--- zzzz-none-000/linux-2.6.19.2/arch/i386/lib/checksum.S	2007-01-10 19:10:37.000000000 +0000
+++ davinci-8020-5505/linux-2.6.19.2/arch/i386/lib/checksum.S	2007-01-19 14:42:56.000000000 +0000
@@ -26,7 +26,8 @@
  */
 
 #include <asm/errno.h>
-				
+#include <asm/segment.h>
+
 /*
  * computes a partial checksum, e.g. for TCP/UDP fragments
  */
@@ -280,12 +281,23 @@
 
 .align 4
 .globl csum_partial_copy_generic
-				
+.globl csum_partial_copy_generic_to_user
+.globl csum_partial_copy_generic_from_user
+
 #ifndef CONFIG_X86_USE_PPRO_CHECKSUM
 
 #define ARGBASE 16		
 #define FP		12
-		
+
+csum_partial_copy_generic_to_user:
+	pushl $(__USER_DS)
+	popl %es
+	jmp csum_partial_copy_generic
+
+csum_partial_copy_generic_from_user:
+	pushl $(__USER_DS)
+	popl %ds
+
 csum_partial_copy_generic:
 	subl  $4,%esp	
 	pushl %edi
@@ -304,7 +316,7 @@
 	jmp 4f
 SRC(1:	movw (%esi), %bx	)
 	addl $2, %esi
-DST(	movw %bx, (%edi)	)
+DST(	movw %bx, %es:(%edi)	)
 	addl $2, %edi
 	addw %bx, %ax	
 	adcl $0, %eax
@@ -316,30 +328,30 @@
 SRC(1:	movl (%esi), %ebx	)
 SRC(	movl 4(%esi), %edx	)
 	adcl %ebx, %eax
-DST(	movl %ebx, (%edi)	)
+DST(	movl %ebx, %es:(%edi)	)
 	adcl %edx, %eax
-DST(	movl %edx, 4(%edi)	)
+DST(	movl %edx, %es:4(%edi)	)
 
 SRC(	movl 8(%esi), %ebx	)
 SRC(	movl 12(%esi), %edx	)
 	adcl %ebx, %eax
-DST(	movl %ebx, 8(%edi)	)
+DST(	movl %ebx, %es:8(%edi)	)
 	adcl %edx, %eax
-DST(	movl %edx, 12(%edi)	)
+DST(	movl %edx, %es:12(%edi)	)
 
 SRC(	movl 16(%esi), %ebx 	)
 SRC(	movl 20(%esi), %edx	)
 	adcl %ebx, %eax
-DST(	movl %ebx, 16(%edi)	)
+DST(	movl %ebx, %es:16(%edi)	)
 	adcl %edx, %eax
-DST(	movl %edx, 20(%edi)	)
+DST(	movl %edx, %es:20(%edi)	)
 
 SRC(	movl 24(%esi), %ebx	)
 SRC(	movl 28(%esi), %edx	)
 	adcl %ebx, %eax
-DST(	movl %ebx, 24(%edi)	)
+DST(	movl %ebx, %es:24(%edi)	)
 	adcl %edx, %eax
-DST(	movl %edx, 28(%edi)	)
+DST(	movl %edx, %es:28(%edi)	)
 
 	lea 32(%esi), %esi
 	lea 32(%edi), %edi
@@ -353,7 +365,7 @@
 	shrl $2, %edx			# This clears CF
 SRC(3:	movl (%esi), %ebx	)
 	adcl %ebx, %eax
-DST(	movl %ebx, (%edi)	)
+DST(	movl %ebx, %es:(%edi)	)
 	lea 4(%esi), %esi
 	lea 4(%edi), %edi
 	dec %edx
@@ -365,12 +377,12 @@
 	jb 5f
 SRC(	movw (%esi), %cx	)
 	leal 2(%esi), %esi
-DST(	movw %cx, (%edi)	)
+DST(	movw %cx, %es:(%edi)	)
 	leal 2(%edi), %edi
 	je 6f
 	shll $16,%ecx
 SRC(5:	movb (%esi), %cl	)
-DST(	movb %cl, (%edi)	)
+DST(	movb %cl, %es:(%edi)	)
 6:	addl %ecx, %eax
 	adcl $0, %eax
 7:
@@ -381,7 +393,7 @@
 
 6001:
 	movl ARGBASE+20(%esp), %ebx	# src_err_ptr
-	movl $-EFAULT, (%ebx)
+	movl $-EFAULT, %ss:(%ebx)
 
 	# zero the complete destination - computing the rest
 	# is too much work 
@@ -394,11 +406,15 @@
 
 6002:
 	movl ARGBASE+24(%esp), %ebx	# dst_err_ptr
-	movl $-EFAULT,(%ebx)
+	movl $-EFAULT,%ss:(%ebx)
 	jmp 5000b
 
 .previous
 
+	pushl %ss
+	popl %ds
+	pushl %ss
+	popl %es
 	popl %ebx
 	popl %esi
 	popl %edi
@@ -410,17 +426,28 @@
 /* Version for PentiumII/PPro */
 
 #define ROUND1(x) \
+	nop; nop; nop;				\
 	SRC(movl x(%esi), %ebx	)	;	\
 	addl %ebx, %eax			;	\
-	DST(movl %ebx, x(%edi)	)	; 
+	DST(movl %ebx, %es:x(%edi));
 
 #define ROUND(x) \
+	nop; nop; nop;				\
 	SRC(movl x(%esi), %ebx	)	;	\
 	adcl %ebx, %eax			;	\
-	DST(movl %ebx, x(%edi)	)	;
+	DST(movl %ebx, %es:x(%edi));
 
 #define ARGBASE 12
-		
+
+csum_partial_copy_generic_to_user:
+	pushl $(__USER_DS)
+	popl %es
+	jmp csum_partial_copy_generic
+
+csum_partial_copy_generic_from_user:
+	pushl $(__USER_DS)
+	popl %ds
+
 csum_partial_copy_generic:
 	pushl %ebx
 	pushl %edi
@@ -439,7 +466,7 @@
 	subl %ebx, %edi  
 	lea  -1(%esi),%edx
 	andl $-32,%edx
-	lea 3f(%ebx,%ebx), %ebx
+	lea 3f(%ebx,%ebx,2), %ebx
 	testl %esi, %esi 
 	jmp *%ebx
 1:	addl $64,%esi
@@ -460,19 +487,19 @@
 	jb 5f
 SRC(	movw (%esi), %dx         )
 	leal 2(%esi), %esi
-DST(	movw %dx, (%edi)         )
+DST(	movw %dx, %es:(%edi)     )
 	leal 2(%edi), %edi
 	je 6f
 	shll $16,%edx
 5:
 SRC(	movb (%esi), %dl         )
-DST(	movb %dl, (%edi)         )
+DST(	movb %dl, %es:(%edi)     )
 6:	addl %edx, %eax
 	adcl $0, %eax
 7:
 .section .fixup, "ax"
 6001:	movl	ARGBASE+20(%esp), %ebx	# src_err_ptr	
-	movl $-EFAULT, (%ebx)
+	movl $-EFAULT, %ss:(%ebx)
 	# zero the complete destination (computing the rest is too much work)
 	movl ARGBASE+8(%esp),%edi	# dst
 	movl ARGBASE+12(%esp),%ecx	# len
@@ -480,10 +507,14 @@
 	rep; stosb
 	jmp 7b
 6002:	movl ARGBASE+24(%esp), %ebx	# dst_err_ptr
-	movl $-EFAULT, (%ebx)
+	movl $-EFAULT, %ss:(%ebx)
 	jmp  7b			
 .previous				
 
+	pushl %ss
+	popl %ds
+	pushl %ss
+	popl %es
 	popl %esi
 	popl %edi
 	popl %ebx