x86 apic: Ack all pending irqs when crashed/on kexec - V5
in [Kernel]
|
Prev: 2.6.35-rc3 - include/linux/fdtable.h:85 invoked rcu_dereference_check() without protection!
Next: [PATCH] vfs repost: fix RCU-lockdep false positive due to /proc access
From: H. Peter Anvin on 16 Jun 2010 20:30 On 06/16/2010 02:11 PM, Pan, Jacob jun wrote: > > W.R.T. the loop limits, is it possible to use a default max_loops value in > case when cpu_khz is not set? The reason is that on Moorestown platform > we need to do an early APIC setup before tsc_init(), so cpu_khz is 0 at the > time we setup local APIC. The result is that we hit WARN_ON(max_loops<= 0) > on Moorestown for early APIC setup. > > The early APIC setup is needed because Moorestown does not have a PIT and the > system timer interrupts are routed via IOAPIC. > Can't MRST install a quick ballpark value into cpu_khz? -hpa -- H. Peter Anvin, Intel Open Source Technology Center I work for Intel. I don't speak on their behalf. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: H. Peter Anvin on 17 Jun 2010 13:10
On 06/17/2010 09:52 AM, Pan, Jacob jun wrote: >> On 06/16/2010 02:11 PM, Pan, Jacob jun wrote: >>> >>> W.R.T. the loop limits, is it possible to use a default max_loops >> value in >>> case when cpu_khz is not set? The reason is that on Moorestown >> platform >>> we need to do an early APIC setup before tsc_init(), so cpu_khz is 0 >> at the >>> time we setup local APIC. The result is that we hit >> WARN_ON(max_loops<= 0) >>> on Moorestown for early APIC setup. >>> >>> The early APIC setup is needed because Moorestown does not have a PIT >> and the >>> system timer interrupts are routed via IOAPIC. >>> >> >> Can't MRST install a quick ballpark value into cpu_khz? >> > yes, we can do that to avoid the warning. the true cpu_khz can then be set > in tsc_init by platform specific calibration code. That is one option. > Seems like a reasonable thing to do to me. -hpa -- H. Peter Anvin, Intel Open Source Technology Center I work for Intel. I don't speak on their behalf. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ |