21.06.2013 14:38, Gerd Hoffmann wrote:
> Signed-off-by: Gerd Hoffmann <kra...@redhat.com>
> ---
>  util/qemu-sockets.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/util/qemu-sockets.c b/util/qemu-sockets.c
> index fdd8dc4..364bd8c 100644
> --- a/util/qemu-sockets.c
> +++ b/util/qemu-sockets.c
> @@ -855,7 +855,7 @@ SocketAddress *socket_parse(const char *str, Error **errp)
>  {
>      SocketAddress *addr = NULL;
>  
> -    addr = g_new(SocketAddress, 1);
> +    addr = g_new0(SocketAddress, 1);

While at it we can remove the =NULL assignment too, guess, someting like this:

--- a/util/qemu-sockets.c
+++ b/util/qemu-sockets.c
@@ -848,9 +848,7 @@ int unix_nonblocking_connect(const char *path,

 SocketAddress *socket_parse(const char *str, Error **errp)
 {
-    SocketAddress *addr = NULL;
-
-    addr = g_new(SocketAddress, 1);
+    SocketAddress *addr = addr = g_new0(SocketAddress, 1);
     if (strstart(str, "unix:", NULL)) {
         if (str[5] == '\0') {
             error_setg(errp, "invalid Unix socket address");

Is that okay with you? :)

Not that it matter much actually (I guess gcc may optimize it
out entirely by its own already).

And not that the original issue is a big issue really, because
in each case each relevant field is initialized.  It's still
nice to see stuff clean in debugger and other places, but the
code which actually uses this struct should work fine without
the change.

/mjt

Reply via email to