In the real SMP hardware topology world, it's much more likely that
we have high cores-per-socket counts and few sockets totally. While
the current preference of sockets over cores in smp parsing results
in a virtual cpu topology with low cores-per-sockets counts and a
large number of sockets, which is just contrary to the real world.
Given that it is better to make the virtual cpu topology be more
reflective of the real world and also for the sake of compatibility,
we start to prefer cores over sockets over threads in smp parsing
since machine type 6.2 for different arches.
In this patch, a boolean "smp_prefer_sockets" is added, and we only
enable the old preference on older machines and enable the new one
since type 6.2 for all arches by using the machine compat mechanism.
Suggested-by: Daniel P. Berrange <berrange@redhat.com>
Signed-off-by: Yanan Wang <wangyanan55@huawei.com>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Pankaj Gupta <pankaj.gupta@ionos.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20210929025816.21076-10-wangyanan55@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Since commit 80d7835749 (qemu-options: rewrite help for -smp options),
the preference of sockets/cores in -smp parsing is considered liable
to change, and actually we are going to change it in a coming commit.
So it'll be more stable to use detailed -smp CLIs in the testcases
that have strong dependency on the parsing results.
Currently, test_def_cpu_split use "-smp 8" and will get 8 CPU sockets
based on current parsing rule. But if we change to prefer cores over
sockets we will get one CPU socket with 8 cores, and this testcase
will not get expected numa set by default on x86_64 (Ok on aarch64).
So now explicitly use "-smp 8,sockets=8" to avoid affect from parsing
logic change.
Signed-off-by: Yanan Wang <wangyanan55@huawei.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20210929025816.21076-9-wangyanan55@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Since commit 80d7835749 (qemu-options: rewrite help for -smp options),
the preference of sockets/cores in -smp parsing is considered liable
to change, and actually we are going to change it in a coming commit.
So it'll be more stable to use detailed -smp CLIs in testing if we
have strong dependency on the parsing results.
pc_dynamic_cpu_cfg currently assumes/needs that there will be 2 CPU
sockets with "-smp 2". To avoid breaking the test because of parsing
logic change, now explicitly use "-smp 2,sockets=2".
Signed-off-by: Yanan Wang <wangyanan55@huawei.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20210929025816.21076-8-wangyanan55@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We have two requirements for a valid SMP configuration:
the product of "sockets * cores * threads" must represent all the
possible cpus, i.e., max_cpus, and then must include the initially
present cpus, i.e., smp_cpus.
So we only need to ensure 1) "sockets * cores * threads == maxcpus"
at first and then ensure 2) "maxcpus >= cpus". With a reasonable
order of the sanity check, we can simplify the error reporting code.
When reporting an error message we also report the exact value of
each topology member to make users easily see what's going on.
Signed-off-by: Yanan Wang <wangyanan55@huawei.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Pankaj Gupta <pankaj.gupta@ionos.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20210929025816.21076-7-wangyanan55@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Currently we directly calculate the omitted cpus based on the given
incomplete collection of parameters. This makes some cmdlines like:
-smp maxcpus=16
-smp sockets=2,maxcpus=16
-smp sockets=2,dies=2,maxcpus=16
-smp sockets=2,cores=4,maxcpus=16
not work. We should probably set the value of cpus to match maxcpus
if it's omitted, which will make above configs start to work.
So the calculation logic of cpus/maxcpus after this patch will be:
When both maxcpus and cpus are omitted, maxcpus will be calculated
from the given parameters and cpus will be set equal to maxcpus.
When only one of maxcpus and cpus is given then the omitted one
will be set to its counterpart's value. Both maxcpus and cpus may
be specified, but maxcpus must be equal to or greater than cpus.
Note: change in this patch won't affect any existing working cmdlines
but allows more incomplete configs to be valid.
Signed-off-by: Yanan Wang <wangyanan55@huawei.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20210929025816.21076-6-wangyanan55@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We are currently using maxcpus to calculate the omitted sockets
but using cpus to calculate the omitted cores/threads. This makes
cmdlines like:
-smp cpus=8,maxcpus=16
-smp cpus=8,cores=4,maxcpus=16
-smp cpus=8,threads=2,maxcpus=16
work fine but the ones like:
-smp cpus=8,sockets=2,maxcpus=16
-smp cpus=8,sockets=2,cores=4,maxcpus=16
-smp cpus=8,sockets=2,threads=2,maxcpus=16
break the sanity check.
Since we require for a valid config that the product of "sockets * cores
* threads" should equal to the maxcpus, we should uniformly use maxcpus
to calculate their omitted values.
Also the if-branch of "cpus == 0 || sockets == 0" was split into two
branches of "cpus == 0" and "sockets == 0" so that we can clearly read
that we are parsing the configuration with a preference on cpus over
sockets over cores over threads.
Note: change in this patch won't affect any existing working cmdlines
but improves consistency and allows more incomplete configs to be valid.
Signed-off-by: Yanan Wang <wangyanan55@huawei.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Pankaj Gupta <pankaj.gupta@ionos.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20210929025816.21076-5-wangyanan55@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
To pave the way for the functional improvement in later patches,
make some refactor/cleanup for the smp parsers, including using
local maxcpus instead of ms->smp.max_cpus in the calculation,
defaulting dies to 0 initially like other members, cleanup the
sanity check for dies.
We actually also fix a hidden defect by avoiding directly using
the provided *zero value* in the calculation, which could cause
a segment fault (e.g. using dies=0 in the calculation).
Signed-off-by: Yanan Wang <wangyanan55@huawei.com>
Reviewed-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20210929025816.21076-4-wangyanan55@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In the SMP configuration, we should either provide a topology
parameter with a reasonable value (greater than zero) or just
omit it and QEMU will compute the missing value.
The users shouldn't provide a configuration with any parameter
of it specified as zero (e.g. -smp 8,sockets=0) which could
possibly cause unexpected results in the -smp parsing. So we
deprecate this kind of configurations since 6.2 by adding the
explicit sanity check.
Signed-off-by: Yanan Wang <wangyanan55@huawei.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20210929025816.21076-3-wangyanan55@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The explanation of @cores should be "number of cores per die" but
not "number of cores per thread". Let's fix it.
Fixes: 1e63fe6858 ("machine: pass QAPI struct to mc->smp_parse")
Signed-off-by: Yanan Wang <wangyanan55@huawei.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20210929025816.21076-2-wangyanan55@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Here's the next batch of ppc related patches for qemu-6.2. Highlights
are:
* Fixes for several TCG math instructions from the El Dorado Institute
* A number of improvements to the powernv machine type
* Support for a new DEVICE_UNPLUG_GUEST_ERROR QAPI event from Daniel
Barboza
* Support for the new FORM2 PAPR NUMA representation. This allows
more specific NUMA distances, as well as asymmetric configurations
* Fix for 64-bit decrementer (used on MicroWatt CPUs)
* Assorted fixes and cleanups
* A number of updates to MAINTAINERS
Note that the DEVICE_UNPLUG_GUEST_ERROR stuff includes changes to
files outside my normal area, but has suitable Acks.
The MAINTAINERS updates are mostly about marking minor platforms
unmaintained / orphaned, and moving some pieces away from myself and
Greg. As we move onto other projects, we're going to need to drop
more of the ppc maintainership, though we're hoping we can avoid too
abrupt a change.
-----BEGIN PGP SIGNATURE-----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=shYQ
-----END PGP SIGNATURE-----
Merge remote-tracking branch 'remotes/dg-gitlab/tags/ppc-for-6.2-20210930' into staging
ppc patch queue for 2021-09-30
Here's the next batch of ppc related patches for qemu-6.2. Highlights
are:
* Fixes for several TCG math instructions from the El Dorado Institute
* A number of improvements to the powernv machine type
* Support for a new DEVICE_UNPLUG_GUEST_ERROR QAPI event from Daniel
Barboza
* Support for the new FORM2 PAPR NUMA representation. This allows
more specific NUMA distances, as well as asymmetric configurations
* Fix for 64-bit decrementer (used on MicroWatt CPUs)
* Assorted fixes and cleanups
* A number of updates to MAINTAINERS
Note that the DEVICE_UNPLUG_GUEST_ERROR stuff includes changes to
files outside my normal area, but has suitable Acks.
The MAINTAINERS updates are mostly about marking minor platforms
unmaintained / orphaned, and moving some pieces away from myself and
Greg. As we move onto other projects, we're going to need to drop
more of the ppc maintainership, though we're hoping we can avoid too
abrupt a change.
# gpg: Signature made Thu 30 Sep 2021 06:42:41 BST
# gpg: using RSA key 75F46586AE61A66CC44E87DC6C38CACA20D9B392
# gpg: Good signature from "David Gibson <david@gibson.dropbear.id.au>" [full]
# gpg: aka "David Gibson (Red Hat) <dgibson@redhat.com>" [full]
# gpg: aka "David Gibson (ozlabs.org) <dgibson@ozlabs.org>" [full]
# gpg: aka "David Gibson (kernel.org) <dwg@kernel.org>" [unknown]
# Primary key fingerprint: 75F4 6586 AE61 A66C C44E 87DC 6C38 CACA 20D9 B392
* remotes/dg-gitlab/tags/ppc-for-6.2-20210930: (44 commits)
MAINTAINERS: Demote sPAPR from "Supported" to "Maintained"
MAINTAINERS: Add information for OpenPIC
MAINTAINERS: Remove David & Greg as reviewers/co-maintainers of powernv
MAINTAINERS: Orphan obscure ppc platforms
MAINTAINERS: Remove David & Greg as reviewers for a number of boards
MAINTAINERS: Remove machine specific files from ppc TCG CPUs entry
spapr/xive: Fix kvm_xive_source_reset trace event
spapr_numa.c: fixes in spapr_numa_FORM2_write_rtas_tables()
hw/intc: openpic: Clean up the styles
hw/intc: openpic: Drop Raven related codes
hw/intc: openpic: Correct the reset value of IPIDR for FSL chipset
target/ppc: Fix 64-bit decrementer
target/ppc: Convert debug to trace events (decrementer and IRQ)
spapr_numa.c: handle auto NUMA node with no distance info
spapr_numa.c: FORM2 NUMA affinity support
spapr: move FORM1 verifications to post CAS
spapr_numa.c: rename numa_assoc_array to FORM1_assoc_array
spapr_numa.c: parametrize FORM1 macros
spapr_numa.c: scrap 'legacy_numa' concept
spapr_numa.c: split FORM1 code into helpers
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
There is no reason why VNC should always be enabled and not be set to
the default value. We already switched the setting in the "configure"
script in commit 3a6a1256d4 ("configure: Allow vnc to get disabled with
--without-default-features"), so let's do that in meson_options.txt now,
too.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20210903081358.956267-3-thuth@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The QAPI schema shouldn't rely on C system headers #define, but on
configure-time project #define, so we can express the build condition in
a C-independent way.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20210907121943.3498701-3-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Trace at memory_region_sync_dirty_bitmap() for log_sync() or global_log_sync()
on memory regions. One trace line should suffice when it finishes, so as to
estimate the time used for each log sync process.
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20210817013706.30986-1-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Provide a name field for all the memory listeners. It can be used to identify
which memory listener is which.
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-Id: <20210817013553.30584-2-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In sev_read_file_base64() we call g_file_get_contents(), which
allocates memory for the file contents. We then base64-decode the
contents (which allocates another buffer for the decoded data), but
forgot to free the memory for the original file data.
Use g_autofree to ensure that the file data is freed.
Fixes: Coverity CID 1459997
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20210820165650.2839-1-peter.maydell@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Skip the test if bzip2 is not available, and run it after they are
uncompressed.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20210923105529.3845741-2-pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The edk2 firmware blobs are needed to run bios-tables-test. Unpack
them if any UEFI-enabled target is selected, so that the test can run.
This is a bit more than is actually necessary, since bios-tables-test
does not run for all UEFI-enabled targets, but it is the easiest
way to write this logic.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20210923105529.3845741-1-pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Libvirt can use query-sgx-capabilities to get the host
sgx capabilities to decide how to allocate SGX EPC size to VM.
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210910102258.46648-3-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The QMP and HMP interfaces can be used by monitor or QMP tools to retrieve
the SGX information from VM side when SGX is enabled on Intel platform.
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210910102258.46648-2-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-34-yang.zhong@intel.com>
[Convert to reStructuredText, and adopt the standard === --- ~~~ headings
suggested for example by Linux. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Since there is no fill_device_info() callback support, and when we
execute "info memory-devices" command in the monitor, the segfault
will be found.
This patch will add this callback support and "info memory-devices"
will show sgx epc memory exposed to guest. The result as below:
qemu) info memory-devices
Memory device [sgx-epc]: ""
memaddr: 0x180000000
size: 29360128
memdev: /objects/mem1
Memory device [sgx-epc]: ""
memaddr: 0x181c00000
size: 10485760
memdev: /objects/mem2
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-33-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Enable SGX EPC virtualization, which is currently only support by KVM.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-22-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Enable SGX EPC virtualization, which is currently only support by KVM.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-21-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The ACPI Device entry for SGX EPC is essentially a hack whose primary
purpose is to provide software with a way to autoprobe SGX support,
e.g. to allow software to implement SGX support as a driver. Details
on the individual EPC sections are not enumerated through ACPI tables,
i.e. software must enumerate the EPC sections via CPUID. Furthermore,
software expects to see only a single EPC Device in the ACPI tables
regardless of the number of EPC sections in the system.
However, several versions of Windows do rely on the ACPI tables to
enumerate the address and size of the EPC. So, regardless of the number
of EPC sections exposed to the guest, create exactly *one* EPC device
with a _CRS entry that spans the entirety of all EPC sections (which are
guaranteed to be contiguous in Qemu).
Note, NUMA support for EPC memory is intentionally not considered as
enumerating EPC NUMA information is not yet defined for bare metal.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-20-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Note that SGX EPC is currently guaranteed to reside in a single
contiguous chunk of memory regardless of the number of EPC sections.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-19-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add helpers to detect if SGX EPC exists above 4g, and if so, where SGX
EPC above 4g ends. Use the helpers to adjust the device memory range
if SGX EPC exists above 4g.
For multiple virtual EPC sections, we just put them together physically
contiguous for the simplicity because we don't support EPC NUMA affinity
now. Once the SGX EPC NUMA support in the kernel SGX driver, we will
support this in the future.
Note that SGX EPC is currently hardcoded to reside above 4g.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-18-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Request SGX an SGX Launch Control to be enabled in FEATURE_CONTROL
when the features are exposed to the guest. Our design is the SGX
Launch Control bit will be unconditionally set in FEATURE_CONTROL,
which is unlike host bios.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-17-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
SGX capabilities are enumerated through CPUID_0x12.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-16-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The SGX sub-leafs are enumerated at CPUID 0x12. Indices 0 and 1 are
always present when SGX is supported, and enumerate SGX features and
capabilities. Indices >=2 are directly correlated with the platform's
EPC sections. Because the number of EPC sections is dynamic and user
defined, the number of SGX sub-leafs is "NULL" terminated.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-15-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If the guest want to fully use SGX, the guest needs to be able to
access provisioning key. Add a new KVM_CAP_SGX_ATTRIBUTE to KVM to
support provisioning key to KVM guests.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-14-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Expose SGX to the guest if and only if KVM is enabled and supports
virtualization of SGX. While the majority of ENCLS can be emulated to
some degree, because SGX uses a hardware-based root of trust, the
attestation aspects of SGX cannot be emulated in software, i.e.
ultimately emulation will fail as software cannot generate a valid
quote/report. The complexity of partially emulating SGX in Qemu far
outweighs the value added, e.g. an SGX specific simulator for userspace
applications can emulate SGX for development and testing purposes.
Note, access to the PROVISIONKEY is not yet advertised to the guest as
KVM blocks access to the PROVISIONKEY by default and requires userspace
to provide additional credentials (via ioctl()) to expose PROVISIONKEY.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-13-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
SGX adds multiple flags to FEATURE_CONTROL to enable SGX and Flexible
Launch Control.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-12-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
On real hardware, on systems that supports SGX Launch Control, those
MSRs are initialized to digest of Intel's signing key; on systems that
don't support SGX Launch Control, those MSRs are not available but
hardware always uses digest of Intel's signing key in EINIT.
KVM advertises SGX LC via CPUID if and only if the MSRs are writable.
Unconditionally initialize those MSRs to digest of Intel's signing key
when CPU is realized and reset to reflect the fact. This avoids
potential bug in case kvm_arch_put_registers() is called before
kvm_arch_get_registers() is called, in which case guest's virtual
SGX_LEPUBKEYHASH MSRs will be set to 0, although KVM initializes those
to digest of Intel's signing key by default, since KVM allows those MSRs
to be updated by Qemu to support live migration.
Save/restore the SGX Launch Enclave Public Key Hash MSRs if SGX Launch
Control (LC) is exposed to the guest. Likewise, migrate the MSRs if they
are writable by the guest.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Kai Huang <kai.huang@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-11-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
CPUID leaf 12_1_EAX is an Intel-defined feature bits leaf enumerating
the platform's SGX capabilities that may be utilized by an enclave, e.g.
whether or not an enclave can gain access to the provision key.
Currently there are six capabilities:
- INIT: set when the enclave has has been initialized by EINIT. Cannot
be set by software, i.e. forced to zero in CPUID.
- DEBUG: permits a debugger to read/write into the enclave.
- MODE64BIT: the enclave runs in 64-bit mode
- PROVISIONKEY: grants has access to the provision key
- EINITTOKENKEY: grants access to the EINIT token key, i.e. the
enclave can generate EINIT tokens
- KSS: Key Separation and Sharing enabled for the enclave.
Note that the entirety of CPUID.0x12.0x1, i.e. all registers, enumerates
the allowed ATTRIBUTES (128 bits), but only bits 31:0 are directly
exposed to the user (via FEAT_12_1_EAX). Bits 63:32 are currently all
reserved and bits 127:64 correspond to the allowed XSAVE Feature Request
Mask, which is calculated based on other CPU features, e.g. XSAVE, MPX,
AVX, etc... and is not exposed to the user.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-10-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
CPUID leaf 12_0_EBX is an Intel-defined feature bits leaf enumerating
the platform's SGX extended capabilities. Currently there is a single
capabilitiy:
- EXINFO: record information about #PFs and #GPs in the enclave's SSA
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-9-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
CPUID leaf 12_0_EAX is an Intel-defined feature bits leaf enumerating
the CPU's SGX capabilities, e.g. supported SGX instruction sets.
Currently there are four enumerated capabilities:
- SGX1 instruction set, i.e. "base" SGX
- SGX2 instruction set for dynamic EPC management
- ENCLV instruction set for VMM oversubscription of EPC
- ENCLS-C instruction set for thread safe variants of ENCLS
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-8-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add CPUID defines for SGX and SGX Launch Control (LC), as well as
defines for their associated FEATURE_CONTROL MSR bits. Define the
Launch Enclave Public Key Hash MSRs (LE Hash MSRs), which exist
when SGX LC is present (in CPUID), and are writable when SGX LC is
enabled (in FEATURE_CONTROL).
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-7-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Because SGX EPC is enumerated through CPUID, EPC "devices" need to be
realized prior to realizing the vCPUs themselves, i.e. long before
generic devices are parsed and realized. From a virtualization
perspective, the CPUID aspect also means that EPC sections cannot be
hotplugged without paravirtualizing the guest kernel (hardware does
not support hotplugging as EPC sections must be locked down during
pre-boot to provide EPC's security properties).
So even though EPC sections could be realized through the generic
-devices command, they need to be created much earlier for them to
actually be usable by the guest. Place all EPC sections in a
contiguous block, somewhat arbitrarily starting after RAM above 4g.
Ensuring EPC is in a contiguous region simplifies calculations, e.g.
device memory base, PCI hole, etc..., allows dynamic calculation of the
total EPC size, e.g. exposing EPC to guests does not require -maxmem,
and last but not least allows all of EPC to be enumerated in a single
ACPI entry, which is expected by some kernels, e.g. Windows 7 and 8.
The new compound properties command for sgx like below:
......
-object memory-backend-epc,id=mem1,size=28M,prealloc=on \
-object memory-backend-epc,id=mem2,size=10M \
-M sgx-epc.0.memdev=mem1,sgx-epc.1.memdev=mem2
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-6-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
SGX EPC is enumerated through CPUID, i.e. EPC "devices" need to be
realized prior to realizing the vCPUs themselves, which occurs long
before generic devices are parsed and realized. Because of this,
do not allow 'sgx-epc' devices to be instantiated after vCPUS have
been created.
The 'sgx-epc' device is essentially a placholder at this time, it will
be fully implemented in a future patch along with a dedicated command
to create 'sgx-epc' devices.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-5-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add the new 'memory-backend-epc' user creatable QOM object in
the ObjectOptions to support SGX since v6.1, or the sgx backend
object cannot bootup.
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-4-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
EPC (Enclave Page Cahe) is a specialized type of memory used by Intel
SGX (Software Guard Extensions). The SDM desribes EPC as:
The Enclave Page Cache (EPC) is the secure storage used to store
enclave pages when they are a part of an executing enclave. For an
EPC page, hardware performs additional access control checks to
restrict access to the page. After the current page access checks
and translations are performed, the hardware checks that the EPC
page is accessible to the program currently executing. Generally an
EPC page is only accessed by the owner of the executing enclave or
an instruction which is setting up an EPC page.
Because of its unique requirements, Linux manages EPC separately from
normal memory. Similar to memfd, the device /dev/sgx_vepc can be
opened to obtain a file descriptor which can in turn be used to mmap()
EPC memory.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-3-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add new CONFIG_SGX for sgx support in the Qemu, and the Kconfig
default enable sgx in the i386 platform.
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-32-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add a new RAMBlock flag to denote "protected" memory, i.e. memory that
looks and acts like RAM but is inaccessible via normal mechanisms,
including DMA. Use the flag to skip protected memory regions when
mapping RAM for DMA in VFIO.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The Linux spi-imx driver does not work on QEMU. The reason is that the
state of m25p80 loops in STATE_READING_DATA state after receiving
RDSR command, the new command is ignored. Before sending a new command,
CS line should be pulled high to make the state of m25p80 back to IDLE.
Currently the SPI flash CS line is connected to the SPI controller, but
on the real board, it's connected to GPIO3_19. This matches the ecspi1
device node in the board dts.
ecspi1 node in imx6qdl-sabrelite.dtsi:
&ecspi1 {
cs-gpios = <&gpio3 19 GPIO_ACTIVE_LOW>;
pinctrl-names = "default";
pinctrl-0 = <&pinctrl_ecspi1>;
status = "okay";
flash: m25p80@0 {
compatible = "sst,sst25vf016b", "jedec,spi-nor";
spi-max-frequency = <20000000>;
reg = <0>;
};
};
Should connect the SSI_GPIO_CS to GPIO3_19 when adding a spi-nor to
spi1 on sabrelite machine.
Verified this patch on Linux v5.14.
Logs:
# echo "01234567899876543210" > test
# mtd_debug erase /dev/mtd0 0x0 0x1000
Erased 4096 bytes from address 0x00000000 in flash
# mtd_debug write /dev/mtdblock0 0x0 20 test
Copied 20 bytes from test to address 0x00000000 in flash
# mtd_debug read /dev/mtdblock0 0x0 20 test_out
Copied 20 bytes from address 0x00000000 in flash to test_out
# cat test_out
01234567899876543210#
Signed-off-by: Xuzhou Cheng <xuzhou.cheng@windriver.com>
Reported-by: Guenter Roeck <linux@roeck-us.net>
Reviewed-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-id: 20210927142825.491-1-xchengl.cn@gmail.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
The function ide_bus_new() does an in-place initialization. Rename
it to ide_bus_init() to follow our _init vs _new convention.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Corey Minyard <cminyard@mvista.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Acked-by: John Snow <jsnow@redhat.com> (Feel free to merge.)
Message-id: 20210923121153.23754-7-peter.maydell@linaro.org
Rename the "allocate and return" qbus creation function to
qbus_new(), to bring it into line with our _init vs _new convention.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Corey Minyard <cminyard@mvista.com>
Message-id: 20210923121153.23754-6-peter.maydell@linaro.org
Rename qbus_create_inplace() to qbus_init(); this is more in line
with our usual naming convention for functions that in-place
initialize objects.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-id: 20210923121153.23754-5-peter.maydell@linaro.org
Rename the pci_root_bus_new_inplace() function to
pci_root_bus_init(); this brings the bus type in to line with a
"_init for in-place init, _new for allocate-and-return" convention.
To do this we need to rename the implementation-internal function
that was using the pci_root_bus_init() name to
pci_root_bus_internal_init().
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Message-id: 20210923121153.23754-4-peter.maydell@linaro.org