On Mon, 14 Jul 2025 09:09:10 -0400
"Michael S. Tsirkin" <m...@redhat.com> wrote:
> On Wed, Jun 04, 2025 at 12:52:32PM +0100, Alireza Sanaee wrote:
> > From: Yicong Yang <yangyic...@hisilicon.com>
> >
> > Currently we build the PPTT starting from the socket node and each
> > socket will be a separate tree. For a multi-socket system it'll
> > be hard for the OS to know the whole system is homogeneous or not
> > (actually we're in the current implementation) since no parent node
> > to telling the identical implementation informentation. Add a
> > root node for indicating this.
> >
> > Signed-off-by: Yicong Yang <yangyic...@hisilicon.com>
> > Reviewed-by: Jonathan Cameron <jonathan.came...@huawei.com>
> > Signed-off-by: Alireza Sanaee <alireza.san...@huawei.com>
> > ---
> > hw/acpi/aml-build.c | 15 ++++++++++++++-
> > 1 file changed, 14 insertions(+), 1 deletion(-)
> >
> > diff --git a/hw/acpi/aml-build.c b/hw/acpi/aml-build.c
> > index 560cee12a2..76a4157a18 100644
> > --- a/hw/acpi/aml-build.c
> > +++ b/hw/acpi/aml-build.c
> > @@ -2153,12 +2153,25 @@ void build_pptt(GArray *table_data,
> > BIOSLinker *linker, MachineState *ms, int64_t socket_id = -1,
> > cluster_id = -1, core_id = -1; uint32_t socket_offset = 0,
> > cluster_offset = 0, core_offset = 0; uint32_t pptt_start =
> > table_data->len;
> > + uint32_t root_offset;
> > int n;
> > AcpiTable table = { .sig = "PPTT", .rev = 2,
> > .oem_id = oem_id, .oem_table_id =
> > oem_table_id };
> > acpi_table_begin(&table, table_data);
> >
> > + /*
> > + * Build a root node for all the processor nodes. Otherwise
> > when
> > + * building a multi-socket system each socket tree is separated
> > + * and will be hard for the OS like Linux to know whether the
> > + * system is homogeneous.
> > + */
> > + root_offset = table_data->len - pptt_start;
> > + build_processor_hierarchy_node(table_data,
> > + (1 << 0) | /* Physical package */
> > + (1 << 4), /* Identical Implementation */
> > + 0, 0, NULL, 0);
> > +
> > /*
> > * This works with the assumption that cpus[n].props.*_id has
> > been
> > * sorted from top to down levels in
> > mc->possible_cpu_arch_ids(). @@ -2175,7 +2188,7 @@ void
> > build_pptt(GArray *table_data, BIOSLinker *linker, MachineState
> > *ms, build_processor_hierarchy_node(table_data, (1 << 0) | /*
> > Physical package */ (1 << 4), /* Identical Implementation */
> > - 0, socket_id, NULL, 0);
> > + root_offset, socket_id, NULL, 0);
> > }
> >
> > if (mc->smp_props.clusters_supported &&
> > mc->smp_props.has_clusters) {
>
>
> This function is also used by loongarch64, but you do not update the
> loongarch64 expected files:
> https://gitlab.com/mstredhat/qemu/-/jobs/10672661860
Hi Michael,
There are new tests you have brought in the
tree after mine.
https://gitlab.com/mstredhat/qemu/-/commit/9e4f80654cefd051f8f5c220d5447201b6cf1810
I can try to fix this and resend with updated PPTT files for
loongarch64. WDYT?
>
> > --
> > 2.43.0
>
>
