On 10/1/20 4:06 AM, Michal Prívozník wrote:
On 10/1/20 1:14 AM, Laine Stump wrote:
> IFLA_VF_MAX was introduced to the Linux kernel in 2.6.35, and was even
> backported to the RHEL*6* 2.6.32 kernel downstream, so it is present
> in all supported versions of all Linux distros that libvirt builds
> on. Additionally, it can't be conditionally compiled out of a
> kernel. There is no reason to conditionalize any piece of code on
> presence of IFLA_VF_MAX - if the platform is Linux, it is supported.
>
> Signed-off-by: Laine Stump <laine(a)redhat.com>
> ---
> src/util/virnetdev.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/src/util/virnetdev.c b/src/util/virnetdev.c
> index e1a4cc2bef..07208a1876 100644
> --- a/src/util/virnetdev.c
> +++ b/src/util/virnetdev.c
> @@ -1473,7 +1473,7 @@ virNetDevSysfsFile(char **pf_sysfs_device_link
> G_GNUC_UNUSED,
> #endif /* !__linux__ */
> -#if defined(__linux__) && defined(WITH_LIBNL) &&
defined(IFLA_VF_MAX)
> +#if defined(__linux__) && defined(WITH_LIBNL)
> static virMacAddr zeroMAC = { .addr = { 0, 0, 0, 0, 0, 0 } };
> @@ -2266,7 +2266,7 @@ virNetDevSetNetConfig(const char *linkdev, int vf,
> }
> -#else /* defined(__linux__) && defined(WITH_LIBNL) &&
> defined(IFLA_VF_MAX) */
> +#else /* defined(__linux__) && defined(WITH_LIBNL) */
> int
> @@ -2309,7 +2309,7 @@ virNetDevSetNetConfig(const char *linkdev
> G_GNUC_UNUSED,
> }
> -#endif /* defined(__linux__) && defined(WITH_LIBNL) &&
> defined(IFLA_VF_MAX) */
> +#endif /* defined(__linux__) && defined(WITH_LIBNL) */
> VIR_ENUM_IMPL(virNetDevIfState,
> VIR_NETDEV_IF_STATE_LAST,
>
This should be squashed in:
diff --git i/meson.build w/meson.build
index 073ea6d49e..592f5882fa 100644
--- i/meson.build
+++ w/meson.build
@@ -1161,8 +1161,7 @@ m_dep = cc.find_library('m', required : false)
use_macvtap = false
if not get_option('macvtap').disabled()
- if (cc.has_header_symbol('linux/if_link.h', 'MACVLAN_MODE_BRIDGE')
and
- cc.has_header_symbol('linux/if_link.h', 'IFLA_VF_MAX'))
+ if cc.has_header_symbol('linux/if_link.h', 'MACVLAN_MODE_BRIDGE') and
Well, really I meant to point out in the commit log that
MACVLAN_MODE_BRIDGE has also been in the kernel since 2.6.33 and so is
always present for all Linuxes.
use_macvtap = true
endif
Ah, you're removing the whole block later in the series. Okay then.
Yeah, that too. I really only left the two patches separate in case
someone complained about removing the user-controlled flag (although I
see 0 use for it, since it doesn't remove any library dependency, and
doesn't reduce the size of any binary by a useful amount).
Michal