[PATCH-4.19-stable 0/2] Backport ENCODE_FRAME_POINTER hint

[PATCH-4.19-stable 0/2] Backport ENCODE_FRAME_POINTER hint

[Dmitry Safonov]

On 4.19.93 the following warning was observed with CONFIG_FRAME_POINTER:
> WARNING: kernel stack frame pointer at 00000000bceb5183 in Coronavirus:3282 has bad value           (null)
>  unwind stack type:0 next_sp:          (null) mask:0x2 graph_idx:0
>  000000009630aa47: ffffc9000126fdb0 (0xffffc9000126fdb0)
>  0000000020360f53: ffffffff81038e33 (__save_stack_trace+0xcb/0xee)
>  00000000675081f2: 0000000000000000 ...
>  0000000043198fe7: ffffc9000126c000 (0xffffc9000126c000)
>  0000000008a46231: ffffc90001270000 (0xffffc90001270000)
[..]

It turns to be missing %rbp hint was making frame pointer unwinder
a bit tipsy.
The observed is WARN_ONCE(), so it one time per boot, but imho, worth to
have in stable too.

Peter Zijlstra (2):
  x86/stackframe: Move ENCODE_FRAME_POINTER to asm/frame.h
  x86/stackframe, x86/ftrace: Add pt_regs frame annotations

 arch/x86/entry/calling.h     | 15 -----------
 arch/x86/entry/entry_32.S    | 16 ------------
 arch/x86/include/asm/frame.h | 49 ++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/ftrace_32.S  |  3 +++
 arch/x86/kernel/ftrace_64.S  |  3 +++
 5 files changed, 55 insertions(+), 31 deletions(-)

-- 
2.25.0

[PATCH-4.19-stable 1/2] x86/stackframe: Move ENCODE_FRAME_POINTER to asm/frame.h

[Dmitry Safonov]

From: Peter Zijlstra <peterz@infradead.org>

[ Upstream commit a9b3c6998d4a7d53a787cf4d0fd4a4c11239e517 ]

In preparation for wider use, move the ENCODE_FRAME_POINTER macros to
a common header and provide inline asm versions.

These macros are used to encode a pt_regs frame for the unwinder; see
unwind_frame.c:decode_frame_pointer().

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Dmitry Safonov <dima@arista.com>
---
 arch/x86/entry/calling.h     | 15 -----------
 arch/x86/entry/entry_32.S    | 16 ------------
 arch/x86/include/asm/frame.h | 49 ++++++++++++++++++++++++++++++++++++
 3 files changed, 49 insertions(+), 31 deletions(-)

diff --git a/arch/x86/entry/calling.h b/arch/x86/entry/calling.h
index 578b5455334f..31fbb4a7d9f6 100644
--- a/arch/x86/entry/calling.h
+++ b/arch/x86/entry/calling.h
@@ -172,21 +172,6 @@ For 32-bit we have the following conventions - kernel is built with
 	.endif
 .endm
 
-/*
- * This is a sneaky trick to help the unwinder find pt_regs on the stack.  The
- * frame pointer is replaced with an encoded pointer to pt_regs.  The encoding
- * is just setting the LSB, which makes it an invalid stack address and is also
- * a signal to the unwinder that it's a pt_regs pointer in disguise.
- *
- * NOTE: This macro must be used *after* PUSH_AND_CLEAR_REGS because it corrupts
- * the original rbp.
- */
-.macro ENCODE_FRAME_POINTER ptregs_offset=0
-#ifdef CONFIG_FRAME_POINTER
-	leaq 1+\ptregs_offset(%rsp), %rbp
-#endif
-.endm
-
 #ifdef CONFIG_PAGE_TABLE_ISOLATION
 
 /*
diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S
index 8059d4fd915c..d07432062ee6 100644
--- a/arch/x86/entry/entry_32.S
+++ b/arch/x86/entry/entry_32.S
@@ -245,22 +245,6 @@
 .Lend_\@:
 .endm
 
-/*
- * This is a sneaky trick to help the unwinder find pt_regs on the stack.  The
- * frame pointer is replaced with an encoded pointer to pt_regs.  The encoding
- * is just clearing the MSB, which makes it an invalid stack address and is also
- * a signal to the unwinder that it's a pt_regs pointer in disguise.
- *
- * NOTE: This macro must be used *after* SAVE_ALL because it corrupts the
- * original rbp.
- */
-.macro ENCODE_FRAME_POINTER
-#ifdef CONFIG_FRAME_POINTER
-	mov %esp, %ebp
-	andl $0x7fffffff, %ebp
-#endif
-.endm
-
 .macro RESTORE_INT_REGS
 	popl	%ebx
 	popl	%ecx
diff --git a/arch/x86/include/asm/frame.h b/arch/x86/include/asm/frame.h
index 5cbce6fbb534..296b346184b2 100644
--- a/arch/x86/include/asm/frame.h
+++ b/arch/x86/include/asm/frame.h
@@ -22,6 +22,35 @@
 	pop %_ASM_BP
 .endm
 
+#ifdef CONFIG_X86_64
+/*
+ * This is a sneaky trick to help the unwinder find pt_regs on the stack.  The
+ * frame pointer is replaced with an encoded pointer to pt_regs.  The encoding
+ * is just setting the LSB, which makes it an invalid stack address and is also
+ * a signal to the unwinder that it's a pt_regs pointer in disguise.
+ *
+ * NOTE: This macro must be used *after* PUSH_AND_CLEAR_REGS because it corrupts
+ * the original rbp.
+ */
+.macro ENCODE_FRAME_POINTER ptregs_offset=0
+	leaq 1+\ptregs_offset(%rsp), %rbp
+.endm
+#else /* !CONFIG_X86_64 */
+/*
+ * This is a sneaky trick to help the unwinder find pt_regs on the stack.  The
+ * frame pointer is replaced with an encoded pointer to pt_regs.  The encoding
+ * is just clearing the MSB, which makes it an invalid stack address and is also
+ * a signal to the unwinder that it's a pt_regs pointer in disguise.
+ *
+ * NOTE: This macro must be used *after* SAVE_ALL because it corrupts the
+ * original ebp.
+ */
+.macro ENCODE_FRAME_POINTER
+	mov %esp, %ebp
+	andl $0x7fffffff, %ebp
+.endm
+#endif /* CONFIG_X86_64 */
+
 #else /* !__ASSEMBLY__ */
 
 #define FRAME_BEGIN				\
@@ -30,12 +59,32 @@
 
 #define FRAME_END "pop %" _ASM_BP "\n"
 
+#ifdef CONFIG_X86_64
+#define ENCODE_FRAME_POINTER			\
+	"lea 1(%rsp), %rbp\n\t"
+#else /* !CONFIG_X86_64 */
+#define ENCODE_FRAME_POINTER			\
+	"movl %esp, %ebp\n\t"			\
+	"andl $0x7fffffff, %ebp\n\t"
+#endif /* CONFIG_X86_64 */
+
 #endif /* __ASSEMBLY__ */
 
 #define FRAME_OFFSET __ASM_SEL(4, 8)
 
 #else /* !CONFIG_FRAME_POINTER */
 
+#ifdef __ASSEMBLY__
+
+.macro ENCODE_FRAME_POINTER ptregs_offset=0
+.endm
+
+#else /* !__ASSEMBLY */
+
+#define ENCODE_FRAME_POINTER
+
+#endif
+
 #define FRAME_BEGIN
 #define FRAME_END
 #define FRAME_OFFSET 0
-- 
2.25.0

[PATCH-4.19-stable 2/2] x86/stackframe, x86/ftrace: Add pt_regs frame annotations

[Dmitry Safonov]

From: Peter Zijlstra <peterz@infradead.org>

[ Upstream commit ea1ed38dba64b64a245ab8ca1406269d17b99485 ]

When CONFIG_FRAME_POINTER, we should mark pt_regs frames.

Fixes user-visible warning for unwinder (i.e, ftrace's stack tracer):
> WARNING: kernel stack frame pointer at 00000000bceb5183 in Coronavirus:3282 has bad value           (null)
>  unwind stack type:0 next_sp:          (null) mask:0x2 graph_idx:0
>  000000009630aa47: ffffc9000126fdb0 (0xffffc9000126fdb0)
>  0000000020360f53: ffffffff81038e33 (__save_stack_trace+0xcb/0xee)
>  00000000675081f2: 0000000000000000 ...
>  0000000043198fe7: ffffc9000126c000 (0xffffc9000126c000)
>  0000000008a46231: ffffc90001270000 (0xffffc90001270000)
[..]

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
[4.19 backport; added user-visible changelog]
Signed-off-by: Dmitry Safonov <dima@arista.com>
---
 arch/x86/kernel/ftrace_32.S | 3 +++
 arch/x86/kernel/ftrace_64.S | 3 +++
 2 files changed, 6 insertions(+)

diff --git a/arch/x86/kernel/ftrace_32.S b/arch/x86/kernel/ftrace_32.S
index 4c8440de3355..83f18e829ac7 100644
--- a/arch/x86/kernel/ftrace_32.S
+++ b/arch/x86/kernel/ftrace_32.S
@@ -9,6 +9,7 @@
 #include <asm/export.h>
 #include <asm/ftrace.h>
 #include <asm/nospec-branch.h>
+#include <asm/frame.h>
 
 #ifdef CC_USING_FENTRY
 # define function_hook	__fentry__
@@ -131,6 +132,8 @@ ENTRY(ftrace_regs_caller)
 	pushl	%ecx
 	pushl	%ebx
 
+	ENCODE_FRAME_POINTER
+
 	movl	12*4(%esp), %eax		/* Load ip (1st parameter) */
 	subl	$MCOUNT_INSN_SIZE, %eax		/* Adjust ip */
 #ifdef CC_USING_FENTRY
diff --git a/arch/x86/kernel/ftrace_64.S b/arch/x86/kernel/ftrace_64.S
index 75f2b36b41a6..24b9abf718e8 100644
--- a/arch/x86/kernel/ftrace_64.S
+++ b/arch/x86/kernel/ftrace_64.S
@@ -9,6 +9,7 @@
 #include <asm/export.h>
 #include <asm/nospec-branch.h>
 #include <asm/unwind_hints.h>
+#include <asm/frame.h>
 
 	.code64
 	.section .entry.text, "ax"
@@ -222,6 +223,8 @@ GLOBAL(ftrace_regs_caller_op_ptr)
 	leaq MCOUNT_REG_SIZE+8*2(%rsp), %rcx
 	movq %rcx, RSP(%rsp)
 
+	ENCODE_FRAME_POINTER
+
 	/* regs go into 4th parameter */
 	leaq (%rsp), %rcx
 
-- 
2.25.0

Re: [PATCH-4.19-stable 0/2] Backport ENCODE_FRAME_POINTER hint

[Greg Kroah-Hartman]

On Mon, Feb 10, 2020 at 02:05:41PM +0000, Dmitry Safonov wrote:
> On 4.19.93 the following warning was observed with CONFIG_FRAME_POINTER:
> > WARNING: kernel stack frame pointer at 00000000bceb5183 in Coronavirus:3282 has bad value           (null)
> >  unwind stack type:0 next_sp:          (null) mask:0x2 graph_idx:0
> >  000000009630aa47: ffffc9000126fdb0 (0xffffc9000126fdb0)
> >  0000000020360f53: ffffffff81038e33 (__save_stack_trace+0xcb/0xee)
> >  00000000675081f2: 0000000000000000 ...
> >  0000000043198fe7: ffffc9000126c000 (0xffffc9000126c000)
> >  0000000008a46231: ffffc90001270000 (0xffffc90001270000)
> [..]
> 
> It turns to be missing %rbp hint was making frame pointer unwinder
> a bit tipsy.
> The observed is WARN_ONCE(), so it one time per boot, but imho, worth to
> have in stable too.

All now queued up, thanks!

greg k-h