Skip to contentSkip to conversation

ACPI: skip IRQ1 override on 3 Ryzen 6000 laptops

PATCH In review Chuanhong Guo 3 participants 6 messages Jun 30, 2022 – Jul 5, 2022

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH v5 1/1 ACPI: skip IRQ1 override on 3 Ryzen 6000 laptops

The IRQ1 of these laptops with Ryzen 6000 and Insyde UEFI are active low and defined in legacy format in ACPI DSDT. The kernel override makes the keyboard interrupt polarity inverted, resulting in non-functional keyboard.

Skip legacy IRQ override for:
Lenovo ThinkBook 14G4+ ARA
Redmi Book Pro 15 2022 Ryzen
Asus Zenbook S 13 OLED UM5302

---
Changes since v1:
Match DMI_PRODUCT_NAME for ThinkBook because the board name is used for other completely different Lenovo laptops. Add a patch for RedmiBook

Changes since v2:
Fix alphabetical order in skip_override_table
Add a patch for Asus Zenbook

Changes since v3:
Merge patches as requested
Fix another alphabetical ordering between two structs

Changes since v4:

 rename the ident in RedmiBook entry.

There's also an Intel version of this series, so rename it to make it specific.
reword commit title

 drivers/acpi/resource.c | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)
drivers/acpi/resource.c +26 -0
index c2d494784425..0491da180fc5 100644
@@ -381,6 +381,31 @@ unsigned int acpi_dev_get_irq_type(int triggering, int polarity)
 }
 EXPORT_SYMBOL_GPL(acpi_dev_get_irq_type);
 
+static const struct dmi_system_id irq1_edge_low_shared[] = {
+	{
+		.ident = "Asus Zenbook S 13 OLED UM5302",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
+			DMI_MATCH(DMI_BOARD_NAME, "UM5302TA"),
+		},
+	},
+	{
+		.ident = "Lenovo ThinkBook 14 G4+ ARA",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "21D0"),
+		},
+	},
+	{
+		.ident = "Redmi Book Pro 15 2022 Ryzen",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "TIMI"),
+			DMI_MATCH(DMI_BOARD_NAME, "TM2113"),
+		},
+	},
+	{ }
+};
+
 static const struct dmi_system_id medion_laptop[] = {
 	{
 		.ident = "MEDION P15651",
@@ -408,6 +433,7 @@ struct irq_override_cmp {
 };
 
 static const struct irq_override_cmp skip_override_table[] = {
+	{ irq1_edge_low_shared, 1, ACPI_EDGE_SENSITIVE, ACPI_ACTIVE_LOW, 1 },
 	{ medion_laptop, 1, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0 },
 };
 
Signature · 1 line
2.36.1
#1 lore raw
Quoted 9 lines · Chuanhong Guo

The IRQ1 of these laptops with Ryzen 6000 and Insyde UEFI are active low and defined in legacy format in ACPI DSDT. The kernel override makes the keyboard interrupt polarity inverted, resulting in non-functional keyboard.

Skip legacy IRQ override for:
Lenovo ThinkBook 14G4+ ARA
Redmi Book Pro 15 2022 Ryzen
Asus Zenbook S 13 OLED UM5302

It's really unfortunate that these laptops have the bug. This was found and fixed in the reference BIOS for Ryzen 6000 too (via an MADT override service) very early in development. It seems these manufacturers didn't pick up (or ignored) the solution.

However I do want to point out that Windows doesn't care about legacy format or not. This bug where keyboard doesn't work only popped up on Linux.

Given the number of systems with the bug is appearing to grow I wonder if the right answer is actually a new heuristic that doesn't apply the kernel override for polarity inversion anymore. Maybe if the system is 2022 or newer? Or on the ACPI version?

Quoted 71 lines

---
Changes since v1:
Match DMI_PRODUCT_NAME for ThinkBook because the board name is used for other completely different Lenovo laptops. Add a patch for RedmiBook

Changes since v2:
Fix alphabetical order in skip_override_table
Add a patch for Asus Zenbook

Changes since v3:
Merge patches as requested
Fix another alphabetical ordering between two structs

Changes since v4:
rename the ident in RedmiBook entry.
There's also an Intel version of this series, so rename it to make it specific.
reword commit title

  drivers/acpi/resource.c | 26 ++++++++++++++++++++++++++
  1 file changed, 26 insertions(+) >
drivers/acpi/resource.c +26 -0
index c2d494784425..0491da180fc5 100644
@@ -381,6 +381,31 @@ unsigned int acpi_dev_get_irq_type(int triggering, int polarity)
  }
  EXPORT_SYMBOL_GPL(acpi_dev_get_irq_type);
 
+static const struct dmi_system_id irq1_edge_low_shared[] = {
+	{
+		.ident = "Asus Zenbook S 13 OLED UM5302",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
+			DMI_MATCH(DMI_BOARD_NAME, "UM5302TA"),
+		},
+	},
+	{
+		.ident = "Lenovo ThinkBook 14 G4+ ARA",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "21D0"),
+		},
+	},
+	{
+		.ident = "Redmi Book Pro 15 2022 Ryzen",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "TIMI"),
+			DMI_MATCH(DMI_BOARD_NAME, "TM2113"),
+		},
+	},
+	{ }
+};
+
  static const struct dmi_system_id medion_laptop[] = {
  	{
  		.ident = "MEDION P15651",
@@ -408,6 +433,7 @@ struct irq_override_cmp {
  };
 
  static const struct irq_override_cmp skip_override_table[] = {
+	{ irq1_edge_low_shared, 1, ACPI_EDGE_SENSITIVE, ACPI_ACTIVE_LOW, 1 },
  	{ medion_laptop, 1, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_LOW, 0 },
  };
 
#2 lore raw
Quoted 8 lines · Mario

However I do want to point out that Windows doesn't care about legacy format or not. This bug where keyboard doesn't work only popped up on Linux.

Given the number of systems with the bug is appearing to grow I wonder if the right answer is actually a new heuristic that doesn't apply the kernel override for polarity inversion anymore. Maybe if the system is 2022 or newer? Or on the ACPI version?

The previous attempt to limit the scope of IRQ override ends up breaking some other buggy devices:
https://patchwork.kernel.org/project/linux-acpi/patch/[email protected]/

It's unfortunate that the original author of this IRQ override doesn't limit the scope to their exact devices.

Hi, Rafael! What do you think? should we skip this IRQ override one-by-one or add a different matching logic to check the bios date instead?

Signature · 2 lines
Regards,
Chuanhong Guo
#3 lore raw
Quoted 16 lines · Chuanhong Guo
Quoted 8 lines · Mario

However I do want to point out that Windows doesn't care about legacy format or not. This bug where keyboard doesn't work only popped up on Linux.

Given the number of systems with the bug is appearing to grow I wonder if the right answer is actually a new heuristic that doesn't apply the kernel override for polarity inversion anymore. Maybe if the system is 2022 or newer? Or on the ACPI version?

The previous attempt to limit the scope of IRQ override ends up breaking some other buggy devices:
https://patchwork.kernel.org/project/linux-acpi/patch/[email protected]/

Quoted 6 lines

It's unfortunate that the original author of this IRQ override doesn't limit the scope to their exact devices.

Hi, Rafael! What do you think? should we skip this IRQ override one-by-one or add a different matching logic to check the bios date instead?

It would be better to find something precise enough to identify the machines in question without pulling in the others and use that for skipping the override instead of listing them all one by one in the blocklist.

#4 lore raw

[Public]

Quoted 41 lines

-----Original Message-----
From: Rafael J. Wysocki <[email protected]>
Sent: Tuesday, July 5, 2022 13:24
To: Chuanhong Guo <[email protected]>
Cc: Rafael J. Wysocki <[email protected]>; Limonciello, Mario <[email protected]>; ACPI Devel Maling List <linux- [email protected]>; Stable <[email protected]>; Tighe Donnelly <[email protected]>; Kent Hou Man <[email protected]>; Len Brown <[email protected]>; open list <[email protected]>
Subject: Re: [PATCH v5] ACPI: skip IRQ1 override on 3 Ryzen 6000 laptops

Quoted 7 lines · Chuanhong Guo
Quoted 4 lines · Mario

However I do want to point out that Windows doesn't care about legacy format or not. This bug where keyboard doesn't work only popped up on Linux.

Quoted 8 lines
Quoted 4 lines

Given the number of systems with the bug is appearing to grow I wonder if the right answer is actually a new heuristic that doesn't apply the kernel override for polarity inversion anymore. Maybe if the system is 2022 or newer? Or on the ACPI version?

The previous attempt to limit the scope of IRQ override ends up breaking some other buggy devices:

https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpatc
hwork.kernel.org%2Fproject%2Flinux- acpi%2Fpatch%2F20210728151958.15205-1- hui.wang%40canonical.com%2F&amp;data=05%7C01%7Cmario.limonciello%4 0amd.com%7C106955e4611344d3bc3808da5eb3971d%7C3dd8961fe4884e608 e11a82d994e183d%7C0%7C0%7C637926422673112765%7CUnknown%7CTWF pbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXV CI6Mn0%3D%7C3000%7C%7C%7C&amp;sdata=xOaRbkCv9EMhpLO%2BGAP mDjEhQ78xjYFBvehLZdg1k1I%3D&amp;reserved=0

Quoted 4 lines

It's unfortunate that the original author of this IRQ override doesn't limit the scope to their exact devices.

Quoted 8 lines
Quoted 3 lines

Hi, Rafael! What do you think? should we skip this IRQ override one-by-one or add a different matching logic to check the bios date instead?

It would be better to find something precise enough to identify the machines in question without pulling in the others and use that for skipping the override instead of listing them all one by one in the blocklist.

How about using the CPU family/model in this case?

#5 lore raw
Quoted 48 lines · Mario

[Public]

Quoted 45 lines

-----Original Message-----
From: Rafael J. Wysocki <[email protected]>
Sent: Tuesday, July 5, 2022 13:24
To: Chuanhong Guo <[email protected]>
Cc: Rafael J. Wysocki <[email protected]>; Limonciello, Mario <[email protected]>; ACPI Devel Maling List <linux- [email protected]>; Stable <[email protected]>; Tighe Donnelly <[email protected]>; Kent Hou Man <[email protected]>; Len Brown <[email protected]>; open list <[email protected]>
Subject: Re: [PATCH v5] ACPI: skip IRQ1 override on 3 Ryzen 6000 laptops

Quoted 7 lines · Chuanhong Guo
Quoted 4 lines · Mario

However I do want to point out that Windows doesn't care about legacy format or not. This bug where keyboard doesn't work only popped up on Linux.

Quoted 8 lines
Quoted 4 lines

Given the number of systems with the bug is appearing to grow I wonder if the right answer is actually a new heuristic that doesn't apply the kernel override for polarity inversion anymore. Maybe if the system is 2022 or newer? Or on the ACPI version?

The previous attempt to limit the scope of IRQ override ends up breaking some other buggy devices:

https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpatc
hwork.kernel.org%2Fproject%2Flinux- acpi%2Fpatch%2F20210728151958.15205-1- hui.wang%40canonical.com%2F&amp;data=05%7C01%7Cmario.limonciello%4 0amd.com%7C106955e4611344d3bc3808da5eb3971d%7C3dd8961fe4884e608 e11a82d994e183d%7C0%7C0%7C637926422673112765%7CUnknown%7CTWF pbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXV CI6Mn0%3D%7C3000%7C%7C%7C&amp;sdata=xOaRbkCv9EMhpLO%2BGAP mDjEhQ78xjYFBvehLZdg1k1I%3D&amp;reserved=0

Quoted 7 lines

It's unfortunate that the original author of this IRQ override doesn't limit the scope to their exact devices.

Hi, Rafael! What do you think? should we skip this IRQ override one-by-one or add a different matching logic to check the bios date instead?

Quoted 6 lines
Quoted 4 lines

It would be better to find something precise enough to identify the machines in question without pulling in the others and use that for skipping the override instead of listing them all one by one in the blocklist.

How about using the CPU family/model in this case?

That would work for me. The code in question is all quirks anyway.