Commit: Update libiberty sources

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Commit: Update libiberty sources

Nick Clifton
Hi Guys,

  I have checked in the attached patch to synchronize our copy of the
  libiberty sources with those from gcc.

Cheers
  Nick

libiberty/ChangeLog
2018-01-03  Jakub Jelinek  <[hidden email]>

        Update copyright years.

2017-01-02  Richard Biener  <[hidden email]>

        PR lto/83452
        * simple-object-elf.c (simple_object_elf_copy_lto_debug_section):
        Do not use UNDEF locals for removed symbols but instead just
        define them in the first prevailing section and with no name.
        Use the same gnu_lto_v1 name for all removed globals we promote to
        WEAK UNDEFs so hpux can use a stub to provide this symbol.  Clear
        sh_info and sh_link in removed sections.

2017-10-30  Richard Biener  <[hidden email]>

        PR lto/82757
        * simple-object-elf.c (simple_object_elf_copy_lto_debug_sections):
        Strip two leading _s from the __gnu_lto_* symbols.

2017-10-24  Alan Modra  <[hidden email]>

        PR lto/82687
        PR lto/82575
        * simple-object-elf.c (simple_object_elf_copy_lto_debug_sections):
        Only make __gnu_lto symbols hidden.

2017-10-20  Alan Modra  <[hidden email]>

        PR lto/82575
        * simple-object-elf.c (simple_object_elf_copy_lto_debug_sections):
        Make discarded non-local symbols weak and hidden.

2017-10-18  Jakub Jelinek  <[hidden email]>

        PR lto/82598
        * simple-object.c (handle_lto_debug_sections): Copy over also
        .note.GNU-stack section with unchanged name.
        * simple-object-elf.c (SHF_EXECINSTR): Define.
        (simple_object_elf_copy_lto_debug_section): Drop SHF_EXECINSTR bit
        on .note.GNU-stack section.

2017-09-25  Nathan Sidwell  <[hidden email]>

        PR demangler/82195
        * cp-demangle.c (d_encoding): Strip return type when name is a
        LOCAL_NAME.
        (d_local_name): Strip return type of enclosing TYPED_NAME.
        * testsuite/demangle-expected: Add and adjust tests.

2017-09-21  Nathan Sidwell  <[hidden email]>

        PR demangler/82195
        * cp-demangle.c (d_name): Revert addition of 'toplevel' parm.
        (has_return_type): Recurse for DEMANGLE_COMPONENT_LOCAL_NAME.
        (d_encoding): Revert d_name change.  Use is_fnqual_component_type
        to strip modifiers that do not belong.
        (d_special_name, d_class_enum_type): Revert d_name call change.
        (d_expresion_1): Commonize DEMANGLE_COMPONENT_UNARY building.
        (d_local_name): Revert parsing of a function type.
        (d_print_comp_inner): An inner LOCAL_NAME might contain a
        TEMPLATE.
        * testsuite/demangle-expected: Add & adjust tests


fred (129K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Commit: Update libiberty sources

Szabolcs Nagy-2
On 10/01/18 13:59, Nick Clifton wrote:
> Hi Guys,
>
>   I have checked in the attached patch to synchronize our copy of the
>   libiberty sources with those from gcc.
>

since this commit i see two new symbols in aarch64 libstdc++ that were not
there before:

$ nm -D libstdc++.so.6.0.25 | grep call_once
0000000000095e70 W _ZZSt9call_onceIMSt6threadFvvEJSt17reference_wrapperIS0_EEEvRSt9once_flagOT_DpOT0_ENKUlvE0_clEv
0000000000095d10 W
_ZZSt9call_onceIMSt6threadFvvEJSt17reference_wrapperIS0_EEEvRSt9once_flagOT_DpOT0_ENUlvE0_4_FUNEv

this makes the aarch64 libstdc++ abi tests fail.

curiously only happens when native ld is used, not when cross ld.
will debug this further, but in case anybody has ideas let me know.


> Cheers
>   Nick
>
> libiberty/ChangeLog
> 2018-01-03  Jakub Jelinek  <[hidden email]>
>
> Update copyright years.
>
> 2017-01-02  Richard Biener  <[hidden email]>
>
> PR lto/83452
> * simple-object-elf.c (simple_object_elf_copy_lto_debug_section):
> Do not use UNDEF locals for removed symbols but instead just
> define them in the first prevailing section and with no name.
> Use the same gnu_lto_v1 name for all removed globals we promote to
> WEAK UNDEFs so hpux can use a stub to provide this symbol.  Clear
> sh_info and sh_link in removed sections.
>
> 2017-10-30  Richard Biener  <[hidden email]>
>
> PR lto/82757
> * simple-object-elf.c (simple_object_elf_copy_lto_debug_sections):
> Strip two leading _s from the __gnu_lto_* symbols.
>
> 2017-10-24  Alan Modra  <[hidden email]>
>
> PR lto/82687
> PR lto/82575
> * simple-object-elf.c (simple_object_elf_copy_lto_debug_sections):
> Only make __gnu_lto symbols hidden.
>
> 2017-10-20  Alan Modra  <[hidden email]>
>
> PR lto/82575
> * simple-object-elf.c (simple_object_elf_copy_lto_debug_sections):
> Make discarded non-local symbols weak and hidden.
>
> 2017-10-18  Jakub Jelinek  <[hidden email]>
>
> PR lto/82598
> * simple-object.c (handle_lto_debug_sections): Copy over also
> .note.GNU-stack section with unchanged name.
> * simple-object-elf.c (SHF_EXECINSTR): Define.
> (simple_object_elf_copy_lto_debug_section): Drop SHF_EXECINSTR bit
> on .note.GNU-stack section.
>
> 2017-09-25  Nathan Sidwell  <[hidden email]>
>
> PR demangler/82195
> * cp-demangle.c (d_encoding): Strip return type when name is a
> LOCAL_NAME.
> (d_local_name): Strip return type of enclosing TYPED_NAME.
> * testsuite/demangle-expected: Add and adjust tests.
>
> 2017-09-21  Nathan Sidwell  <[hidden email]>
>
> PR demangler/82195
> * cp-demangle.c (d_name): Revert addition of 'toplevel' parm.
> (has_return_type): Recurse for DEMANGLE_COMPONENT_LOCAL_NAME.
> (d_encoding): Revert d_name change.  Use is_fnqual_component_type
> to strip modifiers that do not belong.
> (d_special_name, d_class_enum_type): Revert d_name call change.
> (d_expresion_1): Commonize DEMANGLE_COMPONENT_UNARY building.
> (d_local_name): Revert parsing of a function type.
> (d_print_comp_inner): An inner LOCAL_NAME might contain a
> TEMPLATE.
> * testsuite/demangle-expected: Add & adjust tests
>

Reply | Threaded
Open this post in threaded view
|

Re: Commit: Update libiberty sources

H.J. Lu-30
On Mon, Jan 15, 2018 at 8:03 AM, Szabolcs Nagy <[hidden email]> wrote:

> On 10/01/18 13:59, Nick Clifton wrote:
>> Hi Guys,
>>
>>   I have checked in the attached patch to synchronize our copy of the
>>   libiberty sources with those from gcc.
>>
>
> since this commit i see two new symbols in aarch64 libstdc++ that were not
> there before:
>
> $ nm -D libstdc++.so.6.0.25 | grep call_once
> 0000000000095e70 W _ZZSt9call_onceIMSt6threadFvvEJSt17reference_wrapperIS0_EEEvRSt9once_flagOT_DpOT0_ENKUlvE0_clEv
> 0000000000095d10 W
> _ZZSt9call_onceIMSt6threadFvvEJSt17reference_wrapperIS0_EEEvRSt9once_flagOT_DpOT0_ENUlvE0_4_FUNEv
>
> this makes the aarch64 libstdc++ abi tests fail.
>
> curiously only happens when native ld is used, not when cross ld.
> will debug this further, but in case anybody has ideas let me know.
>
>

This is

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83834

--
H.J.