On Sat, 6 Jan 2007 19:35:41 +0000, Alan wrote: > The kernel drivers work on the basis that if you are using drivers/ide > then the IDE driver will grab all the ports, if you are using libata then > the PCI boot quirks will switch to split AHCI and PATA mode and the > libata drivers will take both. > > If you build both IDE and libata support for the same hardware into your > kernel it should generally end up with only one driver claiming the > hardware but the AHCI case doesn't handle this in 2.6.18.
[and in another post:] > You have both the drivers for the Jmicron via drivers/ide and via > drivers/ata (libata) loaded. In that specific combination the two drivers > don't use the same resources so don't spot the other one is busy. Thanks for the explanation. I had suspected such an interaction, especially since browsing the LKML it seems that this doubling of the interfaces with the JMicron controllers is an intentional change: <URL:http://lkml.org/lkml/2006/7/12/133>. So the only problem is how to tell the kernel to not use ide on the SATA channels, and this is what I had tried to do with ide2=noprobe ide3=noprobe. However, this doesn't seem to happen. Am I misinterpreting the meaning of the idex=noprobe parameter? BTW, is there a preference for libata or the jmicron ide driver? Assuming I finally manage to only make it appear once, which one should I go for? -- Giuseppe "Oblomov" Bilotta "I weep for our generation" -- Charlie Brown - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/