All of lore.kernel.org
 help / color / mirror / Atom feed
From: Mika Westerberg <mika.westerberg@linux.intel.com>
To: Grant Morphett <grant@gmorph.com>
Cc: Andreas Noever <andreas.noever@gmail.com>,
	Mika Westerberg <westeri@kernel.org>,
	Yehezkel Bernat <YehezkelShB@gmail.com>,
	linux-usb@vger.kernel.org
Subject: Re: [PATCH v2] thunderbolt: Enable hidden DP OUT adapter on Dell TB16 dock
Date: Mon, 9 Nov 2026 06:27:17 +0100	[thread overview]
Message-ID: <20261109052717.GH176724@black.igk.intel.com> (raw)
In-Reply-To: <20261106062614.27709-1-grant@gmorph.com>

Hi,

On Fri, Nov 06, 2026 at 05:26:14PM +1100, Grant Morphett wrote:
> The Dell TB16 Thunderbolt 3 dock consists of two Alpine Ridge 4C routers
> in a chain: one in the captive cable module (DROM 00d4:b051 "Dell
> Thunderbolt Cable") and one in the dock itself (DROM 00d4:b054 "Dell
> Thunderbolt Dock"). The dock has four video connectors: mini DisplayPort,
> HDMI, full size DisplayPort and VGA.
> 
> Mini DisplayPort and HDMI sit behind an MST hub on DP OUT adapter 8 of
> the dock router and work fine. Full size DisplayPort and VGA are wired
> to DP OUT adapter 8 of the cable router, but the cable router's DROM
> marks that adapter as disabled. The software connection manager
> therefore never initializes it, never tunnels DisplayPort to it, and a
> monitor on the full size DisplayPort or VGA connector stays dark. Only
> two external displays can be used with the dock, while all of them work
> with the firmware connection manager (and under Windows).
> 
> The adapter is real: its adapter config space reports type DP/HDMI OUT
> (0xe0102), it raises hot plug events when a monitor is attached to the
> DisplayPort or VGA connector, and once it is enabled the DP tunnel to it
> is established and the display comes up. The same DROM also fails its
> CRC32 check (expected 0xaf438340, got 0xaf4383c0), so the port
> disabled bit is likely just wrong rather than intentional:
> 
>   1: DROM data CRC32 mismatch (expected: 0xaf438340, got: 0xaf4383c0), continuing
>   1:8: disabled by eeprom
>   acking hot plug event on 1:8
> 
> Add a quirk for the cable router that re-enables adapter 8 if its
> config space confirms it is a DP OUT adapter. It runs in a new
> tb_check_drom_quirks() call right after the DROM has been parsed, so
> the regular port initialisation in tb_switch_add() picks the adapter up
> like any other enabled port.
> 
> With the quirk applied, both DP tunnels come up at boot and all three
> external monitors work:
> 
>   1:8: enabling DP OUT adapter disabled in DROM
>   1:8: DP OUT available
>   0:5 <-> 1:8 (DP): activating
>   301:8: DP OUT available
>   0:6 <-> 301:8 (DP): activating
> 
> Tested on a Lenovo ThinkPad X1 Yoga Gen 8 (Raptor Lake, USB4 host
> 8087:a76d) with a TB16 at dock NVM 27.0 / cable NVM 16.0, driving three
> 1920x1080 monitors on mini DisplayPort, HDMI and full size DisplayPort.
> 
> v2:
> - Add a new tb_check_drom_quirks() call right after DROM parsing, so
>   the quirk corrects the wrong DROM entry before the ports are
>   initialized, instead of making tb_init_port() non-static and calling
>   it from the quirk (Mika Westerberg).

This should go under the '---'

> 
> Assisted-by: LLM

This is not needed.

> Signed-off-by: Grant Morphett <grant@gmorph.com>
> ---
> ---
>  drivers/thunderbolt/quirks.c | 62 ++++++++++++++++++++++++++++++++++++
>  drivers/thunderbolt/switch.c |  1 +
>  drivers/thunderbolt/tb.h     |  1 +
>  3 files changed, 64 insertions(+)
> 
> diff --git a/drivers/thunderbolt/quirks.c b/drivers/thunderbolt/quirks.c
> index 2722918..bba10df 100644
> --- a/drivers/thunderbolt/quirks.c
> +++ b/drivers/thunderbolt/quirks.c
> @@ -65,6 +65,34 @@ static void quirk_stuck_pending(struct tb_switch *sw)
>  	}
>  }
>  
> +/*
> + * The DROM of the Dell TB16 cable module marks adapter 8 as disabled
> + * even though it is the DP OUT adapter wired to the dock's full size
> + * DisplayPort and VGA connectors. Correct the DROM entry so that the
> + * software connection manager initializes the adapter and can tunnel
> + * DisplayPort to it.
> + */
> +static void quirk_dell_tb16_dp_out(struct tb_switch *sw)
> +{
> +	struct tb_port *port;
> +	u32 type;
> +
> +	if (sw->config.max_port_number < 8)
> +		return;
> +
> +	port = &sw->ports[8];
> +	if (!port->disabled)
> +		return;
> +
> +	if (tb_port_read(port, &type, TB_CFG_PORT, 2, 1))
> +		return;
> +	if ((type & 0xffffff) != TB_TYPE_DP_HDMI_OUT)
> +		return;
> +
> +	port->disabled = false;
> +	tb_port_info(port, "enabling DP OUT adapter disabled in DROM\n");
> +}
> +
>  struct tb_quirk {
>  	u16 hw_vendor_id;
>  	u16 hw_device_id;
> @@ -163,3 +191,37 @@ void tb_check_quirks(struct tb_switch *sw)
>  		q->hook(sw);
>  	}
>  }
> +
> +static const struct tb_quirk tb_drom_quirks[] = {
> +	/* Dell TB16 cable module DROM hides the second DP OUT adapter */
> +	{ 0x8086, PCI_DEVICE_ID_INTEL_ALPINE_RIDGE_4C_BRIDGE, 0x00d4, 0xb051,
> +		  quirk_dell_tb16_dp_out },
> +};
> +
> +/**
> + * tb_check_drom_quirks() - Check for quirks to apply after DROM read
> + * @sw: Thunderbolt switch
> + *
> + * Apply any quirks that correct wrong DROM entries before the ports are
> + * initialized.
> + */
> +void tb_check_drom_quirks(struct tb_switch *sw)
> +{
> +	int i;
> +
> +	for (i = 0; i < ARRAY_SIZE(tb_drom_quirks); i++) {
> +		const struct tb_quirk *q = &tb_drom_quirks[i];
> +
> +		if (q->hw_vendor_id && q->hw_vendor_id != sw->config.vendor_id)
> +			continue;
> +		if (q->hw_device_id && q->hw_device_id != sw->config.device_id)
> +			continue;
> +		if (q->vendor && q->vendor != sw->vendor)
> +			continue;
> +		if (q->device && q->device != sw->device)
> +			continue;
> +
> +		tb_sw_dbg(sw, "running %ps\n", q->hook);
> +		q->hook(sw);

Can you make the matching as separate function (tb_apply_quirks()) that is
given the quirk table and then make this and the tb_check_quirks() call
that one? This avoids the duplication.

> +	}
> +}
> diff --git a/drivers/thunderbolt/switch.c b/drivers/thunderbolt/switch.c
> index d22d8db..6268ed6 100644
> --- a/drivers/thunderbolt/switch.c
> +++ b/drivers/thunderbolt/switch.c
> @@ -3343,6 +3343,7 @@ int tb_switch_add(struct tb_switch *sw)
>  		ret = tb_drom_read(sw);
>  		if (ret)
>  			dev_warn(&sw->dev, "reading DROM failed: %d\n", ret);
> +		tb_check_drom_quirks(sw);
>  		tb_sw_dbg(sw, "uid: %#llx\n", sw->uid);
>  
>  		ret = tb_switch_set_uuid(sw);
> diff --git a/drivers/thunderbolt/tb.h b/drivers/thunderbolt/tb.h
> index e868457..7d04376 100644
> --- a/drivers/thunderbolt/tb.h
> +++ b/drivers/thunderbolt/tb.h
> @@ -1510,6 +1510,7 @@ static inline bool usb4_port_device_is_offline(const struct usb4_port *usb4)
>  }
>  
>  void tb_check_quirks(struct tb_switch *sw);
> +void tb_check_drom_quirks(struct tb_switch *sw);
>  
>  #ifdef CONFIG_ACPI
>  bool tb_acpi_add_links(struct tb_nhi *nhi);
> -- 
> 2.53.0

      parent reply	other threads:[~2026-11-09  5:27 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-01 21:23 [PATCH] thunderbolt: Enable hidden DP OUT adapter on Dell TB16 dock Grant Morphett
2026-11-01 21:29 ` sashiko-bot
2026-11-04 12:00 ` Mika Westerberg
2026-11-06  6:26 ` [PATCH v2] " Grant Morphett
2026-11-06  6:33   ` sashiko-bot
2026-11-09  5:27   ` Mika Westerberg [this message]

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=20261109052717.GH176724@black.igk.intel.com \
    --to=mika.westerberg@linux.intel.com \
    --cc=YehezkelShB@gmail.com \
    --cc=andreas.noever@gmail.com \
    --cc=grant@gmorph.com \
    --cc=linux-usb@vger.kernel.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.