[PATCH] powerpc/iseries: Fix double phys_to_abs bug in htab_bolt_mapping

Before the merge I updated create_pte_mapping() to work for iSeries, by
calling iSeries_hpte_bolt_or_insert. (4c55130b2aa93370f1bf52d2304394e91cf8ee39)

Later we changed iSeries_hpte_insert to cope with the bolting case, and called
that instead from create_pte_mapping() (which was renamed to htab_bolt_mapping)
(3c726f8dee6f55e96475574e9f645327e461884c).

Unfortunately that change introduced a subtle bug, where we pass an absolute
address to iSeries_hpte_insert() where it expects a physical address. This
leads to us calling phys_to_abs() twice on the physical address, which is
seriously bogus.

This only causes a problem if the absolute address from the first translation
can be looked up again in the chunk_map, which depends on the size and layout
of memory. I've seen it fail on one box, but not others.

The minimal fix is to pass the physical address to iSeries_hpte_insert(). For
2.6.17 we should make phys_to_abs() BUG if we try to double-translate an
address.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>

diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c
index b1f614c612dd136963cf4c480b90f306fb848b06..e9d589eefc14ce07aa5f670c036b35bfc3124328 100644 (file)
--- a/arch/powerpc/mm/hash_utils_64.c
+++ b/arch/powerpc/mm/hash_utils_64.c
@@ -169,7 +169,7 @@ int htab_bolt_mapping(unsigned long vstart, unsigned long vend,
 #ifdef CONFIG_PPC_ISERIES
                if (_machine == PLATFORM_ISERIES_LPAR)
                        ret = iSeries_hpte_insert(hpteg, va,
-                                                 virt_to_abs(paddr),
+                                                 __pa(vaddr),
                                                  tmp_mode,
                                                  HPTE_V_BOLTED,
                                                  psize);