qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: David Woodhouse <dwmw2@infradead.org>, qemu-devel@nongnu.org
Cc: "Paul Durrant" <paul@xen.org>,
	"Joao Martins" <joao.m.martins@oracle.com>,
	"Ankur Arora" <ankur.a.arora@oracle.com>,
	"Philippe Mathieu-Daudé" <philmd@linaro.org>,
	"Thomas Huth" <thuth@redhat.com>,
	"Alex Bennée" <alex.bennee@linaro.org>,
	"Juan Quintela" <quintela@redhat.com>,
	"Dr . David Alan Gilbert" <dgilbert@redhat.com>,
	"Claudio Fontana" <cfontana@suse.de>
Subject: Re: [RFC PATCH v2 03/22] i386/xen: Add xen-version machine property and init KVM Xen support
Date: Mon, 12 Dec 2022 18:30:21 +0100	[thread overview]
Message-ID: <8495140d-3301-7693-b804-0554166802da@redhat.com> (raw)
In-Reply-To: <20221209095612.689243-4-dwmw2@infradead.org>

On 12/9/22 10:55, David Woodhouse wrote:
> -    m->default_machine_opts = "accel=xen,suppress-vmdesc=on";
> +    if (xen_enabled())
> +            m->default_machine_opts = "accel=xen,suppress-vmdesc=on";
> +    else
> +            m->default_machine_opts = "accel=kvm,xen-version=0x30001";

Please do not modify pc_xen_hvm_init().

"-M xenfv" should be the same as "-M pc-i440fx-...,suppress-vmdesc=on 
-accel xen -device xen-platform".  It must work *without* "-accel xen", 
while here you're basically requiring it.  For now, please make 
KVM-emulated Xen use "-M pc -device xen-platform".  We can figure out 
"-M xenfv" later.

You can instead have:

- a check in xen_init() that checks that xen_mode is XEN_ATTACH.  If 
not, fail.

- an extra enum value for xen_mode, XEN_DISABLED, which is the default 
instead of XEN_EMULATE;

- an accelerator property "-accel kvm,xen-version=...", added in 
kvm_accel_class_init() instead of the machine property.  The property, 
when set to a nonzero value, flips xen_mode from XEN_DISABLED to 
XEN_EMULATE.

The Xen overlay device can be created using the mc->kvm_type function 
(which you can set in DEFINE_PC_MACHINE); at that point, xen_mode has 
switched from XEN_DISABLED to XEN_EMULATE.  Those xen_enabled() checks 
that apply to KVM then become xen_mode != XEN_DISABLED, as long as they 
run during mc->kvm_type or afterwards.

The platform device can be created either in mc->kvm_type or manually 
(not sure if it makes sense to have a "XenVMMXenVMM" CPUID + emulated 
hypercalls but no platform device---would it still use pvclock for 
example?).

Paolo



  parent reply	other threads:[~2022-12-12 17:31 UTC|newest]

Thread overview: 78+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-12-09  9:55 [RFC PATCH v2 00/22] Xen HVM support under KVM David Woodhouse
2022-12-09  9:55 ` [RFC PATCH v2 01/22] include: import xen public headers David Woodhouse
2022-12-12  9:17   ` Paul Durrant
2022-12-09  9:55 ` [RFC PATCH v2 02/22] xen: add CONFIG_XENFV_MACHINE and CONFIG_XEN_EMU options for Xen emulation David Woodhouse
2022-12-12  9:19   ` Paul Durrant
2022-12-12 17:07   ` Paolo Bonzini
2022-12-12 22:22     ` David Woodhouse
2022-12-13  0:39       ` Paolo Bonzini
2022-12-13  0:59         ` David Woodhouse
2022-12-13 22:32           ` Paolo Bonzini
2022-12-16  8:40             ` David Woodhouse
2022-12-09  9:55 ` [RFC PATCH v2 03/22] i386/xen: Add xen-version machine property and init KVM Xen support David Woodhouse
2022-12-12 12:48   ` Paul Durrant
2022-12-12 17:30   ` Paolo Bonzini [this message]
2022-12-12 17:55     ` Paul Durrant
2022-12-13  0:13     ` David Woodhouse
2023-01-17 13:49     ` David Woodhouse
2022-12-09  9:55 ` [RFC PATCH v2 04/22] i386/kvm: handle Xen HVM cpuid leaves David Woodhouse
2022-12-12 13:13   ` Paul Durrant
2022-12-13  9:47     ` David Woodhouse
2022-12-09  9:55 ` [RFC PATCH v2 05/22] xen-platform-pci: allow its creation with XEN_EMULATE mode David Woodhouse
2022-12-12 13:24   ` Paul Durrant
2022-12-12 22:07     ` David Woodhouse
2022-12-09  9:55 ` [RFC PATCH v2 06/22] hw/xen_backend: refactor xen_be_init() David Woodhouse
2022-12-12 13:27   ` Paul Durrant
2022-12-09  9:55 ` [RFC PATCH v2 07/22] pc_piix: handle XEN_EMULATE backend init David Woodhouse
2022-12-12 13:47   ` Paul Durrant
2022-12-12 14:50     ` David Woodhouse
2022-12-09  9:55 ` [RFC PATCH v2 08/22] xen_platform: exclude vfio-pci from the PCI platform unplug David Woodhouse
2022-12-12 13:52   ` Paul Durrant
2022-12-09  9:55 ` [RFC PATCH v2 09/22] pc_piix: allow xenfv machine with XEN_EMULATE David Woodhouse
2022-12-12 14:05   ` Paul Durrant
2022-12-09  9:56 ` [RFC PATCH v2 10/22] i386/xen: handle guest hypercalls David Woodhouse
2022-12-12 14:11   ` Paul Durrant
2022-12-12 14:17     ` David Woodhouse
2022-12-12 17:07   ` Paolo Bonzini
2022-12-09  9:56 ` [RFC PATCH v2 11/22] i386/xen: implement HYPERCALL_xen_version David Woodhouse
2022-12-12 14:17   ` Paul Durrant
2022-12-13  0:06     ` David Woodhouse
2022-12-09  9:56 ` [RFC PATCH v2 12/22] hw/xen: Add xen_overlay device for emulating shared xenheap pages David Woodhouse
2022-12-12 14:29   ` Paul Durrant
2022-12-12 17:14   ` Paolo Bonzini
2022-12-09  9:56 ` [RFC PATCH v2 13/22] i386/xen: implement HYPERVISOR_memory_op David Woodhouse
2022-12-12 14:38   ` Paul Durrant
2022-12-13  0:08     ` David Woodhouse
2022-12-09  9:56 ` [RFC PATCH v2 14/22] i386/xen: implement HYPERVISOR_hvm_op David Woodhouse
2022-12-12 14:41   ` Paul Durrant
2022-12-09  9:56 ` [RFC PATCH v2 15/22] i386/xen: implement HYPERVISOR_vcpu_op David Woodhouse
2022-12-12 14:51   ` Paul Durrant
2022-12-13  0:10     ` David Woodhouse
2022-12-09  9:56 ` [RFC PATCH v2 16/22] i386/xen: handle VCPUOP_register_vcpu_info David Woodhouse
2022-12-12 14:58   ` Paul Durrant
2022-12-13  0:13     ` David Woodhouse
2022-12-14 10:28       ` Paul Durrant
2022-12-14 11:04         ` David Woodhouse
2022-12-09  9:56 ` [RFC PATCH v2 17/22] i386/xen: handle VCPUOP_register_vcpu_time_info David Woodhouse
2022-12-12 15:34   ` Paul Durrant
2022-12-09  9:56 ` [RFC PATCH v2 18/22] i386/xen: handle VCPUOP_register_runstate_memory_area David Woodhouse
2022-12-12 15:38   ` Paul Durrant
2022-12-09  9:56 ` [RFC PATCH v2 19/22] i386/xen: implement HVMOP_set_evtchn_upcall_vector David Woodhouse
2022-12-12 15:52   ` Paul Durrant
2022-12-09  9:56 ` [RFC PATCH v2 20/22] i386/xen: HVMOP_set_param / HVM_PARAM_CALLBACK_IRQ David Woodhouse
2022-12-12 16:16   ` Paul Durrant
2022-12-12 16:26     ` David Woodhouse
2022-12-12 16:39       ` Paul Durrant
2022-12-15 20:54         ` David Woodhouse
2022-12-20 13:56           ` Paul Durrant
2022-12-20 16:27             ` David Woodhouse
2022-12-20 17:25               ` Paul Durrant
2022-12-20 17:29                 ` David Woodhouse
2022-12-28 10:45                   ` David Woodhouse
2022-12-21  1:41     ` David Woodhouse
2022-12-21  9:37       ` Paul Durrant
2022-12-21 12:16         ` David Woodhouse
2022-12-09  9:56 ` [RFC PATCH v2 21/22] i386/xen: implement HYPERVISOR_event_channel_op David Woodhouse
2022-12-12 16:23   ` Paul Durrant
2022-12-09  9:56 ` [RFC PATCH v2 22/22] i386/xen: implement HYPERVISOR_sched_op David Woodhouse
2022-12-12 16:37   ` Paul Durrant

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=8495140d-3301-7693-b804-0554166802da@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=alex.bennee@linaro.org \
    --cc=ankur.a.arora@oracle.com \
    --cc=cfontana@suse.de \
    --cc=dgilbert@redhat.com \
    --cc=dwmw2@infradead.org \
    --cc=joao.m.martins@oracle.com \
    --cc=paul@xen.org \
    --cc=philmd@linaro.org \
    --cc=qemu-devel@nongnu.org \
    --cc=quintela@redhat.com \
    --cc=thuth@redhat.com \
    /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).