Re: [RFA] Fix gdb/669 (fwd)

Re: [RFA] Fix gdb/669 (fwd)

[Michael Elizabeth Chastain]

Oh, good.

I've been getting intermittent failures in gdb.mi/gdb669.exp:
From my 2002-12-12 test runs:

  FAIL: gdb.mi/gdb669.exp: console and MI have same number of threads (try 2)
    # gdb gcc binutils gformat
    HEAD 2.95.3 2.13.1 dwarf-2
    HEAD 2.95.3 2.13.1 stabs+
    HEAD 2.95.3 HEAD dwarf-2
    HEAD 2.95.3 HEAD stabs+
    HEAD 2.95.3 binutils-2_13-branch dwarf-2
    HEAD 2.95.3 binutils-2_13-branch stabs+
    HEAD 2.95.3 vendor stabs+

  FAIL: gdb.mi/gdb669.exp: console and MI have same number of threads (try 3)
    # gdb gcc binutils gformat
    HEAD 3.2.1 binutils-2_13-branch stabs+
    HEAD vendor vendor dwarf-2
    HEAD vendor vendor stabs+

Maybe this patch will fix this.

Michael C

Re: [RFA] Fix gdb/669 (fwd)

[Elena Zannoni]

Michael Elizabeth Chastain writes:
 > Oh, good.
 > 

Yes, it should fix it. Let me know if it doesn't.

Elena


 > I've been getting intermittent failures in gdb.mi/gdb669.exp:
 > >From my 2002-12-12 test runs:
 > 
 >   FAIL: gdb.mi/gdb669.exp: console and MI have same number of threads (try 2)
 >     # gdb gcc binutils gformat
 >     HEAD 2.95.3 2.13.1 dwarf-2
 >     HEAD 2.95.3 2.13.1 stabs+
 >     HEAD 2.95.3 HEAD dwarf-2
 >     HEAD 2.95.3 HEAD stabs+
 >     HEAD 2.95.3 binutils-2_13-branch dwarf-2
 >     HEAD 2.95.3 binutils-2_13-branch stabs+
 >     HEAD 2.95.3 vendor stabs+
 > 
 >   FAIL: gdb.mi/gdb669.exp: console and MI have same number of threads (try 3)
 >     # gdb gcc binutils gformat
 >     HEAD 3.2.1 binutils-2_13-branch stabs+
 >     HEAD vendor vendor dwarf-2
 >     HEAD vendor vendor stabs+
 > 
 > Maybe this patch will fix this.
 > 
 > Michael C

Re: [RFA] Fix gdb/669 (fwd)

[Michael Elizabeth Chastain]

FYI,

gdb.mi/gdb669.exp is all PASS all the time now, in gdb HEAD!

Michael C

[RFA] Fix gdb/669 (fwd)

[Keith Seitz]

Ping.

---------- Forwarded message ----------
Date: Tue, 24 Sep 2002 12:42:24 -0700 (PDT)
From: Keith Seitz <keiths@redhat.com>
To: gdb-patches@sources.redhat.com
Subject: [RFA] Fix gdb/669

Hi,

This patch fixes gdb/669, which demonstrates that the MI command 
"thread-list-ids" and the console's "info thread" get out of sync.

This is caused by the MI command not following the console command, which 
prunes old threads and looks for new ones when it is called.

Keith

ChangeLog
2002-09-05  Keith Seitz  <keiths@redhat.com>

	* thread.c (do_captured_list_thread_ids): Call prune_threads and
	target_find_new_threads.

Patch
Index: thread.c
===================================================================
RCS file: /cvs/src/src/gdb/thread.c,v
retrieving revision 1.23
diff -p -r1.23 thread.c
*** thread.c	2 Aug 2002 20:51:21 -0000	1.23
--- thread.c	24 Sep 2002 19:32:40 -0000
*************** do_captured_list_thread_ids (struct ui_o
*** 262,267 ****
--- 262,269 ----
    struct thread_info *tp;
    int num = 0;
  
+   prune_threads ();
+   target_find_new_threads ();
    ui_out_tuple_begin (uiout, "thread-ids");
  
    for (tp = thread_list; tp; tp = tp->next)

Re: [RFA] Fix gdb/669 (fwd)

[Elena Zannoni]

Keith Seitz writes:
 > Ping.
 > 
 > ---------- Forwarded message ----------
 > Date: Tue, 24 Sep 2002 12:42:24 -0700 (PDT)
 > From: Keith Seitz <keiths@redhat.com>
 > To: gdb-patches@sources.redhat.com
 > Subject: [RFA] Fix gdb/669
 > 
 > Hi,
 > 
 > This patch fixes gdb/669, which demonstrates that the MI command 
 > "thread-list-ids" and the console's "info thread" get out of sync.
 > 
 > This is caused by the MI command not following the console command, which 
 > prunes old threads and looks for new ones when it is called.
 > 


Even though this code is in thread.c, which is Michael's domain, it is
only used from mi. It looks right to me.

Michael?

Elena


 > Keith
 > 
 > ChangeLog
 > 2002-09-05  Keith Seitz  <keiths@redhat.com>
 > 
 > 	* thread.c (do_captured_list_thread_ids): Call prune_threads and
 > 	target_find_new_threads.
 > 
 > Patch
 > Index: thread.c
 > ===================================================================
 > RCS file: /cvs/src/src/gdb/thread.c,v
 > retrieving revision 1.23
 > diff -p -r1.23 thread.c
 > *** thread.c	2 Aug 2002 20:51:21 -0000	1.23
 > --- thread.c	24 Sep 2002 19:32:40 -0000
 > *************** do_captured_list_thread_ids (struct ui_o
 > *** 262,267 ****
 > --- 262,269 ----
 >     struct thread_info *tp;
 >     int num = 0;
 >   
 > +   prune_threads ();
 > +   target_find_new_threads ();
 >     ui_out_tuple_begin (uiout, "thread-ids");
 >   
 >     for (tp = thread_list; tp; tp = tp->next)
 > 

Re: [RFA] Fix gdb/669 (fwd)

[Elena Zannoni]

Elena Zannoni writes:
 > Keith Seitz writes:
 >  > Ping.
 >  > 
 >  > ---------- Forwarded message ----------
 >  > Date: Tue, 24 Sep 2002 12:42:24 -0700 (PDT)
 >  > From: Keith Seitz <keiths@redhat.com>
 >  > To: gdb-patches@sources.redhat.com
 >  > Subject: [RFA] Fix gdb/669
 >  > 
 >  > Hi,
 >  > 
 >  > This patch fixes gdb/669, which demonstrates that the MI command 
 >  > "thread-list-ids" and the console's "info thread" get out of sync.
 >  > 
 >  > This is caused by the MI command not following the console command, which 
 >  > prunes old threads and looks for new ones when it is called.
 >  > 
 > 
 > 
 > Even though this code is in thread.c, which is Michael's domain, it is
 > only used from mi. It looks right to me.
 > 
 > Michael?
 > 

Can this be committed? It is a piece of the MI branch merge.



 > Elena
 > 
 > 
 >  > Keith
 >  > 
 >  > ChangeLog
 >  > 2002-09-05  Keith Seitz  <keiths@redhat.com>
 >  > 
 >  > 	* thread.c (do_captured_list_thread_ids): Call prune_threads and
 >  > 	target_find_new_threads.
 >  > 
 >  > Patch
 >  > Index: thread.c
 >  > ===================================================================
 >  > RCS file: /cvs/src/src/gdb/thread.c,v
 >  > retrieving revision 1.23
 >  > diff -p -r1.23 thread.c
 >  > *** thread.c	2 Aug 2002 20:51:21 -0000	1.23
 >  > --- thread.c	24 Sep 2002 19:32:40 -0000
 >  > *************** do_captured_list_thread_ids (struct ui_o
 >  > *** 262,267 ****
 >  > --- 262,269 ----
 >  >     struct thread_info *tp;
 >  >     int num = 0;
 >  >   
 >  > +   prune_threads ();
 >  > +   target_find_new_threads ();
 >  >     ui_out_tuple_begin (uiout, "thread-ids");
 >  >   
 >  >     for (tp = thread_list; tp; tp = tp->next)
 >  > 

Re: [RFA] Fix gdb/669 (fwd)

[Elena Zannoni]

re-ping.....



Keith Seitz writes:
 > Ping.
 > 
 > ---------- Forwarded message ----------
 > Date: Tue, 24 Sep 2002 12:42:24 -0700 (PDT)
 > From: Keith Seitz <keiths@redhat.com>
 > To: gdb-patches@sources.redhat.com
 > Subject: [RFA] Fix gdb/669
 > 
 > Hi,
 > 
 > This patch fixes gdb/669, which demonstrates that the MI command 
 > "thread-list-ids" and the console's "info thread" get out of sync.
 > 
 > This is caused by the MI command not following the console command, which 
 > prunes old threads and looks for new ones when it is called.
 > 
 > Keith
 > 
 > ChangeLog
 > 2002-09-05  Keith Seitz  <keiths@redhat.com>
 > 
 > 	* thread.c (do_captured_list_thread_ids): Call prune_threads and
 > 	target_find_new_threads.
 > 
 > Patch
 > Index: thread.c
 > ===================================================================
 > RCS file: /cvs/src/src/gdb/thread.c,v
 > retrieving revision 1.23
 > diff -p -r1.23 thread.c
 > *** thread.c	2 Aug 2002 20:51:21 -0000	1.23
 > --- thread.c	24 Sep 2002 19:32:40 -0000
 > *************** do_captured_list_thread_ids (struct ui_o
 > *** 262,267 ****
 > --- 262,269 ----
 >     struct thread_info *tp;
 >     int num = 0;
 >   
 > +   prune_threads ();
 > +   target_find_new_threads ();
 >     ui_out_tuple_begin (uiout, "thread-ids");
 >   
 >     for (tp = thread_list; tp; tp = tp->next)
 > 

Re: [RFA] Fix gdb/669 (fwd)

[Mark Kettenis]

   From: Elena Zannoni <ezannoni@redhat.com>
   Date: Thu, 7 Nov 2002 09:58:24 -0500

   re-ping.....

    > 2002-09-05  Keith Seitz  <keiths@redhat.com>
    > 
    > 	* thread.c (do_captured_list_thread_ids): Call prune_threads and
    > 	target_find_new_threads.

This patch is OK although...

    > Patch
    > Index: thread.c
    > ===================================================================
    > RCS file: /cvs/src/src/gdb/thread.c,v
    > retrieving revision 1.23
    > diff -p -r1.23 thread.c
    > *** thread.c	2 Aug 2002 20:51:21 -0000	1.23
    > --- thread.c	24 Sep 2002 19:32:40 -0000
    > *************** do_captured_list_thread_ids (struct ui_o
    > *** 262,267 ****
    > --- 262,269 ----
    >     struct thread_info *tp;
    >     int num = 0;
    >   
    > +   prune_threads ();
    > +   target_find_new_threads ();

... I'd prefer a blank line here.

    >     ui_out_tuple_begin (uiout, "thread-ids");
    >   
    >     for (tp = thread_list; tp; tp = tp->next)
    > 

Re: [RFA] Fix gdb/669 (fwd)

[Elena Zannoni]

Sigh, forgot all about this one, committed now.

Elena


Mark Kettenis writes:
 >    From: Elena Zannoni <ezannoni@redhat.com>
 >    Date: Thu, 7 Nov 2002 09:58:24 -0500
 > 
 >    re-ping.....
 > 
 >     > 2002-09-05  Keith Seitz  <keiths@redhat.com>
 >     > 
 >     > 	* thread.c (do_captured_list_thread_ids): Call prune_threads and
 >     > 	target_find_new_threads.
 > 
 > This patch is OK although...
 > 
 >     > Patch
 >     > Index: thread.c
 >     > ===================================================================
 >     > RCS file: /cvs/src/src/gdb/thread.c,v
 >     > retrieving revision 1.23
 >     > diff -p -r1.23 thread.c
 >     > *** thread.c	2 Aug 2002 20:51:21 -0000	1.23
 >     > --- thread.c	24 Sep 2002 19:32:40 -0000
 >     > *************** do_captured_list_thread_ids (struct ui_o
 >     > *** 262,267 ****
 >     > --- 262,269 ----
 >     >     struct thread_info *tp;
 >     >     int num = 0;
 >     >   
 >     > +   prune_threads ();
 >     > +   target_find_new_threads ();
 > 
 > ... I'd prefer a blank line here.
 > 
 >     >     ui_out_tuple_begin (uiout, "thread-ids");
 >     >   
 >     >     for (tp = thread_list; tp; tp = tp->next)
 >     >