problem with forecedeth in 2.6.11-2mdk

After 2days, my network connection went down, here is what was written
to the logs :

in /var/log/kernel/warnings :
Apr  3 04:03:18 localhost kernel: Badness in local_bh_enable at
kernel/softirq.c:140
Apr  3 04:03:18 localhost kernel:
Apr  3 04:03:18 localhost kernel: Call Trace:<IRQ>
<ffffffff8018be55>{local_bh_enable+53}
<ffffffff88250cff>{:ip_conntrack:destroy_conntrack+159}
Apr  3 04:03:18 localhost kernel:       
<ffffffff802fb878>{__kfree_skb+216}
<ffffffff881434e5>{:forcedeth:nv_drain_tx+133}
Apr  3 04:03:18 localhost kernel:       
<ffffffff8814389d>{:forcedeth:nv_tx_timeout+93}
<ffffffff80310670>{dev_watchdog+0}
Apr  3 04:03:18 localhost kernel:       
<ffffffff803106d3>{dev_watchdog+99}
<ffffffff8019007e>{run_timer_softirq+366}
Apr  3 04:03:18 localhost kernel:       
<ffffffff8018bd83>{__do_softirq+83} <ffffffff8018be15>{do_softirq+53}
Apr  3 04:03:18 localhost kernel:        <ffffffff80110e07>{do_IRQ+71}
<ffffffff8010e859>{ret_from_intr+0}
Apr  3 04:03:18 localhost kernel:         <EOI>
<ffffffff8010ccd6>{__switch_to+518} <ffffffff8033c2b0>{udp_poll+0}
Apr  3 04:03:18 localhost kernel:       
<ffffffff8010c5a0>{default_idle+0} <ffffffff8010c5c0>{default_idle+32}
Apr  3 04:03:18 localhost kernel:       
<ffffffff8010c6a2>{cpu_idle+50} <ffffffff804c279f>{start_kernel+447}
Apr  3 04:03:18 localhost kernel:        <ffffffff804c2263>{_sinittext+611}

in /var/log/kernel/info :
Apr  3 04:03:18 localhost kernel: NETDEV WATCHDOG: eth0: transmit timed out
Apr  3 04:03:18 localhost kernel: nv_stop_tx: TransmitterStatus
remained busy<7>eth0: tx_timeout: dead entries!

chips is :
forcedeth       : nVidia Corp.|Ethernet controller [BRIDGE_OTHER]
(vendor:10de device:0057 subv:1043 subd:8141)

reloading the driver did not help, after rebooting the machine it is
working again