Problem? intel_iommu=off; perf top shows acpi_os_read_port as extremely busy
in [Kernel]
|
Prev: V4L/DVB updates
Next: [git pull] FireWire fixes
From: Michael Breuer on 28 Nov 2009 13:20 Ok - my only question then is why things appear so different with intel_iommu enabled. Arjan van de Ven wrote: > On Sat, 28 Nov 2009 08:18:08 +0100 > Ingo Molnar <mingo(a)elte.hu> wrote: > > >> * Michael Breuer <mbreuer(a)majjas.com> wrote: >> >> >>> Having given up for now on VT-D, I rebooted 2.6.38 rc8 with >>> intel_iommu=off. Whilst my myriad of broken bios issues cleared, I >>> now see in perf top acpi_os_read_port as continually the busiest >>> function. With intel_iommu enabled, _spin_lock was always on top, >>> and nothing else was notable. >>> >>> This seems odd to me, perhaps this will make sense to someone else. >>> >>> FWIW, I'm running on an Asus p6t deluxe v2; ht enabled; no errors >>> or oddities in dmesg or /var/log/messages. >>> >> Could you post the perf top output please? >> >> Also, could you also post the output of: >> >> perf stat -a --repeat 10 sleep 1 >> >> this will show us how idle the system is. (My guess is that your >> system is idle and perf top shows acpi_os_read_port because the >> system goes to idle via ACPI methods and PIO is slow. In that case >> all is nominal and your system is fine. But it's hard to tell without >> more details.) >> >> > > yeah the os_read_port is part of the idle loop, so if your system is > idle it'll show up big.... not much we can optimize there though... > -- > 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/ > -- 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: Arjan van de Ven on 29 Nov 2009 15:50 On Sat, 28 Nov 2009 13:10:21 -0500 Michael Breuer <mbreuer(a)majjas.com> wrote: > Ok - my only question then is why things appear so different with > intel_iommu enabled. something else is even more expensive then :0 -- Arjan van de Ven Intel Open Source Technology Centre For development, discussion and tips for power savings, visit http://www.lesswatts.org -- 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: Michael Breuer on 30 Nov 2009 00:20
Ok - one more rather odd (to me) data point... I started playing around with various settings, and traced the calls to acpi_os_read_port. To summarize: With intel_iommu=off, I see a large percentage of calls to acpi_os_read_port resulting from user apps (portsentry is #1). With intel_iommu=on, NONE of trace points to any user apps - all derive from the idle loop. To make things more interesting, when I enable intel_iommu and disable vt-d in bios, the system performs much better (20% improvement in glxgears, for example), perf top looks like this: ------------------------------------------------------------------------------ PerfTop: 4863 irqs/sec kernel:62.7% [100000 cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt kernel function _______ _____ _______________ 2213.00 - 5.5% : acpi_idle_enter_bm 2001.00 - 5.0% : acpi_os_read_port 1544.00 - 3.9% : _spin_lock_irqsave 1075.00 - 2.7% : ioread32 928.00 - 2.3% : find_busiest_group 851.00 - 2.1% : _spin_unlock_irqrestore 823.00 - 2.1% : hpet_next_event 810.00 - 2.0% : tg_shares_up 655.00 - 1.6% : fget_light 641.00 - 1.6% : schedule 639.00 - 1.6% : tick_nohz_stop_sched_tick 638.00 - 1.6% : sub_preempt_count 634.00 - 1.6% : add_preempt_count 548.00 - 1.4% : do_sys_poll 446.00 - 1.1% : trace_hardirqs_off And additionally, one recurring boot warning I've seen since I first booted this box has disappeared - first boot message of IRQ16 disabled. I'm thinking that bad VT-D bios is causing trouble even when intel_iommu is disabled. On 11/29/2009 03:47 PM, Arjan van de Ven wrote: > On Sat, 28 Nov 2009 13:10:21 -0500 > Michael Breuer<mbreuer(a)majjas.com> wrote: > > >> Ok - my only question then is why things appear so different with >> intel_iommu enabled. >> > something else is even more expensive then :0 > > > -- 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/ |