Bug 1903 - Error in dmesg after switch txpower
: Error in dmesg after switch txpower
Status: VERIFIED FIXED
: iwlwifi
power management
: iwlwifi-2.6 development tree
: All Cards Fedora 10
: P1 enhancement
Assigned To:
:
:
:
:
:
  Show dependency treegraph
 
Reported: 2009-02-10 23:30 by
Modified: 2009-06-16 19:45 (History)


Attachments
Patch to try (520 bytes, patch)
2009-05-21 09:57, Abhijeet Kolekar
Details | Diff


Note

You need to log in before you can comment on or make changes to this bug.


Description From 2009-02-10 23:30:51
Test Environment
===============================================
Platform        :       Intel SDV M3M31/M3X15/M3M41
Wireless Card   :       Intel(R) WiFi Link 4965/3945/5100/5300
OS              :       Redhat Fedora release 10 (Cambridge) 32/64bit
AP              :       Cisco 1250
uCode           :       iwlwifi-4965-2.ucode 228.57.2.23
                        iwlwifi-3945-2.ucode 15.28.2.8
                        iwlwifi-5000-1.ucode 5.4.1.16
Source          :       commit bc51dd744dc3ce75c4aab96681d6bfa322cb184e
                        2.6.29-rc3 (3945 card)

issues: when set txpower off/on, log has Error message. As we know it is
expected message. But in view of users, Error in dmesg will *confuse them*.
suggest closing relevant order and don't send that signal to show error
message.

Steps to Reproduce:
1. reload driver
2. ifconfig wlan0 up
3. iwconfig wlan0 channel <CHAN> essid <ESSID>
4. ifconfig wlan0 <IP_ADDR>
5. ping <IP_SERVER>
6. iwconfig wlan0 txpower off
7. sleep 20
8. iwconfig wlan0 txpower on---->Error in dmesg is below:

iwlagn 0000:03:00.0: Error sending REPLY_ADD_STA: enqueue_hcmd failed: -5
iwlagn 0000:02:00.0: Error sending REPLY_ADD_STA: time out after 500ms.
iwl3945: Error sending REPLY_LEDS_CMD: iwl3945_enqueue_hcmd failed: -5
iwl3945: Error sending REPLY_TX_PWR_TABLE_CMD: iwl3945_enqueue_hcmd failed: -5
iwl3945: Error sending REPLY_LEDS_CMD: iwl3945_enqueue_hcmd failed: -5
------- Comment #1 From 2009-04-23 02:21:09 -------
detect "iwlagn 0000:01:00.0: Error sending REPLY_CARD_STATE_CMD: time out after
500ms." on 2.6.30-rc2-wl with 4965 card.
commit 9b20f0642b71e36378ba7e53263a18d0d40193d2 (Tue Apr 21 10:06:52 2009)
------- Comment #2 From 2009-05-20 13:46:34 -------
Are you still getting errors for REPLY_ADD_STA.
I think for txpower off you should get the error for REPLY_LEDS_CMD only which
is
actually fine because of SW rfkill.
If you are getting error only for LEDS then pease close this bug.
------- Comment #3 From 2009-05-20 19:32:40 -------
(In reply to comment #2)
> Are you still getting errors for REPLY_ADD_STA.
> I think for txpower off you should get the error for REPLY_LEDS_CMD only which
> is
> actually fine because of SW rfkill.
> If you are getting error only for LEDS then pease close this bug.

Yeah, Only REPLY_LEDS_CMD is always printed on 3945 card. It might not be a
issue, so we set the severity enhancement. Could you tune it and make error
disappear in log even though "SW rfkill"?
------- Comment #4 From 2009-05-21 09:57:08 -------
Created an attachment (id=2007) [details]
Patch to try
------- Comment #5 From 2009-05-21 09:57:42 -------
I have created one patch for you.But this is not the right thing to do.
I don't think we will see this in upstream.
------- Comment #6 From 2009-05-21 13:50:41 -------
One more reason it will only show in 3945 is that we have leds command for 3945
while rest of the chips have register setting. And this is expected behavior
for  3945.
------- Comment #7 From 2009-05-25 22:18:47 -------
I retest it on 5100 and n-band, after switching txpower, dmesg prints "Error
sending REPLY_ADD_STA" and "HW problem - can not stop rx aggregation for tid
0".

Testing Environment
==============================================
Platform        :       Intel SDV M3M41 
Wireless Card   :       Intel(R) WiFi Link 5100
OS              :       Redhat Fedora release 10 (Cambridge) 32bit
AP              :       Cisco 1250
uCode           :       iwlwifi-5000-2.ucode 8.24.2.12
Source          :       commit c612430a75805c0fb5d2b0339653567afa111ab6
                     (2.6.30-rc6-wl)
------- Comment #8 From 2009-06-16 16:32:03 -------
With the new rfkill system we should not be asked to do work when rfkill is
enabled. This should thus be fixed now.
------- Comment #9 From 2009-06-16 19:45:23 -------
Yes. I didn't see it anymore.