qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Igor Mammedov <imammedo@redhat.com>
To: qemu-devel@nongnu.org
Cc: mst@redhat.com, ani@anisinha.ca, berrange@redhat.com
Subject: [PATCH 11/33] pci: fix 'hotplugglable' property behavior
Date: Fri, 24 Feb 2023 16:37:50 +0100	[thread overview]
Message-ID: <20230224153812.4176226-12-imammedo@redhat.com> (raw)
In-Reply-To: <20230224153812.4176226-1-imammedo@redhat.com>

Currently the property may flip its state
during VM bring up or just doesn't work as
the name implies.

In particular with PCIE root port that has
'hotplug={on|off}' property, and when it's
turned off, one would expect
  'hotpluggable' == false
for any devices attached to it.
Which is not the case since qbus_is_hotpluggable()
used by the property just checks for presence
of any hotplug_handler set on bus.

The problem is that name BusState::hotplug_handler
from its inception is misnomer, as it handles
not only hotplug but also in many cases coldplug
as well (i.e. generic wiring interface), and
it's fine to have hotplug_handler set on bus
while it doesn't support hotplug (ex. pcie-slot
with hotplug=off).

Another case of root port flipping 'hotpluggable'
state when ACPI PCI hotplug is enabled in this
case root port with 'hotplug=off' starts as
hotpluggable and then later on, pcihp
hotplug_handler clears hotplug_handler
explicitly after checking root port's 'hotplug'
property.

So root-port hotpluggablity check sort of works
if pcihp is enabled but is broken if pcihp is
disabled.

One way to deal with the issue is to ask
hotplug_handler if bus it controls is hotpluggable
or not. To do that add is_hotpluggable_bus()
hook to HotplugHandler interface and use it in
'hotpluggable' property + teach pcie-slot to
actually look into 'hotplug' property state
before deciding if bus is hotpluggable.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
---
 include/hw/hotplug.h   |  2 ++
 include/hw/qdev-core.h | 13 ++++++++++++-
 hw/pci/pcie_port.c     |  8 ++++++++
 3 files changed, 22 insertions(+), 1 deletion(-)

diff --git a/include/hw/hotplug.h b/include/hw/hotplug.h
index e15f59c8b3..a9840ed485 100644
--- a/include/hw/hotplug.h
+++ b/include/hw/hotplug.h
@@ -48,6 +48,7 @@ typedef void (*hotplug_fn)(HotplugHandler *plug_handler,
  * @unplug: unplug callback.
  *          Used for device removal with devices that implement
  *          asynchronous and synchronous (surprise) removal.
+ * @is_hotpluggable_bus: called to check if bus/its parent allow hotplug on bus
  */
 struct HotplugHandlerClass {
     /* <private> */
@@ -58,6 +59,7 @@ struct HotplugHandlerClass {
     hotplug_fn plug;
     hotplug_fn unplug_request;
     hotplug_fn unplug;
+    bool (*is_hotpluggable_bus)(HotplugHandler *plug_handler, BusState *bus);
 };
 
 /**
diff --git a/include/hw/qdev-core.h b/include/hw/qdev-core.h
index 35fddb19a6..3b3518146b 100644
--- a/include/hw/qdev-core.h
+++ b/include/hw/qdev-core.h
@@ -812,7 +812,18 @@ void qbus_set_bus_hotplug_handler(BusState *bus);
 
 static inline bool qbus_is_hotpluggable(BusState *bus)
 {
-   return bus->hotplug_handler;
+    HotplugHandler *plug_handler = bus->hotplug_handler;
+    bool ret = !!plug_handler;
+
+    if (plug_handler) {
+        HotplugHandlerClass *hdc;
+
+        hdc = HOTPLUG_HANDLER_GET_CLASS(plug_handler);
+        if (hdc->is_hotpluggable_bus) {
+            ret = hdc->is_hotpluggable_bus(plug_handler, bus);
+        }
+    }
+    return ret;
 }
 
 /**
diff --git a/hw/pci/pcie_port.c b/hw/pci/pcie_port.c
index 65a397ad23..000633fec1 100644
--- a/hw/pci/pcie_port.c
+++ b/hw/pci/pcie_port.c
@@ -161,6 +161,13 @@ PCIDevice *pcie_find_port_by_pn(PCIBus *bus, uint8_t pn)
     return NULL;
 }
 
+static bool pcie_slot_is_hotpluggbale_bus(HotplugHandler *plug_handler,
+                                          BusState *bus)
+{
+    PCIESlot *s = PCIE_SLOT(bus->parent);
+    return s->hotplug;
+}
+
 static const TypeInfo pcie_port_type_info = {
     .name = TYPE_PCIE_PORT,
     .parent = TYPE_PCI_BRIDGE,
@@ -188,6 +195,7 @@ static void pcie_slot_class_init(ObjectClass *oc, void *data)
     hc->plug = pcie_cap_slot_plug_cb;
     hc->unplug = pcie_cap_slot_unplug_cb;
     hc->unplug_request = pcie_cap_slot_unplug_request_cb;
+    hc->is_hotpluggable_bus = pcie_slot_is_hotpluggbale_bus;
 }
 
 static const TypeInfo pcie_slot_type_info = {
-- 
2.39.1



  parent reply	other threads:[~2023-02-24 15:40 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-02-24 15:37 [PATCH 00/33] pci(pc/q35): acpi-index support on non-hotpluggable slots Igor Mammedov
2023-02-24 15:37 ` [PATCH 01/33] tests: acpi: whitelist new q35.noacpihp test and pc.hpbrroot Igor Mammedov
2023-02-24 15:37 ` [PATCH 02/33] tests: acpi: add test_acpi_q35_tcg_no_acpi_hotplug test and extend test_acpi_piix4_no_acpi_pci_hotplug Igor Mammedov
2023-02-24 15:37 ` [PATCH 03/33] tests: acpi: update expected blobs Igor Mammedov
2023-02-24 15:37 ` [PATCH 04/33] tests: acpi: whitelist q35/DSDT.multi-bridge before extending testcase Igor Mammedov
2023-02-24 15:37 ` [PATCH 05/33] tests: acpi: extend multi-bridge case with case 'root-port, id=HOHP, hotplug=off root-port, bus=NOHP' Igor Mammedov
2023-02-24 15:37 ` [PATCH 06/33] x86: pcihp: fix missing PCNT callchain when intermediate root-port has 'hotplug=off' set Igor Mammedov
2023-02-24 15:37 ` [PATCH 07/33] tests: acpi: whitelist pc/DSDT.hpbrroot and pc/DSDT.hpbridge tests Igor Mammedov
2023-02-24 15:37 ` [PATCH 08/33] x86: pcihp: fix missing bridge AML when intermediate root-port has 'hotplug=off' set Igor Mammedov
2023-02-24 15:37 ` [PATCH 09/33] tests: acpi: update expected blobs Igor Mammedov
2023-02-24 15:37 ` [PATCH 10/33] pcihp: piix4: do not redirect hotplug controller to piix4 when ACPI hotplug is disabled Igor Mammedov
2023-02-24 15:37 ` Igor Mammedov [this message]
2023-02-24 15:37 ` [PATCH 12/33] tests: acpi: whitelist DSDT blobs before isolating PCI _DSM func 0 prolog Igor Mammedov
2023-02-24 15:37 ` [PATCH 13/33] pcihp: move PCI _DSM function 0 prolog into separate function Igor Mammedov
2023-02-24 15:37 ` [PATCH 14/33] tests: acpi: update expected blobs Igor Mammedov
2023-03-02  0:27   ` Michael S. Tsirkin
2023-03-02 12:02     ` Igor Mammedov
2023-03-02 12:55       ` Fabiano Rosas
2023-03-02 14:57         ` Igor Mammedov
2023-03-02 15:43           ` Fabiano Rosas
2023-03-02 15:59             ` Igor Mammedov
2023-03-02  0:37   ` Michael S. Tsirkin
2023-02-24 15:37 ` [PATCH 15/33] tests: acpi: whitelist DSDT before adding EDSM method Igor Mammedov
2023-02-24 15:37 ` [PATCH 16/33] acpi: pci: add EDSM method to DSDT Igor Mammedov
2023-02-24 15:37 ` [PATCH 17/33] tests: acpi: update expected blobs Igor Mammedov
2023-02-24 15:37 ` [PATCH 18/33] tests: acpi: whitelist DSDT before adding device with acpi-index to testcases Igor Mammedov
2023-02-24 15:37 ` [PATCH 19/33] tests: acpi: add device with acpi-index on non-hotpluggble bus Igor Mammedov
2023-02-24 15:37 ` [PATCH 20/33] acpi: pci: support acpi-index for non-hotpluggable devices Igor Mammedov
2023-02-24 15:38 ` [PATCH 21/33] tests: acpi: update expected blobs Igor Mammedov
2023-02-24 15:38 ` [PATCH 22/33] tests: acpi: whitelist DSDT before exposing non zero functions Igor Mammedov
2023-02-24 15:38 ` [PATCH 23/33] acpi: pci: describe all functions on populated slots Igor Mammedov
2023-02-24 15:38 ` [PATCH 24/33] tests: acpi: update expected blobs Igor Mammedov
2023-02-24 15:38 ` [PATCH 25/33] tests: acpi: whitelist DSDT before adding non-0 function device with acpi-index to testcases Igor Mammedov
2023-02-24 15:38 ` [PATCH 26/33] tests: acpi: add non zero function device with acpi-index on non-hotpluggble bus Igor Mammedov
2023-02-24 15:38 ` [PATCH 27/33] tests: acpi: update expected blobs Igor Mammedov
2023-02-24 15:38 ` [PATCH 28/33] pci: move acpi-index uniqueness check to generic PCI device code Igor Mammedov
2023-02-24 15:38 ` [PATCH 29/33] acpi: pci: drop BSEL usage when deciding that device isn't hotpluggable Igor Mammedov
2023-02-24 15:38 ` [PATCH 30/33] acpi: pci: move BSEL into build_append_pcihp_slots() Igor Mammedov
2023-02-24 15:38 ` [PATCH 31/33] acpi: pci: move out ACPI PCI hotplug generator from generic slot generator build_append_pci_bus_devices() Igor Mammedov
2023-02-24 15:38 ` [PATCH 32/33] pcihp: move fields enabling hotplug into AcpiPciHpState Igor Mammedov
2023-02-24 15:38 ` [PATCH 33/33] pcihp: add ACPI PCI hotplug specific is_hotpluggable_bus() callback Igor Mammedov
2023-03-02 10:59 ` [PATCH 00/33] pci(pc/q35): acpi-index support on non-hotpluggable slots Michael S. Tsirkin
2023-03-02 12:05   ` Jonathan Cameron via
2023-03-02 12:07   ` Igor Mammedov
2023-03-02 23:21     ` Michael S. Tsirkin
2023-03-06 10:49       ` 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=20230224153812.4176226-12-imammedo@redhat.com \
    --to=imammedo@redhat.com \
    --cc=ani@anisinha.ca \
    --cc=berrange@redhat.com \
    --cc=mst@redhat.com \
    --cc=qemu-devel@nongnu.org \
    /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).