Re: USB4/TBT3 support

Hi Tomoaki,

1st of all, thanks a lot for sharing your notes and insights helping to
enrich the discussion

On Sun, Dec 22, 2024 at 7:25 PM Tomoaki AOKI <junchoon@dec.sakura.ne.jp>
wrote:

> On Fri, 20 Dec 2024 17:04:20 +0100
> Mohammad Noureldin <mohammad@thelightbird.com> wrote:
>

<snip>


>
> Hi.
> Thanks for the pointer.
>
> Some notes:
>   *hselasky@ (RIP, Hans...) suffered from the lack of official and open
>    hardware specs for implementing his driver.
>
>   *hselasky@ wasn't going to implement TB3's security features
>    as it seemed NOT to be sufficient for actual security in spec itself.
>
>   *I have Titan Ridge TB3 controller (falles back to Intel Cannon Lake
>    USB 3.1 controller when TB3 device is not connected) in my ThinkPad
>    P52, but hselasky@'s last driver didn't work for it, as his driver
>    only supports older generation of controller chipset (with the lack
>    of information).
>
>   *P52 recognizes TB3-only device (Samsung X5 external SSD I've
>    purchased for test) as internal PCIe drive if it was attached
>    before powerint on the computer, but hot deplugging causes panic,
>    with and without hselasky@'s driver installed.
>
>   *My related info are in Bug 237666 [1].
>
> [1] https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=237666


Believe or not, I was in touch with HPS (may he RIP) over this specific PR
given my general interest in USB and I've even purchased 3 2nd hand
machines to help testing and get to the root cause:
- Dell Precision Tower 5810
-# Which proved not to be useful for this case
- Lenovo ThinkPad P52s
- Lenovo ThinkPad T480
-# Because Bjoern Zeeb (bz@) could reproduce the original reported issue on
this machine and he wrote down some notes about his findings in [1]

But that EuroBSDCon of year 2023 when I was planning to bring one or both
of these Lenovos with me to tinker on them with HPS, I cam to know that he
passed away before the EuroBSDCon's starting date ... RIP HPS.

I've added the *"Problem Reports to be Possibly Revisited"* section [2]
because in addition to [3] I've noticed other similar PR(s) where we can
revisit after a working driver is in place to either:
- Indicate that now this works
- Or, it still not working, we won't support it, and write this down in a
clear way for end users

Also to hopefully make the scope more clear, I've added a Scope sub-section
[4]


>
> --
> Tomoaki AOKI    <junchoon@dec.sakura.ne.jp>
>

Thanks again Tomoaki and happy holidays 👊

[1] https://wiki.freebsd.org/BjoernZeeb/USB
[2]
https://wiki.freebsd.org/MohammadNoureldin/FreeBSDUSB4TBT3Support#Problem_Reports_To_Be_Possibly_Revisited
[3] https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=237666
[4] https://wiki.freebsd.org/MohammadNoureldin/FreeBSDUSB4TBT3Support#Scope

-- 
Thanks
- Mohammad Noureldin
--
"Life is like riding a bicycle. To keep your balance you must keep moving"
- Albert Einstein