<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux-toradex.git/virt/kvm/ioapic.h, branch v2.6.34.9</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>KVM: convert ioapic lock to spinlock</title>
<updated>2010-05-13T04:23:55+00:00</updated>
<author>
<name>Marcelo Tosatti</name>
<email>mtosatti@redhat.com</email>
</author>
<published>2010-04-23T17:03:38+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=46a47b1ed118cda1a08b7f6077b837a00fbc112b'/>
<id>46a47b1ed118cda1a08b7f6077b837a00fbc112b</id>
<content type='text'>
kvm_set_irq is used from non sleepable contexes, so convert ioapic from
mutex to spinlock.

KVM-Stable-Tag.
Tested-by: Ralf Bonenkamp &lt;ralf.bonenkamp@swyx.com&gt;
Signed-off-by: Marcelo Tosatti &lt;mtosatti@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
kvm_set_irq is used from non sleepable contexes, so convert ioapic from
mutex to spinlock.

KVM-Stable-Tag.
Tested-by: Ralf Bonenkamp &lt;ralf.bonenkamp@swyx.com&gt;
Signed-off-by: Marcelo Tosatti &lt;mtosatti@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>KVM: cleanup the failure path of KVM_CREATE_IRQCHIP ioctrl</title>
<updated>2010-03-01T15:36:10+00:00</updated>
<author>
<name>Wei Yongjun</name>
<email>yjwei@cn.fujitsu.com</email>
</author>
<published>2010-02-09T02:33:03+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=72bb2fcd23afe8db53b47e8f9edd736c517ba532'/>
<id>72bb2fcd23afe8db53b47e8f9edd736c517ba532</id>
<content type='text'>
If we fail to init ioapic device or the fail to setup the default irq
routing, the device register by kvm_create_pic() and kvm_ioapic_init()
remain unregister. This patch fixed to do this.

Signed-off-by: Wei Yongjun &lt;yjwei@cn.fujitsu.com&gt;
Signed-off-by: Avi Kivity &lt;avi@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
If we fail to init ioapic device or the fail to setup the default irq
routing, the device register by kvm_create_pic() and kvm_ioapic_init()
remain unregister. This patch fixed to do this.

Signed-off-by: Wei Yongjun &lt;yjwei@cn.fujitsu.com&gt;
Signed-off-by: Avi Kivity &lt;avi@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>KVM: avoid taking ioapic mutex for non-ioapic EOIs</title>
<updated>2010-03-01T15:35:46+00:00</updated>
<author>
<name>Avi Kivity</name>
<email>avi@redhat.com</email>
</author>
<published>2009-12-28T12:08:30+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=46a929bc15fcd48e1e0e770a44040a6949cae133'/>
<id>46a929bc15fcd48e1e0e770a44040a6949cae133</id>
<content type='text'>
When the guest acknowledges an interrupt, it sends an EOI message to the local
apic, which broadcasts it to the ioapic.  To handle the EOI, we need to take
the ioapic mutex.

On large guests, this causes a lot of contention on this mutex.  Since large
guests usually don't route interrupts via the ioapic (they use msi instead),
this is completely unnecessary.

Avoid taking the mutex by introducing a handled_vectors bitmap.  Before taking
the mutex, check if the ioapic was actually responsible for the acked vector.
If not, we can return early.

Signed-off-by: Avi Kivity &lt;avi@redhat.com&gt;
Signed-off-by: Marcelo Tosatti &lt;mtosatti@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
When the guest acknowledges an interrupt, it sends an EOI message to the local
apic, which broadcasts it to the ioapic.  To handle the EOI, we need to take
the ioapic mutex.

On large guests, this causes a lot of contention on this mutex.  Since large
guests usually don't route interrupts via the ioapic (they use msi instead),
this is completely unnecessary.

Avoid taking the mutex by introducing a handled_vectors bitmap.  Before taking
the mutex, check if the ioapic was actually responsible for the acked vector.
If not, we can return early.

Signed-off-by: Avi Kivity &lt;avi@redhat.com&gt;
Signed-off-by: Marcelo Tosatti &lt;mtosatti@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>KVM: Move IO APIC to its own lock</title>
<updated>2009-12-03T07:32:08+00:00</updated>
<author>
<name>Gleb Natapov</name>
<email>gleb@redhat.com</email>
</author>
<published>2009-08-24T08:54:25+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=eba0226bdfffe262e72b8360e4d0d12070e9a0f0'/>
<id>eba0226bdfffe262e72b8360e4d0d12070e9a0f0</id>
<content type='text'>
The allows removal of irq_lock from the injection path.

Signed-off-by: Gleb Natapov &lt;gleb@redhat.com&gt;
Signed-off-by: Avi Kivity &lt;avi@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
The allows removal of irq_lock from the injection path.

Signed-off-by: Gleb Natapov &lt;gleb@redhat.com&gt;
Signed-off-by: Avi Kivity &lt;avi@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>KVM: Move irq sharing information to irqchip level</title>
<updated>2009-12-03T07:32:06+00:00</updated>
<author>
<name>Gleb Natapov</name>
<email>gleb@redhat.com</email>
</author>
<published>2009-08-24T08:54:19+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=1a6e4a8c276e122dbeb6f9c610f29735e4236bfd'/>
<id>1a6e4a8c276e122dbeb6f9c610f29735e4236bfd</id>
<content type='text'>
This removes assumptions that max GSIs is smaller than number of pins.
Sharing is tracked on pin level not GSI level.

[avi: no PIC on ia64]

Signed-off-by: Gleb Natapov &lt;gleb@redhat.com&gt;
Signed-off-by: Avi Kivity &lt;avi@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This removes assumptions that max GSIs is smaller than number of pins.
Sharing is tracked on pin level not GSI level.

[avi: no PIC on ia64]

Signed-off-by: Gleb Natapov &lt;gleb@redhat.com&gt;
Signed-off-by: Avi Kivity &lt;avi@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>KVM: APIC: get rid of deliver_bitmask</title>
<updated>2009-06-10T08:48:27+00:00</updated>
<author>
<name>Gleb Natapov</name>
<email>gleb@redhat.com</email>
</author>
<published>2009-03-05T14:35:04+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=58c2dde17d6eb6c8c0566e52d184aa16755d890f'/>
<id>58c2dde17d6eb6c8c0566e52d184aa16755d890f</id>
<content type='text'>
Deliver interrupt during destination matching loop.

Signed-off-by: Gleb Natapov &lt;gleb@redhat.com&gt;
Acked-by: Xiantao Zhang &lt;xiantao.zhang@intel.com&gt;
Signed-off-by: Marcelo Tosatti &lt;mtosatti@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Deliver interrupt during destination matching loop.

Signed-off-by: Gleb Natapov &lt;gleb@redhat.com&gt;
Acked-by: Xiantao Zhang &lt;xiantao.zhang@intel.com&gt;
Signed-off-by: Marcelo Tosatti &lt;mtosatti@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>KVM: change the way how lowest priority vcpu is calculated</title>
<updated>2009-06-10T08:48:27+00:00</updated>
<author>
<name>Gleb Natapov</name>
<email>gleb@redhat.com</email>
</author>
<published>2009-03-05T14:34:59+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=e1035715ef8d3171e29f9c6aee6f40d57b3fead5'/>
<id>e1035715ef8d3171e29f9c6aee6f40d57b3fead5</id>
<content type='text'>
The new way does not require additional loop over vcpus to calculate
the one with lowest priority as one is chosen during delivery bitmap
construction.

Signed-off-by: Gleb Natapov &lt;gleb@redhat.com&gt;
Signed-off-by: Marcelo Tosatti &lt;mtosatti@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
The new way does not require additional loop over vcpus to calculate
the one with lowest priority as one is chosen during delivery bitmap
construction.

Signed-off-by: Gleb Natapov &lt;gleb@redhat.com&gt;
Signed-off-by: Marcelo Tosatti &lt;mtosatti@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>KVM: consolidate ioapic/ipi interrupt delivery logic</title>
<updated>2009-06-10T08:48:27+00:00</updated>
<author>
<name>Gleb Natapov</name>
<email>gleb@redhat.com</email>
</author>
<published>2009-03-05T14:34:54+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=343f94fe4d16ec898da77720c03da9e09f8523d2'/>
<id>343f94fe4d16ec898da77720c03da9e09f8523d2</id>
<content type='text'>
Use kvm_apic_match_dest() in kvm_get_intr_delivery_bitmask() instead
of duplicating the same code. Use kvm_get_intr_delivery_bitmask() in
apic_send_ipi() to figure out ipi destination instead of reimplementing
the logic.

Signed-off-by: Gleb Natapov &lt;gleb@redhat.com&gt;
Signed-off-by: Marcelo Tosatti &lt;mtosatti@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Use kvm_apic_match_dest() in kvm_get_intr_delivery_bitmask() instead
of duplicating the same code. Use kvm_get_intr_delivery_bitmask() in
apic_send_ipi() to figure out ipi destination instead of reimplementing
the logic.

Signed-off-by: Gleb Natapov &lt;gleb@redhat.com&gt;
Signed-off-by: Marcelo Tosatti &lt;mtosatti@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>KVM: ioapic/msi interrupt delivery consolidation</title>
<updated>2009-06-10T08:48:27+00:00</updated>
<author>
<name>Gleb Natapov</name>
<email>gleb@redhat.com</email>
</author>
<published>2009-03-05T14:34:49+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=a53c17d21c46a752f5ac6695376481bc27865b04'/>
<id>a53c17d21c46a752f5ac6695376481bc27865b04</id>
<content type='text'>
ioapic_deliver() and kvm_set_msi() have code duplication. Move
the code into ioapic_deliver_entry() function and call it from
both places.

Signed-off-by: Gleb Natapov &lt;gleb@redhat.com&gt;
Signed-off-by: Marcelo Tosatti &lt;mtosatti@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
ioapic_deliver() and kvm_set_msi() have code duplication. Move
the code into ioapic_deliver_entry() function and call it from
both places.

Signed-off-by: Gleb Natapov &lt;gleb@redhat.com&gt;
Signed-off-by: Marcelo Tosatti &lt;mtosatti@redhat.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>KVM: Merge kvm_ioapic_get_delivery_bitmask into kvm_get_intr_delivery_bitmask</title>
<updated>2009-06-10T08:48:26+00:00</updated>
<author>
<name>Sheng Yang</name>
<email>sheng@linux.intel.com</email>
</author>
<published>2009-03-04T05:33:02+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=74a3a8f152053394a016518cc2f2fee216897fa4'/>
<id>74a3a8f152053394a016518cc2f2fee216897fa4</id>
<content type='text'>
Gleb fixed bitmap ops usage in kvm_ioapic_get_delivery_bitmask.

Sheng merged two functions, as well as fixed several issues in
kvm_get_intr_delivery_bitmask
1. deliver_bitmask is a bitmap rather than a unsigned long intereger.
2. Lowest priority target bitmap wrong calculated by mistake.
3. Prevent potential NULL reference.
4. Declaration in include/kvm_host.h caused powerpc compilation warning.
5. Add warning for guest broadcast interrupt with lowest priority delivery mode.
6. Removed duplicate bitmap clean up in caller of kvm_get_intr_delivery_bitmask.

Signed-off-by: Gleb Natapov &lt;gleb@redhat.com&gt;
Signed-off-by: Sheng Yang &lt;sheng@linux.intel.com&gt;
Signed-off-by: Marcelo Tosatti &lt;mtosatti@redhat.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Gleb fixed bitmap ops usage in kvm_ioapic_get_delivery_bitmask.

Sheng merged two functions, as well as fixed several issues in
kvm_get_intr_delivery_bitmask
1. deliver_bitmask is a bitmap rather than a unsigned long intereger.
2. Lowest priority target bitmap wrong calculated by mistake.
3. Prevent potential NULL reference.
4. Declaration in include/kvm_host.h caused powerpc compilation warning.
5. Add warning for guest broadcast interrupt with lowest priority delivery mode.
6. Removed duplicate bitmap clean up in caller of kvm_get_intr_delivery_bitmask.

Signed-off-by: Gleb Natapov &lt;gleb@redhat.com&gt;
Signed-off-by: Sheng Yang &lt;sheng@linux.intel.com&gt;
Signed-off-by: Marcelo Tosatti &lt;mtosatti@redhat.com&gt;
</pre>
</div>
</content>
</entry>
</feed>
