On 8/16/2021 09:04, Rafael J. Wysocki wrote: > On Fri, Aug 13, 2021 at 6:19 PM Mario Limonciello > <mario.limonciello(a)amd.com&gt; wrote: >> >> A number of systems are showing "hotplug capable" CPUs when they >> are not really hotpluggable. This is because the MADT has extra >> CPU entries to support different CPUs that may be inserted into >> the socket with different numbers of cores. >> >> Starting with ACPI 6.3 the spec has an Online Capable bit in the >> MADT used to determine whether or not a CPU is hotplug capable >> when the enabled bit is not set. >> >> Link: https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fuefi.or... >> Signed-off-by: Mario Limonciello <mario.limonciello(a)amd.com&gt; >> --- >> arch/x86/kernel/acpi/boot.c | 10 ++++++++++ >> include/acpi/actbl2.h | 1 + >> 2 files changed, 11 insertions(+) >> >> Changes from v1->v2: >> * Check the revision field in MADT to determine if it matches the >> bump from ACPI 6.3 as suggested by Hanjun Guo >> * Update description >> >> diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c >> index e55e0c1fad8c..bfa69a5c9c0b 100644 >> --- a/arch/x86/kernel/acpi/boot.c >> +++ b/arch/x86/kernel/acpi/boot.c >> @@ -53,6 +53,8 @@ int acpi_ioapic; >> int acpi_strict; >> int acpi_disable_cmcff; >> >> +bool acpi_support_online_capable; > > Missing static? Ack, thanks. > >> + >> /* ACPI SCI override configuration */ >> u8 acpi_sci_flags __initdata; >> u32 acpi_sci_override_gsi __initdata = INVALID_ACPI_IRQ; >> @@ -138,6 +140,8 @@ static int __init acpi_parse_madt(struct acpi_table_header *table) >> >> pr_debug("Local APIC address 0x%08x\n", madt->address); >> } >> + if (madt->header.revision >= 5) >> + acpi_support_online_capable = true; >> >> default_acpi_madt_oem_check(madt->header.oem_id, >> madt->header.oem_table_id); >> @@ -239,6 +243,12 @@ acpi_parse_lapic(union acpi_subtable_headers * header, const unsigned long end) >> if (processor->id == 0xff) >> return 0; >> >> + /* don't register processors that can not be onlined */ >> + if (acpi_support_online_capable && >> + !(processor->lapic_flags & ACPI_MADT_ENABLED) && >> + !(processor->lapic_flags & ACPI_MADT_ONLINE_CAPABLE)) >> + return 0; >> + >> /* >> * We need to register disabled CPU as well to permit >> * counting disabled CPUs. This allows us to size >> diff --git a/include/acpi/actbl2.h b/include/acpi/actbl2.h >> index 2069ac38a4e2..fae45e383987 100644 >> --- a/include/acpi/actbl2.h >> +++ b/include/acpi/actbl2.h > > The one below is an ACPICA change and I'd prefer it to be integrated > via the upstream ACPICA. > > Could you prepare an ACPICA pull request for just the bit below and > send it via GitHub? Sure thing. http://github.com/acpica/acpica/pull/708/ They said they would take it later this month or next month. Given that, how do you want to proceed with the first part of this? Should I send a 2 patch series that will add the MADT bit to actbl2.h in advance of their next release, or should I wait to resubmit until after their next release and you've brought it into your tree?