On Fri, May 08, 2020 at 04:26:17PM -0700, Bart Van Assche wrote:
> On 2020-05-04 19:09, Ming Lei wrote:
> > @@ -391,6 +393,7 @@ struct blk_mq_ops {
> > enum {
> > BLK_MQ_F_SHOULD_MERGE = 1 << 0,
> > BLK_MQ_F_TAG_SHARED = 1 << 1,
> > + BLK_MQ_F_NO_MANAGED_IRQ = 1 << 2,
> > BLK_MQ_F_BLOCKING = 1 << 5,
> > BLK_MQ_F_NO_SCHED = 1 << 6,
> > BLK_MQ_F_ALLOC_POLICY_START_BIT = 8,
> > diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h
> > index 77d70b633531..24b3a77810b6 100644
> > --- a/include/linux/cpuhotplug.h
> > +++ b/include/linux/cpuhotplug.h
> > @@ -152,6 +152,7 @@ enum cpuhp_state {
> > CPUHP_AP_SMPBOOT_THREADS,
> > CPUHP_AP_X86_VDSO_VMA_ONLINE,
> > CPUHP_AP_IRQ_AFFINITY_ONLINE,
> > + CPUHP_AP_BLK_MQ_ONLINE,
> > CPUHP_AP_ARM_MVEBU_SYNC_CLOCKS,
> > CPUHP_AP_X86_INTEL_EPB_ONLINE,
> > CPUHP_AP_PERF_ONLINE,
>
> Wouldn't BLK_MQ_F_NO_IRQ be a better name than BLK_MQ_F_NO_MANAGED_IRQ?
No, what this patchset tries to do is to address request timeout or hang
issue in case that managed irq is applied in blk-mq driver.
>
> Please add comments that explain what BLK_MQ_F_NO_MANAGED_IRQ and
> CPUHP_AP_BLK_MQ_ONLINE mean.
I believe both are self-document:
- BLK_MQ_F_NO_MANAGED_IRQ
No managened irq is applied on the driver
- CPUHP_AP_BLK_MQ_ONLINE
It is a standard naming of cpuhp state, which means the startup and
teardown callbacks are called when the passed CPU is still online.
Thanks,
Ming
