What about the shipping costs? Who pays for each direction?
Rob
What about shipping costs? I guess I'm paying to ship it to you, but
what about the return direction?
Regards
Rob
We always pay for the return shipping.
Gerald
We will pay for the return shipment.
Gerald
I dropped my board in the mail last night - I would expect it to
arrive there in a couple of days.
I enclosed a summary of the tests I've run and a reference to this
board.
Sounds good!
Thnak you!
Gerald
Because of the USB HOST problems I’ve switched to the USB OTG port, with various success;
- USB OTG → HUB → HD + RT73 based WIFI dongle
- DD 10GB from HD to /dev/null works as expected (USB HOST → EMI crash after 100MB)
- FTP from WIFI to HD → after few 100MB I get the error bellow:
beagleboard login: usb 2-1.1: reset high speed USB device using musb_hdrc and ad
dress 3
------------[ cut here ]------------
WARNING: at drivers/usb/musb/musb_host.c:128 musb_h_tx_flush_fifo+0x94/0xd4()
Could not flush host TX3 fifo: csr: 0003
Modules linked in: ipv6 rt73
[] (dump_stack+0x0/0x14) from [] (warn_slowpath+0x5c/0x78)
[] (warn_slowpath+0x0/0x78) from [] (musb_h_tx_flush_fifo+0x
94/0xd4)
r3:00000003 r2:c0484cfb
r6:ffffffff r5:00000003 r4:00000003
[] (musb_h_tx_flush_fifo+0x0/0xd4) from [] (musb_cleanup_urb
+0xd4/0x120)
[] (musb_cleanup_urb+0x0/0x120) from [] (musb_urb_dequeue+0x
140/0x174)
[] (musb_urb_dequeue+0x0/0x174) from [] (unlink1+0xb8/0xc4)
[] (unlink1+0x0/0xc4) from [] (usb_hcd_unlink_urb+0x58/0x74)
r8:cf89dfa0 r7:ffffff98 r6:cf9a2560 r5:a0000093 r4:00000000
[] (usb_hcd_unlink_urb+0x0/0x74) from [] (usb_unlink_urb+0x4
0/0x44)
r7:cf89df94 r6:00000500 r5:cf99a000 r4:cf99a35c
[] (usb_unlink_urb+0x0/0x44) from [] (usb_stor_stop_transpor
t+0x3c/0x68)
[] (usb_stor_stop_transport+0x0/0x68) from [] (command_abort
+0x7c/0x90)
r4:cf99a35c
[] (command_abort+0x0/0x90) from [] (scsi_send_eh_cmnd+0x128
/0x21c)
r4:40000013
[] (scsi_send_eh_cmnd+0x0/0x21c) from [] (scsi_eh_tur+0x2c/0
x88)
[] (scsi_eh_tur+0x0/0x88) from [] (scsi_error_handler+0x184/
0x364)
r5:cfab48ac r4:cfab48a0
[] (scsi_error_handler+0x0/0x364) from [] (kthread+0x5c/0x94
)
[] (kthread+0x0/0x94) from [] (do_exit+0x0/0x6a8)
r6:00000000 r5:00000000 r4:00000000
—[ end trace fc1b8fac7574014e ]—
usb 2-1.1: reset high speed USB device using musb_hdrc and address 3
And again:
root@beagleboard:~# usb 2-1.1: reset high speed USB device using musb_hdrc and a
ddress 3
------------[ cut here ]------------
WARNING: at drivers/usb/musb/musb_host.c:128 musb_h_tx_flush_fifo+0x94/0xd4()
Could not flush host TX3 fifo: csr: 0003
Modules linked in: ipv6 rt73
[] (dump_stack+0x0/0x14) from [] (warn_slowpath+0x5c/0x78)
[] (warn_slowpath+0x0/0x78) from [] (musb_h_tx_flush_fifo+0x
94/0xd4)
r3:00000003 r2:c0484cfb
r6:ffffffff r5:00000003 r4:00000003
[] (musb_h_tx_flush_fifo+0x0/0xd4) from [] (musb_cleanup_urb
+0xd4/0x120)
[] (musb_cleanup_urb+0x0/0x120) from [] (musb_urb_dequeue+0x
140/0x174)
[] (musb_urb_dequeue+0x0/0x174) from [] (unlink1+0xb8/0xc4)
[] (unlink1+0x0/0xc4) from [] (usb_hcd_unlink_urb+0x58/0x74)
r8:cf949fa0 r7:ffffff98 r6:cf89c560 r5:a0000093 r4:00000000
[] (usb_hcd_unlink_urb+0x0/0x74) from [] (usb_unlink_urb+0x4
0/0x44)
r7:cf949f94 r6:00000500 r5:cf99c000 r4:cf99c35c
[] (usb_unlink_urb+0x0/0x44) from [] (usb_stor_stop_transpor
t+0x3c/0x68)
[] (usb_stor_stop_transport+0x0/0x68) from [] (command_abort
+0x7c/0x90)
r4:cf99c35c
[] (command_abort+0x0/0x90) from [] (scsi_send_eh_cmnd+0x128
/0x21c)
r4:40000013
[] (scsi_send_eh_cmnd+0x0/0x21c) from [] (scsi_eh_tur+0x2c/0
x88)
[] (scsi_eh_tur+0x0/0x88) from [] (scsi_error_handler+0x184/
0x364)
r5:cfab48ac r4:cfab48a0
[] (scsi_error_handler+0x0/0x364) from [] (kthread+0x5c/0x94
)
[] (kthread+0x0/0x94) from [] (do_exit+0x0/0x6a8)
r6:00000000 r5:00000000 r4:00000000
—[ end trace ace7fd02eed9425e ]—
usb 2-1.1: reset high speed USB device using musb_hdrc and address 3
usb 2-1.1: reset high speed USB device using musb_hdrc and address 3
usb 2-1.1: reset high speed USB device using musb_hdrc and address 3
Does anybody know what’s happening here?
Wkr,
Joep
2009/6/10 Gerald Coley <gerald@beagleboard.org>
Well, my board is now in the hospital, so hopefully the Doctors there
will be able to pop some Vicodin, make some sexist comments, and come
up with a diagnosis as to what the problem is. I hope my reports on
how to reproduce the issue will help.
Yes, the head surgeon has it on his desk and yes it does fail. It definitely has some sort of rash. We are thinking it may be allergic to Linux.
Gerald
I figured out that mine has problems with the d-link dwa-110. The USB
seems only to die when the corresponding module is loaded. This does
not depend on data transfer, the USB can even die when the wifi is not
associated to an AP thus no i/o happening.
HTH
Marcus
Good - in that the problem is reproducible and therefor (hopefully)
identifiable and fixable.
Are you seeing the problem using the tests you'd normally use, or only
using the tests I've recommended (adding one or more interrupt-
endpoint devices)?
I'm really very curious because of your past comments that you've
previously not been able to reproduce the problems, and I'm hoping
that my board is just a very good exemplar of the problem, and not
some other problem unrelated to what others have seen.
Also: I've heard rumblings that the suspicion is on the USB host core
TI licensed, and given that I am looking at a production design where
that core would be useful I'd like to know if that really is the case
or not.
Our normal tests pass, but we have a new stress test that seems to cause the issue to show up on a few boards, including yours. So, we are looking to find the issue as we speak. We do not know if it is HW or SW related, but it does not show up on all boards.
Gerald
Can you share this stress test?
I'm seeing issues with mine and would like to verify myself if I have
the problem or not.
Frans
dd if=/dev/sda of=/dev/null bs=1M
You may have to run it 2-3 times before it fails.
Now, just because you have a board that fails the test, I am not in a position to have a bunch of RMAs coming in to get them “fixed”. Again, we do not know at this point what exactly the issue is and why it just happens on certain boards.
Gerald
Forgive me for asking, but I've found that sometimes it's the
questions you didn't ask that get you in trouble:
Did you run the stress test against the replacement board you sent out
to me? Did it pass?
I am not sure if I should answer that, considering the fact that you felt that you had to ask it.
Gerald
BTW, the board you sent back was one of the first Rev C2 boards that went out. It seems to me to have a lot of problems besides just EHCI. It is very unstable.
Gerald
dd if=/dev/sda of=/dev/null bs=1M
You may have to run it 2-3 times before it fails.
My board fails within 4-5 seconds and the LCD starts immediately to
flicker then goes off.
The last messages on the serial console are:
omap-dss DISPC error: dispc irq error status
00e6 omap-dss DISPC error: dispc irq error status
0040 omap-dss DISPC error: dispc irq error status
00c0 omap-dss DISPC error: dispc irq error status
00e2 omap-dss DISPC error: dispc irq error status
0040 omap-dss DISPC error: dispc irq error status
00c0 omap-dss DISPC error: dispc irq error status
00e2 omap-dss DISPC error: dispc irq error status
00e2 omap-dss DISPC error: dispc irq error status
00e2 omap-dss DISPC error: dispc irq error status
0040 omap-dss DISPC error: Excessive DISPC
errors Turning off lcd and
digit omap-dss DISPC error: Excessive DISPC
errors Turning off lcd and digit
Marcus
Dear Gerald
We saw this post right now after a night of work to understand what we
were doing wrong 'cause the USB was experiencing strange "USB
disconnect" errors on heavy net traffic (we are using an USB-LAN
adapter).
Is it possible to know if the issue has been identified and a
workaround is possible?
Kind regards,
Dario
We saw this post right now after a night of work to understand what we
were doing wrong 'cause the USB was experiencing strange "USB
disconnect" errors on heavy net traffic (we are using an USB-LAN
adapter).
I have 3 REVC boards and im using an USB-LAN adapter (asix based), my
solution to this was to put the beagleboard inside an antistatic
bag... looks stupid i know but it works....
BTW have you tried to compile something in the board? all of mine
boards gives me segfault and internal errors in the gcc.. i suspect
RAM issues....