On Fri, Feb 25, 2000 at 02:46:16PM -0500, Dave Anderson wrote:
> The problem was due to the disconnect between the active_mm pgd
> value and what's actually stored in the kernel task's ptbr value --
> which is what gets loaded into the PTBR register with each alpha
> context switch.
Yes, I can see the potential for a problem there.
> I note that in 2.3.47, the problem looked to have been addressed by
> the addition of the enter_lazy_tlb() call in schedule():
>
> if (!mm) {
> if (next->active_mm) BUG();
> next->active_mm = oldmm;
> atomic_inc(&oldmm->mm_count);
> +++ enter_lazy_tlb(oldmm, next, this_cpu);
> }
>
> Unfortunately the alpha enter_lazy_tlb() doesn't do anything:
I don't know if enter_lazy_tlb is appropriate. It is used in
two contexts: getting ready to swap context (eg that one) and
not planning on swapping context any time soon (eg start_lazy_tlb).
But perhaps there's no issue at all with the later, since the
"host" process can't really die without some sort of swpctx.
Try this.
r~
--- mmu_context.h.orig Fri Mar 3 18:28:58 2000
+++ mmu_context.h Fri Mar 3 18:29:49 2000
@@ -22,11 +22,6 @@
#include <asm/io.h>
#endif
-static inline void
-enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk, unsigned cpu)
-{
-}
-
extern inline unsigned long
__reload_thread(struct thread_struct *pcb)
{
@@ -221,6 +216,12 @@ extern inline void
destroy_context(struct mm_struct *mm)
{
/* Nothing to do. */
+}
+
+static inline void
+enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk, unsigned cpu)
+{
+ tsk->thread.ptbr = ((unsigned long)mm->pgd - IDENT_ADDR) >> PAGE_SHIFT;
}
#ifdef __MMU_EXTERN_INLINE
-
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/
This archive was generated by hypermail 2b29 : Tue Mar 07 2000 - 21:00:16 EST