From: Shanker Donthineni <shankerd@xxxxxxxxxxxxxx>
[ Upstream commit ebe2f8718007d5a1238bb3cb8141b5bb2b4d5773 ]
The ACPI specification says OS shouldn't attempt to use GICC configuration
parameters if the flag ACPI_MADT_ENABLED is cleared. The ARM64-SMP code
skips the disabled GICC entries but not causing any issue. However the
current GICv3 driver probe bails out causing kernel panic() instead of
skipping the disabled GICC interfaces. This issue happens on systems
where redistributor regions are not in the always-on power domain and
one of GICC interface marked with ACPI_MADT_ENABLED=0.
This patch does the two things to fix the panic.
- Don't return an error in gic_acpi_match_gicc() for disabled GICC entry.
- No need to keep GICR region information for disabled GICC entry.
Observed kernel crash on QDF2400 platform GICC entry is disabled.
Kernel crash traces:
Kernel panic - not syncing: No interrupt controller found.
CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.13.5 #26
[<ffff000008087770>] dump_backtrace+0x0/0x218
[<ffff0000080879dc>] show_stack+0x14/0x20
[<ffff00000883b078>] dump_stack+0x98/0xb8
[<ffff0000080c5c14>] panic+0x118/0x26c
[<ffff000008b62348>] init_IRQ+0x24/0x2c
[<ffff000008b609fc>] start_kernel+0x230/0x394
[<ffff000008b601e4>] __primary_switched+0x64/0x6c
---[ end Kernel panic - not syncing: No interrupt controller found.
Disabled GICC subtable example:
Subtable Type : 0B [Generic Interrupt Controller]
Length : 50
Reserved : 0000
CPU Interface Number : 0000003D
Processor UID : 0000003D
Flags (decoded below) : 00000000
Processor Enabled : 0
Performance Interrupt Trig Mode : 0
Virtual GIC Interrupt Trig Mode : 0
Parking Protocol Version : 00000000
Performance Interrupt : 00000017
Parked Address : 0000000000000000
Base Address : 0000000000000000
Virtual GIC Base Address : 0000000000000000
Hypervisor GIC Base Address : 0000000000000000
Virtual GIC Interrupt : 00000019
Redistributor Base Address : 0000FFFF88F40000
ARM MPIDR : 000000000000000D
Efficiency Class : 00
Reserved : 000000
Signed-off-by: Shanker Donthineni <shankerd@xxxxxxxxxxxxxx>
Signed-off-by: Marc Zyngier <marc.zyngier@xxxxxxx>
Signed-off-by: Sasha Levin <alexander.levin@xxxxxxxxxxxxx>
---
drivers/irqchip/irq-gic-v3.c | 11 +++++++++++
1 file changed, 11 insertions(+)
diff --git a/drivers/irqchip/irq-gic-v3.c b/drivers/irqchip/irq-gic-v3.c
index ae9ff72e83ee..848fcdf6a112 100644
--- a/drivers/irqchip/irq-gic-v3.c
+++ b/drivers/irqchip/irq-gic-v3.c
@@ -1297,6 +1297,10 @@ gic_acpi_parse_madt_gicc(struct acpi_subtable_header *header,
u32 size = reg == GIC_PIDR2_ARCH_GICv4 ? SZ_64K * 4 : SZ_64K * 2;
void __iomem *redist_base;
+ /* GICC entry which has !ACPI_MADT_ENABLED is not unusable so skip */
+ if (!(gicc->flags & ACPI_MADT_ENABLED))
+ return 0;
+
redist_base = ioremap(gicc->gicr_base_address, size);
if (!redist_base)
return -ENOMEM;
@@ -1346,6 +1350,13 @@ static int __init gic_acpi_match_gicc(struct acpi_subtable_header *header,
if ((gicc->flags & ACPI_MADT_ENABLED) && gicc->gicr_base_address)
return 0;
+ /*
+ * It's perfectly valid firmware can pass disabled GICC entry, driver
+ * should not treat as errors, skip the entry instead of probe fail.
+ */
+ if (!(gicc->flags & ACPI_MADT_ENABLED))
+ return 0;
+
return -ENODEV;
}
--
2.14.1