Re: libguestfs-winsupport re-enable NTFS
by Richard W.M. Jones
On Wed, Apr 02, 2025 at 12:35:50PM +0200, AL13N wrote:
> Hi,
>
> I've noticed you had disabled NTFS support from guestmount (and the
> likes), but kept it for v2v.
>
> Can you tell me the reason why it was disabled from guestmount in
> the first place?
>
> How would I re-enable it in the spec file today? the winsupport does
> seem like the mount.ntfs is there, is it blocked in guestfs itself?
That's intentional, we're not going to change it, and I can't
go into the reasons.
Rich.
--
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
Fedora Windows cross-compiler. Compile Windows programs, test, and
build Windows installers. Over 100 libraries supported.
http://fedoraproject.org/wiki/MinGW
3 months
[PATCH libguestfs-common] mltools: Unreference various objects
by Richard W.M. Jones
Valgrind was complaining about a memory leak in this function. From
looking at similar code in libosinfo itself, it appears we ought to be
unrefing a lot more references here.
This error started happening after the new 'test-phony-win2k25.sh' was
added, but it seems to be unrelated to the test itself and must have
been a pre-existing problem that was exposed by the test.
==2959521==
==2959521== HEAP SUMMARY:
==2959521== in use at exit: 43,662,865 bytes in 1,194,159 blocks
==2959521== total heap usage: 3,898,497 allocs, 2,704,338 frees, 2,609,865,939 bytes allocated
==2959521==
==2959521== 353 (40 direct, 313 indirect) bytes in 1 blocks are definitely lost in loss record 2,661 of 3,422
==2959521== at 0x52BFAEC: g_type_create_instance (gtype.c:1929)
==2959521== by 0x52A48B3: g_object_new_internal.part.0 (gobject.c:2606)
==2959521== by 0x52A5EFD: g_object_new_with_properties (gobject.c:2603)
==2959521== by 0x52A6F30: g_object_new (gobject.c:2415)
==2959521== by 0x121A52: v2v_osinfo_os_find_os_by_short_id (libosinfo-c.c:166)
==2959521== by 0x2DEE12: caml_c_call (in /home/rjones/d/virt-v2v/v2v/virt-v2v)
==2959521== by 0x6F4D7BF: ???
==2959521== by 0x6F5543F: ???
==2959521==
==2959521== LEAK SUMMARY:
==2959521== definitely lost: 40 bytes in 1 blocks
==2959521== indirectly lost: 87 bytes in 5 blocks
==2959521== possibly lost: 0 bytes in 0 blocks
==2959521== still reachable: 1,976,140 bytes in 98 blocks
==2959521== suppressed: 35,544,294 bytes in 1,133,259 blocks
==2959521== Reachable blocks (those to which a pointer was found) are not shown.
==2959521== To see them, rerun with: --leak-check=full --show-leak-kinds=all
==2959521==
==2959521== For lists of detected and suppressed errors, rerun with: -s
==2959521== ERROR SUMMARY: 1 errors from 1 contexts (suppressed: 10 from 10)
---
mltools/libosinfo-c.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/mltools/libosinfo-c.c b/mltools/libosinfo-c.c
index 11df1fe..00d74c1 100644
--- a/mltools/libosinfo-c.c
+++ b/mltools/libosinfo-c.c
@@ -164,17 +164,22 @@ v2v_osinfo_os_find_os_by_short_id (value dbv, value osv)
os_list = osinfo_db_get_os_list (OsinfoDb_t_val (dbv));
filter = osinfo_filter_new ();
- osinfo_filter_add_constraint (filter, OSINFO_PRODUCT_PROP_SHORT_ID, String_val (osv));
+ osinfo_filter_add_constraint (filter, OSINFO_PRODUCT_PROP_SHORT_ID,
+ String_val (osv));
list = osinfo_list_new_filtered (OSINFO_LIST(os_list), filter);
if (osinfo_list_get_length (list) == 0) {
g_object_unref (list);
+ g_object_unref (filter);
+ g_object_unref (os_list);
caml_raise_not_found ();
}
os = OSINFO_OS(osinfo_list_get_nth (list, 0));
rv = Val_OsinfoOs_t (dbv, os);
g_object_unref (list);
+ g_object_unref (filter);
+ g_object_unref (os_list);
CAMLreturn (rv);
}
--
2.49.0
3 months
