I just needed some extra SATA ports in my home server — and thought it would be easy to solve by adding a DawiControl DC-614e RAID controller card which comes with the Marvell 88SE9215 chipset. Marvellous 🐐, you might think... Turns out, I was wrong!

I’m not using the RAID functionality of the card — I only needed the four extra SATA ports 🤷‍♂️. Unfortunately, the card wasn’t properly recognized out of the box by Linux... 🙄 — Bad penguin! 🐧 Verry baaad, bad penguin! 👎🏻

My best Google searches quickly yielded results. This issue is described in detail in the following blog post: Arbeitsnotizen zu: Debian und Marvell 88SE9215. This excellent site turns out to be by Johannes Keßler (thanks for sharing! — and caring, Johannes 😉), and I found it through a post on good old Reddit.

The author provides a solid solution by applying a kernel patch that adds the necessary PCI ID to the ahci driver. That article is based on kernel version 6.0.8.

You’d expect this to be fixed by now — especially since I’m running Proxmox VE with kernel version 6.8.12 — but... "NEIN! (Du Schwalbe!)" the problem still persists.

Since I didn’t want to bother with compiling a custom kernel, I found a different solution:
I resolved the issue by injecting the PCI ID into the initramfs, which allows the ahci driver to pick up the card automatically during boot 🤓.

In this article, I’ll walk you through how I made that work 👨🏻‍🏫.

1. Lookup details 🔭
Using lspci, I confirmed that the card is visible on the PCI bus:

lspci -nn -kk
[..]
03:00.0 RAID bus controller [0104]: Marvell Technology Group Ltd. 88SE9215 PCIe 2.0 x1 4-port SATA 6 Gb/s Controller [1b4b:9215] (rev 01)
        Subsystem: Dawicontrol GmbH Device [dc93:614e]
        Kernel driver in use: ahci
        Kernel modules: ahci

1b4b → the vendor ID of the manufacturer. In this case, 1b4b stands for Marvell Technology Group Ltd.
9215 → the device ID that identifies the specific model/chipset. Here, that’s the 88SE9215, the SATA controller chip on the card.

2. Ensure AHCI is available during boot 👢
We must ensure that the AHCI driver is available in the initramfs at boot. This can be done by adding ahci to the /etc/initramfs-tools/modules file.

The following command checks whether ahci is already listed in /etc/initramfs-tools/modules. If not, it appends it:

grep '^ahci$' /etc/initramfs-tools/modules || echo ahci >> /etc/initramfs-tools/modules

It should look something like this (just listing ahci):

# List of modules that you want to include in your initramfs.
# They will be loaded at boot time in the order below.
#
# Syntax:  module_name [args ...]
#
# You must run update-initramfs(8) to effect this change.
#
# Examples:
#
# raid1
# sd_mod
ahci

Update the initramfs to include the ahci driver:

update-initramfs -u

3. Init Script 🚀
Ensure the init-top folder exists...

mkdir -p /etc/initramfs-tools/scripts/init-top

...and create a custom init-top script

nano /etc/initramfs-tools/scripts/init-top/dawicontrol

With the following contents:

#!/bin/sh

PREREQ="modules"

prereqs() {
    echo "$PREREQ"
}

case "$1" in
    prereqs)
        prereqs
        exit 0
        ;;
esac

echo "### DawiControl init-top script starting ###"

PCI_ID="1b4b 9215"
TARGET="/sys/bus/pci/drivers/ahci/new_id"

# Wait up to 10 seconds until the target exists and is writable
WAIT=0
while [ ! -w "$TARGET" ] && [ $WAIT -lt 10 ]; do
    echo "Waiting for $TARGET to be writable... ($WAIT)"
    sleep 1
    WAIT=$((WAIT + 1))
done

if [ ! -w "$TARGET" ]; then
    echo "ERROR: $TARGET is not writable after waiting"
    exit 1
fi

# Check if the PCI ID is already registered
if grep -q "$PCI_ID" "$TARGET" 2>/dev/null; then
    echo "PCI ID $PCI_ID already registered"
    exit 0
fi

# Inject the PCI ID into the AHCI driver
echo "$PCI_ID" > "$TARGET" || {
    echo "Failed to inject PCI ID $PCI_ID"
    exit 1
}

echo "PCI ID $PCI_ID successfully injected"

exit 0

Make the script executable...

chmod +x /etc/initramfs-tools/scripts/init-top/dawicontrol

...and rebuild initramfs

update-initramfs -u

After a reboot, the card was correctly initialized and all connected drives became available.

Update 6th of August 2025 – I have performed an in‑place upgrade of my Proxmox version 8 installation to version 9, and my fix survived the upgrade. This means the fix works for both version 8 and version 9!