wrong assumptions about pthread_t being numeric

John Spencer maillist-gdbpatches@barfooze.de
Fri Sep 16 23:01:00 GMT 2011


there are a couple of spots in thread handling which assume that 
thread_t (typedef of pthread_t) is of a numeric type.
according to POSIX 
http://pubs.opengroup.org/onlinepubs/007904875/basedefs/sys/types.h.html 
pthread_t is a non-arithmetic type.

thus whenever a "thread id" of type pthread_t is used in a numeric 
context, it invokes undefined behaviour, since pthread_t could be a 
struct, a pointer, etc.

for example, it is implemented as a pointer to a struct in musl libc for 
efficiency.

i basically wanted to fix my compile error and send a patch, but i think 
this should be discussed first.

thread-db.c: In function 'find_one_thread':
thread-db.c:295:7: error: format '%ld' expects type 'long int', but 
argument 3 has type 'thread_t'
thread-db.c: In function 'attach_thread':
thread-db.c:335:7: error: format '%ld' expects type 'long int', but 
argument 3 has type 'thread_t'
thread-db.c:341:9: error: format '%ld' expects type 'long int', but 
argument 2 has type 'thread_t'

$ grep ti_tid `find gdb-7.3.1 -name '*.c'`
gdb-7.3.1/gdb/sol-thread.c:  return BUILD_THREAD (ti.ti_tid, PIDGET (lwp));
gdb-7.3.1/gdb/sol-thread.c:  ptid = BUILD_THREAD (ti.ti_tid, PIDGET 
(inferior_ptid));
gdb-7.3.1/gdb/sol-thread.c:                    ti.ti_tid, ti.ti_lid);
gdb-7.3.1/gdb/linux-thread-db.c:  gdb_assert (ti_p->ti_tid != 0);
gdb-7.3.1/gdb/linux-thread-db.c:  private->tid = ti_p->ti_tid;
gdb-7.3.1/gdb/linux-thread-db.c:  if (ti.ti_tid == 0 && 
target_has_execution)
gdb-7.3.1/gdb/gdbserver/thread-db.c:         ti.ti_tid, ti.ti_lid);
gdb-7.3.1/gdb/gdbserver/thread-db.c:  if (ti.ti_tid == 0)
gdb-7.3.1/gdb/gdbserver/thread-db.c:         ti_p->ti_tid, ti_p->ti_lid);
gdb-7.3.1/gdb/gdbserver/thread-db.c:           ti_p->ti_tid, ti_p->ti_lid);
gdb-7.3.1/gdb/aix-thread.c:      return thrinf.ti_tid;

-- JS



More information about the Gdb-patches mailing list