On Thu, Apr 06, 2023 at 11:25:47AM +0100, Jonathan Cameron wrote: > On Fri, 17 Mar 2023 16:54:36 +0000 > Jonathan Cameron via <qemu-devel@nongnu.org> wrote: > > > Michael Tsirkin raised that we have recently had churn in the > > bios-tables-test > > and perhaps it was worth factoring some parts of DSDT out as SSDT files. > > This is an attempt to do that for the entries from pxb-pcie and pxb-cxl > > PCI root bridges. > > > > The main PCI root bridge and related elements are left in DSDT as they > > are present in many more tests than PXB. However things brings some > > complexity as some of the DSDT parts are then dependent on building up > > information whilst creating the PXB entries. The ordering constraints > > of RSDT entries prevent easily generating the new SSDT table first > > (see patch 3) > > > > This series works around that by separating that build up of information > > from > > the build up of the PXB parts of the SSDT. That allows the tables to be > > build in the standard order, based on knowledge that the SSDT parts will > > definitely be built later. > > > > Personally, having tried this, I'm not convinced that the advantages of > > simplifying updates to the test data justify the complexity increase needed. > > However I will add that I have a series adding CXL QTG DSM support form Dave > > Jiang in my tree that will only result in updates to SSDT.cxl after this > > patch > > rather than DSDT.cxl reducing chance of a clash with other changes > > in flight. Hence this is an RFC to find out if people think this is > > a good direction to go in. > > > > Suggested-by: Michael S. Tsirkin <m...@redhat.com> > > https://lore.kernel.org/qemu-devel/20230302055544-mutt-send-email-mst@kernel. > > Michael / all, at first glance at least, is sensible to take forwards? > > Whilst I'm not in a rush on this, I'm carrying a bunch of patches > for next cycle that are on top of this at the moment, so I'm just wondering > whether it makes sense reorder things based on what might land first > / not land at all. > > Thanks, > > Jonathan
Yes, I like this. Igor had some reservations about the split-up. Igor can you comment please? > > > > > > Jonathan Cameron (4): > > hw/acpi: Make Aml and / or crs_range_set optional in build_crs > > tests/acpi: Allow changes to DSDT.cxl/viot and SSDT.cxl/viot > > hw/i386/acpi: Separate PXB related parts of DSDT into an SSDT table. > > tests/acpi: Updated DSDT and SSDT due to move of PXB info to SSDT > > > > hw/acpi/aml-build.c | 75 +++++----- > > hw/i386/acpi-build.c | 249 ++++++++++++++++++++++------------ > > hw/pci-host/gpex-acpi.c | 5 +- > > include/hw/acpi/aml-build.h | 4 +- > > tests/data/acpi/q35/DSDT.cxl | Bin 9673 -> 8474 bytes > > tests/data/acpi/q35/DSDT.viot | Bin 9470 -> 8429 bytes > > tests/data/acpi/q35/SSDT.cxl | Bin 0 -> 1235 bytes > > tests/data/acpi/q35/SSDT.viot | Bin 0 -> 1077 bytes > > 8 files changed, 208 insertions(+), 125 deletions(-) > > create mode 100644 tests/data/acpi/q35/SSDT.cxl > > create mode 100644 tests/data/acpi/q35/SSDT.viot > >
