From: Don Slutz <dslutz@verizon.com>
To: Paul Durrant <Paul.Durrant@citrix.com>,
"Slutz, Donald Christopher" <dslutz@verizon.com>,
"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>,
"xen-devel@lists.xen.org" <xen-devel@lists.xen.org>
Cc: Stefano Stabellini <Stefano.Stabellini@citrix.com>,
Don Slutz <don.slutz@gmail.com>,
"Michael S. Tsirkin" <mst@redhat.com>
Subject: Re: [Qemu-devel] [PATCH v2 2/4] Extend device listener interface for PCI to PCI bridges
Date: Tue, 09 Jun 2015 09:53:01 -0400 [thread overview]
Message-ID: <5576EFBD.1070706@Verizon.com> (raw)
In-Reply-To: <9AAE0902D5BC7E449B7C8E4E778ABCD025941186@AMSPEX01CL01.citrite.net>
On 06/09/15 05:08, Paul Durrant wrote:
>> -----Original Message-----
>> From: Don Slutz [mailto:dslutz@verizon.com]
>> Sent: 08 June 2015 22:19
>> To: qemu-devel@nongnu.org; xen-devel@lists.xen.org
>> Cc: Michael S. Tsirkin; Paul Durrant; Stefano Stabellini; Don Slutz; Don Slutz
>> Subject: [PATCH v2 2/4] Extend device listener interface for PCI to PCI
>> bridges
>>
>> The commit 707ff80021ccd7a68f4b3d2c44eebf87efbb41c4 assumed that a
>> PCI device has a static address. This is not true for PCI devices
>> that are on the secondary bus of a PCI to PCI bridge.
>>
>> BIOS and/or guest OS can change the secondary bus number to a new
>> value at any time.
>>
>> Extend the device listener interface to be called when ever the
>> secondary bus number is set to a new value. This new interface
>> is called for all PCI devices that are on the secondary bridge.
>>
>
> Can't you just make unrealize and realize calls back-to-back when the bus number changes, rather than having a new callback?
>
That is what V1 had. Michael S. Tsirkin did not like this.
-Don Slutz
> Paul
>
>> Signed-off-by: Don Slutz <dslutz@verizon.com>
>> CC: Don Slutz <don.slutz@gmail.com>
>> ---
>> hw/core/qdev.c | 7 +++++++
>> hw/pci/pci_bridge.c | 18 ++++++++++++++++++
>> include/hw/qdev-core.h | 3 +++
>> 3 files changed, 28 insertions(+)
>>
>> diff --git a/hw/core/qdev.c b/hw/core/qdev.c
>> index b0f0f84..7728ee7 100644
>> --- a/hw/core/qdev.c
>> +++ b/hw/core/qdev.c
>> @@ -239,6 +239,13 @@ void device_listener_unregister(DeviceListener
>> *listener)
>> QTAILQ_REMOVE(&device_listeners, listener, link);
>> }
>>
>> +void device_listener_change_pci_bus_num(PCIBus *b, PCIDevice *d, void
>> *opaque)
>> +{
>> + uint8_t oldbus = GPOINTER_TO_INT(opaque);
>> +
>> + DEVICE_LISTENER_CALL(change_pci_bus_num, Forward, d, oldbus);
>> +}
>> +
>> static void device_realize(DeviceState *dev, Error **errp)
>> {
>> DeviceClass *dc = DEVICE_GET_CLASS(dev);
>> diff --git a/hw/pci/pci_bridge.c b/hw/pci/pci_bridge.c
>> index 40c97b1..742c4d8 100644
>> --- a/hw/pci/pci_bridge.c
>> +++ b/hw/pci/pci_bridge.c
>> @@ -248,6 +248,7 @@ void pci_bridge_write_config(PCIDevice *d,
>> PCIBridge *s = PCI_BRIDGE(d);
>> uint16_t oldctl = pci_get_word(d->config + PCI_BRIDGE_CONTROL);
>> uint16_t newctl;
>> + uint8_t oldbus = pci_get_byte(d->config + PCI_SECONDARY_BUS);
>>
>> pci_default_write_config(d, address, val, len);
>>
>> @@ -265,6 +266,14 @@ void pci_bridge_write_config(PCIDevice *d,
>> pci_bridge_update_mappings(s);
>> }
>>
>> + if (oldbus != pci_get_byte(d->config + PCI_SECONDARY_BUS)) {
>> + PCIBus *sec_bus = pci_bridge_get_sec_bus(s);
>> +
>> + pci_for_each_device(sec_bus, pci_bus_num(sec_bus),
>> + device_listener_change_pci_bus_num,
>> + GINT_TO_POINTER(oldbus));
>> + }
>> +
>> newctl = pci_get_word(d->config + PCI_BRIDGE_CONTROL);
>> if (~oldctl & newctl & PCI_BRIDGE_CTL_BUS_RESET) {
>> /* Trigger hot reset on 0->1 transition. */
>> @@ -297,6 +306,7 @@ void pci_bridge_reset(DeviceState *qdev)
>> {
>> PCIDevice *dev = PCI_DEVICE(qdev);
>> uint8_t *conf = dev->config;
>> + uint8_t oldbus = conf[PCI_SECONDARY_BUS];
>>
>> conf[PCI_PRIMARY_BUS] = 0;
>> conf[PCI_SECONDARY_BUS] = 0;
>> @@ -329,6 +339,14 @@ void pci_bridge_reset(DeviceState *qdev)
>> pci_set_long(conf + PCI_PREF_LIMIT_UPPER32, 0);
>>
>> pci_set_word(conf + PCI_BRIDGE_CONTROL, 0);
>> +
>> + if (oldbus) {
>> + PCIBus *sec_bus = pci_bridge_get_sec_bus(PCI_BRIDGE(dev));
>> +
>> + pci_for_each_device(sec_bus, pci_bus_num(sec_bus),
>> + device_listener_change_pci_bus_num,
>> + GINT_TO_POINTER(oldbus));
>> + }
>> }
>>
>> /* default qdev initialization function for PCI-to-PCI bridge */
>> diff --git a/include/hw/qdev-core.h b/include/hw/qdev-core.h
>> index d4be92f..154b4c1 100644
>> --- a/include/hw/qdev-core.h
>> +++ b/include/hw/qdev-core.h
>> @@ -168,6 +168,8 @@ struct DeviceState {
>> struct DeviceListener {
>> void (*realize)(DeviceListener *listener, DeviceState *dev);
>> void (*unrealize)(DeviceListener *listener, DeviceState *dev);
>> + void (*change_pci_bus_num)(DeviceListener *listener, PCIDevice *d,
>> + uint8_t oldbus);
>> QTAILQ_ENTRY(DeviceListener) link;
>> };
>>
>> @@ -387,5 +389,6 @@ static inline bool qbus_is_hotpluggable(BusState
>> *bus)
>>
>> void device_listener_register(DeviceListener *listener);
>> void device_listener_unregister(DeviceListener *listener);
>> +void device_listener_change_pci_bus_num(PCIBus *b, PCIDevice *d, void
>> *opaque);
>>
>> #endif
>> --
>> 1.8.4
>
next prev parent reply other threads:[~2015-06-09 13:53 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-06-08 21:18 [Qemu-devel] [PATCH v2 0/4] Fix device listener interface for PCI to PCI bridges Don Slutz
2015-06-08 21:18 ` [Qemu-devel] [BUGFIX][PATCH v2 1/4] exec: Do not use MemoryRegion after free Don Slutz
2015-06-08 21:18 ` [Qemu-devel] [PATCH v2 2/4] Extend device listener interface for PCI to PCI bridges Don Slutz
2015-06-09 9:08 ` Paul Durrant
2015-06-09 13:53 ` Don Slutz [this message]
2015-06-09 13:55 ` Paul Durrant
2015-06-09 14:00 ` Don Slutz
2015-06-08 21:18 ` [Qemu-devel] [PATCH v2 3/4] xen: Add usage of " Don Slutz
2015-06-08 21:18 ` [Qemu-devel] [PATCH v2 4/4] xen: Fix map/unmap of pcidev to ioreq server Don Slutz
2015-06-09 9:05 ` Paul Durrant
2015-06-09 13:51 ` Don Slutz
2015-06-09 14:03 ` Paul Durrant
2015-06-09 14:28 ` Don Slutz
2015-06-09 15:11 ` Paul Durrant
2015-06-09 16:40 ` Don Slutz
2015-06-09 9:13 ` [Qemu-devel] [PATCH v2 0/4] Fix device listener interface for PCI to PCI bridges Michael S. Tsirkin
2015-06-09 9:18 ` Paul Durrant
2015-06-09 10:52 ` Michael S. Tsirkin
2015-06-09 10:58 ` Paul Durrant
2015-06-09 12:29 ` Michael S. Tsirkin
2015-06-09 14:14 ` Paul Durrant
2015-06-09 14:27 ` Michael S. Tsirkin
2015-06-09 15:10 ` Don Slutz
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=5576EFBD.1070706@Verizon.com \
--to=dslutz@verizon.com \
--cc=Paul.Durrant@citrix.com \
--cc=Stefano.Stabellini@citrix.com \
--cc=don.slutz@gmail.com \
--cc=mst@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=xen-devel@lists.xen.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).