HDA controllers and MSI support.
Paul B. Mahol
onemda at gmail.com
Fri Oct 24 20:48:45 UTC 2008
On 10/24/08, Alexander Motin <mav at freebsd.org> wrote:
> Hi.
>
> I have tried and found that my ICH8-based system successfully supports
> Message Signaled Interrupts on HDA controller. It allows to avoid
> interrupt sharing and in theory gives some performance benefits.
>
> It is now disabled in driver by default, but can be enabled via
> hint.hdac.%d.msi device.hints control. You can see that MSI is in use if
> your hdacX device uses interrupt vector above 255.
>
> I am thinking, shouldn't we enable this feature by default?
>
> There should be no problem with both supporting and nonsupporting
> hardware. There could be only problems with broken hardware declaring
> fake features. So I would like to know, is there any of such cases for
> HDA hardware or not?
It works for me:
> vmstat -i
interrupt total rate
irq1: atkbd0 11234 9
irq9: acpi0 1532 1
irq12: psm0 81372 71
irq14: ata0 1 0
irq16: vgapci0 175701 153
irq17: ndis0+ 8240 7
cpu0: timer 2286240 1998
irq256: hdac0 99566 87
cpu1: timer 2286004 1998
Total 4949890 4326
hdac0 at pci0:0:27:0: class=0x040300 card=0x30a2103c chip=0x27d88086
rev=0x01 hdr=0x00
vendor = 'Intel Corporation'
device = '82801G (ICH7 Family) High Definition Audio'
class = multimedia
cap 01[50] = powerspec 2 supports D0 D3 current D0
cap 05[60] = MSI supports 1 message, 64 bit enabled with 1 message
cap 10[70] = PCI-Express 1 type 0
hdac0: <Intel 82801G High Definition Audio Controller> mem
0xf4580000-0xf4583fff irq 16 at device 27.0 on pci0
hdac0: HDA Driver Revision: 20081013_0113
hdac0: [ITHREAD]
hdac0: HDA Codec #0: Analog Devices AD1981HD
hdac0: HDA Codec #1: Lucent/Agere Systems (Unknown)
pcm0: <HDA Analog Devices AD1981HD PCM #0> at cad 0 nid 1 on hdac0
More information about the freebsd-multimedia
mailing list