[PATCH v3 01/29] x86: boot/copy, remove unused functions

[PATCH v3 01/29] x86: boot/copy, remove unused functions

[Jiri Slaby]

copy_from_gs and copy_to_gs are unused in the boot code. They have
actually never been used -- they were always commented out since their
addition in 5be865661516 ("String-handling functions for the new x86
setup code.") in 2007. So remove them -- they can be restored from
history if need be.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: x86@kernel.org
---
 arch/x86/boot/copy.S | 20 --------------------
 1 file changed, 20 deletions(-)

diff --git a/arch/x86/boot/copy.S b/arch/x86/boot/copy.S
index 1eb7d298b47d..15d9f74b0008 100644
--- a/arch/x86/boot/copy.S
+++ b/arch/x86/boot/copy.S
@@ -65,23 +65,3 @@ GLOBAL(copy_to_fs)
 	popw	%es
 	retl
 ENDPROC(copy_to_fs)
-
-#if 0 /* Not currently used, but can be enabled as needed */
-GLOBAL(copy_from_gs)
-	pushw	%ds
-	pushw	%gs
-	popw	%ds
-	calll	memcpy
-	popw	%ds
-	retl
-ENDPROC(copy_from_gs)
-
-GLOBAL(copy_to_gs)
-	pushw	%es
-	pushw	%gs
-	popw	%es
-	calll	memcpy
-	popw	%es
-	retl
-ENDPROC(copy_to_gs)
-#endif
-- 
2.12.2

[PATCH v3 02/29] x86_32: boot, extract efi_pe_entry from startup_32

[Jiri Slaby]

efi_pe_entry body is somehow squashed into startup_32. In the old days,
we forced startup_32 to start at offset 0x00 and efi_pe_entry to start
at 0x10. But this requirement was removed in 99f857db8857 ("x86, build:
Dynamically find entry points in compressed startup code") long time
ago.

The way it is now makes the code less readable and illogical. And given
we can now safely extract the inlined efi_pe_entry body from
startup_32 into a separate function, we do so and we separate it to two
functions as they are marked already: efi_pe_entry + efi32_stub_entry.

We also annotate the functions appropriatelly by ENTRY+ENDPROC.

ABI offset is preserved:
  0000   128 FUNC    GLOBAL DEFAULT    6 startup_32
  0080    60 FUNC    GLOBAL DEFAULT    6 efi_pe_entry
  00bc    68 FUNC    GLOBAL DEFAULT    6 efi32_stub_entry

On the top-level, it looked like:
	ENTRY(startup_32)
	#ifdef CONFIG_EFI_STUB		; start of inlined
		jmp     preferred_addr
	ENTRY(efi_pe_entry)
		... ; a lot of assembly (efi_pe_entry)
	ENTRY(efi32_stub_entry)
		... ; a lot of assembly (efi32_stub_entry)
		leal    preferred_addr(%eax), %eax
		jmp     *%eax
	preferred_addr:
	#endif				; end of inlined
		... ; a lot of assembly (startup_32)
	ENDPROC(startup_32)

And it is converted into:
	ENTRY(startup_32)
		... ; a lot of assembly (startup_32)
	ENDPROC(startup_32)

	#ifdef CONFIG_EFI_STUB
	ENTRY(efi_pe_entry)
		... ; a lot of assembly (efi_pe_entry)
	ENDPROC(efi_pe_entry)

	ENTRY(efi32_stub_entry)
		... ; a lot of assembly (efi32_stub_entry)
		leal    startup_32(%eax), %eax
		jmp     *%eax
	ENDPROC(efi32_stub_entry)
	#endif

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: <x86@kernel.org>
Cc: David Woodhouse <David.Woodhouse@intel.com>
Cc: Matt Fleming <matt.fleming@intel.com>
---
 arch/x86/boot/compressed/head_32.S | 129 ++++++++++++++++++-------------------
 1 file changed, 64 insertions(+), 65 deletions(-)

diff --git a/arch/x86/boot/compressed/head_32.S b/arch/x86/boot/compressed/head_32.S
index d85b9625e836..11c68cf53d4e 100644
--- a/arch/x86/boot/compressed/head_32.S
+++ b/arch/x86/boot/compressed/head_32.S
@@ -61,71 +61,6 @@
 
 	__HEAD
 ENTRY(startup_32)
-#ifdef CONFIG_EFI_STUB
-	jmp	preferred_addr
-
-	/*
-	 * We don't need the return address, so set up the stack so
-	 * efi_main() can find its arguments.
-	 */
-ENTRY(efi_pe_entry)
-	add	$0x4, %esp
-
-	call	1f
-1:	popl	%esi
-	subl	$1b, %esi
-
-	popl	%ecx
-	movl	%ecx, efi32_config(%esi)	/* Handle */
-	popl	%ecx
-	movl	%ecx, efi32_config+8(%esi)	/* EFI System table pointer */
-
-	/* Relocate efi_config->call() */
-	leal	efi32_config(%esi), %eax
-	add	%esi, 40(%eax)
-	pushl	%eax
-
-	call	make_boot_params
-	cmpl	$0, %eax
-	je	fail
-	movl	%esi, BP_code32_start(%eax)
-	popl	%ecx
-	pushl	%eax
-	pushl	%ecx
-	jmp	2f		/* Skip efi_config initialization */
-
-ENTRY(efi32_stub_entry)
-	add	$0x4, %esp
-	popl	%ecx
-	popl	%edx
-
-	call	1f
-1:	popl	%esi
-	subl	$1b, %esi
-
-	movl	%ecx, efi32_config(%esi)	/* Handle */
-	movl	%edx, efi32_config+8(%esi)	/* EFI System table pointer */
-
-	/* Relocate efi_config->call() */
-	leal	efi32_config(%esi), %eax
-	add	%esi, 40(%eax)
-	pushl	%eax
-2:
-	call	efi_main
-	cmpl	$0, %eax
-	movl	%eax, %esi
-	jne	2f
-fail:
-	/* EFI init failed, so hang. */
-	hlt
-	jmp	fail
-2:
-	movl	BP_code32_start(%esi), %eax
-	leal	preferred_addr(%eax), %eax
-	jmp	*%eax
-
-preferred_addr:
-#endif
 	cld
 	/*
 	 * Test KEEP_SEGMENTS flag to see if the bootloader is asking
@@ -208,6 +143,70 @@ preferred_addr:
 	jmp	*%eax
 ENDPROC(startup_32)
 
+#ifdef CONFIG_EFI_STUB
+/*
+ * We don't need the return address, so set up the stack so efi_main() can find
+ * its arguments.
+ */
+ENTRY(efi_pe_entry)
+	add	$0x4, %esp
+
+	call	1f
+1:	popl	%esi
+	subl	$1b, %esi
+
+	popl	%ecx
+	movl	%ecx, efi32_config(%esi)	/* Handle */
+	popl	%ecx
+	movl	%ecx, efi32_config+8(%esi)	/* EFI System table pointer */
+
+	/* Relocate efi_config->call() */
+	leal	efi32_config(%esi), %eax
+	add	%esi, 40(%eax)
+	pushl	%eax
+
+	call	make_boot_params
+	cmpl	$0, %eax
+	je	fail
+	movl	%esi, BP_code32_start(%eax)
+	popl	%ecx
+	pushl	%eax
+	pushl	%ecx
+	jmp	2f		/* Skip efi_config initialization */
+ENDPROC(efi_pe_entry)
+
+ENTRY(efi32_stub_entry)
+	add	$0x4, %esp
+	popl	%ecx
+	popl	%edx
+
+	call	1f
+1:	popl	%esi
+	subl	$1b, %esi
+
+	movl	%ecx, efi32_config(%esi)	/* Handle */
+	movl	%edx, efi32_config+8(%esi)	/* EFI System table pointer */
+
+	/* Relocate efi_config->call() */
+	leal	efi32_config(%esi), %eax
+	add	%esi, 40(%eax)
+	pushl	%eax
+2:
+	call	efi_main
+	cmpl	$0, %eax
+	movl	%eax, %esi
+	jne	2f
+fail:
+	/* EFI init failed, so hang. */
+	hlt
+	jmp	fail
+2:
+	movl	BP_code32_start(%esi), %eax
+	leal	startup_32(%eax), %eax
+	jmp	*%eax
+ENDPROC(efi32_stub_entry)
+#endif
+
 	.text
 relocated:
 
-- 
2.12.2

[PATCH v3 03/29] x86_64: boot, extract efi_pe_entry from startup_64

[Jiri Slaby]

efi_pe_entry body is somehow squashed into startup_64. In the old days,
we forced startup_64 to start at offset 0x200 and efi_pe_entry to start
at 0x210. But this requirement was removed in 99f857db8857 ("x86, build:
Dynamically find entry points in compressed startup code") long time
ago.

The way it is now makes the code less readable and illogical. And given
we can now safely extract the inlined efi_pe_entry body from
startup_64 into a separate function, we do so.

We also annotate the function appropriatelly by ENTRY+ENDPROC.

ABI offsets are preserved:
0000000000000000 T startup_32
0000000000000200 T startup_64
0000000000000390 T efi64_stub_entry

On the top-level, it looked like:
	.org 0x200
	ENTRY(startup_64)
	#ifdef CONFIG_EFI_STUB		; start of inlined
		jmp     preferred_addr
	GLOBAL(efi_pe_entry)
		... ; a lot of assembly (efi_pe_entry)
		leaq    preferred_addr(%rax), %rax
		jmp     *%rax
	preferred_addr:
	#endif				; end of inlined
		... ; a lot of assembly (startup_64)
	ENDPROC(startup_64)

And it is converted into:
	.org 0x200
	ENTRY(startup_64)
		... ; a lot of assembly (startup_64)
	ENDPROC(startup_64)

	#ifdef CONFIG_EFI_STUB
	ENTRY(efi_pe_entry)
		... ; a lot of assembly (efi_pe_entry)
		leaq    startup_64(%rax), %rax
		jmp     *%rax
	ENDPROC(efi_pe_entry)
	#endif

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: <x86@kernel.org>
Cc: David Woodhouse <David.Woodhouse@intel.com>
Cc: Matt Fleming <matt.fleming@intel.com>
---
 arch/x86/boot/compressed/head_64.S | 112 ++++++++++++++++++-------------------
 1 file changed, 53 insertions(+), 59 deletions(-)

diff --git a/arch/x86/boot/compressed/head_64.S b/arch/x86/boot/compressed/head_64.S
index d2ae1f821e0c..146751091801 100644
--- a/arch/x86/boot/compressed/head_64.S
+++ b/arch/x86/boot/compressed/head_64.S
@@ -243,65 +243,6 @@ ENTRY(startup_64)
 	 * that maps our entire kernel(text+data+bss+brk), zero page
 	 * and command line.
 	 */
-#ifdef CONFIG_EFI_STUB
-	/*
-	 * The entry point for the PE/COFF executable is efi_pe_entry, so
-	 * only legacy boot loaders will execute this jmp.
-	 */
-	jmp	preferred_addr
-
-ENTRY(efi_pe_entry)
-	movq	%rcx, efi64_config(%rip)	/* Handle */
-	movq	%rdx, efi64_config+8(%rip) /* EFI System table pointer */
-
-	leaq	efi64_config(%rip), %rax
-	movq	%rax, efi_config(%rip)
-
-	call	1f
-1:	popq	%rbp
-	subq	$1b, %rbp
-
-	/*
-	 * Relocate efi_config->call().
-	 */
-	addq	%rbp, efi64_config+40(%rip)
-
-	movq	%rax, %rdi
-	call	make_boot_params
-	cmpq	$0,%rax
-	je	fail
-	mov	%rax, %rsi
-	leaq	startup_32(%rip), %rax
-	movl	%eax, BP_code32_start(%rsi)
-	jmp	2f		/* Skip the relocation */
-
-handover_entry:
-	call	1f
-1:	popq	%rbp
-	subq	$1b, %rbp
-
-	/*
-	 * Relocate efi_config->call().
-	 */
-	movq	efi_config(%rip), %rax
-	addq	%rbp, 40(%rax)
-2:
-	movq	efi_config(%rip), %rdi
-	call	efi_main
-	movq	%rax,%rsi
-	cmpq	$0,%rax
-	jne	2f
-fail:
-	/* EFI init failed, so hang. */
-	hlt
-	jmp	fail
-2:
-	movl	BP_code32_start(%esi), %eax
-	leaq	preferred_addr(%rax), %rax
-	jmp	*%rax
-
-preferred_addr:
-#endif
 
 	/* Setup data segments. */
 	xorl	%eax, %eax
@@ -371,6 +312,59 @@ preferred_addr:
 	jmp	*%rax
 
 #ifdef CONFIG_EFI_STUB
+
+/* The entry point for the PE/COFF executable is efi_pe_entry. */
+ENTRY(efi_pe_entry)
+	movq	%rcx, efi64_config(%rip)	/* Handle */
+	movq	%rdx, efi64_config+8(%rip) /* EFI System table pointer */
+
+	leaq	efi64_config(%rip), %rax
+	movq	%rax, efi_config(%rip)
+
+	call	1f
+1:	popq	%rbp
+	subq	$1b, %rbp
+
+	/*
+	 * Relocate efi_config->call().
+	 */
+	addq	%rbp, efi64_config+40(%rip)
+
+	movq	%rax, %rdi
+	call	make_boot_params
+	cmpq	$0,%rax
+	je	fail
+	mov	%rax, %rsi
+	leaq	startup_32(%rip), %rax
+	movl	%eax, BP_code32_start(%rsi)
+	jmp	2f		/* Skip the relocation */
+
+handover_entry:
+	call	1f
+1:	popq	%rbp
+	subq	$1b, %rbp
+
+	/*
+	 * Relocate efi_config->call().
+	 */
+	movq	efi_config(%rip), %rax
+	addq	%rbp, 40(%rax)
+2:
+	movq	efi_config(%rip), %rdi
+	call	efi_main
+	movq	%rax,%rsi
+	cmpq	$0,%rax
+	jne	2f
+fail:
+	/* EFI init failed, so hang. */
+	hlt
+	jmp	fail
+2:
+	movl	BP_code32_start(%esi), %eax
+	leaq	startup_64(%rax), %rax
+	jmp	*%rax
+ENDPROC(efi_pe_entry)
+
 	.org 0x390
 ENTRY(efi64_stub_entry)
 	movq	%rdi, efi64_config(%rip)	/* Handle */
-- 
2.12.2

[PATCH v3 04/29] x86: assembly, use ENDPROC for functions

[Jiri Slaby]

Somewhere END was used to end a function. It is not intended to be used
for functions, because it does not mark the actual symbols as functions.
Use ENDPROC in such cases which does the right job.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: <x86@kernel.org>
Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: Juergen Gross <jgross@suse.com>
Reviewed-by: Juergen Gross <jgross@suse.com> [xen parts]
Cc: <xen-devel@lists.xenproject.org>
---
 arch/x86/entry/entry_32.S        | 58 ++++++++++++++++++++--------------------
 arch/x86/entry/entry_64.S        | 40 +++++++++++++--------------
 arch/x86/entry/entry_64_compat.S |  4 +--
 arch/x86/kernel/ftrace_32.S      |  8 +++---
 arch/x86/kernel/ftrace_64.S      | 10 +++----
 arch/x86/xen/xen-pvh.S           |  2 +-
 6 files changed, 61 insertions(+), 61 deletions(-)

diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S
index 50bc26949e9e..a546b84aec01 100644
--- a/arch/x86/entry/entry_32.S
+++ b/arch/x86/entry/entry_32.S
@@ -249,7 +249,7 @@ ENTRY(__switch_to_asm)
 	popl	%ebp
 
 	jmp	__switch_to
-END(__switch_to_asm)
+ENDPROC(__switch_to_asm)
 
 /*
  * A newly forked process directly context switches into this address.
@@ -289,7 +289,7 @@ ENTRY(ret_from_fork)
 	 */
 	movl	$0, PT_EAX(%esp)
 	jmp	2b
-END(ret_from_fork)
+ENDPROC(ret_from_fork)
 
 /*
  * Return to user mode is not as complex as all this looks,
@@ -323,7 +323,7 @@ ENTRY(resume_userspace)
 	movl	%esp, %eax
 	call	prepare_exit_to_usermode
 	jmp	restore_all
-END(ret_from_exception)
+ENDPROC(ret_from_exception)
 
 #ifdef CONFIG_PREEMPT
 ENTRY(resume_kernel)
@@ -335,7 +335,7 @@ ENTRY(resume_kernel)
 	jz	restore_all
 	call	preempt_schedule_irq
 	jmp	.Lneed_resched
-END(resume_kernel)
+ENDPROC(resume_kernel)
 #endif
 
 GLOBAL(__begin_SYSENTER_singlestep_region)
@@ -635,7 +635,7 @@ ENTRY(irq_entries_start)
 	jmp	common_interrupt
 	.align	8
     .endr
-END(irq_entries_start)
+ENDPROC(irq_entries_start)
 
 /*
  * the CPU automatically disables interrupts when executing an IRQ vector,
@@ -684,7 +684,7 @@ ENTRY(coprocessor_error)
 	pushl	$0
 	pushl	$do_coprocessor_error
 	jmp	common_exception
-END(coprocessor_error)
+ENDPROC(coprocessor_error)
 
 ENTRY(simd_coprocessor_error)
 	ASM_CLAC
@@ -698,20 +698,20 @@ ENTRY(simd_coprocessor_error)
 	pushl	$do_simd_coprocessor_error
 #endif
 	jmp	common_exception
-END(simd_coprocessor_error)
+ENDPROC(simd_coprocessor_error)
 
 ENTRY(device_not_available)
 	ASM_CLAC
 	pushl	$-1				# mark this as an int
 	pushl	$do_device_not_available
 	jmp	common_exception
-END(device_not_available)
+ENDPROC(device_not_available)
 
 #ifdef CONFIG_PARAVIRT
 ENTRY(native_iret)
 	iret
 	_ASM_EXTABLE(native_iret, iret_exc)
-END(native_iret)
+ENDPROC(native_iret)
 #endif
 
 ENTRY(overflow)
@@ -719,59 +719,59 @@ ENTRY(overflow)
 	pushl	$0
 	pushl	$do_overflow
 	jmp	common_exception
-END(overflow)
+ENDPROC(overflow)
 
 ENTRY(bounds)
 	ASM_CLAC
 	pushl	$0
 	pushl	$do_bounds
 	jmp	common_exception
-END(bounds)
+ENDPROC(bounds)
 
 ENTRY(invalid_op)
 	ASM_CLAC
 	pushl	$0
 	pushl	$do_invalid_op
 	jmp	common_exception
-END(invalid_op)
+ENDPROC(invalid_op)
 
 ENTRY(coprocessor_segment_overrun)
 	ASM_CLAC
 	pushl	$0
 	pushl	$do_coprocessor_segment_overrun
 	jmp	common_exception
-END(coprocessor_segment_overrun)
+ENDPROC(coprocessor_segment_overrun)
 
 ENTRY(invalid_TSS)
 	ASM_CLAC
 	pushl	$do_invalid_TSS
 	jmp	common_exception
-END(invalid_TSS)
+ENDPROC(invalid_TSS)
 
 ENTRY(segment_not_present)
 	ASM_CLAC
 	pushl	$do_segment_not_present
 	jmp	common_exception
-END(segment_not_present)
+ENDPROC(segment_not_present)
 
 ENTRY(stack_segment)
 	ASM_CLAC
 	pushl	$do_stack_segment
 	jmp	common_exception
-END(stack_segment)
+ENDPROC(stack_segment)
 
 ENTRY(alignment_check)
 	ASM_CLAC
 	pushl	$do_alignment_check
 	jmp	common_exception
-END(alignment_check)
+ENDPROC(alignment_check)
 
 ENTRY(divide_error)
 	ASM_CLAC
 	pushl	$0				# no error code
 	pushl	$do_divide_error
 	jmp	common_exception
-END(divide_error)
+ENDPROC(divide_error)
 
 #ifdef CONFIG_X86_MCE
 ENTRY(machine_check)
@@ -779,7 +779,7 @@ ENTRY(machine_check)
 	pushl	$0
 	pushl	machine_check_vector
 	jmp	common_exception
-END(machine_check)
+ENDPROC(machine_check)
 #endif
 
 ENTRY(spurious_interrupt_bug)
@@ -787,7 +787,7 @@ ENTRY(spurious_interrupt_bug)
 	pushl	$0
 	pushl	$do_spurious_interrupt_bug
 	jmp	common_exception
-END(spurious_interrupt_bug)
+ENDPROC(spurious_interrupt_bug)
 
 #ifdef CONFIG_XEN
 ENTRY(xen_hypervisor_callback)
@@ -888,7 +888,7 @@ ENTRY(trace_page_fault)
 	ASM_CLAC
 	pushl	$trace_do_page_fault
 	jmp	common_exception
-END(trace_page_fault)
+ENDPROC(trace_page_fault)
 #endif
 
 ENTRY(page_fault)
@@ -896,7 +896,7 @@ ENTRY(page_fault)
 	pushl	$do_page_fault
 	ALIGN
 	jmp common_exception
-END(page_fault)
+ENDPROC(page_fault)
 
 common_exception:
 	/* the function address is in %gs's slot on the stack */
@@ -928,7 +928,7 @@ common_exception:
 	movl	%esp, %eax			# pt_regs pointer
 	call	*%edi
 	jmp	ret_from_exception
-END(common_exception)
+ENDPROC(common_exception)
 
 ENTRY(debug)
 	/*
@@ -965,7 +965,7 @@ ENTRY(debug)
 	call	do_debug
 	movl	%ebx, %esp
 	jmp	ret_from_exception
-END(debug)
+ENDPROC(debug)
 
 /*
  * NMI is doubly nasty.  It can happen on the first instruction of
@@ -1033,7 +1033,7 @@ ENTRY(nmi)
 	lss	12+4(%esp), %esp		# back to espfix stack
 	jmp	.Lirq_return
 #endif
-END(nmi)
+ENDPROC(nmi)
 
 ENTRY(int3)
 	ASM_CLAC
@@ -1045,19 +1045,19 @@ ENTRY(int3)
 	movl	%esp, %eax			# pt_regs pointer
 	call	do_int3
 	jmp	ret_from_exception
-END(int3)
+ENDPROC(int3)
 
 ENTRY(general_protection)
 	pushl	$do_general_protection
 	jmp	common_exception
-END(general_protection)
+ENDPROC(general_protection)
 
 #ifdef CONFIG_KVM_GUEST
 ENTRY(async_page_fault)
 	ASM_CLAC
 	pushl	$do_async_page_fault
 	jmp	common_exception
-END(async_page_fault)
+ENDPROC(async_page_fault)
 #endif
 
 ENTRY(rewind_stack_do_exit)
@@ -1069,4 +1069,4 @@ ENTRY(rewind_stack_do_exit)
 
 	call	do_exit
 1:	jmp 1b
-END(rewind_stack_do_exit)
+ENDPROC(rewind_stack_do_exit)
diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
index 607d72c4a485..1fe8758102cb 100644
--- a/arch/x86/entry/entry_64.S
+++ b/arch/x86/entry/entry_64.S
@@ -321,7 +321,7 @@ syscall_return_via_sysret:
 opportunistic_sysret_failed:
 	SWAPGS
 	jmp	restore_c_regs_and_iret
-END(entry_SYSCALL_64)
+ENDPROC(entry_SYSCALL_64)
 
 ENTRY(stub_ptregs_64)
 	/*
@@ -347,13 +347,13 @@ ENTRY(stub_ptregs_64)
 
 1:
 	jmp	*%rax				/* Called from C */
-END(stub_ptregs_64)
+ENDPROC(stub_ptregs_64)
 
 .macro ptregs_stub func
 ENTRY(ptregs_\func)
 	leaq	\func(%rip), %rax
 	jmp	stub_ptregs_64
-END(ptregs_\func)
+ENDPROC(ptregs_\func)
 .endm
 
 /* Instantiate ptregs_stub for each ptregs-using syscall */
@@ -396,7 +396,7 @@ ENTRY(__switch_to_asm)
 	popq	%rbp
 
 	jmp	__switch_to
-END(__switch_to_asm)
+ENDPROC(__switch_to_asm)
 
 /*
  * A newly forked process directly context switches into this address.
@@ -432,7 +432,7 @@ ENTRY(ret_from_fork)
 	 */
 	movq	$0, RAX(%rsp)
 	jmp	2b
-END(ret_from_fork)
+ENDPROC(ret_from_fork)
 
 /*
  * Build the entry stubs with some assembler magic.
@@ -447,7 +447,7 @@ ENTRY(irq_entries_start)
 	jmp	common_interrupt
 	.align	8
     .endr
-END(irq_entries_start)
+ENDPROC(irq_entries_start)
 
 /*
  * Interrupt entry/exit.
@@ -649,7 +649,7 @@ native_irq_return_ldt:
 	 */
 	jmp	native_irq_return_iret
 #endif
-END(common_interrupt)
+ENDPROC(common_interrupt)
 
 /*
  * APIC interrupts.
@@ -661,7 +661,7 @@ ENTRY(\sym)
 .Lcommon_\sym:
 	interrupt \do_sym
 	jmp	ret_from_intr
-END(\sym)
+ENDPROC(\sym)
 .endm
 
 #ifdef CONFIG_TRACING
@@ -827,7 +827,7 @@ ENTRY(\sym)
 
 	jmp	error_exit			/* %ebx: no swapgs flag */
 	.endif
-END(\sym)
+ENDPROC(\sym)
 .endm
 
 #ifdef CONFIG_TRACING
@@ -870,7 +870,7 @@ ENTRY(native_load_gs_index)
 	SWAPGS
 	popfq
 	ret
-END(native_load_gs_index)
+ENDPROC(native_load_gs_index)
 EXPORT_SYMBOL(native_load_gs_index)
 
 	_ASM_EXTABLE(.Lgs_change, bad_gs)
@@ -900,7 +900,7 @@ ENTRY(do_softirq_own_stack)
 	leaveq
 	decl	PER_CPU_VAR(irq_count)
 	ret
-END(do_softirq_own_stack)
+ENDPROC(do_softirq_own_stack)
 
 #ifdef CONFIG_XEN
 idtentry xen_hypervisor_callback xen_do_hypervisor_callback has_error_code=0
@@ -936,7 +936,7 @@ ENTRY(xen_do_hypervisor_callback)		/* do_hypervisor_callback(struct *pt_regs) */
 	call	xen_maybe_preempt_hcall
 #endif
 	jmp	error_exit
-END(xen_do_hypervisor_callback)
+ENDPROC(xen_do_hypervisor_callback)
 
 /*
  * Hypervisor uses this for application faults while it executes.
@@ -982,7 +982,7 @@ ENTRY(xen_failsafe_callback)
 	SAVE_EXTRA_REGS
 	ENCODE_FRAME_POINTER
 	jmp	error_exit
-END(xen_failsafe_callback)
+ENDPROC(xen_failsafe_callback)
 
 apicinterrupt3 HYPERVISOR_CALLBACK_VECTOR \
 	xen_hvm_callback_vector xen_evtchn_do_upcall
@@ -1033,7 +1033,7 @@ ENTRY(paranoid_entry)
 	SWAPGS
 	xorl	%ebx, %ebx
 1:	ret
-END(paranoid_entry)
+ENDPROC(paranoid_entry)
 
 /*
  * "Paranoid" exit path from exception stack.  This is invoked
@@ -1062,7 +1062,7 @@ paranoid_exit_restore:
 	RESTORE_C_REGS
 	REMOVE_PT_GPREGS_FROM_STACK 8
 	INTERRUPT_RETURN
-END(paranoid_exit)
+ENDPROC(paranoid_exit)
 
 /*
  * Save all registers in pt_regs, and switch gs if needed.
@@ -1144,7 +1144,7 @@ ENTRY(error_entry)
 	mov	%rax, %rsp
 	decl	%ebx
 	jmp	.Lerror_entry_from_usermode_after_swapgs
-END(error_entry)
+ENDPROC(error_entry)
 
 
 /*
@@ -1158,7 +1158,7 @@ ENTRY(error_exit)
 	testl	%ebx, %ebx
 	jnz	retint_kernel
 	jmp	retint_user
-END(error_exit)
+ENDPROC(error_exit)
 
 /* Runs on exception stack */
 ENTRY(nmi)
@@ -1506,12 +1506,12 @@ nmi_restore:
 	 * mode, so this cannot result in a fault.
 	 */
 	INTERRUPT_RETURN
-END(nmi)
+ENDPROC(nmi)
 
 ENTRY(ignore_sysret)
 	mov	$-ENOSYS, %eax
 	sysret
-END(ignore_sysret)
+ENDPROC(ignore_sysret)
 
 ENTRY(rewind_stack_do_exit)
 	/* Prevent any naive code from trying to unwind to our caller. */
@@ -1522,4 +1522,4 @@ ENTRY(rewind_stack_do_exit)
 
 	call	do_exit
 1:	jmp 1b
-END(rewind_stack_do_exit)
+ENDPROC(rewind_stack_do_exit)
diff --git a/arch/x86/entry/entry_64_compat.S b/arch/x86/entry/entry_64_compat.S
index e1721dafbcb1..966c09ffd62d 100644
--- a/arch/x86/entry/entry_64_compat.S
+++ b/arch/x86/entry/entry_64_compat.S
@@ -262,7 +262,7 @@ sysret32_from_system_call:
 	movq	RSP-ORIG_RAX(%rsp), %rsp
 	swapgs
 	sysretl
-END(entry_SYSCALL_compat)
+ENDPROC(entry_SYSCALL_compat)
 
 /*
  * 32-bit legacy system call entry.
@@ -340,7 +340,7 @@ ENTRY(entry_INT80_compat)
 	TRACE_IRQS_ON
 	SWAPGS
 	jmp	restore_regs_and_iret
-END(entry_INT80_compat)
+ENDPROC(entry_INT80_compat)
 
 	ALIGN
 GLOBAL(stub32_clone)
diff --git a/arch/x86/kernel/ftrace_32.S b/arch/x86/kernel/ftrace_32.S
index 07f40359c9ea..30bc4af8b0de 100644
--- a/arch/x86/kernel/ftrace_32.S
+++ b/arch/x86/kernel/ftrace_32.S
@@ -31,7 +31,7 @@ EXPORT_SYMBOL(mcount)
 
 ENTRY(function_hook)
 	ret
-END(function_hook)
+ENDPROC(function_hook)
 
 ENTRY(ftrace_caller)
 
@@ -98,7 +98,7 @@ ftrace_graph_call:
 /* This is weak to keep gas from relaxing the jumps */
 WEAK(ftrace_stub)
 	ret
-END(ftrace_caller)
+ENDPROC(ftrace_caller)
 
 ENTRY(ftrace_regs_caller)
 	/*
@@ -202,7 +202,7 @@ ftrace_stub:
 	popl	%ecx
 	popl	%eax
 	jmp	ftrace_stub
-END(function_hook)
+ENDPROC(function_hook)
 #endif /* CONFIG_DYNAMIC_FTRACE */
 
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
@@ -225,7 +225,7 @@ ENTRY(ftrace_graph_caller)
 	popl	%ecx
 	popl	%eax
 	ret
-END(ftrace_graph_caller)
+ENDPROC(ftrace_graph_caller)
 
 .globl return_to_handler
 return_to_handler:
diff --git a/arch/x86/kernel/ftrace_64.S b/arch/x86/kernel/ftrace_64.S
index 1dfac634bbf7..823e31577333 100644
--- a/arch/x86/kernel/ftrace_64.S
+++ b/arch/x86/kernel/ftrace_64.S
@@ -147,7 +147,7 @@ EXPORT_SYMBOL(mcount)
 
 ENTRY(function_hook)
 	retq
-END(function_hook)
+ENDPROC(function_hook)
 
 ENTRY(ftrace_caller)
 	/* save_mcount_regs fills in first two parameters */
@@ -183,7 +183,7 @@ GLOBAL(ftrace_graph_call)
 /* This is weak to keep gas from relaxing the jumps */
 WEAK(ftrace_stub)
 	retq
-END(ftrace_caller)
+ENDPROC(ftrace_caller)
 
 ENTRY(ftrace_regs_caller)
 	/* Save the current flags before any operations that can change them */
@@ -254,7 +254,7 @@ GLOBAL(ftrace_regs_caller_end)
 
 	jmp ftrace_epilogue
 
-END(ftrace_regs_caller)
+ENDPROC(ftrace_regs_caller)
 
 
 #else /* ! CONFIG_DYNAMIC_FTRACE */
@@ -290,7 +290,7 @@ trace:
 	restore_mcount_regs
 
 	jmp fgraph_trace
-END(function_hook)
+ENDPROC(function_hook)
 #endif /* CONFIG_DYNAMIC_FTRACE */
 
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
@@ -312,7 +312,7 @@ ENTRY(ftrace_graph_caller)
 	restore_mcount_regs
 
 	retq
-END(ftrace_graph_caller)
+ENDPROC(ftrace_graph_caller)
 
 GLOBAL(return_to_handler)
 	subq  $24, %rsp
diff --git a/arch/x86/xen/xen-pvh.S b/arch/x86/xen/xen-pvh.S
index 5e246716d58f..512fda03c93f 100644
--- a/arch/x86/xen/xen-pvh.S
+++ b/arch/x86/xen/xen-pvh.S
@@ -133,7 +133,7 @@ ENTRY(pvh_start_xen)
 
 	ljmp $__BOOT_CS, $_pa(startup_32)
 #endif
-END(pvh_start_xen)
+ENDPROC(pvh_start_xen)
 
 	.section ".init.data","aw"
 	.balign 8
-- 
2.12.2

Re: [PATCH v3 04/29] x86: assembly, use ENDPROC for functions

[Josh Poimboeuf]

On Fri, Apr 21, 2017 at 04:12:40PM +0200, Jiri Slaby wrote:
> Somewhere END was used to end a function. It is not intended to be used
> for functions, because it does not mark the actual symbols as functions.
> Use ENDPROC in such cases which does the right job.
> 
> Signed-off-by: Jiri Slaby <jslaby@suse.cz>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: "H. Peter Anvin" <hpa@zytor.com>
> Cc: <x86@kernel.org>
> Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
> Cc: Juergen Gross <jgross@suse.com>
> Reviewed-by: Juergen Gross <jgross@suse.com> [xen parts]
> Cc: <xen-devel@lists.xenproject.org>
> ---
>  arch/x86/entry/entry_32.S        | 58 ++++++++++++++++++++--------------------
>  arch/x86/entry/entry_64.S        | 40 +++++++++++++--------------
>  arch/x86/entry/entry_64_compat.S |  4 +--
>  arch/x86/kernel/ftrace_32.S      |  8 +++---
>  arch/x86/kernel/ftrace_64.S      | 10 +++----
>  arch/x86/xen/xen-pvh.S           |  2 +-
>  6 files changed, 61 insertions(+), 61 deletions(-)
> 
> diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S
> index 50bc26949e9e..a546b84aec01 100644
> --- a/arch/x86/entry/entry_32.S
> +++ b/arch/x86/entry/entry_32.S
> @@ -249,7 +249,7 @@ ENTRY(__switch_to_asm)
>  	popl	%ebp
>  
>  	jmp	__switch_to
> -END(__switch_to_asm)
> +ENDPROC(__switch_to_asm)
>  
>  /*
>   * A newly forked process directly context switches into this address.
> @@ -289,7 +289,7 @@ ENTRY(ret_from_fork)
>  	 */
>  	movl	$0, PT_EAX(%esp)
>  	jmp	2b
> -END(ret_from_fork)
> +ENDPROC(ret_from_fork)
>  
>  /*
>   * Return to user mode is not as complex as all this looks,
> @@ -323,7 +323,7 @@ ENTRY(resume_userspace)
>  	movl	%esp, %eax
>  	call	prepare_exit_to_usermode
>  	jmp	restore_all
> -END(ret_from_exception)
> +ENDPROC(ret_from_exception)

What exactly is the motivation of this patch?  It would be good to
describe that in the commit message.

Is the point to allow objtool to generate CFI for it?  If so, I don't
really see how that would work.  Today, objtool considers ENDPROC to
annotate a *callable* function which conforms to the C calling ABI and
can be called by another function.  The stack is in a known state at
function entry, and so the CFI (or frame pointer info) can be reliably
determined.

But entry code is different.  In most cases, the global symbols aren't
actually called, and they don't follow any conventions.  The code is
spaghetti-esque, with HW handlers and jumps everywhere.  The state of
the stack at symbol entry varies per "function".  That's why objtool
ignores these files.

For special cases (like entry code), I was thinking we'd need manual CFI
annotations, like we had before.  Or maybe there's another way, like
some new macros which tell objtool about the HW entry points and the
state of the registers there.

But I'm having trouble seeing how marking these code snippets with
ENTRY/ENDPROC would help objtool make any sense of the code and where
things are on the stack.

-- 
Josh

Re: [PATCH v3 04/29] x86: assembly, use ENDPROC for functions

[Jiri Slaby]

On 04/26/2017, 03:42 AM, Josh Poimboeuf wrote:
>> @@ -323,7 +323,7 @@ ENTRY(resume_userspace)
>>  	movl	%esp, %eax
>>  	call	prepare_exit_to_usermode
>>  	jmp	restore_all
>> -END(ret_from_exception)
>> +ENDPROC(ret_from_exception)
> 
> What exactly is the motivation of this patch?  It would be good to
> describe that in the commit message.
> 
> Is the point to allow objtool to generate CFI for it?  If so, I don't
> really see how that would work.  Today, objtool considers ENDPROC to
> annotate a *callable* function which conforms to the C calling ABI and
> can be called by another function.  The stack is in a known state at
> function entry, and so the CFI (or frame pointer info) can be reliably
> determined.

Ugh, I haven't checked this in 100 % of cases, but this looks pretty
fragile to me. From reading the code, the use of END or ENDPROC is
rather random -- depending on mood and who wrote the code.

> But entry code is different.  In most cases, the global symbols aren't
> actually called, and they don't follow any conventions.  The code is
> spaghetti-esque, with HW handlers and jumps everywhere.  The state of
> the stack at symbol entry varies per "function".  That's why objtool
> ignores these files.

Unfortunately, this is true.

> For special cases (like entry code), I was thinking we'd need manual CFI
> annotations, like we had before.  Or maybe there's another way, like
> some new macros which tell objtool about the HW entry points and the
> state of the registers there.
> 
> But I'm having trouble seeing how marking these code snippets with
> ENTRY/ENDPROC would help objtool make any sense of the code and where
> things are on the stack.

Ok, my intention was to have every line of assembly code in between of
FUNC_START/FUNC_END. That way, every rsp related push/pop/sub/add can be
annotated very easily. For the C-like functions this is all what needs
to be done.

Then there is the spaghetti code. And I was thinking about manual
annotations like:

  # skip the frame pointer checking between START+END here
  OBJTOOL(SKIP_CHECKING)

  # this fn has unusual frame (like interrupts have),
    and you can find return RIP stored at fp + 0x20
  OBJTOOL(RIP_IS_AT, 0x20)

  # put this raw CFI for this location into eh_frame
  OBJTOOL(RAW_CFI, 0x00, 0x00, 0x00)


Similarly, I have OBJTOOL(START_FUNC) and OBJTOOL(END_FUNC) emitted with
each FUNC_START/FUNC_END. So far, when manually expanded for simplicity,
it looks like this:

#define OBJTOOL_START_FUNC                              \
        .pushsection .discard.asmfunctions ASM_NL       \
        .long 0xfd111111 ASM_NL                         \
        .long 1f - . ASM_NL                             \
        .popsection ASM_NL                              \
        1:

#define OBJTOOL_END_FUNC                                \
        .pushsection .discard.asmfunctions ASM_NL       \
        .long 0xfe111111 ASM_NL                         \
        .long 1f - . ASM_NL                             \
        .popsection ASM_NL                              \
        1:

0xfd111111, 0xfe111111 are "opcodes" for objtool meaning
START_FUNC/END_FUNC. Similar would be SKIP_CHECKING, RIP_IS_AT, and
RAW_CFI from the above.

So on the objtool side, it looks like:
        switch (data->magic) {
        case 0xfd111111:
                pc_begin = rela->addend;
                break;
        case 0xfe111111:
                ret = dwarf_annotate_func(dwarf, rela->sym->sec,
                                pc_begin, rela->addend - pc_begin);
                if (ret < 0)
                        return -1;

                break;

So this was my idea -- having all code marked as function and manually
annotate those which are different.

thanks,
-- 
js
suse labs

Re: [PATCH v3 04/29] x86: assembly, use ENDPROC for functions

[Josh Poimboeuf]

On Fri, May 12, 2017 at 09:53:48AM +0200, Jiri Slaby wrote:
> On 04/26/2017, 03:42 AM, Josh Poimboeuf wrote:
> >> @@ -323,7 +323,7 @@ ENTRY(resume_userspace)
> >>  	movl	%esp, %eax
> >>  	call	prepare_exit_to_usermode
> >>  	jmp	restore_all
> >> -END(ret_from_exception)
> >> +ENDPROC(ret_from_exception)
> > 
> > What exactly is the motivation of this patch?  It would be good to
> > describe that in the commit message.
> > 
> > Is the point to allow objtool to generate CFI for it?  If so, I don't
> > really see how that would work.  Today, objtool considers ENDPROC to
> > annotate a *callable* function which conforms to the C calling ABI and
> > can be called by another function.  The stack is in a known state at
> > function entry, and so the CFI (or frame pointer info) can be reliably
> > determined.
> 
> Ugh, I haven't checked this in 100 % of cases, but this looks pretty
> fragile to me. From reading the code, the use of END or ENDPROC is
> rather random -- depending on mood and who wrote the code.

Yes, it would be fragile, but objtool has a fix for that.  It looks at
every instruction in the object file and warns if it finds a return
instruction outside of an ENDPROC function.  That works because all
callable instructions have return instructions (except when they have
sibling calls, but objtool detects those too).  So objtool will flag any
C-type functions that forgot to use ENDPROC.

> > But entry code is different.  In most cases, the global symbols aren't
> > actually called, and they don't follow any conventions.  The code is
> > spaghetti-esque, with HW handlers and jumps everywhere.  The state of
> > the stack at symbol entry varies per "function".  That's why objtool
> > ignores these files.
> 
> Unfortunately, this is true.
> 
> > For special cases (like entry code), I was thinking we'd need manual CFI
> > annotations, like we had before.  Or maybe there's another way, like
> > some new macros which tell objtool about the HW entry points and the
> > state of the registers there.
> > 
> > But I'm having trouble seeing how marking these code snippets with
> > ENTRY/ENDPROC would help objtool make any sense of the code and where
> > things are on the stack.
> 
> Ok, my intention was to have every line of assembly code in between of
> FUNC_START/FUNC_END. That way, every rsp related push/pop/sub/add can be
> annotated very easily. For the C-like functions this is all what needs
> to be done.
> 
> Then there is the spaghetti code. And I was thinking about manual
> annotations like:
> 
>   # skip the frame pointer checking between START+END here
>   OBJTOOL(SKIP_CHECKING)
> 
>   # this fn has unusual frame (like interrupts have),
>     and you can find return RIP stored at fp + 0x20
>   OBJTOOL(RIP_IS_AT, 0x20)
> 
>   # put this raw CFI for this location into eh_frame
>   OBJTOOL(RAW_CFI, 0x00, 0x00, 0x00)
> 
> 
> Similarly, I have OBJTOOL(START_FUNC) and OBJTOOL(END_FUNC) emitted with
> each FUNC_START/FUNC_END. So far, when manually expanded for simplicity,
> it looks like this:

I like the idea of making objtool smart enough to read the entry code,
and of combining automated annotations (where possible) with manual
annotations (where necessary).  And it does make sense for objtool to
automate every rsp-related push/pop/sub/add annotation.  That will make
the entry code quite a bit cleaner since we don't need 'push_cfi' and
friends anymore.

However, I think trying to force the entry code snippets into being
normal functions would be awkward.  For example, C-type functions all
start off with the following initial CFI state:

     LOC           CFA      ra
  0000000000000000 rsp+8    c-8

That means the previous frame's stack pointer was at rsp+8 and the
return instruction pointer is (rsp).  But those assumptions don't hold
for non-C-type functions, which usually start with pt_regs or iret regs
on the stack, or a blank slate.

So the initial CFI state is different between the two types of
"functions".  And there are a lot of other differences.  C-type
functions have to follow frame pointer conventions, for example.  So
your FUNC_START macro (and objtool) would have to somehow figure out a
way to make a distinction between the two.  So it would probably work
out better if we kept the distinction between C-type functions and other
code.

I think ENDPROC (or FUNC_START/FUNC_END) should mean "this function is
100% standardized to the C ABI and its debuginfo can be completely
automated".  And any code outside of that would be "this code is special
and needs a mix of automated and manual debuginfo annotations."

I'm also not sure we need the objtool-specific macros.  It might be
simpler to have macros which just output the cfi instead.  I guess this
goes back to our previous discussions about whether objtool's CFI access
should be read/write or write-only.  I don't remember, did we ever to
come to a conclusion with that?

Either way, from looking at the entry code, we may be able to get away
with only the following .macros:

- DWARF_EMPTY_FRAME signal=0

  Mark all registers as undefined and potentially mark the frame as a
  signal frame.

- DWARF_SET_CFA base=rsp offset=0 c_regs=0 extra_regs=0 iret_regs=0

  Set the CFA value.  Set c_regs, extra_regs, and/or iret_regs to
  indicate which regs (if any) are stored just below the CFA.

- DWARF_SET_INDIRECT_CFA base=rsp offset=0 val_offset=0

  Set CFA = *(base + offset) + val_offset.  I only saw a few places
  where this is needed, where it switches to the irq stack.  We might be
  able to figure out a way to simplify the code in a non-intrusive way
  to get rid of the need for this one.

And we could create higher-level macros from these primitives if needed.

I think we'd only need the macros in relatively few places in the entry
code.  It would be a lot less intrusive than what we had before.

-- 
Josh

Re: [PATCH v3 04/29] x86: assembly, use ENDPROC for functions

[Jiri Slaby]

On 05/13/2017, 12:15 AM, Josh Poimboeuf wrote:
>> Similarly, I have OBJTOOL(START_FUNC) and OBJTOOL(END_FUNC) emitted with
>> each FUNC_START/FUNC_END. So far, when manually expanded for simplicity,
>> it looks like this:
> 
> I like the idea of making objtool smart enough to read the entry code,
> and of combining automated annotations (where possible) with manual
> annotations (where necessary).  And it does make sense for objtool to
> automate every rsp-related push/pop/sub/add annotation.  That will make
> the entry code quite a bit cleaner since we don't need 'push_cfi' and
> friends anymore.
> 
> However, I think trying to force the entry code snippets into being
> normal functions would be awkward.  For example, C-type functions all
> start off with the following initial CFI state:
> 
>      LOC           CFA      ra
>   0000000000000000 rsp+8    c-8
> 
> That means the previous frame's stack pointer was at rsp+8 and the
> return instruction pointer is (rsp).  But those assumptions don't hold
> for non-C-type functions, which usually start with pt_regs or iret regs
> on the stack, or a blank slate.
> 
> So the initial CFI state is different between the two types of
> "functions".  And there are a lot of other differences.  C-type
> functions have to follow frame pointer conventions, for example.  So
> your FUNC_START macro (and objtool) would have to somehow figure out a
> way to make a distinction between the two.  So it would probably work
> out better if we kept the distinction between C-type functions and other
> code.

Ok, that makes a lot of sense.

> I think ENDPROC (or FUNC_START/FUNC_END) should mean "this function is
> 100% standardized to the C ABI and its debuginfo can be completely
> automated".  And any code outside of that would be "this code is special
> and needs a mix of automated and manual debuginfo annotations."

I only hesitate how to call the others. I assume, SYM_FUNC_START and
SYM_FUNC_END were agreed upon for the C-func-like functions.

For the others, what about simply:
  SYM_FUNC_START_SPECIAL/SYM_FUNC_END_SPECIAL
or
  SYM_CODE_START/SYM_CODE_END
or
  SOMETHING_ELSE
?

> I'm also not sure we need the objtool-specific macros.  It might be
> simpler to have macros which just output the cfi instead.  I guess this
> goes back to our previous discussions about whether objtool's CFI access
> should be read/write or write-only.  I don't remember, did we ever to
> come to a conclusion with that?

Correct, exactly to avoid r-w on dwarfinfo in objtool, I introduced the
special objtool macros. They would just put the same cfis into the
.discard section for objtool to combine them with the automatic injected
annotations and put them to the correct place. For -- almost -- free.

Our last discussion on this topic ended up with w-only for objtool at
the moment. I originally wanted r-w to support inline assembly in C, but
you suggested r-only is quite easier, therefore we should start with it.
So the r-w extension is doable, but the question is whether we want the
complexity now.

> Either way, from looking at the entry code, we may be able to get away
> with only the following .macros:
> 
> - DWARF_EMPTY_FRAME signal=0
> 
>   Mark all registers as undefined and potentially mark the frame as a
>   signal frame.
> 
> - DWARF_SET_CFA base=rsp offset=0 c_regs=0 extra_regs=0 iret_regs=0
> 
>   Set the CFA value.  Set c_regs, extra_regs, and/or iret_regs to
>   indicate which regs (if any) are stored just below the CFA.
> 
> - DWARF_SET_INDIRECT_CFA base=rsp offset=0 val_offset=0
> 
>   Set CFA = *(base + offset) + val_offset.  I only saw a few places
>   where this is needed, where it switches to the irq stack.  We might be
>   able to figure out a way to simplify the code in a non-intrusive way
>   to get rid of the need for this one.

Correct, it corresponds with what I had locally to make DWARF unwinder
working through interrupts, in terms of CFI's:
--- a/arch/x86/entry/entry_64.S
+++ b/arch/x86/entry/entry_64.S
@@ -463,6 +463,7 @@ SYM_FUNC_END(irq_entries_start)
        ALLOC_PT_GPREGS_ON_STACK
        SAVE_C_REGS
        SAVE_EXTRA_REGS
+       DW_CFI(.cfi_rel_offset rbp, RBP+8)
        ENCODE_FRAME_POINTER

        testb   $3, CS(%rsp)
@@ -497,7 +498,17 @@ SYM_FUNC_END(irq_entries_start)
        movq    %rsp, %rdi
        incl    PER_CPU_VAR(irq_count)
        cmovzq  PER_CPU_VAR(irq_stack_ptr), %rsp
+       DW_CFI(.cfi_def_cfa_register rdi)
+
        pushq   %rdi
+       DW_CFI(.cfi_escape 0x0f /* DW_CFA_def_cfa_expression */, 6 /*
block len */, \
+               0x77 /* DW_OP_breg7 (rsp) */, 0 /* offset */, \
+               0x06 /* DW_OP_deref */, \
+               0x08 /* DW_OP_const1u */, SIZEOF_PTREGS, \
+               0x22 /* DW_OP_plus */)
+       DW_CFI(.cfi_offset rsp, -2*8)
+       DW_CFI(.cfi_offset rip, -5*8)
+
        /* We entered an interrupt context - irqs are off: */
        TRACE_IRQS_OFF

@@ -654,9 +665,15 @@ SYM_FUNC_END(common_interrupt)
  * APIC interrupts.
  */
 .macro apicinterrupt3 num sym do_sym
-SYM_FUNC_START(\sym)
+SYM_FUNC_START_ALIAS(\sym)
+       DW_CFI(.cfi_startproc simple)
+       DW_CFI(.cfi_signal_frame)
+       DW_CFI(.cfi_def_cfa rsp, 6*8)
+       DW_CFI(.cfi_rel_offset rsp, 4*8)
+       DW_CFI(.cfi_rel_offset rip, 1*8)
        ASM_CLAC
        pushq   $~(\num)
+       DW_CFI(.cfi_adjust_cfa_offset 8)


(DW_CFI is my local-only macro to kill the annotations by a single
switch whenever I want.)

> And we could create higher-level macros from these primitives if needed.
> 
> I think we'd only need the macros in relatively few places in the entry
> code.  It would be a lot less intrusive than what we had before.

Sure, that's the whole point of this exercise :).

thanks,
-- 
js
suse labs

Re: [PATCH v3 04/29] x86: assembly, use ENDPROC for functions

[Jiri Slaby]

On 05/17/2017, 03:23 PM, Jiri Slaby wrote:
>> So the initial CFI state is different between the two types of
>> "functions".  And there are a lot of other differences.  C-type
>> functions have to follow frame pointer conventions, for example.  So
>> your FUNC_START macro (and objtool) would have to somehow figure out a
>> way to make a distinction between the two.  So it would probably work
>> out better if we kept the distinction between C-type functions and other
>> code.
> 
> Ok, that makes a lot of sense.

A quick question:
Do you consider these to be C-type functions?

  ENTRY(function_hook)
        ret
  END(function_hook)

or this?

  ENTRY(native_load_gs_index)
        pushfq
        DISABLE_INTERRUPTS(CLBR_ANY & ~CLBR_RDI)
        SWAPGS
        movl    %edi, %gs
        SWAPGS
        popfq
        ret
  END(native_load_gs_index)

Both are called from C, but they do not setup frame pointer etc.

thanks,
-- 
js
suse labs

Re: [PATCH v3 04/29] x86: assembly, use ENDPROC for functions

[Josh Poimboeuf]

On Fri, May 19, 2017 at 11:17:24AM +0200, Jiri Slaby wrote:
> On 05/17/2017, 03:23 PM, Jiri Slaby wrote:
> >> So the initial CFI state is different between the two types of
> >> "functions".  And there are a lot of other differences.  C-type
> >> functions have to follow frame pointer conventions, for example.  So
> >> your FUNC_START macro (and objtool) would have to somehow figure out a
> >> way to make a distinction between the two.  So it would probably work
> >> out better if we kept the distinction between C-type functions and other
> >> code.
> > 
> > Ok, that makes a lot of sense.
> 
> A quick question:
> Do you consider these to be C-type functions?
> 
>   ENTRY(function_hook)
>         ret
>   END(function_hook)
> 
> or this?
> 
>   ENTRY(native_load_gs_index)
>         pushfq
>         DISABLE_INTERRUPTS(CLBR_ANY & ~CLBR_RDI)
>         SWAPGS
>         movl    %edi, %gs
>         SWAPGS
>         popfq
>         ret
>   END(native_load_gs_index)
> 
> Both are called from C, but they do not setup frame pointer etc.

Yeah, those are valid C-type functions.  Setting up the frame pointer is
optional for leaf functions (i.e.  functions which don't call other
functions).

-- 
Josh

[PATCH v3 05/29] x86: assembly, add ENDPROC to functions

[Jiri Slaby]

Some functions are annotated as ENTRY or GLOBAL, but their ends are not
annotated at all. This means:
* the annotations are not paired and we cannot deal with such functions
  e.g. in objtool
* the symbols are not marked as functions in the object file
* there are no sizes of the functions in the object file

So fix this by adding ENDPROC to each such case.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: <x86@kernel.org>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Cc: Pavel Machek <pavel@ucw.cz>
Cc: <linux-pm@vger.kernel.org>
Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: Juergen Gross <jgross@suse.com>
Reviewed-by: Juergen Gross <jgross@suse.com> [xen parts]
Cc: <xen-devel@lists.xenproject.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: lguest@lists.ozlabs.org
---
 arch/x86/boot/compressed/head_64.S   | 1 +
 arch/x86/entry/entry_32.S            | 1 +
 arch/x86/entry/entry_64_compat.S     | 1 +
 arch/x86/kernel/acpi/wakeup_32.S     | 2 ++
 arch/x86/kernel/ftrace_32.S          | 1 +
 arch/x86/kernel/ftrace_64.S          | 1 +
 arch/x86/kernel/head_32.S            | 1 +
 arch/x86/lguest/head_32.S            | 4 ++++
 arch/x86/math-emu/div_Xsig.S         | 1 +
 arch/x86/math-emu/div_small.S        | 2 +-
 arch/x86/math-emu/mul_Xsig.S         | 4 +++-
 arch/x86/math-emu/polynom_Xsig.S     | 1 +
 arch/x86/math-emu/reg_norm.S         | 2 ++
 arch/x86/math-emu/reg_round.S        | 2 ++
 arch/x86/math-emu/reg_u_add.S        | 1 +
 arch/x86/math-emu/reg_u_div.S        | 2 ++
 arch/x86/math-emu/reg_u_mul.S        | 1 +
 arch/x86/math-emu/reg_u_sub.S        | 1 +
 arch/x86/math-emu/round_Xsig.S       | 4 ++--
 arch/x86/math-emu/shr_Xsig.S         | 1 +
 arch/x86/math-emu/wm_shrx.S          | 2 ++
 arch/x86/math-emu/wm_sqrt.S          | 1 +
 arch/x86/platform/olpc/xo1-wakeup.S  | 1 +
 arch/x86/power/hibernate_asm_32.S    | 3 +++
 arch/x86/power/hibernate_asm_64.S    | 2 ++
 arch/x86/realmode/rm/reboot.S        | 1 +
 arch/x86/realmode/rm/trampoline_32.S | 2 ++
 arch/x86/realmode/rm/trampoline_64.S | 4 ++++
 arch/x86/realmode/rm/wakeup_asm.S    | 1 +
 arch/x86/xen/xen-asm_32.S            | 3 ++-
 arch/x86/xen/xen-asm_64.S            | 2 ++
 arch/x86/xen/xen-head.S              | 3 ++-
 drivers/lguest/x86/switcher_32.S     | 1 +
 33 files changed, 54 insertions(+), 6 deletions(-)

diff --git a/arch/x86/boot/compressed/head_64.S b/arch/x86/boot/compressed/head_64.S
index 146751091801..3e5752efccff 100644
--- a/arch/x86/boot/compressed/head_64.S
+++ b/arch/x86/boot/compressed/head_64.S
@@ -310,6 +310,7 @@ ENTRY(startup_64)
  */
 	leaq	relocated(%rbx), %rax
 	jmp	*%rax
+ENDPROC(startup_64)
 
 #ifdef CONFIG_EFI_STUB
 
diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S
index a546b84aec01..afeeb389e9aa 100644
--- a/arch/x86/entry/entry_32.S
+++ b/arch/x86/entry/entry_32.S
@@ -356,6 +356,7 @@ GLOBAL(__begin_SYSENTER_singlestep_region)
 ENTRY(xen_sysenter_target)
 	addl	$5*4, %esp			/* remove xen-provided frame */
 	jmp	.Lsysenter_past_esp
+ENDPROC(xen_sysenter_target)
 #endif
 
 /*
diff --git a/arch/x86/entry/entry_64_compat.S b/arch/x86/entry/entry_64_compat.S
index 966c09ffd62d..14dc2f259e2f 100644
--- a/arch/x86/entry/entry_64_compat.S
+++ b/arch/x86/entry/entry_64_compat.S
@@ -353,3 +353,4 @@ GLOBAL(stub32_clone)
 	 */
 	xchg	%r8, %rcx
 	jmp	sys_clone
+ENDPROC(stub32_clone)
diff --git a/arch/x86/kernel/acpi/wakeup_32.S b/arch/x86/kernel/acpi/wakeup_32.S
index 0c26b1b44e51..ea9a29e6a96c 100644
--- a/arch/x86/kernel/acpi/wakeup_32.S
+++ b/arch/x86/kernel/acpi/wakeup_32.S
@@ -38,6 +38,7 @@ wakeup_pmode_return:
 	# jump to place where we left off
 	movl	saved_eip, %eax
 	jmp	*%eax
+ENDPROC(wakeup_pmode_return)
 
 bogus_magic:
 	jmp	bogus_magic
@@ -86,6 +87,7 @@ ret_point:
 	call	restore_registers
 	call	restore_processor_state
 	ret
+ENDPROC(do_suspend_lowlevel)
 
 .data
 ALIGN
diff --git a/arch/x86/kernel/ftrace_32.S b/arch/x86/kernel/ftrace_32.S
index 30bc4af8b0de..89f8324e9a68 100644
--- a/arch/x86/kernel/ftrace_32.S
+++ b/arch/x86/kernel/ftrace_32.S
@@ -168,6 +168,7 @@ GLOBAL(ftrace_regs_call)
 	lea	3*4(%esp), %esp			/* Skip orig_ax, ip and cs */
 
 	jmp	.Lftrace_ret
+ENDPROC(ftrace_regs_caller)
 #else /* ! CONFIG_DYNAMIC_FTRACE */
 
 ENTRY(function_hook)
diff --git a/arch/x86/kernel/ftrace_64.S b/arch/x86/kernel/ftrace_64.S
index 823e31577333..a915729c0246 100644
--- a/arch/x86/kernel/ftrace_64.S
+++ b/arch/x86/kernel/ftrace_64.S
@@ -329,4 +329,5 @@ GLOBAL(return_to_handler)
 	movq (%rsp), %rax
 	addq $24, %rsp
 	jmp *%rdi
+ENDPROC(return_to_handler)
 #endif
diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S
index 1f85ee8f9439..e2acba5b0f2f 100644
--- a/arch/x86/kernel/head_32.S
+++ b/arch/x86/kernel/head_32.S
@@ -173,6 +173,7 @@ num_subarch_entries = (. - subarch_entries) / 4
 #else
 	jmp .Ldefault_entry
 #endif /* CONFIG_PARAVIRT */
+ENDPROC(startup_32)
 
 #ifdef CONFIG_HOTPLUG_CPU
 /*
diff --git a/arch/x86/lguest/head_32.S b/arch/x86/lguest/head_32.S
index d5ae63f5ec5d..b11a989de062 100644
--- a/arch/x86/lguest/head_32.S
+++ b/arch/x86/lguest/head_32.S
@@ -48,6 +48,7 @@ ENTRY(lguest_entry)
 
 	/* Jumps are relative: we're running __PAGE_OFFSET too low. */
 	jmp lguest_init+__PAGE_OFFSET
+ENDPROC(lguest_entry)
 
 /*G:055
  * We create a macro which puts the assembler code between lgstart_ and lgend_
@@ -110,6 +111,7 @@ send_interrupts:
 	/* Put eax back the way we found it. */
 	popl %eax
 	ret
+ENDPROC(lg_irq_enable)
 
 /*
  * Finally, the "popf" or "restore flags" routine.  The %eax register holds the
@@ -131,6 +133,7 @@ ENTRY(lg_restore_fl)
 	jnz send_interrupts
 	/* Again, the normal path has used no extra registers.  Clever, huh? */
 	ret
+ENDPROC(lg_restore_fl)
 /*:*/
 
 /* These demark the EIP where host should never deliver interrupts. */
@@ -190,3 +193,4 @@ ENTRY(lguest_iret)
 	popl	%ss:lguest_data+LGUEST_DATA_irq_enabled
 lguest_noirq_iret:
 	iret
+ENDPROC(lguest_iret)
diff --git a/arch/x86/math-emu/div_Xsig.S b/arch/x86/math-emu/div_Xsig.S
index f77ba3058b31..066996dba6a2 100644
--- a/arch/x86/math-emu/div_Xsig.S
+++ b/arch/x86/math-emu/div_Xsig.S
@@ -363,3 +363,4 @@ L_bugged_2:
 	pop	%ebx
 	jmp	L_exit
 #endif /* PARANOID */ 
+ENDPROC(div_Xsig)
diff --git a/arch/x86/math-emu/div_small.S b/arch/x86/math-emu/div_small.S
index 47099628fa4c..2c71527bd917 100644
--- a/arch/x86/math-emu/div_small.S
+++ b/arch/x86/math-emu/div_small.S
@@ -44,4 +44,4 @@ ENTRY(FPU_div_small)
 
 	leave
 	ret
-
+ENDPROC(FPU_div_small)
diff --git a/arch/x86/math-emu/mul_Xsig.S b/arch/x86/math-emu/mul_Xsig.S
index 717785a53eb4..22e0631bb85a 100644
--- a/arch/x86/math-emu/mul_Xsig.S
+++ b/arch/x86/math-emu/mul_Xsig.S
@@ -62,6 +62,7 @@ ENTRY(mul32_Xsig)
 	popl %esi
 	leave
 	ret
+ENDPROC(mul32_Xsig)
 
 
 ENTRY(mul64_Xsig)
@@ -114,6 +115,7 @@ ENTRY(mul64_Xsig)
 	popl %esi
 	leave
 	ret
+ENDPROC(mul64_Xsig)
 
 
 
@@ -173,4 +175,4 @@ ENTRY(mul_Xsig_Xsig)
 	popl %esi
 	leave
 	ret
-
+ENDPROC(mul_Xsig_Xsig)
diff --git a/arch/x86/math-emu/polynom_Xsig.S b/arch/x86/math-emu/polynom_Xsig.S
index 17315c89ff3d..a9aaf414135d 100644
--- a/arch/x86/math-emu/polynom_Xsig.S
+++ b/arch/x86/math-emu/polynom_Xsig.S
@@ -133,3 +133,4 @@ L_accum_done:
 	popl	%esi
 	leave
 	ret
+ENDPROC(polynomial_Xsig)
diff --git a/arch/x86/math-emu/reg_norm.S b/arch/x86/math-emu/reg_norm.S
index 8b6352efceef..53ac1a343c69 100644
--- a/arch/x86/math-emu/reg_norm.S
+++ b/arch/x86/math-emu/reg_norm.S
@@ -94,6 +94,7 @@ L_overflow:
 	call	arith_overflow
 	pop	%ebx
 	jmp	L_exit
+ENDPROC(FPU_normalize)
 
 
 
@@ -145,3 +146,4 @@ L_exit_nuo_zero:
 	popl	%ebx
 	leave
 	ret
+ENDPROC(FPU_normalize_nuo)
diff --git a/arch/x86/math-emu/reg_round.S b/arch/x86/math-emu/reg_round.S
index d1d4e48b4f67..41af5b208d88 100644
--- a/arch/x86/math-emu/reg_round.S
+++ b/arch/x86/math-emu/reg_round.S
@@ -706,3 +706,5 @@ L_exception_exit:
 	mov	$-1,%eax
 	jmp	fpu_reg_round_special_exit
 #endif /* PARANOID */ 
+
+ENDPROC(FPU_round)
diff --git a/arch/x86/math-emu/reg_u_add.S b/arch/x86/math-emu/reg_u_add.S
index 47c4c2434d85..3b1bc5e9b2f6 100644
--- a/arch/x86/math-emu/reg_u_add.S
+++ b/arch/x86/math-emu/reg_u_add.S
@@ -165,3 +165,4 @@ L_exit:
 	leave
 	ret
 #endif /* PARANOID */
+ENDPROC(FPU_u_add)
diff --git a/arch/x86/math-emu/reg_u_div.S b/arch/x86/math-emu/reg_u_div.S
index cc00654b6f9a..796eb5ab921b 100644
--- a/arch/x86/math-emu/reg_u_div.S
+++ b/arch/x86/math-emu/reg_u_div.S
@@ -469,3 +469,5 @@ L_exit:
 	leave
 	ret
 #endif /* PARANOID */ 
+
+ENDPROC(FPU_u_div)
diff --git a/arch/x86/math-emu/reg_u_mul.S b/arch/x86/math-emu/reg_u_mul.S
index 973f12af97df..6196f68cf3c1 100644
--- a/arch/x86/math-emu/reg_u_mul.S
+++ b/arch/x86/math-emu/reg_u_mul.S
@@ -146,3 +146,4 @@ L_exit:
 	ret
 #endif /* PARANOID */ 
 
+ENDPROC(FPU_u_mul)
diff --git a/arch/x86/math-emu/reg_u_sub.S b/arch/x86/math-emu/reg_u_sub.S
index 1b6c24801d22..d115b900919a 100644
--- a/arch/x86/math-emu/reg_u_sub.S
+++ b/arch/x86/math-emu/reg_u_sub.S
@@ -270,3 +270,4 @@ L_exit:
 	popl	%esi
 	leave
 	ret
+ENDPROC(FPU_u_sub)
diff --git a/arch/x86/math-emu/round_Xsig.S b/arch/x86/math-emu/round_Xsig.S
index bbe0e87718e4..87c99749a495 100644
--- a/arch/x86/math-emu/round_Xsig.S
+++ b/arch/x86/math-emu/round_Xsig.S
@@ -78,7 +78,7 @@ L_exit:
 	popl	%ebx
 	leave
 	ret
-
+ENDPROC(round_Xsig)
 
 
 
@@ -138,4 +138,4 @@ L_n_exit:
 	popl	%ebx
 	leave
 	ret
-
+ENDPROC(norm_Xsig)
diff --git a/arch/x86/math-emu/shr_Xsig.S b/arch/x86/math-emu/shr_Xsig.S
index 31cdd118e918..c8552edeec75 100644
--- a/arch/x86/math-emu/shr_Xsig.S
+++ b/arch/x86/math-emu/shr_Xsig.S
@@ -85,3 +85,4 @@ L_more_than_95:
 	popl	%esi
 	leave
 	ret
+ENDPROC(shr_Xsig)
diff --git a/arch/x86/math-emu/wm_shrx.S b/arch/x86/math-emu/wm_shrx.S
index 518428317985..340dd6897f85 100644
--- a/arch/x86/math-emu/wm_shrx.S
+++ b/arch/x86/math-emu/wm_shrx.S
@@ -92,6 +92,7 @@ L_more_than_95:
 	popl	%esi
 	leave
 	ret
+ENDPROC(FPU_shrx)
 
 
 /*---------------------------------------------------------------------------+
@@ -202,3 +203,4 @@ Ls_more_than_95:
 	popl	%esi
 	leave
 	ret
+ENDPROC(FPU_shrxs)
diff --git a/arch/x86/math-emu/wm_sqrt.S b/arch/x86/math-emu/wm_sqrt.S
index d258f59564e1..695afae38fdf 100644
--- a/arch/x86/math-emu/wm_sqrt.S
+++ b/arch/x86/math-emu/wm_sqrt.S
@@ -468,3 +468,4 @@ sqrt_more_prec_large:
 /* Our estimate is too large */
 	movl	$0x7fffff00,%eax
 	jmp	sqrt_round_result
+ENDPROC(wm_sqrt)
diff --git a/arch/x86/platform/olpc/xo1-wakeup.S b/arch/x86/platform/olpc/xo1-wakeup.S
index 948deb289753..2929091cf7fd 100644
--- a/arch/x86/platform/olpc/xo1-wakeup.S
+++ b/arch/x86/platform/olpc/xo1-wakeup.S
@@ -109,6 +109,7 @@ ret_point:
 	call	restore_registers
 	call	restore_processor_state
 	ret
+ENDPROC(do_olpc_suspend_lowlevel)
 
 .data
 saved_gdt:             .long   0,0
diff --git a/arch/x86/power/hibernate_asm_32.S b/arch/x86/power/hibernate_asm_32.S
index 1d0fa0e24070..9480f4b2df94 100644
--- a/arch/x86/power/hibernate_asm_32.S
+++ b/arch/x86/power/hibernate_asm_32.S
@@ -25,6 +25,7 @@ ENTRY(swsusp_arch_suspend)
 
 	call swsusp_save
 	ret
+ENDPROC(swsusp_arch_suspend)
 
 ENTRY(restore_image)
 	movl	mmu_cr4_features, %ecx
@@ -82,3 +83,5 @@ done:
 	xorl	%eax, %eax
 
 	ret
+ENDPROC(restore_image)
+
diff --git a/arch/x86/power/hibernate_asm_64.S b/arch/x86/power/hibernate_asm_64.S
index ce8da3a0412c..ec6b26fd3a7e 100644
--- a/arch/x86/power/hibernate_asm_64.S
+++ b/arch/x86/power/hibernate_asm_64.S
@@ -68,6 +68,7 @@ ENTRY(restore_image)
 	/* jump to relocated restore code */
 	movq	relocated_restore_code(%rip), %rcx
 	jmpq	*%rcx
+ENDPROC(restore_image)
 
 	/* code below has been relocated to a safe page */
 ENTRY(core_restore_code)
@@ -98,6 +99,7 @@ ENTRY(core_restore_code)
 .Ldone:
 	/* jump to the restore_registers address from the image header */
 	jmpq	*%r8
+ENDPROC(core_restore_code)
 
 	 /* code below belongs to the image kernel */
 	.align PAGE_SIZE
diff --git a/arch/x86/realmode/rm/reboot.S b/arch/x86/realmode/rm/reboot.S
index d66c607bdc58..c8855d50f9c1 100644
--- a/arch/x86/realmode/rm/reboot.S
+++ b/arch/x86/realmode/rm/reboot.S
@@ -62,6 +62,7 @@ GLOBAL(machine_real_restart_paging_off)
 	movl	%ecx, %gs
 	movl	%ecx, %ss
 	ljmpw	$8, $1f
+ENDPROC(machine_real_restart_asm)
 
 /*
  * This is 16-bit protected mode code to disable paging and the cache,
diff --git a/arch/x86/realmode/rm/trampoline_32.S b/arch/x86/realmode/rm/trampoline_32.S
index 48ddd76bc4c3..aa14fea65aac 100644
--- a/arch/x86/realmode/rm/trampoline_32.S
+++ b/arch/x86/realmode/rm/trampoline_32.S
@@ -56,11 +56,13 @@ ENTRY(trampoline_start)
 	lmsw	%dx			# into protected mode
 
 	ljmpl	$__BOOT_CS, $pa_startup_32
+ENDPROC(trampoline_start)
 
 	.section ".text32","ax"
 	.code32
 ENTRY(startup_32)			# note: also used from wakeup_asm.S
 	jmp	*%eax
+ENDPROC(startup_32)
 
 	.bss
 	.balign 8
diff --git a/arch/x86/realmode/rm/trampoline_64.S b/arch/x86/realmode/rm/trampoline_64.S
index dac7b20d2f9d..fe21a26a09fe 100644
--- a/arch/x86/realmode/rm/trampoline_64.S
+++ b/arch/x86/realmode/rm/trampoline_64.S
@@ -79,6 +79,8 @@ ENTRY(trampoline_start)
 no_longmode:
 	hlt
 	jmp no_longmode
+ENDPROC(trampoline_start)
+
 #include "../kernel/verify_cpu.S"
 
 	.section ".text32","ax"
@@ -116,6 +118,7 @@ ENTRY(startup_32)
 	 * the new gdt/idt that has __KERNEL_CS with CS.L = 1.
 	 */
 	ljmpl	$__KERNEL_CS, $pa_startup_64
+ENDPROC(startup_32)
 
 	.section ".text64","ax"
 	.code64
@@ -123,6 +126,7 @@ ENTRY(startup_32)
 ENTRY(startup_64)
 	# Now jump into the kernel using virtual addresses
 	jmpq	*tr_start(%rip)
+ENDPROC(startup_64)
 
 	.section ".rodata","a"
 	# Duplicate the global descriptor table
diff --git a/arch/x86/realmode/rm/wakeup_asm.S b/arch/x86/realmode/rm/wakeup_asm.S
index 9e7e14797a72..08203a187446 100644
--- a/arch/x86/realmode/rm/wakeup_asm.S
+++ b/arch/x86/realmode/rm/wakeup_asm.S
@@ -134,6 +134,7 @@ ENTRY(wakeup_start)
 #else
 	jmp	trampoline_start
 #endif
+ENDPROC(wakeup_start)
 
 bogus_real_magic:
 1:
diff --git a/arch/x86/xen/xen-asm_32.S b/arch/x86/xen/xen-asm_32.S
index feb6d40a0860..748beeff6e7d 100644
--- a/arch/x86/xen/xen-asm_32.S
+++ b/arch/x86/xen/xen-asm_32.S
@@ -138,6 +138,7 @@ xen_iret_end_crit:
 hyper_iret:
 	/* put this out of line since its very rarely used */
 	jmp hypercall_page + __HYPERVISOR_iret * 32
+ENDPROC(xen_iret)
 
 	.globl xen_iret_start_crit, xen_iret_end_crit
 
@@ -220,4 +221,4 @@ ENTRY(xen_iret_crit_fixup)
 
 	lea 4(%edi), %esp		/* point esp to new frame */
 2:	jmp xen_do_upcall
-
+ENDPROC(xen_iret_crit_fixup)
diff --git a/arch/x86/xen/xen-asm_64.S b/arch/x86/xen/xen-asm_64.S
index c3df43141e70..d617bea76039 100644
--- a/arch/x86/xen/xen-asm_64.S
+++ b/arch/x86/xen/xen-asm_64.S
@@ -49,6 +49,7 @@ ENTRY(xen_iret)
 1:	jmp hypercall_iret
 ENDPATCH(xen_iret)
 RELOC(xen_iret, 1b+1)
+ENDPROC(xen_iret)
 
 ENTRY(xen_sysret64)
 	/*
@@ -68,6 +69,7 @@ ENTRY(xen_sysret64)
 1:	jmp hypercall_iret
 ENDPATCH(xen_sysret64)
 RELOC(xen_sysret64, 1b+1)
+ENDPROC(xen_sysret64)
 
 /*
  * Xen handles syscall callbacks much like ordinary exceptions, which
diff --git a/arch/x86/xen/xen-head.S b/arch/x86/xen/xen-head.S
index 72a8e6adebe6..95eb4978791b 100644
--- a/arch/x86/xen/xen-head.S
+++ b/arch/x86/xen/xen-head.S
@@ -33,7 +33,7 @@ ENTRY(startup_xen)
 	mov $init_thread_union+THREAD_SIZE, %_ASM_SP
 
 	jmp xen_start_kernel
-
+ENDPROC(startup_xen)
 	__FINIT
 #endif
 
@@ -41,6 +41,7 @@ ENTRY(startup_xen)
 	.balign PAGE_SIZE
 ENTRY(hypercall_page)
 	.skip PAGE_SIZE
+ENDPROC(hypercall_page)
 
 #define HYPERCALL(n) \
 	.equ xen_hypercall_##n, hypercall_page + __HYPERVISOR_##n * 32; \
diff --git a/drivers/lguest/x86/switcher_32.S b/drivers/lguest/x86/switcher_32.S
index 40634b0db9f7..3318a967f28b 100644
--- a/drivers/lguest/x86/switcher_32.S
+++ b/drivers/lguest/x86/switcher_32.S
@@ -265,6 +265,7 @@ ENTRY(switch_to_guest)
 return_to_host:
 	SWITCH_TO_HOST
 	iret
+ENDPROC(switch_to_guest)
 
 // We are lead to the second path like so:
 // An interrupt, with some cause external
-- 
2.12.2

[PATCH v3 06/29] x86: assembly, annotate functions by ENTRY, not GLOBAL

[Jiri Slaby]

GLOBAL is meant for global symbols, but not functions. Use ENTRY which
is dedicated for global functions:
* it aligns the functions properly
* it is expected to have a corresponding ENDPROC too.

startup_64, which uses .globl explicitly, is converted too.

Besides that, x86's custom GLOBAL macro is going to die very soon.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: <x86@kernel.org>
---
 arch/x86/boot/copy.S             | 8 ++++----
 arch/x86/boot/pmjump.S           | 4 ++--
 arch/x86/entry/entry_64_compat.S | 3 +--
 arch/x86/kernel/ftrace_64.S      | 2 +-
 arch/x86/kernel/head_64.S        | 5 +++--
 5 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/arch/x86/boot/copy.S b/arch/x86/boot/copy.S
index 15d9f74b0008..030a7bde51da 100644
--- a/arch/x86/boot/copy.S
+++ b/arch/x86/boot/copy.S
@@ -17,7 +17,7 @@
 	.code16
 	.text
 
-GLOBAL(memcpy)
+ENTRY(memcpy)
 	pushw	%si
 	pushw	%di
 	movw	%ax, %di
@@ -33,7 +33,7 @@ GLOBAL(memcpy)
 	retl
 ENDPROC(memcpy)
 
-GLOBAL(memset)
+ENTRY(memset)
 	pushw	%di
 	movw	%ax, %di
 	movzbl	%dl, %eax
@@ -48,7 +48,7 @@ GLOBAL(memset)
 	retl
 ENDPROC(memset)
 
-GLOBAL(copy_from_fs)
+ENTRY(copy_from_fs)
 	pushw	%ds
 	pushw	%fs
 	popw	%ds
@@ -57,7 +57,7 @@ GLOBAL(copy_from_fs)
 	retl
 ENDPROC(copy_from_fs)
 
-GLOBAL(copy_to_fs)
+ENTRY(copy_to_fs)
 	pushw	%es
 	pushw	%fs
 	popw	%es
diff --git a/arch/x86/boot/pmjump.S b/arch/x86/boot/pmjump.S
index 3e0edc6d2a20..6528f78a79b5 100644
--- a/arch/x86/boot/pmjump.S
+++ b/arch/x86/boot/pmjump.S
@@ -23,7 +23,7 @@
 /*
  * void protected_mode_jump(u32 entrypoint, u32 bootparams);
  */
-GLOBAL(protected_mode_jump)
+ENTRY(protected_mode_jump)
 	movl	%edx, %esi		# Pointer to boot_params table
 
 	xorl	%ebx, %ebx
@@ -48,7 +48,7 @@ ENDPROC(protected_mode_jump)
 
 	.code32
 	.section ".text32","ax"
-GLOBAL(in_pm32)
+ENTRY(in_pm32)
 	# Set up data segments for flat 32-bit mode
 	movl	%ecx, %ds
 	movl	%ecx, %es
diff --git a/arch/x86/entry/entry_64_compat.S b/arch/x86/entry/entry_64_compat.S
index 14dc2f259e2f..ba6af2fba6a7 100644
--- a/arch/x86/entry/entry_64_compat.S
+++ b/arch/x86/entry/entry_64_compat.S
@@ -342,8 +342,7 @@ ENTRY(entry_INT80_compat)
 	jmp	restore_regs_and_iret
 ENDPROC(entry_INT80_compat)
 
-	ALIGN
-GLOBAL(stub32_clone)
+ENTRY(stub32_clone)
 	/*
 	 * The 32-bit clone ABI is: clone(..., int tls_val, int *child_tidptr).
 	 * The 64-bit clone ABI is: clone(..., int *child_tidptr, int tls_val).
diff --git a/arch/x86/kernel/ftrace_64.S b/arch/x86/kernel/ftrace_64.S
index a915729c0246..76c774a5e792 100644
--- a/arch/x86/kernel/ftrace_64.S
+++ b/arch/x86/kernel/ftrace_64.S
@@ -314,7 +314,7 @@ ENTRY(ftrace_graph_caller)
 	retq
 ENDPROC(ftrace_graph_caller)
 
-GLOBAL(return_to_handler)
+ENTRY(return_to_handler)
 	subq  $24, %rsp
 
 	/* Save the return values */
diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
index ac9d327d2e42..e1da48448cc0 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -46,8 +46,7 @@ L3_START_KERNEL = pud_index(__START_KERNEL_map)
 	.text
 	__HEAD
 	.code64
-	.globl startup_64
-startup_64:
+ENTRY(startup_64)
 	/*
 	 * At this point the CPU runs in 64bit mode CS.L = 1 CS.D = 0,
 	 * and someone has loaded an identity mapped page table
@@ -168,6 +167,8 @@ startup_64:
 .Lskip_fixup:
 	movq	$(early_level4_pgt - __START_KERNEL_map), %rax
 	jmp 1f
+ENDPROC(startup_64)
+
 ENTRY(secondary_startup_64)
 	/*
 	 * At this point the CPU runs in 64bit mode CS.L = 1 CS.D = 0,
-- 
2.12.2

[PATCH v3 07/29] x86: bpf_jit, use ENTRY+ENDPROC

[Jiri Slaby]

Do not use a custom macro FUNC for starts of the global functions, use
ENTRY instead.

And while at it, annotate also ends of the functions by ENDPROC.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: James Morris <jmorris@namei.org>
Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Cc: Patrick McHardy <kaber@trash.net>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Cc: netdev@vger.kernel.org
---
 arch/x86/net/bpf_jit.S | 32 ++++++++++++++++++--------------
 1 file changed, 18 insertions(+), 14 deletions(-)

diff --git a/arch/x86/net/bpf_jit.S b/arch/x86/net/bpf_jit.S
index f2a7faf4706e..762c29fb8832 100644
--- a/arch/x86/net/bpf_jit.S
+++ b/arch/x86/net/bpf_jit.S
@@ -23,16 +23,12 @@
 	32 /* space for rbx,r13,r14,r15 */ + \
 	8 /* space for skb_copy_bits */)
 
-#define FUNC(name) \
-	.globl name; \
-	.type name, @function; \
-	name:
-
-FUNC(sk_load_word)
+ENTRY(sk_load_word)
 	test	%esi,%esi
 	js	bpf_slow_path_word_neg
+ENDPROC(sk_load_word)
 
-FUNC(sk_load_word_positive_offset)
+ENTRY(sk_load_word_positive_offset)
 	mov	%r9d,%eax		# hlen
 	sub	%esi,%eax		# hlen - offset
 	cmp	$3,%eax
@@ -40,12 +36,14 @@ FUNC(sk_load_word_positive_offset)
 	mov     (SKBDATA,%rsi),%eax
 	bswap   %eax  			/* ntohl() */
 	ret
+ENDPROC(sk_load_word_positive_offset)
 
-FUNC(sk_load_half)
+ENTRY(sk_load_half)
 	test	%esi,%esi
 	js	bpf_slow_path_half_neg
+ENDPROC(sk_load_half)
 
-FUNC(sk_load_half_positive_offset)
+ENTRY(sk_load_half_positive_offset)
 	mov	%r9d,%eax
 	sub	%esi,%eax		#	hlen - offset
 	cmp	$1,%eax
@@ -53,16 +51,19 @@ FUNC(sk_load_half_positive_offset)
 	movzwl	(SKBDATA,%rsi),%eax
 	rol	$8,%ax			# ntohs()
 	ret
+ENDPROC(sk_load_half_positive_offset)
 
-FUNC(sk_load_byte)
+ENTRY(sk_load_byte)
 	test	%esi,%esi
 	js	bpf_slow_path_byte_neg
+ENDPROC(sk_load_byte)
 
-FUNC(sk_load_byte_positive_offset)
+ENTRY(sk_load_byte_positive_offset)
 	cmp	%esi,%r9d   /* if (offset >= hlen) goto bpf_slow_path_byte */
 	jle	bpf_slow_path_byte
 	movzbl	(SKBDATA,%rsi),%eax
 	ret
+ENDPROC(sk_load_byte_positive_offset)
 
 /* rsi contains offset and can be scratched */
 #define bpf_slow_path_common(LEN)		\
@@ -119,31 +120,34 @@ bpf_slow_path_word_neg:
 	cmp	SKF_MAX_NEG_OFF, %esi	/* test range */
 	jl	bpf_error	/* offset lower -> error  */
 
-FUNC(sk_load_word_negative_offset)
+ENTRY(sk_load_word_negative_offset)
 	sk_negative_common(4)
 	mov	(%rax), %eax
 	bswap	%eax
 	ret
+ENDPROC(sk_load_word_negative_offset)
 
 bpf_slow_path_half_neg:
 	cmp	SKF_MAX_NEG_OFF, %esi
 	jl	bpf_error
 
-FUNC(sk_load_half_negative_offset)
+ENTRY(sk_load_half_negative_offset)
 	sk_negative_common(2)
 	mov	(%rax),%ax
 	rol	$8,%ax
 	movzwl	%ax,%eax
 	ret
+ENDPROC(sk_load_half_negative_offset)
 
 bpf_slow_path_byte_neg:
 	cmp	SKF_MAX_NEG_OFF, %esi
 	jl	bpf_error
 
-FUNC(sk_load_byte_negative_offset)
+ENTRY(sk_load_byte_negative_offset)
 	sk_negative_common(1)
 	movzbl	(%rax), %eax
 	ret
+ENDPROC(sk_load_byte_negative_offset)
 
 bpf_error:
 # force a return 0 from jit handler
-- 
2.12.2

Re: [PATCH v3 07/29] x86: bpf_jit, use ENTRY+ENDPROC

[Alexei Starovoitov]

On Fri, Apr 21, 2017 at 04:12:43PM +0200, Jiri Slaby wrote:
> Do not use a custom macro FUNC for starts of the global functions, use
> ENTRY instead.
> 
> And while at it, annotate also ends of the functions by ENDPROC.
> 
> Signed-off-by: Jiri Slaby <jslaby@suse.cz>
> Cc: "David S. Miller" <davem@davemloft.net>
> Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
> Cc: James Morris <jmorris@namei.org>
> Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
> Cc: Patrick McHardy <kaber@trash.net>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: "H. Peter Anvin" <hpa@zytor.com>
> Cc: x86@kernel.org
> Cc: netdev@vger.kernel.org
> ---
>  arch/x86/net/bpf_jit.S | 32 ++++++++++++++++++--------------
>  1 file changed, 18 insertions(+), 14 deletions(-)
> 
> diff --git a/arch/x86/net/bpf_jit.S b/arch/x86/net/bpf_jit.S
> index f2a7faf4706e..762c29fb8832 100644
> --- a/arch/x86/net/bpf_jit.S
> +++ b/arch/x86/net/bpf_jit.S
> @@ -23,16 +23,12 @@
>  	32 /* space for rbx,r13,r14,r15 */ + \
>  	8 /* space for skb_copy_bits */)
>  
> -#define FUNC(name) \
> -	.globl name; \
> -	.type name, @function; \
> -	name:
> -
> -FUNC(sk_load_word)
> +ENTRY(sk_load_word)
>  	test	%esi,%esi
>  	js	bpf_slow_path_word_neg
> +ENDPROC(sk_load_word)

this doens't look right.
It will add alignment nops in critical paths of these pseudo functions.
I'm also not sure whether it will still work afterwards.
Was it tested?
I'd prefer if this code kept as-is.

> -FUNC(sk_load_word_positive_offset)
> +ENTRY(sk_load_word_positive_offset)
>  	mov	%r9d,%eax		# hlen
>  	sub	%esi,%eax		# hlen - offset
>  	cmp	$3,%eax
> @@ -40,12 +36,14 @@ FUNC(sk_load_word_positive_offset)
>  	mov     (SKBDATA,%rsi),%eax
>  	bswap   %eax  			/* ntohl() */
>  	ret
> +ENDPROC(sk_load_word_positive_offset)
>  
> -FUNC(sk_load_half)
> +ENTRY(sk_load_half)
>  	test	%esi,%esi
>  	js	bpf_slow_path_half_neg
> +ENDPROC(sk_load_half)
>  
> -FUNC(sk_load_half_positive_offset)
> +ENTRY(sk_load_half_positive_offset)
>  	mov	%r9d,%eax
>  	sub	%esi,%eax		#	hlen - offset
>  	cmp	$1,%eax
> @@ -53,16 +51,19 @@ FUNC(sk_load_half_positive_offset)
>  	movzwl	(SKBDATA,%rsi),%eax
>  	rol	$8,%ax			# ntohs()
>  	ret
> +ENDPROC(sk_load_half_positive_offset)
>  
> -FUNC(sk_load_byte)
> +ENTRY(sk_load_byte)
>  	test	%esi,%esi
>  	js	bpf_slow_path_byte_neg
> +ENDPROC(sk_load_byte)
>  
> -FUNC(sk_load_byte_positive_offset)
> +ENTRY(sk_load_byte_positive_offset)
>  	cmp	%esi,%r9d   /* if (offset >= hlen) goto bpf_slow_path_byte */
>  	jle	bpf_slow_path_byte
>  	movzbl	(SKBDATA,%rsi),%eax
>  	ret
> +ENDPROC(sk_load_byte_positive_offset)
>  
>  /* rsi contains offset and can be scratched */
>  #define bpf_slow_path_common(LEN)		\
> @@ -119,31 +120,34 @@ bpf_slow_path_word_neg:
>  	cmp	SKF_MAX_NEG_OFF, %esi	/* test range */
>  	jl	bpf_error	/* offset lower -> error  */
>  
> -FUNC(sk_load_word_negative_offset)
> +ENTRY(sk_load_word_negative_offset)
>  	sk_negative_common(4)
>  	mov	(%rax), %eax
>  	bswap	%eax
>  	ret
> +ENDPROC(sk_load_word_negative_offset)
>  
>  bpf_slow_path_half_neg:
>  	cmp	SKF_MAX_NEG_OFF, %esi
>  	jl	bpf_error
>  
> -FUNC(sk_load_half_negative_offset)
> +ENTRY(sk_load_half_negative_offset)
>  	sk_negative_common(2)
>  	mov	(%rax),%ax
>  	rol	$8,%ax
>  	movzwl	%ax,%eax
>  	ret
> +ENDPROC(sk_load_half_negative_offset)
>  
>  bpf_slow_path_byte_neg:
>  	cmp	SKF_MAX_NEG_OFF, %esi
>  	jl	bpf_error
>  
> -FUNC(sk_load_byte_negative_offset)
> +ENTRY(sk_load_byte_negative_offset)
>  	sk_negative_common(1)
>  	movzbl	(%rax), %eax
>  	ret
> +ENDPROC(sk_load_byte_negative_offset)
>  
>  bpf_error:
>  # force a return 0 from jit handler
> -- 
> 2.12.2
> 

Re: [PATCH v3 07/29] x86: bpf_jit, use ENTRY+ENDPROC

[Jiri Slaby]

On 04/21/2017, 09:32 PM, Alexei Starovoitov wrote:
> On Fri, Apr 21, 2017 at 04:12:43PM +0200, Jiri Slaby wrote:
>> Do not use a custom macro FUNC for starts of the global functions, use
>> ENTRY instead.
>>
>> And while at it, annotate also ends of the functions by ENDPROC.
>>
>> Signed-off-by: Jiri Slaby <jslaby@suse.cz>
>> Cc: "David S. Miller" <davem@davemloft.net>
>> Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
>> Cc: James Morris <jmorris@namei.org>
>> Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
>> Cc: Patrick McHardy <kaber@trash.net>
>> Cc: Thomas Gleixner <tglx@linutronix.de>
>> Cc: Ingo Molnar <mingo@redhat.com>
>> Cc: "H. Peter Anvin" <hpa@zytor.com>
>> Cc: x86@kernel.org
>> Cc: netdev@vger.kernel.org
>> ---
>>  arch/x86/net/bpf_jit.S | 32 ++++++++++++++++++--------------
>>  1 file changed, 18 insertions(+), 14 deletions(-)
>>
>> diff --git a/arch/x86/net/bpf_jit.S b/arch/x86/net/bpf_jit.S
>> index f2a7faf4706e..762c29fb8832 100644
>> --- a/arch/x86/net/bpf_jit.S
>> +++ b/arch/x86/net/bpf_jit.S
>> @@ -23,16 +23,12 @@
>>  	32 /* space for rbx,r13,r14,r15 */ + \
>>  	8 /* space for skb_copy_bits */)
>>  
>> -#define FUNC(name) \
>> -	.globl name; \
>> -	.type name, @function; \
>> -	name:
>> -
>> -FUNC(sk_load_word)
>> +ENTRY(sk_load_word)
>>  	test	%esi,%esi
>>  	js	bpf_slow_path_word_neg
>> +ENDPROC(sk_load_word)
> 
> this doens't look right.
> It will add alignment nops in critical paths of these pseudo functions.
> I'm also not sure whether it will still work afterwards.
> Was it tested?
> I'd prefer if this code kept as-is.

It cannot stay as-is simply because we want to know where the functions
end to inject debuginfo properly. The code above does not warrant for
any exception.

Executing a nop takes a little and having externally-callable functions
aligned can actually help performance (no, I haven't measured nor tested
the code). But sure, the tool is generic, so I can introduce a local
macros to avoid alignments in the functions:

#define BPF_FUNC_START_LOCAL(name) \
		SYM_START(name, SYM_V_LOCAL, SYM_A_NONE)
#define BPF_FUNC_START(name) \
		SYM_START(name, SYM_V_GLOBAL, SYM_A_NONE)

#define BPF_FUNC_END(name) SYM_FUNC_END(name)

thanks,
-- 
js
suse labs

Re: [PATCH v3 07/29] x86: bpf_jit, use ENTRY+ENDPROC

[David Miller]

From: Jiri Slaby <jslaby@suse.cz>
Date: Mon, 24 Apr 2017 08:45:11 +0200

> On 04/21/2017, 09:32 PM, Alexei Starovoitov wrote:
>> On Fri, Apr 21, 2017 at 04:12:43PM +0200, Jiri Slaby wrote:
>>> Do not use a custom macro FUNC for starts of the global functions, use
>>> ENTRY instead.
>>>
>>> And while at it, annotate also ends of the functions by ENDPROC.
>>>
>>> Signed-off-by: Jiri Slaby <jslaby@suse.cz>
>>> Cc: "David S. Miller" <davem@davemloft.net>
>>> Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
>>> Cc: James Morris <jmorris@namei.org>
>>> Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
>>> Cc: Patrick McHardy <kaber@trash.net>
>>> Cc: Thomas Gleixner <tglx@linutronix.de>
>>> Cc: Ingo Molnar <mingo@redhat.com>
>>> Cc: "H. Peter Anvin" <hpa@zytor.com>
>>> Cc: x86@kernel.org
>>> Cc: netdev@vger.kernel.org
>>> ---
>>>  arch/x86/net/bpf_jit.S | 32 ++++++++++++++++++--------------
>>>  1 file changed, 18 insertions(+), 14 deletions(-)
>>>
>>> diff --git a/arch/x86/net/bpf_jit.S b/arch/x86/net/bpf_jit.S
>>> index f2a7faf4706e..762c29fb8832 100644
>>> --- a/arch/x86/net/bpf_jit.S
>>> +++ b/arch/x86/net/bpf_jit.S
>>> @@ -23,16 +23,12 @@
>>>  	32 /* space for rbx,r13,r14,r15 */ + \
>>>  	8 /* space for skb_copy_bits */)
>>>  
>>> -#define FUNC(name) \
>>> -	.globl name; \
>>> -	.type name, @function; \
>>> -	name:
>>> -
>>> -FUNC(sk_load_word)
>>> +ENTRY(sk_load_word)
>>>  	test	%esi,%esi
>>>  	js	bpf_slow_path_word_neg
>>> +ENDPROC(sk_load_word)
>> 
>> this doens't look right.
>> It will add alignment nops in critical paths of these pseudo functions.
>> I'm also not sure whether it will still work afterwards.
>> Was it tested?
>> I'd prefer if this code kept as-is.
> 
> It cannot stay as-is simply because we want to know where the functions
> end to inject debuginfo properly. The code above does not warrant for
> any exception.

I totally and completely disagree.

> Executing a nop takes a little and having externally-callable functions
> aligned can actually help performance (no, I haven't measured nor tested
> the code). But sure, the tool is generic, so I can introduce a local
> macros to avoid alignments in the functions:

Not for this case, it's a bunch of entry points all packed together
intentionally so that SKB accesses of different access sizes (which is
almost always the case) from BPF programs use the smallest amount of
I-cache as possible.

Re: [PATCH v3 07/29] x86: bpf_jit, use ENTRY+ENDPROC

[Jiri Slaby]

On 04/24/2017, 04:41 PM, David Miller wrote:
>> It cannot stay as-is simply because we want to know where the functions
>> end to inject debuginfo properly. The code above does not warrant for
>> any exception.
> 
> I totally and completely disagree.

You can disagree as you wish but there is really nothing special on the
bpf code with respect to annotations.

>> Executing a nop takes a little and having externally-callable functions
>> aligned can actually help performance (no, I haven't measured nor tested
>> the code). But sure, the tool is generic, so I can introduce a local
>> macros to avoid alignments in the functions:
> 
> Not for this case, it's a bunch of entry points all packed together
> intentionally so that SKB accesses of different access sizes (which is
> almost always the case) from BPF programs use the smallest amount of
> I-cache as possible.

And for that reason I suggested the special macros for the code (see the
macros in the e-mail you replied to again). So what problem do you
actually have with the suggested solution?

thanks,
-- 
js
suse labs

Re: [PATCH v3 07/29] x86: bpf_jit, use ENTRY+ENDPROC

[David Miller]

From: Jiri Slaby <jslaby@suse.cz>
Date: Mon, 24 Apr 2017 16:52:43 +0200

> On 04/24/2017, 04:41 PM, David Miller wrote:
>>> It cannot stay as-is simply because we want to know where the functions
>>> end to inject debuginfo properly. The code above does not warrant for
>>> any exception.
>> 
>> I totally and completely disagree.
> 
> You can disagree as you wish but there is really nothing special on the
> bpf code with respect to annotations.
> 
>>> Executing a nop takes a little and having externally-callable functions
>>> aligned can actually help performance (no, I haven't measured nor tested
>>> the code). But sure, the tool is generic, so I can introduce a local
>>> macros to avoid alignments in the functions:
>> 
>> Not for this case, it's a bunch of entry points all packed together
>> intentionally so that SKB accesses of different access sizes (which is
>> almost always the case) from BPF programs use the smallest amount of
>> I-cache as possible.
> 
> And for that reason I suggested the special macros for the code (see the
> macros in the e-mail you replied to again). So what problem do you
> actually have with the suggested solution?

If you align the entry points, then the code sequence as a whole is
are no longer densely packed.

Or do I misunderstand how your macros work?

Re: [PATCH v3 07/29] x86: bpf_jit, use ENTRY+ENDPROC

[Jiri Slaby]

On 04/24/2017, 05:08 PM, David Miller wrote:
> If you align the entry points, then the code sequence as a whole is
> are no longer densely packed.

Sure.

> Or do I misunderstand how your macros work?

Perhaps. So the suggested macros for the code are:
#define BPF_FUNC_START_LOCAL(name) \
		SYM_START(name, SYM_V_LOCAL, SYM_A_NONE)
#define BPF_FUNC_START(name) \
		SYM_START(name, SYM_V_GLOBAL, SYM_A_NONE)

and they differ from the standard ones:
#define SYM_FUNC_START_LOCAL(name)                      \
        SYM_START(name, SYM_V_LOCAL, SYM_A_ALIGN)
#define SYM_FUNC_START(name)                            \
        SYM_START(name, SYM_V_GLOBAL, SYM_A_ALIGN)


The difference is SYM_A_NONE vs. SYM_A_ALIGN, which means:
#define SYM_A_ALIGN                             ALIGN
#define SYM_A_NONE                              /* nothing */

Does it look OK now?

thanks,
-- 
js
suse labs

Re: [PATCH v3 07/29] x86: bpf_jit, use ENTRY+ENDPROC

[David Miller]

From: Jiri Slaby <jslaby@suse.cz>
Date: Mon, 24 Apr 2017 17:41:06 +0200

> On 04/24/2017, 05:08 PM, David Miller wrote:
>> If you align the entry points, then the code sequence as a whole is
>> are no longer densely packed.
> 
> Sure.
> 
>> Or do I misunderstand how your macros work?
> 
> Perhaps. So the suggested macros for the code are:
> #define BPF_FUNC_START_LOCAL(name) \
> 		SYM_START(name, SYM_V_LOCAL, SYM_A_NONE)
> #define BPF_FUNC_START(name) \
> 		SYM_START(name, SYM_V_GLOBAL, SYM_A_NONE)
> 
> and they differ from the standard ones:
> #define SYM_FUNC_START_LOCAL(name)                      \
>         SYM_START(name, SYM_V_LOCAL, SYM_A_ALIGN)
> #define SYM_FUNC_START(name)                            \
>         SYM_START(name, SYM_V_GLOBAL, SYM_A_ALIGN)
> 
> 
> The difference is SYM_A_NONE vs. SYM_A_ALIGN, which means:
> #define SYM_A_ALIGN                             ALIGN
> #define SYM_A_NONE                              /* nothing */
> 
> Does it look OK now?

I said I'm not OK with the alignment, so personally I am not
with how these macros work and what they will do to the code
generated for BPF packet accesses.

But I'll defer to Alexei on this because I don't have the time
nor the energy to fight this.

Thanks.

Re: [PATCH v3 07/29] x86: bpf_jit, use ENTRY+ENDPROC

[Jiri Slaby]

On 04/24/2017, 05:51 PM, David Miller wrote:
> I said I'm not OK with the alignment

So in short, the suggested macros add no alignment.

-- 
js
suse labs

Re: [PATCH v3 07/29] x86: bpf_jit, use ENTRY+ENDPROC

[Ingo Molnar]

* Jiri Slaby <jslaby@suse.cz> wrote:

> On 04/24/2017, 05:08 PM, David Miller wrote:
> > If you align the entry points, then the code sequence as a whole is
> > are no longer densely packed.
> 
> Sure.
> 
> > Or do I misunderstand how your macros work?
> 
> Perhaps. So the suggested macros for the code are:
> #define BPF_FUNC_START_LOCAL(name) \
> 		SYM_START(name, SYM_V_LOCAL, SYM_A_NONE)
> #define BPF_FUNC_START(name) \
> 		SYM_START(name, SYM_V_GLOBAL, SYM_A_NONE)
> 
> and they differ from the standard ones:
> #define SYM_FUNC_START_LOCAL(name)                      \
>         SYM_START(name, SYM_V_LOCAL, SYM_A_ALIGN)
> #define SYM_FUNC_START(name)                            \
>         SYM_START(name, SYM_V_GLOBAL, SYM_A_ALIGN)
> 
> 
> The difference is SYM_A_NONE vs. SYM_A_ALIGN, which means:
> #define SYM_A_ALIGN                             ALIGN
> #define SYM_A_NONE                              /* nothing */
> 
> Does it look OK now?

No, the patch changes alignment which is undesirable, it needs to preserve the 
existing (non-)alignment of the symbols!

Thanks,

	Ingo

Re: [PATCH v3 07/29] x86: bpf_jit, use ENTRY+ENDPROC

[Jiri Slaby]

On 04/24/2017, 05:55 PM, Ingo Molnar wrote:
> * Jiri Slaby <jslaby@suse.cz> wrote:
> 
>> On 04/24/2017, 05:08 PM, David Miller wrote:
>>> If you align the entry points, then the code sequence as a whole is
>>> are no longer densely packed.
>>
>> Sure.
>>
>>> Or do I misunderstand how your macros work?
>>
>> Perhaps. So the suggested macros for the code are:
>> #define BPF_FUNC_START_LOCAL(name) \
>> 		SYM_START(name, SYM_V_LOCAL, SYM_A_NONE)
>> #define BPF_FUNC_START(name) \
>> 		SYM_START(name, SYM_V_GLOBAL, SYM_A_NONE)
>>
>> and they differ from the standard ones:
>> #define SYM_FUNC_START_LOCAL(name)                      \
>>         SYM_START(name, SYM_V_LOCAL, SYM_A_ALIGN)
>> #define SYM_FUNC_START(name)                            \
>>         SYM_START(name, SYM_V_GLOBAL, SYM_A_ALIGN)
>>
>>
>> The difference is SYM_A_NONE vs. SYM_A_ALIGN, which means:
>> #define SYM_A_ALIGN                             ALIGN
>> #define SYM_A_NONE                              /* nothing */
>>
>> Does it look OK now?
> 
> No, the patch changes alignment which is undesirable, it needs to preserve the 
> existing (non-)alignment of the symbols!

OK, so I am not expressing myself explicitly enough, it seems.

So, correct, the patch v3 adds alignments. I suggested in the discussion
the macros above. They do not add alignments. If everybody is OK with
that, v4 of the patch won't add alignments. OK?

thanks,
-- 
js
suse labs

Re: [PATCH v3 07/29] x86: bpf_jit, use ENTRY+ENDPROC

[Ingo Molnar]

* Jiri Slaby <jslaby@suse.cz> wrote:

> On 04/24/2017, 05:55 PM, Ingo Molnar wrote:
> > * Jiri Slaby <jslaby@suse.cz> wrote:
> > 
> >> On 04/24/2017, 05:08 PM, David Miller wrote:
> >>> If you align the entry points, then the code sequence as a whole is
> >>> are no longer densely packed.
> >>
> >> Sure.
> >>
> >>> Or do I misunderstand how your macros work?
> >>
> >> Perhaps. So the suggested macros for the code are:
> >> #define BPF_FUNC_START_LOCAL(name) \
> >> 		SYM_START(name, SYM_V_LOCAL, SYM_A_NONE)
> >> #define BPF_FUNC_START(name) \
> >> 		SYM_START(name, SYM_V_GLOBAL, SYM_A_NONE)
> >>
> >> and they differ from the standard ones:
> >> #define SYM_FUNC_START_LOCAL(name)                      \
> >>         SYM_START(name, SYM_V_LOCAL, SYM_A_ALIGN)
> >> #define SYM_FUNC_START(name)                            \
> >>         SYM_START(name, SYM_V_GLOBAL, SYM_A_ALIGN)
> >>
> >>
> >> The difference is SYM_A_NONE vs. SYM_A_ALIGN, which means:
> >> #define SYM_A_ALIGN                             ALIGN
> >> #define SYM_A_NONE                              /* nothing */
> >>
> >> Does it look OK now?
> > 
> > No, the patch changes alignment which is undesirable, it needs to preserve the 
> > existing (non-)alignment of the symbols!
> 
> OK, so I am not expressing myself explicitly enough, it seems.
> 
> So, correct, the patch v3 adds alignments. I suggested in the discussion
> the macros above. They do not add alignments. If everybody is OK with
> that, v4 of the patch won't add alignments. OK?

Yes.

Thanks,

	Ingo

Re: [PATCH v3 07/29] x86: bpf_jit, use ENTRY+ENDPROC

[Alexei Starovoitov]

On Mon, Apr 24, 2017 at 06:02:51PM +0200, Jiri Slaby wrote:
> On 04/24/2017, 05:55 PM, Ingo Molnar wrote:
> > * Jiri Slaby <jslaby@suse.cz> wrote:
> > 
> >> On 04/24/2017, 05:08 PM, David Miller wrote:
> >>> If you align the entry points, then the code sequence as a whole is
> >>> are no longer densely packed.
> >>
> >> Sure.
> >>
> >>> Or do I misunderstand how your macros work?
> >>
> >> Perhaps. So the suggested macros for the code are:
> >> #define BPF_FUNC_START_LOCAL(name) \
> >> 		SYM_START(name, SYM_V_LOCAL, SYM_A_NONE)
> >> #define BPF_FUNC_START(name) \
> >> 		SYM_START(name, SYM_V_GLOBAL, SYM_A_NONE)
> >>
> >> and they differ from the standard ones:
> >> #define SYM_FUNC_START_LOCAL(name)                      \
> >>         SYM_START(name, SYM_V_LOCAL, SYM_A_ALIGN)
> >> #define SYM_FUNC_START(name)                            \
> >>         SYM_START(name, SYM_V_GLOBAL, SYM_A_ALIGN)
> >>
> >>
> >> The difference is SYM_A_NONE vs. SYM_A_ALIGN, which means:
> >> #define SYM_A_ALIGN                             ALIGN
> >> #define SYM_A_NONE                              /* nothing */
> >>
> >> Does it look OK now?
> > 
> > No, the patch changes alignment which is undesirable, it needs to preserve the 
> > existing (non-)alignment of the symbols!
> 
> OK, so I am not expressing myself explicitly enough, it seems.
> 
> So, correct, the patch v3 adds alignments. I suggested in the discussion
> the macros above. They do not add alignments. If everybody is OK with
> that, v4 of the patch won't add alignments. OK?

can we go back to what problem this patch set is trying to solve?
Sounds like you want to add _function_ start/end marks to aid debugging?
Debugging with what? What tool will recognize this stuff?

Take a look at what your patch does:
+ENTRY(sk_load_word)
        test    %esi,%esi
        js      bpf_slow_path_word_neg
+ENDPROC(sk_load_word)

Does above two assembler instructions look like a function?

or this:
+ENTRY(sk_load_byte_positive_offset)
        cmp     %esi,%r9d   /* if (offset >= hlen) goto bpf_slow_path_byte */
        jle     bpf_slow_path_byte
        movzbl  (SKBDATA,%rsi),%eax
        ret
+ENDPROC(sk_load_byte_positive_offset)

This assembler code doesn't represent functions. There is no prologue/epilogue
and no stack frame. JITed code uses 'call' insn to jump into them, but they're
not your typical C functions.
Take a look at bpf_slow_path_common() macro that creates the frame before
calling into C code with 'call skb_copy_bits;'

I still think that this code should be left alone.
Even macro names you're proposing:
 #define BPF_FUNC_START_LOCAL
don't sound right. These are not functions.

Re: [PATCH v3 07/29] x86: bpf_jit, use ENTRY+ENDPROC

[Jiri Slaby]

On 04/24/2017, 06:47 PM, Alexei Starovoitov wrote:
> On Mon, Apr 24, 2017 at 06:02:51PM +0200, Jiri Slaby wrote:
>> On 04/24/2017, 05:55 PM, Ingo Molnar wrote:
>>> * Jiri Slaby <jslaby@suse.cz> wrote:
>>>
>>>> On 04/24/2017, 05:08 PM, David Miller wrote:
>>>>> If you align the entry points, then the code sequence as a whole is
>>>>> are no longer densely packed.
>>>>
>>>> Sure.
>>>>
>>>>> Or do I misunderstand how your macros work?
>>>>
>>>> Perhaps. So the suggested macros for the code are:
>>>> #define BPF_FUNC_START_LOCAL(name) \
>>>> 		SYM_START(name, SYM_V_LOCAL, SYM_A_NONE)
>>>> #define BPF_FUNC_START(name) \
>>>> 		SYM_START(name, SYM_V_GLOBAL, SYM_A_NONE)
>>>>
>>>> and they differ from the standard ones:
>>>> #define SYM_FUNC_START_LOCAL(name)                      \
>>>>         SYM_START(name, SYM_V_LOCAL, SYM_A_ALIGN)
>>>> #define SYM_FUNC_START(name)                            \
>>>>         SYM_START(name, SYM_V_GLOBAL, SYM_A_ALIGN)
>>>>
>>>>
>>>> The difference is SYM_A_NONE vs. SYM_A_ALIGN, which means:
>>>> #define SYM_A_ALIGN                             ALIGN
>>>> #define SYM_A_NONE                              /* nothing */
>>>>
>>>> Does it look OK now?
>>>
>>> No, the patch changes alignment which is undesirable, it needs to preserve the 
>>> existing (non-)alignment of the symbols!
>>
>> OK, so I am not expressing myself explicitly enough, it seems.
>>
>> So, correct, the patch v3 adds alignments. I suggested in the discussion
>> the macros above. They do not add alignments. If everybody is OK with
>> that, v4 of the patch won't add alignments. OK?
> 
> can we go back to what problem this patch set is trying to solve?
> Sounds like you want to add _function_ start/end marks to aid debugging?
> Debugging with what? What tool will recognize this stuff?

objtool will generate DWARF debuginfo between every ENTRY and ENDPROC
(dubbed differently at the end of the series). DWARF is understood by
everything, including the kernel proper (we have DWARF unwinder in
SUSE's kernels available for decades).

> Take a look at what your patch does:
> +ENTRY(sk_load_word)
>         test    %esi,%esi
>         js      bpf_slow_path_word_neg
> +ENDPROC(sk_load_word)
> 
> Does above two assembler instructions look like a function?

Yes, you are right, the code is complete mess.

For example what's the point of making the sk_load_word_positive_offset
label a global, callable function? Note that this is exactly the reason
why this particular two hunks look weird to you even though the
annotations only mechanically paraphrase what is in the current code.

> or this:
> +ENTRY(sk_load_byte_positive_offset)
>         cmp     %esi,%r9d   /* if (offset >= hlen) goto bpf_slow_path_byte */
>         jle     bpf_slow_path_byte
>         movzbl  (SKBDATA,%rsi),%eax
>         ret
> +ENDPROC(sk_load_byte_positive_offset)
> 
> This assembler code doesn't represent functions. There is no prologue/epilogue
> and no stack frame. JITed code uses 'call' insn to jump into them, but they're
> not your typical C functions.

I am not looking for C functions everywhere in assembly, actually. (In
the ideal assembly, I would and in most cases I really am.) But you are
right the annotations of the current code aren't right. It results in my
annotations being wrong too -- based on invalid assumptions.

> Take a look at bpf_slow_path_common() macro that creates the frame before
> calling into C code with 'call skb_copy_bits;'
> 
> I still think that this code should be left alone.
> Even macro names you're proposing:
>  #define BPF_FUNC_START_LOCAL
> don't sound right. These are not functions.

Well, what is the reason to call them FUNC in the current code then?

thanks,
-- 
js
suse labs

Re: [PATCH v3 07/29] x86: bpf_jit, use ENTRY+ENDPROC

[David Miller]

From: Jiri Slaby <jslaby@suse.cz>
Date: Mon, 24 Apr 2017 19:51:54 +0200

> For example what's the point of making the sk_load_word_positive_offset
> label a global, callable function? Note that this is exactly the reason
> why this particular two hunks look weird to you even though the
> annotations only mechanically paraphrase what is in the current code.

So that it can be referenced by the eBPF JIT, because these are
helpers for eBPF JIT generated code.  Every architecture implementing
an eBPF JIT has this "mess".

You can't even put a tracepoint or kprobe on these things and expect
to see "arguments" or "return PC" values in the usual spots.  This
code has special calling conventions and register usage as Alexei
explained.

I would suggest that you read and understand how this assembler is
designed, how it is called from the generated JIT code, and what it's
semantics and register usage are, before trying to annotating it.

Thank you.

Re: [PATCH v3 07/29] x86: bpf_jit, use ENTRY+ENDPROC

[Jiri Slaby]

On 04/24/2017, 08:24 PM, David Miller wrote:
> From: Jiri Slaby <jslaby@suse.cz>
> Date: Mon, 24 Apr 2017 19:51:54 +0200
> 
>> For example what's the point of making the sk_load_word_positive_offset
>> label a global, callable function? Note that this is exactly the reason
>> why this particular two hunks look weird to you even though the
>> annotations only mechanically paraphrase what is in the current code.
> 
> So that it can be referenced by the eBPF JIT, because these are
> helpers for eBPF JIT generated code.  Every architecture implementing
> an eBPF JIT has this "mess".

I completely understand the needs for this, but I am complaining about
the way it is written. That is not the best -- unbalanced annotations, C
macros in lowercase (apart from that, C macros in .S need semicolons &
backslashes), FUNC macro, etc.

> You can't even put a tracepoint or kprobe on these things and expect
> to see "arguments" or "return PC" values in the usual spots.  This
> code has special calling conventions and register usage as Alexei
> explained.

Yes, I can see that.

> I would suggest that you read and understand how this assembler is
> designed, how it is called from the generated JIT code, and what it's
> semantics and register usage are, before trying to annotating it.

Of course I studied the code. I only missed macro CHOOSE_LOAD_FUNC which
I see now. So that answers why sk_load_word_positive_offset & similar
are marked as .globl.


But the original question I asked still remains: why do you mind calling
them BPF_FUNC_START & *_END, given:

1) the functions are marked by "FUNC" already:
$ git grep FUNC linus/master arch/x86/net/bpf_jit.S
linus/master:arch/x86/net/bpf_jit.S:#define FUNC(name) \
linus/master:arch/x86/net/bpf_jit.S:FUNC(sk_load_word)
linus/master:arch/x86/net/bpf_jit.S:FUNC(sk_load_word_positive_offset)
linus/master:arch/x86/net/bpf_jit.S:FUNC(sk_load_half)
linus/master:arch/x86/net/bpf_jit.S:FUNC(sk_load_half_positive_offset)
linus/master:arch/x86/net/bpf_jit.S:FUNC(sk_load_byte)
linus/master:arch/x86/net/bpf_jit.S:FUNC(sk_load_byte_positive_offset)
linus/master:arch/x86/net/bpf_jit.S:FUNC(sk_load_word_negative_offset)
linus/master:arch/x86/net/bpf_jit.S:FUNC(sk_load_half_negative_offset)
linus/master:arch/x86/net/bpf_jit.S:FUNC(sk_load_byte_negative_offset)

2) they _are_ all callable from within the JIT code:
EMIT1_off32(0xE8, jmp_offset);

Yes, I fucked up the ENDs. They should be on different locations. But
the pieces are still functions from my POV and should be annotated
accordingly.

thanks,
-- 
js
suse labs

[PATCH v3 08/29] linkage: new macros for assembler symbols

[Jiri Slaby]

Introduce new C macros for annotations of functions and data in
assembly. There is a long-term mess in macros like ENTRY, END, ENDPROC
and similar. They are used in different manners and sometimes
incorrectly.

So introduce macros with clear use to annotate assembly as follows:

a) Support macros
   SYM_T_FUNC -- type used by assembler to mark functions
   SYM_T_OBJECT -- type used by assembler to mark data

   They are defined as STT_FUNC and STT_OBJECT respectively. According
   to the gas manual, this is the most portable way. I am not sure about
   other assemblers, so we can switch this back to %function and %object
   if this turns into a problem. Architectures can also override them by
   something like ", @function" if need be.

   SYM_A_ALIGN, SYM_A_NONE -- align the symbol?
   SYM_V_GLOBAL, SYM_V_WEAK, SYM_V_LOCAL -- visibility of symbols
b) Mostly internal annotations, used by the ones below
   SYM_ENTRY -- use only if you have to for non-paired symbols
   SYM_START -- use only if you have to
   SYM_END -- use only if you have to
c) Generic annotations
d) Annotations for code
   SYM_FUNC_START_LOCAL_ALIAS -- use where there are two local names for
	one code
   SYM_FUNC_START_ALIAS -- use where there are two global names for one
	code
   SYM_FUNC_END_ALIAS -- the end of LOCAL_ALIASed or ALIASed code

   SYM_FUNC_START -- use for global functions
   SYM_FUNC_START_LOCAL -- use for local functions
   SYM_FUNC_START_WEAK -- use for weak functions
   SYM_FUNC_END -- the end of SYM_FUNC_START_LOCAL, SYM_FUNC_START,
	SYM_FUNC_START_WEAK, ...

   SYM_FUNC_INNER_LABEL -- only for labels in the middle of functions
d) For data
   SYM_DATA_START -- global data symbol
   SYM_DATA_END -- the end of SYM_DATA_START symbol
   SYM_DATA_END_LABEL -- the labeled end of SYM_DATA_START symbol
   SYM_DATA_SIMPLE -- start+end wrapper around simple global data
   SYM_DATA_SIMPLE_LOCAL -- start+end wrapper around simple local data

==========

Note that SYM_FUNC_START_WEAK aligns symbols now too.

The macros allow to pair starts and ends of functions and mark function
correctly in the output ELF objects. This will also help a lot to
generate DWARF information automatically during build of asm.

Finally, all users of the old macros will be converted to use these
later.

[v2]
* use SYM_ prefix and sane names
* add SYM_START and SYM_END and parametrize all the macros

[v3]
* add SYM_DATA_SIMPLE, SYM_DATA_SIMPLE_LOCAL, and SYM_DATA_END_LABEL

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: hpa@zytor.com
Cc: Ingo Molnar <mingo@kernel.org>
Cc: jpoimboe@redhat.com
Cc: Juergen Gross <jgross@suse.com>
Cc: Len Brown <len.brown@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org
Cc: linux-pm@vger.kernel.org
Cc: mingo@redhat.com
Cc: Pavel Machek <pavel@ucw.cz>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: xen-devel@lists.xenproject.org
Cc: x86@kernel.org
---
 arch/x86/include/asm/linkage.h |   5 +-
 include/linux/linkage.h        | 163 +++++++++++++++++++++++++++++++++++++++--
 2 files changed, 158 insertions(+), 10 deletions(-)

diff --git a/arch/x86/include/asm/linkage.h b/arch/x86/include/asm/linkage.h
index 0ccb26dda126..de41e51a9ab4 100644
--- a/arch/x86/include/asm/linkage.h
+++ b/arch/x86/include/asm/linkage.h
@@ -12,9 +12,8 @@
 
 #ifdef __ASSEMBLY__
 
-#define GLOBAL(name)	\
-	.globl name;	\
-	name:
+/* deprecated, use SYM_DATA_START, SYM_FUNC_START, or SYM_FUNC_INNER_LABEL */
+#define GLOBAL(name)	SYM_ENTRY(name, SYM_V_GLOBAL, SYM_A_NONE)
 
 #if defined(CONFIG_X86_64) || defined(CONFIG_X86_ALIGNMENT_16)
 #define __ALIGN		.p2align 4, 0x90
diff --git a/include/linux/linkage.h b/include/linux/linkage.h
index a6a42dd02466..31237e9db93f 100644
--- a/include/linux/linkage.h
+++ b/include/linux/linkage.h
@@ -74,25 +74,46 @@
 
 #ifdef __ASSEMBLY__
 
+/* SYM_T_FUNC -- type used by assembler to mark functions */
+#ifndef SYM_T_FUNC
+#define SYM_T_FUNC				STT_FUNC
+#endif
+
+/* SYM_T_OBJECT -- type used by assembler to mark data */
+#ifndef SYM_T_OBJECT
+#define SYM_T_OBJECT				STT_OBJECT
+#endif
+
+/* SYM_A_* -- align the symbol? */
+#define SYM_A_ALIGN				ALIGN
+#define SYM_A_NONE				/* nothing */
+
+/* SYM_V_* -- visibility of symbols */
+#define SYM_V_GLOBAL(name)			.globl name
+#define SYM_V_WEAK(name)			.weak name
+#define SYM_V_LOCAL(name)			/* nothing */
+
 #ifndef LINKER_SCRIPT
 #define ALIGN __ALIGN
 #define ALIGN_STR __ALIGN_STR
 
+/* === DEPRECATED annotations === */
+
 #ifndef ENTRY
+/* deprecated, use SYM_FUNC_START */
 #define ENTRY(name) \
-	.globl name ASM_NL \
-	ALIGN ASM_NL \
-	name:
+	SYM_FUNC_START(name)
 #endif
 #endif /* LINKER_SCRIPT */
 
 #ifndef WEAK
+/* deprecated, use SYM_FUNC_START_WEAK */
 #define WEAK(name)	   \
-	.weak name ASM_NL   \
-	name:
+	SYM_FUNC_START_WEAK(name)
 #endif
 
 #ifndef END
+/* deprecated, use SYM_FUNC_END, SYM_DATA_END, or SYM_END */
 #define END(name) \
 	.size name, .-name
 #endif
@@ -102,11 +123,139 @@
  * static analysis tools such as stack depth analyzer.
  */
 #ifndef ENDPROC
+/* deprecated, use SYM_FUNC_END */
 #define ENDPROC(name) \
-	.type name, @function ASM_NL \
-	END(name)
+	SYM_FUNC_END(name)
+#endif
+
+/* === generic annotations === */
+
+/* SYM_ENTRY -- use only if you have to for non-paired symbols */
+#ifndef SYM_ENTRY
+#define SYM_ENTRY(name, visibility, align...)		\
+	visibility(name) ASM_NL				\
+	align ASM_NL					\
+	name:
+#endif
+
+/* SYM_START -- use only if you have to */
+#ifndef SYM_START
+#define SYM_START(name, visibility, align...)		\
+	SYM_ENTRY(name, visibility, align)
+#endif
+
+/* SYM_END -- use only if you have to */
+#ifndef SYM_END
+#define SYM_END(name, sym_type)				\
+	.type name sym_type ASM_NL			\
+	.size name, .-name
+#endif
+
+/* === code annotations === */
+
+/* SYM_FUNC_START_LOCAL_ALIAS -- use where there are two local names for one code */
+#ifndef SYM_FUNC_START_LOCAL_ALIAS
+#define SYM_FUNC_START_LOCAL_ALIAS(name)		\
+	SYM_START(name, SYM_V_LOCAL, SYM_A_ALIGN)
+#endif
+
+/* SYM_FUNC_START_ALIAS -- use where there are two global names for one code */
+#ifndef SYM_FUNC_START_ALIAS
+#define SYM_FUNC_START_ALIAS(name)			\
+	SYM_START(name, SYM_V_GLOBAL, SYM_A_ALIGN)
+#endif
+
+/* SYM_FUNC_START -- use for global functions */
+#ifndef SYM_FUNC_START
+/*
+ * The same as SYM_FUNC_START_ALIAS, but we will need to distinguish these two
+ * later.
+ */
+#define SYM_FUNC_START(name)				\
+	SYM_START(name, SYM_V_GLOBAL, SYM_A_ALIGN)
+#endif
+
+/* SYM_FUNC_START_LOCAL -- use for local functions */
+#ifndef SYM_FUNC_START_LOCAL
+/* the same as SYM_FUNC_START_LOCAL_ALIAS, see comment near SYM_FUNC_START */
+#define SYM_FUNC_START_LOCAL(name)			\
+	SYM_START(name, SYM_V_LOCAL, SYM_A_ALIGN)
 #endif
 
+/* SYM_FUNC_START_WEAK -- use for weak functions */
+#ifndef SYM_FUNC_START_WEAK
+#define SYM_FUNC_START_WEAK(name)			\
+	SYM_START(name, SYM_V_WEAK, SYM_A_ALIGN)
 #endif
 
+/* SYM_FUNC_END_ALIAS -- the end of LOCAL_ALIASed or ALIASed code */
+#ifndef SYM_FUNC_END_ALIAS
+#define SYM_FUNC_END_ALIAS(name)			\
+	SYM_END(name, SYM_T_FUNC)
+#endif
+
+/*
+ * SYM_FUNC_END -- the end of SYM_FUNC_START_LOCAL, SYM_FUNC_START,
+ * SYM_FUNC_START_WEAK, ...
+ */
+#ifndef SYM_FUNC_END
+/* the same as SYM_FUNC_END_ALIAS, see comment near SYM_FUNC_START */
+#define SYM_FUNC_END(name)				\
+	SYM_END(name, SYM_T_FUNC)
+#endif
+
+/* SYM_FUNC_INNER_LABEL -- only for labels to the middle of functions */
+#ifndef SYM_FUNC_INNER_LABEL
+#define SYM_FUNC_INNER_LABEL(name, visibility)		\
+	.type name SYM_T_FUNC ASM_NL			\
+	SYM_ENTRY(name, visibility, SYM_A_NONE)
+#endif
+
+/* === data annotations === */
+
+/* SYM_DATA_START -- global data symbol */
+#ifndef SYM_DATA_START
+#define SYM_DATA_START(name)				\
+	SYM_START(name, SYM_V_GLOBAL, SYM_A_NONE)
 #endif
+
+/* SYM_DATA_START -- local data symbol */
+#ifndef SYM_DATA_START_LOCAL
+#define SYM_DATA_START_LOCAL(name)			\
+	SYM_START(name, SYM_V_LOCAL, SYM_A_NONE)
+#endif
+
+/* SYM_DATA_END -- the end of SYM_DATA_START symbol */
+#ifndef SYM_DATA_END
+#define SYM_DATA_END(name)				\
+	SYM_END(name, SYM_T_OBJECT)
+#endif
+
+/* SYM_DATA_END_LABEL -- the labeled end of SYM_DATA_START symbol */
+#ifndef SYM_DATA_END_LABEL
+#define SYM_DATA_END_LABEL(name, visibility, label)	\
+	visibility(label) ASM_NL			\
+	.type label SYM_T_OBJECT ASM_NL			\
+	label:						\
+	SYM_END(name, SYM_T_OBJECT)
+#endif
+
+/* SYM_DATA_SIMPLE -- start+end wrapper around simple global data */
+#ifndef SYM_DATA_SIMPLE
+#define SYM_DATA_SIMPLE(name, data)				\
+	SYM_DATA_START(name) ASM_NL				\
+	data ASM_NL						\
+	SYM_DATA_END(name)
+#endif
+
+/* SYM_DATA_SIMPLE_LOCAL -- start+end wrapper around simple local data */
+#ifndef SYM_DATA_SIMPLE_LOCAL
+#define SYM_DATA_SIMPLE_LOCAL(name, data...)			\
+	SYM_DATA_START_LOCAL(name) ASM_NL			\
+	data ASM_NL						\
+	SYM_DATA_END(name)
+#endif
+
+#endif /* __ASSEMBLY__ */
+
+#endif /* _LINUX_LINKAGE_H */
-- 
2.12.2

[PATCH v3 09/29] x86: assembly, use DATA_SIMPLE for data

[Jiri Slaby]

This is a start of series to cleanup macros used for starting functions,
data, globals etc. across x86. When we have all this sorted out, this
will help to inject DWARF unwinding info by objtool later.

The goal is forcing SYM_FUNC_START to emit .cfi_startproc and
SYM_FUNC_END to emit .cfi_endproc. Automatically (by objtool) at best.

This particular patch makes proper use of SYM_DATA_SIMPLE on data
instead of ENTRY. This was not the case on these 4 locations. Rationale:
ENTRY is intended for functions and is to be paired with ENDPROC. It
also aligns symbols which creates unnecessary holes between data.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Cc: Len Brown <len.brown@intel.com>
Cc: Pavel Machek <pavel@ucw.cz>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Cc: linux-pm@vger.kernel.org
---
 arch/x86/kernel/acpi/wakeup_32.S | 10 +++++-----
 arch/x86/kernel/acpi/wakeup_64.S | 14 +++++++-------
 arch/x86/kernel/head_32.S        |  6 ++----
 arch/x86/kernel/head_64.S        |  5 ++---
 4 files changed, 16 insertions(+), 19 deletions(-)

diff --git a/arch/x86/kernel/acpi/wakeup_32.S b/arch/x86/kernel/acpi/wakeup_32.S
index ea9a29e6a96c..11fb24d15f3d 100644
--- a/arch/x86/kernel/acpi/wakeup_32.S
+++ b/arch/x86/kernel/acpi/wakeup_32.S
@@ -91,11 +91,11 @@ ENDPROC(do_suspend_lowlevel)
 
 .data
 ALIGN
-ENTRY(saved_magic)	.long	0
-ENTRY(saved_eip)	.long	0
+SYM_DATA_SIMPLE(saved_magic, .long 0)
+SYM_DATA_SIMPLE_LOCAL(saved_eip, .long 0)
 
 # saved registers
-saved_idt:	.long	0,0
-saved_ldt:	.long	0
-saved_tss:	.long	0
+SYM_DATA_SIMPLE_LOCAL(saved_idt, .long 0,0)
+SYM_DATA_SIMPLE_LOCAL(saved_ldt, .long 0)
+SYM_DATA_SIMPLE_LOCAL(saved_tss, .long 0)
 
diff --git a/arch/x86/kernel/acpi/wakeup_64.S b/arch/x86/kernel/acpi/wakeup_64.S
index 50b8ed0317a3..e53a03648cc7 100644
--- a/arch/x86/kernel/acpi/wakeup_64.S
+++ b/arch/x86/kernel/acpi/wakeup_64.S
@@ -125,12 +125,12 @@ ENTRY(do_suspend_lowlevel)
 ENDPROC(do_suspend_lowlevel)
 
 .data
-ENTRY(saved_rbp)	.quad	0
-ENTRY(saved_rsi)	.quad	0
-ENTRY(saved_rdi)	.quad	0
-ENTRY(saved_rbx)	.quad	0
+SYM_DATA_SIMPLE_LOCAL(saved_rbp, .quad 0)
+SYM_DATA_SIMPLE_LOCAL(saved_rsi, .quad 0)
+SYM_DATA_SIMPLE_LOCAL(saved_rdi, .quad 0)
+SYM_DATA_SIMPLE_LOCAL(saved_rbx, .quad 0)
 
-ENTRY(saved_rip)	.quad	0
-ENTRY(saved_rsp)	.quad	0
+SYM_DATA_SIMPLE_LOCAL(saved_rip, .quad 0)
+SYM_DATA_SIMPLE_LOCAL(saved_rsp, .quad 0)
 
-ENTRY(saved_magic)	.quad	0
+SYM_DATA_SIMPLE(saved_magic, .quad 0)
diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S
index e2acba5b0f2f..b9b6c6ff16a0 100644
--- a/arch/x86/kernel/head_32.S
+++ b/arch/x86/kernel/head_32.S
@@ -548,10 +548,8 @@ GLOBAL(early_recursion_flag)
 
 __REFDATA
 	.align 4
-ENTRY(initial_code)
-	.long i386_start_kernel
-ENTRY(setup_once_ref)
-	.long setup_once
+SYM_DATA_SIMPLE(initial_code, .long i386_start_kernel)
+SYM_DATA_SIMPLE(setup_once_ref, .long setup_once)
 
 /*
  * BSS section
diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
index e1da48448cc0..dace17c6c5fe 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -488,9 +488,8 @@ early_gdt_descr:
 early_gdt_descr_base:
 	.quad	INIT_PER_CPU_VAR(gdt_page)
 
-ENTRY(phys_base)
-	/* This must match the first entry in level2_kernel_pgt */
-	.quad   0x0000000000000000
+/* This must match the first entry in level2_kernel_pgt */
+SYM_DATA_SIMPLE(phys_base, .quad 0x0000000000000000)
 EXPORT_SYMBOL(phys_base)
 
 #include "../../x86/xen/xen-head.S"
-- 
2.12.2

Re: [PATCH v3 09/29] x86: assembly, use DATA_SIMPLE for data

[Pavel Machek]

[-- Attachment #1: Type: text/plain, Size: 716 bytes --]

Hi!

> @@ -91,11 +91,11 @@ ENDPROC(do_suspend_lowlevel)
>  
>  .data
>  ALIGN
> -ENTRY(saved_magic)	.long	0
> -ENTRY(saved_eip)	.long	0
> +SYM_DATA_SIMPLE(saved_magic, .long 0)
> +SYM_DATA_SIMPLE_LOCAL(saved_eip, .long 0)
>  
>  # saved registers
> -saved_idt:	.long	0,0
> -saved_ldt:	.long	0
> -saved_tss:	.long	0
> +SYM_DATA_SIMPLE_LOCAL(saved_idt, .long 0,0)
> +SYM_DATA_SIMPLE_LOCAL(saved_ldt, .long 0)
> +SYM_DATA_SIMPLE_LOCAL(saved_tss, .long 0)

Do we really need to mark local data like that? It complicates the
source code a bit...
								Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]

Re: [PATCH v3 09/29] x86: assembly, use DATA_SIMPLE for data

[Jiri Slaby]

[-- Attachment #1.1: Type: text/plain, Size: 762 bytes --]

Hi,

On 04/27/2017, 01:53 PM, Pavel Machek wrote:
>> @@ -91,11 +91,11 @@ ENDPROC(do_suspend_lowlevel)
>>  
>>  .data
>>  ALIGN
>> -ENTRY(saved_magic)	.long	0
>> -ENTRY(saved_eip)	.long	0
>> +SYM_DATA_SIMPLE(saved_magic, .long 0)
>> +SYM_DATA_SIMPLE_LOCAL(saved_eip, .long 0)
>>  
>>  # saved registers
>> -saved_idt:	.long	0,0
>> -saved_ldt:	.long	0
>> -saved_tss:	.long	0
>> +SYM_DATA_SIMPLE_LOCAL(saved_idt, .long 0,0)
>> +SYM_DATA_SIMPLE_LOCAL(saved_ldt, .long 0)
>> +SYM_DATA_SIMPLE_LOCAL(saved_tss, .long 0)
> 
> Do we really need to mark local data like that? It complicates the
> source code a bit...

Not really -- if the preferred way is not to annotate local symbols, I
can indeed remove that.

thanks,
-- 
js
suse labs


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 854 bytes --]

Re: [PATCH v3 09/29] x86: assembly, use DATA_SIMPLE for data

[Pavel Machek]

[-- Attachment #1: Type: text/plain, Size: 1007 bytes --]

On Thu 2017-04-27 14:30:05, Jiri Slaby wrote:
> Hi,
> 
> On 04/27/2017, 01:53 PM, Pavel Machek wrote:
> >> @@ -91,11 +91,11 @@ ENDPROC(do_suspend_lowlevel)
> >>  
> >>  .data
> >>  ALIGN
> >> -ENTRY(saved_magic)	.long	0
> >> -ENTRY(saved_eip)	.long	0
> >> +SYM_DATA_SIMPLE(saved_magic, .long 0)
> >> +SYM_DATA_SIMPLE_LOCAL(saved_eip, .long 0)
> >>  
> >>  # saved registers
> >> -saved_idt:	.long	0,0
> >> -saved_ldt:	.long	0
> >> -saved_tss:	.long	0
> >> +SYM_DATA_SIMPLE_LOCAL(saved_idt, .long 0,0)
> >> +SYM_DATA_SIMPLE_LOCAL(saved_ldt, .long 0)
> >> +SYM_DATA_SIMPLE_LOCAL(saved_tss, .long 0)
> > 
> > Do we really need to mark local data like that? It complicates the
> > source code a bit...
> 
> Not really -- if the preferred way is not to annotate local symbols, I
> can indeed remove that.

I'd prefer that ;-).
									Pavel

-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]

[PATCH v3 10/29] x86: assembly, annotate relocate_kernel

[Jiri Slaby]

Annotate all functions in relocate_kernel by SYM_FUNC_START,
SYM_FUNC_START_LOCAL, and SYM_FUNC_END properly.

We get in return:
  0000   108 FUNC    GLOBAL DEFAULT    1 relocate_kernel
  0070   165 FUNC    LOCAL  DEFAULT    1 identity_mapped
  0120    53 FUNC    LOCAL  DEFAULT    1 virtual_mapped
  0160   127 FUNC    LOCAL  DEFAULT    1 swap_pages

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
---
 arch/x86/kernel/relocate_kernel_32.S | 13 ++++++++-----
 arch/x86/kernel/relocate_kernel_64.S | 13 ++++++++-----
 2 files changed, 16 insertions(+), 10 deletions(-)

diff --git a/arch/x86/kernel/relocate_kernel_32.S b/arch/x86/kernel/relocate_kernel_32.S
index 77630d57e7bf..4f0f80da0454 100644
--- a/arch/x86/kernel/relocate_kernel_32.S
+++ b/arch/x86/kernel/relocate_kernel_32.S
@@ -37,8 +37,7 @@
 #define CP_PA_BACKUP_PAGES_MAP	DATA(0x1c)
 
 	.text
-	.globl relocate_kernel
-relocate_kernel:
+SYM_FUNC_START(relocate_kernel)
 	/* Save the CPU context, used for jumping back */
 
 	pushl	%ebx
@@ -95,8 +94,9 @@ relocate_kernel:
 	addl    $(identity_mapped - relocate_kernel), %eax
 	pushl   %eax
 	ret
+SYM_FUNC_END(relocate_kernel)
 
-identity_mapped:
+SYM_FUNC_START_LOCAL(identity_mapped)
 	/* set return address to 0 if not preserving context */
 	pushl	$0
 	/* store the start address on the stack */
@@ -193,8 +193,9 @@ identity_mapped:
 	addl	$(virtual_mapped - relocate_kernel), %eax
 	pushl	%eax
 	ret
+SYM_FUNC_END(identity_mapped)
 
-virtual_mapped:
+SYM_FUNC_START_LOCAL(virtual_mapped)
 	movl	CR4(%edi), %eax
 	movl	%eax, %cr4
 	movl	CR3(%edi), %eax
@@ -210,9 +211,10 @@ virtual_mapped:
 	popl	%esi
 	popl	%ebx
 	ret
+SYM_FUNC_END(virtual_mapped)
 
 	/* Do the copies */
-swap_pages:
+SYM_FUNC_START_LOCAL(swap_pages)
 	movl	8(%esp), %edx
 	movl	4(%esp), %ecx
 	pushl	%ebp
@@ -272,6 +274,7 @@ swap_pages:
 	popl	%ebx
 	popl	%ebp
 	ret
+SYM_FUNC_END(swap_pages)
 
 	.globl kexec_control_code_size
 .set kexec_control_code_size, . - relocate_kernel
diff --git a/arch/x86/kernel/relocate_kernel_64.S b/arch/x86/kernel/relocate_kernel_64.S
index 98111b38ebfd..5138b0587816 100644
--- a/arch/x86/kernel/relocate_kernel_64.S
+++ b/arch/x86/kernel/relocate_kernel_64.S
@@ -40,8 +40,7 @@
 	.text
 	.align PAGE_SIZE
 	.code64
-	.globl relocate_kernel
-relocate_kernel:
+SYM_FUNC_START(relocate_kernel)
 	/*
 	 * %rdi indirection_page
 	 * %rsi page_list
@@ -98,8 +97,9 @@ relocate_kernel:
 	addq	$(identity_mapped - relocate_kernel), %r8
 	pushq	%r8
 	ret
+SYM_FUNC_END(relocate_kernel)
 
-identity_mapped:
+SYM_FUNC_START_LOCAL(identity_mapped)
 	/* set return address to 0 if not preserving context */
 	pushq	$0
 	/* store the start address on the stack */
@@ -189,8 +189,9 @@ identity_mapped:
 	movq	$virtual_mapped, %rax
 	pushq	%rax
 	ret
+SYM_FUNC_END(identity_mapped)
 
-virtual_mapped:
+SYM_FUNC_START_LOCAL(virtual_mapped)
 	movq	RSP(%r8), %rsp
 	movq	CR4(%r8), %rax
 	movq	%rax, %cr4
@@ -208,9 +209,10 @@ virtual_mapped:
 	popq	%rbp
 	popq	%rbx
 	ret
+SYM_FUNC_END(virtual_mapped)
 
 	/* Do the copies */
-swap_pages:
+SYM_FUNC_START_LOCAL(swap_pages)
 	movq	%rdi, %rcx 	/* Put the page_list in %rcx */
 	xorl	%edi, %edi
 	xorl	%esi, %esi
@@ -263,6 +265,7 @@ swap_pages:
 	jmp	0b
 3:
 	ret
+SYM_FUNC_END(swap_pages)
 
 	.globl kexec_control_code_size
 .set kexec_control_code_size, . - relocate_kernel
-- 
2.12.2

[PATCH v3 11/29] x86: entry, annotate THUNKs

[Jiri Slaby]

Place SYM_FUNC_START and SYM_FUNC_END around the THUNK macro body, given
it generates functions.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: <x86@kernel.org>
---
 arch/x86/entry/thunk_32.S | 4 ++--
 arch/x86/entry/thunk_64.S | 5 ++---
 2 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/arch/x86/entry/thunk_32.S b/arch/x86/entry/thunk_32.S
index fee6bc79b987..c38852758d41 100644
--- a/arch/x86/entry/thunk_32.S
+++ b/arch/x86/entry/thunk_32.S
@@ -10,8 +10,7 @@
 
 	/* put return address in eax (arg1) */
 	.macro THUNK name, func, put_ret_addr_in_eax=0
-	.globl \name
-\name:
+SYM_FUNC_START(\name)
 	pushl %eax
 	pushl %ecx
 	pushl %edx
@@ -27,6 +26,7 @@
 	popl %eax
 	ret
 	_ASM_NOKPROBE(\name)
+SYM_FUNC_END(\name)
 	.endm
 
 #ifdef CONFIG_TRACE_IRQFLAGS
diff --git a/arch/x86/entry/thunk_64.S b/arch/x86/entry/thunk_64.S
index be36bf4e0957..38b76f5b097d 100644
--- a/arch/x86/entry/thunk_64.S
+++ b/arch/x86/entry/thunk_64.S
@@ -12,9 +12,7 @@
 
 	/* rdi:	arg1 ... normal C conventions. rax is saved/restored. */
 	.macro THUNK name, func, put_ret_addr_in_rdi=0
-	.globl \name
-	.type \name, @function
-\name:
+SYM_FUNC_START(\name)
 	pushq %rbp
 	movq %rsp, %rbp
 
@@ -36,6 +34,7 @@
 	call \func
 	jmp  .L_restore
 	_ASM_NOKPROBE(\name)
+SYM_FUNC_END(\name)
 	.endm
 
 #ifdef CONFIG_TRACE_IRQFLAGS
-- 
2.12.2

[PATCH v3 12/29] x86: assembly, annotate local functions

[Jiri Slaby]

Use the newly added SYM_FUNC_START_LOCAL to annotate starts of all
functions which do not have ".globl" annotation. This is needed to
balance SYM_FUNC_END for tools that are about to generate debuginfo.

Note that some functions already had ENDPROC -- switch to the new
SYM_FUNC_END too.

[v3] annotate more functions

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: <x86@kernel.org>
---
 arch/x86/entry/entry_32.S        |  5 ++---
 arch/x86/entry/entry_64.S        |  3 ++-
 arch/x86/kernel/acpi/wakeup_64.S |  3 ++-
 arch/x86/kernel/head_32.S        | 10 +++++-----
 arch/x86/kernel/head_64.S        |  7 ++++---
 arch/x86/kernel/verify_cpu.S     |  3 ++-
 arch/x86/lib/getuser.S           |  8 ++++----
 arch/x86/lib/putuser.S           |  4 ++--
 arch/x86/net/bpf_jit.S           | 21 ++++++++++++++-------
 9 files changed, 37 insertions(+), 27 deletions(-)

diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S
index afeeb389e9aa..73c34507893b 100644
--- a/arch/x86/entry/entry_32.S
+++ b/arch/x86/entry/entry_32.S
@@ -299,8 +299,7 @@ ENDPROC(ret_from_fork)
  */
 
 	# userspace resumption stub bypassing syscall exit tracing
-	ALIGN
-ret_from_exception:
+SYM_FUNC_START_LOCAL(ret_from_exception)
 	preempt_stop(CLBR_ANY)
 ret_from_intr:
 #ifdef CONFIG_VM86
@@ -323,7 +322,7 @@ ENTRY(resume_userspace)
 	movl	%esp, %eax
 	call	prepare_exit_to_usermode
 	jmp	restore_all
-ENDPROC(ret_from_exception)
+SYM_FUNC_END(ret_from_exception)
 
 #ifdef CONFIG_PREEMPT
 ENTRY(resume_kernel)
diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
index 1fe8758102cb..aff5f8051ce2 100644
--- a/arch/x86/entry/entry_64.S
+++ b/arch/x86/entry/entry_64.S
@@ -876,7 +876,7 @@ EXPORT_SYMBOL(native_load_gs_index)
 	_ASM_EXTABLE(.Lgs_change, bad_gs)
 	.section .fixup, "ax"
 	/* running with kernelgs */
-bad_gs:
+SYM_FUNC_START_LOCAL(bad_gs)
 	SWAPGS					/* switch back to user gs */
 .macro ZAP_GS
 	/* This can't be a string because the preprocessor needs to see it. */
@@ -887,6 +887,7 @@ bad_gs:
 	xorl	%eax, %eax
 	movl	%eax, %gs
 	jmp	2b
+SYM_FUNC_END(bad_gs)
 	.previous
 
 /* Call softirq on interrupt stack. Interrupts are off. */
diff --git a/arch/x86/kernel/acpi/wakeup_64.S b/arch/x86/kernel/acpi/wakeup_64.S
index e53a03648cc7..8fca92dd9144 100644
--- a/arch/x86/kernel/acpi/wakeup_64.S
+++ b/arch/x86/kernel/acpi/wakeup_64.S
@@ -36,8 +36,9 @@ ENTRY(wakeup_long64)
 	jmp	*%rax
 ENDPROC(wakeup_long64)
 
-bogus_64_magic:
+SYM_FUNC_START_LOCAL(bogus_64_magic)
 	jmp	bogus_64_magic
+SYM_FUNC_END(bogus_64_magic)
 
 ENTRY(do_suspend_lowlevel)
 	FRAME_BEGIN
diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S
index b9b6c6ff16a0..00193c67845c 100644
--- a/arch/x86/kernel/head_32.S
+++ b/arch/x86/kernel/head_32.S
@@ -446,7 +446,7 @@ ENTRY(early_idt_handler_array)
 	.endr
 ENDPROC(early_idt_handler_array)
 	
-early_idt_handler_common:
+SYM_FUNC_START_LOCAL(early_idt_handler_common)
 	/*
 	 * The stack is the hardware frame, an error code or zero, and the
 	 * vector number.
@@ -501,11 +501,10 @@ early_idt_handler_common:
 	decl	%ss:early_recursion_flag
 	addl	$4, %esp	/* pop pt_regs->orig_ax */
 	iret
-ENDPROC(early_idt_handler_common)
+SYM_FUNC_END(early_idt_handler_common)
 
 /* This is the default interrupt "handler" :-) */
-	ALIGN
-ignore_int:
+SYM_FUNC_START_LOCAL(ignore_int)
 	cld
 #ifdef CONFIG_PRINTK
 	pushl %eax
@@ -540,7 +539,8 @@ ignore_int:
 hlt_loop:
 	hlt
 	jmp hlt_loop
-ENDPROC(ignore_int)
+SYM_FUNC_END(ignore_int)
+
 __INITDATA
 	.align 4
 GLOBAL(early_recursion_flag)
diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
index dace17c6c5fe..a2230c35fcf9 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -335,8 +335,9 @@ ENDPROC(start_cpu0)
 	.quad  init_thread_union + THREAD_SIZE - SIZEOF_PTREGS
 	__FINITDATA
 
-bad_address:
+SYM_FUNC_START_LOCAL(bad_address)
 	jmp bad_address
+SYM_FUNC_END(bad_address)
 
 	__INIT
 ENTRY(early_idt_handler_array)
@@ -356,7 +357,7 @@ ENTRY(early_idt_handler_array)
 	.endr
 ENDPROC(early_idt_handler_array)
 
-early_idt_handler_common:
+SYM_FUNC_START_LOCAL(early_idt_handler_common)
 	/*
 	 * The stack is the hardware frame, an error code or zero, and the
 	 * vector number.
@@ -397,7 +398,7 @@ early_idt_handler_common:
 20:
 	decl early_recursion_flag(%rip)
 	jmp restore_regs_and_iret
-ENDPROC(early_idt_handler_common)
+SYM_FUNC_END(early_idt_handler_common)
 
 	__INITDATA
 
diff --git a/arch/x86/kernel/verify_cpu.S b/arch/x86/kernel/verify_cpu.S
index 014ea59aa153..fd60f1ac5fec 100644
--- a/arch/x86/kernel/verify_cpu.S
+++ b/arch/x86/kernel/verify_cpu.S
@@ -33,7 +33,7 @@
 #include <asm/cpufeatures.h>
 #include <asm/msr-index.h>
 
-verify_cpu:
+SYM_FUNC_START_LOCAL(verify_cpu)
 	pushf				# Save caller passed flags
 	push	$0			# Kill any dangerous flags
 	popf
@@ -139,3 +139,4 @@ verify_cpu:
 	popf				# Restore caller passed flags
 	xorl %eax, %eax
 	ret
+SYM_FUNC_END(verify_cpu)
diff --git a/arch/x86/lib/getuser.S b/arch/x86/lib/getuser.S
index 37b62d412148..29f0707a3913 100644
--- a/arch/x86/lib/getuser.S
+++ b/arch/x86/lib/getuser.S
@@ -104,21 +104,21 @@ ENDPROC(__get_user_8)
 EXPORT_SYMBOL(__get_user_8)
 
 
-bad_get_user:
+SYM_FUNC_START_LOCAL(bad_get_user)
 	xor %edx,%edx
 	mov $(-EFAULT),%_ASM_AX
 	ASM_CLAC
 	ret
-END(bad_get_user)
+SYM_FUNC_END(bad_get_user)
 
 #ifdef CONFIG_X86_32
-bad_get_user_8:
+SYM_FUNC_START_LOCAL(bad_get_user_8)
 	xor %edx,%edx
 	xor %ecx,%ecx
 	mov $(-EFAULT),%_ASM_AX
 	ASM_CLAC
 	ret
-END(bad_get_user_8)
+SYM_FUNC_END(bad_get_user_8)
 #endif
 
 	_ASM_EXTABLE(1b,bad_get_user)
diff --git a/arch/x86/lib/putuser.S b/arch/x86/lib/putuser.S
index cd5d716d2897..d77883f36875 100644
--- a/arch/x86/lib/putuser.S
+++ b/arch/x86/lib/putuser.S
@@ -88,10 +88,10 @@ ENTRY(__put_user_8)
 ENDPROC(__put_user_8)
 EXPORT_SYMBOL(__put_user_8)
 
-bad_put_user:
+SYM_FUNC_START_LOCAL(bad_put_user)
 	movl $-EFAULT,%eax
 	EXIT
-END(bad_put_user)
+SYM_FUNC_END(bad_put_user)
 
 	_ASM_EXTABLE(1b,bad_put_user)
 	_ASM_EXTABLE(2b,bad_put_user)
diff --git a/arch/x86/net/bpf_jit.S b/arch/x86/net/bpf_jit.S
index 762c29fb8832..823edd6f1db7 100644
--- a/arch/x86/net/bpf_jit.S
+++ b/arch/x86/net/bpf_jit.S
@@ -81,26 +81,29 @@ ENDPROC(sk_load_byte_positive_offset)
 	FRAME_END
 
 
-bpf_slow_path_word:
+SYM_FUNC_START_LOCAL(bpf_slow_path_word)
 	bpf_slow_path_common(4)
 	js	bpf_error
 	mov	- MAX_BPF_STACK + 32(%rbp),%eax
 	bswap	%eax
 	ret
+SYM_FUNC_END(bpf_slow_path_word)
 
-bpf_slow_path_half:
+SYM_FUNC_START_LOCAL(bpf_slow_path_half)
 	bpf_slow_path_common(2)
 	js	bpf_error
 	mov	- MAX_BPF_STACK + 32(%rbp),%ax
 	rol	$8,%ax
 	movzwl	%ax,%eax
 	ret
+SYM_FUNC_END(bpf_slow_path_half)
 
-bpf_slow_path_byte:
+SYM_FUNC_START_LOCAL(bpf_slow_path_byte)
 	bpf_slow_path_common(1)
 	js	bpf_error
 	movzbl	- MAX_BPF_STACK + 32(%rbp),%eax
 	ret
+SYM_FUNC_END(bpf_slow_path_byte)
 
 #define sk_negative_common(SIZE)				\
 	FRAME_BEGIN;						\
@@ -116,9 +119,10 @@ bpf_slow_path_byte:
 	FRAME_END;						\
 	jz	bpf_error
 
-bpf_slow_path_word_neg:
+SYM_FUNC_START_LOCAL(bpf_slow_path_word_neg)
 	cmp	SKF_MAX_NEG_OFF, %esi	/* test range */
 	jl	bpf_error	/* offset lower -> error  */
+SYM_FUNC_END(bpf_slow_path_word_neg)
 
 ENTRY(sk_load_word_negative_offset)
 	sk_negative_common(4)
@@ -127,9 +131,10 @@ ENTRY(sk_load_word_negative_offset)
 	ret
 ENDPROC(sk_load_word_negative_offset)
 
-bpf_slow_path_half_neg:
+SYM_FUNC_START_LOCAL(bpf_slow_path_half_neg)
 	cmp	SKF_MAX_NEG_OFF, %esi
 	jl	bpf_error
+SYM_FUNC_END(bpf_slow_path_half_neg)
 
 ENTRY(sk_load_half_negative_offset)
 	sk_negative_common(2)
@@ -139,9 +144,10 @@ ENTRY(sk_load_half_negative_offset)
 	ret
 ENDPROC(sk_load_half_negative_offset)
 
-bpf_slow_path_byte_neg:
+SYM_FUNC_START_LOCAL(bpf_slow_path_byte_neg)
 	cmp	SKF_MAX_NEG_OFF, %esi
 	jl	bpf_error
+SYM_FUNC_END(bpf_slow_path_byte_neg)
 
 ENTRY(sk_load_byte_negative_offset)
 	sk_negative_common(1)
@@ -149,7 +155,7 @@ ENTRY(sk_load_byte_negative_offset)
 	ret
 ENDPROC(sk_load_byte_negative_offset)
 
-bpf_error:
+SYM_FUNC_START_LOCAL(bpf_error)
 # force a return 0 from jit handler
 	xor	%eax,%eax
 	mov	- MAX_BPF_STACK(%rbp),%rbx
@@ -158,3 +164,4 @@ bpf_error:
 	mov	- MAX_BPF_STACK + 24(%rbp),%r15
 	leaveq
 	ret
+SYM_FUNC_END(bpf_error)
-- 
2.12.2

[PATCH v3 13/29] x86: crypto, annotate local functions

[Jiri Slaby]

Use the newly added SYM_FUNC_START_LOCAL to annotate starts of all
functions which do not have ".globl" annotation, but their ends are
annotated by ENDPROC. This is needed to balance ENDPROC for tools that
are about to generate debuginfo.

We also convert their ENDPROCs to the new SYM_FUNC_END.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: <x86@kernel.org>
Cc: <linux-crypto@vger.kernel.org>
---
 arch/x86/crypto/aesni-intel_asm.S            | 49 ++++++++++++----------------
 arch/x86/crypto/camellia-aesni-avx-asm_64.S  | 20 ++++++------
 arch/x86/crypto/camellia-aesni-avx2-asm_64.S | 20 ++++++------
 arch/x86/crypto/cast5-avx-x86_64-asm_64.S    |  8 ++---
 arch/x86/crypto/cast6-avx-x86_64-asm_64.S    |  8 ++---
 arch/x86/crypto/ghash-clmulni-intel_asm.S    |  4 +--
 arch/x86/crypto/serpent-avx-x86_64-asm_64.S  |  8 ++---
 arch/x86/crypto/serpent-avx2-asm_64.S        |  8 ++---
 arch/x86/crypto/twofish-avx-x86_64-asm_64.S  |  8 ++---
 9 files changed, 62 insertions(+), 71 deletions(-)

diff --git a/arch/x86/crypto/aesni-intel_asm.S b/arch/x86/crypto/aesni-intel_asm.S
index 3c465184ff8a..da76ae01e791 100644
--- a/arch/x86/crypto/aesni-intel_asm.S
+++ b/arch/x86/crypto/aesni-intel_asm.S
@@ -1746,7 +1746,7 @@ ENDPROC(aesni_gcm_enc)
 
 .align 4
 _key_expansion_128:
-_key_expansion_256a:
+SYM_FUNC_START_LOCAL(_key_expansion_256a)
 	pshufd $0b11111111, %xmm1, %xmm1
 	shufps $0b00010000, %xmm0, %xmm4
 	pxor %xmm4, %xmm0
@@ -1757,10 +1757,9 @@ _key_expansion_256a:
 	add $0x10, TKEYP
 	ret
 ENDPROC(_key_expansion_128)
-ENDPROC(_key_expansion_256a)
+SYM_FUNC_END(_key_expansion_256a)
 
-.align 4
-_key_expansion_192a:
+SYM_FUNC_START_LOCAL(_key_expansion_192a)
 	pshufd $0b01010101, %xmm1, %xmm1
 	shufps $0b00010000, %xmm0, %xmm4
 	pxor %xmm4, %xmm0
@@ -1782,10 +1781,9 @@ _key_expansion_192a:
 	movaps %xmm1, 0x10(TKEYP)
 	add $0x20, TKEYP
 	ret
-ENDPROC(_key_expansion_192a)
+SYM_FUNC_END(_key_expansion_192a)
 
-.align 4
-_key_expansion_192b:
+SYM_FUNC_START_LOCAL(_key_expansion_192b)
 	pshufd $0b01010101, %xmm1, %xmm1
 	shufps $0b00010000, %xmm0, %xmm4
 	pxor %xmm4, %xmm0
@@ -1802,10 +1800,9 @@ _key_expansion_192b:
 	movaps %xmm0, (TKEYP)
 	add $0x10, TKEYP
 	ret
-ENDPROC(_key_expansion_192b)
+SYM_FUNC_END(_key_expansion_192b)
 
-.align 4
-_key_expansion_256b:
+SYM_FUNC_START_LOCAL(_key_expansion_256b)
 	pshufd $0b10101010, %xmm1, %xmm1
 	shufps $0b00010000, %xmm2, %xmm4
 	pxor %xmm4, %xmm2
@@ -1815,7 +1812,7 @@ _key_expansion_256b:
 	movaps %xmm2, (TKEYP)
 	add $0x10, TKEYP
 	ret
-ENDPROC(_key_expansion_256b)
+SYM_FUNC_END(_key_expansion_256b)
 
 /*
  * int aesni_set_key(struct crypto_aes_ctx *ctx, const u8 *in_key,
@@ -1968,8 +1965,7 @@ ENDPROC(aesni_enc)
  *	KEY
  *	TKEYP (T1)
  */
-.align 4
-_aesni_enc1:
+SYM_FUNC_START_LOCAL(_aesni_enc1)
 	movaps (KEYP), KEY		# key
 	mov KEYP, TKEYP
 	pxor KEY, STATE		# round 0
@@ -2012,7 +2008,7 @@ _aesni_enc1:
 	movaps 0x70(TKEYP), KEY
 	AESENCLAST KEY STATE
 	ret
-ENDPROC(_aesni_enc1)
+SYM_FUNC_END(_aesni_enc1)
 
 /*
  * _aesni_enc4:	internal ABI
@@ -2032,8 +2028,7 @@ ENDPROC(_aesni_enc1)
  *	KEY
  *	TKEYP (T1)
  */
-.align 4
-_aesni_enc4:
+SYM_FUNC_START_LOCAL(_aesni_enc4)
 	movaps (KEYP), KEY		# key
 	mov KEYP, TKEYP
 	pxor KEY, STATE1		# round 0
@@ -2121,7 +2116,7 @@ _aesni_enc4:
 	AESENCLAST KEY STATE3
 	AESENCLAST KEY STATE4
 	ret
-ENDPROC(_aesni_enc4)
+SYM_FUNC_END(_aesni_enc4)
 
 /*
  * void aesni_dec (struct crypto_aes_ctx *ctx, u8 *dst, const u8 *src)
@@ -2160,8 +2155,7 @@ ENDPROC(aesni_dec)
  *	KEY
  *	TKEYP (T1)
  */
-.align 4
-_aesni_dec1:
+SYM_FUNC_START_LOCAL(_aesni_dec1)
 	movaps (KEYP), KEY		# key
 	mov KEYP, TKEYP
 	pxor KEY, STATE		# round 0
@@ -2204,7 +2198,7 @@ _aesni_dec1:
 	movaps 0x70(TKEYP), KEY
 	AESDECLAST KEY STATE
 	ret
-ENDPROC(_aesni_dec1)
+SYM_FUNC_END(_aesni_dec1)
 
 /*
  * _aesni_dec4:	internal ABI
@@ -2224,8 +2218,7 @@ ENDPROC(_aesni_dec1)
  *	KEY
  *	TKEYP (T1)
  */
-.align 4
-_aesni_dec4:
+SYM_FUNC_START_LOCAL(_aesni_dec4)
 	movaps (KEYP), KEY		# key
 	mov KEYP, TKEYP
 	pxor KEY, STATE1		# round 0
@@ -2313,7 +2306,7 @@ _aesni_dec4:
 	AESDECLAST KEY STATE3
 	AESDECLAST KEY STATE4
 	ret
-ENDPROC(_aesni_dec4)
+SYM_FUNC_END(_aesni_dec4)
 
 /*
  * void aesni_ecb_enc(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
@@ -2591,8 +2584,7 @@ ENDPROC(aesni_cbc_dec)
  *	INC:	== 1, in little endian
  *	BSWAP_MASK == endian swapping mask
  */
-.align 4
-_aesni_inc_init:
+SYM_FUNC_START_LOCAL(_aesni_inc_init)
 	movaps .Lbswap_mask, BSWAP_MASK
 	movaps IV, CTR
 	PSHUFB_XMM BSWAP_MASK CTR
@@ -2600,7 +2592,7 @@ _aesni_inc_init:
 	MOVQ_R64_XMM TCTR_LOW INC
 	MOVQ_R64_XMM CTR TCTR_LOW
 	ret
-ENDPROC(_aesni_inc_init)
+SYM_FUNC_END(_aesni_inc_init)
 
 /*
  * _aesni_inc:		internal ABI
@@ -2617,8 +2609,7 @@ ENDPROC(_aesni_inc_init)
  *	CTR:	== output IV, in little endian
  *	TCTR_LOW: == lower qword of CTR
  */
-.align 4
-_aesni_inc:
+SYM_FUNC_START_LOCAL(_aesni_inc)
 	paddq INC, CTR
 	add $1, TCTR_LOW
 	jnc .Linc_low
@@ -2629,7 +2620,7 @@ _aesni_inc:
 	movaps CTR, IV
 	PSHUFB_XMM BSWAP_MASK IV
 	ret
-ENDPROC(_aesni_inc)
+SYM_FUNC_END(_aesni_inc)
 
 /*
  * void aesni_ctr_enc(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
diff --git a/arch/x86/crypto/camellia-aesni-avx-asm_64.S b/arch/x86/crypto/camellia-aesni-avx-asm_64.S
index f7c495e2863c..8b6a65524067 100644
--- a/arch/x86/crypto/camellia-aesni-avx-asm_64.S
+++ b/arch/x86/crypto/camellia-aesni-avx-asm_64.S
@@ -188,20 +188,20 @@
  * larger and would only be 0.5% faster (on sandy-bridge).
  */
 .align 8
-roundsm16_x0_x1_x2_x3_x4_x5_x6_x7_y0_y1_y2_y3_y4_y5_y6_y7_cd:
+SYM_FUNC_START_LOCAL(roundsm16_x0_x1_x2_x3_x4_x5_x6_x7_y0_y1_y2_y3_y4_y5_y6_y7_cd)
 	roundsm16(%xmm0, %xmm1, %xmm2, %xmm3, %xmm4, %xmm5, %xmm6, %xmm7,
 		  %xmm8, %xmm9, %xmm10, %xmm11, %xmm12, %xmm13, %xmm14, %xmm15,
 		  %rcx, (%r9));
 	ret;
-ENDPROC(roundsm16_x0_x1_x2_x3_x4_x5_x6_x7_y0_y1_y2_y3_y4_y5_y6_y7_cd)
+SYM_FUNC_END(roundsm16_x0_x1_x2_x3_x4_x5_x6_x7_y0_y1_y2_y3_y4_y5_y6_y7_cd)
 
 .align 8
-roundsm16_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab:
+SYM_FUNC_START_LOCAL(roundsm16_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab)
 	roundsm16(%xmm4, %xmm5, %xmm6, %xmm7, %xmm0, %xmm1, %xmm2, %xmm3,
 		  %xmm12, %xmm13, %xmm14, %xmm15, %xmm8, %xmm9, %xmm10, %xmm11,
 		  %rax, (%r9));
 	ret;
-ENDPROC(roundsm16_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab)
+SYM_FUNC_END(roundsm16_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab)
 
 /*
  * IN/OUT:
@@ -721,7 +721,7 @@ ENDPROC(roundsm16_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab)
 .text
 
 .align 8
-__camellia_enc_blk16:
+SYM_FUNC_START_LOCAL(__camellia_enc_blk16)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rax: temporary storage, 256 bytes
@@ -805,10 +805,10 @@ __camellia_enc_blk16:
 		     %xmm15, %rax, %rcx, 24);
 
 	jmp .Lenc_done;
-ENDPROC(__camellia_enc_blk16)
+SYM_FUNC_END(__camellia_enc_blk16)
 
 .align 8
-__camellia_dec_blk16:
+SYM_FUNC_START_LOCAL(__camellia_dec_blk16)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rax: temporary storage, 256 bytes
@@ -890,7 +890,7 @@ __camellia_dec_blk16:
 	      ((key_table + (24) * 8) + 4)(CTX));
 
 	jmp .Ldec_max24;
-ENDPROC(__camellia_dec_blk16)
+SYM_FUNC_END(__camellia_dec_blk16)
 
 ENTRY(camellia_ecb_enc_16way)
 	/* input:
@@ -1119,7 +1119,7 @@ ENDPROC(camellia_ctr_16way)
 	vpxor tmp, iv, iv;
 
 .align 8
-camellia_xts_crypt_16way:
+SYM_FUNC_START_LOCAL(camellia_xts_crypt_16way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst (16 blocks)
@@ -1253,7 +1253,7 @@ camellia_xts_crypt_16way:
 
 	FRAME_END
 	ret;
-ENDPROC(camellia_xts_crypt_16way)
+SYM_FUNC_END(camellia_xts_crypt_16way)
 
 ENTRY(camellia_xts_enc_16way)
 	/* input:
diff --git a/arch/x86/crypto/camellia-aesni-avx2-asm_64.S b/arch/x86/crypto/camellia-aesni-avx2-asm_64.S
index eee5b3982cfd..96b44ad85c59 100644
--- a/arch/x86/crypto/camellia-aesni-avx2-asm_64.S
+++ b/arch/x86/crypto/camellia-aesni-avx2-asm_64.S
@@ -227,20 +227,20 @@
  * larger and would only marginally faster.
  */
 .align 8
-roundsm32_x0_x1_x2_x3_x4_x5_x6_x7_y0_y1_y2_y3_y4_y5_y6_y7_cd:
+SYM_FUNC_START_LOCAL(roundsm32_x0_x1_x2_x3_x4_x5_x6_x7_y0_y1_y2_y3_y4_y5_y6_y7_cd)
 	roundsm32(%ymm0, %ymm1, %ymm2, %ymm3, %ymm4, %ymm5, %ymm6, %ymm7,
 		  %ymm8, %ymm9, %ymm10, %ymm11, %ymm12, %ymm13, %ymm14, %ymm15,
 		  %rcx, (%r9));
 	ret;
-ENDPROC(roundsm32_x0_x1_x2_x3_x4_x5_x6_x7_y0_y1_y2_y3_y4_y5_y6_y7_cd)
+SYM_FUNC_END(roundsm32_x0_x1_x2_x3_x4_x5_x6_x7_y0_y1_y2_y3_y4_y5_y6_y7_cd)
 
 .align 8
-roundsm32_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab:
+SYM_FUNC_START_LOCAL(roundsm32_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab)
 	roundsm32(%ymm4, %ymm5, %ymm6, %ymm7, %ymm0, %ymm1, %ymm2, %ymm3,
 		  %ymm12, %ymm13, %ymm14, %ymm15, %ymm8, %ymm9, %ymm10, %ymm11,
 		  %rax, (%r9));
 	ret;
-ENDPROC(roundsm32_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab)
+SYM_FUNC_END(roundsm32_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab)
 
 /*
  * IN/OUT:
@@ -764,7 +764,7 @@ ENDPROC(roundsm32_x4_x5_x6_x7_x0_x1_x2_x3_y4_y5_y6_y7_y0_y1_y2_y3_ab)
 .text
 
 .align 8
-__camellia_enc_blk32:
+SYM_FUNC_START_LOCAL(__camellia_enc_blk32)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rax: temporary storage, 512 bytes
@@ -848,10 +848,10 @@ __camellia_enc_blk32:
 		     %ymm15, %rax, %rcx, 24);
 
 	jmp .Lenc_done;
-ENDPROC(__camellia_enc_blk32)
+SYM_FUNC_END(__camellia_enc_blk32)
 
 .align 8
-__camellia_dec_blk32:
+SYM_FUNC_START_LOCAL(__camellia_dec_blk32)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rax: temporary storage, 512 bytes
@@ -933,7 +933,7 @@ __camellia_dec_blk32:
 	      ((key_table + (24) * 8) + 4)(CTX));
 
 	jmp .Ldec_max24;
-ENDPROC(__camellia_dec_blk32)
+SYM_FUNC_END(__camellia_dec_blk32)
 
 ENTRY(camellia_ecb_enc_32way)
 	/* input:
@@ -1226,7 +1226,7 @@ ENDPROC(camellia_ctr_32way)
 	vpxor tmp1, iv, iv;
 
 .align 8
-camellia_xts_crypt_32way:
+SYM_FUNC_START_LOCAL(camellia_xts_crypt_32way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst (32 blocks)
@@ -1371,7 +1371,7 @@ camellia_xts_crypt_32way:
 
 	FRAME_END
 	ret;
-ENDPROC(camellia_xts_crypt_32way)
+SYM_FUNC_END(camellia_xts_crypt_32way)
 
 ENTRY(camellia_xts_enc_32way)
 	/* input:
diff --git a/arch/x86/crypto/cast5-avx-x86_64-asm_64.S b/arch/x86/crypto/cast5-avx-x86_64-asm_64.S
index b4a8806234ea..0fe153a87d90 100644
--- a/arch/x86/crypto/cast5-avx-x86_64-asm_64.S
+++ b/arch/x86/crypto/cast5-avx-x86_64-asm_64.S
@@ -224,7 +224,7 @@
 .text
 
 .align 16
-__cast5_enc_blk16:
+SYM_FUNC_START_LOCAL(__cast5_enc_blk16)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	RL1: blocks 1 and 2
@@ -293,10 +293,10 @@ __cast5_enc_blk16:
 	outunpack_blocks(RR4, RL4, RTMP, RX, RKM);
 
 	ret;
-ENDPROC(__cast5_enc_blk16)
+SYM_FUNC_END(__cast5_enc_blk16)
 
 .align 16
-__cast5_dec_blk16:
+SYM_FUNC_START_LOCAL(__cast5_dec_blk16)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	RL1: encrypted blocks 1 and 2
@@ -368,7 +368,7 @@ __cast5_dec_blk16:
 .L__skip_dec:
 	vpsrldq $4, RKR, RKR;
 	jmp .L__dec_tail;
-ENDPROC(__cast5_dec_blk16)
+SYM_FUNC_END(__cast5_dec_blk16)
 
 ENTRY(cast5_ecb_enc_16way)
 	/* input:
diff --git a/arch/x86/crypto/cast6-avx-x86_64-asm_64.S b/arch/x86/crypto/cast6-avx-x86_64-asm_64.S
index 952d3156a933..0d71989fff90 100644
--- a/arch/x86/crypto/cast6-avx-x86_64-asm_64.S
+++ b/arch/x86/crypto/cast6-avx-x86_64-asm_64.S
@@ -262,7 +262,7 @@
 .text
 
 .align 8
-__cast6_enc_blk8:
+SYM_FUNC_START_LOCAL(__cast6_enc_blk8)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	RA1, RB1, RC1, RD1, RA2, RB2, RC2, RD2: blocks
@@ -305,10 +305,10 @@ __cast6_enc_blk8:
 	outunpack_blocks(RA2, RB2, RC2, RD2, RTMP, RX, RKRF, RKM);
 
 	ret;
-ENDPROC(__cast6_enc_blk8)
+SYM_FUNC_END(__cast6_enc_blk8)
 
 .align 8
-__cast6_dec_blk8:
+SYM_FUNC_START_LOCAL(__cast6_dec_blk8)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	RA1, RB1, RC1, RD1, RA2, RB2, RC2, RD2: encrypted blocks
@@ -350,7 +350,7 @@ __cast6_dec_blk8:
 	outunpack_blocks(RA2, RB2, RC2, RD2, RTMP, RX, RKRF, RKM);
 
 	ret;
-ENDPROC(__cast6_dec_blk8)
+SYM_FUNC_END(__cast6_dec_blk8)
 
 ENTRY(cast6_ecb_enc_8way)
 	/* input:
diff --git a/arch/x86/crypto/ghash-clmulni-intel_asm.S b/arch/x86/crypto/ghash-clmulni-intel_asm.S
index f94375a8dcd1..c3db86842578 100644
--- a/arch/x86/crypto/ghash-clmulni-intel_asm.S
+++ b/arch/x86/crypto/ghash-clmulni-intel_asm.S
@@ -47,7 +47,7 @@
  *	T2
  *	T3
  */
-__clmul_gf128mul_ble:
+SYM_FUNC_START_LOCAL(__clmul_gf128mul_ble)
 	movaps DATA, T1
 	pshufd $0b01001110, DATA, T2
 	pshufd $0b01001110, SHASH, T3
@@ -90,7 +90,7 @@ __clmul_gf128mul_ble:
 	pxor T2, T1
 	pxor T1, DATA
 	ret
-ENDPROC(__clmul_gf128mul_ble)
+SYM_FUNC_END(__clmul_gf128mul_ble)
 
 /* void clmul_ghash_mul(char *dst, const u128 *shash) */
 ENTRY(clmul_ghash_mul)
diff --git a/arch/x86/crypto/serpent-avx-x86_64-asm_64.S b/arch/x86/crypto/serpent-avx-x86_64-asm_64.S
index 2925077f8c6a..c2d4a1fc9ee8 100644
--- a/arch/x86/crypto/serpent-avx-x86_64-asm_64.S
+++ b/arch/x86/crypto/serpent-avx-x86_64-asm_64.S
@@ -570,7 +570,7 @@
 	transpose_4x4(x0, x1, x2, x3, t0, t1, t2)
 
 .align 8
-__serpent_enc_blk8_avx:
+SYM_FUNC_START_LOCAL(__serpent_enc_blk8_avx)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	RA1, RB1, RC1, RD1, RA2, RB2, RC2, RD2: blocks
@@ -621,10 +621,10 @@ __serpent_enc_blk8_avx:
 	write_blocks(RA2, RB2, RC2, RD2, RK0, RK1, RK2);
 
 	ret;
-ENDPROC(__serpent_enc_blk8_avx)
+SYM_FUNC_END(__serpent_enc_blk8_avx)
 
 .align 8
-__serpent_dec_blk8_avx:
+SYM_FUNC_START_LOCAL(__serpent_dec_blk8_avx)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	RA1, RB1, RC1, RD1, RA2, RB2, RC2, RD2: encrypted blocks
@@ -675,7 +675,7 @@ __serpent_dec_blk8_avx:
 	write_blocks(RC2, RD2, RB2, RE2, RK0, RK1, RK2);
 
 	ret;
-ENDPROC(__serpent_dec_blk8_avx)
+SYM_FUNC_END(__serpent_dec_blk8_avx)
 
 ENTRY(serpent_ecb_enc_8way_avx)
 	/* input:
diff --git a/arch/x86/crypto/serpent-avx2-asm_64.S b/arch/x86/crypto/serpent-avx2-asm_64.S
index d67888f2a52a..52c527ce4b18 100644
--- a/arch/x86/crypto/serpent-avx2-asm_64.S
+++ b/arch/x86/crypto/serpent-avx2-asm_64.S
@@ -566,7 +566,7 @@
 	transpose_4x4(x0, x1, x2, x3, t0, t1, t2)
 
 .align 8
-__serpent_enc_blk16:
+SYM_FUNC_START_LOCAL(__serpent_enc_blk16)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	RA1, RB1, RC1, RD1, RA2, RB2, RC2, RD2: plaintext
@@ -617,10 +617,10 @@ __serpent_enc_blk16:
 	write_blocks(RA2, RB2, RC2, RD2, RK0, RK1, RK2);
 
 	ret;
-ENDPROC(__serpent_enc_blk16)
+SYM_FUNC_END(__serpent_enc_blk16)
 
 .align 8
-__serpent_dec_blk16:
+SYM_FUNC_START_LOCAL(__serpent_dec_blk16)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	RA1, RB1, RC1, RD1, RA2, RB2, RC2, RD2: ciphertext
@@ -671,7 +671,7 @@ __serpent_dec_blk16:
 	write_blocks(RC2, RD2, RB2, RE2, RK0, RK1, RK2);
 
 	ret;
-ENDPROC(__serpent_dec_blk16)
+SYM_FUNC_END(__serpent_dec_blk16)
 
 ENTRY(serpent_ecb_enc_16way)
 	/* input:
diff --git a/arch/x86/crypto/twofish-avx-x86_64-asm_64.S b/arch/x86/crypto/twofish-avx-x86_64-asm_64.S
index b3f49d286348..1e87dcde342f 100644
--- a/arch/x86/crypto/twofish-avx-x86_64-asm_64.S
+++ b/arch/x86/crypto/twofish-avx-x86_64-asm_64.S
@@ -249,7 +249,7 @@
 	vpxor		x3, wkey, x3;
 
 .align 8
-__twofish_enc_blk8:
+SYM_FUNC_START_LOCAL(__twofish_enc_blk8)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	RA1, RB1, RC1, RD1, RA2, RB2, RC2, RD2: blocks
@@ -288,10 +288,10 @@ __twofish_enc_blk8:
 	outunpack_blocks(RC2, RD2, RA2, RB2, RK1, RX0, RY0, RK2);
 
 	ret;
-ENDPROC(__twofish_enc_blk8)
+SYM_FUNC_END(__twofish_enc_blk8)
 
 .align 8
-__twofish_dec_blk8:
+SYM_FUNC_START_LOCAL(__twofish_dec_blk8)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	RC1, RD1, RA1, RB1, RC2, RD2, RA2, RB2: encrypted blocks
@@ -328,7 +328,7 @@ __twofish_dec_blk8:
 	outunpack_blocks(RA2, RB2, RC2, RD2, RK1, RX0, RY0, RK2);
 
 	ret;
-ENDPROC(__twofish_dec_blk8)
+SYM_FUNC_END(__twofish_dec_blk8)
 
 ENTRY(twofish_ecb_enc_8way)
 	/* input:
-- 
2.12.2

[PATCH v3 14/29] x86: boot, annotate local functions

[Jiri Slaby]

relocated and no_longmode are self-standing local functions, annotate
them as such.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: x86@kernel.org
---
 arch/x86/boot/compressed/head_32.S | 3 ++-
 arch/x86/boot/compressed/head_64.S | 7 +++++--
 2 files changed, 7 insertions(+), 3 deletions(-)

diff --git a/arch/x86/boot/compressed/head_32.S b/arch/x86/boot/compressed/head_32.S
index 11c68cf53d4e..d832ddb78ea2 100644
--- a/arch/x86/boot/compressed/head_32.S
+++ b/arch/x86/boot/compressed/head_32.S
@@ -208,7 +208,7 @@ ENDPROC(efi32_stub_entry)
 #endif
 
 	.text
-relocated:
+SYM_FUNC_START_LOCAL(relocated)
 
 /*
  * Clear BSS (stack is currently empty)
@@ -259,6 +259,7 @@ relocated:
  */
 	xorl	%ebx, %ebx
 	jmp	*%eax
+SYM_FUNC_END(relocated)
 
 #ifdef CONFIG_EFI_STUB
 	.data
diff --git a/arch/x86/boot/compressed/head_64.S b/arch/x86/boot/compressed/head_64.S
index 3e5752efccff..46c132ab5f21 100644
--- a/arch/x86/boot/compressed/head_64.S
+++ b/arch/x86/boot/compressed/head_64.S
@@ -380,7 +380,7 @@ ENDPROC(efi64_stub_entry)
 #endif
 
 	.text
-relocated:
+SYM_FUNC_START_LOCAL(relocated)
 
 /*
  * Clear BSS (stack is currently empty)
@@ -422,13 +422,16 @@ relocated:
  * Jump to the decompressed kernel.
  */
 	jmp	*%rax
+SYM_FUNC_END(relocated)
 
 	.code32
-no_longmode:
+
+SYM_FUNC_START_LOCAL(no_longmode)
 	/* This isn't an x86-64 CPU so hang */
 1:
 	hlt
 	jmp     1b
+SYM_FUNC_END(no_longmode)
 
 #include "../../kernel/verify_cpu.S"
 
-- 
2.12.2

[PATCH v3 15/29] x86: assembly, annotate aliases

[Jiri Slaby]

_key_expansion_128 is an alias to _key_expansion_256a, __memcpy to
memcpy, xen_syscall32_target to xen_sysenter_target, and so on. Annotate
them all using the new SYM_FUNC_START_ALIAS, SYM_FUNC_START_LOCAL_ALIAS,
and SYM_FUNC_END_ALIAS. This will make the tools generating the
debuginfo happy.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: <x86@kernel.org>
Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: Juergen Gross <jgross@suse.com>
Reviewed-by: Juergen Gross <jgross@suse.com> [xen parts]
Cc: <linux-crypto@vger.kernel.org>
Cc: <xen-devel@lists.xenproject.org>
---
 arch/x86/crypto/aesni-intel_asm.S | 5 ++---
 arch/x86/lib/memcpy_64.S          | 4 ++--
 arch/x86/lib/memmove_64.S         | 4 ++--
 arch/x86/lib/memset_64.S          | 4 ++--
 arch/x86/xen/xen-asm_64.S         | 4 ++--
 5 files changed, 10 insertions(+), 11 deletions(-)

diff --git a/arch/x86/crypto/aesni-intel_asm.S b/arch/x86/crypto/aesni-intel_asm.S
index da76ae01e791..3469670df832 100644
--- a/arch/x86/crypto/aesni-intel_asm.S
+++ b/arch/x86/crypto/aesni-intel_asm.S
@@ -1744,8 +1744,7 @@ ENDPROC(aesni_gcm_enc)
 #endif
 
 
-.align 4
-_key_expansion_128:
+SYM_FUNC_START_LOCAL_ALIAS(_key_expansion_128)
 SYM_FUNC_START_LOCAL(_key_expansion_256a)
 	pshufd $0b11111111, %xmm1, %xmm1
 	shufps $0b00010000, %xmm0, %xmm4
@@ -1756,8 +1755,8 @@ SYM_FUNC_START_LOCAL(_key_expansion_256a)
 	movaps %xmm0, (TKEYP)
 	add $0x10, TKEYP
 	ret
-ENDPROC(_key_expansion_128)
 SYM_FUNC_END(_key_expansion_256a)
+SYM_FUNC_END_ALIAS(_key_expansion_128)
 
 SYM_FUNC_START_LOCAL(_key_expansion_192a)
 	pshufd $0b01010101, %xmm1, %xmm1
diff --git a/arch/x86/lib/memcpy_64.S b/arch/x86/lib/memcpy_64.S
index 9a53a06e5a3e..4911b1c61aa8 100644
--- a/arch/x86/lib/memcpy_64.S
+++ b/arch/x86/lib/memcpy_64.S
@@ -26,7 +26,7 @@
  * Output:
  * rax original destination
  */
-ENTRY(__memcpy)
+SYM_FUNC_START_ALIAS(__memcpy)
 ENTRY(memcpy)
 	ALTERNATIVE_2 "jmp memcpy_orig", "", X86_FEATURE_REP_GOOD, \
 		      "jmp memcpy_erms", X86_FEATURE_ERMS
@@ -40,7 +40,7 @@ ENTRY(memcpy)
 	rep movsb
 	ret
 ENDPROC(memcpy)
-ENDPROC(__memcpy)
+SYM_FUNC_END_ALIAS(__memcpy)
 EXPORT_SYMBOL(memcpy)
 EXPORT_SYMBOL(__memcpy)
 
diff --git a/arch/x86/lib/memmove_64.S b/arch/x86/lib/memmove_64.S
index 15de86cd15b0..d22af97e5b27 100644
--- a/arch/x86/lib/memmove_64.S
+++ b/arch/x86/lib/memmove_64.S
@@ -25,7 +25,7 @@
  */
 .weak memmove
 
-ENTRY(memmove)
+SYM_FUNC_START_ALIAS(memmove)
 ENTRY(__memmove)
 
 	/* Handle more 32 bytes in loop */
@@ -207,6 +207,6 @@ ENTRY(__memmove)
 13:
 	retq
 ENDPROC(__memmove)
-ENDPROC(memmove)
+SYM_FUNC_END_ALIAS(memmove)
 EXPORT_SYMBOL(__memmove)
 EXPORT_SYMBOL(memmove)
diff --git a/arch/x86/lib/memset_64.S b/arch/x86/lib/memset_64.S
index 55b95db30a61..0d3a1d341e60 100644
--- a/arch/x86/lib/memset_64.S
+++ b/arch/x86/lib/memset_64.S
@@ -18,7 +18,7 @@
  *
  * rax   original destination
  */
-ENTRY(memset)
+SYM_FUNC_START_ALIAS(memset)
 ENTRY(__memset)
 	/*
 	 * Some CPUs support enhanced REP MOVSB/STOSB feature. It is recommended
@@ -42,8 +42,8 @@ ENTRY(__memset)
 	rep stosb
 	movq %r9,%rax
 	ret
-ENDPROC(memset)
 ENDPROC(__memset)
+SYM_FUNC_END_ALIAS(memset)
 EXPORT_SYMBOL(memset)
 EXPORT_SYMBOL(__memset)
 
diff --git a/arch/x86/xen/xen-asm_64.S b/arch/x86/xen/xen-asm_64.S
index d617bea76039..e1174171ab57 100644
--- a/arch/x86/xen/xen-asm_64.S
+++ b/arch/x86/xen/xen-asm_64.S
@@ -117,13 +117,13 @@ ENDPROC(xen_sysenter_target)
 
 #else /* !CONFIG_IA32_EMULATION */
 
-ENTRY(xen_syscall32_target)
+SYM_FUNC_START_ALIAS(xen_syscall32_target)
 ENTRY(xen_sysenter_target)
 	lea 16(%rsp), %rsp	/* strip %rcx, %r11 */
 	mov $-ENOSYS, %rax
 	pushq $0
 	jmp hypercall_iret
-ENDPROC(xen_syscall32_target)
 ENDPROC(xen_sysenter_target)
+SYM_FUNC_END_ALIAS(xen_syscall32_target)
 
 #endif	/* CONFIG_IA32_EMULATION */
-- 
2.12.2

[PATCH v3 16/29] x86: entry, annotate interrupt symbols properly

[Jiri Slaby]

* annotate local functions properly by SYM_FUNC_START_LOCAL and
  SYM_FUNC_END
* use SYM_FUNC_INNER_LABEL for functions being in the middle of others
* mark the end of common_interrupt by SYM_FUNC_END

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: <x86@kernel.org>
---
 arch/x86/entry/entry_32.S | 15 ++++++++-------
 arch/x86/entry/entry_64.S |  9 ++++-----
 2 files changed, 12 insertions(+), 12 deletions(-)

diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S
index 73c34507893b..acc8d5ad5264 100644
--- a/arch/x86/entry/entry_32.S
+++ b/arch/x86/entry/entry_32.S
@@ -316,7 +316,7 @@ ret_from_intr:
 	cmpl	$USER_RPL, %eax
 	jb	resume_kernel			# not returning to v8086 or userspace
 
-ENTRY(resume_userspace)
+SYM_FUNC_INNER_LABEL(resume_userspace, SYM_V_LOCAL)
 	DISABLE_INTERRUPTS(CLBR_ANY)
 	TRACE_IRQS_OFF
 	movl	%esp, %eax
@@ -547,10 +547,11 @@ restore_all:
 	INTERRUPT_RETURN
 
 .section .fixup, "ax"
-ENTRY(iret_exc	)
+SYM_FUNC_START(iret_exc)
 	pushl	$0				# no error code
 	pushl	$do_iret_error
 	jmp	common_exception
+SYM_FUNC_END(iret_exc)
 .previous
 	_ASM_EXTABLE(.Lirq_return, iret_exc)
 
@@ -642,7 +643,7 @@ ENDPROC(irq_entries_start)
  * so IRQ-flags tracing has to follow that:
  */
 	.p2align CONFIG_X86_L1_CACHE_SHIFT
-common_interrupt:
+SYM_FUNC_START_LOCAL(common_interrupt)
 	ASM_CLAC
 	addl	$-0x80, (%esp)			/* Adjust vector into the [-256, -1] range */
 	SAVE_ALL
@@ -651,7 +652,7 @@ common_interrupt:
 	movl	%esp, %eax
 	call	do_IRQ
 	jmp	ret_from_intr
-ENDPROC(common_interrupt)
+SYM_FUNC_END(common_interrupt)
 
 #define BUILD_INTERRUPT3(name, nr, fn)	\
 ENTRY(name)				\
@@ -811,7 +812,7 @@ ENTRY(xen_hypervisor_callback)
 
 	jmp	xen_iret_crit_fixup
 
-ENTRY(xen_do_upcall)
+SYM_FUNC_INNER_LABEL(xen_do_upcall, SYM_V_GLOBAL)
 1:	mov	%esp, %eax
 	call	xen_evtchn_do_upcall
 #ifndef CONFIG_PREEMPT
@@ -898,7 +899,7 @@ ENTRY(page_fault)
 	jmp common_exception
 ENDPROC(page_fault)
 
-common_exception:
+SYM_FUNC_START_LOCAL(common_exception)
 	/* the function address is in %gs's slot on the stack */
 	pushl	%fs
 	pushl	%es
@@ -928,7 +929,7 @@ common_exception:
 	movl	%esp, %eax			# pt_regs pointer
 	call	*%edi
 	jmp	ret_from_exception
-ENDPROC(common_exception)
+SYM_FUNC_END(common_exception)
 
 ENTRY(debug)
 	/*
diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
index aff5f8051ce2..086f0cb600ff 100644
--- a/arch/x86/entry/entry_64.S
+++ b/arch/x86/entry/entry_64.S
@@ -509,7 +509,7 @@ ENDPROC(irq_entries_start)
 	 * then jump to common_interrupt.
 	 */
 	.p2align CONFIG_X86_L1_CACHE_SHIFT
-common_interrupt:
+SYM_FUNC_START_LOCAL(common_interrupt)
 	ASM_CLAC
 	addq	$-0x80, (%rsp)			/* Adjust vector to [-256, -1] range */
 	interrupt do_IRQ
@@ -562,7 +562,7 @@ restore_c_regs_and_iret:
 	REMOVE_PT_GPREGS_FROM_STACK 8
 	INTERRUPT_RETURN
 
-ENTRY(native_iret)
+SYM_FUNC_INNER_LABEL(native_iret, SYM_V_GLOBAL)
 	/*
 	 * Are we returning to a stack segment from the LDT?  Note: in
 	 * 64-bit mode SS:RSP on the exception stack is always valid.
@@ -572,8 +572,7 @@ ENTRY(native_iret)
 	jnz	native_irq_return_ldt
 #endif
 
-.global native_irq_return_iret
-native_irq_return_iret:
+SYM_FUNC_INNER_LABEL(native_irq_return_iret, SYM_V_GLOBAL)
 	/*
 	 * This may fault.  Non-paranoid faults on return to userspace are
 	 * handled by fixup_bad_iret.  These include #SS, #GP, and #NP.
@@ -649,7 +648,7 @@ native_irq_return_ldt:
 	 */
 	jmp	native_irq_return_iret
 #endif
-ENDPROC(common_interrupt)
+SYM_FUNC_END(common_interrupt)
 
 /*
  * APIC interrupts.
-- 
2.12.2

[PATCH v3 17/29] x86: head, annotate data appropriatelly

[Jiri Slaby]

Use the new SYM_DATA_SIMPLE and SYM_DATA_END in both 32 and 64 bit heads.
In the 64-bit version, define also SYM_DATA_START_PAGE_ALIGNED locally
using the new SYM_START. It is used in the code instead of NEXT_PAGE()
which was defined in this file and has been using the obsolete macro
GLOBAL().

Now, the data in the 64-bit object file look sane:
Value   Size Type    Bind   Vis      Ndx Name
  0000  4096 OBJECT  GLOBAL DEFAULT   15 init_level4_pgt
  1000  4096 OBJECT  GLOBAL DEFAULT   15 level3_kernel_pgt
  2000  2048 OBJECT  GLOBAL DEFAULT   15 level2_kernel_pgt
  3000  4096 OBJECT  GLOBAL DEFAULT   15 level2_fixmap_pgt
  4000  4096 OBJECT  GLOBAL DEFAULT   15 level1_fixmap_pgt
  5000     2 OBJECT  GLOBAL DEFAULT   15 early_gdt_descr
  5002     8 OBJECT  LOCAL  DEFAULT   15 early_gdt_descr_base
  500a     8 OBJECT  GLOBAL DEFAULT   15 phys_base
  0000     8 OBJECT  GLOBAL DEFAULT   17 initial_code
  0008     8 OBJECT  GLOBAL DEFAULT   17 initial_gs
  0010     8 OBJECT  GLOBAL DEFAULT   17 initial_stack
  0000     4 OBJECT  GLOBAL DEFAULT   19 early_recursion_flag
  1000  4096 OBJECT  GLOBAL DEFAULT   19 early_level4_pgt
  2000 0x40000 OBJECT  GLOBAL DEFAULT   19 early_dynamic_pgts
  0000  4096 OBJECT  GLOBAL DEFAULT   22 empty_zero_page

All have correct size and type.

Note, that we can now see that it might be worth pushing
early_recursion_flag after early_dynamic_pgts -- we are wasting almost
4K of .init.data.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
---
 arch/x86/kernel/head_32.S | 31 ++++++++++++----------
 arch/x86/kernel/head_64.S | 67 +++++++++++++++++++++++++----------------------
 2 files changed, 53 insertions(+), 45 deletions(-)

diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S
index 00193c67845c..47a212e3beee 100644
--- a/arch/x86/kernel/head_32.S
+++ b/arch/x86/kernel/head_32.S
@@ -543,8 +543,7 @@ SYM_FUNC_END(ignore_int)
 
 __INITDATA
 	.align 4
-GLOBAL(early_recursion_flag)
-	.long 0
+SYM_DATA_SIMPLE(early_recursion_flag, .long 0)
 
 __REFDATA
 	.align 4
@@ -582,7 +581,7 @@ EXPORT_SYMBOL(empty_zero_page)
 __PAGE_ALIGNED_DATA
 	/* Page-aligned for the benefit of paravirt? */
 	.align PAGE_SIZE
-ENTRY(initial_page_table)
+SYM_DATA_START(initial_page_table)
 	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0	/* low identity map */
 # if KPMDS == 3
 	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0
@@ -600,17 +599,17 @@ ENTRY(initial_page_table)
 #  error "Kernel PMDs should be 1, 2 or 3"
 # endif
 	.align PAGE_SIZE		/* needs to be page-sized too */
+SYM_DATA_END(initial_page_table)
 #endif
 
 .data
 .balign 4
-ENTRY(initial_stack)
-	/*
-	 * The SIZEOF_PTREGS gap is a convention which helps the in-kernel
-	 * unwinder reliably detect the end of the stack.
-	 */
-	.long init_thread_union + THREAD_SIZE - SIZEOF_PTREGS - \
-	      TOP_OF_KERNEL_STACK_PADDING;
+/*
+ * The SIZEOF_PTREGS gap is a convention which helps the in-kernel unwinder
+ * reliably detect the end of the stack.
+ */
+SYM_DATA_SIMPLE(initial_stack, .long init_thread_union + THREAD_SIZE -
+		SIZEOF_PTREGS - TOP_OF_KERNEL_STACK_PADDING)
 
 __INITRODATA
 int_msg:
@@ -632,27 +631,31 @@ int_msg:
 	ALIGN
 # early boot GDT descriptor (must use 1:1 address mapping)
 	.word 0				# 32 bit align gdt_desc.address
-boot_gdt_descr:
+SYM_DATA_START(boot_gdt_descr)
 	.word __BOOT_DS+7
 	.long boot_gdt - __PAGE_OFFSET
+SYM_DATA_END(boot_gdt_descr)
 
 	.word 0				# 32-bit align idt_desc.address
-idt_descr:
+SYM_DATA_START(idt_descr)
 	.word IDT_ENTRIES*8-1		# idt contains 256 entries
 	.long idt_table
+SYM_DATA_END(idt_descr)
 
 # boot GDT descriptor (later on used by CPU#0):
 	.word 0				# 32 bit align gdt_desc.address
-ENTRY(early_gdt_descr)
+SYM_DATA_START(early_gdt_descr)
 	.word GDT_ENTRIES*8-1
 	.long gdt_page			/* Overwritten for secondary CPUs */
+SYM_DATA_END(early_gdt_descr)
 
 /*
  * The boot_gdt must mirror the equivalent in setup.S and is
  * used only for booting.
  */
 	.align L1_CACHE_BYTES
-ENTRY(boot_gdt)
+SYM_DATA_START(boot_gdt)
 	.fill GDT_ENTRY_BOOT_CS,8,0
 	.quad 0x00cf9a000000ffff	/* kernel 4GB code at 0x00000000 */
 	.quad 0x00cf92000000ffff	/* kernel 4GB data at 0x00000000 */
+SYM_DATA_END(boot_gdt)
diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
index a2230c35fcf9..3c9037a65ee9 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -323,16 +323,14 @@ ENDPROC(start_cpu0)
 	/* Both SMP bootup and ACPI suspend change these variables */
 	__REFDATA
 	.balign	8
-	GLOBAL(initial_code)
-	.quad	x86_64_start_kernel
-	GLOBAL(initial_gs)
-	.quad	INIT_PER_CPU_VAR(irq_stack_union)
-	GLOBAL(initial_stack)
-	/*
-	 * The SIZEOF_PTREGS gap is a convention which helps the in-kernel
-	 * unwinder reliably detect the end of the stack.
-	 */
-	.quad  init_thread_union + THREAD_SIZE - SIZEOF_PTREGS
+SYM_DATA_SIMPLE(initial_code,	.quad x86_64_start_kernel)
+SYM_DATA_SIMPLE(initial_gs,	.quad INIT_PER_CPU_VAR(irq_stack_union))
+/*
+ * The SIZEOF_PTREGS gap is a convention which helps the in-kernel unwinder
+ * reliably detect the end of the stack.
+ */
+SYM_DATA_SIMPLE(initial_stack,
+		.quad init_thread_union + THREAD_SIZE - SIZEOF_PTREGS)
 	__FINITDATA
 
 SYM_FUNC_START_LOCAL(bad_address)
@@ -403,12 +401,10 @@ SYM_FUNC_END(early_idt_handler_common)
 	__INITDATA
 
 	.balign 4
-GLOBAL(early_recursion_flag)
-	.long 0
+SYM_DATA_SIMPLE(early_recursion_flag, .long 0)
 
-#define NEXT_PAGE(name) \
-	.balign	PAGE_SIZE; \
-GLOBAL(name)
+#define SYM_DATA_START_PAGE_ALIGNED(name)			\
+	SYM_START(name, SYM_V_GLOBAL, .balign PAGE_SIZE)
 
 /* Automate the creation of 1 to 1 mapping pmd entries */
 #define PMDS(START, PERM, COUNT)			\
@@ -419,44 +415,51 @@ GLOBAL(name)
 	.endr
 
 	__INITDATA
-NEXT_PAGE(early_level4_pgt)
+SYM_DATA_START_PAGE_ALIGNED(early_level4_pgt)
 	.fill	511,8,0
 	.quad	level3_kernel_pgt - __START_KERNEL_map + _PAGE_TABLE
+SYM_DATA_END(early_level4_pgt)
 
-NEXT_PAGE(early_dynamic_pgts)
+SYM_DATA_START_PAGE_ALIGNED(early_dynamic_pgts)
 	.fill	512*EARLY_DYNAMIC_PAGE_TABLES,8,0
+SYM_DATA_END(early_dynamic_pgts)
 
 	.data
 
 #ifndef CONFIG_XEN
-NEXT_PAGE(init_level4_pgt)
+SYM_DATA_START_PAGE_ALIGNED(init_level4_pgt)
 	.fill	512,8,0
+SYM_DATA_END(init_level4_pgt)
 #else
-NEXT_PAGE(init_level4_pgt)
+SYM_DATA_START_PAGE_ALIGNED(init_level4_pgt)
 	.quad   level3_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE
 	.org    init_level4_pgt + L4_PAGE_OFFSET*8, 0
 	.quad   level3_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE
 	.org    init_level4_pgt + L4_START_KERNEL*8, 0
 	/* (2^48-(2*1024*1024*1024))/(2^39) = 511 */
 	.quad   level3_kernel_pgt - __START_KERNEL_map + _PAGE_TABLE
+SYM_DATA_END(init_level4_pgt)
 
-NEXT_PAGE(level3_ident_pgt)
+SYM_DATA_START_PAGE_ALIGNED(level3_ident_pgt)
 	.quad	level2_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE
 	.fill	511, 8, 0
-NEXT_PAGE(level2_ident_pgt)
+SYM_DATA_END(level3_ident_pgt)
+SYM_DATA_START_PAGE_ALIGNED(level2_ident_pgt)
 	/* Since I easily can, map the first 1G.
 	 * Don't set NX because code runs from these pages.
 	 */
 	PMDS(0, __PAGE_KERNEL_IDENT_LARGE_EXEC, PTRS_PER_PMD)
+SYM_DATA_END(level2_ident_pgt)
 #endif
 
-NEXT_PAGE(level3_kernel_pgt)
+SYM_DATA_START_PAGE_ALIGNED(level3_kernel_pgt)
 	.fill	L3_START_KERNEL,8,0
 	/* (2^48-(2*1024*1024*1024)-((2^39)*511))/(2^30) = 510 */
 	.quad	level2_kernel_pgt - __START_KERNEL_map + _KERNPG_TABLE
 	.quad	level2_fixmap_pgt - __START_KERNEL_map + _PAGE_TABLE
+SYM_DATA_END(level3_kernel_pgt)
 
-NEXT_PAGE(level2_kernel_pgt)
+SYM_DATA_START_PAGE_ALIGNED(level2_kernel_pgt)
 	/*
 	 * 512 MB kernel mapping. We spend a full page on this pagetable
 	 * anyway.
@@ -469,25 +472,26 @@ NEXT_PAGE(level2_kernel_pgt)
 	 */
 	PMDS(0, __PAGE_KERNEL_LARGE_EXEC,
 		KERNEL_IMAGE_SIZE/PMD_SIZE)
+SYM_DATA_END(level2_kernel_pgt)
 
-NEXT_PAGE(level2_fixmap_pgt)
+SYM_DATA_START_PAGE_ALIGNED(level2_fixmap_pgt)
 	.fill	506,8,0
 	.quad	level1_fixmap_pgt - __START_KERNEL_map + _PAGE_TABLE
 	/* 8MB reserved for vsyscalls + a 2MB hole = 4 + 1 entries */
 	.fill	5,8,0
+SYM_DATA_END(level2_fixmap_pgt)
 
-NEXT_PAGE(level1_fixmap_pgt)
+SYM_DATA_START_PAGE_ALIGNED(level1_fixmap_pgt)
 	.fill	512,8,0
+SYM_DATA_END(level1_fixmap_pgt)
 
 #undef PMDS
 
 	.data
 	.align 16
-	.globl early_gdt_descr
-early_gdt_descr:
-	.word	GDT_ENTRIES*8-1
-early_gdt_descr_base:
-	.quad	INIT_PER_CPU_VAR(gdt_page)
+
+SYM_DATA_SIMPLE(early_gdt_descr, .word GDT_ENTRIES*8-1)
+SYM_DATA_SIMPLE_LOCAL(early_gdt_descr_base, .quad INIT_PER_CPU_VAR(gdt_page))
 
 /* This must match the first entry in level2_kernel_pgt */
 SYM_DATA_SIMPLE(phys_base, .quad 0x0000000000000000)
@@ -496,7 +500,8 @@ EXPORT_SYMBOL(phys_base)
 #include "../../x86/xen/xen-head.S"
 	
 	__PAGE_ALIGNED_BSS
-NEXT_PAGE(empty_zero_page)
+SYM_DATA_START_PAGE_ALIGNED(empty_zero_page)
 	.skip PAGE_SIZE
+SYM_DATA_END(empty_zero_page)
 EXPORT_SYMBOL(empty_zero_page)
 
-- 
2.12.2

[PATCH v3 18/29] x86: boot, annotate data appropriatelly

[Jiri Slaby]

Use the new SYM_DATA_SIMPLE, SYM_DATA_START, and SYM_DATA_END* macros
for data.

Now, the data in the object file look sane:
Value   Size Type    Bind   Vis      Ndx Name
  0000    10 OBJECT  GLOBAL DEFAULT    3 efi32_boot_gdt
  000a    10 OBJECT  LOCAL  DEFAULT    3 save_gdt
  0014     8 OBJECT  LOCAL  DEFAULT    3 func_rt_ptr
  001c    48 OBJECT  GLOBAL DEFAULT    3 efi_gdt64
  004c     0 OBJECT  LOCAL  DEFAULT    3 efi_gdt64_end

  0000    48 OBJECT  LOCAL  DEFAULT    3 gdt
  0030     0 OBJECT  LOCAL  DEFAULT    3 gdt_end
  0030     8 OBJECT  LOCAL  DEFAULT    3 efi_config
  0038    49 OBJECT  GLOBAL DEFAULT    3 efi32_config
  0069    49 OBJECT  GLOBAL DEFAULT    3 efi64_config

All have correct size and type.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: x86@kernel.org
---
 arch/x86/boot/compressed/efi_thunk_64.S | 21 ++++++++++++---------
 arch/x86/boot/compressed/head_64.S      | 26 ++++++++++++--------------
 2 files changed, 24 insertions(+), 23 deletions(-)

diff --git a/arch/x86/boot/compressed/efi_thunk_64.S b/arch/x86/boot/compressed/efi_thunk_64.S
index 630384a4c14a..86528f120962 100644
--- a/arch/x86/boot/compressed/efi_thunk_64.S
+++ b/arch/x86/boot/compressed/efi_thunk_64.S
@@ -175,16 +175,19 @@ ENDPROC(efi_enter32)
 
 	.data
 	.balign	8
-	.global	efi32_boot_gdt
-efi32_boot_gdt:	.word	0
-		.quad	0
+SYM_DATA_START(efi32_boot_gdt)
+	.word	0
+	.quad	0
+SYM_DATA_END(efi32_boot_gdt)
+
+SYM_DATA_START_LOCAL(save_gdt)
+	.word	0
+	.quad	0
+SYM_DATA_END(save_gdt)
 
-save_gdt:	.word	0
-		.quad	0
-func_rt_ptr:	.quad	0
+SYM_DATA_SIMPLE_LOCAL(func_rt_ptr, .quad 0)
 
-	.global efi_gdt64
-efi_gdt64:
+SYM_DATA_START(efi_gdt64)
 	.word	efi_gdt64_end - efi_gdt64
 	.long	0			/* Filled out by user */
 	.word	0
@@ -193,4 +196,4 @@ efi_gdt64:
 	.quad	0x00cf92000000ffff	/* __KERNEL_DS */
 	.quad	0x0080890000000000	/* TS descriptor */
 	.quad   0x0000000000000000	/* TS continued */
-efi_gdt64_end:
+SYM_DATA_END_LABEL(efi_gdt64, SYM_V_LOCAL, efi_gdt64_end)
diff --git a/arch/x86/boot/compressed/head_64.S b/arch/x86/boot/compressed/head_64.S
index 46c132ab5f21..6eb5a50a301e 100644
--- a/arch/x86/boot/compressed/head_64.S
+++ b/arch/x86/boot/compressed/head_64.S
@@ -436,7 +436,7 @@ SYM_FUNC_END(no_longmode)
 #include "../../kernel/verify_cpu.S"
 
 	.data
-gdt:
+SYM_DATA_START_LOCAL(gdt)
 	.word	gdt_end - gdt
 	.long	gdt
 	.word	0
@@ -445,25 +445,24 @@ gdt:
 	.quad	0x00cf92000000ffff	/* __KERNEL_DS */
 	.quad	0x0080890000000000	/* TS descriptor */
 	.quad   0x0000000000000000	/* TS continued */
-gdt_end:
+SYM_DATA_END_LABEL(gdt, SYM_V_LOCAL, gdt_end)
 
 #ifdef CONFIG_EFI_STUB
-efi_config:
-	.quad	0
+SYM_DATA_SIMPLE_LOCAL(efi_config, .quad 0)
 
 #ifdef CONFIG_EFI_MIXED
-	.global efi32_config
-efi32_config:
+SYM_DATA_START(efi32_config)
 	.fill	5,8,0
 	.quad	efi64_thunk
 	.byte	0
+SYM_DATA_END(efi32_config)
 #endif
 
-	.global efi64_config
-efi64_config:
+SYM_DATA_START(efi64_config)
 	.fill	5,8,0
 	.quad	efi_call
 	.byte	1
+SYM_DATA_END(efi64_config)
 #endif /* CONFIG_EFI_STUB */
 
 /*
@@ -471,16 +470,15 @@ efi64_config:
  */
 	.bss
 	.balign 4
-boot_heap:
-	.fill BOOT_HEAP_SIZE, 1, 0
-boot_stack:
+SYM_DATA_SIMPLE_LOCAL(boot_heap, .fill BOOT_HEAP_SIZE, 1, 0)
+
+SYM_DATA_START_LOCAL(boot_stack)
 	.fill BOOT_STACK_SIZE, 1, 0
-boot_stack_end:
+SYM_DATA_END_LABEL(boot_stack, SYM_V_LOCAL, boot_stack_end)
 
 /*
  * Space for page tables (not in .bss so not zeroed)
  */
 	.section ".pgtable","a",@nobits
 	.balign 4096
-pgtable:
-	.fill BOOT_PGT_SIZE, 1, 0
+SYM_DATA_SIMPLE_LOCAL(pgtable, .fill BOOT_PGT_SIZE, 1, 0)
-- 
2.12.2

[PATCH v3 19/29] x86: um, annotate data appropriatelly

[Jiri Slaby]

Use the new SYM_DATA_START and SYM_DATA_END_LABEL macros for vdso_start.

We get:
  0000  2376 OBJECT  GLOBAL DEFAULT    4 vdso_start
  0948     0 OBJECT  GLOBAL DEFAULT    4 vdso_end

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Richard Weinberger <richard@nod.at>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Cc: user-mode-linux-devel@lists.sourceforge.net
Cc: user-mode-linux-user@lists.sourceforge.net
---
 arch/x86/um/vdso/vdso.S | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/x86/um/vdso/vdso.S b/arch/x86/um/vdso/vdso.S
index 1cb468adacbb..f8b9f46c546d 100644
--- a/arch/x86/um/vdso/vdso.S
+++ b/arch/x86/um/vdso/vdso.S
@@ -1,10 +1,10 @@
 #include <linux/init.h>
+#include <linux/linkage.h>
 
 __INITDATA
 
-	.globl vdso_start, vdso_end
-vdso_start:
+SYM_DATA_START(vdso_start)
 	.incbin "arch/x86/um/vdso/vdso.so"
-vdso_end:
+SYM_DATA_END_LABEL(vdso_start, SYM_V_GLOBAL, vdso_end)
 
 __FINIT
-- 
2.12.2

[PATCH v3 20/29] x86: xen-pvh, annotate data appropriatelly

[Jiri Slaby]

Use the new SYM_DATA_START_LOCAL, and SYM_DATA_END* macros:
  0000     8 OBJECT  LOCAL  DEFAULT    6 gdt
  0008    32 OBJECT  LOCAL  DEFAULT    6 gdt_start
  0028     0 OBJECT  LOCAL  DEFAULT    6 gdt_end
  0028   256 OBJECT  LOCAL  DEFAULT    6 early_stack
  0128     0 OBJECT  LOCAL  DEFAULT    6 early_stack

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: Juergen Gross <jgross@suse.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Cc: xen-devel@lists.xenproject.org
---
 arch/x86/xen/xen-pvh.S | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/arch/x86/xen/xen-pvh.S b/arch/x86/xen/xen-pvh.S
index 512fda03c93f..fa5ba8565646 100644
--- a/arch/x86/xen/xen-pvh.S
+++ b/arch/x86/xen/xen-pvh.S
@@ -137,11 +137,12 @@ ENDPROC(pvh_start_xen)
 
 	.section ".init.data","aw"
 	.balign 8
-gdt:
+SYM_DATA_START_LOCAL(gdt)
 	.word gdt_end - gdt_start
 	.long _pa(gdt_start)
 	.word 0
-gdt_start:
+SYM_DATA_END(gdt)
+SYM_DATA_START_LOCAL(gdt_start)
 	.quad 0x0000000000000000            /* NULL descriptor */
 	.quad 0x0000000000000000            /* reserved */
 #ifdef CONFIG_X86_64
@@ -150,12 +151,12 @@ gdt_start:
 	.quad GDT_ENTRY(0xc09a, 0, 0xfffff) /* __KERNEL_CS */
 #endif
 	.quad GDT_ENTRY(0xc092, 0, 0xfffff) /* __KERNEL_DS */
-gdt_end:
+SYM_DATA_END_LABEL(gdt_start, SYM_V_LOCAL, gdt_end)
 
 	.balign 4
-early_stack:
+SYM_DATA_START_LOCAL(early_stack)
 	.fill 256, 1, 0
-early_stack_end:
+SYM_DATA_END_LABEL(early_stack, SYM_V_LOCAL, early_stack_end)
 
 	ELFNOTE(Xen, XEN_ELFNOTE_PHYS32_ENTRY,
 	             _ASM_PTR (pvh_start_xen - __START_KERNEL_map))
-- 
2.12.2

[PATCH v3 21/29] x86: purgatory, start using annotations

[Jiri Slaby]

purgatory used no annotations. So include linux/linkage.h and annotate
everything:
* functions by SYM_FUNC_*
* data by SYM_DATA_*

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
---
 arch/x86/purgatory/entry64.S      | 21 ++++++++++++---------
 arch/x86/purgatory/setup-x86_64.S | 14 ++++++++------
 arch/x86/purgatory/stack.S        |  7 ++++---
 3 files changed, 24 insertions(+), 18 deletions(-)

diff --git a/arch/x86/purgatory/entry64.S b/arch/x86/purgatory/entry64.S
index d1a4291d3568..d756c53e1d60 100644
--- a/arch/x86/purgatory/entry64.S
+++ b/arch/x86/purgatory/entry64.S
@@ -10,13 +10,13 @@
  * Version 2.  See the file COPYING for more details.
  */
 
+#include <linux/linkage.h>
+
 	.text
 	.balign 16
 	.code64
-	.globl entry64, entry64_regs
-
 
-entry64:
+SYM_FUNC_START(entry64)
 	/* Setup a gdt that should be preserved */
 	lgdt gdt(%rip)
 
@@ -56,10 +56,11 @@ new_cs_exit:
 
 	/* Jump to the new code... */
 	jmpq	*rip(%rip)
+SYM_FUNC_END(entry64)
 
 	.section ".rodata"
 	.balign 4
-entry64_regs:
+SYM_DATA_START(entry64_regs)
 rax:	.quad 0x0
 rcx:	.quad 0x0
 rdx:	.quad 0x0
@@ -77,12 +78,12 @@ r13:	.quad 0x0
 r14:	.quad 0x0
 r15:	.quad 0x0
 rip:	.quad 0x0
-	.size entry64_regs, . - entry64_regs
+SYM_DATA_END(entry64_regs)
 
 	/* GDT */
 	.section ".rodata"
 	.balign 16
-gdt:
+SYM_DATA_START_LOCAL(gdt)
 	/* 0x00 unusable segment
 	 * 0x08 unused
 	 * so use them as gdt ptr
@@ -96,6 +97,8 @@ gdt:
 
 	/* 0x18 4GB flat data segment */
 	.word 0xFFFF, 0x0000, 0x9200, 0x00CF
-gdt_end:
-stack:	.quad   0, 0
-stack_init:
+SYM_DATA_END_LABEL(gdt, SYM_V_LOCAL, gdt_end)
+
+SYM_DATA_START_LOCAL(stack)
+	.quad   0, 0
+SYM_DATA_END_LABEL(stack, SYM_V_LOCAL, stack_init)
diff --git a/arch/x86/purgatory/setup-x86_64.S b/arch/x86/purgatory/setup-x86_64.S
index dfae9b9e60b5..9f1b333c7ffd 100644
--- a/arch/x86/purgatory/setup-x86_64.S
+++ b/arch/x86/purgatory/setup-x86_64.S
@@ -9,14 +9,14 @@
  * This source code is licensed under the GNU General Public License,
  * Version 2.  See the file COPYING for more details.
  */
+#include <linux/linkage.h>
 #include <asm/purgatory.h>
 
 	.text
-	.globl purgatory_start
 	.balign 16
-purgatory_start:
 	.code64
 
+SYM_FUNC_START(purgatory_start)
 	/* Load a gdt so I know what the segment registers are */
 	lgdt	gdt(%rip)
 
@@ -34,10 +34,12 @@ purgatory_start:
 	/* Call the C code */
 	call purgatory
 	jmp	entry64
+SYM_FUNC_END(purgatory_start)
 
 	.section ".rodata"
 	.balign 16
-gdt:	/* 0x00 unusable segment
+SYM_DATA_START_LOCAL(gdt)
+	/* 0x00 unusable segment
 	 * 0x08 unused
 	 * so use them as the gdt ptr
 	 */
@@ -50,10 +52,10 @@ gdt:	/* 0x00 unusable segment
 
 	/* 0x18 4GB flat data segment */
 	.word	0xFFFF, 0x0000, 0x9200, 0x00CF
-gdt_end:
+SYM_DATA_END_LABEL(gdt, SYM_V_LOCAL, gdt_end)
 
 	.bss
 	.balign 4096
-lstack:
+SYM_DATA_START_LOCAL(lstack)
 	.skip 4096
-lstack_end:
+SYM_DATA_END_LABEL(lstack, SYM_V_LOCAL, lstack_end)
diff --git a/arch/x86/purgatory/stack.S b/arch/x86/purgatory/stack.S
index 50a4147f91fb..958ffc769705 100644
--- a/arch/x86/purgatory/stack.S
+++ b/arch/x86/purgatory/stack.S
@@ -7,13 +7,14 @@
  * Version 2.  See the file COPYING for more details.
  */
 
+#include <linux/linkage.h>
+
 	/* A stack for the loaded kernel.
 	 * Separate and in the data section so it can be prepopulated.
 	 */
 	.data
 	.balign 4096
-	.globl stack, stack_end
 
-stack:
+SYM_DATA_START(stack)
 	.skip 4096
-stack_end:
+SYM_DATA_END_LABEL(stack, SYM_V_GLOBAL, stack_end)
-- 
2.12.2

[PATCH v3 22/29] x86: assembly, use SYM_FUNC_INNER_LABEL instead of GLOBAL

[Jiri Slaby]

GLOBAL had several meanings and is going away. In this patch, convert
all the inner functions labels marked with GLOBAL to use
SYM_FUNC_INNER_LABEL instead.

Note that retint_user needs not be global, perhaps since commit
2ec67971facc ("x86/entry/64/compat: Remove most of the fast system call
machinery"), where entry_64_compat's caller was removed. So mark the
label as LOCAL.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Cc: Andy Lutomirski <luto@amacapital.net>
---
 arch/x86/entry/entry_64.S                |  6 +++---
 arch/x86/entry/entry_64_compat.S         |  2 +-
 arch/x86/entry/vdso/vdso32/system_call.S |  2 +-
 arch/x86/kernel/ftrace_32.S              |  2 +-
 arch/x86/kernel/ftrace_64.S              | 16 ++++++++--------
 arch/x86/realmode/rm/reboot.S            |  2 +-
 6 files changed, 15 insertions(+), 15 deletions(-)

diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
index 086f0cb600ff..213127a44c7c 100644
--- a/arch/x86/entry/entry_64.S
+++ b/arch/x86/entry/entry_64.S
@@ -146,7 +146,7 @@ ENTRY(entry_SYSCALL_64)
 	 * after the swapgs, so that it can do the swapgs
 	 * for the guest and jump here on syscall.
 	 */
-GLOBAL(entry_SYSCALL_64_after_swapgs)
+SYM_FUNC_INNER_LABEL(entry_SYSCALL_64_after_swapgs, SYM_V_GLOBAL)
 
 	movq	%rsp, PER_CPU_VAR(rsp_scratch)
 	movq	PER_CPU_VAR(cpu_current_top_of_stack), %rsp
@@ -526,7 +526,7 @@ ret_from_intr:
 	jz	retint_kernel
 
 	/* Interrupt came from user space */
-GLOBAL(retint_user)
+SYM_FUNC_INNER_LABEL(retint_user, SYM_V_LOCAL)
 	mov	%rsp,%rdi
 	call	prepare_exit_to_usermode
 	TRACE_IRQS_IRETQ
@@ -555,7 +555,7 @@ retint_kernel:
  * At this label, code paths which return to kernel and to user,
  * which come from interrupts/exception and from syscalls, merge.
  */
-GLOBAL(restore_regs_and_iret)
+SYM_FUNC_INNER_LABEL(restore_regs_and_iret, SYM_V_GLOBAL)
 	RESTORE_EXTRA_REGS
 restore_c_regs_and_iret:
 	RESTORE_C_REGS
diff --git a/arch/x86/entry/entry_64_compat.S b/arch/x86/entry/entry_64_compat.S
index ba6af2fba6a7..b7934ef3f5bb 100644
--- a/arch/x86/entry/entry_64_compat.S
+++ b/arch/x86/entry/entry_64_compat.S
@@ -131,7 +131,7 @@ ENTRY(entry_SYSENTER_compat)
 	pushq	$X86_EFLAGS_FIXED
 	popfq
 	jmp	.Lsysenter_flags_fixed
-GLOBAL(__end_entry_SYSENTER_compat)
+SYM_FUNC_INNER_LABEL(__end_entry_SYSENTER_compat, SYM_V_GLOBAL)
 ENDPROC(entry_SYSENTER_compat)
 
 /*
diff --git a/arch/x86/entry/vdso/vdso32/system_call.S b/arch/x86/entry/vdso/vdso32/system_call.S
index ed4bc9731cbb..ce5b1dd32337 100644
--- a/arch/x86/entry/vdso/vdso32/system_call.S
+++ b/arch/x86/entry/vdso/vdso32/system_call.S
@@ -61,7 +61,7 @@ __kernel_vsyscall:
 
 	/* Enter using int $0x80 */
 	int	$0x80
-GLOBAL(int80_landing_pad)
+SYM_FUNC_INNER_LABEL(int80_landing_pad, SYM_V_GLOBAL)
 
 	/*
 	 * Restore EDX and ECX in case they were clobbered.  EBP is not
diff --git a/arch/x86/kernel/ftrace_32.S b/arch/x86/kernel/ftrace_32.S
index 89f8324e9a68..8ad07ddfa1c9 100644
--- a/arch/x86/kernel/ftrace_32.S
+++ b/arch/x86/kernel/ftrace_32.S
@@ -139,7 +139,7 @@ ENTRY(ftrace_regs_caller)
 	movl	function_trace_op, %ecx		/* Save ftrace_pos in 3rd parameter */
 	pushl	%esp				/* Save pt_regs as 4th parameter */
 
-GLOBAL(ftrace_regs_call)
+SYM_FUNC_INNER_LABEL(ftrace_regs_call, SYM_V_GLOBAL)
 	call	ftrace_stub
 
 	addl	$4, %esp			/* Skip pt_regs */
diff --git a/arch/x86/kernel/ftrace_64.S b/arch/x86/kernel/ftrace_64.S
index 76c774a5e792..aef60bbe854d 100644
--- a/arch/x86/kernel/ftrace_64.S
+++ b/arch/x86/kernel/ftrace_64.S
@@ -153,14 +153,14 @@ ENTRY(ftrace_caller)
 	/* save_mcount_regs fills in first two parameters */
 	save_mcount_regs
 
-GLOBAL(ftrace_caller_op_ptr)
+SYM_FUNC_INNER_LABEL(ftrace_caller_op_ptr, SYM_V_GLOBAL)
 	/* Load the ftrace_ops into the 3rd parameter */
 	movq function_trace_op(%rip), %rdx
 
 	/* regs go into 4th parameter (but make it NULL) */
 	movq $0, %rcx
 
-GLOBAL(ftrace_call)
+SYM_FUNC_INNER_LABEL(ftrace_call, SYM_V_GLOBAL)
 	call ftrace_stub
 
 	restore_mcount_regs
@@ -173,10 +173,10 @@ GLOBAL(ftrace_call)
 	 * think twice before adding any new code or changing the
 	 * layout here.
 	 */
-GLOBAL(ftrace_epilogue)
+SYM_FUNC_INNER_LABEL(ftrace_epilogue, SYM_V_GLOBAL)
 
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
-GLOBAL(ftrace_graph_call)
+SYM_FUNC_INNER_LABEL(ftrace_graph_call, SYM_V_GLOBAL)
 	jmp ftrace_stub
 #endif
 
@@ -193,7 +193,7 @@ ENTRY(ftrace_regs_caller)
 	save_mcount_regs 8
 	/* save_mcount_regs fills in first two parameters */
 
-GLOBAL(ftrace_regs_caller_op_ptr)
+SYM_FUNC_INNER_LABEL(ftrace_regs_caller_op_ptr, SYM_V_GLOBAL)
 	/* Load the ftrace_ops into the 3rd parameter */
 	movq function_trace_op(%rip), %rdx
 
@@ -220,7 +220,7 @@ GLOBAL(ftrace_regs_caller_op_ptr)
 	/* regs go into 4th parameter */
 	leaq (%rsp), %rcx
 
-GLOBAL(ftrace_regs_call)
+SYM_FUNC_INNER_LABEL(ftrace_regs_call, SYM_V_GLOBAL)
 	call ftrace_stub
 
 	/* Copy flags back to SS, to restore them */
@@ -250,7 +250,7 @@ GLOBAL(ftrace_regs_call)
 	 * The trampoline will add the code to jump
 	 * to the return.
 	 */
-GLOBAL(ftrace_regs_caller_end)
+SYM_FUNC_INNER_LABEL(ftrace_regs_caller_end, SYM_V_GLOBAL)
 
 	jmp ftrace_epilogue
 
@@ -272,7 +272,7 @@ fgraph_trace:
 	jnz ftrace_graph_caller
 #endif
 
-GLOBAL(ftrace_stub)
+SYM_FUNC_INNER_LABEL(ftrace_stub, SYM_V_GLOBAL)
 	retq
 
 trace:
diff --git a/arch/x86/realmode/rm/reboot.S b/arch/x86/realmode/rm/reboot.S
index c8855d50f9c1..cebac961aa42 100644
--- a/arch/x86/realmode/rm/reboot.S
+++ b/arch/x86/realmode/rm/reboot.S
@@ -32,7 +32,7 @@ ENTRY(machine_real_restart_asm)
 	movl	%eax, %cr0
 	ljmpl	$__KERNEL32_CS, $pa_machine_real_restart_paging_off
 
-GLOBAL(machine_real_restart_paging_off)
+SYM_FUNC_INNER_LABEL(machine_real_restart_paging_off, SYM_V_GLOBAL)
 	xorl	%eax, %eax
 	xorl	%edx, %edx
 	movl	$MSR_EFER, %ecx
-- 
2.12.2

[PATCH v3 23/29] x86: realmode, use SYM_DATA_* instead of GLOBAL

[Jiri Slaby]

GLOBAL had several meanings and is going away. In this patch, convert
all the data marked using GLOBAL to use SYM_DATA_START or
SYM_DATA_SIMPLE instead.

Notes:
* SYM_DATA_END_LABEL is used to generate tr_gdt_end too.
* wakeup_idt is marked as LOCAL now as it is used only locally.

Since we removed the last user of END on X86, make sure, END is not
defined there.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
---
 arch/x86/realmode/rm/header.S            |  8 +++-----
 arch/x86/realmode/rm/reboot.S            |  8 ++++----
 arch/x86/realmode/rm/stack.S             | 14 ++++++--------
 arch/x86/realmode/rm/trampoline_32.S     | 10 +++++-----
 arch/x86/realmode/rm/trampoline_64.S     | 17 ++++++++---------
 arch/x86/realmode/rm/trampoline_common.S |  4 ++--
 arch/x86/realmode/rm/wakeup_asm.S        | 12 ++++++------
 arch/x86/realmode/rmpiggy.S              | 10 ++++------
 include/linux/linkage.h                  |  2 ++
 9 files changed, 40 insertions(+), 45 deletions(-)

diff --git a/arch/x86/realmode/rm/header.S b/arch/x86/realmode/rm/header.S
index a28221d94e69..1cbe08c9409e 100644
--- a/arch/x86/realmode/rm/header.S
+++ b/arch/x86/realmode/rm/header.S
@@ -13,7 +13,7 @@
 	.section ".header", "a"
 
 	.balign	16
-GLOBAL(real_mode_header)
+SYM_DATA_START(real_mode_header)
 	.long	pa_text_start
 	.long	pa_ro_end
 	/* SMP trampoline */
@@ -33,11 +33,9 @@ GLOBAL(real_mode_header)
 #ifdef CONFIG_X86_64
 	.long	__KERNEL32_CS
 #endif
-END(real_mode_header)
+SYM_DATA_END(real_mode_header)
 
 	/* End signature, used to verify integrity */
 	.section ".signature","a"
 	.balign 4
-GLOBAL(end_signature)
-	.long	REALMODE_END_SIGNATURE
-END(end_signature)
+SYM_DATA_SIMPLE(end_signature, .long REALMODE_END_SIGNATURE)
diff --git a/arch/x86/realmode/rm/reboot.S b/arch/x86/realmode/rm/reboot.S
index cebac961aa42..370ed1fe34e4 100644
--- a/arch/x86/realmode/rm/reboot.S
+++ b/arch/x86/realmode/rm/reboot.S
@@ -127,13 +127,13 @@ bios:
 	.section ".rodata", "a"
 
 	.balign	16
-GLOBAL(machine_real_restart_idt)
+SYM_DATA_START(machine_real_restart_idt)
 	.word	0xffff		/* Length - real mode default value */
 	.long	0		/* Base - real mode default value */
-END(machine_real_restart_idt)
+SYM_DATA_END(machine_real_restart_idt)
 
 	.balign	16
-GLOBAL(machine_real_restart_gdt)
+SYM_DATA_START(machine_real_restart_gdt)
 	/* Self-pointer */
 	.word	0xffff		/* Length - real mode default value */
 	.long	pa_machine_real_restart_gdt
@@ -153,4 +153,4 @@ GLOBAL(machine_real_restart_gdt)
 	 * semantics we don't have to reload the segments once CR0.PE = 0.
 	 */
 	.quad	GDT_ENTRY(0x0093, 0x100, 0xffff)
-END(machine_real_restart_gdt)
+SYM_DATA_END(machine_real_restart_gdt)
diff --git a/arch/x86/realmode/rm/stack.S b/arch/x86/realmode/rm/stack.S
index 867ae87adfae..7c286e43019b 100644
--- a/arch/x86/realmode/rm/stack.S
+++ b/arch/x86/realmode/rm/stack.S
@@ -5,15 +5,13 @@
 #include <linux/linkage.h>
 
 	.data
-GLOBAL(HEAP)
-	.long	rm_heap
-GLOBAL(heap_end)
-	.long	rm_stack
+SYM_DATA_SIMPLE(HEAP, .long rm_heap)
+SYM_DATA_SIMPLE(heap_end, .long rm_stack)
 
 	.bss
 	.balign	16
-GLOBAL(rm_heap)
-	.space	2048
-GLOBAL(rm_stack)
+SYM_DATA_SIMPLE(rm_heap, .space 2048)
+
+SYM_DATA_START(rm_stack)
 	.space	2048
-GLOBAL(rm_stack_end)
+SYM_DATA_END_LABEL(rm_stack, SYM_V_GLOBAL, rm_stack_end)
diff --git a/arch/x86/realmode/rm/trampoline_32.S b/arch/x86/realmode/rm/trampoline_32.S
index aa14fea65aac..956b1c451661 100644
--- a/arch/x86/realmode/rm/trampoline_32.S
+++ b/arch/x86/realmode/rm/trampoline_32.S
@@ -66,10 +66,10 @@ ENDPROC(startup_32)
 
 	.bss
 	.balign 8
-GLOBAL(trampoline_header)
-	tr_start:		.space	4
-	tr_gdt_pad:		.space	2
-	tr_gdt:			.space	6
-END(trampoline_header)
+SYM_DATA_START(trampoline_header)
+SYM_DATA_SIMPLE_LOCAL(tr_start,		.space 4)
+SYM_DATA_SIMPLE_LOCAL(tr_gdt_pad,	.space 2)
+SYM_DATA_SIMPLE_LOCAL(tr_gdt,		.space 6)
+SYM_DATA_END(trampoline_header)
 	
 #include "trampoline_common.S"
diff --git a/arch/x86/realmode/rm/trampoline_64.S b/arch/x86/realmode/rm/trampoline_64.S
index fe21a26a09fe..f1f2f18fff85 100644
--- a/arch/x86/realmode/rm/trampoline_64.S
+++ b/arch/x86/realmode/rm/trampoline_64.S
@@ -132,25 +132,24 @@ ENDPROC(startup_64)
 	# Duplicate the global descriptor table
 	# so the kernel can live anywhere
 	.balign	16
-	.globl tr_gdt
-tr_gdt:
+SYM_DATA_START(tr_gdt)
 	.short	tr_gdt_end - tr_gdt - 1	# gdt limit
 	.long	pa_tr_gdt
 	.short	0
 	.quad	0x00cf9b000000ffff	# __KERNEL32_CS
 	.quad	0x00af9b000000ffff	# __KERNEL_CS
 	.quad	0x00cf93000000ffff	# __KERNEL_DS
-tr_gdt_end:
+SYM_DATA_END_LABEL(tr_gdt, SYM_V_LOCAL, tr_gdt_end)
 
 	.bss
 	.balign	PAGE_SIZE
-GLOBAL(trampoline_pgd)		.space	PAGE_SIZE
+SYM_DATA_SIMPLE(trampoline_pgd, .space PAGE_SIZE)
 
 	.balign	8
-GLOBAL(trampoline_header)
-	tr_start:		.space	8
-	GLOBAL(tr_efer)		.space	8
-	GLOBAL(tr_cr4)		.space	4
-END(trampoline_header)
+SYM_DATA_START(trampoline_header)
+SYM_DATA_SIMPLE_LOCAL(tr_start,	.space 8)
+SYM_DATA_SIMPLE(tr_efer,	.space 8)
+SYM_DATA_SIMPLE(tr_cr4,		.space 4)
+SYM_DATA_END(trampoline_header)
 
 #include "trampoline_common.S"
diff --git a/arch/x86/realmode/rm/trampoline_common.S b/arch/x86/realmode/rm/trampoline_common.S
index b1ecdb9692ad..3cd403a94245 100644
--- a/arch/x86/realmode/rm/trampoline_common.S
+++ b/arch/x86/realmode/rm/trampoline_common.S
@@ -1,7 +1,7 @@
 	.section ".rodata","a"
 	.balign	16
-tr_idt: .fill 1, 6, 0
+SYM_DATA_SIMPLE_LOCAL(tr_idt, .fill 1, 6, 0)
 
 	.bss
 	.balign	4
-GLOBAL(trampoline_status)	.space	4
+SYM_DATA_SIMPLE(trampoline_status, .space 4)
diff --git a/arch/x86/realmode/rm/wakeup_asm.S b/arch/x86/realmode/rm/wakeup_asm.S
index 08203a187446..41483fd2d247 100644
--- a/arch/x86/realmode/rm/wakeup_asm.S
+++ b/arch/x86/realmode/rm/wakeup_asm.S
@@ -16,7 +16,7 @@
 	.section ".data", "aw"
 
 	.balign	16
-GLOBAL(wakeup_header)
+SYM_DATA_START(wakeup_header)
 	video_mode:	.short	0	/* Video mode number */
 	pmode_entry:	.long	0
 	pmode_cs:	.short	__KERNEL_CS
@@ -30,7 +30,7 @@ GLOBAL(wakeup_header)
 	realmode_flags:	.long	0
 	real_magic:	.long	0
 	signature:	.long	WAKEUP_HEADER_SIGNATURE
-END(wakeup_header)
+SYM_DATA_END(wakeup_header)
 
 	.text
 	.code16
@@ -152,7 +152,7 @@ bogus_real_magic:
 	 */
 
 	.balign	16
-GLOBAL(wakeup_gdt)
+SYM_DATA_START(wakeup_gdt)
 	.word	3*8-1		/* Self-descriptor */
 	.long	pa_wakeup_gdt
 	.word	0
@@ -164,15 +164,15 @@ GLOBAL(wakeup_gdt)
 	.word	0xffff		/* 16-bit data segment @ real_mode_base */
 	.long	0x93000000 + pa_real_mode_base
 	.word	0x008f		/* big real mode */
-END(wakeup_gdt)
+SYM_DATA_END(wakeup_gdt)
 
 	.section ".rodata","a"
 	.balign	8
 
 	/* This is the standard real-mode IDT */
 	.balign	16
-GLOBAL(wakeup_idt)
+SYM_DATA_START_LOCAL(wakeup_idt)
 	.word	0xffff		/* limit */
 	.long	0		/* address */
 	.word	0
-END(wakeup_idt)
+SYM_DATA_END(wakeup_idt)
diff --git a/arch/x86/realmode/rmpiggy.S b/arch/x86/realmode/rmpiggy.S
index 204c6ece0e97..644cc3dc9f23 100644
--- a/arch/x86/realmode/rmpiggy.S
+++ b/arch/x86/realmode/rmpiggy.S
@@ -9,12 +9,10 @@
 
 	.balign PAGE_SIZE
 
-GLOBAL(real_mode_blob)
+SYM_DATA_START(real_mode_blob)
 	.incbin	"arch/x86/realmode/rm/realmode.bin"
-END(real_mode_blob)
+SYM_DATA_END_LABEL(real_mode_blob, SYM_V_GLOBAL, real_mode_blob_end)
 
-GLOBAL(real_mode_blob_end);
-
-GLOBAL(real_mode_relocs)
+SYM_DATA_START(real_mode_relocs)
 	.incbin	"arch/x86/realmode/rm/realmode.relocs"
-END(real_mode_relocs)
+SYM_DATA_END(real_mode_relocs)
diff --git a/include/linux/linkage.h b/include/linux/linkage.h
index 31237e9db93f..27af3543fbc9 100644
--- a/include/linux/linkage.h
+++ b/include/linux/linkage.h
@@ -112,11 +112,13 @@
 	SYM_FUNC_START_WEAK(name)
 #endif
 
+#ifndef CONFIG_X86
 #ifndef END
 /* deprecated, use SYM_FUNC_END, SYM_DATA_END, or SYM_END */
 #define END(name) \
 	.size name, .-name
 #endif
+#endif /* CONFIG_X86 */
 
 /* If symbol 'name' is treated as a subroutine (gets called, and returns)
  * then please use ENDPROC to mark 'name' as STT_FUNC for the benefit of
-- 
2.12.2

[PATCH v3 24/29] x86: assembly, remove GLOBAL macro

[Jiri Slaby]

Convert the remaining 32bit users and remove GLOBAL macro finally. In
particular, this means to use SYM_ENTRY for the singlestepping hack
region.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
---
 arch/x86/entry/entry_32.S      | 4 ++--
 arch/x86/include/asm/linkage.h | 3 ---
 2 files changed, 2 insertions(+), 5 deletions(-)

diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S
index acc8d5ad5264..f1babb84510b 100644
--- a/arch/x86/entry/entry_32.S
+++ b/arch/x86/entry/entry_32.S
@@ -337,7 +337,7 @@ ENTRY(resume_kernel)
 ENDPROC(resume_kernel)
 #endif
 
-GLOBAL(__begin_SYSENTER_singlestep_region)
+SYM_ENTRY(__begin_SYSENTER_singlestep_region, SYM_V_GLOBAL, SYM_A_NONE)
 /*
  * All code from here through __end_SYSENTER_singlestep_region is subject
  * to being single-stepped if a user program sets TF and executes SYSENTER.
@@ -477,7 +477,7 @@ ENTRY(entry_SYSENTER_32)
 	pushl	$X86_EFLAGS_FIXED
 	popfl
 	jmp	.Lsysenter_flags_fixed
-GLOBAL(__end_SYSENTER_singlestep_region)
+SYM_ENTRY(__end_SYSENTER_singlestep_region, SYM_V_GLOBAL, SYM_A_NONE)
 ENDPROC(entry_SYSENTER_32)
 
 /*
diff --git a/arch/x86/include/asm/linkage.h b/arch/x86/include/asm/linkage.h
index de41e51a9ab4..06d0208f73df 100644
--- a/arch/x86/include/asm/linkage.h
+++ b/arch/x86/include/asm/linkage.h
@@ -12,9 +12,6 @@
 
 #ifdef __ASSEMBLY__
 
-/* deprecated, use SYM_DATA_START, SYM_FUNC_START, or SYM_FUNC_INNER_LABEL */
-#define GLOBAL(name)	SYM_ENTRY(name, SYM_V_GLOBAL, SYM_A_NONE)
-
 #if defined(CONFIG_X86_64) || defined(CONFIG_X86_ALIGNMENT_16)
 #define __ALIGN		.p2align 4, 0x90
 #define __ALIGN_STR	__stringify(__ALIGN)
-- 
2.12.2

[PATCH v3 25/29] x86: assembly, make some functions local

[Jiri Slaby]

There is couple of assembly functions, which are invoked only locally in
a file they are defined. In C, we mark them "static". Annotate them here
using SYM_FUNC_START_LOCAL (and switch their ENDPROC to SYM_FUNC_END
too).

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: x86@kernel.org
Cc: Matt Fleming <matt@codeblueprint.co.uk>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: linux-efi@vger.kernel.org
Cc: xen-devel@lists.xenproject.org
---
 arch/x86/boot/compressed/efi_thunk_64.S |  8 ++++----
 arch/x86/boot/pmjump.S                  |  4 ++--
 arch/x86/entry/entry_64.S               | 25 +++++++++++++------------
 arch/x86/lib/copy_page_64.S             |  4 ++--
 arch/x86/lib/memcpy_64.S                | 12 ++++++------
 arch/x86/lib/memset_64.S                |  8 ++++----
 arch/x86/platform/efi/efi_thunk_64.S    | 12 ++++++------
 arch/x86/xen/xen-pvh.S                  |  4 ++--
 8 files changed, 39 insertions(+), 38 deletions(-)

diff --git a/arch/x86/boot/compressed/efi_thunk_64.S b/arch/x86/boot/compressed/efi_thunk_64.S
index 86528f120962..c072711d8d62 100644
--- a/arch/x86/boot/compressed/efi_thunk_64.S
+++ b/arch/x86/boot/compressed/efi_thunk_64.S
@@ -98,12 +98,12 @@ ENTRY(efi64_thunk)
 	ret
 ENDPROC(efi64_thunk)
 
-ENTRY(efi_exit32)
+SYM_FUNC_START_LOCAL(efi_exit32)
 	movq	func_rt_ptr(%rip), %rax
 	push	%rax
 	mov	%rdi, %rax
 	ret
-ENDPROC(efi_exit32)
+SYM_FUNC_END(efi_exit32)
 
 	.code32
 /*
@@ -111,7 +111,7 @@ ENDPROC(efi_exit32)
  *
  * The stack should represent the 32-bit calling convention.
  */
-ENTRY(efi_enter32)
+SYM_FUNC_START_LOCAL(efi_enter32)
 	movl	$__KERNEL_DS, %eax
 	movl	%eax, %ds
 	movl	%eax, %es
@@ -171,7 +171,7 @@ ENTRY(efi_enter32)
 	btsl	$X86_CR0_PG_BIT, %eax
 	movl	%eax, %cr0
 	lret
-ENDPROC(efi_enter32)
+SYM_FUNC_END(efi_enter32)
 
 	.data
 	.balign	8
diff --git a/arch/x86/boot/pmjump.S b/arch/x86/boot/pmjump.S
index 6528f78a79b5..da86f4df8ffb 100644
--- a/arch/x86/boot/pmjump.S
+++ b/arch/x86/boot/pmjump.S
@@ -48,7 +48,7 @@ ENDPROC(protected_mode_jump)
 
 	.code32
 	.section ".text32","ax"
-ENTRY(in_pm32)
+SYM_FUNC_START_LOCAL(in_pm32)
 	# Set up data segments for flat 32-bit mode
 	movl	%ecx, %ds
 	movl	%ecx, %es
@@ -74,4 +74,4 @@ ENTRY(in_pm32)
 	lldt	%cx
 
 	jmpl	*%eax			# Jump to the 32-bit entrypoint
-ENDPROC(in_pm32)
+SYM_FUNC_END(in_pm32)
diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
index 213127a44c7c..ab71baad00fb 100644
--- a/arch/x86/entry/entry_64.S
+++ b/arch/x86/entry/entry_64.S
@@ -323,7 +323,7 @@ opportunistic_sysret_failed:
 	jmp	restore_c_regs_and_iret
 ENDPROC(entry_SYSCALL_64)
 
-ENTRY(stub_ptregs_64)
+SYM_FUNC_START_LOCAL(stub_ptregs_64)
 	/*
 	 * Syscalls marked as needing ptregs land here.
 	 * If we are on the fast path, we need to save the extra regs,
@@ -347,7 +347,7 @@ ENTRY(stub_ptregs_64)
 
 1:
 	jmp	*%rax				/* Called from C */
-ENDPROC(stub_ptregs_64)
+SYM_FUNC_END(stub_ptregs_64)
 
 .macro ptregs_stub func
 ENTRY(ptregs_\func)
@@ -918,7 +918,8 @@ idtentry xen_hypervisor_callback xen_do_hypervisor_callback has_error_code=0
  * existing activation in its critical region -- if so, we pop the current
  * activation and restart the handler using the previous one.
  */
-ENTRY(xen_do_hypervisor_callback)		/* do_hypervisor_callback(struct *pt_regs) */
+/* do_hypervisor_callback(struct *pt_regs) */
+SYM_FUNC_START_LOCAL(xen_do_hypervisor_callback)
 
 /*
  * Since we don't modify %rdi, evtchn_do_upall(struct *pt_regs) will
@@ -936,7 +937,7 @@ ENTRY(xen_do_hypervisor_callback)		/* do_hypervisor_callback(struct *pt_regs) */
 	call	xen_maybe_preempt_hcall
 #endif
 	jmp	error_exit
-ENDPROC(xen_do_hypervisor_callback)
+SYM_FUNC_END(xen_do_hypervisor_callback)
 
 /*
  * Hypervisor uses this for application faults while it executes.
@@ -1020,7 +1021,7 @@ idtentry machine_check					has_error_code=0	paranoid=1 do_sym=*machine_check_vec
  * Use slow, but surefire "are we in kernel?" check.
  * Return: ebx=0: need swapgs on exit, ebx=1: otherwise
  */
-ENTRY(paranoid_entry)
+SYM_FUNC_START_LOCAL(paranoid_entry)
 	cld
 	SAVE_C_REGS 8
 	SAVE_EXTRA_REGS 8
@@ -1033,7 +1034,7 @@ ENTRY(paranoid_entry)
 	SWAPGS
 	xorl	%ebx, %ebx
 1:	ret
-ENDPROC(paranoid_entry)
+SYM_FUNC_END(paranoid_entry)
 
 /*
  * "Paranoid" exit path from exception stack.  This is invoked
@@ -1047,7 +1048,7 @@ ENDPROC(paranoid_entry)
  *
  * On entry, ebx is "no swapgs" flag (1: don't need swapgs, 0: need it)
  */
-ENTRY(paranoid_exit)
+SYM_FUNC_START_LOCAL(paranoid_exit)
 	DISABLE_INTERRUPTS(CLBR_ANY)
 	TRACE_IRQS_OFF_DEBUG
 	testl	%ebx, %ebx			/* swapgs needed? */
@@ -1062,13 +1063,13 @@ paranoid_exit_restore:
 	RESTORE_C_REGS
 	REMOVE_PT_GPREGS_FROM_STACK 8
 	INTERRUPT_RETURN
-ENDPROC(paranoid_exit)
+SYM_FUNC_END(paranoid_exit)
 
 /*
  * Save all registers in pt_regs, and switch gs if needed.
  * Return: EBX=0: came from user mode; EBX=1: otherwise
  */
-ENTRY(error_entry)
+SYM_FUNC_START_LOCAL(error_entry)
 	cld
 	SAVE_C_REGS 8
 	SAVE_EXTRA_REGS 8
@@ -1144,7 +1145,7 @@ ENTRY(error_entry)
 	mov	%rax, %rsp
 	decl	%ebx
 	jmp	.Lerror_entry_from_usermode_after_swapgs
-ENDPROC(error_entry)
+SYM_FUNC_END(error_entry)
 
 
 /*
@@ -1152,13 +1153,13 @@ ENDPROC(error_entry)
  *   1: already in kernel mode, don't need SWAPGS
  *   0: user gsbase is loaded, we need SWAPGS and standard preparation for return to usermode
  */
-ENTRY(error_exit)
+SYM_FUNC_START_LOCAL(error_exit)
 	DISABLE_INTERRUPTS(CLBR_ANY)
 	TRACE_IRQS_OFF
 	testl	%ebx, %ebx
 	jnz	retint_kernel
 	jmp	retint_user
-ENDPROC(error_exit)
+SYM_FUNC_END(error_exit)
 
 /* Runs on exception stack */
 ENTRY(nmi)
diff --git a/arch/x86/lib/copy_page_64.S b/arch/x86/lib/copy_page_64.S
index e8508156c99d..e1ee50bc161a 100644
--- a/arch/x86/lib/copy_page_64.S
+++ b/arch/x86/lib/copy_page_64.S
@@ -20,7 +20,7 @@ ENTRY(copy_page)
 ENDPROC(copy_page)
 EXPORT_SYMBOL(copy_page)
 
-ENTRY(copy_page_regs)
+SYM_FUNC_START_LOCAL(copy_page_regs)
 	subq	$2*8,	%rsp
 	movq	%rbx,	(%rsp)
 	movq	%r12,	1*8(%rsp)
@@ -85,4 +85,4 @@ ENTRY(copy_page_regs)
 	movq	1*8(%rsp), %r12
 	addq	$2*8, %rsp
 	ret
-ENDPROC(copy_page_regs)
+SYM_FUNC_END(copy_page_regs)
diff --git a/arch/x86/lib/memcpy_64.S b/arch/x86/lib/memcpy_64.S
index 4911b1c61aa8..728703c47d58 100644
--- a/arch/x86/lib/memcpy_64.S
+++ b/arch/x86/lib/memcpy_64.S
@@ -27,7 +27,7 @@
  * rax original destination
  */
 SYM_FUNC_START_ALIAS(__memcpy)
-ENTRY(memcpy)
+SYM_FUNC_START_LOCAL(memcpy)
 	ALTERNATIVE_2 "jmp memcpy_orig", "", X86_FEATURE_REP_GOOD, \
 		      "jmp memcpy_erms", X86_FEATURE_ERMS
 
@@ -39,7 +39,7 @@ ENTRY(memcpy)
 	movl %edx, %ecx
 	rep movsb
 	ret
-ENDPROC(memcpy)
+SYM_FUNC_END(memcpy)
 SYM_FUNC_END_ALIAS(__memcpy)
 EXPORT_SYMBOL(memcpy)
 EXPORT_SYMBOL(__memcpy)
@@ -48,14 +48,14 @@ EXPORT_SYMBOL(__memcpy)
  * memcpy_erms() - enhanced fast string memcpy. This is faster and
  * simpler than memcpy. Use memcpy_erms when possible.
  */
-ENTRY(memcpy_erms)
+SYM_FUNC_START_LOCAL(memcpy_erms)
 	movq %rdi, %rax
 	movq %rdx, %rcx
 	rep movsb
 	ret
-ENDPROC(memcpy_erms)
+SYM_FUNC_END(memcpy_erms)
 
-ENTRY(memcpy_orig)
+SYM_FUNC_START_LOCAL(memcpy_orig)
 	movq %rdi, %rax
 
 	cmpq $0x20, %rdx
@@ -180,7 +180,7 @@ ENTRY(memcpy_orig)
 
 .Lend:
 	retq
-ENDPROC(memcpy_orig)
+SYM_FUNC_END(memcpy_orig)
 
 #ifndef CONFIG_UML
 /*
diff --git a/arch/x86/lib/memset_64.S b/arch/x86/lib/memset_64.S
index 0d3a1d341e60..c63ae9987612 100644
--- a/arch/x86/lib/memset_64.S
+++ b/arch/x86/lib/memset_64.S
@@ -58,16 +58,16 @@ EXPORT_SYMBOL(__memset)
  *
  * rax   original destination
  */
-ENTRY(memset_erms)
+SYM_FUNC_START_LOCAL(memset_erms)
 	movq %rdi,%r9
 	movb %sil,%al
 	movq %rdx,%rcx
 	rep stosb
 	movq %r9,%rax
 	ret
-ENDPROC(memset_erms)
+SYM_FUNC_END(memset_erms)
 
-ENTRY(memset_orig)
+SYM_FUNC_START_LOCAL(memset_orig)
 	movq %rdi,%r10
 
 	/* expand byte value  */
@@ -138,4 +138,4 @@ ENTRY(memset_orig)
 	subq %r8,%rdx
 	jmp .Lafter_bad_alignment
 .Lfinal:
-ENDPROC(memset_orig)
+SYM_FUNC_END(memset_orig)
diff --git a/arch/x86/platform/efi/efi_thunk_64.S b/arch/x86/platform/efi/efi_thunk_64.S
index ff85d28c50f2..d18697df1fe9 100644
--- a/arch/x86/platform/efi/efi_thunk_64.S
+++ b/arch/x86/platform/efi/efi_thunk_64.S
@@ -66,7 +66,7 @@ ENDPROC(efi64_thunk)
  *
  * This function must be invoked with a 1:1 mapped stack.
  */
-ENTRY(__efi64_thunk)
+SYM_FUNC_START_LOCAL(__efi64_thunk)
 	movl	%ds, %eax
 	push	%rax
 	movl	%es, %eax
@@ -113,14 +113,14 @@ ENTRY(__efi64_thunk)
 	or	%rcx, %rax
 1:
 	ret
-ENDPROC(__efi64_thunk)
+SYM_FUNC_END(__efi64_thunk)
 
-ENTRY(efi_exit32)
+SYM_FUNC_START_LOCAL(efi_exit32)
 	movq	func_rt_ptr(%rip), %rax
 	push	%rax
 	mov	%rdi, %rax
 	ret
-ENDPROC(efi_exit32)
+SYM_FUNC_END(efi_exit32)
 
 	.code32
 /*
@@ -128,7 +128,7 @@ ENDPROC(efi_exit32)
  *
  * The stack should represent the 32-bit calling convention.
  */
-ENTRY(efi_enter32)
+SYM_FUNC_START_LOCAL(efi_enter32)
 	movl	$__KERNEL_DS, %eax
 	movl	%eax, %ds
 	movl	%eax, %es
@@ -144,7 +144,7 @@ ENTRY(efi_enter32)
 	pushl	%eax
 
 	lret
-ENDPROC(efi_enter32)
+SYM_FUNC_END(efi_enter32)
 
 	.data
 	.balign	8
diff --git a/arch/x86/xen/xen-pvh.S b/arch/x86/xen/xen-pvh.S
index fa5ba8565646..469102d9c812 100644
--- a/arch/x86/xen/xen-pvh.S
+++ b/arch/x86/xen/xen-pvh.S
@@ -54,7 +54,7 @@
  * charge of setting up it's own stack, GDT and IDT.
  */
 
-ENTRY(pvh_start_xen)
+SYM_FUNC_START_LOCAL(pvh_start_xen)
 	cld
 
 	lgdt (_pa(gdt))
@@ -133,7 +133,7 @@ ENTRY(pvh_start_xen)
 
 	ljmp $__BOOT_CS, $_pa(startup_32)
 #endif
-ENDPROC(pvh_start_xen)
+SYM_FUNC_END(pvh_start_xen)
 
 	.section ".init.data","aw"
 	.balign 8
-- 
2.12.2

[PATCH v3 26/29] x86_64: assembly, change all ENTRY to SYM_FUNC_START

[Jiri Slaby]

These are all functions which are invoked from elsewhere, so we annotate
them as global using the new SYM_FUNC_START (and their ENDPROC's by
SYM_FUNC_END.)

And make sure ENTRY/ENDPROC is not defined on X86_64.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: x86@kernel.org
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Cc: Len Brown <len.brown@intel.com>
Cc: Pavel Machek <pavel@ucw.cz>
Cc: Bill Metzenthen <billm@melbpc.org.au>
Cc: Matt Fleming <matt@codeblueprint.co.uk>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: Juergen Gross <jgross@suse.com>
Cc: linux-crypto@vger.kernel.org
Cc: linux-pm@vger.kernel.org
Cc: linux-efi@vger.kernel.org
Cc: xen-devel@lists.xenproject.org
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: James Morris <jmorris@namei.org>
Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Cc: Patrick McHardy <kaber@trash.net>
Cc: netdev@vger.kernel.org
---
 arch/x86/boot/compressed/efi_thunk_64.S            |  4 +-
 arch/x86/boot/compressed/head_64.S                 | 20 ++++----
 arch/x86/boot/copy.S                               | 16 +++---
 arch/x86/boot/pmjump.S                             |  4 +-
 arch/x86/crypto/aes-i586-asm_32.S                  |  8 +--
 arch/x86/crypto/aes-x86_64-asm_64.S                |  4 +-
 arch/x86/crypto/aes_ctrby8_avx-x86_64.S            | 12 ++---
 arch/x86/crypto/aesni-intel_asm.S                  | 44 ++++++++--------
 arch/x86/crypto/aesni-intel_avx-x86_64.S           | 24 ++++-----
 arch/x86/crypto/blowfish-x86_64-asm_64.S           | 16 +++---
 arch/x86/crypto/camellia-aesni-avx-asm_64.S        | 24 ++++-----
 arch/x86/crypto/camellia-aesni-avx2-asm_64.S       | 24 ++++-----
 arch/x86/crypto/camellia-x86_64-asm_64.S           | 16 +++---
 arch/x86/crypto/cast5-avx-x86_64-asm_64.S          | 16 +++---
 arch/x86/crypto/cast6-avx-x86_64-asm_64.S          | 24 ++++-----
 arch/x86/crypto/chacha20-avx2-x86_64.S             |  4 +-
 arch/x86/crypto/chacha20-ssse3-x86_64.S            |  8 +--
 arch/x86/crypto/crc32-pclmul_asm.S                 |  4 +-
 arch/x86/crypto/crc32c-pcl-intel-asm_64.S          |  4 +-
 arch/x86/crypto/crct10dif-pcl-asm_64.S             |  4 +-
 arch/x86/crypto/des3_ede-asm_64.S                  |  8 +--
 arch/x86/crypto/ghash-clmulni-intel_asm.S          |  8 +--
 arch/x86/crypto/poly1305-avx2-x86_64.S             |  4 +-
 arch/x86/crypto/poly1305-sse2-x86_64.S             |  8 +--
 arch/x86/crypto/salsa20-x86_64-asm_64.S            | 12 ++---
 arch/x86/crypto/serpent-avx-x86_64-asm_64.S        | 24 ++++-----
 arch/x86/crypto/serpent-avx2-asm_64.S              | 24 ++++-----
 arch/x86/crypto/serpent-sse2-x86_64-asm_64.S       |  8 +--
 arch/x86/crypto/sha1-mb/sha1_mb_mgr_flush_avx2.S   |  8 +--
 arch/x86/crypto/sha1-mb/sha1_mb_mgr_submit_avx2.S  |  4 +-
 arch/x86/crypto/sha1-mb/sha1_x8_avx2.S             |  4 +-
 arch/x86/crypto/sha1_avx2_x86_64_asm.S             |  4 +-
 arch/x86/crypto/sha1_ni_asm.S                      |  4 +-
 arch/x86/crypto/sha1_ssse3_asm.S                   |  4 +-
 arch/x86/crypto/sha256-avx-asm.S                   |  4 +-
 arch/x86/crypto/sha256-avx2-asm.S                  |  4 +-
 .../crypto/sha256-mb/sha256_mb_mgr_flush_avx2.S    |  8 +--
 .../crypto/sha256-mb/sha256_mb_mgr_submit_avx2.S   |  4 +-
 arch/x86/crypto/sha256-mb/sha256_x8_avx2.S         |  4 +-
 arch/x86/crypto/sha256-ssse3-asm.S                 |  4 +-
 arch/x86/crypto/sha256_ni_asm.S                    |  4 +-
 arch/x86/crypto/sha512-avx-asm.S                   |  4 +-
 arch/x86/crypto/sha512-avx2-asm.S                  |  4 +-
 .../crypto/sha512-mb/sha512_mb_mgr_flush_avx2.S    |  8 +--
 .../crypto/sha512-mb/sha512_mb_mgr_submit_avx2.S   |  4 +-
 arch/x86/crypto/sha512-mb/sha512_x4_avx2.S         |  4 +-
 arch/x86/crypto/sha512-ssse3-asm.S                 |  4 +-
 arch/x86/crypto/twofish-avx-x86_64-asm_64.S        | 24 ++++-----
 arch/x86/crypto/twofish-x86_64-asm_64-3way.S       |  8 +--
 arch/x86/crypto/twofish-x86_64-asm_64.S            |  8 +--
 arch/x86/entry/entry_64.S                          | 58 +++++++++++-----------
 arch/x86/entry/entry_64_compat.S                   | 16 +++---
 arch/x86/kernel/acpi/wakeup_64.S                   |  8 +--
 arch/x86/kernel/ftrace_64.S                        | 24 ++++-----
 arch/x86/kernel/head_64.S                          | 16 +++---
 arch/x86/lib/checksum_32.S                         |  8 +--
 arch/x86/lib/clear_page_64.S                       | 12 ++---
 arch/x86/lib/cmpxchg16b_emu.S                      |  4 +-
 arch/x86/lib/cmpxchg8b_emu.S                       |  4 +-
 arch/x86/lib/copy_page_64.S                        |  4 +-
 arch/x86/lib/copy_user_64.S                        | 16 +++---
 arch/x86/lib/csum-copy_64.S                        |  4 +-
 arch/x86/lib/getuser.S                             | 16 +++---
 arch/x86/lib/hweight.S                             |  8 +--
 arch/x86/lib/iomap_copy_64.S                       |  4 +-
 arch/x86/lib/memcpy_64.S                           |  4 +-
 arch/x86/lib/memmove_64.S                          |  4 +-
 arch/x86/lib/memset_64.S                           |  4 +-
 arch/x86/lib/msr-reg.S                             |  8 +--
 arch/x86/lib/putuser.S                             | 16 +++---
 arch/x86/lib/rwsem.S                               | 20 ++++----
 arch/x86/net/bpf_jit.S                             | 36 +++++++-------
 arch/x86/platform/efi/efi_stub_64.S                |  4 +-
 arch/x86/platform/efi/efi_thunk_64.S               |  4 +-
 arch/x86/platform/olpc/xo1-wakeup.S                |  4 +-
 arch/x86/power/hibernate_asm_64.S                  | 16 +++---
 arch/x86/realmode/rm/reboot.S                      |  4 +-
 arch/x86/realmode/rm/trampoline_64.S               | 12 ++---
 arch/x86/realmode/rm/wakeup_asm.S                  |  4 +-
 arch/x86/xen/xen-asm.S                             | 20 ++++----
 arch/x86/xen/xen-asm_64.S                          | 28 +++++------
 arch/x86/xen/xen-head.S                            |  8 +--
 include/linux/linkage.h                            |  4 ++
 83 files changed, 455 insertions(+), 451 deletions(-)

diff --git a/arch/x86/boot/compressed/efi_thunk_64.S b/arch/x86/boot/compressed/efi_thunk_64.S
index c072711d8d62..b85b49c36da3 100644
--- a/arch/x86/boot/compressed/efi_thunk_64.S
+++ b/arch/x86/boot/compressed/efi_thunk_64.S
@@ -22,7 +22,7 @@
 
 	.code64
 	.text
-ENTRY(efi64_thunk)
+SYM_FUNC_START(efi64_thunk)
 	push	%rbp
 	push	%rbx
 
@@ -96,7 +96,7 @@ ENTRY(efi64_thunk)
 	pop	%rbx
 	pop	%rbp
 	ret
-ENDPROC(efi64_thunk)
+SYM_FUNC_END(efi64_thunk)
 
 SYM_FUNC_START_LOCAL(efi_exit32)
 	movq	func_rt_ptr(%rip), %rax
diff --git a/arch/x86/boot/compressed/head_64.S b/arch/x86/boot/compressed/head_64.S
index 6eb5a50a301e..34386ba83aef 100644
--- a/arch/x86/boot/compressed/head_64.S
+++ b/arch/x86/boot/compressed/head_64.S
@@ -43,7 +43,7 @@
 
 	__HEAD
 	.code32
-ENTRY(startup_32)
+SYM_FUNC_START(startup_32)
 	/*
 	 * 32bit entry is 0 and it is ABI so immutable!
 	 * If we come here directly from a bootloader,
@@ -204,11 +204,11 @@ ENTRY(startup_32)
 
 	/* Jump from 32bit compatibility mode into 64bit mode. */
 	lret
-ENDPROC(startup_32)
+SYM_FUNC_END(startup_32)
 
 #ifdef CONFIG_EFI_MIXED
 	.org 0x190
-ENTRY(efi32_stub_entry)
+SYM_FUNC_START(efi32_stub_entry)
 	add	$0x4, %esp		/* Discard return address */
 	popl	%ecx
 	popl	%edx
@@ -227,12 +227,12 @@ ENTRY(efi32_stub_entry)
 	movl	%eax, efi_config(%ebp)
 
 	jmp	startup_32
-ENDPROC(efi32_stub_entry)
+SYM_FUNC_END(efi32_stub_entry)
 #endif
 
 	.code64
 	.org 0x200
-ENTRY(startup_64)
+SYM_FUNC_START(startup_64)
 	/*
 	 * 64bit entry is 0x200 and it is ABI so immutable!
 	 * We come here either from startup_32 or directly from a
@@ -310,12 +310,12 @@ ENTRY(startup_64)
  */
 	leaq	relocated(%rbx), %rax
 	jmp	*%rax
-ENDPROC(startup_64)
+SYM_FUNC_END(startup_64)
 
 #ifdef CONFIG_EFI_STUB
 
 /* The entry point for the PE/COFF executable is efi_pe_entry. */
-ENTRY(efi_pe_entry)
+SYM_FUNC_START(efi_pe_entry)
 	movq	%rcx, efi64_config(%rip)	/* Handle */
 	movq	%rdx, efi64_config+8(%rip) /* EFI System table pointer */
 
@@ -364,10 +364,10 @@ fail:
 	movl	BP_code32_start(%esi), %eax
 	leaq	startup_64(%rax), %rax
 	jmp	*%rax
-ENDPROC(efi_pe_entry)
+SYM_FUNC_END(efi_pe_entry)
 
 	.org 0x390
-ENTRY(efi64_stub_entry)
+SYM_FUNC_START(efi64_stub_entry)
 	movq	%rdi, efi64_config(%rip)	/* Handle */
 	movq	%rsi, efi64_config+8(%rip) /* EFI System table pointer */
 
@@ -376,7 +376,7 @@ ENTRY(efi64_stub_entry)
 
 	movq	%rdx, %rsi
 	jmp	handover_entry
-ENDPROC(efi64_stub_entry)
+SYM_FUNC_END(efi64_stub_entry)
 #endif
 
 	.text
diff --git a/arch/x86/boot/copy.S b/arch/x86/boot/copy.S
index 030a7bde51da..d44ede19f729 100644
--- a/arch/x86/boot/copy.S
+++ b/arch/x86/boot/copy.S
@@ -17,7 +17,7 @@
 	.code16
 	.text
 
-ENTRY(memcpy)
+SYM_FUNC_START(memcpy)
 	pushw	%si
 	pushw	%di
 	movw	%ax, %di
@@ -31,9 +31,9 @@ ENTRY(memcpy)
 	popw	%di
 	popw	%si
 	retl
-ENDPROC(memcpy)
+SYM_FUNC_END(memcpy)
 
-ENTRY(memset)
+SYM_FUNC_START(memset)
 	pushw	%di
 	movw	%ax, %di
 	movzbl	%dl, %eax
@@ -46,22 +46,22 @@ ENTRY(memset)
 	rep; stosb
 	popw	%di
 	retl
-ENDPROC(memset)
+SYM_FUNC_END(memset)
 
-ENTRY(copy_from_fs)
+SYM_FUNC_START(copy_from_fs)
 	pushw	%ds
 	pushw	%fs
 	popw	%ds
 	calll	memcpy
 	popw	%ds
 	retl
-ENDPROC(copy_from_fs)
+SYM_FUNC_END(copy_from_fs)
 
-ENTRY(copy_to_fs)
+SYM_FUNC_START(copy_to_fs)
 	pushw	%es
 	pushw	%fs
 	popw	%es
 	calll	memcpy
 	popw	%es
 	retl
-ENDPROC(copy_to_fs)
+SYM_FUNC_END(copy_to_fs)
diff --git a/arch/x86/boot/pmjump.S b/arch/x86/boot/pmjump.S
index da86f4df8ffb..1aae6a3b38ab 100644
--- a/arch/x86/boot/pmjump.S
+++ b/arch/x86/boot/pmjump.S
@@ -23,7 +23,7 @@
 /*
  * void protected_mode_jump(u32 entrypoint, u32 bootparams);
  */
-ENTRY(protected_mode_jump)
+SYM_FUNC_START(protected_mode_jump)
 	movl	%edx, %esi		# Pointer to boot_params table
 
 	xorl	%ebx, %ebx
@@ -44,7 +44,7 @@ ENTRY(protected_mode_jump)
 	.byte	0x66, 0xea		# ljmpl opcode
 2:	.long	in_pm32			# offset
 	.word	__BOOT_CS		# segment
-ENDPROC(protected_mode_jump)
+SYM_FUNC_END(protected_mode_jump)
 
 	.code32
 	.section ".text32","ax"
diff --git a/arch/x86/crypto/aes-i586-asm_32.S b/arch/x86/crypto/aes-i586-asm_32.S
index 2849dbc59e11..5b2636c58527 100644
--- a/arch/x86/crypto/aes-i586-asm_32.S
+++ b/arch/x86/crypto/aes-i586-asm_32.S
@@ -223,7 +223,7 @@
 .extern  crypto_ft_tab
 .extern  crypto_fl_tab
 
-ENTRY(aes_enc_blk)
+SYM_FUNC_START(aes_enc_blk)
 	push    %ebp
 	mov     ctx(%esp),%ebp
 
@@ -287,7 +287,7 @@ ENTRY(aes_enc_blk)
 	mov     %r0,(%ebp)
 	pop     %ebp
 	ret
-ENDPROC(aes_enc_blk)
+SYM_FUNC_END(aes_enc_blk)
 
 // AES (Rijndael) Decryption Subroutine
 /* void aes_dec_blk(struct crypto_aes_ctx *ctx, u8 *out_blk, const u8 *in_blk) */
@@ -295,7 +295,7 @@ ENDPROC(aes_enc_blk)
 .extern  crypto_it_tab
 .extern  crypto_il_tab
 
-ENTRY(aes_dec_blk)
+SYM_FUNC_START(aes_dec_blk)
 	push    %ebp
 	mov     ctx(%esp),%ebp
 
@@ -359,4 +359,4 @@ ENTRY(aes_dec_blk)
 	mov     %r0,(%ebp)
 	pop     %ebp
 	ret
-ENDPROC(aes_dec_blk)
+SYM_FUNC_END(aes_dec_blk)
diff --git a/arch/x86/crypto/aes-x86_64-asm_64.S b/arch/x86/crypto/aes-x86_64-asm_64.S
index 910565547163..bd02ef4a10fe 100644
--- a/arch/x86/crypto/aes-x86_64-asm_64.S
+++ b/arch/x86/crypto/aes-x86_64-asm_64.S
@@ -50,7 +50,7 @@
 #define R11	%r11
 
 #define prologue(FUNC,KEY,B128,B192,r1,r2,r3,r4,r5,r6,r7,r8,r9,r10,r11) \
-	ENTRY(FUNC);			\
+	SYM_FUNC_START(FUNC);		\
 	movq	r1,r2;			\
 	movq	r3,r4;			\
 	leaq	KEY+48(r8),r9;		\
@@ -78,7 +78,7 @@
 	movl	r7 ## E,8(r9);		\
 	movl	r8 ## E,12(r9);		\
 	ret;				\
-	ENDPROC(FUNC);
+	SYM_FUNC_END(FUNC);
 
 #define round(TAB,OFFSET,r1,r2,r3,r4,r5,r6,r7,r8,ra,rb,rc,rd) \
 	movzbl	r2 ## H,r5 ## E;	\
diff --git a/arch/x86/crypto/aes_ctrby8_avx-x86_64.S b/arch/x86/crypto/aes_ctrby8_avx-x86_64.S
index 5f6a5af9c489..ec437db1fa54 100644
--- a/arch/x86/crypto/aes_ctrby8_avx-x86_64.S
+++ b/arch/x86/crypto/aes_ctrby8_avx-x86_64.S
@@ -544,11 +544,11 @@ ddq_add_8:
  * aes_ctr_enc_128_avx_by8(void *in, void *iv, void *keys, void *out,
  *			unsigned int num_bytes)
  */
-ENTRY(aes_ctr_enc_128_avx_by8)
+SYM_FUNC_START(aes_ctr_enc_128_avx_by8)
 	/* call the aes main loop */
 	do_aes_ctrmain KEY_128
 
-ENDPROC(aes_ctr_enc_128_avx_by8)
+SYM_FUNC_END(aes_ctr_enc_128_avx_by8)
 
 /*
  * routine to do AES192 CTR enc/decrypt "by8"
@@ -557,11 +557,11 @@ ENDPROC(aes_ctr_enc_128_avx_by8)
  * aes_ctr_enc_192_avx_by8(void *in, void *iv, void *keys, void *out,
  *			unsigned int num_bytes)
  */
-ENTRY(aes_ctr_enc_192_avx_by8)
+SYM_FUNC_START(aes_ctr_enc_192_avx_by8)
 	/* call the aes main loop */
 	do_aes_ctrmain KEY_192
 
-ENDPROC(aes_ctr_enc_192_avx_by8)
+SYM_FUNC_END(aes_ctr_enc_192_avx_by8)
 
 /*
  * routine to do AES256 CTR enc/decrypt "by8"
@@ -570,8 +570,8 @@ ENDPROC(aes_ctr_enc_192_avx_by8)
  * aes_ctr_enc_256_avx_by8(void *in, void *iv, void *keys, void *out,
  *			unsigned int num_bytes)
  */
-ENTRY(aes_ctr_enc_256_avx_by8)
+SYM_FUNC_START(aes_ctr_enc_256_avx_by8)
 	/* call the aes main loop */
 	do_aes_ctrmain KEY_256
 
-ENDPROC(aes_ctr_enc_256_avx_by8)
+SYM_FUNC_END(aes_ctr_enc_256_avx_by8)
diff --git a/arch/x86/crypto/aesni-intel_asm.S b/arch/x86/crypto/aesni-intel_asm.S
index 3469670df832..934d49f71b48 100644
--- a/arch/x86/crypto/aesni-intel_asm.S
+++ b/arch/x86/crypto/aesni-intel_asm.S
@@ -1301,7 +1301,7 @@ _esb_loop_\@:
 * poly = x^128 + x^127 + x^126 + x^121 + 1
 *
 *****************************************************************************/
-ENTRY(aesni_gcm_dec)
+SYM_FUNC_START(aesni_gcm_dec)
 	push	%r12
 	push	%r13
 	push	%r14
@@ -1475,7 +1475,7 @@ _return_T_done_decrypt:
 	pop	%r13
 	pop	%r12
 	ret
-ENDPROC(aesni_gcm_dec)
+SYM_FUNC_END(aesni_gcm_dec)
 
 
 /*****************************************************************************
@@ -1561,7 +1561,7 @@ ENDPROC(aesni_gcm_dec)
 *
 * poly = x^128 + x^127 + x^126 + x^121 + 1
 ***************************************************************************/
-ENTRY(aesni_gcm_enc)
+SYM_FUNC_START(aesni_gcm_enc)
 	push	%r12
 	push	%r13
 	push	%r14
@@ -1739,7 +1739,7 @@ _return_T_done_encrypt:
 	pop	%r13
 	pop	%r12
 	ret
-ENDPROC(aesni_gcm_enc)
+SYM_FUNC_END(aesni_gcm_enc)
 
 #endif
 
@@ -1817,7 +1817,7 @@ SYM_FUNC_END(_key_expansion_256b)
  * int aesni_set_key(struct crypto_aes_ctx *ctx, const u8 *in_key,
  *                   unsigned int key_len)
  */
-ENTRY(aesni_set_key)
+SYM_FUNC_START(aesni_set_key)
 	FRAME_BEGIN
 #ifndef __x86_64__
 	pushl KEYP
@@ -1926,12 +1926,12 @@ ENTRY(aesni_set_key)
 #endif
 	FRAME_END
 	ret
-ENDPROC(aesni_set_key)
+SYM_FUNC_END(aesni_set_key)
 
 /*
  * void aesni_enc(struct crypto_aes_ctx *ctx, u8 *dst, const u8 *src)
  */
-ENTRY(aesni_enc)
+SYM_FUNC_START(aesni_enc)
 	FRAME_BEGIN
 #ifndef __x86_64__
 	pushl KEYP
@@ -1950,7 +1950,7 @@ ENTRY(aesni_enc)
 #endif
 	FRAME_END
 	ret
-ENDPROC(aesni_enc)
+SYM_FUNC_END(aesni_enc)
 
 /*
  * _aesni_enc1:		internal ABI
@@ -2120,7 +2120,7 @@ SYM_FUNC_END(_aesni_enc4)
 /*
  * void aesni_dec (struct crypto_aes_ctx *ctx, u8 *dst, const u8 *src)
  */
-ENTRY(aesni_dec)
+SYM_FUNC_START(aesni_dec)
 	FRAME_BEGIN
 #ifndef __x86_64__
 	pushl KEYP
@@ -2140,7 +2140,7 @@ ENTRY(aesni_dec)
 #endif
 	FRAME_END
 	ret
-ENDPROC(aesni_dec)
+SYM_FUNC_END(aesni_dec)
 
 /*
  * _aesni_dec1:		internal ABI
@@ -2311,7 +2311,7 @@ SYM_FUNC_END(_aesni_dec4)
  * void aesni_ecb_enc(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
  *		      size_t len)
  */
-ENTRY(aesni_ecb_enc)
+SYM_FUNC_START(aesni_ecb_enc)
 	FRAME_BEGIN
 #ifndef __x86_64__
 	pushl LEN
@@ -2365,13 +2365,13 @@ ENTRY(aesni_ecb_enc)
 #endif
 	FRAME_END
 	ret
-ENDPROC(aesni_ecb_enc)
+SYM_FUNC_END(aesni_ecb_enc)
 
 /*
  * void aesni_ecb_dec(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
  *		      size_t len);
  */
-ENTRY(aesni_ecb_dec)
+SYM_FUNC_START(aesni_ecb_dec)
 	FRAME_BEGIN
 #ifndef __x86_64__
 	pushl LEN
@@ -2426,13 +2426,13 @@ ENTRY(aesni_ecb_dec)
 #endif
 	FRAME_END
 	ret
-ENDPROC(aesni_ecb_dec)
+SYM_FUNC_END(aesni_ecb_dec)
 
 /*
  * void aesni_cbc_enc(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
  *		      size_t len, u8 *iv)
  */
-ENTRY(aesni_cbc_enc)
+SYM_FUNC_START(aesni_cbc_enc)
 	FRAME_BEGIN
 #ifndef __x86_64__
 	pushl IVP
@@ -2470,13 +2470,13 @@ ENTRY(aesni_cbc_enc)
 #endif
 	FRAME_END
 	ret
-ENDPROC(aesni_cbc_enc)
+SYM_FUNC_END(aesni_cbc_enc)
 
 /*
  * void aesni_cbc_dec(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
  *		      size_t len, u8 *iv)
  */
-ENTRY(aesni_cbc_dec)
+SYM_FUNC_START(aesni_cbc_dec)
 	FRAME_BEGIN
 #ifndef __x86_64__
 	pushl IVP
@@ -2563,7 +2563,7 @@ ENTRY(aesni_cbc_dec)
 #endif
 	FRAME_END
 	ret
-ENDPROC(aesni_cbc_dec)
+SYM_FUNC_END(aesni_cbc_dec)
 
 #ifdef __x86_64__
 .pushsection .rodata
@@ -2625,7 +2625,7 @@ SYM_FUNC_END(_aesni_inc)
  * void aesni_ctr_enc(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
  *		      size_t len, u8 *iv)
  */
-ENTRY(aesni_ctr_enc)
+SYM_FUNC_START(aesni_ctr_enc)
 	FRAME_BEGIN
 	cmp $16, LEN
 	jb .Lctr_enc_just_ret
@@ -2682,7 +2682,7 @@ ENTRY(aesni_ctr_enc)
 .Lctr_enc_just_ret:
 	FRAME_END
 	ret
-ENDPROC(aesni_ctr_enc)
+SYM_FUNC_END(aesni_ctr_enc)
 
 /*
  * _aesni_gf128mul_x_ble:		internal ABI
@@ -2706,7 +2706,7 @@ ENDPROC(aesni_ctr_enc)
  * void aesni_xts_crypt8(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
  *			 bool enc, u8 *iv)
  */
-ENTRY(aesni_xts_crypt8)
+SYM_FUNC_START(aesni_xts_crypt8)
 	FRAME_BEGIN
 	cmpb $0, %cl
 	movl $0, %ecx
@@ -2810,6 +2810,6 @@ ENTRY(aesni_xts_crypt8)
 
 	FRAME_END
 	ret
-ENDPROC(aesni_xts_crypt8)
+SYM_FUNC_END(aesni_xts_crypt8)
 
 #endif
diff --git a/arch/x86/crypto/aesni-intel_avx-x86_64.S b/arch/x86/crypto/aesni-intel_avx-x86_64.S
index d664382c6e56..2a5228b675b2 100644
--- a/arch/x86/crypto/aesni-intel_avx-x86_64.S
+++ b/arch/x86/crypto/aesni-intel_avx-x86_64.S
@@ -1460,7 +1460,7 @@ _return_T_done\@:
 #        (gcm_data     *my_ctx_data,
 #        u8     *hash_subkey)# /* H, the Hash sub key input. Data starts on a 16-byte boundary. */
 #############################################################
-ENTRY(aesni_gcm_precomp_avx_gen2)
+SYM_FUNC_START(aesni_gcm_precomp_avx_gen2)
         #the number of pushes must equal STACK_OFFSET
         push    %r12
         push    %r13
@@ -1503,7 +1503,7 @@ ENTRY(aesni_gcm_precomp_avx_gen2)
         pop     %r13
         pop     %r12
         ret
-ENDPROC(aesni_gcm_precomp_avx_gen2)
+SYM_FUNC_END(aesni_gcm_precomp_avx_gen2)
 
 ###############################################################################
 #void   aesni_gcm_enc_avx_gen2(
@@ -1521,10 +1521,10 @@ ENDPROC(aesni_gcm_precomp_avx_gen2)
 #        u64     auth_tag_len)# /* Authenticated Tag Length in bytes.
 #				Valid values are 16 (most likely), 12 or 8. */
 ###############################################################################
-ENTRY(aesni_gcm_enc_avx_gen2)
+SYM_FUNC_START(aesni_gcm_enc_avx_gen2)
         GCM_ENC_DEC_AVX     ENC
 	ret
-ENDPROC(aesni_gcm_enc_avx_gen2)
+SYM_FUNC_END(aesni_gcm_enc_avx_gen2)
 
 ###############################################################################
 #void   aesni_gcm_dec_avx_gen2(
@@ -1542,10 +1542,10 @@ ENDPROC(aesni_gcm_enc_avx_gen2)
 #        u64     auth_tag_len)# /* Authenticated Tag Length in bytes.
 #				Valid values are 16 (most likely), 12 or 8. */
 ###############################################################################
-ENTRY(aesni_gcm_dec_avx_gen2)
+SYM_FUNC_START(aesni_gcm_dec_avx_gen2)
         GCM_ENC_DEC_AVX     DEC
 	ret
-ENDPROC(aesni_gcm_dec_avx_gen2)
+SYM_FUNC_END(aesni_gcm_dec_avx_gen2)
 #endif /* CONFIG_AS_AVX */
 
 #ifdef CONFIG_AS_AVX2
@@ -2736,7 +2736,7 @@ _return_T_done\@:
 #        u8     *hash_subkey)# /* H, the Hash sub key input.
 #				Data starts on a 16-byte boundary. */
 #############################################################
-ENTRY(aesni_gcm_precomp_avx_gen4)
+SYM_FUNC_START(aesni_gcm_precomp_avx_gen4)
         #the number of pushes must equal STACK_OFFSET
         push    %r12
         push    %r13
@@ -2779,7 +2779,7 @@ ENTRY(aesni_gcm_precomp_avx_gen4)
         pop     %r13
         pop     %r12
         ret
-ENDPROC(aesni_gcm_precomp_avx_gen4)
+SYM_FUNC_END(aesni_gcm_precomp_avx_gen4)
 
 
 ###############################################################################
@@ -2798,10 +2798,10 @@ ENDPROC(aesni_gcm_precomp_avx_gen4)
 #        u64     auth_tag_len)# /* Authenticated Tag Length in bytes.
 #				Valid values are 16 (most likely), 12 or 8. */
 ###############################################################################
-ENTRY(aesni_gcm_enc_avx_gen4)
+SYM_FUNC_START(aesni_gcm_enc_avx_gen4)
         GCM_ENC_DEC_AVX2     ENC
 	ret
-ENDPROC(aesni_gcm_enc_avx_gen4)
+SYM_FUNC_END(aesni_gcm_enc_avx_gen4)
 
 ###############################################################################
 #void   aesni_gcm_dec_avx_gen4(
@@ -2819,9 +2819,9 @@ ENDPROC(aesni_gcm_enc_avx_gen4)
 #        u64     auth_tag_len)# /* Authenticated Tag Length in bytes.
 #				Valid values are 16 (most likely), 12 or 8. */
 ###############################################################################
-ENTRY(aesni_gcm_dec_avx_gen4)
+SYM_FUNC_START(aesni_gcm_dec_avx_gen4)
         GCM_ENC_DEC_AVX2     DEC
 	ret
-ENDPROC(aesni_gcm_dec_avx_gen4)
+SYM_FUNC_END(aesni_gcm_dec_avx_gen4)
 
 #endif /* CONFIG_AS_AVX2 */
diff --git a/arch/x86/crypto/blowfish-x86_64-asm_64.S b/arch/x86/crypto/blowfish-x86_64-asm_64.S
index 246c67006ed0..fc96846d5707 100644
--- a/arch/x86/crypto/blowfish-x86_64-asm_64.S
+++ b/arch/x86/crypto/blowfish-x86_64-asm_64.S
@@ -118,7 +118,7 @@
 	bswapq 			RX0; \
 	xorq RX0, 		(RIO);
 
-ENTRY(__blowfish_enc_blk)
+SYM_FUNC_START(__blowfish_enc_blk)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -153,9 +153,9 @@ ENTRY(__blowfish_enc_blk)
 .L__enc_xor:
 	xor_block();
 	ret;
-ENDPROC(__blowfish_enc_blk)
+SYM_FUNC_END(__blowfish_enc_blk)
 
-ENTRY(blowfish_dec_blk)
+SYM_FUNC_START(blowfish_dec_blk)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -184,7 +184,7 @@ ENTRY(blowfish_dec_blk)
 	movq %r11, %rbp;
 
 	ret;
-ENDPROC(blowfish_dec_blk)
+SYM_FUNC_END(blowfish_dec_blk)
 
 /**********************************************************************
   4-way blowfish, four blocks parallel
@@ -296,7 +296,7 @@ ENDPROC(blowfish_dec_blk)
 	bswapq 			RX3; \
 	xorq RX3,		24(RIO);
 
-ENTRY(__blowfish_enc_blk_4way)
+SYM_FUNC_START(__blowfish_enc_blk_4way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -342,9 +342,9 @@ ENTRY(__blowfish_enc_blk_4way)
 	popq %rbx;
 	popq %rbp;
 	ret;
-ENDPROC(__blowfish_enc_blk_4way)
+SYM_FUNC_END(__blowfish_enc_blk_4way)
 
-ENTRY(blowfish_dec_blk_4way)
+SYM_FUNC_START(blowfish_dec_blk_4way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -376,4 +376,4 @@ ENTRY(blowfish_dec_blk_4way)
 	popq %rbp;
 
 	ret;
-ENDPROC(blowfish_dec_blk_4way)
+SYM_FUNC_END(blowfish_dec_blk_4way)
diff --git a/arch/x86/crypto/camellia-aesni-avx-asm_64.S b/arch/x86/crypto/camellia-aesni-avx-asm_64.S
index 8b6a65524067..c055e844203f 100644
--- a/arch/x86/crypto/camellia-aesni-avx-asm_64.S
+++ b/arch/x86/crypto/camellia-aesni-avx-asm_64.S
@@ -892,7 +892,7 @@ SYM_FUNC_START_LOCAL(__camellia_dec_blk16)
 	jmp .Ldec_max24;
 SYM_FUNC_END(__camellia_dec_blk16)
 
-ENTRY(camellia_ecb_enc_16way)
+SYM_FUNC_START(camellia_ecb_enc_16way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst (16 blocks)
@@ -915,9 +915,9 @@ ENTRY(camellia_ecb_enc_16way)
 
 	FRAME_END
 	ret;
-ENDPROC(camellia_ecb_enc_16way)
+SYM_FUNC_END(camellia_ecb_enc_16way)
 
-ENTRY(camellia_ecb_dec_16way)
+SYM_FUNC_START(camellia_ecb_dec_16way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst (16 blocks)
@@ -945,9 +945,9 @@ ENTRY(camellia_ecb_dec_16way)
 
 	FRAME_END
 	ret;
-ENDPROC(camellia_ecb_dec_16way)
+SYM_FUNC_END(camellia_ecb_dec_16way)
 
-ENTRY(camellia_cbc_dec_16way)
+SYM_FUNC_START(camellia_cbc_dec_16way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst (16 blocks)
@@ -996,7 +996,7 @@ ENTRY(camellia_cbc_dec_16way)
 
 	FRAME_END
 	ret;
-ENDPROC(camellia_cbc_dec_16way)
+SYM_FUNC_END(camellia_cbc_dec_16way)
 
 #define inc_le128(x, minus_one, tmp) \
 	vpcmpeqq minus_one, x, tmp; \
@@ -1004,7 +1004,7 @@ ENDPROC(camellia_cbc_dec_16way)
 	vpslldq $8, tmp, tmp; \
 	vpsubq tmp, x, x;
 
-ENTRY(camellia_ctr_16way)
+SYM_FUNC_START(camellia_ctr_16way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst (16 blocks)
@@ -1109,7 +1109,7 @@ ENTRY(camellia_ctr_16way)
 
 	FRAME_END
 	ret;
-ENDPROC(camellia_ctr_16way)
+SYM_FUNC_END(camellia_ctr_16way)
 
 #define gf128mul_x_ble(iv, mask, tmp) \
 	vpsrad $31, iv, tmp; \
@@ -1255,7 +1255,7 @@ SYM_FUNC_START_LOCAL(camellia_xts_crypt_16way)
 	ret;
 SYM_FUNC_END(camellia_xts_crypt_16way)
 
-ENTRY(camellia_xts_enc_16way)
+SYM_FUNC_START(camellia_xts_enc_16way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst (16 blocks)
@@ -1267,9 +1267,9 @@ ENTRY(camellia_xts_enc_16way)
 	leaq __camellia_enc_blk16, %r9;
 
 	jmp camellia_xts_crypt_16way;
-ENDPROC(camellia_xts_enc_16way)
+SYM_FUNC_END(camellia_xts_enc_16way)
 
-ENTRY(camellia_xts_dec_16way)
+SYM_FUNC_START(camellia_xts_dec_16way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst (16 blocks)
@@ -1285,4 +1285,4 @@ ENTRY(camellia_xts_dec_16way)
 	leaq __camellia_dec_blk16, %r9;
 
 	jmp camellia_xts_crypt_16way;
-ENDPROC(camellia_xts_dec_16way)
+SYM_FUNC_END(camellia_xts_dec_16way)
diff --git a/arch/x86/crypto/camellia-aesni-avx2-asm_64.S b/arch/x86/crypto/camellia-aesni-avx2-asm_64.S
index 96b44ad85c59..0c1af357c86d 100644
--- a/arch/x86/crypto/camellia-aesni-avx2-asm_64.S
+++ b/arch/x86/crypto/camellia-aesni-avx2-asm_64.S
@@ -935,7 +935,7 @@ SYM_FUNC_START_LOCAL(__camellia_dec_blk32)
 	jmp .Ldec_max24;
 SYM_FUNC_END(__camellia_dec_blk32)
 
-ENTRY(camellia_ecb_enc_32way)
+SYM_FUNC_START(camellia_ecb_enc_32way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst (32 blocks)
@@ -962,9 +962,9 @@ ENTRY(camellia_ecb_enc_32way)
 
 	FRAME_END
 	ret;
-ENDPROC(camellia_ecb_enc_32way)
+SYM_FUNC_END(camellia_ecb_enc_32way)
 
-ENTRY(camellia_ecb_dec_32way)
+SYM_FUNC_START(camellia_ecb_dec_32way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst (32 blocks)
@@ -996,9 +996,9 @@ ENTRY(camellia_ecb_dec_32way)
 
 	FRAME_END
 	ret;
-ENDPROC(camellia_ecb_dec_32way)
+SYM_FUNC_END(camellia_ecb_dec_32way)
 
-ENTRY(camellia_cbc_dec_32way)
+SYM_FUNC_START(camellia_cbc_dec_32way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst (32 blocks)
@@ -1064,7 +1064,7 @@ ENTRY(camellia_cbc_dec_32way)
 
 	FRAME_END
 	ret;
-ENDPROC(camellia_cbc_dec_32way)
+SYM_FUNC_END(camellia_cbc_dec_32way)
 
 #define inc_le128(x, minus_one, tmp) \
 	vpcmpeqq minus_one, x, tmp; \
@@ -1080,7 +1080,7 @@ ENDPROC(camellia_cbc_dec_32way)
 	vpslldq $8, tmp1, tmp1; \
 	vpsubq tmp1, x, x;
 
-ENTRY(camellia_ctr_32way)
+SYM_FUNC_START(camellia_ctr_32way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst (32 blocks)
@@ -1204,7 +1204,7 @@ ENTRY(camellia_ctr_32way)
 
 	FRAME_END
 	ret;
-ENDPROC(camellia_ctr_32way)
+SYM_FUNC_END(camellia_ctr_32way)
 
 #define gf128mul_x_ble(iv, mask, tmp) \
 	vpsrad $31, iv, tmp; \
@@ -1373,7 +1373,7 @@ SYM_FUNC_START_LOCAL(camellia_xts_crypt_32way)
 	ret;
 SYM_FUNC_END(camellia_xts_crypt_32way)
 
-ENTRY(camellia_xts_enc_32way)
+SYM_FUNC_START(camellia_xts_enc_32way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst (32 blocks)
@@ -1386,9 +1386,9 @@ ENTRY(camellia_xts_enc_32way)
 	leaq __camellia_enc_blk32, %r9;
 
 	jmp camellia_xts_crypt_32way;
-ENDPROC(camellia_xts_enc_32way)
+SYM_FUNC_END(camellia_xts_enc_32way)
 
-ENTRY(camellia_xts_dec_32way)
+SYM_FUNC_START(camellia_xts_dec_32way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst (32 blocks)
@@ -1404,4 +1404,4 @@ ENTRY(camellia_xts_dec_32way)
 	leaq __camellia_dec_blk32, %r9;
 
 	jmp camellia_xts_crypt_32way;
-ENDPROC(camellia_xts_dec_32way)
+SYM_FUNC_END(camellia_xts_dec_32way)
diff --git a/arch/x86/crypto/camellia-x86_64-asm_64.S b/arch/x86/crypto/camellia-x86_64-asm_64.S
index 310319c601ed..1886ea733a76 100644
--- a/arch/x86/crypto/camellia-x86_64-asm_64.S
+++ b/arch/x86/crypto/camellia-x86_64-asm_64.S
@@ -190,7 +190,7 @@
 	bswapq				RAB0; \
 	movq RAB0,			4*2(RIO);
 
-ENTRY(__camellia_enc_blk)
+SYM_FUNC_START(__camellia_enc_blk)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -235,9 +235,9 @@ ENTRY(__camellia_enc_blk)
 
 	movq RRBP, %rbp;
 	ret;
-ENDPROC(__camellia_enc_blk)
+SYM_FUNC_END(__camellia_enc_blk)
 
-ENTRY(camellia_dec_blk)
+SYM_FUNC_START(camellia_dec_blk)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -273,7 +273,7 @@ ENTRY(camellia_dec_blk)
 
 	movq RRBP, %rbp;
 	ret;
-ENDPROC(camellia_dec_blk)
+SYM_FUNC_END(camellia_dec_blk)
 
 /**********************************************************************
   2-way camellia
@@ -424,7 +424,7 @@ ENDPROC(camellia_dec_blk)
 		bswapq				RAB1; \
 		movq RAB1,			12*2(RIO);
 
-ENTRY(__camellia_enc_blk_2way)
+SYM_FUNC_START(__camellia_enc_blk_2way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -471,9 +471,9 @@ ENTRY(__camellia_enc_blk_2way)
 	movq RRBP, %rbp;
 	popq %rbx;
 	ret;
-ENDPROC(__camellia_enc_blk_2way)
+SYM_FUNC_END(__camellia_enc_blk_2way)
 
-ENTRY(camellia_dec_blk_2way)
+SYM_FUNC_START(camellia_dec_blk_2way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -511,4 +511,4 @@ ENTRY(camellia_dec_blk_2way)
 	movq RRBP, %rbp;
 	movq RXOR, %rbx;
 	ret;
-ENDPROC(camellia_dec_blk_2way)
+SYM_FUNC_END(camellia_dec_blk_2way)
diff --git a/arch/x86/crypto/cast5-avx-x86_64-asm_64.S b/arch/x86/crypto/cast5-avx-x86_64-asm_64.S
index 0fe153a87d90..bacf6c989d10 100644
--- a/arch/x86/crypto/cast5-avx-x86_64-asm_64.S
+++ b/arch/x86/crypto/cast5-avx-x86_64-asm_64.S
@@ -370,7 +370,7 @@ SYM_FUNC_START_LOCAL(__cast5_dec_blk16)
 	jmp .L__dec_tail;
 SYM_FUNC_END(__cast5_dec_blk16)
 
-ENTRY(cast5_ecb_enc_16way)
+SYM_FUNC_START(cast5_ecb_enc_16way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -402,9 +402,9 @@ ENTRY(cast5_ecb_enc_16way)
 
 	FRAME_END
 	ret;
-ENDPROC(cast5_ecb_enc_16way)
+SYM_FUNC_END(cast5_ecb_enc_16way)
 
-ENTRY(cast5_ecb_dec_16way)
+SYM_FUNC_START(cast5_ecb_dec_16way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -436,9 +436,9 @@ ENTRY(cast5_ecb_dec_16way)
 
 	FRAME_END
 	ret;
-ENDPROC(cast5_ecb_dec_16way)
+SYM_FUNC_END(cast5_ecb_dec_16way)
 
-ENTRY(cast5_cbc_dec_16way)
+SYM_FUNC_START(cast5_cbc_dec_16way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -487,9 +487,9 @@ ENTRY(cast5_cbc_dec_16way)
 
 	FRAME_END
 	ret;
-ENDPROC(cast5_cbc_dec_16way)
+SYM_FUNC_END(cast5_cbc_dec_16way)
 
-ENTRY(cast5_ctr_16way)
+SYM_FUNC_START(cast5_ctr_16way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -562,4 +562,4 @@ ENTRY(cast5_ctr_16way)
 
 	FRAME_END
 	ret;
-ENDPROC(cast5_ctr_16way)
+SYM_FUNC_END(cast5_ctr_16way)
diff --git a/arch/x86/crypto/cast6-avx-x86_64-asm_64.S b/arch/x86/crypto/cast6-avx-x86_64-asm_64.S
index 0d71989fff90..cfb688a40b03 100644
--- a/arch/x86/crypto/cast6-avx-x86_64-asm_64.S
+++ b/arch/x86/crypto/cast6-avx-x86_64-asm_64.S
@@ -352,7 +352,7 @@ SYM_FUNC_START_LOCAL(__cast6_dec_blk8)
 	ret;
 SYM_FUNC_END(__cast6_dec_blk8)
 
-ENTRY(cast6_ecb_enc_8way)
+SYM_FUNC_START(cast6_ecb_enc_8way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -370,9 +370,9 @@ ENTRY(cast6_ecb_enc_8way)
 
 	FRAME_END
 	ret;
-ENDPROC(cast6_ecb_enc_8way)
+SYM_FUNC_END(cast6_ecb_enc_8way)
 
-ENTRY(cast6_ecb_dec_8way)
+SYM_FUNC_START(cast6_ecb_dec_8way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -390,9 +390,9 @@ ENTRY(cast6_ecb_dec_8way)
 
 	FRAME_END
 	ret;
-ENDPROC(cast6_ecb_dec_8way)
+SYM_FUNC_END(cast6_ecb_dec_8way)
 
-ENTRY(cast6_cbc_dec_8way)
+SYM_FUNC_START(cast6_cbc_dec_8way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -415,9 +415,9 @@ ENTRY(cast6_cbc_dec_8way)
 
 	FRAME_END
 	ret;
-ENDPROC(cast6_cbc_dec_8way)
+SYM_FUNC_END(cast6_cbc_dec_8way)
 
-ENTRY(cast6_ctr_8way)
+SYM_FUNC_START(cast6_ctr_8way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -442,9 +442,9 @@ ENTRY(cast6_ctr_8way)
 
 	FRAME_END
 	ret;
-ENDPROC(cast6_ctr_8way)
+SYM_FUNC_END(cast6_ctr_8way)
 
-ENTRY(cast6_xts_enc_8way)
+SYM_FUNC_START(cast6_xts_enc_8way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -466,9 +466,9 @@ ENTRY(cast6_xts_enc_8way)
 
 	FRAME_END
 	ret;
-ENDPROC(cast6_xts_enc_8way)
+SYM_FUNC_END(cast6_xts_enc_8way)
 
-ENTRY(cast6_xts_dec_8way)
+SYM_FUNC_START(cast6_xts_dec_8way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -490,4 +490,4 @@ ENTRY(cast6_xts_dec_8way)
 
 	FRAME_END
 	ret;
-ENDPROC(cast6_xts_dec_8way)
+SYM_FUNC_END(cast6_xts_dec_8way)
diff --git a/arch/x86/crypto/chacha20-avx2-x86_64.S b/arch/x86/crypto/chacha20-avx2-x86_64.S
index 3a2dc3dc6cac..f06f181c572a 100644
--- a/arch/x86/crypto/chacha20-avx2-x86_64.S
+++ b/arch/x86/crypto/chacha20-avx2-x86_64.S
@@ -28,7 +28,7 @@ CTRINC:	.octa 0x00000003000000020000000100000000
 
 .text
 
-ENTRY(chacha20_8block_xor_avx2)
+SYM_FUNC_START(chacha20_8block_xor_avx2)
 	# %rdi: Input state matrix, s
 	# %rsi: 8 data blocks output, o
 	# %rdx: 8 data blocks input, i
@@ -445,4 +445,4 @@ ENTRY(chacha20_8block_xor_avx2)
 	vzeroupper
 	mov		%r8,%rsp
 	ret
-ENDPROC(chacha20_8block_xor_avx2)
+SYM_FUNC_END(chacha20_8block_xor_avx2)
diff --git a/arch/x86/crypto/chacha20-ssse3-x86_64.S b/arch/x86/crypto/chacha20-ssse3-x86_64.S
index 3f511a7d73b8..1e8b93bd2d93 100644
--- a/arch/x86/crypto/chacha20-ssse3-x86_64.S
+++ b/arch/x86/crypto/chacha20-ssse3-x86_64.S
@@ -23,7 +23,7 @@ CTRINC:	.octa 0x00000003000000020000000100000000
 
 .text
 
-ENTRY(chacha20_block_xor_ssse3)
+SYM_FUNC_START(chacha20_block_xor_ssse3)
 	# %rdi: Input state matrix, s
 	# %rsi: 1 data block output, o
 	# %rdx: 1 data block input, i
@@ -143,9 +143,9 @@ ENTRY(chacha20_block_xor_ssse3)
 	movdqu		%xmm3,0x30(%rsi)
 
 	ret
-ENDPROC(chacha20_block_xor_ssse3)
+SYM_FUNC_END(chacha20_block_xor_ssse3)
 
-ENTRY(chacha20_4block_xor_ssse3)
+SYM_FUNC_START(chacha20_4block_xor_ssse3)
 	# %rdi: Input state matrix, s
 	# %rsi: 4 data blocks output, o
 	# %rdx: 4 data blocks input, i
@@ -627,4 +627,4 @@ ENTRY(chacha20_4block_xor_ssse3)
 
 	mov		%r11,%rsp
 	ret
-ENDPROC(chacha20_4block_xor_ssse3)
+SYM_FUNC_END(chacha20_4block_xor_ssse3)
diff --git a/arch/x86/crypto/crc32-pclmul_asm.S b/arch/x86/crypto/crc32-pclmul_asm.S
index f247304299a2..690d429ef5e7 100644
--- a/arch/x86/crypto/crc32-pclmul_asm.S
+++ b/arch/x86/crypto/crc32-pclmul_asm.S
@@ -102,7 +102,7 @@
  *	                     size_t len, uint crc32)
  */
 
-ENTRY(crc32_pclmul_le_16) /* buffer and buffer size are 16 bytes aligned */
+SYM_FUNC_START(crc32_pclmul_le_16) /* buffer and buffer size are 16 bytes aligned */
 	movdqa  (BUF), %xmm1
 	movdqa  0x10(BUF), %xmm2
 	movdqa  0x20(BUF), %xmm3
@@ -243,4 +243,4 @@ fold_64:
 	PEXTRD  0x01, %xmm1, %eax
 
 	ret
-ENDPROC(crc32_pclmul_le_16)
+SYM_FUNC_END(crc32_pclmul_le_16)
diff --git a/arch/x86/crypto/crc32c-pcl-intel-asm_64.S b/arch/x86/crypto/crc32c-pcl-intel-asm_64.S
index 7a7de27c6f41..344ec8d9670b 100644
--- a/arch/x86/crypto/crc32c-pcl-intel-asm_64.S
+++ b/arch/x86/crypto/crc32c-pcl-intel-asm_64.S
@@ -73,7 +73,7 @@
 # unsigned int crc_pcl(u8 *buffer, int len, unsigned int crc_init);
 
 .text
-ENTRY(crc_pcl)
+SYM_FUNC_START(crc_pcl)
 #define    bufp		%rdi
 #define    bufp_dw	%edi
 #define    bufp_w	%di
@@ -310,7 +310,7 @@ do_return:
 	popq    %rdi
 	popq    %rbx
         ret
-ENDPROC(crc_pcl)
+SYM_FUNC_END(crc_pcl)
 
 .section	.rodata, "a", @progbits
         ################################################################
diff --git a/arch/x86/crypto/crct10dif-pcl-asm_64.S b/arch/x86/crypto/crct10dif-pcl-asm_64.S
index de04d3e98d8d..f56b499541e0 100644
--- a/arch/x86/crypto/crct10dif-pcl-asm_64.S
+++ b/arch/x86/crypto/crct10dif-pcl-asm_64.S
@@ -68,7 +68,7 @@
 
 #define        arg1_low32 %edi
 
-ENTRY(crc_t10dif_pcl)
+SYM_FUNC_START(crc_t10dif_pcl)
 .align 16
 
 	# adjust the 16-bit initial_crc value, scale it to 32 bits
@@ -552,7 +552,7 @@ _only_less_than_2:
 
 	jmp	_barrett
 
-ENDPROC(crc_t10dif_pcl)
+SYM_FUNC_END(crc_t10dif_pcl)
 
 .section	.rodata, "a", @progbits
 .align 16
diff --git a/arch/x86/crypto/des3_ede-asm_64.S b/arch/x86/crypto/des3_ede-asm_64.S
index f3e91647ca27..0c39ed072173 100644
--- a/arch/x86/crypto/des3_ede-asm_64.S
+++ b/arch/x86/crypto/des3_ede-asm_64.S
@@ -171,7 +171,7 @@
 	movl   left##d,   (io); \
 	movl   right##d, 4(io);
 
-ENTRY(des3_ede_x86_64_crypt_blk)
+SYM_FUNC_START(des3_ede_x86_64_crypt_blk)
 	/* input:
 	 *	%rdi: round keys, CTX
 	 *	%rsi: dst
@@ -251,7 +251,7 @@ ENTRY(des3_ede_x86_64_crypt_blk)
 	popq %rbp;
 
 	ret;
-ENDPROC(des3_ede_x86_64_crypt_blk)
+SYM_FUNC_END(des3_ede_x86_64_crypt_blk)
 
 /***********************************************************************
  * 3-way 3DES
@@ -425,7 +425,7 @@ ENDPROC(des3_ede_x86_64_crypt_blk)
 #define __movq(src, dst) \
 	movq src, dst;
 
-ENTRY(des3_ede_x86_64_crypt_blk_3way)
+SYM_FUNC_START(des3_ede_x86_64_crypt_blk_3way)
 	/* input:
 	 *	%rdi: ctx, round keys
 	 *	%rsi: dst (3 blocks)
@@ -535,7 +535,7 @@ ENTRY(des3_ede_x86_64_crypt_blk_3way)
 	popq %rbp;
 
 	ret;
-ENDPROC(des3_ede_x86_64_crypt_blk_3way)
+SYM_FUNC_END(des3_ede_x86_64_crypt_blk_3way)
 
 .section	.rodata, "a", @progbits
 .align 16
diff --git a/arch/x86/crypto/ghash-clmulni-intel_asm.S b/arch/x86/crypto/ghash-clmulni-intel_asm.S
index c3db86842578..12e3a850257b 100644
--- a/arch/x86/crypto/ghash-clmulni-intel_asm.S
+++ b/arch/x86/crypto/ghash-clmulni-intel_asm.S
@@ -93,7 +93,7 @@ SYM_FUNC_START_LOCAL(__clmul_gf128mul_ble)
 SYM_FUNC_END(__clmul_gf128mul_ble)
 
 /* void clmul_ghash_mul(char *dst, const u128 *shash) */
-ENTRY(clmul_ghash_mul)
+SYM_FUNC_START(clmul_ghash_mul)
 	FRAME_BEGIN
 	movups (%rdi), DATA
 	movups (%rsi), SHASH
@@ -104,13 +104,13 @@ ENTRY(clmul_ghash_mul)
 	movups DATA, (%rdi)
 	FRAME_END
 	ret
-ENDPROC(clmul_ghash_mul)
+SYM_FUNC_END(clmul_ghash_mul)
 
 /*
  * void clmul_ghash_update(char *dst, const char *src, unsigned int srclen,
  *			   const u128 *shash);
  */
-ENTRY(clmul_ghash_update)
+SYM_FUNC_START(clmul_ghash_update)
 	FRAME_BEGIN
 	cmp $16, %rdx
 	jb .Lupdate_just_ret	# check length
@@ -133,4 +133,4 @@ ENTRY(clmul_ghash_update)
 .Lupdate_just_ret:
 	FRAME_END
 	ret
-ENDPROC(clmul_ghash_update)
+SYM_FUNC_END(clmul_ghash_update)
diff --git a/arch/x86/crypto/poly1305-avx2-x86_64.S b/arch/x86/crypto/poly1305-avx2-x86_64.S
index 3b6e70d085da..68b0f4386dc4 100644
--- a/arch/x86/crypto/poly1305-avx2-x86_64.S
+++ b/arch/x86/crypto/poly1305-avx2-x86_64.S
@@ -83,7 +83,7 @@ ORMASK:	.octa 0x00000000010000000000000001000000
 #define d3 %r12
 #define d4 %r13
 
-ENTRY(poly1305_4block_avx2)
+SYM_FUNC_START(poly1305_4block_avx2)
 	# %rdi: Accumulator h[5]
 	# %rsi: 64 byte input block m
 	# %rdx: Poly1305 key r[5]
@@ -385,4 +385,4 @@ ENTRY(poly1305_4block_avx2)
 	pop		%r12
 	pop		%rbx
 	ret
-ENDPROC(poly1305_4block_avx2)
+SYM_FUNC_END(poly1305_4block_avx2)
diff --git a/arch/x86/crypto/poly1305-sse2-x86_64.S b/arch/x86/crypto/poly1305-sse2-x86_64.S
index c88c670cb5fc..66715fbedc18 100644
--- a/arch/x86/crypto/poly1305-sse2-x86_64.S
+++ b/arch/x86/crypto/poly1305-sse2-x86_64.S
@@ -50,7 +50,7 @@ ORMASK:	.octa 0x00000000010000000000000001000000
 #define d3 %r11
 #define d4 %r12
 
-ENTRY(poly1305_block_sse2)
+SYM_FUNC_START(poly1305_block_sse2)
 	# %rdi: Accumulator h[5]
 	# %rsi: 16 byte input block m
 	# %rdx: Poly1305 key r[5]
@@ -276,7 +276,7 @@ ENTRY(poly1305_block_sse2)
 	pop		%r12
 	pop		%rbx
 	ret
-ENDPROC(poly1305_block_sse2)
+SYM_FUNC_END(poly1305_block_sse2)
 
 
 #define u0 0x00(%r8)
@@ -301,7 +301,7 @@ ENDPROC(poly1305_block_sse2)
 #undef d0
 #define d0 %r13
 
-ENTRY(poly1305_2block_sse2)
+SYM_FUNC_START(poly1305_2block_sse2)
 	# %rdi: Accumulator h[5]
 	# %rsi: 16 byte input block m
 	# %rdx: Poly1305 key r[5]
@@ -581,4 +581,4 @@ ENTRY(poly1305_2block_sse2)
 	pop		%r12
 	pop		%rbx
 	ret
-ENDPROC(poly1305_2block_sse2)
+SYM_FUNC_END(poly1305_2block_sse2)
diff --git a/arch/x86/crypto/salsa20-x86_64-asm_64.S b/arch/x86/crypto/salsa20-x86_64-asm_64.S
index 9279e0b2d60e..a5f3dd15a755 100644
--- a/arch/x86/crypto/salsa20-x86_64-asm_64.S
+++ b/arch/x86/crypto/salsa20-x86_64-asm_64.S
@@ -1,7 +1,7 @@
 #include <linux/linkage.h>
 
 # enter salsa20_encrypt_bytes
-ENTRY(salsa20_encrypt_bytes)
+SYM_FUNC_START(salsa20_encrypt_bytes)
 	mov	%rsp,%r11
 	and	$31,%r11
 	add	$256,%r11
@@ -801,10 +801,10 @@ ENTRY(salsa20_encrypt_bytes)
 	# comment:fp stack unchanged by jump
 	# goto bytesatleast1
 	jmp	._bytesatleast1
-ENDPROC(salsa20_encrypt_bytes)
+SYM_FUNC_END(salsa20_encrypt_bytes)
 
 # enter salsa20_keysetup
-ENTRY(salsa20_keysetup)
+SYM_FUNC_START(salsa20_keysetup)
 	mov	%rsp,%r11
 	and	$31,%r11
 	add	$256,%r11
@@ -890,10 +890,10 @@ ENTRY(salsa20_keysetup)
 	mov	%rdi,%rax
 	mov	%rsi,%rdx
 	ret
-ENDPROC(salsa20_keysetup)
+SYM_FUNC_END(salsa20_keysetup)
 
 # enter salsa20_ivsetup
-ENTRY(salsa20_ivsetup)
+SYM_FUNC_START(salsa20_ivsetup)
 	mov	%rsp,%r11
 	and	$31,%r11
 	add	$256,%r11
@@ -915,4 +915,4 @@ ENTRY(salsa20_ivsetup)
 	mov	%rdi,%rax
 	mov	%rsi,%rdx
 	ret
-ENDPROC(salsa20_ivsetup)
+SYM_FUNC_END(salsa20_ivsetup)
diff --git a/arch/x86/crypto/serpent-avx-x86_64-asm_64.S b/arch/x86/crypto/serpent-avx-x86_64-asm_64.S
index c2d4a1fc9ee8..72de86a8091e 100644
--- a/arch/x86/crypto/serpent-avx-x86_64-asm_64.S
+++ b/arch/x86/crypto/serpent-avx-x86_64-asm_64.S
@@ -677,7 +677,7 @@ SYM_FUNC_START_LOCAL(__serpent_dec_blk8_avx)
 	ret;
 SYM_FUNC_END(__serpent_dec_blk8_avx)
 
-ENTRY(serpent_ecb_enc_8way_avx)
+SYM_FUNC_START(serpent_ecb_enc_8way_avx)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -693,9 +693,9 @@ ENTRY(serpent_ecb_enc_8way_avx)
 
 	FRAME_END
 	ret;
-ENDPROC(serpent_ecb_enc_8way_avx)
+SYM_FUNC_END(serpent_ecb_enc_8way_avx)
 
-ENTRY(serpent_ecb_dec_8way_avx)
+SYM_FUNC_START(serpent_ecb_dec_8way_avx)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -711,9 +711,9 @@ ENTRY(serpent_ecb_dec_8way_avx)
 
 	FRAME_END
 	ret;
-ENDPROC(serpent_ecb_dec_8way_avx)
+SYM_FUNC_END(serpent_ecb_dec_8way_avx)
 
-ENTRY(serpent_cbc_dec_8way_avx)
+SYM_FUNC_START(serpent_cbc_dec_8way_avx)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -729,9 +729,9 @@ ENTRY(serpent_cbc_dec_8way_avx)
 
 	FRAME_END
 	ret;
-ENDPROC(serpent_cbc_dec_8way_avx)
+SYM_FUNC_END(serpent_cbc_dec_8way_avx)
 
-ENTRY(serpent_ctr_8way_avx)
+SYM_FUNC_START(serpent_ctr_8way_avx)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -749,9 +749,9 @@ ENTRY(serpent_ctr_8way_avx)
 
 	FRAME_END
 	ret;
-ENDPROC(serpent_ctr_8way_avx)
+SYM_FUNC_END(serpent_ctr_8way_avx)
 
-ENTRY(serpent_xts_enc_8way_avx)
+SYM_FUNC_START(serpent_xts_enc_8way_avx)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -771,9 +771,9 @@ ENTRY(serpent_xts_enc_8way_avx)
 
 	FRAME_END
 	ret;
-ENDPROC(serpent_xts_enc_8way_avx)
+SYM_FUNC_END(serpent_xts_enc_8way_avx)
 
-ENTRY(serpent_xts_dec_8way_avx)
+SYM_FUNC_START(serpent_xts_dec_8way_avx)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -793,4 +793,4 @@ ENTRY(serpent_xts_dec_8way_avx)
 
 	FRAME_END
 	ret;
-ENDPROC(serpent_xts_dec_8way_avx)
+SYM_FUNC_END(serpent_xts_dec_8way_avx)
diff --git a/arch/x86/crypto/serpent-avx2-asm_64.S b/arch/x86/crypto/serpent-avx2-asm_64.S
index 52c527ce4b18..b866f1632803 100644
--- a/arch/x86/crypto/serpent-avx2-asm_64.S
+++ b/arch/x86/crypto/serpent-avx2-asm_64.S
@@ -673,7 +673,7 @@ SYM_FUNC_START_LOCAL(__serpent_dec_blk16)
 	ret;
 SYM_FUNC_END(__serpent_dec_blk16)
 
-ENTRY(serpent_ecb_enc_16way)
+SYM_FUNC_START(serpent_ecb_enc_16way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -693,9 +693,9 @@ ENTRY(serpent_ecb_enc_16way)
 
 	FRAME_END
 	ret;
-ENDPROC(serpent_ecb_enc_16way)
+SYM_FUNC_END(serpent_ecb_enc_16way)
 
-ENTRY(serpent_ecb_dec_16way)
+SYM_FUNC_START(serpent_ecb_dec_16way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -715,9 +715,9 @@ ENTRY(serpent_ecb_dec_16way)
 
 	FRAME_END
 	ret;
-ENDPROC(serpent_ecb_dec_16way)
+SYM_FUNC_END(serpent_ecb_dec_16way)
 
-ENTRY(serpent_cbc_dec_16way)
+SYM_FUNC_START(serpent_cbc_dec_16way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -738,9 +738,9 @@ ENTRY(serpent_cbc_dec_16way)
 
 	FRAME_END
 	ret;
-ENDPROC(serpent_cbc_dec_16way)
+SYM_FUNC_END(serpent_cbc_dec_16way)
 
-ENTRY(serpent_ctr_16way)
+SYM_FUNC_START(serpent_ctr_16way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst (16 blocks)
@@ -763,9 +763,9 @@ ENTRY(serpent_ctr_16way)
 
 	FRAME_END
 	ret;
-ENDPROC(serpent_ctr_16way)
+SYM_FUNC_END(serpent_ctr_16way)
 
-ENTRY(serpent_xts_enc_16way)
+SYM_FUNC_START(serpent_xts_enc_16way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst (16 blocks)
@@ -789,9 +789,9 @@ ENTRY(serpent_xts_enc_16way)
 
 	FRAME_END
 	ret;
-ENDPROC(serpent_xts_enc_16way)
+SYM_FUNC_END(serpent_xts_enc_16way)
 
-ENTRY(serpent_xts_dec_16way)
+SYM_FUNC_START(serpent_xts_dec_16way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst (16 blocks)
@@ -815,4 +815,4 @@ ENTRY(serpent_xts_dec_16way)
 
 	FRAME_END
 	ret;
-ENDPROC(serpent_xts_dec_16way)
+SYM_FUNC_END(serpent_xts_dec_16way)
diff --git a/arch/x86/crypto/serpent-sse2-x86_64-asm_64.S b/arch/x86/crypto/serpent-sse2-x86_64-asm_64.S
index acc066c7c6b2..bdeee900df63 100644
--- a/arch/x86/crypto/serpent-sse2-x86_64-asm_64.S
+++ b/arch/x86/crypto/serpent-sse2-x86_64-asm_64.S
@@ -634,7 +634,7 @@
 	pxor t0,		x3; \
 	movdqu x3,		(3*4*4)(out);
 
-ENTRY(__serpent_enc_blk_8way)
+SYM_FUNC_START(__serpent_enc_blk_8way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -697,9 +697,9 @@ ENTRY(__serpent_enc_blk_8way)
 	xor_blocks(%rax, RA2, RB2, RC2, RD2, RK0, RK1, RK2);
 
 	ret;
-ENDPROC(__serpent_enc_blk_8way)
+SYM_FUNC_END(__serpent_enc_blk_8way)
 
-ENTRY(serpent_dec_blk_8way)
+SYM_FUNC_START(serpent_dec_blk_8way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -751,4 +751,4 @@ ENTRY(serpent_dec_blk_8way)
 	write_blocks(%rax, RC2, RD2, RB2, RE2, RK0, RK1, RK2);
 
 	ret;
-ENDPROC(serpent_dec_blk_8way)
+SYM_FUNC_END(serpent_dec_blk_8way)
diff --git a/arch/x86/crypto/sha1-mb/sha1_mb_mgr_flush_avx2.S b/arch/x86/crypto/sha1-mb/sha1_mb_mgr_flush_avx2.S
index 93b945597ecf..7623a16c3c5f 100644
--- a/arch/x86/crypto/sha1-mb/sha1_mb_mgr_flush_avx2.S
+++ b/arch/x86/crypto/sha1-mb/sha1_mb_mgr_flush_avx2.S
@@ -103,7 +103,7 @@ offset = \_offset
 
 # JOB* sha1_mb_mgr_flush_avx2(MB_MGR *state)
 # arg 1 : rcx : state
-ENTRY(sha1_mb_mgr_flush_avx2)
+SYM_FUNC_START(sha1_mb_mgr_flush_avx2)
 	FRAME_BEGIN
 	push	%rbx
 
@@ -220,13 +220,13 @@ return:
 return_null:
 	xor     job_rax, job_rax
 	jmp     return
-ENDPROC(sha1_mb_mgr_flush_avx2)
+SYM_FUNC_END(sha1_mb_mgr_flush_avx2)
 
 
 #################################################################
 
 .align 16
-ENTRY(sha1_mb_mgr_get_comp_job_avx2)
+SYM_FUNC_START(sha1_mb_mgr_get_comp_job_avx2)
 	push    %rbx
 
 	## if bit 32+3 is set, then all lanes are empty
@@ -279,7 +279,7 @@ ENTRY(sha1_mb_mgr_get_comp_job_avx2)
 	xor     job_rax, job_rax
 	pop     %rbx
 	ret
-ENDPROC(sha1_mb_mgr_get_comp_job_avx2)
+SYM_FUNC_END(sha1_mb_mgr_get_comp_job_avx2)
 
 .section	.rodata.cst16.clear_low_nibble, "aM", @progbits, 16
 .align 16
diff --git a/arch/x86/crypto/sha1-mb/sha1_mb_mgr_submit_avx2.S b/arch/x86/crypto/sha1-mb/sha1_mb_mgr_submit_avx2.S
index 7a93b1c0d69a..a46e3b04385e 100644
--- a/arch/x86/crypto/sha1-mb/sha1_mb_mgr_submit_avx2.S
+++ b/arch/x86/crypto/sha1-mb/sha1_mb_mgr_submit_avx2.S
@@ -98,7 +98,7 @@ lane_data       = %r10
 # JOB* submit_mb_mgr_submit_avx2(MB_MGR *state, job_sha1 *job)
 # arg 1 : rcx : state
 # arg 2 : rdx : job
-ENTRY(sha1_mb_mgr_submit_avx2)
+SYM_FUNC_START(sha1_mb_mgr_submit_avx2)
 	FRAME_BEGIN
 	push	%rbx
 	push	%r12
@@ -201,7 +201,7 @@ return_null:
 	xor     job_rax, job_rax
 	jmp     return
 
-ENDPROC(sha1_mb_mgr_submit_avx2)
+SYM_FUNC_END(sha1_mb_mgr_submit_avx2)
 
 .section	.rodata.cst16.clear_low_nibble, "aM", @progbits, 16
 .align 16
diff --git a/arch/x86/crypto/sha1-mb/sha1_x8_avx2.S b/arch/x86/crypto/sha1-mb/sha1_x8_avx2.S
index 20f77aa633de..04d763520a82 100644
--- a/arch/x86/crypto/sha1-mb/sha1_x8_avx2.S
+++ b/arch/x86/crypto/sha1-mb/sha1_x8_avx2.S
@@ -294,7 +294,7 @@ W14  = TMP_
 # arg 1 : pointer to array[4] of pointer to input data
 # arg 2 : size (in blocks) ;; assumed to be >= 1
 #
-ENTRY(sha1_x8_avx2)
+SYM_FUNC_START(sha1_x8_avx2)
 
 	# save callee-saved clobbered registers to comply with C function ABI
 	push	%r12
@@ -458,7 +458,7 @@ lloop:
 	pop	%r12
 
 	ret
-ENDPROC(sha1_x8_avx2)
+SYM_FUNC_END(sha1_x8_avx2)
 
 
 .section	.rodata.cst32.K00_19, "aM", @progbits, 32
diff --git a/arch/x86/crypto/sha1_avx2_x86_64_asm.S b/arch/x86/crypto/sha1_avx2_x86_64_asm.S
index 1cd792db15ef..4a477cb80fa5 100644
--- a/arch/x86/crypto/sha1_avx2_x86_64_asm.S
+++ b/arch/x86/crypto/sha1_avx2_x86_64_asm.S
@@ -622,7 +622,7 @@ _loop3:
  * param: function's name
  */
 .macro SHA1_VECTOR_ASM  name
-	ENTRY(\name)
+	SYM_FUNC_START(\name)
 
 	push	%rbx
 	push	%rbp
@@ -673,7 +673,7 @@ _loop3:
 
 	ret
 
-	ENDPROC(\name)
+	SYM_FUNC_END(\name)
 .endm
 
 .section .rodata
diff --git a/arch/x86/crypto/sha1_ni_asm.S b/arch/x86/crypto/sha1_ni_asm.S
index ebbdba72ae07..11efe3a45a1f 100644
--- a/arch/x86/crypto/sha1_ni_asm.S
+++ b/arch/x86/crypto/sha1_ni_asm.S
@@ -95,7 +95,7 @@
  */
 .text
 .align 32
-ENTRY(sha1_ni_transform)
+SYM_FUNC_START(sha1_ni_transform)
 	mov		%rsp, RSPSAVE
 	sub		$FRAME_SIZE, %rsp
 	and		$~0xF, %rsp
@@ -291,7 +291,7 @@ ENTRY(sha1_ni_transform)
 	mov		RSPSAVE, %rsp
 
 	ret
-ENDPROC(sha1_ni_transform)
+SYM_FUNC_END(sha1_ni_transform)
 
 .section	.rodata.cst16.PSHUFFLE_BYTE_FLIP_MASK, "aM", @progbits, 16
 .align 16
diff --git a/arch/x86/crypto/sha1_ssse3_asm.S b/arch/x86/crypto/sha1_ssse3_asm.S
index a4109506a5e8..5a5b433272e5 100644
--- a/arch/x86/crypto/sha1_ssse3_asm.S
+++ b/arch/x86/crypto/sha1_ssse3_asm.S
@@ -71,7 +71,7 @@
  * param: function's name
  */
 .macro SHA1_VECTOR_ASM  name
-	ENTRY(\name)
+	SYM_FUNC_START(\name)
 
 	push	%rbx
 	push	%rbp
@@ -106,7 +106,7 @@
 	pop	%rbx
 	ret
 
-	ENDPROC(\name)
+	SYM_FUNC_END(\name)
 .endm
 
 /*
diff --git a/arch/x86/crypto/sha256-avx-asm.S b/arch/x86/crypto/sha256-avx-asm.S
index e08888a1a5f2..b3208cd5aee9 100644
--- a/arch/x86/crypto/sha256-avx-asm.S
+++ b/arch/x86/crypto/sha256-avx-asm.S
@@ -347,7 +347,7 @@ a = TMP_
 ## arg 3 : Num blocks
 ########################################################################
 .text
-ENTRY(sha256_transform_avx)
+SYM_FUNC_START(sha256_transform_avx)
 .align 32
 	pushq   %rbx
 	pushq   %rbp
@@ -461,7 +461,7 @@ done_hash:
 	popq    %rbp
 	popq    %rbx
 	ret
-ENDPROC(sha256_transform_avx)
+SYM_FUNC_END(sha256_transform_avx)
 
 .section	.rodata.cst256.K256, "aM", @progbits, 256
 .align 64
diff --git a/arch/x86/crypto/sha256-avx2-asm.S b/arch/x86/crypto/sha256-avx2-asm.S
index 89c8f09787d2..63d204bc1148 100644
--- a/arch/x86/crypto/sha256-avx2-asm.S
+++ b/arch/x86/crypto/sha256-avx2-asm.S
@@ -528,7 +528,7 @@ STACK_SIZE	= _RSP      + _RSP_SIZE
 ## arg 3 : Num blocks
 ########################################################################
 .text
-ENTRY(sha256_transform_rorx)
+SYM_FUNC_START(sha256_transform_rorx)
 .align 32
 	pushq	%rbx
 	pushq	%rbp
@@ -721,7 +721,7 @@ done_hash:
 	popq	%rbp
 	popq	%rbx
 	ret
-ENDPROC(sha256_transform_rorx)
+SYM_FUNC_END(sha256_transform_rorx)
 
 .section	.rodata.cst512.K256, "aM", @progbits, 512
 .align 64
diff --git a/arch/x86/crypto/sha256-mb/sha256_mb_mgr_flush_avx2.S b/arch/x86/crypto/sha256-mb/sha256_mb_mgr_flush_avx2.S
index 8fe6338bcc84..9681227be16c 100644
--- a/arch/x86/crypto/sha256-mb/sha256_mb_mgr_flush_avx2.S
+++ b/arch/x86/crypto/sha256-mb/sha256_mb_mgr_flush_avx2.S
@@ -101,7 +101,7 @@ offset = \_offset
 
 # JOB_SHA256* sha256_mb_mgr_flush_avx2(MB_MGR *state)
 # arg 1 : rcx : state
-ENTRY(sha256_mb_mgr_flush_avx2)
+SYM_FUNC_START(sha256_mb_mgr_flush_avx2)
 	FRAME_BEGIN
         push    %rbx
 
@@ -220,12 +220,12 @@ return:
 return_null:
 	xor	job_rax, job_rax
 	jmp	return
-ENDPROC(sha256_mb_mgr_flush_avx2)
+SYM_FUNC_END(sha256_mb_mgr_flush_avx2)
 
 ##############################################################################
 
 .align 16
-ENTRY(sha256_mb_mgr_get_comp_job_avx2)
+SYM_FUNC_START(sha256_mb_mgr_get_comp_job_avx2)
 	push	%rbx
 
 	## if bit 32+3 is set, then all lanes are empty
@@ -282,7 +282,7 @@ ENTRY(sha256_mb_mgr_get_comp_job_avx2)
 	xor	job_rax, job_rax
 	pop	%rbx
 	ret
-ENDPROC(sha256_mb_mgr_get_comp_job_avx2)
+SYM_FUNC_END(sha256_mb_mgr_get_comp_job_avx2)
 
 .section	.rodata.cst16.clear_low_nibble, "aM", @progbits, 16
 .align 16
diff --git a/arch/x86/crypto/sha256-mb/sha256_mb_mgr_submit_avx2.S b/arch/x86/crypto/sha256-mb/sha256_mb_mgr_submit_avx2.S
index b36ae7454084..2213c04a30dc 100644
--- a/arch/x86/crypto/sha256-mb/sha256_mb_mgr_submit_avx2.S
+++ b/arch/x86/crypto/sha256-mb/sha256_mb_mgr_submit_avx2.S
@@ -96,7 +96,7 @@ lane_data	= %r10
 # JOB* sha256_mb_mgr_submit_avx2(MB_MGR *state, JOB_SHA256 *job)
 # arg 1 : rcx : state
 # arg 2 : rdx : job
-ENTRY(sha256_mb_mgr_submit_avx2)
+SYM_FUNC_START(sha256_mb_mgr_submit_avx2)
 	FRAME_BEGIN
 	push	%rbx
 	push	%r12
@@ -206,7 +206,7 @@ return_null:
 	xor	job_rax, job_rax
 	jmp	return
 
-ENDPROC(sha256_mb_mgr_submit_avx2)
+SYM_FUNC_END(sha256_mb_mgr_submit_avx2)
 
 .section	.rodata.cst16.clear_low_nibble, "aM", @progbits, 16
 .align 16
diff --git a/arch/x86/crypto/sha256-mb/sha256_x8_avx2.S b/arch/x86/crypto/sha256-mb/sha256_x8_avx2.S
index 1687c80c5995..042d2381f435 100644
--- a/arch/x86/crypto/sha256-mb/sha256_x8_avx2.S
+++ b/arch/x86/crypto/sha256-mb/sha256_x8_avx2.S
@@ -280,7 +280,7 @@ a = TMP_
 	# general registers preserved in outer calling routine
 	# outer calling routine saves all the XMM registers
 	# save rsp, allocate 32-byte aligned for local variables
-ENTRY(sha256_x8_avx2)
+SYM_FUNC_START(sha256_x8_avx2)
 
 	# save callee-saved clobbered registers to comply with C function ABI
 	push    %r12
@@ -436,7 +436,7 @@ Lrounds_16_xx:
 	pop     %r12
 
 	ret
-ENDPROC(sha256_x8_avx2)
+SYM_FUNC_END(sha256_x8_avx2)
 
 .section	.rodata.K256_8, "a", @progbits
 .align 64
diff --git a/arch/x86/crypto/sha256-ssse3-asm.S b/arch/x86/crypto/sha256-ssse3-asm.S
index 39b83c93e7fd..281633643bb7 100644
--- a/arch/x86/crypto/sha256-ssse3-asm.S
+++ b/arch/x86/crypto/sha256-ssse3-asm.S
@@ -353,7 +353,7 @@ a = TMP_
 ## arg 3 : Num blocks
 ########################################################################
 .text
-ENTRY(sha256_transform_ssse3)
+SYM_FUNC_START(sha256_transform_ssse3)
 .align 32
 	pushq   %rbx
 	pushq   %rbp
@@ -472,7 +472,7 @@ done_hash:
 	popq    %rbx
 
 	ret
-ENDPROC(sha256_transform_ssse3)
+SYM_FUNC_END(sha256_transform_ssse3)
 
 .section	.rodata.cst256.K256, "aM", @progbits, 256
 .align 64
diff --git a/arch/x86/crypto/sha256_ni_asm.S b/arch/x86/crypto/sha256_ni_asm.S
index fb58f58ecfbc..7abade04a3a3 100644
--- a/arch/x86/crypto/sha256_ni_asm.S
+++ b/arch/x86/crypto/sha256_ni_asm.S
@@ -97,7 +97,7 @@
 
 .text
 .align 32
-ENTRY(sha256_ni_transform)
+SYM_FUNC_START(sha256_ni_transform)
 
 	shl		$6, NUM_BLKS		/*  convert to bytes */
 	jz		.Ldone_hash
@@ -327,7 +327,7 @@ ENTRY(sha256_ni_transform)
 .Ldone_hash:
 
 	ret
-ENDPROC(sha256_ni_transform)
+SYM_FUNC_END(sha256_ni_transform)
 
 .section	.rodata.cst256.K256, "aM", @progbits, 256
 .align 64
diff --git a/arch/x86/crypto/sha512-avx-asm.S b/arch/x86/crypto/sha512-avx-asm.S
index 39235fefe6f7..3704ddd7e5d5 100644
--- a/arch/x86/crypto/sha512-avx-asm.S
+++ b/arch/x86/crypto/sha512-avx-asm.S
@@ -277,7 +277,7 @@ frame_size = frame_GPRSAVE + GPRSAVE_SIZE
 # message blocks.
 # L is the message length in SHA512 blocks
 ########################################################################
-ENTRY(sha512_transform_avx)
+SYM_FUNC_START(sha512_transform_avx)
 	cmp $0, msglen
 	je nowork
 
@@ -365,7 +365,7 @@ updateblock:
 
 nowork:
 	ret
-ENDPROC(sha512_transform_avx)
+SYM_FUNC_END(sha512_transform_avx)
 
 ########################################################################
 ### Binary Data
diff --git a/arch/x86/crypto/sha512-avx2-asm.S b/arch/x86/crypto/sha512-avx2-asm.S
index 7f5f6c6ec72e..86cca90edf64 100644
--- a/arch/x86/crypto/sha512-avx2-asm.S
+++ b/arch/x86/crypto/sha512-avx2-asm.S
@@ -568,7 +568,7 @@ frame_size = frame_GPRSAVE + GPRSAVE_SIZE
 #   message blocks.
 # L is the message length in SHA512 blocks
 ########################################################################
-ENTRY(sha512_transform_rorx)
+SYM_FUNC_START(sha512_transform_rorx)
 	# Allocate Stack Space
 	mov	%rsp, %rax
 	sub	$frame_size, %rsp
@@ -679,7 +679,7 @@ done_hash:
 	# Restore Stack Pointer
 	mov	frame_RSPSAVE(%rsp), %rsp
 	ret
-ENDPROC(sha512_transform_rorx)
+SYM_FUNC_END(sha512_transform_rorx)
 
 ########################################################################
 ### Binary Data
diff --git a/arch/x86/crypto/sha512-mb/sha512_mb_mgr_flush_avx2.S b/arch/x86/crypto/sha512-mb/sha512_mb_mgr_flush_avx2.S
index 7c629caebc05..8642f3a04388 100644
--- a/arch/x86/crypto/sha512-mb/sha512_mb_mgr_flush_avx2.S
+++ b/arch/x86/crypto/sha512-mb/sha512_mb_mgr_flush_avx2.S
@@ -107,7 +107,7 @@ offset = \_offset
 
 # JOB* sha512_mb_mgr_flush_avx2(MB_MGR *state)
 # arg 1 : rcx : state
-ENTRY(sha512_mb_mgr_flush_avx2)
+SYM_FUNC_START(sha512_mb_mgr_flush_avx2)
 	FRAME_BEGIN
 	push	%rbx
 
@@ -217,10 +217,10 @@ return:
 return_null:
         xor     job_rax, job_rax
         jmp     return
-ENDPROC(sha512_mb_mgr_flush_avx2)
+SYM_FUNC_END(sha512_mb_mgr_flush_avx2)
 .align 16
 
-ENTRY(sha512_mb_mgr_get_comp_job_avx2)
+SYM_FUNC_START(sha512_mb_mgr_get_comp_job_avx2)
         push    %rbx
 
 	mov     _unused_lanes(state), unused_lanes
@@ -279,7 +279,7 @@ ENTRY(sha512_mb_mgr_get_comp_job_avx2)
         xor     job_rax, job_rax
 	pop     %rbx
         ret
-ENDPROC(sha512_mb_mgr_get_comp_job_avx2)
+SYM_FUNC_END(sha512_mb_mgr_get_comp_job_avx2)
 
 .section	.rodata.cst8.one, "aM", @progbits, 8
 .align 8
diff --git a/arch/x86/crypto/sha512-mb/sha512_mb_mgr_submit_avx2.S b/arch/x86/crypto/sha512-mb/sha512_mb_mgr_submit_avx2.S
index 4ba709ba78e5..62932723d6e9 100644
--- a/arch/x86/crypto/sha512-mb/sha512_mb_mgr_submit_avx2.S
+++ b/arch/x86/crypto/sha512-mb/sha512_mb_mgr_submit_avx2.S
@@ -98,7 +98,7 @@
 # JOB* sha512_mb_mgr_submit_avx2(MB_MGR *state, JOB *job)
 # arg 1 : rcx : state
 # arg 2 : rdx : job
-ENTRY(sha512_mb_mgr_submit_avx2)
+SYM_FUNC_START(sha512_mb_mgr_submit_avx2)
 	FRAME_BEGIN
 	push	%rbx
 	push	%r12
@@ -208,7 +208,7 @@ return:
 return_null:
 	xor     job_rax, job_rax
 	jmp     return
-ENDPROC(sha512_mb_mgr_submit_avx2)
+SYM_FUNC_END(sha512_mb_mgr_submit_avx2)
 
 /* UNUSED?
 .section	.rodata.cst16, "aM", @progbits, 16
diff --git a/arch/x86/crypto/sha512-mb/sha512_x4_avx2.S b/arch/x86/crypto/sha512-mb/sha512_x4_avx2.S
index e22e907643a6..504065d19e03 100644
--- a/arch/x86/crypto/sha512-mb/sha512_x4_avx2.S
+++ b/arch/x86/crypto/sha512-mb/sha512_x4_avx2.S
@@ -239,7 +239,7 @@ a = TMP_
 # void sha512_x4_avx2(void *STATE, const int INP_SIZE)
 # arg 1 : STATE    : pointer to input data
 # arg 2 : INP_SIZE : size of data in blocks (assumed >= 1)
-ENTRY(sha512_x4_avx2)
+SYM_FUNC_START(sha512_x4_avx2)
 	# general registers preserved in outer calling routine
 	# outer calling routine saves all the XMM registers
 	# save callee-saved clobbered registers to comply with C function ABI
@@ -359,7 +359,7 @@ Lrounds_16_xx:
 
 	# outer calling routine restores XMM and other GP registers
 	ret
-ENDPROC(sha512_x4_avx2)
+SYM_FUNC_END(sha512_x4_avx2)
 
 .section	.rodata.K512_4, "a", @progbits
 .align 64
diff --git a/arch/x86/crypto/sha512-ssse3-asm.S b/arch/x86/crypto/sha512-ssse3-asm.S
index 66bbd9058a90..838f984e95d9 100644
--- a/arch/x86/crypto/sha512-ssse3-asm.S
+++ b/arch/x86/crypto/sha512-ssse3-asm.S
@@ -275,7 +275,7 @@ frame_size = frame_GPRSAVE + GPRSAVE_SIZE
 #   message blocks.
 # L is the message length in SHA512 blocks.
 ########################################################################
-ENTRY(sha512_transform_ssse3)
+SYM_FUNC_START(sha512_transform_ssse3)
 
 	cmp $0, msglen
 	je nowork
@@ -364,7 +364,7 @@ updateblock:
 
 nowork:
 	ret
-ENDPROC(sha512_transform_ssse3)
+SYM_FUNC_END(sha512_transform_ssse3)
 
 ########################################################################
 ### Binary Data
diff --git a/arch/x86/crypto/twofish-avx-x86_64-asm_64.S b/arch/x86/crypto/twofish-avx-x86_64-asm_64.S
index 1e87dcde342f..07af53adbc56 100644
--- a/arch/x86/crypto/twofish-avx-x86_64-asm_64.S
+++ b/arch/x86/crypto/twofish-avx-x86_64-asm_64.S
@@ -330,7 +330,7 @@ SYM_FUNC_START_LOCAL(__twofish_dec_blk8)
 	ret;
 SYM_FUNC_END(__twofish_dec_blk8)
 
-ENTRY(twofish_ecb_enc_8way)
+SYM_FUNC_START(twofish_ecb_enc_8way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -348,9 +348,9 @@ ENTRY(twofish_ecb_enc_8way)
 
 	FRAME_END
 	ret;
-ENDPROC(twofish_ecb_enc_8way)
+SYM_FUNC_END(twofish_ecb_enc_8way)
 
-ENTRY(twofish_ecb_dec_8way)
+SYM_FUNC_START(twofish_ecb_dec_8way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -368,9 +368,9 @@ ENTRY(twofish_ecb_dec_8way)
 
 	FRAME_END
 	ret;
-ENDPROC(twofish_ecb_dec_8way)
+SYM_FUNC_END(twofish_ecb_dec_8way)
 
-ENTRY(twofish_cbc_dec_8way)
+SYM_FUNC_START(twofish_cbc_dec_8way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -393,9 +393,9 @@ ENTRY(twofish_cbc_dec_8way)
 
 	FRAME_END
 	ret;
-ENDPROC(twofish_cbc_dec_8way)
+SYM_FUNC_END(twofish_cbc_dec_8way)
 
-ENTRY(twofish_ctr_8way)
+SYM_FUNC_START(twofish_ctr_8way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -420,9 +420,9 @@ ENTRY(twofish_ctr_8way)
 
 	FRAME_END
 	ret;
-ENDPROC(twofish_ctr_8way)
+SYM_FUNC_END(twofish_ctr_8way)
 
-ENTRY(twofish_xts_enc_8way)
+SYM_FUNC_START(twofish_xts_enc_8way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -444,9 +444,9 @@ ENTRY(twofish_xts_enc_8way)
 
 	FRAME_END
 	ret;
-ENDPROC(twofish_xts_enc_8way)
+SYM_FUNC_END(twofish_xts_enc_8way)
 
-ENTRY(twofish_xts_dec_8way)
+SYM_FUNC_START(twofish_xts_dec_8way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -468,4 +468,4 @@ ENTRY(twofish_xts_dec_8way)
 
 	FRAME_END
 	ret;
-ENDPROC(twofish_xts_dec_8way)
+SYM_FUNC_END(twofish_xts_dec_8way)
diff --git a/arch/x86/crypto/twofish-x86_64-asm_64-3way.S b/arch/x86/crypto/twofish-x86_64-asm_64-3way.S
index 1c3b7ceb36d2..ece306e35298 100644
--- a/arch/x86/crypto/twofish-x86_64-asm_64-3way.S
+++ b/arch/x86/crypto/twofish-x86_64-asm_64-3way.S
@@ -216,7 +216,7 @@
 	rorq $32,			RAB2; \
 	outunpack3(mov, RIO, 2, RAB, 2);
 
-ENTRY(__twofish_enc_blk_3way)
+SYM_FUNC_START(__twofish_enc_blk_3way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -270,9 +270,9 @@ ENTRY(__twofish_enc_blk_3way)
 	popq %r14;
 	popq %r15;
 	ret;
-ENDPROC(__twofish_enc_blk_3way)
+SYM_FUNC_END(__twofish_enc_blk_3way)
 
-ENTRY(twofish_dec_blk_3way)
+SYM_FUNC_START(twofish_dec_blk_3way)
 	/* input:
 	 *	%rdi: ctx, CTX
 	 *	%rsi: dst
@@ -309,4 +309,4 @@ ENTRY(twofish_dec_blk_3way)
 	popq %r14;
 	popq %r15;
 	ret;
-ENDPROC(twofish_dec_blk_3way)
+SYM_FUNC_END(twofish_dec_blk_3way)
diff --git a/arch/x86/crypto/twofish-x86_64-asm_64.S b/arch/x86/crypto/twofish-x86_64-asm_64.S
index a350c990dc86..74ef6c55d75f 100644
--- a/arch/x86/crypto/twofish-x86_64-asm_64.S
+++ b/arch/x86/crypto/twofish-x86_64-asm_64.S
@@ -215,7 +215,7 @@
 	xor	%r8d,		d ## D;\
 	ror	$1,		d ## D;
 
-ENTRY(twofish_enc_blk)
+SYM_FUNC_START(twofish_enc_blk)
 	pushq    R1
 
 	/* %rdi contains the ctx address */
@@ -266,9 +266,9 @@ ENTRY(twofish_enc_blk)
 	popq	R1
 	movl	$1,%eax
 	ret
-ENDPROC(twofish_enc_blk)
+SYM_FUNC_END(twofish_enc_blk)
 
-ENTRY(twofish_dec_blk)
+SYM_FUNC_START(twofish_dec_blk)
 	pushq    R1
 
 	/* %rdi contains the ctx address */
@@ -318,4 +318,4 @@ ENTRY(twofish_dec_blk)
 	popq	R1
 	movl	$1,%eax
 	ret
-ENDPROC(twofish_dec_blk)
+SYM_FUNC_END(twofish_dec_blk)
diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
index ab71baad00fb..4de90f9daa1b 100644
--- a/arch/x86/entry/entry_64.S
+++ b/arch/x86/entry/entry_64.S
@@ -14,7 +14,7 @@
  *			at the top of the kernel process stack.
  *
  * Some macro usage:
- * - ENTRY/END:		Define functions in the symbol table.
+ * - SYM_FUNC_START/END:Define functions in the symbol table.
  * - TRACE_IRQ_*:	Trace hardirq state for lock debugging.
  * - idtentry:		Define exception entry points.
  */
@@ -43,10 +43,10 @@
 .section .entry.text, "ax"
 
 #ifdef CONFIG_PARAVIRT
-ENTRY(native_usergs_sysret64)
+SYM_FUNC_START(native_usergs_sysret64)
 	swapgs
 	sysretq
-ENDPROC(native_usergs_sysret64)
+SYM_FUNC_END(native_usergs_sysret64)
 #endif /* CONFIG_PARAVIRT */
 
 .macro TRACE_IRQS_IRETQ
@@ -134,7 +134,7 @@ ENDPROC(native_usergs_sysret64)
  * with them due to bugs in both AMD and Intel CPUs.
  */
 
-ENTRY(entry_SYSCALL_64)
+SYM_FUNC_START(entry_SYSCALL_64)
 	/*
 	 * Interrupts are off on entry.
 	 * We do not frame this tiny irq-off block with TRACE_IRQS_OFF/ON,
@@ -321,7 +321,7 @@ syscall_return_via_sysret:
 opportunistic_sysret_failed:
 	SWAPGS
 	jmp	restore_c_regs_and_iret
-ENDPROC(entry_SYSCALL_64)
+SYM_FUNC_END(entry_SYSCALL_64)
 
 SYM_FUNC_START_LOCAL(stub_ptregs_64)
 	/*
@@ -350,10 +350,10 @@ SYM_FUNC_START_LOCAL(stub_ptregs_64)
 SYM_FUNC_END(stub_ptregs_64)
 
 .macro ptregs_stub func
-ENTRY(ptregs_\func)
+SYM_FUNC_START(ptregs_\func)
 	leaq	\func(%rip), %rax
 	jmp	stub_ptregs_64
-ENDPROC(ptregs_\func)
+SYM_FUNC_END(ptregs_\func)
 .endm
 
 /* Instantiate ptregs_stub for each ptregs-using syscall */
@@ -366,7 +366,7 @@ ENDPROC(ptregs_\func)
  * %rdi: prev task
  * %rsi: next task
  */
-ENTRY(__switch_to_asm)
+SYM_FUNC_START(__switch_to_asm)
 	/*
 	 * Save callee-saved registers
 	 * This must match the order in inactive_task_frame
@@ -396,7 +396,7 @@ ENTRY(__switch_to_asm)
 	popq	%rbp
 
 	jmp	__switch_to
-ENDPROC(__switch_to_asm)
+SYM_FUNC_END(__switch_to_asm)
 
 /*
  * A newly forked process directly context switches into this address.
@@ -405,7 +405,7 @@ ENDPROC(__switch_to_asm)
  * rbx: kernel thread func (NULL for user thread)
  * r12: kernel thread arg
  */
-ENTRY(ret_from_fork)
+SYM_FUNC_START(ret_from_fork)
 	FRAME_BEGIN			/* help unwinder find end of stack */
 	movq	%rax, %rdi
 	call	schedule_tail		/* rdi: 'prev' task parameter */
@@ -432,14 +432,14 @@ ENTRY(ret_from_fork)
 	 */
 	movq	$0, RAX(%rsp)
 	jmp	2b
-ENDPROC(ret_from_fork)
+SYM_FUNC_END(ret_from_fork)
 
 /*
  * Build the entry stubs with some assembler magic.
  * We pack 1 stub into every 8-byte block.
  */
 	.align 8
-ENTRY(irq_entries_start)
+SYM_FUNC_START(irq_entries_start)
     vector=FIRST_EXTERNAL_VECTOR
     .rept (FIRST_SYSTEM_VECTOR - FIRST_EXTERNAL_VECTOR)
 	pushq	$(~vector+0x80)			/* Note: always in signed byte range */
@@ -447,7 +447,7 @@ ENTRY(irq_entries_start)
 	jmp	common_interrupt
 	.align	8
     .endr
-ENDPROC(irq_entries_start)
+SYM_FUNC_END(irq_entries_start)
 
 /*
  * Interrupt entry/exit.
@@ -654,13 +654,13 @@ SYM_FUNC_END(common_interrupt)
  * APIC interrupts.
  */
 .macro apicinterrupt3 num sym do_sym
-ENTRY(\sym)
+SYM_FUNC_START(\sym)
 	ASM_CLAC
 	pushq	$~(\num)
 .Lcommon_\sym:
 	interrupt \do_sym
 	jmp	ret_from_intr
-ENDPROC(\sym)
+SYM_FUNC_END(\sym)
 .endm
 
 #ifdef CONFIG_TRACING
@@ -739,7 +739,7 @@ apicinterrupt IRQ_WORK_VECTOR			irq_work_interrupt		smp_irq_work_interrupt
 #define CPU_TSS_IST(x) PER_CPU_VAR(cpu_tss) + (TSS_ist + ((x) - 1) * 8)
 
 .macro idtentry sym do_sym has_error_code:req paranoid=0 shift_ist=-1
-ENTRY(\sym)
+SYM_FUNC_START(\sym)
 	/* Sanity check */
 	.if \shift_ist != -1 && \paranoid == 0
 	.error "using shift_ist requires paranoid=1"
@@ -826,7 +826,7 @@ ENTRY(\sym)
 
 	jmp	error_exit			/* %ebx: no swapgs flag */
 	.endif
-ENDPROC(\sym)
+SYM_FUNC_END(\sym)
 .endm
 
 #ifdef CONFIG_TRACING
@@ -859,7 +859,7 @@ idtentry simd_coprocessor_error		do_simd_coprocessor_error	has_error_code=0
 	 * Reload gs selector with exception handling
 	 * edi:  new selector
 	 */
-ENTRY(native_load_gs_index)
+SYM_FUNC_START(native_load_gs_index)
 	pushfq
 	DISABLE_INTERRUPTS(CLBR_ANY & ~CLBR_RDI)
 	SWAPGS
@@ -869,7 +869,7 @@ ENTRY(native_load_gs_index)
 	SWAPGS
 	popfq
 	ret
-ENDPROC(native_load_gs_index)
+SYM_FUNC_END(native_load_gs_index)
 EXPORT_SYMBOL(native_load_gs_index)
 
 	_ASM_EXTABLE(.Lgs_change, bad_gs)
@@ -890,7 +890,7 @@ SYM_FUNC_END(bad_gs)
 	.previous
 
 /* Call softirq on interrupt stack. Interrupts are off. */
-ENTRY(do_softirq_own_stack)
+SYM_FUNC_START(do_softirq_own_stack)
 	pushq	%rbp
 	mov	%rsp, %rbp
 	incl	PER_CPU_VAR(irq_count)
@@ -900,7 +900,7 @@ ENTRY(do_softirq_own_stack)
 	leaveq
 	decl	PER_CPU_VAR(irq_count)
 	ret
-ENDPROC(do_softirq_own_stack)
+SYM_FUNC_END(do_softirq_own_stack)
 
 #ifdef CONFIG_XEN
 idtentry xen_hypervisor_callback xen_do_hypervisor_callback has_error_code=0
@@ -952,7 +952,7 @@ SYM_FUNC_END(xen_do_hypervisor_callback)
  * We distinguish between categories by comparing each saved segment register
  * with its current contents: any discrepancy means we in category 1.
  */
-ENTRY(xen_failsafe_callback)
+SYM_FUNC_START(xen_failsafe_callback)
 	movl	%ds, %ecx
 	cmpw	%cx, 0x10(%rsp)
 	jne	1f
@@ -983,7 +983,7 @@ ENTRY(xen_failsafe_callback)
 	SAVE_EXTRA_REGS
 	ENCODE_FRAME_POINTER
 	jmp	error_exit
-ENDPROC(xen_failsafe_callback)
+SYM_FUNC_END(xen_failsafe_callback)
 
 apicinterrupt3 HYPERVISOR_CALLBACK_VECTOR \
 	xen_hvm_callback_vector xen_evtchn_do_upcall
@@ -1162,7 +1162,7 @@ SYM_FUNC_START_LOCAL(error_exit)
 SYM_FUNC_END(error_exit)
 
 /* Runs on exception stack */
-ENTRY(nmi)
+SYM_FUNC_START(nmi)
 	/*
 	 * Fix up the exception frame if we're on Xen.
 	 * PARAVIRT_ADJUST_EXCEPTION_FRAME is guaranteed to push at most
@@ -1507,14 +1507,14 @@ nmi_restore:
 	 * mode, so this cannot result in a fault.
 	 */
 	INTERRUPT_RETURN
-ENDPROC(nmi)
+SYM_FUNC_END(nmi)
 
-ENTRY(ignore_sysret)
+SYM_FUNC_START(ignore_sysret)
 	mov	$-ENOSYS, %eax
 	sysret
-ENDPROC(ignore_sysret)
+SYM_FUNC_END(ignore_sysret)
 
-ENTRY(rewind_stack_do_exit)
+SYM_FUNC_START(rewind_stack_do_exit)
 	/* Prevent any naive code from trying to unwind to our caller. */
 	xorl	%ebp, %ebp
 
@@ -1523,4 +1523,4 @@ ENTRY(rewind_stack_do_exit)
 
 	call	do_exit
 1:	jmp 1b
-ENDPROC(rewind_stack_do_exit)
+SYM_FUNC_END(rewind_stack_do_exit)
diff --git a/arch/x86/entry/entry_64_compat.S b/arch/x86/entry/entry_64_compat.S
index b7934ef3f5bb..c6163b3abc8c 100644
--- a/arch/x86/entry/entry_64_compat.S
+++ b/arch/x86/entry/entry_64_compat.S
@@ -45,7 +45,7 @@
  * ebp  user stack
  * 0(%ebp) arg6
  */
-ENTRY(entry_SYSENTER_compat)
+SYM_FUNC_START(entry_SYSENTER_compat)
 	/* Interrupts are off on entry. */
 	SWAPGS_UNSAFE_STACK
 	movq	PER_CPU_VAR(cpu_current_top_of_stack), %rsp
@@ -132,7 +132,7 @@ ENTRY(entry_SYSENTER_compat)
 	popfq
 	jmp	.Lsysenter_flags_fixed
 SYM_FUNC_INNER_LABEL(__end_entry_SYSENTER_compat, SYM_V_GLOBAL)
-ENDPROC(entry_SYSENTER_compat)
+SYM_FUNC_END(entry_SYSENTER_compat)
 
 /*
  * 32-bit SYSCALL entry.
@@ -181,7 +181,7 @@ ENDPROC(entry_SYSENTER_compat)
  * esp  user stack
  * 0(%esp) arg6
  */
-ENTRY(entry_SYSCALL_compat)
+SYM_FUNC_START(entry_SYSCALL_compat)
 	/* Interrupts are off on entry. */
 	SWAPGS_UNSAFE_STACK
 
@@ -262,7 +262,7 @@ sysret32_from_system_call:
 	movq	RSP-ORIG_RAX(%rsp), %rsp
 	swapgs
 	sysretl
-ENDPROC(entry_SYSCALL_compat)
+SYM_FUNC_END(entry_SYSCALL_compat)
 
 /*
  * 32-bit legacy system call entry.
@@ -290,7 +290,7 @@ ENDPROC(entry_SYSCALL_compat)
  * edi  arg5
  * ebp  arg6
  */
-ENTRY(entry_INT80_compat)
+SYM_FUNC_START(entry_INT80_compat)
 	/*
 	 * Interrupts are off on entry.
 	 */
@@ -340,9 +340,9 @@ ENTRY(entry_INT80_compat)
 	TRACE_IRQS_ON
 	SWAPGS
 	jmp	restore_regs_and_iret
-ENDPROC(entry_INT80_compat)
+SYM_FUNC_END(entry_INT80_compat)
 
-ENTRY(stub32_clone)
+SYM_FUNC_START(stub32_clone)
 	/*
 	 * The 32-bit clone ABI is: clone(..., int tls_val, int *child_tidptr).
 	 * The 64-bit clone ABI is: clone(..., int *child_tidptr, int tls_val).
@@ -352,4 +352,4 @@ ENTRY(stub32_clone)
 	 */
 	xchg	%r8, %rcx
 	jmp	sys_clone
-ENDPROC(stub32_clone)
+SYM_FUNC_END(stub32_clone)
diff --git a/arch/x86/kernel/acpi/wakeup_64.S b/arch/x86/kernel/acpi/wakeup_64.S
index 8fca92dd9144..c159b25bf056 100644
--- a/arch/x86/kernel/acpi/wakeup_64.S
+++ b/arch/x86/kernel/acpi/wakeup_64.S
@@ -13,7 +13,7 @@
 	/*
 	 * Hooray, we are in Long 64-bit mode (but still running in low memory)
 	 */
-ENTRY(wakeup_long64)
+SYM_FUNC_START(wakeup_long64)
 	movq	saved_magic, %rax
 	movq	$0x123456789abcdef0, %rdx
 	cmpq	%rdx, %rax
@@ -34,13 +34,13 @@ ENTRY(wakeup_long64)
 
 	movq	saved_rip, %rax
 	jmp	*%rax
-ENDPROC(wakeup_long64)
+SYM_FUNC_END(wakeup_long64)
 
 SYM_FUNC_START_LOCAL(bogus_64_magic)
 	jmp	bogus_64_magic
 SYM_FUNC_END(bogus_64_magic)
 
-ENTRY(do_suspend_lowlevel)
+SYM_FUNC_START(do_suspend_lowlevel)
 	FRAME_BEGIN
 	subq	$8, %rsp
 	xorl	%eax, %eax
@@ -123,7 +123,7 @@ ENTRY(do_suspend_lowlevel)
 	addq	$8, %rsp
 	FRAME_END
 	jmp	restore_processor_state
-ENDPROC(do_suspend_lowlevel)
+SYM_FUNC_END(do_suspend_lowlevel)
 
 .data
 SYM_DATA_SIMPLE_LOCAL(saved_rbp, .quad 0)
diff --git a/arch/x86/kernel/ftrace_64.S b/arch/x86/kernel/ftrace_64.S
index aef60bbe854d..0970d85693c4 100644
--- a/arch/x86/kernel/ftrace_64.S
+++ b/arch/x86/kernel/ftrace_64.S
@@ -145,11 +145,11 @@ EXPORT_SYMBOL(mcount)
 
 #ifdef CONFIG_DYNAMIC_FTRACE
 
-ENTRY(function_hook)
+SYM_FUNC_START(function_hook)
 	retq
-ENDPROC(function_hook)
+SYM_FUNC_END(function_hook)
 
-ENTRY(ftrace_caller)
+SYM_FUNC_START(ftrace_caller)
 	/* save_mcount_regs fills in first two parameters */
 	save_mcount_regs
 
@@ -183,9 +183,9 @@ SYM_FUNC_INNER_LABEL(ftrace_graph_call, SYM_V_GLOBAL)
 /* This is weak to keep gas from relaxing the jumps */
 WEAK(ftrace_stub)
 	retq
-ENDPROC(ftrace_caller)
+SYM_FUNC_END(ftrace_caller)
 
-ENTRY(ftrace_regs_caller)
+SYM_FUNC_START(ftrace_regs_caller)
 	/* Save the current flags before any operations that can change them */
 	pushfq
 
@@ -254,12 +254,12 @@ SYM_FUNC_INNER_LABEL(ftrace_regs_caller_end, SYM_V_GLOBAL)
 
 	jmp ftrace_epilogue
 
-ENDPROC(ftrace_regs_caller)
+SYM_FUNC_END(ftrace_regs_caller)
 
 
 #else /* ! CONFIG_DYNAMIC_FTRACE */
 
-ENTRY(function_hook)
+SYM_FUNC_START(function_hook)
 	cmpq $ftrace_stub, ftrace_trace_function
 	jnz trace
 
@@ -290,11 +290,11 @@ trace:
 	restore_mcount_regs
 
 	jmp fgraph_trace
-ENDPROC(function_hook)
+SYM_FUNC_END(function_hook)
 #endif /* CONFIG_DYNAMIC_FTRACE */
 
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
-ENTRY(ftrace_graph_caller)
+SYM_FUNC_START(ftrace_graph_caller)
 	/* Saves rbp into %rdx and fills first parameter  */
 	save_mcount_regs
 
@@ -312,9 +312,9 @@ ENTRY(ftrace_graph_caller)
 	restore_mcount_regs
 
 	retq
-ENDPROC(ftrace_graph_caller)
+SYM_FUNC_END(ftrace_graph_caller)
 
-ENTRY(return_to_handler)
+SYM_FUNC_START(return_to_handler)
 	subq  $24, %rsp
 
 	/* Save the return values */
@@ -329,5 +329,5 @@ ENTRY(return_to_handler)
 	movq (%rsp), %rax
 	addq $24, %rsp
 	jmp *%rdi
-ENDPROC(return_to_handler)
+SYM_FUNC_END(return_to_handler)
 #endif
diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S
index 3c9037a65ee9..a08c5d891a80 100644
--- a/arch/x86/kernel/head_64.S
+++ b/arch/x86/kernel/head_64.S
@@ -46,7 +46,7 @@ L3_START_KERNEL = pud_index(__START_KERNEL_map)
 	.text
 	__HEAD
 	.code64
-ENTRY(startup_64)
+SYM_FUNC_START(startup_64)
 	/*
 	 * At this point the CPU runs in 64bit mode CS.L = 1 CS.D = 0,
 	 * and someone has loaded an identity mapped page table
@@ -167,9 +167,9 @@ ENTRY(startup_64)
 .Lskip_fixup:
 	movq	$(early_level4_pgt - __START_KERNEL_map), %rax
 	jmp 1f
-ENDPROC(startup_64)
+SYM_FUNC_END(startup_64)
 
-ENTRY(secondary_startup_64)
+SYM_FUNC_START(secondary_startup_64)
 	/*
 	 * At this point the CPU runs in 64bit mode CS.L = 1 CS.D = 0,
 	 * and someone has loaded a mapped page table.
@@ -304,7 +304,7 @@ ENTRY(secondary_startup_64)
 	pushq	%rax		# target address in negative space
 	lretq
 .Lafter_lret:
-ENDPROC(secondary_startup_64)
+SYM_FUNC_END(secondary_startup_64)
 
 #include "verify_cpu.S"
 
@@ -314,10 +314,10 @@ ENDPROC(secondary_startup_64)
  * up already except stack. We just set up stack here. Then call
  * start_secondary() via .Ljump_to_C_code.
  */
-ENTRY(start_cpu0)
+SYM_FUNC_START(start_cpu0)
 	movq	initial_stack(%rip), %rsp
 	jmp	.Ljump_to_C_code
-ENDPROC(start_cpu0)
+SYM_FUNC_END(start_cpu0)
 #endif
 
 	/* Both SMP bootup and ACPI suspend change these variables */
@@ -338,7 +338,7 @@ SYM_FUNC_START_LOCAL(bad_address)
 SYM_FUNC_END(bad_address)
 
 	__INIT
-ENTRY(early_idt_handler_array)
+SYM_FUNC_START(early_idt_handler_array)
 	# 104(%rsp) %rflags
 	#  96(%rsp) %cs
 	#  88(%rsp) %rip
@@ -353,7 +353,7 @@ ENTRY(early_idt_handler_array)
 	i = i + 1
 	.fill early_idt_handler_array + i*EARLY_IDT_HANDLER_SIZE - ., 1, 0xcc
 	.endr
-ENDPROC(early_idt_handler_array)
+SYM_FUNC_END(early_idt_handler_array)
 
 SYM_FUNC_START_LOCAL(early_idt_handler_common)
 	/*
diff --git a/arch/x86/lib/checksum_32.S b/arch/x86/lib/checksum_32.S
index 4d34bb548b41..a048436ce3ac 100644
--- a/arch/x86/lib/checksum_32.S
+++ b/arch/x86/lib/checksum_32.S
@@ -283,7 +283,7 @@ unsigned int csum_partial_copy_generic (const char *src, char *dst,
 #define ARGBASE 16		
 #define FP		12
 		
-ENTRY(csum_partial_copy_generic)
+SYM_FUNC_START(csum_partial_copy_generic)
 	subl  $4,%esp	
 	pushl %edi
 	pushl %esi
@@ -401,7 +401,7 @@ DST(	movb %cl, (%edi)	)
 	popl %edi
 	popl %ecx			# equivalent to addl $4,%esp
 	ret	
-ENDPROC(csum_partial_copy_generic)
+SYM_FUNC_END(csum_partial_copy_generic)
 
 #else
 
@@ -419,7 +419,7 @@ ENDPROC(csum_partial_copy_generic)
 
 #define ARGBASE 12
 		
-ENTRY(csum_partial_copy_generic)
+SYM_FUNC_START(csum_partial_copy_generic)
 	pushl %ebx
 	pushl %edi
 	pushl %esi
@@ -486,7 +486,7 @@ DST(	movb %dl, (%edi)         )
 	popl %edi
 	popl %ebx
 	ret
-ENDPROC(csum_partial_copy_generic)
+SYM_FUNC_END(csum_partial_copy_generic)
 				
 #undef ROUND
 #undef ROUND1		
diff --git a/arch/x86/lib/clear_page_64.S b/arch/x86/lib/clear_page_64.S
index 81b1635d67de..8c4c26c282fd 100644
--- a/arch/x86/lib/clear_page_64.S
+++ b/arch/x86/lib/clear_page_64.S
@@ -14,15 +14,15 @@
  * Zero a page.
  * %rdi	- page
  */
-ENTRY(clear_page_rep)
+SYM_FUNC_START(clear_page_rep)
 	movl $4096/8,%ecx
 	xorl %eax,%eax
 	rep stosq
 	ret
-ENDPROC(clear_page_rep)
+SYM_FUNC_END(clear_page_rep)
 EXPORT_SYMBOL_GPL(clear_page_rep)
 
-ENTRY(clear_page_orig)
+SYM_FUNC_START(clear_page_orig)
 	xorl   %eax,%eax
 	movl   $4096/64,%ecx
 	.p2align 4
@@ -41,13 +41,13 @@ ENTRY(clear_page_orig)
 	jnz	.Lloop
 	nop
 	ret
-ENDPROC(clear_page_orig)
+SYM_FUNC_END(clear_page_orig)
 EXPORT_SYMBOL_GPL(clear_page_orig)
 
-ENTRY(clear_page_erms)
+SYM_FUNC_START(clear_page_erms)
 	movl $4096,%ecx
 	xorl %eax,%eax
 	rep stosb
 	ret
-ENDPROC(clear_page_erms)
+SYM_FUNC_END(clear_page_erms)
 EXPORT_SYMBOL_GPL(clear_page_erms)
diff --git a/arch/x86/lib/cmpxchg16b_emu.S b/arch/x86/lib/cmpxchg16b_emu.S
index 9b330242e740..b6ba6360b3ca 100644
--- a/arch/x86/lib/cmpxchg16b_emu.S
+++ b/arch/x86/lib/cmpxchg16b_emu.S
@@ -19,7 +19,7 @@
  * %rcx : high 64 bits of new value
  * %al  : Operation successful
  */
-ENTRY(this_cpu_cmpxchg16b_emu)
+SYM_FUNC_START(this_cpu_cmpxchg16b_emu)
 
 #
 # Emulate 'cmpxchg16b %gs:(%rsi)' except we return the result in %al not
@@ -50,4 +50,4 @@ ENTRY(this_cpu_cmpxchg16b_emu)
 	xor %al,%al
 	ret
 
-ENDPROC(this_cpu_cmpxchg16b_emu)
+SYM_FUNC_END(this_cpu_cmpxchg16b_emu)
diff --git a/arch/x86/lib/cmpxchg8b_emu.S b/arch/x86/lib/cmpxchg8b_emu.S
index 03a186fc06ea..77aa18db3968 100644
--- a/arch/x86/lib/cmpxchg8b_emu.S
+++ b/arch/x86/lib/cmpxchg8b_emu.S
@@ -19,7 +19,7 @@
  * %ebx : low 32 bits of new value
  * %ecx : high 32 bits of new value
  */
-ENTRY(cmpxchg8b_emu)
+SYM_FUNC_START(cmpxchg8b_emu)
 
 #
 # Emulate 'cmpxchg8b (%esi)' on UP except we don't
@@ -48,5 +48,5 @@ ENTRY(cmpxchg8b_emu)
 	popfl
 	ret
 
-ENDPROC(cmpxchg8b_emu)
+SYM_FUNC_END(cmpxchg8b_emu)
 EXPORT_SYMBOL(cmpxchg8b_emu)
diff --git a/arch/x86/lib/copy_page_64.S b/arch/x86/lib/copy_page_64.S
index e1ee50bc161a..6ba635d23fcc 100644
--- a/arch/x86/lib/copy_page_64.S
+++ b/arch/x86/lib/copy_page_64.S
@@ -12,12 +12,12 @@
  * prefetch distance based on SMP/UP.
  */
 	ALIGN
-ENTRY(copy_page)
+SYM_FUNC_START(copy_page)
 	ALTERNATIVE "jmp copy_page_regs", "", X86_FEATURE_REP_GOOD
 	movl	$4096/8, %ecx
 	rep	movsq
 	ret
-ENDPROC(copy_page)
+SYM_FUNC_END(copy_page)
 EXPORT_SYMBOL(copy_page)
 
 SYM_FUNC_START_LOCAL(copy_page_regs)
diff --git a/arch/x86/lib/copy_user_64.S b/arch/x86/lib/copy_user_64.S
index c5959576c315..a4253494faa8 100644
--- a/arch/x86/lib/copy_user_64.S
+++ b/arch/x86/lib/copy_user_64.S
@@ -29,7 +29,7 @@
  * Output:
  * eax uncopied bytes or 0 if successful.
  */
-ENTRY(copy_user_generic_unrolled)
+SYM_FUNC_START(copy_user_generic_unrolled)
 	ASM_STAC
 	cmpl $8,%edx
 	jb 20f		/* less then 8 bytes, go to byte copy loop */
@@ -111,7 +111,7 @@ ENTRY(copy_user_generic_unrolled)
 	_ASM_EXTABLE(19b,40b)
 	_ASM_EXTABLE(21b,50b)
 	_ASM_EXTABLE(22b,50b)
-ENDPROC(copy_user_generic_unrolled)
+SYM_FUNC_END(copy_user_generic_unrolled)
 EXPORT_SYMBOL(copy_user_generic_unrolled)
 
 /* Some CPUs run faster using the string copy instructions.
@@ -132,7 +132,7 @@ EXPORT_SYMBOL(copy_user_generic_unrolled)
  * Output:
  * eax uncopied bytes or 0 if successful.
  */
-ENTRY(copy_user_generic_string)
+SYM_FUNC_START(copy_user_generic_string)
 	ASM_STAC
 	cmpl $8,%edx
 	jb 2f		/* less than 8 bytes, go to byte copy loop */
@@ -157,7 +157,7 @@ ENTRY(copy_user_generic_string)
 
 	_ASM_EXTABLE(1b,11b)
 	_ASM_EXTABLE(3b,12b)
-ENDPROC(copy_user_generic_string)
+SYM_FUNC_END(copy_user_generic_string)
 EXPORT_SYMBOL(copy_user_generic_string)
 
 /*
@@ -172,7 +172,7 @@ EXPORT_SYMBOL(copy_user_generic_string)
  * Output:
  * eax uncopied bytes or 0 if successful.
  */
-ENTRY(copy_user_enhanced_fast_string)
+SYM_FUNC_START(copy_user_enhanced_fast_string)
 	ASM_STAC
 	movl %edx,%ecx
 1:	rep
@@ -187,7 +187,7 @@ ENTRY(copy_user_enhanced_fast_string)
 	.previous
 
 	_ASM_EXTABLE(1b,12b)
-ENDPROC(copy_user_enhanced_fast_string)
+SYM_FUNC_END(copy_user_enhanced_fast_string)
 EXPORT_SYMBOL(copy_user_enhanced_fast_string)
 
 /*
@@ -199,7 +199,7 @@ EXPORT_SYMBOL(copy_user_enhanced_fast_string)
  *  - Require 8-byte alignment when size is 8 bytes or larger.
  *  - Require 4-byte alignment when size is 4 bytes.
  */
-ENTRY(__copy_user_nocache)
+SYM_FUNC_START(__copy_user_nocache)
 	ASM_STAC
 
 	/* If size is less than 8 bytes, go to 4-byte copy */
@@ -338,5 +338,5 @@ ENTRY(__copy_user_nocache)
 	_ASM_EXTABLE(31b,.L_fixup_4b_copy)
 	_ASM_EXTABLE(40b,.L_fixup_1b_copy)
 	_ASM_EXTABLE(41b,.L_fixup_1b_copy)
-ENDPROC(__copy_user_nocache)
+SYM_FUNC_END(__copy_user_nocache)
 EXPORT_SYMBOL(__copy_user_nocache)
diff --git a/arch/x86/lib/csum-copy_64.S b/arch/x86/lib/csum-copy_64.S
index 7e48807b2fa1..e93301673bb1 100644
--- a/arch/x86/lib/csum-copy_64.S
+++ b/arch/x86/lib/csum-copy_64.S
@@ -45,7 +45,7 @@
 	.endm
 
 
-ENTRY(csum_partial_copy_generic)
+SYM_FUNC_START(csum_partial_copy_generic)
 	cmpl	$3*64, %edx
 	jle	.Lignore
 
@@ -221,4 +221,4 @@ ENTRY(csum_partial_copy_generic)
 	jz   .Lende
 	movl $-EFAULT, (%rax)
 	jmp .Lende
-ENDPROC(csum_partial_copy_generic)
+SYM_FUNC_END(csum_partial_copy_generic)
diff --git a/arch/x86/lib/getuser.S b/arch/x86/lib/getuser.S
index 29f0707a3913..56b4a13678f5 100644
--- a/arch/x86/lib/getuser.S
+++ b/arch/x86/lib/getuser.S
@@ -35,7 +35,7 @@
 #include <asm/export.h>
 
 	.text
-ENTRY(__get_user_1)
+SYM_FUNC_START(__get_user_1)
 	mov PER_CPU_VAR(current_task), %_ASM_DX
 	cmp TASK_addr_limit(%_ASM_DX),%_ASM_AX
 	jae bad_get_user
@@ -44,10 +44,10 @@ ENTRY(__get_user_1)
 	xor %eax,%eax
 	ASM_CLAC
 	ret
-ENDPROC(__get_user_1)
+SYM_FUNC_END(__get_user_1)
 EXPORT_SYMBOL(__get_user_1)
 
-ENTRY(__get_user_2)
+SYM_FUNC_START(__get_user_2)
 	add $1,%_ASM_AX
 	jc bad_get_user
 	mov PER_CPU_VAR(current_task), %_ASM_DX
@@ -58,10 +58,10 @@ ENTRY(__get_user_2)
 	xor %eax,%eax
 	ASM_CLAC
 	ret
-ENDPROC(__get_user_2)
+SYM_FUNC_END(__get_user_2)
 EXPORT_SYMBOL(__get_user_2)
 
-ENTRY(__get_user_4)
+SYM_FUNC_START(__get_user_4)
 	add $3,%_ASM_AX
 	jc bad_get_user
 	mov PER_CPU_VAR(current_task), %_ASM_DX
@@ -72,10 +72,10 @@ ENTRY(__get_user_4)
 	xor %eax,%eax
 	ASM_CLAC
 	ret
-ENDPROC(__get_user_4)
+SYM_FUNC_END(__get_user_4)
 EXPORT_SYMBOL(__get_user_4)
 
-ENTRY(__get_user_8)
+SYM_FUNC_START(__get_user_8)
 #ifdef CONFIG_X86_64
 	add $7,%_ASM_AX
 	jc bad_get_user
@@ -100,7 +100,7 @@ ENTRY(__get_user_8)
 	ASM_CLAC
 	ret
 #endif
-ENDPROC(__get_user_8)
+SYM_FUNC_END(__get_user_8)
 EXPORT_SYMBOL(__get_user_8)
 
 
diff --git a/arch/x86/lib/hweight.S b/arch/x86/lib/hweight.S
index 23d893cbc200..f520a1a92ef6 100644
--- a/arch/x86/lib/hweight.S
+++ b/arch/x86/lib/hweight.S
@@ -7,7 +7,7 @@
  * unsigned int __sw_hweight32(unsigned int w)
  * %rdi: w
  */
-ENTRY(__sw_hweight32)
+SYM_FUNC_START(__sw_hweight32)
 
 #ifdef CONFIG_X86_64
 	movl %edi, %eax				# w
@@ -32,10 +32,10 @@ ENTRY(__sw_hweight32)
 	shrl $24, %eax				# w = w_tmp >> 24
 	__ASM_SIZE(pop,) %__ASM_REG(dx)
 	ret
-ENDPROC(__sw_hweight32)
+SYM_FUNC_END(__sw_hweight32)
 EXPORT_SYMBOL(__sw_hweight32)
 
-ENTRY(__sw_hweight64)
+SYM_FUNC_START(__sw_hweight64)
 #ifdef CONFIG_X86_64
 	pushq   %rdi
 	pushq   %rdx
@@ -78,5 +78,5 @@ ENTRY(__sw_hweight64)
 	popl    %ecx
 	ret
 #endif
-ENDPROC(__sw_hweight64)
+SYM_FUNC_END(__sw_hweight64)
 EXPORT_SYMBOL(__sw_hweight64)
diff --git a/arch/x86/lib/iomap_copy_64.S b/arch/x86/lib/iomap_copy_64.S
index 33147fef3452..2246fbf32fa8 100644
--- a/arch/x86/lib/iomap_copy_64.S
+++ b/arch/x86/lib/iomap_copy_64.S
@@ -20,8 +20,8 @@
 /*
  * override generic version in lib/iomap_copy.c
  */
-ENTRY(__iowrite32_copy)
+SYM_FUNC_START(__iowrite32_copy)
 	movl %edx,%ecx
 	rep movsd
 	ret
-ENDPROC(__iowrite32_copy)
+SYM_FUNC_END(__iowrite32_copy)
diff --git a/arch/x86/lib/memcpy_64.S b/arch/x86/lib/memcpy_64.S
index 728703c47d58..9bec63e212a8 100644
--- a/arch/x86/lib/memcpy_64.S
+++ b/arch/x86/lib/memcpy_64.S
@@ -188,7 +188,7 @@ SYM_FUNC_END(memcpy_orig)
  * Note that we only catch machine checks when reading the source addresses.
  * Writes to target are posted and don't generate machine checks.
  */
-ENTRY(memcpy_mcsafe_unrolled)
+SYM_FUNC_START(memcpy_mcsafe_unrolled)
 	cmpl $8, %edx
 	/* Less than 8 bytes? Go to byte copy loop */
 	jb .L_no_whole_words
@@ -276,7 +276,7 @@ ENTRY(memcpy_mcsafe_unrolled)
 .L_done_memcpy_trap:
 	xorq %rax, %rax
 	ret
-ENDPROC(memcpy_mcsafe_unrolled)
+SYM_FUNC_END(memcpy_mcsafe_unrolled)
 EXPORT_SYMBOL_GPL(memcpy_mcsafe_unrolled)
 
 	.section .fixup, "ax"
diff --git a/arch/x86/lib/memmove_64.S b/arch/x86/lib/memmove_64.S
index d22af97e5b27..4f3f6359fcf9 100644
--- a/arch/x86/lib/memmove_64.S
+++ b/arch/x86/lib/memmove_64.S
@@ -26,7 +26,7 @@
 .weak memmove
 
 SYM_FUNC_START_ALIAS(memmove)
-ENTRY(__memmove)
+SYM_FUNC_START(__memmove)
 
 	/* Handle more 32 bytes in loop */
 	mov %rdi, %rax
@@ -206,7 +206,7 @@ ENTRY(__memmove)
 	movb %r11b, (%rdi)
 13:
 	retq
-ENDPROC(__memmove)
+SYM_FUNC_END(__memmove)
 SYM_FUNC_END_ALIAS(memmove)
 EXPORT_SYMBOL(__memmove)
 EXPORT_SYMBOL(memmove)
diff --git a/arch/x86/lib/memset_64.S b/arch/x86/lib/memset_64.S
index c63ae9987612..cee5514926e6 100644
--- a/arch/x86/lib/memset_64.S
+++ b/arch/x86/lib/memset_64.S
@@ -19,7 +19,7 @@
  * rax   original destination
  */
 SYM_FUNC_START_ALIAS(memset)
-ENTRY(__memset)
+SYM_FUNC_START(__memset)
 	/*
 	 * Some CPUs support enhanced REP MOVSB/STOSB feature. It is recommended
 	 * to use it when possible. If not available, use fast string instructions.
@@ -42,7 +42,7 @@ ENTRY(__memset)
 	rep stosb
 	movq %r9,%rax
 	ret
-ENDPROC(__memset)
+SYM_FUNC_END(__memset)
 SYM_FUNC_END_ALIAS(memset)
 EXPORT_SYMBOL(memset)
 EXPORT_SYMBOL(__memset)
diff --git a/arch/x86/lib/msr-reg.S b/arch/x86/lib/msr-reg.S
index c81556409bbb..1e8a9cb07b4f 100644
--- a/arch/x86/lib/msr-reg.S
+++ b/arch/x86/lib/msr-reg.S
@@ -11,7 +11,7 @@
  *
  */
 .macro op_safe_regs op
-ENTRY(\op\()_safe_regs)
+SYM_FUNC_START(\op\()_safe_regs)
 	pushq %rbx
 	pushq %rbp
 	movq	%rdi, %r10	/* Save pointer */
@@ -40,13 +40,13 @@ ENTRY(\op\()_safe_regs)
 	jmp     2b
 
 	_ASM_EXTABLE(1b, 3b)
-ENDPROC(\op\()_safe_regs)
+SYM_FUNC_END(\op\()_safe_regs)
 .endm
 
 #else /* X86_32 */
 
 .macro op_safe_regs op
-ENTRY(\op\()_safe_regs)
+SYM_FUNC_START(\op\()_safe_regs)
 	pushl %ebx
 	pushl %ebp
 	pushl %esi
@@ -82,7 +82,7 @@ ENTRY(\op\()_safe_regs)
 	jmp     2b
 
 	_ASM_EXTABLE(1b, 3b)
-ENDPROC(\op\()_safe_regs)
+SYM_FUNC_END(\op\()_safe_regs)
 .endm
 
 #endif
diff --git a/arch/x86/lib/putuser.S b/arch/x86/lib/putuser.S
index d77883f36875..4d015af97968 100644
--- a/arch/x86/lib/putuser.S
+++ b/arch/x86/lib/putuser.S
@@ -35,7 +35,7 @@
 		ret
 
 .text
-ENTRY(__put_user_1)
+SYM_FUNC_START(__put_user_1)
 	ENTER
 	cmp TASK_addr_limit(%_ASM_BX),%_ASM_CX
 	jae bad_put_user
@@ -43,10 +43,10 @@ ENTRY(__put_user_1)
 1:	movb %al,(%_ASM_CX)
 	xor %eax,%eax
 	EXIT
-ENDPROC(__put_user_1)
+SYM_FUNC_END(__put_user_1)
 EXPORT_SYMBOL(__put_user_1)
 
-ENTRY(__put_user_2)
+SYM_FUNC_START(__put_user_2)
 	ENTER
 	mov TASK_addr_limit(%_ASM_BX),%_ASM_BX
 	sub $1,%_ASM_BX
@@ -56,10 +56,10 @@ ENTRY(__put_user_2)
 2:	movw %ax,(%_ASM_CX)
 	xor %eax,%eax
 	EXIT
-ENDPROC(__put_user_2)
+SYM_FUNC_END(__put_user_2)
 EXPORT_SYMBOL(__put_user_2)
 
-ENTRY(__put_user_4)
+SYM_FUNC_START(__put_user_4)
 	ENTER
 	mov TASK_addr_limit(%_ASM_BX),%_ASM_BX
 	sub $3,%_ASM_BX
@@ -69,10 +69,10 @@ ENTRY(__put_user_4)
 3:	movl %eax,(%_ASM_CX)
 	xor %eax,%eax
 	EXIT
-ENDPROC(__put_user_4)
+SYM_FUNC_END(__put_user_4)
 EXPORT_SYMBOL(__put_user_4)
 
-ENTRY(__put_user_8)
+SYM_FUNC_START(__put_user_8)
 	ENTER
 	mov TASK_addr_limit(%_ASM_BX),%_ASM_BX
 	sub $7,%_ASM_BX
@@ -85,7 +85,7 @@ ENTRY(__put_user_8)
 #endif
 	xor %eax,%eax
 	EXIT
-ENDPROC(__put_user_8)
+SYM_FUNC_END(__put_user_8)
 EXPORT_SYMBOL(__put_user_8)
 
 SYM_FUNC_START_LOCAL(bad_put_user)
diff --git a/arch/x86/lib/rwsem.S b/arch/x86/lib/rwsem.S
index bf2c6074efd2..e3905664944d 100644
--- a/arch/x86/lib/rwsem.S
+++ b/arch/x86/lib/rwsem.S
@@ -86,7 +86,7 @@
 #endif
 
 /* Fix up special calling conventions */
-ENTRY(call_rwsem_down_read_failed)
+SYM_FUNC_START(call_rwsem_down_read_failed)
 	FRAME_BEGIN
 	save_common_regs
 	__ASM_SIZE(push,) %__ASM_REG(dx)
@@ -96,9 +96,9 @@ ENTRY(call_rwsem_down_read_failed)
 	restore_common_regs
 	FRAME_END
 	ret
-ENDPROC(call_rwsem_down_read_failed)
+SYM_FUNC_END(call_rwsem_down_read_failed)
 
-ENTRY(call_rwsem_down_write_failed)
+SYM_FUNC_START(call_rwsem_down_write_failed)
 	FRAME_BEGIN
 	save_common_regs
 	movq %rax,%rdi
@@ -106,9 +106,9 @@ ENTRY(call_rwsem_down_write_failed)
 	restore_common_regs
 	FRAME_END
 	ret
-ENDPROC(call_rwsem_down_write_failed)
+SYM_FUNC_END(call_rwsem_down_write_failed)
 
-ENTRY(call_rwsem_down_write_failed_killable)
+SYM_FUNC_START(call_rwsem_down_write_failed_killable)
 	FRAME_BEGIN
 	save_common_regs
 	movq %rax,%rdi
@@ -116,9 +116,9 @@ ENTRY(call_rwsem_down_write_failed_killable)
 	restore_common_regs
 	FRAME_END
 	ret
-ENDPROC(call_rwsem_down_write_failed_killable)
+SYM_FUNC_END(call_rwsem_down_write_failed_killable)
 
-ENTRY(call_rwsem_wake)
+SYM_FUNC_START(call_rwsem_wake)
 	FRAME_BEGIN
 	/* do nothing if still outstanding active readers */
 	__ASM_HALF_SIZE(dec) %__ASM_HALF_REG(dx)
@@ -129,9 +129,9 @@ ENTRY(call_rwsem_wake)
 	restore_common_regs
 1:	FRAME_END
 	ret
-ENDPROC(call_rwsem_wake)
+SYM_FUNC_END(call_rwsem_wake)
 
-ENTRY(call_rwsem_downgrade_wake)
+SYM_FUNC_START(call_rwsem_downgrade_wake)
 	FRAME_BEGIN
 	save_common_regs
 	__ASM_SIZE(push,) %__ASM_REG(dx)
@@ -141,4 +141,4 @@ ENTRY(call_rwsem_downgrade_wake)
 	restore_common_regs
 	FRAME_END
 	ret
-ENDPROC(call_rwsem_downgrade_wake)
+SYM_FUNC_END(call_rwsem_downgrade_wake)
diff --git a/arch/x86/net/bpf_jit.S b/arch/x86/net/bpf_jit.S
index 823edd6f1db7..8e7075280148 100644
--- a/arch/x86/net/bpf_jit.S
+++ b/arch/x86/net/bpf_jit.S
@@ -23,12 +23,12 @@
 	32 /* space for rbx,r13,r14,r15 */ + \
 	8 /* space for skb_copy_bits */)
 
-ENTRY(sk_load_word)
+SYM_FUNC_START(sk_load_word)
 	test	%esi,%esi
 	js	bpf_slow_path_word_neg
-ENDPROC(sk_load_word)
+SYM_FUNC_END(sk_load_word)
 
-ENTRY(sk_load_word_positive_offset)
+SYM_FUNC_START(sk_load_word_positive_offset)
 	mov	%r9d,%eax		# hlen
 	sub	%esi,%eax		# hlen - offset
 	cmp	$3,%eax
@@ -36,14 +36,14 @@ ENTRY(sk_load_word_positive_offset)
 	mov     (SKBDATA,%rsi),%eax
 	bswap   %eax  			/* ntohl() */
 	ret
-ENDPROC(sk_load_word_positive_offset)
+SYM_FUNC_END(sk_load_word_positive_offset)
 
-ENTRY(sk_load_half)
+SYM_FUNC_START(sk_load_half)
 	test	%esi,%esi
 	js	bpf_slow_path_half_neg
-ENDPROC(sk_load_half)
+SYM_FUNC_END(sk_load_half)
 
-ENTRY(sk_load_half_positive_offset)
+SYM_FUNC_START(sk_load_half_positive_offset)
 	mov	%r9d,%eax
 	sub	%esi,%eax		#	hlen - offset
 	cmp	$1,%eax
@@ -51,19 +51,19 @@ ENTRY(sk_load_half_positive_offset)
 	movzwl	(SKBDATA,%rsi),%eax
 	rol	$8,%ax			# ntohs()
 	ret
-ENDPROC(sk_load_half_positive_offset)
+SYM_FUNC_END(sk_load_half_positive_offset)
 
-ENTRY(sk_load_byte)
+SYM_FUNC_START(sk_load_byte)
 	test	%esi,%esi
 	js	bpf_slow_path_byte_neg
-ENDPROC(sk_load_byte)
+SYM_FUNC_END(sk_load_byte)
 
-ENTRY(sk_load_byte_positive_offset)
+SYM_FUNC_START(sk_load_byte_positive_offset)
 	cmp	%esi,%r9d   /* if (offset >= hlen) goto bpf_slow_path_byte */
 	jle	bpf_slow_path_byte
 	movzbl	(SKBDATA,%rsi),%eax
 	ret
-ENDPROC(sk_load_byte_positive_offset)
+SYM_FUNC_END(sk_load_byte_positive_offset)
 
 /* rsi contains offset and can be scratched */
 #define bpf_slow_path_common(LEN)		\
@@ -124,36 +124,36 @@ SYM_FUNC_START_LOCAL(bpf_slow_path_word_neg)
 	jl	bpf_error	/* offset lower -> error  */
 SYM_FUNC_END(bpf_slow_path_word_neg)
 
-ENTRY(sk_load_word_negative_offset)
+SYM_FUNC_START(sk_load_word_negative_offset)
 	sk_negative_common(4)
 	mov	(%rax), %eax
 	bswap	%eax
 	ret
-ENDPROC(sk_load_word_negative_offset)
+SYM_FUNC_END(sk_load_word_negative_offset)
 
 SYM_FUNC_START_LOCAL(bpf_slow_path_half_neg)
 	cmp	SKF_MAX_NEG_OFF, %esi
 	jl	bpf_error
 SYM_FUNC_END(bpf_slow_path_half_neg)
 
-ENTRY(sk_load_half_negative_offset)
+SYM_FUNC_START(sk_load_half_negative_offset)
 	sk_negative_common(2)
 	mov	(%rax),%ax
 	rol	$8,%ax
 	movzwl	%ax,%eax
 	ret
-ENDPROC(sk_load_half_negative_offset)
+SYM_FUNC_END(sk_load_half_negative_offset)
 
 SYM_FUNC_START_LOCAL(bpf_slow_path_byte_neg)
 	cmp	SKF_MAX_NEG_OFF, %esi
 	jl	bpf_error
 SYM_FUNC_END(bpf_slow_path_byte_neg)
 
-ENTRY(sk_load_byte_negative_offset)
+SYM_FUNC_START(sk_load_byte_negative_offset)
 	sk_negative_common(1)
 	movzbl	(%rax), %eax
 	ret
-ENDPROC(sk_load_byte_negative_offset)
+SYM_FUNC_END(sk_load_byte_negative_offset)
 
 SYM_FUNC_START_LOCAL(bpf_error)
 # force a return 0 from jit handler
diff --git a/arch/x86/platform/efi/efi_stub_64.S b/arch/x86/platform/efi/efi_stub_64.S
index cd95075944ab..90936c4e396e 100644
--- a/arch/x86/platform/efi/efi_stub_64.S
+++ b/arch/x86/platform/efi/efi_stub_64.S
@@ -38,7 +38,7 @@
 	mov %rsi, %cr0;			\
 	mov (%rsp), %rsp
 
-ENTRY(efi_call)
+SYM_FUNC_START(efi_call)
 	pushq %rbp
 	movq %rsp, %rbp
 	SAVE_XMM
@@ -54,4 +54,4 @@ ENTRY(efi_call)
 	RESTORE_XMM
 	popq %rbp
 	ret
-ENDPROC(efi_call)
+SYM_FUNC_END(efi_call)
diff --git a/arch/x86/platform/efi/efi_thunk_64.S b/arch/x86/platform/efi/efi_thunk_64.S
index d18697df1fe9..012601609d81 100644
--- a/arch/x86/platform/efi/efi_thunk_64.S
+++ b/arch/x86/platform/efi/efi_thunk_64.S
@@ -24,7 +24,7 @@
 
 	.text
 	.code64
-ENTRY(efi64_thunk)
+SYM_FUNC_START(efi64_thunk)
 	push	%rbp
 	push	%rbx
 
@@ -59,7 +59,7 @@ ENTRY(efi64_thunk)
 	pop	%rbx
 	pop	%rbp
 	retq
-ENDPROC(efi64_thunk)
+SYM_FUNC_END(efi64_thunk)
 
 /*
  * We run this function from the 1:1 mapping.
diff --git a/arch/x86/platform/olpc/xo1-wakeup.S b/arch/x86/platform/olpc/xo1-wakeup.S
index 2929091cf7fd..93ba74de2c55 100644
--- a/arch/x86/platform/olpc/xo1-wakeup.S
+++ b/arch/x86/platform/olpc/xo1-wakeup.S
@@ -89,7 +89,7 @@ restore_registers:
 
 	ret
 
-ENTRY(do_olpc_suspend_lowlevel)
+SYM_FUNC_START(do_olpc_suspend_lowlevel)
 	call	save_processor_state
 	call	save_registers
 
@@ -109,7 +109,7 @@ ret_point:
 	call	restore_registers
 	call	restore_processor_state
 	ret
-ENDPROC(do_olpc_suspend_lowlevel)
+SYM_FUNC_END(do_olpc_suspend_lowlevel)
 
 .data
 saved_gdt:             .long   0,0
diff --git a/arch/x86/power/hibernate_asm_64.S b/arch/x86/power/hibernate_asm_64.S
index ec6b26fd3a7e..e836ce085691 100644
--- a/arch/x86/power/hibernate_asm_64.S
+++ b/arch/x86/power/hibernate_asm_64.S
@@ -23,7 +23,7 @@
 #include <asm/processor-flags.h>
 #include <asm/frame.h>
 
-ENTRY(swsusp_arch_suspend)
+SYM_FUNC_START(swsusp_arch_suspend)
 	movq	$saved_context, %rax
 	movq	%rsp, pt_regs_sp(%rax)
 	movq	%rbp, pt_regs_bp(%rax)
@@ -51,9 +51,9 @@ ENTRY(swsusp_arch_suspend)
 	call swsusp_save
 	FRAME_END
 	ret
-ENDPROC(swsusp_arch_suspend)
+SYM_FUNC_END(swsusp_arch_suspend)
 
-ENTRY(restore_image)
+SYM_FUNC_START(restore_image)
 	/* prepare to jump to the image kernel */
 	movq	restore_jump_address(%rip), %r8
 	movq	restore_cr3(%rip), %r9
@@ -68,10 +68,10 @@ ENTRY(restore_image)
 	/* jump to relocated restore code */
 	movq	relocated_restore_code(%rip), %rcx
 	jmpq	*%rcx
-ENDPROC(restore_image)
+SYM_FUNC_END(restore_image)
 
 	/* code below has been relocated to a safe page */
-ENTRY(core_restore_code)
+SYM_FUNC_START(core_restore_code)
 	/* switch to temporary page tables */
 	movq	%rax, %cr3
 	/* flush TLB */
@@ -99,11 +99,11 @@ ENTRY(core_restore_code)
 .Ldone:
 	/* jump to the restore_registers address from the image header */
 	jmpq	*%r8
-ENDPROC(core_restore_code)
+SYM_FUNC_END(core_restore_code)
 
 	 /* code below belongs to the image kernel */
 	.align PAGE_SIZE
-ENTRY(restore_registers)
+SYM_FUNC_START(restore_registers)
 	/* go back to the original page tables */
 	movq    %r9, %cr3
 
@@ -145,4 +145,4 @@ ENTRY(restore_registers)
 	movq	%rax, in_suspend(%rip)
 
 	ret
-ENDPROC(restore_registers)
+SYM_FUNC_END(restore_registers)
diff --git a/arch/x86/realmode/rm/reboot.S b/arch/x86/realmode/rm/reboot.S
index 370ed1fe34e4..72224849f6c1 100644
--- a/arch/x86/realmode/rm/reboot.S
+++ b/arch/x86/realmode/rm/reboot.S
@@ -18,7 +18,7 @@
  */
 	.section ".text32", "ax"
 	.code32
-ENTRY(machine_real_restart_asm)
+SYM_FUNC_START(machine_real_restart_asm)
 
 #ifdef CONFIG_X86_64
 	/* Switch to trampoline GDT as it is guaranteed < 4 GiB */
@@ -62,7 +62,7 @@ SYM_FUNC_INNER_LABEL(machine_real_restart_paging_off, SYM_V_GLOBAL)
 	movl	%ecx, %gs
 	movl	%ecx, %ss
 	ljmpw	$8, $1f
-ENDPROC(machine_real_restart_asm)
+SYM_FUNC_END(machine_real_restart_asm)
 
 /*
  * This is 16-bit protected mode code to disable paging and the cache,
diff --git a/arch/x86/realmode/rm/trampoline_64.S b/arch/x86/realmode/rm/trampoline_64.S
index f1f2f18fff85..36c0c68709b7 100644
--- a/arch/x86/realmode/rm/trampoline_64.S
+++ b/arch/x86/realmode/rm/trampoline_64.S
@@ -36,7 +36,7 @@
 	.code16
 
 	.balign	PAGE_SIZE
-ENTRY(trampoline_start)
+SYM_FUNC_START(trampoline_start)
 	cli			# We should be safe anyway
 	wbinvd
 
@@ -79,14 +79,14 @@ ENTRY(trampoline_start)
 no_longmode:
 	hlt
 	jmp no_longmode
-ENDPROC(trampoline_start)
+SYM_FUNC_END(trampoline_start)
 
 #include "../kernel/verify_cpu.S"
 
 	.section ".text32","ax"
 	.code32
 	.balign 4
-ENTRY(startup_32)
+SYM_FUNC_START(startup_32)
 	movl	%edx, %ss
 	addl	$pa_real_mode_base, %esp
 	movl	%edx, %ds
@@ -118,15 +118,15 @@ ENTRY(startup_32)
 	 * the new gdt/idt that has __KERNEL_CS with CS.L = 1.
 	 */
 	ljmpl	$__KERNEL_CS, $pa_startup_64
-ENDPROC(startup_32)
+SYM_FUNC_END(startup_32)
 
 	.section ".text64","ax"
 	.code64
 	.balign 4
-ENTRY(startup_64)
+SYM_FUNC_START(startup_64)
 	# Now jump into the kernel using virtual addresses
 	jmpq	*tr_start(%rip)
-ENDPROC(startup_64)
+SYM_FUNC_END(startup_64)
 
 	.section ".rodata","a"
 	# Duplicate the global descriptor table
diff --git a/arch/x86/realmode/rm/wakeup_asm.S b/arch/x86/realmode/rm/wakeup_asm.S
index 41483fd2d247..8501b4d17dca 100644
--- a/arch/x86/realmode/rm/wakeup_asm.S
+++ b/arch/x86/realmode/rm/wakeup_asm.S
@@ -36,7 +36,7 @@ SYM_DATA_END(wakeup_header)
 	.code16
 
 	.balign	16
-ENTRY(wakeup_start)
+SYM_FUNC_START(wakeup_start)
 	cli
 	cld
 
@@ -134,7 +134,7 @@ ENTRY(wakeup_start)
 #else
 	jmp	trampoline_start
 #endif
-ENDPROC(wakeup_start)
+SYM_FUNC_END(wakeup_start)
 
 bogus_real_magic:
 1:
diff --git a/arch/x86/xen/xen-asm.S b/arch/x86/xen/xen-asm.S
index eff224df813f..cfa05b8259ab 100644
--- a/arch/x86/xen/xen-asm.S
+++ b/arch/x86/xen/xen-asm.S
@@ -23,7 +23,7 @@
  * event status with one and operation.  If there are pending events,
  * then enter the hypervisor to get them handled.
  */
-ENTRY(xen_irq_enable_direct)
+SYM_FUNC_START(xen_irq_enable_direct)
 	FRAME_BEGIN
 	/* Unmask events */
 	movb $0, PER_CPU_VAR(xen_vcpu_info) + XEN_vcpu_info_mask
@@ -43,7 +43,7 @@ ENTRY(xen_irq_enable_direct)
 ENDPATCH(xen_irq_enable_direct)
 	FRAME_END
 	ret
-	ENDPROC(xen_irq_enable_direct)
+	SYM_FUNC_END(xen_irq_enable_direct)
 	RELOC(xen_irq_enable_direct, 2b+1)
 
 
@@ -51,11 +51,11 @@ ENDPATCH(xen_irq_enable_direct)
  * Disabling events is simply a matter of making the event mask
  * non-zero.
  */
-ENTRY(xen_irq_disable_direct)
+SYM_FUNC_START(xen_irq_disable_direct)
 	movb $1, PER_CPU_VAR(xen_vcpu_info) + XEN_vcpu_info_mask
 ENDPATCH(xen_irq_disable_direct)
 	ret
-	ENDPROC(xen_irq_disable_direct)
+	SYM_FUNC_END(xen_irq_disable_direct)
 	RELOC(xen_irq_disable_direct, 0)
 
 /*
@@ -67,13 +67,13 @@ ENDPATCH(xen_irq_disable_direct)
  * undefined.  We need to toggle the state of the bit, because Xen and
  * x86 use opposite senses (mask vs enable).
  */
-ENTRY(xen_save_fl_direct)
+SYM_FUNC_START(xen_save_fl_direct)
 	testb $0xff, PER_CPU_VAR(xen_vcpu_info) + XEN_vcpu_info_mask
 	setz %ah
 	addb %ah, %ah
 ENDPATCH(xen_save_fl_direct)
 	ret
-	ENDPROC(xen_save_fl_direct)
+	SYM_FUNC_END(xen_save_fl_direct)
 	RELOC(xen_save_fl_direct, 0)
 
 
@@ -84,7 +84,7 @@ ENDPATCH(xen_save_fl_direct)
  * interrupt mask state, it checks for unmasked pending events and
  * enters the hypervisor to get them delivered if so.
  */
-ENTRY(xen_restore_fl_direct)
+SYM_FUNC_START(xen_restore_fl_direct)
 	FRAME_BEGIN
 #ifdef CONFIG_X86_64
 	testw $X86_EFLAGS_IF, %di
@@ -106,7 +106,7 @@ ENTRY(xen_restore_fl_direct)
 ENDPATCH(xen_restore_fl_direct)
 	FRAME_END
 	ret
-	ENDPROC(xen_restore_fl_direct)
+	SYM_FUNC_END(xen_restore_fl_direct)
 	RELOC(xen_restore_fl_direct, 2b+1)
 
 
@@ -114,7 +114,7 @@ ENDPATCH(xen_restore_fl_direct)
  * Force an event check by making a hypercall, but preserve regs
  * before making the call.
  */
-ENTRY(check_events)
+SYM_FUNC_START(check_events)
 	FRAME_BEGIN
 #ifdef CONFIG_X86_32
 	push %eax
@@ -147,4 +147,4 @@ ENTRY(check_events)
 #endif
 	FRAME_END
 	ret
-ENDPROC(check_events)
+SYM_FUNC_END(check_events)
diff --git a/arch/x86/xen/xen-asm_64.S b/arch/x86/xen/xen-asm_64.S
index e1174171ab57..a29c8a064eda 100644
--- a/arch/x86/xen/xen-asm_64.S
+++ b/arch/x86/xen/xen-asm_64.S
@@ -22,11 +22,11 @@
 
 #include "xen-asm.h"
 
-ENTRY(xen_adjust_exception_frame)
+SYM_FUNC_START(xen_adjust_exception_frame)
 	mov 8+0(%rsp), %rcx
 	mov 8+8(%rsp), %r11
 	ret $16
-ENDPROC(xen_adjust_exception_frame)
+SYM_FUNC_END(xen_adjust_exception_frame)
 
 hypercall_iret = hypercall_page + __HYPERVISOR_iret * 32
 /*
@@ -44,14 +44,14 @@ hypercall_iret = hypercall_page + __HYPERVISOR_iret * 32
  *	r11		}<-- pushed by hypercall page
  * rsp->rax		}
  */
-ENTRY(xen_iret)
+SYM_FUNC_START(xen_iret)
 	pushq $0
 1:	jmp hypercall_iret
 ENDPATCH(xen_iret)
 RELOC(xen_iret, 1b+1)
-ENDPROC(xen_iret)
+SYM_FUNC_END(xen_iret)
 
-ENTRY(xen_sysret64)
+SYM_FUNC_START(xen_sysret64)
 	/*
 	 * We're already on the usermode stack at this point, but
 	 * still with the kernel gs, so we can easily switch back
@@ -69,7 +69,7 @@ ENTRY(xen_sysret64)
 1:	jmp hypercall_iret
 ENDPATCH(xen_sysret64)
 RELOC(xen_sysret64, 1b+1)
-ENDPROC(xen_sysret64)
+SYM_FUNC_END(xen_sysret64)
 
 /*
  * Xen handles syscall callbacks much like ordinary exceptions, which
@@ -96,34 +96,34 @@ ENDPROC(xen_sysret64)
 .endm
 
 /* Normal 64-bit system call target */
-ENTRY(xen_syscall_target)
+SYM_FUNC_START(xen_syscall_target)
 	undo_xen_syscall
 	jmp entry_SYSCALL_64_after_swapgs
-ENDPROC(xen_syscall_target)
+SYM_FUNC_END(xen_syscall_target)
 
 #ifdef CONFIG_IA32_EMULATION
 
 /* 32-bit compat syscall target */
-ENTRY(xen_syscall32_target)
+SYM_FUNC_START(xen_syscall32_target)
 	undo_xen_syscall
 	jmp entry_SYSCALL_compat
-ENDPROC(xen_syscall32_target)
+SYM_FUNC_END(xen_syscall32_target)
 
 /* 32-bit compat sysenter target */
-ENTRY(xen_sysenter_target)
+SYM_FUNC_START(xen_sysenter_target)
 	undo_xen_syscall
 	jmp entry_SYSENTER_compat
-ENDPROC(xen_sysenter_target)
+SYM_FUNC_END(xen_sysenter_target)
 
 #else /* !CONFIG_IA32_EMULATION */
 
 SYM_FUNC_START_ALIAS(xen_syscall32_target)
-ENTRY(xen_sysenter_target)
+SYM_FUNC_START(xen_sysenter_target)
 	lea 16(%rsp), %rsp	/* strip %rcx, %r11 */
 	mov $-ENOSYS, %rax
 	pushq $0
 	jmp hypercall_iret
-ENDPROC(xen_sysenter_target)
+SYM_FUNC_END(xen_sysenter_target)
 SYM_FUNC_END_ALIAS(xen_syscall32_target)
 
 #endif	/* CONFIG_IA32_EMULATION */
diff --git a/arch/x86/xen/xen-head.S b/arch/x86/xen/xen-head.S
index 95eb4978791b..5c541f422055 100644
--- a/arch/x86/xen/xen-head.S
+++ b/arch/x86/xen/xen-head.S
@@ -18,7 +18,7 @@
 
 #ifdef CONFIG_XEN_PV
 	__INIT
-ENTRY(startup_xen)
+SYM_FUNC_START(startup_xen)
 	cld
 
 	/* Clear .bss */
@@ -33,15 +33,15 @@ ENTRY(startup_xen)
 	mov $init_thread_union+THREAD_SIZE, %_ASM_SP
 
 	jmp xen_start_kernel
-ENDPROC(startup_xen)
+SYM_FUNC_END(startup_xen)
 	__FINIT
 #endif
 
 .pushsection .text
 	.balign PAGE_SIZE
-ENTRY(hypercall_page)
+SYM_FUNC_START(hypercall_page)
 	.skip PAGE_SIZE
-ENDPROC(hypercall_page)
+SYM_FUNC_END(hypercall_page)
 
 #define HYPERCALL(n) \
 	.equ xen_hypercall_##n, hypercall_page + __HYPERVISOR_##n * 32; \
diff --git a/include/linux/linkage.h b/include/linux/linkage.h
index 27af3543fbc9..f51928ae175b 100644
--- a/include/linux/linkage.h
+++ b/include/linux/linkage.h
@@ -99,11 +99,13 @@
 
 /* === DEPRECATED annotations === */
 
+#ifndef CONFIG_X86_64
 #ifndef ENTRY
 /* deprecated, use SYM_FUNC_START */
 #define ENTRY(name) \
 	SYM_FUNC_START(name)
 #endif
+#endif /* CONFIG_X86_64 */
 #endif /* LINKER_SCRIPT */
 
 #ifndef WEAK
@@ -120,6 +122,7 @@
 #endif
 #endif /* CONFIG_X86 */
 
+#ifndef CONFIG_X86_64
 /* If symbol 'name' is treated as a subroutine (gets called, and returns)
  * then please use ENDPROC to mark 'name' as STT_FUNC for the benefit of
  * static analysis tools such as stack depth analyzer.
@@ -129,6 +132,7 @@
 #define ENDPROC(name) \
 	SYM_FUNC_END(name)
 #endif
+#endif /* CONFIG_X86_64 */
 
 /* === generic annotations === */
 
-- 
2.12.2

[PATCH v3 27/29] x86_32: assembly, change all ENTRY to SYM_FUNC_START

[Jiri Slaby]

These are all functions which are invoked from elsewhere, so we annotate
them as global using the new SYM_FUNC_START (and their ENDPROC's by
SYM_FUNC_END.)

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: x86@kernel.org
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Cc: Len Brown <len.brown@intel.com>
Cc: Pavel Machek <pavel@ucw.cz>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Bill Metzenthen <billm@melbpc.org.au>
Cc: Matt Fleming <matt@codeblueprint.co.uk>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: Juergen Gross <jgross@suse.com>
Cc: linux-crypto@vger.kernel.org
Cc: linux-pm@vger.kernel.org
Cc: lguest@lists.ozlabs.org
Cc: linux-efi@vger.kernel.org
Cc: xen-devel@lists.xenproject.org
---
 arch/x86/boot/compressed/efi_stub_32.S     |   4 +-
 arch/x86/boot/compressed/head_32.S         |  12 +--
 arch/x86/crypto/salsa20-i586-asm_32.S      |  12 +--
 arch/x86/crypto/serpent-sse2-i586-asm_32.S |   8 +-
 arch/x86/crypto/twofish-i586-asm_32.S      |   8 +-
 arch/x86/entry/entry_32.S                  | 132 ++++++++++++++---------------
 arch/x86/kernel/acpi/wakeup_32.S           |   8 +-
 arch/x86/kernel/ftrace_32.S                |  20 ++---
 arch/x86/kernel/head_32.S                  |  16 ++--
 arch/x86/lguest/head_32.S                  |  16 ++--
 arch/x86/lib/atomic64_386_32.S             |   4 +-
 arch/x86/lib/atomic64_cx8_32.S             |  32 +++----
 arch/x86/lib/checksum_32.S                 |   8 +-
 arch/x86/math-emu/div_Xsig.S               |   4 +-
 arch/x86/math-emu/div_small.S              |   4 +-
 arch/x86/math-emu/mul_Xsig.S               |  12 +--
 arch/x86/math-emu/polynom_Xsig.S           |   4 +-
 arch/x86/math-emu/reg_norm.S               |   8 +-
 arch/x86/math-emu/reg_round.S              |   4 +-
 arch/x86/math-emu/reg_u_add.S              |   4 +-
 arch/x86/math-emu/reg_u_div.S              |   4 +-
 arch/x86/math-emu/reg_u_mul.S              |   4 +-
 arch/x86/math-emu/reg_u_sub.S              |   4 +-
 arch/x86/math-emu/round_Xsig.S             |   8 +-
 arch/x86/math-emu/shr_Xsig.S               |   4 +-
 arch/x86/math-emu/wm_shrx.S                |   8 +-
 arch/x86/math-emu/wm_sqrt.S                |   4 +-
 arch/x86/platform/efi/efi_stub_32.S        |   4 +-
 arch/x86/power/hibernate_asm_32.S          |   8 +-
 arch/x86/realmode/rm/trampoline_32.S       |   8 +-
 arch/x86/xen/xen-asm_32.S                  |   8 +-
 drivers/lguest/x86/switcher_32.S           |   4 +-
 32 files changed, 194 insertions(+), 194 deletions(-)

diff --git a/arch/x86/boot/compressed/efi_stub_32.S b/arch/x86/boot/compressed/efi_stub_32.S
index a53440e81d52..4ceff75b0d2a 100644
--- a/arch/x86/boot/compressed/efi_stub_32.S
+++ b/arch/x86/boot/compressed/efi_stub_32.S
@@ -23,7 +23,7 @@
  */
 
 .text
-ENTRY(efi_call_phys)
+SYM_FUNC_START(efi_call_phys)
 	/*
 	 * 0. The function can only be called in Linux kernel. So CS has been
 	 * set to 0x0010, DS and SS have been set to 0x0018. In EFI, I found
@@ -76,7 +76,7 @@ ENTRY(efi_call_phys)
 	movl	saved_return_addr(%edx), %ecx
 	pushl	%ecx
 	ret
-ENDPROC(efi_call_phys)
+SYM_FUNC_END(efi_call_phys)
 .previous
 
 .data
diff --git a/arch/x86/boot/compressed/head_32.S b/arch/x86/boot/compressed/head_32.S
index d832ddb78ea2..86484c3788f8 100644
--- a/arch/x86/boot/compressed/head_32.S
+++ b/arch/x86/boot/compressed/head_32.S
@@ -60,7 +60,7 @@
 	.hidden _egot
 
 	__HEAD
-ENTRY(startup_32)
+SYM_FUNC_START(startup_32)
 	cld
 	/*
 	 * Test KEEP_SEGMENTS flag to see if the bootloader is asking
@@ -141,14 +141,14 @@ ENTRY(startup_32)
  */
 	leal	relocated(%ebx), %eax
 	jmp	*%eax
-ENDPROC(startup_32)
+SYM_FUNC_END(startup_32)
 
 #ifdef CONFIG_EFI_STUB
 /*
  * We don't need the return address, so set up the stack so efi_main() can find
  * its arguments.
  */
-ENTRY(efi_pe_entry)
+SYM_FUNC_START(efi_pe_entry)
 	add	$0x4, %esp
 
 	call	1f
@@ -173,9 +173,9 @@ ENTRY(efi_pe_entry)
 	pushl	%eax
 	pushl	%ecx
 	jmp	2f		/* Skip efi_config initialization */
-ENDPROC(efi_pe_entry)
+SYM_FUNC_END(efi_pe_entry)
 
-ENTRY(efi32_stub_entry)
+SYM_FUNC_START(efi32_stub_entry)
 	add	$0x4, %esp
 	popl	%ecx
 	popl	%edx
@@ -204,7 +204,7 @@ fail:
 	movl	BP_code32_start(%esi), %eax
 	leal	startup_32(%eax), %eax
 	jmp	*%eax
-ENDPROC(efi32_stub_entry)
+SYM_FUNC_END(efi32_stub_entry)
 #endif
 
 	.text
diff --git a/arch/x86/crypto/salsa20-i586-asm_32.S b/arch/x86/crypto/salsa20-i586-asm_32.S
index 329452b8f794..e9a6703056fc 100644
--- a/arch/x86/crypto/salsa20-i586-asm_32.S
+++ b/arch/x86/crypto/salsa20-i586-asm_32.S
@@ -7,7 +7,7 @@
 .text
 
 # enter salsa20_encrypt_bytes
-ENTRY(salsa20_encrypt_bytes)
+SYM_FUNC_START(salsa20_encrypt_bytes)
 	mov	%esp,%eax
 	and	$31,%eax
 	add	$256,%eax
@@ -934,10 +934,10 @@ ENTRY(salsa20_encrypt_bytes)
 	add	$64,%esi
 	# goto bytesatleast1
 	jmp	._bytesatleast1
-ENDPROC(salsa20_encrypt_bytes)
+SYM_FUNC_END(salsa20_encrypt_bytes)
 
 # enter salsa20_keysetup
-ENTRY(salsa20_keysetup)
+SYM_FUNC_START(salsa20_keysetup)
 	mov	%esp,%eax
 	and	$31,%eax
 	add	$256,%eax
@@ -1060,10 +1060,10 @@ ENTRY(salsa20_keysetup)
 	# leave
 	add	%eax,%esp
 	ret
-ENDPROC(salsa20_keysetup)
+SYM_FUNC_END(salsa20_keysetup)
 
 # enter salsa20_ivsetup
-ENTRY(salsa20_ivsetup)
+SYM_FUNC_START(salsa20_ivsetup)
 	mov	%esp,%eax
 	and	$31,%eax
 	add	$256,%eax
@@ -1111,4 +1111,4 @@ ENTRY(salsa20_ivsetup)
 	# leave
 	add	%eax,%esp
 	ret
-ENDPROC(salsa20_ivsetup)
+SYM_FUNC_END(salsa20_ivsetup)
diff --git a/arch/x86/crypto/serpent-sse2-i586-asm_32.S b/arch/x86/crypto/serpent-sse2-i586-asm_32.S
index d348f1553a79..f3cebd3c6739 100644
--- a/arch/x86/crypto/serpent-sse2-i586-asm_32.S
+++ b/arch/x86/crypto/serpent-sse2-i586-asm_32.S
@@ -512,7 +512,7 @@
 	pxor t0,		x3; \
 	movdqu x3,		(3*4*4)(out);
 
-ENTRY(__serpent_enc_blk_4way)
+SYM_FUNC_START(__serpent_enc_blk_4way)
 	/* input:
 	 *	arg_ctx(%esp): ctx, CTX
 	 *	arg_dst(%esp): dst
@@ -574,9 +574,9 @@ ENTRY(__serpent_enc_blk_4way)
 	xor_blocks(%eax, RA, RB, RC, RD, RT0, RT1, RE);
 
 	ret;
-ENDPROC(__serpent_enc_blk_4way)
+SYM_FUNC_END(__serpent_enc_blk_4way)
 
-ENTRY(serpent_dec_blk_4way)
+SYM_FUNC_START(serpent_dec_blk_4way)
 	/* input:
 	 *	arg_ctx(%esp): ctx, CTX
 	 *	arg_dst(%esp): dst
@@ -628,4 +628,4 @@ ENTRY(serpent_dec_blk_4way)
 	write_blocks(%eax, RC, RD, RB, RE, RT0, RT1, RA);
 
 	ret;
-ENDPROC(serpent_dec_blk_4way)
+SYM_FUNC_END(serpent_dec_blk_4way)
diff --git a/arch/x86/crypto/twofish-i586-asm_32.S b/arch/x86/crypto/twofish-i586-asm_32.S
index 694ea4587ba7..8ecb5234b2b3 100644
--- a/arch/x86/crypto/twofish-i586-asm_32.S
+++ b/arch/x86/crypto/twofish-i586-asm_32.S
@@ -220,7 +220,7 @@
 	xor	%esi,		d ## D;\
 	ror	$1,		d ## D;
 
-ENTRY(twofish_enc_blk)
+SYM_FUNC_START(twofish_enc_blk)
 	push	%ebp			/* save registers according to calling convention*/
 	push    %ebx
 	push    %esi
@@ -274,9 +274,9 @@ ENTRY(twofish_enc_blk)
 	pop	%ebp
 	mov	$1,	%eax
 	ret
-ENDPROC(twofish_enc_blk)
+SYM_FUNC_END(twofish_enc_blk)
 
-ENTRY(twofish_dec_blk)
+SYM_FUNC_START(twofish_dec_blk)
 	push	%ebp			/* save registers according to calling convention*/
 	push    %ebx
 	push    %esi
@@ -331,4 +331,4 @@ ENTRY(twofish_dec_blk)
 	pop	%ebp
 	mov	$1,	%eax
 	ret
-ENDPROC(twofish_dec_blk)
+SYM_FUNC_END(twofish_dec_blk)
diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S
index f1babb84510b..1a4c925c55e7 100644
--- a/arch/x86/entry/entry_32.S
+++ b/arch/x86/entry/entry_32.S
@@ -223,7 +223,7 @@
  * %eax: prev task
  * %edx: next task
  */
-ENTRY(__switch_to_asm)
+SYM_FUNC_START(__switch_to_asm)
 	/*
 	 * Save callee-saved registers
 	 * This must match the order in struct inactive_task_frame
@@ -249,7 +249,7 @@ ENTRY(__switch_to_asm)
 	popl	%ebp
 
 	jmp	__switch_to
-ENDPROC(__switch_to_asm)
+SYM_FUNC_END(__switch_to_asm)
 
 /*
  * A newly forked process directly context switches into this address.
@@ -258,7 +258,7 @@ ENDPROC(__switch_to_asm)
  * ebx: kernel thread func (NULL for user thread)
  * edi: kernel thread arg
  */
-ENTRY(ret_from_fork)
+SYM_FUNC_START(ret_from_fork)
 	FRAME_BEGIN		/* help unwinder find end of stack */
 
 	/*
@@ -289,7 +289,7 @@ ENTRY(ret_from_fork)
 	 */
 	movl	$0, PT_EAX(%esp)
 	jmp	2b
-ENDPROC(ret_from_fork)
+SYM_FUNC_END(ret_from_fork)
 
 /*
  * Return to user mode is not as complex as all this looks,
@@ -325,7 +325,7 @@ SYM_FUNC_INNER_LABEL(resume_userspace, SYM_V_LOCAL)
 SYM_FUNC_END(ret_from_exception)
 
 #ifdef CONFIG_PREEMPT
-ENTRY(resume_kernel)
+SYM_FUNC_START(resume_kernel)
 	DISABLE_INTERRUPTS(CLBR_ANY)
 .Lneed_resched:
 	cmpl	$0, PER_CPU_VAR(__preempt_count)
@@ -334,7 +334,7 @@ ENTRY(resume_kernel)
 	jz	restore_all
 	call	preempt_schedule_irq
 	jmp	.Lneed_resched
-ENDPROC(resume_kernel)
+SYM_FUNC_END(resume_kernel)
 #endif
 
 SYM_ENTRY(__begin_SYSENTER_singlestep_region, SYM_V_GLOBAL, SYM_A_NONE)
@@ -352,10 +352,10 @@ SYM_ENTRY(__begin_SYSENTER_singlestep_region, SYM_V_GLOBAL, SYM_A_NONE)
  * Xen doesn't set %esp to be precisely what the normal SYSENTER
  * entry point expects, so fix it up before using the normal path.
  */
-ENTRY(xen_sysenter_target)
+SYM_FUNC_START(xen_sysenter_target)
 	addl	$5*4, %esp			/* remove xen-provided frame */
 	jmp	.Lsysenter_past_esp
-ENDPROC(xen_sysenter_target)
+SYM_FUNC_END(xen_sysenter_target)
 #endif
 
 /*
@@ -390,7 +390,7 @@ ENDPROC(xen_sysenter_target)
  * ebp  user stack
  * 0(%ebp) arg6
  */
-ENTRY(entry_SYSENTER_32)
+SYM_FUNC_START(entry_SYSENTER_32)
 	movl	TSS_sysenter_sp0(%esp), %esp
 .Lsysenter_past_esp:
 	pushl	$__USER_DS		/* pt_regs->ss */
@@ -478,7 +478,7 @@ ENTRY(entry_SYSENTER_32)
 	popfl
 	jmp	.Lsysenter_flags_fixed
 SYM_ENTRY(__end_SYSENTER_singlestep_region, SYM_V_GLOBAL, SYM_A_NONE)
-ENDPROC(entry_SYSENTER_32)
+SYM_FUNC_END(entry_SYSENTER_32)
 
 /*
  * 32-bit legacy system call entry.
@@ -508,7 +508,7 @@ ENDPROC(entry_SYSENTER_32)
  * edi  arg5
  * ebp  arg6
  */
-ENTRY(entry_INT80_32)
+SYM_FUNC_START(entry_INT80_32)
 	ASM_CLAC
 	pushl	%eax			/* pt_regs->orig_ax */
 	SAVE_ALL pt_regs_ax=$-ENOSYS	/* save rest */
@@ -587,7 +587,7 @@ SYM_FUNC_END(iret_exc)
 	lss	(%esp), %esp			/* switch to espfix segment */
 	jmp	.Lrestore_nocheck
 #endif
-ENDPROC(entry_INT80_32)
+SYM_FUNC_END(entry_INT80_32)
 
 .macro FIXUP_ESPFIX_STACK
 /*
@@ -628,7 +628,7 @@ ENDPROC(entry_INT80_32)
  * We pack 1 stub into every 8-byte block.
  */
 	.align 8
-ENTRY(irq_entries_start)
+SYM_FUNC_START(irq_entries_start)
     vector=FIRST_EXTERNAL_VECTOR
     .rept (FIRST_SYSTEM_VECTOR - FIRST_EXTERNAL_VECTOR)
 	pushl	$(~vector+0x80)			/* Note: always in signed byte range */
@@ -636,7 +636,7 @@ ENTRY(irq_entries_start)
 	jmp	common_interrupt
 	.align	8
     .endr
-ENDPROC(irq_entries_start)
+SYM_FUNC_END(irq_entries_start)
 
 /*
  * the CPU automatically disables interrupts when executing an IRQ vector,
@@ -655,7 +655,7 @@ SYM_FUNC_START_LOCAL(common_interrupt)
 SYM_FUNC_END(common_interrupt)
 
 #define BUILD_INTERRUPT3(name, nr, fn)	\
-ENTRY(name)				\
+SYM_FUNC_START(name)				\
 	ASM_CLAC;			\
 	pushl	$~(nr);			\
 	SAVE_ALL;			\
@@ -664,7 +664,7 @@ ENTRY(name)				\
 	movl	%esp, %eax;		\
 	call	fn;			\
 	jmp	ret_from_intr;		\
-ENDPROC(name)
+SYM_FUNC_END(name)
 
 
 #ifdef CONFIG_TRACING
@@ -680,14 +680,14 @@ ENDPROC(name)
 /* The include is where all of the SMP etc. interrupts come from */
 #include <asm/entry_arch.h>
 
-ENTRY(coprocessor_error)
+SYM_FUNC_START(coprocessor_error)
 	ASM_CLAC
 	pushl	$0
 	pushl	$do_coprocessor_error
 	jmp	common_exception
-ENDPROC(coprocessor_error)
+SYM_FUNC_END(coprocessor_error)
 
-ENTRY(simd_coprocessor_error)
+SYM_FUNC_START(simd_coprocessor_error)
 	ASM_CLAC
 	pushl	$0
 #ifdef CONFIG_X86_INVD_BUG
@@ -699,99 +699,99 @@ ENTRY(simd_coprocessor_error)
 	pushl	$do_simd_coprocessor_error
 #endif
 	jmp	common_exception
-ENDPROC(simd_coprocessor_error)
+SYM_FUNC_END(simd_coprocessor_error)
 
-ENTRY(device_not_available)
+SYM_FUNC_START(device_not_available)
 	ASM_CLAC
 	pushl	$-1				# mark this as an int
 	pushl	$do_device_not_available
 	jmp	common_exception
-ENDPROC(device_not_available)
+SYM_FUNC_END(device_not_available)
 
 #ifdef CONFIG_PARAVIRT
-ENTRY(native_iret)
+SYM_FUNC_START(native_iret)
 	iret
 	_ASM_EXTABLE(native_iret, iret_exc)
-ENDPROC(native_iret)
+SYM_FUNC_END(native_iret)
 #endif
 
-ENTRY(overflow)
+SYM_FUNC_START(overflow)
 	ASM_CLAC
 	pushl	$0
 	pushl	$do_overflow
 	jmp	common_exception
-ENDPROC(overflow)
+SYM_FUNC_END(overflow)
 
-ENTRY(bounds)
+SYM_FUNC_START(bounds)
 	ASM_CLAC
 	pushl	$0
 	pushl	$do_bounds
 	jmp	common_exception
-ENDPROC(bounds)
+SYM_FUNC_END(bounds)
 
-ENTRY(invalid_op)
+SYM_FUNC_START(invalid_op)
 	ASM_CLAC
 	pushl	$0
 	pushl	$do_invalid_op
 	jmp	common_exception
-ENDPROC(invalid_op)
+SYM_FUNC_END(invalid_op)
 
-ENTRY(coprocessor_segment_overrun)
+SYM_FUNC_START(coprocessor_segment_overrun)
 	ASM_CLAC
 	pushl	$0
 	pushl	$do_coprocessor_segment_overrun
 	jmp	common_exception
-ENDPROC(coprocessor_segment_overrun)
+SYM_FUNC_END(coprocessor_segment_overrun)
 
-ENTRY(invalid_TSS)
+SYM_FUNC_START(invalid_TSS)
 	ASM_CLAC
 	pushl	$do_invalid_TSS
 	jmp	common_exception
-ENDPROC(invalid_TSS)
+SYM_FUNC_END(invalid_TSS)
 
-ENTRY(segment_not_present)
+SYM_FUNC_START(segment_not_present)
 	ASM_CLAC
 	pushl	$do_segment_not_present
 	jmp	common_exception
-ENDPROC(segment_not_present)
+SYM_FUNC_END(segment_not_present)
 
-ENTRY(stack_segment)
+SYM_FUNC_START(stack_segment)
 	ASM_CLAC
 	pushl	$do_stack_segment
 	jmp	common_exception
-ENDPROC(stack_segment)
+SYM_FUNC_END(stack_segment)
 
-ENTRY(alignment_check)
+SYM_FUNC_START(alignment_check)
 	ASM_CLAC
 	pushl	$do_alignment_check
 	jmp	common_exception
-ENDPROC(alignment_check)
+SYM_FUNC_END(alignment_check)
 
-ENTRY(divide_error)
+SYM_FUNC_START(divide_error)
 	ASM_CLAC
 	pushl	$0				# no error code
 	pushl	$do_divide_error
 	jmp	common_exception
-ENDPROC(divide_error)
+SYM_FUNC_END(divide_error)
 
 #ifdef CONFIG_X86_MCE
-ENTRY(machine_check)
+SYM_FUNC_START(machine_check)
 	ASM_CLAC
 	pushl	$0
 	pushl	machine_check_vector
 	jmp	common_exception
-ENDPROC(machine_check)
+SYM_FUNC_END(machine_check)
 #endif
 
-ENTRY(spurious_interrupt_bug)
+SYM_FUNC_START(spurious_interrupt_bug)
 	ASM_CLAC
 	pushl	$0
 	pushl	$do_spurious_interrupt_bug
 	jmp	common_exception
-ENDPROC(spurious_interrupt_bug)
+SYM_FUNC_END(spurious_interrupt_bug)
 
 #ifdef CONFIG_XEN
-ENTRY(xen_hypervisor_callback)
+SYM_FUNC_START(xen_hypervisor_callback)
 	pushl	$-1				/* orig_ax = -1 => not a system call */
 	SAVE_ALL
 	ENCODE_FRAME_POINTER
@@ -819,7 +819,7 @@ SYM_FUNC_INNER_LABEL(xen_do_upcall, SYM_V_GLOBAL)
 	call	xen_maybe_preempt_hcall
 #endif
 	jmp	ret_from_intr
-ENDPROC(xen_hypervisor_callback)
+SYM_FUNC_END(xen_hypervisor_callback)
 
 /*
  * Hypervisor uses this for application faults while it executes.
@@ -833,7 +833,7 @@ ENDPROC(xen_hypervisor_callback)
  * to pop the stack frame we end up in an infinite loop of failsafe callbacks.
  * We distinguish between categories by maintaining a status value in EAX.
  */
-ENTRY(xen_failsafe_callback)
+SYM_FUNC_START(xen_failsafe_callback)
 	pushl	%eax
 	movl	$1, %eax
 1:	mov	4(%esp), %ds
@@ -870,7 +870,7 @@ ENTRY(xen_failsafe_callback)
 	_ASM_EXTABLE(2b, 7b)
 	_ASM_EXTABLE(3b, 8b)
 	_ASM_EXTABLE(4b, 9b)
-ENDPROC(xen_failsafe_callback)
+SYM_FUNC_END(xen_failsafe_callback)
 
 BUILD_INTERRUPT3(xen_hvm_callback_vector, HYPERVISOR_CALLBACK_VECTOR,
 		xen_evtchn_do_upcall)
@@ -885,19 +885,19 @@ BUILD_INTERRUPT3(hyperv_callback_vector, HYPERVISOR_CALLBACK_VECTOR,
 #endif /* CONFIG_HYPERV */
 
 #ifdef CONFIG_TRACING
-ENTRY(trace_page_fault)
+SYM_FUNC_START(trace_page_fault)
 	ASM_CLAC
 	pushl	$trace_do_page_fault
 	jmp	common_exception
-ENDPROC(trace_page_fault)
+SYM_FUNC_END(trace_page_fault)
 #endif
 
-ENTRY(page_fault)
+SYM_FUNC_START(page_fault)
 	ASM_CLAC
 	pushl	$do_page_fault
 	ALIGN
 	jmp common_exception
-ENDPROC(page_fault)
+SYM_FUNC_END(page_fault)
 
 SYM_FUNC_START_LOCAL(common_exception)
 	/* the function address is in %gs's slot on the stack */
@@ -931,7 +931,7 @@ SYM_FUNC_START_LOCAL(common_exception)
 	jmp	ret_from_exception
 SYM_FUNC_END(common_exception)
 
-ENTRY(debug)
+SYM_FUNC_START(debug)
 	/*
 	 * #DB can happen at the first instruction of
 	 * entry_SYSENTER_32 or in Xen's SYSENTER prologue.  If this
@@ -966,7 +966,7 @@ ENTRY(debug)
 	call	do_debug
 	movl	%ebx, %esp
 	jmp	ret_from_exception
-ENDPROC(debug)
+SYM_FUNC_END(debug)
 
 /*
  * NMI is doubly nasty.  It can happen on the first instruction of
@@ -975,7 +975,7 @@ ENDPROC(debug)
  * switched stacks.  We handle both conditions by simply checking whether we
  * interrupted kernel code running on the SYSENTER stack.
  */
-ENTRY(nmi)
+SYM_FUNC_START(nmi)
 	ASM_CLAC
 #ifdef CONFIG_X86_ESPFIX32
 	pushl	%eax
@@ -1034,9 +1034,9 @@ ENTRY(nmi)
 	lss	12+4(%esp), %esp		# back to espfix stack
 	jmp	.Lirq_return
 #endif
-ENDPROC(nmi)
+SYM_FUNC_END(nmi)
 
-ENTRY(int3)
+SYM_FUNC_START(int3)
 	ASM_CLAC
 	pushl	$-1				# mark this as an int
 	SAVE_ALL
@@ -1046,22 +1046,22 @@ ENTRY(int3)
 	movl	%esp, %eax			# pt_regs pointer
 	call	do_int3
 	jmp	ret_from_exception
-ENDPROC(int3)
+SYM_FUNC_END(int3)
 
-ENTRY(general_protection)
+SYM_FUNC_START(general_protection)
 	pushl	$do_general_protection
 	jmp	common_exception
-ENDPROC(general_protection)
+SYM_FUNC_END(general_protection)
 
 #ifdef CONFIG_KVM_GUEST
-ENTRY(async_page_fault)
+SYM_FUNC_START(async_page_fault)
 	ASM_CLAC
 	pushl	$do_async_page_fault
 	jmp	common_exception
-ENDPROC(async_page_fault)
+SYM_FUNC_END(async_page_fault)
 #endif
 
-ENTRY(rewind_stack_do_exit)
+SYM_FUNC_START(rewind_stack_do_exit)
 	/* Prevent any naive code from trying to unwind to our caller. */
 	xorl	%ebp, %ebp
 
@@ -1070,4 +1070,4 @@ ENTRY(rewind_stack_do_exit)
 
 	call	do_exit
 1:	jmp 1b
-ENDPROC(rewind_stack_do_exit)
+SYM_FUNC_END(rewind_stack_do_exit)
diff --git a/arch/x86/kernel/acpi/wakeup_32.S b/arch/x86/kernel/acpi/wakeup_32.S
index 11fb24d15f3d..15a61607ccb8 100644
--- a/arch/x86/kernel/acpi/wakeup_32.S
+++ b/arch/x86/kernel/acpi/wakeup_32.S
@@ -8,7 +8,7 @@
 	.code32
 	ALIGN
 
-ENTRY(wakeup_pmode_return)
+SYM_FUNC_START(wakeup_pmode_return)
 wakeup_pmode_return:
 	movw	$__KERNEL_DS, %ax
 	movw	%ax, %ss
@@ -38,7 +38,7 @@ wakeup_pmode_return:
 	# jump to place where we left off
 	movl	saved_eip, %eax
 	jmp	*%eax
-ENDPROC(wakeup_pmode_return)
+SYM_FUNC_END(wakeup_pmode_return)
 
 bogus_magic:
 	jmp	bogus_magic
@@ -72,7 +72,7 @@ restore_registers:
 	popfl
 	ret
 
-ENTRY(do_suspend_lowlevel)
+SYM_FUNC_START(do_suspend_lowlevel)
 	call	save_processor_state
 	call	save_registers
 	pushl	$3
@@ -87,7 +87,7 @@ ret_point:
 	call	restore_registers
 	call	restore_processor_state
 	ret
-ENDPROC(do_suspend_lowlevel)
+SYM_FUNC_END(do_suspend_lowlevel)
 
 .data
 ALIGN
diff --git a/arch/x86/kernel/ftrace_32.S b/arch/x86/kernel/ftrace_32.S
index 8ad07ddfa1c9..9c3db18b2a19 100644
--- a/arch/x86/kernel/ftrace_32.S
+++ b/arch/x86/kernel/ftrace_32.S
@@ -29,11 +29,11 @@ EXPORT_SYMBOL(mcount)
 # define MCOUNT_FRAME			0	/* using frame = false */
 #endif
 
-ENTRY(function_hook)
+SYM_FUNC_START(function_hook)
 	ret
-ENDPROC(function_hook)
+SYM_FUNC_END(function_hook)
 
-ENTRY(ftrace_caller)
+SYM_FUNC_START(ftrace_caller)
 
 #ifdef USING_FRAME_POINTER
 # ifdef CC_USING_FENTRY
@@ -98,9 +98,9 @@ ftrace_graph_call:
 /* This is weak to keep gas from relaxing the jumps */
 WEAK(ftrace_stub)
 	ret
-ENDPROC(ftrace_caller)
+SYM_FUNC_END(ftrace_caller)
 
-ENTRY(ftrace_regs_caller)
+SYM_FUNC_START(ftrace_regs_caller)
 	/*
 	 * i386 does not save SS and ESP when coming from kernel.
 	 * Instead, to get sp, &regs->sp is used (see ptrace.h).
@@ -168,10 +168,10 @@ SYM_FUNC_INNER_LABEL(ftrace_regs_call, SYM_V_GLOBAL)
 	lea	3*4(%esp), %esp			/* Skip orig_ax, ip and cs */
 
 	jmp	.Lftrace_ret
-ENDPROC(ftrace_regs_caller)
+SYM_FUNC_END(ftrace_regs_caller)
 #else /* ! CONFIG_DYNAMIC_FTRACE */
 
-ENTRY(function_hook)
+SYM_FUNC_START(function_hook)
 	cmpl	$__PAGE_OFFSET, %esp
 	jb	ftrace_stub			/* Paging not enabled yet? */
 
@@ -203,11 +203,11 @@ ftrace_stub:
 	popl	%ecx
 	popl	%eax
 	jmp	ftrace_stub
-ENDPROC(function_hook)
+SYM_FUNC_END(function_hook)
 #endif /* CONFIG_DYNAMIC_FTRACE */
 
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
-ENTRY(ftrace_graph_caller)
+SYM_FUNC_START(ftrace_graph_caller)
 	pushl	%eax
 	pushl	%ecx
 	pushl	%edx
@@ -226,7 +226,7 @@ ENTRY(ftrace_graph_caller)
 	popl	%ecx
 	popl	%eax
 	ret
-ENDPROC(ftrace_graph_caller)
+SYM_FUNC_END(ftrace_graph_caller)
 
 .globl return_to_handler
 return_to_handler:
diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S
index 47a212e3beee..74d18d2fef5a 100644
--- a/arch/x86/kernel/head_32.S
+++ b/arch/x86/kernel/head_32.S
@@ -63,7 +63,7 @@ RESERVE_BRK(pagetables, INIT_MAP_SIZE)
  * can.
  */
 __HEAD
-ENTRY(startup_32)
+SYM_FUNC_START(startup_32)
 	movl pa(initial_stack),%ecx
 	
 	/* test KEEP_SEGMENTS flag to see if the bootloader is asking
@@ -173,7 +173,7 @@ num_subarch_entries = (. - subarch_entries) / 4
 #else
 	jmp .Ldefault_entry
 #endif /* CONFIG_PARAVIRT */
-ENDPROC(startup_32)
+SYM_FUNC_END(startup_32)
 
 #ifdef CONFIG_HOTPLUG_CPU
 /*
@@ -181,12 +181,12 @@ ENDPROC(startup_32)
  * up already except stack. We just set up stack here. Then call
  * start_secondary().
  */
-ENTRY(start_cpu0)
+SYM_FUNC_START(start_cpu0)
 	movl initial_stack, %ecx
 	movl %ecx, %esp
 	call *(initial_code)
 1:	jmp 1b
-ENDPROC(start_cpu0)
+SYM_FUNC_END(start_cpu0)
 #endif
 
 /*
@@ -197,7 +197,7 @@ ENDPROC(start_cpu0)
  * If cpu hotplug is not supported then this code can go in init section
  * which will be freed later
  */
-ENTRY(startup_32_smp)
+SYM_FUNC_START(startup_32_smp)
 	cld
 	movl $(__BOOT_DS),%eax
 	movl %eax,%ds
@@ -368,7 +368,7 @@ ENTRY(startup_32_smp)
 
 	call *(initial_code)
 1:	jmp 1b
-ENDPROC(startup_32_smp)
+SYM_FUNC_END(startup_32_smp)
 
 #include "verify_cpu.S"
 
@@ -429,7 +429,7 @@ setup_once:
 	andl $0,setup_once_ref	/* Once is enough, thanks */
 	ret
 
-ENTRY(early_idt_handler_array)
+SYM_FUNC_START(early_idt_handler_array)
 	# 36(%esp) %eflags
 	# 32(%esp) %cs
 	# 28(%esp) %eip
@@ -444,7 +444,7 @@ ENTRY(early_idt_handler_array)
 	i = i + 1
 	.fill early_idt_handler_array + i*EARLY_IDT_HANDLER_SIZE - ., 1, 0xcc
 	.endr
-ENDPROC(early_idt_handler_array)
+SYM_FUNC_END(early_idt_handler_array)
 	
 SYM_FUNC_START_LOCAL(early_idt_handler_common)
 	/*
diff --git a/arch/x86/lguest/head_32.S b/arch/x86/lguest/head_32.S
index b11a989de062..99069ec13511 100644
--- a/arch/x86/lguest/head_32.S
+++ b/arch/x86/lguest/head_32.S
@@ -29,7 +29,7 @@
  * boot.
  */
 .section .init.text, "ax", @progbits
-ENTRY(lguest_entry)
+SYM_FUNC_START(lguest_entry)
 	/*
 	 * We make the "initialization" hypercall now to tell the Host where
 	 * our lguest_data struct is.
@@ -48,7 +48,7 @@ ENTRY(lguest_entry)
 
 	/* Jumps are relative: we're running __PAGE_OFFSET too low. */
 	jmp lguest_init+__PAGE_OFFSET
-ENDPROC(lguest_entry)
+SYM_FUNC_END(lguest_entry)
 
 /*G:055
  * We create a macro which puts the assembler code between lgstart_ and lgend_
@@ -73,7 +73,7 @@ LGUEST_PATCH(pushf, movl lguest_data+LGUEST_DATA_irq_enabled, %eax)
  * in a bit more detail so I'll describe in easy stages.  First, the routine to
  * enable interrupts:
  */
-ENTRY(lg_irq_enable)
+SYM_FUNC_START(lg_irq_enable)
 	/*
 	 * The reverse of irq_disable, this sets lguest_data.irq_enabled to
 	 * X86_EFLAGS_IF (ie. "Interrupts enabled").
@@ -111,14 +111,14 @@ send_interrupts:
 	/* Put eax back the way we found it. */
 	popl %eax
 	ret
-ENDPROC(lg_irq_enable)
+SYM_FUNC_END(lg_irq_enable)
 
 /*
  * Finally, the "popf" or "restore flags" routine.  The %eax register holds the
  * flags (in practice, either X86_EFLAGS_IF or 0): if it's X86_EFLAGS_IF we're
  * enabling interrupts again, if it's 0 we're leaving them off.
  */
-ENTRY(lg_restore_fl)
+SYM_FUNC_START(lg_restore_fl)
 	/* This is just "lguest_data.irq_enabled = flags;" */
 	movl %eax, lguest_data+LGUEST_DATA_irq_enabled
 	/*
@@ -133,7 +133,7 @@ ENTRY(lg_restore_fl)
 	jnz send_interrupts
 	/* Again, the normal path has used no extra registers.  Clever, huh? */
 	ret
-ENDPROC(lg_restore_fl)
+SYM_FUNC_END(lg_restore_fl)
 /*:*/
 
 /* These demark the EIP where host should never deliver interrupts. */
@@ -182,7 +182,7 @@ ENDPROC(lg_restore_fl)
  * data gets updated only after it completes, so we only need to protect
  * one instruction, iret).
  */
-ENTRY(lguest_iret)
+SYM_FUNC_START(lguest_iret)
 	pushl	2*4(%esp)
 	/*
 	 * Note the %ss: segment prefix here.  Normal data accesses use the
@@ -193,4 +193,4 @@ ENTRY(lguest_iret)
 	popl	%ss:lguest_data+LGUEST_DATA_irq_enabled
 lguest_noirq_iret:
 	iret
-ENDPROC(lguest_iret)
+SYM_FUNC_END(lguest_iret)
diff --git a/arch/x86/lib/atomic64_386_32.S b/arch/x86/lib/atomic64_386_32.S
index 9b0ca8fe80fc..9ed71edd9dfe 100644
--- a/arch/x86/lib/atomic64_386_32.S
+++ b/arch/x86/lib/atomic64_386_32.S
@@ -24,10 +24,10 @@
 
 #define BEGIN(op) \
 .macro endp; \
-ENDPROC(atomic64_##op##_386); \
+SYM_FUNC_END(atomic64_##op##_386); \
 .purgem endp; \
 .endm; \
-ENTRY(atomic64_##op##_386); \
+SYM_FUNC_START(atomic64_##op##_386); \
 	LOCK v;
 
 #define ENDP endp
diff --git a/arch/x86/lib/atomic64_cx8_32.S b/arch/x86/lib/atomic64_cx8_32.S
index db3ae85440ff..f02f70890121 100644
--- a/arch/x86/lib/atomic64_cx8_32.S
+++ b/arch/x86/lib/atomic64_cx8_32.S
@@ -20,12 +20,12 @@
 	cmpxchg8b (\reg)
 .endm
 
-ENTRY(atomic64_read_cx8)
+SYM_FUNC_START(atomic64_read_cx8)
 	read64 %ecx
 	ret
-ENDPROC(atomic64_read_cx8)
+SYM_FUNC_END(atomic64_read_cx8)
 
-ENTRY(atomic64_set_cx8)
+SYM_FUNC_START(atomic64_set_cx8)
 1:
 /* we don't need LOCK_PREFIX since aligned 64-bit writes
  * are atomic on 586 and newer */
@@ -33,19 +33,19 @@ ENTRY(atomic64_set_cx8)
 	jne 1b
 
 	ret
-ENDPROC(atomic64_set_cx8)
+SYM_FUNC_END(atomic64_set_cx8)
 
-ENTRY(atomic64_xchg_cx8)
+SYM_FUNC_START(atomic64_xchg_cx8)
 1:
 	LOCK_PREFIX
 	cmpxchg8b (%esi)
 	jne 1b
 
 	ret
-ENDPROC(atomic64_xchg_cx8)
+SYM_FUNC_END(atomic64_xchg_cx8)
 
 .macro addsub_return func ins insc
-ENTRY(atomic64_\func\()_return_cx8)
+SYM_FUNC_START(atomic64_\func\()_return_cx8)
 	pushl %ebp
 	pushl %ebx
 	pushl %esi
@@ -73,14 +73,14 @@ ENTRY(atomic64_\func\()_return_cx8)
 	popl %ebx
 	popl %ebp
 	ret
-ENDPROC(atomic64_\func\()_return_cx8)
+SYM_FUNC_END(atomic64_\func\()_return_cx8)
 .endm
 
 addsub_return add add adc
 addsub_return sub sub sbb
 
 .macro incdec_return func ins insc
-ENTRY(atomic64_\func\()_return_cx8)
+SYM_FUNC_START(atomic64_\func\()_return_cx8)
 	pushl %ebx
 
 	read64 %esi
@@ -98,13 +98,13 @@ ENTRY(atomic64_\func\()_return_cx8)
 	movl %ecx, %edx
 	popl %ebx
 	ret
-ENDPROC(atomic64_\func\()_return_cx8)
+SYM_FUNC_END(atomic64_\func\()_return_cx8)
 .endm
 
 incdec_return inc add adc
 incdec_return dec sub sbb
 
-ENTRY(atomic64_dec_if_positive_cx8)
+SYM_FUNC_START(atomic64_dec_if_positive_cx8)
 	pushl %ebx
 
 	read64 %esi
@@ -123,9 +123,9 @@ ENTRY(atomic64_dec_if_positive_cx8)
 	movl %ecx, %edx
 	popl %ebx
 	ret
-ENDPROC(atomic64_dec_if_positive_cx8)
+SYM_FUNC_END(atomic64_dec_if_positive_cx8)
 
-ENTRY(atomic64_add_unless_cx8)
+SYM_FUNC_START(atomic64_add_unless_cx8)
 	pushl %ebp
 	pushl %ebx
 /* these just push these two parameters on the stack */
@@ -159,9 +159,9 @@ ENTRY(atomic64_add_unless_cx8)
 	jne 2b
 	xorl %eax, %eax
 	jmp 3b
-ENDPROC(atomic64_add_unless_cx8)
+SYM_FUNC_END(atomic64_add_unless_cx8)
 
-ENTRY(atomic64_inc_not_zero_cx8)
+SYM_FUNC_START(atomic64_inc_not_zero_cx8)
 	pushl %ebx
 
 	read64 %esi
@@ -181,4 +181,4 @@ ENTRY(atomic64_inc_not_zero_cx8)
 3:
 	popl %ebx
 	ret
-ENDPROC(atomic64_inc_not_zero_cx8)
+SYM_FUNC_END(atomic64_inc_not_zero_cx8)
diff --git a/arch/x86/lib/checksum_32.S b/arch/x86/lib/checksum_32.S
index a048436ce3ac..78523e028d88 100644
--- a/arch/x86/lib/checksum_32.S
+++ b/arch/x86/lib/checksum_32.S
@@ -49,7 +49,7 @@ unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum)
 	   * Fortunately, it is easy to convert 2-byte alignment to 4-byte
 	   * alignment for the unrolled loop.
 	   */		
-ENTRY(csum_partial)
+SYM_FUNC_START(csum_partial)
 	pushl %esi
 	pushl %ebx
 	movl 20(%esp),%eax	# Function arg: unsigned int sum
@@ -131,13 +131,13 @@ ENTRY(csum_partial)
 	popl %ebx
 	popl %esi
 	ret
-ENDPROC(csum_partial)
+SYM_FUNC_END(csum_partial)
 
 #else
 
 /* Version for PentiumII/PPro */
 
-ENTRY(csum_partial)
+SYM_FUNC_START(csum_partial)
 	pushl %esi
 	pushl %ebx
 	movl 20(%esp),%eax	# Function arg: unsigned int sum
@@ -249,7 +249,7 @@ ENTRY(csum_partial)
 	popl %ebx
 	popl %esi
 	ret
-ENDPROC(csum_partial)
+SYM_FUNC_END(csum_partial)
 				
 #endif
 EXPORT_SYMBOL(csum_partial)
diff --git a/arch/x86/math-emu/div_Xsig.S b/arch/x86/math-emu/div_Xsig.S
index 066996dba6a2..8f2fadf8aed7 100644
--- a/arch/x86/math-emu/div_Xsig.S
+++ b/arch/x86/math-emu/div_Xsig.S
@@ -74,7 +74,7 @@ FPU_result_1:
 
 
 .text
-ENTRY(div_Xsig)
+SYM_FUNC_START(div_Xsig)
 	pushl	%ebp
 	movl	%esp,%ebp
 #ifndef NON_REENTRANT_FPU
@@ -363,4 +363,4 @@ L_bugged_2:
 	pop	%ebx
 	jmp	L_exit
 #endif /* PARANOID */ 
-ENDPROC(div_Xsig)
+SYM_FUNC_END(div_Xsig)
diff --git a/arch/x86/math-emu/div_small.S b/arch/x86/math-emu/div_small.S
index 2c71527bd917..15241e5a7e50 100644
--- a/arch/x86/math-emu/div_small.S
+++ b/arch/x86/math-emu/div_small.S
@@ -18,7 +18,7 @@
 #include "fpu_emu.h"
 
 .text
-ENTRY(FPU_div_small)
+SYM_FUNC_START(FPU_div_small)
 	pushl	%ebp
 	movl	%esp,%ebp
 
@@ -44,4 +44,4 @@ ENTRY(FPU_div_small)
 
 	leave
 	ret
-ENDPROC(FPU_div_small)
+SYM_FUNC_END(FPU_div_small)
diff --git a/arch/x86/math-emu/mul_Xsig.S b/arch/x86/math-emu/mul_Xsig.S
index 22e0631bb85a..eb59220280c3 100644
--- a/arch/x86/math-emu/mul_Xsig.S
+++ b/arch/x86/math-emu/mul_Xsig.S
@@ -24,7 +24,7 @@
 #include "fpu_emu.h"
 
 .text
-ENTRY(mul32_Xsig)
+SYM_FUNC_START(mul32_Xsig)
 	pushl %ebp
 	movl %esp,%ebp
 	subl $16,%esp
@@ -62,10 +62,10 @@ ENTRY(mul32_Xsig)
 	popl %esi
 	leave
 	ret
-ENDPROC(mul32_Xsig)
+SYM_FUNC_END(mul32_Xsig)
 
 
-ENTRY(mul64_Xsig)
+SYM_FUNC_START(mul64_Xsig)
 	pushl %ebp
 	movl %esp,%ebp
 	subl $16,%esp
@@ -115,11 +115,11 @@ ENTRY(mul64_Xsig)
 	popl %esi
 	leave
 	ret
-ENDPROC(mul64_Xsig)
+SYM_FUNC_END(mul64_Xsig)
 
 
 
-ENTRY(mul_Xsig_Xsig)
+SYM_FUNC_START(mul_Xsig_Xsig)
 	pushl %ebp
 	movl %esp,%ebp
 	subl $16,%esp
@@ -175,4 +175,4 @@ ENTRY(mul_Xsig_Xsig)
 	popl %esi
 	leave
 	ret
-ENDPROC(mul_Xsig_Xsig)
+SYM_FUNC_END(mul_Xsig_Xsig)
diff --git a/arch/x86/math-emu/polynom_Xsig.S b/arch/x86/math-emu/polynom_Xsig.S
index a9aaf414135d..fe98ab9acfa6 100644
--- a/arch/x86/math-emu/polynom_Xsig.S
+++ b/arch/x86/math-emu/polynom_Xsig.S
@@ -36,7 +36,7 @@
 #define OVERFLOWED      -16(%ebp)	/* addition overflow flag */
 
 .text
-ENTRY(polynomial_Xsig)
+SYM_FUNC_START(polynomial_Xsig)
 	pushl	%ebp
 	movl	%esp,%ebp
 	subl	$32,%esp
@@ -133,4 +133,4 @@ L_accum_done:
 	popl	%esi
 	leave
 	ret
-ENDPROC(polynomial_Xsig)
+SYM_FUNC_END(polynomial_Xsig)
diff --git a/arch/x86/math-emu/reg_norm.S b/arch/x86/math-emu/reg_norm.S
index 53ac1a343c69..4d71b607f007 100644
--- a/arch/x86/math-emu/reg_norm.S
+++ b/arch/x86/math-emu/reg_norm.S
@@ -21,7 +21,7 @@
 
 
 .text
-ENTRY(FPU_normalize)
+SYM_FUNC_START(FPU_normalize)
 	pushl	%ebp
 	movl	%esp,%ebp
 	pushl	%ebx
@@ -94,12 +94,12 @@ L_overflow:
 	call	arith_overflow
 	pop	%ebx
 	jmp	L_exit
-ENDPROC(FPU_normalize)
+SYM_FUNC_END(FPU_normalize)
 
 
 
 /* Normalise without reporting underflow or overflow */
-ENTRY(FPU_normalize_nuo)
+SYM_FUNC_START(FPU_normalize_nuo)
 	pushl	%ebp
 	movl	%esp,%ebp
 	pushl	%ebx
@@ -146,4 +146,4 @@ L_exit_nuo_zero:
 	popl	%ebx
 	leave
 	ret
-ENDPROC(FPU_normalize_nuo)
+SYM_FUNC_END(FPU_normalize_nuo)
diff --git a/arch/x86/math-emu/reg_round.S b/arch/x86/math-emu/reg_round.S
index 41af5b208d88..4ef24a419bc7 100644
--- a/arch/x86/math-emu/reg_round.S
+++ b/arch/x86/math-emu/reg_round.S
@@ -108,7 +108,7 @@ FPU_denormal:
 .globl fpu_Arith_exit
 
 /* Entry point when called from C */
-ENTRY(FPU_round)
+SYM_FUNC_START(FPU_round)
 	pushl	%ebp
 	movl	%esp,%ebp
 	pushl	%esi
@@ -707,4 +707,4 @@ L_exception_exit:
 	jmp	fpu_reg_round_special_exit
 #endif /* PARANOID */ 
 
-ENDPROC(FPU_round)
+SYM_FUNC_END(FPU_round)
diff --git a/arch/x86/math-emu/reg_u_add.S b/arch/x86/math-emu/reg_u_add.S
index 3b1bc5e9b2f6..9b21cbac6aa1 100644
--- a/arch/x86/math-emu/reg_u_add.S
+++ b/arch/x86/math-emu/reg_u_add.S
@@ -31,7 +31,7 @@
 #include "control_w.h"
 
 .text
-ENTRY(FPU_u_add)
+SYM_FUNC_START(FPU_u_add)
 	pushl	%ebp
 	movl	%esp,%ebp
 	pushl	%esi
@@ -165,4 +165,4 @@ L_exit:
 	leave
 	ret
 #endif /* PARANOID */
-ENDPROC(FPU_u_add)
+SYM_FUNC_END(FPU_u_add)
diff --git a/arch/x86/math-emu/reg_u_div.S b/arch/x86/math-emu/reg_u_div.S
index 796eb5ab921b..cb7c73807de9 100644
--- a/arch/x86/math-emu/reg_u_div.S
+++ b/arch/x86/math-emu/reg_u_div.S
@@ -74,7 +74,7 @@ FPU_ovfl_flag:
 #define DEST	PARAM3
 
 .text
-ENTRY(FPU_u_div)
+SYM_FUNC_START(FPU_u_div)
 	pushl	%ebp
 	movl	%esp,%ebp
 #ifndef NON_REENTRANT_FPU
@@ -470,4 +470,4 @@ L_exit:
 	ret
 #endif /* PARANOID */ 
 
-ENDPROC(FPU_u_div)
+SYM_FUNC_END(FPU_u_div)
diff --git a/arch/x86/math-emu/reg_u_mul.S b/arch/x86/math-emu/reg_u_mul.S
index 6196f68cf3c1..f36d62346785 100644
--- a/arch/x86/math-emu/reg_u_mul.S
+++ b/arch/x86/math-emu/reg_u_mul.S
@@ -44,7 +44,7 @@ FPU_accum_1:
 
 
 .text
-ENTRY(FPU_u_mul)
+SYM_FUNC_START(FPU_u_mul)
 	pushl	%ebp
 	movl	%esp,%ebp
 #ifndef NON_REENTRANT_FPU
@@ -146,4 +146,4 @@ L_exit:
 	ret
 #endif /* PARANOID */ 
 
-ENDPROC(FPU_u_mul)
+SYM_FUNC_END(FPU_u_mul)
diff --git a/arch/x86/math-emu/reg_u_sub.S b/arch/x86/math-emu/reg_u_sub.S
index d115b900919a..87c39c480701 100644
--- a/arch/x86/math-emu/reg_u_sub.S
+++ b/arch/x86/math-emu/reg_u_sub.S
@@ -32,7 +32,7 @@
 #include "control_w.h"
 
 .text
-ENTRY(FPU_u_sub)
+SYM_FUNC_START(FPU_u_sub)
 	pushl	%ebp
 	movl	%esp,%ebp
 	pushl	%esi
@@ -270,4 +270,4 @@ L_exit:
 	popl	%esi
 	leave
 	ret
-ENDPROC(FPU_u_sub)
+SYM_FUNC_END(FPU_u_sub)
diff --git a/arch/x86/math-emu/round_Xsig.S b/arch/x86/math-emu/round_Xsig.S
index 87c99749a495..42d4261d258d 100644
--- a/arch/x86/math-emu/round_Xsig.S
+++ b/arch/x86/math-emu/round_Xsig.S
@@ -22,7 +22,7 @@
 
 
 .text
-ENTRY(round_Xsig)
+SYM_FUNC_START(round_Xsig)
 	pushl	%ebp
 	movl	%esp,%ebp
 	pushl	%ebx		/* Reserve some space */
@@ -78,11 +78,11 @@ L_exit:
 	popl	%ebx
 	leave
 	ret
-ENDPROC(round_Xsig)
+SYM_FUNC_END(round_Xsig)
 
 
 
-ENTRY(norm_Xsig)
+SYM_FUNC_START(norm_Xsig)
 	pushl	%ebp
 	movl	%esp,%ebp
 	pushl	%ebx		/* Reserve some space */
@@ -138,4 +138,4 @@ L_n_exit:
 	popl	%ebx
 	leave
 	ret
-ENDPROC(norm_Xsig)
+SYM_FUNC_END(norm_Xsig)
diff --git a/arch/x86/math-emu/shr_Xsig.S b/arch/x86/math-emu/shr_Xsig.S
index c8552edeec75..ea50aa16add8 100644
--- a/arch/x86/math-emu/shr_Xsig.S
+++ b/arch/x86/math-emu/shr_Xsig.S
@@ -21,7 +21,7 @@
 #include "fpu_emu.h"
 
 .text
-ENTRY(shr_Xsig)
+SYM_FUNC_START(shr_Xsig)
 	push	%ebp
 	movl	%esp,%ebp
 	pushl	%esi
@@ -85,4 +85,4 @@ L_more_than_95:
 	popl	%esi
 	leave
 	ret
-ENDPROC(shr_Xsig)
+SYM_FUNC_END(shr_Xsig)
diff --git a/arch/x86/math-emu/wm_shrx.S b/arch/x86/math-emu/wm_shrx.S
index 340dd6897f85..1c10e993094d 100644
--- a/arch/x86/math-emu/wm_shrx.S
+++ b/arch/x86/math-emu/wm_shrx.S
@@ -32,7 +32,7 @@
  |   Results returned in the 64 bit arg and eax.                             |
  +---------------------------------------------------------------------------*/
 
-ENTRY(FPU_shrx)
+SYM_FUNC_START(FPU_shrx)
 	push	%ebp
 	movl	%esp,%ebp
 	pushl	%esi
@@ -92,7 +92,7 @@ L_more_than_95:
 	popl	%esi
 	leave
 	ret
-ENDPROC(FPU_shrx)
+SYM_FUNC_END(FPU_shrx)
 
 
 /*---------------------------------------------------------------------------+
@@ -111,7 +111,7 @@ ENDPROC(FPU_shrx)
  |   part which has been shifted out of the arg.                             |
  |   Results returned in the 64 bit arg and eax.                             |
  +---------------------------------------------------------------------------*/
-ENTRY(FPU_shrxs)
+SYM_FUNC_START(FPU_shrxs)
 	push	%ebp
 	movl	%esp,%ebp
 	pushl	%esi
@@ -203,4 +203,4 @@ Ls_more_than_95:
 	popl	%esi
 	leave
 	ret
-ENDPROC(FPU_shrxs)
+SYM_FUNC_END(FPU_shrxs)
diff --git a/arch/x86/math-emu/wm_sqrt.S b/arch/x86/math-emu/wm_sqrt.S
index 695afae38fdf..f5ac01472ee3 100644
--- a/arch/x86/math-emu/wm_sqrt.S
+++ b/arch/x86/math-emu/wm_sqrt.S
@@ -74,7 +74,7 @@ FPU_fsqrt_arg_0:
 
 
 .text
-ENTRY(wm_sqrt)
+SYM_FUNC_START(wm_sqrt)
 	pushl	%ebp
 	movl	%esp,%ebp
 #ifndef NON_REENTRANT_FPU
@@ -468,4 +468,4 @@ sqrt_more_prec_large:
 /* Our estimate is too large */
 	movl	$0x7fffff00,%eax
 	jmp	sqrt_round_result
-ENDPROC(wm_sqrt)
+SYM_FUNC_END(wm_sqrt)
diff --git a/arch/x86/platform/efi/efi_stub_32.S b/arch/x86/platform/efi/efi_stub_32.S
index 040192b50d02..9661a191138f 100644
--- a/arch/x86/platform/efi/efi_stub_32.S
+++ b/arch/x86/platform/efi/efi_stub_32.S
@@ -21,7 +21,7 @@
  */
 
 .text
-ENTRY(efi_call_phys)
+SYM_FUNC_START(efi_call_phys)
 	/*
 	 * 0. The function can only be called in Linux kernel. So CS has been
 	 * set to 0x0010, DS and SS have been set to 0x0018. In EFI, I found
@@ -113,7 +113,7 @@ ENTRY(efi_call_phys)
 	movl	(%edx), %ecx
 	pushl	%ecx
 	ret
-ENDPROC(efi_call_phys)
+SYM_FUNC_END(efi_call_phys)
 .previous
 
 .data
diff --git a/arch/x86/power/hibernate_asm_32.S b/arch/x86/power/hibernate_asm_32.S
index 9480f4b2df94..3fdc4d77ead0 100644
--- a/arch/x86/power/hibernate_asm_32.S
+++ b/arch/x86/power/hibernate_asm_32.S
@@ -14,7 +14,7 @@
 
 .text
 
-ENTRY(swsusp_arch_suspend)
+SYM_FUNC_START(swsusp_arch_suspend)
 	movl %esp, saved_context_esp
 	movl %ebx, saved_context_ebx
 	movl %ebp, saved_context_ebp
@@ -25,9 +25,9 @@ ENTRY(swsusp_arch_suspend)
 
 	call swsusp_save
 	ret
-ENDPROC(swsusp_arch_suspend)
+SYM_FUNC_END(swsusp_arch_suspend)
 
-ENTRY(restore_image)
+SYM_FUNC_START(restore_image)
 	movl	mmu_cr4_features, %ecx
 	movl	resume_pg_dir, %eax
 	subl	$__PAGE_OFFSET, %eax
@@ -83,5 +83,5 @@ done:
 	xorl	%eax, %eax
 
 	ret
-ENDPROC(restore_image)
+SYM_FUNC_END(restore_image)
 
diff --git a/arch/x86/realmode/rm/trampoline_32.S b/arch/x86/realmode/rm/trampoline_32.S
index 956b1c451661..8cee44fbd362 100644
--- a/arch/x86/realmode/rm/trampoline_32.S
+++ b/arch/x86/realmode/rm/trampoline_32.S
@@ -28,7 +28,7 @@
 	.code16
 
 	.balign	PAGE_SIZE
-ENTRY(trampoline_start)
+SYM_FUNC_START(trampoline_start)
 	wbinvd			# Needed for NUMA-Q should be harmless for others
 
 	LJMPW_RM(1f)
@@ -56,13 +56,13 @@ ENTRY(trampoline_start)
 	lmsw	%dx			# into protected mode
 
 	ljmpl	$__BOOT_CS, $pa_startup_32
-ENDPROC(trampoline_start)
+SYM_FUNC_END(trampoline_start)
 
 	.section ".text32","ax"
 	.code32
-ENTRY(startup_32)			# note: also used from wakeup_asm.S
+SYM_FUNC_START(startup_32)			# note: also used from wakeup_asm.S
 	jmp	*%eax
-ENDPROC(startup_32)
+SYM_FUNC_END(startup_32)
 
 	.bss
 	.balign 8
diff --git a/arch/x86/xen/xen-asm_32.S b/arch/x86/xen/xen-asm_32.S
index 748beeff6e7d..d4c6f6d50f29 100644
--- a/arch/x86/xen/xen-asm_32.S
+++ b/arch/x86/xen/xen-asm_32.S
@@ -72,7 +72,7 @@ check_events:
 	_ASM_EXTABLE(1b,2b)
 .endm
 
-ENTRY(xen_iret)
+SYM_FUNC_START(xen_iret)
 	/* test eflags for special cases */
 	testl $(X86_EFLAGS_VM | XEN_EFLAGS_NMI), 8(%esp)
 	jnz hyper_iret
@@ -138,7 +138,7 @@ xen_iret_end_crit:
 hyper_iret:
 	/* put this out of line since its very rarely used */
 	jmp hypercall_page + __HYPERVISOR_iret * 32
-ENDPROC(xen_iret)
+SYM_FUNC_END(xen_iret)
 
 	.globl xen_iret_start_crit, xen_iret_end_crit
 
@@ -182,7 +182,7 @@ ENDPROC(xen_iret)
  * SAVE_ALL state before going on, since it's usermode state which we
  * eventually need to restore.
  */
-ENTRY(xen_iret_crit_fixup)
+SYM_FUNC_START(xen_iret_crit_fixup)
 	/*
 	 * Paranoia: Make sure we're really coming from kernel space.
 	 * One could imagine a case where userspace jumps into the
@@ -221,4 +221,4 @@ ENTRY(xen_iret_crit_fixup)
 
 	lea 4(%edi), %esp		/* point esp to new frame */
 2:	jmp xen_do_upcall
-ENDPROC(xen_iret_crit_fixup)
+SYM_FUNC_END(xen_iret_crit_fixup)
diff --git a/drivers/lguest/x86/switcher_32.S b/drivers/lguest/x86/switcher_32.S
index 3318a967f28b..ad49e53e342c 100644
--- a/drivers/lguest/x86/switcher_32.S
+++ b/drivers/lguest/x86/switcher_32.S
@@ -101,7 +101,7 @@ ENTRY(start_switcher_text)
 // Where we save state and still see it from the Guest
 // And %ebx holds the Guest shadow pagetable:
 // Once set we have truly left Host behind.
-ENTRY(switch_to_guest)
+SYM_FUNC_START(switch_to_guest)
 	// We told gcc all its regs could fade,
 	// Clobbered by our journey into the Guest
 	// We could have saved them, if we tried
@@ -265,7 +265,7 @@ ENTRY(switch_to_guest)
 return_to_host:
 	SWITCH_TO_HOST
 	iret
-ENDPROC(switch_to_guest)
+SYM_FUNC_END(switch_to_guest)
 
 // We are lead to the second path like so:
 // An interrupt, with some cause external
-- 
2.12.2

[PATCH v3 28/29] x86_32: lguest, use SYM_ENTRY

[Jiri Slaby]

The last user of ENTRY is in lguest. ENTRY is going away, so switch them
to SYM_ENTRY.

Now, we can finally force ENTRY/ENDPROC to be undefined on X86.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: lguest@lists.ozlabs.org
---
 drivers/lguest/x86/switcher_32.S | 4 ++--
 include/linux/linkage.h          | 8 +++-----
 2 files changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/lguest/x86/switcher_32.S b/drivers/lguest/x86/switcher_32.S
index ad49e53e342c..3cb6aa352330 100644
--- a/drivers/lguest/x86/switcher_32.S
+++ b/drivers/lguest/x86/switcher_32.S
@@ -93,7 +93,7 @@
 // You'll see the trick macro at the end
 // Which interleaves data and text to effect.
 .text
-ENTRY(start_switcher_text)
+SYM_ENTRY(start_switcher_text, SYM_V_GLOBAL, SYM_A_NONE)
 
 // When we reach switch_to_guest we have just left
 // The safe and comforting shores of C code
@@ -386,4 +386,4 @@ handle_nmi:
 // Designed to make your fingers itch to code.
 
 // Here ends the text, the file and poem.
-ENTRY(end_switcher_text)
+SYM_ENTRY(end_switcher_text, SYM_V_GLOBAL, SYM_A_NONE)
diff --git a/include/linux/linkage.h b/include/linux/linkage.h
index f51928ae175b..b0b6b315cd95 100644
--- a/include/linux/linkage.h
+++ b/include/linux/linkage.h
@@ -99,13 +99,13 @@
 
 /* === DEPRECATED annotations === */
 
-#ifndef CONFIG_X86_64
+#ifndef CONFIG_X86
 #ifndef ENTRY
 /* deprecated, use SYM_FUNC_START */
 #define ENTRY(name) \
 	SYM_FUNC_START(name)
 #endif
-#endif /* CONFIG_X86_64 */
+#endif /* CONFIG_X86 */
 #endif /* LINKER_SCRIPT */
 
 #ifndef WEAK
@@ -120,9 +120,7 @@
 #define END(name) \
 	.size name, .-name
 #endif
-#endif /* CONFIG_X86 */
 
-#ifndef CONFIG_X86_64
 /* If symbol 'name' is treated as a subroutine (gets called, and returns)
  * then please use ENDPROC to mark 'name' as STT_FUNC for the benefit of
  * static analysis tools such as stack depth analyzer.
@@ -132,7 +130,7 @@
 #define ENDPROC(name) \
 	SYM_FUNC_END(name)
 #endif
-#endif /* CONFIG_X86_64 */
+#endif /* CONFIG_X86 */
 
 /* === generic annotations === */
 
-- 
2.12.2

[PATCH v3 29/29] x86: assembly, replace WEAK uses

[Jiri Slaby]

Use the new SYM_FUNC_INNER_LABEL for WEAK entries in the middle of x86
assembly functions.

And make sure WEAK is not defined for X86.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
---
 arch/x86/kernel/ftrace_32.S | 2 +-
 arch/x86/kernel/ftrace_64.S | 2 +-
 arch/x86/kernel/head_32.S   | 4 ++--
 include/linux/linkage.h     | 2 +-
 4 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/arch/x86/kernel/ftrace_32.S b/arch/x86/kernel/ftrace_32.S
index 9c3db18b2a19..8ec22538edb9 100644
--- a/arch/x86/kernel/ftrace_32.S
+++ b/arch/x86/kernel/ftrace_32.S
@@ -96,7 +96,7 @@ ftrace_graph_call:
 #endif
 
 /* This is weak to keep gas from relaxing the jumps */
-WEAK(ftrace_stub)
+SYM_FUNC_INNER_LABEL(ftrace_stub, SYM_V_WEAK)
 	ret
 SYM_FUNC_END(ftrace_caller)
 
diff --git a/arch/x86/kernel/ftrace_64.S b/arch/x86/kernel/ftrace_64.S
index 0970d85693c4..a6dc744d069f 100644
--- a/arch/x86/kernel/ftrace_64.S
+++ b/arch/x86/kernel/ftrace_64.S
@@ -181,7 +181,7 @@ SYM_FUNC_INNER_LABEL(ftrace_graph_call, SYM_V_GLOBAL)
 #endif
 
 /* This is weak to keep gas from relaxing the jumps */
-WEAK(ftrace_stub)
+SYM_FUNC_INNER_LABEL(ftrace_stub, SYM_V_WEAK)
 	retq
 SYM_FUNC_END(ftrace_caller)
 
diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S
index 74d18d2fef5a..ce976f1e7a0b 100644
--- a/arch/x86/kernel/head_32.S
+++ b/arch/x86/kernel/head_32.S
@@ -155,8 +155,8 @@ SYM_FUNC_START(startup_32)
 	jmp *%eax
 
 .Lbad_subarch:
-WEAK(lguest_entry)
-WEAK(xen_entry)
+SYM_FUNC_INNER_LABEL(lguest_entry, SYM_V_WEAK)
+SYM_FUNC_INNER_LABEL(xen_entry, SYM_V_WEAK)
 	/* Unknown implementation; there's really
 	   nothing we can do at this point. */
 	ud2a
diff --git a/include/linux/linkage.h b/include/linux/linkage.h
index b0b6b315cd95..9f4d06f04784 100644
--- a/include/linux/linkage.h
+++ b/include/linux/linkage.h
@@ -108,13 +108,13 @@
 #endif /* CONFIG_X86 */
 #endif /* LINKER_SCRIPT */
 
+#ifndef CONFIG_X86
 #ifndef WEAK
 /* deprecated, use SYM_FUNC_START_WEAK */
 #define WEAK(name)	   \
 	SYM_FUNC_START_WEAK(name)
 #endif
 
-#ifndef CONFIG_X86
 #ifndef END
 /* deprecated, use SYM_FUNC_END, SYM_DATA_END, or SYM_END */
 #define END(name) \
-- 
2.12.2