qemu/hw/acpi/pcihp.c
Igor Mammedov 0f689cf5ad acpi: pcihp: allow repeating hot-unplug requests
with Q35 using ACPI PCI hotplug by default, user's request to unplug
device is ignored when it's issued before guest OS has been booted.
And any additional attempt to request device hot-unplug afterwards
results in following error:

  "Device XYZ is already in the process of unplug"

arguably it can be considered as a regression introduced by [2],
before which it was possible to issue unplug request multiple
times.

Accept new uplug requests after timeout (1ms). This brings ACPI PCI
hotplug on par with native PCIe unplug behavior [1] and allows user
to repeat unplug requests at propper times.
Set expire timeout to arbitrary 1msec so user won't be able to
flood guest with SCI interrupts by calling device_del in tight loop.

PS:
ACPI spec doesn't mandate what OSPM can do with GPEx.status
bits set before it's booted => it's impl. depended.
Status bits may be retained (I tested with one Windows version)
or cleared (Linux since 2.6 kernel times) during guest's ACPI
subsystem initialization.
Clearing status bits (though not wrong per se) hides the unplug
event from guest, and it's upto user to repeat device_del later
when guest is able to handle unplug requests.

1) 18416c62e3 ("pcie: expire pending delete")
2)
Fixes: cce8944cc9 ("qdev-monitor: Forbid repeated device_del")
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
CC: mst@redhat.com
CC: anisinha@redhat.com
CC: jusual@redhat.com
CC: kraxel@redhat.com
Message-Id: <20230418090449.2155757-1-imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Ani Sinha <anisinha@redhat.com>
2023-04-24 22:56:55 -04:00

527 lines
16 KiB
C

/*
* QEMU<->ACPI BIOS PCI hotplug interface
*
* QEMU supports PCI hotplug via ACPI. This module
* implements the interface between QEMU and the ACPI BIOS.
* Interface specification - see docs/specs/acpi_pci_hotplug.txt
*
* Copyright (c) 2013, Red Hat Inc, Michael S. Tsirkin (mst@redhat.com)
* Copyright (c) 2006 Fabrice Bellard
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License version 2.1 as published by the Free Software Foundation.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library; if not, see <http://www.gnu.org/licenses/>
*
* Contributions after 2012-01-13 are licensed under the terms of the
* GNU GPL, version 2 or (at your option) any later version.
*/
#include "qemu/osdep.h"
#include "hw/acpi/pcihp.h"
#include "hw/pci-host/i440fx.h"
#include "hw/pci/pci.h"
#include "hw/pci/pci_bridge.h"
#include "hw/pci/pci_host.h"
#include "hw/pci/pcie_port.h"
#include "hw/pci-bridge/xio3130_downstream.h"
#include "hw/i386/acpi-build.h"
#include "hw/acpi/acpi.h"
#include "hw/pci/pci_bus.h"
#include "migration/vmstate.h"
#include "qapi/error.h"
#include "qom/qom-qobject.h"
#include "trace.h"
#define ACPI_PCIHP_SIZE 0x0018
#define PCI_UP_BASE 0x0000
#define PCI_DOWN_BASE 0x0004
#define PCI_EJ_BASE 0x0008
#define PCI_RMV_BASE 0x000c
#define PCI_SEL_BASE 0x0010
#define PCI_AIDX_BASE 0x0014
typedef struct AcpiPciHpFind {
int bsel;
PCIBus *bus;
} AcpiPciHpFind;
static int acpi_pcihp_get_bsel(PCIBus *bus)
{
Error *local_err = NULL;
uint64_t bsel = object_property_get_uint(OBJECT(bus), ACPI_PCIHP_PROP_BSEL,
&local_err);
if (local_err || bsel >= ACPI_PCIHP_MAX_HOTPLUG_BUS) {
if (local_err) {
error_free(local_err);
}
return -1;
} else {
return bsel;
}
}
typedef struct {
unsigned bsel_alloc;
bool has_bridge_hotplug;
} BSELInfo;
/* Assign BSEL property only to buses that support hotplug. */
static void *acpi_set_bsel(PCIBus *bus, void *opaque)
{
BSELInfo *info = opaque;
unsigned *bus_bsel;
DeviceState *br = bus->qbus.parent;
bool is_bridge = IS_PCI_BRIDGE(br);
/* hotplugged bridges can't be described in ACPI ignore them */
if (qbus_is_hotpluggable(BUS(bus))) {
if (!is_bridge || (!br->hotplugged && info->has_bridge_hotplug)) {
bus_bsel = g_malloc(sizeof *bus_bsel);
*bus_bsel = info->bsel_alloc++;
object_property_add_uint32_ptr(OBJECT(bus), ACPI_PCIHP_PROP_BSEL,
bus_bsel, OBJ_PROP_FLAG_READ);
}
}
return info;
}
static void acpi_set_pci_info(bool has_bridge_hotplug)
{
static bool bsel_is_set;
Object *host = acpi_get_i386_pci_host();
PCIBus *bus;
BSELInfo info = { .bsel_alloc = ACPI_PCIHP_BSEL_DEFAULT,
.has_bridge_hotplug = has_bridge_hotplug };
if (bsel_is_set) {
return;
}
bsel_is_set = true;
if (!host) {
return;
}
bus = PCI_HOST_BRIDGE(host)->bus;
if (bus) {
/* Scan all PCI buses. Set property to enable acpi based hotplug. */
pci_for_each_bus_depth_first(bus, acpi_set_bsel, NULL, &info);
}
}
static void acpi_pcihp_test_hotplug_bus(PCIBus *bus, void *opaque)
{
AcpiPciHpFind *find = opaque;
if (find->bsel == acpi_pcihp_get_bsel(bus)) {
find->bus = bus;
}
}
static PCIBus *acpi_pcihp_find_hotplug_bus(AcpiPciHpState *s, int bsel)
{
AcpiPciHpFind find = { .bsel = bsel, .bus = NULL };
if (bsel < 0) {
return NULL;
}
pci_for_each_bus(s->root, acpi_pcihp_test_hotplug_bus, &find);
/* Make bsel 0 eject root bus if bsel property is not set,
* for compatibility with non acpi setups.
* TODO: really needed?
*/
if (!bsel && !find.bus) {
find.bus = s->root;
}
/*
* Check if find.bus is actually hotpluggable. If bsel is set to
* NULL for example on the root bus in order to make it
* non-hotpluggable, find.bus will match the root bus when bsel
* is 0. See acpi_pcihp_test_hotplug_bus() above. Since the
* bus is not hotpluggable however, we should not select the bus.
* Instead, we should set find.bus to NULL in that case. In the check
* below, we generalize this case for all buses, not just the root bus.
* The callers of this function check for a null return value and
* handle them appropriately.
*/
if (find.bus && !qbus_is_hotpluggable(BUS(find.bus))) {
find.bus = NULL;
}
return find.bus;
}
static bool acpi_pcihp_pc_no_hotplug(AcpiPciHpState *s, PCIDevice *dev)
{
DeviceClass *dc = DEVICE_GET_CLASS(dev);
/*
* ACPI doesn't allow hotplug of bridge devices. Don't allow
* hot-unplug of bridge devices unless they were added by hotplug
* (and so, not described by acpi).
*
* Don't allow hot-unplug of SR-IOV Virtual Functions, as they
* will be removed implicitly, when Physical Function is unplugged.
*/
return (IS_PCI_BRIDGE(dev) && !dev->qdev.hotplugged) || !dc->hotpluggable ||
pci_is_vf(dev);
}
static void acpi_pcihp_eject_slot(AcpiPciHpState *s, unsigned bsel, unsigned slots)
{
HotplugHandler *hotplug_ctrl;
BusChild *kid, *next;
int slot = ctz32(slots);
PCIBus *bus = acpi_pcihp_find_hotplug_bus(s, bsel);
trace_acpi_pci_eject_slot(bsel, slot);
if (!bus || slot > 31) {
return;
}
/* Mark request as complete */
s->acpi_pcihp_pci_status[bsel].down &= ~(1U << slot);
s->acpi_pcihp_pci_status[bsel].up &= ~(1U << slot);
QTAILQ_FOREACH_SAFE(kid, &bus->qbus.children, sibling, next) {
DeviceState *qdev = kid->child;
PCIDevice *dev = PCI_DEVICE(qdev);
if (PCI_SLOT(dev->devfn) == slot) {
if (!acpi_pcihp_pc_no_hotplug(s, dev)) {
/*
* partially_hotplugged is used by virtio-net failover:
* failover has asked the guest OS to unplug the device
* but we need to keep some references to the device
* to be able to plug it back in case of failure so
* we don't execute hotplug_handler_unplug().
*/
if (dev->partially_hotplugged) {
/*
* pending_deleted_event is set to true when
* virtio-net failover asks to unplug the device,
* and set to false here when the operation is done
* This is used by the migration loop to detect the
* end of the operation and really start the migration.
*/
qdev->pending_deleted_event = false;
} else {
hotplug_ctrl = qdev_get_hotplug_handler(qdev);
hotplug_handler_unplug(hotplug_ctrl, qdev, &error_abort);
object_unparent(OBJECT(qdev));
}
}
}
}
}
static void acpi_pcihp_update_hotplug_bus(AcpiPciHpState *s, int bsel)
{
BusChild *kid, *next;
PCIBus *bus = acpi_pcihp_find_hotplug_bus(s, bsel);
/* Execute any pending removes during reset */
while (s->acpi_pcihp_pci_status[bsel].down) {
acpi_pcihp_eject_slot(s, bsel, s->acpi_pcihp_pci_status[bsel].down);
}
s->acpi_pcihp_pci_status[bsel].hotplug_enable = ~0;
if (!bus) {
return;
}
QTAILQ_FOREACH_SAFE(kid, &bus->qbus.children, sibling, next) {
DeviceState *qdev = kid->child;
PCIDevice *pdev = PCI_DEVICE(qdev);
int slot = PCI_SLOT(pdev->devfn);
if (acpi_pcihp_pc_no_hotplug(s, pdev)) {
s->acpi_pcihp_pci_status[bsel].hotplug_enable &= ~(1U << slot);
}
}
}
static void acpi_pcihp_update(AcpiPciHpState *s)
{
int i;
for (i = 0; i < ACPI_PCIHP_MAX_HOTPLUG_BUS; ++i) {
acpi_pcihp_update_hotplug_bus(s, i);
}
}
void acpi_pcihp_reset(AcpiPciHpState *s)
{
acpi_set_pci_info(s->use_acpi_hotplug_bridge);
acpi_pcihp_update(s);
}
void acpi_pcihp_device_pre_plug_cb(HotplugHandler *hotplug_dev,
DeviceState *dev, Error **errp)
{
PCIDevice *pdev = PCI_DEVICE(dev);
/* Only hotplugged devices need the hotplug capability. */
if (dev->hotplugged &&
acpi_pcihp_get_bsel(pci_get_bus(pdev)) < 0) {
error_setg(errp, "Unsupported bus. Bus doesn't have property '"
ACPI_PCIHP_PROP_BSEL "' set");
return;
}
}
void acpi_pcihp_device_plug_cb(HotplugHandler *hotplug_dev, AcpiPciHpState *s,
DeviceState *dev, Error **errp)
{
PCIDevice *pdev = PCI_DEVICE(dev);
int slot = PCI_SLOT(pdev->devfn);
PCIDevice *bridge;
PCIBus *bus;
int bsel;
/* Don't send event when device is enabled during qemu machine creation:
* it is present on boot, no hotplug event is necessary. We do send an
* event when the device is disabled later. */
if (!dev->hotplugged) {
/*
* Overwrite the default hotplug handler with the ACPI PCI one
* for cold plugged bridges only.
*/
if (s->use_acpi_hotplug_bridge &&
object_dynamic_cast(OBJECT(dev), TYPE_PCI_BRIDGE)) {
PCIBus *sec = pci_bridge_get_sec_bus(PCI_BRIDGE(pdev));
qbus_set_hotplug_handler(BUS(sec), OBJECT(hotplug_dev));
/* We don't have to overwrite any other hotplug handler yet */
assert(QLIST_EMPTY(&sec->child));
}
return;
}
bus = pci_get_bus(pdev);
bridge = pci_bridge_get_device(bus);
if (object_dynamic_cast(OBJECT(bridge), TYPE_PCIE_ROOT_PORT) ||
object_dynamic_cast(OBJECT(bridge), TYPE_XIO3130_DOWNSTREAM)) {
pcie_cap_slot_enable_power(bridge);
}
bsel = acpi_pcihp_get_bsel(bus);
g_assert(bsel >= 0);
s->acpi_pcihp_pci_status[bsel].up |= (1U << slot);
acpi_send_event(DEVICE(hotplug_dev), ACPI_PCI_HOTPLUG_STATUS);
}
void acpi_pcihp_device_unplug_cb(HotplugHandler *hotplug_dev, AcpiPciHpState *s,
DeviceState *dev, Error **errp)
{
PCIDevice *pdev = PCI_DEVICE(dev);
trace_acpi_pci_unplug(PCI_SLOT(pdev->devfn),
acpi_pcihp_get_bsel(pci_get_bus(pdev)));
qdev_unrealize(dev);
}
void acpi_pcihp_device_unplug_request_cb(HotplugHandler *hotplug_dev,
AcpiPciHpState *s, DeviceState *dev,
Error **errp)
{
PCIDevice *pdev = PCI_DEVICE(dev);
int slot = PCI_SLOT(pdev->devfn);
int bsel = acpi_pcihp_get_bsel(pci_get_bus(pdev));
trace_acpi_pci_unplug_request(bsel, slot);
if (bsel < 0) {
error_setg(errp, "Unsupported bus. Bus doesn't have property '"
ACPI_PCIHP_PROP_BSEL "' set");
return;
}
/*
* pending_deleted_event is used by virtio-net failover to detect the
* end of the unplug operation, the flag is set to false in
* acpi_pcihp_eject_slot() when the operation is completed.
*/
pdev->qdev.pending_deleted_event = true;
/* if unplug was requested before OSPM is initialized,
* linux kernel will clear GPE0.sts[] bits during boot, which effectively
* hides unplug event. And than followup qmp_device_del() calls remain
* blocked by above flag permanently.
* Unblock qmp_device_del() by setting expire limit, so user can
* repeat unplug request later when OSPM has been booted.
*/
pdev->qdev.pending_deleted_expires_ms =
qemu_clock_get_ms(QEMU_CLOCK_VIRTUAL); /* 1 msec */
s->acpi_pcihp_pci_status[bsel].down |= (1U << slot);
acpi_send_event(DEVICE(hotplug_dev), ACPI_PCI_HOTPLUG_STATUS);
}
bool acpi_pcihp_is_hotpluggbale_bus(AcpiPciHpState *s, BusState *bus)
{
Object *o = OBJECT(bus->parent);
if (s->use_acpi_hotplug_bridge &&
object_dynamic_cast(o, TYPE_PCI_BRIDGE)) {
if (object_dynamic_cast(o, TYPE_PCIE_SLOT) && !PCIE_SLOT(o)->hotplug) {
return false;
}
return true;
}
if (s->use_acpi_root_pci_hotplug) {
return true;
}
return false;
}
static uint64_t pci_read(void *opaque, hwaddr addr, unsigned int size)
{
AcpiPciHpState *s = opaque;
uint32_t val = 0;
int bsel = s->hotplug_select;
if (bsel < 0 || bsel >= ACPI_PCIHP_MAX_HOTPLUG_BUS) {
return 0;
}
switch (addr) {
case PCI_UP_BASE:
val = s->acpi_pcihp_pci_status[bsel].up;
if (s->use_acpi_hotplug_bridge) {
s->acpi_pcihp_pci_status[bsel].up = 0;
}
trace_acpi_pci_up_read(val);
break;
case PCI_DOWN_BASE:
val = s->acpi_pcihp_pci_status[bsel].down;
trace_acpi_pci_down_read(val);
break;
case PCI_EJ_BASE:
trace_acpi_pci_features_read(val);
break;
case PCI_RMV_BASE:
val = s->acpi_pcihp_pci_status[bsel].hotplug_enable;
trace_acpi_pci_rmv_read(val);
break;
case PCI_SEL_BASE:
val = s->hotplug_select;
trace_acpi_pci_sel_read(val);
break;
case PCI_AIDX_BASE:
val = s->acpi_index;
s->acpi_index = 0;
trace_acpi_pci_acpi_index_read(val);
break;
default:
break;
}
return val;
}
static void pci_write(void *opaque, hwaddr addr, uint64_t data,
unsigned int size)
{
int slot;
PCIBus *bus;
BusChild *kid, *next;
AcpiPciHpState *s = opaque;
s->acpi_index = 0;
switch (addr) {
case PCI_AIDX_BASE:
/*
* fetch acpi-index for specified slot so that follow up read from
* PCI_AIDX_BASE can return it to guest
*/
slot = ctz32(data);
if (s->hotplug_select >= ACPI_PCIHP_MAX_HOTPLUG_BUS) {
break;
}
bus = acpi_pcihp_find_hotplug_bus(s, s->hotplug_select);
if (!bus) {
break;
}
QTAILQ_FOREACH_SAFE(kid, &bus->qbus.children, sibling, next) {
Object *o = OBJECT(kid->child);
PCIDevice *dev = PCI_DEVICE(o);
if (PCI_SLOT(dev->devfn) == slot) {
s->acpi_index = object_property_get_uint(o, "acpi-index", NULL);
break;
}
}
trace_acpi_pci_acpi_index_write(s->hotplug_select, slot, s->acpi_index);
break;
case PCI_EJ_BASE:
if (s->hotplug_select >= ACPI_PCIHP_MAX_HOTPLUG_BUS) {
break;
}
acpi_pcihp_eject_slot(s, s->hotplug_select, data);
trace_acpi_pci_ej_write(addr, data);
break;
case PCI_SEL_BASE:
s->hotplug_select = s->use_acpi_hotplug_bridge ? data :
ACPI_PCIHP_BSEL_DEFAULT;
trace_acpi_pci_sel_write(addr, data);
default:
break;
}
}
static const MemoryRegionOps acpi_pcihp_io_ops = {
.read = pci_read,
.write = pci_write,
.endianness = DEVICE_LITTLE_ENDIAN,
.valid = {
.min_access_size = 4,
.max_access_size = 4,
},
};
void acpi_pcihp_init(Object *owner, AcpiPciHpState *s, PCIBus *root_bus,
MemoryRegion *address_space_io,
uint16_t io_base)
{
s->io_len = ACPI_PCIHP_SIZE;
s->io_base = io_base;
s->root = root_bus;
memory_region_init_io(&s->io, owner, &acpi_pcihp_io_ops, s,
"acpi-pci-hotplug", s->io_len);
memory_region_add_subregion(address_space_io, s->io_base, &s->io);
object_property_add_uint16_ptr(owner, ACPI_PCIHP_IO_BASE_PROP, &s->io_base,
OBJ_PROP_FLAG_READ);
object_property_add_uint16_ptr(owner, ACPI_PCIHP_IO_LEN_PROP, &s->io_len,
OBJ_PROP_FLAG_READ);
}
const VMStateDescription vmstate_acpi_pcihp_pci_status = {
.name = "acpi_pcihp_pci_status",
.version_id = 1,
.minimum_version_id = 1,
.fields = (VMStateField[]) {
VMSTATE_UINT32(up, AcpiPciHpPciStatus),
VMSTATE_UINT32(down, AcpiPciHpPciStatus),
VMSTATE_END_OF_LIST()
}
};