Skip to contentSkip to conversation

ata: ahci: Add AMD 600 Series Chipset SATA Controller (0x43f6)

PATCH In review Ruslan Valiyev 2 participants 3 messages Mar 10 libata

lore
Share & tools
lore

Ruslan Valiyev

author
#0 lore raw

PATCH 1/1 ata: ahci: Add AMD 600 Series Chipset SATA Controller (0x43f6)

Add PCI ID 0x43f6 for the AMD 600 Series Chipset (B850/X870) SATA controller to ahci_pci_tbl. Without this entry, the AHCI driver does not bind to the controller and SATA devices are invisible at boot.

Users can work around the issue with:
echo "1022 43f6" > /sys/bus/pci/drivers/ahci/new_id

Closes https://bugzilla.kernel.org/show_bug.cgi?id=221193
Signed-off-by Ruslan Valiyev <[email protected]>
---
 drivers/ata/ahci.c | 1 +
 1 file changed, 1 insertion(+)
drivers/ata/ahci.c +1 -0
index 931d0081169b9..a4935f9b80789 100644
@@ -466,6 +466,7 @@ static const struct pci_device_id ahci_pci_tbl[] = {
 	{ PCI_VDEVICE(AMD, 0x7801), board_ahci_no_debounce_delay }, /* AMD Hudson-2 (AHCI mode) */
 	{ PCI_VDEVICE(AMD, 0x7900), board_ahci }, /* AMD CZ */
 	{ PCI_VDEVICE(AMD, 0x7901), board_ahci }, /* AMD Green Sardine */
+	{ PCI_VDEVICE(AMD, 0x43f6), board_ahci }, /* AMD 600 Series Chipset */
 	/* AMD is using RAID class only for ahci controllers */
 	{ PCI_VENDOR_ID_AMD, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
 	  PCI_CLASS_STORAGE_RAID << 8, 0xffffff, board_ahci },
Signature · 1 line
2.43.0
maintainer
#1 lore raw

Hello Ruslan,

Quoted 18 lines · Ruslan Valiyev

Add PCI ID 0x43f6 for the AMD 600 Series Chipset (B850/X870) SATA controller to ahci_pci_tbl. Without this entry, the AHCI driver does not bind to the controller and SATA devices are invisible at boot.

Users can work around the issue with:
echo "1022 43f6" > /sys/bus/pci/drivers/ahci/new_id

Closes https://bugzilla.kernel.org/show_bug.cgi?id=221193
Signed-off-by Ruslan Valiyev <[email protected]>
---
 drivers/ata/ahci.c | 1 +
 1 file changed, 1 insertion(+)
drivers/ata/ahci.c +0 -0
index 931d0081169b9..a4935f9b80789 100644

drivers/ata/ahci.c static const struct pci_device_id ahci_pci_tbl[] = { · line 472

Quoted 7 lines
drivers/ata/ahci.c +1 -0
@@ -466,6 +466,7 @@ static const struct pci_device_id ahci_pci_tbl[] = {
 	{ PCI_VDEVICE(AMD, 0x7801), board_ahci_no_debounce_delay }, /* AMD Hudson-2 (AHCI mode) */
 	{ PCI_VDEVICE(AMD, 0x7900), board_ahci }, /* AMD CZ */
 	{ PCI_VDEVICE(AMD, 0x7901), board_ahci }, /* AMD Green Sardine */
+	{ PCI_VDEVICE(AMD, 0x43f6), board_ahci }, /* AMD 600 Series Chipset */
 	/* AMD is using RAID class only for ahci controllers */
 	{ PCI_VENDOR_ID_AMD, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
 	  PCI_CLASS_STORAGE_RAID << 8, 0xffffff, board_ahci },

The AHCI driver already has these two entries:

        /* AMD is using RAID class only for ahci controllers */
        { PCI_VENDOR_ID_AMD, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
          PCI_CLASS_STORAGE_RAID << 8, 0xffffff, board_ahci },

and

        /* Generic, PCI class code for AHCI */
        { PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
          PCI_CLASS_STORAGE_SATA_AHCI, 0xffffff, board_ahci },

If the ahci driver fails to bind, that means that the class code for this AMD AHCI contorller is
not PCI_CLASS_STORAGE_RAID and is not PCI_CLASS_STORAGE_SATA_AHCI.

I'm honestly curious... what is the PCI class code of this controller then? :)

My AMD AHCI controller is:

$ lspci -v | grep AHCI
6a:00.0 SATA controller: Advanced Micro Devices, Inc. [AMD] 600 Series Chipset SATA Controller (rev 01) (prog-if 01 [AHCI 1.0])
$ sudo lspci -nnvs 6a:00.0
6a:00.0 SATA controller [0106]: Advanced Micro Devices, Inc. [AMD] 600 Series Chipset SATA Controller [1022:43f6] (rev 01) (prog-if 01 [AHCI 1.0])
        Subsystem: ASMedia Technology Inc. Device [1b21:1062]
        Flags: bus master, fast devsel, latency 0, IRQ 48, IOMMU group 20
        Memory at 81080000 (32-bit, non-prefetchable) [size=1K]
        Expansion ROM at 81000000 [disabled] [size=512K]
        Capabilities: [50] MSI: Enable+ Count=1/1 Maskable- 64bit+
        Capabilities: [70] Power Management version 3
        Capabilities: [80] Express Legacy Endpoint, IntMsgNum 0
        Capabilities: [100] Advanced Error Reporting
        Kernel driver in use: ahci
include/linux/pci_ids.h:#define PCI_CLASS_STORAGE_SCSI          0x0100
include/linux/pci_ids.h:#define PCI_CLASS_STORAGE_IDE           0x0101
include/linux/pci_ids.h:#define PCI_CLASS_STORAGE_FLOPPY        0x0102
include/linux/pci_ids.h:#define PCI_CLASS_STORAGE_IPI           0x0103
include/linux/pci_ids.h:#define PCI_CLASS_STORAGE_RAID          0x0104
include/linux/pci_ids.h:#define PCI_CLASS_STORAGE_SATA          0x0106
include/linux/pci_ids.h:#define PCI_CLASS_STORAGE_SATA_AHCI     0x010601
include/linux/pci_ids.h:#define PCI_CLASS_STORAGE_SAS           0x0107
include/linux/pci_ids.h:#define PCI_CLASS_STORAGE_EXPRESS       0x010802
include/linux/pci_ids.h:#define PCI_CLASS_STORAGE_OTHER         0x0180

Kind regards,
Niklas

author
#2 lore raw
Quoted 4 lines · Niklas Cassel

If the ahci driver fails to bind, that means that the class code for this AMD AHCI controller is not PCI_CLASS_STORAGE_RAID and is not PCI_CLASS_STORAGE_SATA_AHCI. I am honestly curious... what is the PCI class code of this controller then? :)

That is exactly the right question, and I should have asked it before sending this patch.

I based this on a bugzilla report where the user confirmed the new_id workaround works, but I did not ask for lspci -nnvs output first. I do not have the class code from the affected system, which means I cannot justify the patch as-is.

I will go back to the reporter and get the full lspci output. If the prog-if differs from yours (e.g. 0x00 instead of 0x01), I will resubmit with proper justification. If not, I will drop it.

Sorry for the noise.

Ruslan