Eiichi Tsukata found a small bug from the fixup of the stack code

Removing ULONG_MAX as the marker for the user stack trace end,
 made the tracing code not know where the end is. The end is now
 marked with a zero (NULL) pointer. Eiichi fixed this in the tracing
 code.
 -----BEGIN PGP SIGNATURE-----
 
 iIoEABYIADIWIQRRSw7ePDh/lE+zeZMp5XQQmuv6qgUCXTHsuRQccm9zdGVkdEBn
 b29kbWlzLm9yZwAKCRAp5XQQmuv6qgETAQDqRtu1KhJM6ujNlPY1aw6e9ncDAqWn
 6GaumMgAdBqEcAEAxJSjr5UlzXuJsCjUjwE0txLfTscyNwljKW77h4/WNwA=
 =bwtH
 -----END PGP SIGNATURE-----

Merge tag 'trace-v5.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace

Pull tracing fix from Steven Rostedt:
 "Eiichi Tsukata found a small bug from the fixup of the stack code

  Removing ULONG_MAX as the marker for the user stack trace end, made
  the tracing code not know where the end is. The end is now marked with
  a zero (NULL) pointer. Eiichi fixed this in the tracing code"

* tag 'trace-v5.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
  tracing: Fix user stack trace "??" output
This commit is contained in:
Linus Torvalds 2019-07-19 12:18:46 -07:00
commit 41ba485ef1

View File

@ -1109,17 +1109,10 @@ static enum print_line_t trace_user_stack_print(struct trace_iterator *iter,
for (i = 0; i < FTRACE_STACK_ENTRIES; i++) {
unsigned long ip = field->caller[i];
if (ip == ULONG_MAX || trace_seq_has_overflowed(s))
if (!ip || trace_seq_has_overflowed(s))
break;
trace_seq_puts(s, " => ");
if (!ip) {
trace_seq_puts(s, "??");
trace_seq_putc(s, '\n');
continue;
}
seq_print_user_ip(s, mm, ip, flags);
trace_seq_putc(s, '\n');
}