Fw: DisplayPort - IFD HACK CB:38770(BCM4331-802.11a.b.g.n;);

Wed Oct 14 17:54:32 BST 2026

Received: (at 1147573-done) by bugs.debian.org; 14 Sep 2026 07:32:27 +0000
>From sylvestredeb at ledru.info Mon Sep 14 07:32:27 2026
X-Spam-Checker-Version: SpamAssassin 4.0.1-bugs.debian.org_2005_01_02
        (2024-03-25) on buxtehude.debian.org
X-Spam-Level:
X-Spam-Status: No, score=-6.9 required=4.0 tests=BAYES_00,DKIM_SIGNED,
        DKIM_VALID,DKIM_VALID_AU,DKIM_VALID_EF,FOURLA,HAS_BUG_NUMBER,
        RCVD_IN_DNSWL_LOW,SPF_HELO_PASS,SPF_PASS autolearn=ham
        autolearn_force=no version=4.0.1-bugs.debian.org_2005_01_02
X-Spam-Bayes: score:0.0000 Tokens: new, 30; hammy, 104; neutral, 21; spammy,
        1. spammytokens:0.940-+--H*r:bugs.debian.org
        hammytokens:0.000-+--H*Ad:N*Matthias, 0.000-+--HFeedback-ID:user,
        0.000-+--HFeedback-ID:proton, 0.000-+--sk:pkg-llv, 0.000-+--sk:pkgllv
Return-path: <sylvestredeb at ledru.info>
Received: from mail-4327.protonmail.ch ([185.70.43.27]:55087)
        by buxtehude.debian.org with esmtps (TLS1.3:ECDHE_X25519__RSA_PSS_RSAE_SHA256__AES_256_GCM:256)
        (Exim 4.96)
        (envelope-from <sylvestredeb at ledru.info>)
        id 1x61Av-000dKP-1D
        for 1147573-done at bugs.debian.org;
        Mon, 14 Sep 2026 07:32:27 +0000
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ledru.info;
        s=protonmail2; t=1789371091; x=1789630291;
        bh=1Sd7NbZBWVoxL6uw6pOyjFfYObFAkQSuI7qL1wRy3KQ=;
        h=Date:To:From:Cc:Subject:Message-ID:In-Reply-To:References:
         Feedback-ID:From:To:Cc:Date:Subject:Reply-To:Feedback-ID:
         Message-ID:BIMI-Selector;
        b=LTuy/s1Quff7s2V/ETIrt7l4B+ctwhKP6WkqH2fTgwpT9OHG21xWlMQ92wrgNNNwA
         Uy5yIMg9bzk8KDERBRCA8p2CFTcYhnl3m/T3yBiwJQdQMovwFxkaJuNMLjtGS+umDG
         x+uQiamLcWMRWeNeK8psb1iEkr8U0qvuWy1IubqKgr1Ymtd42Bc+iaJBb9X2YNjsW2
         qcl/c31jmstNtVaFRyP0GfplwrZBb30wfSKYjovKmoEPr2rF5NdHMF+MWWGSto0Ch1
         sQ/SJP6AlaEIDNtRcSBnAUY+oFH6ZMFok+bY04Cqfz1KKugTr+gkG68Z7K+SxiZUAy
         22AZe8ic3jEfQ==
Date: Mon, 14 Sep 2026 07:31:26 +0000
To: Matthias Klose <doko at debian.org>
From: sylvestredeb at ledru.info
Cc: "1147573-done at bugs.debian.org" <1147573-done at bugs.debian.org>
Subject: Re : Bug#1147573: packaging of the fuzzer libraries is broken with LLVM 21, 22 and 23
Message-ID: <SdSwTre45FN_qR5VYK_1cNYiR1ePdxqq_bRbkr496BnrpJzioRlkfpG_q-UjZM_JQnHIcjDh6OgPBSmX_CYhfzmoiF41DyV3Y697X935W78=@ledru.info>
In-Reply-To: <1eabb575-b1f0-4324-ad8b-b06903b12919 at debian.org>
References: <22d47882-0523-4f5b-a43b-76f2e55a4d7c at debian.org> <1eabb575-b1f0-4324-ad8b-b06903b12919 at debian.org>
Feedback-ID: 110699482:user:proton
X-Pm-Message-ID: 10eed78087b8de25d477fde19e6ba86e4bcba3f6
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

________________________________________
From: Christian A Trinidad <ctmartinez5 at outlook.com>
Sent: Sunday, October 11, 2026 6:00 PM
To: walterav1984 at gmail.com
Cc: jxw5883 at runbox.io
Subject: Re: DisplayPort - IFD HACK CB:38770(autoport);

Broadcom BCM4331 802.11a/b/g/n (b43 *29*.fw firmware required

________________________________
From: Christian A Trinidad <ctmartinez5 at outlook.com>
Sent: Sunday, October 11, 2026 5:57 PM
To: jxw5883 at runbox.io <jxw5883 at runbox.io>
Subject: Re: DisplayPort - IFD HACK CB:38770(autoport);

src/mainboard/dell/precision_m4800/devicetree.cb

mb/dell: Add Dell Precision M4800 (Haswell)

Port of the Dell Precision M4800, a haswell mobile workstation with an
MXM type-A slot. The MXM slot works with any other MXM GPUs, including
GPUs this board does not boot on vendor firmware. Libgfxinit does not
appear to work, possibly due to some unusual graphics setup but a
workaround is to use EDK2 and add a Gop driver to the Tianocore build.
The EC is the SMSC MEC5055 which seems to be compatible with the
MEC5035 code. Based on autoport.

Working:
- USB EHCI debug (left side usb port closest to the touchpad)
- Keyboard
- Touchpad/trackpoint
- Ethernet
- Headphone port
- SD card reader
- mPCIe WiFi (With the MEC5035 radio enables commit)
- SeaBIOS 1.16.2
- edk2

mb/apple: Add Mac mini 6,1 (Late 2012) {Ivybridge / Thunderbolt 1}

Apple: Mac mini 6,1 (Late 2012)
Model No.: A1347 (820-3227-B)
EMC: 631 - 1993

Pure autoport (initial commit) doesn't boot, second patch brings up the
complete machine, mostly cherry picking unmerged other Apple coreboot
commits. This machine is very similar to a MacbookPro9,2 comparing logs
(autoport) and code wise see commit CB:88242<https://review.coreboot.org/88242> also awaiting review.

Flash instructions:
Only higher regions of the SPI are locked, a tiny 1MB coreboot ROM image
must be used for internal flashing (modified IFD, SeaBIOS, shrunken ME).
Use the extensive MacbookPro 8,x or 10,x two step IFD HACK CB:38770<https://review.coreboot.org/38770>
by Evgeny Zinoviev and adjust coreboot compilation allong the process to
create a full 8MB ROM image which can be flashed after running coreboot.
The used SPI ROM IC is a MXIC MX 25L6406E which flashrom doesn't detect
automatically, specifying using -c "MX25L6406E/MX25L6408E" is mandatory.
External in situ SOIC-8 SPI flash recovery verified using pomona probe
with a ch341a_spi (3,3v mod) while only using power of the programmer
itself and the AC powersupply disconnected!

Tested:
 - coreboot 26.09-14-ge0eec9336f54 as base
 - EDK2 (MrChromebox/26.09.2)
 - SeaBIOS 1.17
 - iPXE (2025.12)
 - libgfxinit textmode (SeaBIOS) / framebuffer (EDK2)
 - miniDP / miniDP++ (VGA) and HDMI available during POST, BOOT and OS
 - i5-3210M
 - RAM 2 slots DDR3 1.5v 1600MHz SODIMMs total 8GB
        Slot 0: 4GB DDR3-1600 - SK Hynix HMT351S6CFR8C-PB (2013-W12)
        Slot 2: 4GB DDR3-1600 - SK Hynix HMT351S6CFR8C-PB (2013-W12)
 - Fedora MATE 42 (Kernel 6.14)
 - KDE NEON 6.4 (Kernel 6.11)
 - Kubuntu 26.04.1 (Kernel 7)
 - MS Windows 10 / 11 (Install only limited testing)
 - Audio Input: Line In
 - Audio Outputs: DP++, HDMI, Headphone, Speaker (left channel only)
 - Audio SPDIF optical: In & Out
 - Built-in IR Receiver [USB] (Apple Remote volume keys work)
 - Broadcom BCM2046 Bluetooth [USB] (scans devices)
 - Broadcom NetXtreme BCM57766 Gigabit Ethernet PCIe Gb LAN
 - Broadcom BCM57765/57785 SDXC/MMC Card Reader [PCIe]
 - Broadcom BCM4331 802.11a/b/g/n (b43 *29*.fw firmware required)
 - Firewire 800 LSI Corporation FW643 (SanDisk SDDRX4-CF Reader)
 - SATA 2 ports
 - Thunderbolt 1 Controller DSL3510 [Cactus Ridge 4C 2012] (see note!)
 - USB2/3 all 4 ports
 - USB EHCI DEBUG (2nd port counted from left looking at backside)
 - SMSC 1428-7 fan control and temperature sensors
 - LED POWER (breath during suspend)
 - PowerButton (PowerOn/PowerOff/Wake)
 - Shutdown/Reboot/Suspend (wakes from USB / Powerbutton)
 - power_on_after_fail= Disable / Enable / Keep
 - Strip down the Intel ME/TXE firmware (make menuconfig)
 - flashrom -p internal -c "MX25L6406E/MX25L6408E" #read & write

Not tested:
 - VBIOS

Not working:
 - Audio Input: Microphone inside Headphone jack
 - Built-in Speaker (right channel missing)
 - show MEI device (its always hidden in lspci, intelmetool can find it)
 - Wake On LAN (neither Apple Thunderbolt Ethernet / Sonnet SEL X540-AT)

Fan note:
Works by default at ~1800 RPM very silent. Couldn't get it hot enough to
notice auto rampup of fan speed (tresholds to high?). However installing
"mbpfan" will static force a RPM or works by adjusting temp. thresholds.

Thunderbolt 1 note:
Enabling Thunderbolt will interfere with suspend, not only TB devices
but also iGPU may not always wake after sleep, therefore off by default.
In devicetree.cb turn on the "peg10" device and compile coreboot with
"Enable PCIe Hotplug Support" (only tested EDK2). Hotplug a Thunderbolt
device and authorize it with the OS Thunderbolt manager. This will make
it show in dmesg as a Thunderbolt device itself but sometimes not yet as
its PCIe function device that loads a driver listed in 'lspci' therefore
a 'echo 1 > /sys/bus/pci/rescan' may be needed for enabling each device!
Tested both Apple Thunderbolt to Ethernet / Firewire and Sonnet Echo
Express SEL PCIe 2.0 5GT/s case with Intel X540-AT Dual 10Gb ethernet.
DisplayPort also remains available through Thunderbolt daisy chaining.

VBT note:
The data.vbt blob was extracted using debugfs from Apple EFI firmware
v1.1 BIOS version 429.0.0.0.0 03/18/2022 enabling both miniDisplayPort
/ miniDisplayPort++ (VGA) and HDMI video outputs.

Change-Id: Ie5ed23dd07b4c7305fa90a685e511bd6b1e1e20e<https://review.coreboot.org/q/Ie5ed23dd07b4c7305fa90a685e511bd6b1e1e20e>
Signed-off-by: Walter Sonius <walterav1984 at gmail.com>

(MrChromebox's fork, uefipayload_202309);

- dGPU (NVIDIA Quadro K2100m, NVIDIA Quadro M2200m, NVIDIA GeForce GTX
1050)
- SATA DVD Drive
- SATA SSD
- mSATA
- DisplayPort
- Internal flashing
- Broadwell MRC
- Haswell MRC
- Dell PR03X Dock (Thanks to Jan Philipp Gross for testing)

Not working:
- S3 suspend: may be a similar issue to the E6430
- Libgfxinit
- Physical Wireless switch (does not enable/disable wifi, wifi must be
enabled with mec5035 command)
    - See commit 2f48eaf36f02a6607b4f94b6e8e48ec21b910e87<https://review.coreboot.org/q/commit:2f48eaf36f02a6607b4f94b6e8e48ec21b910e87>
      ("ec/dell/mec5035: Add command to control radio state")
- Brightness hotkeys
- Lid close suspend
- Battery reporting
- Speakers (Thanks to Jan Philipp Gross for testing)

Unknown/Untested:
- ExpressCard
- TPM

Change-Id: I49870b28be943dcb1932909f9d3ec4207cc11436<https://review.coreboot.org/q/I49870b28be943dcb1932909f9d3ec4207cc11436>
Signed-off-by: Justin Wu <jxw5883 at runbox.io>

________________________________
From: owner-freebsd-arch at FreeBSD.org <owner-freebsd-arch at FreeBSD.org> on behalf of Poul-Henning Kamp <phk at phk.freebsd.dk>
Sent: Monday, October 5, 2026 6:17 AM
To: Yuri Tkachenko <yura.tkachenko at gmail.com>
Cc: freebsd-arch at freebsd.org <freebsd-arch at freebsd.org>
Subject: Re: [RFC] Kernel-Enforced Jail Scoping, Lifecycle States, and Automated QoS/Cleanup

--------
Yuri Tkachenko writes:

> When containers crash or orchestrators abort unexpectedly, this userland-centric
> approach frequently leads to orphaned VNET interfaces, zombie ZFS datasets,
> and stale rctl rules.
> […]
> To solve this, I've put together an architectural RFC for a kernel-managed
> lifecycle framework. […]

My considered opinion is that this proposal suffers from the "everything is
better/easier/smarter in the kernel" misconception which is almost directly
the opposite of UNIX philosophy.

Yes, when things go wrong in userland, you can end up with debris,
which you, crucially, have to the tools to properly dispose.

If you tangle all this up in the kernel, you'd better hope nothing of the
sort ever happens, because you will have no other cure than a reboot.

Not a fan.

--
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.