On Mon, Apr 16, 2018 at 8:24 AM, Joerg Roedel <j...@8bytes.org> wrote: > > I tested this version again with my load-test of running > perf-top/various x86-selftests/kernel-compile in a loop for > a couple of hours. This showed no issues. I also briefly > tested a 64bit kernel and this also worked as expected.
Andy, can you send your extra x86 self-tests to Joerg too? Also, it would be nice to have performance numbers, and check that the global page issues are at least fixed on 32-bit too, since we screwed that up on x86-64 initially. On x86-32, the global pages are likely a bigger deal since there's no PCID. Linus