sbp
Posts: 128
Joined: Wed Sep 26, 2012 7:54 pm

Re: USB FIQ testing....

Wed Jun 12, 2013 2:57 pm

Hi
I'm trying to cross compile the fiq_split kernel (3.8.13) on my Debian computer.

It starts to compile fine, but just before it is finished I get this error and no kernel is produced:

Code: Select all

  CC      lib/rwsem-spinlock.o
  CC      lib/sha1.o
  CC      lib/show_mem.o
  CC      lib/string.o
  CC      lib/timerqueue.o
  CC      lib/vsprintf.o
  CC [M]  lib/crc-ccitt.o
  CC [M]  lib/crc7.o
  CC [M]  lib/ts_kmp.o
  CC [M]  lib/ts_bm.o
  CC [M]  lib/ts_fsm.o
  GEN     lib/crc32table.h
  CC      lib/crc32.o
  AR      lib/lib.a
  LD      lib/built-in.o
  LINK    vmlinux
: command not found
make: *** [vmlinux] Error 127
[email protected]:/mrproper1/linux-rpi-fiq.split#
Does anybody know what is causing this error

Can you please share a working .config file.

Steen
piCorePlayer webpage: https://sites.google.com/site/picoreplayer/home

gsh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 1452
Joined: Sat Sep 10, 2011 11:43 am

Re: USB FIQ testing....

Wed Jun 12, 2013 3:03 pm

the bcmrpi_defconfig works fine for me...

Gordon
--
Gordon Hollingworth PhD
Raspberry Pi - Director of Software Engineering

sbp
Posts: 128
Joined: Wed Sep 26, 2012 7:54 pm

Re: USB FIQ testing....

Wed Jun 12, 2013 4:18 pm

Thanks

Using the bcmrpi_defconfig gives this error:

Code: Select all

[email protected]:/mrproper1/linux-rpi-fiq.split# make ARCH=arm -j 3
scripts/kconfig/conf --silentoldconfig Kconfig
.config:891:warning: symbol value 'm' invalid for ZSMALLOC
*
* Restart config...
*
*
* General setup
piCorePlayer webpage: https://sites.google.com/site/picoreplayer/home

dom
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 5380
Joined: Wed Aug 17, 2011 7:41 pm
Location: Cambridge

Re: USB FIQ testing....

Wed Jun 12, 2013 8:27 pm

sbp wrote: Using the bcmrpi_defconfig gives this error:
That's a warning and is harmless if you don't need ZSMALLOC (which you almost certainly don't).

kalehrl
Posts: 350
Joined: Tue Jul 24, 2012 10:49 am

Re: USB FIQ testing....

Wed Jun 19, 2013 1:10 pm

Ethernet dies invariably after a few days with this kernel.
No connection via ssh possible.

gsh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 1452
Joined: Sat Sep 10, 2011 11:43 am

Re: USB FIQ testing....

Wed Jun 19, 2013 1:21 pm

Any debug available?

How do you know Ethernet dies?

Gorodn
--
Gordon Hollingworth PhD
Raspberry Pi - Director of Software Engineering

kalehrl
Posts: 350
Joined: Tue Jul 24, 2012 10:49 am

Re: USB FIQ testing....

Wed Jun 19, 2013 6:21 pm

I can no longer connect to it via ssh.
Also, the ping to google doesn't work.
I noticed that today I restarted my linux pc 2 times.
On that pc I have nfs server running and raspi connects to it.
In the logs I attach there are a lot of nfs related messages.
Could it be that ethernet dies when raspi tries to connect to an nfs server but cannot establish a connection?
Attachments
Desktop.zip
(15.67 KiB) Downloaded 173 times

tom.k.cook
Posts: 47
Joined: Fri Jun 22, 2012 8:51 am

Re: USB FIQ testing....

Fri Jun 21, 2013 2:39 pm

Hi Gordon,

One issue with the current source on github (I'm using the 3.8.y branch, but I believe it is common to all of them) is that there is no protection from data aborts in the FIQ handler. This means that attempts to use ftrace without first filtering out the dwc_otg FIQ handler cause a kernel oops (bad_mode is called).

It is possible to manually filter out the handler function using set_ftrace_notrace in the debugfs. But there must be a better way of doing it - after all, IRQ mode also doesn't handle data aborts, but they don't cause oopses when ftrace is enabled. I'm not sure what the right way to go about this is though.

Regards,
Tom

gsh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 1452
Joined: Sat Sep 10, 2011 11:43 am

Re: USB FIQ testing....

Fri Jun 21, 2013 2:53 pm

Yes I've seen that as well, although have recently just been tracing a small filter of functions and therefore get around it...

I don't know what the right solution is, but I assume you can suitably decorate a function to stop the ftrace from trying to insert a function call at the beginning (the problem is the beginning of the function there is no stack set up and therefore everything fails!)

If you've got the time to go search it would be a very useful pull request! (Otherwise it'll take me quite some time to fix)

Thanks

Gordon
--
Gordon Hollingworth PhD
Raspberry Pi - Director of Software Engineering

kalehrl
Posts: 350
Joined: Tue Jul 24, 2012 10:49 am

Re: USB FIQ testing....

Mon Jun 24, 2013 8:39 am

kalehrl wrote:I can no longer connect to it via ssh.
Also, the ping to google doesn't work.
I noticed that today I restarted my linux pc 2 times.
On that pc I have nfs server running and raspi connects to it.
In the logs I attach there are a lot of nfs related messages.
Could it be that ethernet dies when raspi tries to connect to an nfs server but cannot establish a connection?
Any progress on this?

ddv2005
Posts: 23
Joined: Fri Jul 20, 2012 2:17 am

Re: USB FIQ testing....

Wed Jul 03, 2013 2:18 am

How about kernel crash on USB audio? I saw that you did commit to the main kernel brunch...Did you change anything or just post alpha version to main kernel?

gsh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 1452
Joined: Sat Sep 10, 2011 11:43 am

Re: USB FIQ testing....

Wed Jul 03, 2013 5:59 am

The only crash I've seen on audio was the one you posted in auddemo, that's not the kernel...

I can only believe that the problem is the application isn't handling it's buffers correctly and the response it is getting from the USB has changed, would need more useful debug information to understand what the problem is...

Gordon
--
Gordon Hollingworth PhD
Raspberry Pi - Director of Software Engineering

ddv2005
Posts: 23
Joined: Fri Jul 20, 2012 2:17 am

Re: USB FIQ testing....

Thu Jul 04, 2013 3:29 pm

gsh wrote:The only crash I've seen on audio was the one you posted in auddemo, that's not the kernel...
What??? This application (absolutely SAME ELF) work fine on other 5 platforms and crashed only on YOUR kernel.

Code: Select all

Message from [email protected] at May 29 20:11:06 ...
 kernel:[  281.335164] Internal error: Oops - BUG: 0 [#1] PREEMPT ARM
This is KERNEL INTERNAL ERROR!!!

P.S. Why you think that you best programmer??? I looked your code and it is...not good...The only difference that you have access to very expensive hardware and has documentation.

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 24656
Joined: Sat Jul 30, 2011 7:41 pm

Re: USB FIQ testing....

Thu Jul 04, 2013 3:41 pm

ddv2005 wrote:
gsh wrote:The only crash I've seen on audio was the one you posted in auddemo, that's not the kernel...
What??? This application (absolutely SAME ELF) work fine on other 5 platforms and crashed only on YOUR kernel.

Code: Select all

Message from [email protected] at May 29 20:11:06 ...
 kernel:[  281.335164] Internal error: Oops - BUG: 0 [#1] PREEMPT ARM
This is KERNEL INTERNAL ERROR!!!

P.S. Why you think that you best programmer??? I looked your code and it is...not good...The only difference that you have access to very expensive hardware and has documentation.
Banned for week. I WILL NOT accept postings like that.
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Contrary to popular belief, humorous signatures are allowed. Here's an example...
“I own the world’s worst thesaurus. Not only is it awful, it’s awful."

gsh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 1452
Joined: Sat Sep 10, 2011 11:43 am

Re: USB FIQ testing....

Thu Jul 04, 2013 4:14 pm

ddv2005 wrote: P.S. Why you think that you best programmer??? I looked your code and it is...not good...The only difference that you have access to very expensive hardware and has documentation.
I never said I was the worlds best programmer, in fact I don't think I am. I'm fairly sure I'd probably have a fair amount of support from the people I've worked with and those that have worked for me to say that I do know what I'm talking about.

Just so you know many of the issues fixed in the USB driver were done by P33M who has never seen the 'official' documentation for the hardware or has any expensive hardware to test it with (he used a GPIO to understand how the USB worked). If he can do it then there's no reason why anyone else can't. The documentation tells you almost nothing more than the software and I almost never refer to it (because quite often it's wrong) plus the analyser I have is very rarely used.

You are clearly a far better programmer than I so I would assume you'll have the problem all fixed and your audio working in the next week.

Thanks

Gordon
--
Gordon Hollingworth PhD
Raspberry Pi - Director of Software Engineering

User avatar
liz
Raspberry Pi Foundation Employee & Forum Moderator
Raspberry Pi Foundation Employee & Forum Moderator
Posts: 5202
Joined: Thu Jul 28, 2011 7:22 pm
Contact: Website

Re: USB FIQ testing....

Thu Jul 04, 2013 10:03 pm

James, you're much too nice. I'm not. ddv2005, I've just issued you a permanent ban - the rule here is that if you wouldn't say it to someone in their living room, you don't say it here. Don't let the door etc. etc.

And anyway, it should be clear to us all that the world's best programmer is whoever wrote Candy Crush Saga.
Director of Communications, Raspberry Pi

GrayShade
Posts: 39
Joined: Fri Jul 05, 2013 6:13 am

Re: USB FIQ testing....

Fri Jul 05, 2013 6:33 am

Hello,

I'm not sure that this is the right place and that it hasn't already been reported, but I'll try.

I have a headless Raspberry Pi with a USB DAC and I'm streaming raw audio over the network. The ALSA client is not very well written, but it worked with standard kernels. I've previously tried the fiq-split branch and I was getting choppy audio with it, so I reverted to the mainline.

The problem is that I've upgraded from #484 to #488 and I'm getting choppy audio again. I've also noticed in dmesg a message about the FIQ fix being enabled, which I think didn't appear with previous versions.

I tried setting dwc_otg.fiq_fix_enable=0 and my Raspberry wouldn't boot anymore. I found this in the log and although it seems that it did start after a while, it wasn't accepting SSH connections.

[code]Jul 5 08:20:32 raspberrypi kernel: [ 2.344817] usb 1-1: new high-speed USB device number 2 using dwc_otg
Jul 5 08:20:32 raspberrypi kernel: [ 2.755886] [<c0013a7c>] (unwind_backtrace+0x0/0xf0) from [<c0079ff4>] (__report_bad_irq+0x20/0xe4)
Jul 5 08:20:32 raspberrypi kernel: [ 2.767704] [<c0079ff4>] (__report_bad_irq+0x20/0xe4) from [<c007a2f0>] (note_interrupt+0x1d0/0x230)
Jul 5 08:20:32 raspberrypi kernel: [ 2.779680] [<c007a2f0>] (note_interrupt+0x1d0/0x230) from [<c00785e0>] (handle_irq_event_percpu+0xac/0x1b0)
Jul 5 08:20:32 raspberrypi kernel: [ 2.792490] [<c00785e0>] (handle_irq_event_percpu+0xac/0x1b0) from [<c0078738>] (handle_irq_event+0x54/0x84)
Jul 5 08:20:32 raspberrypi kernel: [ 2.805453] [<c0078738>] (handle_irq_event+0x54/0x84) from [<c007aad4>] (handle_level_irq+0x98/0x108)
Jul 5 08:20:32 raspberrypi kernel: [ 2.817965] [<c007aad4>] (handle_level_irq+0x98/0x108) from [<c0077e7c>] (generic_handle_irq+0x28/0x40)
Jul 5 08:20:32 raspberrypi kernel: [ 2.830698] [<c0077e7c>] (generic_handle_irq+0x28/0x40) from [<c000e91c>] (handle_IRQ+0x30/0x84)
Jul 5 08:20:32 raspberrypi kernel: [ 2.842904] [<c000e91c>] (handle_IRQ+0x30/0x84) from [<c039abb4>] (__irq_svc+0x34/0xc8)
Jul 5 08:20:32 raspberrypi kernel: [ 2.854395] [<c039abb4>] (__irq_svc+0x34/0xc8) from [<c0024b64>] (msecs_to_jiffies+0x0/0x20)
Jul 5 08:20:32 raspberrypi kernel: [ 2.866475] [<c0024b64>] (msecs_to_jiffies+0x0/0x20) from [<c0275d2c>] (usb_start_wait_urb+0x94/0xbc)
Jul 5 08:20:32 raspberrypi kernel: [ 2.879420] [<c0275d2c>] (usb_start_wait_urb+0x94/0xbc) from [<c0275f08>] (usb_control_msg+0xb4/0xe8)
Jul 5 08:20:32 raspberrypi kernel: [ 2.892404] [<c0275f08>] (usb_control_msg+0xb4/0xe8) from [<c026df00>] (hub_port_init+0x3a0/0x92c)
Jul 5 08:20:32 raspberrypi kernel: [ 2.905110] [<c026df00>] (hub_port_init+0x3a0/0x92c) from [<c0270944>] (hub_thread+0x3d4/0x10d4)
Jul 5 08:20:32 raspberrypi kernel: [ 2.917677] [<c0270944>] (hub_thread+0x3d4/0x10d4) from [<c003a7ec>] (kthread+0x88/0x94)
Jul 5 08:20:32 raspberrypi kernel: [ 2.929548] [<c003a7ec>] (kthread+0x88/0x94) from [<c000e9fc>] (kernel_thread_exit+0x0/0x8)
Jul 5 08:20:32 raspberrypi kernel: [ 3.003716] EXT4-fs (mmcblk0p2): recovery complete
Jul 5 08:20:32 raspberrypi kernel: [ 3.014985] EXT4-fs (mmcblk0p2): mounted filesystem with ordered data mode. Opts: (null)
Jul 5 08:20:32 raspberrypi kernel: [ 3.026343] VFS: Mounted root (ext4 filesystem) on device 179:2.
Jul 5 08:20:32 raspberrypi kernel: [ 3.036861] devtmpfs: mounted
Jul 5 08:20:32 raspberrypi kernel: [ 3.041880] Freeing init memory: 128K
Jul 5 08:20:32 raspberrypi kernel: [ 17.964870] Indeed it is in host mode hprt0 = 00001101
Jul 5 08:20:32 raspberrypi kernel: [ 33.254814] Indeed it is in host mode hprt0 = 00001101
Jul 5 08:20:32 raspberrypi kernel: [ 33.544825] Indeed it is in host mode hprt0 = 00001101
Jul 5 08:20:32 raspberrypi kernel: [ 33.724792] usb 1-1: new high-speed USB device number 3 using dwc_otg
Jul 5 08:20:32 raspberrypi kernel: [ 48.724802] Indeed it is in host mode hprt0 = 00001101
Jul 5 08:20:32 raspberrypi kernel: [ 64.014808] Indeed it is in host mode hprt0 = 00001101
Jul 5 08:20:32 raspberrypi kernel: [ 64.304824] Indeed it is in host mode hprt0 = 00001101
Jul 5 08:20:32 raspberrypi kernel: [ 64.484784] usb 1-1: new high-speed USB device number 4 using dwc_otg
Jul 5 08:20:32 raspberrypi kernel: [ 74.912415] Indeed it is in host mode hprt0 = 00001101
Jul 5 08:20:32 raspberrypi kernel: [ 75.094789] usb 1-1: new high-speed USB device number 5 using dwc_otg
Jul 5 08:20:32 raspberrypi kernel: [ 87.622986] EXT4-fs (mmcblk0p2): re-mounted. Opts: (null)
Jul 5 08:20:32 raspberrypi kernel: [ 88.067052] EXT4-fs (mmcblk0p2): re-mounted. Opts: (null)
Jul 5 08:20:34 raspberrypi kernel: [ 99.855271] Adding 102396k swap on /var/swap. Priority:-1 extents:2 across:507900k SS
[/code]

I don't have a display to investigate this further, but you might want to know about it.

PS: How do I get a specific version with rpi-update? I tried overwriting the files from /boot with the ones in /boot.bak, but I got an older version than my previous one.

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 24656
Joined: Sat Jul 30, 2011 7:41 pm

Re: USB FIQ testing....

Fri Jul 05, 2013 8:14 am

liz wrote:James, you're much too nice. I'm not. ddv2005, I've just issued you a permanent ban - the rule here is that if you wouldn't say it to someone in their living room, you don't say it here. Don't let the door etc. etc.

And anyway, it should be clear to us all that the world's best programmer is whoever wrote Candy Crush Saga.
Er, Bubble Bobble, with second place going to Mr Do!

James 'Mr Nice' H
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Contrary to popular belief, humorous signatures are allowed. Here's an example...
“I own the world’s worst thesaurus. Not only is it awful, it’s awful."

gsh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 1452
Joined: Sat Sep 10, 2011 11:43 am

Re: USB FIQ testing....

Fri Jul 05, 2013 10:13 am

GrayShade wrote:Hello,

PS: How do I get a specific version with rpi-update? I tried overwriting the files from /boot with the ones in /boot.bak, but I got an older version than my previous one.
I assume you're using BRANCH=next already? We've integrated the new fiq_split fixes into the next branch (3.9) which may be causing the problem... try fiq_split_enable=0 on the command line if you're using that...

Also dmesg | grep fiq

Would be useful output

Gordon
--
Gordon Hollingworth PhD
Raspberry Pi - Director of Software Engineering

GrayShade
Posts: 39
Joined: Fri Jul 05, 2013 6:13 am

Re: USB FIQ testing....

Fri Jul 05, 2013 10:21 am

No, I didn't set the BRANCH. Could it be related to this https://github.com/raspberrypi/firmware ... 79305c63db ?

I looked into dmesg, but there isn't anything else besides "FIQ enabled" and "FIQ split fix enabled". I have yet to try fiq_split_enable=0.

gsh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 1452
Joined: Sat Sep 10, 2011 11:43 am

Re: USB FIQ testing....

Fri Jul 05, 2013 10:34 am

Ah OK,

I thought that was only checked into the next branch...

Try fiq_split_enable=0 and see if that fixes your problem (or causes more issues) I was hoping this will help more people than it hinders!

Didn't see your report of choppy audio in this forum... That's why it's now made its way to the head branch without being looked into... How are you playing the audio, I've only got a Daffodil device here, what device is it you are using?

Need reproduction information

Gordon
--
Gordon Hollingworth PhD
Raspberry Pi - Director of Software Engineering

GrayShade
Posts: 39
Joined: Fri Jul 05, 2013 6:13 am

Re: USB FIQ testing....

Fri Jul 05, 2013 11:00 am

It's a Behringer UCA-202 (http://www.behringer.com/EN/Products/UCA202.aspx), but it's rather finicky: if I plug it in after my computer has started, it sometimes crashes -- both on Windows (no BSOD, just a hang) and the Raspberry (no idea what happens).

Code: Select all

[    0.940772] usbcore: registered new interface driver usbfs
[    0.946360] usbcore: registered new interface driver hub
[    0.951956] usbcore: registered new device driver usb
[    1.349796] usbcore: registered new interface driver smsc95xx
[    1.826248] dwc_otg bcm2708_usb: DWC OTG Controller
[    1.832978] dwc_otg bcm2708_usb: new USB bus registered, assigned bus number 1
[    1.842008] dwc_otg bcm2708_usb: irq 32, io mem 0x00000000
[    1.860005] usb usb1: New USB device found, idVendor=1d6b, idProduct=0002
[    1.868665] usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1
[    1.877749] usb usb1: Product: DWC OTG Controller
[    1.884315] usb usb1: Manufacturer: Linux 3.6.11+ dwc_otg_hcd
[    1.891883] usb usb1: SerialNumber: bcm2708_usb
[    1.898980] hub 1-0:1.0: USB hub found
[    1.910812] Initializing USB Mass Storage driver...
[    1.917706] usbcore: registered new interface driver usb-storage
[    1.925508] USB Mass Storage support registered.
[    1.931989] usbcore: registered new interface driver libusual
[    2.076926] usbcore: registered new interface driver usbhid
[    2.084382] usbhid: USB HID core driver
[    2.362476] usb 1-1: new high-speed USB device number 2 using dwc_otg
[    2.592762] usb 1-1: New USB device found, idVendor=0424, idProduct=9512
[    2.602208] usb 1-1: New USB device strings: Mfr=0, Product=0, SerialNumber=0
[    2.612940] hub 1-1:1.0: USB hub found
[    2.902701] usb 1-1.1: new high-speed USB device number 3 using dwc_otg
[    3.033086] usb 1-1.1: New USB device found, idVendor=0424, idProduct=ec00
[    3.043028] usb 1-1.1: New USB device strings: Mfr=0, Product=0, SerialNumber=0
[    3.127553] smsc95xx 1-1.1:1.0: eth0: register 'smsc95xx' at usb-bcm2708_usb-1.1, smsc95xx USB 2.0 Ethernet, b8:27:eb:92:3a:8e
[    3.252772] usb 1-1.3: new full-speed USB device number 4 using dwc_otg
[    3.369290] usb 1-1.3: New USB device found, idVendor=08bb, idProduct=2902
[    3.389855] usb 1-1.3: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[    3.409244] usb 1-1.3: Product: USB Audio CODEC
[    3.421885] usb 1-1.3: Manufacturer: Burr-Brown from TI
[    3.446675] input: Burr-Brown from TI               USB Audio CODEC  as /devices/platform/bcm2708_usb/usb1/1-1/1-1.3/1-1.3:1.3/input/input0
[    3.483091] hid-generic 0003:08BB:2902.0001: input,hidraw0: USB HID v1.00 Device [Burr-Brown from TI               USB Audio CODEC ] on usb-bcm2708_usb-1.3/input3
[    6.796567] usbcore: registered new interface driver snd-usb-audio
I had to tweak my asound.conf file to change the sampling rate from 48000 Hz to 44100 Hz, since it was horribly choppy with the defaults:

Code: Select all

pcm.!default {
    type plug
    slave {
        pcm {
            type hw
            card 0
        }
        rate 44100
    }
}
I didn't write a report because I didn't really get to read about the patch, its purpose and problems people encountered. Actually, I'm not even sure what it does besides fixing some keyboard problems :).

I'll try to do a few more tests: IIRC, the fiq-split branch had that problem even when using aplay, but yesterday aplay seemed to be working fine, while my application wasn't. I know that the network card is connected to the USB controller, and since I'm streaming a lot of data it might have an effect.

gsh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 1452
Joined: Sat Sep 10, 2011 11:43 am

Re: USB FIQ testing....

Fri Jul 05, 2013 11:05 am

OK,

I'll order one... What's the CPU usage like? I assume you're not hitting significant usage?

Gordon
--
Gordon Hollingworth PhD
Raspberry Pi - Director of Software Engineering

GrayShade
Posts: 39
Joined: Fri Jul 05, 2013 6:13 am

Re: USB FIQ testing....

Fri Jul 05, 2013 11:12 am

My app uses around 5%, which is more than I would like; I didn't look too much at the system CPU usage, but it wasn't exceptionally high.

GrayShade
Posts: 39
Joined: Fri Jul 05, 2013 6:13 am

Re: USB FIQ testing....

Sat Jul 06, 2013 10:38 am

Ok, I think I found a way to reproduce this on #494. On the Raspberry:

Code: Select all

aplay -f cd ss.raw
nc -kl 0.0.0.0 7777 > /dev/null
From another computer, send a large file to the listening netcat instance. The audio will be choppy while data is being transferred. If I set fiq_split_enable=0, the audio plays fine.

I also had a hang or oops during boot with the FIQ split fix enabled, but I don't see anything in the logs: it looks like it booted properly, but network didn't work.

Thanks for looking into this :). One more thing: I was getting lots of errors like "delay: estimated 0, actual 177" and I set nrpacks=1 for snd-usb-audio.

Return to “Troubleshooting”