
3
[ThinLTO] Using two different IRMovers for the same composite module? (related to PR28180)
It seems that the patch works for me as well, though the linker crashes with another error after that. Thanks!
Mehdi, I couldn’t quite understand what do you mean by you don’t have a repro so you couldn’t upstream the patch. Aren’t .ll files you attached sufficient to submit along with the patch? If there is anything I can help you to upstream it, please let me know.
-- Taewook
On 7/22/16,