Skip to contentSkip to conversation

ACPI: resource: do IRQ override on LENOVO IdeaPad

PATCH In review Jiri Slaby (SUSE) 4 participants 6 messages Oct 4, 2022 – Oct 13, 2022 ACPI

lore
Share & tools
lore
#0 lore raw

PATCH 1/2 ACPI: resource: do IRQ override on LENOVO IdeaPad

LENOVO IdeaPad Flex 5 is ryzen-5 based and the commit below removed IRQ overriding for those. This broke touchscreen and trackpad:
i2c_designware AMDI0010:00: controller timed out i2c_designware AMDI0010:03: controller timed out i2c_hid_acpi i2c-MSFT0001:00: failed to reset device: -61 i2c_designware AMDI0010:03: controller timed out
...
i2c_hid_acpi i2c-MSFT0001:00: can't add hid device: -61
i2c_hid_acpi: probe of i2c-MSFT0001:00 failed with error -61

White-list this specific model in the override_table.

For this to work, the ZEN test needs to be put below the table walk.

Fixes 37c81d9f1d1b (ACPI: resource: skip IRQ override on AMD Zen platforms)
Cc "Rafael J. Wysocki" <[email protected]>
Cc Len Brown <[email protected]>
Cc Chuanhong Guo <[email protected]>
Cc Tighe Donnelly <[email protected]>
Cc Mario Limonciello <[email protected]>
Cc Fridrich Strba <[email protected]>
---
 drivers/acpi/resource.c | 42 +++++++++++++++++++++++++++--------------
 1 file changed, 28 insertions(+), 14 deletions(-)
drivers/acpi/resource.c +28 -14
index 514d89656dde..8d13e94bb921 100644
@@ -424,17 +424,31 @@ static const struct dmi_system_id asus_laptop[] = {
 	{ }
 };
 
+static const struct dmi_system_id lenovo_82ra[] = {
+	{
+		.ident = "LENOVO IdeaPad Flex 5 16ALC7",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "82RA"),
+		},
+	},
+	{ }
+};
+
 struct irq_override_cmp {
 	const struct dmi_system_id *system;
 	unsigned char irq;
 	unsigned char triggering;
 	unsigned char polarity;
 	unsigned char shareable;
+	bool override;
 };
 
-static const struct irq_override_cmp skip_override_table[] = {
-	{ medion_laptop, 1, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0 },
-	{ asus_laptop, 1, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0 },
+static const struct irq_override_cmp override_table[] = {
+	{ medion_laptop, 1, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0, false },
+	{ asus_laptop, 1, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0, false },
+	{ lenovo_82ra, 6, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0, true },
+	{ lenovo_82ra, 10, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0, true },
 };
 
 static bool acpi_dev_irq_override(u32 gsi, u8 triggering, u8 polarity,
@@ -442,6 +456,17 @@ static bool acpi_dev_irq_override(u32 gsi, u8 triggering, u8 polarity,
 {
 	int i;
 
+	for (i = 0; i < ARRAY_SIZE(override_table); i++) {
+		const struct irq_override_cmp *entry = &override_table[i];
+
+		if (dmi_check_system(entry->system) &&
+		    entry->irq == gsi &&
+		    entry->triggering == triggering &&
+		    entry->polarity == polarity &&
+		    entry->shareable == shareable)
+			return entry->override;
+	}
+
 #ifdef CONFIG_X86
 	/*
 	 * IRQ override isn't needed on modern AMD Zen systems and
@@ -452,17 +477,6 @@ static bool acpi_dev_irq_override(u32 gsi, u8 triggering, u8 polarity,
 		return false;
 #endif
 
-	for (i = 0; i < ARRAY_SIZE(skip_override_table); i++) {
-		const struct irq_override_cmp *entry = &skip_override_table[i];
-
-		if (dmi_check_system(entry->system) &&
-		    entry->irq == gsi &&
-		    entry->triggering == triggering &&
-		    entry->polarity == polarity &&
-		    entry->shareable == shareable)
-			return false;
-	}
-
 	return true;
 }
 
Signature · 1 line
2.37.3
#1 lore raw

PATCH 2/2 ACPI: resource: note more about IRQ override

Use an exclamation mark to note which of the properties was overridden.

Cc "Rafael J. Wysocki" <[email protected]>
Cc Len Brown <[email protected]>
Cc Chuanhong Guo <[email protected]>
Cc Tighe Donnelly <[email protected]>
Cc Mario Limonciello <[email protected]>
---
 drivers/acpi/resource.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/acpi/resource.c +5 -2
index 8d13e94bb921..497d626185e4 100644
@@ -508,8 +508,11 @@ static void acpi_dev_get_irqresource(struct resource *res, u32 gsi,
 		u8 pol = p ? ACPI_ACTIVE_LOW : ACPI_ACTIVE_HIGH;
 
 		if (triggering != trig || polarity != pol) {
-			pr_warn("ACPI: IRQ %d override to %s, %s\n", gsi,
-				t ? "level" : "edge", p ? "low" : "high");
+			pr_warn("ACPI: IRQ %d override to %s%s, %s%s\n", gsi,
+				t ? "level" : "edge",
+				trig == triggering ? "" : "(!)",
+				p ? "low" : "high",
+				pol == polarity ? "" : "(!)");
 			triggering = trig;
 			polarity = pol;
 		}
Signature · 1 line
2.37.3
#2 lore raw
In reply to Jiri Slaby (SUSE)

[Public]

Quoted 21 lines

-----Original Message-----
From: Jiri Slaby (SUSE) <[email protected]>
Sent: Tuesday, October 4, 2022 05:34
To: [email protected]
Cc: [email protected]; Jiri Slaby (SUSE) <[email protected]>; Rafael J. Wysocki <[email protected]>; Len Brown <[email protected]>; linux- [email protected]; Chuanhong Guo <[email protected]>; Tighe Donnelly <[email protected]>; Limonciello, Mario <[email protected]>; Fridrich Strba <[email protected]>
Subject: [PATCH 1/2] ACPI: resource: do IRQ override on LENOVO IdeaPad

LENOVO IdeaPad Flex 5 is ryzen-5 based and the commit below removed IRQ overriding for those. This broke touchscreen and trackpad:
i2c_designware AMDI0010:00: controller timed out i2c_designware AMDI0010:03: controller timed out i2c_hid_acpi i2c-MSFT0001:00: failed to reset device: -61 i2c_designware AMDI0010:03: controller timed out
...
i2c_hid_acpi i2c-MSFT0001:00: can't add hid device: -61
i2c_hid_acpi: probe of i2c-MSFT0001:00 failed with error -61

Quoted 3 lines

White-list this specific model in the override_table.

For this to work, the ZEN test needs to be put below the table walk.

Unfortunately this is the second case that popped up very recently. Another one is listed here:
https://bugzilla.kernel.org/show_bug.cgi?id=216552

I don't think we have a good solution to cover the intersection of these bugs. The existing heuristic to look at legacy syntax and the IOAPIC doesn't work properly on all the Lenovo and ASUS Ryzen 6000 systems, but it does on these other two.

We're going to be adding more to this table either way. I /suspect/ the better solution is to revert 37c81d9f1d1b and add to the table all those that are broken.

If there is an agreement I'll try to gather a list of the ones I knew about and we can build that table.

To make that list not grow very large, we also need to get the IBVs to move to the modern extended IRQ descriptor syntax for future systems though. I can certainly have that conversation with the people who talk to them, but it's not going to be effective for a generation or two if they agree.

Quoted 103 lines
Fixes 37c81d9f1d1b (ACPI: resource: skip IRQ override on AMD Zen platforms)

Link:
https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fbugzilla.
suse.com%2Fshow_bug.cgi%3Fid%3D1203794&amp;data=05%7C01%7Cmario.li monciello%40amd.com%7Cdcaa6ca41a97434ebb7308daa5f3ebe1%7C3dd8961f e4884e608e11a82d994e183d%7C0%7C0%7C638004764302852606%7CUnknow n%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haW wiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&amp;sdata=S2udkzf131LnGABs9huj %2Fw%2BwyHp9PT0bPaUyaH4rIDY%3D&amp;reserved=0

Cc "Rafael J. Wysocki" <[email protected]>
Cc Len Brown <[email protected]>
Cc Chuanhong Guo <[email protected]>
Cc Tighe Donnelly <[email protected]>
Cc Mario Limonciello <[email protected]>
Cc Fridrich Strba <[email protected]>
---
 drivers/acpi/resource.c | 42 +++++++++++++++++++++++++++--------------
 1 file changed, 28 insertions(+), 14 deletions(-)
drivers/acpi/resource.c +17 -3
index 514d89656dde..8d13e94bb921 100644
@@ -424,17 +424,31 @@ static const struct dmi_system_id asus_laptop[] = {
 	{ }
 };
 
+static const struct dmi_system_id lenovo_82ra[] = {
+	{
+		.ident = "LENOVO IdeaPad Flex 5 16ALC7",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "82RA"),
+		},
+	},
+	{ }
+};
+
 struct irq_override_cmp {
 	const struct dmi_system_id *system;
 	unsigned char irq;
 	unsigned char triggering;
 	unsigned char polarity;
 	unsigned char shareable;
+	bool override;
 };
 
-static const struct irq_override_cmp skip_override_table[] = {
-	{ medion_laptop, 1, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0 },
-	{ asus_laptop, 1, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0 },
+static const struct irq_override_cmp override_table[] = {
+	{ medion_laptop, 1, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0,
 false },
+	{ asus_laptop, 1, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0, false },
+	{ lenovo_82ra, 6, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0, true },
+	{ lenovo_82ra, 10, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0, true
 },
 };

static bool acpi_dev_irq_override(u32 gsi, u8 triggering, u8 polarity,

drivers/acpi/resource.c +11 -0
@@ -442,6 +456,17 @@ static bool acpi_dev_irq_override(u32 gsi, u8
 triggering, u8 polarity,
 {
 	int i;
 
+	for (i = 0; i < ARRAY_SIZE(override_table); i++) {
+		const struct irq_override_cmp *entry = &override_table[i];
+
+		if (dmi_check_system(entry->system) &&
+		    entry->irq == gsi &&
+		    entry->triggering == triggering &&
+		    entry->polarity == polarity &&
+		    entry->shareable == shareable)
+			return entry->override;
+	}
+
 #ifdef CONFIG_X86
 	/*
drivers/acpi/resource.c +0 -12
 	 * IRQ override isn't needed on modern AMD Zen systems and
@@ -452,17 +477,6 @@ static bool acpi_dev_irq_override(u32 gsi, u8
 triggering, u8 polarity,
 		return false;
 #endif
 
-	for (i = 0; i < ARRAY_SIZE(skip_override_table); i++) {
-		const struct irq_override_cmp *entry = &skip_override_table[i];
-
-		if (dmi_check_system(entry->system) &&
-		    entry->irq == gsi &&
-		    entry->triggering == triggering &&
-		    entry->polarity == polarity &&
-		    entry->shareable == shareable)
-			return false;
-	}
-
 	return true;
 }
 
--

2.37.3

#3 lore raw

Hi!

Quoted 9 lines · Mario

[...]

Quoted 4 lines

White-list this specific model in the override_table.

For this to work, the ZEN test needs to be put below the table walk.

Unfortunately this is the second case that popped up very recently. Another one is listed here:
https://bugzilla.kernel.org/show_bug.cgi?id=216552

Now I'm really curious how Windows is able to handle all these vendor crap...

Quoted 3 lines

I don't think we have a good solution to cover the intersection of these bugs. The existing heuristic to look at legacy syntax and the IOAPIC doesn't work properly on all the Lenovo and ASUS Ryzen 6000 systems, but it does on these other two.

These legacy IRQ declarations are obsolete, but they aren't really wrong. Meanwhile the two devices popped up until now both got IRQ declarations which don't match the actual device configuration.

Quoted 2 lines

We're going to be adding more to this table either way. I /suspect/ the better solution is to revert 37c81d9f1d1b and add to the table all those that are broken.

I think we should have a list of only the wrong IRQ declaration and apply the fix
just for them, instead of applying the fix to all devices and skip it for selected
devices the fix breaks.

Signature · 2 lines
Regards,
Chuanhong Guo
#4 lore raw

[Public]

Quoted 37 lines

-----Original Message-----
From: Chuanhong Guo <[email protected]>
Sent: Tuesday, October 4, 2022 23:02
To: Limonciello, Mario <[email protected]>
Cc: Jiri Slaby (SUSE) <[email protected]>; [email protected]; linux- [email protected]; Rafael J. Wysocki <[email protected]>; Len Brown <[email protected]>; [email protected]; Tighe Donnelly <[email protected]>; Fridrich Strba <[email protected]>
Subject: Re: [PATCH 1/2] ACPI: resource: do IRQ override on LENOVO IdeaPad

Hi!

Quoted 9 lines · Mario

[...]

Quoted 4 lines

White-list this specific model in the override_table.

For this to work, the ZEN test needs to be put below the table walk.

Unfortunately this is the second case that popped up very recently. Another one is listed here:

https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fbugz
illa.kernel.org%2Fshow_bug.cgi%3Fid%3D216552&amp;data=05%7C01%7CM ario.Limonciello%40amd.com%7C27a32c2395ed4d2a85e208daa68666bb%7C3 dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C638005393451041667%7C Unknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJB TiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&amp;sdata=iLj95fy44% 2BT2KCahzTD8HP2bl2dD6gXVOcVnHylPWJc%3D&amp;reserved=0

Now I'm really curious how Windows is able to handle all these vendor crap...

Quoted 1 line

I don't think we have a good solution to cover the intersection of these

bugs. The

Quoted 1 line

existing heuristic to look at legacy syntax and the IOAPIC doesn't work

Quoted 8 lines

properly

Quoted 1 line

on all the Lenovo and ASUS Ryzen 6000 systems, but it does on these other

two.

These legacy IRQ declarations are obsolete, but they aren't really wrong. Meanwhile the two devices popped up until now both got IRQ declarations which
don't match the actual device configuration.

You're right; both of these are technically BIOS DSDT bugs if you had assumed that this workaround wasn't in place.

Quoted 10 lines
Quoted 2 lines

We're going to be adding more to this table either way. I /suspect/ the

better solution

Quoted 1 line

is to revert 37c81d9f1d1b and add to the table all those that are broken.

I think we should have a list of only the wrong IRQ declaration and apply the fix
just for them, instead of applying the fix to all devices and skip it for selected
devices the fix breaks.

OK. In that case Jiri I think your patch series makes sense.

Quoted 4 lines

--
Regards,
Chuanhong Guo

#5 lore raw
In reply to Jiri Slaby (SUSE)
Quoted 95 lines · Jiri Slaby (SUSE)

LENOVO IdeaPad Flex 5 is ryzen-5 based and the commit below removed IRQ overriding for those. This broke touchscreen and trackpad:
i2c_designware AMDI0010:00: controller timed out i2c_designware AMDI0010:03: controller timed out i2c_hid_acpi i2c-MSFT0001:00: failed to reset device: -61 i2c_designware AMDI0010:03: controller timed out
...
i2c_hid_acpi i2c-MSFT0001:00: can't add hid device: -61
i2c_hid_acpi: probe of i2c-MSFT0001:00 failed with error -61

White-list this specific model in the override_table.

For this to work, the ZEN test needs to be put below the table walk.

Fixes 37c81d9f1d1b (ACPI: resource: skip IRQ override on AMD Zen platforms)
Cc "Rafael J. Wysocki" <[email protected]>
Cc Len Brown <[email protected]>
Cc Chuanhong Guo <[email protected]>
Cc Tighe Donnelly <[email protected]>
Cc Mario Limonciello <[email protected]>
Cc Fridrich Strba <[email protected]>
---
 drivers/acpi/resource.c | 42 +++++++++++++++++++++++++++--------------
 1 file changed, 28 insertions(+), 14 deletions(-)
drivers/acpi/resource.c +28 -8
index 514d89656dde..8d13e94bb921 100644
@@ -424,17 +424,31 @@ static const struct dmi_system_id asus_laptop[] = {
        { }
 };
 
+static const struct dmi_system_id lenovo_82ra[] = {
+       {
+               .ident = "LENOVO IdeaPad Flex 5 16ALC7",
+               .matches = {
+                       DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
+                       DMI_MATCH(DMI_PRODUCT_NAME, "82RA"),
+               },
+       },
+       { }
+};
+
 struct irq_override_cmp {
        const struct dmi_system_id *system;
        unsigned char irq;
        unsigned char triggering;
        unsigned char polarity;
        unsigned char shareable;
+       bool override;
 };
 
-static const struct irq_override_cmp skip_override_table[] = {
-       { medion_laptop, 1, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0 },
-       { asus_laptop, 1, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0 },
+static const struct irq_override_cmp override_table[] = {
+       { medion_laptop, 1, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0, false },
+       { asus_laptop, 1, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0, false },
+       { lenovo_82ra, 6, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0, true },
+       { lenovo_82ra, 10, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0, true },
 };
 
 static bool acpi_dev_irq_override(u32 gsi, u8 triggering, u8 polarity,
@@ -442,6 +456,17 @@ static bool acpi_dev_irq_override(u32 gsi, u8 triggering, u8 polarity,
 {
        int i;
 
+       for (i = 0; i < ARRAY_SIZE(override_table); i++) {
+               const struct irq_override_cmp *entry = &override_table[i];
+
+               if (dmi_check_system(entry->system) &&
+                   entry->irq == gsi &&
+                   entry->triggering == triggering &&
+                   entry->polarity == polarity &&
+                   entry->shareable == shareable)
+                       return entry->override;
+       }
+
 #ifdef CONFIG_X86
        /*
         * IRQ override isn't needed on modern AMD Zen systems and
@@ -452,17 +477,6 @@ static bool acpi_dev_irq_override(u32 gsi, u8 triggering, u8 polarity,
                return false;
 #endif
 
-       for (i = 0; i < ARRAY_SIZE(skip_override_table); i++) {
-               const struct irq_override_cmp *entry = &skip_override_table[i];
-
-               if (dmi_check_system(entry->system) &&
-                   entry->irq == gsi &&

drivers/acpi/resource.c static bool acpi_dev_irq_override(u32 gsi, u8 triggering, u8 polarity,

Quoted 10 lines
drivers/acpi/resource.c +0 -7
-                   entry->triggering == triggering &&
-                   entry->polarity == polarity &&
-                   entry->shareable == shareable)
-                       return false;
-       }
-
        return true;
 }
 
--

Applied along with the [2/2] as 6.1-rc material, thanks!