Skip to contentSkip to conversation

Use ACPI PCI hot-plug for q35

RFC PATCH Superseded Julia Suvorova 5 participants 22 messages Jul 8, 2020 – Jul 15, 2020

v1 of 6 Superseded Latest is v6 →

lore
Share & tools
lore
#0 lore raw

cover Use ACPI PCI hot-plug for q35

PCIe native hot-plug has numerous problems with racing events and unpredictable guest behaviour (Windows). Switching to ACPI hot-plug for now.

Tested on RHEL 8 and Windows 2019.

pxb-pcie is not yet supported.

Julia Suvorova (5):
hw/acpi/pcihp: Introduce find_host()
hw/acpi/ich9: Trace ich9_gpe_readb()/writeb() hw/i386/acpi-build: Add ACPI PCI hot-plug methods to q35 hw/i386/acpi-build: Turn off support of PCIe native hot-plug and SHPC
in _OSC
hw/acpi/ich9: Enable ACPI PCI hot-plug

 hw/i386/acpi-build.h   |  2 ++
 include/hw/acpi/ich9.h |  3 +++
 hw/acpi/ich9.c         | 52 +++++++++++++++++++++++++++++++++++++++++-
 hw/acpi/pcihp.c        | 16 ++++++++++++-
 hw/i386/acpi-build.c   | 34 +++++++++++++++++----------
 hw/i386/pc.c           |  4 +++-
 hw/acpi/trace-events   |  4 ++++
 7 files changed, 100 insertions(+), 15 deletions(-)
Signature · 1 line
2.25.4
#1 lore raw

RFC PATCH 1/5 hw/acpi/pcihp: Introduce find_host()

Returns the current host bus with ACPI PCI hot-plug support: q35 or i440fx.

---
 hw/i386/acpi-build.h |  2 ++
 hw/acpi/pcihp.c      | 13 +++++++++++++
 hw/i386/acpi-build.c |  2 +-
 3 files changed, 16 insertions(+), 1 deletion(-)
hw/i386/acpi-build.h +2 -0
index 74df5fc612..0696b4e48d 100644
@@ -7,4 +7,6 @@ extern const struct AcpiGenericAddress x86_nvdimm_acpi_dsmio;
 
 void acpi_setup(void);
 
+Object *acpi_get_i386_pci_host(void);
+
 #endif
hw/acpi/pcihp.c +13 -0
index d42906ea19..3d4ee3af72 100644
@@ -33,10 +33,12 @@
 #include "hw/acpi/acpi.h"
 #include "exec/address-spaces.h"
 #include "hw/pci/pci_bus.h"
+#include "hw/pci/pci_host.h"
 #include "migration/vmstate.h"
 #include "qapi/error.h"
 #include "qom/qom-qobject.h"
 #include "trace.h"
+#include "hw/i386/acpi-build.h"
 
 #define ACPI_PCIHP_ADDR 0xae00
 #define ACPI_PCIHP_SIZE 0x0014
@@ -86,6 +88,17 @@ static void *acpi_set_bsel(PCIBus *bus, void *opaque)
     return bsel_alloc;
 }
 
+static PCIBus *find_host(void)
+{
+    Object *obj = acpi_get_i386_pci_host();
+
+    if (obj) {
+        return PCI_HOST_BRIDGE(obj)->bus;
+    }
+
+    return NULL;
+}
+
 static void acpi_set_pci_info(void)
 {
     static bool bsel_is_set;
hw/i386/acpi-build.c +1 -1
index 900f786d08..11c598f955 100644
@@ -270,7 +270,7 @@ static void acpi_get_misc_info(AcpiMiscInfo *info)
  * Because of the PXB hosts we cannot simply query TYPE_PCI_HOST_BRIDGE.
  * On i386 arch we only have two pci hosts, so we can look only for them.
  */
-static Object *acpi_get_i386_pci_host(void)
+Object *acpi_get_i386_pci_host(void)
 {
     PCIHostState *host;
 
Signature · 1 line
2.25.4
#2 lore raw

Re: [RFC PATCH 1/5] hw/acpi/pcihp: Introduce find_host()

Quoted 41 lines · On Thu, 9 Jul 2020 00:46:11 +0200 Julia Suvorova <[email protected]> wrote:

Returns the current host bus with ACPI PCI hot-plug support: q35 or i440fx.

---
 hw/i386/acpi-build.h |  2 ++
 hw/acpi/pcihp.c      | 13 +++++++++++++
 hw/i386/acpi-build.c |  2 +-
 3 files changed, 16 insertions(+), 1 deletion(-)
hw/i386/acpi-build.h +2 -0
index 74df5fc612..0696b4e48d 100644
@@ -7,4 +7,6 @@ extern const struct AcpiGenericAddress x86_nvdimm_acpi_dsmio;
 
 void acpi_setup(void);
 
+Object *acpi_get_i386_pci_host(void);
+
 #endif
hw/acpi/pcihp.c +2 -0
index d42906ea19..3d4ee3af72 100644
@@ -33,10 +33,12 @@
 #include "hw/acpi/acpi.h"
 #include "exec/address-spaces.h"
 #include "hw/pci/pci_bus.h"
+#include "hw/pci/pci_host.h"
 #include "migration/vmstate.h"
 #include "qapi/error.h"
 #include "qom/qom-qobject.h"
 #include "trace.h"
+#include "hw/i386/acpi-build.h"
 
 #define ACPI_PCIHP_ADDR 0xae00
 #define ACPI_PCIHP_SIZE 0x0014
@@ -86,6 +88,17 @@ static void *acpi_set_bsel(PCIBus *bus, void *opaque)
     return bsel_alloc;
 }
 

hw/acpi/pcihp.c static void *acpi_set_bsel(PCIBus *bus, void *opaque) · line 100

Quoted 10 lines
hw/acpi/pcihp.c +10 -0
+static PCIBus *find_host(void)
+{
+    Object *obj = acpi_get_i386_pci_host();
+
+    if (obj) {
+        return PCI_HOST_BRIDGE(obj)->bus;
+    }
+
+    return NULL;
+}

My guess you are adding it for 5/5, with a function name a bit off compared to what you are doing (probably you've tried to reuse find_i440fx() idea)

I'd just make acpi_get_i386_pci_host() public, drop find_host and use

host = acpi_get_i386_pci_host()
bus = PCI_HOST_BRIDGE(pci_host)->bus

like it's done elsewhere

Quoted 16 lines
hw/acpi/pcihp.c +0 -0
 static void acpi_set_pci_info(void)
 {
     static bool bsel_is_set;
hw/i386/acpi-build.c +1 -1
index 900f786d08..11c598f955 100644
@@ -270,7 +270,7 @@ static void acpi_get_misc_info(AcpiMiscInfo *info)
  * Because of the PXB hosts we cannot simply query TYPE_PCI_HOST_BRIDGE.
  * On i386 arch we only have two pci hosts, so we can look only for them.
  */
-static Object *acpi_get_i386_pci_host(void)
+Object *acpi_get_i386_pci_host(void)
 {
     PCIHostState *host;
 
#3 lore raw

RFC PATCH 2/5 hw/acpi/ich9: Trace ich9_gpe_readb()/writeb()

In reply to Julia Suvorova

Add trace events similar to piix4_gpe_readb() to check gpe status.

---
 hw/acpi/ich9.c       | 7 ++++++-
 hw/acpi/trace-events | 4 ++++
 2 files changed, 10 insertions(+), 1 deletion(-)
hw/acpi/ich9.c +6 -1
index 2d204babc6..0fdd736da4 100644
@@ -36,6 +36,7 @@
 #include "hw/acpi/acpi.h"
 #include "hw/acpi/tco.h"
 #include "exec/address-spaces.h"
+#include "trace.h"
 
 #include "hw/i386/ich9.h"
 #include "hw/mem/pc-dimm.h"
@@ -59,13 +60,17 @@ static void ich9_pm_update_sci_fn(ACPIREGS *regs)
 static uint64_t ich9_gpe_readb(void *opaque, hwaddr addr, unsigned width)
 {
     ICH9LPCPMRegs *pm = opaque;
-    return acpi_gpe_ioport_readb(&pm->acpi_regs, addr);
+    uint64_t val = acpi_gpe_ioport_readb(&pm->acpi_regs, addr);
+
+    trace_ich9_gpe_readb(addr, width, val);
+    return val;
 }
 
 static void ich9_gpe_writeb(void *opaque, hwaddr addr, uint64_t val,
                             unsigned width)
 {
     ICH9LPCPMRegs *pm = opaque;
+    trace_ich9_gpe_writeb(addr, width, val);
     acpi_gpe_ioport_writeb(&pm->acpi_regs, addr, val);
     acpi_update_sci(&pm->acpi_regs, pm->irq);
 }
hw/acpi/trace-events +4 -0
index afbc77de1c..b9f4827afc 100644
@@ -32,6 +32,10 @@ cpuhp_acpi_ejecting_cpu(uint32_t idx) "0x%"PRIx32
 cpuhp_acpi_write_ost_ev(uint32_t slot, uint32_t ev) "idx[0x%"PRIx32"] OST EVENT: 0x%"PRIx32
 cpuhp_acpi_write_ost_status(uint32_t slot, uint32_t st) "idx[0x%"PRIx32"] OST STATUS: 0x%"PRIx32
 
+# ich9.c
+ich9_gpe_readb(uint64_t addr, unsigned width, uint64_t val) "addr: 0x%" PRIx64 " width: %d ==> 0x%" PRIx64
+ich9_gpe_writeb(uint64_t addr, unsigned width, uint64_t val) "addr: 0x%" PRIx64 " width: %d <== 0x%" PRIx64
+
 # pcihp.c
 acpi_pci_eject_slot(unsigned bsel, unsigned slot) "bsel: %u slot: %u"
 acpi_pci_unplug(int bsel, int slot) "bsel: %d slot: %d"
Signature · 1 line
2.25.4
Reviewed-by on v1 2/5
#4 lore raw

Re: [RFC PATCH 2/5] hw/acpi/ich9: Trace ich9_gpe_readb()/writeb()

Quoted 3 lines · On Thu, 9 Jul 2020 00:46:12 +0200 Julia Suvorova <[email protected]> wrote:

Add trace events similar to piix4_gpe_readb() to check gpe status.

Quoted 51 lines
---
 hw/acpi/ich9.c       | 7 ++++++-
 hw/acpi/trace-events | 4 ++++
 2 files changed, 10 insertions(+), 1 deletion(-)
hw/acpi/ich9.c +6 -1
index 2d204babc6..0fdd736da4 100644
@@ -36,6 +36,7 @@
 #include "hw/acpi/acpi.h"
 #include "hw/acpi/tco.h"
 #include "exec/address-spaces.h"
+#include "trace.h"
 
 #include "hw/i386/ich9.h"
 #include "hw/mem/pc-dimm.h"
@@ -59,13 +60,17 @@ static void ich9_pm_update_sci_fn(ACPIREGS *regs)
 static uint64_t ich9_gpe_readb(void *opaque, hwaddr addr, unsigned width)
 {
     ICH9LPCPMRegs *pm = opaque;
-    return acpi_gpe_ioport_readb(&pm->acpi_regs, addr);
+    uint64_t val = acpi_gpe_ioport_readb(&pm->acpi_regs, addr);
+
+    trace_ich9_gpe_readb(addr, width, val);
+    return val;
 }
 
 static void ich9_gpe_writeb(void *opaque, hwaddr addr, uint64_t val,
                             unsigned width)
 {
     ICH9LPCPMRegs *pm = opaque;
+    trace_ich9_gpe_writeb(addr, width, val);
     acpi_gpe_ioport_writeb(&pm->acpi_regs, addr, val);
     acpi_update_sci(&pm->acpi_regs, pm->irq);
 }
hw/acpi/trace-events +4 -0
index afbc77de1c..b9f4827afc 100644
@@ -32,6 +32,10 @@ cpuhp_acpi_ejecting_cpu(uint32_t idx) "0x%"PRIx32
 cpuhp_acpi_write_ost_ev(uint32_t slot, uint32_t ev) "idx[0x%"PRIx32"] OST EVENT: 0x%"PRIx32
 cpuhp_acpi_write_ost_status(uint32_t slot, uint32_t st) "idx[0x%"PRIx32"] OST STATUS: 0x%"PRIx32
 
+# ich9.c
+ich9_gpe_readb(uint64_t addr, unsigned width, uint64_t val) "addr: 0x%" PRIx64 " width: %d ==> 0x%" PRIx64
+ich9_gpe_writeb(uint64_t addr, unsigned width, uint64_t val) "addr: 0x%" PRIx64 " width: %d <== 0x%" PRIx64
+
 # pcihp.c
 acpi_pci_eject_slot(unsigned bsel, unsigned slot) "bsel: %u slot: %u"
 acpi_pci_unplug(int bsel, int slot) "bsel: %d slot: %d"
#5 lore raw

RFC PATCH 3/5 hw/i386/acpi-build: Add ACPI PCI hot-plug methods to q35

In reply to Julia Suvorova

Implement notifications and gpe to support q35 ACPI PCI hot-plug. The addresses specified in [1] remain the same to make fewer changes.

[1] docs/spec/acpi_pci_hotplug.txt

---
 hw/i386/acpi-build.c | 20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)
hw/i386/acpi-build.c +13 -7
index 11c598f955..5c5ad88ad6 100644
@@ -201,10 +201,6 @@ static void acpi_get_pm_info(MachineState *machine, AcpiPmInfo *pm)
         /* w2k requires FADT(rev1) or it won't boot, keep PC compatible */
         pm->fadt.rev = 1;
         pm->cpu_hp_io_base = PIIX4_CPU_HOTPLUG_IO_BASE;
-        pm->pcihp_io_base =
-            object_property_get_uint(obj, ACPI_PCIHP_IO_BASE_PROP, NULL);
-        pm->pcihp_io_len =
-            object_property_get_uint(obj, ACPI_PCIHP_IO_LEN_PROP, NULL);
     }
     if (lpc) {
         struct AcpiGenericAddress r = { .space_id = AML_AS_SYSTEM_IO,
@@ -214,6 +210,10 @@ static void acpi_get_pm_info(MachineState *machine, AcpiPmInfo *pm)
         pm->fadt.flags |= 1 << ACPI_FADT_F_RESET_REG_SUP;
         pm->cpu_hp_io_base = ICH9_CPU_HOTPLUG_IO_BASE;
     }
+    pm->pcihp_io_base =
+        object_property_get_uint(obj, ACPI_PCIHP_IO_BASE_PROP, NULL);
+    pm->pcihp_io_len =
+        object_property_get_uint(obj, ACPI_PCIHP_IO_LEN_PROP, NULL);
 
     /* The above need not be conditional on machine type because the reset port
      * happens to be the same on PIIX (pc) and ICH9 (q35). */
@@ -472,7 +472,7 @@ static void build_append_pci_bus_devices(Aml *parent_scope, PCIBus *bus,
         QLIST_FOREACH(sec, &bus->child, sibling) {
             int32_t devfn = sec->parent_dev->devfn;
 
-            if (pci_bus_is_root(sec) || pci_bus_is_express(sec)) {
+            if (pci_bus_is_root(sec)) {
                 continue;
             }
 
@@ -1586,7 +1586,12 @@ static void build_piix4_pci_hotplug(Aml *table)
     aml_append(table, scope);
 }
 
-static Aml *build_q35_osc_method(void)
+static void build_q35_pci_hotplug(Aml *table)
+{
+    build_piix4_pci_hotplug(table);
+}
+
+static Aml *build_q35_osc_method(AcpiPmInfo *pm)
 {
     Aml *if_ctx;
     Aml *if_ctx2;
@@ -1698,6 +1703,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
         build_hpet_aml(dsdt);
         build_q35_isa_bridge(dsdt);
         build_isa_devices_aml(dsdt);
+        build_q35_pci_hotplug(dsdt);
         build_q35_pci0_int(dsdt);
         if (pcms->smbus && !pcmc->do_not_add_smb_acpi) {
             build_smb0(dsdt, pcms->smbus, ICH9_SMB_DEV, ICH9_SMB_FUNC);
@@ -1724,7 +1730,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
     {
         aml_append(scope, aml_name_decl("_HID", aml_string("ACPI0006")));
 
-        if (misc->is_piix4) {
+        if (misc->is_piix4 || pm->pcihp_bridge_en) {
             method = aml_method("_E01", 0, AML_NOTSERIALIZED);
             aml_append(method,
                 aml_acquire(aml_name("\\_SB.PCI0.BLCK"), 0xFFFF));
Signature · 1 line
2.25.4
#6 lore raw

Re: [RFC PATCH 3/5] hw/i386/acpi-build: Add ACPI PCI hot-plug methods to q35

Quoted 4 lines · On Thu, 9 Jul 2020 00:46:13 +0200 Julia Suvorova <[email protected]> wrote:

Implement notifications and gpe to support q35 ACPI PCI hot-plug. The addresses specified in [1] remain the same to make fewer changes.

[1] docs/spec/acpi_pci_hotplug.txt

CCing Gerd his opinion on reusing piix4 IO port range for q35

Quoted 41 lines
---
 hw/i386/acpi-build.c | 20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)
hw/i386/acpi-build.c +5 -5
index 11c598f955..5c5ad88ad6 100644
@@ -201,10 +201,6 @@ static void acpi_get_pm_info(MachineState *machine, AcpiPmInfo *pm)
         /* w2k requires FADT(rev1) or it won't boot, keep PC compatible */
         pm->fadt.rev = 1;
         pm->cpu_hp_io_base = PIIX4_CPU_HOTPLUG_IO_BASE;
-        pm->pcihp_io_base =
-            object_property_get_uint(obj, ACPI_PCIHP_IO_BASE_PROP, NULL);
-        pm->pcihp_io_len =
-            object_property_get_uint(obj, ACPI_PCIHP_IO_LEN_PROP, NULL);
     }
     if (lpc) {
         struct AcpiGenericAddress r = { .space_id = AML_AS_SYSTEM_IO,
@@ -214,6 +210,10 @@ static void acpi_get_pm_info(MachineState *machine, AcpiPmInfo *pm)
         pm->fadt.flags |= 1 << ACPI_FADT_F_RESET_REG_SUP;
         pm->cpu_hp_io_base = ICH9_CPU_HOTPLUG_IO_BASE;
     }
+    pm->pcihp_io_base =
+        object_property_get_uint(obj, ACPI_PCIHP_IO_BASE_PROP, NULL);
+    pm->pcihp_io_len =
+        object_property_get_uint(obj, ACPI_PCIHP_IO_LEN_PROP, NULL);
 
     /* The above need not be conditional on machine type because the reset port
      * happens to be the same on PIIX (pc) and ICH9 (q35). */
@@ -472,7 +472,7 @@ static void build_append_pci_bus_devices(Aml *parent_scope, PCIBus *bus,
         QLIST_FOREACH(sec, &bus->child, sibling) {
             int32_t devfn = sec->parent_dev->devfn;
 
-            if (pci_bus_is_root(sec) || pci_bus_is_express(sec)) {
+            if (pci_bus_is_root(sec)) {
                 continue;
             }
 

hw/i386/acpi-build.c static void build_piix4_pci_hotplug(Aml *table) · line 1592

Quoted 8 lines
hw/i386/acpi-build.c +4 -1
@@ -1586,7 +1586,12 @@ static void build_piix4_pci_hotplug(Aml *table)
     aml_append(table, scope);
 }
 
-static Aml *build_q35_osc_method(void)
+static void build_q35_pci_hotplug(Aml *table)
+{
+    build_piix4_pci_hotplug(table);
+}

s/build_piix4_pci_hotplug/build_i386_acpi_pci_hotplug/

and reuse it in both cases, instead of adding wrapper?

Quoted 22 lines
hw/i386/acpi-build.c +4 -1
+
+static Aml *build_q35_osc_method(AcpiPmInfo *pm)
 {
     Aml *if_ctx;
     Aml *if_ctx2;
@@ -1698,6 +1703,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
         build_hpet_aml(dsdt);
         build_q35_isa_bridge(dsdt);
         build_isa_devices_aml(dsdt);
+        build_q35_pci_hotplug(dsdt);
         build_q35_pci0_int(dsdt);
         if (pcms->smbus && !pcmc->do_not_add_smb_acpi) {
             build_smb0(dsdt, pcms->smbus, ICH9_SMB_DEV, ICH9_SMB_FUNC);
@@ -1724,7 +1730,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
     {
         aml_append(scope, aml_name_decl("_HID", aml_string("ACPI0006")));
 
-        if (misc->is_piix4) {
+        if (misc->is_piix4 || pm->pcihp_bridge_en) {
             method = aml_method("_E01", 0, AML_NOTSERIALIZED);
             aml_append(method,
                 aml_acquire(aml_name("\\_SB.PCI0.BLCK"), 0xFFFF));
#7 lore raw

Re: [RFC PATCH 3/5] hw/i386/acpi-build: Add ACPI PCI hot-plug methods to q35

Quoted 63 lines · Igor Mammedov
Quoted 4 lines · On Thu, 9 Jul 2020 00:46:13 +0200 Julia Suvorova <[email protected]> wrote:

Implement notifications and gpe to support q35 ACPI PCI hot-plug. The addresses specified in [1] remain the same to make fewer changes.

[1] docs/spec/acpi_pci_hotplug.txt

CCing Gerd his opinion on reusing piix4 IO port range for q35

Quoted 41 lines
---
 hw/i386/acpi-build.c | 20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)
hw/i386/acpi-build.c +5 -5
index 11c598f955..5c5ad88ad6 100644
@@ -201,10 +201,6 @@ static void acpi_get_pm_info(MachineState *machine, AcpiPmInfo *pm)
         /* w2k requires FADT(rev1) or it won't boot, keep PC compatible */
         pm->fadt.rev = 1;
         pm->cpu_hp_io_base = PIIX4_CPU_HOTPLUG_IO_BASE;
-        pm->pcihp_io_base =
-            object_property_get_uint(obj, ACPI_PCIHP_IO_BASE_PROP, NULL);
-        pm->pcihp_io_len =
-            object_property_get_uint(obj, ACPI_PCIHP_IO_LEN_PROP, NULL);
     }
     if (lpc) {
         struct AcpiGenericAddress r = { .space_id = AML_AS_SYSTEM_IO,
@@ -214,6 +210,10 @@ static void acpi_get_pm_info(MachineState *machine, AcpiPmInfo *pm)
         pm->fadt.flags |= 1 << ACPI_FADT_F_RESET_REG_SUP;
         pm->cpu_hp_io_base = ICH9_CPU_HOTPLUG_IO_BASE;
     }
+    pm->pcihp_io_base =
+        object_property_get_uint(obj, ACPI_PCIHP_IO_BASE_PROP, NULL);
+    pm->pcihp_io_len =
+        object_property_get_uint(obj, ACPI_PCIHP_IO_LEN_PROP, NULL);
 
     /* The above need not be conditional on machine type because the reset port
      * happens to be the same on PIIX (pc) and ICH9 (q35). */
@@ -472,7 +472,7 @@ static void build_append_pci_bus_devices(Aml *parent_scope, PCIBus *bus,
         QLIST_FOREACH(sec, &bus->child, sibling) {
             int32_t devfn = sec->parent_dev->devfn;
 
-            if (pci_bus_is_root(sec) || pci_bus_is_express(sec)) {
+            if (pci_bus_is_root(sec)) {
                 continue;
             }
 

hw/i386/acpi-build.c static void build_piix4_pci_hotplug(Aml *table) · line 1592

Quoted 8 lines
hw/i386/acpi-build.c +4 -1
@@ -1586,7 +1586,12 @@ static void build_piix4_pci_hotplug(Aml *table)
     aml_append(table, scope);
 }
 
-static Aml *build_q35_osc_method(void)
+static void build_q35_pci_hotplug(Aml *table)
+{
+    build_piix4_pci_hotplug(table);
+}

s/build_piix4_pci_hotplug/build_i386_acpi_pci_hotplug/

Quoted 1 line

and reuse it in both cases, instead of adding wrapper?

I'm not sure about that - we have microvm too ...

Quoted 22 lines
Quoted 22 lines
hw/i386/acpi-build.c +4 -1
+
+static Aml *build_q35_osc_method(AcpiPmInfo *pm)
 {
     Aml *if_ctx;
     Aml *if_ctx2;
@@ -1698,6 +1703,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
         build_hpet_aml(dsdt);
         build_q35_isa_bridge(dsdt);
         build_isa_devices_aml(dsdt);
+        build_q35_pci_hotplug(dsdt);
         build_q35_pci0_int(dsdt);
         if (pcms->smbus && !pcmc->do_not_add_smb_acpi) {
             build_smb0(dsdt, pcms->smbus, ICH9_SMB_DEV, ICH9_SMB_FUNC);
@@ -1724,7 +1730,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
     {
         aml_append(scope, aml_name_decl("_HID", aml_string("ACPI0006")));
 
-        if (misc->is_piix4) {
+        if (misc->is_piix4 || pm->pcihp_bridge_en) {
             method = aml_method("_E01", 0, AML_NOTSERIALIZED);
             aml_append(method,
                 aml_acquire(aml_name("\\_SB.PCI0.BLCK"), 0xFFFF));
#8 lore raw

Re: [RFC PATCH 3/5] hw/i386/acpi-build: Add ACPI PCI hot-plug methods to q35

Quoted 64 lines · On Tue, 14 Jul 2020 05:22:20 -0400 "Michael S. Tsirkin" <[email protected]> wrote:
Quoted 63 lines · Igor Mammedov
Quoted 4 lines · On Thu, 9 Jul 2020 00:46:13 +0200 Julia Suvorova <[email protected]> wrote:

Implement notifications and gpe to support q35 ACPI PCI hot-plug. The addresses specified in [1] remain the same to make fewer changes.

[1] docs/spec/acpi_pci_hotplug.txt

CCing Gerd his opinion on reusing piix4 IO port range for q35

Quoted 41 lines
---
 hw/i386/acpi-build.c | 20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)
hw/i386/acpi-build.c +5 -5
index 11c598f955..5c5ad88ad6 100644
@@ -201,10 +201,6 @@ static void acpi_get_pm_info(MachineState *machine, AcpiPmInfo *pm)
         /* w2k requires FADT(rev1) or it won't boot, keep PC compatible */
         pm->fadt.rev = 1;
         pm->cpu_hp_io_base = PIIX4_CPU_HOTPLUG_IO_BASE;
-        pm->pcihp_io_base =
-            object_property_get_uint(obj, ACPI_PCIHP_IO_BASE_PROP, NULL);
-        pm->pcihp_io_len =
-            object_property_get_uint(obj, ACPI_PCIHP_IO_LEN_PROP, NULL);
     }
     if (lpc) {
         struct AcpiGenericAddress r = { .space_id = AML_AS_SYSTEM_IO,
@@ -214,6 +210,10 @@ static void acpi_get_pm_info(MachineState *machine, AcpiPmInfo *pm)
         pm->fadt.flags |= 1 << ACPI_FADT_F_RESET_REG_SUP;
         pm->cpu_hp_io_base = ICH9_CPU_HOTPLUG_IO_BASE;
     }
+    pm->pcihp_io_base =
+        object_property_get_uint(obj, ACPI_PCIHP_IO_BASE_PROP, NULL);
+    pm->pcihp_io_len =
+        object_property_get_uint(obj, ACPI_PCIHP_IO_LEN_PROP, NULL);
 
     /* The above need not be conditional on machine type because the reset port
      * happens to be the same on PIIX (pc) and ICH9 (q35). */
@@ -472,7 +472,7 @@ static void build_append_pci_bus_devices(Aml *parent_scope, PCIBus *bus,
         QLIST_FOREACH(sec, &bus->child, sibling) {
             int32_t devfn = sec->parent_dev->devfn;
 
-            if (pci_bus_is_root(sec) || pci_bus_is_express(sec)) {
+            if (pci_bus_is_root(sec)) {
                 continue;
             }
 

hw/i386/acpi-build.c static void build_piix4_pci_hotplug(Aml *table) · line 1592

Quoted 8 lines
hw/i386/acpi-build.c +4 -1
@@ -1586,7 +1586,12 @@ static void build_piix4_pci_hotplug(Aml *table)
     aml_append(table, scope);
 }
 
-static Aml *build_q35_osc_method(void)
+static void build_q35_pci_hotplug(Aml *table)
+{
+    build_piix4_pci_hotplug(table);
+}  

s/build_piix4_pci_hotplug/build_i386_acpi_pci_hotplug/

Quoted 3 lines
Quoted 1 line

and reuse it in both cases, instead of adding wrapper?

I'm not sure about that - we have microvm too ...

it doesn't have pci if I'm not mistaken,
and when/if it gains one someday, it may or maynot use hotplug it will be up to its specific DSDT to include this call.

What doesn't make sense is using board specific naming here for the same code. I don't insist on the variant I've suggested, only on consolidating it instead of adding dummy wrapper

Quoted 23 lines
Quoted 22 lines
Quoted 22 lines
hw/i386/acpi-build.c +4 -1
+
+static Aml *build_q35_osc_method(AcpiPmInfo *pm)
 {
     Aml *if_ctx;
     Aml *if_ctx2;
@@ -1698,6 +1703,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
         build_hpet_aml(dsdt);
         build_q35_isa_bridge(dsdt);
         build_isa_devices_aml(dsdt);
+        build_q35_pci_hotplug(dsdt);
         build_q35_pci0_int(dsdt);
         if (pcms->smbus && !pcmc->do_not_add_smb_acpi) {
             build_smb0(dsdt, pcms->smbus, ICH9_SMB_DEV, ICH9_SMB_FUNC);
@@ -1724,7 +1730,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
     {
         aml_append(scope, aml_name_decl("_HID", aml_string("ACPI0006")));
 
-        if (misc->is_piix4) {
+        if (misc->is_piix4 || pm->pcihp_bridge_en) {
             method = aml_method("_E01", 0, AML_NOTSERIALIZED);
             aml_append(method,
                 aml_acquire(aml_name("\\_SB.PCI0.BLCK"), 0xFFFF));  
#9 lore raw

Re: [RFC PATCH 3/5] hw/i386/acpi-build: Add ACPI PCI hot-plug methods to q35

In reply to Igor Mammedov
Quoted 9 lines · Igor Mammedov
Quoted 4 lines · On Thu, 9 Jul 2020 00:46:13 +0200 Julia Suvorova <[email protected]> wrote:

Implement notifications and gpe to support q35 ACPI PCI hot-plug. The addresses specified in [1] remain the same to make fewer changes.

[1] docs/spec/acpi_pci_hotplug.txt

CCing Gerd his opinion on reusing piix4 IO port range for q35

Oh no, please don't. Grabbing random io ports is asking for trouble, especially on q35 where you only need enough pcie devices to have the port range 0xae00 -> 0xae0f actually overlap with a pci bridge window.

Ideally we'll find some unused spot in the hidden pci bar of the ich9 pm device. At least the qemu implementation has some room:

    0000000000000600-000000000000067f (prio 0, i/o): ich9-pm
      0000000000000600-0000000000000603 (prio 0, i/o): acpi-evt
      0000000000000604-0000000000000605 (prio 0, i/o): acpi-cnt
      0000000000000608-000000000000060b (prio 0, i/o): acpi-tmr
      0000000000000620-000000000000062f (prio 0, i/o): acpi-gpe0
      0000000000000630-0000000000000637 (prio 0, i/o): acpi-smi
      0000000000000660-000000000000067f (prio 0, i/o): sm-tco

Offset 0x40 seems to be unused for example (but better check the chipset specs to see if that is really unused or whenever the qemu emulation is incomplete).

If that doesn't work out pick an io range which is unlikely to conflict with something. That excludes anything above > 0x1000 (pci bridge windows) and anything below 0x3ff (legacy isa). From the remaining range 0x400 -> 0xfff the 0xc00 -> 0xcff block is the best candidate I think. Because of the pci config registers being there it is unlikely that the firmware places something in that range.

Oh, I see the cpu hotplug registers are already there:

    [ ... ]
    0000000000000700-000000000000073f (prio 1, i/o): pm-smbus
    0000000000000cd8-0000000000000ce3 (prio 0, i/o): acpi-cpu-hotplug
    0000000000000cf8-0000000000000cfb (prio 0, i/o): pci-conf-idx
    0000000000000cf9-0000000000000cf9 (prio 1, i/o): lpc-reset-control
    0000000000000cfc-0000000000000cff (prio 0, i/o): pci-conf-data
    [ ... ]

So placing the pci hotplug registers next to them (say at 0xcc8) looks like a good pick to me.

While being at it: Shouldn't we reserve these port ranges somehow? Using an acpi device for example, simliar to fw_cfg? The guest OS should better know there is something at those ports ...

take care,
Gerd

#10 lore raw

Re: [RFC PATCH 3/5] hw/i386/acpi-build: Add ACPI PCI hot-plug methods to q35

Quoted 10 lines · On Wed, 15 Jul 2020 08:57:51 +0200 Gerd Hoffmann <[email protected]> wrote:
Quoted 9 lines · Igor Mammedov
Quoted 4 lines · On Thu, 9 Jul 2020 00:46:13 +0200 Julia Suvorova <[email protected]> wrote:

Implement notifications and gpe to support q35 ACPI PCI hot-plug. The addresses specified in [1] remain the same to make fewer changes.

[1] docs/spec/acpi_pci_hotplug.txt

CCing Gerd his opinion on reusing piix4 IO port range for q35

[...]

Quoted 3 lines

While being at it: Shouldn't we reserve these port ranges somehow? Using an acpi device for example, simliar to fw_cfg? The guest OS should better know there is something at those ports ...

we do it at ACPI level in DSDT, look for comment
/* reserve PCIHP resources */

It should make Windows trip over in case of another range overlap with reserved ports. (linux kernel is more tolerant and may silently ignore or print a warning)

Quoted 3 lines

take care,
Gerd

#11 lore raw

Re: [RFC PATCH 3/5] hw/i386/acpi-build: Add ACPI PCI hot-plug methods to q35

Quoted 6 lines
Quoted 3 lines

While being at it: Shouldn't we reserve these port ranges somehow? Using an acpi device for example, simliar to fw_cfg? The guest OS should better know there is something at those ports ...

we do it at ACPI level in DSDT, look for comment
/* reserve PCIHP resources */

Ah, good.

Quoted 3 lines

It should make Windows trip over in case of another range overlap with reserved ports. (linux kernel is more tolerant and may silently ignore or print a warning)

Hmm, linux doesn't list the device, can't see it neither in the kernel log nor in /proc/ioports ...

take care,
Gerd

#12 lore raw

RFC PATCH 4/5 hw/i386/acpi-build: Turn off support of PCIe native hot-plug and SHPC in _OSC

In reply to Julia Suvorova

Other methods may be used if the system is capable of this and the _OSC bit is set. Disable them explicitly to force ACPI PCI hot-plug use. The older versions will still use PCIe native.

---
 hw/i386/acpi-build.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)
hw/i386/acpi-build.c +8 -4
index 5c5ad88ad6..0e2891c3ea 100644
@@ -1599,6 +1599,7 @@ static Aml *build_q35_osc_method(AcpiPmInfo *pm)
     Aml *method;
     Aml *a_cwd1 = aml_name("CDW1");
     Aml *a_ctrl = aml_local(0);
+    unsigned osc_ctrl;
 
     method = aml_method("_OSC", 4, AML_NOTSERIALIZED);
     aml_append(method, aml_create_dword_field(aml_arg(3), aml_int(0), "CDW1"));
@@ -1612,9 +1613,12 @@ static Aml *build_q35_osc_method(AcpiPmInfo *pm)
 
     /*
      * Always allow native PME, AER (no dependencies)
-     * Allow SHPC (PCI bridges can have SHPC controller)
+     * Need to disable native and SHPC hot-plug to use acpihp
+     *
+     * PCI Firmware Specification, Revision 3.2
      */
-    aml_append(if_ctx, aml_and(a_ctrl, aml_int(0x1F), a_ctrl));
+    osc_ctrl = pm->pcihp_bridge_en ? 0x1C : 0x1F;
+    aml_append(if_ctx, aml_and(a_ctrl, aml_int(osc_ctrl), a_ctrl));
 
     if_ctx2 = aml_if(aml_lnot(aml_equal(aml_arg(1), aml_int(1))));
     /* Unknown revision */
@@ -1696,7 +1700,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
         aml_append(dev, aml_name_decl("_CID", aml_eisaid("PNP0A03")));
         aml_append(dev, aml_name_decl("_ADR", aml_int(0)));
         aml_append(dev, aml_name_decl("_UID", aml_int(1)));
-        aml_append(dev, build_q35_osc_method());
+        aml_append(dev, build_q35_osc_method(pm));
         aml_append(sb_scope, dev);
         aml_append(dsdt, sb_scope);
 
@@ -1771,7 +1775,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
             if (pci_bus_is_express(bus)) {
                 aml_append(dev, aml_name_decl("_HID", aml_eisaid("PNP0A08")));
                 aml_append(dev, aml_name_decl("_CID", aml_eisaid("PNP0A03")));
-                aml_append(dev, build_q35_osc_method());
+                aml_append(dev, build_q35_osc_method(pm));
             } else {
                 aml_append(dev, aml_name_decl("_HID", aml_eisaid("PNP0A03")));
             }
Signature · 1 line
2.25.4
#13 lore raw

Re: [RFC PATCH 4/5] hw/i386/acpi-build: Turn off support of PCIe native hot-plug and SHPC in _OSC

Quoted 22 lines · On Thu, 9 Jul 2020 00:46:14 +0200 Julia Suvorova <[email protected]> wrote:

Other methods may be used if the system is capable of this and the _OSC bit is set. Disable them explicitly to force ACPI PCI hot-plug use. The older versions will still use PCIe native.

---
 hw/i386/acpi-build.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)
hw/i386/acpi-build.c +1 -0
index 5c5ad88ad6..0e2891c3ea 100644
@@ -1599,6 +1599,7 @@ static Aml *build_q35_osc_method(AcpiPmInfo *pm)
     Aml *method;
     Aml *a_cwd1 = aml_name("CDW1");
     Aml *a_ctrl = aml_local(0);
+    unsigned osc_ctrl;
 
     method = aml_method("_OSC", 4, AML_NOTSERIALIZED);
     aml_append(method, aml_create_dword_field(aml_arg(3), aml_int(0), "CDW1"));

hw/i386/acpi-build.c static Aml *build_q35_osc_method(AcpiPmInfo *pm) · line 1618

Quoted 7 lines
hw/i386/acpi-build.c +3 -1
@@ -1612,9 +1613,12 @@ static Aml *build_q35_osc_method(AcpiPmInfo *pm)
 
     /*
      * Always allow native PME, AER (no dependencies)
-     * Allow SHPC (PCI bridges can have SHPC controller)
+     * Need to disable native and SHPC hot-plug to use acpihp
+     *
+     * PCI Firmware Specification, Revision 3.2

seems incomplete, were you going to point to a concrete chapter that requires this change?

hw/i386/acpi-build.c static Aml *build_q35_osc_method(AcpiPmInfo *pm) · line 1620

Quoted 3 lines
hw/i386/acpi-build.c +1 -1
      */
-    aml_append(if_ctx, aml_and(a_ctrl, aml_int(0x1F), a_ctrl));
+    osc_ctrl = pm->pcihp_bridge_en ? 0x1C : 0x1F;

Since you are touching this, how about converting this magic number to something more readable?
i.e.

            set_bit(ACPI_OSC_SHPC_EN,  osc_ctrl)
          or
            osc_ctrl |= BIT(SOME_FEATURE)
Quoted 22 lines
hw/i386/acpi-build.c +3 -2
+    aml_append(if_ctx, aml_and(a_ctrl, aml_int(osc_ctrl), a_ctrl));
 
     if_ctx2 = aml_if(aml_lnot(aml_equal(aml_arg(1), aml_int(1))));
     /* Unknown revision */
@@ -1696,7 +1700,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
         aml_append(dev, aml_name_decl("_CID", aml_eisaid("PNP0A03")));
         aml_append(dev, aml_name_decl("_ADR", aml_int(0)));
         aml_append(dev, aml_name_decl("_UID", aml_int(1)));
-        aml_append(dev, build_q35_osc_method());
+        aml_append(dev, build_q35_osc_method(pm));
         aml_append(sb_scope, dev);
         aml_append(dsdt, sb_scope);
 
@@ -1771,7 +1775,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
             if (pci_bus_is_express(bus)) {
                 aml_append(dev, aml_name_decl("_HID", aml_eisaid("PNP0A08")));
                 aml_append(dev, aml_name_decl("_CID", aml_eisaid("PNP0A03")));
-                aml_append(dev, build_q35_osc_method());
+                aml_append(dev, build_q35_osc_method(pm));
             } else {
                 aml_append(dev, aml_name_decl("_HID", aml_eisaid("PNP0A03")));
             }
#14 lore raw

Re: [RFC PATCH 4/5] hw/i386/acpi-build: Turn off support of PCIe native hot-plug and SHPC in _OSC

Quoted 6 lines · Igor Mammedov
Quoted 3 lines · On Thu, 9 Jul 2020 00:46:14 +0200 Julia Suvorova <[email protected]> wrote:

Other methods may be used if the system is capable of this and the _OSC bit is set. Disable them explicitly to force ACPI PCI hot-plug use. The older versions will still use PCIe native.

Do we need that later part btw?

Quoted 20 lines
Quoted 20 lines
---
 hw/i386/acpi-build.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)
hw/i386/acpi-build.c +1 -0
index 5c5ad88ad6..0e2891c3ea 100644
@@ -1599,6 +1599,7 @@ static Aml *build_q35_osc_method(AcpiPmInfo *pm)
     Aml *method;
     Aml *a_cwd1 = aml_name("CDW1");
     Aml *a_ctrl = aml_local(0);
+    unsigned osc_ctrl;
 
     method = aml_method("_OSC", 4, AML_NOTSERIALIZED);
     aml_append(method, aml_create_dword_field(aml_arg(3), aml_int(0), "CDW1"));
@@ -1612,9 +1613,12 @@ static Aml *build_q35_osc_method(AcpiPmInfo *pm)
 
Quoted 6 lines

hw/i386/acpi-build.c static Aml *build_q35_osc_method(AcpiPmInfo *pm) · line 1618

Quoted 6 lines
hw/i386/acpi-build.c +3 -1
     /*
      * Always allow native PME, AER (no dependencies)
-     * Allow SHPC (PCI bridges can have SHPC controller)
+     * Need to disable native and SHPC hot-plug to use acpihp
+     *
+     * PCI Firmware Specification, Revision 3.2

I don't think you have to add a reference as part of this patchset. The spec in question documents _OSC so it's not a bad idea to add it, but it's a bit more work, e.g. we generally try to list the earliest spec that documents the given feature, since So I suspect this is 3.0 actually.

Quoted 1 line

seems incomplete, were you going to point to a concrete chapter that requires this change?

It doesn't as such. A better description would be:

/ * Guests seem to generally prefer native hotplug control. As we want them to
* use ACPI, don't enable it.
*/

Quoted 10 lines

hw/i386/acpi-build.c static Aml *build_q35_osc_method(AcpiPmInfo *pm) · line 1620

Quoted 3 lines
hw/i386/acpi-build.c +1 -1
      */
-    aml_append(if_ctx, aml_and(a_ctrl, aml_int(0x1F), a_ctrl));
+    osc_ctrl = pm->pcihp_bridge_en ? 0x1C : 0x1F;

Since you are touching this, how about converting this magic number to something more readable?
i.e.

            set_bit(ACPI_OSC_SHPC_EN,  osc_ctrl)
          or
            osc_ctrl |= BIT(SOME_FEATURE)

... if there is such a macro. If not I suspect it's better as a comment:

	0x10 /* PCI Express Capability Structure control */ |
	0x80 /* PCI Express Advanced Error Reporting control */ |
	0x40 /* PCI Express Native Power Management Events control */
Quoted 22 lines
Quoted 22 lines
hw/i386/acpi-build.c +3 -2
+    aml_append(if_ctx, aml_and(a_ctrl, aml_int(osc_ctrl), a_ctrl));
 
     if_ctx2 = aml_if(aml_lnot(aml_equal(aml_arg(1), aml_int(1))));
     /* Unknown revision */
@@ -1696,7 +1700,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
         aml_append(dev, aml_name_decl("_CID", aml_eisaid("PNP0A03")));
         aml_append(dev, aml_name_decl("_ADR", aml_int(0)));
         aml_append(dev, aml_name_decl("_UID", aml_int(1)));
-        aml_append(dev, build_q35_osc_method());
+        aml_append(dev, build_q35_osc_method(pm));
         aml_append(sb_scope, dev);
         aml_append(dsdt, sb_scope);
 
@@ -1771,7 +1775,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
             if (pci_bus_is_express(bus)) {
                 aml_append(dev, aml_name_decl("_HID", aml_eisaid("PNP0A08")));
                 aml_append(dev, aml_name_decl("_CID", aml_eisaid("PNP0A03")));
-                aml_append(dev, build_q35_osc_method());
+                aml_append(dev, build_q35_osc_method(pm));
             } else {
                 aml_append(dev, aml_name_decl("_HID", aml_eisaid("PNP0A03")));
             }
#15 lore raw

Re: [RFC PATCH 4/5] hw/i386/acpi-build: Turn off support of PCIe native hot-plug and SHPC in _OSC

Quoted 66 lines · On Tue, 14 Jul 2020 04:39:53 -0400 "Michael S. Tsirkin" <[email protected]> wrote:
Quoted 6 lines · Igor Mammedov
Quoted 3 lines · On Thu, 9 Jul 2020 00:46:14 +0200 Julia Suvorova <[email protected]> wrote:

Other methods may be used if the system is capable of this and the _OSC bit is set. Disable them explicitly to force ACPI PCI hot-plug use. The older versions will still use PCIe native.

Do we need that later part btw?

Quoted 20 lines
Quoted 20 lines
---
 hw/i386/acpi-build.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)
hw/i386/acpi-build.c +1 -0
index 5c5ad88ad6..0e2891c3ea 100644
@@ -1599,6 +1599,7 @@ static Aml *build_q35_osc_method(AcpiPmInfo *pm)
     Aml *method;
     Aml *a_cwd1 = aml_name("CDW1");
     Aml *a_ctrl = aml_local(0);
+    unsigned osc_ctrl;
 
     method = aml_method("_OSC", 4, AML_NOTSERIALIZED);
     aml_append(method, aml_create_dword_field(aml_arg(3), aml_int(0), "CDW1"));
@@ -1612,9 +1613,12 @@ static Aml *build_q35_osc_method(AcpiPmInfo *pm)
 
Quoted 6 lines

hw/i386/acpi-build.c static Aml *build_q35_osc_method(AcpiPmInfo *pm) · line 1618

Quoted 6 lines
hw/i386/acpi-build.c +3 -1
     /*
      * Always allow native PME, AER (no dependencies)
-     * Allow SHPC (PCI bridges can have SHPC controller)
+     * Need to disable native and SHPC hot-plug to use acpihp
+     *
+     * PCI Firmware Specification, Revision 3.2  

I don't think you have to add a reference as part of this patchset. The spec in question documents _OSC so it's not a bad idea to add it, but it's a bit more work, e.g. we generally try to list the earliest spec that documents the given feature, since So I suspect this is 3.0 actually.

Quoted 1 line

seems incomplete, were you going to point to a concrete chapter that requires this change?

It doesn't as such. A better description would be:

/ * Guests seem to generally prefer native hotplug control. As we want them to
* use ACPI, don't enable it.
*/

Quoted 10 lines

hw/i386/acpi-build.c static Aml *build_q35_osc_method(AcpiPmInfo *pm) · line 1620

Quoted 3 lines
hw/i386/acpi-build.c +1 -1
      */
-    aml_append(if_ctx, aml_and(a_ctrl, aml_int(0x1F), a_ctrl));
+    osc_ctrl = pm->pcihp_bridge_en ? 0x1C : 0x1F;  

Since you are touching this, how about converting this magic number to something more readable?
i.e.

            set_bit(ACPI_OSC_SHPC_EN,  osc_ctrl)
          or
            osc_ctrl |= BIT(SOME_FEATURE)
Quoted 5 lines

... if there is such a macro. If not I suspect it's better as a comment:

	0x10 /* PCI Express Capability Structure control */ |
	0x80 /* PCI Express Advanced Error Reporting control */ |
	0x40 /* PCI Express Native Power Management Events control */

if that is a spec quoted nearby than, I like comments idea as it follows the same style which we use with ACPI numbers. We just need split single magic number onto separate features bits so it would be readable.

Quoted 27 lines
Quoted 25 lines
Quoted 25 lines
hw/i386/acpi-build.c +3 -2
+    aml_append(if_ctx, aml_and(a_ctrl, aml_int(osc_ctrl), a_ctrl));
 
     if_ctx2 = aml_if(aml_lnot(aml_equal(aml_arg(1), aml_int(1))));
     /* Unknown revision */
@@ -1696,7 +1700,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
         aml_append(dev, aml_name_decl("_CID", aml_eisaid("PNP0A03")));
         aml_append(dev, aml_name_decl("_ADR", aml_int(0)));
         aml_append(dev, aml_name_decl("_UID", aml_int(1)));
-        aml_append(dev, build_q35_osc_method());
+        aml_append(dev, build_q35_osc_method(pm));
         aml_append(sb_scope, dev);
         aml_append(dsdt, sb_scope);
 
@@ -1771,7 +1775,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
             if (pci_bus_is_express(bus)) {
                 aml_append(dev, aml_name_decl("_HID", aml_eisaid("PNP0A08")));
                 aml_append(dev, aml_name_decl("_CID", aml_eisaid("PNP0A03")));
-                aml_append(dev, build_q35_osc_method());
+                aml_append(dev, build_q35_osc_method(pm));
             } else {
                 aml_append(dev, aml_name_decl("_HID", aml_eisaid("PNP0A03")));
             }  
#16 lore raw

RFC PATCH 5/5 hw/acpi/ich9: Enable ACPI PCI hot-plug

In reply to Julia Suvorova

Add acpi_pcihp to ich9_pm and use ACPI PCI hot-plug by default.

---
 include/hw/acpi/ich9.h |  3 +++
 hw/acpi/ich9.c         | 45 ++++++++++++++++++++++++++++++++++++++++++
 hw/acpi/pcihp.c        |  3 ++-
 hw/i386/pc.c           |  4 +++-
 4 files changed, 53 insertions(+), 2 deletions(-)
include/hw/acpi/ich9.h +3 -0
index 28a53181cb..d345da6b74 100644
@@ -24,6 +24,7 @@
 #include "hw/acpi/acpi.h"
 #include "hw/acpi/cpu_hotplug.h"
 #include "hw/acpi/cpu.h"
+#include "hw/acpi/pcihp.h"
 #include "hw/acpi/memory_hotplug.h"
 #include "hw/acpi/acpi_dev_interface.h"
 #include "hw/acpi/tco.h"
@@ -53,6 +54,8 @@ typedef struct ICH9LPCPMRegs {
     AcpiCpuHotplug gpe_cpu;
     CPUHotplugState cpuhp_state;
 
+    bool use_acpi_pci_hotplug;
+    AcpiPciHpState acpi_pci_hotplug;
     MemHotplugState acpi_memory_hotplug;
 
     uint8_t disable_s3;
hw/acpi/ich9.c +45 -0

99 lines not on this page · raw message

hw/acpi/pcihp.c +2 -1
index 3d4ee3af72..d905d1b8f2 100644
@@ -110,7 +110,8 @@ static void acpi_set_pci_info(void)
     }
     bsel_is_set = true;
 
-    bus = find_i440fx(); /* TODO: Q35 support */
+    bus = find_host();
+
     if (bus) {
         /* Scan all PCI buses. Set property to enable acpi based hotplug. */
         pci_for_each_bus_depth_first(bus, acpi_set_bsel, NULL, &bsel_alloc);
hw/i386/pc.c +3 -1
index 143ac1c354..21c6eb779e 100644
@@ -96,7 +96,9 @@
 #include "fw_cfg.h"
 #include "trace.h"
 
-GlobalProperty pc_compat_5_0[] = {};
+GlobalProperty pc_compat_5_0[] = {
+    { "ICH9-LPC", "acpi-pci-hotplug-with-bridge-support", "off" },
+};
 const size_t pc_compat_5_0_len = G_N_ELEMENTS(pc_compat_5_0);
 
 GlobalProperty pc_compat_4_2[] = {
Signature · 1 line
2.25.4
#17 lore raw

Re: [RFC PATCH 5/5] hw/acpi/ich9: Enable ACPI PCI hot-plug

Quoted 23 lines · On Thu, 9 Jul 2020 00:46:15 +0200 Julia Suvorova <[email protected]> wrote:

Add acpi_pcihp to ich9_pm and use ACPI PCI hot-plug by default.

---
 include/hw/acpi/ich9.h |  3 +++
 hw/acpi/ich9.c         | 45 ++++++++++++++++++++++++++++++++++++++++++
 hw/acpi/pcihp.c        |  3 ++-
 hw/i386/pc.c           |  4 +++-
 4 files changed, 53 insertions(+), 2 deletions(-)
include/hw/acpi/ich9.h +1 -0
index 28a53181cb..d345da6b74 100644
@@ -24,6 +24,7 @@
 #include "hw/acpi/acpi.h"
 #include "hw/acpi/cpu_hotplug.h"
 #include "hw/acpi/cpu.h"
+#include "hw/acpi/pcihp.h"
 #include "hw/acpi/memory_hotplug.h"
 #include "hw/acpi/acpi_dev_interface.h"
 #include "hw/acpi/tco.h"

include/hw/acpi/ich9.h typedef struct ICH9LPCPMRegs { · line 57

Quoted 4 lines
include/hw/acpi/ich9.h +1 -0
@@ -53,6 +54,8 @@ typedef struct ICH9LPCPMRegs {
     AcpiCpuHotplug gpe_cpu;
     CPUHotplugState cpuhp_state;
 
+    bool use_acpi_pci_hotplug;

a bit outdated naming,
see 0affda043675c7619248a924a89bfd3781759f63, and rename it to match piix4

Quoted 56 lines
include/hw/acpi/ich9.h +1 -0
+    AcpiPciHpState acpi_pci_hotplug;
     MemHotplugState acpi_memory_hotplug;
 
     uint8_t disable_s3;
hw/acpi/ich9.c +26 -0
index 0fdd736da4..e0291373f2 100644
@@ -265,6 +265,7 @@ static void pm_reset(void *opaque)
     }
     pm->smi_en_wmask = ~0;
 
+    acpi_pcihp_reset(&pm->acpi_pci_hotplug);
     acpi_update_sci(&pm->acpi_regs, pm->irq);
 }
 
@@ -303,6 +304,17 @@ void ich9_pm_init(PCIDevice *lpc_pci, ICH9LPCPMRegs *pm,
     pm->enable_tco = true;
     acpi_pm_tco_init(&pm->tco_regs, &pm->io);
 
+    if (pm->use_acpi_pci_hotplug) {
+        acpi_pcihp_init(OBJECT(lpc_pci),
+                        &pm->acpi_pci_hotplug,
+                        pci_get_bus(lpc_pci),
+                        pci_address_space_io(lpc_pci),
+                        true);
+        qbus_set_hotplug_handler(BUS(pci_get_bus(lpc_pci)),
+                                 OBJECT(lpc_pci),
+                                 &error_abort);
+    }
+
     pm->irq = sci_irq;
     qemu_register_reset(pm_reset, pm);
     pm->powerdown_notifier.notify = pm_powerdown_req;
@@ -374,6 +386,20 @@ static void ich9_pm_set_enable_tco(Object *obj, bool value, Error **errp)
     s->pm.enable_tco = value;
 }
 
+static bool ich9_pm_get_acpi_pci_hotplug(Object *obj, Error **errp)
+{
+    ICH9LPCState *s = ICH9_LPC_DEVICE(obj);
+
+    return s->pm.use_acpi_pci_hotplug;
+}
+
+static void ich9_pm_set_acpi_pci_hotplug(Object *obj, bool value,
+                                               Error **errp)
+{
+    ICH9LPCState *s = ICH9_LPC_DEVICE(obj);
+
+    s->pm.use_acpi_pci_hotplug = value;
+}
 void ich9_pm_add_properties(Object *obj, ICH9LPCPMRegs *pm)
 {
     static const uint32_t gpe0_len = ICH9_PMIO_GPE0_LEN;

hw/acpi/ich9.c void ich9_pm_add_properties(Object *obj, ICH9LPCPMRegs *pm) · line 411

Quoted 4 lines
hw/acpi/ich9.c +1 -0
@@ -382,6 +408,7 @@ void ich9_pm_add_properties(Object *obj, ICH9LPCPMRegs *pm)
     pm->disable_s3 = 0;
     pm->disable_s4 = 0;
     pm->s4_val = 2;
+    pm->use_acpi_pci_hotplug = true;

I'm not sure about making crutch on by default though.

Does linux guests work fine with native hotplug or does it have the same issues as Windows?

Also you had an idea how to workaround native hotplug issues on QEMU side.
Can you dump here a quick summary why it didn't work out in the end?

Quoted 85 lines
hw/acpi/ich9.c +18 -0
     object_property_add_uint32_ptr(obj, ACPI_PM_PROP_PM_IO_BASE,
                                    &pm->pm_io_base, OBJ_PROP_FLAG_READ);
@@ -405,6 +432,9 @@ void ich9_pm_add_properties(Object *obj, ICH9LPCPMRegs *pm)
     object_property_add_bool(obj, ACPI_PM_PROP_TCO_ENABLED,
                              ich9_pm_get_enable_tco,
                              ich9_pm_set_enable_tco);
+    object_property_add_bool(obj, "acpi-pci-hotplug-with-bridge-support",
+                             ich9_pm_get_acpi_pci_hotplug,
+                             ich9_pm_set_acpi_pci_hotplug);
 }
 
 void ich9_pm_device_pre_plug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
@@ -412,6 +442,11 @@ void ich9_pm_device_pre_plug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
 {
     ICH9LPCState *lpc = ICH9_LPC_DEVICE(hotplug_dev);
 
+    if (object_dynamic_cast(OBJECT(dev), TYPE_PCI_DEVICE)) {
+        acpi_pcihp_device_pre_plug_cb(hotplug_dev, dev, errp);
+        return;
+    }
+
     if (object_dynamic_cast(OBJECT(dev), TYPE_PC_DIMM) &&
         !lpc->pm.acpi_memory_hotplug.is_enabled)
         error_setg(errp,
@@ -437,6 +472,9 @@ void ich9_pm_device_plug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
         } else {
             acpi_cpu_plug_cb(hotplug_dev, &lpc->pm.cpuhp_state, dev, errp);
         }
+    } else if (object_dynamic_cast(OBJECT(dev), TYPE_PCI_DEVICE)) {
+        acpi_pcihp_device_plug_cb(hotplug_dev, &lpc->pm.acpi_pci_hotplug,
+                                  dev, errp);
     } else {
         error_setg(errp, "acpi: device plug request for not supported device"
                    " type: %s", object_get_typename(OBJECT(dev)));
@@ -457,6 +495,10 @@ void ich9_pm_device_unplug_request_cb(HotplugHandler *hotplug_dev,
                !lpc->pm.cpu_hotplug_legacy) {
         acpi_cpu_unplug_request_cb(hotplug_dev, &lpc->pm.cpuhp_state,
                                    dev, errp);
+    } else if (object_dynamic_cast(OBJECT(dev), TYPE_PCI_DEVICE)) {
+        acpi_pcihp_device_unplug_request_cb(hotplug_dev,
+                                            &lpc->pm.acpi_pci_hotplug,
+                                            dev, errp);
     } else {
         error_setg(errp, "acpi: device unplug request for not supported device"
                    " type: %s", object_get_typename(OBJECT(dev)));
@@ -474,6 +516,9 @@ void ich9_pm_device_unplug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
     } 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 if (object_dynamic_cast(OBJECT(dev), TYPE_PCI_DEVICE)) {
+        acpi_pcihp_device_unplug_cb(hotplug_dev, &lpc->pm.acpi_pci_hotplug,
+                                    dev, errp);
     } else {
         error_setg(errp, "acpi: device unplug for not supported device"
                    " type: %s", object_get_typename(OBJECT(dev)));
hw/acpi/pcihp.c +2 -1
index 3d4ee3af72..d905d1b8f2 100644
@@ -110,7 +110,8 @@ static void acpi_set_pci_info(void)
     }
     bsel_is_set = true;
 
-    bus = find_i440fx(); /* TODO: Q35 support */
+    bus = find_host();
+
     if (bus) {
         /* Scan all PCI buses. Set property to enable acpi based hotplug. */
         pci_for_each_bus_depth_first(bus, acpi_set_bsel, NULL, &bsel_alloc);
hw/i386/pc.c +3 -1
index 143ac1c354..21c6eb779e 100644
@@ -96,7 +96,9 @@
 #include "fw_cfg.h"
 #include "trace.h"
 
-GlobalProperty pc_compat_5_0[] = {};
+GlobalProperty pc_compat_5_0[] = {
+    { "ICH9-LPC", "acpi-pci-hotplug-with-bridge-support", "off" },
+};
 const size_t pc_compat_5_0_len = G_N_ELEMENTS(pc_compat_5_0);
 
 GlobalProperty pc_compat_4_2[] = {
#18 lore raw

Re: [RFC PATCH 5/5] hw/acpi/ich9: Enable ACPI PCI hot-plug

Quoted 96 lines · Igor Mammedov
Quoted 23 lines · On Thu, 9 Jul 2020 00:46:15 +0200 Julia Suvorova <[email protected]> wrote:

Add acpi_pcihp to ich9_pm and use ACPI PCI hot-plug by default.

---
 include/hw/acpi/ich9.h |  3 +++
 hw/acpi/ich9.c         | 45 ++++++++++++++++++++++++++++++++++++++++++
 hw/acpi/pcihp.c        |  3 ++-
 hw/i386/pc.c           |  4 +++-
 4 files changed, 53 insertions(+), 2 deletions(-)
include/hw/acpi/ich9.h +1 -0
index 28a53181cb..d345da6b74 100644
@@ -24,6 +24,7 @@
 #include "hw/acpi/acpi.h"
 #include "hw/acpi/cpu_hotplug.h"
 #include "hw/acpi/cpu.h"
+#include "hw/acpi/pcihp.h"
 #include "hw/acpi/memory_hotplug.h"
 #include "hw/acpi/acpi_dev_interface.h"
 #include "hw/acpi/tco.h"

include/hw/acpi/ich9.h typedef struct ICH9LPCPMRegs { · line 57

Quoted 4 lines
include/hw/acpi/ich9.h +1 -0
@@ -53,6 +54,8 @@ typedef struct ICH9LPCPMRegs {
     AcpiCpuHotplug gpe_cpu;
     CPUHotplugState cpuhp_state;
 
+    bool use_acpi_pci_hotplug;

a bit outdated naming,
see 0affda043675c7619248a924a89bfd3781759f63, and rename it to match piix4

Quoted 56 lines
include/hw/acpi/ich9.h +1 -0
+    AcpiPciHpState acpi_pci_hotplug;
     MemHotplugState acpi_memory_hotplug;
 
     uint8_t disable_s3;
hw/acpi/ich9.c +26 -0
index 0fdd736da4..e0291373f2 100644
@@ -265,6 +265,7 @@ static void pm_reset(void *opaque)
     }
     pm->smi_en_wmask = ~0;
 
+    acpi_pcihp_reset(&pm->acpi_pci_hotplug);
     acpi_update_sci(&pm->acpi_regs, pm->irq);
 }
 
@@ -303,6 +304,17 @@ void ich9_pm_init(PCIDevice *lpc_pci, ICH9LPCPMRegs *pm,
     pm->enable_tco = true;
     acpi_pm_tco_init(&pm->tco_regs, &pm->io);
 
+    if (pm->use_acpi_pci_hotplug) {
+        acpi_pcihp_init(OBJECT(lpc_pci),
+                        &pm->acpi_pci_hotplug,
+                        pci_get_bus(lpc_pci),
+                        pci_address_space_io(lpc_pci),
+                        true);
+        qbus_set_hotplug_handler(BUS(pci_get_bus(lpc_pci)),
+                                 OBJECT(lpc_pci),
+                                 &error_abort);
+    }
+
     pm->irq = sci_irq;
     qemu_register_reset(pm_reset, pm);
     pm->powerdown_notifier.notify = pm_powerdown_req;
@@ -374,6 +386,20 @@ static void ich9_pm_set_enable_tco(Object *obj, bool value, Error **errp)
     s->pm.enable_tco = value;
 }
 
+static bool ich9_pm_get_acpi_pci_hotplug(Object *obj, Error **errp)
+{
+    ICH9LPCState *s = ICH9_LPC_DEVICE(obj);
+
+    return s->pm.use_acpi_pci_hotplug;
+}
+
+static void ich9_pm_set_acpi_pci_hotplug(Object *obj, bool value,
+                                               Error **errp)
+{
+    ICH9LPCState *s = ICH9_LPC_DEVICE(obj);
+
+    s->pm.use_acpi_pci_hotplug = value;
+}
 void ich9_pm_add_properties(Object *obj, ICH9LPCPMRegs *pm)
 {
     static const uint32_t gpe0_len = ICH9_PMIO_GPE0_LEN;

hw/acpi/ich9.c void ich9_pm_add_properties(Object *obj, ICH9LPCPMRegs *pm) · line 411

Quoted 4 lines
hw/acpi/ich9.c +1 -0
@@ -382,6 +408,7 @@ void ich9_pm_add_properties(Object *obj, ICH9LPCPMRegs *pm)
     pm->disable_s3 = 0;
     pm->disable_s4 = 0;
     pm->s4_val = 2;
+    pm->use_acpi_pci_hotplug = true;

I'm not sure about making crutch on by default though.

Quoted 7 lines

Does linux guests work fine with native hotplug or does it have the same issues as Windows?

Also you had an idea how to workaround native hotplug issues on QEMU side.
Can you dump here a quick summary why it didn't work out in the end?

To help out with that a bit:

I think native hotplug just does not match what management needs from QEMU. QEMU wants what ACPI calls a "VCR-style" ejection where eject happens automatically after software allows it.

Native hotplug is designed with a human in the loop.

Things like multi-second delay after an eject request, inability to differentiate between power down and eject, ability to power device back up after power down are all architectural in hative hotplug.

Interface is also designed such that there are subtle races making it hard not to loose hotplug events are also harmless on bare-metal since a human can just press the eject button again, but with causes no end of issues for us.

I agree it's a good idea to include the motivation here or in the commit log, though.

Quoted 86 lines
Quoted 86 lines
hw/acpi/ich9.c +18 -0
     object_property_add_uint32_ptr(obj, ACPI_PM_PROP_PM_IO_BASE,
                                    &pm->pm_io_base, OBJ_PROP_FLAG_READ);
@@ -405,6 +432,9 @@ void ich9_pm_add_properties(Object *obj, ICH9LPCPMRegs *pm)
     object_property_add_bool(obj, ACPI_PM_PROP_TCO_ENABLED,
                              ich9_pm_get_enable_tco,
                              ich9_pm_set_enable_tco);
+    object_property_add_bool(obj, "acpi-pci-hotplug-with-bridge-support",
+                             ich9_pm_get_acpi_pci_hotplug,
+                             ich9_pm_set_acpi_pci_hotplug);
 }
 
 void ich9_pm_device_pre_plug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
@@ -412,6 +442,11 @@ void ich9_pm_device_pre_plug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
 {
     ICH9LPCState *lpc = ICH9_LPC_DEVICE(hotplug_dev);
 
+    if (object_dynamic_cast(OBJECT(dev), TYPE_PCI_DEVICE)) {
+        acpi_pcihp_device_pre_plug_cb(hotplug_dev, dev, errp);
+        return;
+    }
+
     if (object_dynamic_cast(OBJECT(dev), TYPE_PC_DIMM) &&
         !lpc->pm.acpi_memory_hotplug.is_enabled)
         error_setg(errp,
@@ -437,6 +472,9 @@ void ich9_pm_device_plug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
         } else {
             acpi_cpu_plug_cb(hotplug_dev, &lpc->pm.cpuhp_state, dev, errp);
         }
+    } else if (object_dynamic_cast(OBJECT(dev), TYPE_PCI_DEVICE)) {
+        acpi_pcihp_device_plug_cb(hotplug_dev, &lpc->pm.acpi_pci_hotplug,
+                                  dev, errp);
     } else {
         error_setg(errp, "acpi: device plug request for not supported device"
                    " type: %s", object_get_typename(OBJECT(dev)));
@@ -457,6 +495,10 @@ void ich9_pm_device_unplug_request_cb(HotplugHandler *hotplug_dev,
                !lpc->pm.cpu_hotplug_legacy) {
         acpi_cpu_unplug_request_cb(hotplug_dev, &lpc->pm.cpuhp_state,
                                    dev, errp);
+    } else if (object_dynamic_cast(OBJECT(dev), TYPE_PCI_DEVICE)) {
+        acpi_pcihp_device_unplug_request_cb(hotplug_dev,
+                                            &lpc->pm.acpi_pci_hotplug,
+                                            dev, errp);
     } else {
         error_setg(errp, "acpi: device unplug request for not supported device"
                    " type: %s", object_get_typename(OBJECT(dev)));
@@ -474,6 +516,9 @@ void ich9_pm_device_unplug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
     } 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 if (object_dynamic_cast(OBJECT(dev), TYPE_PCI_DEVICE)) {
+        acpi_pcihp_device_unplug_cb(hotplug_dev, &lpc->pm.acpi_pci_hotplug,
+                                    dev, errp);
     } else {
         error_setg(errp, "acpi: device unplug for not supported device"
                    " type: %s", object_get_typename(OBJECT(dev)));
hw/acpi/pcihp.c +2 -1
index 3d4ee3af72..d905d1b8f2 100644
@@ -110,7 +110,8 @@ static void acpi_set_pci_info(void)
     }
     bsel_is_set = true;
 
-    bus = find_i440fx(); /* TODO: Q35 support */
+    bus = find_host();
+
     if (bus) {
         /* Scan all PCI buses. Set property to enable acpi based hotplug. */
         pci_for_each_bus_depth_first(bus, acpi_set_bsel, NULL, &bsel_alloc);
hw/i386/pc.c +3 -1
index 143ac1c354..21c6eb779e 100644
@@ -96,7 +96,9 @@
 #include "fw_cfg.h"
 #include "trace.h"
 
-GlobalProperty pc_compat_5_0[] = {};
+GlobalProperty pc_compat_5_0[] = {
+    { "ICH9-LPC", "acpi-pci-hotplug-with-bridge-support", "off" },
+};
 const size_t pc_compat_5_0_len = G_N_ELEMENTS(pc_compat_5_0);
 
 GlobalProperty pc_compat_4_2[] = {
#19 lore raw

Re: [RFC PATCH 5/5] hw/acpi/ich9: Enable ACPI PCI hot-plug

Quoted 118 lines · On Tue, 14 Jul 2020 04:54:02 -0400 "Michael S. Tsirkin" <[email protected]> wrote:
Quoted 96 lines · Igor Mammedov
Quoted 23 lines · On Thu, 9 Jul 2020 00:46:15 +0200 Julia Suvorova <[email protected]> wrote:

Add acpi_pcihp to ich9_pm and use ACPI PCI hot-plug by default.

---
 include/hw/acpi/ich9.h |  3 +++
 hw/acpi/ich9.c         | 45 ++++++++++++++++++++++++++++++++++++++++++
 hw/acpi/pcihp.c        |  3 ++-
 hw/i386/pc.c           |  4 +++-
 4 files changed, 53 insertions(+), 2 deletions(-)
include/hw/acpi/ich9.h +1 -0
index 28a53181cb..d345da6b74 100644
@@ -24,6 +24,7 @@
 #include "hw/acpi/acpi.h"
 #include "hw/acpi/cpu_hotplug.h"
 #include "hw/acpi/cpu.h"
+#include "hw/acpi/pcihp.h"
 #include "hw/acpi/memory_hotplug.h"
 #include "hw/acpi/acpi_dev_interface.h"
 #include "hw/acpi/tco.h"

include/hw/acpi/ich9.h typedef struct ICH9LPCPMRegs { · line 57

Quoted 4 lines
include/hw/acpi/ich9.h +1 -0
@@ -53,6 +54,8 @@ typedef struct ICH9LPCPMRegs {
     AcpiCpuHotplug gpe_cpu;
     CPUHotplugState cpuhp_state;
 
+    bool use_acpi_pci_hotplug;  

a bit outdated naming,
see 0affda043675c7619248a924a89bfd3781759f63, and rename it to match piix4

Quoted 56 lines
include/hw/acpi/ich9.h +1 -0
+    AcpiPciHpState acpi_pci_hotplug;
     MemHotplugState acpi_memory_hotplug;
 
     uint8_t disable_s3;
hw/acpi/ich9.c +26 -0
index 0fdd736da4..e0291373f2 100644
@@ -265,6 +265,7 @@ static void pm_reset(void *opaque)
     }
     pm->smi_en_wmask = ~0;
 
+    acpi_pcihp_reset(&pm->acpi_pci_hotplug);
     acpi_update_sci(&pm->acpi_regs, pm->irq);
 }
 
@@ -303,6 +304,17 @@ void ich9_pm_init(PCIDevice *lpc_pci, ICH9LPCPMRegs *pm,
     pm->enable_tco = true;
     acpi_pm_tco_init(&pm->tco_regs, &pm->io);
 
+    if (pm->use_acpi_pci_hotplug) {
+        acpi_pcihp_init(OBJECT(lpc_pci),
+                        &pm->acpi_pci_hotplug,
+                        pci_get_bus(lpc_pci),
+                        pci_address_space_io(lpc_pci),
+                        true);
+        qbus_set_hotplug_handler(BUS(pci_get_bus(lpc_pci)),
+                                 OBJECT(lpc_pci),
+                                 &error_abort);
+    }
+
     pm->irq = sci_irq;
     qemu_register_reset(pm_reset, pm);
     pm->powerdown_notifier.notify = pm_powerdown_req;
@@ -374,6 +386,20 @@ static void ich9_pm_set_enable_tco(Object *obj, bool value, Error **errp)
     s->pm.enable_tco = value;
 }
 
+static bool ich9_pm_get_acpi_pci_hotplug(Object *obj, Error **errp)
+{
+    ICH9LPCState *s = ICH9_LPC_DEVICE(obj);
+
+    return s->pm.use_acpi_pci_hotplug;
+}
+
+static void ich9_pm_set_acpi_pci_hotplug(Object *obj, bool value,
+                                               Error **errp)
+{
+    ICH9LPCState *s = ICH9_LPC_DEVICE(obj);
+
+    s->pm.use_acpi_pci_hotplug = value;
+}
 void ich9_pm_add_properties(Object *obj, ICH9LPCPMRegs *pm)
 {
     static const uint32_t gpe0_len = ICH9_PMIO_GPE0_LEN;

hw/acpi/ich9.c void ich9_pm_add_properties(Object *obj, ICH9LPCPMRegs *pm) · line 411

Quoted 4 lines
hw/acpi/ich9.c +1 -0
@@ -382,6 +408,7 @@ void ich9_pm_add_properties(Object *obj, ICH9LPCPMRegs *pm)
     pm->disable_s3 = 0;
     pm->disable_s4 = 0;
     pm->s4_val = 2;
+    pm->use_acpi_pci_hotplug = true;  

I'm not sure about making crutch on by default though.

Quoted 7 lines

Does linux guests work fine with native hotplug or does it have the same issues as Windows?

Also you had an idea how to workaround native hotplug issues on QEMU side.
Can you dump here a quick summary why it didn't work out in the end?

To help out with that a bit:

I think native hotplug just does not match what management needs from QEMU. QEMU wants what ACPI calls a "VCR-style" ejection where eject happens automatically after software allows it.

Native hotplug is designed with a human in the loop.

Things like multi-second delay after an eject request, inability to differentiate between power down and eject, ability to power device back up after power down are all architectural in hative hotplug.

Quoted 4 lines

Interface is also designed such that there are subtle races making it hard not to loose hotplug events are also harmless on bare-metal since a human can just press the eject button again, but with causes no end of issues for us.

that is a generic excuse for adding ACPI hotplug to q35,
but what I've asked for, is to document (for history reasons) alternative ideas that were explored to make native unplug work despite spec issues. That would be also a good justification for ACPI approach showing that we've exhausted any other approach before resorting to ACPI hotplug

Jilia had an idea about 'caching' unplug req and depending on cached value behave differently on power off event. Which looked to me as a thing that might solve unplug issue.

Quoted 90 lines

I agree it's a good idea to include the motivation here or in the commit log, though.

Quoted 86 lines
Quoted 86 lines
hw/acpi/ich9.c +18 -0
     object_property_add_uint32_ptr(obj, ACPI_PM_PROP_PM_IO_BASE,
                                    &pm->pm_io_base, OBJ_PROP_FLAG_READ);
@@ -405,6 +432,9 @@ void ich9_pm_add_properties(Object *obj, ICH9LPCPMRegs *pm)
     object_property_add_bool(obj, ACPI_PM_PROP_TCO_ENABLED,
                              ich9_pm_get_enable_tco,
                              ich9_pm_set_enable_tco);
+    object_property_add_bool(obj, "acpi-pci-hotplug-with-bridge-support",
+                             ich9_pm_get_acpi_pci_hotplug,
+                             ich9_pm_set_acpi_pci_hotplug);
 }
 
 void ich9_pm_device_pre_plug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
@@ -412,6 +442,11 @@ void ich9_pm_device_pre_plug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
 {
     ICH9LPCState *lpc = ICH9_LPC_DEVICE(hotplug_dev);
 
+    if (object_dynamic_cast(OBJECT(dev), TYPE_PCI_DEVICE)) {
+        acpi_pcihp_device_pre_plug_cb(hotplug_dev, dev, errp);
+        return;
+    }
+
     if (object_dynamic_cast(OBJECT(dev), TYPE_PC_DIMM) &&
         !lpc->pm.acpi_memory_hotplug.is_enabled)
         error_setg(errp,
@@ -437,6 +472,9 @@ void ich9_pm_device_plug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
         } else {
             acpi_cpu_plug_cb(hotplug_dev, &lpc->pm.cpuhp_state, dev, errp);
         }
+    } else if (object_dynamic_cast(OBJECT(dev), TYPE_PCI_DEVICE)) {
+        acpi_pcihp_device_plug_cb(hotplug_dev, &lpc->pm.acpi_pci_hotplug,
+                                  dev, errp);
     } else {
         error_setg(errp, "acpi: device plug request for not supported device"
                    " type: %s", object_get_typename(OBJECT(dev)));
@@ -457,6 +495,10 @@ void ich9_pm_device_unplug_request_cb(HotplugHandler *hotplug_dev,
                !lpc->pm.cpu_hotplug_legacy) {
         acpi_cpu_unplug_request_cb(hotplug_dev, &lpc->pm.cpuhp_state,
                                    dev, errp);
+    } else if (object_dynamic_cast(OBJECT(dev), TYPE_PCI_DEVICE)) {
+        acpi_pcihp_device_unplug_request_cb(hotplug_dev,
+                                            &lpc->pm.acpi_pci_hotplug,
+                                            dev, errp);
     } else {
         error_setg(errp, "acpi: device unplug request for not supported device"
                    " type: %s", object_get_typename(OBJECT(dev)));
@@ -474,6 +516,9 @@ void ich9_pm_device_unplug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
     } 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 if (object_dynamic_cast(OBJECT(dev), TYPE_PCI_DEVICE)) {
+        acpi_pcihp_device_unplug_cb(hotplug_dev, &lpc->pm.acpi_pci_hotplug,
+                                    dev, errp);
     } else {
         error_setg(errp, "acpi: device unplug for not supported device"
                    " type: %s", object_get_typename(OBJECT(dev)));
hw/acpi/pcihp.c +2 -1
index 3d4ee3af72..d905d1b8f2 100644
@@ -110,7 +110,8 @@ static void acpi_set_pci_info(void)
     }
     bsel_is_set = true;
 
-    bus = find_i440fx(); /* TODO: Q35 support */
+    bus = find_host();
+
     if (bus) {
         /* Scan all PCI buses. Set property to enable acpi based hotplug. */
         pci_for_each_bus_depth_first(bus, acpi_set_bsel, NULL, &bsel_alloc);
hw/i386/pc.c +3 -1
index 143ac1c354..21c6eb779e 100644
@@ -96,7 +96,9 @@
 #include "fw_cfg.h"
 #include "trace.h"
 
-GlobalProperty pc_compat_5_0[] = {};
+GlobalProperty pc_compat_5_0[] = {
+    { "ICH9-LPC", "acpi-pci-hotplug-with-bridge-support", "off" },
+};
 const size_t pc_compat_5_0_len = G_N_ELEMENTS(pc_compat_5_0);
 
 GlobalProperty pc_compat_4_2[] = {  
#20 lore raw
In reply to Julia Suvorova
Patchew URL: https://patchew.org/QEMU/[email protected]/

Patchew URL: https://patchew.org/QEMU/[email protected]/

Hi,

This series failed the docker-quick@centos7 build test. Please find the testing commands and their output below. If you have Docker installed, you can probably reproduce it locally.

=== TEST SCRIPT BEGIN ===
#!/bin/bash
make docker-image-centos7 V=1 NETWORK=1
time make docker-test-quick@centos7 SHOW_ENV=1 J=14 NETWORK=1 === TEST SCRIPT END ===

CC hw/acpi/tpm.o
CC hw/acpi/ipmi.o
/tmp/qemu-test/src/hw/acpi/ich9.c: In function 'ich9_pm_init':
/tmp/qemu-test/src/hw/acpi/ich9.c:315:34: error: too many arguments to function 'qbus_set_hotplug_handler'

                                  &error_abort);
                                  ^
In file included from /tmp/qemu-test/src/include/hw/isa/isa.h:8:0,
---
 void qbus_set_hotplug_handler(BusState *bus, Object *handler);
      ^

CC hw/acpi/acpi-stub.o
make: *** [hw/acpi/ich9.o] Error 1
make: *** Waiting for unfinished jobs....
CC hw/acpi/aml-build-stub.o
Traceback (most recent call last):
---
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['sudo', '-n', 'docker', 'run', '--label', 'com.qemu.instance.uuid=e8dad3af07f046e6acc07d5aab1854b9', '-u', '1001', '--security-opt', 'seccomp=unconfined', '--rm', '-e', 'TARGET_LIST=', '-e', 'EXTRA_CONFIGURE_OPTS=', '-e', 'V=', '-e', 'J=14', '-e', 'DEBUG=', '-e', 'SHOW_ENV=1', '-e', 'CCACHE_DIR=/var/tmp/ccache', '-v', '/home/patchew/.cache/qemu-docker-ccache:/var/tmp/ccache:z', '-v', '/var/tmp/patchew-tester-tmp-_j3k8sfi/src/docker-src.2020-07-08-19.27.02.11252:/var/tmp/qemu:z,ro', 'qemu:centos7', '/var/tmp/qemu/run', 'test-quick']' returned non-zero exit status 2. filter=--filter=label=com.qemu.instance.uuid=e8dad3af07f046e6acc07d5aab1854b9
make[1]: *** [docker-run] Error 1
make[1]: Leaving directory `/var/tmp/patchew-tester-tmp-_j3k8sfi/src'
make: *** [docker-run-test-quick@centos7] Error 2

real 2m25.350s
user 0m8.686s

The full log is available at
http://patchew.org/logs/[email protected]/testing.docker-quick@centos7/?type=message.
---
Email generated automatically by Patchew [https://patchew.org/].
Please send your feedback to [email protected]

#21 lore raw
In reply to Julia Suvorova
Patchew URL: https://patchew.org/QEMU/[email protected]/

Patchew URL: https://patchew.org/QEMU/[email protected]/

Hi,

This series failed the docker-mingw@fedora build test. Please find the testing commands and their output below. If you have Docker installed, you can probably reproduce it locally.

=== TEST SCRIPT BEGIN ===
#! /bin/bash
export ARCH=x86_64
make docker-image-fedora V=1 NETWORK=1
time make docker-test-mingw@fedora J=14 NETWORK=1
=== TEST SCRIPT END ===

CC hw/acpi/acpi_interface.o
CC hw/acpi/bios-linker-loader.o
/tmp/qemu-test/src/hw/acpi/ich9.c: In function 'ich9_pm_init':
/tmp/qemu-test/src/hw/acpi/ich9.c:313:9: error: too many arguments to function 'qbus_set_hotplug_handler'

  313 |         qbus_set_hotplug_handler(BUS(pci_get_bus(lpc_pci)),
      |         ^~~~~~~~~~~~~~~~~~~~~~~~
In file included from /tmp/qemu-test/src/include/hw/isa/isa.h:8,
---
/tmp/qemu-test/src/include/hw/qdev-core.h:538:6: note: declared here
  538 | void qbus_set_hotplug_handler(BusState *bus, Object *handler);
      |      ^~~~~~~~~~~~~~~~~~~~~~~~

make: *** [/tmp/qemu-test/src/rules.mak:69: hw/acpi/ich9.o] Error 1
make: *** Waiting for unfinished jobs....
CC hw/acpi/aml-build.o
Traceback (most recent call last):
---
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['sudo', '-n', 'docker', 'run', '--label', 'com.qemu.instance.uuid=f40198b991eb4bf4a15cccbf81eccbf5', '-u', '1003', '--security-opt', 'seccomp=unconfined', '--rm', '-e', 'TARGET_LIST=', '-e', 'EXTRA_CONFIGURE_OPTS=', '-e', 'V=', '-e', 'J=14', '-e', 'DEBUG=', '-e', 'SHOW_ENV=', '-e', 'CCACHE_DIR=/var/tmp/ccache', '-v', '/home/patchew2/.cache/qemu-docker-ccache:/var/tmp/ccache:z', '-v', '/var/tmp/patchew-tester-tmp-tvkth6at/src/docker-src.2020-07-08-19.30.00.18172:/var/tmp/qemu:z,ro', 'qemu:fedora', '/var/tmp/qemu/run', 'test-mingw']' returned non-zero exit status 2. filter=--filter=label=com.qemu.instance.uuid=f40198b991eb4bf4a15cccbf81eccbf5
make[1]: *** [docker-run] Error 1
make[1]: Leaving directory `/var/tmp/patchew-tester-tmp-tvkth6at/src'
make: *** [docker-run-test-mingw@fedora] Error 2

real 3m32.203s
user 0m8.424s

The full log is available at
http://patchew.org/logs/[email protected]/testing.docker-mingw@fedora/?type=message.
---
Email generated automatically by Patchew [https://patchew.org/].
Please send your feedback to [email protected]