selftests/bpf: Enable private stack tests for arm64

As arm64 JIT now supports private stack, make sure all relevant tests
run on arm64 architecture.

Relevant tests:

#415/1 struct_ops_private_stack/private_stack:OK
#415/2 struct_ops_private_stack/private_stack_fail:OK
#415/3 struct_ops_private_stack/private_stack_recur:OK
#415 struct_ops_private_stack:OK
#549/1 verifier_private_stack/Private stack, single prog:OK
#549/2 verifier_private_stack/Private stack, subtree > MAX_BPF_STACK:OK
#549/3 verifier_private_stack/No private stack:OK
#549/4 verifier_private_stack/Private stack, callback:OK
#549/5 verifier_private_stack/Private stack, exception in mainprog:OK
#549/6 verifier_private_stack/Private stack, exception in subprog:OK
#549/7 verifier_private_stack/Private stack, async callback, not nested:OK
#549/8 verifier_private_stack/Private stack, async callback, potential nesting:OK
#549 verifier_private_stack:OK
Summary: 2/11 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Puranjay Mohan <[e-mail]>
Signed-off-by: Daniel Borkmann <[e-mail]>
Acked-by: Yonghong Song <[e-mail]>
Link: https://lore.kernel.org/bpf/[e-mail]

tools/testing/selftests/bpf/progs/struct_ops_private_stack.c

···
 
 char _license[] SEC("license") = "GPL";
 
-#if defined(__TARGET_ARCH_x86)
+#if defined(__TARGET_ARCH_x86) || defined(__TARGET_ARCH_arm64)
 bool skip __attribute((__section__(".data"))) = false;
 #else
 bool skip = true;

tools/testing/selftests/bpf/progs/struct_ops_private_stack_fail.c

···
 
 char _license[] SEC("license") = "GPL";
 
-#if defined(__TARGET_ARCH_x86)
+#if defined(__TARGET_ARCH_x86) || defined(__TARGET_ARCH_arm64)
 bool skip __attribute((__section__(".data"))) = false;
 #else
 bool skip = true;

tools/testing/selftests/bpf/progs/struct_ops_private_stack_recur.c

···
 
 char _license[] SEC("license") = "GPL";
 
-#if defined(__TARGET_ARCH_x86)
+#if defined(__TARGET_ARCH_x86) || defined(__TARGET_ARCH_arm64)
 bool skip __attribute((__section__(".data"))) = false;
 #else
 bool skip = true;

tools/testing/selftests/bpf/progs/verifier_private_stack.c

···
 /* From include/linux/filter.h */
 #define MAX_BPF_STACK    512
 
-#if defined(__TARGET_ARCH_x86)
+#if defined(__TARGET_ARCH_x86) || defined(__TARGET_ARCH_arm64)
 
 struct elem {
 	struct bpf_timer t;
···
 __jited("	addq	%gs:{{.*}}, %r9")
 __jited("	movl	$0x2a, %edi")
 __jited("	movq	%rdi, -0x100(%r9)")
+__arch_arm64
+__jited("	stp	x25, x27, [sp, {{.*}}]!")
+__jited("	mov	x27, {{.*}}")
+__jited("	movk	x27, {{.*}}, lsl #16")
+__jited("	movk	x27, {{.*}}")
+__jited("	mrs	x10, TPIDR_EL{{[0-1]}}")
+__jited("	add	x27, x27, x10")
+__jited("	add	x25, x27, {{.*}}")
+__jited("	mov	x0, #0x2a")
+__jited("	str	x0, [x27]")
+__jited("...")
+__jited("	ldp	x25, x27, [sp], {{.*}}")
 __naked void private_stack_single_prog(void)
 {
 	asm volatile ("			\
···
 __success
 __arch_x86_64
 __jited("	subq	$0x8, %rsp")
+__arch_arm64
+__jited("	mov	x25, sp")
+__jited("	sub	sp, sp, #0x10")
 __naked void no_private_stack_nested(void)
 {
 	asm volatile ("			\
···
 __jited("	callq	0x{{.*}}")
 __jited("	popq	%r9")
 __jited("	xorl	%eax, %eax")
+__arch_arm64
+__jited("	stp	x25, x27, [sp, {{.*}}]!")
+__jited("	mov	x27, {{.*}}")
+__jited("	movk	x27, {{.*}}, lsl #16")
+__jited("	movk	x27, {{.*}}")
+__jited("	mrs	x10, TPIDR_EL{{[0-1]}}")
+__jited("	add	x27, x27, x10")
+__jited("	add	x25, x27, {{.*}}")
+__jited("	mov	x0, #0x2a")
+__jited("	str	x0, [x27]")
+__jited("	bl	{{.*}}")
+__jited("...")
+__jited("	ldp	x25, x27, [sp], {{.*}}")
 __naked void private_stack_nested_1(void)
 {
 	asm volatile ("				\
···
 __jited("	pushq	%r9")
 __jited("	callq")
 __jited("	popq	%r9")
+__arch_arm64
+__jited("func #1")
+__jited("...")
+__jited("	stp	x25, x27, [sp, {{.*}}]!")
+__jited("	mov	x27, {{.*}}")
+__jited("	movk	x27, {{.*}}, lsl #16")
+__jited("	movk	x27, {{.*}}")
+__jited("	mrs	x10, TPIDR_EL{{[0-1]}}")
+__jited("	add	x27, x27, x10")
+__jited("	add	x25, x27, {{.*}}")
+__jited("	bl	0x{{.*}}")
+__jited("	add	x7, x0, #0x0")
+__jited("	mov	x0, #0x2a")
+__jited("	str	x0, [x27]")
+__jited("	bl	0x{{.*}}")
+__jited("	add	x7, x0, #0x0")
+__jited("	mov	x7, #0x0")
+__jited("	ldp	x25, x27, [sp], {{.*}}")
 __naked void private_stack_callback(void)
 {
 	asm volatile ("			\
···
 __jited("	pushq	%r9")
 __jited("	callq")
 __jited("	popq	%r9")
+__arch_arm64
+__jited("	stp	x29, x30, [sp, #-0x10]!")
+__jited("	mov	x29, sp")
+__jited("	stp	xzr, x26, [sp, #-0x10]!")
+__jited("	mov	x26, sp")
+__jited("	stp	x19, x20, [sp, #-0x10]!")
+__jited("	stp	x21, x22, [sp, #-0x10]!")
+__jited("	stp	x23, x24, [sp, #-0x10]!")
+__jited("	stp	x25, x26, [sp, #-0x10]!")
+__jited("	stp	x27, x28, [sp, #-0x10]!")
+__jited("	mov	x27, {{.*}}")
+__jited("	movk	x27, {{.*}}, lsl #16")
+__jited("	movk	x27, {{.*}}")
+__jited("	mrs	x10, TPIDR_EL{{[0-1]}}")
+__jited("	add	x27, x27, x10")
+__jited("	add	x25, x27, {{.*}}")
+__jited("	mov	x0, #0x2a")
+__jited("	str	x0, [x27]")
+__jited("	mov	x0, #0x0")
+__jited("	bl	0x{{.*}}")
+__jited("	add	x7, x0, #0x0")
+__jited("	ldp	x27, x28, [sp], #0x10")
 int private_stack_exception_main_prog(void)
 {
 	asm volatile ("			\
···
 __jited("	pushq	%r9")
 __jited("	callq")
 __jited("	popq	%r9")
+__arch_arm64
+__jited("	stp	x27, x28, [sp, #-0x10]!")
+__jited("	mov	x27, {{.*}}")
+__jited("	movk	x27, {{.*}}, lsl #16")
+__jited("	movk	x27, {{.*}}")
+__jited("	mrs	x10, TPIDR_EL{{[0-1]}}")
+__jited("	add	x27, x27, x10")
+__jited("	add	x25, x27, {{.*}}")
+__jited("	mov	x0, #0x2a")
+__jited("	str	x0, [x27]")
+__jited("	bl	0x{{.*}}")
+__jited("	add	x7, x0, #0x0")
+__jited("	ldp	x27, x28, [sp], #0x10")
 int private_stack_exception_sub_prog(void)
 {
 	asm volatile ("			\
···
 __success __retval(0)
 __arch_x86_64
 __jited("	movabsq	$0x{{.*}}, %r9")
+__arch_arm64
+__jited("	mrs	x10, TPIDR_EL{{[0-1]}}")
+__jited("	add	x27, x27, x10")
+__jited("	add	x25, x27, {{.*}}")
 int private_stack_async_callback_1(void)
 {
 	struct bpf_timer *arr_timer;
···
 __success __retval(0)
 __arch_x86_64
 __jited("	subq	$0x100, %rsp")
+__arch_arm64
+__jited("	sub	sp, sp, #0x100")
 int private_stack_async_callback_2(void)
 {
 	struct bpf_timer *arr_timer;