diff options
| author | Ben Pfaff <blp@nicira.com> | 2014-06-13 15:28:29 -0700 |
|---|---|---|
| committer | Ben Pfaff <blp@nicira.com> | 2014-08-04 11:11:44 -0700 |
| commit | 6a54dedc66931ef01c3cabb6d6fd481461c99687 (patch) | |
| tree | d4e3bb094193c88a639fae7302ac460c9389062c /build-aux/extract-odp-netlink-h | |
| parent | 837eefc76b3c79bb790a4c4c2d0a314d81b71a28 (diff) | |
| download | openvswitch-6a54dedc66931ef01c3cabb6d6fd481461c99687.tar.gz | |
odp-netlink.h: Use 32-bit aligned 64-bit types.
Open vSwitch userspace uses special types to indicate that a particular
object may not be naturally aligned. Netlink is one source of such
problems: in Netlink, 64-bit integers are often aligned only on 32-bit
boundaries. This commit changes the odp-netlink.h that is transformed from
<linux/openvswitch.h> to use these types to make it harder to accidentally
access a misaligned 64-bit member.
Signed-off-by: Ben Pfaff <blp@nicira.com>
Acked-by: Jesse Gross <jesse@nicira.com>
Diffstat (limited to 'build-aux/extract-odp-netlink-h')
| -rwxr-xr-x | build-aux/extract-odp-netlink-h | 12 |
1 files changed, 8 insertions, 4 deletions
diff --git a/build-aux/extract-odp-netlink-h b/build-aux/extract-odp-netlink-h index c61a1b73a..068fde43b 100755 --- a/build-aux/extract-odp-netlink-h +++ b/build-aux/extract-odp-netlink-h @@ -14,18 +14,22 @@ # Avoid using reserved names in header guards. s/_LINUX_OPENVSWITCH_H/ODP_NETLINK_H/ -# Transform Linux-specific __u<N> types into C99 uint<N>_t types, -# and Linux-specific __be<N> into Open vSwitch ovs_be<N>, +# Transform most Linux-specific __u<N> types into C99 uint<N>_t types, +# and most Linux-specific __be<N> into Open vSwitch ovs_be<N>, # and use the appropriate userspace header. s,<linux/types\.h>,"openvswitch/types.h", -s/__u64/uint64_t/g s/__u32/uint32_t/g s/__u16/uint16_t/g s/__u8/uint8_t/g -s/__be64/ovs_be64/g s/__be32/ovs_be32/g s/__be16/ovs_be16/g +# Transform 64-bit Linux-specific types into Open vSwitch specialized +# types for 64-bit numbers that might only be aligned on a 32-bit +# boundary. +s/__u64/ovs_32aligned_u64/g +s/__be64/ovs_32aligned_be64/g + # Use OVS's own ETH_ADDR_LEN instead of Linux-specific ETH_ALEN. s,<linux/if_ether\.h>,"packets.h", s/ETH_ALEN/ETH_ADDR_LEN/ |