binutils-gdb/gdb/linux-nat.h
Pedro Alves 8784d56326 Linux: on attach, attach to lwps listed under /proc/$pid/task/
... instead of relying on libthread_db.

I wrote a test that attaches to a program that constantly spawns
short-lived threads, which exposed several issues.  This is one of
them.

On Linux, we need to attach to all threads of a process (thread group)
individually.  We currently rely on libthread_db to list the threads,
but that is problematic, because libthread_db relies on reading data
structures out of the inferior (which may well be corrupted).  If
threads are being created or exiting just while we try to attach, we
may trip on inconsistencies in the inferior's thread list.  To work
around that, when we see a seemingly corrupt list, we currently retry
a few times:

 static void
 thread_db_find_new_threads_2 (ptid_t ptid, int until_no_new)
 {
 ...
   if (until_no_new)
     {
       /* Require 4 successive iterations which do not find any new threads.
	  The 4 is a heuristic: there is an inherent race here, and I have
	  seen that 2 iterations in a row are not always sufficient to
	  "capture" all threads.  */
 ...

That heuristic may well fail, and when it does, we end up with threads
in the program that aren't under GDB's control.  That's obviously bad
and results in quite mistifying failures, like e.g., the process dying
for seeminly no reason when a thread that wasn't attached trips on a
breakpoint.

There's really no reason to rely on libthread_db for this nowadays
when we have /proc mounted.  In that case, which is the usual case, we
can list the LWPs from /proc/PID/task/.  In fact, GDBserver is already
doing this.  The patch factors out that code that knows to walk the
task/ directory out of GDBserver, and makes GDB use it too.

Like GDBserver, the patch makes GDB attach to LWPs and _not_ wait for
them to stop immediately.  Instead, we just tag the LWP as having an
expected stop.  Because we can only set the ptrace options when the
thread stops, we need a new flag in the lwp structure to keep track of
whether we've already set the ptrace options, just like in GDBserver.
Note that nothing issues any ptrace command to the threads between the
PTRACE_ATTACH and the stop, so this is safe (unlike one scenario
described in gdbserver's linux-low.c).

When we attach to a program that has threads exiting while we attach,
it's easy to race with a thread just exiting as we try to attach to
it, like:

  #1 - get current list of threads
  #2 - attach to each listed thread
  #3 - ooops, attach failed, thread is already gone

As this is pretty normal, we shouldn't be issuing a scary warning in
step #3.

When #3 happens, PTRACE_ATTACH usually fails with ESRCH, but sometimes
we'll see EPERM as well.  That happens when the kernel still has the
thread in its task list, but the thread is marked as dead.
Unfortunately, EPERM is ambiguous and we'll get it also on other
scenarios where the thread isn't dead, and in those cases, it's useful
to get a warning.  To distiguish the cases, when we get an EPERM
failure, we open /proc/PID/status, and check the thread's state -- if
the /proc file no longer exists, or the state is "Z (Zombie)" or "X
(Dead)", we ignore the EPERM error silently; otherwise, we'll warn.
Unfortunately, there seems to be a kernel race here.  Sometimes I get
EPERM, and then the /proc state still indicates "R (Running)"...  If
we wait a bit and retry, we do end up seeing X or Z state, or get an
ESRCH.  I thought of making GDB retry the attach a few times, but even
with a 500ms wait and 4 retries, I still see the warning sometimes.  I
haven't been able to identify the kernel path that causes this yet,
but in any case, it looks like a kernel bug to me.  As this just
results failure to suppress a warning that we've been printing since
about forever anyway, I'm just making the test cope with it, and issue
an XFAIL.

gdb/gdbserver/
2015-01-09  Pedro Alves  <palves@redhat.com>

	* linux-low.c (linux_attach_fail_reason_string): Move to
	nat/linux-ptrace.c, and rename.
	(linux_attach_lwp): Update comment.
	(attach_proc_task_lwp_callback): New function.
	(linux_attach): Adjust to rename and use
	linux_proc_attach_tgid_threads.
	(linux_attach_fail_reason_string): Delete declaration.

gdb/
2015-01-09  Pedro Alves  <palves@redhat.com>

	* linux-nat.c (attach_proc_task_lwp_callback): New function.
	(linux_nat_attach): Use linux_proc_attach_tgid_threads.
	(wait_lwp, linux_nat_filter_event): If not set yet, set the lwp's
	ptrace option flags.
	* linux-nat.h (struct lwp_info) <must_set_ptrace_flags>: New
	field.
	* nat/linux-procfs.c: Include <dirent.h>.
	(linux_proc_get_int): New parameter "warn".  Handle it.
	(linux_proc_get_tgid): Adjust.
	(linux_proc_get_tracerpid): Rename to ...
	(linux_proc_get_tracerpid_nowarn): ... this.
	(linux_proc_pid_get_state): New function, factored out from
	(linux_proc_pid_has_state): ... this.  Add new parameter "warn"
	and handle it.
	(linux_proc_pid_is_gone): New function.
	(linux_proc_pid_is_stopped): Adjust.
	(linux_proc_pid_is_zombie_maybe_warn)
	(linux_proc_pid_is_zombie_nowarn): New functions.
	(linux_proc_pid_is_zombie): Use
	linux_proc_pid_is_zombie_maybe_warn.
	(linux_proc_attach_tgid_threads): New function.
	* nat/linux-procfs.h (linux_proc_get_tgid): Update comment.
	(linux_proc_get_tracerpid): Rename to ...
	(linux_proc_get_tracerpid_nowarn): ... this, and update comment.
	(linux_proc_pid_is_gone): New declaration.
	(linux_proc_pid_is_zombie): Update comment.
	(linux_proc_pid_is_zombie_nowarn): New declaration.
	(linux_proc_attach_lwp_func): New typedef.
	(linux_proc_attach_tgid_threads): New declaration.
	* nat/linux-ptrace.c (linux_ptrace_attach_fail_reason): Adjust to
	use nowarn functions.
	(linux_ptrace_attach_fail_reason_string): Move here from
	gdbserver/linux-low.c and rename.
	(ptrace_supports_feature): If the current ptrace options are not
	known yet, check them now, instead of asserting.
	* nat/linux-ptrace.h (linux_ptrace_attach_fail_reason_string):
	Declare.
2015-01-09 11:39:49 +00:00

197 lines
6.6 KiB
C

/* Native debugging support for GNU/Linux (LWP layer).
Copyright (C) 2000-2015 Free Software Foundation, Inc.
This file is part of GDB.
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>. */
#include "target.h"
#include <signal.h>
struct arch_lwp_info;
/* Structure describing an LWP. This is public only for the purposes
of ALL_LWPS; target-specific code should generally not access it
directly. */
struct lwp_info
{
/* The process id of the LWP. This is a combination of the LWP id
and overall process id. */
ptid_t ptid;
/* If this flag is set, we need to set the event request flags the
next time we see this LWP stop. */
int must_set_ptrace_flags;
/* Non-zero if this LWP is cloned. In this context "cloned" means
that the LWP is reporting to its parent using a signal other than
SIGCHLD. */
int cloned;
/* Non-zero if we sent this LWP a SIGSTOP (but the LWP didn't report
it back yet). */
int signalled;
/* Non-zero if this LWP is stopped. */
int stopped;
/* Non-zero if this LWP will be/has been resumed. Note that an LWP
can be marked both as stopped and resumed at the same time. This
happens if we try to resume an LWP that has a wait status
pending. We shouldn't let the LWP run until that wait status has
been processed, but we should not report that wait status if GDB
didn't try to let the LWP run. */
int resumed;
/* The last resume GDB requested on this thread. */
enum resume_kind last_resume_kind;
/* If non-zero, a pending wait status. */
int status;
/* Non-zero if we were stepping this LWP. */
int step;
/* STOPPED_BY_WATCHPOINT is non-zero if this LWP stopped with a data
watchpoint trap. */
int stopped_by_watchpoint;
/* On architectures where it is possible to know the data address of
a triggered watchpoint, STOPPED_DATA_ADDRESS_P is non-zero, and
STOPPED_DATA_ADDRESS contains such data address. Otherwise,
STOPPED_DATA_ADDRESS_P is false, and STOPPED_DATA_ADDRESS is
undefined. Only valid if STOPPED_BY_WATCHPOINT is true. */
int stopped_data_address_p;
CORE_ADDR stopped_data_address;
/* Non-zero if we expect a duplicated SIGINT. */
int ignore_sigint;
/* If WAITSTATUS->KIND != TARGET_WAITKIND_SPURIOUS, the waitstatus
for this LWP's last event. This may correspond to STATUS above,
or to a local variable in lin_lwp_wait. */
struct target_waitstatus waitstatus;
/* Signal wether we are in a SYSCALL_ENTRY or
in a SYSCALL_RETURN event.
Values:
- TARGET_WAITKIND_SYSCALL_ENTRY
- TARGET_WAITKIND_SYSCALL_RETURN */
int syscall_state;
/* The processor core this LWP was last seen on. */
int core;
/* Arch-specific additions. */
struct arch_lwp_info *arch_private;
/* Next LWP in list. */
struct lwp_info *next;
};
/* The global list of LWPs, for ALL_LWPS. Unlike the threads list,
there is always at least one LWP on the list while the GNU/Linux
native target is active. */
extern struct lwp_info *lwp_list;
/* Iterate over each active thread (light-weight process). */
#define ALL_LWPS(LP) \
for ((LP) = lwp_list; \
(LP) != NULL; \
(LP) = (LP)->next)
/* Attempt to initialize libthread_db. */
void check_for_thread_db (void);
int thread_db_attach_lwp (ptid_t ptid);
/* Return the set of signals used by the threads library. */
extern void lin_thread_get_thread_signals (sigset_t *mask);
/* Find process PID's pending signal set from /proc/pid/status. */
void linux_proc_pending_signals (int pid, sigset_t *pending,
sigset_t *blocked, sigset_t *ignored);
extern int lin_lwp_attach_lwp (ptid_t ptid);
extern void linux_stop_lwp (struct lwp_info *lwp);
/* Iterator function for lin-lwp's lwp list. */
struct lwp_info *iterate_over_lwps (ptid_t filter,
int (*callback) (struct lwp_info *,
void *),
void *data);
/* Create a prototype generic GNU/Linux target. The client can
override it with local methods. */
struct target_ops * linux_target (void);
/* Create a generic GNU/Linux target using traditional
ptrace register access. */
struct target_ops *
linux_trad_target (CORE_ADDR (*register_u_offset)(struct gdbarch *, int, int));
/* Register the customized GNU/Linux target. This should be used
instead of calling add_target directly. */
void linux_nat_add_target (struct target_ops *);
/* Register a method to call whenever a new thread is attached. */
void linux_nat_set_new_thread (struct target_ops *, void (*) (struct lwp_info *));
/* Register a method to call whenever a new fork is attached. */
typedef void (linux_nat_new_fork_ftype) (struct lwp_info *parent,
pid_t child_pid);
void linux_nat_set_new_fork (struct target_ops *ops,
linux_nat_new_fork_ftype *fn);
/* Register a method to call whenever a process is killed or
detached. */
typedef void (linux_nat_forget_process_ftype) (pid_t pid);
void linux_nat_set_forget_process (struct target_ops *ops,
linux_nat_forget_process_ftype *fn);
/* Call the method registered with the function above. PID is the
process to forget about. */
void linux_nat_forget_process (pid_t pid);
/* Register a method that converts a siginfo object between the layout
that ptrace returns, and the layout in the architecture of the
inferior. */
void linux_nat_set_siginfo_fixup (struct target_ops *,
int (*) (siginfo_t *,
gdb_byte *,
int));
/* Register a method to call prior to resuming a thread. */
void linux_nat_set_prepare_to_resume (struct target_ops *,
void (*) (struct lwp_info *));
/* Update linux-nat internal state when changing from one fork
to another. */
void linux_nat_switch_fork (ptid_t new_ptid);
/* Store the saved siginfo associated with PTID in *SIGINFO.
Return 1 if it was retrieved successfully, 0 otherwise (*SIGINFO is
uninitialized in such case). */
int linux_nat_get_siginfo (ptid_t ptid, siginfo_t *siginfo);
/* Set alternative SIGTRAP-like events recognizer. */
void linux_nat_set_status_is_event (struct target_ops *t,
int (*status_is_event) (int status));