In commit fbf6449f84bf ("x86/sev-es: Set x86_virt_bits to the correct
value straight away, instead of a two-phase approach"), the initialization
of c->x86_phys_bits was moved after this_cpu->c_early_init(c). This is
incorrect because early_init_amd() expected to be able to reduce the
value according to the contents of CPUID leaf 0x8000001f.
Fortunately, the bug was negated by init_amd()'s call to early_init_amd(),
which does reduce x86_phys_bits in the end. However, this is very
late in the boot process and, most notably, the wrong value is used for
x86_phys_bits when setting up MTRRs.
To fix this, call get_cpu_address_sizes() as soon as X86_FEATURE_CPUID is
set/cleared, and c->extended_cpuid_level is retrieved.
Fixes: fbf6449f84bf ("x86/sev-es: Set x86_virt_bits to the correct value straight away, instead of a two-phase approach")
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc:stable@vger.kernel.org
Link: https://lore.kernel.org/all/20240131230902.1867092-2-pbonzini%40redhat.com
Pull x86 TDX updates from Dave Hansen: "This contains the initial support for host-side TDX support so that KVM can run TDX-protected guests. This does not include the actual KVM-side support which will come from the KVM folks. The TDX host interactions with kexec also needs to be ironed out before this is ready for prime time, so this code is currently Kconfig'd off when kexec is on. The majority of the code here is the kernel telling the TDX module which memory to protect and handing some additional memory over to it to use to store TDX module metadata. That sounds pretty simple, but the TDX architecture is rather flexible and it takes quite a bit of back-and-forth to say, "just protect all memory, please." There is also some code tacked on near the end of the series to handle a hardware erratum. The erratum can make software bugs such as a kernel write to TDX-protected memory cause a machine check and masquerade as a real hardware failure. The erratum handling watches out for these and tries to provide nicer user errors" * tag 'x86_tdx_for_6.8' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (21 commits) x86/virt/tdx: Make TDX host depend on X86_MCE x86/virt/tdx: Disable TDX host support when kexec is enabled Documentation/x86: Add documentation for TDX host support x86/mce: Differentiate real hardware #MCs from TDX erratum ones x86/cpu: Detect TDX partial write machine check erratum x86/virt/tdx: Handle TDX interaction with sleep and hibernation x86/virt/tdx: Initialize all TDMRs x86/virt/tdx: Configure global KeyID on all packages x86/virt/tdx: Configure TDX module with the TDMRs and global KeyID x86/virt/tdx: Designate reserved areas for all TDMRs x86/virt/tdx: Allocate and set up PAMTs for TDMRs x86/virt/tdx: Fill out TDMRs to cover all TDX memory regions x86/virt/tdx: Add placeholder to construct TDMRs to cover all TDX memory regions x86/virt/tdx: Get module global metadata for module initialization x86/virt/tdx: Use all system memory when initializing TDX module as TDX memory x86/virt/tdx: Add skeleton to enable TDX on demand x86/virt/tdx: Add SEAMCALL error printing for module initialization x86/virt/tdx: Handle SEAMCALL no entropy error in common code x86/virt/tdx: Make INTEL_TDX_HOST depend on X86_X2APIC x86/virt/tdx: Define TDX supported page sizes as macros ...
Pull x86 asm updates from Ingo Molnar:
"Replace magic numbers in GDT descriptor definitions & handling:
- Introduce symbolic names via macros for descriptor
types/fields/flags, and then use these symbolic names.
- Clean up definitions a bit, such as GDT_ENTRY_INIT()
- Fix/clean up details that became visibly inconsistent after the
symbol-based code was introduced:
- Unify accessed flag handling
- Set the D/B size flag consistently & according to the HW
specification"
* tag 'x86-asm-2024-01-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/asm: Add DB flag to 32-bit percpu GDT entry
x86/asm: Always set A (accessed) flag in GDT descriptors
x86/asm: Replace magic numbers in GDT descriptors, script-generated change
x86/asm: Replace magic numbers in GDT descriptors, preparations
x86/asm: Provide new infrastructure for GDT descriptors
We have no known use for having the CPU track whether GDT descriptors have been accessed or not. Simplify the code by adding the flag to the common flags and removing it everywhere else. Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com> Signed-off-by: Ingo Molnar <mingo@kernel.org> Acked-by: Linus Torvalds <torvalds@linux-foundation.org> Link: https://lore.kernel.org/r/20231219151200.2878271-5-vegard.nossum@oracle.com
Actually replace the numeric values by the new symbolic values.
I used this to find all the existing users of the GDT_ENTRY*() macros:
$ git grep -P 'GDT_ENTRY(_INIT)?\('
Some of the lines will exceed 80 characters, but some of them will be
shorter again in the next couple of patches.
Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/20231219151200.2878271-4-vegard.nossum@oracle.com
We'd like to replace all the magic numbers in various GDT descriptors with new, semantically meaningful, symbolic values. In order to be able to verify that the change doesn't cause any actual changes to the compiled binary code, I've split the change into two patches: - Part 1 (this commit): everything _but_ actually replacing the numbers - Part 2 (the following commit): _only_ replacing the numbers The reason we need this split for verification is that including new headers causes some spurious changes to the object files, mostly line number changes in the debug info but occasionally other subtle codegen changes. Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com> Signed-off-by: Ingo Molnar <mingo@kernel.org> Acked-by: Linus Torvalds <torvalds@linux-foundation.org> Link: https://lore.kernel.org/r/20231219151200.2878271-3-vegard.nossum@oracle.com
Intel Trust Domain Extensions (TDX) protects guest VMs from malicious host and certain physical attacks. A CPU-attested software module called 'the TDX module' runs inside a new isolated memory range as a trusted hypervisor to manage and run protected VMs. Pre-TDX Intel hardware has support for a memory encryption architecture called MKTME. The memory encryption hardware underpinning MKTME is also used for Intel TDX. TDX ends up "stealing" some of the physical address space from the MKTME architecture for crypto-protection to VMs. The BIOS is responsible for partitioning the "KeyID" space between legacy MKTME and TDX. The KeyIDs reserved for TDX are called 'TDX private KeyIDs' or 'TDX KeyIDs' for short. During machine boot, TDX microcode verifies that the BIOS programmed TDX private KeyIDs consistently and correctly programmed across all CPU packages. The MSRs are locked in this state after verification. This is why MSR_IA32_MKTME_KEYID_PARTITIONING gets used for TDX enumeration: it indicates not just that the hardware supports TDX, but that all the boot-time security checks passed. The TDX module is expected to be loaded by the BIOS when it enables TDX, but the kernel needs to properly initialize it before it can be used to create and run any TDX guests. The TDX module will be initialized by the KVM subsystem when KVM wants to use TDX. Detect platform TDX support by detecting TDX private KeyIDs. The TDX module itself requires one TDX KeyID as the 'TDX global KeyID' to protect its metadata. Each TDX guest also needs a TDX KeyID for its own protection. Just use the first TDX KeyID as the global KeyID and leave the rest for TDX guests. If no TDX KeyID is left for TDX guests, disable TDX as initializing the TDX module alone is useless. [ dhansen: add X86_FEATURE, replace helper function ] Signed-off-by: Kai Huang <kai.huang@intel.com> Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com> Reviewed-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Reviewed-by: Isaku Yamahata <isaku.yamahata@intel.com> Reviewed-by: David Hildenbrand <david@redhat.com> Reviewed-by: Dave Hansen <dave.hansen@linux.intel.com> Reviewed-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com> Link: https://lore.kernel.org/all/20231208170740.53979-1-dave.hansen%40intel.com
AMD does not have the requirement for a synchronization barrier when
acccessing a certain group of MSRs. Do not incur that unnecessary
penalty there.
There will be a CPUID bit which explicitly states that a MFENCE is not
needed. Once that bit is added to the APM, this will be extended with
it.
While at it, move to processor.h to avoid include hell. Untangling that
file properly is a matter for another day.
Some notes on the performance aspect of why this is relevant, courtesy
of Kishon VijayAbraham <Kishon.VijayAbraham@amd.com>:
On a AMD Zen4 system with 96 cores, a modified ipi-bench[1] on a VM
shows x2AVIC IPI rate is 3% to 4% lower than AVIC IPI rate. The
ipi-bench is modified so that the IPIs are sent between two vCPUs in the
same CCX. This also requires to pin the vCPU to a physical core to
prevent any latencies. This simulates the use case of pinning vCPUs to
the thread of a single CCX to avoid interrupt IPI latency.
In order to avoid run-to-run variance (for both x2AVIC and AVIC), the
below configurations are done:
1) Disable Power States in BIOS (to prevent the system from going to
lower power state)
2) Run the system at fixed frequency 2500MHz (to prevent the system
from increasing the frequency when the load is more)
With the above configuration:
*) Performance measured using ipi-bench for AVIC:
Average Latency: 1124.98ns [Time to send IPI from one vCPU to another vCPU]
Cumulative throughput: 42.6759M/s [Total number of IPIs sent in a second from
48 vCPUs simultaneously]
*) Performance measured using ipi-bench for x2AVIC:
Average Latency: 1172.42ns [Time to send IPI from one vCPU to another vCPU]
Cumulative throughput: 40.9432M/s [Total number of IPIs sent in a second from
48 vCPUs simultaneously]
From above, x2AVIC latency is ~4% more than AVIC. However, the expectation is
x2AVIC performance to be better or equivalent to AVIC. Upon analyzing
the perf captures, it is observed significant time is spent in
weak_wrmsr_fence() invoked by x2apic_send_IPI().
With the fix to skip weak_wrmsr_fence()
*) Performance measured using ipi-bench for x2AVIC:
Average Latency: 1117.44ns [Time to send IPI from one vCPU to another vCPU]
Cumulative throughput: 42.9608M/s [Total number of IPIs sent in a second from
48 vCPUs simultaneously]
Comparing the performance of x2AVIC with and without the fix, it can be seen
the performance improves by ~4%.
Performance captured using an unmodified ipi-bench using the 'mesh-ipi' option
with and without weak_wrmsr_fence() on a Zen4 system also showed significant
performance improvement without weak_wrmsr_fence(). The 'mesh-ipi' option ignores
CCX or CCD and just picks random vCPU.
Average throughput (10 iterations) with weak_wrmsr_fence(),
Cumulative throughput: 4933374 IPI/s
Average throughput (10 iterations) without weak_wrmsr_fence(),
Cumulative throughput: 6355156 IPI/s
[1] https://github.com/bytedance/kvm-utils/tree/master/microbenchmark/ipi-bench
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230622095212.20940-1-bp@alien8.de
Pull x86 microcode loading updates from Borislac Petkov:
"Major microcode loader restructuring, cleanup and improvements by
Thomas Gleixner:
- Restructure the code needed for it and add a temporary initrd
mapping on 32-bit so that the loader can access the microcode
blobs. This in itself is a preparation for the next major
improvement:
- Do not load microcode on 32-bit before paging has been enabled.
Handling this has caused an endless stream of headaches, issues,
ugly code and unnecessary hacks in the past. And there really
wasn't any sensible reason to do that in the first place. So switch
the 32-bit loading to happen after paging has been enabled and turn
the loader code "real purrty" again
- Drop mixed microcode steppings loading on Intel - there, a single
patch loaded on the whole system is sufficient
- Rework late loading to track which CPUs have updated microcode
successfully and which haven't, act accordingly
- Move late microcode loading on Intel in NMI context in order to
guarantee concurrent loading on all threads
- Make the late loading CPU-hotplug-safe and have the offlined
threads be woken up for the purpose of the update
- Add support for a minimum revision which determines whether late
microcode loading is safe on a machine and the microcode does not
change software visible features which the machine cannot use
anyway since feature detection has happened already. Roughly, the
minimum revision is the smallest revision number which must be
loaded currently on the system so that late updates can be allowed
- Other nice leanups, fixess, etc all over the place"
* tag 'x86_microcode_for_v6.7_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (40 commits)
x86/microcode/intel: Add a minimum required revision for late loading
x86/microcode: Prepare for minimal revision check
x86/microcode: Handle "offline" CPUs correctly
x86/apic: Provide apic_force_nmi_on_cpu()
x86/microcode: Protect against instrumentation
x86/microcode: Rendezvous and load in NMI
x86/microcode: Replace the all-in-one rendevous handler
x86/microcode: Provide new control functions
x86/microcode: Add per CPU control field
x86/microcode: Add per CPU result state
x86/microcode: Sanitize __wait_for_cpus()
x86/microcode: Clarify the late load logic
x86/microcode: Handle "nosmt" correctly
x86/microcode: Clean up mc_cpu_down_prep()
x86/microcode: Get rid of the schedule work indirection
x86/microcode: Mop up early loading leftovers
x86/microcode/amd: Use cached microcode for AP load
x86/microcode/amd: Cache builtin/initrd microcode early
x86/microcode/amd: Cache builtin microcode too
x86/microcode/amd: Use correct per CPU ucode_cpu_info
...
Pull x86 core updates from Thomas Gleixner: - Limit the hardcoded topology quirk for Hygon CPUs to those which have a model ID less than 4. The newer models have the topology CPUID leaf 0xB correctly implemented and are not affected. - Make SMT control more robust against enumeration failures SMT control was added to allow controlling SMT at boottime or runtime. The primary purpose was to provide a simple mechanism to disable SMT in the light of speculation attack vectors. It turned out that the code is sensible to enumeration failures and worked only by chance for XEN/PV. XEN/PV has no real APIC enumeration which means the primary thread mask is not set up correctly. By chance a XEN/PV boot ends up with smp_num_siblings == 2, which makes the hotplug control stay at its default value "enabled". So the mask is never evaluated. The ongoing rework of the topology evaluation caused XEN/PV to end up with smp_num_siblings == 1, which sets the SMT control to "not supported" and the empty primary thread mask causes the hotplug core to deny the bringup of the APS. Make the decision logic more robust and take 'not supported' and 'not implemented' into account for the decision whether a CPU should be booted or not. - Fake primary thread mask for XEN/PV Pretend that all XEN/PV vCPUs are primary threads, which makes the usage of the primary thread mask valid on XEN/PV. That is consistent with because all of the topology information on XEN/PV is fake or even non-existent. - Encapsulate topology information in cpuinfo_x86 Move the randomly scattered topology data into a separate data structure for readability and as a preparatory step for the topology evaluation overhaul. - Consolidate APIC ID data type to u32 It's fixed width hardware data and not randomly u16, int, unsigned long or whatever developers decided to use. - Cure the abuse of cpuinfo for persisting logical IDs. Per CPU cpuinfo is used to persist the logical package and die IDs. That's really not the right place simply because cpuinfo is subject to be reinitialized when a CPU goes through an offline/online cycle. Use separate per CPU data for the persisting to enable the further topology management rework. It will be removed once the new topology management is in place. - Provide a debug interface for inspecting topology information Useful in general and extremly helpful for validating the topology management rework in terms of correctness or "bug" compatibility. * tag 'x86-core-2023-10-29-v2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (23 commits) x86/apic, x86/hyperv: Use u32 in hv_snp_boot_ap() too x86/cpu: Provide debug interface x86/cpu/topology: Cure the abuse of cpuinfo for persisting logical ids x86/apic: Use u32 for wakeup_secondary_cpu[_64]() x86/apic: Use u32 for [gs]et_apic_id() x86/apic: Use u32 for phys_pkg_id() x86/apic: Use u32 for cpu_present_to_apicid() x86/apic: Use u32 for check_apicid_used() x86/apic: Use u32 for APIC IDs in global data x86/apic: Use BAD_APICID consistently x86/cpu: Move cpu_l[l2]c_id into topology info x86/cpu: Move logical package and die IDs into topology info x86/cpu: Remove pointless evaluation of x86_coreid_bits x86/cpu: Move cu_id into topology info x86/cpu: Move cpu_core_id into topology info hwmon: (fam15h_power) Use topology_core_id() scsi: lpfc: Use topology_core_id() x86/cpu: Move cpu_die_id into topology info x86/cpu: Move phys_proc_id into topology info x86/cpu: Encapsulate topology information in cpuinfo_x86 ...