[prev in list] [next in list] [prev in thread] [next in thread]
List: illumos-developer
Subject: =?UTF-8?B?W2RldmVsb3Blcl0gUmV2aWV3IOKAkyAxNjQ2OSBgbWFjX3J4X2Zs?= =?UTF-8?B?b3dgIHNob3VsZCBhdHRlbXB0I
From: kyle () oxide ! computer
Date: 2024-04-12 12:44:30
Message-ID: 17129258700.9C211c.6551 () composer ! illumos ! topicbox ! com
[Download RAW message or body]
--17129258701.7b0bcfb.6551
Date: Fri, 12 Apr 2024 08:44:30 -0400
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Hello,
Could you please review the below change?
https://www.illumos.org/issues/16469
https://code.illumos.org/c/illumos-gate/+/3419
This change ports the chain processing logic from
mac_rx_srs_subflow_process into mac_rx_flow, which allows for mblk_t
chains to be received when a mac ring is set to MAC_SW_CLASSIFIER.
This does include some minor logical differences -- packets which do not
match a flent in the mac_impl_t's flow table will not cause the
in-progress chain to be flushed, as all of these packets are already
placed into a separate chain to be freed. This has the benefit of
allowing larger chains if the device also happens to be in promiscuous
mode.
Thanks,
--Kyle
------------------------------------------
illumos: illumos-developer
Permalink: https://illumos.topicbox.com/groups/developer/Tb83bcb1f6eef354f-M29c812fe65863e5d344af22b
Delivery options: https://illumos.topicbox.com/groups/developer/subscription
--17129258701.7b0bcfb.6551
Date: Fri, 12 Apr 2024 08:44:30 -0400
MIME-Version: 1.0
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE html>
<html><html><html><body><div>Hello,<br /></div><div><br /></div><div>Could you please \
review the below change?<br /></div><div><br /></div><div><a \
href="https://www.illumos.org/issues/16469">https://www.illumos.org/issues/16469</a><br \
/></div><div><a href="https://code.illumos.org/c/illumos-gate/+/3419">https://code.illumos.org/c/illumos-gate/+/3419</a><br \
/></div><div><br /></div><div>This change ports the chain processing logic from<br \
/></div><div>mac_rx_srs_subflow_process into mac_rx_flow, which allows for mblk_t<br \
/></div><div>chains to be received when a mac ring is set to MAC_SW_CLASSIFIER.<br \
/></div><div><br /></div><div>This does include some minor logical differences -- \
packets which do not<br /></div><div>match a flent in the mac_impl_t's flow table \
will not cause the<br /></div><div>in-progress chain to be flushed, as all of these \
packets are already<br /></div><div>placed into a separate chain to be freed. This \
has the benefit of<br /></div><div>allowing larger chains if the device also happens \
to be in promiscuous<br /></div><div>mode.<br /></div><div><br \
/></div><div>Thanks,<br /></div><div>--Kyle<br /></div><div id="topicbox-footer" \
style="margin:10px 0 0;border-top:1px solid \
#ddd;border-color:rgba(0,0,0,.15);padding:7px 0;">
<strong><a href="https://illumos.topicbox.com/latest" \
style="color:inherit;text-decoration:none">illumos</a></strong> / illumos-developer \
/ see <a href="https://illumos.topicbox.com/groups/developer">discussions</a>
+
<a href="https://illumos.topicbox.com/groups/developer/members">participants</a>
+
<a href="https://illumos.topicbox.com/groups/developer/subscription">delivery options</a>
<a href="https://illumos.topicbox.com/groups/developer/Tb83bcb1f6eef354f-M29c812fe65863e5d344af22b" \
style="float:right">Permalink</a> </div>
</body></html></html></html>
--17129258701.7b0bcfb.6551--
[prev in list] [next in list] [prev in thread] [next in thread]
Configure |
About |
News |
Add a list |
Sponsored by KoreLogic