Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH v2 3/4] KVM: x86: kvm_hv_flush_tlb use inputs from XMM registers
Date:   Tue, 13 Apr 2021 11:09:41 +0200
From:   Siddharth Chandrasekaran <sidcha@amazon.de>
To:     Wei Liu <wei.liu@kernel.org>
CC:     "K. Y. Srinivasan" <kys@microsoft.com>,
        Haiyang Zhang <haiyangz@microsoft.com>,
        Stephen Hemminger <sthemmin@microsoft.com>,
        "Thomas Gleixner" <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>,
        "Borislav Petkov" <bp@alien8.de>, <x86@kernel.org>,
        "H. Peter Anvin" <hpa@zytor.com>,
        Paolo Bonzini <pbonzini@redhat.com>,
        Sean Christopherson <seanjc@google.com>,
        Vitaly Kuznetsov <vkuznets@redhat.com>,
        Wanpeng Li <wanpengli@tencent.com>,
        Jim Mattson <jmattson@google.com>,
        Joerg Roedel <joro@8bytes.org>,
        "Alexander Graf" <graf@amazon.com>,
        Evgeny Iakovlev <eyakovl@amazon.de>,
        Liran Alon <liran@amazon.com>,
        Ioannis Aslanidis <iaslan@amazon.de>,
        <linux-hyperv@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
        <kvm@vger.kernel.org>
Message-ID: <20210413090906.GA25751@u366d62d47e3651.ant.amazon.com>
References: <cover.1618244920.git.sidcha@amazon.de>
 <da036c786700032b32e68ebece06fd1a6b6bf344.1618244920.git.sidcha@amazon.de>
 <20210412201319.lnohqg6r54p6embb@liuwe-devbox-debian-v2>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Disposition: inline
In-Reply-To: <20210412201319.lnohqg6r54p6embb@liuwe-devbox-debian-v2>
User-Agent: Mutt/1.9.4 (2018-02-28)
Precedence: bulk

On Mon, Apr 12, 2021 at 08:13:19PM +0000, Wei Liu wrote:
> On Mon, Apr 12, 2021 at 07:00:16PM +0200, Siddharth Chandrasekaran wrote:
> > +
> > +static inline void kvm_hv_hypercall_read_xmm(struct kvm_hv_hcall *hc)
> 
> Do you really need inline here? The compiler should be smart enough to
> inline this function if necessary.

Removed.

> > +{
> > +     int reg;
> > +
> > +     kvm_fpu_get();
> > +     for (reg = 0; reg < KVM_HV_HYPERCALL_MAX_XMM_REGISTERS; reg++)
> > +             _kvm_read_sse_reg(reg, &hc->xmm[reg]);
> > +     kvm_fpu_put();
> > +     hc->xmm_dirty = false;
> 
> There is no code that sets xmm_dirty to true? What am I missing? I guess
> that's because you haven't implemented the hypercalls that need writing
> back to guest?

Yes, when a hypercall want to return data via XMM registers, it should
update hc->xmm[] and set hc->dirty (I plan on using this in a future
patch). The reason why I didn't differ this change to actual patch
needs it is that it pairs nicely with the read/write xmm_reg() calls in
kvm_hv_hypercall().

~ Sid.


Amazon Development Center Germany GmbH
Krausenstr. 38
10117 Berlin
Geschaeftsfuehrung: Christian Schlaeger, Jonathan Weiss
Eingetragen am Amtsgericht Charlottenburg unter HRB 149173 B
Sitz: Berlin
Ust-ID: DE 289 237 879