[linux-dvb] How might a SATA drive interfere with DVB-S

[Reinhard Nissl]

Hi,

Jon Burgess wrote:

>> Attached you'll find my lspci -vv. Interrupts seem to be different, 
>> but is a latency of 0 ok for the SATA device?
>>
>> ------------------------------------------------------------------------
>>
>> 00:1f.2 IDE interface: Intel Corporation 82801EB (ICH5) SATA 
>> Controller (rev 02) (prog-if 8a [Master SecP PriP])
>>     Subsystem: Fujitsu Siemens Computer GmbH: Unknown device 101c
>>     Control: I/O+ Mem- BusMaster+ SpecCycle- MemWINV- VGASnoop- 
>> ParErr- Stepping- SERR- FastB2B-
>>     Status: Cap- 66MHz+ UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- 
>> <TAbort- <MAbort- >SERR- <PERR-
>>     Latency: 0
>>     Interrupt: pin A routed to IRQ 169
>>     Region 0: I/O ports at <unassigned>
>>     Region 1: I/O ports at <unassigned>
>>     Region 2: I/O ports at <unassigned>
>>     Region 3: I/O ports at <unassigned>
>>     Region 4: I/O ports at 2c00 [size=16]
> 
> Since this controller is integrated into the ICH SouthBridge chip 
> instead of being on a discrete PCI bus I'm not sure if the latency 
> figure makes any difference. I had a quick look at a machine with an 
> '845 chipset and that had similar 0 latency values.
> 
> I think however that some of the settings the above indicate that you're 
> running in legacy/combined mode rather than using the native sata_piix 
> driver. Does the disk appear as /dev/sdX or /dev/hdX? If using the 
> libata drivers it should appear as /dev/sdX.

SATA hard disk: /dev/sda
PATA dvd drive: /dev/sr0

> If not, try going into the BIOS and try to find a SATA setting referred 
> to as Native/Enhanced (it may currently be set to something like 
> legacy/combined).
> 
> I've seen one machine where using the legacy mode resulted in a 
> performance of around 1MB/s at 99% CPU utilisation. Switching to the 
> enhanced mode got the performance up to a much more normal 40MB/s @ ~20% 
> CPU.

CPU load is about 15 % with software decoding, using VDR, vdr-xine and xine.

Bye.
-- 
Dipl.-Inform. (FH) Reinhard Nissl
mailto:rnissl at gmx.de