Bug 14417
| Summary: | [Regression] Wireless driver iwlagn+iwlcore doesn't work after resume (needs reloading) | ||
|---|---|---|---|
| Product: | Drivers | Reporter: | Eddy Petrișor (eddy.petrisor+linbug) |
| Component: | network-wireless | Assignee: | Reinette Chatre (reinette.chatre) |
| Status: | CLOSED INSUFFICIENT_DATA | ||
| Severity: | normal | CC: | linville, reinette.chatre, rjw |
| Priority: | P1 | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Kernel Version: | 2.6.32-rc3 | Subsystem: | |
| Regression: | Yes | Bisected commit-id: | |
| Bug Depends on: | |||
| Bug Blocks: | 14230 | ||
| Attachments: |
syslog which contains all information in the boot-hibernate-resume-reconnect cycle
dmesg after resume with kernel 2.6.32-rc3 turn off device when not used reset device before unload |
||
|
Description
Eddy Petrișor
2009-10-16 11:07:05 UTC
Logs from boot through hibernate and until after the failed post-resume connection would seem interesting...thanks! 0 eddy@heidi ~/tmp/iwlagn-hibernate-errors $ dmesg > before-hibernate.dmesg 0 eddy@heidi ~/tmp/iwlagn-hibernate-errors $ ping bugzilla.kernel.org -c 3 PING bugzilla.kernel.org (140.211.167.41) 56(84) bytes of data. 64 bytes from 140.211.167.41: icmp_seq=1 ttl=44 time=207 ms 64 bytes from 140.211.167.41: icmp_seq=2 ttl=44 time=205 ms 64 bytes from 140.211.167.41: icmp_seq=3 ttl=44 time=204 ms --- bugzilla.kernel.org ping statistics --- 3 packets transmitted, 3 received, 0% packet loss, time 10459ms rtt min/avg/max/mdev = 204.338/205.881/207.795/1.435 ms Hibernated and resumed... 0 eddy@heidi ~/tmp/iwlagn-hibernate-errors $ ping bugzilla.kernel.org -c 3 ping: unknown host bugzilla.kernel.org 2 eddy@heidi ~/tmp/iwlagn-hibernate-errors $ dmesg > after-hibernate.dmesg 0 eddy@heidi ~/tmp/iwlagn-hibernate-errors $ cp /var/log/syslog . 0 eddy@heidi ~/tmp/iwlagn-hibernate-errors $ su Password: heidi:/home/eddy/tmp/iwlagn-hibernate-errors# rmmod iwlagn heidi:/home/eddy/tmp/iwlagn-hibernate-errors# rmmod iwlcore heidi:/home/eddy/tmp/iwlagn-hibernate-errors# modprobe iwlagn heidi:/home/eddy/tmp/iwlagn-hibernate-errors# exit 0 eddy@heidi ~/tmp/iwlagn-hibernate-errors $ # waited a few seconds and wicd autoconnected 0 eddy@heidi ~/tmp/iwlagn-hibernate-errors $ ping bugzilla.kernel.org -c 3 PING bugzilla.kernel.org (140.211.167.41) 56(84) bytes of data. 64 bytes from 140.211.167.41: icmp_seq=1 ttl=44 time=210 ms 64 bytes from 140.211.167.41: icmp_seq=2 ttl=44 time=206 ms 64 bytes from 140.211.167.41: icmp_seq=3 ttl=44 time=205 ms --- bugzilla.kernel.org ping statistics --- 3 packets transmitted, 3 received, 0% packet loss, time 10459ms rtt min/avg/max/mdev = 205.875/207.652/210.682/2.153 ms Created attachment 23465 [details]
syslog which contains all information in the boot-hibernate-resume-reconnect cycle
Created attachment 23466 [details]
dmesg after resume with kernel 2.6.32-rc3
(In reply to comment #3) > Created an attachment (id=23465) [details] > syslog which contains all information in the boot-hibernate-resume-reconnect > cycle This syslog is obtained after the rmmod-modprobe cycle and contains only the information from this boot cycle. Created attachment 23535 [details]
turn off device when not used
Two patches to try, here is #1.
Created attachment 23536 [details]
reset device before unload
The two patches ensure that the device power is managed correctly during the unloading and interface up/down stages. Could you please apply them on top of linux-2.6 and test?
(In reply to comment #7) > Created an attachment (id=23536) [details] > reset device before unload > > The two patches ensure that the device power is managed correctly during the > unloading and interface up/down stages. Could you please apply them on top of > linux-2.6 and test? On top of which version? I have been using 2.6.32-rc3 to debug some other problems/regressions. And AFAIK is better to test on known stable ground, so I'll use 2.6.32-rc3 or a newer released version if the patches fail to apply cleanly. (In reply to comment #8) > (In reply to comment #7) > > Created an attachment (id=23536) [details] [details] > > reset device before unload > > > > The two patches ensure that the device power is managed correctly during > the > > unloading and interface up/down stages. Could you please apply them on top > of > > linux-2.6 and test? > > On top of which version? I tested that these patches apply on linux-2.6 - I am not sure which rc it was at the time. The driver code did not change in this area though so it should apply cleanly to whatever rc you are running. > I have been using 2.6.32-rc3 to debug some other > problems/regressions. And AFAIK is better to test on known stable ground, Right, that is why I based the patches on linux-2.6 > I'll use 2.6.32-rc3 or a newer released version if the patches fail to apply > cleanly. Thanks (In reply to comment #9) > (In reply to comment #8) > > (In reply to comment #7) > > > Created an attachment (id=23536) [details] [details] [details] > > > reset device before unload > > > > > > The two patches ensure that the device power is managed correctly during > the > > > unloading and interface up/down stages. Could you please apply them on > top of > > > linux-2.6 and test? > > > > On top of which version? > > I tested that these patches apply on linux-2.6 - I am not sure which rc it > was > at the time. The driver code did not change in this area though so it should > apply cleanly to whatever rc you are running. > > > I have been using 2.6.32-rc3 to debug some other > > problems/regressions. And AFAIK is better to test on known stable ground, > > > Right, that is why I based the patches on linux-2.6 > > > I'll use 2.6.32-rc3 or a newer released version if the patches fail to > apply > > cleanly. I applied the patches over 2.6.32-rc3 and the wireless didn't work properly after resume, I still had to 'rmmod iwlagn && rmmod iwlcore && modprobe iwlagn' to get networking. (In reply to comment #4) > Created an attachment (id=23466) [details] > dmesg after resume with kernel 2.6.32-rc3 This log shows that the AP denies the association during resume. I see it here: [ 475.091604] wlan0: direct probe to AP 00:1b:fc:45:33:70 (try 1) [ 475.094272] wlan0: direct probe responded [ 475.094282] wlan0: authenticate with AP 00:1b:fc:45:33:70 (try 1) [ 475.096205] wlan0: authenticated [ 475.096245] wlan0: associate with AP 00:1b:fc:45:33:70 (try 1) [ 475.098501] wlan0: RX AssocResp from 00:1b:fc:45:33:70 (capab=0x411 status=12 aid=0) [ 475.098508] wlan0: AP denied association (code=12) Is it possible to get some debug logs from wicd? Alternatively, is it possible to run your test with wpa_supplicant instead of wicd and obtain its full debug logs? I am a little busy these days. I've read the messages, but I don't have the time right now to do this. I'll give an answer next week. (In reply to comment #12) > I am a little busy these days. I've read the messages, but I don't have the > time right now to do this. I'll give an answer next week. It has been more than two months since you wrote the above - if this is still an issue, please do provide the information, if this is not an issue then please close the bug. We have been waiting more than two months now so resolving this with "insufficient data" - if this is still an issue please open a new bug and provide the logs. |