public inbox for iwd@lists.linux.dev
 help / color / mirror / Atom feed
From: Arend van Spriel <arend.vanspriel@broadcom.com>
To: Denis Kenzior <denkenz@gmail.com>,
	Martin Petzold <martin.petzold@tavla.de>
Cc: "iwd@lists.linux.dev" <iwd@lists.linux.dev>
Subject: Re: IWD 1.27 with brcmfmac not working for roaming
Date: Wed, 16 Oct 2024 10:32:39 +0200	[thread overview]
Message-ID: <6b9d7715-4e2c-444e-91fe-2820fd33c41c@broadcom.com> (raw)
In-Reply-To: <1761198e-9f41-4e5a-b2b9-a1652732346d@gmail.com>



On 10/16/2024 4:04 AM, Denis Kenzior wrote:
> Hi Arend,
> 
>>>
>>> One may question why -67 AP was preferred over -29 one?  We'd need 
>>> logs to answer that question.  It could be the -29 one was simply not 
>>> seen in the scan results at the time of the initial scan / connection.
>>
>> The roaming algorithm is not that intuitive. Roaming and rate 
>> selection are not based on signal strength alone. They may look at PER 
>> (packet error ratio) to decide. -29 dBm might actually be too strong. 
>> I always try to get between -40 and -60 dBm.
> 
> I was mostly thinking of the initial BSS selection which iwd can 
> influence at initial CMD_CONNECT.  iwd sorts APs according to highest 
> expected throughput, but that generally means that if 2 APs are 
> identical, the one with better RSSI wins.  Still, without logs there's 
> no way of knowing why the lower rssi one was picked.

Right. The firmware is probably release version so it will only have 
error prints and that kinda details are not showing. If brcmfmac is 
compiled with CONFIG_BRCMDBG you can enable firmware console logging 
(FWCON) with the 'debug' module parameter [1].

> I'm curious: what are the downsides to using RSSI above -40 dBm?

ear damage ;-) Not entirely joking though. Too strong signal can affect 
reception as well. However, I am not sure at what level those issues 
would show. So my -40 dBm is not backed by technical data. Maybe my 
co-workers doing phy stuff have a better explanation.

>>
>> The nl80211 API actually offer the possibility to affect the BSS 
>> selection. The NL80211_CMD_CONNECT command can have the attribute 
>> NL80211_ATTR_BSS_SELECT for that [2]. When not provided the firmware 
>> will obviously use its default behavior whatever that is. Not sure if 
>> IWD or wpa_supplicant support this, but I am fairly sure brcmfmac 
>> supports it. However I do not know if it also applies to roaming.
> 
> iwd does not.  Doesn't look like wpa_s does either.

Should have done that when I added the nl80211 api change ;-)

Regards,
Arend

[1] 
https://elixir.bootlin.com/linux/v6.11.3/source/drivers/net/wireless/broadcom/brcm80211/brcmfmac/debug.h

  reply	other threads:[~2024-10-16  8:32 UTC|newest]

Thread overview: 68+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-09  9:43 IWD 1.27 with brcmfmac not working for roaming Martin Petzold
2024-10-09 16:07 ` Denis Kenzior
2024-10-09 16:50   ` Arend Van Spriel
2024-10-09 17:54     ` Martin Petzold
2024-10-10  8:06     ` Martin Petzold
2024-10-12 11:06       ` Martin Petzold
2024-10-12 11:51         ` Arend van Spriel
2024-10-13 15:43           ` Martin Petzold
2024-10-30 19:19           ` Martin Petzold
2024-10-30 19:23             ` James Prestwood
2024-10-09 16:58   ` Arend Van Spriel
2024-10-10 13:20   ` Martin Petzold
2024-10-10 13:36     ` James Prestwood
2024-10-10 13:47       ` Martin Petzold
2024-10-10 13:55         ` James Prestwood
2024-10-11  8:35           ` Martin Petzold
2024-10-11 10:46             ` Martin Petzold
2024-10-12 10:59               ` Martin Petzold
2024-10-15 14:43   ` Martin Petzold
2024-10-15 15:17     ` Denis Kenzior
2024-10-15 19:13       ` Arend van Spriel
2024-10-16  2:04         ` Denis Kenzior
2024-10-16  8:32           ` Arend van Spriel [this message]
2024-10-17 10:58           ` Martin Petzold
2024-10-19 14:04             ` Martin Petzold
2024-10-19 14:41               ` Denis Kenzior
2024-10-21 13:34                 ` Martin Petzold
2024-10-21 14:40                   ` Arend Van Spriel
2024-10-21 14:53                     ` Martin Petzold
2024-10-21 15:23                     ` James Prestwood
2024-10-21 17:08                       ` Arend Van Spriel
2024-10-21 17:20                         ` Martin Petzold
2024-10-21 17:40                         ` Denis Kenzior
2024-10-21 18:26                           ` Arend van Spriel
2024-10-21 18:45                             ` Martin Petzold
2024-10-21 18:48                               ` Martin Petzold
2024-10-21 18:55                             ` Denis Kenzior
2024-10-21 19:08                               ` Jeremy Blum
2024-10-22 15:26                                 ` Denis Kenzior
2024-10-22 16:38                                   ` Jeremy Blum
2024-10-21 19:15                               ` Martin Petzold
2024-10-21 19:11                             ` James Prestwood
2024-10-21 20:23                             ` Martin Petzold
2024-10-22  6:08                               ` Arend Van Spriel
2024-10-21 22:01                             ` KeithG
2024-10-21 22:10                               ` Martin Petzold
2024-10-22 17:40                                 ` James Prestwood
2024-10-22 18:04                                   ` Martin Petzold
2024-10-22 18:21                                     ` Martin Petzold
2024-10-22 18:24                                     ` James Prestwood
2024-10-22 18:32                                       ` Martin Petzold
2024-10-22 18:44                                       ` Denis Kenzior
2024-10-22 18:47                                         ` Martin Petzold
2024-10-22 19:10                                           ` James Prestwood
2024-10-22 18:47                                         ` James Prestwood
2024-10-22 18:49                                           ` Martin Petzold
2024-10-22 18:52                                           ` Denis Kenzior
2024-10-23 12:02                             ` Martin Petzold
2024-10-23 12:13                               ` James Prestwood
2024-10-23 12:19                                 ` Martin Petzold
2024-10-23 13:22                                   ` James Prestwood
2024-10-23 13:34                                     ` Martin Petzold
2024-10-23 15:22                                     ` Martin Petzold
2024-10-23 15:27                                       ` Martin Petzold
2024-10-23 15:30                                         ` James Prestwood
2024-10-23 15:37                                           ` Martin Petzold
2024-10-23 15:28                                       ` James Prestwood
2024-10-23 15:11                               ` Arend Van Spriel

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=6b9d7715-4e2c-444e-91fe-2820fd33c41c@broadcom.com \
    --to=arend.vanspriel@broadcom.com \
    --cc=denkenz@gmail.com \
    --cc=iwd@lists.linux.dev \
    --cc=martin.petzold@tavla.de \
    /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