From: "Radim Krčmář" <rkrcmar@redhat.com>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: linux-kernel@vger.kernel.org, kvm@vger.kernel.org, "Lan,
Tianyu" <tianyu.lan@intel.com>,
Igor Mammedov <imammedo@redhat.com>,
Jan Kiszka <jan.kiszka@web.de>, Peter Xu <peterx@redhat.com>,
Yang Zhang <yang.zhang.wz@gmail.com>
Subject: Re: [PATCH v3 12/14] KVM: x86: add a flag to disable KVM x2apic broadcast quirk
Date: Wed, 13 Jul 2016 17:14:13 +0200 [thread overview]
Message-ID: <20160713151413.GD22438@potion> (raw)
In-Reply-To: <4b30aa90-24bb-c5d0-18c0-b1fb4a62014f@redhat.com>
2016-07-13 10:38+0200, Paolo Bonzini:
> On 12/07/2016 22:09, Radim Krčmář wrote:
>> @@ -619,14 +619,17 @@ static bool kvm_apic_match_logical_addr(struct kvm_lapic *apic, u32 mda)
>> /* KVM APIC implementation has two quirks
>> * - dest always begins at 0 while xAPIC MDA has offset 24,
>> * - IOxAPIC messages have to be delivered (directly) to x2APIC.
>> + *
>> + * The broadcast quirk can be disabled with KVM_CAP_X2APIC_API.
>
> --verbose version:
>
> /* KVM APIC implementation has two quirks
> - * - dest always begins at 0 while xAPIC MDA has offset 24,
> - * - IOxAPIC messages have to be delivered (directly) to x2APIC.
> + * - dest always begins at 0 while xAPIC MDA has offset 24. This is
> + * just a quirk in the API and is not problematic.
> *
> - * The broadcast quirk can be disabled with KVM_CAP_X2APIC_API.
> + * - in-kernel IOAPIC messages have to be delivered directly to
> + * x2APIC, because the kernel does not support interrupt remapping.
> + * In order to support broadcast without interrupt remapping, x2APIC
> + * rewrites the destination of non-IPI messages (and also of IPIs sent
> + * from xAPIC-mode LAPICs) from APIC_BROADCAST to X2APIC_BROADCAST.
KVM doesn't do the operation in parentheses since 394457a928e0 ("KVM:
x86: some apic broadcast modes does not work").
Following patches added only !ipi case (I already regret posting them).
> + *
> + * The broadcast quirk can be disabled with KVM_CAP_X2APIC_API. This is
> + * important when userspace wants to use x2APIC-format MSIs, because
> + * APIC_BROADCAST (0xff) is a legal route for "cluster 0, CPUs 0-7".
> */
>
> Sounds good?
Except that one bit. (Code alone really can't explain the what happens
here.)
next prev parent reply other threads:[~2016-07-13 15:14 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-07-12 20:09 [PATCH v3 00/14] KVM: x86: break the xAPIC barrier Radim Krčmář
2016-07-12 20:09 ` [PATCH v3 01/14] KVM: x86: bump KVM_SOFT_MAX_VCPUS to 240 Radim Krčmář
2016-07-12 20:09 ` [PATCH v3 02/14] KVM: x86: add kvm_apic_map_get_dest_lapic Radim Krčmář
2016-07-12 20:09 ` [PATCH v3 03/14] KVM: x86: use physical LAPIC array for logical x2APIC Radim Krčmář
2016-08-02 13:46 ` Wanpeng Li
2016-08-02 14:22 ` Radim Krčmář
2016-08-02 16:40 ` Nadav Amit
2016-07-12 20:09 ` [PATCH v3 04/14] KVM: x86: dynamic kvm_apic_map Radim Krčmář
2016-07-13 8:29 ` Paolo Bonzini
2016-07-13 14:37 ` Radim Krčmář
2016-08-02 11:39 ` Wanpeng Li
2016-08-02 12:22 ` Radim Krčmář
2016-08-02 13:15 ` Wanpeng Li
2016-07-12 20:09 ` [PATCH v3 05/14] KVM: x86: use generic function for MSI parsing Radim Krčmář
2016-07-12 20:09 ` [PATCH v3 06/14] KVM: x86: use hardware-compatible format for APIC ID register Radim Krčmář
2016-07-12 20:09 ` [PATCH v3 07/14] KVM: x86: reset APIC ID when enabling LAPIC Radim Krčmář
2016-08-02 13:26 ` Wanpeng Li
2016-08-02 13:52 ` Radim Krčmář
2016-07-12 20:09 ` [PATCH v3 08/14] KVM: VMX: optimize APIC ID read with APICv Radim Krčmář
2016-07-12 20:09 ` [PATCH v3 09/14] KVM: x86: reset lapic base in kvm_lapic_reset Radim Krčmář
2016-07-12 20:09 ` [PATCH v3 10/14] KVM: pass struct kvm to kvm_set_routing_entry Radim Krčmář
2016-07-12 20:09 ` [PATCH v3 11/14] KVM: x86: add KVM_CAP_X2APIC_API Radim Krčmář
2016-07-13 8:41 ` Paolo Bonzini
2016-07-13 14:40 ` Radim Krčmář
2016-07-12 20:09 ` [PATCH v3 12/14] KVM: x86: add a flag to disable KVM x2apic broadcast quirk Radim Krčmář
2016-07-13 8:38 ` Paolo Bonzini
2016-07-13 15:14 ` Radim Krčmář [this message]
2016-07-13 15:30 ` Paolo Bonzini
2016-07-13 10:15 ` Paolo Bonzini
2016-07-13 14:52 ` Radim Krčmář
2016-07-12 20:09 ` [PATCH v3 13/14] KVM: x86: bump MAX_VCPUS to 288 Radim Krčmář
2016-07-12 20:09 ` [PATCH v3 14/14] KVM: x86: bump KVM_MAX_VCPU_ID to 1023 Radim Krčmář
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=20160713151413.GD22438@potion \
--to=rkrcmar@redhat.com \
--cc=imammedo@redhat.com \
--cc=jan.kiszka@web.de \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pbonzini@redhat.com \
--cc=peterx@redhat.com \
--cc=tianyu.lan@intel.com \
--cc=yang.zhang.wz@gmail.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).