ahci: asm1064: correct count of reported ports

PATCH Applied Niklas Cassel 2 participants 5 messages Feb 14, 2024 – Feb 19, 2024 libata

lore

Share & tools

lore

Niklas Cassel

author maintainer

2024-02-14 16:58 UTC #0 lore raw

PATCH v2 1/1 ahci: asm1064: correct count of reported ports

From: "Andrey Jr. Melnikov" <[email protected]>

The ASM1064 SATA host controller always reports wrongly, that it has 24 ports. But in reality, it only has four ports.

before:
ahci 0000:04:00.0: SSS flag set, parallel bus scan disabled
ahci 0000:04:00.0: AHCI 0001.0301 32 slots 24 ports 6 Gbps 0xffff0f impl SATA mode ahci 0000:04:00.0: flags: 64bit ncq sntf stag pm led only pio sxs deso sadm sds apst

after:
ahci 0000:04:00.0: ASM1064 has only four ports
ahci 0000:04:00.0: forcing port_map 0xffff0f -> 0xf
ahci 0000:04:00.0: SSS flag set, parallel bus scan disabled
ahci 0000:04:00.0: AHCI 0001.0301 32 slots 24 ports 6 Gbps 0xf impl SATA mode ahci 0000:04:00.0: flags: 64bit ncq sntf stag pm led only pio sxs deso sadm sds apst

Signed-off-by Andrey Jr. Melnikov <[email protected]>

Signed-off-by Niklas Cassel <[email protected]>

---
Changes since V1: use switch case.

 drivers/ata/ahci.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

drivers/ata/ahci.c +11 -3

index da2e74fce2d9..682ff550ccfb 100644
@@ -671,9 +671,17 @@ MODULE_PARM_DESC(mobile_lpm_policy, "Default LPM policy for mobile chipsets");
 static void ahci_pci_save_initial_config(struct pci_dev *pdev,
 					 struct ahci_host_priv *hpriv)
 {
-	if (pdev->vendor == PCI_VENDOR_ID_ASMEDIA && pdev->device == 0x1166) {
-		dev_info(&pdev->dev, "ASM1166 has only six ports\n");
-		hpriv->saved_port_map = 0x3f;
+	if (pdev->vendor == PCI_VENDOR_ID_ASMEDIA) {
+		switch (pdev->device) {
+		case 0x1166:
+			dev_info(&pdev->dev, "ASM1166 has only six ports\n");
+			hpriv->saved_port_map = 0x3f;
+			break;
+		case 0x1064:
+			dev_info(&pdev->dev, "ASM1064 has only four ports\n");
+			hpriv->saved_port_map = 0xf;
+			break;
+		}
 	}
 
 	if (pdev->vendor == PCI_VENDOR_ID_JMICRON && pdev->device == 0x2361) {

Signature · 1 line

2.43.0

Andrey Melnikov

2024-02-14 17:24 UTC #1 lore raw

ср, 14 февр. 2024 г. в 19:58, Niklas Cassel <[email protected]>:

Quoted 18 lines

From: "Andrey Jr. Melnikov" <[email protected]>
The ASM1064 SATA host controller always reports wrongly, that it has 24 ports. But in reality, it only has four ports.
before:
ahci 0000:04:00.0: SSS flag set, parallel bus scan disabled
ahci 0000:04:00.0: AHCI 0001.0301 32 slots 24 ports 6 Gbps 0xffff0f impl SATA mode ahci 0000:04:00.0: flags: 64bit ncq sntf stag pm led only pio sxs deso sadm sds apst
after:
ahci 0000:04:00.0: ASM1064 has only four ports
ahci 0000:04:00.0: forcing port_map 0xffff0f -> 0xf
ahci 0000:04:00.0: SSS flag set, parallel bus scan disabled
ahci 0000:04:00.0: AHCI 0001.0301 32 slots 24 ports 6 Gbps 0xf impl SATA mode ahci 0000:04:00.0: flags: 64bit ncq sntf stag pm led only pio sxs deso sadm sds apst

Quoted 4 lines

Signed-off-by Andrey Jr. Melnikov <[email protected]>
Signed-off-by Niklas Cassel <[email protected]>
---
Changes since V1: use switch case.

I think we should stop abusing saved_port_map and convert all quirk's to use mask_port_map.

Quoted 31 lines

 drivers/ata/ahci.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

drivers/ata/ahci.c +11 -4

index da2e74fce2d9..682ff550ccfb 100644
@@ -671,9 +671,17 @@ MODULE_PARM_DESC(mobile_lpm_policy, "Default LPM policy for mobile chipsets");
 static void ahci_pci_save_initial_config(struct pci_dev *pdev,
                                         struct ahci_host_priv *hpriv)
 {
-       if (pdev->vendor == PCI_VENDOR_ID_ASMEDIA && pdev->device == 0x1166) {
-               dev_info(&pdev->dev, "ASM1166 has only six ports\n");
-               hpriv->saved_port_map = 0x3f;
+       if (pdev->vendor == PCI_VENDOR_ID_ASMEDIA) {
+               switch (pdev->device) {
+               case 0x1166:
+                       dev_info(&pdev->dev, "ASM1166 has only six ports\n");
+                       hpriv->saved_port_map = 0x3f;
+                       break;
+               case 0x1064:
+                       dev_info(&pdev->dev, "ASM1064 has only four ports\n");
+                       hpriv->saved_port_map = 0xf;
+                       break;
+               }
        }
 
        if (pdev->vendor == PCI_VENDOR_ID_JMICRON && pdev->device == 0x2361) {
--

2.43.0

Niklas Cassel

author maintainer

2024-02-14 18:13 UTC #2 lore raw

Hello Andrey,

Quoted 19 lines · Andrey Melnikov

ср, 14 февр. 2024 г. в 19:58, Niklas Cassel <[email protected]>:
Quoted 18 lines
From: "Andrey Jr. Melnikov" <[email protected]>
The ASM1064 SATA host controller always reports wrongly, that it has 24 ports. But in reality, it only has four ports.
before:
ahci 0000:04:00.0: SSS flag set, parallel bus scan disabled
ahci 0000:04:00.0: AHCI 0001.0301 32 slots 24 ports 6 Gbps 0xffff0f impl SATA mode ahci 0000:04:00.0: flags: 64bit ncq sntf stag pm led only pio sxs deso sadm sds apst
after:
ahci 0000:04:00.0: ASM1064 has only four ports
ahci 0000:04:00.0: forcing port_map 0xffff0f -> 0xf
ahci 0000:04:00.0: SSS flag set, parallel bus scan disabled
ahci 0000:04:00.0: AHCI 0001.0301 32 slots 24 ports 6 Gbps 0xf impl SATA mode ahci 0000:04:00.0: flags: 64bit ncq sntf stag pm led only pio sxs deso sadm sds apst

Quoted 7 lines

Quoted 4 lines
Signed-off-by Andrey Jr. Melnikov <[email protected]>
Signed-off-by Niklas Cassel <[email protected]>
---
Changes since V1: use switch case.
I think we should stop abusing saved_port_map and convert all quirk's to use mask_port_map.

That sounds like a great idea to me.
Do you plan to send some patches?

If so, I could hold off this patch until I get something from you.

Kind regards,
Niklas

Niklas Cassel

author maintainer Applied

2024-02-19 09:30 UTC #3 lore raw

In reply to Niklas Cassel

Quoted 17 lines · Niklas Cassel

From: "Andrey Jr. Melnikov" <[email protected]>
The ASM1064 SATA host controller always reports wrongly, that it has 24 ports. But in reality, it only has four ports.
before:
ahci 0000:04:00.0: SSS flag set, parallel bus scan disabled
ahci 0000:04:00.0: AHCI 0001.0301 32 slots 24 ports 6 Gbps 0xffff0f impl SATA mode ahci 0000:04:00.0: flags: 64bit ncq sntf stag pm led only pio sxs deso sadm sds apst
after:
ahci 0000:04:00.0: ASM1064 has only four ports
ahci 0000:04:00.0: forcing port_map 0xffff0f -> 0xf
ahci 0000:04:00.0: SSS flag set, parallel bus scan disabled
ahci 0000:04:00.0: AHCI 0001.0301 32 slots 24 ports 6 Gbps 0xf impl SATA mode ahci 0000:04:00.0: flags: 64bit ncq sntf stag pm led only pio sxs deso sadm sds apst

Quoted 3 lines

Signed-off-by Andrey Jr. Melnikov <[email protected]>
Signed-off-by Niklas Cassel <[email protected]>
---

While I agree with Andrey that we should change the quirks that use saved_port_map to instead use force_port_map (so that we don't have two different ways to apply port_map quirks), that clean up can be done on top of this fix.

---

Applied:
https://git.kernel.org/pub/scm/linux/kernel/git/libata/linux.git/log/?h=for-6.8-fixes

Niklas Cassel

author maintainer

2024-02-19 10:37 UTC #4 lore raw

Quoted 1 line · Niklas Cassel

On Wed, Feb 14, 2024 at 05:57:57PM +0100, Niklas Cassel wrote:

(snip)

Quoted 4 lines

While I agree with Andrey that we should change the quirks that use saved_port_map to instead use force_port_map (so that we don't have two different ways to apply port_map quirks), that clean up can be done on top of this fix.

s/force_port_map/mask_port_map/

(force_port_map was the third way of applying these quirks, but it has been removed already.)

Kind regards,
Niklas