On Tue, Nov 01, 2005 at 01:04:02PM +, David Woodhouse wrote:
> On Tue, 2005-11-01 at 13:40 +0100, Duncan Sands wrote:
> > this code looks like a 'orrible hack to work around a common problem
> > with USB modem's of this type: if the modem is plugged in while the
> > system boots, the driver may
On Thu, May 12, 2005 at 06:56:34PM +0200, Oliver Neukum wrote:
> static inline int usbatm_submit_urb(struct urb *urb)
>
> You'll enter this with irqs off and will leave it with irqs on.
Mind elaborating a bit? I can see that we may enable irqs with
spin_unlock_irq() on the error path, but how do
On Wed, May 04, 2005 at 01:37:30PM -0700, David Brownell wrote:
> On Wednesday 04 May 2005 12:19 pm, Roman Kagan wrote:
> > struct urb {
> > /* private, usb core and host controller only fields in the urb */
> > ...
> > struct list_head urb_list; /* list
On Wed, May 04, 2005 at 11:19:15AM -0400, Alan Stern wrote:
> Here's an idea I had recently to improve the klist library; tell me what
> you think. As a space optimization, instead of storing a struct
> completion in every klist_node, just put a wait_queue_header in struct
> klist. Under normal
Hi,
As the recommended practice in 2.6 is to keep the endpoint queue busy,
drivers (should) normally want to use multiple urbs on an endpoint.
Given that, it would be very handy to have a list head in struct urb, so
that drivers could conveniently manipulate multiple urbs.
(As a real world exam
On Wed, May 04, 2005 at 10:30:57AM +0400, Roman Kagan wrote:
> On Fri, Apr 29, 2005 at 02:34:59PM -0400, Alan Stern wrote:
> > Does the patch look good? If you like it, I'll submit it to Greg.
>
> Yes it looks reasonable. At the moment I can't test the problematic
>
On Fri, Apr 29, 2005 at 12:28:30PM -0700, David Brownell wrote:
> On Friday 29 April 2005 11:23 am, Roman Kagan wrote:
> >
> > ... instead of trying to make sure the attributes are available via
> > sysfs at hotplug time, we can use another means to pass them to hotplug:
>
On Fri, Apr 29, 2005 at 02:34:59PM -0400, Alan Stern wrote:
> Yes, we do. Okay, are you still concerned about the new code being
> asymmetric somehow?
No :)
> Does the patch look good? If you like it, I'll submit it to Greg.
Yes it looks reasonable. At the moment I can't test the problematic
On Wed, Apr 27, 2005 at 04:14:08PM -0700, Patrick Mansfield wrote:
> On Wed, Apr 27, 2005 at 05:21:10PM -0400, Alan Stern wrote:
>
> > David's right. Why did kobject_hotplug() move out of kobject_add() and
> > into its callers sometime after 2.6.11? In particular the invocation in
> > device_ad
On Thu, Apr 28, 2005 at 02:36:39PM -0400, Alan Stern wrote:
> Here is a fix for driver_detach(). It's a little ugly because it avoids
> using the klist iterator, but there's no way around it -- the iterator
> simply can't be made to work here. Not only does it prevent
> device_release_driver() fr
On Tue, Apr 26, 2005 at 11:57:34AM -0400, Alan Stern wrote:
> On Tue, 26 Apr 2005, Roman Kagan wrote:
>
> > On Mon, Apr 25, 2005 at 04:01:07PM -0400, Alan Stern wrote:
> > > I think usb_driver_claim_interface is correct as it stands. It was a
> >
On Mon, Apr 25, 2005 at 12:33:22PM -0400, Alan Stern wrote:
> On Mon, 25 Apr 2005, Roman Kagan wrote:
> > http://marc.theaimsgroup.com/?l=linux-usb-devel&m=111341625900343
>
> I see. This may not be a problem so much with klists as with usbcore.
> That is, device_release_
On Mon, Apr 25, 2005 at 04:01:07PM -0400, Alan Stern wrote:
> I think usb_driver_claim_interface is correct as it stands. It was a
> mistake to leave out from usb_driver_release_interface originally the line
> setting iface->condition to USB_INTERFACE_UNBINDING.
But it would be asymmetric then.
On Mon, Apr 25, 2005 at 02:39:55PM -0400, Alan Stern wrote:
> On Mon, 25 Apr 2005, Roman Kagan wrote:
> > Well for non-klist version the test !list_empty(&dev->driver_list) in
> > usb_driver_release_interface() was a good enough guard against recursion
> > into device_
On Mon, Apr 25, 2005 at 10:52:20AM -0400, Alan Stern wrote:
> On Mon, 25 Apr 2005, Roman Kagan wrote:
> > This patch fixes a real and easy to trigger problem, so IMHO it may be
> > worth applying before 2.6.12-final is out...
>
> You might also consider submitting this for
On Wed, Apr 13, 2005 at 09:40:17PM +0400, Roman Kagan wrote:
> With 2.6.11 and 2.6.12-rc2 (and perhaps a few versions before) usb
> drivers for multi-interface devices, which do
> usb_driver_release_interface() in their disconnect(), make rmmod hang.
>
> It turns out to be due to a
The patch below adjusts the MODALIAS generated by the usb hotplug
function to match the proposed change to scripts/mod/file2alias.c.
Please consider applying.
Roman.
Signed-off-by: Roman Kagan <[EMAIL PROTECTED]>
drivers/usb/core/usb.c |6 ++
1 files changed, 2 insertions
m3 with an offset. The
matching patch to fix the MODALIAS environment variable now generated by
the usb hotplug function follows.
Please consider applying.
Roman.
Signed-off-by: Roman Kagan <[EMAIL PROTECTED]>
file2alias.c | 111 ++-
On Mon, Apr 18, 2005 at 08:55:24AM -0700, Greg KH wrote:
> On Mon, Apr 18, 2005 at 07:40:25PM +0400, Roman Kagan wrote:
> > On Mon, Apr 18, 2005 at 08:02:25AM -0700, Greg KH wrote:
> > > And your patch modified the map file which is created, right?
> >
> > By n
On Mon, Apr 18, 2005 at 08:02:25AM -0700, Greg KH wrote:
> On Mon, Apr 18, 2005 at 08:41:58AM +0400, Roman Kagan wrote:
> > On Sun, Apr 17, 2005 at 04:19:15PM -0700, Greg KH wrote:
> > > On Sat, Apr 09, 2005 at 09:55:21AM +0400, Roman Kagan wrote:
> > > > On Fri, Ap
On Sun, Apr 17, 2005 at 04:19:15PM -0700, Greg KH wrote:
> On Sat, Apr 09, 2005 at 09:55:21AM +0400, Roman Kagan wrote:
> > On Fri, Apr 08, 2005 at 02:49:13PM -0700, Greg KH wrote:
> > > On Thu, Mar 31, 2005 at 07:40:46PM +0400, Roman Kagan wrote:
> > > > Any c
wn by another two
points, here's a cosmetic patch on top of that one.
Roman.
Replace container_of by list_entry for list entries; replace
list_for_each with list_for_each_entry.
Signed-off-by: Roman Kagan <[EMAIL PROTECTED]>
drivers/base/bus.c |8 +++-
1 files changed, 3 inser
On Wed, Apr 13, 2005 at 09:40:17PM +0400, Roman Kagan wrote:
> With 2.6.11 and 2.6.12-rc2 (and perhaps a few versions before) usb
> drivers for multi-interface devices, which do
> usb_driver_release_interface() in their disconnect(), make rmmod hang.
>
> It turns out to be due to a
klists will be discussed
in another mail.
Signed-off-by: Roman Kagan <[EMAIL PROTECTED]>
bus.c |5 ++---
1 files changed, 2 insertions(+), 3 deletions(-)
--- linux-2.6.12-rc2/drivers/base/bus.c.orig2005-03-18 04:34:06.0
+0300
+++ linux-2.6.12-rc2/drivers/base/bus.c 2005
On Fri, Apr 08, 2005 at 02:49:13PM -0700, Greg KH wrote:
> On Thu, Mar 31, 2005 at 07:40:46PM +0400, Roman Kagan wrote:
> > On Wed, Mar 30, 2005 at 01:47:55PM -0800, Greg KH wrote:
> > > ChangeSet 1.2181.4.57, 2005/03/24 14:34:40-08:00, [EMAIL PROTECTED]
> > >
> >
On Wed, Apr 06, 2005 at 11:34:16AM +0200, maximilian attems wrote:
> On Wed, 06 Apr 2005, Duncan Sands wrote:
> > > Apr 5 12:07:30 sputnik kernel: usb 1-1: USB disconnect, address 7
> > > Apr 5 12:07:30 sputnik kernel: usb_unlink_urb() is deprecated for
> > > synchronous unlinks. Use usb_kill_u
On Mon, Mar 28, 2005 at 12:19:36AM +0400, Roman Kagan wrote:
> On Fri, Mar 25, 2005 at 09:40:48AM -0800, Greg KH wrote:
> > On Fri, Mar 25, 2005 at 10:43:01AM +0300, Roman Kagan wrote:
> > > > > The patch expects a matching change in scripts/mod/file2alias.c I sent
> >
On Thu, Mar 31, 2005 at 09:50:20PM +0200, trem wrote:
> when I boot:
> localhost kernel: usb 3-6.1: no stage 1 firmware found!<6>parport0:
> PC-style at 0x378 [PCSPP,TRISTATE,EPP]
> localhost modprobe: FATAL: Error inserting speedtch
> (/lib/modules/2.6.12-rc1/kernel/drivers/usb/atm/speedtch.ko):
On Wed, Mar 30, 2005 at 01:15:44PM -0800, David Brownell wrote:
> On Wednesday 30 March 2005 12:12 pm, Duncan Sands wrote:
> > This change was discussed some time ago on the usbatm list -
> > David Woodhouse didn't like it IIRC, but perhaps he has changed
> > his mind? I think his main objection t
, allowing for its straightforward use with modprobe.
>
> Signed-off-by: Roman Kagan <[EMAIL PROTECTED]>
> Signed-off-by: Greg Kroah-Hartman <[EMAIL PROTECTED]>
Any chance to get my other patch re. modifying the module alias patterns
in usb modules to correctly handle bcdDevice ra
On Thu, Mar 31, 2005 at 09:03:11AM +0100, David Woodhouse wrote:
> On Thu, 2005-03-31 at 09:50 +0200, Duncan Sands wrote:
> > I'd rather have "select". I'm considering having it select ATM as well...
> > I think select is more logical: if someone has this modem, it's natural
> > that they simply c
On Fri, Mar 25, 2005 at 09:40:48AM -0800, Greg KH wrote:
> On Fri, Mar 25, 2005 at 10:43:01AM +0300, Roman Kagan wrote:
> > > > The patch expects a matching change in scripts/mod/file2alias.c I sent
> > > > to linux-hotplug-devel list a few days ago (i.e. puts
> >
On Thu, Mar 24, 2005 at 02:34:59PM -0800, Greg KH wrote:
> On Fri, Mar 04, 2005 at 01:01:58PM +0300, Roman Kagan wrote:
> > The patch below adds MODNAME environment variable to usb hotplug
> > callout, allowing for its straightforward use with modprobe.
>
> Sorry for the dela
that dldh).
Please consider applying.
Roman.
Signed-off-by: Roman Kagan <[EMAIL PROTECTED]>
drivers/usb/core/usb.c | 25 +
1 files changed, 25 insertions(+)
--- linux-2.6.11/drivers/usb/core/usb.c.modname 2005-03-02 10:38:09.0
+0300
+++ linux-
On Wed, Feb 16, 2005 at 12:27:51PM -0800, David Brownell wrote:
> On Wednesday 16 February 2005 12:01 pm, Roman Kagan wrote:
> > BTW the sanity checks added to usbfs handlers in 2.6.11 seem to allow
> > exactly the opposite behavior, i.e. both bulk and interrupt URBs on
> > in
On Wed, Feb 16, 2005 at 09:01:11AM -0800, Srdjan Sobajic wrote:
> Regarding the actual problem, I doubt that the endpoints change type.
Sure they don't.
> They're bulk, one in, one out. However, on Windows the API for
> accessing bulk and interrupt endpoints is the same, while you have a
> differ
On Wed, Feb 16, 2005 at 10:28:17AM -0500, Alan Stern wrote:
> You're not doing anything wrong. The question is, what is USB Snoopy
> doing wrong?
>
> If you decode the TransferBuffer contents for URB 3 coming back, you can
> easily see that both endpoints are listed with bmAttributes = 0x02 = B
Hi Duncan,
On Sat, 2005-01-15 at 13:37 +0100, Duncan Sands wrote:
> I agree that it makes more sense to only create the ATM device when the modem
> is fully
> initialised.
Great. This should simplify the initialisation stuff dramatically,
making the complicated interaction between usb_atm and
Hi Duncan,
On Mon, 2005-01-10 at 17:44 +0100, Duncan Sands wrote:
> (3) Setting the MAC address in cxacru_got_firmware. This one worries me: we
> create
> the ATM device when the device is probed. The firmware is loaded at some
> later time
> (in a kernel thread, because it would take too lo
Hi,
On Mon, 2005-01-10 at 17:44 +0100, Duncan Sands wrote:
> (1) cxacru_blocking_completion, cxacru_timeout_kill and cxacru_start_wait_urb
> will have to go. I think Josep put them in because in linux 2.4 there was no
> way to cancel the generic versions in message.c on disconnect. In 2.6 any
>
h Duncan's patch applied. Please
consider applying and pushing to Linus before 2.6.10-final is out.
drivers/usb/atm/usb_atm.c | 10 +-
1 files changed, 5 insertions(+), 5 deletions(-)
Signed-off-by: Roman Kagan <[EMAIL PROTECTED]>
--- linux-2.6.10-rc3.debug/drivers/usb/atm/us
ore 2.6.10-final is out.
drivers/usb/atm/usb_atm.c | 10 +-
1 files changed, 5 insertions(+), 5 deletions(-)
Signed-off-by: Roman Kagan <[EMAIL PROTECTED]>
--- linux-2.6.10-rc3.debug/drivers/usb/atm/usb_atm.c2004-12-22
19:30:54.0 +0300
+++ linux-2.6.10-rc3.test
On Wed, 2004-12-22 at 15:00 +0100, Duncan Sands wrote:
> Hi Roman, can you please check that this patch also fixes the debugging
> problem
> you were presumably seeing.
Yes. Now uncommenting the appropriate define I get the right debugging
level.
One nitpick:
> +#ifdef VERBOSE_DEBUG
> +static
On Wed, Dec 22, 2004 at 12:09:05PM +0300, Michael Tokarev wrote:
> Roman Kagan wrote:
> >Attached is the driver for USB ADSL modems based on the Conexant
> >AccessRunner chipset using the recently introduced usb_atm
> >infrastructure. []
>
> Not to say anything about
Attached is the patch to include the cxacru.c in the build.
drivers/usb/atm/Kconfig | 13 +
drivers/usb/atm/Makefile |1 +
2 files changed, 14 insertions(+)
Signed-off-by: Roman Kagan <[EMAIL PROTECTED]>
Roman.
--- linux-2.6.10-rc3.orig/drivers/usb/atm/Kconfig 2
W_LOADER
#endif
#define DRIVER_AUTHOR "Roman Kagan, Josep Comas, David Woodhouse, Duncan
Sands"
#define DRIVER_VERSION "0.1"
#define DRIVER_DESC "Conexant AccessRunner USB driver version "
DRIVER_VERSION
static const char cxacru_driver_name[] = "cxacru&
ed, 7 insertions(+), 7 deletions(-)
Signed-off-by: Roman Kagan <[EMAIL PROTECTED]>
Roman.
--- linux-2.6.10-rc3.orig/drivers/usb/atm/usb_atm.c 2004-12-04
00:55:02.0 +0300
+++ linux-2.6.10-rc3/drivers/usb/atm/usb_atm.c 2004-12-21 21:39:21.0
+0300
@@ -81,8 +81,6 @@
#inc
ent).
The patch is against 2.6.10-rc3. Please consider applying and pushing
to Linus before 2.6.10-final is out.
drivers/usb/atm/usb_atm.c | 14 +++---
1 files changed, 7 insertions(+), 7 deletions(-)
Signed-off-by: Roman Kagan <[EMAIL PROTECTED]>
Roman.
--- linux-2.6.10
48 matches
Mail list logo
