diff --git a/arch/arm64/include/asm/ftrace.h b/arch/arm64/include/asm/ftrace.h
index dc9cf0bd2a4c..5ccff4de7f09 100644
--- a/arch/arm64/include/asm/ftrace.h
+++ b/arch/arm64/include/asm/ftrace.h
@@ -54,8 +54,11 @@ extern void return_to_handler(void);
 unsigned long ftrace_call_adjust(unsigned long addr);
 
 #ifdef CONFIG_DYNAMIC_FTRACE_WITH_ARGS
+#define HAVE_ARCH_FTRACE_REGS
 struct dyn_ftrace;
 struct ftrace_ops;
+struct ftrace_regs;
+#define arch_ftrace_regs(fregs) ((struct __arch_ftrace_regs *)(fregs))
 
 #define arch_ftrace_get_regs(regs) NULL
 
@@ -63,7 +66,7 @@ struct ftrace_ops;
  * Note: sizeof(struct ftrace_regs) must be a multiple of 16 to ensure correct
  * stack alignment
  */
-struct ftrace_regs {
+struct __arch_ftrace_regs {
 	/* x0 - x8 */
 	unsigned long regs[9];
 
@@ -83,47 +86,47 @@ struct ftrace_regs {
 static __always_inline unsigned long
 ftrace_regs_get_instruction_pointer(const struct ftrace_regs *fregs)
 {
-	return fregs->pc;
+	return arch_ftrace_regs(fregs)->pc;
 }
 
 static __always_inline void
 ftrace_regs_set_instruction_pointer(struct ftrace_regs *fregs,
 				    unsigned long pc)
 {
-	fregs->pc = pc;
+	arch_ftrace_regs(fregs)->pc = pc;
 }
 
 static __always_inline unsigned long
 ftrace_regs_get_stack_pointer(const struct ftrace_regs *fregs)
 {
-	return fregs->sp;
+	return arch_ftrace_regs(fregs)->sp;
 }
 
 static __always_inline unsigned long
 ftrace_regs_get_argument(struct ftrace_regs *fregs, unsigned int n)
 {
 	if (n < 8)
-		return fregs->regs[n];
+		return arch_ftrace_regs(fregs)->regs[n];
 	return 0;
 }
 
 static __always_inline unsigned long
 ftrace_regs_get_return_value(const struct ftrace_regs *fregs)
 {
-	return fregs->regs[0];
+	return arch_ftrace_regs(fregs)->regs[0];
 }
 
 static __always_inline void
 ftrace_regs_set_return_value(struct ftrace_regs *fregs,
 			     unsigned long ret)
 {
-	fregs->regs[0] = ret;
+	arch_ftrace_regs(fregs)->regs[0] = ret;
 }
 
 static __always_inline void
 ftrace_override_function_with_return(struct ftrace_regs *fregs)
 {
-	fregs->pc = fregs->lr;
+	arch_ftrace_regs(fregs)->pc = arch_ftrace_regs(fregs)->lr;
 }
 
 int ftrace_regs_query_register_offset(const char *name);
@@ -143,7 +146,7 @@ static inline void arch_ftrace_set_direct_caller(struct ftrace_regs *fregs,
 	 * The ftrace trampoline will return to this address instead of the
 	 * instrumented function.
 	 */
-	fregs->direct_tramp = addr;
+	arch_ftrace_regs(fregs)->direct_tramp = addr;
 }
 #endif /* CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS */
 
diff --git a/arch/arm64/kernel/asm-offsets.c b/arch/arm64/kernel/asm-offsets.c
index f89df36ca4bd..29bf85dacffe 100644
--- a/arch/arm64/kernel/asm-offsets.c
+++ b/arch/arm64/kernel/asm-offsets.c
@@ -80,19 +80,19 @@ int main(void)
   DEFINE(PT_REGS_SIZE,		sizeof(struct pt_regs));
   BLANK();
 #ifdef CONFIG_DYNAMIC_FTRACE_WITH_ARGS
-  DEFINE(FREGS_X0,		offsetof(struct ftrace_regs, regs[0]));
-  DEFINE(FREGS_X2,		offsetof(struct ftrace_regs, regs[2]));
-  DEFINE(FREGS_X4,		offsetof(struct ftrace_regs, regs[4]));
-  DEFINE(FREGS_X6,		offsetof(struct ftrace_regs, regs[6]));
-  DEFINE(FREGS_X8,		offsetof(struct ftrace_regs, regs[8]));
-  DEFINE(FREGS_FP,		offsetof(struct ftrace_regs, fp));
-  DEFINE(FREGS_LR,		offsetof(struct ftrace_regs, lr));
-  DEFINE(FREGS_SP,		offsetof(struct ftrace_regs, sp));
-  DEFINE(FREGS_PC,		offsetof(struct ftrace_regs, pc));
+  DEFINE(FREGS_X0,		offsetof(struct __arch_ftrace_regs, regs[0]));
+  DEFINE(FREGS_X2,		offsetof(struct __arch_ftrace_regs, regs[2]));
+  DEFINE(FREGS_X4,		offsetof(struct __arch_ftrace_regs, regs[4]));
+  DEFINE(FREGS_X6,		offsetof(struct __arch_ftrace_regs, regs[6]));
+  DEFINE(FREGS_X8,		offsetof(struct __arch_ftrace_regs, regs[8]));
+  DEFINE(FREGS_FP,		offsetof(struct __arch_ftrace_regs, fp));
+  DEFINE(FREGS_LR,		offsetof(struct __arch_ftrace_regs, lr));
+  DEFINE(FREGS_SP,		offsetof(struct __arch_ftrace_regs, sp));
+  DEFINE(FREGS_PC,		offsetof(struct __arch_ftrace_regs, pc));
 #ifdef CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS
-  DEFINE(FREGS_DIRECT_TRAMP,	offsetof(struct ftrace_regs, direct_tramp));
+  DEFINE(FREGS_DIRECT_TRAMP,	offsetof(struct __arch_ftrace_regs, direct_tramp));
 #endif
-  DEFINE(FREGS_SIZE,		sizeof(struct ftrace_regs));
+  DEFINE(FREGS_SIZE,		sizeof(struct __arch_ftrace_regs));
   BLANK();
 #endif
   DEFINE(CPU_BOOT_TASK,		offsetof(struct secondary_data, task));
diff --git a/arch/arm64/kernel/ftrace.c b/arch/arm64/kernel/ftrace.c
index a650f5e11fc5..b2d947175cbe 100644
--- a/arch/arm64/kernel/ftrace.c
+++ b/arch/arm64/kernel/ftrace.c
@@ -23,10 +23,10 @@ struct fregs_offset {
 	int offset;
 };
 
-#define FREGS_OFFSET(n, field)				\
-{							\
-	.name = n,					\
-	.offset = offsetof(struct ftrace_regs, field),	\
+#define FREGS_OFFSET(n, field)					\
+{								\
+	.name = n,						\
+	.offset = offsetof(struct __arch_ftrace_regs, field),	\
 }
 
 static const struct fregs_offset fregs_offsets[] = {
@@ -481,7 +481,7 @@ void prepare_ftrace_return(unsigned long self_addr, unsigned long *parent,
 void ftrace_graph_func(unsigned long ip, unsigned long parent_ip,
 		       struct ftrace_ops *op, struct ftrace_regs *fregs)
 {
-	prepare_ftrace_return(ip, &fregs->lr, fregs->fp);
+	prepare_ftrace_return(ip, &arch_ftrace_regs(fregs)->lr, arch_ftrace_regs(fregs)->fp);
 }
 #else
 /*
diff --git a/arch/loongarch/include/asm/ftrace.h b/arch/loongarch/include/asm/ftrace.h
index c0a682808e07..8f13eaeaa325 100644
--- a/arch/loongarch/include/asm/ftrace.h
+++ b/arch/loongarch/include/asm/ftrace.h
@@ -44,40 +44,19 @@ void prepare_ftrace_return(unsigned long self_addr, unsigned long *parent);
 #ifdef CONFIG_HAVE_DYNAMIC_FTRACE_WITH_ARGS
 struct ftrace_ops;
 
-struct ftrace_regs {
-	struct pt_regs regs;
-};
+#include <linux/ftrace_regs.h>
 
 static __always_inline struct pt_regs *arch_ftrace_get_regs(struct ftrace_regs *fregs)
 {
-	return &fregs->regs;
-}
-
-static __always_inline unsigned long
-ftrace_regs_get_instruction_pointer(struct ftrace_regs *fregs)
-{
-	return instruction_pointer(&fregs->regs);
+	return &arch_ftrace_regs(fregs)->regs;
 }
 
 static __always_inline void
 ftrace_regs_set_instruction_pointer(struct ftrace_regs *fregs, unsigned long ip)
 {
-	instruction_pointer_set(&fregs->regs, ip);
+	instruction_pointer_set(&arch_ftrace_regs(fregs)->regs, ip);
 }
 
-#define ftrace_regs_get_argument(fregs, n) \
-	regs_get_kernel_argument(&(fregs)->regs, n)
-#define ftrace_regs_get_stack_pointer(fregs) \
-	kernel_stack_pointer(&(fregs)->regs)
-#define ftrace_regs_return_value(fregs) \
-	regs_return_value(&(fregs)->regs)
-#define ftrace_regs_set_return_value(fregs, ret) \
-	regs_set_return_value(&(fregs)->regs, ret)
-#define ftrace_override_function_with_return(fregs) \
-	override_function_with_return(&(fregs)->regs)
-#define ftrace_regs_query_register_offset(name) \
-	regs_query_register_offset(name)
-
 #define ftrace_graph_func ftrace_graph_func
 void ftrace_graph_func(unsigned long ip, unsigned long parent_ip,
 		       struct ftrace_ops *op, struct ftrace_regs *fregs);
@@ -90,7 +69,7 @@ __arch_ftrace_set_direct_caller(struct pt_regs *regs, unsigned long addr)
 }
 
 #define arch_ftrace_set_direct_caller(fregs, addr) \
-	__arch_ftrace_set_direct_caller(&(fregs)->regs, addr)
+	__arch_ftrace_set_direct_caller(&arch_ftrace_regs(fregs)->regs, addr)
 #endif /* CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS */
 
 #endif
diff --git a/arch/loongarch/kernel/ftrace_dyn.c b/arch/loongarch/kernel/ftrace_dyn.c
index bff058317062..18056229e22e 100644
--- a/arch/loongarch/kernel/ftrace_dyn.c
+++ b/arch/loongarch/kernel/ftrace_dyn.c
@@ -241,7 +241,7 @@ void prepare_ftrace_return(unsigned long self_addr, unsigned long *parent)
 void ftrace_graph_func(unsigned long ip, unsigned long parent_ip,
 		       struct ftrace_ops *op, struct ftrace_regs *fregs)
 {
-	struct pt_regs *regs = &fregs->regs;
+	struct pt_regs *regs = &arch_ftrace_regs(fregs)->regs;
 	unsigned long *parent = (unsigned long *)&regs->regs[1];
 
 	prepare_ftrace_return(ip, (unsigned long *)parent);
diff --git a/arch/powerpc/include/asm/ftrace.h b/arch/powerpc/include/asm/ftrace.h
index 559560286e6d..0edfb874eb02 100644
--- a/arch/powerpc/include/asm/ftrace.h
+++ b/arch/powerpc/include/asm/ftrace.h
@@ -32,42 +32,21 @@ struct dyn_arch_ftrace {
 int ftrace_init_nop(struct module *mod, struct dyn_ftrace *rec);
 #define ftrace_init_nop ftrace_init_nop
 
-struct ftrace_regs {
-	struct pt_regs regs;
-};
+#include <linux/ftrace_regs.h>
 
 static __always_inline struct pt_regs *arch_ftrace_get_regs(struct ftrace_regs *fregs)
 {
 	/* We clear regs.msr in ftrace_call */
-	return fregs->regs.msr ? &fregs->regs : NULL;
+	return arch_ftrace_regs(fregs)->regs.msr ? &arch_ftrace_regs(fregs)->regs : NULL;
 }
 
 static __always_inline void
 ftrace_regs_set_instruction_pointer(struct ftrace_regs *fregs,
 				    unsigned long ip)
 {
-	regs_set_return_ip(&fregs->regs, ip);
+	regs_set_return_ip(&arch_ftrace_regs(fregs)->regs, ip);
 }
 
-static __always_inline unsigned long
-ftrace_regs_get_instruction_pointer(struct ftrace_regs *fregs)
-{
-	return instruction_pointer(&fregs->regs);
-}
-
-#define ftrace_regs_get_argument(fregs, n) \
-	regs_get_kernel_argument(&(fregs)->regs, n)
-#define ftrace_regs_get_stack_pointer(fregs) \
-	kernel_stack_pointer(&(fregs)->regs)
-#define ftrace_regs_return_value(fregs) \
-	regs_return_value(&(fregs)->regs)
-#define ftrace_regs_set_return_value(fregs, ret) \
-	regs_set_return_value(&(fregs)->regs, ret)
-#define ftrace_override_function_with_return(fregs) \
-	override_function_with_return(&(fregs)->regs)
-#define ftrace_regs_query_register_offset(name) \
-	regs_query_register_offset(name)
-
 struct ftrace_ops;
 
 #define ftrace_graph_func ftrace_graph_func
diff --git a/arch/powerpc/kernel/trace/ftrace.c b/arch/powerpc/kernel/trace/ftrace.c
index d8d6b4fd9a14..df41f4a7c738 100644
--- a/arch/powerpc/kernel/trace/ftrace.c
+++ b/arch/powerpc/kernel/trace/ftrace.c
@@ -421,7 +421,7 @@ int __init ftrace_dyn_arch_init(void)
 void ftrace_graph_func(unsigned long ip, unsigned long parent_ip,
 		       struct ftrace_ops *op, struct ftrace_regs *fregs)
 {
-	unsigned long sp = fregs->regs.gpr[1];
+	unsigned long sp = arch_ftrace_regs(fregs)->regs.gpr[1];
 	int bit;
 
 	if (unlikely(ftrace_graph_is_dead()))
@@ -439,6 +439,6 @@ void ftrace_graph_func(unsigned long ip, unsigned long parent_ip,
 
 	ftrace_test_recursion_unlock(bit);
 out:
-	fregs->regs.link = parent_ip;
+	arch_ftrace_regs(fregs)->regs.link = parent_ip;
 }
 #endif /* CONFIG_FUNCTION_GRAPH_TRACER */
diff --git a/arch/powerpc/kernel/trace/ftrace_64_pg.c b/arch/powerpc/kernel/trace/ftrace_64_pg.c
index 12fab1803bcf..d3c5552e4984 100644
--- a/arch/powerpc/kernel/trace/ftrace_64_pg.c
+++ b/arch/powerpc/kernel/trace/ftrace_64_pg.c
@@ -829,7 +829,7 @@ __prepare_ftrace_return(unsigned long parent, unsigned long ip, unsigned long sp
 void ftrace_graph_func(unsigned long ip, unsigned long parent_ip,
 		       struct ftrace_ops *op, struct ftrace_regs *fregs)
 {
-	fregs->regs.link = __prepare_ftrace_return(parent_ip, ip, fregs->regs.gpr[1]);
+	arch_ftrace_regs(fregs)->regs.link = __prepare_ftrace_return(parent_ip, ip, arch_ftrace_regs(fregs)->regs.gpr[1]);
 }
 #else
 unsigned long prepare_ftrace_return(unsigned long parent, unsigned long ip,
diff --git a/arch/riscv/include/asm/ftrace.h b/arch/riscv/include/asm/ftrace.h
index 2cddd79ff21b..3d66437a1029 100644
--- a/arch/riscv/include/asm/ftrace.h
+++ b/arch/riscv/include/asm/ftrace.h
@@ -125,8 +125,12 @@ int ftrace_init_nop(struct module *mod, struct dyn_ftrace *rec);
 
 #ifdef CONFIG_DYNAMIC_FTRACE_WITH_ARGS
 #define arch_ftrace_get_regs(regs) NULL
+#define HAVE_ARCH_FTRACE_REGS
 struct ftrace_ops;
-struct ftrace_regs {
+struct ftrace_regs;
+#define arch_ftrace_regs(fregs) ((struct __arch_ftrace_regs *)(fregs))
+
+struct __arch_ftrace_regs {
 	unsigned long epc;
 	unsigned long ra;
 	unsigned long sp;
@@ -150,42 +154,42 @@ struct ftrace_regs {
 static __always_inline unsigned long ftrace_regs_get_instruction_pointer(const struct ftrace_regs
 									 *fregs)
 {
-	return fregs->epc;
+	return arch_ftrace_regs(fregs)->epc;
 }
 
 static __always_inline void ftrace_regs_set_instruction_pointer(struct ftrace_regs *fregs,
 								unsigned long pc)
 {
-	fregs->epc = pc;
+	arch_ftrace_regs(fregs)->epc = pc;
 }
 
 static __always_inline unsigned long ftrace_regs_get_stack_pointer(const struct ftrace_regs *fregs)
 {
-	return fregs->sp;
+	return arch_ftrace_regs(fregs)->sp;
 }
 
 static __always_inline unsigned long ftrace_regs_get_argument(struct ftrace_regs *fregs,
 							      unsigned int n)
 {
 	if (n < 8)
-		return fregs->args[n];
+		return arch_ftrace_regs(fregs)->args[n];
 	return 0;
 }
 
 static __always_inline unsigned long ftrace_regs_get_return_value(const struct ftrace_regs *fregs)
 {
-	return fregs->a0;
+	return arch_ftrace_regs(fregs)->a0;
 }
 
 static __always_inline void ftrace_regs_set_return_value(struct ftrace_regs *fregs,
 							 unsigned long ret)
 {
-	fregs->a0 = ret;
+	arch_ftrace_regs(fregs)->a0 = ret;
 }
 
 static __always_inline void ftrace_override_function_with_return(struct ftrace_regs *fregs)
 {
-	fregs->epc = fregs->ra;
+	arch_ftrace_regs(fregs)->epc = arch_ftrace_regs(fregs)->ra;
 }
 
 int ftrace_regs_query_register_offset(const char *name);
@@ -196,7 +200,7 @@ void ftrace_graph_func(unsigned long ip, unsigned long parent_ip,
 
 static inline void arch_ftrace_set_direct_caller(struct ftrace_regs *fregs, unsigned long addr)
 {
-	fregs->t1 = addr;
+	arch_ftrace_regs(fregs)->t1 = addr;
 }
 #endif /* CONFIG_DYNAMIC_FTRACE_WITH_ARGS */
 
diff --git a/arch/riscv/kernel/asm-offsets.c b/arch/riscv/kernel/asm-offsets.c
index c2f3129a8e5c..e89455a6a0e5 100644
--- a/arch/riscv/kernel/asm-offsets.c
+++ b/arch/riscv/kernel/asm-offsets.c
@@ -496,19 +496,19 @@ void asm_offsets(void)
 	OFFSET(STACKFRAME_RA, stackframe, ra);
 
 #ifdef CONFIG_DYNAMIC_FTRACE_WITH_ARGS
-	DEFINE(FREGS_SIZE_ON_STACK, ALIGN(sizeof(struct ftrace_regs), STACK_ALIGN));
-	DEFINE(FREGS_EPC,	    offsetof(struct ftrace_regs, epc));
-	DEFINE(FREGS_RA,	    offsetof(struct ftrace_regs, ra));
-	DEFINE(FREGS_SP,	    offsetof(struct ftrace_regs, sp));
-	DEFINE(FREGS_S0,	    offsetof(struct ftrace_regs, s0));
-	DEFINE(FREGS_T1,	    offsetof(struct ftrace_regs, t1));
-	DEFINE(FREGS_A0,	    offsetof(struct ftrace_regs, a0));
-	DEFINE(FREGS_A1,	    offsetof(struct ftrace_regs, a1));
-	DEFINE(FREGS_A2,	    offsetof(struct ftrace_regs, a2));
-	DEFINE(FREGS_A3,	    offsetof(struct ftrace_regs, a3));
-	DEFINE(FREGS_A4,	    offsetof(struct ftrace_regs, a4));
-	DEFINE(FREGS_A5,	    offsetof(struct ftrace_regs, a5));
-	DEFINE(FREGS_A6,	    offsetof(struct ftrace_regs, a6));
-	DEFINE(FREGS_A7,	    offsetof(struct ftrace_regs, a7));
+	DEFINE(FREGS_SIZE_ON_STACK, ALIGN(sizeof(struct __arch_ftrace_regs), STACK_ALIGN));
+	DEFINE(FREGS_EPC,	    offsetof(struct __arch_ftrace_regs, epc));
+	DEFINE(FREGS_RA,	    offsetof(struct __arch_ftrace_regs, ra));
+	DEFINE(FREGS_SP,	    offsetof(struct __arch_ftrace_regs, sp));
+	DEFINE(FREGS_S0,	    offsetof(struct __arch_ftrace_regs, s0));
+	DEFINE(FREGS_T1,	    offsetof(struct __arch_ftrace_regs, t1));
+	DEFINE(FREGS_A0,	    offsetof(struct __arch_ftrace_regs, a0));
+	DEFINE(FREGS_A1,	    offsetof(struct __arch_ftrace_regs, a1));
+	DEFINE(FREGS_A2,	    offsetof(struct __arch_ftrace_regs, a2));
+	DEFINE(FREGS_A3,	    offsetof(struct __arch_ftrace_regs, a3));
+	DEFINE(FREGS_A4,	    offsetof(struct __arch_ftrace_regs, a4));
+	DEFINE(FREGS_A5,	    offsetof(struct __arch_ftrace_regs, a5));
+	DEFINE(FREGS_A6,	    offsetof(struct __arch_ftrace_regs, a6));
+	DEFINE(FREGS_A7,	    offsetof(struct __arch_ftrace_regs, a7));
 #endif
 }
diff --git a/arch/riscv/kernel/ftrace.c b/arch/riscv/kernel/ftrace.c
index 4b95c574fd04..5081ad886841 100644
--- a/arch/riscv/kernel/ftrace.c
+++ b/arch/riscv/kernel/ftrace.c
@@ -214,7 +214,7 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr,
 void ftrace_graph_func(unsigned long ip, unsigned long parent_ip,
 		       struct ftrace_ops *op, struct ftrace_regs *fregs)
 {
-	prepare_ftrace_return(&fregs->ra, ip, fregs->s0);
+	prepare_ftrace_return(&arch_ftrace_regs(fregs)->ra, ip, arch_ftrace_regs(fregs)->s0);
 }
 #else /* CONFIG_DYNAMIC_FTRACE_WITH_ARGS */
 extern void ftrace_graph_call(void);
diff --git a/arch/s390/include/asm/ftrace.h b/arch/s390/include/asm/ftrace.h
index 406746666eb7..fc97d75dc752 100644
--- a/arch/s390/include/asm/ftrace.h
+++ b/arch/s390/include/asm/ftrace.h
@@ -51,13 +51,11 @@ static inline unsigned long ftrace_call_adjust(unsigned long addr)
 	return addr;
 }
 
-struct ftrace_regs {
-	struct pt_regs regs;
-};
+#include <linux/ftrace_regs.h>
 
 static __always_inline struct pt_regs *arch_ftrace_get_regs(struct ftrace_regs *fregs)
 {
-	struct pt_regs *regs = &fregs->regs;
+	struct pt_regs *regs = &arch_ftrace_regs(fregs)->regs;
 
 	if (test_pt_regs_flag(regs, PIF_FTRACE_FULL_REGS))
 		return regs;
@@ -81,32 +79,13 @@ static __always_inline unsigned long fgraph_ret_regs_frame_pointer(struct fgraph
 }
 #endif /* CONFIG_FUNCTION_GRAPH_TRACER */
 
-static __always_inline unsigned long
-ftrace_regs_get_instruction_pointer(const struct ftrace_regs *fregs)
-{
-	return fregs->regs.psw.addr;
-}
-
 static __always_inline void
 ftrace_regs_set_instruction_pointer(struct ftrace_regs *fregs,
 				    unsigned long ip)
 {
-	fregs->regs.psw.addr = ip;
+	arch_ftrace_regs(fregs)->regs.psw.addr = ip;
 }
 
-#define ftrace_regs_get_argument(fregs, n) \
-	regs_get_kernel_argument(&(fregs)->regs, n)
-#define ftrace_regs_get_stack_pointer(fregs) \
-	kernel_stack_pointer(&(fregs)->regs)
-#define ftrace_regs_return_value(fregs) \
-	regs_return_value(&(fregs)->regs)
-#define ftrace_regs_set_return_value(fregs, ret) \
-	regs_set_return_value(&(fregs)->regs, ret)
-#define ftrace_override_function_with_return(fregs) \
-	override_function_with_return(&(fregs)->regs)
-#define ftrace_regs_query_register_offset(name) \
-	regs_query_register_offset(name)
-
 #ifdef CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS
 /*
  * When an ftrace registered caller is tracing a function that is
@@ -117,7 +96,7 @@ ftrace_regs_set_instruction_pointer(struct ftrace_regs *fregs,
  */
 static inline void arch_ftrace_set_direct_caller(struct ftrace_regs *fregs, unsigned long addr)
 {
-	struct pt_regs *regs = &fregs->regs;
+	struct pt_regs *regs = &arch_ftrace_regs(fregs)->regs;
 	regs->orig_gpr2 = addr;
 }
 #endif /* CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS */
diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c
index 1d7ed0faff8b..862a9140528e 100644
--- a/arch/s390/kernel/asm-offsets.c
+++ b/arch/s390/kernel/asm-offsets.c
@@ -181,8 +181,8 @@ int main(void)
 	OFFSET(__FGRAPH_RET_FP, fgraph_ret_regs, fp);
 	DEFINE(__FGRAPH_RET_SIZE, sizeof(struct fgraph_ret_regs));
 #endif
-	OFFSET(__FTRACE_REGS_PT_REGS, ftrace_regs, regs);
-	DEFINE(__FTRACE_REGS_SIZE, sizeof(struct ftrace_regs));
+	OFFSET(__FTRACE_REGS_PT_REGS, __arch_ftrace_regs, regs);
+	DEFINE(__FTRACE_REGS_SIZE, sizeof(struct __arch_ftrace_regs));
 
 	OFFSET(__PCPU_FLAGS, pcpu, flags);
 	return 0;
diff --git a/arch/s390/kernel/ftrace.c b/arch/s390/kernel/ftrace.c
index 0b6e62d1d8b8..51439a71e392 100644
--- a/arch/s390/kernel/ftrace.c
+++ b/arch/s390/kernel/ftrace.c
@@ -318,7 +318,7 @@ void kprobe_ftrace_handler(unsigned long ip, unsigned long parent_ip,
 	if (bit < 0)
 		return;
 
-	kmsan_unpoison_memory(fregs, sizeof(*fregs));
+	kmsan_unpoison_memory(fregs, ftrace_regs_size());
 	regs = ftrace_get_regs(fregs);
 	p = get_kprobe((kprobe_opcode_t *)ip);
 	if (!regs || unlikely(!p) || kprobe_disabled(p))
diff --git a/arch/s390/lib/test_unwind.c b/arch/s390/lib/test_unwind.c
index 8b7f981e6f34..6e42100875e7 100644
--- a/arch/s390/lib/test_unwind.c
+++ b/arch/s390/lib/test_unwind.c
@@ -270,9 +270,9 @@ static void notrace __used test_unwind_ftrace_handler(unsigned long ip,
 						      struct ftrace_ops *fops,
 						      struct ftrace_regs *fregs)
 {
-	struct unwindme *u = (struct unwindme *)fregs->regs.gprs[2];
+	struct unwindme *u = (struct unwindme *)arch_ftrace_regs(fregs)->regs.gprs[2];
 
-	u->ret = test_unwind(NULL, (u->flags & UWM_REGS) ? &fregs->regs : NULL,
+	u->ret = test_unwind(NULL, (u->flags & UWM_REGS) ? &arch_ftrace_regs(fregs)->regs : NULL,
 			     (u->flags & UWM_SP) ? u->sp : 0);
 }
 
diff --git a/arch/x86/include/asm/ftrace.h b/arch/x86/include/asm/ftrace.h
index b4d719de2c84..6e8cf0fa48fc 100644
--- a/arch/x86/include/asm/ftrace.h
+++ b/arch/x86/include/asm/ftrace.h
@@ -35,37 +35,21 @@ static inline unsigned long ftrace_call_adjust(unsigned long addr)
 }
 
 #ifdef CONFIG_HAVE_DYNAMIC_FTRACE_WITH_ARGS
-struct ftrace_regs {
-	struct pt_regs		regs;
-};
+
+#include <linux/ftrace_regs.h>
 
 static __always_inline struct pt_regs *
 arch_ftrace_get_regs(struct ftrace_regs *fregs)
 {
 	/* Only when FL_SAVE_REGS is set, cs will be non zero */
-	if (!fregs->regs.cs)
+	if (!arch_ftrace_regs(fregs)->regs.cs)
 		return NULL;
-	return &fregs->regs;
+	return &arch_ftrace_regs(fregs)->regs;
 }
 
 #define ftrace_regs_set_instruction_pointer(fregs, _ip)	\
-	do { (fregs)->regs.ip = (_ip); } while (0)
+	do { arch_ftrace_regs(fregs)->regs.ip = (_ip); } while (0)
 
-#define ftrace_regs_get_instruction_pointer(fregs) \
-	((fregs)->regs.ip)
-
-#define ftrace_regs_get_argument(fregs, n) \
-	regs_get_kernel_argument(&(fregs)->regs, n)
-#define ftrace_regs_get_stack_pointer(fregs) \
-	kernel_stack_pointer(&(fregs)->regs)
-#define ftrace_regs_return_value(fregs) \
-	regs_return_value(&(fregs)->regs)
-#define ftrace_regs_set_return_value(fregs, ret) \
-	regs_set_return_value(&(fregs)->regs, ret)
-#define ftrace_override_function_with_return(fregs) \
-	override_function_with_return(&(fregs)->regs)
-#define ftrace_regs_query_register_offset(name) \
-	regs_query_register_offset(name)
 
 struct ftrace_ops;
 #define ftrace_graph_func ftrace_graph_func
@@ -90,7 +74,7 @@ __arch_ftrace_set_direct_caller(struct pt_regs *regs, unsigned long addr)
 	regs->orig_ax = addr;
 }
 #define arch_ftrace_set_direct_caller(fregs, addr) \
-	__arch_ftrace_set_direct_caller(&(fregs)->regs, addr)
+	__arch_ftrace_set_direct_caller(&arch_ftrace_regs(fregs)->regs, addr)
 #endif /* CONFIG_DYNAMIC_FTRACE_WITH_DIRECT_CALLS */
 
 #ifdef CONFIG_DYNAMIC_FTRACE
diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c
index 8da0e66ca22d..adb09f78edb2 100644
--- a/arch/x86/kernel/ftrace.c
+++ b/arch/x86/kernel/ftrace.c
@@ -647,7 +647,7 @@ void prepare_ftrace_return(unsigned long ip, unsigned long *parent,
 void ftrace_graph_func(unsigned long ip, unsigned long parent_ip,
 		       struct ftrace_ops *op, struct ftrace_regs *fregs)
 {
-	struct pt_regs *regs = &fregs->regs;
+	struct pt_regs *regs = &arch_ftrace_regs(fregs)->regs;
 	unsigned long *stack = (unsigned long *)kernel_stack_pointer(regs);
 
 	prepare_ftrace_return(ip, (unsigned long *)stack, 0);
diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index fd5e84d0ec47..aa9ddd1e4bb6 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -113,14 +113,54 @@ static inline int ftrace_mod_get_kallsym(unsigned int symnum, unsigned long *val
 
 #ifdef CONFIG_FUNCTION_TRACER
 
+#include <linux/ftrace_regs.h>
+
 extern int ftrace_enabled;
 
-#ifndef CONFIG_HAVE_DYNAMIC_FTRACE_WITH_ARGS
-
+/**
+ * ftrace_regs - ftrace partial/optimal register set
+ *
+ * ftrace_regs represents a group of registers which is used at the
+ * function entry and exit. There are three types of registers.
+ *
+ * - Registers for passing the parameters to callee, including the stack
+ *   pointer. (e.g. rcx, rdx, rdi, rsi, r8, r9 and rsp on x86_64)
+ * - Registers for passing the return values to caller.
+ *   (e.g. rax and rdx on x86_64)
+ * - Registers for hooking the function call and return including the
+ *   frame pointer (the frame pointer is architecture/config dependent)
+ *   (e.g. rip, rbp and rsp for x86_64)
+ *
+ * Also, architecture dependent fields can be used for internal process.
+ * (e.g. orig_ax on x86_64)
+ *
+ * On the function entry, those registers will be restored except for
+ * the stack pointer, so that user can change the function parameters
+ * and instruction pointer (e.g. live patching.)
+ * On the function exit, only registers which is used for return values
+ * are restored.
+ *
+ * NOTE: user *must not* access regs directly, only do it via APIs, because
+ * the member can be changed according to the architecture.
+ * This is why the structure is empty here, so that nothing accesses
+ * the ftrace_regs directly.
+ */
 struct ftrace_regs {
-	struct pt_regs		regs;
+	/* Nothing to see here, use the accessor functions! */
 };
-#define arch_ftrace_get_regs(fregs) (&(fregs)->regs)
+
+#define ftrace_regs_size()	sizeof(struct __arch_ftrace_regs)
+
+#ifndef CONFIG_HAVE_DYNAMIC_FTRACE_WITH_ARGS
+/*
+ * Architectures that define HAVE_DYNAMIC_FTRACE_WITH_ARGS must define their own
+ * arch_ftrace_get_regs() where it only returns pt_regs *if* it is fully
+ * populated. It should return NULL otherwise.
+ */
+static inline struct pt_regs *arch_ftrace_get_regs(struct ftrace_regs *fregs)
+{
+	return &arch_ftrace_regs(fregs)->regs;
+}
 
 /*
  * ftrace_regs_set_instruction_pointer() is to be defined by the architecture
@@ -150,23 +190,6 @@ static __always_inline bool ftrace_regs_has_args(struct ftrace_regs *fregs)
 	return ftrace_get_regs(fregs) != NULL;
 }
 
-#ifndef CONFIG_HAVE_DYNAMIC_FTRACE_WITH_ARGS
-#define ftrace_regs_get_instruction_pointer(fregs) \
-	instruction_pointer(ftrace_get_regs(fregs))
-#define ftrace_regs_get_argument(fregs, n) \
-	regs_get_kernel_argument(ftrace_get_regs(fregs), n)
-#define ftrace_regs_get_stack_pointer(fregs) \
-	kernel_stack_pointer(ftrace_get_regs(fregs))
-#define ftrace_regs_return_value(fregs) \
-	regs_return_value(ftrace_get_regs(fregs))
-#define ftrace_regs_set_return_value(fregs, ret) \
-	regs_set_return_value(ftrace_get_regs(fregs), ret)
-#define ftrace_override_function_with_return(fregs) \
-	override_function_with_return(ftrace_get_regs(fregs))
-#define ftrace_regs_query_register_offset(name) \
-	regs_query_register_offset(name)
-#endif
-
 typedef void (*ftrace_func_t)(unsigned long ip, unsigned long parent_ip,
 			      struct ftrace_ops *op, struct ftrace_regs *fregs);
 
@@ -1014,6 +1037,17 @@ struct ftrace_graph_ent {
 	int depth;
 } __packed;
 
+/*
+ * Structure that defines an entry function trace with retaddr.
+ * It's already packed but the attribute "packed" is needed
+ * to remove extra padding at the end.
+ */
+struct fgraph_retaddr_ent {
+	unsigned long func; /* Current function */
+	int depth;
+	unsigned long retaddr;  /* Return address */
+} __packed;
+
 /*
  * Structure that defines a return function trace.
  * It's already packed but the attribute "packed" is needed
@@ -1039,7 +1073,8 @@ typedef void (*trace_func_graph_ret_t)(struct ftrace_graph_ret *,
 typedef int (*trace_func_graph_ent_t)(struct ftrace_graph_ent *,
 				      struct fgraph_ops *); /* entry */
 
-extern int ftrace_graph_entry_stub(struct ftrace_graph_ent *trace, struct fgraph_ops *gops);
+extern int ftrace_graph_entry_stub(struct ftrace_graph_ent *trace,
+				   struct fgraph_ops *gops);
 bool ftrace_pids_enabled(struct ftrace_ops *ops);
 
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
@@ -1055,6 +1090,7 @@ struct fgraph_ops {
 
 void *fgraph_reserve_data(int idx, int size_bytes);
 void *fgraph_retrieve_data(int idx, int *size_bytes);
+void *fgraph_retrieve_parent_data(int idx, int *size_bytes, int depth);
 
 /*
  * Stack of return addresses for functions
@@ -1064,10 +1100,6 @@ void *fgraph_retrieve_data(int idx, int *size_bytes);
 struct ftrace_ret_stack {
 	unsigned long ret;
 	unsigned long func;
-	unsigned long long calltime;
-#ifdef CONFIG_FUNCTION_PROFILER
-	unsigned long long subtime;
-#endif
 #ifdef HAVE_FUNCTION_GRAPH_FP_TEST
 	unsigned long fp;
 #endif
@@ -1087,6 +1119,7 @@ function_graph_enter(unsigned long ret, unsigned long func,
 
 struct ftrace_ret_stack *
 ftrace_graph_get_ret_stack(struct task_struct *task, int skip);
+unsigned long ftrace_graph_top_ret_addr(struct task_struct *task);
 
 unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx,
 				    unsigned long ret, unsigned long *retp);
diff --git a/include/linux/ftrace_regs.h b/include/linux/ftrace_regs.h
new file mode 100644
index 000000000000..be1ed0c891d0
--- /dev/null
+++ b/include/linux/ftrace_regs.h
@@ -0,0 +1,36 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _LINUX_FTRACE_REGS_H
+#define _LINUX_FTRACE_REGS_H
+
+/*
+ * For archs that just copy pt_regs in ftrace regs, it can use this default.
+ * If an architecture does not use pt_regs, it must define all the below
+ * accessor functions.
+ */
+#ifndef HAVE_ARCH_FTRACE_REGS
+struct __arch_ftrace_regs {
+	struct pt_regs		regs;
+};
+
+#define arch_ftrace_regs(fregs) ((struct __arch_ftrace_regs *)(fregs))
+
+struct ftrace_regs;
+
+#define ftrace_regs_get_instruction_pointer(fregs) \
+	instruction_pointer(&arch_ftrace_regs(fregs)->regs)
+#define ftrace_regs_get_argument(fregs, n) \
+	regs_get_kernel_argument(&arch_ftrace_regs(fregs)->regs, n)
+#define ftrace_regs_get_stack_pointer(fregs) \
+	kernel_stack_pointer(&arch_ftrace_regs(fregs)->regs)
+#define ftrace_regs_get_return_value(fregs) \
+	regs_return_value(&arch_ftrace_regs(fregs)->regs)
+#define ftrace_regs_set_return_value(fregs, ret) \
+	regs_set_return_value(&arch_ftrace_regs(fregs)->regs, ret)
+#define ftrace_override_function_with_return(fregs) \
+	override_function_with_return(&arch_ftrace_regs(fregs)->regs)
+#define ftrace_regs_query_register_offset(name) \
+	regs_query_register_offset(name)
+
+#endif /* HAVE_ARCH_FTRACE_REGS */
+
+#endif /* _LINUX_FTRACE_REGS_H */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 5a2cdd9d8b10..f0e9e00d3cf5 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1441,6 +1441,7 @@ struct task_struct {
 
 	/* Timestamp for last schedule: */
 	unsigned long long		ftrace_timestamp;
+	unsigned long long		ftrace_sleeptime;
 
 	/*
 	 * Number of functions that haven't been traced
diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index 721c3b221048..74c2b1d43bb9 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -242,6 +242,16 @@ config FUNCTION_GRAPH_RETVAL
 	  enable it via the trace option funcgraph-retval.
 	  See Documentation/trace/ftrace.rst
 
+config FUNCTION_GRAPH_RETADDR
+	bool "Kernel Function Graph Return Address"
+	depends on FUNCTION_GRAPH_TRACER
+	default n
+	help
+	  Support recording and printing the function return address when
+	  using function graph tracer. It can be helpful to locate code line that
+	  the function is called. This feature is off by default, and you can
+	  enable it via the trace option funcgraph-retaddr.
+
 config DYNAMIC_FTRACE
 	bool "enable/disable function tracing dynamically"
 	depends on FUNCTION_TRACER
diff --git a/kernel/trace/fgraph.c b/kernel/trace/fgraph.c
index 69e226a48daa..0bf78517b5d4 100644
--- a/kernel/trace/fgraph.c
+++ b/kernel/trace/fgraph.c
@@ -153,7 +153,7 @@ enum {
  * SHADOW_STACK_OFFSET:	The size in long words of the shadow stack
  * SHADOW_STACK_MAX_OFFSET: The max offset of the stack for a new frame to be added
  */
-#define SHADOW_STACK_SIZE	(PAGE_SIZE)
+#define SHADOW_STACK_SIZE	(4096)
 #define SHADOW_STACK_OFFSET	(SHADOW_STACK_SIZE / sizeof(long))
 /* Leave on a buffer at the end */
 #define SHADOW_STACK_MAX_OFFSET				\
@@ -172,6 +172,8 @@ enum {
 DEFINE_STATIC_KEY_FALSE(kill_ftrace_graph);
 int ftrace_graph_active;
 
+static struct kmem_cache *fgraph_stack_cachep;
+
 static struct fgraph_ops *fgraph_array[FGRAPH_ARRAY_SIZE];
 static unsigned long fgraph_array_bitmask;
 
@@ -390,21 +392,7 @@ void *fgraph_reserve_data(int idx, int size_bytes)
  */
 void *fgraph_retrieve_data(int idx, int *size_bytes)
 {
-	int offset = current->curr_ret_stack - 1;
-	unsigned long val;
-
-	val = get_fgraph_entry(current, offset);
-	while (__get_type(val) == FGRAPH_TYPE_DATA) {
-		if (__get_data_index(val) == idx)
-			goto found;
-		offset -= __get_data_size(val) + 1;
-		val = get_fgraph_entry(current, offset);
-	}
-	return NULL;
-found:
-	if (size_bytes)
-		*size_bytes = __get_data_size(val) * sizeof(long);
-	return get_data_type_data(current, offset);
+	return fgraph_retrieve_parent_data(idx, size_bytes, 0);
 }
 
 /**
@@ -460,8 +448,56 @@ get_ret_stack(struct task_struct *t, int offset, int *frame_offset)
 	return RET_STACK(t, offset);
 }
 
+/**
+ * fgraph_retrieve_parent_data - get data from a parent function
+ * @idx: The index into the fgraph_array (fgraph_ops::idx)
+ * @size_bytes: A pointer to retrieved data size
+ * @depth: The depth to find the parent (0 is the current function)
+ *
+ * This is similar to fgraph_retrieve_data() but can be used to retrieve
+ * data from a parent caller function.
+ *
+ * Return: a pointer to the specified parent data or NULL if not found
+ */
+void *fgraph_retrieve_parent_data(int idx, int *size_bytes, int depth)
+{
+	struct ftrace_ret_stack *ret_stack = NULL;
+	int offset = current->curr_ret_stack;
+	unsigned long val;
+
+	if (offset <= 0)
+		return NULL;
+
+	for (;;) {
+		int next_offset;
+
+		ret_stack = get_ret_stack(current, offset, &next_offset);
+		if (!ret_stack || --depth < 0)
+			break;
+		offset = next_offset;
+	}
+
+	if (!ret_stack)
+		return NULL;
+
+	offset--;
+
+	val = get_fgraph_entry(current, offset);
+	while (__get_type(val) == FGRAPH_TYPE_DATA) {
+		if (__get_data_index(val) == idx)
+			goto found;
+		offset -= __get_data_size(val) + 1;
+		val = get_fgraph_entry(current, offset);
+	}
+	return NULL;
+found:
+	if (size_bytes)
+		*size_bytes = __get_data_size(val) * sizeof(long);
+	return get_data_type_data(current, offset);
+}
+
 /* Both enabled by default (can be cleared by function_graph tracer flags */
-static bool fgraph_sleep_time = true;
+bool fgraph_sleep_time = true;
 
 #ifdef CONFIG_DYNAMIC_FTRACE
 /*
@@ -524,7 +560,6 @@ ftrace_push_return_trace(unsigned long ret, unsigned long func,
 			 int fgraph_idx)
 {
 	struct ftrace_ret_stack *ret_stack;
-	unsigned long long calltime;
 	unsigned long val;
 	int offset;
 
@@ -554,8 +589,6 @@ ftrace_push_return_trace(unsigned long ret, unsigned long func,
 		return -EBUSY;
 	}
 
-	calltime = trace_clock_local();
-
 	offset = READ_ONCE(current->curr_ret_stack);
 	ret_stack = RET_STACK(current, offset);
 	offset += FGRAPH_FRAME_OFFSET;
@@ -589,7 +622,6 @@ ftrace_push_return_trace(unsigned long ret, unsigned long func,
 
 	ret_stack->ret = ret;
 	ret_stack->func = func;
-	ret_stack->calltime = calltime;
 #ifdef HAVE_FUNCTION_GRAPH_FP_TEST
 	ret_stack->fp = frame_pointer;
 #endif
@@ -723,7 +755,6 @@ ftrace_pop_return_trace(struct ftrace_graph_ret *trace, unsigned long *ret,
 	*offset += FGRAPH_FRAME_OFFSET;
 	*ret = ret_stack->ret;
 	trace->func = ret_stack->func;
-	trace->calltime = ret_stack->calltime;
 	trace->overrun = atomic_read(&current->trace_overrun);
 	trace->depth = current->curr_ret_depth;
 	/*
@@ -867,6 +898,29 @@ ftrace_graph_get_ret_stack(struct task_struct *task, int idx)
 	return ret_stack;
 }
 
+/**
+ * ftrace_graph_top_ret_addr - return the top return address in the shadow stack
+ * @task: The task to read the shadow stack from.
+ *
+ * Return the first return address on the shadow stack of the @task, which is
+ * not the fgraph's return_to_handler.
+ */
+unsigned long ftrace_graph_top_ret_addr(struct task_struct *task)
+{
+	unsigned long return_handler = (unsigned long)dereference_kernel_function_descriptor(return_to_handler);
+	struct ftrace_ret_stack *ret_stack = NULL;
+	int offset = task->curr_ret_stack;
+
+	if (offset < 0)
+		return 0;
+
+	do {
+		ret_stack = get_ret_stack(task, offset, &offset);
+	} while (ret_stack && ret_stack->ret == return_handler);
+
+	return ret_stack ? ret_stack->ret : 0;
+}
+
 /**
  * ftrace_graph_ret_addr - return the original value of the return address
  * @task: The task the unwinder is being executed on
@@ -892,7 +946,7 @@ unsigned long ftrace_graph_ret_addr(struct task_struct *task, int *idx,
 {
 	struct ftrace_ret_stack *ret_stack;
 	unsigned long return_handler = (unsigned long)dereference_kernel_function_descriptor(return_to_handler);
-	int i = task->curr_ret_stack;
+	int i;
 
 	if (ret != return_handler)
 		return ret;
@@ -970,8 +1024,11 @@ static int alloc_retstack_tasklist(unsigned long **ret_stack_list)
 	int start = 0, end = FTRACE_RETSTACK_ALLOC_SIZE;
 	struct task_struct *g, *t;
 
+	if (WARN_ON_ONCE(!fgraph_stack_cachep))
+		return -ENOMEM;
+
 	for (i = 0; i < FTRACE_RETSTACK_ALLOC_SIZE; i++) {
-		ret_stack_list[i] = kmalloc(SHADOW_STACK_SIZE, GFP_KERNEL);
+		ret_stack_list[i] = kmem_cache_alloc(fgraph_stack_cachep, GFP_KERNEL);
 		if (!ret_stack_list[i]) {
 			start = 0;
 			end = i;
@@ -1002,7 +1059,7 @@ static int alloc_retstack_tasklist(unsigned long **ret_stack_list)
 	rcu_read_unlock();
 free:
 	for (i = start; i < end; i++)
-		kfree(ret_stack_list[i]);
+		kmem_cache_free(fgraph_stack_cachep, ret_stack_list[i]);
 	return ret;
 }
 
@@ -1012,9 +1069,7 @@ ftrace_graph_probe_sched_switch(void *ignore, bool preempt,
 				struct task_struct *next,
 				unsigned int prev_state)
 {
-	struct ftrace_ret_stack *ret_stack;
 	unsigned long long timestamp;
-	int offset;
 
 	/*
 	 * Does the user want to count the time a function was asleep.
@@ -1031,17 +1086,7 @@ ftrace_graph_probe_sched_switch(void *ignore, bool preempt,
 	if (!next->ftrace_timestamp)
 		return;
 
-	/*
-	 * Update all the counters in next to make up for the
-	 * time next was sleeping.
-	 */
-	timestamp -= next->ftrace_timestamp;
-
-	for (offset = next->curr_ret_stack; offset > 0; ) {
-		ret_stack = get_ret_stack(next, offset, &offset);
-		if (ret_stack)
-			ret_stack->calltime += timestamp;
-	}
+	next->ftrace_sleeptime += timestamp - next->ftrace_timestamp;
 }
 
 static DEFINE_PER_CPU(unsigned long *, idle_ret_stack);
@@ -1077,9 +1122,12 @@ void ftrace_graph_init_idle_task(struct task_struct *t, int cpu)
 	if (ftrace_graph_active) {
 		unsigned long *ret_stack;
 
+		if (WARN_ON_ONCE(!fgraph_stack_cachep))
+			return;
+
 		ret_stack = per_cpu(idle_ret_stack, cpu);
 		if (!ret_stack) {
-			ret_stack = kmalloc(SHADOW_STACK_SIZE, GFP_KERNEL);
+			ret_stack = kmem_cache_alloc(fgraph_stack_cachep, GFP_KERNEL);
 			if (!ret_stack)
 				return;
 			per_cpu(idle_ret_stack, cpu) = ret_stack;
@@ -1099,7 +1147,10 @@ void ftrace_graph_init_task(struct task_struct *t)
 	if (ftrace_graph_active) {
 		unsigned long *ret_stack;
 
-		ret_stack = kmalloc(SHADOW_STACK_SIZE, GFP_KERNEL);
+		if (WARN_ON_ONCE(!fgraph_stack_cachep))
+			return;
+
+		ret_stack = kmem_cache_alloc(fgraph_stack_cachep, GFP_KERNEL);
 		if (!ret_stack)
 			return;
 		graph_init_task(t, ret_stack);
@@ -1114,7 +1165,11 @@ void ftrace_graph_exit_task(struct task_struct *t)
 	/* NULL must become visible to IRQs before we free it: */
 	barrier();
 
-	kfree(ret_stack);
+	if (ret_stack) {
+		if (WARN_ON_ONCE(!fgraph_stack_cachep))
+			return;
+		kmem_cache_free(fgraph_stack_cachep, ret_stack);
+	}
 }
 
 #ifdef CONFIG_DYNAMIC_FTRACE
@@ -1254,6 +1309,14 @@ int register_ftrace_graph(struct fgraph_ops *gops)
 
 	guard(mutex)(&ftrace_lock);
 
+	if (!fgraph_stack_cachep) {
+		fgraph_stack_cachep = kmem_cache_create("fgraph_stack",
+							SHADOW_STACK_SIZE,
+							SHADOW_STACK_SIZE, 0, NULL);
+		if (!fgraph_stack_cachep)
+			return -ENOMEM;
+	}
+
 	if (!fgraph_initialized) {
 		ret = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "fgraph:online",
 					fgraph_cpu_init, NULL);
@@ -1318,17 +1381,17 @@ void unregister_ftrace_graph(struct fgraph_ops *gops)
 {
 	int command = 0;
 
-	mutex_lock(&ftrace_lock);
+	guard(mutex)(&ftrace_lock);
 
 	if (unlikely(!ftrace_graph_active))
-		goto out;
+		return;
 
 	if (unlikely(gops->idx < 0 || gops->idx >= FGRAPH_ARRAY_SIZE ||
 		     fgraph_array[gops->idx] != gops))
-		goto out;
+		return;
 
 	if (fgraph_lru_release_index(gops->idx) < 0)
-		goto out;
+		return;
 
 	fgraph_array[gops->idx] = &fgraph_stub;
 
@@ -1350,7 +1413,5 @@ void unregister_ftrace_graph(struct fgraph_ops *gops)
 		unregister_pm_notifier(&ftrace_suspend_notifier);
 		unregister_trace_sched_switch(ftrace_graph_probe_sched_switch, NULL);
 	}
- out:
 	gops->saved_func = NULL;
-	mutex_unlock(&ftrace_lock);
 }
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index 4c28dd177ca6..71e53eaba8bc 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -820,10 +820,16 @@ void ftrace_graph_graph_time_control(bool enable)
 	fgraph_graph_time = enable;
 }
 
+struct profile_fgraph_data {
+	unsigned long long		calltime;
+	unsigned long long		subtime;
+	unsigned long long		sleeptime;
+};
+
 static int profile_graph_entry(struct ftrace_graph_ent *trace,
 			       struct fgraph_ops *gops)
 {
-	struct ftrace_ret_stack *ret_stack;
+	struct profile_fgraph_data *profile_data;
 
 	function_profile_call(trace->func, 0, NULL, NULL);
 
@@ -831,9 +837,13 @@ static int profile_graph_entry(struct ftrace_graph_ent *trace,
 	if (!current->ret_stack)
 		return 0;
 
-	ret_stack = ftrace_graph_get_ret_stack(current, 0);
-	if (ret_stack)
-		ret_stack->subtime = 0;
+	profile_data = fgraph_reserve_data(gops->idx, sizeof(*profile_data));
+	if (!profile_data)
+		return 0;
+
+	profile_data->subtime = 0;
+	profile_data->sleeptime = current->ftrace_sleeptime;
+	profile_data->calltime = trace_clock_local();
 
 	return 1;
 }
@@ -841,33 +851,42 @@ static int profile_graph_entry(struct ftrace_graph_ent *trace,
 static void profile_graph_return(struct ftrace_graph_ret *trace,
 				 struct fgraph_ops *gops)
 {
-	struct ftrace_ret_stack *ret_stack;
+	struct profile_fgraph_data *profile_data;
 	struct ftrace_profile_stat *stat;
 	unsigned long long calltime;
+	unsigned long long rettime = trace_clock_local();
 	struct ftrace_profile *rec;
 	unsigned long flags;
+	int size;
 
 	local_irq_save(flags);
 	stat = this_cpu_ptr(&ftrace_profile_stats);
 	if (!stat->hash || !ftrace_profile_enabled)
 		goto out;
 
+	profile_data = fgraph_retrieve_data(gops->idx, &size);
+
 	/* If the calltime was zero'd ignore it */
-	if (!trace->calltime)
+	if (!profile_data || !profile_data->calltime)
 		goto out;
 
-	calltime = trace->rettime - trace->calltime;
+	calltime = rettime - profile_data->calltime;
+
+	if (!fgraph_sleep_time) {
+		if (current->ftrace_sleeptime)
+			calltime -= current->ftrace_sleeptime - profile_data->sleeptime;
+	}
 
 	if (!fgraph_graph_time) {
+		struct profile_fgraph_data *parent_data;
 
 		/* Append this call time to the parent time to subtract */
-		ret_stack = ftrace_graph_get_ret_stack(current, 1);
-		if (ret_stack)
-			ret_stack->subtime += calltime;
+		parent_data = fgraph_retrieve_parent_data(gops->idx, &size, 1);
+		if (parent_data)
+			parent_data->subtime += calltime;
 
-		ret_stack = ftrace_graph_get_ret_stack(current, 0);
-		if (ret_stack && ret_stack->subtime < calltime)
-			calltime -= ret_stack->subtime;
+		if (profile_data->subtime && profile_data->subtime < calltime)
+			calltime -= profile_data->subtime;
 		else
 			calltime = 0;
 	}
@@ -883,6 +902,10 @@ static void profile_graph_return(struct ftrace_graph_ret *trace,
 }
 
 static struct fgraph_ops fprofiler_ops = {
+	.ops = {
+		.flags = FTRACE_OPS_FL_INITIALIZED,
+		INIT_OPS_HASH(fprofiler_ops.ops)
+	},
 	.entryfunc = &profile_graph_entry,
 	.retfunc = &profile_graph_return,
 };
@@ -3663,7 +3686,8 @@ static int ftrace_hash_move_and_update_subops(struct ftrace_ops *subops,
 }
 
 
-static u64		ftrace_update_time;
+u64			ftrace_update_time;
+u64			ftrace_total_mod_time;
 unsigned long		ftrace_update_tot_cnt;
 unsigned long		ftrace_number_of_pages;
 unsigned long		ftrace_number_of_groups;
@@ -3683,7 +3707,7 @@ static int ftrace_update_code(struct module *mod, struct ftrace_page *new_pgs)
 	bool init_nop = ftrace_need_init_nop();
 	struct ftrace_page *pg;
 	struct dyn_ftrace *p;
-	u64 start, stop;
+	u64 start, stop, update_time;
 	unsigned long update_cnt = 0;
 	unsigned long rec_flags = 0;
 	int i;
@@ -3727,7 +3751,11 @@ static int ftrace_update_code(struct module *mod, struct ftrace_page *new_pgs)
 	}
 
 	stop = ftrace_now(raw_smp_processor_id());
-	ftrace_update_time = stop - start;
+	update_time = stop - start;
+	if (mod)
+		ftrace_total_mod_time += update_time;
+	else
+		ftrace_update_time = update_time;
 	ftrace_update_tot_cnt += update_cnt;
 
 	return 0;
@@ -4806,15 +4834,13 @@ match_records(struct ftrace_hash *hash, char *func, int len, char *mod)
 		mod_g.len = strlen(mod_g.search);
 	}
 
-	mutex_lock(&ftrace_lock);
+	guard(mutex)(&ftrace_lock);
 
 	if (unlikely(ftrace_disabled))
-		goto out_unlock;
+		return 0;
 
-	if (func_g.type == MATCH_INDEX) {
-		found = add_rec_by_index(hash, &func_g, clear_filter);
-		goto out_unlock;
-	}
+	if (func_g.type == MATCH_INDEX)
+		return add_rec_by_index(hash, &func_g, clear_filter);
 
 	do_for_each_ftrace_rec(pg, rec) {
 
@@ -4823,16 +4849,12 @@ match_records(struct ftrace_hash *hash, char *func, int len, char *mod)
 
 		if (ftrace_match_record(rec, &func_g, mod_match, exclude_mod)) {
 			ret = enter_record(hash, rec, clear_filter);
-			if (ret < 0) {
-				found = ret;
-				goto out_unlock;
-			}
+			if (ret < 0)
+				return ret;
 			found = 1;
 		}
 		cond_resched();
 	} while_for_each_ftrace_rec();
- out_unlock:
-	mutex_unlock(&ftrace_lock);
 
 	return found;
 }
@@ -4930,14 +4952,14 @@ static int cache_mod(struct trace_array *tr,
 {
 	struct ftrace_mod_load *ftrace_mod, *n;
 	struct list_head *head = enable ? &tr->mod_trace : &tr->mod_notrace;
-	int ret;
 
-	mutex_lock(&ftrace_lock);
+	guard(mutex)(&ftrace_lock);
 
 	/* We do not cache inverse filters */
 	if (func[0] == '!') {
+		int ret = -EINVAL;
+
 		func++;
-		ret = -EINVAL;
 
 		/* Look to remove this hash */
 		list_for_each_entry_safe(ftrace_mod, n, head, list) {
@@ -4953,20 +4975,15 @@ static int cache_mod(struct trace_array *tr,
 				continue;
 			}
 		}
-		goto out;
+		return ret;
 	}
 
-	ret = -EINVAL;
 	/* We only care about modules that have not been loaded yet */
 	if (module_exists(module))
-		goto out;
+		return -EINVAL;
 
 	/* Save this string off, and execute it when the module is loaded */
-	ret = ftrace_add_mod(tr, func, module, enable);
- out:
-	mutex_unlock(&ftrace_lock);
-
-	return ret;
+	return ftrace_add_mod(tr, func, module, enable);
 }
 
 static int
@@ -5276,7 +5293,7 @@ static void release_probe(struct ftrace_func_probe *probe)
 {
 	struct ftrace_probe_ops *probe_ops;
 
-	mutex_lock(&ftrace_lock);
+	guard(mutex)(&ftrace_lock);
 
 	WARN_ON(probe->ref <= 0);
 
@@ -5294,7 +5311,6 @@ static void release_probe(struct ftrace_func_probe *probe)
 		list_del(&probe->list);
 		kfree(probe);
 	}
-	mutex_unlock(&ftrace_lock);
 }
 
 static void acquire_probe_locked(struct ftrace_func_probe *probe)
@@ -6805,12 +6821,10 @@ ftrace_graph_set_hash(struct ftrace_hash *hash, char *buffer)
 
 	func_g.len = strlen(func_g.search);
 
-	mutex_lock(&ftrace_lock);
+	guard(mutex)(&ftrace_lock);
 
-	if (unlikely(ftrace_disabled)) {
-		mutex_unlock(&ftrace_lock);
+	if (unlikely(ftrace_disabled))
 		return -ENODEV;
-	}
 
 	do_for_each_ftrace_rec(pg, rec) {
 
@@ -6826,7 +6840,7 @@ ftrace_graph_set_hash(struct ftrace_hash *hash, char *buffer)
 				if (entry)
 					continue;
 				if (add_hash_entry(hash, rec->ip) == NULL)
-					goto out;
+					return 0;
 			} else {
 				if (entry) {
 					free_hash_entry(hash, entry);
@@ -6835,13 +6849,8 @@ ftrace_graph_set_hash(struct ftrace_hash *hash, char *buffer)
 			}
 		}
 	} while_for_each_ftrace_rec();
-out:
-	mutex_unlock(&ftrace_lock);
 
-	if (fail)
-		return -EINVAL;
-
-	return 0;
+	return fail ? -EINVAL : 0;
 }
 
 static ssize_t
@@ -7920,7 +7929,7 @@ __ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip,
 void arch_ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip,
 			       struct ftrace_ops *op, struct ftrace_regs *fregs)
 {
-	kmsan_unpoison_memory(fregs, sizeof(*fregs));
+	kmsan_unpoison_memory(fregs, ftrace_regs_size());
 	__ftrace_ops_list_func(ip, parent_ip, NULL, fregs);
 }
 #else
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 6a891e00aa7f..103b30b1c73b 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -8587,15 +8587,22 @@ tracing_read_dyn_info(struct file *filp, char __user *ubuf,
 	char *buf;
 	int r;
 
-	/* 256 should be plenty to hold the amount needed */
-	buf = kmalloc(256, GFP_KERNEL);
+	/* 512 should be plenty to hold the amount needed */
+#define DYN_INFO_BUF_SIZE	512
+
+	buf = kmalloc(DYN_INFO_BUF_SIZE, GFP_KERNEL);
 	if (!buf)
 		return -ENOMEM;
 
-	r = scnprintf(buf, 256, "%ld pages:%ld groups: %ld\n",
+	r = scnprintf(buf, DYN_INFO_BUF_SIZE,
+		      "%ld pages:%ld groups: %ld\n"
+		      "ftrace boot update time = %llu (ns)\n"
+		      "ftrace module total update time = %llu (ns)\n",
 		      ftrace_update_tot_cnt,
 		      ftrace_number_of_pages,
-		      ftrace_number_of_groups);
+		      ftrace_number_of_groups,
+		      ftrace_update_time,
+		      ftrace_total_mod_time);
 
 	ret = simple_read_from_buffer(ubuf, cnt, ppos, buf, r);
 	kfree(buf);
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index c866991b9c78..3307dad4d917 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -46,6 +46,7 @@ enum trace_type {
 	TRACE_BRANCH,
 	TRACE_GRAPH_RET,
 	TRACE_GRAPH_ENT,
+	TRACE_GRAPH_RETADDR_ENT,
 	TRACE_USER_STACK,
 	TRACE_BLK,
 	TRACE_BPUTS,
@@ -512,6 +513,8 @@ extern void __ftrace_bad_type(void);
 		IF_ASSIGN(var, ent, struct trace_branch, TRACE_BRANCH); \
 		IF_ASSIGN(var, ent, struct ftrace_graph_ent_entry,	\
 			  TRACE_GRAPH_ENT);		\
+		IF_ASSIGN(var, ent, struct fgraph_retaddr_ent_entry,\
+			  TRACE_GRAPH_RETADDR_ENT);		\
 		IF_ASSIGN(var, ent, struct ftrace_graph_ret_entry,	\
 			  TRACE_GRAPH_RET);		\
 		IF_ASSIGN(var, ent, struct func_repeats_entry,		\
@@ -772,6 +775,8 @@ extern void trace_event_follow_fork(struct trace_array *tr, bool enable);
 extern unsigned long ftrace_update_tot_cnt;
 extern unsigned long ftrace_number_of_pages;
 extern unsigned long ftrace_number_of_groups;
+extern u64 ftrace_update_time;
+extern u64 ftrace_total_mod_time;
 void ftrace_init_trace_array(struct trace_array *tr);
 #else
 static inline void ftrace_init_trace_array(struct trace_array *tr) { }
@@ -879,6 +884,7 @@ static __always_inline bool ftrace_hash_empty(struct ftrace_hash *hash)
 #define TRACE_GRAPH_GRAPH_TIME          0x400
 #define TRACE_GRAPH_PRINT_RETVAL        0x800
 #define TRACE_GRAPH_PRINT_RETVAL_HEX    0x1000
+#define TRACE_GRAPH_PRINT_RETADDR       0x2000
 #define TRACE_GRAPH_PRINT_FILL_SHIFT	28
 #define TRACE_GRAPH_PRINT_FILL_MASK	(0x3 << TRACE_GRAPH_PRINT_FILL_SHIFT)
 
@@ -900,6 +906,10 @@ extern void graph_trace_close(struct trace_iterator *iter);
 extern int __trace_graph_entry(struct trace_array *tr,
 			       struct ftrace_graph_ent *trace,
 			       unsigned int trace_ctx);
+extern int __trace_graph_retaddr_entry(struct trace_array *tr,
+				struct ftrace_graph_ent *trace,
+				unsigned int trace_ctx,
+				unsigned long retaddr);
 extern void __trace_graph_return(struct trace_array *tr,
 				 struct ftrace_graph_ret *trace,
 				 unsigned int trace_ctx);
@@ -1048,6 +1058,7 @@ static inline void ftrace_graph_addr_finish(struct fgraph_ops *gops, struct ftra
 #endif /* CONFIG_DYNAMIC_FTRACE */
 
 extern unsigned int fgraph_max_depth;
+extern bool fgraph_sleep_time;
 
 static inline bool
 ftrace_graph_ignore_func(struct fgraph_ops *gops, struct ftrace_graph_ent *trace)
diff --git a/kernel/trace/trace_entries.h b/kernel/trace/trace_entries.h
index c47422b20908..82fd174ebbe0 100644
--- a/kernel/trace/trace_entries.h
+++ b/kernel/trace/trace_entries.h
@@ -85,9 +85,35 @@ FTRACE_ENTRY_PACKED(funcgraph_entry, ftrace_graph_ent_entry,
 	F_printk("--> %ps (%d)", (void *)__entry->func, __entry->depth)
 );
 
-/* Function return entry */
+#ifdef CONFIG_FUNCTION_GRAPH_RETADDR
+
+/* Function call entry with a return address */
+FTRACE_ENTRY_PACKED(fgraph_retaddr_entry, fgraph_retaddr_ent_entry,
+
+	TRACE_GRAPH_RETADDR_ENT,
+
+	F_STRUCT(
+		__field_struct(	struct fgraph_retaddr_ent,	graph_ent	)
+		__field_packed(	unsigned long,	graph_ent,	func		)
+		__field_packed(	int,		graph_ent,	depth		)
+		__field_packed(	unsigned long,	graph_ent,	retaddr		)
+	),
+
+	F_printk("--> %ps (%d) <- %ps", (void *)__entry->func, __entry->depth,
+		(void *)__entry->retaddr)
+);
+
+#else
+
+#ifndef fgraph_retaddr_ent_entry
+#define fgraph_retaddr_ent_entry ftrace_graph_ent_entry
+#endif
+
+#endif
+
 #ifdef CONFIG_FUNCTION_GRAPH_RETVAL
 
+/* Function return entry */
 FTRACE_ENTRY_PACKED(funcgraph_exit, ftrace_graph_ret_entry,
 
 	TRACE_GRAPH_RET,
@@ -110,6 +136,7 @@ FTRACE_ENTRY_PACKED(funcgraph_exit, ftrace_graph_ret_entry,
 
 #else
 
+/* Function return entry */
 FTRACE_ENTRY_PACKED(funcgraph_exit, ftrace_graph_ret_entry,
 
 	TRACE_GRAPH_RET,
diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c
index a569daaac4c4..03c5a0d300a5 100644
--- a/kernel/trace/trace_functions_graph.c
+++ b/kernel/trace/trace_functions_graph.c
@@ -31,7 +31,10 @@ struct fgraph_data {
 	struct fgraph_cpu_data __percpu *cpu_data;
 
 	/* Place to preserve last processed entry. */
-	struct ftrace_graph_ent_entry	ent;
+	union {
+		struct ftrace_graph_ent_entry	ent;
+		struct fgraph_retaddr_ent_entry	rent;
+	} ent;
 	struct ftrace_graph_ret_entry	ret;
 	int				failed;
 	int				cpu;
@@ -63,6 +66,10 @@ static struct tracer_opt trace_opts[] = {
 	{ TRACER_OPT(funcgraph-retval, TRACE_GRAPH_PRINT_RETVAL) },
 	/* Display function return value in hexadecimal format ? */
 	{ TRACER_OPT(funcgraph-retval-hex, TRACE_GRAPH_PRINT_RETVAL_HEX) },
+#endif
+#ifdef CONFIG_FUNCTION_GRAPH_RETADDR
+	/* Display function return address ? */
+	{ TRACER_OPT(funcgraph-retaddr, TRACE_GRAPH_PRINT_RETADDR) },
 #endif
 	/* Include sleep time (scheduled out) between entry and return */
 	{ TRACER_OPT(sleep-time, TRACE_GRAPH_SLEEP_TIME) },
@@ -83,6 +90,11 @@ static struct tracer_flags tracer_flags = {
 	.opts = trace_opts
 };
 
+static bool tracer_flags_is_set(u32 flags)
+{
+	return (tracer_flags.val & flags) == flags;
+}
+
 /*
  * DURATION column is being also used to display IRQ signs,
  * following values are used by print_graph_irq and others
@@ -119,6 +131,38 @@ int __trace_graph_entry(struct trace_array *tr,
 	return 1;
 }
 
+#ifdef CONFIG_FUNCTION_GRAPH_RETADDR
+int __trace_graph_retaddr_entry(struct trace_array *tr,
+				struct ftrace_graph_ent *trace,
+				unsigned int trace_ctx,
+				unsigned long retaddr)
+{
+	struct ring_buffer_event *event;
+	struct trace_buffer *buffer = tr->array_buffer.buffer;
+	struct fgraph_retaddr_ent_entry *entry;
+
+	event = trace_buffer_lock_reserve(buffer, TRACE_GRAPH_RETADDR_ENT,
+					  sizeof(*entry), trace_ctx);
+	if (!event)
+		return 0;
+	entry	= ring_buffer_event_data(event);
+	entry->graph_ent.func = trace->func;
+	entry->graph_ent.depth = trace->depth;
+	entry->graph_ent.retaddr = retaddr;
+	trace_buffer_unlock_commit_nostack(buffer, event);
+
+	return 1;
+}
+#else
+int __trace_graph_retaddr_entry(struct trace_array *tr,
+				struct ftrace_graph_ent *trace,
+				unsigned int trace_ctx,
+				unsigned long retaddr)
+{
+	return 1;
+}
+#endif
+
 static inline int ftrace_graph_ignore_irqs(void)
 {
 	if (!ftrace_graph_skip_irqs || trace_recursion_test(TRACE_IRQ_BIT))
@@ -127,12 +171,18 @@ static inline int ftrace_graph_ignore_irqs(void)
 	return in_hardirq();
 }
 
+struct fgraph_times {
+	unsigned long long		calltime;
+	unsigned long long		sleeptime; /* may be optional! */
+};
+
 int trace_graph_entry(struct ftrace_graph_ent *trace,
 		      struct fgraph_ops *gops)
 {
 	unsigned long *task_var = fgraph_get_task_var(gops);
 	struct trace_array *tr = gops->private;
 	struct trace_array_cpu *data;
+	struct fgraph_times *ftimes;
 	unsigned long flags;
 	unsigned int trace_ctx;
 	long disabled;
@@ -167,6 +217,19 @@ int trace_graph_entry(struct ftrace_graph_ent *trace,
 	if (ftrace_graph_ignore_irqs())
 		return 0;
 
+	if (fgraph_sleep_time) {
+		/* Only need to record the calltime */
+		ftimes = fgraph_reserve_data(gops->idx, sizeof(ftimes->calltime));
+	} else {
+		ftimes = fgraph_reserve_data(gops->idx, sizeof(*ftimes));
+		if (ftimes)
+			ftimes->sleeptime = current->ftrace_sleeptime;
+	}
+	if (!ftimes)
+		return 0;
+
+	ftimes->calltime = trace_clock_local();
+
 	/*
 	 * Stop here if tracing_threshold is set. We only write function return
 	 * events to the ring buffer.
@@ -180,7 +243,13 @@ int trace_graph_entry(struct ftrace_graph_ent *trace,
 	disabled = atomic_inc_return(&data->disabled);
 	if (likely(disabled == 1)) {
 		trace_ctx = tracing_gen_ctx_flags(flags);
-		ret = __trace_graph_entry(tr, trace, trace_ctx);
+		if (unlikely(IS_ENABLED(CONFIG_FUNCTION_GRAPH_RETADDR) &&
+			tracer_flags_is_set(TRACE_GRAPH_PRINT_RETADDR))) {
+			unsigned long retaddr = ftrace_graph_top_ret_addr(current);
+
+			ret = __trace_graph_retaddr_entry(tr, trace, trace_ctx, retaddr);
+		} else
+			ret = __trace_graph_entry(tr, trace, trace_ctx);
 	} else {
 		ret = 0;
 	}
@@ -238,15 +307,27 @@ void __trace_graph_return(struct trace_array *tr,
 		trace_buffer_unlock_commit_nostack(buffer, event);
 }
 
+static void handle_nosleeptime(struct ftrace_graph_ret *trace,
+			       struct fgraph_times *ftimes,
+			       int size)
+{
+	if (fgraph_sleep_time || size < sizeof(*ftimes))
+		return;
+
+	ftimes->calltime += current->ftrace_sleeptime - ftimes->sleeptime;
+}
+
 void trace_graph_return(struct ftrace_graph_ret *trace,
 			struct fgraph_ops *gops)
 {
 	unsigned long *task_var = fgraph_get_task_var(gops);
 	struct trace_array *tr = gops->private;
 	struct trace_array_cpu *data;
+	struct fgraph_times *ftimes;
 	unsigned long flags;
 	unsigned int trace_ctx;
 	long disabled;
+	int size;
 	int cpu;
 
 	ftrace_graph_addr_finish(gops, trace);
@@ -256,6 +337,14 @@ void trace_graph_return(struct ftrace_graph_ret *trace,
 		return;
 	}
 
+	ftimes = fgraph_retrieve_data(gops->idx, &size);
+	if (!ftimes)
+		return;
+
+	handle_nosleeptime(trace, ftimes, size);
+
+	trace->calltime = ftimes->calltime;
+
 	local_irq_save(flags);
 	cpu = raw_smp_processor_id();
 	data = per_cpu_ptr(tr->array_buffer.data, cpu);
@@ -271,6 +360,9 @@ void trace_graph_return(struct ftrace_graph_ret *trace,
 static void trace_graph_thresh_return(struct ftrace_graph_ret *trace,
 				      struct fgraph_ops *gops)
 {
+	struct fgraph_times *ftimes;
+	int size;
+
 	ftrace_graph_addr_finish(gops, trace);
 
 	if (trace_recursion_test(TRACE_GRAPH_NOTRACE_BIT)) {
@@ -278,8 +370,16 @@ static void trace_graph_thresh_return(struct ftrace_graph_ret *trace,
 		return;
 	}
 
+	ftimes = fgraph_retrieve_data(gops->idx, &size);
+	if (!ftimes)
+		return;
+
+	handle_nosleeptime(trace, ftimes, size);
+
+	trace->calltime = ftimes->calltime;
+
 	if (tracing_thresh &&
-	    (trace->rettime - trace->calltime < tracing_thresh))
+	    (trace->rettime - ftimes->calltime < tracing_thresh))
 		return;
 	else
 		trace_graph_return(trace, gops);
@@ -457,7 +557,7 @@ get_return_for_leaf(struct trace_iterator *iter,
 	 * then we just reuse the data from before.
 	 */
 	if (data && data->failed) {
-		curr = &data->ent;
+		curr = &data->ent.ent;
 		next = &data->ret;
 	} else {
 
@@ -487,7 +587,10 @@ get_return_for_leaf(struct trace_iterator *iter,
 			 * Save current and next entries for later reference
 			 * if the output fails.
 			 */
-			data->ent = *curr;
+			if (unlikely(curr->ent.type == TRACE_GRAPH_RETADDR_ENT))
+				data->ent.rent = *(struct fgraph_retaddr_ent_entry *)curr;
+			else
+				data->ent.ent = *curr;
 			/*
 			 * If the next event is not a return type, then
 			 * we only care about what type it is. Otherwise we can
@@ -651,52 +754,96 @@ print_graph_duration(struct trace_array *tr, unsigned long long duration,
 }
 
 #ifdef CONFIG_FUNCTION_GRAPH_RETVAL
-
 #define __TRACE_GRAPH_PRINT_RETVAL TRACE_GRAPH_PRINT_RETVAL
+#else
+#define __TRACE_GRAPH_PRINT_RETVAL 0
+#endif
 
-static void print_graph_retval(struct trace_seq *s, unsigned long retval,
-				bool leaf, void *func, bool hex_format)
+#ifdef CONFIG_FUNCTION_GRAPH_RETADDR
+#define __TRACE_GRAPH_PRINT_RETADDR TRACE_GRAPH_PRINT_RETADDR
+static void print_graph_retaddr(struct trace_seq *s, struct fgraph_retaddr_ent_entry *entry,
+				u32 trace_flags, bool comment)
+{
+	if (comment)
+		trace_seq_puts(s, " /*");
+
+	trace_seq_puts(s, " <-");
+	seq_print_ip_sym(s, entry->graph_ent.retaddr, trace_flags | TRACE_ITER_SYM_OFFSET);
+
+	if (comment)
+		trace_seq_puts(s, " */");
+}
+#else
+#define __TRACE_GRAPH_PRINT_RETADDR 0
+#define print_graph_retaddr(_seq, _entry, _tflags, _comment)		do { } while (0)
+#endif
+
+#if defined(CONFIG_FUNCTION_GRAPH_RETVAL) || defined(CONFIG_FUNCTION_GRAPH_RETADDR)
+
+static void print_graph_retval(struct trace_seq *s, struct ftrace_graph_ent_entry *entry,
+				struct ftrace_graph_ret *graph_ret, void *func,
+				u32 opt_flags, u32 trace_flags)
 {
 	unsigned long err_code = 0;
+	unsigned long retval = 0;
+	bool print_retaddr = false;
+	bool print_retval = false;
+	bool hex_format = !!(opt_flags & TRACE_GRAPH_PRINT_RETVAL_HEX);
 
-	if (retval == 0 || hex_format)
-		goto done;
+#ifdef CONFIG_FUNCTION_GRAPH_RETVAL
+	retval = graph_ret->retval;
+	print_retval = !!(opt_flags & TRACE_GRAPH_PRINT_RETVAL);
+#endif
 
-	/* Check if the return value matches the negative format */
-	if (IS_ENABLED(CONFIG_64BIT) && (retval & BIT(31)) &&
-		(((u64)retval) >> 32) == 0) {
-		/* sign extension */
-		err_code = (unsigned long)(s32)retval;
-	} else {
-		err_code = retval;
+#ifdef CONFIG_FUNCTION_GRAPH_RETADDR
+	print_retaddr = !!(opt_flags & TRACE_GRAPH_PRINT_RETADDR);
+#endif
+
+	if (print_retval && retval && !hex_format) {
+		/* Check if the return value matches the negative format */
+		if (IS_ENABLED(CONFIG_64BIT) && (retval & BIT(31)) &&
+			(((u64)retval) >> 32) == 0) {
+			err_code = sign_extend64(retval, 31);
+		} else {
+			err_code = retval;
+		}
+
+		if (!IS_ERR_VALUE(err_code))
+			err_code = 0;
 	}
 
-	if (!IS_ERR_VALUE(err_code))
-		err_code = 0;
+	if (entry) {
+		if (entry->ent.type != TRACE_GRAPH_RETADDR_ENT)
+			print_retaddr = false;
 
-done:
-	if (leaf) {
-		if (hex_format || (err_code == 0))
-			trace_seq_printf(s, "%ps(); /* = 0x%lx */\n",
-					func, retval);
+		trace_seq_printf(s, "%ps();", func);
+		if (print_retval || print_retaddr)
+			trace_seq_puts(s, " /*");
 		else
-			trace_seq_printf(s, "%ps(); /* = %ld */\n",
-					func, err_code);
+			trace_seq_putc(s, '\n');
 	} else {
-		if (hex_format || (err_code == 0))
-			trace_seq_printf(s, "} /* %ps = 0x%lx */\n",
-					func, retval);
-		else
-			trace_seq_printf(s, "} /* %ps = %ld */\n",
-					func, err_code);
+		print_retaddr = false;
+		trace_seq_printf(s, "} /* %ps", func);
 	}
+
+	if (print_retaddr)
+		print_graph_retaddr(s, (struct fgraph_retaddr_ent_entry *)entry,
+				    trace_flags, false);
+
+	if (print_retval) {
+		if (hex_format || (err_code == 0))
+			trace_seq_printf(s, " ret=0x%lx", retval);
+		else
+			trace_seq_printf(s, " ret=%ld", err_code);
+	}
+
+	if (!entry || print_retval || print_retaddr)
+		trace_seq_puts(s, " */\n");
 }
 
 #else
 
-#define __TRACE_GRAPH_PRINT_RETVAL 0
-
-#define print_graph_retval(_seq, _retval, _leaf, _func, _format) do {} while (0)
+#define print_graph_retval(_seq, _ent, _ret, _func, _opt_flags, _trace_flags) do {} while (0)
 
 #endif
 
@@ -748,14 +895,15 @@ print_graph_entry_leaf(struct trace_iterator *iter,
 		trace_seq_putc(s, ' ');
 
 	/*
-	 * Write out the function return value if the option function-retval is
-	 * enabled.
+	 * Write out the function return value or return address
 	 */
-	if (flags & __TRACE_GRAPH_PRINT_RETVAL)
-		print_graph_retval(s, graph_ret->retval, true, (void *)func,
-				!!(flags & TRACE_GRAPH_PRINT_RETVAL_HEX));
-	else
+	if (flags & (__TRACE_GRAPH_PRINT_RETVAL | __TRACE_GRAPH_PRINT_RETADDR)) {
+		print_graph_retval(s, entry, graph_ret,
+				   (void *)graph_ret->func + iter->tr->text_delta,
+				   flags, tr->trace_flags);
+	} else {
 		trace_seq_printf(s, "%ps();\n", (void *)func);
+	}
 
 	print_graph_irq(iter, graph_ret->func, TRACE_GRAPH_RET,
 			cpu, iter->ent->pid, flags);
@@ -796,7 +944,12 @@ print_graph_entry_nested(struct trace_iterator *iter,
 
 	func = call->func + iter->tr->text_delta;
 
-	trace_seq_printf(s, "%ps() {\n", (void *)func);
+	trace_seq_printf(s, "%ps() {", (void *)func);
+	if (flags & __TRACE_GRAPH_PRINT_RETADDR  &&
+		entry->ent.type == TRACE_GRAPH_RETADDR_ENT)
+		print_graph_retaddr(s, (struct fgraph_retaddr_ent_entry *)entry,
+			tr->trace_flags, true);
+	trace_seq_putc(s, '\n');
 
 	if (trace_seq_has_overflowed(s))
 		return TRACE_TYPE_PARTIAL_LINE;
@@ -1043,11 +1196,10 @@ print_graph_return(struct ftrace_graph_ret *trace, struct trace_seq *s,
 
 	/*
 	 * Always write out the function name and its return value if the
-	 * function-retval option is enabled.
+	 * funcgraph-retval option is enabled.
 	 */
 	if (flags & __TRACE_GRAPH_PRINT_RETVAL) {
-		print_graph_retval(s, trace->retval, false, (void *)func,
-			!!(flags & TRACE_GRAPH_PRINT_RETVAL_HEX));
+		print_graph_retval(s, NULL, trace, (void *)func, flags, tr->trace_flags);
 	} else {
 		/*
 		 * If the return function does not have a matching entry,
@@ -1162,7 +1314,7 @@ print_graph_function_flags(struct trace_iterator *iter, u32 flags)
 	 * to print out the missing entry which would never go out.
 	 */
 	if (data && data->failed) {
-		field = &data->ent;
+		field = &data->ent.ent;
 		iter->cpu = data->cpu;
 		ret = print_graph_entry(field, s, iter, flags);
 		if (ret == TRACE_TYPE_HANDLED && iter->cpu != cpu) {
@@ -1186,6 +1338,16 @@ print_graph_function_flags(struct trace_iterator *iter, u32 flags)
 		saved = *field;
 		return print_graph_entry(&saved, s, iter, flags);
 	}
+#ifdef CONFIG_FUNCTION_GRAPH_RETADDR
+	case TRACE_GRAPH_RETADDR_ENT: {
+		struct fgraph_retaddr_ent_entry saved;
+		struct fgraph_retaddr_ent_entry *rfield;
+
+		trace_assign_type(rfield, entry);
+		saved = *rfield;
+		return print_graph_entry((struct ftrace_graph_ent_entry *)&saved, s, iter, flags);
+	}
+#endif
 	case TRACE_GRAPH_RET: {
 		struct ftrace_graph_ret_entry *field;
 		trace_assign_type(field, entry);
@@ -1380,6 +1542,13 @@ static struct trace_event graph_trace_entry_event = {
 	.funcs		= &graph_functions,
 };
 
+#ifdef CONFIG_FUNCTION_GRAPH_RETADDR
+static struct trace_event graph_trace_retaddr_entry_event = {
+	.type		= TRACE_GRAPH_RETADDR_ENT,
+	.funcs		= &graph_functions,
+};
+#endif
+
 static struct trace_event graph_trace_ret_event = {
 	.type		= TRACE_GRAPH_RET,
 	.funcs		= &graph_functions
@@ -1466,6 +1635,13 @@ static __init int init_graph_trace(void)
 		return 1;
 	}
 
+#ifdef CONFIG_FUNCTION_GRAPH_RETADDR
+	if (!register_trace_event(&graph_trace_retaddr_entry_event)) {
+		pr_warn("Warning: could not register graph trace retaddr events\n");
+		return 1;
+	}
+#endif
+
 	if (!register_trace_event(&graph_trace_ret_event)) {
 		pr_warn("Warning: could not register graph trace events\n");
 		return 1;
diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c
index 1469dd8075fa..38b5754790c9 100644
--- a/kernel/trace/trace_selftest.c
+++ b/kernel/trace/trace_selftest.c
@@ -17,6 +17,7 @@ static inline int trace_valid_entry(struct trace_entry *entry)
 	case TRACE_PRINT:
 	case TRACE_BRANCH:
 	case TRACE_GRAPH_ENT:
+	case TRACE_GRAPH_RETADDR_ENT:
 	case TRACE_GRAPH_RET:
 		return 1;
 	}
diff --git a/tools/testing/selftests/ftrace/test.d/ftrace/fgraph-retval.tc b/tools/testing/selftests/ftrace/test.d/ftrace/fgraph-retval.tc
index e34c0bdef3ed..4307d4eef417 100644
--- a/tools/testing/selftests/ftrace/test.d/ftrace/fgraph-retval.tc
+++ b/tools/testing/selftests/ftrace/test.d/ftrace/fgraph-retval.tc
@@ -29,7 +29,7 @@ set -e
 
 : "Test printing the error code in signed decimal format"
 echo 0 > options/funcgraph-retval-hex
-count=`cat trace | grep 'proc_reg_write' | grep '= -5' | wc -l`
+count=`cat trace | grep 'proc_reg_write' | grep -e '=-5 ' -e '= -5 '  | wc -l`
 if [ $count -eq 0 ]; then
     fail "Return value can not be printed in signed decimal format"
 fi