powerpc: Avoid unaligned loads and stores in boot memcpy code

The 601 processor will generate an alignment exception for accesses
which cross a page boundary.  In the boot wrapper code, OF is still
handling all exceptions, and it doesn't have an alignment exception
handler that emulates the instruction and continues.

This changes the memcpy and memmove routines in the boot wrapper to
avoid doing unaligned accesses.  If the source and destination are
misaligned with respect to each other, we just copy one byte at a
time.

Signed-off-by: Paul Mackerras <paulus@samba.org>

diff --git a/arch/powerpc/boot/string.S b/arch/powerpc/boot/string.S
index b1eeaed7db177e273c6e014c4d0c882e82cee739..ac3d43b6a324c97549d8c181e2896ee2f41d377f 100644 (file)
--- a/arch/powerpc/boot/string.S
+++ b/arch/powerpc/boot/string.S
@@ -107,10 +107,12 @@ memcpy:
        rlwinm. r7,r5,32-3,3,31         /* r7 = r5 >> 3 */
        addi    r6,r3,-4
        addi    r4,r4,-4
-       beq     2f                      /* if less than 8 bytes to do */
+       beq     3f                      /* if less than 8 bytes to do */
        andi.   r0,r6,3                 /* get dest word aligned */
        mtctr   r7
        bne     5f
+       andi.   r0,r4,3                 /* check src word aligned too */
+       bne     3f
 1:     lwz     r7,4(r4)
        lwzu    r8,8(r4)
        stw     r7,4(r6)
@@ -132,6 +134,11 @@ memcpy:
        bdnz    4b
        blr
 5:     subfic  r0,r0,4
+       cmpw    cr1,r0,r5
+       add     r7,r0,r4
+       andi.   r7,r7,3                 /* will source be word-aligned too? */
+       ble     cr1,3b
+       bne     3b                      /* do byte-by-byte if not */
        mtctr   r0
 6:     lbz     r7,4(r4)
        addi    r4,r4,1
@@ -149,10 +156,12 @@ backwards_memcpy:
        rlwinm. r7,r5,32-3,3,31         /* r7 = r5 >> 3 */
        add     r6,r3,r5
        add     r4,r4,r5
-       beq     2f
+       beq     3f
        andi.   r0,r6,3
        mtctr   r7
        bne     5f
+       andi.   r0,r4,3
+       bne     3f
 1:     lwz     r7,-4(r4)
        lwzu    r8,-8(r4)
        stw     r7,-4(r6)
@@ -171,7 +180,12 @@ backwards_memcpy:
        stbu    r0,-1(r6)
        bdnz    4b
        blr
-5:     mtctr   r0
+5:     cmpw    cr1,r0,r5
+       subf    r7,r0,r4
+       andi.   r7,r7,3
+       ble     cr1,3b
+       bne     3b
+       mtctr   r0
 6:     lbzu    r7,-1(r4)
        stbu    r7,-1(r6)
        bdnz    6b