[PATCH v2] PCI: Add func1 DMA quirk for Marvell 88SE9125 SATA controller

[PATCH v2] PCI: Add func1 DMA quirk for Marvell 88SE9125 SATA controller

[Yifeng Li]

Like other SATA controller chips in the Marvell 88SE91xx series, the
Marvell 88SE9125 has the same DMA requester ID hardware bug that prevents
it from working under IOMMU. This patch adds its device ID 0x9125 to the
Function 1 DMA alias quirk list.

This patch should not be confused with an earlier patch, commit 059983790a4c
("PCI: Add function 1 DMA alias quirk for Marvell 9215 SATA controller"),
which applies to a different chip with a similar model number, 88SE9215.

Without this patch, device initialization fails with DMA errors.

    ata8: softreset failed (1st FIS failed)
    DMAR: DRHD: handling fault status reg 2
    DMAR: [DMA Write NO_PASID] Request device [03:00.1] fault addr 0xfffc0000 [fault reason 0x02] Present bit in context entry is clear
    DMAR: DRHD: handling fault status reg 2
    DMAR: [DMA Read NO_PASID] Request device [03:00.1] fault addr 0xfffc0000 [fault reason 0x02] Present bit in context entry is clear

After applying the patch, the controller can be successfully initialized.

    ata8: SATA link up 1.5 Gbps (SStatus 113 SControl 330)
    ata8.00: ATAPI: PIONEER BD-RW   BDR-207M, 1.21, max UDMA/100
    ata8.00: configured for UDMA/100
    scsi 7:0:0:0: CD-ROM            PIONEER  BD-RW   BDR-207M 1.21 PQ: 0 ANSI: 5

Cc: stable@vger.kernel.org
Reported-by: sbingner <sam@bingner.com>
Tested-by: sbingner <sam@bingner.com>
Tested-by: Yifeng Li <tomli@tomli.me>
Signed-off-by: Yifeng Li <tomli@tomli.me>
---

Notes:
    I accidentally sent an earlier version of the commit without CCing
    stable@vger.kernel.org. The mail itself was also rejected by
    many servers due to a DKIM issue. Thus [PATCH v2], sorry for the
    noise.

 drivers/pci/quirks.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index 003950c73..20a932690 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -4103,6 +4103,9 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_MARVELL_EXT, 0x9120,
 			 quirk_dma_func1_alias);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_MARVELL_EXT, 0x9123,
 			 quirk_dma_func1_alias);
+/* https://bugzilla.kernel.org/show_bug.cgi?id=42679#c136 */
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_MARVELL_EXT, 0x9125,
+			 quirk_dma_func1_alias);
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_MARVELL_EXT, 0x9128,
 			 quirk_dma_func1_alias);
 /* https://bugzilla.kernel.org/show_bug.cgi?id=42679#c14 */
-- 
2.31.1

Re: [PATCH v2] PCI: Add func1 DMA quirk for Marvell 88SE9125 SATA controller

[Krzysztof Wilczyński]

Hi,

Thank you for sending the patch over to fix this!

> Like other SATA controller chips in the Marvell 88SE91xx series, the
> Marvell 88SE9125 has the same DMA requester ID hardware bug that prevents
> it from working under IOMMU. This patch adds its device ID 0x9125 to the
> Function 1 DMA alias quirk list.

[...]
> Reported-by: sbingner <sam@bingner.com>
> Tested-by: sbingner <sam@bingner.com>

Both of these tags would require a proper full name, if possible, rather
than a name that is abbreviated and/or a username.

Reviewed-by: Krzysztof Wilczyński <kw@liunx.com>

	Krzysztof

Re: [PATCH v2] PCI: Add func1 DMA quirk for Marvell 88SE9125 SATA controller

[Tom Li]

On Tue, Nov 16, 2021 at 04:49:54PM +0100, Krzysztof Wilczyński wrote:
> > Reported-by: sbingner <sam@bingner.com>
> > Tested-by: sbingner <sam@bingner.com>
> 
> Both of these tags would require a proper full name, if possible, rather
> than a name that is abbreviated and/or a username.
> 
> Reviewed-by: Krzysztof Wilczyński <kw@liunx.com>
> 
> 	Krzysztof

No problem, I'll send a revision to correct the tags immediately.

Cheers,
Yifeng Li

Re: [PATCH v2] PCI: Add func1 DMA quirk for Marvell 88SE9125 SATA controller

[Krzysztof Wilczyński]

Hi,

> On Tue, Nov 16, 2021 at 04:49:54PM +0100, Krzysztof Wilczyński wrote:
> > > Reported-by: sbingner <sam@bingner.com>
> > > Tested-by: sbingner <sam@bingner.com>
> > 
> > Both of these tags would require a proper full name, if possible, rather
> > than a name that is abbreviated and/or a username.
> > 
> > Reviewed-by: Krzysztof Wilczyński <kw@liunx.com>
> > 
> > 	Krzysztof
> 
> No problem, I'll send a revision to correct the tags immediately.

No worries!  Thank you!

	Krzysztof

Re: [PATCH v2] PCI: Add func1 DMA quirk for Marvell 88SE9125 SATA controller

[Christoph Hellwig]

On Tue, Nov 16, 2021 at 02:32:26PM +0000, Yifeng Li wrote:
> Like other SATA controller chips in the Marvell 88SE91xx series, the
> Marvell 88SE9125 has the same DMA requester ID hardware bug that prevents
> it from working under IOMMU. This patch adds its device ID 0x9125 to the
> Function 1 DMA alias quirk list.

Btw, do we need to prevent vfio assignment for all devices with this
quirk?

Re: [PATCH v2] PCI: Add func1 DMA quirk for Marvell 88SE9125 SATA controller

[Alex Williamson]

On Wed, 17 Nov 2021 02:12:04 -0800
Christoph Hellwig <hch@infradead.org> wrote:

> On Tue, Nov 16, 2021 at 02:32:26PM +0000, Yifeng Li wrote:
> > Like other SATA controller chips in the Marvell 88SE91xx series, the
> > Marvell 88SE9125 has the same DMA requester ID hardware bug that prevents
> > it from working under IOMMU. This patch adds its device ID 0x9125 to the
> > Function 1 DMA alias quirk list.  
> 
> Btw, do we need to prevent vfio assignment for all devices with this
> quirk?

No, the alias is taken into account with grouping and IOMMU
programming, it should work with vfio.  Thanks,

Alex