Messages in this thread | | | Subject | Re: Opinion on ordering of writel vs. stores to RAM | From | Benjamin Herrenschmidt <> | Date | Mon, 11 Sep 2006 08:18:28 +1000 |
| |
On Sun, 2006-09-10 at 23:23 +0100, Alan Cox wrote: > Ar Llu, 2006-09-11 am 07:25 +1000, ysgrifennodd Benjamin Herrenschmidt: > > I'm copying that from a private discussion I had. Please let me know if > > you have comments. This proposal includes some questions too so please > > answer :) > > Looks sane and Linus seems to like mmiowb. Only other question: what are > the guarantees of memcpy_to/fromio. Does it access the memory in ordered > fashion or not, does it guarantee only ordering at the end of the copy > or during it ?
Well, Linus is also ok with writel not ordering memory an IO accesses :) Though he also mentioned that if we go that route (which is what we have now in fact), we take the burden of having to test and fix drivers who don't get it...
That's why I think a compromise is in order, thus my proposal :)
Ben.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |