lkml.org 
[lkml]   [2014]   [Nov]   [24]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH v3 06/13] irqchip: GICv3: ITS: LPI allocator
On 24/11/14 14:57, Jiang Liu wrote:
>
>
> On 2014/11/24 22:35, Marc Zyngier wrote:
>> LPIs are the type of interrupts that are used by the ITS. Given
>> the size of the namespace (anywhere between 16 and 32bit), interrupt
>> IDs are allocated in chunks of 32.
>>
>> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
>> ---
>> drivers/irqchip/irq-gic-v3-its.c | 103 +++++++++++++++++++++++++++++++++++++++
>> 1 file changed, 103 insertions(+)
>>
>> diff --git a/drivers/irqchip/irq-gic-v3-its.c b/drivers/irqchip/irq-gic-v3-its.c
>> index d24bebd..4154a16 100644
>> --- a/drivers/irqchip/irq-gic-v3-its.c
>> +++ b/drivers/irqchip/irq-gic-v3-its.c
>> @@ -586,3 +586,106 @@ static struct irq_chip its_irq_chip = {
>> .irq_eoi = its_eoi_irq,
>> .irq_set_affinity = its_set_affinity,
>> };
>> +
>> +/*
>> + * How we allocate LPIs:
>> + *
>> + * The GIC has id_bits bits for interrupt identifiers. From there, we
>> + * must subtract 8192 which are reserved for SGIs/PPIs/SPIs. Then, as
>> + * we allocate LPIs by chunks of 32, we can shift the whole thing by 5
>> + * bits to the right.
> Just curious, why 32? sizeof(long) is 4 on ARM64?

No, sizeof(long) == 8, as on any sane 64bit architecture.

There are two reasons for this:
- the ID space is rather large (at least 16 bits, possibly 32 bits), so
we're trying not to allocate the whole bitmap in one go.
- 32 is the maximum a MSI-capable device can request. Allocating 32
interrupts in one go makes sure that these interrupts are contiguous and
satisfy the MSI requirements.

Hope this helps,

M.
--
Jazz is not dead. It just smells funny...


\
 
 \ /
  Last update: 2014-11-24 17:01    [W:0.179 / U:0.080 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site