https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=231151
Stephen Hurd changed:
What|Removed |Added
Status|New |Closed
Resolution|---
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=231151
--- Comment #13 from commit-h...@freebsd.org ---
A commit references this bug:
Author: shurd
Date: Thu Sep 20 19:35:35 UTC 2018
New revision: 338838
URL: https://svnweb.freebsd.org/changeset/base/338838
Log:
Fix capabilities handling for
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=231151
--- Comment #12 from Kaho Toshikazu ---
(In reply to Shirkdog from comment #11)
> I did notice that I no longer observed the any of the IPv6 flags
> when I try to enable them,
It is correct because em doesn't have ipv6 offloads. You can
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=231151
--- Comment #11 from Shirkdog ---
(In reply to Stephen Hurd from comment #9)
rebuilt and tested using the update and they appear to be honored (running
ifconfig em0 promisc up at boot:
# ifconfig em0
em0: flags=28943
metric 0 mtu 1500
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=231151
--- Comment #10 from Kaho Toshikazu ---
(In reply to Stephen Hurd from comment #8)
> I'm trying to figure out why the old code was forcing only
> IFCAP_WOL_MCAST|IFCAP_WOL_MAGIC if any WoL flag (including IFCAL_WOL_UCAST)
> was set.
I sus
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=231151
Stephen Hurd changed:
What|Removed |Added
Attachment #197054|0 |1
is obsolete|
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=231151
--- Comment #8 from Stephen Hurd ---
(In reply to Kaho Toshikazu from comment #7)
> the variable setmask should be also masked by isc_capabilities, I think.
Everything going into setmask should be masked by mask already... I've got that
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=231151
--- Comment #7 from Kaho Toshikazu ---
(In reply to Stephen Hurd from comment #5)
> + mask &= ctx->ifc_softc_ctx.isc_capabilities;
An em class device doesn't have rxcsum6 capability and this line is useful.
If a device doesn't
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=231151
--- Comment #6 from Stephen Hurd ---
(In reply to Kaho Toshikazu from comment #3)
> I think iflib should not do a special thing, and it is better that
> an individual driver has some workarounds if need.
The driver isn't involved in this
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=231151
--- Comment #5 from Stephen Hurd ---
Created attachment 197054
--> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=197054&action=edit
Allow rxcsum to be disabled, but keep it tied to rxcsum6
This patch allows rxcsum to be disabled, b
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=231151
--- Comment #4 from Shirkdog ---
(In reply to Kaho Toshikazu from comment #3)
I think that is it okay for this, if someone wants to disable a NIC feature for
IPv4, there is no reason to leave it enabled for IPv6.
I am also rebuilding a k
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=231151
--- Comment #3 from Kaho Toshikazu ---
(In reply to Stephen Hurd from comment #2)
> I'll see if I can figure out why r307562 made this change, and come up
> with a Better Way to twi the two flags together.
I don't know exactly but igb de
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=231151
--- Comment #2 from Stephen Hurd ---
Yeah, for the RX checksum stuff, it's pretty clear that if *either* rxcsum or
rxcsum6 is set in the target capabilities, both will be forced on. Since
ifconfig does a separate ioctl for each command-lin
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=231151
Kaho Toshikazu changed:
What|Removed |Added
CC||k...@elam.kais.kyoto-u.ac.j
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=231151
Mark Linimon changed:
What|Removed |Added
CC||sbr...@freebsd.org
Keywor
15 matches
Mail list logo
