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

tor callgrinds



Thought you guys might find this interesting. I did a couple of callgrind
runs on 2 different tor builds, 1 using -Os and the other using -O3. The
difference between the two optimization options is not particularly
interesting, but the visualization of the tor profile overall is. The
latter -O3 run was profiled for 26 hours w/ a base bandwidth of 1MB/sec
(although my graphs show only around 600kB/sec average) and the former -Os
only an hour before I shut it down to do a longer -O3 run.

http://tor.anodized.com/callgrind/

cflags are for tor.
cflags.lev are for libevent.

One thing I notice is zlib taking up a lot of ground, that's for sure.
I also made a couple of interesting snapshots around the rjindaelEncrypt()
function in aes.c

The grinds are downloadable if you want to check things out in a local
kcachegrind visualizer.

-cl