On 10/24/2017 12:52 AM, Ram Pai wrote:
On Thu, Sep 14, 2017 at 06:13:57PM +1000, Benjamin Herrenschmidt wrote:
On Fri, 2017-09-08 at 15:44 -0700, Ram Pai wrote:
The second part of the PTE will hold
(H_PAGE_F_SECOND|H_PAGE_F_GIX) at bit 60,61,62,63.
NOTE: None of the bits in the secondary PTE were not used
by 64k-HPTE backed PTE.
Have you measured the performance impact of this ? The second part of
the PTE being in a different cache line there could be one...
hmm..missed responding to this comment.
I did a preliminay measurement running mmap bench in the selftest.
Ran it multiple times. almost always the numbers were either equal-to
or better-than without the patch-series.
mmap bench doesn't do any fault. It is just mmap/munmap in loop.
-aneesh
