linux-kernel - Re: [PATCH] KVM: x86: Use fast path for Xen timer delivery

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] [day] [month] [year] [list]

Date:   Wed, 27 Sep 2023 08:42:40 +0200
From:   David Woodhouse <dwmw2@...radead.org>
To:     kvm <kvm@...r.kernel.org>
Cc:     Paul Durrant <paul@....org>,
        Sean Christopherson <seanjc@...gle.com>,
        Paolo Bonzini <pbonzini@...hat.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        Dave Hansen <dave.hansen@...ux.intel.com>, x86@...nel.org,
        "H. Peter Anvin" <hpa@...or.com>, linux-kernel@...r.kernel.org,
        "Griffoul, Fred" <fgriffo@...zon.com>
Subject: Re: [PATCH] KVM: x86: Use fast path for Xen timer delivery

On Fri, 2023-09-22 at 10:20 +0100, David Woodhouse wrote:
> From: David Woodhouse <dwmw@...zon.co.uk>
> 
> Most of the time there's no need to kick the vCPU and deliver the timer
> event through kvm_xen_inject_timer_irqs(). Use kvm_xen_set_evtchn_fast()
> directly from the timer callback, and only fall back to the slow path
> when it's necessary to do so.
> 
> This gives a significant improvement in timer latency testing (using
> nanosleep() for various periods and then measuring the actual time
> elapsed).
> 
> Signed-off-by: David Woodhouse <dwmw@...zon.co.uk>

Hm, scratch that. It brings back the race condition Paolo talks about in
https://lore.kernel.org/kvm/0709ac62f664c0f3123fcdeabed3b79038cef3b6.camel@infradead.org/T/

Well, not *quite*... there's no race with clearing timer_expires
because I forgot to clear it at all :)

I think I'll be able to fix it up with an hrtimer_cancel() in the rare
ioctl attr_get path to avoid the race. 

Download attachment "smime.p7s" of type "application/pkcs7-signature" (5965 bytes)