Hello,
I came across a build issue for qemu-system-native on dunfell branch when
using spice PACKAGECONFIG.
I found that the culprit is the CVE-2022-4144.patch introduced mid-march
and that was merged in upstream
end of november 2022:
https://gitlab.com/qemu-project/qemu/-/commit/6dbbf055148c6f1b7d8a3251a65bd6f3d1e1f622
because it misses a variable declaration:
| CC hw/dma/xlnx-zynq-devcfg.o
| CC hw/dma/xlnx-zdma.o
| CC hw/dma/bcm2835_dma.o
| CC hw/gpio/max7310.o
| CC hw/gpio/pl061.o
| /home/test/qemu-4.2.0/hw/display/qxl.c: In function ‘qxl_phys2virt’:
| /home/test/qemu-4.2.0/hw/display/qxl.c:1508:67: error: ‘size’ undeclared
(first use in this function); did you mean ‘gsize’?
| 1508 | if (!qxl_get_check_slot_offset(qxl, pqxl, &slot, &offset,
size)) {
| |
^~~~
| |
gsize
| /home/test/qemu-4.2.0/hw/display/qxl.c:1508:67: note: each undeclared
identifier is reported only once for each function it appears in
| make: *** [/home/test/qemu-4.2.0/rules.mak:69: hw/display/qxl.o] Error 1
| make: *** Waiting for unfinished jobs....
| ERROR: oe_runmake failed
| WARNING: exit code 1 from a shell command.
Looking more to the qemu git, I think the backported CVE is not an atomic
commit one
and is a part of larger patches series, it needs at least this one:
https://gitlab.com/qemu-project/qemu/-/commit/8efec0ef8bbc1e75a7ebf6e325a35806ece9b39f
to be able to build.
Moreover, I saw 5 commits that were merged by the same issuer and that
seems to
be related to the same code path.
Can someone who knows qemu give advice on what to do ?
Modify the single CVE patch ? take the other commit which corrects the
mistake ? take the whole 5 commits ?
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#179811):
https://lists.openembedded.org/g/openembedded-core/message/179811
Mute This Topic: https://lists.openembedded.org/mt/98120051/21656
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-
