PATCH: Re: Memory leaks.

[Sven Wischnowsky <wischnow@xxxxxxxxxxxxxxxxxxxxxxx>]

Felix Rosencrantz wrote:

> ...
> 
> Memory leak from bin_compadd():
> 	malloc         [rtlib.o]
> 	zcalloc        [mem.c:469]
> 	parse_cmatcher [complete.c:294]
> 	bin_compadd    [complete.c:580]
> 	execbuiltin    [builtin.c:368]
> 	execcmd        [exec.c:2257]

Oops. parse_cmatcher() should use heap memory, maybe this got confused 
when removing useheap.

> Another leak in bin_compadd:
>         malloc         [rtlib.o]
>         zalloc         [mem.c:453]
>         ztrdup         [mem.c:530]
>         bin_compadd    [complete.c:574]
>         execbuiltin    [builtin.c:368]
>         execcmd        [exec.c:2257]
> 
> I think this is due to the fact that we do a tricat on complete.c:572, without
> freeing the previous values of mstr.

Right.

> Memory leak from addvars():
> 	malloc         [rtlib.o]
> 	zalloc         [mem.c:453]
> 	ztrdup         [mem.c:530]
> 	addvars        [exec.c:1510]
> 	execsimple     [exec.c:750]
> 	execlist       [exec.c:801]

Hmm. Maybe we get this because some parameter setfn() neither uses nor 
frees the string it gets.

> Memory leak out of mkautofn():
> 	malloc         [rtlib.o]
> 	zalloc         [mem.c:453]
> 	mkautofn       [builtin.c:2161]
> 	bin_functions  [builtin.c:2143]
> 	execbuiltin    [builtin.c:368]
> 	execcmd        [exec.c:2257]

I can't see where this comes from. mkautofn() creates the
autofn-program wich is then freed in loadautofn() (or
freeshfuncnode(), with ksh-autoloading).

All these autofn-progs won't be freed at the end, though (together
with many other things).

> ...
> 
> Memory leak from memory allocated in permmatches():
> 	malloc         [rtlib.o]
> 	zcalloc        [mem.c:469]
> 	permmatches    [compcore.c:2771]
> 	get_nmatches   [complete.c:1057]
> 	getstrvalue    [params.c:1367]
> 	getarg         [params.c:967]

Whoa. Right.

> Memory leak from set_gmatcher():
>         malloc         [rtlib.o]
>         zcalloc        [mem.c:469]
>         parse_cmatcher [complete.c:294]
>         set_gmatcher   [compctl.c:313]
>         get_gmatcher   [compctl.c:341]
>         bin_compctl    [compctl.c:1587]

This is the same one as the first.

> (I guess this is still in my startup files...)

Tststs ;-)

Bye
 Sven

Index: Src/Zle/compcore.c
===================================================================
RCS file: /cvsroot/zsh/zsh/Src/Zle/compcore.c,v
retrieving revision 1.27
diff -u -r1.27 compcore.c
--- Src/Zle/compcore.c	2000/05/31 09:56:12	1.27
+++ Src/Zle/compcore.c	2000/06/02 08:07:26
@@ -2879,6 +2879,7 @@
 
 	for (m = g->matches; *m; m++)
 	    freematch(*m, g->nbrbeg, g->nbrend);
+	free(g->matches);
 
 	if (g->ylist)
 	    freearray(g->ylist);
Index: Src/Zle/complete.c
===================================================================
RCS file: /cvsroot/zsh/zsh/Src/Zle/complete.c,v
retrieving revision 1.10
diff -u -r1.10 complete.c
--- Src/Zle/complete.c	2000/06/02 01:54:33	1.10
+++ Src/Zle/complete.c	2000/06/02 08:07:26
@@ -291,7 +291,7 @@
 	if (err)
 	    return pcm_err;
 
-	n = (Cmatcher) zcalloc(sizeof(*ret));
+	n = (Cmatcher) hcalloc(sizeof(*ret));
 	n->next = NULL;
 	n->flags = fl;
 	n->line = line;
@@ -568,9 +568,11 @@
 		    return 1;
 		}
 		if (dm) {
-		    if (mstr)
-			mstr = tricat(mstr, " ", m);
-		    else
+		    if (mstr) {
+			char *tmp = tricat(mstr, " ", m);
+			zsfree(mstr);
+			mstr = tmp;
+		    } else
 			mstr = ztrdup(m);
 		    m = NULL;
 		}

--
Sven Wischnowsky                         wischnow@xxxxxxxxxxxxxxxxxxxxxxx