diff options
| author | Marc Zyngier <Marc.Zyngier@arm.com> | 2013-06-21 12:07:27 +0100 |
|---|---|---|
| committer | Russell King <rmk+kernel@arm.linux.org.uk> | 2013-06-24 15:27:35 +0100 |
| commit | 0d0752bca1f9a91fb646647aa4abbb21156f316c (patch) | |
| tree | d5cedcc1ba2f3fd4ed4c095c98851ed1ecf4e110 /arch/arm/kernel | |
| parent | b8e4a4740fa2b17c0a447b3ab783b3dc10702e27 (diff) | |
| download | linux-next-0d0752bca1f9a91fb646647aa4abbb21156f316c.tar.gz | |
ARM: 7769/1: Cortex-A15: fix erratum 798181 implementation
Looking into the active_asids array is not enough, as we also need
to look into the reserved_asids array (they both represent processes
that are currently running).
Also, not holding the ASID allocator lock is racy, as another CPU
could schedule that process and trigger a rollover, making the erratum
workaround miss an IPI.
Exposing this outside of context.c is a little ugly on the side, so
let's define a new entry point that the erratum workaround can call
to obtain the cpumask.
Cc: <stable@vger.kernel.org> # 3.9
Acked-by: Will Deacon <will.deacon@arm.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Diffstat (limited to 'arch/arm/kernel')
| -rw-r--r-- | arch/arm/kernel/smp_tlb.c | 18 |
1 files changed, 2 insertions, 16 deletions
diff --git a/arch/arm/kernel/smp_tlb.c b/arch/arm/kernel/smp_tlb.c index 9a52a07aa40e..a98b62dca2fa 100644 --- a/arch/arm/kernel/smp_tlb.c +++ b/arch/arm/kernel/smp_tlb.c @@ -103,7 +103,7 @@ static void broadcast_tlb_a15_erratum(void) static void broadcast_tlb_mm_a15_erratum(struct mm_struct *mm) { - int cpu, this_cpu; + int this_cpu; cpumask_t mask = { CPU_BITS_NONE }; if (!erratum_a15_798181()) @@ -111,21 +111,7 @@ static void broadcast_tlb_mm_a15_erratum(struct mm_struct *mm) dummy_flush_tlb_a15_erratum(); this_cpu = get_cpu(); - for_each_online_cpu(cpu) { - if (cpu == this_cpu) - continue; - /* - * We only need to send an IPI if the other CPUs are running - * the same ASID as the one being invalidated. There is no - * need for locking around the active_asids check since the - * switch_mm() function has at least one dmb() (as required by - * this workaround) in case a context switch happens on - * another CPU after the condition below. - */ - if (atomic64_read(&mm->context.id) == - atomic64_read(&per_cpu(active_asids, cpu))) - cpumask_set_cpu(cpu, &mask); - } + a15_erratum_get_cpumask(this_cpu, mm, &mask); smp_call_function_many(&mask, ipi_flush_tlb_a15_erratum, NULL, 1); put_cpu(); } |