Commit Graph

3 Commits

Author SHA1 Message Date
Samuel Thibault
f9c3cf2864 support: Make support_process_state_wait wait less
Tests such as posix/tst-waitid.c make heavy use of
support_process_state_wait, and thus on non-Linux where it falls back
to sleeping, a 2s sleep makes such test time out, while 1s remains
fine enough.
2020-12-28 23:17:56 +01:00
Florian Weimer
cea56af185 support: Change xgetline to return 0 on EOF
The advantage is that the buffer will always contain the number
of characters as returned from the function, which allows one to use
a sequence like

  /* No more audit module output.  */
  line_length = xgetline (&buffer, &buffer_length, fp);
  TEST_COMPARE_BLOB ("", 0, buffer, line_length);

to check for an expected EOF, while also reporting any unexpected
extra data encountered.

Reviewed-by: Carlos O'Donell <carlos@redhat.com>
2020-04-03 16:26:10 +02:00
Adhemerval Zanella
dfe9aa9156 support: Add support_process_state_wait
It allows parent process to wait for child state using a polling
strategy over procfs on Linux.  The polling is used over ptrace to
avoid the need to handle signals on the target pid and to handle some
system specific limitation (such as YAMA).

The polling has some limitations, such as resource consumption due
the procfs read in a loop and the lack of synchronization after the
state is obtained.

The interface idea is to simplify some sleep synchronization waitid
tests and is to reduce timeouts by replacing arbitrary waits.
2020-02-27 15:52:35 -03:00