mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Bjorn Helgaas <helgaas@kernel.org>
To: Stefan Ghetu <stefan.ghetu@nxp.com>
Cc: linux-pci@vger.kernel.org, bhelgaas@google.com,
	linux-kernel@vger.kernel.org, hongxing.zhu@nxp.com,
	Jason Liu <jason.hui.liu@nxp.com>
Subject: Re: [PATCH] LF-18678 PCI: Add quirk to fix byte-swapped class code on ARA240 device
Date: Wed, 18 Nov 2026 10:03:41 -0600	[thread overview]
Message-ID: <20261118160341.GA583443@bhelgaas> (raw)
In-Reply-To: <20261118100406.600995-1-stefan.ghetu@nxp.com>

On Wed, Nov 18, 2026 at 12:04:06PM +0200, Stefan Ghetu wrote:
> From: Richard Zhu <hongxing.zhu@nxp.com>
> 
> The ARA240 PCIe endpoint device (1e58:0002) reports an incorrect,
> byte-swapped class code of 0x000012 instead of 0x120000 (Processing
> Accelerator). This causes the device to appear as an unclassified
> device (class 0x00) rather than a processing accelerator (class 0x12).
> 
> On ARM platforms, the kernel may refuse to assign BARs to unclassified
> devices, preventing the ARA240 from functioning properly. The device
> cannot be initialized and remains unusable.

Out of curiosity, what's the ARM connection here?  The question of
assigning resources for unclassified devices seems like it should be
generic, not ARM-specific.

If it *is* in fact generic, we should remove the ARM references.

> Add a PCI quirk to force the correct class code during early device
> enumeration, allowing proper resource assignment and device operation.
> 
> Signed-off-by: Richard Zhu <hongxing.zhu@nxp.com>
> Acked-by: Jason Liu <jason.hui.liu@nxp.com>
> Signed-off-by: Stefan Ghetu <stefan.ghetu@nxp.com>
> ---
>  drivers/pci/quirks.c | 12 ++++++++++++
>  1 file changed, 12 insertions(+)
> 
> diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
> index 3d92b99a49238..fc95d87d4af0a 100644
> --- a/drivers/pci/quirks.c
> +++ b/drivers/pci/quirks.c
> @@ -559,6 +559,18 @@ static void quirk_citrine(struct pci_dev *dev)
>  }
>  DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_IBM,	PCI_DEVICE_ID_IBM_CITRINE,	quirk_citrine);
>  
> +/*
> + * The ARA240 device reports a byte-swapped class code (0x000012) instead of
> + * the correct value (0x120000 - Processing Accelerator). This causes resource
> + * assignment issues on ARM platforms. Apply a quirk to force the correct
> + * class code.
> + */
> +static void quirk_ara240_class_code(struct pci_dev *dev)
> +{
> +	dev->class = 0x120000;
> +}
> +DECLARE_PCI_FIXUP_HEADER(0x1e58, 0x0002, quirk_ara240_class_code);
> +
>  /*
>   * This chip can cause bus lockups if config addresses above 0x600
>   * are read or written.
> -- 
> 2.53.0
> 

      parent reply	other threads:[~2026-11-18 16:03 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-18 10:04 Stefan Ghetu
2026-11-18 12:12 ` Ilpo Järvinen
2026-11-18 16:03 ` Bjorn Helgaas [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=20261118160341.GA583443@bhelgaas \
    --to=helgaas@kernel.org \
    --cc=bhelgaas@google.com \
    --cc=hongxing.zhu@nxp.com \
    --cc=jason.hui.liu@nxp.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=stefan.ghetu@nxp.com \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®