ARM: move ftrace assembly code to separate file

The ftrace assembly code doesn't need to live in entry-common.S and
be surrounded with #ifdef CONFIG_FUNCTION_TRACER.  Instead, move it
to its own file and conditionally assemble it.

Tested-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

1 files changed, 243 insertions, 0 deletions

diff --git a/arch/arm/kernel/entry-ftrace.S b/arch/arm/kernel/entry-ftrace.S
new file mode 100644
index 000000000000..fe57c73e70a4
--- /dev/null
+++ b/arch/arm/kernel/entry-ftrace.S
@@ -0,0 +1,243 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <asm/assembler.h>
+#include <asm/ftrace.h>
+#include <asm/unwind.h>
+
+#include "entry-header.S"
+
+/*
+ * When compiling with -pg, gcc inserts a call to the mcount routine at the
+ * start of every function.  In mcount, apart from the function's address (in
+ * lr), we need to get hold of the function's caller's address.
+ *
+ * Older GCCs (pre-4.4) inserted a call to a routine called mcount like this:
+ *
+ *	bl	mcount
+ *
+ * These versions have the limitation that in order for the mcount routine to
+ * be able to determine the function's caller's address, an APCS-style frame
+ * pointer (which is set up with something like the code below) is required.
+ *
+ *	mov     ip, sp
+ *	push    {fp, ip, lr, pc}
+ *	sub     fp, ip, #4
+ *
+ * With EABI, these frame pointers are not available unless -mapcs-frame is
+ * specified, and if building as Thumb-2, not even then.
+ *
+ * Newer GCCs (4.4+) solve this problem by introducing a new version of mcount,
+ * with call sites like:
+ *
+ *	push	{lr}
+ *	bl	__gnu_mcount_nc
+ *
+ * With these compilers, frame pointers are not necessary.
+ *
+ * mcount can be thought of as a function called in the middle of a subroutine
+ * call.  As such, it needs to be transparent for both the caller and the
+ * callee: the original lr needs to be restored when leaving mcount, and no
+ * registers should be clobbered.  (In the __gnu_mcount_nc implementation, we
+ * clobber the ip register.  This is OK because the ARM calling convention
+ * allows it to be clobbered in subroutines and doesn't use it to hold
+ * parameters.)
+ *
+ * When using dynamic ftrace, we patch out the mcount call by a "mov r0, r0"
+ * for the mcount case, and a "pop {lr}" for the __gnu_mcount_nc case (see
+ * arch/arm/kernel/ftrace.c).
+ */
+
+#ifndef CONFIG_OLD_MCOUNT
+#if (__GNUC__ < 4 || (__GNUC__ == 4 && __GNUC_MINOR__ < 4))
+#error Ftrace requires CONFIG_FRAME_POINTER=y with GCC older than 4.4.0.
+#endif
+#endif
+
+.macro mcount_adjust_addr rd, rn
+	bic	\rd, \rn, #1		@ clear the Thumb bit if present
+	sub	\rd, \rd, #MCOUNT_INSN_SIZE
+.endm
+
+.macro __mcount suffix
+	mcount_enter
+	ldr	r0, =ftrace_trace_function
+	ldr	r2, [r0]
+	adr	r0, .Lftrace_stub
+	cmp	r0, r2
+	bne	1f
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+	ldr     r1, =ftrace_graph_return
+	ldr     r2, [r1]
+	cmp     r0, r2
+	bne     ftrace_graph_caller\suffix
+
+	ldr     r1, =ftrace_graph_entry
+	ldr     r2, [r1]
+	ldr     r0, =ftrace_graph_entry_stub
+	cmp     r0, r2
+	bne     ftrace_graph_caller\suffix
+#endif
+
+	mcount_exit
+
+1: 	mcount_get_lr	r1			@ lr of instrumented func
+	mcount_adjust_addr	r0, lr		@ instrumented function
+	adr	lr, BSYM(2f)
+	mov	pc, r2
+2:	mcount_exit
+.endm
+
+.macro __ftrace_caller suffix
+	mcount_enter
+
+	mcount_get_lr	r1			@ lr of instrumented func
+	mcount_adjust_addr	r0, lr		@ instrumented function
+
+	.globl ftrace_call\suffix
+ftrace_call\suffix:
+	bl	ftrace_stub
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+	.globl ftrace_graph_call\suffix
+ftrace_graph_call\suffix:
+	mov	r0, r0
+#endif
+
+	mcount_exit
+.endm
+
+.macro __ftrace_graph_caller
+	sub	r0, fp, #4		@ &lr of instrumented routine (&parent)
+#ifdef CONFIG_DYNAMIC_FTRACE
+	@ called from __ftrace_caller, saved in mcount_enter
+	ldr	r1, [sp, #16]		@ instrumented routine (func)
+	mcount_adjust_addr	r1, r1
+#else
+	@ called from __mcount, untouched in lr
+	mcount_adjust_addr	r1, lr	@ instrumented routine (func)
+#endif
+	mov	r2, fp			@ frame pointer
+	bl	prepare_ftrace_return
+	mcount_exit
+.endm
+
+#ifdef CONFIG_OLD_MCOUNT
+/*
+ * mcount
+ */
+
+.macro mcount_enter
+	stmdb	sp!, {r0-r3, lr}
+.endm
+
+.macro mcount_get_lr reg
+	ldr	\reg, [fp, #-4]
+.endm
+
+.macro mcount_exit
+	ldr	lr, [fp, #-4]
+	ldmia	sp!, {r0-r3, pc}
+.endm
+
+ENTRY(mcount)
+#ifdef CONFIG_DYNAMIC_FTRACE
+	stmdb	sp!, {lr}
+	ldr	lr, [fp, #-4]
+	ldmia	sp!, {pc}
+#else
+	__mcount _old
+#endif
+ENDPROC(mcount)
+
+#ifdef CONFIG_DYNAMIC_FTRACE
+ENTRY(ftrace_caller_old)
+	__ftrace_caller _old
+ENDPROC(ftrace_caller_old)
+#endif
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+ENTRY(ftrace_graph_caller_old)
+	__ftrace_graph_caller
+ENDPROC(ftrace_graph_caller_old)
+#endif
+
+.purgem mcount_enter
+.purgem mcount_get_lr
+.purgem mcount_exit
+#endif
+
+/*
+ * __gnu_mcount_nc
+ */
+
+.macro mcount_enter
+/*
+ * This pad compensates for the push {lr} at the call site.  Note that we are
+ * unable to unwind through a function which does not otherwise save its lr.
+ */
+ UNWIND(.pad	#4)
+	stmdb	sp!, {r0-r3, lr}
+ UNWIND(.save	{r0-r3, lr})
+.endm
+
+.macro mcount_get_lr reg
+	ldr	\reg, [sp, #20]
+.endm
+
+.macro mcount_exit
+	ldmia	sp!, {r0-r3, ip, lr}
+	ret	ip
+.endm
+
+ENTRY(__gnu_mcount_nc)
+UNWIND(.fnstart)
+#ifdef CONFIG_DYNAMIC_FTRACE
+	mov	ip, lr
+	ldmia	sp!, {lr}
+	ret	ip
+#else
+	__mcount
+#endif
+UNWIND(.fnend)
+ENDPROC(__gnu_mcount_nc)
+
+#ifdef CONFIG_DYNAMIC_FTRACE
+ENTRY(ftrace_caller)
+UNWIND(.fnstart)
+	__ftrace_caller
+UNWIND(.fnend)
+ENDPROC(ftrace_caller)
+#endif
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+ENTRY(ftrace_graph_caller)
+UNWIND(.fnstart)
+	__ftrace_graph_caller
+UNWIND(.fnend)
+ENDPROC(ftrace_graph_caller)
+#endif
+
+.purgem mcount_enter
+.purgem mcount_get_lr
+.purgem mcount_exit
+
+#ifdef CONFIG_FUNCTION_GRAPH_TRACER
+	.globl return_to_handler
+return_to_handler:
+	stmdb	sp!, {r0-r3}
+	mov	r0, fp			@ frame pointer
+	bl	ftrace_return_to_handler
+	mov	lr, r0			@ r0 has real ret addr
+	ldmia	sp!, {r0-r3}
+	ret	lr
+#endif
+
+ENTRY(ftrace_stub)
+.Lftrace_stub:
+	ret	lr
+ENDPROC(ftrace_stub)