I tried to write a patch, but then I ran into a problem:
On UP, the kernel calls flush_tlb_all() if kernel pages have changed, I
could add a WBINVD instruction to flush_tlb_all(). flush_tlb_all() is
rare, the WBINVD would not be a performance problem.
But on SMP, there is no "flush_tlb_all()"-IPI, ie I must add a new IPI
[with smp_call_function()].
Do you think that
1) it's not worth the effort.
2) the patch would be a victim of the feature freeze, wait for 2.5
3) write the patch and wait until Linus returns.
-- Manfred- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/