[PATCH RFC 11/28] ARM: entry: Merge the common and trace entry code

[Linus Walleij]

The trace entry code now can handle recursive and complex
calls in C using stack.

Move the common code over to using that approach.

We now use the ret_fast_syscall return path also when tracing,
which appears to work just fine.

Signed-off-by: Linus Walleij <linus.walleij@xxxxxxxxxx>
---
 arch/arm/include/asm/syscall.h |  3 +--
 arch/arm/kernel/entry-common.S | 61 +++++++++---------------------------------
 arch/arm/kernel/syscall.c      | 34 +++++++++++------------
 3 files changed, 28 insertions(+), 70 deletions(-)

diff --git a/arch/arm/include/asm/syscall.h b/arch/arm/include/asm/syscall.h
index cb0073c4151b..9c664d8c5718 100644
--- a/arch/arm/include/asm/syscall.h
+++ b/arch/arm/include/asm/syscall.h
@@ -19,8 +19,7 @@
 
 extern const unsigned long sys_call_table[];
 
-int invoke_syscall(void *table, struct pt_regs *regs, int scno, void *retp);
-void invoke_syscall_trace(void *table, struct pt_regs *regs);
+int invoke_syscall_trace(void *table, struct pt_regs *regs, int scno);
 
 static inline int syscall_get_nr(struct task_struct *task,
 				 struct pt_regs *regs)
diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S
index dbc947d301ec..f0f1f8723965 100644
--- a/arch/arm/kernel/entry-common.S
+++ b/arch/arm/kernel/entry-common.S
@@ -39,7 +39,6 @@ saved_pc	.req	lr
  * from those features make this path too inefficient.
  */
 ret_fast_syscall:
-__ret_fast_syscall:
  UNWIND(.fnstart	)
  UNWIND(.cantunwind	)
 	disable_irq_notrace			@ disable interrupts
@@ -47,13 +46,13 @@ __ret_fast_syscall:
 	movs	r1, r1, lsl #16
 	bne	fast_work_pending
 
-	restore_user_regs fast = 1, offset = S_OFF
+	restore_user_regs fast = 0, offset = S_OFF
  UNWIND(.fnend		)
 ENDPROC(ret_fast_syscall)
 
 	/* Ok, we need to do extra processing, enter the slow path. */
 fast_work_pending:
-	str	r0, [sp, #S_R0+S_OFF]!		@ returned r0
+	add	sp, sp, #(S_R0 + S_OFF)
 	/* fall through to work_pending */
 #else
 /*
@@ -63,10 +62,9 @@ fast_work_pending:
  * call.
  */
 ret_fast_syscall:
-__ret_fast_syscall:
  UNWIND(.fnstart	)
  UNWIND(.cantunwind	)
-	str	r0, [sp, #S_R0 + S_OFF]!	@ save returned r0
+	add	sp, sp, #(S_R0 + S_OFF)
 #if IS_ENABLED(CONFIG_DEBUG_RSEQ)
 	/* do_rseq_syscall needs interrupts enabled. */
 	mov	r0, sp				@ 'regs'
@@ -83,7 +81,9 @@ ENDPROC(ret_fast_syscall)
 #endif
 
 	tst	r1, #_TIF_SYSCALL_WORK
-	bne	__sys_trace_return_nosave
+	beq	slow_work_pending
+	b	ret_to_user
+
 slow_work_pending:
 	mov	r0, sp				@ 'regs'
 	bl	do_work_pending
@@ -257,19 +257,15 @@ ENTRY(vector_swi)
 	str	scno, [tsk, #TI_ABI_SYSCALL]
 #endif
 	mov	r1, sp				@ put regs into r1
-	ldr	r10, [tsk, #TI_FLAGS]		@ check for syscall tracing
 	stmdb	sp!, {r4, r5}			@ push fifth and sixth args
-
-	tst	r10, #_TIF_SYSCALL_WORK		@ are we tracing syscalls?
-	bne	__sys_trace
-
 	mov	r0, tbl
-	/* r1 already contains regs */
 	mov	r2, scno			@ syscall number from r7
-	/* We return here no matter what, also pass this as an argument */
-	badr	lr, __ret_fast_syscall
-	mov	r3, lr
-	b 	invoke_syscall
+	bl	invoke_syscall_trace
+	cmp	r0, #0
+	beq	ret_fast_syscall
+	/* This path taken when tracing */
+	add	sp, sp, #(S_R0 + S_OFF)
+	b	ret_to_user
 
 #if defined(CONFIG_OABI_COMPAT) || !defined(CONFIG_AEABI)
 	/*
@@ -289,20 +285,6 @@ ENTRY(vector_swi)
 ENDPROC(vector_swi)
 	.ltorg
 
-	/*
-	 * This is the really slow path.  We're going to be doing
-	 * context switches, and waiting for our parent to respond.
-	 */
-__sys_trace:
-	add	r1, sp, #S_R0 + S_OFF		@ pointer to regs
-	mov	r0, tbl
-	bl	invoke_syscall_trace
-	add	sp, sp, #S_OFF			@ restore stack pointer
-	b	ret_to_user
-
-__sys_trace_return_nosave:
-	b	ret_to_user
-
 	.macro	syscall_table_start, sym
 	.equ	__sys_nr, 0
 	.type	\sym, #object
@@ -402,25 +384,6 @@ sys_mmap2:
 		b	sys_mmap_pgoff
 ENDPROC(sys_mmap2)
 
-/*
- * This call wants:
- * r0: syscall table
- * r1: regs
- * r2: syscall number
- * r3: pointer to return function
- */
-SYM_TYPED_FUNC_START(invoke_syscall_asm)
-#ifdef CONFIG_CPU_SPECTRE
-	csdb
-#endif
-	mov	tbl, r0
-	mov	scno, r2
-	mov	lr, r3				@ return address
-	ldmia 	r1, {r0 - r3}			@ reload r0-r3
-	/* Arguments 5 and 6 are (hopefully) on the stack */
-	ldr	pc, [tbl, scno, lsl #2]		@ call sys_* routine
-SYM_FUNC_END(invoke_syscall_asm)
-
 /*
  * This call wants:
  * r0: syscall table
diff --git a/arch/arm/kernel/syscall.c b/arch/arm/kernel/syscall.c
index 3ee367958298..ab9e66da0a80 100644
--- a/arch/arm/kernel/syscall.c
+++ b/arch/arm/kernel/syscall.c
@@ -3,48 +3,44 @@
 #include <linux/syscalls.h>
 #include <asm/syscall.h>
 
-int invoke_syscall_asm(void *table, struct pt_regs *regs, int scno, void *retp);
-
-__visible int invoke_syscall(void *table, struct pt_regs *regs, int scno, void *retp)
+static inline bool has_syscall_work(unsigned long flags)
 {
-	if (scno < NR_syscalls)
-		/* Doing this with return makes sure the stack gets pop:ed */
-		return invoke_syscall_asm(table, regs, scno, retp);
-
-	if (scno >= __ARM_NR_BASE)
-		return arm_syscall(scno, regs);
-
-	return sys_ni_syscall();
+	return unlikely(flags & _TIF_SYSCALL_WORK);
 }
 
 int invoke_syscall_trace_asm(void *table, struct pt_regs *regs, int scno);
 
-__visible void invoke_syscall_trace(void *table, struct pt_regs *regs)
+__visible int invoke_syscall_trace(void *table, struct pt_regs *regs, int scno)
 {
-	int scno;
+	unsigned long flags = read_thread_flags();
 	int ret;
 
-	scno = syscall_trace_enter(regs);
-	if (scno == -1)
-		goto trace_exit_nosave;
+	if (has_syscall_work(flags)) {
+		scno = syscall_trace_enter(regs);
+		if (scno == -1)
+			goto trace_exit_nosave;
+	}
 
 	if (scno < NR_syscalls) {
 		ret = invoke_syscall_trace_asm(table, regs, scno);
-		goto trace_exit_save;
+		goto exit_save;
 	}
 
 	if (scno >= __ARM_NR_BASE) {
 		ret = arm_syscall(scno, regs);
-		goto trace_exit_save;
+		goto exit_save;
 	}
 
 	ret = sys_ni_syscall();
 
-trace_exit_save:
+exit_save:
 	/* Save return value from syscall */
 	regs->ARM_r0 = ret;
+	if (!has_syscall_work(flags))
+		return 0;
 
 trace_exit_nosave:
 	local_irq_enable();
 	syscall_trace_exit(regs);
+	return 1;
 }

-- 
2.46.2