On Fri, Nov 14, 2008 at 10:02:03AM -0700, Grant Grundler wrote: > Some ideas: > 1) file name is IRQ#, contents are addr/vector pair (ie MSIX table in > config space) We can already read the MSI-X table from userspace should we need to actually get the address and data ... why would we want it? > 2) File name is index# into MSIX table and contents is IRQ#. Yes, this was what I proposed, and Greg agreed to. > 3) forget the directory. Just one file with a list of MSIX IRQ #s? You might want to review the first patch in this thread ... > 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. No, the driver needs to map current CPU -> best queue. The interrupt number belongs to the queue and doesn't need to form part of the driver's decision. -- Matthew Wilcox Intel Open Source Technology Centre "Bill, look, we understand that you're interested in selling us this operating system, but compare it to ours. We can't possibly take such a retrograde step." -- 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
