> From: Russell King - ARM Linux [mailto:linux@xxxxxxxxxxxxxxxx] > Sent: Wednesday, September 03, 2008 3:30 PM <snip> > On Wed, Sep 03, 2008 at 02:58:10PM -0500, Woodruff, Richard wrote: > > Fixed translations do have some benefits. You can ensure that you are > > using section or super section descriptors to cover large areas. This > > does result in better TLB usage. Along with freeing up TLB entries > you > > also generally avoid TLB misses on IO calls which touch a variety of > > internal spaces as part of the IRQ sequence. > > These are valid points, and you can arrange for ioremap to give you the > same effect. That way, you end up using _standard_ Linux interfaces > which everyone understands and still having your cake. Is there some example in current code which does this? I recall ioremap's only using 4k and not combining. Also, fixed maps are a lot easier to debug. > > Each chip has very different performance targets and is really better > > built with an optimized tool chain (ARMv5, ARMv6, ARMv7). Doing > > multi-boots with in the same architecture family seems really good > > but across seems less so. > > This thread isn't about multi-boots across differnet architecture; > please take that discussion elsewhere - it's off topic. No but it is a consideration in how fix up is done. Regards, Richard W. -- 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
