[PATCH v3 0/2] arm64 sigreturn unwinding fixes

[PATCH v3 0/2] arm64 sigreturn unwinding fixes

[Will Deacon]

Hi folks,

Here is v3 of my attempt at fixing the vdso sigreturn code for unwinders.
Previous versions are available here:

  v1: https://lore.kernel.org/r/20200519121818.14511-1-will@kernel.org
  v2: https://lore.kernel.org/r/20200519162821.16857-1-will@kernel.org

Changes since v2 include:

  * Fix the .cfi directives to identify the link register correctly
  * Even more comments

Daniel, please can you give this a spin with the LLVM unwinder? It should work
this time.

Cheers,

Will

Cc: Dave Martin <dave.martin@arm.com>
Cc: Tamas Zsoldos <tamas.zsoldos@arm.com> 
Cc: Daniel Kiss <daniel.kiss@arm.com>
Cc: Mark Brown <broonie@kernel.org>
Cc: <kernel-team@android.com>

--->8

Will Deacon (2):
  arm64: vdso: Don't prefix sigreturn trampoline with a BTI C
    instruction
  arm64: vdso: Fix CFI directives in sigreturn trampoline

 arch/arm64/include/asm/linkage.h     |  6 ++--
 arch/arm64/kernel/vdso/sigreturn.S   | 51 ++++++++++++++++++++++++----
 arch/arm64/kernel/vdso32/sigreturn.S | 19 ++++++-----
 3 files changed, 58 insertions(+), 18 deletions(-)

-- 
2.26.2.761.g0e0b3e54be-goog


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH v3 1/2] arm64: vdso: Don't prefix sigreturn trampoline with a BTI C instruction

[Will Deacon]

For better or worse, GDB relies on the exact instruction sequence in the
VDSO sigreturn trampoline in order to unwind from signals correctly.
Commit c91db232da48 ("arm64: vdso: Convert to modern assembler annotations")
unfortunately added a BTI C instruction to the start of __kernel_rt_sigreturn,
which breaks this check. Thankfully, it's also not required, since the
trampoline is called from a RET instruction when returning from the signal
handler

Remove the unnecessary BTI C instruction from __kernel_rt_sigreturn,
and do the same for the 32-bit VDSO as well for good measure.

Cc: Daniel Kiss <daniel.kiss@arm.com>
Cc: Tamas Zsoldos <tamas.zsoldos@arm.com>
Reviewed-by: Dave Martin <dave.martin@arm.com>
Reviewed-by: Mark Brown <broonie@kernel.org>
Fixes: c91db232da48 ("arm64: vdso: Convert to modern assembler annotations")
Signed-off-by: Will Deacon <will@kernel.org>
---
 arch/arm64/include/asm/linkage.h     |  6 +++---
 arch/arm64/kernel/vdso/sigreturn.S   | 11 +++++++++--
 arch/arm64/kernel/vdso32/sigreturn.S | 16 ++++++++--------
 3 files changed, 20 insertions(+), 13 deletions(-)

diff --git a/arch/arm64/include/asm/linkage.h b/arch/arm64/include/asm/linkage.h
index b5a7998a6b2a..81fefd2a1d02 100644
--- a/arch/arm64/include/asm/linkage.h
+++ b/arch/arm64/include/asm/linkage.h
@@ -15,9 +15,9 @@
 #define BTI_J hint 36 ;
 
 /*
- * When using in-kernel BTI we need to ensure that assembly functions
- * have suitable annotations.  Override SYM_FUNC_START to insert a BTI
- * landing pad at the start of everything.
+ * When using in-kernel BTI we need to ensure that PCS-conformant assembly
+ * functions have suitable annotations.  Override SYM_FUNC_START to insert
+ * a BTI landing pad at the start of everything.
  */
 #define SYM_FUNC_START(name)				\
 	SYM_START(name, SYM_L_GLOBAL, SYM_A_ALIGN)	\
diff --git a/arch/arm64/kernel/vdso/sigreturn.S b/arch/arm64/kernel/vdso/sigreturn.S
index 3fb13b81f780..0c921130002a 100644
--- a/arch/arm64/kernel/vdso/sigreturn.S
+++ b/arch/arm64/kernel/vdso/sigreturn.S
@@ -15,7 +15,14 @@
 	.text
 
 	nop
-SYM_FUNC_START(__kernel_rt_sigreturn)
+/*
+ * GDB relies on being able to identify the sigreturn instruction sequence to
+ * unwind from signal handlers. We cannot, therefore, use SYM_FUNC_START()
+ * here, as it will emit a BTI C instruction and break the unwinder. Thankfully,
+ * this function is only ever called from a RET and so omitting the landing pad
+ * is perfectly fine.
+ */
+SYM_CODE_START(__kernel_rt_sigreturn)
 	.cfi_startproc
 	.cfi_signal_frame
 	.cfi_def_cfa	x29, 0
@@ -24,6 +31,6 @@ SYM_FUNC_START(__kernel_rt_sigreturn)
 	mov	x8, #__NR_rt_sigreturn
 	svc	#0
 	.cfi_endproc
-SYM_FUNC_END(__kernel_rt_sigreturn)
+SYM_CODE_END(__kernel_rt_sigreturn)
 
 emit_aarch64_feature_1_and
diff --git a/arch/arm64/kernel/vdso32/sigreturn.S b/arch/arm64/kernel/vdso32/sigreturn.S
index 620524969696..b36d4e2267a3 100644
--- a/arch/arm64/kernel/vdso32/sigreturn.S
+++ b/arch/arm64/kernel/vdso32/sigreturn.S
@@ -17,39 +17,39 @@
 	.save {r0-r15}
 	.pad #COMPAT_SIGFRAME_REGS_OFFSET
 	nop
-SYM_FUNC_START(__kernel_sigreturn_arm)
+SYM_CODE_START(__kernel_sigreturn_arm)
 	mov r7, #__NR_compat_sigreturn
 	svc #0
 	.fnend
-SYM_FUNC_END(__kernel_sigreturn_arm)
+SYM_CODE_END(__kernel_sigreturn_arm)
 
 	.fnstart
 	.save {r0-r15}
 	.pad #COMPAT_RT_SIGFRAME_REGS_OFFSET
 	nop
-SYM_FUNC_START(__kernel_rt_sigreturn_arm)
+SYM_CODE_START(__kernel_rt_sigreturn_arm)
 	mov r7, #__NR_compat_rt_sigreturn
 	svc #0
 	.fnend
-SYM_FUNC_END(__kernel_rt_sigreturn_arm)
+SYM_CODE_END(__kernel_rt_sigreturn_arm)
 
 	.thumb
 	.fnstart
 	.save {r0-r15}
 	.pad #COMPAT_SIGFRAME_REGS_OFFSET
 	nop
-SYM_FUNC_START(__kernel_sigreturn_thumb)
+SYM_CODE_START(__kernel_sigreturn_thumb)
 	mov r7, #__NR_compat_sigreturn
 	svc #0
 	.fnend
-SYM_FUNC_END(__kernel_sigreturn_thumb)
+SYM_CODE_END(__kernel_sigreturn_thumb)
 
 	.fnstart
 	.save {r0-r15}
 	.pad #COMPAT_RT_SIGFRAME_REGS_OFFSET
 	nop
-SYM_FUNC_START(__kernel_rt_sigreturn_thumb)
+SYM_CODE_START(__kernel_rt_sigreturn_thumb)
 	mov r7, #__NR_compat_rt_sigreturn
 	svc #0
 	.fnend
-SYM_FUNC_END(__kernel_rt_sigreturn_thumb)
+SYM_CODE_END(__kernel_rt_sigreturn_thumb)
-- 
2.26.2.761.g0e0b3e54be-goog


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH v3 2/2] arm64: vdso: Fix CFI directives in sigreturn trampoline

[Will Deacon]

Daniel reports that the .cfi_startproc is misplaced for the sigreturn
trampoline, which causes LLVM's unwinder to misbehave:

  | I run into this with LLVM’s unwinder.
  | This combination was always broken.

This prompted Dave to question our use of CFI directives more generally,
and I ended up going down a rabbit hole trying to figure out how this
very poorly documented stuff gets used.

Move the CFI directives so that the "mysterious NOP" is included in
the .cfi_{start,end}proc block and add a bunch of comments so that I
can save myself another headache in future.

Cc: Tamas Zsoldos <tamas.zsoldos@arm.com>
Reported-by: Dave Martin <dave.martin@arm.com>
Reported-by: Daniel Kiss <daniel.kiss@arm.com>
Signed-off-by: Will Deacon <will@kernel.org>
---
 arch/arm64/kernel/vdso/sigreturn.S   | 44 +++++++++++++++++++++++-----
 arch/arm64/kernel/vdso32/sigreturn.S |  3 ++
 2 files changed, 40 insertions(+), 7 deletions(-)

diff --git a/arch/arm64/kernel/vdso/sigreturn.S b/arch/arm64/kernel/vdso/sigreturn.S
index 0c921130002a..620a3ef837b7 100644
--- a/arch/arm64/kernel/vdso/sigreturn.S
+++ b/arch/arm64/kernel/vdso/sigreturn.S
@@ -1,7 +1,11 @@
 /* SPDX-License-Identifier: GPL-2.0-only */
 /*
  * Sigreturn trampoline for returning from a signal when the SA_RESTORER
- * flag is not set.
+ * flag is not set. It serves primarily as a hall of shame for crappy
+ * unwinders and features an exciting but mysterious NOP instruction.
+ *
+ * It's also fragile as hell, so please think twice before changing anything
+ * in here.
  *
  * Copyright (C) 2012 ARM Limited
  *
@@ -14,7 +18,38 @@
 
 	.text
 
-	nop
+/* Ensure that the mysterious NOP can be associated with a function. */
+	.cfi_startproc
+
+/*
+ * .cfi_signal_frame causes the corresponding Frame Description Entry in the
+ * .eh_frame section to be annotated as a signal frame. This allows DWARF
+ * unwinders (e.g. libstdc++) to implement _Unwind_GetIPInfo(), which permits
+ * unwinding out of the signal trampoline without the need for the mysterious
+ * NOP.
+ */
+	.cfi_signal_frame
+
+/*
+ * Tell the unwinder where to locate the frame record linking back to the
+ * interrupted context. We don't provide unwind info for registers other
+ * than the frame pointer and the link register here; in practice, this
+ * is sufficient for unwinding in C/C++ based runtimes and the values in
+ * the sigcontext may have been modified by this point anyway. Debuggers
+ * already have baked-in strategies for attempting to unwind out of signals.
+ */
+	.cfi_def_cfa    x29, 0
+	.cfi_offset     x29, 0 * 8
+	.cfi_offset     x30, 1 * 8
+
+/*
+ * This mysterious NOP is required for some unwinders (e.g. libc++) that
+ * unconditionally subtract one from the result of _Unwind_GetIP() in order to
+ * identify the calling function.
+ * Hack borrowed from arch/powerpc/kernel/vdso64/sigtramp.S.
+ */
+	nop	// Mysterious NOP
+
 /*
  * GDB relies on being able to identify the sigreturn instruction sequence to
  * unwind from signal handlers. We cannot, therefore, use SYM_FUNC_START()
@@ -23,11 +58,6 @@
  * is perfectly fine.
  */
 SYM_CODE_START(__kernel_rt_sigreturn)
-	.cfi_startproc
-	.cfi_signal_frame
-	.cfi_def_cfa	x29, 0
-	.cfi_offset	x29, 0 * 8
-	.cfi_offset	x30, 1 * 8
 	mov	x8, #__NR_rt_sigreturn
 	svc	#0
 	.cfi_endproc
diff --git a/arch/arm64/kernel/vdso32/sigreturn.S b/arch/arm64/kernel/vdso32/sigreturn.S
index b36d4e2267a3..b0091064c3d6 100644
--- a/arch/arm64/kernel/vdso32/sigreturn.S
+++ b/arch/arm64/kernel/vdso32/sigreturn.S
@@ -3,6 +3,9 @@
  * This file provides both A32 and T32 versions, in accordance with the
  * arm sigreturn code.
  *
+ * Please read the comments in arch/arm64/kernel/vdso/sigreturn.S to
+ * understand some of the craziness in here.
+ *
  * Copyright (C) 2018 ARM Limited
  */
 
-- 
2.26.2.761.g0e0b3e54be-goog


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH v3 0/2] arm64 sigreturn unwinding fixes

[Daniel Kiss]

Hi Will,

Works okay, LGTM.

Thanks for the patch.

Cheers,
Daniel

> On 20 May 2020, at 19:55, Will Deacon <will@kernel.org> wrote:
> 
> Hi folks,
> 
> Here is v3 of my attempt at fixing the vdso sigreturn code for unwinders.
> Previous versions are available here:
> 
>  v1: https://lore.kernel.org/r/20200519121818.14511-1-will@kernel.org
>  v2: https://lore.kernel.org/r/20200519162821.16857-1-will@kernel.org
> 
> Changes since v2 include:
> 
>  * Fix the .cfi directives to identify the link register correctly
>  * Even more comments
> 
> Daniel, please can you give this a spin with the LLVM unwinder? It should work
> this time.
> 
> Cheers,
> 
> Will
> 
> Cc: Dave Martin <dave.martin@arm.com>
> Cc: Tamas Zsoldos <tamas.zsoldos@arm.com> 
> Cc: Daniel Kiss <daniel.kiss@arm.com>
> Cc: Mark Brown <broonie@kernel.org>
> Cc: <kernel-team@android.com>
> 
> --->8
> 
> Will Deacon (2):
>  arm64: vdso: Don't prefix sigreturn trampoline with a BTI C
>    instruction
>  arm64: vdso: Fix CFI directives in sigreturn trampoline
> 
> arch/arm64/include/asm/linkage.h     |  6 ++--
> arch/arm64/kernel/vdso/sigreturn.S   | 51 ++++++++++++++++++++++++----
> arch/arm64/kernel/vdso32/sigreturn.S | 19 ++++++-----
> 3 files changed, 58 insertions(+), 18 deletions(-)
> 
> -- 
> 2.26.2.761.g0e0b3e54be-goog
> 


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel