From: "Toke Høiland-Jørgensen" <toke@redhat.com>
To: cake@lists.bufferbloat.net
Subject: Re: [Cake] [PATCH net] sch_cake: Take advantage of skb->hash where appropriate
Date: Fri, 29 May 2020 15:02:03 +0200 [thread overview]
Message-ID: <871rn228tw.fsf@toke.dk> (raw)
In-Reply-To: <20200529124344.355785-1-toke@redhat.com>
[-- Attachment #1: Type: text/plain, Size: 565 bytes --]
> The immediate motivation for fixing this was the recent patch to WireGuard
> to preserve the skb->hash on encapsulation. As such, this is also what I
> tested against; with this patch, added latency under load for competing
> flows drops from ~8 ms to sub-1ms on an RRUL test over a WireGuard tunnel
> going through a virtual link shaped to 1Gbps using sch_cake. This matches
> the results we saw with a similar setup using sch_fq_codel when testing the
> WireGuard patch.
See attached Flent data files for the full results (dropping netdev@ for
these).
-Toke
[-- Attachment #2: rrul-2020-05-29T125220.474228.cake_no-patch_wg_patched.flent.gz --]
[-- Type: application/gzip, Size: 116901 bytes --]
[-- Attachment #3: rrul-2020-05-29T140517.282425.cake_patch_wg_patched.flent.gz --]
[-- Type: application/gzip, Size: 102927 bytes --]
[-- Attachment #4: rtt_fair_var-2020-05-29T134324.452771.cake_no-patch_wg_patched_nonat.flent.gz --]
[-- Type: application/gzip, Size: 82092 bytes --]
[-- Attachment #5: rtt_fair_var-2020-05-29T135038.694646.cake_no-patch_wg_patched_nonat_dual-srchost.flent.gz --]
[-- Type: application/gzip, Size: 82338 bytes --]
[-- Attachment #6: rtt_fair_var-2020-05-29T135627.430220.cake_no-patch_wg_patched_nonat_dual-srchost.flent.gz --]
[-- Type: application/gzip, Size: 100479 bytes --]
[-- Attachment #7: rtt_fair_var-2020-05-29T135809.540527.cake_no-patch_wg_patched_nat_dual-srchost.flent.gz --]
[-- Type: application/gzip, Size: 101197 bytes --]
[-- Attachment #8: rtt_fair_var-2020-05-29T140711.738176.cake_patch_wg_patched_nonat.flent.gz --]
[-- Type: application/gzip, Size: 80263 bytes --]
[-- Attachment #9: rtt_fair_var-2020-05-29T141034.197074.cake_patch_wg_patched_nonat_dual-srchost.flent.gz --]
[-- Type: application/gzip, Size: 98289 bytes --]
[-- Attachment #10: rtt_fair_var-2020-05-29T141217.145841.cake_patch_wg_patched_nat_dual-srchost.flent.gz --]
[-- Type: application/gzip, Size: 98860 bytes --]
[-- Attachment #11: rtt_fair_var-2020-05-29T141551.629034.cake_patch_wg_patched_nat_dual-srchost.flent.gz --]
[-- Type: application/gzip, Size: 115860 bytes --]
[-- Attachment #12: tcp_nup-2020-05-29T123116.704500.cake_no-patch_wg_patch.flent.gz --]
[-- Type: application/gzip, Size: 216035 bytes --]
[-- Attachment #13: tcp_nup-2020-05-29T140336.810381.cake_patch_wg_patch.flent.gz --]
[-- Type: application/gzip, Size: 353354 bytes --]
next prev parent reply other threads:[~2020-05-29 13:02 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-05-29 12:43 Toke Høiland-Jørgensen
2020-05-29 13:02 ` Toke Høiland-Jørgensen [this message]
2020-05-29 17:57 ` Jakub Kicinski
2020-05-29 18:31 ` Toke Høiland-Jørgensen
2020-05-31 4:52 ` David Miller
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
List information: https://lists.bufferbloat.net/postorius/lists/cake.lists.bufferbloat.net/
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=871rn228tw.fsf@toke.dk \
--to=toke@redhat.com \
--cc=cake@lists.bufferbloat.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