From: Michael Snyder <msnyder@vmware.com>
To: Eli Zaretskii <eliz@gnu.org>
Cc: "pedro@codesourcery.com" <pedro@codesourcery.com>,
"gdb-patches@sourceware.org" <gdb-patches@sourceware.org>,
"tromey@redhat.com" <tromey@redhat.com>
Subject: Re: [rfc] 'thread tid' command
Date: Sun, 13 Feb 2011 00:01:00 -0000 [thread overview]
Message-ID: <4D571F42.4040907@vmware.com> (raw)
In-Reply-To: <4D570F07.2090802@vmware.com>
[-- Attachment #1: Type: text/plain, Size: 1480 bytes --]
Michael Snyder wrote:
> Eli Zaretskii wrote:
>>> Date: Fri, 11 Feb 2011 16:32:54 -0800
>>> From: Michael Snyder <msnyder@vmware.com>
>>> CC: "gdb-patches@sourceware.org" <gdb-patches@sourceware.org>
>>>
>>> Pedro Alves wrote:
>>>> On Friday 11 February 2011 22:12:39, Michael Snyder wrote:
>>>>> This is purposefully rough, 'cause I hope for discussion.
>>>>>
>>>>> This is a new command to help manage large thread lists.
>>>>> I started with the idea "I want to find out which thread
>>>>> has target id 12345", then extended it to also handle the
>>>>> new concept of thread names (which thread has name 'foo'),
>>>>> and extra info as well (which thread has extra info that
>>>>> includes the string "xyz").
>>>>>
>>>>> The syntax (I'm open to renaming etc.):
>>>>>
>>>>> thread tid [NAME | TARGET_ID | EXTRA_INFO]
>>>> Shouldn't this be under "info" or "show"?
>>> Suits me, but
>>> 1) I'd prefer it wasn't "show", because there is no "set".
>>> 2) Info what? tid? thread-id?
>> thread show [NAME | TARGET_ID | EXTRA_INFO]
>> thread info [NAME | TARGET_ID | EXTRA_INFO]
>
> How about "thread find"?
Here's a new diff combining both patches (since they sort of relate),
and implementing some of your suggestions.
* info thread can now take a list of thread ids (closer to a thread set)
* thread find (so called) can now take a regular expression.
Tom, since you implemented the thread name, I'd like to know if this
scheme fits in with what you had in mind.
[-- Attachment #2: tid.txt --]
[-- Type: text/plain, Size: 4474 bytes --]
2011-02-10 Michael Snyder <msnyder@msnyder-server.eng.vmware.com>
* thread.c (info_threads_command): Process arg as thread id,
or list of thread ids.
(thread_find_command): New command.
(_initialize_thread): Document argument for info threads.
Document 'thread find' command.
Index: thread.c
===================================================================
RCS file: /cvs/src/src/gdb/thread.c,v
retrieving revision 1.131
diff -u -p -u -p -r1.131 thread.c
--- thread.c 19 Jan 2011 17:21:36 -0000 1.131
+++ thread.c 12 Feb 2011 23:53:18 -0000
@@ -43,6 +43,7 @@
#include "observer.h"
#include "annotate.h"
#include "cli/cli-decode.h"
+#include "gdb_regex.h"
/* Definition of struct thread_info exported to gdbthread.h. */
@@ -957,15 +958,35 @@ No selected thread. See `help thread'.\
/* Print information about currently known threads
- * Note: this has the drawback that it _really_ switches
- * threads, which frees the frame cache. A no-side
- * effects info-threads command would be nicer.
- */
+ Optional ARG is a thread id, or list of thread ids.
+
+ Note: this has the drawback that it _really_ switches
+ threads, which frees the frame cache. A no-side
+ effects info-threads command would be nicer. */
static void
info_threads_command (char *arg, int from_tty)
{
- print_thread_info (uiout, -1, -1);
+ int tid = -1;
+
+ if (arg == NULL || *arg == '\0')
+ {
+ print_thread_info (uiout, -1, -1);
+ return;
+ }
+
+ while (arg != NULL && *arg != '\0')
+ {
+ int tmp_tid
+ = (int) parse_and_eval_long (scan_expression_with_cleanup (&arg,
+ NULL));
+
+ if (tmp_tid != 0)
+ {
+ tid = tmp_tid;
+ print_thread_info (uiout, tid, -1);
+ }
+ }
}
/* Switch from one thread to another. */
@@ -1313,6 +1334,56 @@ thread_name_command (char *arg, int from
info->name = arg ? xstrdup (arg) : NULL;
}
+/* Find thread ids with a name, target pid, or extra info matching ARG. */
+
+static void
+thread_find_command (char *arg, int from_tty)
+{
+ struct thread_info *tp;
+ char *tmp;
+ unsigned long match = 0;
+
+ if (arg == NULL || *arg == '\0')
+ error (_("Command requires an argument."));
+
+ tmp = re_comp (arg);
+ if (tmp != 0)
+ error (_("Invalid regexp (%s): %s"), tmp, arg);
+
+ update_thread_list ();
+ for (tp = thread_list; tp; tp = tp->next)
+ {
+ if (tp->name != NULL && re_exec (tp->name))
+ {
+ printf_filtered ("Thread %d has name '%s'\n", tp->num, tp->name);
+ match++;
+ }
+
+ tmp = target_thread_name (tp);
+ if (tmp != NULL && re_exec (tmp))
+ {
+ printf_filtered ("Thread %d has target name '%s'\n", tp->num, tmp);
+ match++;
+ }
+
+ tmp = target_pid_to_str (tp->ptid);
+ if (tmp != NULL && re_exec (tmp))
+ {
+ printf_filtered ("Thread %d has target id '%s'\n", tp->num, tmp);
+ match++;
+ }
+
+ tmp = target_extra_thread_info (tp);
+ if (tmp != NULL && re_exec (tmp))
+ {
+ printf_filtered ("Thread %d has extra info '%s'\n", tp->num, tmp);
+ match++;
+ }
+ }
+ if (!match)
+ printf_filtered ("No threads match '%s'\n", arg);
+}
+
/* Print notices when new threads are attached and detached. */
int print_thread_events = 1;
static void
@@ -1403,8 +1474,10 @@ _initialize_thread (void)
{
static struct cmd_list_element *thread_apply_list = NULL;
- add_info ("threads", info_threads_command,
- _("IDs of currently known threads."));
+ add_info ("threads", info_threads_command, _("\
+Usage: info threads [ID [ID]...]. Display currently known threads.\n\
+If ID is given, display only that thread or those threads;\
+otherwise, all threads are displayed."));
add_prefix_cmd ("thread", class_run, thread_command, _("\
Use this command to switch between threads.\n\
@@ -1418,11 +1491,15 @@ The new thread ID must be currently know
add_cmd ("all", class_run, thread_apply_all_command,
_("Apply a command to all threads."), &thread_apply_list);
- add_cmd ("name", class_run, thread_name_command,
+ add_cmd ("name", no_class, thread_name_command,
_("Set the current thread's name.\n\
Usage: thread name [NAME]\n\
If NAME is not given, then any existing name is removed."), &thread_cmd_list);
+ add_cmd ("find", no_class, thread_find_command, _("\
+Find thread ids with a name, target pid, or extra info matching REGEXP."),
+ &thread_cmd_list);
+
if (!xdb_commands)
add_com_alias ("t", "thread", class_run, 1);
next prev parent reply other threads:[~2011-02-13 0:01 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-02-11 22:12 Michael Snyder
2011-02-11 23:56 ` Pedro Alves
2011-02-12 0:33 ` Michael Snyder
2011-02-12 7:51 ` Eli Zaretskii
2011-02-12 22:52 ` Michael Snyder
2011-02-13 0:01 ` Michael Snyder [this message]
2011-02-12 7:44 ` Eli Zaretskii
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=4D571F42.4040907@vmware.com \
--to=msnyder@vmware.com \
--cc=eliz@gnu.org \
--cc=gdb-patches@sourceware.org \
--cc=pedro@codesourcery.com \
--cc=tromey@redhat.com \
/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