leak fixes

Pete Zaitcev zaitcev at redhat.com
Tue Dec 14 06:22:29 UTC 2010


On Sun, 12 Dec 2010 11:18:44 +0100
Jim Meyering <jim at meyering.net> wrote:

> While working on allowing dynamic provider management,
> I've been watching for leaks and have finally fixed
> two lingering ones that are unrelated to my work:

> +++ b/rest.c
> @@ -1511,6 +1511,7 @@ proxy_object_post (void *cctx, struct MHD_Connection *conn, const char *url,
>  				MHD_NO,MHD_NO);
>  			if (!resp) {
>  				fprintf(stderr,"MHD_crfd failed\n");
> +				free_ms(ms);
>  				return MHD_NO;
>  			}
>  			MHD_queue_response(conn,rc,resp);

It looks to me like you just run valgrind randomly and then stub
whatever pops up, while many identical places remain (say, in
proxy_bucket_post and control_api_root). This is not the way to
go about it, we need look at this in a comprehensive way.

I am going to start auditing the whole thing.

-- Pete


More information about the iwhd-devel mailing list