net: ethernet: Align handling of different ethernet protocols in recv #83967
Conversation
Remove unused family variable from receive function in ethernet layer 2. It is checked only once under such conditions, that do not allow the variable to be different. Signed-off-by: Cla Mattia Galliard <[email protected]>
zephyrbot
requested review from
jukkar,
pdgendt,
rlubos,
ssharks and
tbursztyka
clamattia
force-pushed
the
lldp_unite_with_other_ethernet_protocols
branch
5 times, most recently
from
0506b12 to
dfad2e3
Compare
subsys/net/l2/ethernet/ethernet.c
Outdated
| !net_eth_is_addr_multicast((struct net_eth_addr *)lladdr->addr) && | ||
| !net_eth_is_addr_lldp_multicast( | ||
| (struct net_eth_addr *)lladdr->addr) && | ||
| !net_eth_is_addr_ptp_multicast((struct net_eth_addr *)lladdr->addr) && |
There was a problem hiding this comment.
I am wondering how the group address check replaces the individual address checks. As this PR looks to be more related to refactoring the code, then this change looks to be doing a bit more than that.
Perhaps we should change this code in future PR where the actual ptype functionality is changed to be extensible.
There was a problem hiding this comment.
Fully agree, will revert.
This is needed, as soon as we have made ethernet more extensible. We will either have to provide extentions points for this filter also. Or, what I want to propose, only filter out all non-broadcast-, non-multicast- (in the wider sense), non-unicast- (for us) frames.
There was a problem hiding this comment.
I have handled this address checking issue in #84037
clamattia
force-pushed
the
lldp_unite_with_other_ethernet_protocols
branch
from
dfad2e3 to
8eb1186
Compare
Distribute packets to all ethernet protocols in a unified way. Signed-off-by: Cla Mattia Galliard <[email protected]>
clamattia
force-pushed
the
lldp_unite_with_other_ethernet_protocols
branch
from
8eb1186 to
4ecfdb0
Compare
|
Closing in favor of : #84037 |
Distribute packets to all ethernet protocols in a unified way. This would
also fix an issue, where lldp frames would get dropped too soon.
Refactor to handle
lldpin a unified way witharp,acdandptp.Depends on #83916
Note, that the behavior is slightly different for
lldpas theethernet headeris stripped. But thesrcanddstare now available fromnet_pkt_lladdr_src(pkt)->addr, I believe. This needs to be checked, if correct. Let me know, if you know more.