x86/entry: Fix the end of the stack for newly forked tasks

When unwinding a task, the end of the stack is always at the same offset
right below the saved pt_regs, regardless of which syscall was used to
enter the kernel.  That convention allows the unwinder to verify that a
stack is sane.

However, newly forked tasks don't always follow that convention, as
reported by the following unwinder warning seen by Dave Jones:

  WARNING: kernel stack frame pointer at ffffc90001443f30 in kworker/u8:8:30468 has bad value           (null)

The warning was due to the following call chain:

  (ftrace handler)
  call_usermodehelper_exec_async+0x5/0x140
  ret_from_fork+0x22/0x30

The problem is that ret_from_fork() doesn't create a stack frame before
calling other functions.  Fix that by carefully using the frame pointer
macros.

In addition to conforming to the end of stack convention, this also
makes related stack traces more sensible by making it clear to the user
that ret_from_fork() was involved.

Reported-by: Dave Jones <davej@codemonkey.org.uk>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Miroslav Benes <mbenes@suse.cz>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/8854cdaab980e9700a81e9ebf0d4238e4bbb68ef.1483978430.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>

2 files changed, 18 insertions, 23 deletions

diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S
index 701d29f8e4d3..57f7ec35216e 100644
--- a/arch/x86/entry/entry_32.S
+++ b/arch/x86/entry/entry_32.S
@@ -255,23 +255,6 @@ ENTRY(__switch_to_asm)
 END(__switch_to_asm)
 
 /*
- * The unwinder expects the last frame on the stack to always be at the same
- * offset from the end of the page, which allows it to validate the stack.
- * Calling schedule_tail() directly would break that convention because its an
- * asmlinkage function so its argument has to be pushed on the stack.  This
- * wrapper creates a proper "end of stack" frame header before the call.
- */
-ENTRY(schedule_tail_wrapper)
-	FRAME_BEGIN
-
-	pushl	%eax
-	call	schedule_tail
-	popl	%eax
-
-	FRAME_END
-	ret
-ENDPROC(schedule_tail_wrapper)
-/*
  * A newly forked process directly context switches into this address.
  *
  * eax: prev task we switched from
@@ -279,15 +262,24 @@ ENDPROC(schedule_tail_wrapper)
  * edi: kernel thread arg
  */
 ENTRY(ret_from_fork)
-	call	schedule_tail_wrapper
+	FRAME_BEGIN		/* help unwinder find end of stack */
+
+	/*
+	 * schedule_tail() is asmlinkage so we have to put its 'prev' argument
+	 * on the stack.
+	 */
+	pushl	%eax
+	call	schedule_tail
+	popl	%eax
 
 	testl	%ebx, %ebx
 	jnz	1f		/* kernel threads are uncommon */
 
 2:
 	/* When we fork, we trace the syscall return in the child, too. */
-	movl    %esp, %eax
+	leal	FRAME_OFFSET(%esp), %eax
 	call    syscall_return_slowpath
+	FRAME_END
 	jmp     restore_all
 
 	/* kernel thread */
diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
index 5b219707c2f2..044d18ebc43c 100644
--- a/arch/x86/entry/entry_64.S
+++ b/arch/x86/entry/entry_64.S
@@ -36,6 +36,7 @@
 #include <asm/smap.h>
 #include <asm/pgtable_types.h>
 #include <asm/export.h>
+#include <asm/frame.h>
 #include <linux/err.h>
 
 .code64
@@ -408,17 +409,19 @@ END(__switch_to_asm)
  * r12: kernel thread arg
  */
 ENTRY(ret_from_fork)
+	FRAME_BEGIN			/* help unwinder find end of stack */
 	movq	%rax, %rdi
-	call	schedule_tail			/* rdi: 'prev' task parameter */
+	call	schedule_tail		/* rdi: 'prev' task parameter */
 
-	testq	%rbx, %rbx			/* from kernel_thread? */
-	jnz	1f				/* kernel threads are uncommon */
+	testq	%rbx, %rbx		/* from kernel_thread? */
+	jnz	1f			/* kernel threads are uncommon */
 
 2:
-	movq	%rsp, %rdi
+	leaq	FRAME_OFFSET(%rsp),%rdi	/* pt_regs pointer */
 	call	syscall_return_slowpath	/* returns with IRQs disabled */
 	TRACE_IRQS_ON			/* user mode is traced as IRQS on */
 	SWAPGS
+	FRAME_END
 	jmp	restore_regs_and_iret
 
 1: