From: Sean Christopherson <seanjc@google.com>
To: Peter Xu <peterx@redhat.com>
Cc: linux-kernel@vger.kernel.org, kvm@vger.kernel.org,
Maxim Levitsky <mlevitsk@redhat.com>,
Vitaly Kuznetsov <vkuznets@redhat.com>,
Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: [PATCH v2 9/9] KVM: X86: Optimize zapping rmap
Date: Wed, 28 Jul 2021 22:31:51 +0000 [thread overview]
Message-ID: <YQHa1xuNKhqRr4Fq@google.com> (raw)
In-Reply-To: <YQHTocEdMzsJQuzL@t490s>
On Wed, Jul 28, 2021, Peter Xu wrote:
> On Wed, Jul 28, 2021 at 09:39:02PM +0000, Sean Christopherson wrote:
> > On Fri, Jun 25, 2021, Peter Xu wrote:
> > Why implement this as a generic method with a callback? gcc is suprisingly
> > astute in optimizing callback(), but I don't see the point of adding a complex
> > helper that has a single caller, and is extremely unlikely to gain new callers.
> > Or is there another "zap everything" case I'm missing?
>
> No other case; it's just that pte_list_*() helpers will be more self-contained.
Eh, but this flow is as much about rmaps as it is about pte_list.
> If that'll be a performance concern, no objection to hard code it.
It's more about unnecessary complexity than it is about performance, e.g. gcc-10
generates identical code for both version (which did surprise the heck out of me).
If we really want to isolate pte_list_destroy(), I would vote for something like
this (squashed in). pte_list_remove() already calls mmu_spte_clear_track_bits(),
so that particular separation of concerns has already gone out the window.
-/* Return true if rmap existed and callback called, false otherwise */
-static bool pte_list_destroy(struct kvm_rmap_head *rmap_head,
- void (*callback)(u64 *sptep))
+static bool pte_list_destroy(struct kvm_rmap_head *rmap_head)
{
struct pte_list_desc *desc, *next;
int i;
@@ -1013,20 +1011,16 @@ static bool pte_list_destroy(struct kvm_rmap_head *rmap_head,
return false;
if (!(rmap_head->val & 1)) {
- if (callback)
- callback((u64 *)rmap_head->val);
+ mmu_spte_clear_track_bits((u64 *)rmap_head->val);
goto out;
}
desc = (struct pte_list_desc *)(rmap_head->val & ~1ul);
-
- while (desc) {
- if (callback)
- for (i = 0; i < desc->spte_count; i++)
- callback(desc->sptes[i]);
+ for ( ; desc; desc = next) {
+ for (i = 0; i < desc->spte_count; i++)
+ mmu_spte_clear_track_bits(desc->sptes[i]);
next = desc->more;
mmu_free_pte_list_desc(desc);
- desc = next;
}
out:
/* rmap_head is meaningless now, remember to reset it */
@@ -1422,22 +1416,17 @@ static bool rmap_write_protect(struct kvm_vcpu *vcpu, u64 gfn)
return kvm_mmu_slot_gfn_write_protect(vcpu->kvm, slot, gfn, PG_LEVEL_4K);
}
-static void mmu_spte_clear_track_bits_cb(u64 *sptep)
-{
- mmu_spte_clear_track_bits(sptep);
-}
-
static bool kvm_zap_rmapp(struct kvm *kvm, struct kvm_rmap_head *rmap_head,
const struct kvm_memory_slot *slot)
{
- return pte_list_destroy(rmap_head, mmu_spte_clear_track_bits_cb);
+ return pte_list_destroy(rmap_head);
}
static bool kvm_unmap_rmapp(struct kvm *kvm, struct kvm_rmap_head *rmap_head,
struct kvm_memory_slot *slot, gfn_t gfn, int level,
pte_t unused)
{
- return kvm_zap_rmapp(kvm, rmap_head, slot);
+ return pte_list_destroy(rmap_head);
}
static bool kvm_set_pte_rmapp(struct kvm *kvm, struct kvm_rmap_head *rmap_head,
next prev parent reply other threads:[~2021-07-28 22:31 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-06-25 15:32 [PATCH v2 0/9] KVM: X86: Some light optimizations on rmap logic Peter Xu
2021-06-25 15:32 ` [PATCH v2 1/9] KVM: X86: Add per-vm stat for max rmap list size Peter Xu
2021-06-25 15:32 ` [PATCH v2 2/9] KVM: Introduce kvm_get_kvm_safe() Peter Xu
2021-07-26 13:42 ` Paolo Bonzini
2021-06-25 15:32 ` [PATCH v2 3/9] KVM: Allow to have arch-specific per-vm debugfs files Peter Xu
2021-06-25 15:32 ` [PATCH v2 4/9] KVM: X86: Introduce pte_list_count() helper Peter Xu
2021-06-25 15:32 ` [PATCH v2 5/9] KVM: X86: Introduce kvm_mmu_slot_lpages() helpers Peter Xu
2021-06-25 15:32 ` [PATCH v2 6/9] KVM: X86: Introduce mmu_rmaps_stat per-vm debugfs file Peter Xu
2021-06-25 15:34 ` [PATCH v2 7/9] KVM: X86: MMU: Tune PTE_LIST_EXT to be bigger Peter Xu
2021-07-28 21:01 ` Sean Christopherson
2021-06-25 15:34 ` [PATCH v2 8/9] KVM: X86: Optimize pte_list_desc with per-array counter Peter Xu
2021-07-28 21:04 ` Sean Christopherson
2021-07-28 21:51 ` Peter Xu
2021-07-29 9:33 ` Paolo Bonzini
2021-07-29 15:53 ` Peter Xu
2021-07-30 15:45 ` Peter Xu
2021-06-25 15:34 ` [PATCH v2 9/9] KVM: X86: Optimize zapping rmap Peter Xu
2021-07-28 21:39 ` Sean Christopherson
2021-07-28 22:01 ` Peter Xu
2021-07-28 22:31 ` Sean Christopherson [this message]
2021-07-29 9:35 ` Paolo Bonzini
2021-07-26 13:05 ` [PATCH v2 0/9] KVM: X86: Some light optimizations on rmap logic Paolo Bonzini
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=YQHa1xuNKhqRr4Fq@google.com \
--to=seanjc@google.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mlevitsk@redhat.com \
--cc=pbonzini@redhat.com \
--cc=peterx@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).