https://gcc.gnu.org/bugzilla/show_bug.cgi?id=125129
--- Comment #6 from ro at CeBiTec dot Uni-Bielefeld.DE <ro at CeBiTec dot
Uni-Bielefeld.DE> ---
> --- Comment #5 from Rainer Orth <ro at gcc dot gnu.org> ---
> I've found two other issues with the patch:
[...]
> * Another issue occured in a Solaris/SPARC bootstrap: at some point the build
> hang with two python processes:
>
> 18687: python3 /vol/gcc/src/hg/master/local/gcc/m2/tools-src/def2doc.py -t
> -b/var/gcc/
> argv[0]: python3
> argv[1]: /vol/gcc/src/hg/master/local/gcc/m2/tools-src/def2doc.py
> argv[2]: -t
> argv[3]: -b/var/gcc/regression/master/11.4-gcc/build/gcc/m2
> argv[4]:
> -f/var/gcc/regression/master/11.4-gcc/build/gcc/m2/gm2-libs/SYSTEM.def
> argv[5]: -o
> argv[6]: /var/gcc/regression/master/11.4-gcc/build/gcc/m2/SYSTEM-pim.texi
>
> Both were in loops trying to read from fd 4 (unfortunately I forgot to check
> what that was): the 8 kB reads only returned 0, but the process kept
> trying...
This just happened again, but now I've managed to collect some
additional information:
There are 2 python processes:
PID USERNAME NLWP PRI NICE SIZE RES STATE TIME CPU COMMAND
10128 ro 1 0 0 20M 18M cpu/154 211:31 0.52% python3.11
10125 ro 1 0 0 20M 18M cpu/105 211:31 0.52% python3.11
namely
10128 python3
/vol/gcc/src/hg/master/local/gcc/m2/tools-src/def2doc.py -t
-b/var/gcc/regression/master/11.4-gcc/build/gcc/m2
-f/var/gcc/regression/master/11.4-gcc/build/gcc/m2/gm2-libs/SYSTEM.def -o
/var/gcc/regression/master/11.4-gcc/build/gcc/m2/SYSTEM-pim.texi
10125 python3
/vol/gcc/src/hg/master/local/gcc/m2/tools-src/def2doc.py -t
-b/var/gcc/regression/master/11.4-gcc/build/gcc/m2
-f/var/gcc/regression/master/11.4-gcc/build/gcc/m2/gm2-libs-iso/SYSTEM.def -o
/var/gcc/regression/master/11.4-gcc/build/gcc/m2/SYSTEM-iso.texi
Both are trying to read in a loop although not getting anything, like
14:38:01.688168 read(4, 0x7DEF87CA0, 8192) = 0
14:38:01.688200 read(4, 0x7DEF87CA0, 8192) = 0
14:38:01.688233 read(4, 0x7DEF87CA0, 8192) = 0
14:38:01.688265 read(4, 0x7DEF87CA0, 8192) = 0
and so on, blocking the build.
fd 4 is
4: S_IFREG mode:0644 dev:262,65548 ino:1239531 uid:2110 gid:4620 size:0
O_RDONLY|O_LARGEFILE FD_CLOEXEC
/var/gcc/regression/master/11.4-gcc/build/gcc/m2/gm2-libs/SYSTEM.def
offset:0
which is an empty file:
-rw-r--r-- 1 ro gcc 0 Jul 6 11:04
/var/gcc/regression/master/11.4-gcc/build/gcc/m2/gm2-libs/SYSTEM.def
