From: "Kirill A. Shutemov" <kirill@shutemov.name>
To: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@kernel.org>,
"Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>,
Linus Torvalds <torvalds@linux-foundation.org>,
Andrew Morton <akpm@linux-foundation.org>,
x86@kernel.org, Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@redhat.com>, Arnd Bergmann <arnd@arndb.de>,
"H. Peter Anvin" <hpa@zytor.com>,
Dave Hansen <dave.hansen@intel.com>,
Andy Lutomirski <luto@amacapital.net>,
linux-arch@vger.kernel.org, linux-mm@kvack.org,
linux-kernel@vger.kernel.org
Subject: Re: [RFC, PATCHv1 00/28] 5-level paging
Date: Fri, 9 Dec 2016 20:21:18 +0300 [thread overview]
Message-ID: <20161209172118.GB8932@node.shutemov.name> (raw)
Message-ID: <20161209172118.LkcjIYa0uvMSg5D3MDyU_2IcyUxaNNyVZiPWb9XVuBk@z> (raw)
In-Reply-To: <20161209164011.GL8388@tassilo.jf.intel.com>
On Fri, Dec 09, 2016 at 08:40:11AM -0800, Andi Kleen wrote:
> > On other hand, large virtual address space would put more pressure on
> > cache -- at least one more page table per process, if we make 56-bit VA
> > default.
>
> The top level page always has to be there unless you disable it at boot time
> (unless you go for a scheme where some processes share top level pages, and
> others do not, which would likely be very complicated)
>
> But even with that it is more than one: A typical set up has at least two extra
> 4K pages overhead, one for the bottom and one for the top mappings. Could easily be
> more.
So, right, one page for pgd, which we can't easily avoid.
If we limit VA to 47-bits by default, we would have one p4d page as the
range will be covered by one entry in pgd.
If we go to 56-bits VA by default, we would have at least two p4d pages
even for small processes. This where mine "at least one more page table
per process" comes from.
That's waste of memory and potentially cache. I don't think it's
justified.
--
Kirill A. Shutemov
next prev parent reply other threads:[~2016-12-09 17:21 UTC|newest]
Thread overview: 113+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-12-08 16:21 [RFC, PATCHv1 00/28] 5-level paging Kirill A. Shutemov
2016-12-08 16:21 ` [QEMU, PATCH] x86: implement la57 paging mode Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:48 ` [Qemu-devel] " no-reply
2016-12-08 16:48 ` no-reply
2016-12-08 16:21 ` [RFC, PATCHv1 01/28] asm-generic: introduce 5level-fixup.h Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 02/28] asm-generic: introduce __ARCH_USE_5LEVEL_HACK Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 03/28] arch, mm: convert all architectures to use 5level-fixup.h Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 04/28] asm-generic: introduce <asm-generic/pgtable-nop4d.h> Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 05/28] mm: convert generic code to 5-level paging Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 06/28] x86: basic changes into headers for " Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 07/28] x86: trivial portion of 5-level paging conversion Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 08/28] x86/gup: add 5-level paging support Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 09/28] x86/ident_map: " Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 10/28] x86/mm: add support of p4d_t in vmalloc_fault() Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 11/28] x86/power: support p4d_t in hibernate code Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 12/28] x86/kexec: support p4d_t Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 13/28] x86: convert the rest of the code to " Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 14/28] mm: introduce __p4d_alloc() Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 15/28] x86: detect 5-level paging support Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 20:05 ` Borislav Petkov
2016-12-08 20:08 ` Linus Torvalds
2016-12-08 20:08 ` Linus Torvalds
2016-12-08 20:20 ` Borislav Petkov
2016-12-13 22:44 ` H. Peter Anvin
2016-12-13 22:44 ` H. Peter Anvin
2016-12-13 23:07 ` Boris Petkov
2016-12-13 23:07 ` Boris Petkov
2016-12-15 14:39 ` Borislav Petkov
2016-12-15 14:39 ` Borislav Petkov
2016-12-15 17:52 ` hpa
2016-12-15 17:52 ` hpa
2016-12-15 19:09 ` Borislav Petkov
2016-12-15 19:20 ` Andi Kleen
2016-12-15 19:20 ` Andi Kleen
2016-12-15 20:52 ` hpa
2016-12-15 20:52 ` hpa
2016-12-15 20:57 ` hpa
2016-12-15 20:57 ` hpa
2016-12-09 15:32 ` Kirill A. Shutemov
2016-12-09 16:33 ` Borislav Petkov
2016-12-13 22:50 ` H. Peter Anvin
2016-12-13 22:50 ` H. Peter Anvin
2016-12-08 16:21 ` [RFC, PATCHv1 16/28] x86/asm: remove __VIRTUAL_MASK_SHIFT==47 assert Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 18:39 ` Andy Lutomirski
2016-12-08 19:22 ` Kirill A. Shutemov
2016-12-08 19:22 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 17/28] x86/mm: define virtual memory map for 5-level paging Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 18:56 ` Randy Dunlap
2016-12-08 19:24 ` Kirill A. Shutemov
2016-12-08 19:24 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 18/28] x86/paravirt: make paravirt code support " Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 19/28] x86/mm: basic defines/helpers for CONFIG_X86_5LEVEL Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 20/28] x86/dump_pagetables: support 5-level paging Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 21/28] x86/mm: extend kasan to " Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 22/28] x86/espfix: " Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 18:40 ` Andy Lutomirski
2016-12-08 18:40 ` Andy Lutomirski
2016-12-12 14:22 ` Kirill A. Shutemov
2016-12-12 14:22 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 23/28] x86/mm: add support of additional page table level during early boot Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 24/28] x86/mm: add sync_global_pgds() for configuration with 5-level paging Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 18:42 ` Andy Lutomirski
2016-12-08 19:33 ` Kirill A. Shutemov
2016-12-08 19:33 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 25/28] x86/mm: make kernel_physical_mapping_init() support " Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 26/28] x86/mm: add support for 5-level paging for KASLR Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 27/28] x86: enable la57 support Kirill A. Shutemov
2016-12-08 16:21 ` Kirill A. Shutemov
2016-12-08 16:21 ` [RFC, PATCHv1 28/28] TESTING-ONLY: bump TASK_SIZE_MAX Kirill A. Shutemov
2016-12-08 18:16 ` [RFC, PATCHv1 00/28] 5-level paging Linus Torvalds
2016-12-08 18:16 ` Linus Torvalds
2016-12-08 18:26 ` hpa
2016-12-08 18:26 ` hpa
2016-12-08 19:20 ` Kirill A. Shutemov
2016-12-08 19:20 ` Kirill A. Shutemov
2016-12-09 5:01 ` Ingo Molnar
2016-12-09 5:01 ` Ingo Molnar
2016-12-09 10:24 ` Arnd Bergmann
2016-12-09 10:51 ` Catalin Marinas
2016-12-09 10:51 ` Catalin Marinas
2016-12-09 10:37 ` Kirill A. Shutemov
2016-12-09 10:37 ` Kirill A. Shutemov
2016-12-09 16:40 ` Andi Kleen
2016-12-09 17:21 ` Kirill A. Shutemov [this message]
2016-12-09 17:21 ` Kirill A. Shutemov
2016-12-09 16:49 ` Dave Hansen
2016-12-09 16:49 ` Dave Hansen
2016-12-13 21:06 ` Dave Hansen
2016-12-13 21:06 ` 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=20161209172118.GB8932@node.shutemov.name \
--to=kirill@shutemov.name \
--cc=ak@linux.intel.com \
--cc=akpm@linux-foundation.org \
--cc=arnd@arndb.de \
--cc=dave.hansen@intel.com \
--cc=hpa@zytor.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=luto@amacapital.net \
--cc=mingo@kernel.org \
--cc=mingo@redhat.com \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--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;
as well as URLs for NNTP newsgroup(s).