[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: [tor-talk] tor-0.2.2.32 compilation failure



> Wow.  Your compiler doesn't like 'double d; printf("%lf", d);' .
> That's pretty... special.  :)

I'm sure at the time gcc 2.95.4 was pretty typical. The fact that
I've got such a relic, yeah... special :)

This may be where it's coming from, gcc(1): -Wformat is included
in -Wall.

===== 1.c =====
#include <stdio.h>
#include <stdlib.h>
int main () {
double d=2;
printf("%lf\n",d);
exit(0); }
=====
rm -f ./a.out ; gcc -Wall -pedantic -std=c89 1.c ; ./a.out
1.c: In function 'main':
1.c:8: warning: ISO C90 does not support the '%lf' printf format
2.000000
rm -f ./a.out ; gcc -Wall -pedantic -std=c99 1.c ; ./a.out
2.000000
=====

> Also, if they don't work, try sending the output of "make -k" so
> that we get to see all the errors at once, and not just one file
> at a time.

OK. Will also post a cumulative patch so you know what gave the
output.

> Hm.  No lrint?

Not anywhere in /usr/include or man.

However on RELENG_8, yes, both lrint(3) and lround(3). See math(3).
http://www.freebsd.org/cgi/man.cgi

> I think it should be okay.

I was thinking about the change in type/size doing the wrong thing
like overflowing something and having a running but poor anonymity
result.

> RELENG_4 -- is not getting security patches any more

Everything RELENG_6 and earlier is EOL.
http://security.freebsd.org/

> Make sure that you've patched all your services!

Chroot is already containing some damage ;-)

> The log_mutex one is okay if this is really a no-threads build?
> I guess it would have to be.

I do have pthread(3) and gcc -pthread for linking libc_r. Though
the resulting Tor makefiles and compilation don't indicate that it
is in use.

Sorry gmail wraps output lines. I don't see an option to defeat it.
_______________________________________________
tor-talk mailing list
tor-talk@xxxxxxxxxxxxxxxxxxxx
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk