linux-stable.git/drivers/net, branch linux-2.6.19.y

sky2: fix ram buffer allocation settings

2007-03-03T00:32:46+00:00

Different chipsets have different amount of ram buffer (some have none),
so need to make sure that driver does proper setup for all cases from 0 on
to 48K, in units of 1K.

This is a backport of the code from 2.6.19 or later

Signed-off-by: Stephen Hemminger 
Signed-off-by: Greg Kroah-Hartman

sky2: dont flush good pause frames

2007-03-03T00:32:46+00:00

Don't mark pause frames as errors. This caused a bug that would
lockup transmitter, as well as all the other ports on a switch.

Signed-off-by: Stephen Hemminger 
Signed-off-by: Greg Kroah-Hartman

sky2: transmit timeout deadlock

2007-03-03T00:32:46+00:00

The code in transmit timeout incorrectly assumed that netif_tx_lock
was not set.

Signed-off-by: Stephen Hemminger 
Signed-off-by: Greg Kroah-Hartman

sky2: kfree_skb with IRQ with netconsole

2007-03-03T00:32:46+00:00

When using netconsole, it is possible for the driver to try and
free skb's with IRQ's disabled. This could cause problems if the skb
had a destructor function that didn't expect this. The documented semantics
is that skb->destructor callback happens in softirq.

Signed-off-by: Stephen Hemminger 
Signed-off-by: Greg Kroah-Hartman

sky2: fix for NAPI with dual port boards

2007-03-03T00:32:46+00:00

This driver uses port 0 to handle receives on both ports. So
the netif_poll_disable call in dev_close would end up stopping the
second port on dual port cards.

Signed-off-by: Stephen Hemminger 
Signed-off-by: Greg Kroah-Hartman

sky2: pci config blocks phy power control

2007-03-03T00:32:46+00:00

PHY power control won't work because writes to PCI config space
are not re-enabled.

Signed-off-by: Stephen Hemminger 
Signed-off-by: Greg Kroah-Hartman

netpoll: drivers must not enable IRQ unconditionally in their NAPI handler

2007-03-03T00:32:45+00:00

netpoll: drivers must not enable IRQ unconditionally in their NAPI handler

net/core/netpoll.c::netpoll_send_skb() calls the poll handler when
it is available. As netconsole can be used from almost any context,
IRQ must not be enabled blindly in the NAPI handler of a driver which
supports netpoll.

b57bd06655a028aba7b92e1c19c2093e7fcfb341 fixed the issue for the
8139too.c driver.

Signed-off-by: Francois Romieu 
Signed-off-by: Jeff Garzik 
Signed-off-by: Greg Kroah-Hartman

net, 8139too.c: fix netpoll deadlock

2007-02-24T00:24:27+00:00

fix deadlock in the 8139too driver: poll handlers should never forcibly
enable local interrupts, because they might be used by netpoll/printk
from IRQ context.

  =================================
  [ INFO: inconsistent lock state ]
  2.6.19 #11
  ---------------------------------
  inconsistent {softirq-on-W} -> {in-softirq-W} usage.
  swapper/1 [HC0[0]:SC1[1]:HE1:SE0] takes:
   (&npinfo->poll_lock){-+..}, at: [] net_rx_action+0x64/0x1de
  {softirq-on-W} state was registered at:
    [] mark_lock+0x5b/0x39c
    [] mark_held_locks+0x4b/0x68
    [] trace_hardirqs_on+0x115/0x139
    [] rtl8139_poll+0x3d7/0x3f4
    [] netpoll_poll+0x82/0x32f
    [] netpoll_send_skb+0xc9/0x12f
    [] netpoll_send_udp+0x253/0x25b
    [] write_msg+0x40/0x65
    [] __call_console_drivers+0x45/0x51
    [] _call_console_drivers+0x5d/0x61
    [] release_console_sem+0x11f/0x1d8
    [] register_console+0x1ac/0x1b3
    [] init_netconsole+0x55/0x67
    [] init+0x9a/0x24e
    [] kernel_thread_helper+0x7/0x10
    [] 0xffffffff
  irq event stamp: 819992
  hardirqs last  enabled at (819992): [] net_rx_action+0x39/0x1de
  hardirqs last disabled at (819991): [] net_rx_action+0x141/0x1de
  softirqs last  enabled at (817552): [] __do_softirq+0xa3/0xa8
  softirqs last disabled at (819987): [] do_softirq+0x5b/0xc9

  other info that might help us debug this:
  no locks held by swapper/1.

  stack backtrace:
   [] dump_trace+0x63/0x1e8
   [] show_trace_log_lvl+0x19/0x2e
   [] show_trace+0x12/0x14
   [] dump_stack+0x14/0x16
   [] print_usage_bug+0x23c/0x246
   [] mark_lock+0x108/0x39c
   [] __lock_acquire+0x361/0x9ed
   [] lock_acquire+0x56/0x72
   [] _spin_lock+0x35/0x42
   [] net_rx_action+0x64/0x1de
   [] __do_softirq+0x52/0xa8
   [] do_softirq+0x5b/0xc9
   [] irq_exit+0x3c/0x48
   [] do_IRQ+0xa4/0xbd
   [] common_interrupt+0x2e/0x34
   [] vprintk+0x2c0/0x309
   [] printk+0x1b/0x1d
   [] init+0x80/0x24e
   [] kernel_thread_helper+0x7/0x10
   =======================

Signed-off-by: Ingo Molnar 
Acked-by: Jeff Garzik 
Cc: Chuck Ebbert 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

prism54: correct assignment of DOT1XENABLE in WE-19 codepaths

2007-02-24T00:24:27+00:00

Correct assignment of DOT1XENABLE in WE-19 codepaths.
RX_UNENCRYPTED_EAPOL = 1 really means setting DOT1XENABLE _off_, and
vice versa.  The original WE-19 patch erroneously reversed that.  This
patch fixes association with unencrypted and WEP networks when using
wpa_supplicant.

It also adds two missing break statements that, left out, could result
in incorrect card configuration.

Applies to (I think) 2.6.19 and later.

Signed-off-by: Dan Williams 
Signed-off-by: Greg Kroah-Hartman

bcm43xx: Fix for oops on ampdu status

2007-02-24T00:24:26+00:00

If bcm43xx were to process an afterburner (ampdu) status response, Linux would oops. The
ampdu and intermediate status bits are properly named.

Signed-off-by: Michael Buesch 
Signed-off-by: Larry Finger 
Signed-off-by: Greg Kroah-Hartman