similar to: [PATCH v2] v2v: Free XML objects in the correct order.

If you free an xmlDocPtr before any xmlXPathObjectPtrs that reference the doc, you'll get valgrind errors like this: ==7390== Invalid read of size 4 ==7390== at 0x4EB8BC6: xmlXPathFreeNodeSet (xpath.c:4185) ==7390== by 0x4EB8CC5: xmlXPathFreeObject (xpath.c:5492) ==7390== by 0x400A56: main (in /tmp/test) ==7390== Address 0x60c0928 is 8 bytes inside a block of size 120

If you free an xmlDocPtr before any xmlXPathObjectPtr 's that reference the doc, you'll get valgrind errors like this: ==7390== Invalid read of size 4 ==7390== at 0x4EB8BC6: xmlXPathFreeNodeSet (xpath.c:4185) ==7390== by 0x4EB8CC5: xmlXPathFreeObject (xpath.c:5492) ==7390== by 0x400A56: main (in /tmp/test) ==7390== Address 0x60c0928 is 8 bytes inside a block of size

Hi all, Here is a new version of the virt-builder-repository series taking care of Pino's comments. It has also been rebased on recent master. Cédric Bosdonnat (7): mllib: factorize code to add Checksum.get_checksum function Move xml and xpath_helpers OCAML code to mllib mllib: expose libosinfo DB reading functions in mllib builder: rename docs test script builder: add

Hi guys, Here is a v3 of the series, including changes to answer Richard's comments. Cédric Bosdonnat (10): mllib: factorize code to add Checksum.get_checksum function Move xml and xpath_helpers OCAML code to mllib mllib: add Xml.parse_file helper lib/osinfo.c: Extract xml processing into a callback lib: extract osinfo DB traversing API mllib: ocaml wrapper for lib/osinfo

It turns out that RHEL 5 Xen conversions don't work if the source disk is located on a block device. See patch 1/4 for the gory details. This patch series proposes a new tool called virt-v2v-copy-to-local which essentially is a way to make new virt-v2v work like the old virt-v2v, ie. copy first, convert after. Of course this is very slow and would only be used as a last resort, but I

Hi, I am trying to add xmlXPathRegisterNs() to v2v/xml-c.c but I get a seg fault, the signature should be: http://xmlsoft.org/html/libxml-xpathInternals.html#xmlXPathRegisterNs I think I am wrong in CAMLparam and CAMLlocal..., Following is the patch: =================================================================================== diff --git a/v2v/xml-c.c b/v2v/xml-c.c index 4c9bc77..a917c24

On Thursday, 23 March 2017 10:02:47 CEST Cédric Bosdonnat wrote: > This function will allow more OCAML-ish processing of xpath queries > with multiple results. s/OCAML/OCaml/ s/xpath/XPath/ > --- > mllib/xpath_helpers.ml | 9 +++++++++ > mllib/xpath_helpers.mli | 4 ++++ > 2 files changed, 13 insertions(+) > > diff --git a/mllib/xpath_helpers.ml

When importing from VMware via the libvirt driver, the libvirt driver can add an empty source bridge name: <interface type='bridge'> <mac address='00:01:02:03:04:05:06'/> <source bridge=''/> </interface> Replicate what we do on the -i ova path, and map these to "eth0", "eth1" etc. This also includes a bunch

https://bugzilla.redhat.com/show_bug.cgi?id=912499 (especially comments 7 & 10) This patch set is the final fix so that we can access disks in use by other guests when SELinux and sVirt are enabled. Previously such disks were inaccessible because sVirt labels the disks with a random SELinux label to prevent other instances of qemu from being able to read them. So naturally the libguestfs

Link to version 1: https://www.redhat.com/archives/libguestfs/2013-February/thread.html#00122 Changes since version 1: - I've pushed two (of the three) code refactoring patches. The third one proved rather hard to move. - selinuxnorelabel option is no more. Instead there is a second internal API (internal_set_libvirt_selinux_norelabel_disks). - fixed bogus commit message -

To allow other pieces of code to process XML files easily, move the xml.ml* and xpath_helpers.ml* from v2v to mllib. --- docs/C_SOURCE_FILES | 2 +- mllib/Makefile.am | 9 ++++++-- {v2v => mllib}/xml-c.c | 47 +++++++++++++++++++------------------- {v2v => mllib}/xml.ml | 49 ++++++++++++++++++++++------------------ {v2v =>

The subject says ‘xpath_get_nodes()‘, but this function doesn't actually take a unit parameter, so it's better to drop ‘()’. On Thu, Oct 05, 2017 at 04:58:29PM +0200, Cédric Bosdonnat wrote: > + > +let xpath_get_nodes xpathctx expr = > + let obj = Xml.xpath_eval_expression xpathctx expr in > + let nodes = ref [] in > + for i = 0 to Xml.xpathobj_nr_nodes obj - 1 do > +

Hi all, Here is the latest version of the series. Diffs to v10: * Make Index.arch a (string, string option) maybe and use it to guess arch at parse time * Compute the image size at parse time when the template flag is set and the value is missing. * Add virt-repository_main slow test * Other fixes from Richard's comments Cédric Bosdonnat (7): Ignore builder/*.out and *.img

In case you configure libguestfs with a custom QEMU, e.g.: $ ./configure [...] QEMU=/path/to/qemu then the libvirt backend did not use to override it, launching the appliance with the default QEMU for libvirt. This does not change the manual emulator overriding using set-hv. Pino Toscano (2): launch: libvirt: get default QEMU from domcapabilities launch: libvirt: fix custom hypervisor

A couple of patches cleaning up the -i ova code. These are both just refactoring (or should be at any rate). The second patch is best viewed with 'git show -w' to exclude whitespace changes. Rich.

Hi all, Here is the v5 of my patches series applying the latest comments from Pino. Cédric Bosdonnat (10): lib/osinfo.c: Extract xml processing into a callback lib: extract osinfo DB traversing API mllib: ocaml wrapper for lib/osinfo builder: rename docs test script builder: add a template parameter to get_index builder: add Index.write_entry function dib: move do_cp to

Starting with 5.2.0, libvirt has a way to select the firmware by specifying its type, provided configuration files for the firmware are shipped. Currently we start the appliance as UEFI if any of the firmware are found, so instead we can try to just set the firmware type iff: - the libvirt autoselection works - the 'efi' firmware is available The only behaviour change is that the default

In Std_utils we already extend Char and String. These commits take it a little further by extending List and adding a new Option submodule. All basically simple refactoring. Rich.

As I'm now working through the enormous virt-v2v/virt-p2v bug list, we need a high quality set of tests to ensure that we don't accidentally regress some old OS/hypervisor combination while making changes. The test cases are going to be huge, so we cannot possibly distribute them in libguestfs. Furthermore many of them have licensing problems which means we cannot redistribute them at

This is tangentially related to: https://bugzilla.redhat.com/show_bug.cgi?id=1372668 The problem in that bug is that we didn't pass the source CPU model (Sandybridge in that case) through to the target RHV hypervisor. So when the Windows guest booted on the target it gives an error about CPU hardware being disconnected (although it otherwise boots and works fine). This patch series