Hi,
this fixes the completion handler for the interrupt urb in hpusbscsi.
Please apply.
Regards
Oliver
You can import this changeset into BK by piping this whole message to:
'| bk receive [path to repository]' or apply the patch as usual.
Hi,
I just want to add some experience I did in testing. I am not so
deep inside for code analysis at the moment.
1.) pwc on 2.6.0-test1-ac1: It does not stop when loading the system, but
I had a stop after a number of hours (3). The received counter at
closing the device seems not to
Am Mittwoch, 16. Juli 2003 09:59 schrieb Detlef Schmicker \(Siebertz Electronic GmbH\):
Hi,
I just want to add some experience I did in testing. I am not so
deep inside for code analysis at the moment.
1.) pwc on 2.6.0-test1-ac1: It does not stop when loading the system, but
I had a
Hi,
usblcd registers a device before all buffers are allocated leading
to a race resulting in NULL pointers being followed.
This fixes it.
Regards
Oliver
You can import this changeset into BK by piping this whole message to:
'| bk receive [path to repository]' or apply
Oliver Neukum wrote:
Am Mittwoch, 16. Juli 2003 09:59 schrieb Detlef Schmicker \(Siebertz Electronic
GmbH\):
Hi,
I just want to add some experience I did in testing. I am not so
deep inside for code analysis at the moment.
1.) pwc on 2.6.0-test1-ac1: It does not stop when loading
On Tue, 15 Jul 2003, Harald Dunkel wrote:
Here is some more syslog with debugging output enabled. Again,
I tried to format /dev/sdc1 using mkreiserfs after removing
the old partition table and creating a maxsize partition. Since
this problem affects file I/O, it is possible that the
syslog
Hi All,
In the UHCI or OHCI host controller driver, it builds a tree structure for
periodic requests. Is there any mechanism in EHCI host controller driver for HS
periodic requests and where are the codes ??
Thanks a lot.
+wzf¢+,¦ì¢·o%LYªÞZ+aTÌ쨹Ƨ®éæº[b¦W¨¥êÚ¶)à³+-zk(«
Hi,
registering a device only partially initialised is quite bad an
idea.
Regards
Oliver
You can import this changeset into BK by piping this whole message to:
'| bk receive [path to repository]' or apply the patch as usual.
On Tue, 15 Jul 2003, Charles Lepple wrote:
Before I round up the ksymoops log (it requires setting up a serial
console), should I try testing with a later version of usbtest? Is this a
known issue?
The log was easier to get than I expected. It's attached.
On second glance, it looks
Alan Stern said:
This is a problem in uhci-hcd. A struct urbp (private data structure) is
being accessed after it has been freed. More specifically, the list of
unlinked URBs (uhci-urb_remove_list) has been corrupted: it points to a
deallocated urbp. It would be nice to know how this could
On Tue, Jul 15, 2003 at 07:28:56PM -0700, Matthew Dharm wrote:
Greg, please apply this patch from Alan to 2.5
Applied, thanks.
greg k-h
---
This SF.net email is sponsored by: VM Ware
With VMware you can run multiple operating systems on a
On Wed, Jul 16, 2003 at 01:00:39PM +0200, Oliver Neukum wrote:
Hi,
usblcd registers a device before all buffers are allocated leading
to a race resulting in NULL pointers being followed.
This fixes it.
Applied, thanks.
greg k-h
---
This
On Wed, Jul 16, 2003 at 05:43:23PM +0200, Oliver Neukum wrote:
Hi,
registering a device only partially initialised is quite bad an
idea.
Applied, thanks.
greg k-h
---
This SF.net email is sponsored by: VM Ware
With VMware you can run
On Wed, Jul 16, 2003 at 08:17:56AM +0200, Oliver Neukum wrote:
Hi,
this fixes the completion handler for the interrupt urb in hpusbscsi.
Please apply.
Applied, thanks,
greg k-h
---
This SF.net email is sponsored by: VM Ware
With VMware
Alan Stern said:
If you haven't done that, try running the test again with USB debugging
on. Or if you already have, look through the system log for messages from
the uhci driver.
Attached is the log with debugging enabled (first with firmware loaded,
then after rebooting, without firmware).
Hi
When connecting the usb storage device to USB 2 VIA host controller
I get these messages repeated (with address incremented by one)
indefinitely:
hub.c: new USB device 00:12.2-1, assigned address 2
usb.c: USB device not accepting new address=2 (error=-71)
hub.c: new USB device 00:12.2-1,
On Wed, 16 Jul 2003, Charles Lepple wrote:
Alan Stern said:
This is a problem in uhci-hcd. A struct urbp (private data structure) is
being accessed after it has been freed. More specifically, the list of
unlinked URBs (uhci-urb_remove_list) has been corrupted: it points to a
Charles Lepple wrote:
I was mistaken before when I said that usbtest would respond to signals
when the firmware was loaded; the only way I could see to terminate it was
to unplug the USB device.
Yes, that's a minor annoyance. Or sometimes not-so-minor, when
khubd can't un-block. It'd be a Good
On Wed, 16 Jul 2003, Charles Lepple wrote:
Attached is the log with debugging enabled (first with firmware loaded,
then after rebooting, without firmware). usbtest is currently wedged in
test_ctrl_queue (according to /proc/pid/wchan).
The debugging information isn't much help. All it shows
Hi,
the driver is registering the device too early, so that open may
see a partially initialised device.
Regards
Oliver
You can import this changeset into BK by piping this whole message to:
'| bk receive [path to repository]' or apply the patch as usual.
Bernd Porr wrote:
The transfer mode is isochronous.
Which is probably part of the reason for the crash; that
code has hardly been used, and there are no tests to make
sure that it works right either..
Related to this problem: how often is the completion routine called?
Every microframe (125us)
Alan Stern said:
It might be that the problem is triggered by a particular pattern of URBs
being unlinked and completed. The exact timing may matter as well.
Anyway, this bug shouldn't occur no matter what hardware or firmware you
use.
Definitely timing related-- the extra printks seem to
Am Mittwoch, 16. Juli 2003 22:07 schrieb Alan Stern:
Oliver:
Along with all these races between probe and open that you've been fixing
recently, have you considered the race between disconnect and open? I
added code to handle that in usb-skeleton.c a while ago, but I didn't look
at any
On Wed, 16 Jul 2003, Oliver Neukum wrote:
Also, races with disconnect are not confined to open.
Really? I thought that open was special because it was the only place
that the usage count could be increased.
Then there are a couple of drivers who can be killed by ENOMEM in the
wrong moment.
Am Mittwoch, 16. Juli 2003 23:21 schrieb Alan Stern:
On Wed, 16 Jul 2003, Oliver Neukum wrote:
Also, races with disconnect are not confined to open.
Really? I thought that open was special because it was the only place
that the usage count could be increased.
Right. Therefore only open
Hi,
if probe fails, allocated URBs and buffers must be freed.
Regards
Oliver
You can import this changeset into BK by piping this whole message to:
'| bk receive [path to repository]' or apply the patch as usual.
Hello,
On Tuesday 15 July 2003 10:43, Oliver Neukum wrote:
Particularly after taking a quick glance at the code
and not seeing anything that much cared about nonzero
fault codes when the urbs are returned.
Hi nemosoft,
looking at the 2.5 code error handling seems fishy:
if
On Thu, Jul 17, 2003 at 01:05:53AM +0200, Oliver Neukum wrote:
Hi,
if probe fails, allocated URBs and buffers must be freed.
Regards
Oliver
Nice catch. However, you forgot about self-speed_urb.
Thanks, I'll push a fixed version to Jeff with my next
28 matches
Mail list logo
