On Sat, Sep 12, 2009 at 10:55 PM, Jung-Ik (John) Lee <jilee@xxxxxxxxxx> wrote: > I agree in that it's hard to measure/signify the additional overhead, > since those io insts are already too slow. > Anyways, the two extra "if"s and one PIO_MASK on every ioread/iowrite > are pure overhead on top of in/out insts. I don't think that any platform currently breaks if you use the readX/writeX instead of ioreadX/iowriteX on iomapped MMIO addresses (and there's certainly enough code in libata that does this), but the API definition really expects that you should be using ioread/iowrite so it would probably be best not to add any more of these.. -- To unsubscribe from this list: send the line "unsubscribe linux-ide" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html