proc: fix test for "vsyscall=xonly" boot option

Booting with vsyscall=xonly results in the following vsyscall VMA:

ffffffffff600000-ffffffffff601000 --xp ... [vsyscall]


Test does read from fixed vsyscall address to determine if kernel
supports vsyscall page but it doesn't work because, well, vsyscall
page is execute only.

Fix test by trying to execute from the first byte of the page which
contains gettimeofday() stub. This should work because vsyscall
entry points have stable addresses by design.

Alexey, avoiding parsing .config, /proc/config.gz and
/proc/cmdline at all costs.

Link: https://lkml.kernel.org/r/[e-mail]
Signed-off-by: Alexey Dobriyan <[e-mail]>
Cc: Shuah Khan <[e-mail]>
Cc: <[e-mail]>
Signed-off-by: Andrew Morton <[e-mail]>

tools/testing/selftests/proc/proc-pid-vm.c

···
 }
 #endif
 
-static bool g_vsyscall = false;
+/*
+ * 0: vsyscall VMA doesn't exist	vsyscall=none
+ * 1: vsyscall VMA is r-xp		vsyscall=emulate
+ * 2: vsyscall VMA is --xp		vsyscall=xonly
+ */
+static int g_vsyscall;
+static const char *str_vsyscall;
 
-static const char str_vsyscall[] =
+static const char str_vsyscall_0[] = "";
+static const char str_vsyscall_1[] =
 "ffffffffff600000-ffffffffff601000 r-xp 00000000 00:00 0                  [vsyscall]\n";
+static const char str_vsyscall_2[] =
+"ffffffffff600000-ffffffffff601000 --xp 00000000 00:00 0                  [vsyscall]\n";
 
 #ifdef __x86_64__
 static void sigaction_SIGSEGV(int _, siginfo_t *__, void *___)
···
 }
 
 /*
- * vsyscall page can't be unmapped, probe it with memory load.
+ * vsyscall page can't be unmapped, probe it directly.
  */
 static void vsyscall(void)
 {
···
 		act.sa_sigaction = sigaction_SIGSEGV;
 		(void)sigaction(SIGSEGV, &act, NULL);
 
+		/* gettimeofday(NULL, NULL); */
+		asm volatile (
+			"call %P0"
+			:
+			: "i" (0xffffffffff600000), "D" (NULL), "S" (NULL)
+			: "rax", "rcx", "r11"
+		);
+		exit(0);
+	}
+	waitpid(pid, &wstatus, 0);
+	if (WIFEXITED(wstatus) && WEXITSTATUS(wstatus) == 0) {
+		/* vsyscall page exists and is executable. */
+	} else {
+		/* vsyscall page doesn't exist. */
+		g_vsyscall = 0;
+		return;
+	}
+
+	pid = fork();
+	if (pid < 0) {
+		fprintf(stderr, "fork, errno %d\n", errno);
+		exit(1);
+	}
+	if (pid == 0) {
+		struct rlimit rlim = {0, 0};
+		(void)setrlimit(RLIMIT_CORE, &rlim);
+
+		/* Hide "segfault at ffffffffff600000" messages. */
+		struct sigaction act;
+		memset(&act, 0, sizeof(struct sigaction));
+		act.sa_flags = SA_SIGINFO;
+		act.sa_sigaction = sigaction_SIGSEGV;
+		(void)sigaction(SIGSEGV, &act, NULL);
+
 		*(volatile int *)0xffffffffff600000UL;
 		exit(0);
 	}
 	waitpid(pid, &wstatus, 0);
 	if (WIFEXITED(wstatus) && WEXITSTATUS(wstatus) == 0) {
-		g_vsyscall = true;
+		/* vsyscall page is readable and executable. */
+		g_vsyscall = 1;
+		return;
 	}
+
+	/* vsyscall page is executable but unreadable. */
+	g_vsyscall = 2;
 }
 
 int main(void)
···
 	int exec_fd;
 
 	vsyscall();
+	switch (g_vsyscall) {
+	case 0:
+		str_vsyscall = str_vsyscall_0;
+		break;
+	case 1:
+		str_vsyscall = str_vsyscall_1;
+		break;
+	case 2:
+		str_vsyscall = str_vsyscall_2;
+		break;
+	default:
+		abort();
+	}
 
 	atexit(ate);
 
···
 
 	/* Test /proc/$PID/maps */
 	{
-		const size_t len = strlen(buf0) + (g_vsyscall ? strlen(str_vsyscall) : 0);
+		const size_t len = strlen(buf0) + strlen(str_vsyscall);
 		char buf[256];
 		ssize_t rv;
 		int fd;
···
 		rv = read(fd, buf, sizeof(buf));
 		assert(rv == len);
 		assert(memcmp(buf, buf0, strlen(buf0)) == 0);
-		if (g_vsyscall) {
+		if (g_vsyscall > 0) {
 			assert(memcmp(buf + strlen(buf0), str_vsyscall, strlen(str_vsyscall)) == 0);
 		}
 	}
···
 			assert(memmem(buf, rv, S[i], strlen(S[i])));
 		}
 
-		if (g_vsyscall) {
+		if (g_vsyscall > 0) {
 			assert(memmem(buf, rv, str_vsyscall, strlen(str_vsyscall)));
 		}
 	}