[Bug lto/81968] [8 regression] early lto debug objects make Solaris ld SEGV

["rguenther at suse dot de" <gcc-bugzilla at gcc dot gnu dot org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81968

--- Comment #33 from rguenther at suse dot de <rguenther at suse dot de> ---
On December 21, 2017 4:20:35 PM GMT+01:00, "ro at CeBiTec dot Uni-Bielefeld.DE"
<gcc-bugzilla@gcc.gnu.org> wrote:
>https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81968
>
>--- Comment #32 from ro at CeBiTec dot Uni-Bielefeld.DE <ro at CeBiTec
>dot Uni-Bielefeld.DE> ---
>Here are the additional types of failures found on Solaris/x86:
>
>* Invalid sh_info:
>
>FAIL: g++.dg/lto/pr42987 cp_lto_pr42987_0.o-cp_lto_pr42987_1.o link, 
>-flto
>-flto-partition=none -g 
>FAIL: g++.dg/lto/pr42987 cp_lto_pr42987_0.o-cp_lto_pr42987_1.o link, 
>-flto -g 
>
>ld: fatal: file /var/tmp//ccXwdzGddebugobjtem: section
>[6].rel.gnu.debuglto_.debug_info: has invalid sh_info: 9
>ld: fatal: file /var/tmp//ccuVJOVbdebugobjtem: section
>[6].rel.gnu.debuglto_.debug_info: has invalid sh_info: 8
>
>  seen during the ld -r -o ccrIZb_ddebugobj ccSerOuddebugobjtem
>  cc5b3Arcdebugobjtem step:
>
>ccSerOuddebugobjtem: .rel.gnu.debuglto_.debug_info: sh_info: 9: does
>not point
>to a valid section
>
>Section Header[6]:  sh_name: .rel.gnu.debuglto_.debug_info
>    sh_addr:      0               sh_flags:   0
>    sh_size:      0x8             sh_type:    [ SHT_REL ]
>    sh_offset:    0x11c8          sh_entsize: 0x8 (1 entry)
>    sh_link:      5               sh_info:    9
>    sh_addralign: 0x4       
>
>  But there's no section 9 any longer.

I fixed that in the posted patch for the hpux issue(s). 

>  In the input object (cp_lto_pr42987_0.o):
>
>Section Header[47]:  sh_name: .rel.gnu.debuglto_.debug_info
>    sh_addr:      0               sh_flags:   0
>    sh_size:      0x8             sh_type:    [ SHT_REL ]
>    sh_offset:    0x4434          sh_entsize: 0x8 (1 entry)
>    sh_link:      45              sh_info:    9
>    sh_addralign: 0x4       
>
>  with sh_info pointing at:
>
>Section Header[9]:  sh_name: .gnu.debuglto_.debug_info
>    sh_addr:      0               sh_flags:   0
>    sh_size:      0x3d4           sh_type:    [ SHT_PROGBITS ]
>    sh_offset:    0x7fb           sh_entsize: 0
>    sh_link:      0               sh_info:    0
>    sh_addralign: 0x1       
>
>*
>
>UNRESOLVED: g++.dg/lto/pr51650-1
>cp_lto_pr51650-1_0.o-cp_lto_pr51650-1_0.o
>execute  -flto -g 
>FAIL: g++.dg/lto/pr51650-1 cp_lto_pr51650-1_0.o-cp_lto_pr51650-1_0.o
>link, 
>-flto -g 
>
>ld: fatal: file /var/tmp//ccvDXoBddebugobjtem: section
>[6].rel.gnu.debuglto_.debug_info: relocations can not be applied
>against
>section [5].symtab
>
>  during ld -r:
>
>ld -r -o ccyWaCJadebugobj ccNQIWVcdebugobjtem
>ld: fatal: file ccNQIWVcdebugobjtem: section
>[6].rel.gnu.debuglto_.debug_info:
>relocations can not be applied against section [5].symtab
>
>Relocation Section:  .rel.gnu.debuglto_.debug_info
>  index  type      offset value  section symbol
>    [0]  R_386_32     0x6     0  .symtab .debug_abbrev (section)
>
>  input object (cp_lto_pr51650-1_0.o):
>
>Relocation Section:  .rel.gnu.debuglto_.debug_info
>  index  type      offset value  section / symbol
>    [0]  R_386_32     0x6     0  .gnu.debuglto_.debug_info
>.gnu.debuglto_.debug_abbrev (section)
>
>  I don't fully understand what the linker error is trying to say, but
>  obviously there's something amiss with reloc section remapping (or
>  lack thereof).
>
>        Rainer