regressions.lists.linux.dev archive mirror
 help / color / mirror / Atom feed
From: Vegard Nossum <vegard.nossum@oracle.com>
To: Alexander Wetzel <alexander@wetzel-home.de>,
	Thomas Mann <rauchwolke@gmx.net>,
	linux-wireless@vger.kernel.org
Cc: Johannes Berg <johannes.berg@intel.com>,
	regressions@leemhuis.info, regressions@lists.linux.dev
Subject: iwlwifi frequent drops between v6.2-rc3 and v6.3-rc1
Date: Fri, 17 Mar 2023 09:53:55 +0100	[thread overview]
Message-ID: <9d11ed29-1114-055d-5b26-0899a5fc0d7f@oracle.com> (raw)


Hi,

Somewhere between e8f60cd7db24 (v6.2-rc3) and 2eb29d59ddf0 (v6.3-rc1) my
wifi started dropping a lot. I'm using iwlwifi and I've checked the old
logs that the firmware hasn't changed:

Feb 25 16:22:33 kernel: iwlwifi 0000:04:00.0: loaded firmware version 
36.e91976c0.0 8000C-36.ucode op_mode iwlmvm

My logs look like this:

09:01:50 kernel: wlp4s0: Connection to AP [REDACTED]:0b:25 lost
09:01:50 wpa_supplicant[1279]: wlp4s0: CTRL-EVENT-DISCONNECTED 
bssid=[REDACTED]:0b:25 reason=4 locally_generated=1
09:01:50 NetworkManager[8368]: <warn>  [1679040110.6548] 
sup-iface[0x5628f2495990,wlp4s0]: connection disconnected (reason -4)
09:01:50 NetworkManager[8368]: <info>  [1679040110.6656] device 
(wlp4s0): supplicant interface state: completed -> disconnected
09:01:50 NetworkManager[8368]: <info>  [1679040110.7609] device 
(wlp4s0): supplicant interface state: disconnected -> scanning
09:01:53 wpa_supplicant[1279]: wlp4s0: SME: Trying to authenticate with 
[REDACTED]:0b:25 (SSID='[REDACTED]' freq=5500 MHz)
09:01:53 kernel: wlp4s0: authenticate with [REDACTED]:0b:25
09:01:53 kernel: wlp4s0: send auth to [REDACTED]:0b:25 (try 1/3)
09:01:53 kernel: wlp4s0: authenticated
09:01:53 wpa_supplicant[1279]: wlp4s0: Trying to associate with 
[REDACTED]:0b:25 (SSID='[REDACTED]' freq=5500 MHz)
09:01:53 kernel: wlp4s0: associate with [REDACTED]:0b:25 (try 1/3)
09:01:53 kernel: wlp4s0: RX AssocResp from [REDACTED]:0b:25 
(capab=0x1511 status=0 aid=36)
09:01:53 NetworkManager[8368]: <info>  [1679040113.4553] device 
(wlp4s0): supplicant interface state: scanning -> authenticating
09:01:53 kernel: wlp4s0: associated
09:01:53 wpa_supplicant[1279]: wlp4s0: Associated with [REDACTED]:0b:25
09:01:53 wpa_supplicant[1279]: wlp4s0: CTRL-EVENT-SUBNET-STATUS-UPDATE 
status=0
09:01:53 NetworkManager[8368]: <info>  [1679040113.4645] device 
(wlp4s0): supplicant interface state: authenticating -> associating
09:01:53 NetworkManager[8368]: <info>  [1679040113.4829] device 
(wlp4s0): supplicant interface state: associating -> associated
09:01:53 NetworkManager[8368]: <info>  [1679040113.5175] device 
(wlp4s0): supplicant interface state: associated -> 4-way handshake
09:01:53 kernel: wlp4s0: Connection to AP [REDACTED]:0b:25 lost
09:01:53 wpa_supplicant[1279]: wlp4s0: CTRL-EVENT-DISCONNECTED 
bssid=[REDACTED]:0b:25 reason=4 locally_generated=1
09:01:53 wpa_supplicant[1279]: wlp4s0: WPA: 4-Way Handshake failed - 
pre-shared key may be incorrect
09:01:53 wpa_supplicant[1279]: wlp4s0: CTRL-EVENT-SSID-TEMP-DISABLED 
id=0 ssid="[REDACTED]" auth_failures=1 duration=10 reason=WRONG_KEY
09:01:53 NetworkManager[8368]: <warn>  [1679040113.5685] 
sup-iface[0x5628f2495990,wlp4s0]: connection disconnected (reason -4)
09:01:53 NetworkManager[8368]: <info>  [1679040113.5737] device 
(wlp4s0): supplicant interface state: 4-way handshake -> disconnected
09:01:53 NetworkManager[8368]: <info>  [1679040113.6759] device 
(wlp4s0): supplicant interface state: disconnected -> scanning

I did see that somebody else reported a similar regression, but I don't
know if it could be the same problem or not (I see the driver is
different, but the fix looks generic):

https://linux-regtracking.leemhuis.info/regzbot/regression/217119/

The buggy commit that was identified there does seem to be within the
range of potential culprits:

$ git log --oneline e8f60cd7db24..2eb29d59ddf0 | grep resumption
4444bc2116ae wifi: mac80211: Proper mark iTXQs for resumption

If people think it's the same, I could try the proposed fix -- otherwise
let me know what else I can do to help track this down.

I don't think bisecting this is feasible for me since there are hours
between the drops. But it's really noticeable over the stretch of
multiple days that something has changed -- looking at the logs, the
disconnect rate has basically tripled.

Thanks,


Vegard

             reply	other threads:[~2023-03-17  8:54 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-03-17  8:53 Vegard Nossum [this message]
2023-03-17 15:29 ` iwlwifi frequent drops between v6.2-rc3 and v6.3-rc1 Alexander Wetzel
2023-03-25 20:33   ` Alexander Wetzel
2023-04-12 11:38     ` Linux regression tracking (Thorsten Leemhuis)
2023-04-16 12:40       ` Linux regression tracking #update (Thorsten Leemhuis)
2023-03-22 13:59 ` Linux regression tracking #adding (Thorsten Leemhuis)

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=9d11ed29-1114-055d-5b26-0899a5fc0d7f@oracle.com \
    --to=vegard.nossum@oracle.com \
    --cc=alexander@wetzel-home.de \
    --cc=johannes.berg@intel.com \
    --cc=linux-wireless@vger.kernel.org \
    --cc=rauchwolke@gmx.net \
    --cc=regressions@leemhuis.info \
    --cc=regressions@lists.linux.dev \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).