All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Michael S. Tsirkin" <mst@redhat.com>
To: kvm@vger.kernel.org
Cc: Avi Kivity <avi@redhat.com>,
	Marcelo Tosatti <mtosatti@redhat.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@redhat.com>, "H. Peter Anvin" <hpa@zytor.com>,
	x86@kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] kvm: optimize ISR lookups
Date: Mon, 21 May 2012 21:44:28 +0300	[thread overview]
Message-ID: <20120521184352.GA14463@redhat.com> (raw)
In-Reply-To: <20120521163727.GA13337@redhat.com>

On Mon, May 21, 2012 at 07:37:27PM +0300, Michael S. Tsirkin wrote:
> We perform ISR lookups twice: during interrupt
> injection and on EOI. Typical workloads only have
> a single bit set there. So we can avoid ISR scans by
> 1. counting bits as we set/clear them in ISR
> 2. if count is 1, caching the vector number
> 3. if count != 1, invalidating the cache
> 
> The real purpose of this is enabling PV EOI
> which needs to quickly validate the vector.
> But non PV guests also benefit.
> 
> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

BTW host to guest netperf result (which is 100% CPU bound)
seems to increase pretty drastically:

before

TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 11.0.0.104
(11.0.0.104) port 0 AF_INET
Recv   Send    Send                          Utilization       Service
Demand
Socket Socket  Message  Elapsed              Send     Recv     Send
Recv
Size   Size    Size     Time     Throughput  local    remote   local
remote
bytes  bytes   bytes    secs.    10^6bits/s  % S      % S      us/KB
us/KB

 87380  16384  16384    10.01      7905.79   11.26    99.50    2.801 1.031  

after

TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 11.0.0.104
(11.0.0.104) port 0 AF_INET
Recv   Send    Send                          Utilization       Service
Demand
Socket Socket  Message  Elapsed              Send     Recv     Send
Recv
Size   Size    Size     Time     Throughput  local    remote   local
remote
bytes  bytes   bytes    secs.    10^6bits/s  % S      % S      us/KB
us/KB

 87380  16384  16384    10.01      8826.59   11.36    99.20    2.531 0.921  

> ---
>  arch/x86/kvm/lapic.c |   51 ++++++++++++++++++++++++++++++++++++++++++++++++-
>  arch/x86/kvm/lapic.h |    2 +
>  2 files changed, 51 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
> index 93c1574..232950a 100644
> --- a/arch/x86/kvm/lapic.c
> +++ b/arch/x86/kvm/lapic.c
> @@ -107,6 +107,16 @@ static inline void apic_clear_vector(int vec, void *bitmap)
>  	clear_bit(VEC_POS(vec), (bitmap) + REG_POS(vec));
>  }
>  
> +static inline int __apic_test_and_set_vector(int vec, void *bitmap)
> +{
> +	return __test_and_set_bit(VEC_POS(vec), (bitmap) + REG_POS(vec));
> +}
> +
> +static inline int __apic_test_and_clear_vector(int vec, void *bitmap)
> +{
> +	return __test_and_clear_bit(VEC_POS(vec), (bitmap) + REG_POS(vec));
> +}
> +
>  static inline int apic_hw_enabled(struct kvm_lapic *apic)
>  {
>  	return (apic)->vcpu->arch.apic_base & MSR_IA32_APICBASE_ENABLE;
> @@ -210,6 +220,16 @@ static int find_highest_vector(void *bitmap)
>  		return fls(word[word_offset << 2]) - 1 + (word_offset << 5);
>  }
>  
> +static u8 count_vectors(void *bitmap)
> +{
> +	u32 *word = bitmap;
> +	int word_offset;
> +	u8 count = 0;
> +	for (word_offset = 0; word_offset < MAX_APIC_VECTOR >> 5; ++word_offset)
> +		count += hweight32(word[word_offset << 2]);
> +	return count;
> +}
> +
>  static inline int apic_test_and_set_irr(int vec, struct kvm_lapic *apic)
>  {
>  	apic->irr_pending = true;
> @@ -242,6 +262,25 @@ static inline void apic_clear_irr(int vec, struct kvm_lapic *apic)
>  		apic->irr_pending = true;
>  }
>  
> +static inline void apic_set_isr(int vec, struct kvm_lapic *apic)
> +{
> +	if (!__apic_test_and_set_vector(vec, apic->regs + APIC_ISR))
> +		++apic->isr_count;
> +	ASSERT(apic->isr_count > MAX_APIC_VECTOR);
> +	if (likely(apic->isr_count == 1))
> +		apic->isr_cache = vec;
> +	else
> +		apic->isr_cache = -1;
> +}
> +
> +static inline void apic_clear_isr(int vec, struct kvm_lapic *apic)
> +{
> +	if (__apic_test_and_clear_vector(vec, apic->regs + APIC_ISR))
> +		--apic->isr_count;
> +	ASSERT(apic->isr_count < 0);
> +	apic->isr_cache = -1;
> +}
> +
>  int kvm_lapic_find_highest_irr(struct kvm_vcpu *vcpu)
>  {
>  	struct kvm_lapic *apic = vcpu->arch.apic;
> @@ -273,6 +312,10 @@ int kvm_apic_set_irq(struct kvm_vcpu *vcpu, struct kvm_lapic_irq *irq)
>  static inline int apic_find_highest_isr(struct kvm_lapic *apic)
>  {
>  	int result;
> +	if (!apic->isr_count)
> +		return -1;
> +	if (likely(apic->isr_cache != -1))
> +		return apic->isr_cache;
>  
>  	result = find_highest_vector(apic->regs + APIC_ISR);
>  	ASSERT(result == -1 || result >= 16);
> @@ -492,7 +535,7 @@ static void apic_set_eoi(struct kvm_lapic *apic)
>  	if (vector == -1)
>  		return;
>  
> -	apic_clear_vector(vector, apic->regs + APIC_ISR);
> +	apic_clear_isr(vector, apic);
>  	apic_update_ppr(apic);
>  
>  	if (!(apic_get_reg(apic, APIC_SPIV) & APIC_SPIV_DIRECTED_EOI) &&
> @@ -1081,6 +1124,8 @@ void kvm_lapic_reset(struct kvm_vcpu *vcpu)
>  		apic_set_reg(apic, APIC_TMR + 0x10 * i, 0);
>  	}
>  	apic->irr_pending = false;
> +	apic->isr_count = 0;
> +	apic->isr_cache = -1;
>  	update_divide_count(apic);
>  	atomic_set(&apic->lapic_timer.pending, 0);
>  	if (kvm_vcpu_is_bsp(vcpu))
> @@ -1248,7 +1293,7 @@ int kvm_get_apic_interrupt(struct kvm_vcpu *vcpu)
>  	if (vector == -1)
>  		return -1;
>  
> -	apic_set_vector(vector, apic->regs + APIC_ISR);
> +	apic_set_isr(vector, apic);
>  	apic_update_ppr(apic);
>  	apic_clear_irr(vector, apic);
>  	return vector;
> @@ -1267,6 +1312,8 @@ void kvm_apic_post_state_restore(struct kvm_vcpu *vcpu)
>  	update_divide_count(apic);
>  	start_apic_timer(apic);
>  	apic->irr_pending = true;
> +	apic->isr_count = count_vectors(apic->regs + APIC_ISR);
> +	apic->isr_cache = -1;
>  	kvm_make_request(KVM_REQ_EVENT, vcpu);
>  }
>  
> diff --git a/arch/x86/kvm/lapic.h b/arch/x86/kvm/lapic.h
> index 6f4ce25..9f8deff 100644
> --- a/arch/x86/kvm/lapic.h
> +++ b/arch/x86/kvm/lapic.h
> @@ -13,6 +13,8 @@ struct kvm_lapic {
>  	u32 divide_count;
>  	struct kvm_vcpu *vcpu;
>  	bool irr_pending;
> +	s16 isr_count;
> +	int isr_cache;
>  	void *regs;
>  	gpa_t vapic_addr;
>  	struct page *vapic_page;
> -- 
> MST

  reply	other threads:[~2012-05-21 18:44 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-05-21 16:37 [PATCH] kvm: optimize ISR lookups Michael S. Tsirkin
2012-05-21 18:44 ` Michael S. Tsirkin [this message]
2012-05-21 21:04 ` Thomas Gleixner
2012-05-21 21:51   ` Michael S. Tsirkin
2012-05-21 22:14     ` Thomas Gleixner
2012-05-21 22:24       ` Michael S. Tsirkin
2012-05-21 22:44         ` Thomas Gleixner
2012-05-21 22:50           ` Michael S. Tsirkin
2012-05-21 23:01         ` Thomas Gleixner
2012-05-22 10:46           ` Avi Kivity
2012-05-23 14:48             ` Ingo Molnar
2012-05-23 15:03               ` Avi Kivity
2012-05-23 20:10                 ` Thomas Gleixner
2012-05-23 20:46                   ` Michael S. Tsirkin
2012-05-23 23:02                     ` Thomas Gleixner
2012-05-23 15:14               ` Michael S. Tsirkin
2012-05-23 19:22                 ` H. Peter Anvin
2012-05-21 23:11         ` Thomas Gleixner
2012-05-21 23:06   ` Michael S. Tsirkin
2012-05-21 23:12     ` H. Peter Anvin
2012-05-21 23:36     ` Thomas Gleixner
2012-05-22 10:59   ` Avi Kivity
2012-05-22 17:26     ` Thomas Gleixner
2012-05-23 15:10       ` Avi Kivity
2012-05-23 18:37         ` Thomas Gleixner
2012-05-23 19:25           ` H. Peter Anvin
2012-05-23 22:00             ` Thomas Gleixner
2012-05-30 14:18               ` Avi Kivity

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=20120521184352.GA14463@redhat.com \
    --to=mst@redhat.com \
    --cc=avi@redhat.com \
    --cc=hpa@zytor.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=mtosatti@redhat.com \
    --cc=tglx@linutronix.de \
    --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 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.