All of lore.kernel.org
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: Christian Borntraeger <borntraeger@de.ibm.com>,
	Christoffer Dall <christoffer.dall@linaro.org>,
	kvmarm@lists.cs.columbia.edu, kvm@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org
Cc: Marc Zyngier <marc.zyngier@arm.com>
Subject: Re: [PATCH v2] arm/arm64: KVM: Properly account for guest CPU time
Date: Mon, 01 Jun 2015 13:52:18 +0200	[thread overview]
Message-ID: <556C4772.1070506@redhat.com> (raw)
In-Reply-To: <556C4510.80704@de.ibm.com>



On 01/06/2015 13:42, Christian Borntraeger wrote:
> Am 01.06.2015 um 13:34 schrieb Paolo Bonzini:
>>
>>
>> On 01/06/2015 09:47, Christian Borntraeger wrote:
>>>
>>> 1: "disable", "guest", "disable again and save", "restore to disable", "enable"
>>> and now it is
>>> 2: "disable", "guest", "enable"
>>> and with your patch it is
>>> 3: "disable", "guest", "enable", "disable, "enable"
>>>
>>> I assume that 3 and 1 are similar in its costs, so this is probably ok.
>>
>> At least on x86, 3 and 2 are similar, but 3 is much more expensive than
>> 1!  See https://lkml.org/lkml/2015/5/5/835:
> 
> That does not make sense. If 3 and 2 are similar, then 2 must be much more
> expensive than 1 as well. As 2 is a strict subset of 1 it must be cheaper, no?

Yes, it must.  I meant 3 is much cheaper than 1.

Paolo

>> Cost of: CLI                         insn  same-IF :     0 cycles
>> Cost of: CLI                         insn  flip-IF :     0 cycles
>> Cost of: STI                         insn  same-IF :     0 cycles
>> Cost of: STI                         insn  flip-IF :     0 cycles
>> Cost of: PUSHF                       insn          :     0 cycles
>> Cost of: POPF                        insn  same-IF :    20 cycles
>> Cost of: POPF                        insn  flip-IF :    28 cycles
>> Cost of: local_irq_save()            fn            :    20 cycles
>> Cost of: local_irq_restore()         fn    same-IF :    24 cycles
>> Cost of: local_irq_restore()         fn    flip-IF :    28 cycles
>> Cost of: irq_save()+restore()        fn    same-IF :    48 cycles
>> Cost of: irq_save()+restore()        fn    flip-IF :    48 cycles
> 
> Yes its similar on s390. local_irq_save/restore is noticable in guest exit
> hot loops (thats what inspired my patch), but a simple irq disable is
> just single cycle pipelined. Given the design of aggressive out-out order
> designs with all the architectural ordering this makes sense.
> 
> Christian
> 

WARNING: multiple messages have this Message-ID (diff)
From: pbonzini@redhat.com (Paolo Bonzini)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v2] arm/arm64: KVM: Properly account for guest CPU time
Date: Mon, 01 Jun 2015 13:52:18 +0200	[thread overview]
Message-ID: <556C4772.1070506@redhat.com> (raw)
In-Reply-To: <556C4510.80704@de.ibm.com>



On 01/06/2015 13:42, Christian Borntraeger wrote:
> Am 01.06.2015 um 13:34 schrieb Paolo Bonzini:
>>
>>
>> On 01/06/2015 09:47, Christian Borntraeger wrote:
>>>
>>> 1: "disable", "guest", "disable again and save", "restore to disable", "enable"
>>> and now it is
>>> 2: "disable", "guest", "enable"
>>> and with your patch it is
>>> 3: "disable", "guest", "enable", "disable, "enable"
>>>
>>> I assume that 3 and 1 are similar in its costs, so this is probably ok.
>>
>> At least on x86, 3 and 2 are similar, but 3 is much more expensive than
>> 1!  See https://lkml.org/lkml/2015/5/5/835:
> 
> That does not make sense. If 3 and 2 are similar, then 2 must be much more
> expensive than 1 as well. As 2 is a strict subset of 1 it must be cheaper, no?

Yes, it must.  I meant 3 is much cheaper than 1.

Paolo

>> Cost of: CLI                         insn  same-IF :     0 cycles
>> Cost of: CLI                         insn  flip-IF :     0 cycles
>> Cost of: STI                         insn  same-IF :     0 cycles
>> Cost of: STI                         insn  flip-IF :     0 cycles
>> Cost of: PUSHF                       insn          :     0 cycles
>> Cost of: POPF                        insn  same-IF :    20 cycles
>> Cost of: POPF                        insn  flip-IF :    28 cycles
>> Cost of: local_irq_save()            fn            :    20 cycles
>> Cost of: local_irq_restore()         fn    same-IF :    24 cycles
>> Cost of: local_irq_restore()         fn    flip-IF :    28 cycles
>> Cost of: irq_save()+restore()        fn    same-IF :    48 cycles
>> Cost of: irq_save()+restore()        fn    flip-IF :    48 cycles
> 
> Yes its similar on s390. local_irq_save/restore is noticable in guest exit
> hot loops (thats what inspired my patch), but a simple irq disable is
> just single cycle pipelined. Given the design of aggressive out-out order
> designs with all the architectural ordering this makes sense.
> 
> Christian
> 

  reply	other threads:[~2015-06-01 11:52 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-05-28 18:49 [PATCH v2] arm/arm64: KVM: Properly account for guest CPU time Christoffer Dall
2015-05-28 18:49 ` Christoffer Dall
2015-05-29 22:34 ` Mario Smarduch
2015-05-29 22:34   ` Mario Smarduch
2015-05-31  6:59   ` Christoffer Dall
2015-05-31  6:59     ` Christoffer Dall
2015-06-01 15:48     ` Mario Smarduch
2015-06-01 15:48       ` Mario Smarduch
2015-06-02  9:27       ` Christoffer Dall
2015-06-02  9:27         ` Christoffer Dall
2015-06-02 11:55         ` Christoffer Dall
2015-06-02 11:55           ` Christoffer Dall
2015-06-05 12:24         ` Mario Smarduch
2015-06-05 12:24           ` Mario Smarduch
2015-06-08 11:35           ` Christoffer Dall
2015-06-08 11:35             ` Christoffer Dall
2015-06-09 23:04             ` Mario Smarduch
2015-06-09 23:04               ` Mario Smarduch
2015-06-01  7:47 ` Christian Borntraeger
2015-06-01  7:47   ` Christian Borntraeger
2015-06-01  9:08   ` Christoffer Dall
2015-06-01  9:08     ` Christoffer Dall
2015-06-01  9:21     ` Christian Borntraeger
2015-06-01  9:21       ` Christian Borntraeger
2015-06-01 13:35       ` Christoffer Dall
2015-06-01 13:35         ` Christoffer Dall
2015-06-01 13:37         ` Christian Borntraeger
2015-06-01 13:37           ` Christian Borntraeger
2015-06-02  9:28           ` Christoffer Dall
2015-06-02  9:28             ` Christoffer Dall
2015-06-01 11:34   ` Paolo Bonzini
2015-06-01 11:34     ` Paolo Bonzini
2015-06-01 11:42     ` Christian Borntraeger
2015-06-01 11:42       ` Christian Borntraeger
2015-06-01 11:52       ` Paolo Bonzini [this message]
2015-06-01 11:52         ` Paolo Bonzini
2015-06-08 17:50 ` Marc Zyngier
2015-06-08 17:50   ` Marc Zyngier
2015-06-09 14:43   ` Christoffer Dall
2015-06-09 14:43     ` Christoffer Dall
2015-06-09 16:39     ` Marc Zyngier
2015-06-09 16:39       ` Marc Zyngier

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=556C4772.1070506@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=borntraeger@de.ibm.com \
    --cc=christoffer.dall@linaro.org \
    --cc=kvm@vger.kernel.org \
    --cc=kvmarm@lists.cs.columbia.edu \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=marc.zyngier@arm.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.