25 May
2011
25 May
'11
12:03 p.m.
Ouch...
Unfortunately I have a laptop, so or I get to work... or I change it
altogether (and I just bought it two months ago...especially to play
music!).
What gives me hope though is that:
- I have en XHCI USB3 port on the same laptop... Module snd-usb-audio
doesn't work on it yet, but I don't despair.
- Most important is that yesterday, I compiled the 2.6.39 kernel with
irqthreads, setup the rtirq script, removed every module I wasn't
using to free the irqs (xhci for the moment, wifi, other sound cards,
etc...) and I was down to only 2/3 crackles (and 3 messages in dmesg)
when playing a song with Ardour3 and Hydrogen playing along.
I still have to test and turn off KMS as I have read another user who
reported the same crackles (but no error in dmesg, because he has to
have alsa compiled with debug option to get them) and improved a lot
the thing with this option off...
Strange that the HP Envy 17, which is apparently designed with music
production in mind isn't set up properly hardware wise to support USB
sound cards!
25 May
25 May
4:45 p.m.
New subject: Cracking sound via USB (ALSA urb.c:480: frame 5 active: -18)
Yeah, I'm not convinced are issues are the same, though they seem
related. From reading your posts your crackling was happening on a
timed interval, where as mine is more of a random static.
I have a laptop as well, but it is going on 5 years old, and this
problem is new since kernels .33+. I don't think I have this new USB
bus/hub that you are talking about.
That said, doing 2.6.39 with rtirq and nokms is the only way to solve it
in my case. That or use a really old release of a distro, which gives
me other issues non audio related.
-BB
On 5/25/11 5:03 AM, Aurélien Leblond wrote:
Ouch...
Unfortunately I have a laptop, so or I get to work... or I change it
altogether (and I just bought it two months ago...especially to play
music!).
What gives me hope though is that:
- I have en XHCI USB3 port on the same laptop... Module snd-usb-audio
doesn't work on it yet, but I don't despair.
- Most important is that yesterday, I compiled the 2.6.39 kernel with
irqthreads, setup the rtirq script, removed every module I wasn't
using to free the irqs (xhci for the moment, wifi, other sound cards,
etc...) and I was down to only 2/3 crackles (and 3 messages in dmesg)
when playing a song with Ardour3 and Hydrogen playing along.
I still have to test and turn off KMS as I have read another user who
reported the same crackles (but no error in dmesg, because he has to
have alsa compiled with debug option to get them) and improved a lot
the thing with this option off...
Strange that the HP Envy 17, which is apparently designed with music
production in mind isn't set up properly hardware wise to support USB
sound cards!
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
5:37 p.m.
New subject: Cracking sound via USB (ALSA urb.c:480: frame 5 active: -18)
I'm currently testing rt performance with (( 2.6.39 SMP PREEMPT ))
Compiled with tickless timer, config-hz 1000, hr-timer, low-latency preemtion.
Crackling sound, here too ! But with a PCI card.
Maybe it's due to the rtirq script, which doesn't seem to work as
expected with .39 ?
Maybe I selected any incompatible options ?
What is the KMS option ?
According to jack, it's running in real-time mode.
I'm waiting for a working .config now.
--
E.R.
6:54 p.m.
New subject: Cracking sound via USB (ALSA urb.c:480: frame 5 active: -18)
On 05/25/2011 05:36 PM, Emanuel Rumpf wrote:
I'm currently testing rt performance with ((
2.6.39 SMP PREEMPT ))
Compiled with tickless timer, config-hz 1000, hr-timer, low-latency preemtion.
same here.
Crackling sound, here too ! But with a PCI card.
No problems here with UA-25 USB and internal HDA Intel.
Maybe it's due to the rtirq script, which
doesn't seem to work as
expected with .39 ?
You need to add 'threadirqs' to the kernel-boot option to enable
threaded IRQ handlers with 2.6.39.
rtirq works just fine.
I'm waiting for a working .config now.
I'm waiting for feedback from Daniel about a 2.6.39 kernel for
openDAW/64studio. Stay tuned.
The .config I'm currently using successfully on i386 thinkpad X60s is:
http://rg42.org/_media/wiki/kernel/config-2.6.39.gz
Cheers!
robin
7:49 p.m.
New subject: Cracking sound via USB (ALSA urb.c:480: frame 5 active: -18)
2011/5/25 Robin Gareus <robin(a)gareus.org>rg>:
You need to add 'threadirqs' to the kernel-boot option to enable
threaded IRQ handlers with 2.6.39.
Things have changed, I'm not quite sure how to do that these days.
Add it to GRUB_CMDLINE_LINUX_DEFAULT in /etc/default/grub ? ? ?
I'm waiting for feedback from Daniel about a
2.6.39 kernel for
openDAW/64studio. Stay tuned.
The .config I'm currently using successfully on i386 thinkpad X60s is:
http://rg42.org/_media/wiki/kernel/config-2.6.39.gz
Thanks.
There's a tool called irqbalance
Running that, is it still recommended with threadirqs ?
--
E.R.
8:22 p.m.
New subject: Cracking sound via USB (ALSA urb.c:480: frame 5 active: -18)
2011/5/25 Emanuel Rumpf <xbran(a)web.de>de>:
2011/5/25 Robin Gareus <robin(a)gareus.org>rg>:
Yes! And don't forget update-grub
The rtqrq script is working now.
Performance is a lot better, but
there are still xruns, especially with the ICE1712 driver.
Strange, but the internal snd-hda-intel causes far less xruns.
(testet with 8ms latency)
--
E.R.
You need to add 'threadirqs' to the kernel-boot option to enable
threaded IRQ handlers with 2.6.39.
Things have changed, I'm not quite sure how to do that these days.
Add it to GRUB_CMDLINE_LINUX_DEFAULT in /etc/default/grub ? ? ?
8:23 p.m.
New subject: Cracking sound via USB (ALSA urb.c:480: frame 5 active: -18)
On 05/25/2011 07:48 PM, Emanuel Rumpf wrote:
I don't know. I do run irqbalance here; but it does not seem to have
effect on hardware-irqs [anymore] (see dumps below).
But I'm getting off-topic.
There are no crackles nor xruns here with and without irqbalance.
jackd2-svn-r4393, The UA-25 is on IRQ 18;
Are your crackles gone with the 'threadirqs' option?
robin
# cat /proc/interrupts
CPU0 CPU1
0: 278696424 6 IO-APIC-edge timer
1: 765597 0 IO-APIC-edge i8042
8: 1 0 IO-APIC-edge rtc0
9: 19624127 0 IO-APIC-fasteoi acpi
12: 9265100 0 IO-APIC-edge i8042
14: 2635256 4 IO-APIC-edge ata_piix
15: 0 0 IO-APIC-edge ata_piix
16: 1238945 0 IO-APIC-fasteoi yenta, uhci_hcd:usb2,
i915@pci:0000:00:02.0
17: 6413 0 IO-APIC-fasteoi uhci_hcd:usb3, firewire_ohci
18: 2273102 2 IO-APIC-fasteoi uhci_hcd:usb4, mmc0
19: 7270696 23 IO-APIC-fasteoi ehci_hcd:usb1, uhci_hcd:usb5
40: 0 0 PCI-MSI-edge PCIe PME
41: 0 0 PCI-MSI-edge PCIe PME
42: 0 0 PCI-MSI-edge PCIe PME
43: 0 0 PCI-MSI-edge PCIe PME
44: 8880802 38 PCI-MSI-edge hda_intel
45: 20916424 16 PCI-MSI-edge iwl3945
NMI: 2647 2909 Non-maskable interrupts
LOC: 77842296 233977838 Local timer interrupts
SPU: 0 0 Spurious interrupts
PMI: 2647 2909 Performance monitoring interrupts
IWI: 0 0 IRQ work interrupts
RES: 85016190 124517149 Rescheduling interrupts
CAL: 332967 582697 Function call interrupts
TLB: 849663 604581 TLB shootdowns
TRM: 12 12 Thermal event interrupts
THR: 0 0 Threshold APIC interrupts
MCE: 0 0 Machine check exceptions
MCP: 1144 1124 Machine check polls
ERR: 0
MIS: 0
# cat /proc/softirqs
CPU0 CPU1
HI: 71 0
TIMER: 78744168 77532112
NET_TX: 93266 813476
NET_RX: 104790 97888
BLOCK: 1482548 3411882
BLOCK_IOPOLL: 71 0
TASKLET: 10734408 19778443
SCHED: 4012801 4799134
HRTIMER: 215170 248879
RCU: 21928395 24010120
# /etc/init.d/rtirq status
PID CLS RTPRIO NI PRI %CPU STAT COMMAND
970 FF 88 - 128 0.0 S irq/18-uhci_hcd
317 FF 84 - 124 0.2 S irq/44-hda_inte
610 FF 80 - 120 0.0 S irq/1-i8042
609 FF 79 - 119 0.0 S irq/12-i8042
159 FF 50 - 90 0.1 S irq/9-acpi
318 FF 50 - 90 0.0 S irq/18-mmc0
321 FF 50 - 90 0.0 S irq/16-i915@pci
322 FF 50 - 90 0.0 S irq/17-firewire
443 FF 50 - 90 0.0 S irq/40-PCIe PME
444 FF 50 - 90 0.0 S irq/41-PCIe PME
445 FF 50 - 90 0.0 S irq/42-PCIe PME
446 FF 50 - 90 0.0 S irq/43-PCIe PME
569 FF 50 - 90 0.0 S irq/14-ata_piix
570 FF 50 - 90 0.0 S irq/15-ata_piix
589 FF 50 - 90 0.0 S irq/16-yenta
929 FF 50 - 90 0.0 S irq/8-rtc0
963 FF 50 - 90 0.0 S irq/19-ehci_hcd
967 FF 50 - 90 0.0 S irq/16-uhci_hcd
969 FF 50 - 90 0.0 S irq/17-uhci_hcd
972 FF 50 - 90 0.0 S irq/19-uhci_hcd
29970 FF 50 - 90 0.5 S irq/45-iwl3945
3 TS - 0 19 0.1 S ksoftirqd/0
315 TS - 0 19 0.2 S ksoftirqd/1
3620 TS - -20 39 0.0 S< kvm-irqfd-clean
24746 TS - 0 19 0.0 S+ rtirq
2011/5/25 Robin Gareus <robin(a)gareus.org>rg>:
Yes, but it depends on your bootloader and distro.
I use GRUB_CMDLINE_LINUX="i915.modeset=1 threadirqs"
in /etc/default/grub. run `sudo update-grub` after saving changes to
that file.
If you install the kernel as debian package the 'update-grub' part is
taken care of by dpkg's postinst script.
You need to add 'threadirqs' to the kernel-boot option to enable
threaded IRQ handlers with 2.6.39.
Things have changed, I'm not quite sure how to do that these days.
Add it to GRUB_CMDLINE_LINUX_DEFAULT in /etc/default/grub ? ? ? I'm
waiting for feedback from Daniel about a 2.6.39 kernel for
openDAW/64studio. Stay tuned.
The .config I'm currently using successfully on i386 thinkpad X60s is:
http://rg42.org/_media/wiki/kernel/config-2.6.39.gz
Thanks.
There's a tool called irqbalance
Running that, is it still recommended with threadirqs ?
27 May
27 May
6:49 p.m.
New subject: Cracking sound via USB (ALSA urb.c:480: frame 5 active: -18)
2011/5/25 Robin Gareus <robin(a)gareus.org>rg>:
Uff. Near to crazy here.
Your config performs much better latency wise,
but no matter what - I'm unable to get my usb-wlan-stick (realtec or
compatible rt2870) working.
There are unresolved symbols:
rt2x00usb: Unknown symbol rt2x00lib_suspend (err 0)
rt2x00usb: Unknown symbol rt2x00lib_probe_dev (err 0)
rt2x00usb: Unknown symbol rt2x00lib_dmadone (err 0)
rt2x00usb: Unknown symbol rt2x00lib_txdone_noinfo (err 0)
rt2x00usb: Unknown symbol ieee80211_free_hw (err 0)
rt2x00usb: Unknown symbol ieee80211_alloc_hw (err 0)
rt2x00usb: Unknown symbol rt2x00lib_rxdone (err 0)
rt2x00usb: Unknown symbol rt2x00queue_get_entry (err 0)
rt2x00usb: Unknown symbol rt2x00lib_remove_dev (err 0)
rt2x00usb: Unknown symbol rt2x00queue_for_each_entry (err 0)
rt2x00usb: Unknown symbol rt2x00queue_flush_queue (err 0)
rt2x00usb: Unknown symbol rt2x00lib_dmastart (err 0)
rt2x00usb: Unknown symbol rt2x00lib_resume (err 0)
I'm currently clueless how to resolve it.
Shouldn't selecting the driver in the kernel config include the
required stuff ??
With my original config, there's no such problem, btw.
--
E.R.
The .config I'm currently using successfully on i386 thinkpad X60s is:
http://rg42.org/_media/wiki/kernel/config-2.6.39.gz
Thanks.
4740
days inactive
4742
days old
linux-audio-user@lists.linuxaudio.org
7 comments
4 participants
participants (4)
-
Aurélien Leblond -
Brian Bergstrom -
Emanuel Rumpf -
Robin Gareus