net: skbuff: reorder bytes 2 and 3 of the bitfield
authorJakub Kicinski <kuba@kernel.org>
Tue, 21 Mar 2023 01:41:14 +0000 (18:41 -0700)
committerMartin KaFai Lau <martin.lau@kernel.org>
Tue, 21 Mar 2023 04:57:49 +0000 (21:57 -0700)
commitb94e032b7ad6318b36615f0e0cc3b0d61a5531e8
treeb895dc3f9d3caa60a32f5b28849fd8e141506cad
parent04aae213e719ec2bb310158c4025316ace50589b
net: skbuff: reorder bytes 2 and 3 of the bitfield

BPF needs to know the offsets of fields it tries to access.
Zero-length fields are added to make offsetof() work.
This unfortunately partitions the bitfield (fields across
the zero-length members can't be coalesced).

Reorder bytes 2 and 3, BPF needs to know the offset of fields
previously in byte 3 and some fields in byte 2 should really
be optional.

The two bytes are always in the same cacheline so it should
not matter.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/r/20230321014115.997841-3-kuba@kernel.org
Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
include/linux/skbuff.h