sh: Fix multiple UTLB hit on UP SH-4.
authorHideo Saito <saito@densan.co.jp>
Thu, 14 Feb 2008 05:45:08 +0000 (14:45 +0900)
committerPaul Mundt <lethal@linux-sh.org>
Thu, 14 Feb 2008 05:45:08 +0000 (14:45 +0900)
This acts as a reversion of 1c6b2ca5e0939bf8b5d1a11f1646f25189ecd447 in
the case of UP SH-4, where we still have the risk of a multiple hit
between the slow and fast paths. As seen on SH7780.

Signed-off-by: Hideo Saito <saito@densan.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
arch/sh/mm/fault_32.c

index 4ef0a1f1a9ab48701beab038e48ca1d697b3b23e..d1fa27594c6e4f0d7340670597e0c5ee4517b9c5 100644 (file)
@@ -299,6 +299,14 @@ asmlinkage int __kprobes __do_page_fault(struct pt_regs *regs,
                entry = pte_mkdirty(entry);
        entry = pte_mkyoung(entry);
 
+#if defined(CONFIG_CPU_SH4) && !defined(CONFIG_SMP)
+       /*
+        * ITLB is not affected by "ldtlb" instruction.
+        * So, we need to flush the entry by ourselves.
+        */
+       local_flush_tlb_one(get_asid(), address & PAGE_MASK);
+#endif
+
        set_pte(pte, entry);
        update_mmu_cache(NULL, address, entry);