FreeChainXenon/binutils-gdb
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Remove support for LinuxThreads and vendor 2.4 kernels w/ backported NPTL

Browse source 
Since we now rely on PTRACE_EVENT_CLONE being available (added in
Linux 2.5.46), we're relying on NPTL.

This commit removes the support for older LinuxThreads, as well as the
workarounds for vendor 2.4 kernels with NPTL backported.

 - Rely on tkill being available.

 - Assume gdb doesn't get cancel signals.

 - Remove code that checks the LinuxThreads restart and cancel signals
   in the inferior.

 - Assume that __WALL is available.

 - Assume that non-leader threads report WIFEXITED.

 - Thus, no longer need to send signal 0 to check whether threads are
   still alive.

 - Update comments throughout.

Tested on x86_64 Fedora 20, native and gdbserver.

gdb/ChangeLog:

	* configure.ac: Remove tkill checks.
	* configure, config.in: Regenerate.
	* linux-nat.c: Remove HAVE_TKILL_SYSCALL check.  Update top level
	comments.
	(linux_nat_post_attach_wait): Remove 'cloned' parameter.  Use
	__WALL.
	(attach_proc_task_lwp_callback): Don't set the cloned flag.
	(linux_nat_attach): Adjust.
	(kill_lwp): Remove HAVE_TKILL_SYSCALL check.  No longer fall back
	to 'kill'.
	(linux_handle_extended_wait): Use __WALL.  Don't set the cloned
	flag.
	(wait_lwp): Use __WALL.  Update comments.
	(running_callback, stop_and_resume_callback): Delete.
	(linux_nat_filter_event): Don't stop and resume all lwps. Don't
	check if the event LWP has previously exited.
	(check_zombie_leaders): Update comments.
	(linux_nat_wait_1): Use __WALL.
	(kill_wait_callback): Don't handle clone processes separately.
	Use __WALL instead.
	(linux_thread_alive): Delete.
	(linux_nat_thread_alive): Return true as long as the LWP is in the
	LWP list.
	(linux_nat_update_thread_list): Assume the kernel supports
	PTRACE_EVENT_CLONE.
	(get_signo): Delete.
	(lin_thread_get_thread_signals): Remove LinuxThreads references.
	No longer check __pthread_sig_restart / __pthread_sig_cancel in
	the inferior.
	* linux-nat.h (struct lwp_info) <cloned>: Delete field.
	* linux-thread-db.c: Update comments.
	(_initialize_thread_db): Remove LinuxThreads references.
	* nat/linux-waitpid.c (my_waitpid): No longer emulate __WALL.
	Pass down flags unmodified.
	* linux-waitpid.h (my_waitpid): Update documentation.

gdb/gdbserver/ChangeLog:

	* linux-low.c (linux_kill_one_lwp): Remove references to
	LinuxThreads.
	(kill_lwp): Remove HAVE_TKILL_SYSCALL check.  No longer fall back
	to 'kill'.
	(linux_init_signals): Delete.
	(initialize_low): Adjust.
	* thread-db.c (thread_db_init): Remove LinuxThreads reference.
This commit is contained in:
Pedro Alves 2015-12-17 14:20:51 +00:00
parent c3c874459b
commit 4a6ed09b0f
12 changed files with 168 additions and 538 deletions
Download patch file Download diff file Expand all files Collapse all files

60
gdb/gdbserver/linux-low.c
View file

@ -878,10 +878,6 @@ linux_create_inferior (char *program, char **allargs)
close_most_fds ();
ptrace (PTRACE_TRACEME, 0, (PTRACE_TYPE_ARG3) 0, (PTRACE_TYPE_ARG4) 0);
#ifndef __ANDROID__ /* Bionic doesn't use SIGRTMIN the way glibc does. */
signal (__SIGRTMIN + 1, SIG_DFL);
#endif
setpgid (0, 0);
/* If gdbserver is connected to gdb via stdio, redirect the inferior's
@ -1183,13 +1179,12 @@ linux_kill_one_lwp (struct lwp_info *lwp)
ptrace(CONT, pid, 0,0) and just resumes the tracee. A better
alternative is to kill with SIGKILL. We only need one SIGKILL
per process, not one for each thread. But since we still support
linuxthreads, and we also support debugging programs using raw
clone without CLONE_THREAD, we send one for each thread. For
years, we used PTRACE_KILL only, so we're being a bit paranoid
about some old kernels where PTRACE_KILL might work better
(dubious if there are any such, but that's why it's paranoia), so
we try SIGKILL first, PTRACE_KILL second, and so we're fine
everywhere. */
support debugging programs using raw clone without CLONE_THREAD,
we send one for each thread. For years, we used PTRACE_KILL
only, so we're being a bit paranoid about some old kernels where
PTRACE_KILL might work better (dubious if there are any such, but
that's why it's paranoia), so we try SIGKILL first, PTRACE_KILL
second, and so we're fine everywhere. */
errno = 0;
kill_lwp (pid, SIGKILL);
@ -3320,8 +3315,6 @@ linux_wait_1 (ptid_t ptid,
stepping - they may require special handling to skip the signal
handler. Also never ignore signals that could be caused by a
breakpoint. */
/* FIXME drow/2002-06-09: Get signal numbers from the inferior's
thread library? */
if (WIFSTOPPED (w)
&& current_thread->last_resume_kind != resume_step
&& (
@ -3663,27 +3656,17 @@ linux_wait (ptid_t ptid,
static int
kill_lwp (unsigned long lwpid, int signo)
{
/* Use tkill, if possible, in case we are using nptl threads. If tkill
fails, then we are not using nptl threads and we should be using kill. */
int ret;
#ifdef __NR_tkill
{
static int tkill_failed;
if (!tkill_failed)
{
int ret;
errno = 0;
ret = syscall (__NR_tkill, lwpid, signo);
if (errno != ENOSYS)
return ret;
tkill_failed = 1;
}
}
#endif
return kill (lwpid, signo);
errno = 0;
ret = syscall (__NR_tkill, lwpid, signo);
if (errno == ENOSYS)
{
/* If tkill fails, then we are not using nptl threads, a
configuration we no longer support. */
perror_with_name (("tkill"));
}
return ret;
}
void
@ -7196,16 +7179,6 @@ static struct target_ops linux_target_ops = {
linux_supports_software_single_step
};
static void
linux_init_signals ()
{
/* FIXME drow/2002-06-09: As above, we should check with LinuxThreads
to find what the cancel signal actually is. */
#ifndef __ANDROID__ /* Bionic doesn't use SIGRTMIN the way glibc does. */
signal (__SIGRTMIN+1, SIG_IGN);
#endif
}
#ifdef HAVE_LINUX_REGSETS
void
initialize_regsets_info (struct regsets_info *info)
@ -7225,7 +7198,6 @@ initialize_low (void)
memset (&sigchld_action, 0, sizeof (sigchld_action));
set_target_ops (&linux_target_ops);
linux_init_signals ();
linux_ptrace_init_warnings ();
sigchld_action.sa_handler = sigchld_handler;