Re: [PATCH 3/4] mm/tlb, x86/mm: Support invalidating TLB caches for RCU_TABLE_FREE

From: Rik van Riel
Date: Mon Aug 27 2018 - 09:37:08 EST

On Mon, 2018-08-27 at 18:04 +1000, Nicholas Piggin wrote:

> It could do that. It requires a tlbie that matches the page size,
> so it means 3 sizes. I think possibly even that would be better
> than current code, but we could do better if we had a few specific
> fields in there.

Would it cause a noticeable overhead to keep track
of which page sizes were removed, and to simply flush
the whole TLB in the (unlikely?) event that multiple
page sizes were removed in the same munmap?

Once the unmap is so large that multiple page sizes
were covered, you may already be looking at so many
individual flush operations that a full flush might
be faster.

Is there a point on PPC where simply flushing the
whole TLB, and having other things be reloaded later,
is faster than flushing every individual page mapping
that got unmapped?

All Rights Reversed.

Attachment: signature.asc
Description: This is a digitally signed message part