Re: Problems with big sequences

[Antti T Tuominen <attuomin cc helsinki fi>]

Hi,

> Actually that looks like a bug in the linux writev implementation. I'm not

Yes, something odd seemed to happen to the vector.

> just dumps core.... I changed it to use at most 1000 iovecs.
> Please try the following new version. It worked here:

Works fine! Thank you.

> http://i81pc4.ira.uka.de/ORBit/ORBit-mt-0.5.3a.tar.gz

I'm running ORBit-mt-0.5.4 now, setting ORBIT_MICRO_VERSION=3a didn't seem
like a good idea since it lead to assignments like

const guint orbit_micro_version = ORBIT_MICRO_VERSION;
(in src/orb/orbit.c, line 43)

Hmm... maybe 0x3a :-)

Regards,
Antti
-- 
main(G,C,c){char*d="@?dddTO OHHG@HH?I HH?I@oA@@@p`a?a`p NAAO@FIIF OHLHG@oA@"
"OHHG FMMD@OHHG@",*n="@@@@@@@@@@ PHDBA";for(puts("\33[2J"),C=72;C++;usleep(
35000))for(G=0;G++<75;)for(c=5;c<21;)printf("%c\33[%d;%dH",d[(C+G)%75]&n[(int
)(5*sin(C%628/1e1)*sin((C-G)/(8+3*sin(C%1257/20.0)))+c++)%16]&63?56:32,c,G);}