Re: How is testing going?

From: Bjoern A. Zeeb <bz_at_freebsd.org>
Date: Tue, 31 Mar 2026 15:55:55 UTC
On Tue, 31 Mar 2026, David Wolfskill wrote:

> On Fri, Mar 20, 2026 at 11:04:22PM +0000, Bjoern A. Zeeb wrote:
>> Hi,
>>
>> to those who have updated.  Did you encounter further problems than reported,
>> any instabilities, etc. now that it's been a few more days?
>> ....
>
> I didn't *think* I had anything (useful) to report; it's quite possible
> that what follows is not actually useful... but it reflects something I
> was not expecting, so there's that.
>
> TL;DR: My main laptop currently has an Intel "Centrino Ultimate-N 6300"
> card, using the iwn(4) driver.  I track both stable/15 and head on it,
> daily.  I have had no notable issues using wireless network access from
> the laptop at home or at work (though it is (very!) rare that the
> laptop is running head while I am at work).
>
> Yesterday, I traveled to a hotel (we are having a Gathering for work);
> no issues with wireless at the hotel while running stable/15, but (when
> I rebooted to head this morning for my morning updates), a total
> inability to associate.
>
> Some details:
> Card:
> iwn0@pci0:3:0:0:        class=0x028000 rev=0x35 hdr=0x00 vendor=0x8086 device=0x422b subvendor=0x8086 subdevice=0x1121
>    vendor     = 'Intel Corporation'
>    device     = 'Centrino Ultimate-N 6300'
>    class      = network
>
> stable/15: yesterday: stable/15-n282731-dafac65367c9
>               today: stable/15-n282761-3d85a2a7a8a9
>
> head:      yesterday: main-n284770-e220af9cee74
>               today: main-n284796-33e095ee097d

wpa in main and stab;e/15 is the same

sys/net80211 has two differences:  one is Adrian's crypto comments and abstractions as well as some time difference check fixes, the other one is Gleb's BPF work.

sys/dev/iwn has some extra work from Adrian for sequence number offload.


If you are still there and in the same situation, disable HT and try a
11a or 11g association only and see if that makes a difference.

Otherwise I have no clue.

-- 
Bjoern A. Zeeb                                                     r15:7