Your message dated Sat, 07 Nov 2020 14:35:18 +0100
with message-id <9a8ae98b8253a791d4eef4ed6177f0cf4bce8bc0.ca...@gmail.com>
and subject line Fixed on giac's side
has caused the Debian Bug report #973588,
regarding Update from pari 2.11 to 2.13 breaks giac compilation
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact ow...@bugs.debian.org
immediately.)
--
973588: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=973588
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems
--- Begin Message ---
Package: libpari-dev
Version: 2.13.0-2
Severity: grave
The last pari upload broke giac on amd64, arm64 and mipsel :
https://buildd.debian.org/status/fetch.php?pkg=giac&arch=amd64&ver=1.6.0.25%2Bdfsg1-2%2Bb1&stamp=1604135053&raw=0
The failing test on amd64 is a segfault. I could check that running the
failing test by hand while exporting PARI_SIZE=1024000 makes the
problem disappear, and PARI_SIZE=512000 or nothing makes it come back ;
here is an abstract from giac's pari.cc:
static struct pari_constants_initialization {
pari_constants_initialization () {
if (getenv("PARI_SIZE")){
string pari_size_s(getenv("PARI_SIZE"));
pari_mem_size= atoi(pari_size_s.c_str());
}
entree * ptr=functions_basic;
for (;ptr->name;++ptr){
pari_function_table[ptr->name]=ptr;
}
}
} pari_constants_initialization_singleton;
static void call_pari_init()
{
// do not initialize INIT_JMP so that PARI error do not exit
pari_init_opts(pari_mem_size,pari_maxprime,INIT_SIGm | INIT_DFTm);
paristack_setsize(pari_mem_size, (1<<30));
// initialize variable ordering x,y,z
gp_read_str("[x,y,z,t]");
}
Since it worked with pari 2.11 and breaks with pari 2.13, I guess the
problem is on pari's side.
Setting severity to grave because it breaks another package.
I hope it helps,
JP
--- End Message ---
--- Begin Message ---
Hi,
the problem seems in large part to be coming giac's initialisation
code, and the recent giac 1.6.0.25+dfsg1-3 upload made it disappear.
I'm hence closing this report against pari.
Thanks,
JP
--- End Message ---
