From: Simon Marchi <simon.marchi@ericsson.com>
To: Pedro Alves <palves@redhat.com>,
Simon Marchi <simon.marchi@polymtl.ca>,
"gdb-patches@sourceware.org" <gdb-patches@sourceware.org>
Subject: Re: [PATCH 3/3] Avoid GDB SIGTTOU on catch exec + set follow-exec-mode new (PR 23368)
Date: Tue, 23 Oct 2018 09:49:00 -0000 [thread overview]
Message-ID: <aaaca3a2-150a-6dc6-7d9e-40a7ae070696@ericsson.com> (raw)
In-Reply-To: <9e8dd186-86bd-5a6a-16a1-5aab96725a7b@redhat.com>
On 2018-10-22 4:56 p.m., Pedro Alves wrote:
> On 10/20/2018 04:38 PM, Simon Marchi wrote:
>> On 2018-10-20 11:13 a.m., Simon Marchi wrote:
>>> On 2018-10-17 1:04 p.m., Pedro Alves wrote:
>>>> I think the "for both inferiors" part is what's dubious here.
>>>>
>>>> The process lives on in the new inferior, but we lost its
>>>> terminal settings! Seems to me that they should be migrated
>>>> from the old inferior to the new one. And then the problem
>>>> sorts itself out, because then the new inferior will have
>>>> target_terminal_state::is_inferior state.
>>>
>>> Yeah that makes sense. This crossed my mind when I look at the issue,
>>> but for some reason I didn't went that route. But now that you say it,
>>> it appears obvious that this should be the fix.
>>>
>>> I just tried copying the inferior::terminal_state value from the old
>>> inferior to the new inferior, and it fixes the problem. But
>>> actually, we should also be transferring all of the struct terminal_info
>>> associated to the inferior, is that right?
>>
>> This would be the updated patch (testing on the buildbot at the moment).
>
> Yes. Looks good. I think you could avoid the
> copying-with-garanteed-deleting by adding a "swap_terminal_info" function
> that swaps the terminal_info and terminal_state between the inferiors.
Like this? (commit message and ChangeLog to be updated)
From a43edf2aada8d5312370429a4d7fad7cfb1ec85c Mon Sep 17 00:00:00 2001
From: Simon Marchi <simon.marchi@polymtl.ca>
Date: Sat, 20 Oct 2018 10:57:33 -0400
Subject: [PATCH] Avoid GDB SIGTTOU on catch exec + set follow-exec-mode new
(PR 23368)
Here's a summary of PR 23368:
#include <unistd.h>
int main (void)
{
char *exec_args[] = { "/bin/ls", NULL };
execve (exec_args[0], exec_args, NULL);
}
$ gdb -nx t -ex "catch exec" -ex "set follow-exec-mode new" -ex run
...
[1] + 13146 suspended (tty output) gdb -q -nx t -ex "catch exec" -ex "set follow-exec-mode new" -ex run
$
Here's what happens: when the inferior execs with "follow-exec-mode
new", we first "mourn" it before creating the new one. This ends up
calling inflow_inferior_exit, which sets the per-inferior terminal state
to "is_ours":
inf->terminal_state = target_terminal_state::is_ours;
At this point, the inferior's terminal_state is is_ours, while the
"reality", tracked by gdb_tty_state, is is_inferior (GDB doesn't own the
terminal).
Later, we continue processing the exec inferior event and decide we want
to stop (because of the "catch exec") and call target_terminal::ours to
make sure we own the terminal. However, we don't actually go to the
target backend to change the settings, because the core thinks that no
inferior owns the terminal (inf->terminal_state is
target_terminal_state::is_ours, as checked in
target_terminal_is_ours_kind, for both inferiors). When something in
readline tries to mess with the terminal settings, it generates a
SIGTTOU.
This patch fixes this by tranferring the state of the terminal from the
old inferior to the new inferior.
gdb/ChangeLog:
* infrun.c (follow_exec): In the follow_exec_mode_new case,
transfer terminal state from old new new inferior.
---
gdb/inflow.c | 16 ++++++++++++++++
gdb/infrun.c | 9 +++++----
gdb/terminal.h | 3 +++
3 files changed, 24 insertions(+), 4 deletions(-)
diff --git a/gdb/inflow.c b/gdb/inflow.c
index e355f4aa9fc..d673bef8b43 100644
--- a/gdb/inflow.c
+++ b/gdb/inflow.c
@@ -700,6 +700,22 @@ copy_terminal_info (struct inferior *to, struct inferior *from)
to->terminal_state = from->terminal_state;
}
+/* See terminal.h. */
+
+void
+swap_terminal_info (inferior *a, inferior *b)
+{
+ terminal_info *info_a
+ = (struct terminal_info *) inferior_data (a, inflow_inferior_data);
+ terminal_info *info_b
+ = (struct terminal_info *) inferior_data (a, inflow_inferior_data);
+
+ set_inferior_data (a, inflow_inferior_data, info_b);
+ set_inferior_data (b, inflow_inferior_data, info_a);
+
+ std::swap (a->terminal_state, b->terminal_state);
+}
+
void
info_terminal_command (const char *arg, int from_tty)
{
diff --git a/gdb/infrun.c b/gdb/infrun.c
index 72e24961767..cedaaecd91a 100644
--- a/gdb/infrun.c
+++ b/gdb/infrun.c
@@ -1189,12 +1189,13 @@ follow_exec (ptid_t ptid, char *exec_file_target)
/* The user wants to keep the old inferior and program spaces
around. Create a new fresh one, and switch to it. */
- /* Do exit processing for the original inferior before adding
- the new inferior so we don't have two active inferiors with
- the same ptid, which can confuse find_inferior_ptid. */
+ /* Do exit processing for the original inferior before setting the new
+ inferior's pid. Having two inferiors with the same pid would confuse
+ find_inferior_p(t)id. */
+ inf = add_inferior_with_spaces ();
+ swap_terminal_info (inf, current_inferior ());
exit_inferior_silent (current_inferior ());
- inf = add_inferior_with_spaces ();
inf->pid = pid;
target_follow_exec (inf, exec_file_target);
diff --git a/gdb/terminal.h b/gdb/terminal.h
index 7774eefcd46..da27e46f6ec 100644
--- a/gdb/terminal.h
+++ b/gdb/terminal.h
@@ -29,6 +29,9 @@ extern void new_tty_postfork (void);
extern void copy_terminal_info (struct inferior *to, struct inferior *from);
+/* Exchange the terminal info and state between inferiors A and B. */
+extern void swap_terminal_info (inferior *a, inferior *b);
+
extern pid_t create_tty_session (void);
/* Set up a serial structure describing standard input. In inflow.c. */
--
2.19.1
next prev parent reply other threads:[~2018-10-23 9:49 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-10-16 3:38 [PATCH 1/3] Pass inferior to terminal_save_inferior Simon Marchi
2018-10-16 3:38 ` [PATCH 2/3] Pass inferior to terminal_inferior Simon Marchi
2018-10-22 20:54 ` Pedro Alves
2018-10-23 9:14 ` Simon Marchi
2018-10-16 3:38 ` [PATCH 3/3] Avoid GDB SIGTTOU on catch exec + set follow-exec-mode new (PR 23368) Simon Marchi
[not found] ` <37bde004-853a-3ccc-3777-03cc43b36147@redhat.com>
2018-10-20 15:14 ` Simon Marchi
2018-10-20 15:38 ` Simon Marchi
2018-10-22 20:56 ` Pedro Alves
2018-10-23 9:49 ` Simon Marchi [this message]
2018-10-23 10:55 ` Pedro Alves
[not found] ` <mvmftwvr8ao.fsf@suse.de>
2018-10-24 14:08 ` Simon Marchi
2018-10-22 20:54 ` [PATCH 1/3] Pass inferior to terminal_save_inferior Pedro Alves
2018-10-23 9:01 ` Simon Marchi
2018-10-23 11:11 ` Pedro Alves
2018-10-23 20:56 ` Simon Marchi
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=aaaca3a2-150a-6dc6-7d9e-40a7ae070696@ericsson.com \
--to=simon.marchi@ericsson.com \
--cc=gdb-patches@sourceware.org \
--cc=palves@redhat.com \
--cc=simon.marchi@polymtl.ca \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox