bcopyxx: remove 16-bitisms

Remove a couple of 16-bitisms in the bcopy code, since it is now
running in perfectly ordinary 32-bit mode.  In particular,
prefer 32-bit registers to 16-bit registers, and drop "a32" prefixes
(which do nothing.)

Signed-off-by: H. Peter Anvin <hpa@zytor.com>

diff --git a/core/bcopyxx.inc b/core/bcopyxx.inc
index 22b32b5..89ae4f4 100644 (file)
--- a/core/bcopyxx.inc
+++ b/core/bcopyxx.inc
@@ -57,36 +57,36 @@ pm_bcopy:
 
 .forward:
                ; Initial alignment
-               mov dx,di
-               shr dx,1
+               mov edx,edi
+               shr edx,1
                jnc .faa1
-               a32 movsb
+               movsb
                dec ecx
 .faa1:
                mov al,cl
                cmp ecx,2
                jb .f_tiny
 
-               shr dx,1
+               shr edx,1
                jnc .faa2
-               a32 movsw
+               movsw
                sub ecx,2
 .faa2:
 
                ; Bulk transfer
                mov al,cl               ; Save low bits
                shr ecx,2               ; Convert to dwords
-               a32 rep movsd           ; Do our business
+               rep movsd               ; Do our business
                ; At this point ecx == 0
 
                test al,2
                jz .fab2
-               a32 movsw
+               movsw
 .fab2:
 .f_tiny:
                test al,1
                jz .fab1
-               a32 movsb
+               movsb
 .fab1:
 .done:
                pop eax
@@ -101,10 +101,10 @@ pm_bcopy:
                lea edi,[edi+ecx-1]
 
                ; Initial alignment
-               mov dx,di
-               shr dx,1
+               mov edx,edi
+               shr edx,1
                jc .raa1
-               a32 movsb
+               movsb
                dec ecx
 .raa1:
 
@@ -113,9 +113,9 @@ pm_bcopy:
                mov al,cl
                cmp ecx,2
                jb .r_tiny
-               shr dx,1
+               shr edx,1
                jc .raa2
-               a32 movsw
+               movsw
                sub ecx,2
 .raa2:
 
@@ -124,7 +124,7 @@ pm_bcopy:
                sub edi,2
                mov al,cl               ; Save low bits
                shr ecx,2
-               a32 rep movsd
+               rep movsd
 
                ; Final alignment
 .r_final:
@@ -132,14 +132,14 @@ pm_bcopy:
                add edi,2
                test al,2
                jz .rab2
-               a32 movsw
+               movsw
 .rab2:
 .r_tiny:
                inc esi
                inc edi
                test al,1
                jz .rab1
-               a32 movsb
+               movsb
 .rab1:
                cld
                jmp short .done
@@ -148,35 +148,35 @@ pm_bcopy:
                xor eax,eax
 
                ; Initial alignment
-               mov dx,di
-               shr dx,1
+               mov edx,edi
+               shr edx,1
                jnc .zaa1
-               a32 stosb
+               stosb
                dec ecx
 .zaa1:
 
                mov bl,cl
                cmp ecx,2
                jb .z_tiny
-               shr dx,1
+               shr edx,1
                jnc .zaa2
-               a32 stosw
+               stosw
                sub ecx,2
 .zaa2:
 
                ; Bulk
                mov bl,cl               ; Save low bits
                shr ecx,2
-               a32 rep stosd
+               rep stosd
 
                test bl,2
                jz .zab2
-               a32 stosw
+               stosw
 .zab2:
 .z_tiny:
                test bl,1
                jz .zab1
-               a32 stosb
+               stosb
 .zab1:
                jmp short .done