Here is a trial to do something suggested by Patrick McHardy.
[NET_SCHED] sch_htb:
- turn intermediate classes into leaves again when their last child is deleted
(qdisc of deleted class is reused; struct htb_class changed)
- sch_tree_lock added in htb_put before htb_destroy_class
(for con
Here are some fixes proposals suggested by Patrick McHardy.
[NET_SCHED] sch_cbq:
- deactivating of active classes when grafting
- purging of queue/q.qlen adjustment when deleting an active class
- deactivating of active classes when q.qlen drops to zero in ->drop()
- a redundant instruction re
On Fri, Nov 24, 2006 at 02:07:08PM +0100, Patrick McHardy wrote:
> Jarek Poplawski wrote:
...
> > Probably it is as usual something with my eyes but
> > it seems cbq needs in cbq_delete or cbq_destroy_class
> > some treatment (about qdisc_destroy) similar to htb.
>
> No, you're right about this on
On Mon, Nov 27, 2006 at 01:59:34AM +0100, Adrian Bunk wrote:
> NET_SCH_ATM doesn't need ipcommon.o
>
> Signed-off-by: Adrian Bunk <[EMAIL PROTECTED]>
>
> --- linux-2.6.19-rc6-mm1/net/atm/Makefile.old 2006-11-26 08:50:05.0
> +0100
> +++ linux-2.6.19-rc6-mm1/net/atm/Makefile 2006-11-26
The return value of crypto_alloc_comp() should be checked by
IS_ERR().
Cc: David Miller <[EMAIL PROTECTED]>
Cc: Hideaki YOSHIFUJI <[EMAIL PROTECTED]>
Signed-off-by: Akinobu Mita <[EMAIL PROTECTED]>
---
net/ipv4/ipcomp.c |2 +-
net/ipv6/ipcomp6.c |2 +-
2 files changed, 2 insertions(+),
The return value of crypto_alloc_hash() should be checked by
IS_ERR().
Cc: David Miller <[EMAIL PROTECTED]>
Signed-off-by: Akinobu Mita <[EMAIL PROTECTED]>
---
net/sctp/socket.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
Index: work-fault-inject/net/sctp/socket.c
=
Johannes Berg wrote:
On Sun, 2006-11-26 at 13:25 -0500, Daniel Drake wrote:
No, we could stick with our existing setup if the stack did something
different.
But you do need full refcounting I guess.
No, I don't think we would need to move away from our existing method:
spin_lock_irqsave(
NET_SCH_ATM doesn't need ipcommon.o
Signed-off-by: Adrian Bunk <[EMAIL PROTECTED]>
--- linux-2.6.19-rc6-mm1/net/atm/Makefile.old 2006-11-26 08:50:05.0
+0100
+++ linux-2.6.19-rc6-mm1/net/atm/Makefile 2006-11-26 08:56:29.0
+0100
@@ -10,7 +10,6 @@
atm-$(subst m,y,$(CONFIG_
Adrian Bunk wrote:
Considering that NET_CLS_POLICE has been marked as obsolete for more
than one year, would a patch to remove it be acceptable?
cu
Adrian
People still request the "2.4 policer" on LARTC as unlike its
replacement it hooks after prerouting/denat so they can police on local
a
> before i go opening bugs with the distribution folks, could someone chime
> in as to what is the recommended approach these days? did grat arp fall
> out of favour, or is it just a case of userland not keeping up?
The ifup script in iproute2 does it in user land, but nobody uses it
directly
Herbert Poetzl <[EMAIL PROTECTED]> writes:
> On Sat, Nov 25, 2006 at 01:21:39AM -0700, Eric W. Biederman wrote:
>> There are two techniques in real use.
>> - Bind/Accept filtering
>>
>> Which layer 3 addresses a socket can bind/accept are filtered,
>> but otherwise the network stack remains
Herbert Poetzl wrote:
On Sat, Nov 25, 2006 at 01:21:39AM -0700, Eric W. Biederman wrote:
Then the question is how do we reduce the overhead when we don't have
enough physical network interfaces to go around. My feeling is that
we could push the work to the network adapters and allow single
p
Adrian Bunk <[EMAIL PROTECTED]> writes:
> It's still not used...
>
> So let's unexport it again until some driver that actually uses it shows up.
Fair enough.
--
Krzysztof Halasa
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
Mor
Johannes Berg wrote:
On Sun, 2006-11-26 at 13:25 -0500, Daniel Drake wrote:
No, we could stick with our existing setup if the stack did something
different.
But you do need full refcounting I guess.
Because for
the stack it wouldn't be required if it'd simply not start scanning when
queue i
On Sun, 2006-11-26 at 13:25 -0500, Daniel Drake wrote:
> No, we could stick with our existing setup if the stack did something
> different.
But you do need full refcounting I guess.
> > Because for
> > the stack it wouldn't be required if it'd simply not start scanning when
> > queue is disable
On Sat, Nov 25, 2006 at 01:21:39AM -0700, Eric W. Biederman wrote:
>
> jamal <[EMAIL PROTECTED]> writes:
>
> > On Fri, 2006-27-10 at 11:10 +0200, Daniel Lezcano wrote:
> >
> >> No, it uses virtualization at layer 2 and I had already mention it
> >> before (see the first email of the thread), but
Johannes Berg wrote:
Would you actually need a fully refcounted enable/disable?
No, we could stick with our existing setup if the stack did something
different.
Because for
the stack it wouldn't be required if it'd simply not start scanning when
queue is disabled and stop scanning immediate
On Sun, 26 Nov 2006, James Courtier-Dutton wrote:
> dean gaudet wrote:
> > On Sun, 26 Nov 2006, James Courtier-Dutton wrote:
> >
> > > dean gaudet wrote:
> > > > hi...
> > > >
> > > > i ran into some problems recently which would have been avoided if my
> > > > box
> > > > did a gratuitous arp a
On Sun, 2006-11-26 at 11:37 -0500, Daniel Drake wrote:
> However the other reason for the patch (transmit
> queue needed for active scanning) is bogus,
I think that was just a misunderstanding.
> and the patch introduces a
> problem where session frames may be transmitted during scanning (usin
Johannes Berg wrote:
The problem is that queue disabling isn't refcounted so that a scan that
collides with bcm43xx having disabled the queue for calibration might
re-enable the queue while bcm43xx is still calibrating.
I agree with that part. However the other reason for the patch (transmit
q
Arjan van de Ven wrote:
On Thu, 2006-11-23 at 04:35 +, Al Viro wrote:
I would even say 10 function calls deep to allocate file blocks
is overkill, but 22 it just astronomically bad.
Especially since a large part is due to cxfs...
-
it's a bit sad to see XFS this crippled in linux due to a
[NET_SCHED]: policer: restore compatibility with old iproute binaries
The tc actions increased the size of struct tc_police, which broke
compatibility with old iproute binaries since both the act_police
and the old NET_CLS_POLICE code check for an exact size match.
Since the new members are not e
On Sun, 26 Nov 2006, James Courtier-Dutton wrote:
dean gaudet wrote:
On Sun, 26 Nov 2006, James Courtier-Dutton wrote:
dean gaudet wrote:
hi...
i ran into some problems recently which would have been avoided if my box
did a gratuitous arp as it brought up all interfaces (the router took
f
Adrian Bunk wrote:
> Considering that NET_CLS_POLICE has been marked as obsolete for more
> than one year, would a patch to remove it be acceptable?
Looking at this, the only reason I thought why we kept this
option was to keep old iproute binaries working. But struct
tc_police in new kernels is
dean gaudet wrote:
On Sun, 26 Nov 2006, James Courtier-Dutton wrote:
dean gaudet wrote:
hi...
i ran into some problems recently which would have been avoided if my box
did a gratuitous arp as it brought up all interfaces (the router took
forever to timeout the ARP entries for interface aliase
On Sunday 26 November 2006 05:05, Daniel Drake wrote:
> Larry Finger wrote:
> > In the scan section of ieee80211softmac, network transmits are disabled.
> > Clearly, this does not make any sense as transmit is necessary for active
> > scanning, and transmits are not used when passive scanning.
>
>
On Sunday 26 November 2006 11:03, Johannes Berg wrote:
> On Sat, 2006-11-25 at 23:05 -0500, Daniel Drake wrote:
>
> > Surely disabling the queue actually makes sense, in order to avoid
> > frames designated for the "current session" being transmitted on
> > different channels during scanning?
>
On Sat, 2006-11-25 at 23:05 -0500, Daniel Drake wrote:
> Surely disabling the queue actually makes sense, in order to avoid
> frames designated for the "current session" being transmitted on
> different channels during scanning?
Also, for bcm43xx this isn't a problem since the firmware (optiona
On Sat, 2006-11-25 at 23:05 -0500, Daniel Drake wrote:
> Surely disabling the queue actually makes sense, in order to avoid
> frames designated for the "current session" being transmitted on
> different channels during scanning?
The problem is that queue disabling isn't refcounted so that a sca
29 matches
Mail list logo
