[Bug mi/26014] New: Ctrl-C is not being passed to remote gdbserver

["equationssolver at yahoo dot com" <sourceware-bugzilla@sourceware.org>]

https://sourceware.org/bugzilla/show_bug.cgi?id=26014

            Bug ID: 26014
           Summary: Ctrl-C is not being passed to remote gdbserver
           Product: gdb
           Version: HEAD
            Status: UNCONFIRMED
          Severity: critical
          Priority: P2
         Component: mi
          Assignee: unassigned at sourceware dot org
          Reporter: equationssolver at yahoo dot com
  Target Milestone: ---

Ctrl-C is not being passed to remote gdbserver and hence there is no remote
stopped packet from the remote gdbserver.
Remote stop packet is require to generate MI record to tell the Eclipse IDE
that target has stopped.

This is an issue related to eclipse IDE where,  if user suspend the remote
gdbserver connection with dprint , then IDE infinitely waits and expect of stop
remote packet reply from gdbserver,  but there is no MI stopped remote packet
and hence IDE suspend button grayed out and user cannot do further debugging.

Looking into gdb traces shows that ^C is not passed to gdbserver that's why
there is no MI record stopped packet in an reply of ^C.  
This is only happened under Windows OS with latest GDB.

This can be reproduce with CLI

i.e
C test program

#1 #include <stdio.h>
#2
#3 int main ()
#4 {
#5   while(1)
#6     printf ("hello world! \n");
#7  return 0;
#8 }

$ gcc -ggdb -g3 -O0  -o test test.c

Launch gdbserver
$ gdbserver :10000 test

Launch gdb
(gdb) file test
Reading symbols from test...done.
(gdb) set dprintf-style call
(gdb) dprintf 6, "print dprintf\n"
Dprintf 1 at 0x63e: file test.c, line 6.
(gdb) set debug remote 1
(gdb) tar rem :10000
(gdb) c

press ^C on terminal

Sending packet:
$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...^CPacket
received: OK
Sending packet:
$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...Packet
received: OK
Sending packet: $z0,7ffff7a22c93,1#cd...Packet received: OK
Sending packet: $z0,7ffff7b18bc3,1#fc...Packet received: OK
Sending packet: $z0,7ffff7a22c20,1#c3...Packet received: OK
Sending packet: $z0,7fffffffdaff,1#c5...Packet received: OK
Quit
(gdb)

In above reproduce able scenario, pressing the ^C, "Quit" the gdb debugging and
^C is not passed to remote gdbserver. 
So how we can improve GDB that it always pass ^C to remote target in order to
create remote stop packet.

-- 
You are receiving this mail because:
You are on the CC list for the bug.