Messages in this thread | | | Date | Tue, 30 Oct 2001 17:54:17 +0100 | From | Andrea Arcangeli <> | Subject | Re: please revert bogus patch to vmscan.c |
| |
On Tue, Oct 30, 2001 at 05:13:53PM +0100, Giuliano Pochini wrote: > > >> But of course going from page flush to the mm flush is fine from my part > >> too. As Linus noted a few days ago during swapout we're going to block > >> and reschedule all the time, so the range flush is going to be a noop in > > > > Only on architectures where the TLB (or equivalent) is > > small and only capable of holding entries for one address > > space at a time. > > > > It's simply not true on eg PPC. > > #ifdef ?
yes, but not for ppc, for alpha and all other archs without accessed bit provided in hardware (and cached in the tlb). the flush_mm proposed by Ben looks fine for x86 too, it's a waste only for archs without accessed bit.
I think an #ifndef HAVE_NO_ACCESS_BIT_IN_TLB or something like that, then define that in asm-alpha/ and the other archs without accessed bit.
OTOH, it probably doesn't make much difference so maybe it doesn't worth the effort.
Andrea - 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/
| |