00:27 <stintel> DannyB: reverting all commits in drivers/net/wireless/broadcom/brcm80211/brcmfmac/ up to and including c8161a3ae79314781c1a31e0937f9ed4c26646c1 fixes the disconnect every minunte

00:27 <stintel> but SAE does not work

00:28 <stintel> so it's definitely two distinct problems

00:33 <DannyB> the SAE is due to the supplicants not handling all cases of partial SAE offload, we can just disable partial offload and that will fix that

00:38 <DannyB> wait a second, do you have ce49746a75918e4ed1d3f4792aefddaf496c7810 and 75846bce03328e59867589737f2d3ca097b735d7 in your tree?

00:38 <DannyB> if not, that's one of the disconnect issues

00:39 <DannyB> actually need

00:39 <DannyB> 5fafe88bd152a64d0bc9f5fae9b2adf17a716ab0

00:39 <DannyB> 75846bce03328e59867589737f2d3ca097b735d7

00:39 <DannyB> and

00:39 <DannyB> ce49746a75918e4ed1d3f4792aefddaf496c7810

00:41 <DannyB> or else the chaninfo_list won't work

00:43 <stintel> I've all 3

00:47 <DannyB> okay

00:48 <DannyB> when you get a chance, turn on BRCMFDBG and BRCM_TRACING, and then load the module with debug=1102854

00:48 <DannyB> that will produce a *lot* of dmesg output

00:48 <DannyB> but we need it

00:54 <DannyB> Reverting the chan_info_list pieces should change nothing, they just change how it gets the list of available channels. if that is changing something, something is broken in the firmware

00:55 <DannyB> I would try reverting 488b9c8d92d1b06dcc174dab2c89a4ea478345ac instead

00:55 <DannyB> but that's a guess

00:55 <DannyB> let's get the firmware losg when you get a chance on friday

00:56 <DannyB> and go from ther ewith what you have

01:02 <stintel> yeah reverting 488b9c8d92d1b06dcc174dab2c89a4ea478345ac didn't help the disconnect

01:02 <stintel> grabbing the firmware logs will not be for tonight anymore, my home automation just reminded me it's time to go to bed by switching of all the lights :P

01:12 <DannyB> You can apply the block i gave above that made SAE work for you, i'll add disabling that to the asahi tree for now.

01:14 <stintel> yeah, so I'm currently back at all commits reverted up to including c8161a3ae79314781c1a31e0937f9ed4c26646c1, that fixes the disconnect, and SAE, well not super important

01:35 <DannyB> okay, so give me firmware logs with debugging whne oyu get a chance and i'll fix it for real :)

01:38 <stintel> check, thanks!

01:38 <stintel> damn I need to go to bed. I need to make that reminder more annoying :D

14:18 <DannyB> stintel: before i forget - those logs will have your wifi password, so either redact it or send it to me privately (i'd be disbarred if I did something, so you don't have to worry about my side :P)

14:21 <stintel> DannyB: I'll do a multi-AP test setup first, the password will be something like changeme! :)

22:54 <DannyB> stintel: Great. I have a patch that may fix it all

22:55 <DannyB> and may make it worse, we'll see!