From: "Michael S. Tsirkin" <mst@redhat.com>
To: Igor Mammedov <imammedo@redhat.com>
Cc: Marcel Apfelbaum <marcel@redhat.com>,
qemu-devel@nongnu.org, rkrcmar@redhat.com, ehabkost@redhat.com,
drjones@redhat.com, armbru@redhat.com
Subject: Re: [Qemu-devel] [PATCH 20/33] pc: q35: initialize new CPU hotplug hw
Date: Tue, 31 May 2016 15:52:35 +0300 [thread overview]
Message-ID: <20160531155201-mutt-send-email-mst@redhat.com> (raw)
In-Reply-To: <20160531145103.13db12ac@nial.brq.redhat.com>
On Tue, May 31, 2016 at 02:51:03PM +0200, Igor Mammedov wrote:
> On Tue, 31 May 2016 13:21:20 +0300
> Marcel Apfelbaum <marcel@redhat.com> wrote:
>
> > On 05/31/2016 01:06 PM, Igor Mammedov wrote:
> > > On Mon, 30 May 2016 22:02:31 +0300
> > > Marcel Apfelbaum <marcel@redhat.com> wrote:
> > >
> > >> On 05/17/2016 05:43 PM, Igor Mammedov wrote:
> > >>> add necessary wiring to init new CPU hotplug hardware
> > >>> if ICH9-LPC.cpu-hotplug-legacy is "off".
> > >>> Set ICH9-LPC.cpu-hotplug-legacy to "off" by default and
> > >>> switch legacy hotplug to enabled only for 2.6 and older
> > >>> machine types.
> > >>>
> > >>> Signed-off-by: Igor Mammedov <imammedo@redhat.com>
> > >>> ---
> > >>> v1:
> > >>> - drop ICH9-LPC.cpu-hotplug property
> > >>> ---
> > >>> hw/acpi/ich9.c | 42 +++++++++++++++++++++++++++++++++++++-----
> > >>> include/hw/acpi/ich9.h | 6 +++++-
> > >>> include/hw/compat.h | 6 +++++-
> > >>> 3 files changed, 47 insertions(+), 7 deletions(-)
> > >>>
> > >>> diff --git a/hw/acpi/ich9.c b/hw/acpi/ich9.c
> > >>> index 121e30c..b0285b9 100644
> > >>> --- a/hw/acpi/ich9.c
> > >>> +++ b/hw/acpi/ich9.c
> > >>> @@ -170,6 +170,24 @@ static const VMStateDescription vmstate_memhp_state = {
> > >>> }
> > >>> };
> > >>>
> > >>> +static bool vmstate_test_use_cpuhp(void *opaque)
> > >>> +{
> > >>> + ICH9LPCPMRegs *s = opaque;
> > >>> + return !s->cpu_hotplug_legacy;
> > >>> +}
> > >>> +
> > >>> +static const VMStateDescription vmstate_cpuhp_state = {
> > >>> + .name = "ich9_pm/cpuhp",
> > >>> + .version_id = 1,
> > >>> + .minimum_version_id = 1,
> > >>> + .minimum_version_id_old = 1,
> > >>> + .needed = vmstate_test_use_cpuhp,
> > >>> + .fields = (VMStateField[]) {
> > >>> + VMSTATE_CPU_HOTPLUG(cpuhp.state, ICH9LPCPMRegs),
> > >>> + VMSTATE_END_OF_LIST()
> > >>> + }
> > >>> +};
> > >>> +
> > >>> static bool vmstate_test_use_tco(void *opaque)
> > >>> {
> > >>> ICH9LPCPMRegs *s = opaque;
> > >>> @@ -209,6 +227,7 @@ const VMStateDescription vmstate_ich9_pm = {
> > >>> .subsections = (const VMStateDescription*[]) {
> > >>> &vmstate_memhp_state,
> > >>> &vmstate_tco_io_state,
> > >>> + &vmstate_cpuhp_state,
> > >>> NULL
> > >>> }
> > >>> };
> > >>> @@ -275,7 +294,10 @@ void ich9_pm_init(PCIDevice *lpc_pci, ICH9LPCPMRegs *pm,
> > >>>
> > >>> if (pm->cpu_hotplug_legacy) {
> > >>> legacy_acpi_cpu_hotplug_init(pci_address_space_io(lpc_pci),
> > >>> - OBJECT(lpc_pci), &pm->gpe_cpu, ICH9_CPU_HOTPLUG_IO_BASE);
> > >>> + OBJECT(lpc_pci), &pm->cpuhp.legacy, ICH9_CPU_HOTPLUG_IO_BASE);
> > >>> + } else {
> > >>> + cpu_hotplug_hw_init(pci_address_space_io(lpc_pci), OBJECT(lpc_pci),
> > >>> + &pm->cpuhp.state, ICH9_CPU_HOTPLUG_IO_BASE);
> > >>> }
> > >>>
> > >>> if (pm->acpi_memory_hotplug.is_enabled) {
> > >>> @@ -414,7 +436,6 @@ void ich9_pm_add_properties(Object *obj, ICH9LPCPMRegs *pm, Error **errp)
> > >>> {
> > >>> static const uint32_t gpe0_len = ICH9_PMIO_GPE0_LEN;
> > >>> pm->acpi_memory_hotplug.is_enabled = true;
> > >>> - pm->cpu_hotplug_legacy = true;
> > >>
> > >> Maybe is too soon to disable legacy CPU hotplug? Will the new cpu hotplug mechanism
> > >> work from this patch on, or it will break git bisect?
> > > it doesn't break bisect compile wise nor 'make check' wise,
> > > but it's hotplug won't be functional till
> > > acpi: add CPU hotplug methods to DSDT
> > > perhaps I need to move it there or into a separate patch after it.
> > >
> >
> > Exactly, I would move the "enabling" code after the new hotplug mechanism
> > is functional. The same for the next patch for piix4.
> ok, will do on v2 respin.
I have everything up to
pc: use AcpiDeviceIfClass.send_event to issue GPE events
tentatively in my tree - do you plan to rework any of these?
> >
> > Thanks,
> > Marcel
> >
> >
> > >>
> > >> Thanks,
> > >> Marcel
> > >>
> > >>> pm->disable_s3 = 0;
> > >>> pm->disable_s4 = 0;
> > >>> pm->s4_val = 2;
> > >>> @@ -461,9 +482,13 @@ void ich9_pm_device_plug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
> > >>> object_dynamic_cast(OBJECT(dev), TYPE_PC_DIMM)) {
> > >>> acpi_memory_plug_cb(hotplug_dev, &lpc->pm.acpi_memory_hotplug,
> > >>> dev, errp);
> > >>> - } else if (lpc->pm.cpu_hotplug_legacy &&
> > >>> - object_dynamic_cast(OBJECT(dev), TYPE_CPU)) {
> > >>> - legacy_acpi_cpu_plug_cb(hotplug_dev, &lpc->pm.gpe_cpu, dev, errp);
> > >>> + } else if (object_dynamic_cast(OBJECT(dev), TYPE_CPU)) {
> > >>> + if (lpc->pm.cpu_hotplug_legacy) {
> > >>> + legacy_acpi_cpu_plug_cb(hotplug_dev, &lpc->pm.cpuhp.legacy, dev,
> > >>> + errp);
> > >>> + } else {
> > >>> + acpi_cpu_plug_cb(hotplug_dev, &lpc->pm.cpuhp.state, dev, errp);
> > >>> + }
> > >>> } else {
> > >>> error_setg(errp, "acpi: device plug request for not supported device"
> > >>> " type: %s", object_get_typename(OBJECT(dev)));
> > >>> @@ -480,6 +505,10 @@ void ich9_pm_device_unplug_request_cb(HotplugHandler *hotplug_dev,
> > >>> acpi_memory_unplug_request_cb(hotplug_dev,
> > >>> &lpc->pm.acpi_memory_hotplug, dev,
> > >>> errp);
> > >>> + } else if (object_dynamic_cast(OBJECT(dev), TYPE_CPU) &&
> > >>> + !lpc->pm.cpu_hotplug_legacy) {
> > >>> + acpi_cpu_unplug_request_cb(hotplug_dev, &lpc->pm.cpuhp.state,
> > >>> + dev, errp);
> > >>> } else {
> > >>> error_setg(errp, "acpi: device unplug request for not supported device"
> > >>> " type: %s", object_get_typename(OBJECT(dev)));
> > >>> @@ -494,6 +523,9 @@ void ich9_pm_device_unplug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
> > >>> if (lpc->pm.acpi_memory_hotplug.is_enabled &&
> > >>> object_dynamic_cast(OBJECT(dev), TYPE_PC_DIMM)) {
> > >>> acpi_memory_unplug_cb(&lpc->pm.acpi_memory_hotplug, dev, errp);
> > >>> + } else if (object_dynamic_cast(OBJECT(dev), TYPE_CPU) &&
> > >>> + !lpc->pm.cpu_hotplug_legacy) {
> > >>> + acpi_cpu_unplug_cb(&lpc->pm.cpuhp.state, dev, errp);
> > >>> } else {
> > >>> error_setg(errp, "acpi: device unplug for not supported device"
> > >>> " type: %s", object_get_typename(OBJECT(dev)));
> > >>> diff --git a/include/hw/acpi/ich9.h b/include/hw/acpi/ich9.h
> > >>> index e29a856..198c017 100644
> > >>> --- a/include/hw/acpi/ich9.h
> > >>> +++ b/include/hw/acpi/ich9.h
> > >>> @@ -23,6 +23,7 @@
> > >>>
> > >>> #include "hw/acpi/acpi.h"
> > >>> #include "hw/acpi/cpu_hotplug.h"
> > >>> +#include "hw/acpi/cpu.h"
> > >>> #include "hw/acpi/memory_hotplug.h"
> > >>> #include "hw/acpi/acpi_dev_interface.h"
> > >>> #include "hw/acpi/tco.h"
> > >>> @@ -49,7 +50,10 @@ typedef struct ICH9LPCPMRegs {
> > >>> Notifier powerdown_notifier;
> > >>>
> > >>> bool cpu_hotplug_legacy;
> > >>> - AcpiCpuHotplug gpe_cpu;
> > >>> + union {
> > >>> + AcpiCpuHotplug legacy; /* used for keeping legacy state */
> > >>> + CPUHotplugState state;
> > >>> + } cpuhp;
> > >>>
> > >>> MemHotplugState acpi_memory_hotplug;
> > >>>
> > >>> diff --git a/include/hw/compat.h b/include/hw/compat.h
> > >>> index 636befe..a65d9d2 100644
> > >>> --- a/include/hw/compat.h
> > >>> +++ b/include/hw/compat.h
> > >>> @@ -2,7 +2,11 @@
> > >>> #define HW_COMPAT_H
> > >>>
> > >>> #define HW_COMPAT_2_6 \
> > >>> - /* empty */
> > >>> + {\
> > >>> + .driver = "ICH9-LPC",\
> > >>> + .property = "cpu-hotplug-legacy",\
> > >>> + .value = "on",\
> > >>> + },\
> > >>>
> > >>> #define HW_COMPAT_2_5 \
> > >>> {\
> > >>>
> > >>
> > >
> >
next prev parent reply other threads:[~2016-05-31 12:52 UTC|newest]
Thread overview: 83+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-17 14:42 [Qemu-devel] [PATCH 00/33] ACPI CPU hotplug refactoring to support more than 255 CPUs and PXM/OST methods Igor Mammedov
2016-05-17 14:42 ` [Qemu-devel] [PATCH 01/33] tests: acpi: report names of expected files in verbose mode Igor Mammedov
2016-05-24 16:58 ` Marcel Apfelbaum
2016-05-26 9:46 ` [Qemu-devel] [PATCH v2 1/33] " Igor Mammedov
2016-05-30 18:14 ` Marcel Apfelbaum
2016-06-02 11:02 ` Marcel Apfelbaum
2016-05-17 14:42 ` [Qemu-devel] [PATCH 02/33] acpi: add aml_debug() Igor Mammedov
2016-05-24 16:59 ` Marcel Apfelbaum
2016-05-17 14:42 ` [Qemu-devel] [PATCH 03/33] acpi: add aml_refof() Igor Mammedov
2016-05-24 17:00 ` Marcel Apfelbaum
2016-05-17 14:42 ` [Qemu-devel] [PATCH 04/33] pc: acpi: remove AML for empty/not used GPE handlers Igor Mammedov
2016-05-25 9:11 ` Marcel Apfelbaum
2016-05-25 13:19 ` Igor Mammedov
2016-05-31 10:06 ` Marcel Apfelbaum
2016-05-17 14:42 ` [Qemu-devel] [PATCH 05/33] pc: acpi: consolidate CPU hotplug AML Igor Mammedov
2016-05-30 18:18 ` Marcel Apfelbaum
2016-05-31 7:50 ` Igor Mammedov
2016-05-31 10:18 ` Marcel Apfelbaum
2016-05-31 12:49 ` Igor Mammedov
2016-05-17 14:42 ` [Qemu-devel] [PATCH 06/33] pc: acpi: consolidate \GPE._E02 with the rest of " Igor Mammedov
2016-05-30 18:22 ` Marcel Apfelbaum
2016-05-17 14:42 ` [Qemu-devel] [PATCH 07/33] pc: acpi: cpu-hotplug: make AML CPU_foo defines local to cpu_hotplug_acpi_table.c Igor Mammedov
2016-05-30 18:23 ` Marcel Apfelbaum
2016-05-17 14:43 ` [Qemu-devel] [PATCH 08/33] pc: acpi: mark current CPU hotplug functions as legacy Igor Mammedov
2016-05-30 18:28 ` Marcel Apfelbaum
2016-05-17 14:43 ` [Qemu-devel] [PATCH 09/33] pc: acpi: consolidate legacy CPU hotplug in one file Igor Mammedov
2016-05-30 18:31 ` Marcel Apfelbaum
2016-05-17 14:43 ` [Qemu-devel] [PATCH 10/33] pc: acpi: simplify build_legacy_cpu_hotplug_aml() signature Igor Mammedov
2016-05-30 18:31 ` Marcel Apfelbaum
2016-05-17 14:43 ` [Qemu-devel] [PATCH 11/33] pc: acpi: cpuhp-legacy: switch ProcessorID to possible_cpus idx Igor Mammedov
2016-05-30 18:39 ` Marcel Apfelbaum
2016-05-31 13:03 ` Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 12/33] tests: acpi: update tables with consolidated legacy cpu-hotplug AML Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 13/33] acpi: extend ACPI interface to provide send_event hook Igor Mammedov
2016-05-30 18:45 ` Marcel Apfelbaum
2016-05-31 9:57 ` [Qemu-devel] [PATCH v2 " Igor Mammedov
2016-06-02 11:09 ` Marcel Apfelbaum
2016-06-02 11:19 ` Igor Mammedov
2016-06-02 11:21 ` Marcel Apfelbaum
2016-05-17 14:43 ` [Qemu-devel] [PATCH 14/33] pc: use AcpiDeviceIfClass.send_event to issue GPE events Igor Mammedov
2016-05-31 10:01 ` [Qemu-devel] [PATCH v2 " Igor Mammedov
2016-06-02 11:13 ` Marcel Apfelbaum
2016-06-02 11:29 ` Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 15/33] docs: update ACPI CPU hotplug spec with new protocol Igor Mammedov
2016-05-31 4:49 ` Michael S. Tsirkin
2016-05-31 15:07 ` Igor Mammedov
2016-05-31 21:09 ` Michael S. Tsirkin
2016-06-06 9:57 ` Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 16/33] acpi: hardware side of CPU hotplug Igor Mammedov
2016-05-30 18:50 ` Marcel Apfelbaum
2016-05-31 13:24 ` Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 17/33] pc: add generic CPU unplug callbacks Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 18/33] pc: add 2.7 machine Igor Mammedov
2016-05-30 18:53 ` Marcel Apfelbaum
2016-05-30 19:04 ` Eduardo Habkost
2016-05-17 14:43 ` [Qemu-devel] [PATCH 19/33] pc: piix4/ich9: add 'cpu-hotplug-legacy' property Igor Mammedov
2016-05-30 18:59 ` Marcel Apfelbaum
2016-05-17 14:43 ` [Qemu-devel] [PATCH 20/33] pc: q35: initialize new CPU hotplug hw Igor Mammedov
2016-05-30 19:02 ` Marcel Apfelbaum
2016-05-31 10:06 ` Igor Mammedov
2016-05-31 10:21 ` Marcel Apfelbaum
2016-05-31 12:51 ` Igor Mammedov
2016-05-31 12:52 ` Michael S. Tsirkin [this message]
2016-05-31 13:18 ` Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 21/33] pc: piix4: " Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 22/33] pc: acpi: introduce AcpiDeviceIfClass.madt_cpu hook Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 23/33] acpi: add CPU devices AML to DSDT Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 24/33] acpi: add CPU hotplug methods " Igor Mammedov
2016-05-31 4:38 ` Michael S. Tsirkin
2016-05-31 8:45 ` Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 25/33] qdev: hotplug: Introduce HotplugHandler.pre_plug() callback Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 26/33] target-i386: add X86CPU.node property Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 27/33] pc: numa: replace node_cpu indexing by apic_id with possible_cpus index Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 28/33] pc: set X86CPU.node property if QEMU starts with numa enabled Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 29/33] acpi: cpuhp: provide cpu._PXM method if running in numa mode Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 30/33] acpi: cpuhp: add cpu._OST handling Igor Mammedov
2016-05-17 15:29 ` Eric Blake
2016-05-18 8:09 ` Igor Mammedov
2016-05-30 18:21 ` Michael S. Tsirkin
2016-05-31 12:53 ` Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 31/33] tests: acpi: update expected tables with new cpu-hotplug methods enabled by default Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 32/33] tests: acpi: add CPU hotplug testcase Igor Mammedov
2016-05-17 14:43 ` [Qemu-devel] [PATCH 33/33] tests: acpi: add DSDT/MADT expected tables for cpu-hotplug case Igor Mammedov
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=20160531155201-mutt-send-email-mst@redhat.com \
--to=mst@redhat.com \
--cc=armbru@redhat.com \
--cc=drjones@redhat.com \
--cc=ehabkost@redhat.com \
--cc=imammedo@redhat.com \
--cc=marcel@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=rkrcmar@redhat.com \
/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).