x86/boot: Remove run-time relocations from .head.text code

The assembly code in head_{32,64}.S, while meant to be
position-independent, generates run-time relocations because it uses
instructions such as:

	leal	gdt(%edx), %eax

which make the assembler and linker think that the code is using %edx as
an index into gdt, and hence gdt needs to be relocated to its run-time
address.

On 32-bit, with lld Dmitry Golovin reports that this results in a
link-time error with default options (i.e. unless -z notext is
explicitly passed):

  LD      arch/x86/boot/compressed/vmlinux
  ld.lld: error: can't create dynamic relocation R_386_32 against local
  symbol in readonly segment; recompile object files with -fPIC or pass
  '-Wl,-z,notext' to allow text relocations in the output

With the BFD linker, this generates a warning during the build, if
--warn-shared-textrel is enabled, which at least Gentoo enables by
default:

  LD      arch/x86/boot/compressed/vmlinux
  ld: arch/x86/boot/compressed/head_32.o: warning: relocation in read-only section `.head.text'
  ld: warning: creating a DT_TEXTREL in object

On 64-bit, it is not possible to link the kernel as -pie with lld, and
it is only possible with a BFD linker that supports -z noreloc-overflow,
i.e. versions >2.26. This is because these instructions cannot really be
relocated: the displacement field is only 32-bits wide, and thus cannot
be relocated for a 64-bit load address. The -z noreloc-overflow option
simply overrides the linker error, and results in R_X86_64_RELATIVE
relocations that apply a 64-bit relocation to a 32-bit field anyway.
This happens to work because nothing will process these run-time
relocations.

Start fixing this by removing relocations from .head.text:

- On 32-bit, use a base register that holds the address of the GOT and
  reference symbol addresses using @Gotoff, i.e.
	leal	gdt@GOTOFF(%edx), %eax

- On 64-bit, most of the code can (and already does) use %rip-relative
  addressing, however the .code32 bits can't, and the 64-bit code also
  needs to reference symbol addresses as they will be after moving the
  compressed kernel to the end of the decompression buffer.
  For these cases, reference the symbols as an offset to startup_32 to
  avoid creating relocations, i.e.:

	leal	(gdt-startup_32)(%bp), %eax

  This only works in .head.text as the subtraction cannot be represented
  as a PC-relative relocation unless startup_32 is in the same section
  as the code. Move efi32_pe_entry into .head.text so that it can use
  the same method to avoid relocations.

Reported-by: Dmitry Golovin <[email protected]>
Signed-off-by: Arvind Sankar <[email protected]>
Signed-off-by: Kees Cook <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Tested-by: Nick Desaulniers <[email protected]>
Tested-by: Sedat Dilek <[email protected]>
Reviewed-by: Kees Cook <[email protected]>
Reviewed-by: Ard Biesheuvel <[email protected]>
Reviewed-by: Fangrui Song <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

arch/x86/boot/compressed/head_32.S

@@ -33,26 +33,10 @@
 #include <asm/bootparam.h>
 
 /*
- * The 32-bit x86 assembler in binutils 2.26 will generate R_386_GOT32X
- * relocation to get the symbol address in PIC.  When the compressed x86
- * kernel isn't built as PIC, the linker optimizes R_386_GOT32X
- * relocations to their fixed symbol addresses.  However, when the
- * compressed x86 kernel is loaded at a different address, it leads
- * to the following load failure:
- *
- *   Failed to allocate space for phdrs
- *
- * during the decompression stage.
- *
- * If the compressed x86 kernel is relocatable at run-time, it should be
- * compiled with -fPIE, instead of -fPIC, if possible and should be built as
- * Position Independent Executable (PIE) so that linker won't optimize
- * R_386_GOT32X relocation to its fixed symbol address.  Older
- * linkers generate R_386_32 relocations against locally defined symbols,
- * _bss, _ebss and _end, in PIE.  It isn't wrong, just less optimal than
- * R_386_RELATIVE.  But the x86 kernel fails to properly handle R_386_32
- * relocations when relocating the kernel.  To generate R_386_RELATIVE
- * relocations, we mark _bss, _ebss and _end as hidden:
+ * These symbols needed to be marked as .hidden to prevent the BFD linker from
+ * generating R_386_32 (rather than R_386_RELATIVE) relocations for them when
+ * the 32-bit compressed kernel is linked as PIE. This is no longer necessary,
+ * but it doesn't hurt to keep them .hidden.
  */
 	.hidden _bss
 	.hidden _ebss
@@ -74,10 +58,10 @@ SYM_FUNC_START(startup_32)
 	leal	(BP_scratch+4)(%esi), %esp
 	call	1f
 1:	popl	%edx
-	subl	$1b, %edx
+	addl	$_GLOBAL_OFFSET_TABLE_+(.-1b), %edx
 
 	/* Load new GDT */
-	leal	gdt(%edx), %eax
+	leal	gdt@GOTOFF(%edx), %eax
 	movl	%eax, 2(%eax)
 	lgdt	(%eax)
 
@@ -90,14 +74,16 @@ SYM_FUNC_START(startup_32)
 	movl	%eax, %ss
 
 /*
- * %edx contains the address we are loaded at by the boot loader and %ebx
- * contains the address where we should move the kernel image temporarily
- * for safe in-place decompression. %ebp contains the address that the kernel
- * will be decompressed to.
+ * %edx contains the address we are loaded at by the boot loader (plus the
+ * offset to the GOT).  The below code calculates %ebx to be the address where
+ * we should move the kernel image temporarily for safe in-place decompression
+ * (again, plus the offset to the GOT).
+ *
+ * %ebp is calculated to be the address that the kernel will be decompressed to.
  */
 
 #ifdef CONFIG_RELOCATABLE
-	movl	%edx, %ebx
+	leal	startup_32@GOTOFF(%edx), %ebx
 
 #ifdef CONFIG_EFI_STUB
 /*
@@ -108,7 +94,7 @@ SYM_FUNC_START(startup_32)
  *	image_offset = startup_32 - image_base
  * Otherwise image_offset will be zero and has no effect on the calculations.
  */
-	subl    image_offset(%edx), %ebx
+	subl    image_offset@GOTOFF(%edx), %ebx
 #endif
 
 	movl	BP_kernel_alignment(%esi), %eax
@@ -125,10 +111,10 @@ SYM_FUNC_START(startup_32)
 	movl	%ebx, %ebp	// Save the output address for later
 	/* Target address to relocate to for decompression */
 	addl    BP_init_size(%esi), %ebx
-	subl    $_end, %ebx
+	subl    $_end@GOTOFF, %ebx
 
 	/* Set up the stack */
-	leal	boot_stack_end(%ebx), %esp
+	leal	boot_stack_end@GOTOFF(%ebx), %esp
 
 	/* Zero EFLAGS */
 	pushl	$0
@@ -139,8 +125,8 @@ SYM_FUNC_START(startup_32)
  * where decompression in place becomes safe.
  */
 	pushl	%esi
-	leal	(_bss-4)(%edx), %esi
-	leal	(_bss-4)(%ebx), %edi
+	leal	(_bss@GOTOFF-4)(%edx), %esi
+	leal	(_bss@GOTOFF-4)(%ebx), %edi
 	movl	$(_bss - startup_32), %ecx
 	shrl	$2, %ecx
 	std
@@ -153,14 +139,14 @@ SYM_FUNC_START(startup_32)
 	 * during extract_kernel below. To avoid any issues, repoint the GDTR
 	 * to the new copy of the GDT.
 	 */
-	leal	gdt(%ebx), %eax
+	leal	gdt@GOTOFF(%ebx), %eax
 	movl	%eax, 2(%eax)
 	lgdt	(%eax)
 
 /*
  * Jump to the relocated address.
  */
-	leal	.Lrelocated(%ebx), %eax
+	leal	.Lrelocated@GOTOFF(%ebx), %eax
 	jmp	*%eax
 SYM_FUNC_END(startup_32)
 
@@ -170,7 +156,7 @@ SYM_FUNC_START_ALIAS(efi_stub_entry)
 	add	$0x4, %esp
 	movl	8(%esp), %esi	/* save boot_params pointer */
 	call	efi_main
-	leal	startup_32(%eax), %eax
+	/* efi_main returns the possibly relocated address of startup_32 */
 	jmp	*%eax
 SYM_FUNC_END(efi32_stub_entry)
 SYM_FUNC_END_ALIAS(efi_stub_entry)
@@ -183,8 +169,8 @@ SYM_FUNC_START_LOCAL_NOALIGN(.Lrelocated)
  * Clear BSS (stack is currently empty)
  */
 	xorl	%eax, %eax
-	leal	_bss(%ebx), %edi
-	leal	_ebss(%ebx), %ecx
+	leal	_bss@GOTOFF(%ebx), %edi
+	leal	_ebss@GOTOFF(%ebx), %ecx
 	subl	%edi, %ecx
 	shrl	$2, %ecx
 	rep	stosl
@@ -198,9 +184,9 @@ SYM_FUNC_START_LOCAL_NOALIGN(.Lrelocated)
 	pushl	%ebp		/* output address */
 
 	pushl	$z_input_len	/* input_len */
-	leal	input_data(%ebx), %eax
+	leal	input_data@GOTOFF(%ebx), %eax
 	pushl	%eax		/* input_data */
-	leal	boot_heap(%ebx), %eax
+	leal	boot_heap@GOTOFF(%ebx), %eax
 	pushl	%eax		/* heap area */
 	pushl	%esi		/* real mode pointer */
 	call	extract_kernel	/* returns kernel location in %eax */