diff options
author | Michael S. Tsirkin <mst@redhat.com> | 2023-03-02 01:45:28 -0500 |
---|---|---|
committer | Michael S. Tsirkin <mst@redhat.com> | 2023-03-02 19:13:52 -0500 |
commit | ee92a56b08d0b59016a4a9bc1bf3a3de1fbe3956 (patch) | |
tree | f2eeb17e5a59811624de5c30f5f01d1eaa0bf690 /docs/spin | |
parent | e16add2b6b888faa2591de035b6cda5f5aef7ae7 (diff) | |
download | qemu-ee92a56b08d0b59016a4a9bc1bf3a3de1fbe3956.zip qemu-ee92a56b08d0b59016a4a9bc1bf3a3de1fbe3956.tar.gz qemu-ee92a56b08d0b59016a4a9bc1bf3a3de1fbe3956.tar.bz2 |
tests/data/acpi/virt: drop (most) duplicate files.
When virt ACPI files were added, lots of duplicates were created because
we forgot that there's a no-prefix fallback: e.g. if
tests/data/acpi/virt/APIC.memhp is not there then test will use
tests/data/acpi/virt/APIC.
Drop these.
These were found with
$find tests/data/acpi/ -type f -exec sha256sum '{}' ';'|sort -d|uniq -w 64 --all-repeated=separate
(trick: -d does a dictionary sort so a no-suffix file ends up first).
Note: there are still a bunch of issues with duplicates left even after this.
First pc and q35 are often identical.
Second, sometimes files are identical but not identical to the default
fallback, e.g.
tests/data/acpi/pc/SLIT.cphp and tests/data/acpi/pc/SLIT.memhp
or
tests/data/acpi/q35/HMAT.acpihmat-noinitiator and tests/data/acpi/virt/HMAT.acpihmatvirt
Finding a way to deduplicate these is still a TODO item - softlinks
maybe?
We also need to make rebuild-expected-aml.sh smarter about not creating
these duplicates in the 1st place.
And maybe we should use softlinks instead of relying on a fallback
to make it explicit what version does each test expect?
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Diffstat (limited to 'docs/spin')
0 files changed, 0 insertions, 0 deletions