Since 6.2 QEMU will assert when SLIC table is passed with
help of -acpitable. This series fixes the issue and adds a test
case for it.
PS: gitlab whining:
* the issue was reported and ivestigated via shiny gitlab issue tracker,
the problem is that all that discussion is buried there and is not stored
in qemu-devel mail list. So when gitlab is gone, so will be all the history
and one won't have (nicely and locally stored) mail archive to search in
a convenient way.
Also I'd notice the report earlier if it were forwarded to qemu-devel.
I wonder if there is a way to bridge issue tracker discussions to mail list?
* another issue is that gitlab hides user's emails, with a bit of detective
work
one can find email if the user has committed a patch via gitlab, but that
doesn't
work for every user. So I can't CC/properly credit reporter when posting
formal
patch.
CC: dlen...@gmail.com
CC: m...@redhat.com
Igor Mammedov (4):
acpi: fix QEMU crash when started with SLIC table
tests: acpi: whitelist expected blobs before changing them
tests: acpi: add SLIC table test
tests: acpi: SLIC: update expected blobs
hw/acpi/core.c | 4 ++--
hw/i386/acpi-build.c | 2 ++
tests/data/acpi/q35/FACP.slic | Bin 0 -> 244 bytes
tests/data/acpi/q35/SLIC.slic | Bin 0 -> 36 bytes
tests/qtest/bios-tables-test.c | 15 +++++++++++++++
5 files changed, 19 insertions(+), 2 deletions(-)
create mode 100644 tests/data/acpi/q35/FACP.slic
create mode 100644 tests/data/acpi/q35/SLIC.slic
--
2.31.1
