kernel_optimize_test/arch/arm/mm/copypage-v4wb.c
Nicolas Pitre 43ae286b7d [ARM] fix a couple clear_user_highpage assembly constraints
In all cases the kaddr is assigned an input register even though it is
modified in the assembly code.  Let's assign a new variable to the
modified value and mark those inline asm with volatile otherwise they
get optimized away because the output variable is otherwise not used.

Also fix a few conversion errors in copypage-feroceon.c and
copypage-v4mc.c.

Signed-off-by: Nicolas Pitre <nico@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2008-11-28 15:36:43 +00:00

95 lines
2.8 KiB
C

/*
* linux/arch/arm/mm/copypage-v4wb.c
*
* Copyright (C) 1995-1999 Russell King
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License version 2 as
* published by the Free Software Foundation.
*/
#include <linux/init.h>
#include <linux/highmem.h>
/*
* ARMv4 optimised copy_user_highpage
*
* We flush the destination cache lines just before we write the data into the
* corresponding address. Since the Dcache is read-allocate, this removes the
* Dcache aliasing issue. The writes will be forwarded to the write buffer,
* and merged as appropriate.
*
* Note: We rely on all ARMv4 processors implementing the "invalidate D line"
* instruction. If your processor does not supply this, you have to write your
* own copy_user_highpage that does the right thing.
*/
static void __attribute__((naked))
v4wb_copy_user_page(void *kto, const void *kfrom)
{
asm("\
stmfd sp!, {r4, lr} @ 2\n\
mov r2, %0 @ 1\n\
ldmia r1!, {r3, r4, ip, lr} @ 4\n\
1: mcr p15, 0, r0, c7, c6, 1 @ 1 invalidate D line\n\
stmia r0!, {r3, r4, ip, lr} @ 4\n\
ldmia r1!, {r3, r4, ip, lr} @ 4+1\n\
stmia r0!, {r3, r4, ip, lr} @ 4\n\
ldmia r1!, {r3, r4, ip, lr} @ 4\n\
mcr p15, 0, r0, c7, c6, 1 @ 1 invalidate D line\n\
stmia r0!, {r3, r4, ip, lr} @ 4\n\
ldmia r1!, {r3, r4, ip, lr} @ 4\n\
subs r2, r2, #1 @ 1\n\
stmia r0!, {r3, r4, ip, lr} @ 4\n\
ldmneia r1!, {r3, r4, ip, lr} @ 4\n\
bne 1b @ 1\n\
mcr p15, 0, r1, c7, c10, 4 @ 1 drain WB\n\
ldmfd sp!, {r4, pc} @ 3"
:
: "I" (PAGE_SIZE / 64));
}
void v4wb_copy_user_highpage(struct page *to, struct page *from,
unsigned long vaddr)
{
void *kto, *kfrom;
kto = kmap_atomic(to, KM_USER0);
kfrom = kmap_atomic(from, KM_USER1);
v4wb_copy_user_page(kto, kfrom);
kunmap_atomic(kfrom, KM_USER1);
kunmap_atomic(kto, KM_USER0);
}
/*
* ARMv4 optimised clear_user_page
*
* Same story as above.
*/
void v4wb_clear_user_highpage(struct page *page, unsigned long vaddr)
{
void *ptr, *kaddr = kmap_atomic(page, KM_USER0);
asm volatile("\
mov r1, %2 @ 1\n\
mov r2, #0 @ 1\n\
mov r3, #0 @ 1\n\
mov ip, #0 @ 1\n\
mov lr, #0 @ 1\n\
1: mcr p15, 0, %0, c7, c6, 1 @ 1 invalidate D line\n\
stmia %0!, {r2, r3, ip, lr} @ 4\n\
stmia %0!, {r2, r3, ip, lr} @ 4\n\
mcr p15, 0, %0, c7, c6, 1 @ 1 invalidate D line\n\
stmia %0!, {r2, r3, ip, lr} @ 4\n\
stmia %0!, {r2, r3, ip, lr} @ 4\n\
subs r1, r1, #1 @ 1\n\
bne 1b @ 1\n\
mcr p15, 0, r1, c7, c10, 4 @ 1 drain WB"
: "=r" (ptr)
: "0" (kaddr), "I" (PAGE_SIZE / 64)
: "r1", "r2", "r3", "ip", "lr");
kunmap_atomic(kaddr, KM_USER0);
}
struct cpu_user_fns v4wb_user_fns __initdata = {
.cpu_clear_user_highpage = v4wb_clear_user_highpage,
.cpu_copy_user_highpage = v4wb_copy_user_highpage,
};