On Thu, Feb 17, 2011 at 12:05:59PM -0800, Colin Cross wrote: > On Thu, Feb 17, 2011 at 2:41 AM, Lorenzo Pieralisi > <lorenzo.pieralisi@xxxxxxx> wrote: > > I think that on resume you invalidate L1 in the reset vector before > > jumping to cpu_resume, correct ? > > I manually invalidate the L1 Dcache, but not the Icache. Well... ARMv5 and older invalidates the L1 Dcache. ARMv6 cleans and invalidates the L1 Dcache. ARMv7 does nothing with the L1 Dcache. That's rather inconsistent, but that's what people are doing with their current suspend/resume paths. We really ought to have this well-defined. I don't see any reason why there should be any dirty data in the Dcache, so I think the safe thing we should be doing in every case is a L1 Dcache invalidate on resume. That's a separate change though. -- To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
