Date: Fri, 03 Oct 2003 21:13:50 +1000 From: Stephen McKay <smckay@internode.on.net> To: Josef Karthauser <joe@tao.org.uk> Cc: Stephen McKay <smckay@internode.on.net> Subject: Re: 4.9-rc + usb + asus a7n8x == bad news Message-ID: <200310031113.h93BDoJW013160@dungeon.home> In-Reply-To: <20031003101411.GA19458@genius.tao.org.uk> from Josef Karthauser at "Fri, 03 Oct 2003 11:14:11 %2B0100" References: <200310030944.h939isJW010930@dungeon.home> <20031003101411.GA19458@genius.tao.org.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
On Friday, 3rd October 2003, Josef Karthauser wrote:
>The usb stack in 4.9 is the same as that in 4.7 and so I'd expect no
>improvements.
And I thought I'd waited long enough before buying a USB device. :-)
>Please check the -stable mail list for a post by me a
>month or so ago with a patch that brings the -current USB stack into
>stable. I would expect that to be much better. We missed getting it
>into 4.9, but it will be committed to RELENG_4 as soon as the 4.9 tag
>have been laid down.
I tried this just now. The result is different, but not really any better.
On my first attempt to plug in my USB memory stick, I got:
umass0: IMGSYS-3 Security X-Disk V1.02, rev 1.10/2.01, addr 2
umass0: at uhub0 port 3 (addr 2) disconnected
umass0: detached
umass0: IMGSYS-3 Security X-Disk V1.02, rev 1.10/2.01, addr 2
panic: Removing other than first element
Debugger("panic")
Stopped at Debugger+0x34: movb $0,in_Debugger.429
db>
This was just a single insertion. It looks like there's some sort of
"bounce" involved (ie insert, remove, insert detected even though it's
only one insertion).
DDB seems to work, and just listing the call stack, I get:
Debugger
panic
usb_transfer_complete
ohci_abort_xfer
ohci_timeout_task
usb_task_thread(0)
fork_trampoline()
No, I don't currently have a serial console set up. Then, when I try to
get a core dump to check this further I get (funny, ha, ha) this message:
dump failed, reason: device doesn't support a dump routine
I remember the good old days when core dumps worked because everybody
needed them. :-)
Stephen.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200310031113.h93BDoJW013160>