All of lore.kernel.org
 help / color / mirror / Atom feed
From: Elias El Yandouzi <eliasely@amazon.com>
To: <xen-devel@lists.xenproject.org>
Cc: julien@xen.org, pdurrant@amazon.com, dwmw@amazon.com,
	"Elias El Yandouzi" <eliasely@amazon.com>,
	"Andrew Cooper" <andrew.cooper3@citrix.com>,
	"George Dunlap" <george.dunlap@citrix.com>,
	"Jan Beulich" <jbeulich@suse.com>,
	"Stefano Stabellini" <sstabellini@kernel.org>,
	"Wei Liu" <wl@xen.org>,
	"Bertrand Marquis" <bertrand.marquis@arm.com>,
	"Michal Orzel" <michal.orzel@amd.com>,
	"Volodymyr Babchuk" <Volodymyr_Babchuk@epam.com>,
	"Roger Pau Monné" <roger.pau@citrix.com>,
	"Lukasz Hawrylko" <lukasz@hawrylko.pl>,
	"Daniel P. Smith" <dpsmith@apertussolutions.com>,
	"Mateusz Mówka" <mateusz.mowka@intel.com>
Subject: [PATCH v2 (resend) 00/27] Remove the directmap
Date: Tue, 16 Jan 2024 19:25:44 +0000	[thread overview]
Message-ID: <20240116192611.41112-1-eliasely@amazon.com> (raw)

Hi all,

A few years ago, Wei Liu implemented a PoC to remove the directmap
from Xen. The last version was sent by Hongyan Xia [1].

I will start with thanking both Wei and Hongyan for the initial work
to upstream the feature. A lot of patches already went in and this is
the last few patches missing to effectively enable the feature.

=== What is the directmap? ===

At the moment, on both arm64 and x86, most of the RAM is mapped
in Xen address space. This means that domain memory is easily
accessible in Xen.

=== Why do we want to remove the directmap? ===

(Summarizing my understanding of the previous discussion)

Speculation attacks (like Spectre SP1) rely on loading piece of memory
in the cache. If the region is not mapped then it can't be loaded.

So removing reducing the amount of memory mapped in Xen will also
reduce the surface attack.

=== What's the performance impact? ===

As the guest memory is not always mapped, then the cost of mapping
will increase. I haven't done the numbers with this new version, but
some measurement were provided in the previous version for x86.

=== Improvement possible ===

The known area to improve on x86 are:
   * Mapcache: There was a patch sent by Hongyan:
     https://lore.kernel.org/xen-devel/4058e92ce21627731c49b588a95809dc0affd83a.1581015491.git.hongyxia@amazon.com/
   * EPT: At the moment an guest page-tabel walk requires about 20 map/unmap.
     This will have an very high impact on the performance. We need to decide
     whether keep the EPT always mapped is a problem

The original series didn't have support for Arm64. But as there were
some interest, I have provided a PoC.

There are more extra work for Arm64:
   * The mapcache is quite simple. We would investigate the performance
   * The mapcache should be made compliant to the Arm Arm (this is now
     more critical).
   * We will likely have the same problem as for the EPT.
   * We have no support for merging table to a superpage, neither
     free empty page-tables. (See more below)

=== Implementation ===

The subject is probably a misnomer. The directmap is still present but
the RAM is not mapped by default. Instead, the region will still be used
to map pages allocate via alloc_xenheap_pages().

The advantage is the solution is simple (so IHMO good enough for been
merged as a tech preview). The disadvantage is the page allocator is not
trying to keep all the xenheap pages together. So we may end up to have
an increase of page table usage.

In the longer term, we should consider to remove the direct map
completely and switch to vmap(). The main problem with this approach
is it is frequent to use mfn_to_virt() in the code. So we would need
to cache the mapping (maybe in the struct page_info).

=== Why arm32 is not covered? ===

On Arm32, the domheap and xenheap is always separated. So by design
the guest memory is not mapped by default.

At this stage, it seems unnecessary to have to map/unmap xenheap pages
every time they are allocated.

=== Why not using a separate domheap and xenheap? ===

While a separate xenheap/domheap reduce the page-table usage (all
xenheap pages are contiguous and could be always mapped), it is also
currently less scalable because the split is fixed at boot time (XXX:
Can this be dynamic?).

=== Future of secret-free hypervisor ===

There are some information in an e-mail from Andrew a few years ago:

https://lore.kernel.org/xen-devel/e3219697-0759-39fc-2486-715cdec1ca9e@citrix.com/

Cheers,

[1] https://lore.kernel.org/xen-devel/cover.1588278317.git.hongyxia@amazon.com/

*** BLURB HERE ***

Elias El Yandouzi (3):
  xen/x86: Add build assertion for fixmap entries
  Rename mfn_to_virt() calls
  Rename maddr_to_virt() calls

Hongyan Xia (13):
  acpi: vmap pages in acpi_os_alloc_memory
  xen/numa: vmap the pages for memnodemap
  x86/srat: vmap the pages for acpi_slit
  x86: Map/unmap pages in restore_all_guests
  x86/pv: Rewrite how building PV dom0 handles domheap mappings
  x86/pv: Map L4 page table for shim domain
  x86/mapcache: Initialise the mapcache for the idle domain
  x86: Add a boot option to enable and disable the direct map
  x86/domain_page: Remove the fast paths when mfn is not in the
    directmap
  xen/page_alloc: Add a path for xenheap when there is no direct map
  x86/setup: Leave early boot slightly earlier
  x86/setup: vmap heap nodes when they are outside the direct map
  x86/setup: Do not create valid mappings when directmap=no

Julien Grall (8):
  xen/vmap: Check the page has been mapped in vm_init_type()
  xen/vmap: Introduce vmap_size() and use it
  xen/arm: fixmap: Rename the fixmap slots to follow the x86 convention
  xen/x86: Add support for the PMAP
  xen/arm32: mm: Rename 'first' to 'root' in init_secondary_pagetables()
  xen/arm64: mm: Use per-pCPU page-tables
  xen/arm64: Implement a mapcache for arm64
  xen/arm64: Allow the admin to enable/disable the directmap

Wei Liu (3):
  x86/setup: Move vm_init() before acpi calls
  x86/pv: Domheap pages should be mapped while relocating initrd
  x86: Lift mapcache variable to the arch level

 docs/misc/xen-command-line.pandoc             |  12 ++
 xen/arch/arm/Kconfig                          |   3 +-
 xen/arch/arm/acpi/lib.c                       |  18 +--
 xen/arch/arm/arm64/mmu/mm.c                   |  45 +++++-
 xen/arch/arm/domain_page.c                    |  50 ++++++-
 xen/arch/arm/include/asm/arm32/mm.h           |   8 --
 xen/arch/arm/include/asm/arm64/mm.h           |   7 +-
 xen/arch/arm/include/asm/domain_page.h        |  13 ++
 xen/arch/arm/include/asm/early_printk.h       |   2 +-
 xen/arch/arm/include/asm/fixmap.h             |  16 +--
 xen/arch/arm/include/asm/mm.h                 |   9 ++
 xen/arch/arm/include/asm/mmu/layout.h         |  13 +-
 xen/arch/arm/include/asm/mmu/mm.h             |   2 +
 xen/arch/arm/mm.c                             |   1 +
 xen/arch/arm/mmu/pt.c                         |  12 +-
 xen/arch/arm/mmu/setup.c                      |  33 ++---
 xen/arch/arm/mmu/smpboot.c                    |  32 ++---
 xen/arch/arm/setup.c                          |   6 +-
 xen/arch/x86/Kconfig                          |   2 +
 xen/arch/x86/dmi_scan.c                       |   4 +-
 xen/arch/x86/domain.c                         |  12 +-
 xen/arch/x86/domain_page.c                    |  74 +++++++---
 xen/arch/x86/hvm/dom0_build.c                 |   4 +-
 xen/arch/x86/include/asm/config.h             |  12 +-
 xen/arch/x86/include/asm/domain.h             |  13 +-
 xen/arch/x86/include/asm/fixmap.h             |   9 ++
 .../x86/include/asm/mach-default/bios_ebda.h  |   2 +-
 xen/arch/x86/include/asm/mm.h                 |  10 +-
 xen/arch/x86/include/asm/page.h               |   8 +-
 xen/arch/x86/include/asm/pmap.h               |  25 ++++
 xen/arch/x86/include/asm/x86_64/page.h        |   2 +-
 xen/arch/x86/mm.c                             |  18 ++-
 xen/arch/x86/mpparse.c                        |   2 +-
 xen/arch/x86/pv/dom0_build.c                  |  73 +++++++---
 xen/arch/x86/pv/domain.c                      |  34 +++++
 xen/arch/x86/setup.c                          | 129 +++++++++++++++---
 xen/arch/x86/srat.c                           |   4 +-
 xen/arch/x86/tboot.c                          |   2 +-
 xen/arch/x86/x86_64/asm-offsets.c             |   1 +
 xen/arch/x86/x86_64/entry.S                   |   8 ++
 xen/arch/x86/x86_64/mm.c                      |  26 ++--
 xen/common/Kconfig                            |  17 +++
 xen/common/efi/boot.c                         |  23 ++--
 xen/common/numa.c                             |   9 +-
 xen/common/page_alloc.c                       |  89 ++++++++++--
 xen/common/pmap.c                             |   8 +-
 xen/common/trace.c                            |   8 +-
 xen/common/vmap.c                             |  62 ++++++---
 xen/drivers/acpi/osl.c                        |  14 +-
 xen/include/xen/mm.h                          |   7 +
 xen/include/xen/vmap.h                        |   4 +
 51 files changed, 753 insertions(+), 244 deletions(-)
 create mode 100644 xen/arch/arm/include/asm/domain_page.h
 create mode 100644 xen/arch/x86/include/asm/pmap.h

-- 
2.40.1



             reply	other threads:[~2024-01-16 19:26 UTC|newest]

Thread overview: 60+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-01-16 19:25 Elias El Yandouzi [this message]
2024-01-16 19:25 ` [PATCH v2 (resend) 01/27] xen/vmap: Check the page has been mapped in vm_init_type() Elias El Yandouzi
2024-01-25 16:14   ` Jan Beulich
2024-01-16 19:25 ` [PATCH v2 (resend) 02/27] x86/setup: Move vm_init() before acpi calls Elias El Yandouzi
2024-01-25 16:17   ` Jan Beulich
2024-02-05 22:55     ` Stefano Stabellini
2024-01-16 19:25 ` [PATCH v2 (resend) 03/27] xen/vmap: Introduce vmap_size() and use it Elias El Yandouzi
2024-01-25 16:26   ` Jan Beulich
2024-01-16 19:25 ` [PATCH v2 (resend) 04/27] acpi: vmap pages in acpi_os_alloc_memory Elias El Yandouzi
2024-01-25 16:28   ` Jan Beulich
2024-06-26 13:54   ` Alejandro Vallejo
2024-06-26 15:17     ` Jan Beulich
2024-06-26 16:33       ` Alejandro Vallejo
2024-01-16 19:25 ` [PATCH v2 (resend) 05/27] xen/numa: vmap the pages for memnodemap Elias El Yandouzi
2024-01-25 16:30   ` Jan Beulich
2024-01-16 19:25 ` [PATCH v2 (resend) 06/27] x86/srat: vmap the pages for acpi_slit Elias El Yandouzi
2024-01-25 16:32   ` Jan Beulich
2024-01-16 19:25 ` [PATCH v2 (resend) 07/27] x86: Map/unmap pages in restore_all_guests Elias El Yandouzi
2024-02-20  9:51   ` Jan Beulich
2024-04-30 16:08     ` Elias El Yandouzi
2024-05-02  6:48       ` Jan Beulich
2024-01-16 19:25 ` [PATCH v2 (resend) 08/27] x86/pv: Domheap pages should be mapped while relocating initrd Elias El Yandouzi
2024-02-20 10:07   ` Jan Beulich
2024-01-16 19:25 ` [PATCH v2 (resend) 09/27] x86/pv: Rewrite how building PV dom0 handles domheap mappings Elias El Yandouzi
2024-02-20 10:28   ` Jan Beulich
2024-05-07 15:21     ` Elias El Yandouzi
2024-05-14  9:52       ` Jan Beulich
2024-01-16 19:25 ` [PATCH v2 (resend) 10/27] x86/pv: Map L4 page table for shim domain Elias El Yandouzi
2024-02-20 10:37   ` Jan Beulich
2024-01-16 19:25 ` [PATCH v2 (resend) 11/27] x86: Lift mapcache variable to the arch level Elias El Yandouzi
2024-02-20 10:46   ` Jan Beulich
2024-05-07 15:22     ` Elias El Yandouzi
2024-05-14  9:53       ` Jan Beulich
2024-01-16 19:25 ` [PATCH v2 (resend) 12/27] x86/mapcache: Initialise the mapcache for the idle domain Elias El Yandouzi
2024-02-20 10:51   ` Jan Beulich
2024-05-07 15:25     ` Elias El Yandouzi
2024-05-13  9:35     ` Elias El Yandouzi
2024-05-14 10:08       ` Jan Beulich
2024-01-16 19:25 ` [PATCH v2 (resend) 13/27] x86: Add a boot option to enable and disable the direct map Elias El Yandouzi
2024-02-20 11:14   ` Jan Beulich
2024-05-13 10:50     ` Elias El Yandouzi
2024-05-14 10:19       ` Jan Beulich
2024-01-16 19:25 ` [PATCH v2 (resend) 14/27] xen/arm: fixmap: Rename the fixmap slots to follow the x86 convention Elias El Yandouzi
2024-01-16 19:25 ` [PATCH v2 (resend) 15/27] xen/x86: Add support for the PMAP Elias El Yandouzi
2024-01-16 19:26 ` [PATCH v2 (resend) 16/27] xen/x86: Add build assertion for fixmap entries Elias El Yandouzi
2024-01-16 19:26 ` [PATCH v2 (resend) 17/27] x86/domain_page: Remove the fast paths when mfn is not in the directmap Elias El Yandouzi
2024-01-16 19:26 ` [PATCH v2 (resend) 18/27] xen/page_alloc: Add a path for xenheap when there is no direct map Elias El Yandouzi
2024-01-16 19:26 ` [PATCH v2 (resend) 19/27] x86/setup: Leave early boot slightly earlier Elias El Yandouzi
2024-01-16 19:26 ` [PATCH v2 (resend) 20/27] x86/setup: vmap heap nodes when they are outside the direct map Elias El Yandouzi
2024-01-16 19:26 ` [PATCH v2 (resend) 21/27] x86/setup: Do not create valid mappings when directmap=no Elias El Yandouzi
2024-01-16 19:26 ` [PATCH v2 (resend) 22/27] Rename mfn_to_virt() calls Elias El Yandouzi
2024-01-16 19:26 ` [PATCH v2 (resend) 23/27] Rename maddr_to_virt() calls Elias El Yandouzi
2024-01-16 19:26 ` [PATCH v2 (resend) 24/27] xen/arm32: mm: Rename 'first' to 'root' in init_secondary_pagetables() Elias El Yandouzi
2024-01-16 19:26 ` [PATCH v2 (resend) 25/27] xen/arm64: mm: Use per-pCPU page-tables Elias El Yandouzi
2024-01-16 19:26 ` [PATCH v2 (resend) 26/27] xen/arm64: Implement a mapcache for arm64 Elias El Yandouzi
2024-01-16 19:26 ` [PATCH v2 (resend) 27/27] xen/arm64: Allow the admin to enable/disable the directmap Elias El Yandouzi
2024-01-29  8:28 ` [PATCH v2 (resend) 00/27] Remove " Jan Beulich
2024-02-05 11:11   ` Elias El Yandouzi
2024-02-16 17:17     ` Julien Grall
2024-03-25 10:31 ` Jan Beulich

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=20240116192611.41112-1-eliasely@amazon.com \
    --to=eliasely@amazon.com \
    --cc=Volodymyr_Babchuk@epam.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=bertrand.marquis@arm.com \
    --cc=dpsmith@apertussolutions.com \
    --cc=dwmw@amazon.com \
    --cc=george.dunlap@citrix.com \
    --cc=jbeulich@suse.com \
    --cc=julien@xen.org \
    --cc=lukasz@hawrylko.pl \
    --cc=mateusz.mowka@intel.com \
    --cc=michal.orzel@amd.com \
    --cc=pdurrant@amazon.com \
    --cc=roger.pau@citrix.com \
    --cc=sstabellini@kernel.org \
    --cc=wl@xen.org \
    --cc=xen-devel@lists.xenproject.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.