On 02/17/2015 04:13 PM, grarpamp wrote:
> FreeBSD linux kernel module does not yet support 64bit.
> For reference, 32bit:
> https://svnweb.freebsd.org/base/head/sys/compat/linux/
> https://svnweb.freebsd.org/base/head/sys/i386/linux/
> https://svnweb.freebsd.org/base/head/sys/amd64/linux32/
OpenBS
Has anyone looked into this? I talked to the maintainer of the OpenBSD
Firefox port, but he wasn't very interested and pointed out the
difficulty caused by the deterministic build system.
I can verify that it doesn't work out of the box, but haven't had time
to play with it much more than that. I
On 01/03/2015 02:36 AM, Yawning Angel wrote:
> This all is kind of a moot point because even if the relevant time
> calls did take ~2 usec it still doesn't explain the performance issues,
> and my curiosity is close to being exhausted. But, for what it's worth.
>
> Forcing the timecounter hardwar
On 01/02/2015 05:20 AM, Yawning Angel wrote:
> So while optimization is cool and all, I'm not seeing why this
> specifically is the underlying issue.
A lot of people have been reporting underperformance on OpenBSD, and
time syscalls are a very common source of performance discrepancy
between Linux
27;ll look into it myself as well.
On 01/01/2015 11:42 PM, Libertas wrote:
> On 01/01/2015 03:47 PM, Nick Mathewson wrote:
>> What part of tor is accounting for all these time/gettimeofday calls?
>
> I set up an OpenBSD 5.6 VM and built Tor 0.2.5.10 on it. I couldn't get
> ch
thoughts.
Let me know if I've made any mistakes thus far. I'll start looking into
potential improvements. I'm eager to work with someone on this if
anyone's interested. Also, if you have any ideas or suggestions, please
share.
Thanks
On 01/01/2015 08:07 AM, teor wrote:
> I'd like to get a sense of how many calls per second this represents.
> (400,000 would seem to be 100 to 1000 per second, unless you were debugging
> for a long while.)
My kdump suggested that there were almost 7,000 calls a second to
gettimeofday(). The deta
I'm working on question #2 now - I'm less well equipped to work on #1 at
the moment.
It's probably also worth mentioning that the dump with >400,000 calls to
gettimeofday() also had 30,714 calls to clock_gettime().
On 01/01/2015 03:47 PM, Nick Mathewson wrote:
>
> On Dec 31, 2014 10:38 PM, "Yawn
Also, I can do some testing with chutney on the OpenBSD VM in a few hours.
signature.asc
Description: OpenPGP digital signature
___
tor-dev mailing list
tor-dev@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-dev
To clarify, I used "strings /usr/local/bin/tor | grep approx_time".
signature.asc
Description: OpenPGP digital signature
___
tor-dev mailing list
tor-dev@lists.torproject.org
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-dev
> I wonder if TIME_IS_FAST is being defined as 1 in any of these Linux or BSD
> builds.
I just found that the current OpenBSD release (5.6) has TIME_IS_FAST =
0. Details are in a ticket comment I just posted.
> Libertas, can you search your tor binary or tor debug symbols for the
&
I just ran ktrace/kdump (used for observing system calls) on the Tor
process of my exit node, which relays about 800 KB/s. It listed >400,000
calls to gettimeofday(). The list was swamped with them.
I think I remember reading somewhere that that sort of system call is
way slower in OpenBSD than Li
reading it, you can
probably guess which *nix flavors I'm familiar with.
Enjoy,
Libertas
-BEGIN PGP SIGNATURE-
Version: GnuPG v1
iQIcBAEBCAAGBQJUc7qtAAoJELxHvGCsI27NOwsP/1bP+g4W2y1NHefR5bd+5hwo
V927TXzarNneNFxfhKJsVewVnU593Ex13iHv+WgA+XM39xiK8KkhpxOaw2IpahZp
pIRCk4/XX78YaEc/K3ci
in OpenBSD, so Tor will
probably default to 1024 regardless of what openfiles-max is.
I'm still a beginner, so let me know if I made any mistakes. Also, let
me know if I should send this to the relays list as well.
I can write up a short draft paragraph if people think this is
accura
14 matches
Mail list logo
