RE: [Linuxarm] Re: [PATCH for-next 00/32] spin lock usage optimization for SCSI drivers

From: Finn Thain
Date: Thu Feb 11 2021 - 18:59:03 EST


On Thu, 11 Feb 2021, Song Bao Hua (Barry Song) wrote:

> > On Wed, 10 Feb 2021, Song Bao Hua (Barry Song) wrote:
> >
> > > > On Wed, 10 Feb 2021, Song Bao Hua (Barry Song) wrote:
> > > >
> > > > > TBH, that is why m68k is so confusing. irqs_disabled() on m68k
> > > > > should just reflect the status of all interrupts have been
> > > > > disabled except NMI.
> > > > >
> > > > > irqs_disabled() should be consistent with the calling of APIs
> > > > > such as local_irq_disable, local_irq_save, spin_lock_irqsave
> > > > > etc.
> > > > >
> > > >
> > > > When irqs_disabled() returns true, we cannot infer that
> > > > arch_local_irq_disable() was called. But I have not yet found
> > > > driver code or core kernel code attempting that inference.
> > > >
> > > > > >
> > > > > > > Isn't arch_irqs_disabled() a status reflection of irq
> > > > > > > disable API?
> > > > > > >
> > > > > >
> > > > > > Why not?
> > > > >
> > > > > If so, arch_irqs_disabled() should mean all interrupts have been
> > > > > masked except NMI as NMI is unmaskable.
> > > > >
> > > >
> > > > Can you support that claim with a reference to core kernel code or
> > > > documentation? (If some arch code agrees with you, that's neither
> > > > here nor there.)
> > >
> > > I think those links I share you have supported this. Just you don't
> > > believe :-)
> > >
> >
> > Your links show that the distinction between fast and slow handlers
> > was removed. Your links don't support your claim that
> > "arch_irqs_disabled() should mean all interrupts have been masked".
> > Where is the code that makes that inference? Where is the
> > documentation that supports your claim?
>
> (1)
> https://lwn.net/Articles/380931/
> Looking at all these worries, one might well wonder if a system which
> *disabled interrupts for all handlers* would function well at all. So it
> is interesting to note one thing: any system which has the lockdep
> locking checker enabled has been running all handlers that way for some
> years now. Many developers and testers run lockdep-enabled kernels, and
> they are available for some of the more adventurous distributions
> (Rawhide, for example) as well. So we have quite a bit of test coverage
> for this mode of operation already.
>

IIUC, your claim is that CONFIG_LOCKDEP involves code that contains the
inference, "arch_irqs_disabled() means all interrupts have been masked".

Unfortunately, m68k lacks CONFIG_LOCKDEP support so I can't easily confirm
this. I suppose there may be other architectures that support both LOCKDEP
and nested interrupts (?)

Anyway, if you would point to the code that contains said inference, that
would help a lot.

> (2)
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=b738a50a
>
> "We run all handlers *with interrupts disabled* and expect them not to
> enable them. Warn when we catch one who does."
>

Again, you don't see that warning because irqs_disabled() correctly
returns true. You can confirm this fact in QEMU or Aranym if you are
interested.

> (3)
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=e58aa3d2d0cc
> genirq: Run irq handlers *with interrupts disabled*
>
> Running interrupt handlers with interrupts enabled can cause stack
> overflows. That has been observed with multiqueue NICs delivering all
> their interrupts to a single core. We might band aid that somehow by
> checking the interrupt stacks, but the real safe fix is to *run the irq
> handlers with interrupts disabled*.
>

Again, the stack overflow issue is not applicable. 68000 uses a priority
mask, like ARM GIC. So there's no arbitrary nesting of interrupt handlers.

In practice stack overflows simply don't occur on m68k. Please do try it.

>
> All these documents say we are running irq handler with interrupts
> disabled. but it seems you think high-prio interrupts don't belong
> to "interrupts" in those documents :-)
>
> that is why we can't get agreement. I think "interrupts" mean all except
> NMI in these documents, but you insist high-prio IRQ is an exception.
>

We can't get agreement because you seek to remove functionality without
justification.

> > > >
> > > > > >
> > > > > > Are all interrupts (including NMI) masked whenever
> > > > > > arch_irqs_disabled() returns true on your platforms?
> > > > >
> > > > > On my platform, once irqs_disabled() is true, all interrupts are
> > > > > masked except NMI. NMI just ignore spin_lock_irqsave or
> > > > > local_irq_disable.
> > > > >
> > > > > On ARM64, we also have high-priority interrupts, but they are
> > > > > running as PESUDO_NMI:
> > > > > https://lwn.net/Articles/755906/
> > > > >
> > > >
> > > > A glance at the ARM GIC specification suggests that your hardware
> > > > works much like 68000 hardware.
> > > >
> > > > When enabled, a CPU interface takes the highest priority
> > > > pending interrupt for its connected processor and determines
> > > > whether the interrupt has sufficient priority for it to signal
> > > > the interrupt request to the processor. [...]
> > > >
> > > > When the processor acknowledges the interrupt at the CPU
> > > > interface, the Distributor changes the status of the interrupt
> > > > from pending to either active, or active and pending. At this
> > > > point the CPU interface can signal another interrupt to the
> > > > processor, to preempt interrupts that are active on the
> > > > processor. If there is no pending interrupt with sufficient
> > > > priority for signaling to the processor, the interface
> > > > deasserts the interrupt request signal to the processor.
> > > >
> > > > https://developer.arm.com/documentation/ihi0048/b/
> > > >
> > > > Have you considered that Linux/arm might benefit if it could fully
> > > > exploit hardware features already available, such as the interrupt
> > > > priority masking feature in the GIC in existing arm systems?
> > >
> > > I guess no:-) there are only two levels: IRQ and NMI. Injecting a
> > > high-prio IRQ level between them makes no sense.
> > >
> > > To me, arm64's design is quite clear and has no any confusion.
> > >
> >
> > Are you saying that the ARM64 hardware design is confusing because it
> > implements a priority mask, and that's why you had to simplify it with
> > a pseudo-nmi scheme in software?
>
> No, I was not saying this. I think both m68k and arm64 have good
> hardware design. Just Linux's implementation is running irq-handlers
> with interrupts disabled. So ARM64's pseudo-nmi is adapted to Linux
> better.
>

So, a platform should do what all the other platforms do because to
deviate would be too dangerous?

> > > >
> > > > > On m68k, it seems you mean:
> > > > > irq_disabled() is true, but high-priority interrupts can still
> > > > > come; local_irq_disable() can disable high-priority interrupts,
> > > > > and at that time, irq_disabled() is also true.
> > > > >
> > > > > TBH, this is wrong and confusing on m68k.
> > > > >
> > > >
> > > > Like you, I was surprised when I learned about it. But that
> > > > doesn't mean it's wrong. The fact that it works should tell you
> > > > something.
> > > >
> > >
> > > The fact is that m68k lets arch_irq_disabled() return true to
> > > pretend all IRQs are disabled while high-priority IRQ is still open,
> > > thus "pass" all sanitizing check in genirq and kernel core.
> > >
> >
> > The fact is that m68k has arch_irq_disabled() return false when all
> > IRQs are enabled. So there is no bug.
>
> But it has arch_irq_disabled() return true while some interrupts(not
> NMI) are still open.
>
> >
> > > > Things could always be made simpler. But discarding features isn't
> > > > necessarily an improvement.
> > >
> > > This feature could be used by calling local_irq_enable_in_hardirq()
> > > in those IRQ handlers who hope high-priority interrupts to preempt
> > > it for a while.
> > >
> >
> > So, if one handler is sensitive to interrupt latency, all other
> > handlers should be modified? I don't think that's workable.
>
> I think we just enable preempt_rt or force threaded_irq, and then
> improve the priority of the irq thread who is sensitive to latency. No
> need to touch all threads.
>
> I also understand your point, we let one high-prio interrupt preempt low
> priority interrupt, then we don't need to change the whole system. But I
> think Linux prefers the method of threaded_irq or preempt_rt for this
> kind of problems.
>

So, some interrupt (or exception) processing happens atomically and the
rest is deferred to a different execution context. (Not a new idea.)

If you introduce latency in the former context you can't win it back in
the latter. Your solution fails because it adds latency to high priority
handlers.

> >
> > In anycase, what you're describing is a completely different nested
> > interrupt scheme that would defeat the priority level mechanism that
> > the hardware provides us with.
>
> Yes. Indeed.
>
> >
> > > It shouldn't hide somewhere and make confusion.
> > >
> >
> > The problem is hiding so well that no-one has found it! I say it
> > doesn't exist.
>
> Long long ago(before 2.6.38), we had a kernel supporting IRQF_DISABLED
> and nested interrupts were widely supported, but system also ran well in
> most cases. That means nested interrupts don't really matter in most
> cases. That is why m68k is also running well even though it is still
> nesting.
>

No, m68k runs well because it uses priority masking. It is not because
some cases are untested.

Your hardware may not have been around for 4 decades but it implements the
same capability because the design is known to work.

> >
> > > On the other hand, those who care about realtime should use threaded
> > > IRQ and let IRQ threads preempt each other.
> > >
> >
> > Yes. And those threads also have priority levels.
>
> Finn, I am not a m68k guy, would you help check if this could activate a
> warning on m68k. maybe we can discuss this question in genirq maillist from
> this warning if you are happy. Thanks very much.
>
> diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h
> index 7c9d6a2d7e90..b8ca27555c76 100644
> --- a/include/linux/hardirq.h
> +++ b/include/linux/hardirq.h
> @@ -32,6 +32,7 @@ static __always_inline void rcu_irq_enter_check_tick(void)
> */
> #define __irq_enter() \
> do { \
> + WARN_ONCE(in_hardirq() && irqs_disabled(), "nested interrupts\n"); \
> preempt_count_add(HARDIRQ_OFFSET); \
> lockdep_hardirq_enter(); \
> account_hardirq_enter(current); \
> @@ -44,6 +45,7 @@ static __always_inline void rcu_irq_enter_check_tick(void)
> */
> #define __irq_enter_raw() \
> do { \
> + WARN_ONCE(in_hardirq() && irqs_disabled(), "nested interrupts\n"); \
> preempt_count_add(HARDIRQ_OFFSET); \
> lockdep_hardirq_enter(); \
> } while (0)
>

If you think that lockdep or some other code somewhere should be protected
in this way, perhaps you can point to that code. Otherwise, your patch
seems to lack any justification.

> Best Regards
> Barry
>
>