I'm sorry to inform even with 3.17.0-031700rc3-generic the problem does not seemed to be fixed.
Today I used the wireless in question for the first time since my last post and after about 15 minutes usage I got bad connectivity and found the following in dmesg.
[10806.700419] iwlwifi 0000:01:00.0: fail to flush all tx fifo queues Q 2
[10806.700424] iwlwifi 0000:01:00.0: Current SW read_ptr 168 write_ptr 237
[10806.700510] iwl data: 00000000: 00 00 00 00 00 00 00 00 00 ff ff 00 00 00 00 00 ................
[10806.700564] iwlwifi 0000:01:00.0: FH TRBs(0) = 0x00000000
and so on.
kernel is still
3.17.0-031700rc3-generic #201408312235 SMP Mon Sep 1 02:36:10 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux
However today I resumed the machine from memory. Last time I used the network for about an hour without any problems I did a fresh boot. So maybe the problem is in wake from sleep part of the module.
Also I'm afraid I currently don't have the free time to do the mentioned bisect if I find a Kernel that works. However I will continue using the latest kernel from mainline and report back if I find one without the flush problem.
I'm sorry to inform even with 3.17.0- 031700rc3- generic the problem does not seemed to be fixed.
Today I used the wireless in question for the first time since my last post and after about 15 minutes usage I got bad connectivity and found the following in dmesg.
[10806.700419] iwlwifi 0000:01:00.0: fail to flush all tx fifo queues Q 2
[10806.700424] iwlwifi 0000:01:00.0: Current SW read_ptr 168 write_ptr 237
[10806.700510] iwl data: 00000000: 00 00 00 00 00 00 00 00 00 ff ff 00 00 00 00 00 ................
[10806.700564] iwlwifi 0000:01:00.0: FH TRBs(0) = 0x00000000
and so on.
kernel is still 031700rc3- generic #201408312235 SMP Mon Sep 1 02:36:10 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux
3.17.0-
However today I resumed the machine from memory. Last time I used the network for about an hour without any problems I did a fresh boot. So maybe the problem is in wake from sleep part of the module.
Also I'm afraid I currently don't have the free time to do the mentioned bisect if I find a Kernel that works. However I will continue using the latest kernel from mainline and report back if I find one without the flush problem.