OpenE2K
/
binutils-gdb
12
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

* lynx-nat.c (child_wait): Don't restart new threads and loop 

to the top of child_wait; let the machine independent code in
        wait_for_inferior deal with new thread notifications.

lynx-6100 (partial)
Fixes SEGV when multi-threaded program hits a breakpoint in a new
thread.
This commit is contained in:
Jeff Law 1995-06-25 02:33:33 +00:00
parent cd5431634a
commit 1657d50b38
2 changed files with 15 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
gdb/ChangeLog
View File

@ -1,3 +1,9 @@
Sat Jun 24 19:27:37 1995 Jeffrey A. Law <law@rtl.cygnus.com>
* lynx-nat.c (child_wait): Don't restart new threads and loop
to the top of child_wait; let the machine independent code in
wait_for_inferior deal with new thread notifications.
Fri Jun 23 11:51:58 1995 Kung Hsu <kung@mexican.cygnus.com>
* remote-nindy (nindy_load): Put in target specific load, it's

15
gdb/lynx-nat.c
View File

@ -649,13 +649,16 @@ child_wait (pid, ourstatus)
if (realsig == SIGNEWTHREAD)
{
/* Simply ignore new thread notification, as we can't do anything
useful with such threads. All ptrace calls at this point just
fail for no apparent reason. The thread will eventually get a
real signal when it becomes real. */
child_resume (pid, 0, TARGET_SIGNAL_0);
continue;
/* It's a new thread notification. Nothing to do here since
the machine independent code in wait_for_inferior will
add the thread to the thread list and restart the thread
when pid != inferior_pid and pid is not in the thread
list. We don't even want to much with realsig -- the
code in wait_for_inferior expects SIGTRAP. */
;
}
else
error ("Signal for unknown thread was not SIGNEWTHREAD");
}
#ifdef SPARC