I had to put this aside for a bit, but am back looking at it.
The build paths QA failure comes from the use of glib-mkenums, which
definitely generates files with absolute paths. I don't see any
options available in its documentation for sysroot or relative
patches, so I'll need to come up with some sort of fixup.
I'm also now seeing a failure on the latest OE core master (and
master-next of meta-virt), like this:
nfig-domain-capabilities-os.c.o.d -o
libvirt-gconfig/libvirt-gconfig-1.0.so.0.4000.0.p/libvirt-gconfig-domain-capabilities-os.c.o
-c ../git/libvirt-gconfig/libvirt-gconfig-domain-capabilities-os.c
| In file included from
/opt/poky/build/tmp/work/core2-64-poky-linux/libvirt-glib/4.0.0-r0/recipe-sysroot/usr/include/glib-2.0/glib.h:52,
| from
/opt/poky/build/tmp/work/core2-64-poky-linux/libvirt-glib/4.0.0-r0/recipe-sysroot/usr/include/glib-2.0/gobject/gbinding.h:30,
| from
/opt/poky/build/tmp/work/core2-64-poky-linux/libvirt-glib/4.0.0-r0/recipe-sysroot/usr/include/glib-2.0/glib-object.h:24,
| from ../git/libvirt-gconfig/libvirt-gconfig.h:27,
| from
../git/libvirt-gconfig/libvirt-gconfig-domain-capabilities-os.c:23:
| ../git/libvirt-gconfig/libvirt-gconfig-domain-capabilities-os.c: In
function 'search_firmwares':
| ../git/libvirt-gconfig/libvirt-gconfig-domain-capabilities-os.c:70:26:
error: pointer targets in passing argument 1 of 'strcmp' differ in
signedness [-Werror=pointer-sign]
| 70 | if (!g_str_equal(node->name, "enum"))
|
/opt/poky/build/tmp/work/core2-64-poky-linux/libvirt-glib/4.0.0-r0/recipe-sysroot/usr/include/glib-2.0/glib/ghash.h:165:39:
note: in definition of macro 'g_str_equal'
| 165 | #define g_str_equal(v1, v2) (strcmp ((v1), (v2)) == 0)
| | ^~
| In file included from
/opt/poky/build/tmp/work/core2-64-poky-linux/libvirt-glib/4.0.0-r0/recipe-sysroot/usr/include/glib-2.0/glib/galloca.h:35,
| from
/opt/poky/build/tmp/work/core2-64-poky-linux/libvirt-glib/4.0.0-r0/recipe-sysroot/usr/include/glib-2.0/glib.h:32:
|
/opt/poky/build/tmp/work/core2-64-poky-linux/libvirt-glib/4.0.0-r0/recipe-sysroot/usr/include/string.h:156:32:
note: expected 'const char *' but argument is of type 'const xmlChar
*' {aka 'const unsigned char *'}
| 156 | extern int strcmp (const char *__s1, const char *__s2)
| | ~~~~~~~~~~~~^~~~
| cc1: all warnings being treated as errors
If you have any cycles free, I'd be interested to know if this is
specific to my builders, or if others are seeing the same thing.
Bruce
On Fri, Nov 4, 2022 at 3:23 AM Markus Volk <[email protected]> wrote:
>
> Am Do, 3. Nov 2022 um 11:23:25 -0400 schrieb Bruce Ashfield
> <[email protected]>:
>
> but libvirt still has other TMPDIR QA warnings that still haven't been fixed,
> so I don't consider this an absolute blocker. I'll have a look and try to fix
> it.
>
>
> Thanks for the help!
>
> Markus
--
- Thou shalt not follow the NULL pointer, for chaos and madness await
thee at its end
- "Use the force Harry" - Gandalf, Star Trek II
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#7695):
https://lists.yoctoproject.org/g/meta-virtualization/message/7695
Mute This Topic: https://lists.yoctoproject.org/mt/94667263/21656
Group Owner: [email protected]
Unsubscribe: https://lists.yoctoproject.org/g/meta-virtualization/unsub
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-
