When the best selected cpu is offline, work_on_cpu will stuck
forever. Therefore, in this case, we should call
local_pci_probe instead of work_on_cpu.
Cc: <[email protected]>
Signed-off-by: Huacai Chen <[email protected]>
Signed-off-by: Hongchen Zhang <[email protected]>
---
drivers/pci/pci-driver.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index af2996d0d17f..32a99828e6a3 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -386,7 +386,7 @@ static int pci_call_probe(struct pci_driver *drv, struct pci_dev *dev,
free_cpumask_var(wq_domain_mask);
}
- if (cpu < nr_cpu_ids)
+ if ((cpu < nr_cpu_ids) && cpu_online(cpu))
error = work_on_cpu(cpu, local_pci_probe, &ddi);
else
error = local_pci_probe(&ddi);
--
2.33.0
Hi, Hongchen,
On Wed, May 29, 2024 at 7:19 PM Hongchen Zhang
<[email protected]> wrote:
The title should be better like this:
PCI: Use local_pci_probe() when best selected CPU is offline
>
> When the best selected cpu is offline, work_on_cpu will stuck
> forever. Therefore, in this case, we should call
> local_pci_probe instead of work_on_cpu.
It is better to reword like this:
When the best selected CPU is offline, work_on_cpu() will stuck forever.
This can be happen if a node is online while all its CPUs are offline
(we can use "maxcpus=1" without "nr_cpus=1" to reproduce it), Therefore,
in this case, we should call local_pci_probe() instead of work_on_cpu().
Huacai
>
> Cc: <[email protected]>
> Signed-off-by: Huacai Chen <[email protected]>
> Signed-off-by: Hongchen Zhang <[email protected]>
> ---
> drivers/pci/pci-driver.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
> index af2996d0d17f..32a99828e6a3 100644
> --- a/drivers/pci/pci-driver.c
> +++ b/drivers/pci/pci-driver.c
> @@ -386,7 +386,7 @@ static int pci_call_probe(struct pci_driver *drv, struct pci_dev *dev,
> free_cpumask_var(wq_domain_mask);
> }
>
> - if (cpu < nr_cpu_ids)
> + if ((cpu < nr_cpu_ids) && cpu_online(cpu))
> error = work_on_cpu(cpu, local_pci_probe, &ddi);
> else
> error = local_pci_probe(&ddi);
> --
> 2.33.0
>
>
On 2024/6/4 下午10:57, Huacai Chen wrote:
> Hi, Hongchen,
>
> On Wed, May 29, 2024 at 7:19 PM Hongchen Zhang
> <[email protected]> wrote:
> The title should be better like this:
> PCI: Use local_pci_probe() when best selected CPU is offline
>
>>
>> When the best selected cpu is offline, work_on_cpu will stuck
>> forever. Therefore, in this case, we should call
>> local_pci_probe instead of work_on_cpu.
>
> It is better to reword like this:
>
> When the best selected CPU is offline, work_on_cpu() will stuck forever.
> This can be happen if a node is online while all its CPUs are offline
> (we can use "maxcpus=1" without "nr_cpus=1" to reproduce it), Therefore,
> in this case, we should call local_pci_probe() instead of work_on_cpu().
>
OK, Let me modify the log and send V2.
> Huacai
>
>>
>> Cc: <[email protected]>
>> Signed-off-by: Huacai Chen <[email protected]>
>> Signed-off-by: Hongchen Zhang <[email protected]>
>> ---
>> drivers/pci/pci-driver.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
>> index af2996d0d17f..32a99828e6a3 100644
>> --- a/drivers/pci/pci-driver.c
>> +++ b/drivers/pci/pci-driver.c
>> @@ -386,7 +386,7 @@ static int pci_call_probe(struct pci_driver *drv, struct pci_dev *dev,
>> free_cpumask_var(wq_domain_mask);
>> }
>>
>> - if (cpu < nr_cpu_ids)
>> + if ((cpu < nr_cpu_ids) && cpu_online(cpu))
>> error = work_on_cpu(cpu, local_pci_probe, &ddi);
>> else
>> error = local_pci_probe(&ddi);
>> --
>> 2.33.0
>>
>>
--
Best Regards
Hongchen Zhang