[Bug debug/65822] New: [4.8.2 regression] Used variant fun names in dwarf info for CTORs

[Bug debug/65822] New: [4.8.2 regression] Used variant fun names in dwarf info for CTORs

[chihin.ko at oracle dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65822

            Bug ID: 65822
           Summary: [4.8.2 regression] Used variant fun names in dwarf
                    info for CTORs
           Product: gcc
           Version: 4.8.2
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: debug
          Assignee: unassigned at gcc dot gnu.org
          Reporter: chihin.ko at oracle dot com

Created attachment 35372
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=35372&action=edit
test case 1/1

for attached *.cc, there are 3 A CTORS and 3 variant A CTORS:

 nm a.out | grep _ZN1AC
0000000000400a8c W _ZN1AC1Ei
0000000000400ab2 W _ZN1AC1Eifci
0000000000400a5a W _ZN1AC1Ev
0000000000400a8c W _ZN1AC2Ei
0000000000400ab2 W _ZN1AC2Eifci
0000000000400a5a W _ZN1AC2Ev

In dwarf dump, _ZN1AC2... was used :

 grep _ZN1AC a.out.dw
                      DW_AT_MIPS_linkage_name     "_ZN1AC2Ev"
                      DW_AT_MIPS_linkage_name     "_ZN1AC2Ei"
                      DW_AT_MIPS_linkage_name     "_ZN1AC2Eifci"
name at offset 0x0000011a, length    9 is '_ZN1AC2Ei'
name at offset 0x00000143, length   12 is '_ZN1AC2Eifci'
name at offset 0x0000015a, length    9 is '_ZN1AC2Ev'

I don't see _ZN1AC1... were used in a.out.dw

[Bug debug/65822] [4.8/4.9/5/6 Regression] Used variant fun names in dwarf info for CTORs

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65822

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jakub at gcc dot gnu.org,
                   |                            |jason at gcc dot gnu.org
            Summary|[4.8.2 regression] Used     |[4.8/4.9/5/6 Regression]
                   |variant fun names in dwarf  |Used variant fun names in
                   |info for CTORs              |dwarf info for CTORs

--- Comment #1 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
The C1 and C2 don't stand for CTORs and variant CTORs, but for complete object
constructor and base object constructor.  If the compiler chooses them to emit
them as aliases (because they are the same), you only get one of them in the
debug info, emitting both the same way would just severely bloat the debug
info, unless there is a way to represent the aliases without duplication. 
Dunno if e.g. a DW_TAG_subprogram
with DW_AT_name/DW_AT_linkage_name and just DW_AT_abstract_origin on the alias
and nothing else would do the job.

[Bug debug/65822] [4.8/4.9/5/6 Regression] Used variant fun names in dwarf info for CTORs

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65822

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.8.5

--- Comment #2 from Richard Biener <rguenth at gcc dot gnu.org> ---
I suppose it does do the job.

[Bug debug/65822] [4.8/4.9/5/6 Regression] Used variant fun names in dwarf info for CTORs

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65822

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|4.8.5                       |4.9.3

--- Comment #3 from Richard Biener <rguenth at gcc dot gnu.org> ---
The gcc-4_8-branch is being closed, re-targeting regressions to 4.9.3.

[Bug debug/65822] [4.9/5/6 Regression] Used variant fun names in dwarf info for CTORs

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65822

--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
GCC 4.9.3 has been released.

[Bug debug/65822] [4.9/5/6 Regression] Used variant fun names in dwarf info for CTORs

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65822

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|4.9.3                       |4.9.4