qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: David Hildenbrand <david@redhat.com>
To: qemu-devel@nongnu.org
Cc: "Juan Quintela" <quintela@redhat.com>,
	"Murilo Opsfelder Araujo" <muriloo@linux.ibm.com>,
	"Cornelia Huck" <cohuck@redhat.com>,
	"Eduardo Habkost" <ehabkost@redhat.com>,
	"Michael S. Tsirkin" <mst@redhat.com>,
	"Stefan Weil" <sw@weilnetz.de>,
	"David Hildenbrand" <david@redhat.com>,
	"Richard Henderson" <richard.henderson@linaro.org>,
	"Dr. David Alan Gilbert" <dgilbert@redhat.com>,
	"Peter Xu" <peterx@redhat.com>, "Greg Kurz" <groug@kaod.org>,
	"Halil Pasic" <pasic@linux.ibm.com>,
	"Christian Borntraeger" <borntraeger@de.ibm.com>,
	"Stefan Hajnoczi" <stefanha@redhat.com>,
	"Igor Mammedov" <imammedo@redhat.com>,
	"Thomas Huth" <thuth@redhat.com>,
	"Paolo Bonzini" <pbonzini@redhat.com>,
	"Philippe Mathieu-Daudé" <philmd@redhat.com>,
	"Igor Kotrasinski" <i.kotrasinsk@partner.samsung.com>
Subject: [PATCH v1 0/9] RAM_NORESERVE, MAP_NORESERVE and hostmem "reserve" property
Date: Tue,  9 Feb 2021 14:49:30 +0100	[thread overview]
Message-ID: <20210209134939.13083-1-david@redhat.com> (raw)

Some cleanups previously sent in other context (resizeable allocations),
followed by RAM_NORESERVE, implementing it under POSIX using MAP_NORESERVE,
and letting users configure it for memory backens using the "reserve"
property (default: true).

MAP_NORESERVE under Linux has in the context of QEMU an effect on
1) Private anonymous memory
-> memory-backend-ram,id=mem0,size=10G
2) Private file-based mappings
-> memory-backend-file,id=mem0,size=10G,mem-path=/dev/shm/0
3) Private/shared hugetlbfs memory
-> memory-backend-memfd,id=mem0,size=10G,hugetlb=on,hugetlbsize=2M

With MAP_NORESERVE/"reserve=off", we won't be reserving swap space (1/2) or
huge pages (3) for the whole memory region.

The target use case is virtio-mem, which dynamically exposes memory
inside a large, sparse memory area to the VM. MAP_NORESERVE tells the OS
"this mapping might be very sparse". This essentially allows
avoiding to set "/proc/sys/vm/overcommit_memory == 0") when using
virtio-mem and also supporting hugetlbfs in the future.

virtio-mem currently only supports anonymous memory, in the future we want
to also support shared file-based and shared hugetlbfs mappings. We most
probably won't be supporting private mappings as they can end up behaving
very weird when it comes to memory consumption.

Future work for virtio-mem I am currently working on includes
1. Introducing a prealloc option for virtio-mem (e.g., using fallocate()
   when plugging blocks) to fail nicely when running out of
   backing storage like huge pages.
2. Supporting resizable RAM block/memoryr egions, such that we won't always
   expose a large, sparse memory region to the VM.
3. Protecting unplugged memory e.g., using userfaultfd.
4. (resizeable allocations / optimized mmap handling when resizing RAM
    blocks)

Cc: Peter Xu <peterx@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
Cc: Richard Henderson <richard.henderson@linaro.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Igor Mammedov <imammedo@redhat.com>
Cc: "Philippe Mathieu-Daudé" <philmd@redhat.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Cc: Murilo Opsfelder Araujo <muriloo@linux.ibm.com>
Cc: Greg Kurz <groug@kaod.org>
Cc: Liam Merwick <liam.merwick@oracle.com>

David Hildenbrand (9):
  softmmu/physmem: drop "shared" parameter from ram_block_add()
  util/mmap-alloc: factor out calculation of the pagesize for the guard
    page
  util/mmap-alloc: factor out reserving of a memory region to
    mmap_reserve()
  util/mmap-alloc: factor out activating of memory to mmap_activate()
  softmmu/memory: pass ram_flags into qemu_ram_alloc_from_fd()
  softmmu/memory: pass ram_flags into
    memory_region_init_ram_shared_nomigrate()
  memory: introduce RAM_NORESERVE and wire it up in qemu_ram_mmap()
  util/mmap-alloc: support RAM_NORESERVE via MAP_NORESERVE
  hostmem: wire up RAM_NORESERVE via "reserve" property

 backends/hostmem-file.c                       |  11 +-
 backends/hostmem-memfd.c                      |   8 +-
 backends/hostmem-ram.c                        |   7 +-
 backends/hostmem.c                            |  33 +++
 hw/m68k/next-cube.c                           |   4 +-
 hw/misc/ivshmem.c                             |   4 +-
 include/exec/cpu-common.h                     |   1 +
 include/exec/memory.h                         |  43 ++--
 include/exec/ram_addr.h                       |   9 +-
 include/qemu/mmap-alloc.h                     |   4 +-
 include/qemu/osdep.h                          |   3 +-
 include/sysemu/hostmem.h                      |   2 +-
 include/sysemu/kvm.h                          |   3 +-
 migration/ram.c                               |   3 +-
 .../memory-region-housekeeping.cocci          |   8 +-
 softmmu/memory.c                              |  27 ++-
 softmmu/physmem.c                             |  48 +++--
 target/s390x/kvm.c                            |   6 +-
 util/mmap-alloc.c                             | 188 ++++++++++++------
 util/oslib-posix.c                            |   5 +-
 util/oslib-win32.c                            |  13 +-
 21 files changed, 281 insertions(+), 149 deletions(-)

-- 
2.29.2



             reply	other threads:[~2021-02-09 13:52 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-02-09 13:49 David Hildenbrand [this message]
2021-02-09 13:49 ` [PATCH v1 1/9] softmmu/physmem: drop "shared" parameter from ram_block_add() David Hildenbrand
2021-02-09 13:49 ` [PATCH v1 2/9] util/mmap-alloc: factor out calculation of the pagesize for the guard page David Hildenbrand
2021-02-09 13:49 ` [PATCH v1 3/9] util/mmap-alloc: factor out reserving of a memory region to mmap_reserve() David Hildenbrand
2021-02-09 13:49 ` [PATCH v1 4/9] util/mmap-alloc: factor out activating of memory to mmap_activate() David Hildenbrand
2021-02-09 13:49 ` [PATCH v1 5/9] softmmu/memory: pass ram_flags into qemu_ram_alloc_from_fd() David Hildenbrand
2021-03-02 17:17   ` Peter Xu
2021-02-09 13:49 ` [PATCH v1 6/9] softmmu/memory: pass ram_flags into memory_region_init_ram_shared_nomigrate() David Hildenbrand
2021-03-02 17:17   ` Peter Xu
2021-02-09 13:49 ` [PATCH v1 7/9] memory: introduce RAM_NORESERVE and wire it up in qemu_ram_mmap() David Hildenbrand
2021-03-02 17:32   ` Peter Xu
2021-03-02 19:02     ` David Hildenbrand
2021-03-02 20:54       ` Peter Xu
2021-03-02 20:58         ` David Hildenbrand
2021-03-03 11:35       ` Cornelia Huck
2021-03-03 11:37         ` David Hildenbrand
2021-03-03 12:12           ` Thomas Huth
2021-03-03 12:24             ` David Hildenbrand
2021-03-03 11:39         ` Thomas Huth
2021-03-03 11:41           ` David Hildenbrand
2021-02-09 13:49 ` [PATCH v1 8/9] util/mmap-alloc: support RAM_NORESERVE via MAP_NORESERVE David Hildenbrand
2021-03-02 17:51   ` Peter Xu
2021-03-02 19:01     ` David Hildenbrand
2021-03-02 21:44       ` Peter Xu
2021-03-03 10:14         ` David Hildenbrand
2021-03-03 17:05           ` Peter Xu
2021-03-04 16:15             ` David Hildenbrand
2021-02-09 13:49 ` [PATCH v1 9/9] hostmem: wire up RAM_NORESERVE via "reserve" property David Hildenbrand
2021-03-02 17:55   ` Peter Xu
2021-03-02 13:12 ` [PATCH v1 0/9] RAM_NORESERVE, MAP_NORESERVE and hostmem " David Hildenbrand

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=20210209134939.13083-1-david@redhat.com \
    --to=david@redhat.com \
    --cc=borntraeger@de.ibm.com \
    --cc=cohuck@redhat.com \
    --cc=dgilbert@redhat.com \
    --cc=ehabkost@redhat.com \
    --cc=groug@kaod.org \
    --cc=i.kotrasinsk@partner.samsung.com \
    --cc=imammedo@redhat.com \
    --cc=mst@redhat.com \
    --cc=muriloo@linux.ibm.com \
    --cc=pasic@linux.ibm.com \
    --cc=pbonzini@redhat.com \
    --cc=peterx@redhat.com \
    --cc=philmd@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=quintela@redhat.com \
    --cc=richard.henderson@linaro.org \
    --cc=stefanha@redhat.com \
    --cc=sw@weilnetz.de \
    --cc=thuth@redhat.com \
    /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).