linux.git/net/iucv, branch v2.6.26

iucv: Delay bus registration until core is ready.

2008-04-10T09:12:45+00:00

If we register the iucv bus after the infrastructure is ready,
userspace can start relying on it when it receives the uevent
for the bus.

Signed-off-by: Cornelia Huck 
Signed-off-by: Heiko Carstens 
Signed-off-by: Ursula Braun 
Signed-off-by: David S. Miller

iucv: get rid of in_atomic() use.

2008-04-10T09:12:03+00:00

This BUG_ON is not needed, since all (debug) checks are also done
in smp_call_function() which gets called by this function.

Signed-off-by: Heiko Carstens 
Signed-off-by: Ursula Braun 
Signed-off-by: David S. Miller

af_iucv: Use non-deprecated __RW_LOCK_UNLOCKED macro.

2008-04-10T09:11:24+00:00

Signed-off-by: Robert P. J. Day 
Signed-off-by: Ursula Braun 
Signed-off-by: David S. Miller

iucv: fix build error on !SMP

2008-03-03T20:12:33+00:00

Since a5fbb6d1064be885d2a6b82f625186753cf74848
"KVM: fix !SMP build error" smp_call_function isn't a define anymore
that folds into nothing but a define that calls up_smp_call_function
with all parameters. Hence we cannot #ifdef out the unused code
anymore...
This seems to be the preferred method, so do this for s390 as well.

net/iucv/iucv.c: In function 'iucv_cleanup_queue':
net/iucv/iucv.c:657: error: '__iucv_cleanup_queue' undeclared

Signed-off-by: Heiko Carstens 
Signed-off-by: Josef 'Jeff' Sipek 
Signed-off-by: Ursula Braun 
Signed-off-by: David S. Miller

[AF_IUCV]: defensive programming of iucv_callback_txdone

2008-02-08T02:07:44+00:00

The loop in iucv_callback_txdone presumes existence of an entry
with msg->tag in the send_skb_q list. In error cases this
assumption might be wrong and might cause an endless loop.
Loop is rewritten to guarantee loop end in case of missing
msg->tag entry in send_skb_q.

Signed-off-by: Ursula Braun 
Signed-off-by: David S. Miller

[AF_IUCV]: broken send_skb_q results in endless loop

2008-02-08T02:07:19+00:00

A race has been detected in iucv_callback_txdone().
skb_unlink has to be done inside the locked area.

In addition checkings for successful allocations are inserted.

Signed-off-by: Ursula Braun 
Signed-off-by: David S. Miller

[IUCV]: wrong irq-disabling locking at module load time

2008-02-08T02:06:52+00:00

Linux may hang when running af_iucv socket programs concurrently
with a load of module netiucv. iucv_register() tries to take the
iucv_table_lock with spin_lock_irq. This conflicts with
iucv_connect() which has a need for an smp_call_function while
holding the iucv_table_lock.
Solution: use bh-disabling locking in iucv_register()

Signed-off-by: Ursula Braun 
Signed-off-by: David S. Miller

[IUCV]: use LIST_HEAD instead of LIST_HEAD_INIT

2008-01-28T22:56:54+00:00

these three list_head are all local variables, but can also use
LIST_HEAD.

Signed-off-by: Denis Cheng 
Signed-off-by: David S. Miller

[NET]: Convert init_timer into setup_timer

2008-01-28T22:53:35+00:00

Many-many code in the kernel initialized the timer->function
and  timer->data together with calling init_timer(timer). There
is already a helper for this. Use it for networking code.

The patch is HUGE, but makes the code 130 lines shorter
(98 insertions(+), 228 deletions(-)).

Signed-off-by: Pavel Emelyanov 
Acked-by: Arnaldo Carvalho de Melo 
Signed-off-by: David S. Miller

[S390] Explicitly code allocpercpu calls in iucv

2007-11-20T10:13:47+00:00

The iucv is the only user of the various functions that are used to bring
parts of cpus up and down. Its the only allocpercpu user that will do
I/O on per cpu objects (which is difficult to do with virtually mapped memory).
And its the only use of allocpercpu where a GFP_DMA allocation is done.

Remove the allocpercpu calls from iucv and code the allocation and freeing
manually. After this patch it is possible to remove a large part of
the allocpercpu API.

Signed-off-by: Christoph Lameter 
Signed-off-by: Martin Schwidefsky