sh64: Invert page fault fast-path error path values.

This brings the sh64 version in line with the sh32 one with regards to
how errors are handled. Base work for further unification of the
implementations.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

diff --git a/arch/sh/kernel/cpu/sh5/entry.S b/arch/sh/kernel/cpu/sh5/entry.S
index de60dc8d737dd71645207fbfbadf749ac13569ea..ff1f0e6e9becd18f111d3a3d83730fd4860dfe02 100644 (file)
--- a/arch/sh/kernel/cpu/sh5/entry.S
+++ b/arch/sh/kernel/cpu/sh5/entry.S
@@ -335,7 +335,7 @@ tlb_miss:
        /* If the fast path handler fixed the fault, just drop through quickly
           to the restore code right away to return to the excepting context.
           */
-       beqi/u  r2, 0, tr1
+       bnei/u  r2, 0, tr1
 
 fast_tlb_miss_restore:
        ld.q    SP, SAVED_TR0, r2

diff --git a/arch/sh/mm/tlbex_64.c b/arch/sh/mm/tlbex_64.c
index 98b64278f8c7b5fc7684d1801d41170820f4bf7b..59cb058217a0254195497d21b5267075921a5cba 100644 (file)
--- a/arch/sh/mm/tlbex_64.c
+++ b/arch/sh/mm/tlbex_64.c
@@ -53,23 +53,23 @@ static int handle_vmalloc_fault(struct mm_struct *mm,
 
        pud = pud_offset(dir, address);
        if (pud_none_or_clear_bad(pud))
-               return 0;
+               return 1;
 
        pmd = pmd_offset(pud, address);
        if (pmd_none_or_clear_bad(pmd))
-               return 0;
+               return 1;
 
        pte = pte_offset_kernel(pmd, address);
        entry = *pte;
 
        if (pte_none(entry) || !pte_present(entry))
-               return 0;
+               return 1;
        if ((pte_val(entry) & protection_flags) != protection_flags)
-               return 0;
+               return 1;
 
        update_mmu_cache(NULL, address, pte);
 
-       return 1;
+       return 0;
 }
 
 static int handle_tlbmiss(struct mm_struct *mm,
@@ -94,27 +94,27 @@ static int handle_tlbmiss(struct mm_struct *mm,
           the next test is necessary.  - RPC */
        if (address >= (unsigned long) TASK_SIZE)
                /* upper half - never has page table entries. */
-               return 0;
+               return 1;
 
        dir = pgd_offset(mm, address);
        if (pgd_none(*dir) || !pgd_present(*dir))
-               return 0;
+               return 1;
        if (!pgd_present(*dir))
-               return 0;
+               return 1;
 
        pud = pud_offset(dir, address);
        if (pud_none(*pud) || !pud_present(*pud))
-               return 0;
+               return 1;
 
        pmd = pmd_offset(pud, address);
        if (pmd_none(*pmd) || !pmd_present(*pmd))
-               return 0;
+               return 1;
 
        pte = pte_offset_kernel(pmd, address);
        entry = *pte;
 
        if (pte_none(entry) || !pte_present(entry))
-               return 0;
+               return 1;
 
        /*
         * If the page doesn't have sufficient protection bits set to
@@ -123,11 +123,11 @@ static int handle_tlbmiss(struct mm_struct *mm,
         * handler.
         */
        if ((pte_val(entry) & protection_flags) != protection_flags)
-               return 0;
+               return 1;
 
        update_mmu_cache(NULL, address, pte);
 
-       return 1;
+       return 0;
 }
 
 /*
@@ -214,12 +214,12 @@ asmlinkage int do_fast_page_fault(unsigned long long ssr_md,
                         * Process-contexts can never have this address
                         * range mapped
                         */
-                       if (handle_vmalloc_fault(mm, protection_flags, address))
-                               return 1;
+                       if (handle_vmalloc_fault(mm, protection_flags, address) == 0)
+                               return 0;
        } else if (!in_interrupt() && mm) {
-               if (handle_tlbmiss(mm, protection_flags, address))
-                       return 1;
+               if (handle_tlbmiss(mm, protection_flags, address) == 0)
+                       return 0;
        }
 
-       return 0;
+       return 1;
 }