On Thu, Nov 13, 2008 at 06:10:14PM -0800, Greg KH wrote: > On Fri, Nov 14, 2008 at 12:25:05PM +1100, Vincent Rizza wrote: > > Thanks Greg, I wasn't sure what to do there. I've updated it to no > > permissions. > > Hm, no, I don't think that's a valid thing. This would be the first "a > file is present, but it doesn't contain anything" type of a file in > sysfs, and I don't think we want to start doing this. I also don't like "msix_irq" name. Can we call it "msix_map"? > Any other ideas of how to structure this? Some ideas: 1) file name is IRQ#, contents are addr/vector pair (ie MSIX table in config space) 2) File name is index# into MSIX table and contents is IRQ#. 3) forget the directory. Just one file with a list of MSIX IRQ #s? I'm preferring (2) because we might be able to use the index to communicate other information to the driver. I'm still trying to figure out how irqbalanced can communicate the CPU mask for each MSIX vector to the driver. When a multi-queue driver starts an IO, it will want to place the IO on the queue which will result in the completion interrupt coming back on the same (or nearby) CPU. So the driver needs a quick way to reverse map the current_CPU->MSIX_index. thanks, grant -- To unsubscribe from this list: send the line "unsubscribe linux-pci" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
