linux-kernel - Re: [PATCH v2 8/9] KVM: X86: Optimize pte_list

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <YQHGXhOc5gO9aYsL@google.com>
Date:   Wed, 28 Jul 2021 21:04:30 +0000
From:   Sean Christopherson <seanjc@...gle.com>
To:     Peter Xu <peterx@...hat.com>
Cc:     linux-kernel@...r.kernel.org, kvm@...r.kernel.org,
        Maxim Levitsky <mlevitsk@...hat.com>,
        Vitaly Kuznetsov <vkuznets@...hat.com>,
        Paolo Bonzini <pbonzini@...hat.com>
Subject: Re: [PATCH v2 8/9] KVM: X86: Optimize pte_list_desc with per-array
 counter

On Fri, Jun 25, 2021, Peter Xu wrote:
> Add a counter field into pte_list_desc, so as to simplify the add/remove/loop
> logic.  E.g., we don't need to loop over the array any more for most reasons.
> 
> This will make more sense after we've switched the array size to be larger
> otherwise the counter will be a waste.
> 
> Initially I wanted to store a tail pointer at the head of the array list so we
> don't need to traverse the list at least for pushing new ones (if without the
> counter we traverse both the list and the array).  However that'll need
> slightly more change without a huge lot benefit, e.g., after we grow entry
> numbers per array the list traversing is not so expensive.
> 
> So let's be simple but still try to get as much benefit as we can with just
> these extra few lines of changes (not to mention the code looks easier too
> without looping over arrays).
> 
> I used the same a test case to fork 500 child and recycle them ("./rmap_fork
> 500" [1]), this patch further speeds up the total fork time of about 14%, which
> is a total of 38% of vanilla kernel:
> 
>         Vanilla:      367.20 (+-4.58%)
>         3->15 slots:  302.00 (+-5.30%)
>         Add counter:  265.20 (+-9.88%)
> 
> [1] https://github.com/xzpeter/clibs/commit/825436f825453de2ea5aaee4bdb1c92281efe5b3
> 
> Signed-off-by: Peter Xu <peterx@...hat.com>
> ---
>  arch/x86/kvm/mmu/mmu.c | 33 ++++++++++++++++++---------------
>  1 file changed, 18 insertions(+), 15 deletions(-)
> 
> diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
> index 9b093985a2ef..ba0258bdebc4 100644
> --- a/arch/x86/kvm/mmu/mmu.c
> +++ b/arch/x86/kvm/mmu/mmu.c
> @@ -138,10 +138,15 @@ module_param(dbg, bool, 0644);
>  #include <trace/events/kvm.h>
>  
>  /* make pte_list_desc fit well in cache lines */
> -#define PTE_LIST_EXT 15
> +#define PTE_LIST_EXT 14

Doh, I looked at kvm/queue code before looking at the full series.

>  struct pte_list_desc {
>  	u64 *sptes[PTE_LIST_EXT];
> +	/*
> +	 * Stores number of entries stored in the pte_list_desc.  No need to be
> +	 * u64 but just for easier alignment.  When PTE_LIST_EXT, means full.
> +	 */
> +	u64 spte_count;

Per my feedback to the previous patch, this should be above sptes[] so that rmaps
with <8 SPTEs only touch one cache line.  No idea if it actually matters in
practice, but I can't see how it would harm anything.

>  	struct pte_list_desc *more;
>  };