From: "Michael S. Tsirkin" <mst@redhat.com>
To: Jan Kiszka <jan.kiszka@siemens.com>
Cc: Alex Williamson <alex.williamson@redhat.com>,
qemu-devel <qemu-devel@nongnu.org>
Subject: Re: [Qemu-devel] [PATCH 03/13] pci: Introduce cached device INTx routing
Date: Sun, 10 Jun 2012 14:28:01 +0300 [thread overview]
Message-ID: <20120610112801.GL6250@redhat.com> (raw)
In-Reply-To: <aa4ff6ec67c299a0a975c61d7b4698d32ba60877.1338799935.git.jan.kiszka@siemens.com>
On Mon, Jun 04, 2012 at 10:52:11AM +0200, Jan Kiszka wrote:
> Based on the original version by Michael Tsirkin: Instead of traversing
> the PCI bus hierarchy from a device to the host bridge, cache this path
> in PCIDevice and use it directly on interrupt delivery. This will
> specifically pay off with more complex topologies than the current
> single host bus.
>
> Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
OK, I applied this. I still think we should go the extra
mile and cache the final IRQ not just the root intx.
But that can be a patch on top.
> ---
> hw/pci.c | 53 ++++++++++++++++++++++++++++++++++++++++++++---------
> hw/pci.h | 4 ++++
> 2 files changed, 48 insertions(+), 9 deletions(-)
>
> diff --git a/hw/pci.c b/hw/pci.c
> index 33452ab..771fb39 100644
> --- a/hw/pci.c
> +++ b/hw/pci.c
> @@ -68,6 +68,8 @@ static void pci_update_mappings(PCIDevice *d);
> static void pci_set_irq(void *opaque, int irq_num, int level);
> static int pci_add_option_rom(PCIDevice *pdev, bool is_default_rom);
> static void pci_del_option_rom(PCIDevice *pdev);
> +static void pci_for_each_device_under_bus(PCIBus *bus,
> + void (*fn)(PCIBus *b, PCIDevice *d));
>
> static uint16_t pci_default_sub_vendor_id = PCI_SUBVENDOR_ID_REDHAT_QUMRANET;
> static uint16_t pci_default_sub_device_id = PCI_SUBDEVICE_ID_QEMU;
> @@ -112,18 +114,49 @@ static inline void pci_set_irq_state(PCIDevice *d, int irq_num, int level)
> d->irq_state |= level << irq_num;
> }
>
> -static void pci_change_irq_level(PCIDevice *pci_dev, int irq_num, int change)
> +static void pci_set_device_intx_routing(PCIBus *bus, PCIDevice *dev)
> {
> - PCIBus *bus;
> + int pin, output_pin;
> + PCIDevice *pci_dev;
> +
> + /* We might be too early, i.e. before pci_bus_irqs was called.
> + * We will be called again when this happened. */
> + if (!bus->map_irq) {
> + return;
> + }
> +
> + for (pin = 0; pin < PCI_NUM_PINS; pin++) {
> + pci_dev = dev;
> + output_pin = pin;
> + do {
> + bus = pci_dev->bus;
> + output_pin = bus->map_irq(pci_dev, output_pin);
> + pci_dev = bus->parent_dev;
> + } while (pci_dev);
> +
> + dev->host_intx_pin[pin] = output_pin;
> + dev->host_bus = bus;
> + }
> +}
> +
> +static void pci_set_bus_intx_routing(PCIBus *bus)
> +{
> + PCIBus *sec;
> +
> + pci_for_each_device_under_bus(bus, pci_set_device_intx_routing);
>
> - do {
> - bus = pci_dev->bus;
> - irq_num = bus->map_irq(pci_dev, irq_num);
> - pci_dev = bus->parent_dev;
> - } while (pci_dev);
> + QLIST_FOREACH(sec, &bus->child, sibling) {
> + pci_set_bus_intx_routing(sec);
> + }
> +}
> +
> +static void pci_change_irq_level(PCIDevice *dev, int pin, int change)
> +{
> + PCIBus *bus = dev->host_bus;
> + int output_pin = dev->host_intx_pin[pin];
>
> - bus->irq_count[irq_num] += change;
> - bus->set_irq(bus->irq_opaque, irq_num, bus->irq_count[irq_num] != 0);
> + bus->irq_count[output_pin] += change;
> + bus->set_irq(bus->irq_opaque, output_pin, bus->irq_count[output_pin] != 0);
> }
>
> int pci_bus_get_irq_level(PCIBus *bus, int irq_num)
> @@ -293,6 +326,7 @@ void pci_bus_irqs(PCIBus *bus, pci_set_irq_fn set_irq, pci_map_irq_fn map_irq,
> bus->irq_opaque = irq_opaque;
> bus->nirq = nirq;
> bus->irq_count = g_malloc0(nirq * sizeof(bus->irq_count[0]));
> + pci_set_bus_intx_routing(bus);
> }
>
> void pci_bus_hotplug(PCIBus *bus, pci_hotplug_fn hotplug, DeviceState *qdev)
> @@ -798,6 +832,7 @@ static PCIDevice *do_pci_register_device(PCIDevice *pci_dev, PCIBus *bus,
> bus->devices[devfn] = pci_dev;
> pci_dev->irq = qemu_allocate_irqs(pci_set_irq, pci_dev, PCI_NUM_PINS);
> pci_dev->version_id = 2; /* Current pci device vmstate version */
> + pci_set_device_intx_routing(bus, pci_dev);
> return pci_dev;
> }
>
> diff --git a/hw/pci.h b/hw/pci.h
> index 7eaf90b..c4fd863 100644
> --- a/hw/pci.h
> +++ b/hw/pci.h
> @@ -211,6 +211,10 @@ struct PCIDevice {
> /* Current IRQ levels. Used internally by the generic PCI code. */
> uint8_t irq_state;
>
> + /* Used internally by PCI code to cache the interrupt routing */
> + PCIBus *host_bus;
> + int host_intx_pin[PCI_NUM_PINS];
> +
A better name would be root_XXXX: host is an overloaded term.
I applied such a patch.
> /* Capability bits */
> uint32_t cap_present;
>
> --
> 1.7.3.4
next prev parent reply other threads:[~2012-06-10 11:27 UTC|newest]
Thread overview: 88+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-04 8:52 [Qemu-devel] [PATCH 00/13] pci: Cleanups & preparations for KVM device assignment Jan Kiszka
2012-06-04 8:52 ` [Qemu-devel] [PATCH 01/13] pci: Refactor pci_change_irq_level Jan Kiszka
2012-06-10 11:28 ` Michael S. Tsirkin
2012-06-04 8:52 ` [Qemu-devel] [PATCH 02/13] pci: Fold pci_bus_new_inplace into pci_bus_new Jan Kiszka
2012-06-07 12:51 ` Andreas Färber
2012-06-07 15:07 ` Jan Kiszka
2012-06-04 8:52 ` [Qemu-devel] [PATCH 03/13] pci: Introduce cached device INTx routing Jan Kiszka
2012-06-10 11:28 ` Michael S. Tsirkin [this message]
2012-06-04 8:52 ` [Qemu-devel] [PATCH 04/13] pci: Rename map_irq to route_pin Jan Kiszka
2012-06-10 11:28 ` Michael S. Tsirkin
2012-06-10 13:23 ` Michael S. Tsirkin
2012-06-04 8:52 ` [Qemu-devel] [PATCH 05/13] pci: Add pci_device_route_intx_to_irq Jan Kiszka
2012-06-07 14:32 ` Michael S. Tsirkin
2012-06-07 15:10 ` Jan Kiszka
2012-06-07 16:28 ` Michael S. Tsirkin
2012-06-07 16:46 ` Jan Kiszka
2012-06-07 16:55 ` Michael S. Tsirkin
2012-06-10 9:55 ` Michael S. Tsirkin
2012-06-10 10:08 ` Jan Kiszka
2012-06-10 10:41 ` Michael S. Tsirkin
2012-06-10 10:49 ` Jan Kiszka
2012-06-10 10:53 ` Michael S. Tsirkin
2012-06-10 14:19 ` Alex Williamson
2012-06-10 14:43 ` Michael S. Tsirkin
2012-06-10 15:25 ` Alex Williamson
2012-06-10 15:55 ` Michael S. Tsirkin
2012-06-10 16:30 ` Jan Kiszka
2012-06-10 16:50 ` Michael S. Tsirkin
2012-06-10 17:04 ` Michael S. Tsirkin
2012-06-04 8:52 ` [Qemu-devel] [PATCH 06/13] pci: Add INTx routing notifier Jan Kiszka
2012-06-07 13:14 ` Michael S. Tsirkin
2012-06-07 15:13 ` Jan Kiszka
2012-06-08 12:47 ` [Qemu-devel] [PATCH v2 " Jan Kiszka
2012-06-10 9:48 ` [Qemu-devel] [PATCH " Michael S. Tsirkin
2012-06-10 10:05 ` Jan Kiszka
2012-06-10 10:33 ` Michael S. Tsirkin
2012-06-10 10:44 ` Jan Kiszka
2012-06-10 11:11 ` Michael S. Tsirkin
2012-06-10 11:18 ` Jan Kiszka
2012-06-10 11:39 ` Michael S. Tsirkin
2012-06-10 12:09 ` Jan Kiszka
2012-06-10 12:16 ` Michael S. Tsirkin
2012-06-10 12:33 ` Jan Kiszka
2012-06-10 12:42 ` Michael S. Tsirkin
2012-06-10 12:47 ` Jan Kiszka
2012-06-10 13:19 ` Michael S. Tsirkin
2012-06-10 12:32 ` Michael S. Tsirkin
2012-06-04 8:52 ` [Qemu-devel] [PATCH 07/13] pci: Make domain and bus unsigned in pci_read_devaddr Jan Kiszka
2012-06-04 8:52 ` [Qemu-devel] [PATCH 08/13] pci: Export pci_parse_devaddr instead of pci_read_devaddr Jan Kiszka
2012-06-04 8:52 ` [Qemu-devel] [PATCH 09/13] pci: Introduce and apply PCIDeviceAddress Jan Kiszka
2012-06-10 9:37 ` Michael S. Tsirkin
2012-06-10 10:10 ` Jan Kiszka
2012-06-04 8:52 ` [Qemu-devel] [PATCH 10/13] pci: Fix coding style of pci_parse_devaddr Jan Kiszka
2012-06-04 8:52 ` [Qemu-devel] [PATCH 11/13] Move pci_parse_devaddr to qdev-properties Jan Kiszka
2012-06-07 12:57 ` Andreas Färber
2012-06-07 15:11 ` Jan Kiszka
2012-06-07 15:56 ` Andreas Färber
2012-06-08 10:57 ` Jan Kiszka
2012-06-08 12:03 ` Andreas Färber
2012-06-08 12:14 ` Jan Kiszka
2012-06-08 12:18 ` Andreas Färber
2012-06-08 12:45 ` Jan Kiszka
2012-06-08 14:17 ` Michael S. Tsirkin
2012-06-08 14:20 ` Anthony Liguori
2012-06-08 13:55 ` Anthony Liguori
2012-06-08 19:21 ` Andreas Färber
2012-06-04 8:52 ` [Qemu-devel] [PATCH 12/13] qdev-properties: Use qemu_parse_pci_devaddr for pci-devfn property Jan Kiszka
2012-06-04 8:52 ` [Qemu-devel] [PATCH 13/13] qdev-properties: Add pci-devaddr property Jan Kiszka
2012-06-10 9:35 ` Michael S. Tsirkin
2012-06-10 10:14 ` Jan Kiszka
2012-06-10 10:49 ` Michael S. Tsirkin
2012-06-10 10:52 ` Jan Kiszka
2012-06-10 10:58 ` Michael S. Tsirkin
2012-06-10 11:00 ` Jan Kiszka
2012-06-10 11:17 ` Michael S. Tsirkin
2012-06-10 11:25 ` Jan Kiszka
2012-06-10 12:01 ` Michael S. Tsirkin
2012-06-10 13:41 ` Alex Williamson
2012-06-10 14:03 ` Michael S. Tsirkin
2012-06-10 14:41 ` Alex Williamson
2012-06-10 14:54 ` Michael S. Tsirkin
2012-06-10 15:15 ` Alex Williamson
2012-06-10 15:37 ` Michael S. Tsirkin
2012-06-10 15:58 ` Alex Williamson
2012-06-10 16:22 ` Michael S. Tsirkin
2012-06-10 17:29 ` Alex Williamson
2012-06-10 17:57 ` Michael S. Tsirkin
2012-06-10 13:49 ` Michael S. Tsirkin
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=20120610112801.GL6250@redhat.com \
--to=mst@redhat.com \
--cc=alex.williamson@redhat.com \
--cc=jan.kiszka@siemens.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).