From: Roman Kagan <rkagan@virtuozzo.com>
To: Vitaly Kuznetsov <vkuznets@redhat.com>
Cc: "kvm@vger.kernel.org" <kvm@vger.kernel.org>,
Michael Kelley <mikelley@microsoft.com>,
Lan Tianyu <Tianyu.Lan@microsoft.com>,
Joerg Roedel <jroedel@suse.de>,
"K. Y. Srinivasan" <kys@microsoft.com>,
Haiyang Zhang <haiyangz@microsoft.com>,
Stephen Hemminger <sthemmin@microsoft.com>,
Sasha Levin <sashal@kernel.org>,
Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
"H. Peter Anvin" <hpa@zytor.com>,
"x86@kernel.org" <x86@kernel.org>,
"linux-hyperv@vger.kernel.org" <linux-hyperv@vger.kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v2] x86/hyperv: make vapic support x2apic mode
Date: Thu, 3 Oct 2019 12:52:39 +0000 [thread overview]
Message-ID: <20191003125236.GA2424@rkaganb.sw.ru> (raw)
In-Reply-To: <87muei14ms.fsf@vitty.brq.redhat.com>
On Thu, Oct 03, 2019 at 12:54:03PM +0200, Vitaly Kuznetsov wrote:
> Roman Kagan <rkagan@virtuozzo.com> writes:
>
> > Now that there's Hyper-V IOMMU driver, Linux can switch to x2apic mode
> > when supported by the vcpus.
> >
> > However, the apic access functions for Hyper-V enlightened apic assume
> > xapic mode only.
> >
> > As a result, Linux fails to bring up secondary cpus when run as a guest
> > in QEMU/KVM with both hv_apic and x2apic enabled.
> >
> > I didn't manage to make my instance of Hyper-V expose x2apic to the
> > guest; nor does Hyper-V spec document the expected behavior. However,
> > a Windows guest running in QEMU/KVM with hv_apic and x2apic and a big
> > number of vcpus (so that it turns on x2apic mode) does use enlightened
> > apic MSRs passing unshifted 32bit destination id and falls back to the
> > regular x2apic MSRs for less frequently used apic fields.
> >
> > So implement the same behavior, by replacing enlightened apic access
> > functions (only those where it makes a difference) with their
> > x2apic-aware versions when x2apic is in use.
> >
> > Fixes: 29217a474683 ("iommu/hyper-v: Add Hyper-V stub IOMMU driver")
> > Fixes: 6b48cb5f8347 ("X86/Hyper-V: Enlighten APIC access")
> > Cc: stable@vger.kernel.org
> > Signed-off-by: Roman Kagan <rkagan@virtuozzo.com>
> > ---
> > v1 -> v2:
> > - add ifdefs to handle !CONFIG_X86_X2APIC
> >
> > arch/x86/hyperv/hv_apic.c | 54 ++++++++++++++++++++++++++++++++++++---
> > 1 file changed, 51 insertions(+), 3 deletions(-)
> >
> > diff --git a/arch/x86/hyperv/hv_apic.c b/arch/x86/hyperv/hv_apic.c
> > index 5c056b8aebef..eb1434ae9e46 100644
> > --- a/arch/x86/hyperv/hv_apic.c
> > +++ b/arch/x86/hyperv/hv_apic.c
> > @@ -84,6 +84,44 @@ static void hv_apic_write(u32 reg, u32 val)
> > }
> > }
> >
> > +#ifdef CONFIG_X86_X2APIC
> > +static void hv_x2apic_icr_write(u32 low, u32 id)
> > +{
> > + wrmsr(HV_X64_MSR_ICR, low, id);
> > +}
>
> AFAIU you're trying to mirror native_x2apic_icr_write() here but this is
> different from what hv_apic_icr_write() does
> (SET_APIC_DEST_FIELD(id)).
Right. In xapic mode the ICR2 aka the high 4 bytes of ICR is programmed
with the destination id in the highest byte; in x2apic mode the whole
ICR2 is set to the 32bit destination id.
> Is it actually correct? (I think you've tested this and it is but)
As I wrote in the commit log, I haven't tested it in the sense that I
ran a Linux guest in a Hyper-V VM exposing x2apic to the guest, because
I didn't manage to configure it to do so. OTOH I did run a Windows
guest in QEMU/KVM with hv_apic and x2apic enabled and saw it write
destination ids unshifted to the ICR2 part of ICR, so I assume it's
correct.
> Michael, could you please shed some light here?
Would be appreciated, indeed.
> > +static u32 hv_x2apic_read(u32 reg)
> > +{
> > + u32 reg_val, hi;
> > +
> > + switch (reg) {
> > + case APIC_EOI:
> > + rdmsr(HV_X64_MSR_EOI, reg_val, hi);
> > + return reg_val;
> > + case APIC_TASKPRI:
> > + rdmsr(HV_X64_MSR_TPR, reg_val, hi);
> > + return reg_val;
> > +
> > + default:
> > + return native_apic_msr_read(reg);
> > + }
> > +}
> > +
> > +static void hv_x2apic_write(u32 reg, u32 val)
> > +{
> > + switch (reg) {
> > + case APIC_EOI:
> > + wrmsr(HV_X64_MSR_EOI, val, 0);
> > + break;
> > + case APIC_TASKPRI:
> > + wrmsr(HV_X64_MSR_TPR, val, 0);
> > + break;
> > + default:
> > + native_apic_msr_write(reg, val);
> > + }
> > +}
> > +#endif /* CONFIG_X86_X2APIC */
> > +
> > static void hv_apic_eoi_write(u32 reg, u32 val)
> > {
> > struct hv_vp_assist_page *hvp = hv_vp_assist_page[smp_processor_id()];
> > @@ -262,9 +300,19 @@ void __init hv_apic_init(void)
> > if (ms_hyperv.hints & HV_X64_APIC_ACCESS_RECOMMENDED) {
> > pr_info("Hyper-V: Using MSR based APIC access\n");
> > apic_set_eoi_write(hv_apic_eoi_write);
> > - apic->read = hv_apic_read;
> > - apic->write = hv_apic_write;
> > - apic->icr_write = hv_apic_icr_write;
> > +#ifdef CONFIG_X86_X2APIC
> > + if (x2apic_enabled()) {
> > + apic->read = hv_x2apic_read;
> > + apic->write = hv_x2apic_write;
> > + apic->icr_write = hv_x2apic_icr_write;
> > + } else {
> > +#endif
> > + apic->read = hv_apic_read;
> > + apic->write = hv_apic_write;
> > + apic->icr_write = hv_apic_icr_write;
>
> (just wondering): Is it always safe to assume that we cannot switch
> between apic_flat/x2apic in runtime?
I guess so. All apic choices are made early at __init, obviously before
the secondary CPUs are brought up, and aren't reconsidered afterwards.
> Moreover, the only difference
> between hv_apic_read/hv_apic_write and hv_x2apic_read/hv_x2apic_write is
> native_apic_mem_{read,write} -> native_apic_msr_{read,write}. Would it
> make sense to move if (x2apic_enabled()) and merge these functions?
x2apic_enabled() is too heavy for that: it reads MSR_IA32_APICBASE. One
could probably use a read-mostly global variable instead but I'm not
sure it would make the code better.
Thanks,
Roman.
next prev parent reply other threads:[~2019-10-03 12:52 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-02 10:19 [PATCH v2] x86/hyperv: make vapic support x2apic mode Roman Kagan
2019-10-03 10:54 ` Vitaly Kuznetsov
2019-10-03 12:52 ` Roman Kagan [this message]
2019-10-04 3:01 ` Michael Kelley
2019-10-04 9:18 ` Roman Kagan
2019-10-04 22:33 ` Michael Kelley
2019-10-08 20:54 ` Michael Kelley
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=20191003125236.GA2424@rkaganb.sw.ru \
--to=rkagan@virtuozzo.com \
--cc=Tianyu.Lan@microsoft.com \
--cc=bp@alien8.de \
--cc=haiyangz@microsoft.com \
--cc=hpa@zytor.com \
--cc=jroedel@suse.de \
--cc=kvm@vger.kernel.org \
--cc=kys@microsoft.com \
--cc=linux-hyperv@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mikelley@microsoft.com \
--cc=mingo@redhat.com \
--cc=sashal@kernel.org \
--cc=sthemmin@microsoft.com \
--cc=tglx@linutronix.de \
--cc=vkuznets@redhat.com \
--cc=x86@kernel.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).