--- Begin Message ---
- To: "Jonathan Nieder" <jrnieder@gmail.com>
- Subject: Re: general protection fault in a ttyACM0 access
- From: "Albert van der Horst" <spenarnc@xs4all.nl>
- Date: Thu, 29 Nov 2012 16:36:04 +0100
- Message-id: <63c5e744ee30be765788dcef7038c590.squirrel@webmail.xs4all.nl>
- In-reply-to: <20121124221239.GA23157@elie.Belkin>
- References: <20121018142105.3885.6395.reportbug@cherry.spenarnc.xs4all.nl> <20121124221239.GA23157@elie.Belkin>
> severity 690848 important
> # guessing
> found 690848 linux-2.6/2.6.32-46
> quit
>
> Hi Albert,
>
> Albert van der Horst wrote:
>
>> A Texas Instruments Launcpad is connected as the sole
>> USB device, resulting in device /dev/ttyACM0 showing up.
>> No TI drivers were installed, or any specific usb-drivers.
>>
>> A simple program in Forth does some direct read and write on ttyACM0,
>> while the device is responding. It functions properly, except for
>> situation where an output stream of the device is interrupted from the
>> terminal.
> [...]
>> kernel:[15594.923330] general protection fault: 0000 [#1] SMP
> [...]
>> kernel:[15594.923490] Code: 85 c0 75 15 65 48 8b 04 25 c8 cb 00 00 48
>> 2d d8 1f 00 00 48
>> 89 43 18 31 c0 5b c3 53 48 89 fb 48 83 ec 10 e8 31 f8 ff ff 48 89 df
>> <f0> ff 0f 79 05 e8
>> 48 ff ff ff 65 48 8b 04 25 c8 cb 00 00 48 2d
>
> Thanks for reporting it. Is this reproducible? If so, please attach
> full "dmesg" output from booting and reproducing the bug, since that
> should give us a fuller trace.
I have experienced it one more time, running an old version of my program.
[If I run a now debugged and functional version, things run smoothly.]
I will expend effort to try to make a reproducable report, maybe including a
copy of the program. This may take a while, however.
Please note that this program exercises the USB driver using system calls
from a statically linked binary, and the device may send while it is being
read from. This may provide a clue.
>
> Does a 3.2.y or newer kernel from wheezy, sid, or squeeze-backports
> reproduce the same behavior? The only packages from outside squeeze
> that would be needed for this test are the kernel image itself,
> linux-base, and initramfs-tools.
This is a bit over my head. Could you identify an Ubuntu version where you
would have like it have tested? I can probably accomodate that.
I use Debian stable for serious work. But I have a recent Ubuntu system for
You Tube nonsense and experimenting.
>
> Sorry for the slow reply and hope that helps,
You're welcome.
> Jonathan
>
--- End Message ---