On Thu, May 17, 2012 at 11:08 PM, Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote: > On Thu, 2012-05-17 at 16:51 +0200, Peter Zijlstra wrote: >> >> Also, does it even work if the range happens to be backed by huge pages? >> IIRC we try and do the identity map with large pages wherever possible. > > OK, the Intel SDM seems to suggest it will indeed invalidate ANY mapping > to that linear address, which would include 2M and 1G pages. As for as I know, the 1GB TLB fold into 2MB now, and 2MB page still maybe fold into 4K page in TLB flush. We are tracking this. > > -- > To unsubscribe, send a message with 'unsubscribe linux-mm' in > the body to majordomo@xxxxxxxxx. For more info on Linux MM, > see: http://www.linux-mm.org/ . > Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ > Don't email: <a hrefmailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a> -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: <a href