From: Liu Jiang <jiang.liu@xxxxxxxxxx> This bug interfered with another bug in QLA FC driver has caused big trouble to us during the past month when validating a 32P Poulson system with plenty of IO devices. Hope it could find its way into the mainstream kernel. Thanks! --- When handling Interrupt Source Override in MADT table, the default ISA IRQ trigger model and poliarity should be edge-rising. Current IA64 implmentation doesn't follow the specification and set default ISA IRQ trigger model as level-low. With that wrong configuration and when system runs out of interrupt vectors, it will cause vector sharing among edge triggered ISA IRQ and level triggered PCI IRQ, then interrupt storm. So change the code to follow the specification. Signed-off-by: Liu Jiang <jiang.liu@xxxxxxxxxx> --- arch/ia64/kernel/acpi.c | 8 ++++---- 1 files changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c index bfb4d01..18c33d6 100644 --- a/arch/ia64/kernel/acpi.c +++ b/arch/ia64/kernel/acpi.c @@ -349,11 +349,11 @@ acpi_parse_int_src_ovr(struct acpi_subtable_header * header, iosapic_override_isa_irq(p->source_irq, p->global_irq, ((p->inti_flags & ACPI_MADT_POLARITY_MASK) == - ACPI_MADT_POLARITY_ACTIVE_HIGH) ? - IOSAPIC_POL_HIGH : IOSAPIC_POL_LOW, + ACPI_MADT_POLARITY_ACTIVE_LOW) ? + IOSAPIC_POL_LOW : IOSAPIC_POL_HIGH, ((p->inti_flags & ACPI_MADT_TRIGGER_MASK) == - ACPI_MADT_TRIGGER_EDGE) ? - IOSAPIC_EDGE : IOSAPIC_LEVEL); + ACPI_MADT_TRIGGER_LEVEL) ? + IOSAPIC_LEVEL : IOSAPIC_EDGE); return 0; } -- 1.7.1 -- To unsubscribe from this list: send the line "unsubscribe linux-ia64" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
|