Bug#664461: atl1c: AR8152: "transmit queue 0 timed out" and network is unusable until reset

[Jonathan Nieder <jrnieder@gmail.com>]

forcemerge 664461 613874
quit

axst@users.sourceforge.net wrote:

> Sorry to say it does work. I've tried it several times and for hours, doing a big backup,
> and the wired network access runs smoothly.

Thanks; nice to hear.

[...]
> As we know which module it is, can't the authors tell you the exact
> version and patch?

Assuming it's a bug in the driver, here's the list of patches to sift
through.

 d94202bc7eb5 atl1c: dont use highprio tx queue
 9e903e085262 net: add skb frag size accessors
 5d6bcdfe38ce net: use DMA_x_DEVICE and dma_mapping_error with skb_frag_dma_map
 8d1bb865575d atl1c: convert to SKB paged frag API.
 afc4b13df143 net: remove use of ndo_set_multicast_list in drivers
 2b133ad6e9e9 atl*: Move the Atheros drivers
 46facce9130e atl1c: do vlan cleanup
 95cb3656c09f net: Remove unneeded version.h includes from drivers/net/
 a6b7a407865a net: remove interrupt.h inclusion from netdevice.h
 d187c1aab8d3 atl1c: atl1c_resume() is only used when CONFIG_PM_SLEEP is defined.
 707394972093 ethtool: cosmetic: Use ethtool ethtool_cmd_speed API
 25db0338813a ethtool: Use full 32 bit speed range in ethtool's set_settings
 cb771838715b atl1c: Fix work event interrupt/task races
 0c78641d7f67 atl1c: Fix set-but-unused variable.
 782d640afd15 net: atl*: convert to hw_features
 81b504b81437 atl1[ce]: fix sparse warnings
 0303adeee3d6 atl1c: Do not call device_init_wakeup() in atl1c_probe()
 94dde7e451fa atl1c: Add missing PCI device ID
 34aac66cc2c3 atl1c: remove private #define.
 5a3a7658863f Comment typo fixes for 'descriptor'
 762e30239cce atl1c: Do not use legacy PCI power management
 0d0b16727f24 net: Fix drivers advertising HW_CSUM feature to use csum_start
 0fb1e54ed177 atl1c: make functions static
 eab6d18d20fc vlan: Don't check for vlan group before vlan_tx_tag_present.
 bc8acf2c8c3e drivers/net: avoid some skb->ip_summed initializations
 2bc9e2346580 drivers/net/atl1c: Remove double test
 a2c483a19a96 atl1c: use net_device_stats from struct net_device
 8f574b35f22f atl1c: Add AR8151 v2 support and change L0s/L1 routine
 a4b770972b8f drivers/net: Remove unnecessary returns from void function()s
 22bedad3ce11 net: convert multicast list to list_head
 b05b7d9563f1 net: remove redundant code
 5a0e3ad6af86 include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
 0ddf477b8a9b net: convert multiple drivers to use netdev_for_each_mc_addr, part3
 a3aa18842a53 drivers/net/: use DEFINE_PCI_DEVICE_TABLE()
 829911725f4b drivers/net/atl1c/atl1c_main.c: use %pM to show MAC address
 4b45e3424e72 atl1c:add pci map direction in atl1c_buffer flags
 af901ca181d9 tree-wide: fix assorted typos all over the place
 9aff7e922b85 drivers/net/atl1c: remove exceptional & on function name
 c6060be46fbd atl1c: change atl1c_buffer struct and restructure clean atl1c_buffer procedure

Information about whether 2.6.38 from snapshot.debian.org reproduces
the problem would cut that list in half. ;-)

Thanks,
Jonathan