x86, boot: straighten out ranges to copy/zero in compressed/head*.S

Both on 32 and 64 bits, we copy all the way up to the end of bss,
except that on 64 bits there is a hack to avoid copying on top of the
page tables.  There is no point in copying bss at all, especially
since we are just about to zero it all anyway.

To clean up and unify the handling, we now do:

  - copy from startup_32 to _bss.
  - zero from _bss to _ebss.
  - the _ebss symbol is aligned to an 8-byte boundary.
  - the page tables are moved to a separate section.

Use _bss as the copy endpoint since _edata may be misaligned.

[ Impact: cleanup, trivial performance improvement ]

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

arch/x86/boot/compressed/head_32.S

@@ -93,9 +93,9 @@ ENTRY(startup_32)
  * where decompression in place becomes safe.
  */
 	pushl	%esi
-	leal	_ebss(%ebp), %esi
-	leal	_ebss(%ebx), %edi
-	movl	$(_ebss - startup_32), %ecx
+	leal	_bss(%ebp), %esi
+	leal	_bss(%ebx), %edi
+	movl	$(_bss - startup_32), %ecx
 	std
 	rep	movsb
 	cld
@@ -125,7 +125,7 @@ relocated:
  * Clear BSS
  */
 	xorl	%eax, %eax
-	leal	_edata(%ebx), %edi
+	leal	_bss(%ebx), %edi
 	leal	_ebss(%ebx), %ecx
 	subl	%edi, %ecx
 	cld

arch/x86/boot/compressed/head_64.S

@@ -253,9 +253,9 @@ ENTRY(startup_64)
  * Copy the compressed kernel to the end of our buffer
  * where decompression in place becomes safe.
  */
-	leaq	_end_before_pgt(%rip), %r8
-	leaq	_end_before_pgt(%rbx), %r9
-	movq	$_end_before_pgt /* - $startup_32 */, %rcx
+	leaq	_bss(%rip), %r8
+	leaq	_bss(%rbx), %r9
+	movq	$_bss /* - $startup_32 */, %rcx
 1:	subq	$8, %r8
 	subq	$8, %r9
 	movq	0(%r8), %rax
@@ -276,8 +276,8 @@ relocated:
  * Clear BSS
  */
 	xorq	%rax, %rax
-	leaq    _edata(%rbx), %rdi
-	leaq    _end_before_pgt(%rbx), %rcx
+	leaq    _bss(%rbx), %rdi
+	leaq    _ebss(%rbx), %rcx
 	subq	%rdi, %rcx
 	cld
 	rep	stosb
@@ -329,3 +329,11 @@ boot_heap:
 boot_stack:
 	.fill BOOT_STACK_SIZE, 1, 0
 boot_stack_end:
+
+/*
+ * Space for page tables (not in .bss so not zeroed)
+ */
+	.section ".pgtable","a",@nobits
+	.balign 4096
+pgtable:
+	.fill 6*4096, 1, 0

arch/x86/boot/compressed/vmlinux.lds.S

@@ -2,6 +2,8 @@ OUTPUT_FORMAT(CONFIG_OUTPUT_FORMAT, CONFIG_OUTPUT_FORMAT, CONFIG_OUTPUT_FORMAT)
 
 #undef i386
 
+#include <asm/page_types.h>
+
 #ifdef CONFIG_X86_64
 OUTPUT_ARCH(i386:x86-64)
 ENTRY(startup_64)
@@ -48,13 +50,16 @@ SECTIONS
 		*(.bss)
 		*(.bss.*)
 		*(COMMON)
-#ifdef CONFIG_X86_64
-		. = ALIGN(8);
-		_end_before_pgt = . ;
-		. = ALIGN(4096);
-		pgtable = . ;
-		. = . + 4096 * 6;
-#endif
+		. = ALIGN(8);	/* For convenience during zeroing */
 		_ebss = .;
 	}
+#ifdef CONFIG_X86_64
+       . = ALIGN(PAGE_SIZE);
+       .pgtable : {
+		_pgtable = . ;
+		*(.pgtable)
+		_epgtable = . ;
+	}
+#endif
+	_end = .;
 }