diff options
author | Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com> | 2008-02-25 14:32:22 +0900 |
---|---|---|
committer | Tony Luck <tony.luck@intel.com> | 2008-03-04 14:16:20 -0800 |
commit | a6cd6322d594014240465210ccb290971469c6e8 (patch) | |
tree | 4be5c34dc636fa85e7b6e39e0b3ff82510c7bb8a /include | |
parent | 86dffa4cd1a1d61fed68ab64c674d4094f2bdfe4 (diff) | |
download | linux-a6cd6322d594014240465210ccb290971469c6e8.tar.bz2 |
[IA64] Fix irq migration in multiple vector domain
Fix the problem that the following error message is sometimes displayed
at irq migration when vector domain is enabled.
"Unexpected interrupt vector %d on CPU %d is not mapped to any IRQ!"
The cause of this problem is an interrupt is sent to the previous
target CPU after cleaning up vector to irq mapping table. To clean up
vector to irq map on the previous target CPU safty, change the irq
migration in multiple vector domain as follows. The original idea is
from x86 interrupt management code.
- Delay vector to irq table cleanup until the interrupts are sent
to new target CPUs. By this, it is ensured that target CPU is
completely changed on the interrupt controller side.
- Even after the interrupts are sent to new target CPUs, there can
be pended interrupts remaining on the previous target CPU. So we
need to delay clearning up vector to irq table until the pended
interrupt is handled. For this, send IPI to the previous target
CPU with lower priority vector and clean up vector to irq table
in its handler.
This patch affects only to irq migration code with multiple vector
domain is enabled.
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Diffstat (limited to 'include')
-rw-r--r-- | include/asm-ia64/hw_irq.h | 12 |
1 files changed, 11 insertions, 1 deletions
diff --git a/include/asm-ia64/hw_irq.h b/include/asm-ia64/hw_irq.h index 7e6e3779670a..76366dc9c1a0 100644 --- a/include/asm-ia64/hw_irq.h +++ b/include/asm-ia64/hw_irq.h @@ -93,6 +93,9 @@ extern __u8 isa_irq_to_vector_map[16]; struct irq_cfg { ia64_vector vector; cpumask_t domain; + cpumask_t old_domain; + unsigned move_cleanup_count; + u8 move_in_progress : 1; }; extern spinlock_t vector_lock; extern struct irq_cfg irq_cfg[NR_IRQS]; @@ -106,12 +109,19 @@ extern int assign_irq_vector (int irq); /* allocate a free vector */ extern void free_irq_vector (int vector); extern int reserve_irq_vector (int vector); extern void __setup_vector_irq(int cpu); -extern int reassign_irq_vector(int irq, int cpu); extern void ia64_send_ipi (int cpu, int vector, int delivery_mode, int redirect); extern void register_percpu_irq (ia64_vector vec, struct irqaction *action); extern int check_irq_used (int irq); extern void destroy_and_reserve_irq (unsigned int irq); +#if defined(CONFIG_SMP) && (defined(CONFIG_IA64_GENERIC) || defined(CONFIG_IA64_DIG)) +extern int irq_prepare_move(int irq, int cpu); +extern void irq_complete_move(unsigned int irq); +#else +static inline int irq_prepare_move(int irq, int cpu) { return 0; } +static inline void irq_complete_move(unsigned int irq) {} +#endif + static inline void ia64_resend_irq(unsigned int vector) { platform_send_ipi(smp_processor_id(), vector, IA64_IPI_DM_INT, 0); |