From: Wanpeng Li <kernellwp@gmail.com>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krcmar <rkrcmar@redhat.com>,
LKML <linux-kernel@vger.kernel.org>, kvm <kvm@vger.kernel.org>,
Vitaly Kuznetsov <vkuznets@redhat.com>
Subject: Re: [PATCH v3 2/6] KVM: X86: Implement PV IPIs in linux guest
Date: Fri, 20 Jul 2018 13:58:49 +0800 [thread overview]
Message-ID: <CANRm+CxFy_R9VpxhBDuJNwZxp6Sh6wKkemaosDJ436cESk5s8A@mail.gmail.com> (raw)
In-Reply-To: <1357a0b5-0cd8-8cb5-6c61-c9662219bed0@redhat.com>
On Fri, 20 Jul 2018 at 00:47, Paolo Bonzini <pbonzini@redhat.com> wrote:
>
> On 19/07/2018 18:28, Radim Krčmář wrote:
> >> +
> >> + kvm_hypercall3(KVM_HC_SEND_IPI, ipi_bitmap_low, ipi_bitmap_high, vector);
> > and
> >
> > kvm_hypercall3(KVM_HC_SEND_IPI, ipi_bitmap[0], ipi_bitmap[1], vector);
> >
> > Still, the main problem is that we can only address 128 APICs.
> >
> > A simple improvement would reuse the vector field (as we need only 8
> > bits) and put a 'offset' in the rest. The offset would say which
> > cluster of 128 are we addressing. 24 bits of offset results in 2^31
> > total addressable CPUs (we probably should even use that many bits).
> > The downside of this is that we can only address 128 at a time.
> >
> > It's basically the same as x2apic cluster mode, only with 128 cluster
> > size instead of 16, so the code should be a straightforward port.
> > And because x2apic code doesn't seem to use any division by the cluster
> > size, we could even try to use kvm_hypercall4, add ipi_bitmap[2], and
> > make the cluster size 192. :)
>
> I did suggest an offset earlier in the discussion.
>
> The main problem is that consecutive CPU ids do not map to consecutive
> APIC ids. But still, we could do an hypercall whenever the total range
> exceeds 64. Something like
>
> u64 ipi_bitmap = 0;
> for_each_cpu(cpu, mask)
> if (!ipi_bitmap) {
> min = max = cpu;
> } else if (cpu < min && max - cpu < 64) {
> ipi_bitmap <<= min - cpu;
> min = cpu;
> } else if (id < min + 64) {
> max = cpu < max ? max : cpu;
> } else {
> /* ... send hypercall... */
> min = max = cpu;
> ipi_bitmap = 0;
> }
> __set_bit(ipi_bitmap, cpu - min);
> }
> if (ipi_bitmap) {
> /* ... send hypercall... */
> }
>
> We could keep the cluster size of 128, but it would be more complicated
> to do the left shift in the first "else if". If the limit is 64, you
> can keep the two arguments in the hypercall, and just pass 0 as the
> "high" bitmap on 64-bit kernels.
As David pointed out, we need to scale to higher APIC IDs. I will add
the cpu id to apic id transfer in the for loop. How about:
kvm_hypercall2(KVM_HC_SEND_IPI, ipi_bitmap, vector); directly. In
addition, why need to pass the 0 as the "high" bitmap even if for 128
vCPUs case?
Regards,
Wanpeng Li
next prev parent reply other threads:[~2018-07-20 6:00 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-07-03 6:21 [PATCH v3 0/6] KVM: X86: Implement PV IPIs support Wanpeng Li
2018-07-03 6:21 ` [PATCH v3 1/6] KVM: X86: Add kvm hypervisor init time platform setup callback Wanpeng Li
2018-07-03 6:21 ` [PATCH v3 2/6] KVM: X86: Implement PV IPIs in linux guest Wanpeng Li
2018-07-19 16:28 ` Radim Krčmář
2018-07-19 16:47 ` Paolo Bonzini
2018-07-19 17:22 ` Radim Krčmář
2018-07-20 3:35 ` Wanpeng Li
2018-07-20 5:58 ` Wanpeng Li [this message]
2018-07-20 8:06 ` Paolo Bonzini
2018-07-20 3:33 ` Wanpeng Li
2018-07-20 9:51 ` Radim Krcmar
2018-07-20 10:17 ` Wanpeng Li
2018-07-19 23:05 ` David Matlack
2018-07-20 3:45 ` Wanpeng Li
2018-07-20 13:12 ` Radim Krcmar
2018-07-03 6:21 ` [PATCH v3 3/6] KVM: X86: Fallback to original apic hooks when bad happens Wanpeng Li
2018-07-03 6:21 ` [PATCH v3 4/6] KVM: X86: Implement PV IPIs send hypercall Wanpeng Li
2018-07-19 16:47 ` Paolo Bonzini
2018-07-20 3:49 ` Wanpeng Li
2018-07-03 6:21 ` [PATCH v3 5/6] KVM: X86: Add NMI support to PV IPIs Wanpeng Li
2018-07-19 16:31 ` Radim Krčmář
2018-07-20 3:53 ` Wanpeng Li
2018-07-20 8:04 ` Paolo Bonzini
2018-07-20 13:26 ` Radim Krcmar
2018-07-23 0:52 ` Wanpeng Li
2018-07-03 6:21 ` [PATCH v3 6/6] KVM: X86: Expose PV_SEND_IPI CPUID feature bit to guest Wanpeng Li
2018-07-18 3:00 ` [PATCH v3 0/6] KVM: X86: Implement PV IPIs support Wanpeng Li
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=CANRm+CxFy_R9VpxhBDuJNwZxp6Sh6wKkemaosDJ436cESk5s8A@mail.gmail.com \
--to=kernellwp@gmail.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pbonzini@redhat.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).