From: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
To: Paolo Bonzini <pbonzini@redhat.com>,
"rkagan@virtuozzo.com" <rkagan@virtuozzo.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"kvm@vger.kernel.org" <kvm@vger.kernel.org>,
"rkrcmar@redhat.com" <rkrcmar@redhat.com>,
"joro@8bytes.org" <joro@8bytes.org>,
"vkuznets@redhat.com" <vkuznets@redhat.com>,
"graf@amazon.com" <graf@amazon.com>,
"jschoenh@amazon.de" <jschoenh@amazon.de>,
"karahmed@amazon.de" <karahmed@amazon.de>,
"rimasluk@amazon.com" <rimasluk@amazon.com>,
"Grimm, Jon" <Jon.Grimm@amd.com>
Subject: Re: [PATCH v4 13/17] kvm: i8254: Deactivate APICv when using in-kernel PIT re-injection mode.
Date: Mon, 11 Nov 2019 11:37:53 -0600 [thread overview]
Message-ID: <9361adbc-77e8-4964-c859-8956e1fbb182@amd.com> (raw)
In-Reply-To: <ac4313a6-df96-2223-bed3-33c3a8555c98@redhat.com>
Roman/Paolo
On 11/5/2019 4:47 PM, Paolo Bonzini wrote:
> On 05/11/19 00:17, Roman Kagan wrote:
>>> This is not too nice for Intel which does support (through the EOI exit
>>> mask) APICv even if PIT reinjection active.
>> Hmm, it's tempting to just make svm_load_eoi_exitmap() disable AVIC when
>> given a non-empty eoi_exit_bitmap, and enable it back on a clear
>> eoi_exit_bitmap. This may remove the need to add special treatment to
>> PIT etc.
>
> That is a very nice idea---we can make that a single disable reason,
> like APICV_DEACTIVATE_REASON_EOI, and Intel can simply never use it.
I took at look at the svm_load_eoi_exitmap() and it is called via:
kvm_make_scan_ioapic_request() ->
KVM_REQ_SCAN_IOAPIC -> vcpu_scan_ioapic() ->
KVM_REQ_LOAD_EOI_EXITMAP -> vcpu_load_eoi_exitmap()
The kvm_make_scan_ioapic_request() is called from multiple places:
arch/x86/kvm/irq_comm.c:
* kvm_arch_post_irq_routing_update() : Called from kvm_set_irq_routing()
arch/x86/kvm/ioapic.c:
* kvm_arch_post_irq_ack_notifier_list_update() : (Un)registering irq ack notifier
* kvm_set_ioapic() : Setting ioapic irqchip
* ioapic_mmio_write() -> ioapic_write_indirect()
arch/x86/kvm/lapic.c:
* recalculate_apic_map()
Most calls would be from ioapic_mmio_write()->ioapic_write_indirect().
In case of AMD AVIC, the svm_load_e::vsoi_exitmap() is called several times, and requesting
APICV (de)activate from here when the eoi_exit_bitmap is set/clear would introduce
large overhead especially with SMP machine. So, for now, let's just disable APICv
when in-kernel PIT is in reinject (delay) mode.
I'll also add the logic to avoid unnecessary overhead for Intel.
Thanks,
Suravee
next prev parent reply other threads:[~2019-11-11 17:38 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-11-01 22:41 [PATCH v4 00/17] kvm: x86: Support AMD SVM AVIC w/ in-kernel irqchip mode Suthikulpanit, Suravee
2019-11-01 22:41 ` [PATCH v4 01/17] kvm: x86: Modify kvm_x86_ops.get_enable_apicv() to use struct kvm parameter Suthikulpanit, Suravee
2019-11-01 22:41 ` [PATCH v4 02/17] kvm: lapic: Introduce APICv update helper function Suthikulpanit, Suravee
2019-11-01 22:41 ` [PATCH v4 03/17] kvm: x86: Introduce APICv deactivate bits Suthikulpanit, Suravee
2019-11-02 9:51 ` Paolo Bonzini
2019-11-01 22:41 ` [PATCH v4 04/17] kvm: x86: Add support for activate/de-activate APICv at runtime Suthikulpanit, Suravee
2019-11-02 9:52 ` Paolo Bonzini
2019-11-04 19:22 ` Suthikulpanit, Suravee
2019-11-01 22:41 ` [PATCH v4 05/17] kvm: x86: Add APICv activate/deactivate request trace points Suthikulpanit, Suravee
2019-11-01 22:41 ` [PATCH v4 06/17] kvm: x86: svm: Add support to activate/deactivate posted interrupts Suthikulpanit, Suravee
2019-11-01 22:41 ` [PATCH v4 07/17] svm: Add support for setup/destroy virutal APIC backing page for AVIC Suthikulpanit, Suravee
2019-11-04 21:53 ` Roman Kagan
2019-11-12 0:05 ` Suravee Suthikulpanit
2019-11-01 22:41 ` [PATCH v4 08/17] kvm: x86: Introduce APICv pre-update hook Suthikulpanit, Suravee
2019-11-04 22:05 ` Roman Kagan
2019-11-12 0:08 ` Suravee Suthikulpanit
2019-11-12 11:12 ` Roman Kagan
2019-11-01 22:41 ` [PATCH v4 09/17] svm: Add support for activate/deactivate AVIC at runtime Suthikulpanit, Suravee
2019-11-01 22:41 ` [PATCH v4 10/17] kvm: x86: hyperv: Use APICv update request interface Suthikulpanit, Suravee
2019-11-01 22:41 ` [PATCH v4 11/17] svm: Deactivate AVIC when launching guest with nested SVM support Suthikulpanit, Suravee
2019-11-01 22:41 ` [PATCH v4 12/17] svm: Temporary deactivate AVIC during ExtINT handling Suthikulpanit, Suravee
2019-11-02 10:01 ` Paolo Bonzini
2019-11-01 22:41 ` [PATCH v4 13/17] kvm: i8254: Deactivate APICv when using in-kernel PIT re-injection mode Suthikulpanit, Suravee
2019-11-02 9:57 ` Paolo Bonzini
2019-11-04 18:54 ` Suthikulpanit, Suravee
2019-11-04 21:49 ` Paolo Bonzini
2019-11-05 7:05 ` Graf (AWS), Alexander
2019-11-05 8:40 ` Roman Kagan
2019-11-04 23:17 ` Roman Kagan
2019-11-05 22:47 ` Paolo Bonzini
2019-11-11 17:37 ` Suravee Suthikulpanit [this message]
2019-11-12 12:22 ` Roman Kagan
2019-11-01 22:41 ` [PATCH v4 14/17] kvm: lapic: Clean up APIC predefined macros Suthikulpanit, Suravee
2019-11-01 22:41 ` [PATCH v4 15/17] kvm: ioapic: Refactor kvm_ioapic_update_eoi() Suthikulpanit, Suravee
2019-11-01 22:41 ` [PATCH v4 16/17] kvm: ioapic: Lazy update IOAPIC EOI Suthikulpanit, Suravee
2019-11-01 22:41 ` [PATCH v4 17/17] svm: Allow AVIC with in-kernel irqchip mode Suthikulpanit, Suravee
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=9361adbc-77e8-4964-c859-8956e1fbb182@amd.com \
--to=suravee.suthikulpanit@amd.com \
--cc=Jon.Grimm@amd.com \
--cc=graf@amazon.com \
--cc=joro@8bytes.org \
--cc=jschoenh@amazon.de \
--cc=karahmed@amazon.de \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pbonzini@redhat.com \
--cc=rimasluk@amazon.com \
--cc=rkagan@virtuozzo.com \
--cc=rkrcmar@redhat.com \
--cc=vkuznets@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).