[Bug 855365] New: After suspend/restart wpa_supplicant uses 100 percent of CPU
https://bugzilla.novell.com/show_bug.cgi?id=855365 https://bugzilla.novell.com/show_bug.cgi?id=855365#c0 Summary: After suspend/restart wpa_supplicant uses 100 percent of CPU Classification: openSUSE Product: openSUSE 13.1 Version: Final Platform: x86-64 OS/Version: openSUSE 13.1 Status: NEW Severity: Normal Priority: P5 - None Component: Network AssignedTo: bnc-team-screening@forge.provo.novell.com ReportedBy: dan@spacemandan.net QAContact: qa-bugs@suse.de Found By: --- Blocker: --- User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/31.0.1650.57 Safari/537.36 If I suspend to RAM, then power back up, wpa_supplicant immediately starts to use 100 percent of one CPU. wpa_supplicant does seem to be working in that state, because NetworkManager can still connect to a wireless network. It also seems to be using the CPU in nice mode, but it still uses far too much power and heats up the computer eventually. If I kill wpa_supplicant, it restarts immediately and still at runs at 100%. Sending SIGSTOP pauses it correctly, but then wireless networking doesn't work right. Restarting the network doesn't help, but after a complete reboot, the program runs properly again until the next suspend. card: Netgear WG111v3 USB driver module: rtl8187 Reproducible: Always Steps to Reproduce: 1. Connect to wireless network with NetworkManager 2. Suspend to RAM 3. Wake computer Actual Results: wpa_supplicant uses 100% of CPU. Expected Results: wpa_supplicant uses no processor power except when needed. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=855365
https://bugzilla.novell.com/show_bug.cgi?id=855365#c
zhang jiajun
https://bugzilla.novell.com/show_bug.cgi?id=855365
https://bugzilla.novell.com/show_bug.cgi?id=855365#c1
--- Comment #1 from Dan Gerhards
https://bugzilla.novell.com/show_bug.cgi?id=855365
https://bugzilla.novell.com/show_bug.cgi?id=855365#c2
Libor Pecháček
https://bugzilla.novell.com/show_bug.cgi?id=855365
https://bugzilla.novell.com/show_bug.cgi?id=855365#c3
Gary Ching-Pang Lin
https://bugzilla.novell.com/show_bug.cgi?id=855365
https://bugzilla.novell.com/show_bug.cgi?id=855365#c4
--- Comment #4 from Dan Gerhards
https://bugzilla.novell.com/show_bug.cgi?id=855365
https://bugzilla.novell.com/show_bug.cgi?id=855365#c5
--- Comment #5 from Dan Gerhards
2. Kill wpa_supplicant Killing it doesn't work. It restarts immediately. However, I have been sending
https://bugzilla.novell.com/show_bug.cgi?id=855365
https://bugzilla.novell.com/show_bug.cgi?id=855365#c6
--- Comment #6 from Dan Gerhards
4. Attach /var/log/wpa_supplicant.log if wpa_supplicant occupies CPU. The whole file was very large, so the file I attached just includes the part after the resume. (Yes, wpa_supplicant occupied the CPU.)
Please also notice if there are any strange message in dmesg after resume. I attached the output because it might be strange.
-- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=855365
https://bugzilla.novell.com/show_bug.cgi?id=855365#c7
--- Comment #7 from Gary Ching-Pang Lin
2. Kill wpa_supplicant Killing it doesn't work. It restarts immediately. However, I have been sending the TERM signal. I did try to send KILL, and when it respawned, the supplicant took over the CPU just like when I do a suspend/resume! That gives me an easier way to reproduce the bug, and maybe it gives you some more information.
So you can reproduce the bug by sending KILL to wpa_supplicant? This might be a clue.
4. Attach /var/log/wpa_supplicant.log if wpa_supplicant occupies CPU. The whole file was very large, so the file I attached just includes the part after the resume. (Yes, wpa_supplicant occupied the CPU.)
I found the following messages: Could not set interface wlan0 flags (UP): Input/output error wlan0: Failed to initialize driver interface - try next driver wrapper There is another wifi device wlan1 showed in the log, and this looks strange. Do you have two wifi cards in the machine? Please attach the output of "iw dev", "lspci", and "lsusb".
Please also notice if there are any strange message in dmesg after resume. I attached the output because it might be strange. ieee80211 phy0: rt61pci_wait_bbp_ready: Error - BBP register access failed, aborting ieee80211 phy0: rt61pci_set_device_state: Error - Device failed to enter state 4 (-5)
I suspect this is a driver issue. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=855365
https://bugzilla.novell.com/show_bug.cgi?id=855365#c8
--- Comment #8 from Dan Gerhards
Do you have two wifi cards in the machine? I did until 10 minutes ago. It's a desktop machine, and the internal NIC couldn't get a good signal from where it sits, so I put a USB NIC at the end of a 10-foot cable. That setup worked fine until I upgraded to 13.1.
Based on your comment, I tried taking the internal NIC out, and the bug disappeared! I wish someone had suggested that *seven months* ago. It never occurred to me that a driver could cause a problem like that, and I didn't know where to look for messages. Anyway, this is solved as far as I'm concerned. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=855365
https://bugzilla.novell.com/show_bug.cgi?id=855365#c9
Gary Ching-Pang Lin
participants (1)
-
bugzilla_noreply@novell.com