linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Linus Torvalds <torvalds@linux-foundation.org>
To: Dave Hansen <dave.hansen@linux.intel.com>
Cc: "Linux Kernel Mailing List" <linux-kernel@vger.kernel.org>,
	linux-mm <linux-mm@kvack.org>,
	"Andrea Arcangeli" <aarcange@redhat.com>,
	"Andrew Lutomirski" <luto@kernel.org>,
	"Kees Cook" <keescook@google.com>,
	"Hugh Dickins" <hughd@google.com>,
	"Jürgen Groß" <jgross@suse.com>,
	"the arch/x86 maintainers" <x86@kernel.org>,
	namit@vmware.com
Subject: Re: [PATCH 00/11] Use global pages with PTI
Date: Fri, 23 Mar 2018 17:46:42 -0700	[thread overview]
Message-ID: <CA+55aFxn=NiAhtz77nrx1_10em8bume-M0UzYZU2eVm5n71juA@mail.gmail.com> (raw)
In-Reply-To: <be2e683c-bf0a-e9ce-2f02-4905f6bd56d3@linux.intel.com>

On Fri, Mar 23, 2018 at 5:40 PM, Dave Hansen
<dave.hansen@linux.intel.com> wrote:
>
> Well, rats.  This somehow makes things slower with PCIDs on.

.. what happens when you enable global pages with PCID? You disabled
them explicitly because you thought they wouldn't matter..

Even with PCID, a global TLB entry for the shared pages would make
sense, because it's now just *one* entry in the TLB rather that "one
per PCID and one for the kernel mapping".

So even if in theory the lifetime of the TLB entry is the same, when
you have capacity misses it most definitely isn't.

And for process tear-down and build-up the per-PCID TLB entry does
nothing at all. While for a true global entry, it gets shared even
across process creation/deletion. So even ignoring TLB capacity
issues, with lots of shortlived processes global TLB entries are much
better.

It is, of course, possible that I misunderstood what you actually
benchmarked. But I assume the above benchmark numbers are with the
whole "don't even do global entries if you have PCID".

               Linus

  reply	other threads:[~2018-03-24  0:46 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-03-23 17:44 Dave Hansen
2018-03-23 17:44 ` [PATCH 01/11] x86/mm: factor out pageattr _PAGE_GLOBAL setting Dave Hansen
2018-03-23 17:44 ` [PATCH 02/11] x86/mm: undo double _PAGE_PSE clearing Dave Hansen
2018-03-23 17:44 ` [PATCH 03/11] x86/mm: introduce "default" kernel PTE mask Dave Hansen
2018-03-23 17:44 ` [PATCH 04/11] x86/espfix: document use of _PAGE_GLOBAL Dave Hansen
2018-03-23 17:44 ` [PATCH 05/11] x86/mm: do not auto-massage page protections Dave Hansen
2018-03-23 19:15   ` Nadav Amit
2018-03-23 19:26     ` Dave Hansen
2018-03-23 19:34       ` Nadav Amit
2018-03-23 19:38         ` Dave Hansen
2018-03-24 15:10   ` kbuild test robot
2018-03-24 15:21   ` kbuild test robot
2018-03-23 17:44 ` [PATCH 06/11] x86/mm: remove extra filtering in pageattr code Dave Hansen
2018-03-23 17:44 ` [PATCH 07/11] x86/mm: comment _PAGE_GLOBAL mystery Dave Hansen
2018-03-23 17:44 ` [PATCH 08/11] x86/mm: do not forbid _PAGE_RW before init for __ro_after_init Dave Hansen
2018-03-23 17:45 ` [PATCH 09/11] x86/pti: enable global pages for shared areas Dave Hansen
2018-03-23 19:12   ` Nadav Amit
2018-03-23 19:36     ` Dave Hansen
2018-03-23 17:45 ` [PATCH 10/11] x86/pti: clear _PAGE_GLOBAL for kernel image Dave Hansen
2018-03-23 17:45 ` [PATCH 11/11] x86/pti: leave kernel text global for !PCID Dave Hansen
2018-03-23 18:26 ` [PATCH 00/11] Use global pages with PTI Linus Torvalds
2018-03-24  0:40   ` Dave Hansen
2018-03-24  0:46     ` Linus Torvalds [this message]
2018-03-24  0:54       ` Linus Torvalds
2018-03-24 11:05     ` Ingo Molnar
2018-03-27 13:36     ` Thomas Gleixner
2018-03-27 16:32       ` Dave Hansen
2018-03-27 17:51         ` Thomas Gleixner
2018-03-27 20:07           ` Ingo Molnar
2018-03-27 20:19             ` Dave Hansen
2018-03-29  0:17             ` Dave Hansen
2018-03-30 12:09               ` Ingo Molnar
2018-03-30 12:17                 ` Ingo Molnar
2018-03-30 20:26                   ` Dave Hansen
2018-03-30 20:32                     ` Thomas Gleixner
2018-03-30 21:40                       ` Dave Hansen
2018-03-31  5:39                         ` Ingo Molnar
2018-03-31 18:19                           ` Dave Hansen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CA+55aFxn=NiAhtz77nrx1_10em8bume-M0UzYZU2eVm5n71juA@mail.gmail.com' \
    --to=torvalds@linux-foundation.org \
    --cc=aarcange@redhat.com \
    --cc=dave.hansen@linux.intel.com \
    --cc=hughd@google.com \
    --cc=jgross@suse.com \
    --cc=keescook@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=luto@kernel.org \
    --cc=namit@vmware.com \
    --cc=x86@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox