[tor-dev] Fwd: [tor-relays] tcmalloc in FreeBSD

Nick Mathewson nickm at alum.mit.edu
Wed Aug 8 23:43:56 UTC 2012


On Wed, Aug 8, 2012 at 7:09 PM, Jordi Espasa Clofent
<jespasac at minibofh.org> wrote:
> Here is the log:
>
> ===>  Installing for tor-0.2.2.37
> ===>   tor-0.2.2.37 depends on file: /usr/local/lib/libcrypto.so.7 - found
> ===>   tor-0.2.2.37 depends on shared library: event-2.0 - found
> ===>   tor-0.2.2.37 depends on shared library: tcmalloc - found

So here's the line that makes Tor:

> cc  -O2 -pipe  -fno-strict-aliasing -D_FORTIFY_SOURCE=2
> -fstack-protector-all -fwrapv -fPIE -Wstack-protector --param
> ssp-buffer-size=1 -Wall -fno-strict-aliasing -L/usr/local/lib
> -L/usr/local/lib/event2 -rpath=/usr/local/lib -pie -Wl,-z,relro -Wl,-z,now
> -o  tor tor_main.o ./libtor.a ../common/libor.a ../common/libor-crypto.a
> ../common/libor-event.a  -lz -lm -levent-2.0 -lssl -lcrypto -pthread

and yeah, there's no tcmalloc there.

This is a little strange; when I try it with Tor master, I get
(removing my pile of warning options):

gcc  -g -O2 -D_FORTIFY_SOURCE=2 -fstack-protector-all
-Wstack-protector -fwrapv -fPIE -Wall -fno-strict-aliasing -ltcmalloc
-pie -z relro -z now -o tor tor_main.o ./libtor.a ../common/libor.a
../common/libor-crypto.a ../common/libor-event.a -lz -lm -levent -lssl
-lcrypto  -lpthread -lrt -ldl

 [...]
> Is plain to see the script is checking if the tcmalloc is there, but more
> than that... I cannot see the compiler itself is calling the mentioned
> library.

Can you see what arguments are being passed to configure, and what
configure does with them?  Is the freebsd build process passing
--with-tcmalloc to the configure script?

-- 
Nick


More information about the tor-dev mailing list