Re: [PATCH v2 6/6] irqchip: gicv3: skip ITS init when no ITS available

From: Yun Wu (Abel)
Date: Mon Feb 16 2015 - 09:57:57 EST


Hi Murzin,
On 2015/2/16 18:05, Vladimir Murzin wrote:

> Hi Yun,
>
> On 15/02/15 09:32, Yun Wu wrote:
>> There is one more condition that needs to be considered when judging
>> whether LPI feature is enabled or not, which is whether there is any
>> ITS available and correctly enabled.
>>
>> This patch will fix this by caching ITS enabling status in the GIC
>> chip data structure.
>
> I posted patch for that before [1] and it landed in Marc's tree
> (irq/gic-fixes). It is not clear from the commit message what the "one
> more condition" is, but I guess it is the same dts stuff. Do you see
> issue without your patch applied?

Oh yes, your patch perfectly solved this problem, so this one is no longer
needed. And sorry for not noticing your patch. :)

Thanks,
Abel

>
> [1]
> http://lists.infradead.org/pipermail/linux-arm-kernel/2015-January/314752.html
>
> Thanks
> Vladimir
>
>>
>> Signed-off-by: Yun Wu <wuyun.wu@xxxxxxxxxx>
>> ---
>> drivers/irqchip/irq-gic-v3.c | 18 +++++++++---------
>> 1 file changed, 9 insertions(+), 9 deletions(-)
>>
>> diff --git a/drivers/irqchip/irq-gic-v3.c b/drivers/irqchip/irq-gic-v3.c
>> index 1a146cc..e17faca 100644
>> --- a/drivers/irqchip/irq-gic-v3.c
>> +++ b/drivers/irqchip/irq-gic-v3.c
>> @@ -47,6 +47,7 @@ struct gic_chip_data {
>> u64 redist_stride;
>> u32 nr_redist_regions;
>> unsigned int irq_nr;
>> + int lpi_enabled;
>> };
>>
>> static struct gic_chip_data gic_data __read_mostly;
>> @@ -390,11 +391,6 @@ static void gic_cpu_sys_reg_init(void)
>> gic_write_grpen1(1);
>> }
>>
>> -static int gic_dist_supports_lpis(void)
>> -{
>> - return !!(readl_relaxed(gic_data.dist_base + GICD_TYPER) & GICD_TYPER_LPIS);
>> -}
>> -
>> static void gic_cpu_init(void)
>> {
>> void __iomem *rbase;
>> @@ -410,7 +406,7 @@ static void gic_cpu_init(void)
>> gic_cpu_config(rbase, gic_redist_wait_for_rwp);
>>
>> /* Give LPIs a spin */
>> - if (IS_ENABLED(CONFIG_ARM_GIC_V3_ITS) && gic_dist_supports_lpis())
>> + if (gic_data.lpi_enabled)
>> its_cpu_init();
>>
>> /* initialise system registers */
>> @@ -629,7 +625,7 @@ static int gic_irq_domain_map(struct irq_domain *d, unsigned int irq,
>> }
>> /* LPIs */
>> if (hw >= 8192 && hw < GIC_ID_NR) {
>> - if (!gic_dist_supports_lpis())
>> + if (!gic_data.lpi_enabled)
>> return -EPERM;
>> irq_domain_set_info(d, irq, hw, &gic_chip, d->host_data,
>> handle_fasteoi_irq, NULL, NULL);
>> @@ -785,8 +781,12 @@ static int __init gic_of_init(struct device_node *node, struct device_node *pare
>>
>> set_handle_irq(gic_handle_irq);
>>
>> - if (IS_ENABLED(CONFIG_ARM_GIC_V3_ITS) && gic_dist_supports_lpis())
>> - its_init(node, &gic_data.rdists, gic_data.domain);
>> + if (IS_ENABLED(CONFIG_ARM_GIC_V3_ITS) &&
>> + !!(readl_relaxed(dist_base + GICD_TYPER) & GICD_TYPER_LPIS) &&
>> + !its_init(node, &gic_data.rdists, gic_data.domain))
>> + gic_data.lpi_enabled = 1;
>> + else
>> + gic_data.lpi_enabled = 0;
>>
>> gic_smp_init();
>> gic_dist_init();
>> --
>> 1.8.0
>>


--
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/