Messages in this thread | | | Date | Wed, 09 Jul 2008 14:26:08 -0700 | From | Jeremy Fitzhardinge <> | Subject | Re: [RFC 00/15] x86_64: Optimize percpu accesses |
| |
H. Peter Anvin wrote: > Either way, I really suspect that the right thing to do is to use > negative offsets, with the possible exception of a handful of things > (40 bytes or less, perhaps like current) which can get small positive > offsets and end up in the "super hot" cacheline. > > The sucky part is that I don't believe GNU ld has native support for a > "hanging down" section (one which has a fixed endpoint rather than a > starting point), so it requires extra magic around the link (or > finding some way to do it with linker script functions.)
If you're going to do another linker pass, you could have a script to extract all the percpu symbols and generate a set of derived zero-based ones and then link against that.
Or generate a vmlinux with relocations and "relocate" all the percpu symbols down to 0.
J
| |