LVM 0.9 vgscan problem

OK, Jan's patch fixed the vgextend problem on my machine, at least. I manually made the changes to the lvm.c that ships with 2.4.0 and recompiled the module, and I succesfully did a few vgextend operations. We're just commenting out one line and adding two lines in a different place. Jan's patch does not appear to be against the version of lvm.c that ships with 2.4.0, so I didn't try to apply it. A patch against the version of lvm.c that ships with 2.4.0 follows (sorry if the format is wrong, I don't usually do this sort of thing):

---------------

@@ -2021,9 +2021,10 @@ static int lvm_do_vg_extend(vg_t *vg_ptr
                for (p = 0; p < vg_ptr->pv_max; p++) {
                        if ( ( pv_ptr = vg_ptr->pv[p]) == NULL) {
                                ret = lvm_do_pv_create(arg, vg_ptr, p);
-                               lvm_do_create_proc_entry_of_pv ( vg_ptr,

pv_ptr);
+/* lvm_do_create_proc_entry_of_pv ( vg_ptr,
pv_ptr); */

                                if ( ret != 0) return ret;
-       
+                               pv_ptr = vg_ptr->pv[p];
+                               lvm_do_create_proc_entry_of_pv ( vg_ptr,
pv_ptr);
                                /* We don't need the PE list
                                   in kernel space like LVs pe_t list */
                                pv_ptr->pe = NULL;

Please note that I didn't write that patch - I didn't even try vgextend with 0.9, yet. The only thing I did was finding the patch in the public cvs archive.

Jan