cet: update map_shadow_stack syscall number due to kernel code update

[xupengfe]

User space SHSTK has been merged into Linux kernel main line: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=df57721f9a63e8a1fb9b9b2e70de4aa4c7e0cd2e

SHSTK syscall map_shadow_stack number is 453(not 451 or 452) in Linux mainline kernel:
https://github.com/torvalds/linux/blob/0468be89b3fa5b9249cee1097f1d50314950a2df/arch/x86/entry/syscalls/syscall_64.tbl#L377 All test kernel will follow the main line kernel, so update map_shadow_stack to 453 for future kernel tests.

[xupengfe]

The 3 warnings are as expected, please ignore:
"
#61: FILE: /root/test_shadow_stack.c:61:
+#define ARCH_SHSTK_WRSS (1ULL << 1)

WARNING: Prefer __always_inline over attribute((always_inline))
#84: FILE: /root/test_shadow_stack.c:84:
+static inline unsigned long attribute((always_inline)) get_ssp(void)

WARNING: Use of volatile is usually wrong: see Documentation/process/volatile-considered-harmful.rst
#206: FILE: /root/test_shadow_stack.c:206:
+volatile bool segv_triggered;

WARNING: Prefer noinline over attribute((noinline))
#209: FILE: /root/test_shadow_stack.c:209:
+void attribute((noinline)) violate_ss(void)
"
Thanks!

[xupengfe]

/* It's a test code not kernel code and it can't use always_inline. */
static inline unsigned long attribute((always_inline)) get_ssp(void)

/* The volatile is necessary for the tests. */
volatile bool segv_triggered;

/* It's a test code not kernel code and it can't use noinline. */
void attribute((noinline)) violate_ss(void)

[ysun]

Thanks Pengfei's explaination. The patch looks good to me.

[xupengfe]

Thanks Sun Yi's review!