x86, boot: use appropriate rep string for move and clear

In the pre-decompression code, use the appropriate largest possible
rep movs and rep stos to move code and clear bss, respectively.  For
reverse copy, do note that the initial values are supposed to be the
address of the first (highest) copy datum, not one byte beyond the end
of the buffer.

rep strings are not necessarily the fastest way to perform these
operations on all current processors, but are likely to be in the
future, and perhaps more importantly, we want to encourage the
architecturally right thing to do here.

This also fixes a couple of trivial inefficiencies on 64 bits.

[ Impact: trivial performance enhancement, increase code similarity ]
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

arch/x86/boot/compressed/head_32.S

@@ -100,11 +100,12 @@ ENTRY(startup_32)
  * where decompression in place becomes safe.
  */
 	pushl	%esi
-	leal	_bss(%ebp), %esi
-	leal	_bss(%ebx), %edi
+	leal	(_bss-4)(%ebp), %esi
+	leal	(_bss-4)(%ebx), %edi
 	movl	$(_bss - startup_32), %ecx
+	shrl	$2, %ecx
 	std
-	rep	movsb
+	rep	movsl
 	cld
 	popl	%esi
 
@@ -135,8 +136,8 @@ relocated:
 	leal	_bss(%ebx), %edi
 	leal	_ebss(%ebx), %ecx
 	subl	%edi, %ecx
-	cld
-	rep	stosb
+	shrl	$2, %ecx
+	rep	stosl
 
 /*
  * Do the decompression, and jump to the new kernel..

arch/x86/boot/compressed/head_64.S

@@ -260,15 +260,15 @@ ENTRY(startup_64)
  * Copy the compressed kernel to the end of our buffer
  * where decompression in place becomes safe.
  */
-	leaq	_bss(%rip), %r8
-	leaq	_bss(%rbx), %r9
+	pushq	%rsi
+	leaq	(_bss-8)(%rip), %rsi
+	leaq	(_bss-8)(%rbx), %rdi
 	movq	$_bss /* - $startup_32 */, %rcx
-1:	subq	$8, %r8
-	subq	$8, %r9
-	movq	0(%r8), %rax
-	movq	%rax, 0(%r9)
-	subq	$8, %rcx
-	jnz	1b
+	shrq	$3, %rcx
+	std
+	rep	movsq
+	cld
+	popq	%rsi
 
 /*
  * Jump to the relocated address.
@@ -282,12 +282,12 @@ relocated:
 /*
  * Clear BSS (stack is currently empty)
  */
-	xorq	%rax, %rax
-	leaq    _bss(%rbx), %rdi
-	leaq    _ebss(%rbx), %rcx
+	xorl	%eax, %eax
+	leaq    _bss(%rip), %rdi
+	leaq    _ebss(%rip), %rcx
 	subq	%rdi, %rcx
-	cld
-	rep	stosb
+	shrq	$3, %rcx
+	rep	stosq
 
 /*
  * Do the decompression, and jump to the new kernel..