Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 13 Apr 2007 13:56:13 -0700
From:      Nate Lawson <nate@root.org>
To:        Eygene Ryabinkin <rea-fbsd@codelabs.ru>
Cc:        max@love2party.net, freebsd-current@freebsd.org, pf@freebsd.org
Subject:   Re: call for testers: altq in current
Message-ID:  <461FEE6D.4030201@root.org>
In-Reply-To: <20070413204237.GG49158@codelabs.ru>
References:  <4617D3A6.8000201@root.org> <20070409094010.GL26348@codelabs.ru> <461FDD28.6030502@root.org> <20070413204237.GG49158@codelabs.ru>

next in thread | previous in thread | raw e-mail | index | archive | help
Eygene Ryabinkin wrote:
> Nate, good day!
> 
> Fri, Apr 13, 2007 at 12:42:32PM -0700, Nate Lawson wrote:
>>> I see no difference between the -CURRENT from today and from 30th
>>> March (I see that your commit was made at 26th of March, but I am
>>> not sure that mu current was updated after it for the -CURRENT
>>> compiled at 30th of March).
>>>
>>> The bad news are that the ALTQ behaves wrong: when the CPU frequency
>>> is changed the bandwidth changes too. Either I am doing something
>>> wrong, or your commit should be polished a bit.
>> First, add a printf at line 915 (end of function tsc_freq_changed() in
>> sys/contrib/altq/altq/altq_subr.c):
>> 	printf("machclk_freq now %d\n", machclk_freq);
>                                  ^^
> Should be %lu, I believe?

Sure, whatever's right.

>> Does it trigger when you change the cpu freq?  Is the number printent
>> correct (i.e. 400 million for 400 Mhz)?
> 
> Yes, the numbers are perfectly correct. I will try to redo the
> tests on Monday (when I will be able to use the LAN link) and
> will watch for this debug information. Any other recommendations
> are, of course, welcome.

Ok, that is good to know the code is running and the freq values are
correct.  Can you verify through some other cpu benchmark test that the
freq actually did change to the value printed?

Also, make sure you're not using the TSC timecounter.  sysctl
kern.timecounter

>>> First two logs, ifstat.bw3Kb.old.wan.log and ifstat.bw3Kb.new.wan.log
>>> do show the WAN results. The 100 Kbps corresponds to 400 MHz, 200
>>> Kbps -- to 800 MHz, 410 Kbps -- to 1600 MHz and 560 Kbps -- to 2200
>>> MHz CPU speed. I thought that I was bounded by the WAN link here.
>> What was the CPU speed on bootup?
> 
> 2200 MHz.

I don't understand those values.  Didn't you setup a constant 3 Kb/sec
link?  so why would you be getting even 100 Kbps at 400 Mhz?

On the new code but without loading cpufreq and leaving the freq at 2200
Mhz, do you get the right numbers?  Are they constant?

-- 
Nate



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?461FEE6D.4030201>