1 files changed, 27 insertions, 28 deletions

diff --git a/firmware/common/memset_a.S b/firmware/common/memset_a.S
index 96fece4421..6dbdab9595 100644
--- a/firmware/common/memset_a.S
+++ b/firmware/common/memset_a.S
@@ -241,30 +241,29 @@ memset:
 
 #elif defined(CPU_ARM)
 
-/*  The following code is taken from the Linux kernel version 2.6.15.3
+/*  The following code is based on code found in Linux kernel version 2.6.15.3
  *  linux/arch/arm/lib/memset.S
  *
  *  Copyright (C) 1995-2000 Russell King
  */
 
-@       .word   0
-1:      subs    r2, r2, #4              @ 1 do we have enough
+/* This code will align a pointer for memset, if needed */
+1:      cmp     r2, #4                  @ 1 do we have enough
         blt     5f                      @ 1 bytes to align with?
         cmp     r3, #2                  @ 1
-        strltb  r1, [r0], #1            @ 1
-        strleb  r1, [r0], #1            @ 1
-        strb    r1, [r0], #1            @ 1
-        add     r2, r2, r3              @ 1 (r2 = r2 - (4 - r3))
-/*
- * The pointer is now aligned and the length is adjusted.  Try doing the
- * memzero again.
- */
+        strgtb  r1, [r0, #-1]!          @ 1
+        strgeb  r1, [r0, #-1]!          @ 1
+        strb    r1, [r0, #-1]!          @ 1
+        sub     r2, r2, r3              @ 1 r2 = r2 - r3
+        b 2f
 
         .global     memset
         .type       memset,%function
 memset:
+        add     r0, r0, r2              @ we'll write backwards in memory
         ands    r3, r0, #3              @ 1 unaligned?
         bne     1b                      @ 1
+2:
 /*
  * we know that the pointer in r0 is aligned to a word boundary.
  */
@@ -272,7 +271,7 @@ memset:
         orr     r1, r1, r1, lsl #16
         mov     r3, r1
         cmp     r2, #16
-        blt     4f
+        blt     5f
 /*
  * We need an extra register for this loop - save the return address and
  * use the LR
@@ -281,36 +280,36 @@ memset:
         mov     ip, r1
         mov     lr, r1
 
-2:      subs    r2, r2, #64
-        stmgeia r0!, {r1, r3, ip, lr}   @ 64 bytes at a time.
-        stmgeia r0!, {r1, r3, ip, lr}
-        stmgeia r0!, {r1, r3, ip, lr}
-        stmgeia r0!, {r1, r3, ip, lr}
-        bgt     2b
+3:      subs    r2, r2, #64
+        stmgedb r0!, {r1, r3, ip, lr}   @ 64 bytes at a time.
+        stmgedb r0!, {r1, r3, ip, lr}
+        stmgedb r0!, {r1, r3, ip, lr}
+        stmgedb r0!, {r1, r3, ip, lr}
+        bgt     3b
         ldmeqfd sp!, {pc}               @ Now <64 bytes to go.
 /*
  * No need to correct the count; we're only testing bits from now on
  */
         tst     r2, #32
-        stmneia r0!, {r1, r3, ip, lr}
-        stmneia r0!, {r1, r3, ip, lr}
+        stmnedb r0!, {r1, r3, ip, lr}
+        stmnedb r0!, {r1, r3, ip, lr}
         tst     r2, #16
-        stmneia r0!, {r1, r3, ip, lr}
+        stmnedb r0!, {r1, r3, ip, lr}
         ldr     lr, [sp], #4
 
-4:      tst     r2, #8
-        stmneia r0!, {r1, r3}
+5:      tst     r2, #8
+        stmnedb r0!, {r1, r3}
         tst     r2, #4
-        strne   r1, [r0], #4
+        strne   r1, [r0, #-4]!
 /*
  * When we get here, we've got less than 4 bytes to zero.  We
  * may have an unaligned pointer as well.
  */
-5:      tst     r2, #2
-        strneb  r1, [r0], #1
-        strneb  r1, [r0], #1
+6:      tst     r2, #2
+        strneb  r1, [r0, #-1]!
+        strneb  r1, [r0, #-1]!
         tst     r2, #1
-        strneb  r1, [r0], #1
+        strneb  r1, [r0, #-1]!
         mov     pc, lr
 end:
         .size   memset,.end-memset