Bug 16176
| Summary: | Microcode errors with iwl3945 | ||
|---|---|---|---|
| Product: | Drivers | Reporter: | Steinar H. Gunderson (steinar+kernel) |
| Component: | network-wireless | Assignee: | Reinette Chatre (reinette.chatre) |
| Status: | CLOSED UNREPRODUCIBLE | ||
| Severity: | normal | CC: | dode, linville, maciej.rutecki, reinette.chatre, rjw |
| Priority: | P1 | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Kernel Version: | 2.6.35-rc2 | Subsystem: | |
| Regression: | Yes | Bisected commit-id: | |
| Bug Depends on: | |||
| Bug Blocks: | 16055 | ||
| Attachments: | more internal scan debug | ||
|
Description
Steinar H. Gunderson
2010-06-10 19:28:47 UTC
We recently pushed some fixes for this issue upstream but they did not make it into 2.6.35-rc2 - they will be in rc3. Could you try with the latest linux-2.6? On Fri, Jun 11, 2010 at 06:46:59PM +0000, bugzilla-daemon@bugzilla.kernel.org wrote: > https://bugzilla.kernel.org/show_bug.cgi?id=16176 > We recently pushed some fixes for this issue upstream but they did not make > it > into 2.6.35-rc2 - they will be in rc3. Could you try with the latest > linux-2.6? I tried -rc3, and it was much, much worse, but in a different way. Basically, holding up TCP connections for more than a few seconds at a time was completely impossible. I could not even get through the captive portal here; everything would just hang randomly. (Tried N browsers and telnet.) /* Steinar */ Created attachment 26785 [details] more internal scan debug The commit I refer to in comment #1 is in rc3 and should take care of the BUG you encounter. You mention that things are behaving in a different way - are you still seeing the same errors? The fix I pointed out does not address the SYSASSERT. On a closer look that SYSASSERT is the hardware telling the driver that it is not legal to scan on channel zero. This is its regulatory enforcement. To look into this further I'd like to see some scan and also some general debugging. Please ensure your driver is compiled with debug support (CONFIG_IWLWIFI_DEBUG) and then load the module with debug flags as follows: modprobe iwl3945 debug=0x43fff I attach an additional debug patch that will get us a little more data if you are able to get it included in your driver that will be great. OK, well, I guess I have bad news :-) I didn't have time to check this last week, and now I've traveled back from the US to Switzerland again. I don't see these problems anymore — the oopses were fixed already in 2.6.35-rc3, but the TCP problems are nowhere to be seen. Of course, if this really was a regulatory domain problem, I'd probably have to travel back to the US and the offending APs to reproduce... IPv6 routing is now, on the other hand, completely broken (I can only ping stuff on my local network, nothing that has to go through a default route), but I take it this is not iwl3945's fault. IOW, this is now unreproducible for me, so if nobody else steps forward with the same problems, I guess the bug has to be closed. (In reply to comment #4) > IPv6 routing is now, on the other hand, completely broken (I can only ping > stuff on my local network, nothing that has to go through a default route), > but > I take it this is not iwl3945's fault. You may want to try http://git.kernel.org/?p=linux/kernel/git/iwlwifi/iwlwifi-2.6.git;a=commit;h=4d23e4e5eb50431426facf192354ad2506e2dd40 which fixes a multicast issue in iwl3945. (In reply to comment #4) > I don't see > these problems anymore — the oopses were fixed already in 2.6.35-rc3, but the > TCP problems are nowhere to be seen. Of course, if this really was a > regulatory > domain problem, I'd probably have to travel back to the US and the offending > APs to reproduce... > ... > IOW, this is now unreproducible for me, so if nobody else steps forward with > the same problems, I guess the bug has to be closed. Exactly from the moment where I built NetworkManager 0.8.9997 and replaced 0.8.4 with it, I have a lot of the below in dmesg/syslog, and scanning Wifi networks doesn't seem to work. I might have not built or set up NetworkManager entirely correct (even though everything else than scanning seems to work fine), but I am wondering how another version of NetworkManager can cause the iwl3945 Microcode SW error to happen? I am running Ubuntu 11.04 2.6.38-8-generic #42-Ubuntu. [ 5420.614860] iwl3945 0000:04:00.0: Microcode SW error detected. Restarting 0x82000008. [ 5420.614872] iwl3945 0000:04:00.0: Loaded firmware version: 15.32.2.9 [ 5420.614908] iwl3945 0000:04:00.0: Start IWL Error Log Dump: [ 5420.614914] iwl3945 0000:04:00.0: Status: 0x0002A2E4, count: 1 [ 5420.614919] iwl3945 0000:04:00.0: Desc Time asrtPC blink2 ilink1 nmiPC Line [ 5420.615153] iwl3945 0000:04:00.0: SYSASSERT (0x5) 2932061891 0x008B6 0x13756 0x0031C 0x00000 764 [ 5420.615156] [ 5420.615289] iwl3945 0000:04:00.0: Start IWL Event Log Dump: display last 20 count [ 5420.615334] iwl3945 0000:04:00.0: 2932060759 0x00000006 0463 [ 5420.615360] iwl3945 0000:04:00.0: 2932060760 0x00000006 0462 [ 5420.615385] iwl3945 0000:04:00.0: 2932060795 0x00000000 0201 [ 5420.615411] iwl3945 0000:04:00.0: 2932060806 0x0000001c 0206 [ 5420.615437] iwl3945 0000:04:00.0: 2932060808 0x00400001 0204 [ 5420.615462] iwl3945 0000:04:00.0: 2932060812 0x00400001 0219 [ 5420.615488] iwl3945 0000:04:00.0: 2932060813 0x00020041 0211 [ 5420.615513] iwl3945 0000:04:00.0: 2932060818 0x00000000 0212 [ 5420.615539] iwl3945 0000:04:00.0: 2932061149 0x00000000 0215 [ 5420.615565] iwl3945 0000:04:00.0: 2932061223 0x00400008 0220 [ 5420.615590] iwl3945 0000:04:00.0: 2932061238 0x00000000 0301 [ 5420.615616] iwl3945 0000:04:00.0: 2932061507 0x000000d4 0303 [ 5420.615641] iwl3945 0000:04:00.0: 2932061834 0x442d0080 0401 [ 5420.615667] iwl3945 0000:04:00.0: 2932061848 0x00000017 0452 [ 5420.615692] iwl3945 0000:04:00.0: 2932061853 0x00103b6e 0450 [ 5420.615717] iwl3945 0000:04:00.0: 2932061859 0x00103b6e 0450 [ 5420.615743] iwl3945 0000:04:00.0: 2932061864 0x00103b6e 0450 [ 5420.615769] iwl3945 0000:04:00.0: 2932061869 0x00103b6e 0450 [ 5420.615794] iwl3945 0000:04:00.0: 2932061874 0x00103b6e 0450 [ 5420.615820] iwl3945 0000:04:00.0: 2932061893 0x00000100 0125 [ 5420.615848] iwl3945 0000:04:00.0: Error Reply type 0x000002FC cmd REPLY_SCAN_CMD (0x80) seq 0x442D ser 0x00340000 [ 5420.620278] iwl3945 0000:04:00.0: Can't stop Rx DMA. |