[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Running the test suite faster.

From: Philip Martin <philip_at_codematters.co.uk>
Date: 2007-11-03 12:43:52 CET

"Justin Erenkrantz" <justin@erenkrantz.com> writes:

> On Nov 1, 2007 5:00 PM, Karl Fogel <kfogel@red-bean.com> wrote:
>> Is 'make check' really slowly for you? Does it take hours and hours
>> to complete a full three-RA-layer check?
>
> I just did some timings with the basic tests (Ubuntu 7.10):
>
> ra_serf via davautocheck.sh basic:
>
> /dev/random: 1 minute 33 seconds
> /dev/urandom: 1 minute 31 seconds
>
> ra_local via 'make check TESTS=subversion/tests/cmdline/basic_tests.py':
>
> /dev/random: 1 minute 27 seconds
> /dev/urandom: 1 minute 26 seconds
>
> That isn't statistically significant.
>
> What am I doing wrong? Is it different for you?
>
> Or, is the slowdown *not* associated with basic tests - perhaps
> there's a set of tests that are prone to the entropy source? This
> would lead me to believe there's a set of tests (or code paths) that
> are at fault. But, if it doesn't manifest itself in the basic tests,
> then I think it's something quite code path specific and not endemic
> to the overall system.
>
> FWIW, my Mac OS X 10.4 install takes over twice as long to do basic
> tests - about 3:28.

It depends on how fast the hardware generates entropy in comparison to
how fast the testsuite runs. I used to use an older dual processor
machine with multiple SCSI disks, multiple network cards, a separate
video card, etc. and it ran the testsuite without stalling. I now use
a newer machine with onboard video/LAN/SATA and a single disk and it
runs the testsuite much faster but will stall when it runs out of
entropy. These stalls are most visible when using ra_svn. The times
you give for the tests are relatively slow compared to the times I get
out of my new machine where basic_tests over ra_local or ra_dav runs
in under 20 seconds. Perhaps your machine is simply too slow to drain
its entropy pool? On my machine ra_svn is as fast as the other layers
when using /dev/urandom, under 20 seconds, but takes over 90 seconds
using /dev/random.

On Linux you can use

  cat /proc/sys/kernel/random/entropy_avail
 
to see how much entropy is available, and ra_svn does drain it more
noticeably than the other layers.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Sat Nov 3 12:44:15 2007

This is an archived mail posted to the Subversion Dev mailing list.