No description
The preempt mode postcopy has been introduced for a while. From latency POV, it should always win the vanilla postcopy. However there's one thing missing when preempt mode is enabled right now, which is the spatial locality hint when there're page requests from the destination side. In vanilla postcopy, as long as a page request was unqueued, it will update the PSS of the precopy background stream, so that after a page request the background thread will move the pages after whatever was requested. It's pretty much a natural behavior when there's only one channel anyway, and one scanner to send the pages. Preempt mode didn't follow that, because preempt mode has its own channel and its own PSS (which doesn't linearly scan the guest memory, but dedicated to resolve page requested from destination). So the page request process and the background migration process are completely separate. This patch adds the hint explicitly for preempt mode. With that, whenever the preempt mode receives a page request on the source, it will service the remote page fault in the return path, then it'll provide a hint to the background thread so that we'll start sending the pages right after the requested ones in the background, assuming the follow up pages have a higher chance to be accessed later. NOTE: since the background migration thread and return path thread run completely concurrently, it doesn't always mean the hint will be applied every single time. For example, it's possible that the return path thread receives multiple page requests in a row without the background thread getting the chance to consume one. In such case, the preempt thread only provide the hint if the previous hint has been consumed. After all, there's no point queuing hints when we only have one linear scanner. This could measureably improve the simple sequential memory access pattern during postcopy (when preempt is on). For random accesses, I can measure a slight increase of remote page fault latency from ~500us -> ~600us, that could be a trade-off to have such hint mechanism, and after all that's still greatly improved comparing to vanilla postcopy on random (~10ms). The patch is verified by our QE team in a video streaming test case, to reduce the pause of the video from ~1min to a few seconds when switching over to postcopy with preempt mode. Reported-by: Xiaohui Li <xiaohli@redhat.com> Tested-by: Xiaohui Li <xiaohli@redhat.com> Reviewed-by: Juraj Marcin <jmarcin@redhat.com> Link: https://lore.kernel.org/r/20250424220705.195544-1-peterx@redhat.com Signed-off-by: Peter Xu <peterx@redhat.com> |
||
|---|---|---|
| .github/workflows | ||
| .gitlab/issue_templates | ||
| .gitlab-ci.d | ||
| accel | ||
| audio | ||
| authz | ||
| backends | ||
| block | ||
| bsd-user | ||
| chardev | ||
| common-user | ||
| configs | ||
| contrib | ||
| crypto | ||
| disas | ||
| docs | ||
| dump | ||
| ebpf | ||
| fpu | ||
| fsdev | ||
| gdb-xml | ||
| gdbstub | ||
| host/include | ||
| hw | ||
| include | ||
| io | ||
| libdecnumber | ||
| linux-headers | ||
| linux-user | ||
| migration | ||
| monitor | ||
| nbd | ||
| net | ||
| pc-bios | ||
| plugins | ||
| po | ||
| python | ||
| qapi | ||
| qga | ||
| qobject | ||
| qom | ||
| replay | ||
| roms | ||
| rust | ||
| scripts | ||
| scsi | ||
| semihosting | ||
| stats | ||
| storage-daemon | ||
| stubs | ||
| subprojects | ||
| system | ||
| target | ||
| tcg | ||
| tests | ||
| tools | ||
| trace | ||
| ui | ||
| util | ||
| .b4-config | ||
| .dir-locals.el | ||
| .editorconfig | ||
| .exrc | ||
| .gdbinit | ||
| .git-blame-ignore-revs | ||
| .gitattributes | ||
| .gitignore | ||
| .gitlab-ci.yml | ||
| .gitmodules | ||
| .gitpublish | ||
| .mailmap | ||
| .patchew.yml | ||
| .readthedocs.yml | ||
| .travis.yml | ||
| block.c | ||
| blockdev-nbd.c | ||
| blockdev.c | ||
| blockjob.c | ||
| configure | ||
| COPYING | ||
| COPYING.LIB | ||
| cpu-common.c | ||
| cpu-target.c | ||
| event-loop-base.c | ||
| gitdm.config | ||
| hmp-commands-info.hx | ||
| hmp-commands.hx | ||
| iothread.c | ||
| job-qmp.c | ||
| job.c | ||
| Kconfig | ||
| Kconfig.host | ||
| LICENSE | ||
| MAINTAINERS | ||
| Makefile | ||
| meson.build | ||
| meson_options.txt | ||
| module-common.c | ||
| os-posix.c | ||
| os-win32.c | ||
| page-target.c | ||
| page-vary-common.c | ||
| page-vary-target.c | ||
| pythondeps.toml | ||
| qemu-bridge-helper.c | ||
| qemu-edid.c | ||
| qemu-img-cmds.hx | ||
| qemu-img.c | ||
| qemu-io-cmds.c | ||
| qemu-io.c | ||
| qemu-keymap.c | ||
| qemu-nbd.c | ||
| qemu-options.hx | ||
| qemu.nsi | ||
| qemu.sasl | ||
| README.rst | ||
| replication.c | ||
| target-info-stub.c | ||
| target-info.c | ||
| trace-events | ||
| VERSION | ||
| version.rc | ||
=========== QEMU README =========== QEMU is a generic and open source machine & userspace emulator and virtualizer. QEMU is capable of emulating a complete machine in software without any need for hardware virtualization support. By using dynamic translation, it achieves very good performance. QEMU can also integrate with the Xen and KVM hypervisors to provide emulated hardware while allowing the hypervisor to manage the CPU. With hypervisor support, QEMU can achieve near native performance for CPUs. When QEMU emulates CPUs directly it is capable of running operating systems made for one machine (e.g. an ARMv7 board) on a different machine (e.g. an x86_64 PC board). QEMU is also capable of providing userspace API virtualization for Linux and BSD kernel interfaces. This allows binaries compiled against one architecture ABI (e.g. the Linux PPC64 ABI) to be run on a host using a different architecture ABI (e.g. the Linux x86_64 ABI). This does not involve any hardware emulation, simply CPU and syscall emulation. QEMU aims to fit into a variety of use cases. It can be invoked directly by users wishing to have full control over its behaviour and settings. It also aims to facilitate integration into higher level management layers, by providing a stable command line interface and monitor API. It is commonly invoked indirectly via the libvirt library when using open source applications such as oVirt, OpenStack and virt-manager. QEMU as a whole is released under the GNU General Public License, version 2. For full licensing details, consult the LICENSE file. Documentation ============= Documentation can be found hosted online at `<https://www.qemu.org/documentation/>`_. The documentation for the current development version that is available at `<https://www.qemu.org/docs/master/>`_ is generated from the ``docs/`` folder in the source tree, and is built by `Sphinx <https://www.sphinx-doc.org/en/master/>`_. Building ======== QEMU is multi-platform software intended to be buildable on all modern Linux platforms, OS-X, Win32 (via the Mingw64 toolchain) and a variety of other UNIX targets. The simple steps to build QEMU are: .. code-block:: shell mkdir build cd build ../configure make Additional information can also be found online via the QEMU website: * `<https://wiki.qemu.org/Hosts/Linux>`_ * `<https://wiki.qemu.org/Hosts/Mac>`_ * `<https://wiki.qemu.org/Hosts/W32>`_ Submitting patches ================== The QEMU source code is maintained under the GIT version control system. .. code-block:: shell git clone https://gitlab.com/qemu-project/qemu.git When submitting patches, one common approach is to use 'git format-patch' and/or 'git send-email' to format & send the mail to the qemu-devel@nongnu.org mailing list. All patches submitted must contain a 'Signed-off-by' line from the author. Patches should follow the guidelines set out in the `style section <https://www.qemu.org/docs/master/devel/style.html>`_ of the Developers Guide. Additional information on submitting patches can be found online via the QEMU website: * `<https://wiki.qemu.org/Contribute/SubmitAPatch>`_ * `<https://wiki.qemu.org/Contribute/TrivialPatches>`_ The QEMU website is also maintained under source control. .. code-block:: shell git clone https://gitlab.com/qemu-project/qemu-web.git * `<https://www.qemu.org/2017/02/04/the-new-qemu-website-is-up/>`_ A 'git-publish' utility was created to make above process less cumbersome, and is highly recommended for making regular contributions, or even just for sending consecutive patch series revisions. It also requires a working 'git send-email' setup, and by default doesn't automate everything, so you may want to go through the above steps manually for once. For installation instructions, please go to: * `<https://github.com/stefanha/git-publish>`_ The workflow with 'git-publish' is: .. code-block:: shell $ git checkout master -b my-feature $ # work on new commits, add your 'Signed-off-by' lines to each $ git publish Your patch series will be sent and tagged as my-feature-v1 if you need to refer back to it in the future. Sending v2: .. code-block:: shell $ git checkout my-feature # same topic branch $ # making changes to the commits (using 'git rebase', for example) $ git publish Your patch series will be sent with 'v2' tag in the subject and the git tip will be tagged as my-feature-v2. Bug reporting ============= The QEMU project uses GitLab issues to track bugs. Bugs found when running code built from QEMU git or upstream released sources should be reported via: * `<https://gitlab.com/qemu-project/qemu/-/issues>`_ If using QEMU via an operating system vendor pre-built binary package, it is preferable to report bugs to the vendor's own bug tracker first. If the bug is also known to affect latest upstream code, it can also be reported via GitLab. For additional information on bug reporting consult: * `<https://wiki.qemu.org/Contribute/ReportABug>`_ ChangeLog ========= For version history and release notes, please visit `<https://wiki.qemu.org/ChangeLog/>`_ or look at the git history for more detailed information. Contact ======= The QEMU community can be contacted in a number of ways, with the two main methods being email and IRC: * `<mailto:qemu-devel@nongnu.org>`_ * `<https://lists.nongnu.org/mailman/listinfo/qemu-devel>`_ * #qemu on irc.oftc.net Information on additional methods of contacting the community can be found online via the QEMU website: * `<https://wiki.qemu.org/Contribute/StartHere>`_