All of lore.kernel.org
 help / color / mirror / Atom feed
From: Mika Westerberg <mika.westerberg@linux.intel.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: andreas.noever@gmail.com, michael.jamet@intel.com,
	westeri@kernel.org, YehezkelShB@gmail.com,
	linux-usb@vger.kernel.org
Subject: Re: Thunderbolt display chaining
Date: Tue, 22 Apr 2025 08:05:01 +0300	[thread overview]
Message-ID: <20250422050501.GF88033@black.fi.intel.com> (raw)
In-Reply-To: <20250421163549.GA40412@noisy.programming.kicks-ass.net>

Hi,

On Mon, Apr 21, 2025 at 06:35:49PM +0200, Peter Zijlstra wrote:
> Hi guys,
> 
> I got two Dell U2725QE screens, that can do 4k@120Hz. I'm trying to
> daisy chain them on thunderbolt 4.
> 
> Using a windows laptop (Thinkpad X1) it all 'just' works, and both
> screens are running 4k@60Hz, all good.
> 
> Using a Linux laptop (Thinkpad X13 Gen2i -- tigerlake) it sorta works.
> It drives the first at 4k@120Hz and the second at 4k@30Hz, which is
> kinda terrible. Trying to manually set the first screen to 60Hz works,
> but doesn't appear to enable me to then manually set the second screen
> to 60, it's stuck at 30 :-(
> 
> Using some Tigerlake NUC, I only get a single screen at 4k120Hz and
> nothing at all on the second screen.
> 
> I tried having a quick look at tb.c, and it has this bandwidth
> calculation in tb_available_bandwidth(), but I'm not quite sure where to
> start poking.

If this is Tigerlake then that file is never used - DP tunneling is all
firmware. So what you have is:

  TGL <-> M <-> M

The USB4/TB link is 40G with 10% reserved so you get 36G for DP for the
first link (since both go through that one). 4k @ 120Hz takes 25.82G which
leaves 36 - 28.82 = ~7G for the second monitor that's exactly the 4k at
30Hz you see.

Bandwidth allocation mode can help here but it is not supported until MTL
and I suspect this is what the Windows laptop is using.

You can remedy this by not daisy chaining if you have multiple ports:

  TGL <-> M
   ^
   |
   v
   M

Let me know if you need more information.

  reply	other threads:[~2025-04-22  5:05 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-04-21 16:35 Thunderbolt display chaining Peter Zijlstra
2025-04-22  5:05 ` Mika Westerberg [this message]
2025-04-22  9:12   ` Peter Zijlstra

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=20250422050501.GF88033@black.fi.intel.com \
    --to=mika.westerberg@linux.intel.com \
    --cc=YehezkelShB@gmail.com \
    --cc=andreas.noever@gmail.com \
    --cc=linux-usb@vger.kernel.org \
    --cc=michael.jamet@intel.com \
    --cc=peterz@infradead.org \
    --cc=westeri@kernel.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.