Date: Sat, 05 Feb 2000 11:57:34 +0100
From: Manfred Spraul <manfreds@colorfullife.com>
I have a unrelated question about TLB flushes: I think I found a
problem for architectures that access "current->active_mm" from the
tlb flush IPI: Between switch_mm() and switch_to(),
"current->active_mm" and the loaded mm can differ.
i386 was broken, and I sent a patch to Linus last week. Could you
check if Alpha/Sparc have the same problem?
Sparc64 should be ok, the only way we ever check current->active_mm is
to compare it against the mm passed into the flush routine on SMP and
we perform certain xcall optimizations only of they are equal.
Later,
David S. Miller
davem@redhat.com
-
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 : Mon Feb 07 2000 - 21:00:12 EST