qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Jason Baron <jbaron@redhat.com>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: agraf@suse.de, aliguori@us.ibm.com, juzhang@redhat.com,
	jan.kiszka@siemens.com, qemu-devel@nongnu.org, armbru@redhat.com,
	blauwirbel@gmail.com, yamahata@valinux.co.jp,
	alex.williamson@redhat.com, kevin@koconnor.net, avi@redhat.com,
	mkletzan@redhat.com, pbonzini@redhat.com, lcapitulino@redhat.com,
	afaerber@suse.de, kraxel@redhat.com
Subject: Re: [Qemu-devel] [PATCH v2 21/21] q35: add acpi-based pci hotplug.
Date: Thu, 11 Oct 2012 11:34:08 -0400	[thread overview]
Message-ID: <20121011153408.GI3592@redhat.com> (raw)
In-Reply-To: <20121011144656.GF8983@redhat.com>

On Thu, Oct 11, 2012 at 04:46:56PM +0200, Michael S. Tsirkin wrote:
> On Thu, Oct 11, 2012 at 10:21:22AM -0400, Jason Baron wrote:
> > On Thu, Oct 11, 2012 at 12:57:06PM +0200, Michael S. Tsirkin wrote:
> > > On Mon, Oct 08, 2012 at 11:30:39PM -0400, Jason Baron wrote:
> > > > From: Jason Baron <jbaron@redhat.com>
> > > > 
> > > > Add piix style acpi hotplug to q35.
> > > > 
> > > > Signed-off-by: Jason Baron <jbaron@redhat.com>
> > > 
> > > Something I don't understand here: this only handles hotplug
> > > of devices behind the root, no?
> > > Don't we need support for hotplug/hot remove of devices behind
> > > bridges?
> > > 
> > 
> > Yes, this only handles acpi hotplug of devices behind the root. I'm
> > trying to reach minimal set of q35 patches that we can build upon. I
> > think that this patch gives us at least the same functionality as piix
> > does. (Plus there is pcie hotplug).
> > 
> > As you know, I have a proof of concept patch series providing a second
> > level of hotplug behind bridges. Unfortunately, it was based on the
> > static acpi tables, before Paolo made the hotplug table generation
> > dynamic. So it needs quite a bit of re-work. But I know that it should
> > work :)
> > 
> > Thanks,
> > 
> > -Jason
> 
> Yes. Reason I ask is because q35 is adding bridges by default now.
> Would it be possible to only add them if requested on command line by user
> instead? I realize some guests expect devices at specific slots
> but this does not apply to bridges I think?
> 

I just tried out getting rid of the bridges by default.

So 'lspci' goes from:

00:00.0 Host bridge: Intel Corporation 82G33/G31/P35/P31 Express DRAM Controller
00:01.0 PCI bridge: Intel Corporation 5500 Non-Legacy I/O Hub PCI Express Root Port 0 (rev 02)
00:02.0 VGA compatible controller: Cirrus Logic GD 5446
00:03.0 Ethernet controller: Intel Corporation 82540EM Gigabit Ethernet Controller (rev 03)
00:17.0 PCI bridge: Intel Corporation 5500 Non-Legacy I/O Hub PCI Express Root Port 0 (rev 02)
00:18.0 PCI bridge: Intel Corporation 5500 Non-Legacy I/O Hub PCI Express Root Port 0 (rev 02)
00:1c.0 PCI bridge: Intel Corporation 5500 Non-Legacy I/O Hub PCI Express Root Port 0 (rev 02)
00:1c.1 PCI bridge: Intel Corporation 5500 Non-Legacy I/O Hub PCI Express Root Port 0 (rev 02)
00:1c.2 PCI bridge: Intel Corporation 5500 Non-Legacy I/O Hub PCI Express Root Port 0 (rev 02)
00:1c.3 PCI bridge: Intel Corporation 5500 Non-Legacy I/O Hub PCI Express Root Port 0 (rev 02)
00:1c.4 PCI bridge: Intel Corporation 5500 Non-Legacy I/O Hub PCI Express Root Port 0 (rev 02)
00:1c.5 PCI bridge: Intel Corporation 5500 Non-Legacy I/O Hub PCI Express Root Port 0 (rev 02)
00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev 92)
00:1f.0 ISA bridge: Intel Corporation 82801IB (ICH9) LPC Interface Controller (rev 02)
00:1f.2 SATA controller: Intel Corporation 82801IR/IO/IH (ICH9R/DO/DH) 6 port SATA AHCI Controller (rev 02)
00:1f.3 SMBus: Intel Corporation 82801I (ICH9 Family) SMBus Controller (rev 02)
03:00.0 PCI bridge: Texas Instruments XIO3130 PCI Express Switch (Upstream) (rev 02)
04:00.0 PCI bridge: Texas Instruments XIO3130 PCI Express Switch (Downstream) (rev 01)
0c:1c.0 PCI bridge: Digital Equipment Corporation DECchip 21154 (rev 05)
0c:1d.0 PCI bridge: Digital Equipment Corporation DECchip 21154 (rev 05)
0c:1e.0 PCI bridge: Digital Equipment Corporation DECchip 21154 (rev 05)
0c:1f.0 PCI bridge: Digital Equipment Corporation DECchip 21154 (rev 05)

To:

00:00.0 Host bridge: Intel Corporation 82G33/G31/P35/P31 Express DRAM Controller
00:01.0 VGA compatible controller: Cirrus Logic GD 5446
00:02.0 Ethernet controller: Intel Corporation 82540EM Gigabit Ethernet Controller (rev 03)
00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev 92)
00:1f.0 ISA bridge: Intel Corporation 82801IB (ICH9) LPC Interface Controller (rev 02)
00:1f.2 SATA controller: Intel Corporation 82801IR/IO/IH (ICH9R/DO/DH) 6 port SATA AHCI Controller (rev 02)
00:1f.3 SMBus: Intel Corporation 82801I (ICH9 Family) SMBus Controller (rev 02)


Windows and Linux guests seem fine with either layout. Slots 1-2 are
specific to my setup. So this is a pretty minimal set.

I think that providing the minimal set of devices is good, since it
allows the user to configure things as much as possible. So I am in
favor of this more minimal set. My only hesitation is that we pull out,
or that I have not included some important piece h/w at a specific slot
that a guest might need. Thus potentially breaking existing setups.
Perhaps, that might mean a new machine type in the future, if we've
messed up?

These devices and slots are pulled from the Intel docs on ICH9 and Q35
specs. See:

http://www.intel.com/content/www/us/en/io/io-controller-hub-9-datasheet.html

Perhaps, Yamahata can comment further on the specific set of bridges?

> It would also be nice to add comments explaining why
> specific slots were selected e.g. /* BSD XYZ fails to boot unless ahci is at alow 2 */
> etc.

Right, its basically just pulled from the Intel spec as mentioned above.

> 
> Also - will adding this code now mean that when adding bridges
> we'll need to add compatibility code in bios/qemu in the future?
> 

I don't think so, but maybe you can elaborate this concern more
specifically?

Thanks,

-Jason

  parent reply	other threads:[~2012-10-11 15:34 UTC|newest]

Thread overview: 65+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-10-09  3:30 [Qemu-devel] [PATCH v2 00/21] q35 qemu support Jason Baron
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 01/21] blockdev: Introduce a default machine blockdev interface field, QEMUMachine->mach_if Jason Baron
2012-10-09  7:34   ` Paolo Bonzini
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 03/21] pci: pci capability must be in PCI space Jason Baron
2012-10-09  7:36   ` Paolo Bonzini
2012-10-13  8:29   ` Blue Swirl
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 02/21] blockdev: Introduce IF_AHCI Jason Baron
2012-10-09  7:36   ` Paolo Bonzini
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 04/21] pci: introduce pci_swizzle_map_irq_fn() for standardized interrupt pin swizzle Jason Baron
2012-10-09  7:39   ` Paolo Bonzini
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 05/21] pc, pc_piix: split out pc nic initialization Jason Baron
2012-10-09  7:39   ` Paolo Bonzini
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 06/21] pc: Move ioapic_init() from pc_piix.c to pc.c Jason Baron
2012-10-09  7:44   ` Paolo Bonzini
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 07/21] pc/piix_pci: factor out smram/pam logic Jason Baron
2012-10-09  7:47   ` Paolo Bonzini
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 08/21] pci_ids: add intel 82801BA pci-to-pci bridge id and PCI_CLASS_SERIAL_SMBUS Jason Baron
2012-10-09  7:48   ` Paolo Bonzini
2012-10-13  8:31   ` Blue Swirl
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 10/21] pcie: pass pcie window size to pcie_host_mmcfg_update() Jason Baron
2012-10-09  7:52   ` Paolo Bonzini
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 09/21] pci: Add class 0xc05 as 'SMBus' Jason Baron
2012-10-09  7:49   ` Paolo Bonzini
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 12/21] q35: Introduce q35 pc based chipset emulator Jason Baron
2012-10-11 14:47   ` Michael S. Tsirkin
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 11/21] pcie: Convert PCIExpressHost to use the QOM Jason Baron
2012-10-09  7:52   ` Paolo Bonzini
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 13/21] q35: Re-base q35 Jason Baron
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 14/21] q35: Suppress SMM BIOS initialization under KVM Jason Baron
2012-10-09  7:53   ` Paolo Bonzini
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 15/21] q35: Fix non-PCI IRQ processing in ich9_lpc_update_apic Jason Baron
2012-10-09  7:53   ` Paolo Bonzini
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 16/21] q35: smbus: Remove PCI_STATUS_SIG_SYSTEM_ERROR and PCI_STATUS_DETECTED_PARITY from w1cmask Jason Baron
2012-10-09  7:54   ` Paolo Bonzini
2012-10-11 14:53   ` Michael S. Tsirkin
2012-10-19 15:13     ` Jason Baron
2012-10-19 16:17       ` Isaku Yamahata
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 17/21] q35: Add kvmclock support Jason Baron
2012-10-09  7:54   ` Paolo Bonzini
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 18/21] q35: Fix irr initialization for slots 25..31 Jason Baron
2012-10-09  7:58   ` Paolo Bonzini
2012-10-11 14:49   ` Michael S. Tsirkin
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 19/21] Add a fallback bios file search, if -L fails Jason Baron
2012-10-09  7:59   ` Paolo Bonzini
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 20/21] q35: automatically load the q35 dsdt table Jason Baron
2012-10-09  8:02   ` Paolo Bonzini
2012-10-09  8:29     ` Paolo Bonzini
2012-10-09 20:06     ` Jason Baron
2012-10-13  8:33   ` Blue Swirl
2012-10-09  3:30 ` [Qemu-devel] [PATCH v2 21/21] q35: add acpi-based pci hotplug Jason Baron
2012-10-09  8:04   ` Paolo Bonzini
2012-10-11 10:57   ` Michael S. Tsirkin
2012-10-11 14:21     ` Jason Baron
2012-10-11 14:46       ` Michael S. Tsirkin
2012-10-11 14:54         ` Paolo Bonzini
2012-10-11 15:40           ` Jason Baron
2012-10-11 15:34         ` Jason Baron [this message]
2012-10-11 20:40           ` Michael S. Tsirkin
2012-10-12 15:27             ` Jason Baron
2012-10-13 23:03               ` Michael S. Tsirkin
2012-10-12  7:27           ` Gerd Hoffmann
2012-10-12  9:39             ` Michael S. Tsirkin
2012-10-12 10:06               ` Gerd Hoffmann
2012-10-12 10:39                 ` Michael S. Tsirkin
2012-10-12 15:00                 ` Jason Baron

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=20121011153408.GI3592@redhat.com \
    --to=jbaron@redhat.com \
    --cc=afaerber@suse.de \
    --cc=agraf@suse.de \
    --cc=alex.williamson@redhat.com \
    --cc=aliguori@us.ibm.com \
    --cc=armbru@redhat.com \
    --cc=avi@redhat.com \
    --cc=blauwirbel@gmail.com \
    --cc=jan.kiszka@siemens.com \
    --cc=juzhang@redhat.com \
    --cc=kevin@koconnor.net \
    --cc=kraxel@redhat.com \
    --cc=lcapitulino@redhat.com \
    --cc=mkletzan@redhat.com \
    --cc=mst@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=yamahata@valinux.co.jp \
    /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).