[PATCH v9 0/6] x86/vdso: VDSO updates and fixes for sframes

[PATCH v9 0/6] x86/vdso: VDSO updates and fixes for sframes

[Jens Remus]

This enables generation of SFrame V3 stack trace information for VDSO on
x86-64.  It's a continuation of Josh's and Steve's work:

   https://lore.kernel.org/all/cover.1737511963.git.jpoimboe@kernel.org/
   https://lore.kernel.org/all/20250425023750.669174660@goodmis.org/

This series focuses only on the VDSO code. They are helpful fixes
and updates that doesn't rely on sframes (although the last patch
is sframe related).

This series applies on top of tip:x86/entry:

  git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git  x86/entry

Like the unwind user sframe series [1] it depends on the binutils 2.46
release to be used to build the VDSO with SFrame V3 stack trace
information (using the assembler option --gsframe-3).

[1]: [PATCH v13 00/18] unwind_deferred: Implement sframe handling,
    https://lore.kernel.org/all/20260127150554.2760964-1-jremus@linux.ibm.com/

Changes in v9 (see indivicual patch notes):
- Always define KEEP_SFRAME to either true/false in specific VDSO linker
  scripts and use #if instead of #ifdef in common one. (Peter)
- Reword patch 6 commit message to provide more details.
- Note: Binutils 2.46 with SFrame V3 support has been released.

Changes in v8:
- Discard .sframe for x32 and x86-32 VDSOs. (Josh/Indu)
- Define CFI_SIGNAL_FRAME for !BUILD_VDSO.
- Drop .cfi_sections .sframe in dwarf2.h in favor of the explicitly
  specified more specific assembler option --gsframe-3.
- Incorporate missing changes and review feedback from Steven's v6
  (I erroneously based my v6 on Steven's v5):
  - Reword patch 3 commit subject to Steven's v6 one.
  - Remove SYM_F_ALIGN in __vdso_sgx_enter_enclave(). (Josh)

Changes in v7:
- Rebase on H. Peter Anvin's vDSO changes on tip:x86/entry. (Peter)
- Simplify adding assembler option -Wa,--gsframe-3.  Add for vdso64
  only.
- Align to .eh_frame and mark .sframe as KEEP in vDSO linker script.
  Note that GNU linker 2.46 will mark .sframe as KEEP in its default
  linker script as well.

Changes in v6:
- SFrame V3 support (SFrame V2 is not supported).
- Prevent GNU_SFRAME program table entry to empty .sframe section.
- Integrate v5 review feedback. (Josh)

Regards,
Jens


Josh Poimboeuf (6):
  x86/vdso: Fix DWARF generation for getrandom()
  x86/asm: Avoid emitting DWARF CFI for non-VDSO
  x86/asm: Use CFI_* macros in SYM_FUNC_* macros so they can be added to
    VDSO
  x86/vdso: Use SYM_FUNC_{START,END} in __kernel_vsyscall()
  x86/vdso: Use CFI macros in __vdso_sgx_enter_enclave()
  x86/vdso: Enable sframe generation in VDSO

 arch/Kconfig                                  |  7 +++
 arch/x86/entry/vdso/common/vdso-layout.lds.S  | 17 +++++-
 arch/x86/entry/vdso/vdso32/system_call.S      | 10 +---
 arch/x86/entry/vdso/vdso32/vdso32.lds.S       |  3 ++
 arch/x86/entry/vdso/vdso64/Makefile           |  1 +
 arch/x86/entry/vdso/vdso64/vdso64.lds.S       |  2 +
 arch/x86/entry/vdso/vdso64/vdsox32.lds.S      |  6 +++
 .../x86/entry/vdso/vdso64/vgetrandom-chacha.S |  3 +-
 arch/x86/entry/vdso/vdso64/vsgx.S             | 18 +++----
 arch/x86/include/asm/dwarf2.h                 | 52 +++++++++++++------
 arch/x86/include/asm/linkage.h                | 33 +++++++++---
 arch/x86/include/asm/vdso.h                   |  1 -
 12 files changed, 107 insertions(+), 46 deletions(-)

-- 
2.51.0

[PATCH v9 1/6] x86/vdso: Fix DWARF generation for getrandom()

[Jens Remus]

From: Josh Poimboeuf <jpoimboe@kernel.org>

Add CFI annotations to the VDSO implementation of getrandom() so it will
have valid DWARF unwinding metadata.

Fixes: 33385150ac45 ("x86: vdso: Wire up getrandom() vDSO implementation")
Signed-off-by: Josh Poimboeuf <jpoimboe@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Acked-by: H. Peter Anvin (Intel) <hpa@zytor.com>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
---
 arch/x86/entry/vdso/vdso64/vgetrandom-chacha.S | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/arch/x86/entry/vdso/vdso64/vgetrandom-chacha.S b/arch/x86/entry/vdso/vdso64/vgetrandom-chacha.S
index bcba5639b8ee..cc82da9216fb 100644
--- a/arch/x86/entry/vdso/vdso64/vgetrandom-chacha.S
+++ b/arch/x86/entry/vdso/vdso64/vgetrandom-chacha.S
@@ -4,7 +4,7 @@
  */
 
 #include <linux/linkage.h>
-#include <asm/frame.h>
+#include <asm/dwarf2.h>
 
 .section	.rodata, "a"
 .align 16
@@ -22,7 +22,7 @@ CONSTANTS:	.octa 0x6b20657479622d323320646e61707865
  *	rcx: number of 64-byte blocks to write to output
  */
 SYM_FUNC_START(__arch_chacha20_blocks_nostack)
-
+	CFI_STARTPROC
 .set	output,		%rdi
 .set	key,		%rsi
 .set	counter,	%rdx
@@ -175,4 +175,5 @@ SYM_FUNC_START(__arch_chacha20_blocks_nostack)
 	pxor		temp,temp
 
 	ret
+	CFI_ENDPROC
 SYM_FUNC_END(__arch_chacha20_blocks_nostack)
-- 
2.51.0

[PATCH v9 2/6] x86/asm: Avoid emitting DWARF CFI for non-VDSO

[Jens Remus]

From: Josh Poimboeuf <jpoimboe@kernel.org>

It was decided years ago that .cfi_* annotations aren't maintainable in
the kernel.  They were replaced by objtool unwind hints.  For the kernel
proper, ensure the CFI_* macros don't do anything.

On the other hand the VDSO library *does* use them, so user space can
unwind through it.

Make sure these macros only work for VDSO.  They aren't actually being
used outside of VDSO anyway, so there's no functional change.

[ Jens Remus: Define CFI_SIGNAL_FRAME for !BUILD_VDSO. ]

Signed-off-by: Josh Poimboeuf <jpoimboe@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Acked-by: H. Peter Anvin (Intel) <hpa@zytor.com>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
---
 arch/x86/include/asm/dwarf2.h | 52 ++++++++++++++++++++++++-----------
 1 file changed, 36 insertions(+), 16 deletions(-)

diff --git a/arch/x86/include/asm/dwarf2.h b/arch/x86/include/asm/dwarf2.h
index 09c9684d3ad6..13e2e64ef265 100644
--- a/arch/x86/include/asm/dwarf2.h
+++ b/arch/x86/include/asm/dwarf2.h
@@ -6,6 +6,15 @@
 #warning "asm/dwarf2.h should be only included in pure assembly files"
 #endif
 
+#ifdef BUILD_VDSO
+
+	/*
+	 * For the vDSO, emit both runtime unwind information and debug
+	 * symbols for the .dbg file.
+	 */
+
+	.cfi_sections .eh_frame, .debug_frame
+
 #define CFI_STARTPROC		.cfi_startproc
 #define CFI_ENDPROC		.cfi_endproc
 #define CFI_DEF_CFA		.cfi_def_cfa
@@ -22,21 +31,32 @@
 #define CFI_ESCAPE		.cfi_escape
 #define CFI_SIGNAL_FRAME	.cfi_signal_frame
 
-#ifndef BUILD_VDSO
-	/*
-	 * Emit CFI data in .debug_frame sections, not .eh_frame sections.
-	 * The latter we currently just discard since we don't do DWARF
-	 * unwinding at runtime.  So only the offline DWARF information is
-	 * useful to anyone.  Note we should not use this directive if we
-	 * ever decide to enable DWARF unwinding at runtime.
-	 */
-	.cfi_sections .debug_frame
-#else
-	 /*
-	  * For the vDSO, emit both runtime unwind information and debug
-	  * symbols for the .dbg file.
-	  */
-	.cfi_sections .eh_frame, .debug_frame
-#endif
+#else /* !BUILD_VDSO */
+
+/*
+ * On x86, these macros aren't used outside VDSO.  As well they shouldn't be:
+ * they're fragile and very difficult to maintain.
+ */
+
+.macro nocfi args:vararg
+.endm
+
+#define CFI_STARTPROC		nocfi
+#define CFI_ENDPROC		nocfi
+#define CFI_DEF_CFA		nocfi
+#define CFI_DEF_CFA_REGISTER	nocfi
+#define CFI_DEF_CFA_OFFSET	nocfi
+#define CFI_ADJUST_CFA_OFFSET	nocfi
+#define CFI_OFFSET		nocfi
+#define CFI_REL_OFFSET		nocfi
+#define CFI_REGISTER		nocfi
+#define CFI_RESTORE		nocfi
+#define CFI_REMEMBER_STATE	nocfi
+#define CFI_RESTORE_STATE	nocfi
+#define CFI_UNDEFINED		nocfi
+#define CFI_ESCAPE		nocfi
+#define CFI_SIGNAL_FRAME	nocfi
+
+#endif /* !BUILD_VDSO */
 
 #endif /* _ASM_X86_DWARF2_H */
-- 
2.51.0

[PATCH v9 3/6] x86/asm: Use CFI_* macros in SYM_FUNC_* macros so they can be added to VDSO

[Jens Remus]

From: Josh Poimboeuf <jpoimboe@kernel.org>

Add CFI_STARTPROC and CFI_ENDPROC annotations to the SYM_FUNC_* macros
so the VDSO asm functions don't need to add them manually.  Note this
only affects VDSO, the CFI_* macros are empty for the kernel proper.

Signed-off-by: Josh Poimboeuf <jpoimboe@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Acked-by: H. Peter Anvin (Intel) <hpa@zytor.com>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
---
 arch/x86/entry/vdso/common/vdso-layout.lds.S  |  2 +-
 .../x86/entry/vdso/vdso64/vgetrandom-chacha.S |  2 --
 arch/x86/entry/vdso/vdso64/vsgx.S             |  4 ---
 arch/x86/include/asm/linkage.h                | 33 +++++++++++++++----
 arch/x86/include/asm/vdso.h                   |  1 -
 5 files changed, 28 insertions(+), 14 deletions(-)

diff --git a/arch/x86/entry/vdso/common/vdso-layout.lds.S b/arch/x86/entry/vdso/common/vdso-layout.lds.S
index a1e30be3e83d..856b8b9d278c 100644
--- a/arch/x86/entry/vdso/common/vdso-layout.lds.S
+++ b/arch/x86/entry/vdso/common/vdso-layout.lds.S
@@ -1,5 +1,5 @@
 /* SPDX-License-Identifier: GPL-2.0 */
-#include <asm/vdso.h>
+#include <asm/page_types.h>
 #include <asm/vdso/vsyscall.h>
 #include <vdso/datapage.h>
 
diff --git a/arch/x86/entry/vdso/vdso64/vgetrandom-chacha.S b/arch/x86/entry/vdso/vdso64/vgetrandom-chacha.S
index cc82da9216fb..a33212594731 100644
--- a/arch/x86/entry/vdso/vdso64/vgetrandom-chacha.S
+++ b/arch/x86/entry/vdso/vdso64/vgetrandom-chacha.S
@@ -22,7 +22,6 @@ CONSTANTS:	.octa 0x6b20657479622d323320646e61707865
  *	rcx: number of 64-byte blocks to write to output
  */
 SYM_FUNC_START(__arch_chacha20_blocks_nostack)
-	CFI_STARTPROC
 .set	output,		%rdi
 .set	key,		%rsi
 .set	counter,	%rdx
@@ -175,5 +174,4 @@ SYM_FUNC_START(__arch_chacha20_blocks_nostack)
 	pxor		temp,temp
 
 	ret
-	CFI_ENDPROC
 SYM_FUNC_END(__arch_chacha20_blocks_nostack)
diff --git a/arch/x86/entry/vdso/vdso64/vsgx.S b/arch/x86/entry/vdso/vdso64/vsgx.S
index 37a3d4c02366..c0342238c976 100644
--- a/arch/x86/entry/vdso/vdso64/vsgx.S
+++ b/arch/x86/entry/vdso/vdso64/vsgx.S
@@ -24,8 +24,6 @@
 .section .text, "ax"
 
 SYM_FUNC_START(__vdso_sgx_enter_enclave)
-	/* Prolog */
-	.cfi_startproc
 	push	%rbp
 	.cfi_adjust_cfa_offset	8
 	.cfi_rel_offset		%rbp, 0
@@ -143,8 +141,6 @@ SYM_FUNC_START(__vdso_sgx_enter_enclave)
 	jle	.Lout
 	jmp	.Lenter_enclave
 
-	.cfi_endproc
-
 _ASM_VDSO_EXTABLE_HANDLE(.Lenclu_eenter_eresume, .Lhandle_exception)
 
 SYM_FUNC_END(__vdso_sgx_enter_enclave)
diff --git a/arch/x86/include/asm/linkage.h b/arch/x86/include/asm/linkage.h
index 9d38ae744a2e..9d7f90c57451 100644
--- a/arch/x86/include/asm/linkage.h
+++ b/arch/x86/include/asm/linkage.h
@@ -40,6 +40,10 @@
 
 #ifdef __ASSEMBLER__
 
+#ifndef LINKER_SCRIPT
+#include <asm/dwarf2.h>
+#endif
+
 #if defined(CONFIG_MITIGATION_RETHUNK) && !defined(__DISABLE_EXPORTS) && !defined(BUILD_VDSO)
 #define RET	jmp __x86_return_thunk
 #else /* CONFIG_MITIGATION_RETPOLINE */
@@ -112,34 +116,51 @@
 # define SYM_FUNC_ALIAS_MEMFUNC	SYM_FUNC_ALIAS
 #endif
 
+#define __SYM_FUNC_START				\
+	CFI_STARTPROC ASM_NL
+
+#define __SYM_FUNC_END					\
+	CFI_ENDPROC ASM_NL
+
 /* SYM_TYPED_FUNC_START -- use for indirectly called globals, w/ CFI type */
 #define SYM_TYPED_FUNC_START(name)				\
 	SYM_TYPED_START(name, SYM_L_GLOBAL, SYM_F_ALIGN)	\
+	__SYM_FUNC_START					\
 	ENDBR
 
 /* SYM_FUNC_START -- use for global functions */
 #define SYM_FUNC_START(name)				\
-	SYM_START(name, SYM_L_GLOBAL, SYM_F_ALIGN)
+	SYM_START(name, SYM_L_GLOBAL, SYM_F_ALIGN)	\
+	__SYM_FUNC_START
 
 /* SYM_FUNC_START_NOALIGN -- use for global functions, w/o alignment */
 #define SYM_FUNC_START_NOALIGN(name)			\
-	SYM_START(name, SYM_L_GLOBAL, SYM_A_NONE)
+	SYM_START(name, SYM_L_GLOBAL, SYM_A_NONE)	\
+	__SYM_FUNC_START
 
 /* SYM_FUNC_START_LOCAL -- use for local functions */
 #define SYM_FUNC_START_LOCAL(name)			\
-	SYM_START(name, SYM_L_LOCAL, SYM_F_ALIGN)
+	SYM_START(name, SYM_L_LOCAL, SYM_F_ALIGN)	\
+	__SYM_FUNC_START
 
 /* SYM_FUNC_START_LOCAL_NOALIGN -- use for local functions, w/o alignment */
 #define SYM_FUNC_START_LOCAL_NOALIGN(name)		\
-	SYM_START(name, SYM_L_LOCAL, SYM_A_NONE)
+	SYM_START(name, SYM_L_LOCAL, SYM_A_NONE)	\
+	__SYM_FUNC_START
 
 /* SYM_FUNC_START_WEAK -- use for weak functions */
 #define SYM_FUNC_START_WEAK(name)			\
-	SYM_START(name, SYM_L_WEAK, SYM_F_ALIGN)
+	SYM_START(name, SYM_L_WEAK, SYM_F_ALIGN)	\
+	__SYM_FUNC_START
 
 /* SYM_FUNC_START_WEAK_NOALIGN -- use for weak functions, w/o alignment */
 #define SYM_FUNC_START_WEAK_NOALIGN(name)		\
-	SYM_START(name, SYM_L_WEAK, SYM_A_NONE)
+	SYM_START(name, SYM_L_WEAK, SYM_A_NONE)		\
+	__SYM_FUNC_START
+
+#define SYM_FUNC_END(name)				\
+	__SYM_FUNC_END					\
+	SYM_END(name, SYM_T_FUNC)
 
 /*
  * Expose 'sym' to the startup code in arch/x86/boot/startup/, by emitting an
diff --git a/arch/x86/include/asm/vdso.h b/arch/x86/include/asm/vdso.h
index e8afbe9faa5b..498ac423741c 100644
--- a/arch/x86/include/asm/vdso.h
+++ b/arch/x86/include/asm/vdso.h
@@ -2,7 +2,6 @@
 #ifndef _ASM_X86_VDSO_H
 #define _ASM_X86_VDSO_H
 
-#include <asm/page_types.h>
 #include <linux/linkage.h>
 #include <linux/init.h>
 
-- 
2.51.0

[PATCH v9 4/6] x86/vdso: Use SYM_FUNC_{START,END} in __kernel_vsyscall()

[Jens Remus]

From: Josh Poimboeuf <jpoimboe@kernel.org>

Use SYM_FUNC_{START,END} instead of all the boilerplate.  No functional
change.

Signed-off-by: Josh Poimboeuf <jpoimboe@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Acked-by: H. Peter Anvin (Intel) <hpa@zytor.com>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
---
 arch/x86/entry/vdso/vdso32/system_call.S | 10 ++--------
 1 file changed, 2 insertions(+), 8 deletions(-)

diff --git a/arch/x86/entry/vdso/vdso32/system_call.S b/arch/x86/entry/vdso/vdso32/system_call.S
index 9157cf9c5749..a90f4f7de396 100644
--- a/arch/x86/entry/vdso/vdso32/system_call.S
+++ b/arch/x86/entry/vdso/vdso32/system_call.S
@@ -9,11 +9,7 @@
 #include <asm/alternative.h>
 
 	.text
-	.globl __kernel_vsyscall
-	.type __kernel_vsyscall,@function
-	ALIGN
-__kernel_vsyscall:
-	CFI_STARTPROC
+SYM_FUNC_START(__kernel_vsyscall)
 
 	/*
 	 * If using int $0x80, there is no reason to muck about with the
@@ -85,7 +81,5 @@ SYM_INNER_LABEL(int80_landing_pad, SYM_L_GLOBAL)
 	CFI_RESTORE		ecx
 	CFI_ADJUST_CFA_OFFSET	-4
 	RET
-	CFI_ENDPROC
-
-	.size __kernel_vsyscall,.-__kernel_vsyscall
+SYM_FUNC_END(__kernel_vsyscall)
 	.previous
-- 
2.51.0

[PATCH v9 5/6] x86/vdso: Use CFI macros in __vdso_sgx_enter_enclave()

[Jens Remus]

From: Josh Poimboeuf <jpoimboe@kernel.org>

Use the CFI macros instead of the raw .cfi_* directives to be consistent
with the rest of the VDSO asm.  It's also easier on the eyes.

No functional changes.

Signed-off-by: Josh Poimboeuf <jpoimboe@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Acked-by: H. Peter Anvin (Intel) <hpa@zytor.com>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
---
 arch/x86/entry/vdso/vdso64/vsgx.S | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)

diff --git a/arch/x86/entry/vdso/vdso64/vsgx.S b/arch/x86/entry/vdso/vdso64/vsgx.S
index c0342238c976..76efbeb1e287 100644
--- a/arch/x86/entry/vdso/vdso64/vsgx.S
+++ b/arch/x86/entry/vdso/vdso64/vsgx.S
@@ -25,12 +25,12 @@
 
 SYM_FUNC_START(__vdso_sgx_enter_enclave)
 	push	%rbp
-	.cfi_adjust_cfa_offset	8
-	.cfi_rel_offset		%rbp, 0
+	CFI_ADJUST_CFA_OFFSET	8
+	CFI_REL_OFFSET		%rbp, 0
 	mov	%rsp, %rbp
-	.cfi_def_cfa_register	%rbp
+	CFI_DEF_CFA_REGISTER	%rbp
 	push	%rbx
-	.cfi_rel_offset		%rbx, -8
+	CFI_REL_OFFSET		%rbx, -8
 
 	mov	%ecx, %eax
 .Lenter_enclave:
@@ -77,13 +77,11 @@ SYM_FUNC_START(__vdso_sgx_enter_enclave)
 .Lout:
 	pop	%rbx
 	leave
-	.cfi_def_cfa		%rsp, 8
+	CFI_DEF_CFA		%rsp, 8
 	RET
 
-	/* The out-of-line code runs with the pre-leave stack frame. */
-	.cfi_def_cfa		%rbp, 16
-
 .Linvalid_input:
+	CFI_DEF_CFA		%rbp, 16
 	mov	$(-EINVAL), %eax
 	jmp	.Lout
 
-- 
2.51.0

[PATCH v9 6/6] x86/vdso: Enable sframe generation in VDSO

[Jens Remus]

From: Josh Poimboeuf <jpoimboe@kernel.org>

Enable .sframe generation in the VDSO library so kernel and user space
can unwind through it.

Starting with binutils 2.46 both GNU assembler and GNU linker
exclusively support generating and merging .sframe in SFrame V3 format.
For x86 SFrame is only supported for x86-64.  Not for x86-32 nor x32.

Test whether the assembler supports option '--gsframe-3' to explicitly
select SFrame V3 format.  Note that testing using Kconfig macro
'as-option' is not sufficient, as GNU assembler will accept the option
for any target, regardless of whether it is actually capable to generate
.sframe for it, as long the input does not trigger the generation.
Therefore it is necessary to use Kconfig macro 'as-instr' to provide
minimal CFI directives that trigger generation of .sframe.

For x86-64 VDSO, only if supported by the assembler, generate .sframe,
collect it, mark it as KEEP, and generate a GNU_SFRAME program table
entry.

For x86-32 and x32 VDSOs, given SFrame is not supported, do not generate
any .sframe nor GNU_SFRAME program table entry.  Instead explicitly
discard any .sframe.  The latter is required for x32 VDSO, as it is
built from x86-64 VDSO objects (potentially with .sframe) converted to
x32.  In this regard discarding .sframe also prevents potential
issues with linkers, such as GNU linker prior to binutils 2.46 commit
7487c98ff07a ("x32: Allow R_X86_64_PC64 for SFrame V3"), that do not
support R_X86_64_PC64 relocations in x32, like those found in .sframe
in SFrame V3 format.

[ Jens Remus: Add support for SFrame V3.  Prevent GNU_SFRAME program
table entry to empty .sframe section.  Reword commit message. ]

Signed-off-by: Josh Poimboeuf <jpoimboe@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
---

Notes (jremus):
    Changes in v9:
    - Always define KEEP_SFRAME to either true/false in specific VDSO linker
      scripts and use #if instead of #ifdef in common one. (Peter)
    - Reword commit message to provide more details.

 arch/Kconfig                                 |  7 +++++++
 arch/x86/entry/vdso/common/vdso-layout.lds.S | 15 +++++++++++++++
 arch/x86/entry/vdso/vdso32/vdso32.lds.S      |  3 +++
 arch/x86/entry/vdso/vdso64/Makefile          |  1 +
 arch/x86/entry/vdso/vdso64/vdso64.lds.S      |  2 ++
 arch/x86/entry/vdso/vdso64/vdsox32.lds.S     |  6 ++++++
 6 files changed, 34 insertions(+)

diff --git a/arch/Kconfig b/arch/Kconfig
index 31220f512b16..8170e492a44c 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -479,6 +479,13 @@ config HAVE_HARDLOCKUP_DETECTOR_ARCH
 	  It uses the same command line parameters, and sysctl interface,
 	  as the generic hardlockup detectors.
 
+config AS_SFRAME
+	bool
+
+config AS_SFRAME3
+	def_bool $(as-instr,.cfi_startproc\n.cfi_endproc,-Wa$(comma)--gsframe-3)
+	select AS_SFRAME
+
 config UNWIND_USER
 	bool
 
diff --git a/arch/x86/entry/vdso/common/vdso-layout.lds.S b/arch/x86/entry/vdso/common/vdso-layout.lds.S
index 856b8b9d278c..c486b07b195a 100644
--- a/arch/x86/entry/vdso/common/vdso-layout.lds.S
+++ b/arch/x86/entry/vdso/common/vdso-layout.lds.S
@@ -60,6 +60,13 @@ SECTIONS
 		*(.eh_frame.*)
 	}					:text
 
+#if KEEP_SFRAME
+	.sframe		: {
+		KEEP (*(.sframe))
+		*(.sframe.*)
+	}					:text	:sframe
+#endif
+
 	/*
 	 * Text is well-separated from actual data: there's plenty of
 	 * stuff that isn't used at runtime in between.
@@ -80,6 +87,10 @@ SECTIONS
 		*(.discard)
 		*(.discard.*)
 		*(__bug_table)
+#if !KEEP_SFRAME
+		*(.sframe)
+		*(.sframe.*)
+#endif
 	}
 }
 
@@ -89,6 +100,7 @@ SECTIONS
 #define PT_GNU_EH_FRAME	0x6474e550
 #define PT_GNU_STACK	0x6474e551
 #define PT_GNU_PROPERTY	0x6474e553
+#define PT_GNU_SFRAME	0x6474e554
 
 /*
  * We must supply the ELF program headers explicitly to get just one
@@ -104,6 +116,9 @@ PHDRS
 	dynamic		PT_DYNAMIC	PF_R;
 	note		PT_NOTE		PF_R;
 	eh_frame_hdr	PT_GNU_EH_FRAME PF_R;
+#if KEEP_SFRAME
+	sframe		PT_GNU_SFRAME	PF_R;
+#endif
 	gnu_stack	PT_GNU_STACK	PF_RW;
 	gnu_property	PT_GNU_PROPERTY	PF_R;
 }
diff --git a/arch/x86/entry/vdso/vdso32/vdso32.lds.S b/arch/x86/entry/vdso/vdso32/vdso32.lds.S
index 8a853543fc0d..3a161f8e9c2d 100644
--- a/arch/x86/entry/vdso/vdso32/vdso32.lds.S
+++ b/arch/x86/entry/vdso/vdso32/vdso32.lds.S
@@ -11,6 +11,9 @@
 
 #define BUILD_VDSO32
 
+/* Discard .sframe if any.  SFrame does not support x86-32. */
+#define KEEP_SFRAME	0
+
 #include "common/vdso-layout.lds.S"
 
 /* The ELF entry point can be used to set the AT_SYSINFO value.  */
diff --git a/arch/x86/entry/vdso/vdso64/Makefile b/arch/x86/entry/vdso/vdso64/Makefile
index bfffaf1aeecc..459f8026531e 100644
--- a/arch/x86/entry/vdso/vdso64/Makefile
+++ b/arch/x86/entry/vdso/vdso64/Makefile
@@ -14,6 +14,7 @@ vobjs-$(CONFIG_X86_SGX)		+= vsgx.o
 
 # Compilation flags
 flags-y				:= -DBUILD_VDSO64 -m64 -mcmodel=small
+flags-$(CONFIG_AS_SFRAME3)	+= -Wa,--gsframe-3
 
 # The location of this include matters!
 include $(src)/../common/Makefile.include
diff --git a/arch/x86/entry/vdso/vdso64/vdso64.lds.S b/arch/x86/entry/vdso/vdso64/vdso64.lds.S
index 5ce3f2b6373a..6685cf385fc1 100644
--- a/arch/x86/entry/vdso/vdso64/vdso64.lds.S
+++ b/arch/x86/entry/vdso/vdso64/vdso64.lds.S
@@ -9,6 +9,8 @@
 
 #define BUILD_VDSO64
 
+#define KEEP_SFRAME	IS_ENABLED(CONFIG_AS_SFRAME)
+
 #include "common/vdso-layout.lds.S"
 
 /*
diff --git a/arch/x86/entry/vdso/vdso64/vdsox32.lds.S b/arch/x86/entry/vdso/vdso64/vdsox32.lds.S
index 3dbd20c8dacc..5270fd0bdd0f 100644
--- a/arch/x86/entry/vdso/vdso64/vdsox32.lds.S
+++ b/arch/x86/entry/vdso/vdso64/vdsox32.lds.S
@@ -9,6 +9,12 @@
 
 #define BUILD_VDSOX32
 
+/*
+ * Discard .sframe from x86-64 compiles.  SFrame does not support x32 and
+ * it contains R_X86_64_PC64 relocations, which linkers may not expect.
+ */
+#define KEEP_SFRAME	0
+
 #include "common/vdso-layout.lds.S"
 
 /*
-- 
2.51.0