Date: Mon, 21 Nov 2005 11:46:17 -0500 From: John Baldwin <jhb@freebsd.org> To: guru@sisis.de Cc: freebsd-mobile@freebsd.org Subject: Re: high CPU activity for interrupts Message-ID: <200511211146.18646.jhb@freebsd.org> In-Reply-To: <20051119073610.GA1409@rebelion.Sisis.de> References: <20051118123825.GA16774@rebelion.Sisis.de> <200511181340.42888.jhb@freebsd.org> <20051119073610.GA1409@rebelion.Sisis.de>
next in thread | previous in thread | raw e-mail | index | archive | help
On Saturday 19 November 2005 02:36 am, guru@sisis.de wrote: > El d=EDa Friday, November 18, 2005 a las 01:40:42PM -0500, John Baldwin e= scribi=F3: > > > The high rate of interrupts are coming from cardbus somehow: > > > > > > $ vmstat -i > > > interrupt total rate > > > irq0: clk 1526386 99 > > > irq1: atkbd0 19720 1 > > > irq4: sio0 2 0 > > > irq6: fdc0 10 0 > > > irq7: 3 0 > > > stray irq7 3 0 > > > irq8: rtc 1953511 127 > > > irq9: acpi0 1102 0 > > > irq11: cbb1 pcm0++* 126785936 8304 <*********** > > > irq12: psm0 116294 7 > > > irq14: ata0 70657 4 > > > irq15: ata1 33673 2 > > > Total 130507297 8548 > > > > The '++*' part means you have at least 4 other devices sharing that IRQ > > line as well. Can you do a "dmesg | grep 'irq 11'" and post the output? > > Thanks for the feedback and here it is; > > matthias > > fwohci0: <SiS 7007> mem 0xec000000-0xec000fff irq 11 at device 2.3 on pci0 > ohci0: <SiS 5571 USB controller> mem 0xec001000-0xec001fff irq 11 at devi= ce 3.0 on pci0 > ohci1: <SiS 5571 USB controller> mem 0xec002000-0xec002fff irq 11 at devi= ce 3.1 on pci0 > ohci2: <SiS 5571 USB controller> mem 0xec003000-0xec003fff irq 11 at devi= ce 3.2 on pci0 > sis0: <SiS 900> 10/100BaseTX> port 0x2000-0x20ff mem 0xec005000-0xec005ff= f irq 11 at > device 4.0 on pci0 > pcm0: <SiS 7012> port 0x1800-0x187f,0x1c00-0x1cff irq 11 at device 2.7 on= pci0 > pcm0: <SiS 7012> port 0x1800-0x187f,0x1c00-0x1cff irq 11 at device 2.7 on= pci0 > drm0: <ATI Radeon Lf R250 Mobility 9000 M9> port 0xa000-0xa0ff mem 0xec10= 0000-0xec10ffff,0xf0000000-0xf7ffffff irq 11 at > device 0.0 on pci1 Ok, so the high number of interrupts can be coming from any one of these de= vices. You can try building a custom kernel that doesn't include the various devic= es (such as ohci, pcm, or radeon_drm) to see which devices are generating the = large number of interrupts. =2D-=20 John Baldwin <jhb@FreeBSD.org> <>< http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve" =3D http://www.FreeBSD.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200511211146.18646.jhb>