Re: kern/52117: ptrace(2) PTRACE_FORK fails on some platforms

["Kamil Rytarowski" <n54%gmx.com@localhost>]

The following reply was made to PR kern/52117; it has been noted by GNATS.

From: "Kamil Rytarowski" <n54%gmx.com@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: 
Subject: Re: kern/52117: ptrace(2) PTRACE_FORK fails on some platforms
Date: Tue, 28 Mar 2017 14:56:53 +0200

 Example taken from evbarm:
 
 Test case: kernel/t_ptrace_wait4/fork1
 
 Duration: 0.030507 seconds
 Termination reason
 
 FAILED: /usr/src/tests/kernel/t_ptrace_wait.h:295: Reported exited process
 Standard output stream
 
 Before forking process PID=12298
 Parent process PID=12298, child's PID=19554
 Before calling PT_TRACE_ME from child 19554
 Before calling wait4() for the child
 Before raising Stopped (signal) from child
 Enable PTRACE_FORK in EVENT_MASK for the child 19554
 Before resuming the child process where it left off and without signal to be sent
 Before calling wait4() for the child 19554
 Reported PTRACE_FORK event with forkee 11702
 Before calling wait4() for the forkee 11702 of the child 19554
 
 
 It means that fork(2) must generate two SIGTAP events:
  - first for child (traced parent) with TRAP_CHLD, pe_report_event=PTRACE_FORK, pe_other_pid=child2
  - second for child2 (traced child of child) with TRAP_CHLD, pe_report_event=PTRACE_FORK, pe_other_pid=child
 
 The test waits infinitely for the second SIGRAP.
 There is a hand-shake for forking.
 
 Debuggers calls wait*(2) twice, one after another, without operation of resuming
 its traced processes between the wait*(2) calls.