From: Paul Durrant <xadimgnik@gmail.com>
To: David Woodhouse <dwmw2@infradead.org>, qemu-devel@nongnu.org
Cc: "Paolo Bonzini" <pbonzini@redhat.com>,
"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>,
"Julien Grall" <julien@xen.org>
Subject: Re: [PATCH v6 15/51] i386/xen: manage and save/restore Xen guest long_mode setting
Date: Mon, 16 Jan 2023 17:20:45 +0000 [thread overview]
Message-ID: <9384a737-02b7-a32f-2805-4319542373bc@xen.org> (raw)
In-Reply-To: <20230110122042.1562155-16-dwmw2@infradead.org>
On 10/01/2023 12:20, David Woodhouse wrote:
> From: David Woodhouse <dwmw@amazon.co.uk>
>
> Xen will "latch" the guest's 32-bit or 64-bit ("long mode") setting when
> the guest writes the MSR to fill in the hypercall page, or when the guest
> sets the event channel callback in HVM_PARAM_CALLBACK_IRQ.
>
> KVM handles the former and sets the kernel's long_mode flag accordingly.
> The latter will be handled in userspace. Keep them in sync by noticing
> when a hypercall is made in a mode that doesn't match qemu's idea of
> the guest mode, and resyncing from the kernel. Do that same sync right
> before serialization too, in case the guest has set the hypercall page
> but hasn't yet made a system call.
>
> Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Reviewed-by: Paul Durrant <paul@xen.org>
... with one suggestion...
> ---
> hw/i386/kvm/xen_overlay.c | 65 +++++++++++++++++++++++++++++++++++++++
> hw/i386/kvm/xen_overlay.h | 4 +++
> target/i386/kvm/xen-emu.c | 12 ++++++++
> 3 files changed, 81 insertions(+)
>
> diff --git a/hw/i386/kvm/xen_overlay.c b/hw/i386/kvm/xen_overlay.c
> index 3e85bf912f..6fd63ff906 100644
> --- a/hw/i386/kvm/xen_overlay.c
> +++ b/hw/i386/kvm/xen_overlay.c
> @@ -44,6 +44,7 @@ struct XenOverlayState {
> MemoryRegion shinfo_mem;
> void *shinfo_ptr;
> uint64_t shinfo_gpa;
> + bool long_mode;
> };
>
> struct XenOverlayState *xen_overlay_singleton;
> @@ -96,9 +97,21 @@ static void xen_overlay_realize(DeviceState *dev, Error **errp)
>
> s->shinfo_ptr = memory_region_get_ram_ptr(&s->shinfo_mem);
> s->shinfo_gpa = INVALID_GPA;
> + s->long_mode = false;
> memset(s->shinfo_ptr, 0, XEN_PAGE_SIZE);
> }
>
> +static int xen_overlay_pre_save(void *opaque)
> +{
> + /*
> + * Fetch the kernel's idea of long_mode to avoid the race condition
> + * where the guest has set the hypercall page up in 64-bit mode but
> + * not yet made a hypercall by the time migration happens, so qemu
> + * hasn't yet noticed.
> + */
> + return xen_sync_long_mode();
> +}
> +
> static int xen_overlay_post_load(void *opaque, int version_id)
> {
> XenOverlayState *s = opaque;
> @@ -107,6 +120,9 @@ static int xen_overlay_post_load(void *opaque, int version_id)
> xen_overlay_map_page_locked(&s->shinfo_mem, s->shinfo_gpa);
> xen_overlay_set_be_shinfo(s->shinfo_gpa >> XEN_PAGE_SHIFT);
> }
> + if (s->long_mode) {
> + xen_set_long_mode(true);
> + }
>
> return 0;
> }
> @@ -121,9 +137,11 @@ static const VMStateDescription xen_overlay_vmstate = {
> .version_id = 1,
> .minimum_version_id = 1,
> .needed = xen_overlay_is_needed,
> + .pre_save = xen_overlay_pre_save,
> .post_load = xen_overlay_post_load,
> .fields = (VMStateField[]) {
> VMSTATE_UINT64(shinfo_gpa, XenOverlayState),
> + VMSTATE_BOOL(long_mode, XenOverlayState),
> VMSTATE_END_OF_LIST()
> }
> };
> @@ -198,3 +216,50 @@ void *xen_overlay_get_shinfo_ptr(void)
>
> return s->shinfo_ptr;
> }
> +
> +int xen_sync_long_mode(void)
> +{
> + int ret;
> + struct kvm_xen_hvm_attr xa = {
> + .type = KVM_XEN_ATTR_TYPE_LONG_MODE,
> + };
> +
> + if (!xen_overlay_singleton) {
> + return -ENOENT;
> + }
> +
> + ret = kvm_vm_ioctl(kvm_state, KVM_XEN_HVM_GET_ATTR, &xa);
> + if (!ret) {
> + xen_overlay_singleton->long_mode = xa.u.long_mode;
> + }
> +
> + return ret;
> +}
> +
> +int xen_set_long_mode(bool long_mode)
> +{
> + int ret;
> + struct kvm_xen_hvm_attr xa = {
> + .type = KVM_XEN_ATTR_TYPE_LONG_MODE,
> + .u.long_mode = long_mode,
> + };
> +
> + if (!xen_overlay_singleton) {
> + return -ENOENT;
> + }
> +
> + ret = kvm_vm_ioctl(kvm_state, KVM_XEN_HVM_SET_ATTR, &xa);
> + if (!ret) {
> + xen_overlay_singleton->long_mode = xa.u.long_mode;
> + }
> +
> + return ret;
> +}
> +
> +bool xen_is_long_mode(void)
> +{
> + if (xen_overlay_singleton) {
> + return xen_overlay_singleton->long_mode;
> + }
> + return false;
return xen_overlay_singleton && xen_overlay_singleton->long_mode;
perhaps?
> +}
> diff --git a/hw/i386/kvm/xen_overlay.h b/hw/i386/kvm/xen_overlay.h
> index 00cff05bb0..5c46a0b036 100644
> --- a/hw/i386/kvm/xen_overlay.h
> +++ b/hw/i386/kvm/xen_overlay.h
> @@ -17,4 +17,8 @@ void xen_overlay_create(void);
> int xen_overlay_map_shinfo_page(uint64_t gpa);
> void *xen_overlay_get_shinfo_ptr(void);
>
> +int xen_sync_long_mode(void);
> +int xen_set_long_mode(bool long_mode);
> +bool xen_is_long_mode(void);
> +
> #endif /* QEMU_XEN_OVERLAY_H */
> diff --git a/target/i386/kvm/xen-emu.c b/target/i386/kvm/xen-emu.c
> index 80005ea527..80f09f33df 100644
> --- a/target/i386/kvm/xen-emu.c
> +++ b/target/i386/kvm/xen-emu.c
> @@ -19,6 +19,8 @@
> #include "trace.h"
> #include "sysemu/runstate.h"
>
> +#include "hw/i386/kvm/xen_overlay.h"
> +
> #include "standard-headers/xen/version.h"
> #include "standard-headers/xen/sched.h"
>
> @@ -274,6 +276,16 @@ int kvm_xen_handle_exit(X86CPU *cpu, struct kvm_xen_exit *exit)
> return -1;
> }
>
> + /*
> + * The kernel latches the guest 32/64 mode when the MSR is used to fill
> + * the hypercall page. So if we see a hypercall in a mode that doesn't
> + * match our own idea of the guest mode, fetch the kernel's idea of the
> + * "long mode" to remain in sync.
> + */
> + if (exit->u.hcall.longmode != xen_is_long_mode()) {
> + xen_sync_long_mode();
> + }
> +
> if (!do_kvm_xen_handle_exit(cpu, exit)) {
> /*
> * Some hypercalls will be deliberately "implemented" by returning
next prev parent reply other threads:[~2023-01-16 17:21 UTC|newest]
Thread overview: 94+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-01-10 12:19 [PATCH v6 00/51] Xen support under KVM David Woodhouse
2023-01-10 12:19 ` [PATCH v6 01/51] include: import Xen public headers to include/standard-headers/ David Woodhouse
2023-01-10 12:19 ` [PATCH v6 02/51] xen: add CONFIG_XENFV_MACHINE and CONFIG_XEN_EMU options for Xen emulation David Woodhouse
2023-01-10 12:19 ` [PATCH v6 03/51] xen: Add XEN_DISABLED mode and make it default David Woodhouse
2023-01-10 12:19 ` [PATCH v6 04/51] i386/kvm: Add xen-version KVM accelerator property and init KVM Xen support David Woodhouse
2023-01-10 12:19 ` [PATCH v6 05/51] i386/kvm: handle Xen HVM cpuid leaves David Woodhouse
2023-01-10 12:19 ` [PATCH v6 06/51] i386/hvm: Set Xen vCPU ID in KVM David Woodhouse
2023-01-10 12:19 ` [PATCH v6 07/51] xen-platform: exclude vfio-pci from the PCI platform unplug David Woodhouse
2023-01-10 12:19 ` [PATCH v6 08/51] xen-platform: allow its creation with XEN_EMULATE mode David Woodhouse
2023-01-16 16:20 ` Paul Durrant
2023-01-16 17:56 ` David Woodhouse
2023-01-10 12:20 ` [PATCH v6 09/51] i386/xen: handle guest hypercalls David Woodhouse
2023-01-16 16:24 ` Paul Durrant
2023-01-16 17:57 ` David Woodhouse
2023-01-10 12:20 ` [PATCH v6 10/51] i386/xen: implement HYPERVISOR_xen_version David Woodhouse
2023-01-10 12:20 ` [PATCH v6 11/51] i386/xen: implement HYPERVISOR_sched_op, SCHEDOP_shutdown David Woodhouse
2023-01-16 16:27 ` Paul Durrant
2023-01-10 12:20 ` [PATCH v6 12/51] i386/xen: Implement SCHEDOP_poll and SCHEDOP_yield David Woodhouse
2023-01-16 16:36 ` Paul Durrant
2023-01-10 12:20 ` [PATCH v6 13/51] hw/xen: Add xen_overlay device for emulating shared xenheap pages David Woodhouse
2023-01-16 16:57 ` Paul Durrant
2023-01-10 12:20 ` [PATCH v6 14/51] i386/xen: add pc_machine_kvm_type to initialize XEN_EMULATE mode David Woodhouse
2023-01-16 17:17 ` Paul Durrant
2023-01-16 19:45 ` David Woodhouse
2023-01-10 12:20 ` [PATCH v6 15/51] i386/xen: manage and save/restore Xen guest long_mode setting David Woodhouse
2023-01-16 17:20 ` Paul Durrant [this message]
2023-01-10 12:20 ` [PATCH v6 16/51] i386/xen: implement HYPERVISOR_memory_op David Woodhouse
2023-01-16 17:28 ` Paul Durrant
2023-01-10 12:20 ` [PATCH v6 17/51] i386/xen: implement XENMEM_add_to_physmap_batch David Woodhouse
2023-01-16 17:36 ` Paul Durrant
2023-01-10 12:20 ` [PATCH v6 18/51] i386/xen: implement HYPERVISOR_hvm_op David Woodhouse
2023-01-16 17:39 ` Paul Durrant
2023-01-10 12:20 ` [PATCH v6 19/51] i386/xen: implement HYPERVISOR_vcpu_op David Woodhouse
2023-01-16 17:40 ` Paul Durrant
2023-01-10 12:20 ` [PATCH v6 20/51] i386/xen: handle VCPUOP_register_vcpu_info David Woodhouse
2023-01-16 17:46 ` Paul Durrant
2023-01-10 12:20 ` [PATCH v6 21/51] i386/xen: handle VCPUOP_register_vcpu_time_info David Woodhouse
2023-01-16 17:53 ` Paul Durrant
2023-01-10 12:20 ` [PATCH v6 22/51] i386/xen: handle VCPUOP_register_runstate_memory_area David Woodhouse
2023-01-16 17:56 ` Paul Durrant
2023-01-10 12:20 ` [PATCH v6 23/51] i386/xen: implement HYPERVISOR_event_channel_op David Woodhouse
2023-01-16 17:59 ` Paul Durrant
2023-01-16 19:54 ` David Woodhouse
2023-01-10 12:20 ` [PATCH v6 24/51] i386/xen: implement HVMOP_set_evtchn_upcall_vector David Woodhouse
2023-01-10 12:20 ` [PATCH v6 25/51] i386/xen: implement HVMOP_set_param David Woodhouse
2023-01-16 18:00 ` Paul Durrant
2023-01-10 12:20 ` [PATCH v6 26/51] hw/xen: Add xen_evtchn device for event channel emulation David Woodhouse
2023-01-10 12:20 ` [PATCH v6 27/51] i386/xen: Add support for Xen event channel delivery to vCPU David Woodhouse
2023-01-10 12:20 ` [PATCH v6 28/51] hw/xen: Implement EVTCHNOP_status David Woodhouse
2023-01-10 12:20 ` [PATCH v6 29/51] hw/xen: Implement EVTCHNOP_close David Woodhouse
2023-01-10 12:20 ` [PATCH v6 30/51] hw/xen: Implement EVTCHNOP_unmask David Woodhouse
2023-01-10 12:20 ` [PATCH v6 31/51] hw/xen: Implement EVTCHNOP_bind_virq David Woodhouse
2023-01-10 12:20 ` [PATCH v6 32/51] hw/xen: Implement EVTCHNOP_bind_ipi David Woodhouse
2023-01-10 12:20 ` [PATCH v6 33/51] hw/xen: Implement EVTCHNOP_send David Woodhouse
2023-01-10 12:20 ` [PATCH v6 34/51] hw/xen: Implement EVTCHNOP_alloc_unbound David Woodhouse
2023-01-10 12:20 ` [PATCH v6 35/51] hw/xen: Implement EVTCHNOP_bind_interdomain David Woodhouse
2023-01-10 12:20 ` [PATCH v6 36/51] hw/xen: Implement EVTCHNOP_bind_vcpu David Woodhouse
2023-01-10 12:20 ` [PATCH v6 37/51] hw/xen: Implement EVTCHNOP_reset David Woodhouse
2023-01-10 12:20 ` [PATCH v6 38/51] i386/xen: add monitor commands to test event injection David Woodhouse
2023-01-11 14:28 ` Dr. David Alan Gilbert
2023-01-11 14:57 ` David Woodhouse
2023-01-10 12:20 ` [PATCH v6 39/51] hw/xen: Support HVM_PARAM_CALLBACK_TYPE_GSI callback David Woodhouse
2023-01-10 12:20 ` [PATCH v6 40/51] hw/xen: Support HVM_PARAM_CALLBACK_TYPE_PCI_INTX callback David Woodhouse
2023-01-10 12:20 ` [PATCH v6 41/51] kvm/i386: Add xen-gnttab-max-frames property David Woodhouse
2023-01-10 12:20 ` [PATCH v6 42/51] hw/xen: Add xen_gnttab device for grant table emulation David Woodhouse
2023-01-10 12:20 ` [PATCH v6 43/51] hw/xen: Support mapping grant frames David Woodhouse
2023-01-10 12:20 ` [PATCH v6 44/51] i386/xen: Implement HYPERVISOR_grant_table_op and GNTTABOP_[gs]et_verson David Woodhouse
2023-01-10 12:20 ` [PATCH v6 45/51] hw/xen: Implement GNTTABOP_query_size David Woodhouse
2023-01-10 12:20 ` [PATCH v6 46/51] i386/xen: handle PV timer hypercalls David Woodhouse
2023-01-10 12:20 ` [PATCH v6 47/51] i386/xen: Reserve Xen special pages for console, xenstore rings David Woodhouse
2023-01-10 12:20 ` [PATCH v6 48/51] i386/xen: handle HVMOP_get_param David Woodhouse
2023-01-10 12:20 ` [PATCH v6 49/51] hw/xen: Add backend implementation of interdomain event channel support David Woodhouse
2023-01-10 12:20 ` [PATCH v6 50/51] hw/xen: Add xen_xenstore device for xenstore emulation David Woodhouse
2023-01-10 12:20 ` [PATCH v6 51/51] hw/xen: Add basic ring handling to xenstore David Woodhouse
2023-01-10 12:37 ` [RFC PATCH v1 00/15] Xen PV backend support for KVM/Xen guests David Woodhouse
2023-01-10 12:37 ` [RFC PATCH v1 01/15] hw/xen: Add evtchn operations to allow redirection to internal emulation David Woodhouse
2023-01-10 12:37 ` [RFC PATCH v1 02/15] hw/xen: Add emulated evtchn ops David Woodhouse
2023-01-10 12:37 ` [RFC PATCH v1 03/15] hw/xen: Add gnttab operations to allow redirection to internal emulation David Woodhouse
2023-01-10 12:37 ` [RFC PATCH v1 04/15] hw/xen: Pass grant ref to gnttab unmap David Woodhouse
2023-01-10 12:37 ` [RFC PATCH v1 05/15] hw/xen: Add foreignmem operations to allow redirection to internal emulation David Woodhouse
2023-01-10 12:37 ` [RFC PATCH v1 06/15] hw/xen: Add xenstore " David Woodhouse
2023-01-10 12:37 ` [RFC PATCH v1 07/15] hw/xen: Move xenstore_store_pv_console_info to xen_console.c David Woodhouse
2023-01-10 12:37 ` [RFC PATCH v1 08/15] hw/xen: Use XEN_PAGE_SIZE in PV backend drivers David Woodhouse
2023-01-10 12:37 ` [RFC PATCH v1 09/15] hw/xen: Rename xen_common.h to xen_native.h David Woodhouse
2023-01-10 12:37 ` [RFC PATCH v1 10/15] hw/xen: Build PV backend drivers for XENFV_MACHINE David Woodhouse
2023-01-10 12:37 ` [RFC PATCH v1 11/15] hw/xen: Map guest XENSTORE_PFN grant in emulated Xenstore David Woodhouse
2023-01-10 12:37 ` [RFC PATCH v1 12/15] hw/xen: Add backend implementation of grant table operations David Woodhouse
2023-01-10 12:37 ` [RFC PATCH v1 13/15] hw/xen: Implement soft reset for emulated gnttab David Woodhouse
2023-01-10 12:37 ` [RFC PATCH v1 14/15] hw/xen: Remove old version of Xen headers David Woodhouse
2023-01-10 12:37 ` [RFC PATCH v1 15/15] i386/xen: Initialize XenBus and legacy backends from pc_init1() David Woodhouse
2023-01-10 15:43 ` [RFC PATCH v1 00/15] Xen PV backend support for KVM/Xen guests Joao Martins
2023-01-10 15:47 ` Joao Martins
2023-01-10 16:52 ` David Woodhouse
2023-01-10 17:26 ` Joao Martins
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=9384a737-02b7-a32f-2805-4319542373bc@xen.org \
--to=xadimgnik@gmail.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=julien@xen.org \
--cc=paul@xen.org \
--cc=pbonzini@redhat.com \
--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).