On Wed, Jul 30, 2014 at 12:03:38PM -0700, Venky Venkatesan wrote: > Neil, > > >Hey all- > > I've been trying to update the fedora dpdk package to support VFIO > >enabled drivers and ran into a problem in which ixgbe didn't compile because > >the > >rxtx_vec code uses sse4.2 instruction intrinsics, which aren't supported in > >the > >default config I have. I tried to remedy this by replacing the intrinsics > >with > >the __builtin macros, but it was pointed out (correctly), that this doesn't > >work > >properly. So this is my second attempt, which I actually like a bit better. > > I > >noted that code that uses intrinsics (ixgbe and the acl library), don't need > >to > >have those instructions turned on build-wide. Rather, we can just enable the > >instructions in the specific code we want to build with support for that, and > >test for instruction support dynamically at run time. This allows me to > >build > >the dpdk for a generic platform, but in such a way that some optimizations > >can > >be used if the executing cpu supports them at run time. > > > >Change notes: > > > >v2) > > * Added Log messages to run time check failures per Konstantin > > * Removed run time check caching in acl per Konstantin > > > >Signed-off-by: Neil Horman <nhorman at tuxdriver.com> > >CC: Thomas Monjalon <thomas.monjalon at 6wind.com> > > > > One possible change to look at - the _mm_shuffle_epi8 is something that is > available all the way from SSSE3, so you should be able to leave that in. > The SSE4.2 dependency is only for the popcount (and __builtin_popcountll > will work for that anyway). That should make things simpler. > > -Venky >
ugh, actually both the ixgbe and acl libraries use the popcnt instruction, so I think I'd rather just keep the sse4.2 instruction set as it is, unless you think the __builtin_popcnt is equally performant. Neil
