From: Avi Kivity <avi@redhat.com>
To: Alexander Graf <agraf-l3A5Bk7waGM@public.gmane.org>
Cc: kvm-ppc-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
KVM list <kvm-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
linuxppc-dev
<linuxppc-dev-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org>
Subject: Re: [PATCH 4/7] KVM: PPC: Add book3s_32 tlbie flush acceleration
Date: Sun, 01 Aug 2010 14:08:49 +0000 [thread overview]
Message-ID: <4C557FF1.4000400@redhat.com> (raw)
In-Reply-To: <1280408662-10328-5-git-send-email-agraf-l3A5Bk7waGM@public.gmane.org>
On 07/29/2010 04:04 PM, Alexander Graf wrote:
> On Book3s_32 the tlbie instruction flushed effective addresses by the mask
> 0x0ffff000. This is pretty hard to reflect with a hash that hashes ~0xfff, so
> to speed up that target we should also keep a special hash around for it.
>
>
> static inline u64 kvmppc_mmu_hash_vpte(u64 vpage)
> {
> return hash_64(vpage& 0xfffffffffULL, HPTEG_HASH_BITS_VPTE);
> @@ -66,6 +72,11 @@ void kvmppc_mmu_hpte_cache_map(struct kvm_vcpu *vcpu, struct hpte_cache *pte)
> index = kvmppc_mmu_hash_pte(pte->pte.eaddr);
> hlist_add_head_rcu(&pte->list_pte,&vcpu->arch.hpte_hash_pte[index]);
>
> + /* Add to ePTE_long list */
> + index = kvmppc_mmu_hash_pte_long(pte->pte.eaddr);
> + hlist_add_head_rcu(&pte->list_pte_long,
> + &vcpu->arch.hpte_hash_pte_long[index]);
> +
Isn't it better to make operations on this list conditional on
Book3s_32? Hashes are expensive since they usually cost cache misses.
Can of course be done later as an optimization.
--
error compiling committee.c: too many arguments to function
WARNING: multiple messages have this Message-ID (diff)
From: Avi Kivity <avi@redhat.com>
To: Alexander Graf <agraf@suse.de>
Cc: linuxppc-dev <linuxppc-dev@lists.ozlabs.org>,
KVM list <kvm@vger.kernel.org>,
kvm-ppc@vger.kernel.org
Subject: Re: [PATCH 4/7] KVM: PPC: Add book3s_32 tlbie flush acceleration
Date: Sun, 01 Aug 2010 17:08:49 +0300 [thread overview]
Message-ID: <4C557FF1.4000400@redhat.com> (raw)
In-Reply-To: <1280408662-10328-5-git-send-email-agraf@suse.de>
On 07/29/2010 04:04 PM, Alexander Graf wrote:
> On Book3s_32 the tlbie instruction flushed effective addresses by the mask
> 0x0ffff000. This is pretty hard to reflect with a hash that hashes ~0xfff, so
> to speed up that target we should also keep a special hash around for it.
>
>
> static inline u64 kvmppc_mmu_hash_vpte(u64 vpage)
> {
> return hash_64(vpage& 0xfffffffffULL, HPTEG_HASH_BITS_VPTE);
> @@ -66,6 +72,11 @@ void kvmppc_mmu_hpte_cache_map(struct kvm_vcpu *vcpu, struct hpte_cache *pte)
> index = kvmppc_mmu_hash_pte(pte->pte.eaddr);
> hlist_add_head_rcu(&pte->list_pte,&vcpu->arch.hpte_hash_pte[index]);
>
> + /* Add to ePTE_long list */
> + index = kvmppc_mmu_hash_pte_long(pte->pte.eaddr);
> + hlist_add_head_rcu(&pte->list_pte_long,
> + &vcpu->arch.hpte_hash_pte_long[index]);
> +
Isn't it better to make operations on this list conditional on
Book3s_32? Hashes are expensive since they usually cost cache misses.
Can of course be done later as an optimization.
--
error compiling committee.c: too many arguments to function
WARNING: multiple messages have this Message-ID (diff)
From: Avi Kivity <avi-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
To: Alexander Graf <agraf-l3A5Bk7waGM@public.gmane.org>
Cc: kvm-ppc-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
KVM list <kvm-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
linuxppc-dev
<linuxppc-dev-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org>
Subject: Re: [PATCH 4/7] KVM: PPC: Add book3s_32 tlbie flush acceleration
Date: Sun, 01 Aug 2010 17:08:49 +0300 [thread overview]
Message-ID: <4C557FF1.4000400@redhat.com> (raw)
In-Reply-To: <1280408662-10328-5-git-send-email-agraf-l3A5Bk7waGM@public.gmane.org>
On 07/29/2010 04:04 PM, Alexander Graf wrote:
> On Book3s_32 the tlbie instruction flushed effective addresses by the mask
> 0x0ffff000. This is pretty hard to reflect with a hash that hashes ~0xfff, so
> to speed up that target we should also keep a special hash around for it.
>
>
> static inline u64 kvmppc_mmu_hash_vpte(u64 vpage)
> {
> return hash_64(vpage& 0xfffffffffULL, HPTEG_HASH_BITS_VPTE);
> @@ -66,6 +72,11 @@ void kvmppc_mmu_hpte_cache_map(struct kvm_vcpu *vcpu, struct hpte_cache *pte)
> index = kvmppc_mmu_hash_pte(pte->pte.eaddr);
> hlist_add_head_rcu(&pte->list_pte,&vcpu->arch.hpte_hash_pte[index]);
>
> + /* Add to ePTE_long list */
> + index = kvmppc_mmu_hash_pte_long(pte->pte.eaddr);
> + hlist_add_head_rcu(&pte->list_pte_long,
> + &vcpu->arch.hpte_hash_pte_long[index]);
> +
Isn't it better to make operations on this list conditional on
Book3s_32? Hashes are expensive since they usually cost cache misses.
Can of course be done later as an optimization.
--
error compiling committee.c: too many arguments to function
next prev parent reply other threads:[~2010-08-01 14:08 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-07-29 13:04 [PATCH 0/7] Rest of my KVM-PPC patch queue Alexander Graf
2010-07-29 13:04 ` Alexander Graf
2010-07-29 13:04 ` Alexander Graf
[not found] ` <1280408662-10328-1-git-send-email-agraf-l3A5Bk7waGM@public.gmane.org>
2010-07-29 13:04 ` [PATCH 1/7] KVM: PPC: Book3S_32 MMU debug compile fixes Alexander Graf
2010-07-29 13:04 ` Alexander Graf
2010-07-29 13:04 ` Alexander Graf
2010-07-29 13:04 ` [PATCH 2/7] KVM: PPC: RCU'ify the Book3s MMU Alexander Graf
2010-07-29 13:04 ` Alexander Graf
2010-07-29 13:04 ` Alexander Graf
2010-07-29 13:04 ` [PATCH 4/7] KVM: PPC: Add book3s_32 tlbie flush acceleration Alexander Graf
2010-07-29 13:04 ` Alexander Graf
2010-07-29 13:04 ` Alexander Graf
[not found] ` <1280408662-10328-5-git-send-email-agraf-l3A5Bk7waGM@public.gmane.org>
2010-08-01 14:08 ` Avi Kivity [this message]
2010-08-01 14:08 ` Avi Kivity
2010-08-01 14:08 ` Avi Kivity
2010-08-01 20:20 ` Alexander Graf
2010-08-01 20:20 ` Alexander Graf
2010-08-01 20:20 ` Alexander Graf
2010-07-29 13:04 ` [PATCH 6/7] KVM: PPC: Make long relocations be ulong Alexander Graf
2010-07-29 13:04 ` Alexander Graf
2010-07-29 13:04 ` Alexander Graf
2010-07-29 13:04 ` [PATCH 3/7] KVM: PPC: correctly check gfn_to_pfn() return value Alexander Graf
2010-07-29 13:04 ` Alexander Graf
2010-07-29 13:04 ` Alexander Graf
2010-07-29 13:04 ` [PATCH 5/7] KVM: PPC: Use MSR_DR for external load_up Alexander Graf
2010-07-29 13:04 ` Alexander Graf
2010-07-29 13:04 ` Alexander Graf
2010-07-29 13:04 ` [PATCH 7/7] KVM: PPC: Move KVM trampolines before __end_interrupts Alexander Graf
2010-07-29 13:04 ` Alexander Graf
2010-07-29 13:04 ` Alexander Graf
2010-08-01 14:09 ` [PATCH 0/7] Rest of my KVM-PPC patch queue Avi Kivity
2010-08-01 14:09 ` Avi Kivity
2010-08-01 14:09 ` 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=4C557FF1.4000400@redhat.com \
--to=avi@redhat.com \
--cc=agraf-l3A5Bk7waGM@public.gmane.org \
--cc=kvm-ppc-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=kvm-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linuxppc-dev-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.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.