diff --git a/arch/arm64/kernel/vdso/Makefile b/arch/arm64/kernel/vdso/Makefile
index 556d424c6f52f1..1e5a940532da37 100644
--- a/arch/arm64/kernel/vdso/Makefile
+++ b/arch/arm64/kernel/vdso/Makefile
@@ -24,7 +24,7 @@ btildflags-$(CONFIG_ARM64_BTI_KERNEL) += -z force-bti
 # routines, as x86 does (see 6f121e548f83 ("x86, vdso: Reimplement vdso.so
 # preparation in build-time C")).
 ldflags-y := -shared -nostdlib -soname=linux-vdso.so.1 --hash-style=sysv \
-		-Bsymbolic --eh-frame-hdr --build-id -n $(btildflags-y) -T
+		-Bsymbolic --no-eh-frame-hdr --build-id -n $(btildflags-y) -T
 
 ccflags-y := -fno-common -fno-builtin -fno-stack-protector -ffixed-x18
 ccflags-y += -DDISABLE_BRANCH_PROFILING
diff --git a/arch/arm64/kernel/vdso/sigreturn.S b/arch/arm64/kernel/vdso/sigreturn.S
index 620a3ef837b7c9..0e18729abc3b78 100644
--- a/arch/arm64/kernel/vdso/sigreturn.S
+++ b/arch/arm64/kernel/vdso/sigreturn.S
@@ -18,29 +18,40 @@
 
 	.text
 
+/*
+ * NOTE!!!  You may notice that all of the .cfi directives in this file have
+ * been commented out. This is because they have been shown to trigger segfaults
+ * in libgcc when unwinding out of a SIGCANCEL handler to invoke pthread
+ * cleanup handlers during the thread cancellation dance. By omitting the
+ * directives, we trigger an arm64-specific fallback path in the unwinder which
+ * recognises the signal frame and restores many of the registers directly from
+ * the sigcontext. Re-enabling the cfi directives here therefore needs to be
+ * much more comprehensive to reduce the risk of further regressions.
+ */
+
 /* Ensure that the mysterious NOP can be associated with a function. */
-	.cfi_startproc
+//	.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 causes the corresponding Frame Description Entry (FDE) in
+ * the .eh_frame section to be annotated as a signal frame. This allows DWARF
+ * unwinders (e.g. libstdc++) to implement _Unwind_GetIPInfo() and identify
+ * the next frame using the unmodified return address instead of subtracting 1,
+ * which may yield the wrong FDE.
  */
-	.cfi_signal_frame
+//	.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
+ * interrupted context. We don't provide unwind info for registers other than
+ * the frame pointer and the link register here; in practice, this is likely to
+ * be insufficient for unwinding in C/C++ based runtimes, especially without a
+ * means to restore the stack pointer. Thankfully, unwinders and 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
+//	.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
@@ -51,16 +62,19 @@
 	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()
- * 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.
+ * GDB, libgcc and libunwind rely 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)
+//	PLEASE DO NOT MODIFY
 	mov	x8, #__NR_rt_sigreturn
+//	PLEASE DO NOT MODIFY
 	svc	#0
-	.cfi_endproc
+//	PLEASE DO NOT MODIFY
+//	.cfi_endproc
 SYM_CODE_END(__kernel_rt_sigreturn)
 
 emit_aarch64_feature_1_and