diff options
| author | Dale Rahn <drahn@cvs.openbsd.org> | 2002-01-06 06:03:00 +0000 |
|---|---|---|
| committer | Dale Rahn <drahn@cvs.openbsd.org> | 2002-01-06 06:03:00 +0000 |
| commit | 15b2ac946094abab1577c44b1b2418930eff2693 (patch) | |
| tree | 17a713e40c95cf76f8ff6ad103b394c751e0b48b | |
| parent | b09d348ee766f933bd11c133593fb79c4e2010e3 (diff) | |
That was no fix, that broke things. If the pte entry is currently found
in the po lists, it will NOT have the PTE_VALID bit set. Thus valid
mappings could be ignored if enough mappings existed for that PTEG pair.
This explains the bus_dma panics.
| -rw-r--r-- | sys/arch/powerpc/powerpc/pmap.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/sys/arch/powerpc/powerpc/pmap.c b/sys/arch/powerpc/powerpc/pmap.c index d80145eb13c..40b11085298 100644 --- a/sys/arch/powerpc/powerpc/pmap.c +++ b/sys/arch/powerpc/powerpc/pmap.c @@ -1,4 +1,4 @@ -/* $OpenBSD: pmap.c,v 1.53 2002/01/06 04:04:48 drahn Exp $ */ +/* $OpenBSD: pmap.c,v 1.54 2002/01/06 06:02:59 drahn Exp $ */ /* $NetBSD: pmap.c,v 1.1 1996/09/30 16:34:52 ws Exp $ */ /* @@ -1493,7 +1493,7 @@ pmap_extract(pm, va, pap) int s = splimp(); boolean_t ret; - if (!(ptp = pte_find(pm, va)) || (ptp->pte_hi & PTE_VALID) == 0) { + if (!(ptp = pte_find(pm, va))) { /* return address 0 if not mapped??? */ ret = FALSE; if (pm == pmap_kernel() && va < 0x80000000){ |