Hi, I have a question regarding the per-cpu tlbstate logic that is used to lazily switch to the swapper_pgdir when running a process with no mm_struct of its own.
There is a comment in arch/i386/kernel/smp.c:leave_mm() that states 'We need to reload %cr3 since the page tables may be going away from under us'. AFAICT this is not true -- the currently-running task holds a reference on the active_mm until it is context-switched off the CPU, at which point the reference is dropped in sched.c:finish_task_switch(). Until that point the pgd cannot be freed and so kernel mappings should remain valid to use. Although the corresponding function in arch/x86_64 doesn't include this comment, Andi Kleen recently modified it to switch to the swapper_pg_dir, instead of doing a simple __flush_tlb. Does this mean that I am missing something, and the comment in arch/i386 is in fact correct? Thanks, Keir - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
