From: Michael Buesch <[EMAIL PROTECTED]>
This fixes a netdev watchdog timeout problem.
The software needs to call netif_tx_disable before running the
hardware calibration code. The problem condition can be shown by the
following timegraph.
|---5secs - ~10 jiffies time---|---|OOPS
^
On Saturday 28 October 2006 2:13 pm, Christoph Hellwig wrote:
>
> I still don't quite like the approach. What about simply putting
> the mii using functions into usbnet-mii.c and let makefile doing
> all the work? This would require a second set of ethtool ops,
> but I'd actually consider that a
On Sat, Oct 28, 2006 at 02:10:09PM -0700, David Brownell wrote:
> On Saturday 28 October 2006 4:21 am, Christoph Hellwig wrote:
>
> > This is really awkward and against what we do in any other driver.
>
> Awkward, yes -- which is why I posted the non-awkward version,
> which is repeated below. (
On Sat, Oct 28, 2006 at 02:10:09PM -0700, David Brownell wrote:
> On Saturday 28 October 2006 4:21 am, Christoph Hellwig wrote:
>
> > This is really awkward and against what we do in any other driver.
>
> Awkward, yes -- which is why I posted the non-awkward version,
> which is repeated below. (
On Saturday 28 October 2006 4:21 am, Christoph Hellwig wrote:
> This is really awkward and against what we do in any other driver.
Awkward, yes -- which is why I posted the non-awkward version,
which is repeated below. (No thanks to "diff" for making the
patch ugly though; the resulting code is
Hi Stephen,
In my own kernels, I've added your backport of SKGE to 2.4 that I found
here :
http://developer.osdl.org/shemminger/releases/skge-sky2-backport.tar.bz2
It seems to work pretty well compared to the original syskonnect driver
(up to and including 8.36). Several people around me have
Hi John,
Please
git pull http://bu3sch.de/git/wireless-dev.git for-linville
This will pull the following things since my last
pull request:
Michael Buesch:
bcm43xx-d80211: Fix DMA engine TX buffer unmap crash.
bcm43xx-d80211: Don't ignore return value of pci_enable_device()
drivers
On Sat, Oct 28, 2006 at 03:34:52PM +0200, Eric Dumazet ([EMAIL PROTECTED])
wrote:
> >>>+ list_del(&k->ready_entry);
> >>Arg... no
> >>
> >>You cannot call list_del() , then check overflow_kevent.
> >>
> >>I you call list_del on what happens to be the kevent pointed by
> >>overflow_kevent, you lo
Evgeniy Polyakov a e'crit :
On Sat, Oct 28, 2006 at 03:23:40PM +0200, Eric Dumazet ([EMAIL PROTECTED])
wrote:
diff --git a/kernel/kevent/kevent_user.c b/kernel/kevent/kevent_user.c
index 711a8a8..ecee668 100644
--- a/kernel/kevent/kevent_user.c
+++ b/kernel/kevent/kevent_user.c
@@ -235,6 +235,3
On Sat, Oct 28, 2006 at 03:23:40PM +0200, Eric Dumazet ([EMAIL PROTECTED])
wrote:
> >diff --git a/kernel/kevent/kevent_user.c b/kernel/kevent/kevent_user.c
> >index 711a8a8..ecee668 100644
> >--- a/kernel/kevent/kevent_user.c
> >+++ b/kernel/kevent/kevent_user.c
> >@@ -235,6 +235,36 @@ static void
Evgeniy Polyakov a e'crit :
On Sat, Oct 28, 2006 at 02:36:31PM +0200, Eric Dumazet ([EMAIL PROTECTED])
wrote:
Evgeniy Polyakov a e'crit :
On Sat, Oct 28, 2006 at 12:28:12PM +0200, Eric Dumazet
([EMAIL PROTECTED]) wrote:
I really dont understand how you manage to queue multiple kevents in the
On Sat, Oct 28, 2006 at 02:36:31PM +0200, Eric Dumazet ([EMAIL PROTECTED])
wrote:
> Evgeniy Polyakov a e'crit :
> >On Sat, Oct 28, 2006 at 12:28:12PM +0200, Eric Dumazet
> >([EMAIL PROTECTED]) wrote:
> >>
> >>I really dont understand how you manage to queue multiple kevents in the
> >>'overflow
Evgeniy Polyakov a e'crit :
On Sat, Oct 28, 2006 at 12:28:12PM +0200, Eric Dumazet ([EMAIL PROTECTED])
wrote:
I really dont understand how you manage to queue multiple kevents in the
'overflow list'. You just queue one kevent at most. What am I missing ?
There is no overflow list - it is a
On Wed, Oct 25, 2006 at 04:58:58PM -0700, Randy Dunlap wrote:
> On Wed, 25 Oct 2006 15:27:09 -0700 David Brownell wrote:
>
> > Instead, "usbnet.c" should #ifdef the relevant ethtool hooks
> > according to CONFIG_MII ... since it's completely legit to
> > use usbnet with peripherals that don't need
On Sat, Oct 28, 2006 at 12:28:12PM +0200, Eric Dumazet ([EMAIL PROTECTED])
wrote:
> +/*
> + * Called under kevent_user->ready_lock, so updates are always protected.
> + */
> +int kevent_user_ring_add_event(struct kevent *k)
> +{
> + unsigned int pidx, off;
> + struct kevent_mring *ring, *c
+/*
+ * Called under kevent_user->ready_lock, so updates are always protected.
+ */
+int kevent_user_ring_add_event(struct kevent *k)
+{
+ unsigned int pidx, off;
+ struct kevent_mring *ring, *copy_ring;
+
+ ring = k->user->pring[0];
+
+ if ((ring->kidx + 1 == ring->
On Sat, Oct 28, 2006 at 12:04:10PM +0200, Eric Dumazet ([EMAIL PROTECTED])
wrote:
> Evgeniy Polyakov a écrit :
>
> >+file = fget(k->event.id.raw[0]);
> >+if (!file)
> >+return -ENODEV;
>
> Please, do us a favor, and use EBADF instead of ENODEV.
>
> EBADF : /* Bad file number
Evgeniy Polyakov a écrit :
+ file = fget(k->event.id.raw[0]);
+ if (!file)
+ return -ENODEV;
Please, do us a favor, and use EBADF instead of ENODEV.
EBADF : /* Bad file number */
ENODEV : /* No such device */
You have many ENODEV uses in your patches and that reall
On Fri, Oct 27, 2006 at 11:47:52PM +0200, Yakov Lerner wrote:
> I'd like to find/gather estimates about 2.4 vs 2.6 share in [small]
> linux routers in 2006. Can anyone offer estimates and/or references ?
For ARM devices, 2.4 is still definitely in the majority.
The reason for that appears to be
19 matches
Mail list logo
