Hello, I have some ugly numbers given by zttest for ztdummy on an AMD64 box running linux-2.6.15 compiled for Athlon64. linux-2.6.15, zaptel/branches/1.2 r900, jiffies ./zttest Opened pseudo zap interface, measuring accuracy... 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.987793% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.987793% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.987793% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% 99.975586% --- Results after 136 passes --- Best: 99.987793 -- Worst: 99.975586 -- Average: 99.975853 linux-2.6.15, zaptel/branches/1.2 r900, RTC Opened pseudo zap interface, measuring accuracy... 99.938965% 99.963379% 99.938965% 99.963379% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.963379% 99.938965% 99.963379% 99.951172% 99.963379% 99.938965% 99.963379% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.963379% 99.938965% 99.963379% 99.938965% 99.951172% 99.938965% 99.963379% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% 99.938965% 99.963379% --- Results after 96 passes --- Best: 99.963379 -- Worst: 99.938965 -- Average: 99.952942 linux-2.6.15, zaptel/branches/1.2 r900+patch bugs.digium.com/view.php?id=5971, RTC Opened pseudo zap interface, measuring accuracy... 99.987793% 99.719238% 99.707031% 100.000000% 99.890137% 99.865723% 100.000000% 99.987793% 100.000000% 99.975586% 100.000000% 99.987793% 99.975586% 100.000000% 100.000000% 99.987793% 99.975586% 99.768066% 99.768066% 99.987793% 99.926758% 99.926758% 99.987793% 99.975586% 100.000000% 99.975586% 99.987793% 100.000000% 100.000000% 99.975586% 99.938965% 100.000000% 99.975586% 99.816895% 99.816895% 100.000000% 99.987793% 99.975586% 100.000000% 99.975586% 100.000000% 99.987793% 100.000000% 99.975586% 99.987793% 100.000000% 99.719238% 99.707031% 99.987793% 99.877930% 99.865723% 100.000000% 99.987793% 99.975586% 100.000000% 99.987793% 100.000000% 99.975586% 100.000000% 99.987793% 99.987793% 100.000000% 99.768066% 99.768066% 99.975586% 99.938965% 99.926758% 99.975586% 100.000000% 99.987793% 100.000000% 99.975586% 100.000000% 99.975586% 99.987793% 100.000000% 100.000000% 99.987793% 99.829102% 99.816895% 99.975586% 99.987793% 99.975586% 100.000000% 99.987793% 99.975586% 100.000000% 99.975586% 99.987793% 100.000000% 100.000000% 99.719238% 99.694824% 100.000000% 99.890137% 99.877930% 99.987793% 100.000000% 99.975586% 99.987793% 100.000000% 99.975586% 99.987793% 99.987793% 100.000000% 99.975586% 100.000000% 99.780273% 99.755859% 100.000000% 99.938965% 99.938965% 99.975586% 100.000000% 99.987793% 99.975586% 100.000000% 99.987793% 100.000000% 99.975586% 100.000000% 99.987793% 99.975586% 99.816895% 99.816895% 100.000000% 100.000000% 99.975586% 99.975586% 99.987793% 100.000000% 100.000000% 99.987793% 99.975586% 100.000000% 99.975586% --- Results after 136 passes --- Best: 100.000000 -- Worst: 99.694824 -- Average: 99.951973 HW: Tyan Tomcat K8E, Athlon64 3000+, 1GB RAM, 3ware 8006, 2x Maxtor HDD SW: Ubuntu 5.10, linux-2.6.15, zaptel from 1.2 branch Any idea what can be wrong? Thanks in advance, Tamas
On Mon, 2006-01-16 at 17:43 +0100, Tamas wrote:> Hello, > > I have some ugly numbers given by zttest for ztdummy on an AMD64 box > running linux-2.6.15 compiled for Athlon64. > > linux-2.6.15, zaptel/branches/1.2 r900, jiffies > ./zttest > Opened pseudo zap interface, measuring accuracy...[snip]> --- Results after 136 passes --- > Best: 99.987793 -- Worst: 99.975586 -- Average: 99.975853 > > linux-2.6.15, zaptel/branches/1.2 r900, RTC > Opened pseudo zap interface, measuring accuracy...> [snip]> --- Results after 96 passes --- > Best: 99.963379 -- Worst: 99.938965 -- Average: 99.952942 > > linux-2.6.15, zaptel/branches/1.2 r900+patch > bugs.digium.com/view.php?id=5971, RTC > > Opened pseudo zap interface, measuring accuracy...[snip]> --- Results after 136 passes --- > Best: 100.000000 -- Worst: 99.694824 -- Average: 99.951973 > > HW: > Tyan Tomcat K8E, Athlon64 3000+, 1GB RAM, 3ware 8006, 2x Maxtor HDD > > SW: > Ubuntu 5.10, linux-2.6.15, zaptel from 1.2 branch > > Any idea what can be wrong?What does your /proc/interrupts say? On my asterisk box, I was seeing crappy interrupt handling like this only when I was using XT-PIC interrupt handling, when I moved to IO-APIC, things got much better... Steve
Tony Mountifield
2006-Jan-16 14:59 UTC
[Asterisk-Users] Re: ztdummy inaccuracy on linux-2.6
In article <43CBCD3F.9050006@gmail.com>, Tamas <jalsot@gmail.com> wrote:> Hello, > > I have some ugly numbers given by zttest for ztdummy on an AMD64 box > running linux-2.6.15 compiled for Athlon64.Don't be misled by the apparent ugliness of the numbers! Are you hearing any effects in the audio? If not, then stop worrying! The output of zttest is misleading, since it is just relying on the kernel time of day clock for measuring. Because of this, the "jiffies" version will appear to show more consistent results, and the RTC version will be a little more variable. But that is not because the RTC version of ztdummy is less accurate. On the contrary, it is more accurate, on average (8000 calls out of every 8192 per second). It is the reference for the calculations that is less accurate. I'm glad that your post mentioned bug number 5971, as I hadn't been aware of that one until now. I think the patch in that bug is seriously flawed, and have just posted on it to that effect.> [... snip results ...]> > HW: > Tyan Tomcat K8E, Athlon64 3000+, 1GB RAM, 3ware 8006, 2x Maxtor HDD > > SW: > Ubuntu 5.10, linux-2.6.15, zaptel from 1.2 branch > > Any idea what can be wrong?Nothing! Please ignore the figures, and use ztdummy with USE_RTC without the patch from 5971. Then just describe any operational effect you might be seeing, if any.> Thanks in advance, > TamasCheers Tony -- Tony Mountifield Work: tony@softins.co.uk - http://www.softins.co.uk Play: tony@mountifield.org - http://tony.mountifield.org