From: "DERUMIER, Alexandre" <alexandre.derumier@groupe-cyllene.com>
To: "pve-devel@lists.proxmox.com" <pve-devel@lists.proxmox.com>,
"aderumier@odiso.com" <aderumier@odiso.com>,
"f.ebner@proxmox.com" <f.ebner@proxmox.com>
Subject: Re: [pve-devel] [PATCH-SERIES v3 qemu-server/manager/common] add and set x86-64-v2 as default model for new vms and detect best cpumodel
Date: Fri, 2 Jun 2023 09:13:16 +0000 [thread overview]
Message-ID: <81d002bd8aeb44d29f268d44a80ec7a544914791.camel@groupe-cyllene.com> (raw)
In-Reply-To: <c034772d-9088-ac01-efc3-03249661694f@proxmox.com>
>
> Note that migration between CPUs of different vendors is not a
> supported
> use case (it will always depend on specific models, kernel versions,
> etc.), so we can only justify not adding it to the new default model
> if
> it doesn't make life worse for everybody else.
>
> And I'd be a bit careful to jump to general conclusions just from one
> forum post.
>
> It seems like you were the one adding the flag ;)
>
>
yes, I known ^_^ (It's the default too on rhev, but they are not
supporting amd-intel migration too).
> and the LWN-archived mail linked in the commit message says
>
> > Ticket locks have an inherent problem in a virtualized case,
> > because
> > the vCPUs are scheduled rather than running concurrently (ignoring
> > gang scheduled vCPUs). This can result in catastrophic performance
> > collapses when the vCPU scheduler doesn't schedule the correct
> > "next"
> > vCPU, and ends up scheduling a vCPU which burns its entire
> > timeslice
> > spinning. (Note that this is not the same problem as lock-holder
> > preemption, which this series also addresses; that's also a
> > problem,
> > but not catastrophic).
>
> "catastrophic performance collapses" doesn't sound very promising :/
>
I have found another thread here:
https://lore.kernel.org/all/0484ea3f-4ba7-4b93-e976-098c5717166e@redhat.com/
where paolo have done benchmark with only 3% difference.
but yes, still slower anyway.
at minimum, it could be interesting to expose the flag in the gui, for
users really needed intel-amd migration.
> But if we find that
> kvm64,enforce,+kvm_pv_eoi,+kvm_pv_unhalt,+sep,+lahf_lm,+popcnt,+sse4.
> 1,+sse4.2,+ssse3
> causes issues (even if not cross-vendor live-migrating) with the
> +kvm_pv_unhalt flag, but not without, it would be a much more
> convincing
> reason against adding the flag for the new default.
>
ok ! (I will send a new patch version today)
next prev parent reply other threads:[~2023-06-02 9:13 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-05-22 10:25 Alexandre Derumier
2023-05-22 10:25 ` [pve-devel] [PATCH v3 qemu-server 1/7] cpuconfig: add new x86-64-vX models Alexandre Derumier
2023-05-31 11:08 ` Fiona Ebner
2023-05-31 15:08 ` DERUMIER, Alexandre
2023-06-01 9:17 ` Fiona Ebner
2023-06-01 11:27 ` DERUMIER, Alexandre
2023-05-22 10:25 ` [pve-devel] [PATCH v2 pve-manager 1/1] qemu: processor : set x86-64-v2 as default cputype for create wizard Alexandre Derumier
2023-05-22 10:25 ` [pve-devel] [PATCH pve-common 1/1] read_cpuinfo: add msr support Alexandre Derumier
2023-05-22 10:25 ` [pve-devel] [PATCH v3 qemu-server 2/7] cpumodel: add cpu models with flags Alexandre Derumier
2023-05-22 10:25 ` [pve-devel] [PATCH v3 qemu-server 3/7] cpumodel: compute qemu supported flags Alexandre Derumier
2023-05-22 10:25 ` [pve-devel] [PATCH v3 qemu-server 4/7] cpuconfig: add get_host_cpu_flags Alexandre Derumier
2023-05-22 10:25 ` [pve-devel] [PATCH v3 qemu-server 5/7] cpuconfig: add find_best_cpumodel Alexandre Derumier
2023-05-22 10:25 ` [pve-devel] [PATCH v3 qemu-server 6/7] cpuconfig: add find_hosts_common_flags Alexandre Derumier
2023-05-22 10:25 ` [pve-devel] [PATCH v3 qemu-server 7/7] add best cpu model detection tests Alexandre Derumier
2023-05-31 11:36 ` [pve-devel] [PATCH-SERIES v3 qemu-server/manager/common] add and set x86-64-v2 as default model for new vms and detect best cpumodel Fiona Ebner
2023-05-31 14:34 ` DERUMIER, Alexandre
2023-06-01 8:34 ` Fiona Ebner
2023-06-01 9:06 ` DERUMIER, Alexandre
2023-06-03 14:14 ` Thomas Lamprecht
2023-06-04 6:29 ` DERUMIER, Alexandre
2023-06-03 14:05 ` Thomas Lamprecht
2023-06-01 9:34 ` Fiona Ebner
2023-06-01 11:37 ` DERUMIER, Alexandre
2023-06-01 13:53 ` DERUMIER, Alexandre
2023-06-01 15:56 ` Fiona Ebner
2023-06-01 21:15 ` DERUMIER, Alexandre
2023-06-02 7:28 ` Fiona Ebner
2023-06-02 9:13 ` DERUMIER, Alexandre [this message]
2023-06-02 11:13 ` Fiona Ebner
2023-06-02 11:44 ` DERUMIER, Alexandre
2023-06-03 14:21 ` Thomas Lamprecht
[not found] ` <8277a27b-a70f-b731-69f7-fc9ae69b2da2@binovo.es>
2023-06-01 16:00 ` Fiona Ebner
2023-06-02 12:41 ` Aaron Lauterer
2023-06-02 14:15 ` DERUMIER, Alexandre
2023-06-02 16:09 ` Aaron Lauterer
2023-06-02 16:27 ` DERUMIER, Alexandre
[not found] ` <fa3565e5-3a9c-9348-f291-554a0e0d6628@binovo.es>
2023-06-06 9:15 ` Fiona Ebner
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=81d002bd8aeb44d29f268d44a80ec7a544914791.camel@groupe-cyllene.com \
--to=alexandre.derumier@groupe-cyllene.com \
--cc=aderumier@odiso.com \
--cc=f.ebner@proxmox.com \
--cc=pve-devel@lists.proxmox.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