Re: High interrupt count on OpenBSD 4.3/amd64 with driver azalia

2008-09-03 Thread Gabriel Linder 
On Thu, 28 Aug 2008 10:49:48 +0200
Gabriel Linder [EMAIL PROTECTED] wrote:

 Hi,
 
 I use a desktop system powered by OpenBSD 4.3-stable with GENERIC.MP
 kernel (amd64, up to date as of now). I listen music with mplayer
 while I work, and sometime my system feel very slow. A top show that
 CPU0 is spending 80% time in interrupts, and leaving mplayer calm the
 system down. Is there anything I can do to avoid that, or any
 information I can provide to help fix it ?

Just tried with a GENERIC kernel, which froze the system after ten
minutes of uptime. Dmesg extracted from /var/log/messages follows. I
will try with -current as soon as I can.

Sep  3 08:52:04 oblivion syslogd: start
Sep  3 08:52:04 oblivion /bsd: OpenBSD 4.3-stable (bsd) #3: Tue Sep  2 14:19:19 
CEST 2008
Sep  3 08:52:04 oblivion /bsd: [EMAIL PROTECTED]:/root/bsd
Sep  3 08:52:04 oblivion /bsd: real mem = 2111426560 (2013MB)
Sep  3 08:52:04 oblivion /bsd: avail mem = 2038685696 (1944MB)
Sep  3 08:52:04 oblivion /bsd: RTC BIOS diagnostic error 11memory_size
Sep  3 08:52:04 oblivion /bsd: mainbus0 at root
Sep  3 08:52:04 oblivion /bsd: bios0 at mainbus0: SMBIOS rev. 2.5 @ 0xf0450 (80 
entries)
Sep  3 08:52:04 oblivion /bsd: bios0: vendor Dell Inc. version A09 date 
03/11/2008
Sep  3 08:52:04 oblivion /bsd: bios0: Dell Inc. OptiPlex 755
Sep  3 08:52:04 oblivion /bsd: acpi0 at bios0: rev 2
Sep  3 08:52:04 oblivion /bsd: acpi0: tables DSDT FACP SSDT APIC BOOT ASF! MCFG 
HPET TCPA  SLIC SSDT SSDT SSDT
Sep  3 08:52:04 oblivion /bsd: acpi0: wakeup devices VBTN(S4) PCI0(S5) PCI4(S5) 
PCI2(S5) PCI3(S5) PCI1(S5) PCI5(S5) PCI6(S5) MOU_(S3) USB0(S3) USB1(S3) 
USB2(S3) USB3(S3) USB4(S3) USB5(S3)
Sep  3 08:52:04 oblivion /bsd: acpitimer0 at acpi0: 3579545 Hz, 24 bits
Sep  3 08:52:04 oblivion /bsd: acpihpet0 at acpi0: 14318179 Hz
Sep  3 08:52:04 oblivion /bsd: acpiprt0 at acpi0: bus 3 (PCI4)
Sep  3 08:52:04 oblivion /bsd: acpiprt1 at acpi0: bus 2 (PCI2)
Sep  3 08:52:04 oblivion /bsd: acpiprt2 at acpi0: bus -1 (PCI3)
Sep  3 08:52:04 oblivion /bsd: acpiprt3 at acpi0: bus 1 (PCI1)
Sep  3 08:52:04 oblivion /bsd: acpiprt4 at acpi0: bus -1 (PCI5)
Sep  3 08:52:04 oblivion /bsd: acpiprt5 at acpi0: bus -1 (PCI6)
Sep  3 08:52:04 oblivion /bsd: acpiprt6 at acpi0: bus 0 (PCI0)
Sep  3 08:52:04 oblivion /bsd: acpicpu0 at acpi0: FVS, 2667, 2333, 2000 MHz
Sep  3 08:52:04 oblivion /bsd: acpibtn0 at acpi0: VBTN
Sep  3 08:52:04 oblivion /bsd: cpu0 at mainbus0: (uniprocessor)
Sep  3 08:52:04 oblivion /bsd: cpu0: Intel(R) Core(TM)2 Duo CPU E8200 @ 
2.66GHz, 2660.33 MHz
Sep  3 08:52:04 oblivion /bsd: cpu0: 
FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,SBF,SSE3,MWAIT,DS-CPL,VMX,EST,TM2,CX16,xTPR,NXE,LONG
Sep  3 08:52:04 oblivion /bsd: cpu0: 6MB 64b/line 16-way L2 cache
Sep  3 08:52:04 oblivion /bsd: pci0 at mainbus0 bus 0: configuration mode 1
Sep  3 08:52:04 oblivion /bsd: pchb0 at pci0 dev 0 function 0 Intel 82Q35 
Host rev 0x02
Sep  3 08:52:04 oblivion /bsd: ppb0 at pci0 dev 1 function 0 Intel 82Q35 PCIE 
rev 0x02: irq 11
Sep  3 08:52:04 oblivion /bsd: pci1 at ppb0 bus 1
Sep  3 08:52:04 oblivion /bsd: vga1 at pci1 dev 0 function 0 ATI Radeon HD 
2400 XT rev 0x00
Sep  3 08:52:04 oblivion /bsd: wsdisplay0 at vga1 mux 1: console (80x25, vt100 
emulation)
Sep  3 08:52:04 oblivion /bsd: wsdisplay0: screen 1-5 added (80x25, vt100 
emulation)
Sep  3 08:52:04 oblivion /bsd: Intel 82Q35 HECI rev 0x02 at pci0 dev 3 
function 0 not configured
Sep  3 08:52:04 oblivion /bsd: pciide0 at pci0 dev 3 function 2 Intel 82Q35 PT 
IDER rev 0x02: DMA (unsupported), channel 0 wired to native-PCI, channel 1 
wired to native-PCI
Sep  3 08:52:04 oblivion /bsd: pciide0: using irq 9 for native-PCI interrupt
Sep  3 08:52:04 oblivion /bsd: pciide0: channel 0 ignored (not responding; 
disabled or no drives?)
Sep  3 08:52:04 oblivion /bsd: pciide0: channel 1 ignored (not responding; 
disabled or no drives?)
Sep  3 08:52:04 oblivion /bsd: Intel 82Q35 KT rev 0x02 at pci0 dev 3 function 
3 not configured
Sep  3 08:52:04 oblivion /bsd: em0 at pci0 dev 25 function 0 Intel ICH9 IGP 
AMT rev 0x02: irq 4, address 00:1e:4f:c5:11:9f
Sep  3 08:52:04 oblivion /bsd: uhci0 at pci0 dev 26 function 0 Intel 82801I 
USB rev 0x02: irq 11
Sep  3 08:52:04 oblivion /bsd: uhci1 at pci0 dev 26 function 1 Intel 82801I 
USB rev 0x02: irq 5
Sep  3 08:52:04 oblivion /bsd: ehci0 at pci0 dev 26 function 7 Intel 82801I 
USB rev 0x02: irq 5
Sep  3 08:52:04 oblivion /bsd: usb0 at ehci0: USB revision 2.0
Sep  3 08:52:04 oblivion /bsd: uhub0 at usb0 Intel EHCI root hub rev 
2.00/1.00 addr 1
Sep  3 08:52:04 oblivion /bsd: azalia0 at pci0 dev 27 function 0 Intel 82801I 
HD Audio rev 0x02: irq 11
Sep  3 08:52:04 oblivion /bsd: azalia0: RIRB time out
Sep  3 08:52:04 oblivion /bsd: azalia0: RIRB time out
Sep  3 08:52:04 oblivion /bsd: azalia0: codec[s]: Analog Devices AD1984
Sep  3 08:52:04 oblivion /bsd: audio0 at azalia0
Sep  3 08:52:04 oblivion /bsd: ppb1 at pci0 dev 28

Re: High interrupt count on OpenBSD 4.3/amd64 with driver azalia

2008-09-03 Thread Ariane van der Steldt 
On Wed, Sep 03, 2008 at 09:42:06AM +0200, Gabriel Linder wrote:
 On Thu, 28 Aug 2008 10:49:48 +0200
 Gabriel Linder [EMAIL PROTECTED] wrote:
  I use a desktop system powered by OpenBSD 4.3-stable with GENERIC.MP
  kernel (amd64, up to date as of now). I listen music with mplayer
  while I work, and sometime my system feel very slow. A top show that
  CPU0 is spending 80% time in interrupts, and leaving mplayer calm the
  system down. Is there anything I can do to avoid that, or any
  information I can provide to help fix it ?
 
 Just tried with a GENERIC kernel, which froze the system after ten
 minutes of uptime. Dmesg extracted from /var/log/messages follows. I
 will try with -current as soon as I can.

I think the interrupt load with a -current or -snapshot will be better.
I can confirm your high interrupt load (my opteron has an interrupt load
hovering between 70% and 80% in GENERIC.MP).

Ariane

Re: High interrupt count on OpenBSD 4.3/amd64 with driver azalia

2008-09-03 Thread Gabriel Linder 
On Wed, 3 Sep 2008 10:37:28 +0200
Ariane van der Steldt [EMAIL PROTECTED] wrote:

 I think the interrupt load with a -current or -snapshot will be
 better. I can confirm your high interrupt load (my opteron has an
 interrupt load hovering between 70% and 80% in GENERIC.MP).

Thanks for your feedback. I am running -current for five hours now,
and azalia has a normal interrupt count.

OpenBSD oblivion.jeuxvideo.com 4.4 GENERIC.MP#1827 amd64