<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux-toradex.git/arch/arm/include/asm/kvm_hyp.h, branch v4.10</title>
<subtitle>Linux kernel for Apalis and Colibri modules</subtitle>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/'/>
<entry>
<title>arm/arm64: KVM: Perform local TLB invalidation when multiplexing vcpus on a single CPU</title>
<updated>2016-11-04T17:56:28+00:00</updated>
<author>
<name>Marc Zyngier</name>
<email>marc.zyngier@arm.com</email>
</author>
<published>2016-10-18T17:37:49+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=94d0e5980d6791b9f98a9b6c586c1f7cb76b2178'/>
<id>94d0e5980d6791b9f98a9b6c586c1f7cb76b2178</id>
<content type='text'>
Architecturally, TLBs are private to the (physical) CPU they're
associated with. But when multiple vcpus from the same VM are
being multiplexed on the same CPU, the TLBs are not private
to the vcpus (and are actually shared across the VMID).

Let's consider the following scenario:

- vcpu-0 maps PA to VA
- vcpu-1 maps PA' to VA

If run on the same physical CPU, vcpu-1 can hit TLB entries generated
by vcpu-0 accesses, and access the wrong physical page.

The solution to this is to keep a per-VM map of which vcpu ran last
on each given physical CPU, and invalidate local TLBs when switching
to a different vcpu from the same VM.

Reviewed-by: Christoffer Dall &lt;christoffer.dall@linaro.org&gt;
Signed-off-by: Marc Zyngier &lt;marc.zyngier@arm.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Architecturally, TLBs are private to the (physical) CPU they're
associated with. But when multiple vcpus from the same VM are
being multiplexed on the same CPU, the TLBs are not private
to the vcpus (and are actually shared across the VMID).

Let's consider the following scenario:

- vcpu-0 maps PA to VA
- vcpu-1 maps PA' to VA

If run on the same physical CPU, vcpu-1 can hit TLB entries generated
by vcpu-0 accesses, and access the wrong physical page.

The solution to this is to keep a per-VM map of which vcpu ran last
on each given physical CPU, and invalidate local TLBs when switching
to a different vcpu from the same VM.

Reviewed-by: Christoffer Dall &lt;christoffer.dall@linaro.org&gt;
Signed-off-by: Marc Zyngier &lt;marc.zyngier@arm.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>ARM: KVM: Support vgic-v3</title>
<updated>2016-09-22T11:22:21+00:00</updated>
<author>
<name>Vladimir Murzin</name>
<email>vladimir.murzin@arm.com</email>
</author>
<published>2016-09-12T14:49:24+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=acda5430bee4621f218391d0bcfbe4412adb3554'/>
<id>acda5430bee4621f218391d0bcfbe4412adb3554</id>
<content type='text'>
This patch allows to build and use vgic-v3 in 32-bit mode.

Unfortunately, it can not be split in several steps without extra
stubs to keep patches independent and bisectable.  For instance,
virt/kvm/arm/vgic/vgic-v3.c uses function from vgic-v3-sr.c, handling
access to GICv3 cpu interface from the guest requires vgic_v3.vgic_sre
to be already defined.

It is how support has been done:

* handle SGI requests from the guest

* report configured SRE on access to GICv3 cpu interface from the guest

* required vgic-v3 macros are provided via uapi.h

* static keys are used to select GIC backend

* to make vgic-v3 build KVM_ARM_VGIC_V3 guard is removed along with
  the static inlines

Acked-by: Marc Zyngier &lt;marc.zyngier@arm.com&gt;
Reviewed-by: Christoffer Dall &lt;christoffer.dall@linaro.org&gt;
Signed-off-by: Vladimir Murzin &lt;vladimir.murzin@arm.com&gt;
Signed-off-by: Christoffer Dall &lt;christoffer.dall@linaro.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This patch allows to build and use vgic-v3 in 32-bit mode.

Unfortunately, it can not be split in several steps without extra
stubs to keep patches independent and bisectable.  For instance,
virt/kvm/arm/vgic/vgic-v3.c uses function from vgic-v3-sr.c, handling
access to GICv3 cpu interface from the guest requires vgic_v3.vgic_sre
to be already defined.

It is how support has been done:

* handle SGI requests from the guest

* report configured SRE on access to GICv3 cpu interface from the guest

* required vgic-v3 macros are provided via uapi.h

* static keys are used to select GIC backend

* to make vgic-v3 build KVM_ARM_VGIC_V3 guard is removed along with
  the static inlines

Acked-by: Marc Zyngier &lt;marc.zyngier@arm.com&gt;
Reviewed-by: Christoffer Dall &lt;christoffer.dall@linaro.org&gt;
Signed-off-by: Vladimir Murzin &lt;vladimir.murzin@arm.com&gt;
Signed-off-by: Christoffer Dall &lt;christoffer.dall@linaro.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>ARM: Move system register accessors to asm/cp15.h</title>
<updated>2016-09-22T11:22:10+00:00</updated>
<author>
<name>Vladimir Murzin</name>
<email>vladimir.murzin@arm.com</email>
</author>
<published>2016-09-12T14:49:22+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=4f2546384150e78cad8045e59a9587fabcd9f9fe'/>
<id>4f2546384150e78cad8045e59a9587fabcd9f9fe</id>
<content type='text'>
Headers linux/irqchip/arm-gic.v3.h and arch/arm/include/asm/kvm_hyp.h
are included in virt/kvm/arm/hyp/vgic-v3-sr.c and both define macros
called __ACCESS_CP15 and __ACCESS_CP15_64 which obviously creates a
conflict. These macros were introduced independently for GIC and KVM
and, in fact, do the same thing.

As an option we could add prefixes to KVM and GIC version of macros so
they won't clash, but it'd introduce code duplication.  Alternatively,
we could keep macro in, say, GIC header and include it in KVM one (or
vice versa), but such dependency would not look nicer.

So we follow arm64 way (it handles this via sysreg.h) and move only
single set of macros to asm/cp15.h

Cc: Russell King &lt;rmk+kernel@armlinux.org.uk&gt;
Acked-by: Marc Zyngier &lt;marc.zyngier@arm.com&gt;
Signed-off-by: Vladimir Murzin &lt;vladimir.murzin@arm.com&gt;
Signed-off-by: Christoffer Dall &lt;christoffer.dall@linaro.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Headers linux/irqchip/arm-gic.v3.h and arch/arm/include/asm/kvm_hyp.h
are included in virt/kvm/arm/hyp/vgic-v3-sr.c and both define macros
called __ACCESS_CP15 and __ACCESS_CP15_64 which obviously creates a
conflict. These macros were introduced independently for GIC and KVM
and, in fact, do the same thing.

As an option we could add prefixes to KVM and GIC version of macros so
they won't clash, but it'd introduce code duplication.  Alternatively,
we could keep macro in, say, GIC header and include it in KVM one (or
vice versa), but such dependency would not look nicer.

So we follow arm64 way (it handles this via sysreg.h) and move only
single set of macros to asm/cp15.h

Cc: Russell King &lt;rmk+kernel@armlinux.org.uk&gt;
Acked-by: Marc Zyngier &lt;marc.zyngier@arm.com&gt;
Signed-off-by: Vladimir Murzin &lt;vladimir.murzin@arm.com&gt;
Signed-off-by: Christoffer Dall &lt;christoffer.dall@linaro.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>arm/arm64: Get rid of KERN_TO_HYP</title>
<updated>2016-07-03T21:41:27+00:00</updated>
<author>
<name>Marc Zyngier</name>
<email>marc.zyngier@arm.com</email>
</author>
<published>2016-06-30T17:40:51+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=6c41a413fd44af8eae2949869d4d57ce681a0c30'/>
<id>6c41a413fd44af8eae2949869d4d57ce681a0c30</id>
<content type='text'>
We have both KERN_TO_HYP and kern_hyp_va, which do the exact same
thing. Let's standardize on the latter.

Signed-off-by: Marc Zyngier &lt;marc.zyngier@arm.com&gt;
Signed-off-by: Christoffer Dall &lt;christoffer.dall@linaro.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
We have both KERN_TO_HYP and kern_hyp_va, which do the exact same
thing. Let's standardize on the latter.

Signed-off-by: Marc Zyngier &lt;marc.zyngier@arm.com&gt;
Signed-off-by: Christoffer Dall &lt;christoffer.dall@linaro.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>arm/arm64: KVM: Remove hyp_kern_va helper</title>
<updated>2016-07-03T21:41:27+00:00</updated>
<author>
<name>Marc Zyngier</name>
<email>marc.zyngier@arm.com</email>
</author>
<published>2016-06-30T17:40:36+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=3f0f8830d440e3edf5580424519a7c3434891c64'/>
<id>3f0f8830d440e3edf5580424519a7c3434891c64</id>
<content type='text'>
hyp_kern_va is now completely unused, so let's remove it entirely.

Signed-off-by: Marc Zyngier &lt;marc.zyngier@arm.com&gt;
Signed-off-by: Christoffer Dall &lt;christoffer.dall@linaro.org&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
hyp_kern_va is now completely unused, so let's remove it entirely.

Signed-off-by: Marc Zyngier &lt;marc.zyngier@arm.com&gt;
Signed-off-by: Christoffer Dall &lt;christoffer.dall@linaro.org&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>ARM: KVM: Use common version of timer-sr.c</title>
<updated>2016-02-29T18:34:19+00:00</updated>
<author>
<name>Marc Zyngier</name>
<email>marc.zyngier@arm.com</email>
</author>
<published>2016-01-28T14:48:42+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=68130cb5db09cb8a285a59f70ac72d2bfa8685fd'/>
<id>68130cb5db09cb8a285a59f70ac72d2bfa8685fd</id>
<content type='text'>
Using the common HYP timer code is a bit more tricky, since we
use system register names. Nothing a set of macros cannot
work around...

Acked-by: Christoffer Dall &lt;christoffer.dall@linaro.org&gt;
Signed-off-by: Marc Zyngier &lt;marc.zyngier@arm.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Using the common HYP timer code is a bit more tricky, since we
use system register names. Nothing a set of macros cannot
work around...

Acked-by: Christoffer Dall &lt;christoffer.dall@linaro.org&gt;
Signed-off-by: Marc Zyngier &lt;marc.zyngier@arm.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>ARM: KVM: Move kvm/hyp/hyp.h to include/asm/kvm_hyp.h</title>
<updated>2016-02-29T18:34:18+00:00</updated>
<author>
<name>Marc Zyngier</name>
<email>marc.zyngier@arm.com</email>
</author>
<published>2016-01-28T14:31:37+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=f1c9cad7c508f59fedd9f77eb36e5859e11ce5ab'/>
<id>f1c9cad7c508f59fedd9f77eb36e5859e11ce5ab</id>
<content type='text'>
In order to be able to use the code located in virt/kvm/arm/hyp,
we need to make the global hyp.h file accessible from include/asm,
similar to what we did for arm64.

Acked-by: Christoffer Dall &lt;christoffer.dall@linaro.org&gt;
Signed-off-by: Marc Zyngier &lt;marc.zyngier@arm.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
In order to be able to use the code located in virt/kvm/arm/hyp,
we need to make the global hyp.h file accessible from include/asm,
similar to what we did for arm64.

Acked-by: Christoffer Dall &lt;christoffer.dall@linaro.org&gt;
Signed-off-by: Marc Zyngier &lt;marc.zyngier@arm.com&gt;
</pre>
</div>
</content>
</entry>
</feed>
