Re: QPassSignals patch to go with proposed protocol

[Daniel Jacobowitz <drow@false.org>]

On Wed, Nov 15, 2006 at 08:17:04PM +0200, Eli Zaretskii wrote:
> This is okay with me, thanks.

Thanks, checked in.  Since we just got a query about this functionality
on gdb@ (ironic - first one I can remember seeing, half a day before
checking it in) I wonder if it ought to be mentioned in NEWS, as
attached?

-- 
Daniel Jacobowitz
CodeSourcery

2006-11-16  Daniel Jacobowitz  <dan@codesourcery.com>

	* NEWS: Mention QPassSignals.

Index: NEWS
===================================================================
RCS file: /cvs/src/src/gdb/NEWS,v
retrieving revision 1.199
diff -u -p -r1.199 NEWS
--- NEWS	14 Nov 2006 21:53:59 -0000	1.199
+++ NEWS	16 Nov 2006 15:12:31 -0000
@@ -21,6 +21,9 @@ broken in GDB 6.3, 6.4, and 6.5.
 * Support for GNU/Linux Thread Local Storage (TLS, per-thread variables) no
 longer requires symbolic debug information (e.g. DWARF-2).
 
+* The GDB remote stub, gdbserver, now has lower overhead for high 
+frequency signals (e.g. SIGALRM) via the QPassSignals packet.
+
 * New commands
 
 set substitute-path
@@ -59,6 +62,10 @@ qXfer:auxv:read:
   Fetch an OS auxilliary vector from the remote stub.  This packet is a
   more efficient replacement for qPart:auxv:read.
 
+QPassSignals:
+  Ignore the specified signals; pass them directly to the debugged program
+  without stopping other threads or reporting them to GDB.
+
 * Removed remote packets
 
 qPart:auxv:read: