The patch titled
m68knommu: remove is_in_rom() function
has been added to the -mm tree. Its filename is
m68knommu-remove-is_in_rom-function.patch
*** Remember to use Documentation/SubmitChecklist when testing your code ***
See http://www.zip.com.au/~akpm/linux/patches/stuff/added-to-mm.txt to find
out what to do about this
------------------------------------------------------
Subject: m68knommu: remove is_in_rom() function
From: Greg Ungerer <gerg@xxxxxxxxxxxx>
Remove is_in_rom() function. It doesn't actually serve the purpose it was
intended to. If you look at the use of it _access_ok() (which is the only use
of it) then it is obvious that most of memory is marked as access_ok. No
point having is_in_rom() then, so remove it.
Signed-off-by: Greg Ungerer <gerg@xxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
arch/m68knommu/kernel/m68k_ksyms.c | 2 --
arch/m68knommu/mm/memory.c | 20 --------------------
include/asm-m68knommu/pgtable.h | 1 -
include/asm-m68knommu/uaccess.h | 11 +++++++----
4 files changed, 7 insertions(+), 27 deletions(-)
diff -puN arch/m68knommu/kernel/m68k_ksyms.c~m68knommu-remove-is_in_rom-function arch/m68knommu/kernel/m68k_ksyms.c
--- a/arch/m68knommu/kernel/m68k_ksyms.c~m68knommu-remove-is_in_rom-function
+++ a/arch/m68knommu/kernel/m68k_ksyms.c
@@ -81,8 +81,6 @@ EXPORT_SYMBOL(__mulsi3);
EXPORT_SYMBOL(__udivsi3);
EXPORT_SYMBOL(__umodsi3);
-EXPORT_SYMBOL(is_in_rom);
-
#ifdef CONFIG_COLDFIRE
extern unsigned int *dma_device_address;
extern unsigned long dma_base_addr, _ramend;
diff -puN arch/m68knommu/mm/memory.c~m68knommu-remove-is_in_rom-function arch/m68knommu/mm/memory.c
--- a/arch/m68knommu/mm/memory.c~m68knommu-remove-is_in_rom-function
+++ a/arch/m68knommu/mm/memory.c
@@ -33,23 +33,3 @@ unsigned long kernel_map(unsigned long p
return paddr;
}
-
-int is_in_rom(unsigned long addr)
-{
- extern unsigned long _ramstart, _ramend;
-
- /*
- * What we are really trying to do is determine if addr is
- * in an allocated kernel memory region. If not then assume
- * we cannot free it or otherwise de-allocate it. Ideally
- * we could restrict this to really being in a ROM or flash,
- * but that would need to be done on a board by board basis,
- * not globally.
- */
- if ((addr < _ramstart) || (addr >= _ramend))
- return(1);
-
- /* Default case, not in ROM */
- return(0);
-}
-
diff -puN include/asm-m68knommu/pgtable.h~m68knommu-remove-is_in_rom-function include/asm-m68knommu/pgtable.h
--- a/include/asm-m68knommu/pgtable.h~m68knommu-remove-is_in_rom-function
+++ a/include/asm-m68knommu/pgtable.h
@@ -49,7 +49,6 @@ static inline int pte_file(pte_t pte) {
* These would be in other places but having them here reduces the diffs.
*/
extern unsigned int kobjsize(const void *objp);
-extern int is_in_rom(unsigned long);
/*
* No page table caches to initialise.
diff -puN include/asm-m68knommu/uaccess.h~m68knommu-remove-is_in_rom-function include/asm-m68knommu/uaccess.h
--- a/include/asm-m68knommu/uaccess.h~m68knommu-remove-is_in_rom-function
+++ a/include/asm-m68knommu/uaccess.h
@@ -15,12 +15,15 @@
#define access_ok(type,addr,size) _access_ok((unsigned long)(addr),(size))
+/*
+ * It is not enough to just have access_ok check for a real RAM address.
+ * This would disallow the case of code/ro-data running XIP in flash/rom.
+ * Ideally we would check the possible flash ranges too, but that is
+ * currently not so easy.
+ */
static inline int _access_ok(unsigned long addr, unsigned long size)
{
- extern unsigned long memory_start, memory_end;
-
- return (((addr >= memory_start) && (addr+size < memory_end)) ||
- (is_in_rom(addr) && is_in_rom(addr+size)));
+ return 1;
}
/*
_
Patches currently in -mm which might be from gerg@xxxxxxxxxxxx are
origin.patch
git-mtd.patch
m68knommu-generic-irq-handling.patch
m68knommu-start-dump-from-exception-stack.patch
m68knommu-remove-is_in_rom-function.patch
-
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
