From: "Alex Bennée" <alex.bennee@linaro.org>
To: "Daniel P. Berrangé" <berrange@redhat.com>
Cc: fam@euphon.net, Richard Henderson <richard.henderson@linaro.org>,
qemu-devel@nongnu.org,
Christian Ehrhardt <christian.ehrhardt@canonical.com>,
f4bug@amsat.org, cota@braap.org,
Paolo Bonzini <pbonzini@redhat.com>,
aurelien@aurel32.net, Richard Henderson <rth@twiddle.net>
Subject: Re: [PATCH v2 06/12] accel/tcg: better handle memory constrained systems
Date: Thu, 23 Jul 2020 10:22:25 +0100 [thread overview]
Message-ID: <87d04mtwku.fsf@linaro.org> (raw)
In-Reply-To: <20200723090027.GB2615312@redhat.com>
Daniel P. Berrangé <berrange@redhat.com> writes:
> On Wed, Jul 22, 2020 at 12:02:59PM -0700, Richard Henderson wrote:
>> On 7/22/20 9:44 AM, Daniel P. Berrangé wrote:
>> > OpenStack uses TCG in alot of their CI infrastructure for example
>> > and runs multiple VMs. If there's 4 VMs, that's another 4 GB of
>> > RAM usage just silently added on top of the explicit -m value.
>> >
>> > I wouldn't be surprised if this pushes CI into OOM, even without
>> > containers or cgroups being involved, as they have plenty of other
>> > services consuming RAM in the CI VMs.
>>
>> I would hope that CI would also supply a -tb_size to go along with that -m
>> value. Because we really can't guess on their behalf.
>
> I've never even seen mention of -tb_size argument before myself, nor
> seen anyone else using it and libvirt doesn't set it, so I think
> this is not a valid assumption.
>
>
>> > The commit 600e17b261555c56a048781b8dd5ba3985650013 talks about this
>> > minimizing codegen cache flushes, but doesn't mention the real world
>> > performance impact of eliminating those flushes ?
>>
>> Somewhere on the mailing list was this info. It was so dreadfully slow it was
>> *really* noticable. Timeouts everywhere.
>>
>> > Presumably this makes the guest OS boot faster, but what's the before
>> > and after time ? And what's the time like for values in between the
>> > original 32mb and the new 1 GB ?
>>
>> But it wasn't "the original 32MB".
>> It was the original "ram_size / 4", until that broke due to argument parsing
>> ordering.
>
> Hmm, 600e17b261555c56a048781b8dd5ba3985650013 says it was 32 MB as the
> default in its commit message, which seems to match the code doing
>
> #define DEFAULT_CODE_GEN_BUFFER_SIZE_1 (32 * MiB)
You need to look earlier in the sequence (see the tag pull-tcg-20200228):
47a2def4533a2807e48954abd50b32ecb1aaf29a
so when the argument ordering broke the guest ram_size heuristic we
started getting reports of performance regressions because we fell back
to that size. Before then it was always based on guest ram size within
the min/max bounds set by those defines.
>> I don't know what CI usually uses, but I usually use at least -m 4G, sometimes
>> more. What's the libvirt default?
>
> There's no default memory size - its up to whomever/whatever creates the
> VMs to choose how much RAM is given.
>
> Regards,
> Daniel
--
Alex Bennée
next prev parent reply other threads:[~2020-07-23 9:23 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-07-22 6:28 [PATCH v2 00/12] candidate fixes for 5.1-rc1 (testing, semihosting, OOM tcg, x86 fpu) Alex Bennée
2020-07-22 6:28 ` [PATCH v2 01/12] shippable: add one more qemu to registry url Alex Bennée
2020-07-22 6:28 ` [PATCH v2 02/12] semihosting: defer connect_chardevs a little more to use serialx Alex Bennée
2020-07-22 6:28 ` [PATCH v2 03/12] semihosting: don't send the trailing '\0' Alex Bennée
2020-07-22 6:28 ` [PATCH v2 04/12] util: add qemu_get_host_physmem utility function Alex Bennée
2020-07-22 15:51 ` Richard Henderson
2020-07-22 6:28 ` [PATCH v2 05/12] util/oslib-win32: add qemu_get_host_physmem implementation Alex Bennée
2020-07-22 6:49 ` Philippe Mathieu-Daudé
2020-07-22 10:03 ` Stefan Weil
2020-07-22 10:13 ` Daniel P. Berrangé
2020-07-22 11:33 ` Alex Bennée
2020-07-22 11:38 ` Daniel P. Berrangé
2020-07-22 10:32 ` 罗勇刚(Yonggang Luo)
2020-07-22 10:50 ` Stefan Weil
2020-07-22 11:31 ` Alex Bennée
2020-07-22 11:41 ` Daniel P. Berrangé
2020-07-22 6:28 ` [PATCH v2 06/12] accel/tcg: better handle memory constrained systems Alex Bennée
2020-07-22 15:57 ` Richard Henderson
2020-07-22 16:29 ` Alex Bennée
2020-07-22 16:44 ` Daniel P. Berrangé
2020-07-22 19:02 ` Richard Henderson
2020-07-23 9:00 ` Daniel P. Berrangé
2020-07-23 9:22 ` Alex Bennée [this message]
2020-07-23 9:31 ` Daniel P. Berrangé
2020-07-23 10:06 ` Alex Bennée
2020-07-22 6:28 ` [PATCH v2 07/12] target/i386: floatx80: avoid compound literals in static initializers Alex Bennée
2020-07-22 6:45 ` Philippe Mathieu-Daudé
2020-07-22 6:28 ` [PATCH v2 08/12] linux-user: don't use MAP_FIXED in pgd_find_hole_fallback Alex Bennée
2020-07-22 16:00 ` Richard Henderson
2020-07-22 6:28 ` [PATCH v2 09/12] tests/docker: fix update command due to python3 str/bytes distinction Alex Bennée
2020-07-22 6:29 ` [PATCH v2 10/12] tests/docker: fix binfmt_misc image building Alex Bennée
2020-07-22 6:35 ` Philippe Mathieu-Daudé
2020-07-22 6:29 ` [PATCH v2 11/12] tests/docker: add support for DEB_KEYRING Alex Bennée
2020-07-22 14:27 ` Philippe Mathieu-Daudé
2020-07-22 16:03 ` Richard Henderson
2020-07-22 6:29 ` [PATCH v2 12/12] linux-user: fix clock_nanosleep() Alex Bennée
2020-07-22 6:49 ` Laurent Vivier
2020-07-22 8:33 ` Laurent Vivier
2020-07-22 8:55 ` Alex Bennée
2020-07-22 12:03 ` Laurent Vivier
2020-07-22 7:05 ` Philippe Mathieu-Daudé
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=87d04mtwku.fsf@linaro.org \
--to=alex.bennee@linaro.org \
--cc=aurelien@aurel32.net \
--cc=berrange@redhat.com \
--cc=christian.ehrhardt@canonical.com \
--cc=cota@braap.org \
--cc=f4bug@amsat.org \
--cc=fam@euphon.net \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=richard.henderson@linaro.org \
--cc=rth@twiddle.net \
/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).