<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux-dev/tools/testing/selftests/kvm/aarch64, branch master</title>
<subtitle>Linux kernel development work - see feature branches</subtitle>
<id>https://git.zx2c4.com/linux-dev/atom/tools/testing/selftests/kvm/aarch64?h=master</id>
<link rel='self' href='https://git.zx2c4.com/linux-dev/atom/tools/testing/selftests/kvm/aarch64?h=master'/>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/'/>
<updated>2022-10-10T08:41:39Z</updated>
<entry>
<title>KVM: arm64: selftests: Fix multiple versions of GIC creation</title>
<updated>2022-10-10T08:41:39Z</updated>
<author>
<name>Zenghui Yu</name>
<email>yuzenghui@huawei.com</email>
</author>
<published>2022-10-09T03:31:31Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=8a6ffcbe26fd14d58075dcf3cbdf1b5b69b20402'/>
<id>urn:sha1:8a6ffcbe26fd14d58075dcf3cbdf1b5b69b20402</id>
<content type='text'>
Commit 98f94ce42ac6 ("KVM: selftests: Move KVM_CREATE_DEVICE_TEST code to
separate helper") wrongly converted a "real" GIC device creation to
__kvm_test_create_device() and caused the test failure on my D05 (which
supports v2 emulation). Fix it.

Fixes: 98f94ce42ac6 ("KVM: selftests: Move KVM_CREATE_DEVICE_TEST code to separate helper")
Signed-off-by: Zenghui Yu &lt;yuzenghui@huawei.com&gt;
Reviewed-by: Oliver Upton &lt;oliver.upton@linux.dev&gt;
Signed-off-by: Marc Zyngier &lt;maz@kernel.org&gt;
Link: https://lore.kernel.org/r/20221009033131.365-1-yuzenghui@huawei.com
</content>
</entry>
<entry>
<title>Merge branch kvm-arm64/misc-6.1 into kvmarm-master/next</title>
<updated>2022-10-01T09:19:36Z</updated>
<author>
<name>Marc Zyngier</name>
<email>maz@kernel.org</email>
</author>
<published>2022-10-01T09:19:36Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=b302ca52ba8235ff0e18c0fa1fa92b51784aef6a'/>
<id>urn:sha1:b302ca52ba8235ff0e18c0fa1fa92b51784aef6a</id>
<content type='text'>
* kvm-arm64/misc-6.1:
  : .
  : Misc KVM/arm64 fixes and improvement for v6.1
  :
  : - Simplify the affinity check when moving a GICv3 collection
  :
  : - Tone down the shouting when kvm-arm.mode=protected is passed
  :   to a guest
  :
  : - Fix various comments
  :
  : - Advertise the new kvmarm@lists.linux.dev and deprecate the
  :   old Columbia list
  : .
  KVM: arm64: Advertise new kvmarm mailing list
  KVM: arm64: Fix comment typo in nvhe/switch.c
  KVM: selftests: Update top-of-file comment in psci_test
  KVM: arm64: Ignore kvm-arm.mode if !is_hyp_mode_available()
  KVM: arm64: vgic: Remove duplicate check in update_affinity_collection()

Signed-off-by: Marc Zyngier &lt;maz@kernel.org&gt;
</content>
</entry>
<entry>
<title>KVM: selftests: Update top-of-file comment in psci_test</title>
<updated>2022-09-28T15:47:59Z</updated>
<author>
<name>Oliver Upton</name>
<email>oliver.upton@linux.dev</email>
</author>
<published>2022-08-19T16:21:00Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=448e711693e48d03f7933ab3673334701b0c3f41'/>
<id>urn:sha1:448e711693e48d03f7933ab3673334701b0c3f41</id>
<content type='text'>
Fix the comment to accurately describe the test and recently added
SYSTEM_SUSPEND test case.

What was once psci_cpu_on_test was renamed and extended to squeeze in a
test case for PSCI SYSTEM_SUSPEND. Nonetheless, the author of those
changes (whoever they may be...) failed to update the file comment to
reflect what had changed.

Reported-by: Reiji Watanabe &lt;reijiw@google.com&gt;
Signed-off-by: Oliver Upton &lt;oliver.upton@linux.dev&gt;
Signed-off-by: Marc Zyngier &lt;maz@kernel.org&gt;
Link: https://lore.kernel.org/r/20220819162100.213854-1-oliver.upton@linux.dev
</content>
</entry>
<entry>
<title>Merge branch kvm-arm64/single-step-async-exception into kvmarm-master/next</title>
<updated>2022-09-19T09:59:29Z</updated>
<author>
<name>Marc Zyngier</name>
<email>maz@kernel.org</email>
</author>
<published>2022-09-19T09:59:29Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=bb0cca240a16dd9721c6ee1d55865465a3fb7211'/>
<id>urn:sha1:bb0cca240a16dd9721c6ee1d55865465a3fb7211</id>
<content type='text'>
* kvm-arm64/single-step-async-exception:
  : .
  : Single-step fixes from Reiji Watanabe:
  :
  : "This series fixes two bugs of single-step execution enabled by
  : userspace, and add a test case for KVM_GUESTDBG_SINGLESTEP to
  : the debug-exception test to verify the single-step behavior."
  : .
  KVM: arm64: selftests: Add a test case for KVM_GUESTDBG_SINGLESTEP
  KVM: arm64: selftests: Refactor debug-exceptions to make it amenable to new test cases
  KVM: arm64: Clear PSTATE.SS when the Software Step state was Active-pending
  KVM: arm64: Preserve PSTATE.SS for the guest while single-step is enabled

Signed-off-by: Marc Zyngier &lt;maz@kernel.org&gt;
</content>
</entry>
<entry>
<title>KVM: arm64: selftests: Add a test case for KVM_GUESTDBG_SINGLESTEP</title>
<updated>2022-09-19T09:48:53Z</updated>
<author>
<name>Reiji Watanabe</name>
<email>reijiw@google.com</email>
</author>
<published>2022-09-17T01:06:00Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=b18e4d4aebdddd05810ceb2f73d7f72afcd11b41'/>
<id>urn:sha1:b18e4d4aebdddd05810ceb2f73d7f72afcd11b41</id>
<content type='text'>
Add a test case for KVM_GUESTDBG_SINGLESTEP to the debug-exceptions test.
The test enables single-step execution from userspace, and check if the
exit to userspace occurs for each instruction that is stepped.
Set the default number of the test iterations to a number of iterations
sufficient to always reproduce the problem that the previous patch fixes
on an Ampere Altra machine.

Signed-off-by: Reiji Watanabe &lt;reijiw@google.com&gt;
Signed-off-by: Marc Zyngier &lt;maz@kernel.org&gt;
Link: https://lore.kernel.org/r/20220917010600.532642-5-reijiw@google.com
</content>
</entry>
<entry>
<title>KVM: arm64: selftests: Refactor debug-exceptions to make it amenable to new test cases</title>
<updated>2022-09-19T09:48:53Z</updated>
<author>
<name>Reiji Watanabe</name>
<email>reijiw@google.com</email>
</author>
<published>2022-09-17T01:05:59Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=ff00e737090e0f015059e59829aaa58565b16321'/>
<id>urn:sha1:ff00e737090e0f015059e59829aaa58565b16321</id>
<content type='text'>
Split up the current test into a helper, but leave the debug version
checking in main(), to make it convenient to add a new debug exception
test case in a subsequent patch.

Signed-off-by: Reiji Watanabe &lt;reijiw@google.com&gt;
Signed-off-by: Marc Zyngier &lt;maz@kernel.org&gt;
Link: https://lore.kernel.org/r/20220917010600.532642-4-reijiw@google.com
</content>
</entry>
<entry>
<title>KVM: selftests: Add test for AArch32 ID registers</title>
<updated>2022-09-14T10:36:16Z</updated>
<author>
<name>Oliver Upton</name>
<email>oliver.upton@linux.dev</email>
</author>
<published>2022-09-13T09:44:40Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=797b84517c190053597e3f7e03ead15da872e04d'/>
<id>urn:sha1:797b84517c190053597e3f7e03ead15da872e04d</id>
<content type='text'>
Add a test to assert that KVM handles the AArch64 views of the AArch32
ID registers as RAZ/WI (writable only from userspace). For registers
that were already hidden or unallocated, expect RAZ + invariant
behavior.

Signed-off-by: Oliver Upton &lt;oliver.upton@linux.dev&gt;
Signed-off-by: Marc Zyngier &lt;maz@kernel.org&gt;
Link: https://lore.kernel.org/r/20220913094441.3957645-8-oliver.upton@linux.dev
</content>
</entry>
<entry>
<title>Merge tag 'kvmarm-5.20' of git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm into HEAD</title>
<updated>2022-08-01T07:24:12Z</updated>
<author>
<name>Paolo Bonzini</name>
<email>pbonzini@redhat.com</email>
</author>
<published>2022-07-29T14:07:59Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=c4edb2babc750d829cb551a6af7cb39b2afc9fb5'/>
<id>urn:sha1:c4edb2babc750d829cb551a6af7cb39b2afc9fb5</id>
<content type='text'>
KVM/arm64 updates for 5.20:

- Unwinder implementations for both nVHE modes (classic and
  protected), complete with an overflow stack

- Rework of the sysreg access from userspace, with a complete
  rewrite of the vgic-v3 view to allign with the rest of the
  infrastructure

- Disagregation of the vcpu flags in separate sets to better track
  their use model.

- A fix for the GICv2-on-v3 selftest

- A small set of cosmetic fixes
</content>
</entry>
<entry>
<title>KVM: arm64: selftests: Add support for GICv2 on v3</title>
<updated>2022-07-15T10:01:00Z</updated>
<author>
<name>Marc Zyngier</name>
<email>maz@kernel.org</email>
</author>
<published>2022-07-14T15:41:08Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=6a4f7fcd750497cb2fa870f799e8b23270bec6e3'/>
<id>urn:sha1:6a4f7fcd750497cb2fa870f799e8b23270bec6e3</id>
<content type='text'>
The current vgic_init test wrongly assumes that the host cannot
multiple versions of the GIC architecture, while v2 emulation
on v3 has almost always been supported (it was supported before
the standalone v3 emulation).

Tweak the test to support multiple GIC incarnations.

Signed-off-by: Marc Zyngier &lt;maz@kernel.org&gt;
Fixes: 3f4db37e203b ("KVM: arm64: selftests: Make vgic_init gic version agnostic")
Reviewed-by: Ricardo Koller &lt;ricarkol@google.com&gt;
Link: https://lore.kernel.org/r/20220714154108.3531213-1-maz@kernel.org
</content>
</entry>
<entry>
<title>KVM: selftests: Fix filename reporting in guest asserts</title>
<updated>2022-07-14T01:14:08Z</updated>
<author>
<name>Colton Lewis</name>
<email>coltonlewis@google.com</email>
</author>
<published>2022-06-15T19:31:16Z</published>
<link rel='alternate' type='text/html' href='https://git.zx2c4.com/linux-dev/commit/?id=594a1c271c159c9c5f0ff2d92ebfda469e94e48d'/>
<id>urn:sha1:594a1c271c159c9c5f0ff2d92ebfda469e94e48d</id>
<content type='text'>
Fix filename reporting in guest asserts by ensuring the GUEST_ASSERT
macro records __FILE__ and substituting REPORT_GUEST_ASSERT for many
repetitive calls to TEST_FAIL.

Previously filename was reported by using __FILE__ directly in the
selftest, wrongly assuming it would always be the same as where the
assertion failed.

Signed-off-by: Colton Lewis &lt;coltonlewis@google.com&gt;
Reported-by: Ricardo Koller &lt;ricarkol@google.com&gt;
Fixes: 4e18bccc2e5544f0be28fc1c4e6be47a469d6c60
Link: https://lore.kernel.org/r/20220615193116.806312-5-coltonlewis@google.com
[sean: convert more TEST_FAIL =&gt; REPORT_GUEST_ASSERT instances]
Signed-off-by: Sean Christopherson &lt;seanjc@google.com&gt;
</content>
</entry>
</feed>
