Re: genirq: Ensure we locate the passed IRQ in irq_alloc_descs()

From: Milton Miller
Date: Fri Jun 03 2011 - 05:24:47 EST


On Thu, 02 Jun 2011 17:55:13 -0000, Mark Brown wrote:
> When irq_alloc_descs() is called with no base IRQ specified then it will
> search for a range of IRQs starting from a specified base address. In the
> case where an IRQ is specified it still does this search in order to ensure
> that none of the requested range is already allocated and it still uses the
> from parameter to specify the base for the search. This means that in the
> case where a base is specified but from is zero (which is reasonable as
> any IRQ number is in the range specified by a zero from) the function will
> get confused and try to allocate the first suitably sized block of free IRQs
> it finds.
>
> Instead use a specified IRQ as the base address for the search, and insist
> that any from that is specified can support that IRQ.
>
> Signed-off-by: Mark Brown <broonie@xxxxxxxxxxxxxxxxxxxxxxxxxxx>
>
> ---
> kernel/irq/irqdesc.c | 6 ++++++
> 1 files changed, 6 insertions(+), 0 deletions(-)
>
> diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c
> index 886e803..bb53d6c 100644
> --- a/kernel/irq/irqdesc.c
> +++ b/kernel/irq/irqdesc.c
> @@ -346,6 +346,12 @@ irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node)
> if (!cnt)
> return -EINVAL;
>
> + if (irq >= 0) {
> + if (from > irq)
> + return -EINVAL;
> + from = irq;
> + }
> +
> mutex_lock(&sparse_irq_lock);
>
> start = bitmap_find_next_zero_area(allocated_irqs, IRQ_BITMAP_BITS,

and then right after this the code continues:

ret = -EEXIST;
if (irq >=0 && start != irq)
goto err;


This patch enables exactly the calls I want to forbid ! Why do
you need to verify that there are no irqs between from and irq ?
What is your use case?

Change your caller to specify the irq twice if you need a specific irq
block, or if you only need one then use the helper irq_alloc_desc_at.

If you want to change irq_alloc_descs, please make it return -EINVAL
if irq >=0 && from != irq (like I did).

See http://lkml.indiana.edu/hypermail/linux/kernel/1105.3/00739.html
[PATCH RFC 4/4] irq: allow a per-allocation upper limit when allocating irqs

(and yes, I have made the changes based on the feedback but haven't
had time to get back to the series).

Thanks,
milton
QUIT
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/