Re: MSI interrupts and disable_irq

["Yinghai Lu" <yhlu.kernel@gmail.com>]

On 10/5/07, Eric W. Biederman <ebiederm@xmission.com> wrote:
> Stephen Hemminger <shemminger@linux-foundation.org> writes:
>
> > On Fri, 28 Sep 2007 22:47:16 -0400
> > Jeff Garzik <jgarzik@pobox.com> wrote:
> >
> >> Ayaz Abdulla wrote:
> >> > I am trying to track down a forcedeth driver issue described by bug 9047
> >> > in bugzilla (2.6.23-rc7-git1 forcedeth w/ MCP55 oops under heavy load).
> >> > I added a patch to synchronize the timer handlers so that one handler
> >> > doesn't accidently enable the IRQ while another timer handler is running
> >> > (see attachment 'Add timer lock' in bug report) and for other processing
> >> > protection.
> >> >
> >> > However, the system still had an Oops. So I added a lock around the
> >> > nv_rx_process_optimized() and the Oops has not happened (see attachment
> >> > 'New patch for locking' in bug report). This would imply a
> >> > synchronization issue. However, the only callers of that function are
> >> > the IRQ handler and the timer handlers (in non-NAPI case). The timer
> >> > handlers  use disable_irq so that the IRQ handler does not contend with
> >> > them. It looks as if disable_irq is not working properly.
> >> >
> >> > This issue repros only with MSI interrupt and not legacy INTx
> >> > interrupts. Any ideas?
> >>
> >> (added linux-kernel to CC, since I think it's more of a general kernel
> >> issue)
>
> I didn't see anything in disable_irq that would cause it to fail in
> the suggested way.  But I couldn't quite convince myself we were
> race free either.  I didn't see anything that was specific to MSI
> that would cause something.  But switching from level to edge
> triggered, and to a lower latency delivery path may have caused
> some behavior changes.
>
> >> To be brutally frank, I always thought this disable_irq() mess was a
> >> hack both ugly and fragile.  This disable_irq() work that appeared in a
> >> couple net drivers was correct at the time, so I didn't feel I had the
> >> justification to reject it, but it still gave me a bad feeling.
> >>
> >> I think the scenario you outline is an illustration of the approach's
> >> fragility:  disable_irq() is a heavy hammer that originated with INTx,
> >> and it relies on a chip-specific disable method (kernel/irq/manage.c)
> >> that practically guarantees behavior will vary across MSI/INTx/etc.
> >>
> >>
> >> Based on your report, it is certainly possible that there is a problem
> >> with MSI's desc->chip->disable() method...  but I would actually
> >> recommend working around the problem by making the forcedeth locking
> >> more standardized by removing all those disable_irq() hacks.
> >>
> >
> > I'll try and clean it up if the author doesn't get to it first.
>
> I took a look at the underlying side of this.
>
> I don't know if the MSI capability for the forcedeth supports a mask
> bit or not.  Mine doesn't even have a msi capability.  If it doesn't
> support a mask bit the pci spec provides not valid way to mask the
> interrupt, so what we do is actually disable the msi capability.
> At which point we might get weird INTx interactions.
>
> We have a similar case with ioapics and INTx that also turns
> a hardware level disable into a reroute to another irq command.
> So I'm going to take a look and see how infrequently we can use
> hardware level disabled.

also the driver need to have seperate handlers for ioapic, msi, msi-x...
current nv_nic_irq_optimized keep checking msi_flag...

also nv_enable_hw_interrupts and nv_disable_hw_interrupts is not
corresponding in MSI side..

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