qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
To: mst@redhat.com, imammedo@redhat.com, ani@anisinha.ca,
	jean-philippe@linaro.org, qemu-devel@nongnu.org
Subject: [PATCH v2 0/6] hw/acpi/viot: generate stable VIOT ACPI tables
Date: Sun, 22 May 2022 16:50:02 +0100	[thread overview]
Message-ID: <20220522155008.21832-1-mark.cave-ayland@ilande.co.uk> (raw)

I was working away at some improvements for PS2 devices when I noticed that one
small change to the instantiation of a PS2 mouse device caused a regression in
tests/qtest/bios-tables-test, specifically the /x86_64/acpi/q35/viot subtest.

Closer examination of the failed test output showed the problem was that the
order of the PCI host bridge entries had changed within the table causing the
generated binary to fail to match the version in tests/data/acpi/q35/VIOT.viot.

The error occurs because there is no guarantee in the order of PCI host bridges
being returned from object_child_foreach_recursive() used within
hw/acpi/viot.c's build_viot() function, so any change to the QOM tree can
potentially change the generated ACPI VIOT table ordering and cause the
regression tests to fail.

Fortunately the solution is fairly easy: change build_viot() to build an array
of PCI host bridges and then sort them first before generating the final ACPI
VIOT table. I've chosen to sort the PCI host bridges based upon the min_bus
number which seems to work okay here.

The changes in this patchset were heavily inspired by the build_iort() function
in hw/arm/virt-acpi-build.c which already does the right thing here. Patches 1-5
make the required changes before patch 6 updates the VIOT binary to match the
updated ACPI VIOT table ordering.

Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>

v2:
- Rebase onto master
- Rename pci_host_bridges() to enumerate_pci_host_bridges() in patch 1
- Change return type of pci_host_range_compare() from int to gint in patch 5
- Tweak subject line in patch 5: s/PCI host bus/PCI host bridge/
- Add Acked-by and Reviewed-by tags from Ani


Mark Cave-Ayland (6):
  hw/acpi/viot: rename build_pci_range_node() to
    enumerate_pci_host_bridges()
  hw/acpi/viot: move the individual PCI host bridge entry generation to
    a new function
  hw/acpi/viot: build array of PCI host bridges before generating VIOT
    ACPI table
  tests/acpi: virt: allow VIOT acpi table changes
  hw/acpi/viot: sort VIOT ACPI table entries by PCI host bridge min_bus
  tests/acpi: virt: update golden masters for VIOT

 hw/acpi/viot.c                | 107 +++++++++++++++++++++-------------
 tests/data/acpi/q35/VIOT.viot | Bin 112 -> 112 bytes
 2 files changed, 68 insertions(+), 39 deletions(-)

-- 
2.20.1



             reply	other threads:[~2022-05-22 15:53 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-22 15:50 Mark Cave-Ayland [this message]
2022-05-22 15:50 ` [PATCH v2 1/6] hw/acpi/viot: rename build_pci_range_node() to enumerate_pci_host_bridges() Mark Cave-Ayland
2022-05-23  6:32   ` Ani Sinha
2022-05-22 15:50 ` [PATCH v2 2/6] hw/acpi/viot: move the individual PCI host bridge entry generation to a new function Mark Cave-Ayland
2022-05-22 15:50 ` [PATCH v2 3/6] hw/acpi/viot: build array of PCI host bridges before generating VIOT ACPI table Mark Cave-Ayland
2022-05-22 15:50 ` [PATCH v2 4/6] tests/acpi: virt: allow VIOT acpi table changes Mark Cave-Ayland
2022-05-22 15:50 ` [PATCH v2 5/6] hw/acpi/viot: sort VIOT ACPI table entries by PCI host bridge min_bus Mark Cave-Ayland
2022-05-22 15:50 ` [PATCH v2 6/6] tests/acpi: virt: update golden masters for VIOT Mark Cave-Ayland

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220522155008.21832-1-mark.cave-ayland@ilande.co.uk \
    --to=mark.cave-ayland@ilande.co.uk \
    --cc=ani@anisinha.ca \
    --cc=imammedo@redhat.com \
    --cc=jean-philippe@linaro.org \
    --cc=mst@redhat.com \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).