https://gcc.gnu.org/bugzilla/show_bug.cgi?id=119966
Robins Tharakan <tharakan at gmail dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |tharakan at gmail dot com
--- Comment #16 from Robins Tharakan <tharakan at gmail dot com> ---
Issue: GCC master is unable to build Postgres HEAD and is constantly failing
since 19th June. The signature matches the one listed above, thus posting as a
comment.
If it helps in triaging:
- The last successful GCC commit (that was able to build postgres) was
a0ffc9b1bfd.
- The first postgres build failure was seen at this GCC commit 4eb11136389.
=========
Background: Some Postgres builfarm animals keep an ~hourly GCC build.
Leafhopper (an aarch64 machine) is failing since yesterday.
Notably only aarch64 is failing[1]. Here's another aarch64 machine - snakefly
[2] - is also failing since June 19th.
Some other animals (x86_64 machines - also on gcc hourly) are able to build
successfully [3].
=========
during RTL pass: reload
formatting.c: In function ‘int_to_roman’:
formatting.c:5116:1: internal compiler error: Segmentation fault
5116 | }
| ^
gcc -Wall -Wmissing-prototypes -Wpointer-arith -Wdeclaration-after-statement
-Werror=vla -Wendif-labels -Wmissing-format-attribute -Wimplicit-fallthrough=3
-Wcast-function-type -Wshadow=compatible-local -Wformat-security
-Wmissing-variable-declarations -fno-strict-aliasing -fwrapv
-fexcess-precision=standard -Wno-format-truncation -Wno-stringop-truncation -g
-O2 -I. -I. -I../../../../src/include -D_GNU_SOURCE -I/usr/include/libxml2
-c -o int8.o int8.c
gcc -Wall -Wmissing-prototypes -Wpointer-arith -Wdeclaration-after-statement
-Werror=vla -Wendif-labels -Wmissing-format-attribute -Wimplicit-fallthrough=3
-Wcast-function-type -Wshadow=compatible-local -Wformat-security
-Wmissing-variable-declarations -fno-strict-aliasing -fwrapv
-fexcess-precision=standard -Wno-format-truncation -Wno-stringop-truncation -g
-O2 -I. -I. -I../../../../src/include -D_GNU_SOURCE -I/usr/include/libxml2
-c -o pgstat_wal.o pgstat_wal.c
0x217455b internal_error(char const*, ...)
/opt/gcc/source/gcc/diagnostic-global-context.cc:517
0xfecc1b crash_signal
/opt/gcc/source/gcc/toplev.cc:321
0xdcd718 add_regs_to_insn_regno_info
/opt/gcc/source/gcc/lra.cc:1484
0xdcd94f add_regs_to_insn_regno_info
/opt/gcc/source/gcc/lra.cc:1568
0xdcdb73 lra_update_insn_regno_info(rtx_insn*)
/opt/gcc/source/gcc/lra.cc:1661
0xdebcef process_insn_for_elimination
/opt/gcc/source/gcc/lra-eliminations.cc:1398
0xdebcef lra_eliminate(bool, bool)
/opt/gcc/source/gcc/lra-eliminations.cc:1497
0xde2e8b lra_constraints(bool)
/opt/gcc/source/gcc/lra-constraints.cc:5368
0xdce38f lra(_IO_FILE*, int)
/opt/gcc/source/gcc/lra.cc:2455
0xd813cf do_reload
/opt/gcc/source/gcc/ira.cc:5986
0xd813cf execute
/opt/gcc/source/gcc/ira.cc:6174
=========
Reference:
1. aarch64 machine builds failing since 2025/06/19 -
https://buildfarm.postgresql.org/cgi-bin/show_history.pl?nm=leafhopper&br=master
2. Another aarch64 build failing since 2025/6/19 -
https://buildfarm.postgresql.org/cgi-bin/show_history.pl?nm=snakefly&br=master
3. x86_64 machine builds are working okay -
https://buildfarm.postgresql.org/cgi-bin/show_history.pl?nm=alligator&br=master
