Update KVM-TDX.README.md with BKC reference wiki link #2

hongyuni · 2022-06-29T09:14:37Z

Update KVM-TDX.README.md with BKC reference wiki link info

The next step of TDX guest creation is to create vcpu. Allocate TDX vcpu structures, initialize it. Allocate pages of TDX vcpu for the TDX module. In the case of the conventional case, cpuid is empty at the initialization. and cpuid is configured after the vcpu initialization. Because TDX supports only X2APIC mode, cpuid is forcibly initialized to support X2APIC on the vcpu initialization. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

TD guest vcpu need to be configured before ready to run which requests addtional information from Device model (e.g. qemu), one 64bit value is passed to vcpu's RCX as an initial value. Repurpose KVM_MEMORY_ENCRYPT_OP to vcpu-scope and add new sub-commands KVM_TDX_INIT_VCPU under it for such additional vcpu configuration. Add callback for kvm vCPU-scoped operations of KVM_MEMORY_ENCRYPT_OP and add a new subcommand, KVM_TDX_INIT_VCPU, for further vcpu initialization. Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com> Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

This empty commit is to mark the start of patch series of KVM MMU GPA shared bits. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

To Keep the case of non TDX intact, introduce a new config option for private KVM MMU support. At the moment, this is synonym for CONFIG_INTEL_TDX_HOST && CONFIG_KVM_INTEL. The new flag make it clear that the config is only for x86 KVM MMU. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

TDX repurposes one GPA bits (51 bit or 47 bit based on configuration) to indicate the GPA is private(if cleared) or shared (if set) with VMM. If GPA.shared is set, GPA is converted existing conventional EPT pointed by EPTP. If GPA.shared bit is cleared, GPA is converted by Secure-EPT(S-EPT) TDX module manages. VMM has to issue SEAM call to TDX module to operate on S-EPT. e.g. populating/zapping guest page or shadow page by TDH.PAGE.{ADD, REMOVE} for guest page, TDH.PAGE.SEPT.{ADD, REMOVE} S-EPT etc. Several hooks needs to be added to KVM MMU to support TDX. Add a function to check if KVM MMU is running for TDX and several functions for address conversation between private-GPA and shared-GPA. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

This empty commit is to mark the start of patch series of KVM TDP refactoring for TDX. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

Explicitly check for an MMIO spte in the fast page fault flow. TDX will use a not-present entry for MMIO sptes, which can be mistaken for an access-tracked spte since both have SPTE_SPECIAL_MASK set. MMIO sptes are handled in handle_mmio_page_fault for non-TDX VMs, so this patch does not affect them. TDX will handle MMIO emulation through a hypercall instead. Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com> Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

TDX introduced a new ETP, Secure-EPT, in addition to the existing EPT. Secure-EPT maps protected guest memory, which is called private. Since Secure-EPT page tables is also protected, those page tables is also called private. The existing EPT is often called shared EPT to distinguish from Secure-EPT. And also page tables for share EPT is also called shared. Virtualization Exception, #VE, is a new processor exception in VMX non-root operation. In certain virtualizatoin-related conditions, #VE is injected into guest instead of exiting from guest to VMM so that guest is given a chance to inspect it. One important one is EPT violation. When "ETP-violation #VE" VM-execution is set, "#VE suppress bit" in EPT entry is cleared, #VE is injected instead of EPT violation. Because guest memory is protected with TDX, VMM can't parse instructions in the guest memory. Instead, MMIO hypercall is used for guest to pass necessary information to VMM. To make unmodified device driver work, guest TD expects #VE on accessing shared GPA. The #VE handler converts MMIO access into MMIO hypercall with the EPT entry of enabled "#VE" by clearing "suppress #VE" bit. Before VMM enabling #VE, it needs to figure out the given GPA is for MMIO by EPT violation. So the execution flow looks like - Allocate unused shared EPT entry with suppress #VE bit set. - EPT violation on that GPA. - VMM figures out the faulted GPA is for MMIO. - VMM clears the suppress #VE bit. - Guest TD gets #VE, and converts MMIO access into MMIO hypercall. - If the GPA maps guest memory, VMM resolves it with guest pages. For both cases, SPTE needs suppress #VE" bit set initially when it is allocated or zapped, therefore non-zero non-present value for SPTE needs to be allowed. This change requires to update FNAME(sync_page) for shadow EPT. "if(!sp->spte[i])" in FNAME(sync_page) means that the spte entry is the initial value. With the introduction of shadow_nonpresent_value which can be non-zero, it doesn't hold any more. Replace zero check with "!is_shadow_present_pte() && !is_mmio_spte()". When "if (!spt[i])" doesn't hold, but the entry value is shadow_nonpresent_value, the entry is wrongly synchronized from non-present to non-present with (wrongly) pfn changed and tries to remove rmap wrongly and BUG_ON() is hit. TDP MMU uses REMOVED_SPTE = 0x5a0ULL as special constant to indicate the intermediate value to indicate one thread is operating on it and the value should be semi-arbitrary value. For TDX (more correctly to use #VE), the value should include suppress #VE value which is SHADOW_NONPRESENT_VALUE. Rename REMOVED_SPTE to __REMOVED_SPTE and define REMOVED_SPTE as SHADOW_NONPRESENT_VALUE | REMOVED_SPTE to set suppress #VE bit. Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com> Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

TDX will use a different shadow PTE entry value for MMIO from VMX. Add members to kvm_arch and track value for MMIO per-VM instead of global variables. By using the per-VM EPT entry value for MMIO, the existing VMX logic is kept working. In the case of VMX VM case, the EPT entry for MMIO is non-present PTE (present bit cleared) without backing guest physical address (on EPT violation, KVM searches backing guest memory and it finds there is no backing guest page.) or the value to trigger EPT misconfiguration. Once MMIO is triggered on the EPT entry, the EPT entry is updated to trigger EPT misconfiguration for the future MMIO on the same GPA. It allows KVM to understand the memory access is for MMIO without searching backing guest pages.). And then KVM parses guest instruction to figure out address/value/width for MMIO. In the case of the guest TD, the guest memory is protected so that VMM can't parse guest instruction to understand the value and access width for MMIO. Instead, VMM sets up (Shared) EPT to trigger #VE by clearing the VE-suppress bit. When the guest TD issues MMIO, #VE is injected. Guest VE handler converts MMIO access into MMIO hypercall to pass address/value/width for MMIO to VMM. (or directly paravirtualize MMIO into hypercall.) Then VMM can handle the MMIO hypercall without parsing guest instructions. Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com> Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

TDX requires TDX SEAMCALL to operate Secure EPT instead of direct memory access and TDX SEAMCALL is heavy operation. Fast page fault on private GPA doesn't make sense. Disallow fast page fault on private GPA. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>

TODO: This is a transient workaround patch until the large page support for TDX is implemented. Support large page for TDX and remove this patch. At this point, large page for TDX isn't supported, and need to allow guest TD to work only with 4K pages. On the other hand, conventional VMX VMs should continue to work with large page. Allow per-VM override of the TDP max page level. In the existing x86 KVM MMU code, there is already max_level member in struct kvm_page_fault with KVM_MAX_HUGEPAGE_LEVEL initial value. The KVM page fault handler denies page size larger than max_level. Add per-VM member to indicate the allowed maximum page size with KVM_MAX_HUGEPAGE_LEVEL as default value and initialize max_level in struct kvm_page_fault with it. For the guest TD, the set per-VM value for allows maximum page size to 4K page size. Then only allowed page size is 4K. It means large page is disabled. Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com> Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

…te mmu For kvm mmu that has shared bit mask, zap only leaf SPTEs when deleting/moving a memslot. The existing kvm_mmu_zap_memslot() depends on role.invalid with read lock of mmu_lock so that other vcpu can operate on kvm mmu concurrently. Mark the root page table invalid, unlink it from page table pointer of CPU, process the page table. It doesn't work for private page table to unlink the root page table because it requires all SPTE entry to be non-present. Instead, with write-lock of mmu_lock and zap only leaf SPTEs for kvm mmu with shared bit mask. Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com> Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

To support TDX, KVM is enhanced to operate with #VE. For TDX, KVM programs to inject #VE conditionally and set #VE suppress bit in EPT entry. For VMX case, #VE isn't used. If #VE happens for VMX, it's a bug. To be defensive (test that VMX case isn't broken), introduce option ept_violation_ve_test and when it's set, set error. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

This empty commit is to mark the start of patch series of KVM TDP MMU hooks. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

In this patch series, TDX supports only TDP MMU and doesn't support legacy MMU. Forcibly use TDP MMU for TDX irrelevant of kernel parameter to disable TDP MMU. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

For private GPA, CPU refers a private page table whose contents are encrypted. The dedicated APIs to operate on it (e.g. updating/reading its PTE entry) are used and their cost is expensive. When KVM resolves KVM page fault, it walks the page tables. To reuse the existing KVM MMU code and mitigate the heavy cost to directly walk encrypted private page table, allocate a more page to mirror the existing KVM page table. Resolve KVM page fault with the existing code, and do additional operations necessary for the mirrored private page table. To distinguish such cases, the existing KVM page table is called a shared page table (i.e. no mirrored private page table), and the KVM page table with mirrored private page table is called a private page table. The relationship is depicted below. Add private pointer to struct kvm_mmu_page for mirrored private page table and add helper functions to allocate/initialize/free a mirrored private page table page. Also, add helper functions to check if a given kvm_mmu_page is private. The later patch introduces hooks to operate on the mirrored private page table. KVM page fault | | | V | -------------+---------- | | | | V V | shared GPA private GPA | | | | V V | CPU/KVM shared PT root KVM private PT root | CPU private PT root | | | | V V | V shared PT private PT <----mirror----> mirrored private PT | | | | | \-----------------+------\ | | | | | V | V V shared guest page | private guest page | non-encrypted memory | encrypted memory | PT: page table Both CPU and KVM refer to CPU/KVM shared page table. Private page table is used only by KVM. CPU refers to mirrored private page table. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

Factor out non-leaf SPTE population logic from kvm_tdp_mmu_map(). MapGPA hypercall needs to populate non-leaf SPTE to record which GPA, private or shared, is allowed in the leaf EPT entry. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

Allocate mirrored private page table for private page table, and add hooks to operate on mirrored private page table. This patch adds only hooks. As kvm_gfn_shared_mask() returns false always, those hooks aren't called yet. Because private guest page is protected, page copy with mmu_notifier to migrate page doesn't work. Callback from backing store is needed. When the faulting GPA is private, the KVM fault is also called private. When resolving private KVM, allocate mirrored private page table and call hooks to operate on mirrored private page table. On the change of the private PTE entry, invoke kvm_x86_ops hook in __handle_changed_spte() to propagate the change to mirrored private page table. The following depicts the relationship. private KVM page fault | | | V | private GPA | | | V | KVM private PT root | CPU private PT root | | | V | V private PT ---hook to mirror--->mirrored private PT | | | \--------------------+------\ | | | | | V V | private guest page | | non-encrypted memory | encrypted memory | PT: page table The existing KVM TDP MMU code uses atomic update of SPTE. On populating the EPT entry, atomically set the entry. However, it requires TLB shootdown to zap SPTE. To address it, the entry is frozen with the special SPTE value that clears the present bit. After the TLB shootdown, the entry is set to the eventual value (unfreeze). For mirrored private page table, hooks are called to update mirrored private page table in addition to direct access to the private SPTE. For the zapping case, it works to freeze the SPTE. It can call hooks in addition to TLB shootdown. For populating the private SPTE entry, there can be a race condition without further protection vcpu 1: populating 2M private SPTE vcpu 2: populating 4K private SPTE vcpu 2: TDX SEAMCALL to update 4K mirrored private SPTE => error vcpu 1: TDX SEAMCALL to update 2M mirrored private SPTE To avoid the race, the frozen SPTE is utilized. Instead of atomic update of the private entry, freeze the entry, call the hook that update mirrored private SPTE, set the entry to the final value. Support 4K page only at this stage. 2M page support can be done in future patches. Add is_private member to kvm_page_fault to indicate the fault is private. Also is_private member to struct tdp_inter to propagate it. Co-developed-by: Kai Huang <kai.huang@intel.com> Signed-off-by: Kai Huang <kai.huang@intel.com> Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

This empty commit is to mark the start of patch series of TDX EPT violation. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

TDX doesn't support dirty logging. Report dirty logging isn't supported so that device model, for example qemu, can properly handle it. Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com> Signed-off-by: Xiaoyao Li <xiaoyao.li@intel.com> Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>

Some KVM MMU operations (dirty page logging, page migration, aging page) aren't supported for private GFNs (yet) with the first generation of TDX. Silently return on unsupported TDX KVM MMU operations. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

The difference of TDX EPT violation is how to retrieve information, GPA, and exit qualification. To share the code to handle EPT violation, split out the guts of EPT violation handler so that VMX/TDX exit handler can call it after retrieving GPA and exit qualification. Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com> Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>

EPT MMU masks are used commonly for VMX and TDX. The value needs to be initialized in common code before both VMX/TDX-specific initialization code. Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com> Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

For virtual IO, the guest TD shares guest pages with VMM without encryption. Shared EPT is used to map guest pages in unprotected way. Add the VMCS field encoding for the shared EPTP, which will be used by TDX to have separate EPT walks for private GPAs (existing EPTP) versus shared GPAs (new shared EPTP). Set shared EPT pointer value for the TDX guest to initialize TDX MMU. Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com> Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>

TDX doesn't need APIC page depending on vapic and its callback is WARN_ON_ONCE(is_tdx). To avoid unnecessary overhead and WARN_ON_ONCE(), skip requesting KVM_REQ_APIC_PAGE_RELOAD when TD. ------------[ cut here ]------------ WARNING: CPU: 134 PID: 42205 at arch/x86/kvm/vmx/main.c:696 vt_set_apic_access_page_addr+0x3c/0x50 [kvm_intel] Modules linked in: squashfs nls_iso8859_1 nls_cp437 vhost_vsock vhost vhost_iotlb tdx_debug kvm_intel kvm irqbypass crct10dif_pclmul crc32_pclmul ghash_clmulni_intel aesni_intel crypto_simd cryptd i2c_i801 i2c_smbus i2c_ismt CPU: 134 PID: 42205 Comm: tdx_vm_tests Tainted: G W 5.17.0-rc8 #165 4baba67c36c7c1001d782c47f2964b779a5659c7 Hardware name: Intel Corporation EAGLESTREAM/EAGLESTREAM, BIOS EGSDCRB1.SYS.0066.D24.2110072326 10/07/2021 RIP: 0010:vt_set_apic_access_page_addr+0x3c/0x50 [kvm_intel] Code: e7 d5 49 8b 1c 24 48 8d bb 78 15 00 00 e8 4c 78 e7 d5 48 83 bb 78 15 00 00 01 74 0d 4c 89 e7 e8 7a 9b fd ff 5b 41 5c 5d c3 90 <0f 0b 90 5b 41 5c 5d c3 66 66 2e 0f 1f 84 00 00 00 00 00 90 0f 1f RSP: 0018:ffa0000027477b68 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffa00000572d9000 RCX: ffffffffde6864d4 RDX: dffffc0000000000 RSI: 0000000000000008 RDI: ffa00000572da578 RBP: ffa0000027477b78 R08: 0000000000000001 R09: ffe21c006df80008 R10: ff1100036fc0003f R11: ffe21c006df80007 R12: ff1100036fc00000 R13: ff1100036fc000d8 R14: ff1100036fc00038 R15: ff1100036fc00000 FS: 00007fdf1ad32740(0000) GS:ff11000e1ed00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007fdf15f1b000 CR3: 000000011e462005 CR4: 0000000000773ee0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe07f0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: <TASK> vcpu_enter_guest+0x145d/0x24d0 [kvm] ? inject_pending_event+0x750/0x750 [kvm] ? xsaves+0x31/0x40 ? rcu_read_lock_held_common+0x1e/0x60 ? rcu_read_lock_sched_held+0x60/0xe0 ? rcu_read_lock_bh_held+0xc0/0xc0 kvm_arch_vcpu_ioctl_run+0x25d/0xcc0 [kvm] kvm_vcpu_ioctl+0x414/0xa30 [kvm]] ? kvm_clear_dirty_log_protect+0x4d0/0x4d0 [kvm] ? userfaultfd_unmap_prep+0x240/0x240 ? __up_read+0x17f/0x530 ? rwsem_wake+0x110/0x110 ? __do_munmap+0x437/0x7c0 ? rcu_read_lock_held_common+0x1e/0x60 ? rcu_read_lock_sched_held+0x60/0xe0 ? rcu_read_lock_sched_held+0x60/0xe0 ? __kasan_check_read+0x11/0x20 ? __fget_light+0xa9/0x100 __x64_sys_ioctl+0xc0/0x100 do_syscall_64+0x39/0xc0 entry_SYSCALL_64_after_hwframe+0x44/0xae RIP: 0033:0x7fdf1ae493db Code: 0f 1e fa 48 8b 05 b5 7a 0d 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 0f 1f 44 00 00 f3 0f 1e fa b8 10 00 00 00 0f 05 <48 3d 01 f0 ff ff 73 01 c3 48 8b 0d 85 7a 0d 00 f7 d8 64 89 01 48 RSP: 002b:00007ffcf8bdfb38 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 RAX: ffffffffffffffda RBX: 00000000006f26d0 RCX: 00007fdf1ae493db RDX: 0000000000000000 RSI: 000000000000ae80 RDI: 0000000000000007 RBP: 0000000000000000 R08: 0000000000411d36 R09: 0000000000000000 R10: fffffffffffffb69 R11: 0000000000000246 R12: 0000000000402410 R13: 00000000006f02b0 R14: 0000000000000000 R15: 0000000000000000 </TASK> irq event stamp: 0 hardirqs last enabled at (0): [<0000000000000000>] 0x0 hardirqs last disabled at (0): [<ffffffffb40c809a>] copy_process+0xaca/0x3270 softirqs last enabled at (0): [<ffffffffb40c809a>] copy_process+0xaca/0x3270 softirqs last disabled at (0): [<0000000000000000>] 0x0 ---[ end trace 0000000000000000 ]--- Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

Implement hooks of TDP MMU for TDX backend. TLB flush, TLB shootdown, propagating the change private EPT entry to Secure EPT and freeing Secure EPT page. TLB flush handles both shared EPT and private EPT. It flushes shared EPT same as VMX. It also waits for the TDX TLB shootdown. For the hook to free Secure EPT page, unlinks the Secure EPT page from the Secure EPT so that the page can be freed to OS. Propagating the entry change to Secure EPT. The possible entry changes are present -> non-present(zapping) and non-present -> present(population). On population just link the Secure EPT page or the private guest page to the Secure EPT by TDX SEAMCALL. Because TDP MMU allows concurrent zapping/population, zapping requires synchronous TLB shootdown with the frozen EPT entry. It zaps the secure entry, increments TLB counter, sends IPI to remote vcpus to trigger TLB flush, and then unlinks the private guest page from the Secure EPT. For simplicity, batched zapping with exclude lock is handled as concurrent zapping. Although it's inefficient, it can be optimized in the future. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

This empty commit is to mark the start of patch series of KVM TDP MMU MapGPA. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

…d or not With TDX, all GFNs are private at guest boot time. At run time guest TD can explicitly change it to shared from private or vice-versa by MapGPA hypercall. If it's specified, the given GFN can't be used as otherwise. That's is, if a guest tells KVM that the GFN is shared, it can't be used as private. or vice-versa. Steal software usable bit, SPTE_SHARED_MASK, for it from MMIO counter to record it. Use the bit SPTE_SHARED_MASK in shared or private EPT to determine which mapping, shared or private, is allowed. If requested mapping isn't allowed, return RET_PF_RETRY to wait for other vcpu to change it. The bit is recorded in both shared and private shadow page to avoid traverse one more shadow page when resolving KVM page fault. The bit needs to be kept over zapping the EPT entry. Currently the EPT entry is initialized SHADOW_NONPRESENT_VALUE unconditionally to clear SPTE_SHARED_MASK bit. To carry SPTE_SHARED_MASK bit, introduce a helper function to get initial value for zapped entry with SPTE_SHARED_MASK bit. Replace SHADOW_NONPRESENT_VALUE with it. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

The TDX Guest-Hypervisor communication interface(GHCI) specification defines MapGPA hypercall for guest TD to request the host VMM to map given GPA range as private or shared. It means the guest TD uses the GPA as shared (or private). The GPA won't be used as private (or shared). VMM should enforce GPA usage. VMM doesn't have to map the GPA on the hypercall request. - Allocate 4k PTE to record SPTE_SHARED_MASK bit. - Zap the aliased region. If shared (or private) GPA is requested, zap private (or shared) GPA (modulo shared bit). - Record the request GPA is shared (or private) by SPTE_SHARED_MASK in SPTE in both shared and private EPT tables. - With SPTE_SHARED_MASK set, a shared GPA is allowed. - With SPTE_SHARED_MASK cleared, a private GPA is allowed. The reason to record SPTE_SHARED_MASK in both shared and private EPT is to optimize EPT violation path for normal guest TD execution path and penalize map_gpa hypercall. If the guest TD faults on not-allowed GPA (modulo shared bit), the KVM doesn't resolve EPT violation and let vcpu retry. vcpu will keep faulting until other vcpu maps the region with MapGPA hypercall. With the nonpresent value of spte(shadow_nonpresent_value), SPTE_SHARED_MASK is cleared. So the default behavior doesn't change. - don't map GPA. The GPA is mapped on the next EPT violation. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

Introduce a helper to directly (pun intended) fault-in a TDP page without having to go through the full page fault path. This allows TDX to get the resulting pfn and also allows the RET_PF_* enums to stay in mmu.c where they belong. Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com> Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

Intel TDX protects guest VM's from malicious host and certain physical attacks. TDX introduces a new operation mode, Secure Arbitration Mode (SEAM) to isolate and protect guest VM's. A TDX guest VM runs in SEAM and, unlike VMX, direct control and interaction with the guest by the host VMM is not possible. Instead, Intel TDX Module, which also runs in SEAM, provides a SEAMCALL API. The SEAMCALL that provides the ability to enter a guest is TDH.VP.ENTER. The TDX Module processes TDH.VP.ENTER, and enters the guest via VMX VMLAUNCH/VMRESUME instructions. When a guest VM-exit requires host VMM interaction, the TDH.VP.ENTER SEAMCALL returns to the host VMM (KVM). Add tdh_vp_enter() to wrap the SEAMCALL invocation of TDH.VP.ENTER. TDH.VP.ENTER is different from other SEAMCALLS in several ways: - it may take some time to return as the guest executes - it uses more arguments - after it returns some host state may need to be restored TDH.VP.ENTER arguments are passed through General Purpose Registers (GPRs). For the special case of the TD guest invoking TDG.VP.VMCALL, nearly any GPR can be used, as well as XMM0 to XMM15. Notably, RBP is not used, and Linux mandates the TDX Module feature NO_RBP_MOD, which is enforced elsewhere. Additionally, XMM registers are not required for the existing Guest Hypervisor Communication Interface and are handled by existing KVM code should they be modified by the guest. There are 2 input formats and 5 output formats for TDH.VP.ENTER arguments. Input #1 : Initial entry or following a previous async. TD Exit Input #2 : Following a previous TDCALL(TDG.VP.VMCALL) Output #1 : On Error (No TD Entry) Output #2 : Async. Exits with a VMX Architectural Exit Reason Output #3 : Async. Exits with a non-VMX TD Exit Status Output #4 : Async. Exits with Cross-TD Exit Details Output #5 : On TDCALL(TDG.VP.VMCALL) Currently, to keep things simple, the wrapper function does not attempt to support different formats, and just passes all the GPRs that could be used. The GPR values are held by KVM in the area set aside for guest GPRs. KVM code uses the guest GPR area (vcpu->arch.regs[]) to set up for or process results of tdh_vp_enter(). Therefore changing tdh_vp_enter() to use more complex argument formats would also alter the way KVM code interacts with tdh_vp_enter(). Signed-off-by: Kai Huang <kai.huang@intel.com> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Message-ID: <20241121201448.36170-2-adrian.hunter@intel.com> Acked-by: Dave Hansen <dave.hansen@linux.intel.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Intel TDX protects guest VM's from malicious host and certain physical attacks. TDX introduces a new operation mode, Secure Arbitration Mode (SEAM) to isolate and protect guest VM's. A TDX guest VM runs in SEAM and, unlike VMX, direct control and interaction with the guest by the host VMM is not possible. Instead, Intel TDX Module, which also runs in SEAM, provides a SEAMCALL API. The SEAMCALL that provides the ability to enter a guest is TDH.VP.ENTER. The TDX Module processes TDH.VP.ENTER, and enters the guest via VMX VMLAUNCH/VMRESUME instructions. When a guest VM-exit requires host VMM interaction, the TDH.VP.ENTER SEAMCALL returns to the host VMM (KVM). Add tdh_vp_enter() to wrap the SEAMCALL invocation of TDH.VP.ENTER; tdh_vp_enter() needs to be noinstr because VM entry in KVM is noinstr as well, which is for two reasons: * marking the area as CT_STATE_GUEST via guest_state_enter_irqoff() and guest_state_exit_irqoff() * IRET must be avoided between VM-exit and NMI handling, in order to avoid prematurely releasing the NMI inhibit. TDH.VP.ENTER is different from other SEAMCALLs in several ways: it uses more arguments, and after it returns some host state may need to be restored. Therefore tdh_vp_enter() uses __seamcall_saved_ret() instead of __seamcall_ret(); since it is the only caller of __seamcall_saved_ret(), it can be made noinstr also. TDH.VP.ENTER arguments are passed through General Purpose Registers (GPRs). For the special case of the TD guest invoking TDG.VP.VMCALL, nearly any GPR can be used, as well as XMM0 to XMM15. Notably, RBP is not used, and Linux mandates the TDX Module feature NO_RBP_MOD, which is enforced elsewhere. Additionally, XMM registers are not required for the existing Guest Hypervisor Communication Interface and are handled by existing KVM code should they be modified by the guest. There are 2 input formats and 5 output formats for TDH.VP.ENTER arguments. Input #1 : Initial entry or following a previous async. TD Exit Input #2 : Following a previous TDCALL(TDG.VP.VMCALL) Output #1 : On Error (No TD Entry) Output #2 : Async. Exits with a VMX Architectural Exit Reason Output #3 : Async. Exits with a non-VMX TD Exit Status Output #4 : Async. Exits with Cross-TD Exit Details Output #5 : On TDCALL(TDG.VP.VMCALL) Currently, to keep things simple, the wrapper function does not attempt to support different formats, and just passes all the GPRs that could be used. The GPR values are held by KVM in the area set aside for guest GPRs. KVM code uses the guest GPR area (vcpu->arch.regs[]) to set up for or process results of tdh_vp_enter(). Therefore changing tdh_vp_enter() to use more complex argument formats would also alter the way KVM code interacts with tdh_vp_enter(). Signed-off-by: Kai Huang <kai.huang@intel.com> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Message-ID: <20241121201448.36170-2-adrian.hunter@intel.com> Acked-by: Dave Hansen <dave.hansen@linux.intel.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

into HEAD KVM x86 fixes for 6.14-rcN #2 - Set RFLAGS.IF in C code on SVM to get VMRUN out of the STI shadow. - Ensure DEBUGCTL is context switched on AMD to avoid running the guest with the host's value, which can lead to unexpected bus lock #DBs. - Suppress DEBUGCTL.BTF on AMD (to match Intel), as KVM doesn't properly emulate BTF. KVM's lack of context switching has meant BTF has always been broken to some extent. - Always save DR masks for SNP vCPUs if DebugSwap is *supported*, as the guest can enable DebugSwap without KVM's knowledge. - Fix a bug in mmu_stress_tests where a vCPU could finish the "writes to RO memory" phase without actually generating a write-protection fault. - Fix a printf() goof in the SEV smoke test that causes build failures with -Werror. - Explicitly zero EAX and EBX in CPUID.0x8000_0022 output when PERFMON_V2 isn't supported by KVM.

Intel TDX protects guest VM's from malicious host and certain physical attacks. TDX introduces a new operation mode, Secure Arbitration Mode (SEAM) to isolate and protect guest VM's. A TDX guest VM runs in SEAM and, unlike VMX, direct control and interaction with the guest by the host VMM is not possible. Instead, Intel TDX Module, which also runs in SEAM, provides a SEAMCALL API. The SEAMCALL that provides the ability to enter a guest is TDH.VP.ENTER. The TDX Module processes TDH.VP.ENTER, and enters the guest via VMX VMLAUNCH/VMRESUME instructions. When a guest VM-exit requires host VMM interaction, the TDH.VP.ENTER SEAMCALL returns to the host VMM (KVM). Add tdh_vp_enter() to wrap the SEAMCALL invocation of TDH.VP.ENTER; tdh_vp_enter() needs to be noinstr because VM entry in KVM is noinstr as well, which is for two reasons: * marking the area as CT_STATE_GUEST via guest_state_enter_irqoff() and guest_state_exit_irqoff() * IRET must be avoided between VM-exit and NMI handling, in order to avoid prematurely releasing the NMI inhibit. TDH.VP.ENTER is different from other SEAMCALLs in several ways: it uses more arguments, and after it returns some host state may need to be restored. Therefore tdh_vp_enter() uses __seamcall_saved_ret() instead of __seamcall_ret(); since it is the only caller of __seamcall_saved_ret(), it can be made noinstr also. TDH.VP.ENTER arguments are passed through General Purpose Registers (GPRs). For the special case of the TD guest invoking TDG.VP.VMCALL, nearly any GPR can be used, as well as XMM0 to XMM15. Notably, RBP is not used, and Linux mandates the TDX Module feature NO_RBP_MOD, which is enforced elsewhere. Additionally, XMM registers are not required for the existing Guest Hypervisor Communication Interface and are handled by existing KVM code should they be modified by the guest. There are 2 input formats and 5 output formats for TDH.VP.ENTER arguments. Input #1 : Initial entry or following a previous async. TD Exit Input #2 : Following a previous TDCALL(TDG.VP.VMCALL) Output #1 : On Error (No TD Entry) Output #2 : Async. Exits with a VMX Architectural Exit Reason Output #3 : Async. Exits with a non-VMX TD Exit Status Output #4 : Async. Exits with Cross-TD Exit Details Output #5 : On TDCALL(TDG.VP.VMCALL) Currently, to keep things simple, the wrapper function does not attempt to support different formats, and just passes all the GPRs that could be used. The GPR values are held by KVM in the area set aside for guest GPRs. KVM code uses the guest GPR area (vcpu->arch.regs[]) to set up for or process results of tdh_vp_enter(). Therefore changing tdh_vp_enter() to use more complex argument formats would also alter the way KVM code interacts with tdh_vp_enter(). Signed-off-by: Kai Huang <kai.huang@intel.com> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Message-ID: <20241121201448.36170-2-adrian.hunter@intel.com> Acked-by: Dave Hansen <dave.hansen@linux.intel.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Since commit 6037802 ("power: supply: core: implement extension API") there is the following ABBA deadlock (simplified) between the LED trigger code and the power-supply code: 1) When registering a power-supply class device, power_supply_register() calls led_trigger_register() from power_supply_create_triggers() in a scoped_guard(rwsem_read, &psy->extensions_sem) context. led_trigger_register() then in turn takes a LED subsystem lock. So here we have the following locking order: * Read-lock extensions_sem * Lock LED subsystem lock(s) 2) When registering a LED class device, with its default trigger set to a power-supply LED trigger (which has already been registered) The LED class code calls power_supply_led_trigger_activate() when setting up the default trigger. power_supply_led_trigger_activate() calls power_supply_get_property() to determine the initial value of to assign to the LED and that read-locks extensions_sem. So now we have the following locking order: * Lock LED subsystem lock(s) * Read-lock extensions_sem Fixing this is easy, there is no need to hold the extensions_sem when calling power_supply_create_triggers() since all triggers are always created rather then checking for the presence of certain attributes as power_supply_add_hwmon_sysfs() does. Move power_supply_create_triggers() out of the guard block to fix this. Here is the lockdep report fixed by this change: [ 31.249343] ====================================================== [ 31.249378] WARNING: possible circular locking dependency detected [ 31.249413] 6.13.0-rc6+ #251 Tainted: G C E [ 31.249440] ------------------------------------------------------ [ 31.249471] (udev-worker)/553 is trying to acquire lock: [ 31.249501] ffff892adbcaf660 (&psy->extensions_sem){.+.+}-{4:4}, at: power_supply_get_property.part.0+0x22/0x150 [ 31.249574] but task is already holding lock: [ 31.249603] ffff892adbc0bad0 (&led_cdev->trigger_lock){+.+.}-{4:4}, at: led_trigger_set_default+0x34/0xe0 [ 31.249657] which lock already depends on the new lock. [ 31.249696] the existing dependency chain (in reverse order) is: [ 31.249735] -> #2 (&led_cdev->trigger_lock){+.+.}-{4:4}: [ 31.249778] down_write+0x3b/0xd0 [ 31.249803] led_trigger_set_default+0x34/0xe0 [ 31.249833] led_classdev_register_ext+0x311/0x3a0 [ 31.249863] input_leds_connect+0x1dc/0x2a0 [ 31.249889] input_attach_handler.isra.0+0x75/0x90 [ 31.249921] input_register_device.cold+0xa1/0x150 [ 31.249955] hidinput_connect+0x8a2/0xb80 [ 31.249982] hid_connect+0x582/0x5c0 [ 31.250007] hid_hw_start+0x3f/0x60 [ 31.250030] hid_device_probe+0x122/0x1f0 [ 31.250053] really_probe+0xde/0x340 [ 31.250080] __driver_probe_device+0x78/0x110 [ 31.250105] driver_probe_device+0x1f/0xa0 [ 31.250132] __device_attach_driver+0x85/0x110 [ 31.250160] bus_for_each_drv+0x78/0xc0 [ 31.250184] __device_attach+0xb0/0x1b0 [ 31.250207] bus_probe_device+0x94/0xb0 [ 31.250230] device_add+0x64a/0x860 [ 31.250252] hid_add_device+0xe5/0x240 [ 31.250279] usbhid_probe+0x4dc/0x620 [ 31.250303] usb_probe_interface+0xe4/0x2a0 [ 31.250329] really_probe+0xde/0x340 [ 31.250353] __driver_probe_device+0x78/0x110 [ 31.250377] driver_probe_device+0x1f/0xa0 [ 31.250404] __device_attach_driver+0x85/0x110 [ 31.250431] bus_for_each_drv+0x78/0xc0 [ 31.250455] __device_attach+0xb0/0x1b0 [ 31.250478] bus_probe_device+0x94/0xb0 [ 31.250501] device_add+0x64a/0x860 [ 31.250523] usb_set_configuration+0x606/0x8a0 [ 31.250552] usb_generic_driver_probe+0x3e/0x60 [ 31.250579] usb_probe_device+0x3d/0x120 [ 31.250605] really_probe+0xde/0x340 [ 31.250629] __driver_probe_device+0x78/0x110 [ 31.250653] driver_probe_device+0x1f/0xa0 [ 31.250680] __device_attach_driver+0x85/0x110 [ 31.250707] bus_for_each_drv+0x78/0xc0 [ 31.250731] __device_attach+0xb0/0x1b0 [ 31.250753] bus_probe_device+0x94/0xb0 [ 31.250776] device_add+0x64a/0x860 [ 31.250798] usb_new_device.cold+0x141/0x38f [ 31.250828] hub_event+0x1166/0x1980 [ 31.250854] process_one_work+0x20f/0x580 [ 31.250879] worker_thread+0x1d1/0x3b0 [ 31.250904] kthread+0xee/0x120 [ 31.250926] ret_from_fork+0x30/0x50 [ 31.250954] ret_from_fork_asm+0x1a/0x30 [ 31.250982] -> #1 (triggers_list_lock){++++}-{4:4}: [ 31.251022] down_write+0x3b/0xd0 [ 31.251045] led_trigger_register+0x40/0x1b0 [ 31.251074] power_supply_register_led_trigger+0x88/0x150 [ 31.251107] power_supply_create_triggers+0x55/0xe0 [ 31.251135] __power_supply_register.part.0+0x34e/0x4a0 [ 31.251164] devm_power_supply_register+0x70/0xc0 [ 31.251190] bq27xxx_battery_setup+0x1a1/0x6d0 [bq27xxx_battery] [ 31.251235] bq27xxx_battery_i2c_probe+0xe5/0x17f [bq27xxx_battery_i2c] [ 31.251272] i2c_device_probe+0x125/0x2b0 [ 31.251299] really_probe+0xde/0x340 [ 31.251324] __driver_probe_device+0x78/0x110 [ 31.251348] driver_probe_device+0x1f/0xa0 [ 31.251375] __driver_attach+0xba/0x1c0 [ 31.251398] bus_for_each_dev+0x6b/0xb0 [ 31.251421] bus_add_driver+0x111/0x1f0 [ 31.251445] driver_register+0x6e/0xc0 [ 31.251470] i2c_register_driver+0x41/0xb0 [ 31.251498] do_one_initcall+0x5e/0x3a0 [ 31.251522] do_init_module+0x60/0x220 [ 31.251550] __do_sys_init_module+0x15f/0x190 [ 31.251575] do_syscall_64+0x93/0x180 [ 31.251598] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ 31.251629] -> #0 (&psy->extensions_sem){.+.+}-{4:4}: [ 31.251668] __lock_acquire+0x13ce/0x21c0 [ 31.251694] lock_acquire+0xcf/0x2e0 [ 31.251719] down_read+0x3e/0x170 [ 31.251741] power_supply_get_property.part.0+0x22/0x150 [ 31.251774] power_supply_update_leds+0x8d/0x230 [ 31.251804] power_supply_led_trigger_activate+0x18/0x20 [ 31.251837] led_trigger_set+0x1fc/0x300 [ 31.251863] led_trigger_set_default+0x90/0xe0 [ 31.251892] led_classdev_register_ext+0x311/0x3a0 [ 31.251921] devm_led_classdev_multicolor_register_ext+0x6e/0xb80 [led_class_multicolor] [ 31.251969] ktd202x_probe+0x464/0x5c0 [leds_ktd202x] [ 31.252002] i2c_device_probe+0x125/0x2b0 [ 31.252027] really_probe+0xde/0x340 [ 31.252052] __driver_probe_device+0x78/0x110 [ 31.252076] driver_probe_device+0x1f/0xa0 [ 31.252103] __driver_attach+0xba/0x1c0 [ 31.252125] bus_for_each_dev+0x6b/0xb0 [ 31.252148] bus_add_driver+0x111/0x1f0 [ 31.252172] driver_register+0x6e/0xc0 [ 31.252197] i2c_register_driver+0x41/0xb0 [ 31.252225] do_one_initcall+0x5e/0x3a0 [ 31.252248] do_init_module+0x60/0x220 [ 31.252274] __do_sys_init_module+0x15f/0x190 [ 31.253986] do_syscall_64+0x93/0x180 [ 31.255826] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ 31.257614] other info that might help us debug this: [ 31.257619] Chain exists of: &psy->extensions_sem --> triggers_list_lock --> &led_cdev->trigger_lock [ 31.257630] Possible unsafe locking scenario: [ 31.257632] CPU0 CPU1 [ 31.257633] ---- ---- [ 31.257634] lock(&led_cdev->trigger_lock); [ 31.257637] lock(triggers_list_lock); [ 31.257640] lock(&led_cdev->trigger_lock); [ 31.257643] rlock(&psy->extensions_sem); [ 31.257646] *** DEADLOCK *** [ 31.289433] 4 locks held by (udev-worker)/553: [ 31.289443] #0: ffff892ad9658108 (&dev->mutex){....}-{4:4}, at: __driver_attach+0xaf/0x1c0 [ 31.289463] #1: ffff892adbc0bbc8 (&led_cdev->led_access){+.+.}-{4:4}, at: led_classdev_register_ext+0x1c7/0x3a0 [ 31.289476] #2: ffffffffad0e30b0 (triggers_list_lock){++++}-{4:4}, at: led_trigger_set_default+0x2c/0xe0 [ 31.289487] #3: ffff892adbc0bad0 (&led_cdev->trigger_lock){+.+.}-{4:4}, at: led_trigger_set_default+0x34/0xe0 Fixes: 6037802 ("power: supply: core: implement extension API") Cc: Thomas Weißschuh <linux@weissschuh.net> Cc: Armin Wolf <W_Armin@gmx.de> Signed-off-by: Hans de Goede <hdegoede@redhat.com> Reviewed-by: Thomas Weißschuh <linux@weissschuh.net> Link: https://lore.kernel.org/r/20250130140035.20636-1-hdegoede@redhat.com Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>

…ea as VM_ALLOC Erhard reported the following KASAN hit while booting his PowerMac G4 with a KASAN-enabled kernel 6.13-rc6: BUG: KASAN: vmalloc-out-of-bounds in copy_to_kernel_nofault+0xd8/0x1c8 Write of size 8 at addr f1000000 by task chronyd/1293 CPU: 0 UID: 123 PID: 1293 Comm: chronyd Tainted: G W 6.13.0-rc6-PMacG4 #2 Tainted: [W]=WARN Hardware name: PowerMac3,6 7455 0x80010303 PowerMac Call Trace: [c2437590] [c1631a84] dump_stack_lvl+0x70/0x8c (unreliable) [c24375b0] [c0504998] print_report+0xdc/0x504 [c2437610] [c050475c] kasan_report+0xf8/0x108 [c2437690] [c0505a3c] kasan_check_range+0x24/0x18c [c24376a0] [c03fb5e4] copy_to_kernel_nofault+0xd8/0x1c8 [c24376c0] [c004c014] patch_instructions+0x15c/0x16c [c2437710] [c00731a8] bpf_arch_text_copy+0x60/0x7c [c2437730] [c0281168] bpf_jit_binary_pack_finalize+0x50/0xac [c2437750] [c0073cf4] bpf_int_jit_compile+0xb30/0xdec [c2437880] [c0280394] bpf_prog_select_runtime+0x15c/0x478 [c24378d0] [c1263428] bpf_prepare_filter+0xbf8/0xc14 [c2437990] [c12677ec] bpf_prog_create_from_user+0x258/0x2b4 [c24379d0] [c027111c] do_seccomp+0x3dc/0x1890 [c2437ac0] [c001d8e0] system_call_exception+0x2dc/0x420 [c2437f30] [c00281ac] ret_from_syscall+0x0/0x2c --- interrupt: c00 at 0x5a1274 NIP: 005a1274 LR: 006a3b3c CTR: 005296c8 REGS: c2437f40 TRAP: 0c00 Tainted: G W (6.13.0-rc6-PMacG4) MSR: 0200f932 <VEC,EE,PR,FP,ME,IR,DR,RI> CR: 24004422 XER: 00000000 GPR00: 00000166 af8f3fa0 a7ee3540 00000001 00000000 013b6500 005a5858 0200f932 GPR08: 00000000 00001fe9 013d5fc8 005296c8 2822244c 00b2fcd8 00000000 af8f4b57 GPR16: 00000000 00000001 00000000 00000000 00000000 00000001 00000000 00000002 GPR24: 00afdbb0 00000000 00000000 00000000 006e0004 013ce060 006e7c1c 00000001 NIP [005a1274] 0x5a1274 LR [006a3b3c] 0x6a3b3c --- interrupt: c00 The buggy address belongs to the virtual mapping at [f1000000, f1002000) created by: text_area_cpu_up+0x20/0x190 The buggy address belongs to the physical page: page: refcount:1 mapcount:0 mapping:00000000 index:0x0 pfn:0x76e30 flags: 0x80000000(zone=2) raw: 80000000 00000000 00000122 00000000 00000000 00000000 ffffffff 00000001 raw: 00000000 page dumped because: kasan: bad access detected Memory state around the buggy address: f0ffff00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 f0ffff80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 >f1000000: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 ^ f1000080: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f1000100: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 ================================================================== f8 corresponds to KASAN_VMALLOC_INVALID which means the area is not initialised hence not supposed to be used yet. Powerpc text patching infrastructure allocates a virtual memory area using get_vm_area() and flags it as VM_ALLOC. But that flag is meant to be used for vmalloc() and vmalloc() allocated memory is not supposed to be used before a call to __vmalloc_node_range() which is never called for that area. That went undetected until commit e4137f0 ("mm, kasan, kmsan: instrument copy_from/to_kernel_nofault") The area allocated by text_area_cpu_up() is not vmalloc memory, it is mapped directly on demand when needed by map_kernel_page(). There is no VM flag corresponding to such usage, so just pass no flag. That way the area will be unpoisonned and usable immediately. Reported-by: Erhard Furtner <erhard_f@mailbox.org> Closes: https://lore.kernel.org/all/20250112135832.57c92322@yea/ Fixes: 37bc3e5 ("powerpc/lib/code-patching: Use alternate map for patch_instruction()") Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu> Signed-off-by: Madhavan Srinivasan <maddy@linux.ibm.com> Link: https://patch.msgid.link/06621423da339b374f48c0886e3a5db18e896be8.1739342693.git.christophe.leroy@csgroup.eu

In the Intel PMU counters test, only validate the counts for architectural events that are supported in hardware. If an arch event isn't supported, the event selector may enable a completely different event, and thus the logic for the expected count is bogus. This fixes test failures on pre-Icelake systems due to the encoding for the architectural Top-Down Slots event corresponding to something else (at least on the Skylake family of CPUs). Note, validation relies on *hardware* support, not KVM support and not guest support. Architectural events are all about enumerating the event selector encoding; lack of enumeration for an architectural event doesn't mean the event itself is unsupported, i.e. the event should still count as expected even if KVM and/or guest CPUID doesn't enumerate the event as being "architectural". Note #2, it's desirable to _program_ the architectural event encoding even if hardware doesn't support the event. The count can't be validated when the event is fully enabled, but KVM should still let the guest program the event selector, and the PMC shouldn't count if the event is disabled. Fixes: 4f1bd6b ("KVM: selftests: Test Intel PMU architectural events on gp counters") Reported-by: kernel test robot <oliver.sang@intel.com> Closes: https://lore.kernel.org/oe-lkp/202501141009.30c629b4-lkp@intel.com Debugged-by: Dapeng Mi <dapeng1.mi@linux.intel.com> Link: https://lore.kernel.org/r/20250117234204.2600624-3-seanjc@google.com Signed-off-by: Sean Christopherson <seanjc@google.com>

Sync the new iteration to the guest prior to restarting the vCPU, otherwise it's possible for the vCPU to dirty memory for the next iteration using the current iteration's value. Note, because the guest can be interrupted between the vCPU's load of the iteration and its write to memory, it's still possible for the guest to store the previous iteration to memory as the previous iteration may be cached in a CPU register (which the test accounts for). Note #2, the test's current approach of collecting dirty entries *before* stopping the vCPU also results dirty memory having the previous iteration. E.g. if page is dirtied in the previous iteration, but not the current iteration, the verification phase will observe the previous iteration's value in memory. That wart will be remedied in the near future, at which point synchronizing the iteration before restarting the vCPU will guarantee the only way for verification to observe stale iterations is due to the CPU register caching case, or due to a dirty entry being collected before the store retires. Link: https://lore.kernel.org/r/20250111003004.1235645-3-seanjc@google.com Signed-off-by: Sean Christopherson <seanjc@google.com>

Handle "guest stopped" requests once per guest time update in preparation of restoring KVM's historical behavior of setting PVCLOCK_GUEST_STOPPED for kvmclock and only kvmclock. For now, simply move the code to minimize the probability of an unintentional change in functionally. Note, in practice, all clocks are guaranteed to see the request (or not) even though each PV clock processes the request individual, as KVM holds vcpu->mutex (blocks KVM_KVMCLOCK_CTRL) and it should be impossible for KVM's suspend notifier to run while KVM is handling requests. And because the helper updates the reference flags, all subsequent PV clock updates will pick up PVCLOCK_GUEST_STOPPED. Note #2, once PVCLOCK_GUEST_STOPPED is restricted to kvmclock, the horrific #ifdef will go away. Cc: Paul Durrant <pdurrant@amazon.com> Cc: David Woodhouse <dwmw@amazon.co.uk> Reviewed-by: Paul Durrant <paul@xen.org> Link: https://lore.kernel.org/r/20250201013827.680235-5-seanjc@google.com Signed-off-by: Sean Christopherson <seanjc@google.com>

Use raw_spinlock in order to fix spurious messages about invalid context when spinlock debugging is enabled. The lock is only used to serialize register access. [ 4.239592] ============================= [ 4.239595] [ BUG: Invalid wait context ] [ 4.239599] 6.13.0-rc7-arm64-renesas-05496-gd088502a519f #35 Not tainted [ 4.239603] ----------------------------- [ 4.239606] kworker/u8:5/76 is trying to lock: [ 4.239609] ffff0000091898a0 (&p->lock){....}-{3:3}, at: gpio_rcar_config_interrupt_input_mode+0x34/0x164 [ 4.239641] other info that might help us debug this: [ 4.239643] context-{5:5} [ 4.239646] 5 locks held by kworker/u8:5/76: [ 4.239651] #0: ffff0000080fb148 ((wq_completion)async){+.+.}-{0:0}, at: process_one_work+0x190/0x62c [ 4.250180] OF: /soc/sound@ec500000/ports/port@0/endpoint: Read of boolean property 'frame-master' with a value. [ 4.254094] #1: ffff80008299bd80 ((work_completion)(&entry->work)){+.+.}-{0:0}, at: process_one_work+0x1b8/0x62c [ 4.254109] #2: ffff00000920c8f8 [ 4.258345] OF: /soc/sound@ec500000/ports/port@1/endpoint: Read of boolean property 'bitclock-master' with a value. [ 4.264803] (&dev->mutex){....}-{4:4}, at: __device_attach_async_helper+0x3c/0xdc [ 4.264820] #3: ffff00000a50ca40 (request_class#2){+.+.}-{4:4}, at: __setup_irq+0xa0/0x690 [ 4.264840] #4: [ 4.268872] OF: /soc/sound@ec500000/ports/port@1/endpoint: Read of boolean property 'frame-master' with a value. [ 4.273275] ffff00000a50c8c8 (lock_class){....}-{2:2}, at: __setup_irq+0xc4/0x690 [ 4.296130] renesas_sdhi_internal_dmac ee100000.mmc: mmc1 base at 0x00000000ee100000, max clock rate 200 MHz [ 4.304082] stack backtrace: [ 4.304086] CPU: 1 UID: 0 PID: 76 Comm: kworker/u8:5 Not tainted 6.13.0-rc7-arm64-renesas-05496-gd088502a519f #35 [ 4.304092] Hardware name: Renesas Salvator-X 2nd version board based on r8a77965 (DT) [ 4.304097] Workqueue: async async_run_entry_fn [ 4.304106] Call trace: [ 4.304110] show_stack+0x14/0x20 (C) [ 4.304122] dump_stack_lvl+0x6c/0x90 [ 4.304131] dump_stack+0x14/0x1c [ 4.304138] __lock_acquire+0xdfc/0x1584 [ 4.426274] lock_acquire+0x1c4/0x33c [ 4.429942] _raw_spin_lock_irqsave+0x5c/0x80 [ 4.434307] gpio_rcar_config_interrupt_input_mode+0x34/0x164 [ 4.440061] gpio_rcar_irq_set_type+0xd4/0xd8 [ 4.444422] __irq_set_trigger+0x5c/0x178 [ 4.448435] __setup_irq+0x2e4/0x690 [ 4.452012] request_threaded_irq+0xc4/0x190 [ 4.456285] devm_request_threaded_irq+0x7c/0xf4 [ 4.459398] ata1: link resume succeeded after 1 retries [ 4.460902] mmc_gpiod_request_cd_irq+0x68/0xe0 [ 4.470660] mmc_start_host+0x50/0xac [ 4.474327] mmc_add_host+0x80/0xe4 [ 4.477817] tmio_mmc_host_probe+0x2b0/0x440 [ 4.482094] renesas_sdhi_probe+0x488/0x6f4 [ 4.486281] renesas_sdhi_internal_dmac_probe+0x60/0x78 [ 4.491509] platform_probe+0x64/0xd8 [ 4.495178] really_probe+0xb8/0x2a8 [ 4.498756] __driver_probe_device+0x74/0x118 [ 4.503116] driver_probe_device+0x3c/0x154 [ 4.507303] __device_attach_driver+0xd4/0x160 [ 4.511750] bus_for_each_drv+0x84/0xe0 [ 4.515588] __device_attach_async_helper+0xb0/0xdc [ 4.520470] async_run_entry_fn+0x30/0xd8 [ 4.524481] process_one_work+0x210/0x62c [ 4.528494] worker_thread+0x1ac/0x340 [ 4.532245] kthread+0x10c/0x110 [ 4.535476] ret_from_fork+0x10/0x20 Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se> Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be> Tested-by: Geert Uytterhoeven <geert+renesas@glider.be> Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20250121135833.3769310-1-niklas.soderlund+renesas@ragnatech.se Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>

…cal section A circular lock dependency splat has been seen involving down_trylock(): ====================================================== WARNING: possible circular locking dependency detected 6.12.0-41.el10.s390x+debug ------------------------------------------------------ dd/32479 is trying to acquire lock: 0015a20accd0d4f8 ((console_sem).lock){-.-.}-{2:2}, at: down_trylock+0x26/0x90 but task is already holding lock: 000000017e461698 (&zone->lock){-.-.}-{2:2}, at: rmqueue_bulk+0xac/0x8f0 the existing dependency chain (in reverse order) is: -> #4 (&zone->lock){-.-.}-{2:2}: -> #3 (hrtimer_bases.lock){-.-.}-{2:2}: -> #2 (&rq->__lock){-.-.}-{2:2}: -> #1 (&p->pi_lock){-.-.}-{2:2}: -> #0 ((console_sem).lock){-.-.}-{2:2}: The console_sem -> pi_lock dependency is due to calling try_to_wake_up() while holding the console_sem raw_spinlock. This dependency can be broken by using wake_q to do the wakeup instead of calling try_to_wake_up() under the console_sem lock. This will also make the semaphore's raw_spinlock become a terminal lock without taking any further locks underneath it. The hrtimer_bases.lock is a raw_spinlock while zone->lock is a spinlock. The hrtimer_bases.lock -> zone->lock dependency happens via the debug_objects_fill_pool() helper function in the debugobjects code. -> #4 (&zone->lock){-.-.}-{2:2}: __lock_acquire+0xe86/0x1cc0 lock_acquire.part.0+0x258/0x630 lock_acquire+0xb8/0xe0 _raw_spin_lock_irqsave+0xb4/0x120 rmqueue_bulk+0xac/0x8f0 __rmqueue_pcplist+0x580/0x830 rmqueue_pcplist+0xfc/0x470 rmqueue.isra.0+0xdec/0x11b0 get_page_from_freelist+0x2ee/0xeb0 __alloc_pages_noprof+0x2c2/0x520 alloc_pages_mpol_noprof+0x1fc/0x4d0 alloc_pages_noprof+0x8c/0xe0 allocate_slab+0x320/0x460 ___slab_alloc+0xa58/0x12b0 __slab_alloc.isra.0+0x42/0x60 kmem_cache_alloc_noprof+0x304/0x350 fill_pool+0xf6/0x450 debug_object_activate+0xfe/0x360 enqueue_hrtimer+0x34/0x190 __run_hrtimer+0x3c8/0x4c0 __hrtimer_run_queues+0x1b2/0x260 hrtimer_interrupt+0x316/0x760 do_IRQ+0x9a/0xe0 do_irq_async+0xf6/0x160 Normally a raw_spinlock to spinlock dependency is not legitimate and will be warned if CONFIG_PROVE_RAW_LOCK_NESTING is enabled, but debug_objects_fill_pool() is an exception as it explicitly allows this dependency for non-PREEMPT_RT kernel without causing PROVE_RAW_LOCK_NESTING lockdep splat. As a result, this dependency is legitimate and not a bug. Anyway, semaphore is the only locking primitive left that is still using try_to_wake_up() to do wakeup inside critical section, all the other locking primitives had been migrated to use wake_q to do wakeup outside of the critical section. It is also possible that there are other circular locking dependencies involving printk/console_sem or other existing/new semaphores lurking somewhere which may show up in the future. Let just do the migration now to wake_q to avoid headache like this. Reported-by: yzbot+ed801a886dfdbfe7136d@syzkaller.appspotmail.com Signed-off-by: Waiman Long <longman@redhat.com> Signed-off-by: Boqun Feng <boqun.feng@gmail.com> Signed-off-by: Ingo Molnar <mingo@kernel.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Link: https://lore.kernel.org/r/20250307232717.1759087-3-boqun.feng@gmail.com

A blocking notification chain uses a read-write semaphore to protect the integrity of the chain. The semaphore is acquired for writing when adding / removing notifiers to / from the chain and acquired for reading when traversing the chain and informing notifiers about an event. In case of the blocking switchdev notification chain, recursive notifications are possible which leads to the semaphore being acquired twice for reading and to lockdep warnings being generated [1]. Specifically, this can happen when the bridge driver processes a SWITCHDEV_BRPORT_UNOFFLOADED event which causes it to emit notifications about deferred events when calling switchdev_deferred_process(). Fix this by converting the notification chain to a raw notification chain in a similar fashion to the netdev notification chain. Protect the chain using the RTNL mutex by acquiring it when modifying the chain. Events are always informed under the RTNL mutex, but add an assertion in call_switchdev_blocking_notifiers() to make sure this is not violated in the future. Maintain the "blocking" prefix as events are always emitted from process context and listeners are allowed to block. [1]: WARNING: possible recursive locking detected 6.14.0-rc4-custom-g079270089484 #1 Not tainted -------------------------------------------- ip/52731 is trying to acquire lock: ffffffff850918d8 ((switchdev_blocking_notif_chain).rwsem){++++}-{4:4}, at: blocking_notifier_call_chain+0x58/0xa0 but task is already holding lock: ffffffff850918d8 ((switchdev_blocking_notif_chain).rwsem){++++}-{4:4}, at: blocking_notifier_call_chain+0x58/0xa0 other info that might help us debug this: Possible unsafe locking scenario: CPU0 ---- lock((switchdev_blocking_notif_chain).rwsem); lock((switchdev_blocking_notif_chain).rwsem); *** DEADLOCK *** May be due to missing lock nesting notation 3 locks held by ip/52731: #0: ffffffff84f795b0 (rtnl_mutex){+.+.}-{4:4}, at: rtnl_newlink+0x727/0x1dc0 #1: ffffffff8731f628 (&net->rtnl_mutex){+.+.}-{4:4}, at: rtnl_newlink+0x790/0x1dc0 #2: ffffffff850918d8 ((switchdev_blocking_notif_chain).rwsem){++++}-{4:4}, at: blocking_notifier_call_chain+0x58/0xa0 stack backtrace: ... ? __pfx_down_read+0x10/0x10 ? __pfx_mark_lock+0x10/0x10 ? __pfx_switchdev_port_attr_set_deferred+0x10/0x10 blocking_notifier_call_chain+0x58/0xa0 switchdev_port_attr_notify.constprop.0+0xb3/0x1b0 ? __pfx_switchdev_port_attr_notify.constprop.0+0x10/0x10 ? mark_held_locks+0x94/0xe0 ? switchdev_deferred_process+0x11a/0x340 switchdev_port_attr_set_deferred+0x27/0xd0 switchdev_deferred_process+0x164/0x340 br_switchdev_port_unoffload+0xc8/0x100 [bridge] br_switchdev_blocking_event+0x29f/0x580 [bridge] notifier_call_chain+0xa2/0x440 blocking_notifier_call_chain+0x6e/0xa0 switchdev_bridge_port_unoffload+0xde/0x1a0 ... Fixes: f7a70d6 ("net: bridge: switchdev: Ensure deferred event delivery on unoffload") Signed-off-by: Amit Cohen <amcohen@nvidia.com> Reviewed-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: Simon Horman <horms@kernel.org> Reviewed-by: Vladimir Oltean <olteanv@gmail.com> Tested-by: Vladimir Oltean <olteanv@gmail.com> Link: https://patch.msgid.link/20250305121509.631207-1-amcohen@nvidia.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>

When on a MANA VM hibernation is triggered, as part of hibernate_snapshot(), mana_gd_suspend() and mana_gd_resume() are called. If during this mana_gd_resume(), a failure occurs with HWC creation, mana_port_debugfs pointer does not get reinitialized and ends up pointing to older, cleaned-up dentry. Further in the hibernation path, as part of power_down(), mana_gd_shutdown() is triggered. This call, unaware of the failures in resume, tries to cleanup the already cleaned up mana_port_debugfs value and hits the following bug: [ 191.359296] mana 7870:00:00.0: Shutdown was called [ 191.359918] BUG: kernel NULL pointer dereference, address: 0000000000000098 [ 191.360584] #PF: supervisor write access in kernel mode [ 191.361125] #PF: error_code(0x0002) - not-present page [ 191.361727] PGD 1080ea067 P4D 0 [ 191.362172] Oops: Oops: 0002 [#1] SMP NOPTI [ 191.362606] CPU: 11 UID: 0 PID: 1674 Comm: bash Not tainted 6.14.0-rc5+ #2 [ 191.363292] Hardware name: Microsoft Corporation Virtual Machine/Virtual Machine, BIOS Hyper-V UEFI Release v4.1 11/21/2024 [ 191.364124] RIP: 0010:down_write+0x19/0x50 [ 191.364537] Code: 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 0f 1f 44 00 00 55 48 89 e5 53 48 89 fb e8 de cd ff ff 31 c0 ba 01 00 00 00 <f0> 48 0f b1 13 75 16 65 48 8b 05 88 24 4c 6a 48 89 43 08 48 8b 5d [ 191.365867] RSP: 0000:ff45fbe0c1c037b8 EFLAGS: 00010246 [ 191.366350] RAX: 0000000000000000 RBX: 0000000000000098 RCX: ffffff8100000000 [ 191.366951] RDX: 0000000000000001 RSI: 0000000000000064 RDI: 0000000000000098 [ 191.367600] RBP: ff45fbe0c1c037c0 R08: 0000000000000000 R09: 0000000000000001 [ 191.368225] R10: ff45fbe0d2b01000 R11: 0000000000000008 R12: 0000000000000000 [ 191.368874] R13: 000000000000000b R14: ff43dc27509d67c0 R15: 0000000000000020 [ 191.369549] FS: 00007dbc5001e740(0000) GS:ff43dc663f380000(0000) knlGS:0000000000000000 [ 191.370213] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 191.370830] CR2: 0000000000000098 CR3: 0000000168e8e002 CR4: 0000000000b73ef0 [ 191.371557] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 191.372192] DR3: 0000000000000000 DR6: 00000000fffe07f0 DR7: 0000000000000400 [ 191.372906] Call Trace: [ 191.373262] <TASK> [ 191.373621] ? show_regs+0x64/0x70 [ 191.374040] ? __die+0x24/0x70 [ 191.374468] ? page_fault_oops+0x290/0x5b0 [ 191.374875] ? do_user_addr_fault+0x448/0x800 [ 191.375357] ? exc_page_fault+0x7a/0x160 [ 191.375971] ? asm_exc_page_fault+0x27/0x30 [ 191.376416] ? down_write+0x19/0x50 [ 191.376832] ? down_write+0x12/0x50 [ 191.377232] simple_recursive_removal+0x4a/0x2a0 [ 191.377679] ? __pfx_remove_one+0x10/0x10 [ 191.378088] debugfs_remove+0x44/0x70 [ 191.378530] mana_detach+0x17c/0x4f0 [ 191.378950] ? __flush_work+0x1e2/0x3b0 [ 191.379362] ? __cond_resched+0x1a/0x50 [ 191.379787] mana_remove+0xf2/0x1a0 [ 191.380193] mana_gd_shutdown+0x3b/0x70 [ 191.380642] pci_device_shutdown+0x3a/0x80 [ 191.381063] device_shutdown+0x13e/0x230 [ 191.381480] kernel_power_off+0x35/0x80 [ 191.381890] hibernate+0x3c6/0x470 [ 191.382312] state_store+0xcb/0xd0 [ 191.382734] kobj_attr_store+0x12/0x30 [ 191.383211] sysfs_kf_write+0x3e/0x50 [ 191.383640] kernfs_fop_write_iter+0x140/0x1d0 [ 191.384106] vfs_write+0x271/0x440 [ 191.384521] ksys_write+0x72/0xf0 [ 191.384924] __x64_sys_write+0x19/0x20 [ 191.385313] x64_sys_call+0x2b0/0x20b0 [ 191.385736] do_syscall_64+0x79/0x150 [ 191.386146] ? __mod_memcg_lruvec_state+0xe7/0x240 [ 191.386676] ? __lruvec_stat_mod_folio+0x79/0xb0 [ 191.387124] ? __pfx_lru_add+0x10/0x10 [ 191.387515] ? queued_spin_unlock+0x9/0x10 [ 191.387937] ? do_anonymous_page+0x33c/0xa00 [ 191.388374] ? __handle_mm_fault+0xcf3/0x1210 [ 191.388805] ? __count_memcg_events+0xbe/0x180 [ 191.389235] ? handle_mm_fault+0xae/0x300 [ 191.389588] ? do_user_addr_fault+0x559/0x800 [ 191.390027] ? irqentry_exit_to_user_mode+0x43/0x230 [ 191.390525] ? irqentry_exit+0x1d/0x30 [ 191.390879] ? exc_page_fault+0x86/0x160 [ 191.391235] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ 191.391745] RIP: 0033:0x7dbc4ff1c574 [ 191.392111] Code: c7 00 16 00 00 00 b8 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 f3 0f 1e fa 80 3d d5 ea 0e 00 00 74 13 b8 01 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 54 c3 0f 1f 00 55 48 89 e5 48 83 ec 20 48 89 [ 191.393412] RSP: 002b:00007ffd95a23ab8 EFLAGS: 00000202 ORIG_RAX: 0000000000000001 [ 191.393990] RAX: ffffffffffffffda RBX: 0000000000000005 RCX: 00007dbc4ff1c574 [ 191.394594] RDX: 0000000000000005 RSI: 00005a6eeadb0ce0 RDI: 0000000000000001 [ 191.395215] RBP: 00007ffd95a23ae0 R08: 00007dbc50003b20 R09: 0000000000000000 [ 191.395805] R10: 0000000000000001 R11: 0000000000000202 R12: 0000000000000005 [ 191.396404] R13: 00005a6eeadb0ce0 R14: 00007dbc500045c0 R15: 00007dbc50001ee0 [ 191.396987] </TASK> To fix this, we explicitly set such mana debugfs variables to NULL after debugfs_remove() is called. Fixes: 6607c17 ("net: mana: Enable debugfs files for MANA device") Cc: stable@vger.kernel.org Signed-off-by: Shradha Gupta <shradhagupta@linux.microsoft.com> Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com> Reviewed-by: Michal Kubiak <michal.kubiak@intel.com> Link: https://patch.msgid.link/1741688260-28922-1-git-send-email-shradhagupta@linux.microsoft.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>

During a module removal, kvm_exit invokes arch specific disable call which disables AIA. However, we invoke aia_exit before kvm_exit resulting in the following warning. KVM kernel module can't be inserted afterwards due to inconsistent state of IRQ. [25469.031389] percpu IRQ 31 still enabled on CPU0! [25469.031732] WARNING: CPU: 3 PID: 943 at kernel/irq/manage.c:2476 __free_percpu_irq+0xa2/0x150 [25469.031804] Modules linked in: kvm(-) [25469.031848] CPU: 3 UID: 0 PID: 943 Comm: rmmod Not tainted 6.14.0-rc5-06947-g91c763118f47-dirty #2 [25469.031905] Hardware name: riscv-virtio,qemu (DT) [25469.031928] epc : __free_percpu_irq+0xa2/0x150 [25469.031976] ra : __free_percpu_irq+0xa2/0x150 [25469.032197] epc : ffffffff8007db1e ra : ffffffff8007db1e sp : ff2000000088bd50 [25469.032241] gp : ffffffff8131cef8 tp : ff60000080b96400 t0 : ff2000000088baf8 [25469.032285] t1 : fffffffffffffffc t2 : 5249207570637265 s0 : ff2000000088bd90 [25469.032329] s1 : ff60000098b21080 a0 : 037d527a15eb4f00 a1 : 037d527a15eb4f00 [25469.032372] a2 : 0000000000000023 a3 : 0000000000000001 a4 : ffffffff8122dbf8 [25469.032410] a5 : 0000000000000fff a6 : 0000000000000000 a7 : ffffffff8122dc10 [25469.032448] s2 : ff60000080c22eb0 s3 : 0000000200000022 s4 : 000000000000001f [25469.032488] s5 : ff60000080c22e00 s6 : ffffffff80c351c0 s7 : 0000000000000000 [25469.032582] s8 : 0000000000000003 s9 : 000055556b7fb490 s10: 00007ffff0e12fa0 [25469.032621] s11: 00007ffff0e13e9a t3 : ffffffff81354ac7 t4 : ffffffff81354ac7 [25469.032664] t5 : ffffffff81354ac8 t6 : ffffffff81354ac7 [25469.032698] status: 0000000200000100 badaddr: ffffffff8007db1e cause: 0000000000000003 [25469.032738] [<ffffffff8007db1e>] __free_percpu_irq+0xa2/0x150 [25469.032797] [<ffffffff8007dbfc>] free_percpu_irq+0x30/0x5e [25469.032856] [<ffffffff013a57dc>] kvm_riscv_aia_exit+0x40/0x42 [kvm] [25469.033947] [<ffffffff013b4e82>] cleanup_module+0x10/0x32 [kvm] [25469.035300] [<ffffffff8009b150>] __riscv_sys_delete_module+0x18e/0x1fc [25469.035374] [<ffffffff8000c1ca>] syscall_handler+0x3a/0x46 [25469.035456] [<ffffffff809ec9a4>] do_trap_ecall_u+0x72/0x134 [25469.035536] [<ffffffff809f5e18>] handle_exception+0x148/0x156 Invoke aia_exit and other arch specific cleanup functions after kvm_exit so that disable gets a chance to be called first before exit. Fixes: 54e4332 ("RISC-V: KVM: Initial skeletal support for AIA") Fixes: eded675 ("riscv: KVM: add basic support for host vs guest profiling") Signed-off-by: Atish Patra <atishp@rivosinc.com> Reviewed-by: Anup Patel <anup@brainfault.org> Reviewed-by: Sean Christopherson <seanjc@google.com> Link: https://lore.kernel.org/r/20250317-kvm_exit_fix-v1-1-aa5240c5dbd2@rivosinc.com Signed-off-by: Anup Patel <anup@brainfault.org>

Commit 7da55c2 ("drm/amd/display: Remove incorrect FP context start") removes the FP context protection of dml2_create(), and it said "All the DC_FP_START/END should be used before call anything from DML2". However, dml2_init()/dml21_init() are not protected from their callers, causing such errors: do_fpu invoked from kernel context![#1]: CPU: 0 UID: 0 PID: 239 Comm: kworker/0:5 Not tainted 6.14.0-rc6+ #2 Workqueue: events work_for_cpu_fn pc ffff80000319de80 ra ffff80000319de5c tp 900000010575c000 sp 900000010575f840 a0 0000000000000000 a1 900000012f210130 a2 900000012f000000 a3 ffff80000357e268 a4 ffff80000357e260 a5 900000012ea52cf0 a6 0000000400000004 a7 0000012c00001388 t0 00001900000015e0 t1 ffff80000379d000 t2 0000000010624dd3 t3 0000006400000014 t4 00000000000003e8 t5 0000005000000018 t6 0000000000000020 t7 0000000f00000064 t8 000000000000002f u0 5f5e9200f8901912 s9 900000012d380010 s0 900000012ea51fd8 s1 900000012f000000 s2 9000000109296000 s3 0000000000000001 s4 0000000000001fd8 s5 0000000000000001 s6 ffff800003415000 s7 900000012d390000 s8 ffff800003211f80 ra: ffff80000319de5c dml21_apply_soc_bb_overrides+0x3c/0x960 [amdgpu] ERA: ffff80000319de80 dml21_apply_soc_bb_overrides+0x60/0x960 [amdgpu] CRMD: 000000b0 (PLV0 -IE -DA +PG DACF=CC DACM=CC -WE) PRMD: 00000004 (PPLV0 +PIE -PWE) EUEN: 00000000 (-FPE -SXE -ASXE -BTE) ECFG: 00071c1d (LIE=0,2-4,10-12 VS=7) ESTAT: 000f0000 [FPD] (IS= ECode=15 EsubCode=0) PRID: 0014d010 (Loongson-64bit, Loongson-3C6000/S) Process kworker/0:5 (pid: 239, threadinfo=00000000927eadc6, task=000000008fd31682) Stack : 00040dc000003164 0000000000000001 900000012f210130 900000012eabeeb8 900000012f000000 ffff80000319fe48 900000012f210000 900000012f210130 900000012f000000 900000012eabeeb8 0000000000000001 ffff8000031a0064 900000010575f9f0 900000012f210130 900000012eac0000 900000012ea80000 900000012f000000 ffff8000031cefc4 900000010575f9f0 ffff8000035859c0 ffff800003414000 900000010575fa78 900000012f000000 ffff8000031b4c50 0000000000000000 9000000101c9d700 9000000109c40000 5f5e9200f8901912 900000012d3c4bd0 900000012d3c5000 ffff8000034aed18 900000012d380010 900000012d3c4bd0 ffff800003414000 900000012d380000 ffff800002ea49dc 0000000000000001 900000012d3c6000 00000000ffffe423 0000000000010000 ... Call Trace: [<ffff80000319de80>] dml21_apply_soc_bb_overrides+0x60/0x960 [amdgpu] [<ffff80000319fe44>] dml21_init+0xa4/0x280 [amdgpu] [<ffff8000031a0060>] dml21_create+0x40/0x80 [amdgpu] [<ffff8000031cefc0>] dc_state_create+0x100/0x160 [amdgpu] [<ffff8000031b4c4c>] dc_create+0x44c/0x640 [amdgpu] [<ffff800002ea49d8>] amdgpu_dm_init+0x3f8/0x2060 [amdgpu] [<ffff800002ea6658>] dm_hw_init+0x18/0x60 [amdgpu] [<ffff800002b16738>] amdgpu_device_init+0x1938/0x27e0 [amdgpu] [<ffff800002b18e80>] amdgpu_driver_load_kms+0x20/0xa0 [amdgpu] [<ffff800002b0c8f0>] amdgpu_pci_probe+0x1b0/0x580 [amdgpu] [<900000000448eae4>] local_pci_probe+0x44/0xc0 [<9000000003b02b18>] work_for_cpu_fn+0x18/0x40 [<9000000003b05da0>] process_one_work+0x160/0x300 [<9000000003b06718>] worker_thread+0x318/0x440 [<9000000003b11b8c>] kthread+0x12c/0x220 [<9000000003ac1484>] ret_from_kernel_thread+0x8/0xa4 Unfortunately, protecting dml2_init()/dml21_init() out of DML2 causes "sleeping function called from invalid context", so protect them with DC_FP_START() and DC_FP_END() inside. Fixes: 7da55c2 ("drm/amd/display: Remove incorrect FP context start") Cc: stable@vger.kernel.org Signed-off-by: Huacai Chen <chenhuacai@loongson.cn> Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

On SME platforms, at hardware level dirty cachelines with and without encryption bit of the same memory can coexist, and the CPU can flush them back to memory in random order. During kexec, the caches must be flushed before jumping to the new kernel to avoid silent memory corruption when a cacheline with a different encryption property is written back over whatever encryption properties the new kernel is using. TDX also needs cache flush during kexec for the same reason. It would be good to implement a generic way to flush cache instead of scattering checks for each feature all around. During kexec, the kexec-ing CPU sends IPIs to all remote CPUs to stop them before it boots to the new kernel. For SME the kernel basically encrypts all memory including the kernel itself by manipulating page tables. A simple memory write from the kernel could dirty cachelines. Currently, the kernel uses WBINVD to flush cache for SME during kexec in two places: 1) the one in stop_this_cpu() for all remote CPUs when the kexec-ing CPU stops them; 2) the one in the relocate_kernel() where the kexec-ing CPU jumps to the new kernel. Unlike SME, TDX can only dirty cachelines when it is used (i.e., when SEAMCALLs are performed). Since there are no more SEAMCALLs after the aforementioned WBINVDs, leverage them for TDX. In order to have a generic way to cover both SME and TDX, use a percpu boolean to indicate the cache may be in an incoherent state thus cache flush is needed during kexec, and turn on the boolean for SME. TDX can then leverage it by also turning the boolean on. A percpu boolean isn't strictly needed for SME since it is activated at early boot time and on all CPUs. A global flag would be sufficient. But using a percpu variable has two benefits. Foremost, the state that is being tracked here (percpu cache coherency situation requiring a flush) is percpu, so a percpu state is a more direct and natural fit. Secondly, it will fit TDX's usage better since the percpu var can be set when a CPU makes a SEAMCALL, and cleared when another WBINVID on the CPU obviates the need for a kexec-time WBINVID. Saving kexec-time WBINVD is valuable, because there is an existing race[*] where kexec could proceed while another CPU is active. WBINVD could make this race worse, so it's worth skipping it when possible. Today the first WBINVD in the stop_this_cpu() is performed when the platform supports SME and the second WBINVD is done in relocate_kernel() when SME is activated by the kernel. Make things simple by changing to do the second WBINVD when the platform supports SME. This allows the kernel to simply turn on this percpu boolean when bringing up a CPU by checking whether the platform supports SME. No other functional change intended. Also, currently machine_check() has a comment to explain why no function call is allowed after load_segments(). After changing to use the new percpu boolean to control whether to perform WBINVD when calling the relocate_kernel(), that comment isn't needed anymore. But it is still a useful comment, so expand the comment around load_segments() to mention that due to depth tracking no function call can be made after load_segments(). [*] The "race" in native_stop_other_cpus() Commit 1f5e7eb: ("x86/smp: Make stop_other_cpus() more robust") introduced a new 'cpus_stop_mask' to resolve an "intermittent lockups on poweroff" issue which was caused by the WBINVD in stop_this_cpu(). Specifically, the new cpumask resolved the below problem mentioned in that commit: CPU0 CPU1 stop_other_cpus() send_IPIs(REBOOT); stop_this_cpu() while (num_online_cpus() > 1); set_online(false); proceed... -> hang wbinvd() While it fixed the reported issue, that commit explained the new cpumask "cannot plug all holes either". This is because it doesn't address the "race" mentioned in the #3 in the comment in native_stop_other_cpus(): /* * 1) Send an IPI on the reboot vector to all other CPUs. * * The other CPUs should react on it after leaving critical * sections and re-enabling interrupts. They might still hold * locks, but there is nothing which can be done about that. * * 2) Wait for all other CPUs to report that they reached the * HLT loop in stop_this_cpu() * * 3) If #2 timed out send an NMI to the CPUs which did not * yet report * * 4) Wait for all other CPUs to report that they reached the * HLT loop in stop_this_cpu() * * #3 can obviously race against a CPU reaching the HLT loop late. * That CPU will have reported already and the "have all CPUs * reached HLT" condition will be true despite the fact that the * other CPU is still handling the NMI. Again, there is no * protection against that as "disabled" APICs still respond to * NMIs. */ Consider below case: CPU 0 CPU 1 native_stop_other_cpus() stop_this_cpu() // sends REBOOT IPI to stop remote CPUs ... wbinvd(); // wait timesout, try NMI if (!cpumask_empty(&cpus_stop_mask)) { for_each_cpu(cpu, &cpus_stop_mask) { ... cpumask_clear_cpu(cpu, &cpus_stop_mask); hlt; // send NMI ---> wakeup from hlt and run stop_this_cpu(): // WAIT CPUs TO STOP while (!cpumask_empty( &cpus_stop_mask) && ...) } ... proceed ... wbinvd(); ... hlt; The "WAIT CPUs TO STOP" is supposed to wait until all remote CPUs are stopped, but actually it quits immediately because the remote CPUs have been cleared in cpus_stop_mask when stop_this_cpu() is called from the REBOOT IPI. Doing WBINVD in stop_this_cpu() could potentially increase the chance to trigger the above "race" despite it's still rare to happen. Signed-off-by: Kai Huang <kai.huang@intel.com>

The ieee80211 skb control block key (set when skb was queued) could have been removed before ieee80211_tx_dequeue() call. ieee80211_tx_dequeue() already called ieee80211_tx_h_select_key() to get the current key, but the latter do not update the key in skb control block in case it is NULL. Because some drivers actually use this key in their TX callbacks (e.g. ath1{1,2}k_mac_op_tx()) this could lead to the use after free below: BUG: KASAN: slab-use-after-free in ath11k_mac_op_tx+0x590/0x61c Read of size 4 at addr ffffff803083c248 by task kworker/u16:4/1440 CPU: 3 UID: 0 PID: 1440 Comm: kworker/u16:4 Not tainted 6.13.0-ge128f627f404 #2 Hardware name: HW (DT) Workqueue: bat_events batadv_send_outstanding_bcast_packet Call trace: show_stack+0x14/0x1c (C) dump_stack_lvl+0x58/0x74 print_report+0x164/0x4c0 kasan_report+0xac/0xe8 __asan_report_load4_noabort+0x1c/0x24 ath11k_mac_op_tx+0x590/0x61c ieee80211_handle_wake_tx_queue+0x12c/0x1c8 ieee80211_queue_skb+0xdcc/0x1b4c ieee80211_tx+0x1ec/0x2bc ieee80211_xmit+0x224/0x324 __ieee80211_subif_start_xmit+0x85c/0xcf8 ieee80211_subif_start_xmit+0xc0/0xec4 dev_hard_start_xmit+0xf4/0x28c __dev_queue_xmit+0x6ac/0x318c batadv_send_skb_packet+0x38c/0x4b0 batadv_send_outstanding_bcast_packet+0x110/0x328 process_one_work+0x578/0xc10 worker_thread+0x4bc/0xc7c kthread+0x2f8/0x380 ret_from_fork+0x10/0x20 Allocated by task 1906: kasan_save_stack+0x28/0x4c kasan_save_track+0x1c/0x40 kasan_save_alloc_info+0x3c/0x4c __kasan_kmalloc+0xac/0xb0 __kmalloc_noprof+0x1b4/0x380 ieee80211_key_alloc+0x3c/0xb64 ieee80211_add_key+0x1b4/0x71c nl80211_new_key+0x2b4/0x5d8 genl_family_rcv_msg_doit+0x198/0x240 <...> Freed by task 1494: kasan_save_stack+0x28/0x4c kasan_save_track+0x1c/0x40 kasan_save_free_info+0x48/0x94 __kasan_slab_free+0x48/0x60 kfree+0xc8/0x31c kfree_sensitive+0x70/0x80 ieee80211_key_free_common+0x10c/0x174 ieee80211_free_keys+0x188/0x46c ieee80211_stop_mesh+0x70/0x2cc ieee80211_leave_mesh+0x1c/0x60 cfg80211_leave_mesh+0xe0/0x280 cfg80211_leave+0x1e0/0x244 <...> Reset SKB control block key before calling ieee80211_tx_h_select_key() to avoid that. Fixes: bb42f2d ("mac80211: Move reorder-sensitive TX handlers to after TXQ dequeue") Signed-off-by: Remi Pommarel <repk@triplefau.lt> Link: https://patch.msgid.link/06aa507b853ca385ceded81c18b0a6dd0f081bc8.1742833382.git.repk@triplefau.lt Signed-off-by: Johannes Berg <johannes.berg@intel.com>

SMC consists of two sockets: smc_sock and kernel TCP socket. Currently, there are two ways of creating the sockets, and syzbot reported a lockdep splat [0] for the newer way introduced by commit d25a92c ("net/smc: Introduce IPPROTO_SMC"). socket(AF_SMC , SOCK_STREAM, SMCPROTO_SMC or SMCPROTO_SMC6) socket(AF_INET or AF_INET6, SOCK_STREAM, IPPROTO_SMC) When a socket is allocated, sock_lock_init() sets a lockdep lock class to sk->sk_lock.slock based on its protocol family. In the IPPROTO_SMC case, AF_INET or AF_INET6 lock class is assigned to smc_sock. The repro sets IPV6_JOIN_ANYCAST for IPv6 UDP and SMC socket and exercises smc_switch_to_fallback() for IPPROTO_SMC. 1. smc_switch_to_fallback() is called under lock_sock() and holds smc->clcsock_release_lock. sk_lock-AF_INET6 -> &smc->clcsock_release_lock (sk_lock-AF_SMC) 2. Setting IPV6_JOIN_ANYCAST to SMC holds smc->clcsock_release_lock and calls setsockopt() for the kernel TCP socket, which holds RTNL and the kernel socket's lock_sock(). &smc->clcsock_release_lock -> rtnl_mutex (-> k-sk_lock-AF_INET6) 3. Setting IPV6_JOIN_ANYCAST to UDP holds RTNL and lock_sock(). rtnl_mutex -> sk_lock-AF_INET6 Then, lockdep detects a false-positive circular locking, .-> sk_lock-AF_INET6 -> &smc->clcsock_release_lock -> rtnl_mutex -. `-----------------------------------------------------------------' but IPPROTO_SMC should have the same locking rule as AF_SMC. sk_lock-AF_SMC -> &smc->clcsock_release_lock -> rtnl_mutex -> k-sk_lock-AF_INET6 Let's set the same lock class for smc_sock. Given AF_SMC uses the same lock class for SMCPROTO_SMC and SMCPROTO_SMC6, we do not need to separate the class for AF_INET and AF_INET6. [0]: WARNING: possible circular locking dependency detected 6.14.0-rc3-syzkaller-00267-gff202c5028a1 #0 Not tainted syz.4.1528/11571 is trying to acquire lock: ffffffff8fef8de8 (rtnl_mutex){+.+.}-{4:4}, at: ipv6_sock_ac_close+0xd9/0x110 net/ipv6/anycast.c:220 but task is already holding lock: ffff888027f596a8 (&smc->clcsock_release_lock){+.+.}-{4:4}, at: smc_clcsock_release+0x75/0xe0 net/smc/smc_close.c:30 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #2 (&smc->clcsock_release_lock){+.+.}-{4:4}: __mutex_lock_common kernel/locking/mutex.c:585 [inline] __mutex_lock+0x19b/0xb10 kernel/locking/mutex.c:730 smc_switch_to_fallback+0x2d/0xa00 net/smc/af_smc.c:903 smc_sendmsg+0x13d/0x520 net/smc/af_smc.c:2781 sock_sendmsg_nosec net/socket.c:718 [inline] __sock_sendmsg net/socket.c:733 [inline] ____sys_sendmsg+0xaaf/0xc90 net/socket.c:2573 ___sys_sendmsg+0x135/0x1e0 net/socket.c:2627 __sys_sendmsg+0x16e/0x220 net/socket.c:2659 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xcd/0x250 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f -> #1 (sk_lock-AF_INET6){+.+.}-{0:0}: lock_sock_nested+0x3a/0xf0 net/core/sock.c:3645 lock_sock include/net/sock.h:1624 [inline] sockopt_lock_sock net/core/sock.c:1133 [inline] sockopt_lock_sock+0x54/0x70 net/core/sock.c:1124 do_ipv6_setsockopt+0x2160/0x4520 net/ipv6/ipv6_sockglue.c:567 ipv6_setsockopt+0xcb/0x170 net/ipv6/ipv6_sockglue.c:993 udpv6_setsockopt+0x7d/0xd0 net/ipv6/udp.c:1850 do_sock_setsockopt+0x222/0x480 net/socket.c:2303 __sys_setsockopt+0x1a0/0x230 net/socket.c:2328 __do_sys_setsockopt net/socket.c:2334 [inline] __se_sys_setsockopt net/socket.c:2331 [inline] __x64_sys_setsockopt+0xbd/0x160 net/socket.c:2331 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xcd/0x250 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f -> #0 (rtnl_mutex){+.+.}-{4:4}: check_prev_add kernel/locking/lockdep.c:3163 [inline] check_prevs_add kernel/locking/lockdep.c:3282 [inline] validate_chain kernel/locking/lockdep.c:3906 [inline] __lock_acquire+0x249e/0x3c40 kernel/locking/lockdep.c:5228 lock_acquire.part.0+0x11b/0x380 kernel/locking/lockdep.c:5851 __mutex_lock_common kernel/locking/mutex.c:585 [inline] __mutex_lock+0x19b/0xb10 kernel/locking/mutex.c:730 ipv6_sock_ac_close+0xd9/0x110 net/ipv6/anycast.c:220 inet6_release+0x47/0x70 net/ipv6/af_inet6.c:485 __sock_release net/socket.c:647 [inline] sock_release+0x8e/0x1d0 net/socket.c:675 smc_clcsock_release+0xb7/0xe0 net/smc/smc_close.c:34 __smc_release+0x5c2/0x880 net/smc/af_smc.c:301 smc_release+0x1fc/0x5f0 net/smc/af_smc.c:344 __sock_release+0xb0/0x270 net/socket.c:647 sock_close+0x1c/0x30 net/socket.c:1398 __fput+0x3ff/0xb70 fs/file_table.c:464 task_work_run+0x14e/0x250 kernel/task_work.c:227 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline] exit_to_user_mode_loop kernel/entry/common.c:114 [inline] exit_to_user_mode_prepare include/linux/entry-common.h:329 [inline] __syscall_exit_to_user_mode_work kernel/entry/common.c:207 [inline] syscall_exit_to_user_mode+0x27b/0x2a0 kernel/entry/common.c:218 do_syscall_64+0xda/0x250 arch/x86/entry/common.c:89 entry_SYSCALL_64_after_hwframe+0x77/0x7f other info that might help us debug this: Chain exists of: rtnl_mutex --> sk_lock-AF_INET6 --> &smc->clcsock_release_lock Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&smc->clcsock_release_lock); lock(sk_lock-AF_INET6); lock(&smc->clcsock_release_lock); lock(rtnl_mutex); *** DEADLOCK *** 2 locks held by syz.4.1528/11571: #0: ffff888077e88208 (&sb->s_type->i_mutex_key#10){+.+.}-{4:4}, at: inode_lock include/linux/fs.h:877 [inline] #0: ffff888077e88208 (&sb->s_type->i_mutex_key#10){+.+.}-{4:4}, at: __sock_release+0x86/0x270 net/socket.c:646 #1: ffff888027f596a8 (&smc->clcsock_release_lock){+.+.}-{4:4}, at: smc_clcsock_release+0x75/0xe0 net/smc/smc_close.c:30 stack backtrace: CPU: 0 UID: 0 PID: 11571 Comm: syz.4.1528 Not tainted 6.14.0-rc3-syzkaller-00267-gff202c5028a1 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 02/12/2025 Call Trace: <TASK> __dump_stack lib/dump_stack.c:94 [inline] dump_stack_lvl+0x116/0x1f0 lib/dump_stack.c:120 print_circular_bug+0x490/0x760 kernel/locking/lockdep.c:2076 check_noncircular+0x31a/0x400 kernel/locking/lockdep.c:2208 check_prev_add kernel/locking/lockdep.c:3163 [inline] check_prevs_add kernel/locking/lockdep.c:3282 [inline] validate_chain kernel/locking/lockdep.c:3906 [inline] __lock_acquire+0x249e/0x3c40 kernel/locking/lockdep.c:5228 lock_acquire.part.0+0x11b/0x380 kernel/locking/lockdep.c:5851 __mutex_lock_common kernel/locking/mutex.c:585 [inline] __mutex_lock+0x19b/0xb10 kernel/locking/mutex.c:730 ipv6_sock_ac_close+0xd9/0x110 net/ipv6/anycast.c:220 inet6_release+0x47/0x70 net/ipv6/af_inet6.c:485 __sock_release net/socket.c:647 [inline] sock_release+0x8e/0x1d0 net/socket.c:675 smc_clcsock_release+0xb7/0xe0 net/smc/smc_close.c:34 __smc_release+0x5c2/0x880 net/smc/af_smc.c:301 smc_release+0x1fc/0x5f0 net/smc/af_smc.c:344 __sock_release+0xb0/0x270 net/socket.c:647 sock_close+0x1c/0x30 net/socket.c:1398 __fput+0x3ff/0xb70 fs/file_table.c:464 task_work_run+0x14e/0x250 kernel/task_work.c:227 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline] exit_to_user_mode_loop kernel/entry/common.c:114 [inline] exit_to_user_mode_prepare include/linux/entry-common.h:329 [inline] __syscall_exit_to_user_mode_work kernel/entry/common.c:207 [inline] syscall_exit_to_user_mode+0x27b/0x2a0 kernel/entry/common.c:218 do_syscall_64+0xda/0x250 arch/x86/entry/common.c:89 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f8b4b38d169 Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 a8 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007ffe4efd22d8 EFLAGS: 00000246 ORIG_RAX: 00000000000001b4 RAX: 0000000000000000 RBX: 00000000000b14a3 RCX: 00007f8b4b38d169 RDX: 0000000000000000 RSI: 000000000000001e RDI: 0000000000000003 RBP: 00007f8b4b5a7ba0 R08: 0000000000000001 R09: 000000114efd25cf R10: 00007f8b4b200000 R11: 0000000000000246 R12: 00007f8b4b5a5fac R13: 00007f8b4b5a5fa0 R14: ffffffffffffffff R15: 00007ffe4efd23f0 </TASK> Fixes: d25a92c ("net/smc: Introduce IPPROTO_SMC") Reported-by: syzbot+be6f4b383534d88989f7@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=be6f4b383534d88989f7 Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com> Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com> Link: https://patch.msgid.link/20250407170332.26959-1-kuniyu@amazon.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Ido Schimmel says: ==================== fib_rules: Fix iif / oif matching on L3 master device Patch #1 fixes a recently reported regression regarding FIB rules that match on iif / oif being a VRF device. Patch #2 adds test cases to the FIB rules selftest. ==================== Link: https://patch.msgid.link/20250414172022.242991-1-idosch@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>

There is a potential deadlock if we do report zones in an IO context, detailed in below lockdep report. When one process do a report zones and another process freezes the block device, the report zones side cannot allocate a tag because the freeze is already started. This can thus result in new block group creation to hang forever, blocking the write path. Thankfully, a new block group should be created on empty zones. So, reporting the zones is not necessary and we can set the write pointer = 0 and load the zone capacity from the block layer using bdev_zone_capacity() helper. ====================================================== WARNING: possible circular locking dependency detected 6.14.0-rc1 #252 Not tainted ------------------------------------------------------ modprobe/1110 is trying to acquire lock: ffff888100ac83e0 ((work_completion)(&(&wb->dwork)->work)){+.+.}-{0:0}, at: __flush_work+0x38f/0xb60 but task is already holding lock: ffff8881205b6f20 (&q->q_usage_counter(queue)#16){++++}-{0:0}, at: sd_remove+0x85/0x130 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #3 (&q->q_usage_counter(queue)#16){++++}-{0:0}: blk_queue_enter+0x3d9/0x500 blk_mq_alloc_request+0x47d/0x8e0 scsi_execute_cmd+0x14f/0xb80 sd_zbc_do_report_zones+0x1c1/0x470 sd_zbc_report_zones+0x362/0xd60 blkdev_report_zones+0x1b1/0x2e0 btrfs_get_dev_zones+0x215/0x7e0 [btrfs] btrfs_load_block_group_zone_info+0x6d2/0x2c10 [btrfs] btrfs_make_block_group+0x36b/0x870 [btrfs] btrfs_create_chunk+0x147d/0x2320 [btrfs] btrfs_chunk_alloc+0x2ce/0xcf0 [btrfs] start_transaction+0xce6/0x1620 [btrfs] btrfs_uuid_scan_kthread+0x4ee/0x5b0 [btrfs] kthread+0x39d/0x750 ret_from_fork+0x30/0x70 ret_from_fork_asm+0x1a/0x30 -> #2 (&fs_info->dev_replace.rwsem){++++}-{4:4}: down_read+0x9b/0x470 btrfs_map_block+0x2ce/0x2ce0 [btrfs] btrfs_submit_chunk+0x2d4/0x16c0 [btrfs] btrfs_submit_bbio+0x16/0x30 [btrfs] btree_write_cache_pages+0xb5a/0xf90 [btrfs] do_writepages+0x17f/0x7b0 __writeback_single_inode+0x114/0xb00 writeback_sb_inodes+0x52b/0xe00 wb_writeback+0x1a7/0x800 wb_workfn+0x12a/0xbd0 process_one_work+0x85a/0x1460 worker_thread+0x5e2/0xfc0 kthread+0x39d/0x750 ret_from_fork+0x30/0x70 ret_from_fork_asm+0x1a/0x30 -> #1 (&fs_info->zoned_meta_io_lock){+.+.}-{4:4}: __mutex_lock+0x1aa/0x1360 btree_write_cache_pages+0x252/0xf90 [btrfs] do_writepages+0x17f/0x7b0 __writeback_single_inode+0x114/0xb00 writeback_sb_inodes+0x52b/0xe00 wb_writeback+0x1a7/0x800 wb_workfn+0x12a/0xbd0 process_one_work+0x85a/0x1460 worker_thread+0x5e2/0xfc0 kthread+0x39d/0x750 ret_from_fork+0x30/0x70 ret_from_fork_asm+0x1a/0x30 -> #0 ((work_completion)(&(&wb->dwork)->work)){+.+.}-{0:0}: __lock_acquire+0x2f52/0x5ea0 lock_acquire+0x1b1/0x540 __flush_work+0x3ac/0xb60 wb_shutdown+0x15b/0x1f0 bdi_unregister+0x172/0x5b0 del_gendisk+0x841/0xa20 sd_remove+0x85/0x130 device_release_driver_internal+0x368/0x520 bus_remove_device+0x1f1/0x3f0 device_del+0x3bd/0x9c0 __scsi_remove_device+0x272/0x340 scsi_forget_host+0xf7/0x170 scsi_remove_host+0xd2/0x2a0 sdebug_driver_remove+0x52/0x2f0 [scsi_debug] device_release_driver_internal+0x368/0x520 bus_remove_device+0x1f1/0x3f0 device_del+0x3bd/0x9c0 device_unregister+0x13/0xa0 sdebug_do_remove_host+0x1fb/0x290 [scsi_debug] scsi_debug_exit+0x17/0x70 [scsi_debug] __do_sys_delete_module.isra.0+0x321/0x520 do_syscall_64+0x93/0x180 entry_SYSCALL_64_after_hwframe+0x76/0x7e other info that might help us debug this: Chain exists of: (work_completion)(&(&wb->dwork)->work) --> &fs_info->dev_replace.rwsem --> &q->q_usage_counter(queue)#16 Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&q->q_usage_counter(queue)#16); lock(&fs_info->dev_replace.rwsem); lock(&q->q_usage_counter(queue)#16); lock((work_completion)(&(&wb->dwork)->work)); *** DEADLOCK *** 5 locks held by modprobe/1110: #0: ffff88811f7bc108 (&dev->mutex){....}-{4:4}, at: device_release_driver_internal+0x8f/0x520 #1: ffff8881022ee0e0 (&shost->scan_mutex){+.+.}-{4:4}, at: scsi_remove_host+0x20/0x2a0 #2: ffff88811b4c4378 (&dev->mutex){....}-{4:4}, at: device_release_driver_internal+0x8f/0x520 #3: ffff8881205b6f20 (&q->q_usage_counter(queue)#16){++++}-{0:0}, at: sd_remove+0x85/0x130 #4: ffffffffa3284360 (rcu_read_lock){....}-{1:3}, at: __flush_work+0xda/0xb60 stack backtrace: CPU: 0 UID: 0 PID: 1110 Comm: modprobe Not tainted 6.14.0-rc1 #252 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.3-3.fc41 04/01/2014 Call Trace: <TASK> dump_stack_lvl+0x6a/0x90 print_circular_bug.cold+0x1e0/0x274 check_noncircular+0x306/0x3f0 ? __pfx_check_noncircular+0x10/0x10 ? mark_lock+0xf5/0x1650 ? __pfx_check_irq_usage+0x10/0x10 ? lockdep_lock+0xca/0x1c0 ? __pfx_lockdep_lock+0x10/0x10 __lock_acquire+0x2f52/0x5ea0 ? __pfx___lock_acquire+0x10/0x10 ? __pfx_mark_lock+0x10/0x10 lock_acquire+0x1b1/0x540 ? __flush_work+0x38f/0xb60 ? __pfx_lock_acquire+0x10/0x10 ? __pfx_lock_release+0x10/0x10 ? mark_held_locks+0x94/0xe0 ? __flush_work+0x38f/0xb60 __flush_work+0x3ac/0xb60 ? __flush_work+0x38f/0xb60 ? __pfx_mark_lock+0x10/0x10 ? __pfx___flush_work+0x10/0x10 ? __pfx_wq_barrier_func+0x10/0x10 ? __pfx___might_resched+0x10/0x10 ? mark_held_locks+0x94/0xe0 wb_shutdown+0x15b/0x1f0 bdi_unregister+0x172/0x5b0 ? __pfx_bdi_unregister+0x10/0x10 ? up_write+0x1ba/0x510 del_gendisk+0x841/0xa20 ? __pfx_del_gendisk+0x10/0x10 ? _raw_spin_unlock_irqrestore+0x35/0x60 ? __pm_runtime_resume+0x79/0x110 sd_remove+0x85/0x130 device_release_driver_internal+0x368/0x520 ? kobject_put+0x5d/0x4a0 bus_remove_device+0x1f1/0x3f0 device_del+0x3bd/0x9c0 ? __pfx_device_del+0x10/0x10 __scsi_remove_device+0x272/0x340 scsi_forget_host+0xf7/0x170 scsi_remove_host+0xd2/0x2a0 sdebug_driver_remove+0x52/0x2f0 [scsi_debug] ? kernfs_remove_by_name_ns+0xc0/0xf0 device_release_driver_internal+0x368/0x520 ? kobject_put+0x5d/0x4a0 bus_remove_device+0x1f1/0x3f0 device_del+0x3bd/0x9c0 ? __pfx_device_del+0x10/0x10 ? __pfx___mutex_unlock_slowpath+0x10/0x10 device_unregister+0x13/0xa0 sdebug_do_remove_host+0x1fb/0x290 [scsi_debug] scsi_debug_exit+0x17/0x70 [scsi_debug] __do_sys_delete_module.isra.0+0x321/0x520 ? __pfx___do_sys_delete_module.isra.0+0x10/0x10 ? __pfx_slab_free_after_rcu_debug+0x10/0x10 ? kasan_save_stack+0x2c/0x50 ? kasan_record_aux_stack+0xa3/0xb0 ? __call_rcu_common.constprop.0+0xc4/0xfb0 ? kmem_cache_free+0x3a0/0x590 ? __x64_sys_close+0x78/0xd0 do_syscall_64+0x93/0x180 ? lock_is_held_type+0xd5/0x130 ? __call_rcu_common.constprop.0+0x3c0/0xfb0 ? lockdep_hardirqs_on+0x78/0x100 ? __call_rcu_common.constprop.0+0x3c0/0xfb0 ? __pfx___call_rcu_common.constprop.0+0x10/0x10 ? kmem_cache_free+0x3a0/0x590 ? lockdep_hardirqs_on_prepare+0x16d/0x400 ? do_syscall_64+0x9f/0x180 ? lockdep_hardirqs_on+0x78/0x100 ? do_syscall_64+0x9f/0x180 ? __pfx___x64_sys_openat+0x10/0x10 ? lockdep_hardirqs_on_prepare+0x16d/0x400 ? do_syscall_64+0x9f/0x180 ? lockdep_hardirqs_on+0x78/0x100 ? do_syscall_64+0x9f/0x180 entry_SYSCALL_64_after_hwframe+0x76/0x7e RIP: 0033:0x7f436712b68b RSP: 002b:00007ffe9f1a8658 EFLAGS: 00000206 ORIG_RAX: 00000000000000b0 RAX: ffffffffffffffda RBX: 00005559b367fd80 RCX: 00007f436712b68b RDX: 0000000000000000 RSI: 0000000000000800 RDI: 00005559b367fde8 RBP: 00007ffe9f1a8680 R08: 1999999999999999 R09: 0000000000000000 R10: 00007f43671a5fe0 R11: 0000000000000206 R12: 0000000000000000 R13: 00007ffe9f1a86b0 R14: 0000000000000000 R15: 0000000000000000 </TASK> Reported-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com> CC: <stable@vger.kernel.org> # 6.13+ Tested-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com> Reviewed-by: Damien Le Moal <dlemoal@kernel.org> Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Signed-off-by: Naohiro Aota <naohiro.aota@wdc.com> Signed-off-by: David Sterba <dsterba@suse.com>

Communicating with the hypervisor using the shared GHCB page requires clearing the C bit in the mapping of that page. When executing in the context of the EFI boot services, the page tables are owned by the firmware, and this manipulation is not possible. So switch to a different API for accepting memory in SEV-SNP guests, one which is actually supported at the point during boot where the EFI stub may need to accept memory, but the SEV-SNP init code has not executed yet. For simplicity, also switch the memory acceptance carried out by the decompressor when not booting via EFI - this only involves the allocation for the decompressed kernel, and is generally only called after kexec, as normal boot will jump straight into the kernel from the EFI stub. Fixes: 6c32117 ("x86/sev: Add SNP-specific unaccepted memory support") Tested-by: Tom Lendacky <thomas.lendacky@amd.com> Co-developed-by: Tom Lendacky <thomas.lendacky@amd.com> Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com> Signed-off-by: Ard Biesheuvel <ardb@kernel.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Cc: <stable@vger.kernel.org> Cc: Dionna Amalie Glaze <dionnaglaze@google.com> Cc: Kevin Loughlin <kevinloughlin@google.com> Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: linux-efi@vger.kernel.org Link: https://lore.kernel.org/r/20250404082921.2767593-8-ardb+git@google.com # discussion thread #1 Link: https://lore.kernel.org/r/20250410132850.3708703-2-ardb+git@google.com # discussion thread #2 Link: https://lore.kernel.org/r/20250417202120.1002102-2-ardb+git@google.com # final submission

[BUG] There is a bug report that a syzbot reproducer can lead to the following busy inode at unmount time: BTRFS info (device loop1): last unmount of filesystem 1680000e-3c1e-4c46-84b6-56bd3909af50 VFS: Busy inodes after unmount of loop1 (btrfs) ------------[ cut here ]------------ kernel BUG at fs/super.c:650! Oops: invalid opcode: 0000 [#1] SMP KASAN NOPTI CPU: 0 UID: 0 PID: 48168 Comm: syz-executor Not tainted 6.15.0-rc2-00471-g119009db2674 #2 PREEMPT(full) Hardware name: QEMU Ubuntu 24.04 PC (i440FX + PIIX, 1996), BIOS 1.16.3-debian-1.16.3-2 04/01/2014 RIP: 0010:generic_shutdown_super+0x2e9/0x390 fs/super.c:650 Call Trace: <TASK> kill_anon_super+0x3a/0x60 fs/super.c:1237 btrfs_kill_super+0x3b/0x50 fs/btrfs/super.c:2099 deactivate_locked_super+0xbe/0x1a0 fs/super.c:473 deactivate_super fs/super.c:506 [inline] deactivate_super+0xe2/0x100 fs/super.c:502 cleanup_mnt+0x21f/0x440 fs/namespace.c:1435 task_work_run+0x14d/0x240 kernel/task_work.c:227 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline] exit_to_user_mode_loop kernel/entry/common.c:114 [inline] exit_to_user_mode_prepare include/linux/entry-common.h:329 [inline] __syscall_exit_to_user_mode_work kernel/entry/common.c:207 [inline] syscall_exit_to_user_mode+0x269/0x290 kernel/entry/common.c:218 do_syscall_64+0xd4/0x250 arch/x86/entry/syscall_64.c:100 entry_SYSCALL_64_after_hwframe+0x77/0x7f </TASK> [CAUSE] When btrfs_alloc_path() failed, btrfs_iget() directly returned without releasing the inode already allocated by btrfs_iget_locked(). This results the above busy inode and trigger the kernel BUG. [FIX] Fix it by calling iget_failed() if btrfs_alloc_path() failed. If we hit error inside btrfs_read_locked_inode(), it will properly call iget_failed(), so nothing to worry about. Although the iget_failed() cleanup inside btrfs_read_locked_inode() is a break of the normal error handling scheme, let's fix the obvious bug and backport first, then rework the error handling later. Reported-by: Penglei Jiang <superman.xpt@gmail.com> Link: https://lore.kernel.org/linux-btrfs/20250421102425.44431-1-superman.xpt@gmail.com/ Fixes: 7c855e1 ("btrfs: remove conditional path allocation in btrfs_read_locked_inode()") CC: stable@vger.kernel.org # 6.13+ Reviewed-by: Qu Wenruo <wqu@suse.com> Signed-off-by: Penglei Jiang <superman.xpt@gmail.com> Signed-off-by: David Sterba <dsterba@suse.com>

…ux/kernel/git/kvmarm/kvmarm into HEAD KVM/arm64 fixes for 6.15, round #2 - Single fix for broken usage of 'multi-MIDR' infrastructure in PI code, adding an open-coded erratum check for everyone's favorite pile of sand: Cavium ThunderX

…_USERCOPY=y crash Borislav Petkov reported the following boot crash on x86-32, with CONFIG_HARDENED_USERCOPY=y: | usercopy: Kernel memory overwrite attempt detected to SLUB object 'task_struct' (offset 2112, size 160)! | ... | kernel BUG at mm/usercopy.c:102! So the useroffset and usersize arguments are what control the allowed window of copying in/out of the "task_struct" kmem cache: /* create a slab on which task_structs can be allocated */ task_struct_whitelist(&useroffset, &usersize); task_struct_cachep = kmem_cache_create_usercopy("task_struct", arch_task_struct_size, align, SLAB_PANIC|SLAB_ACCOUNT, useroffset, usersize, NULL); task_struct_whitelist() positions this window based on the location of the thread_struct within task_struct, and gets the arch-specific details via arch_thread_struct_whitelist(offset, size): static void __init task_struct_whitelist(unsigned long *offset, unsigned long *size) { /* Fetch thread_struct whitelist for the architecture. */ arch_thread_struct_whitelist(offset, size); /* * Handle zero-sized whitelist or empty thread_struct, otherwise * adjust offset to position of thread_struct in task_struct. */ if (unlikely(*size == 0)) *offset = 0; else *offset += offsetof(struct task_struct, thread); } Commit cb7ca40 ("x86/fpu: Make task_struct::thread constant size") removed the logic for the window, leaving: static inline void arch_thread_struct_whitelist(unsigned long *offset, unsigned long *size) { *offset = 0; *size = 0; } So now there is no window that usercopy hardening will allow to be copied in/out of task_struct. But as reported above, there *is* a copy in copy_uabi_to_xstate(). (It seems there are several, actually.) int copy_sigframe_from_user_to_xstate(struct task_struct *tsk, const void __user *ubuf) { return copy_uabi_to_xstate(x86_task_fpu(tsk)->fpstate, NULL, ubuf, &tsk->thread.pkru); } This appears to be writing into x86_task_fpu(tsk)->fpstate. With or without CONFIG_X86_DEBUG_FPU, this resolves to: ((struct fpu *)((void *)(task) + sizeof(*(task)))) i.e. the memory "after task_struct" is cast to "struct fpu", and the uses the "fpstate" pointer. How that pointer gets set looks to be variable, but I think the one we care about here is: fpu->fpstate = &fpu->__fpstate; And struct fpu::__fpstate says: struct fpstate __fpstate; /* * WARNING: '__fpstate' is dynamically-sized. Do not put * anything after it here. */ So we're still dealing with a dynamically sized thing, even if it's not within the literal struct task_struct -- it's still in the kmem cache, though. Looking at the kmem cache size, it has allocated "arch_task_struct_size" bytes, which is calculated in fpu__init_task_struct_size(): int task_size = sizeof(struct task_struct); task_size += sizeof(struct fpu); /* * Subtract off the static size of the register state. * It potentially has a bunch of padding. */ task_size -= sizeof(union fpregs_state); /* * Add back the dynamically-calculated register state * size. */ task_size += fpu_kernel_cfg.default_size; /* * We dynamically size 'struct fpu', so we require that * 'state' be at the end of 'it: */ CHECK_MEMBER_AT_END_OF(struct fpu, __fpstate); arch_task_struct_size = task_size; So, this is still copying out of the kmem cache for task_struct, and the window seems unchanged (still fpu regs). This is what the window was before: void fpu_thread_struct_whitelist(unsigned long *offset, unsigned long *size) { *offset = offsetof(struct thread_struct, fpu.__fpstate.regs); *size = fpu_kernel_cfg.default_size; } And the same commit I mentioned above removed it. I think the misunderstanding is here: | The fpu_thread_struct_whitelist() quirk to hardened usercopy can be removed, | now that the FPU structure is not embedded in the task struct anymore, which | reduces text footprint a bit. Yes, FPU is no longer in task_struct, but it IS in the kmem cache named "task_struct", since the fpstate is still being allocated there. Partially revert the earlier mentioned commit, along with a recalculation of the fpstate regs location. Fixes: cb7ca40 ("x86/fpu: Make task_struct::thread constant size") Reported-by: Borislav Petkov (AMD) <bp@alien8.de> Tested-by: Borislav Petkov (AMD) <bp@alien8.de> Signed-off-by: Kees Cook <kees@kernel.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Cc: Chang S. Bae <chang.seok.bae@intel.com> Cc: Gustavo A. R. Silva <gustavoars@kernel.org> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Andy Lutomirski <luto@kernel.org> Cc: Brian Gerst <brgerst@gmail.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: linux-hardening@vger.kernel.org Link: https://lore.kernel.org/all/20250409211127.3544993-1-mingo@kernel.org/ # Discussion #1 Link: https://lore.kernel.org/r/202505041418.F47130C4C8@keescook # Discussion #2

On SME platforms, at hardware level dirty cachelines with and without encryption bit of the same memory can coexist, and the CPU can flush them back to memory in random order. During kexec, the caches must be flushed before jumping to the new kernel to avoid silent memory corruption when a cacheline with a different encryption property is written back over whatever encryption properties the new kernel is using. TDX also needs cache flush during kexec for the same reason. It would be good to implement a generic way to flush cache instead of scattering checks for each feature all around. During kexec, the kexec-ing CPU sends IPIs to all remote CPUs to stop them before it boots to the new kernel. For SME the kernel basically encrypts all memory including the kernel itself by manipulating page tables. A simple memory write from the kernel could dirty cachelines. Currently, the kernel uses WBINVD to flush cache for SME during kexec in two places: 1) the one in stop_this_cpu() for all remote CPUs when the kexec-ing CPU stops them; 2) the one in the relocate_kernel() where the kexec-ing CPU jumps to the new kernel. Unlike SME, TDX can only dirty cachelines when it is used (i.e., when SEAMCALLs are performed). Since there are no more SEAMCALLs after the aforementioned WBINVDs, leverage this for TDX. In order to have a generic way to cover both SME and TDX, use a percpu boolean to indicate the cache may be in an incoherent state thus cache flush is needed during kexec, and turn on the boolean for SME. TDX can then leverage it by also turning the boolean on. A percpu boolean isn't strictly needed for SME since it is activated at very early boot time and on all CPUs. A global flag would be sufficient. A percpu boolean isn't strictly needed for SME since it is activated at very early boot time and on all CPUs. A global flag would be sufficient. But using a percpu variable has two benefits. Foremost, the state that is being tracked here (percpu cache coherency situation requiring a flush) is percpu, so a percpu state is a more direct and natural fit. Secondly, it will fit TDX's usage better since the percpu var can be set when a CPU makes a SEAMCALL, and cleared when another WBINVD on the CPU obviates the need for a kexec-time WBINVD. Saving kexec-time WBINVD is valuable, because there is an existing race[*] where kexec could proceed while another CPU is active. WBINVD could make this race worse, so it's worth skipping it when possible. Today the first WBINVD in the stop_this_cpu() is performed when SME is *supported* by the platform, and the second WBINVD is done in relocate_kernel() when SME is *activated* by the kernel. Make things simple by changing to do the second WBINVD when the platform supports SME. This allows the kernel to simply turn on this percpu boolean when bringing up a CPU by checking whether the platform supports SME. No other functional change intended. Also, currently machine_check() has a comment to explain why no function call is allowed after load_segments(). After changing to use the new percpu boolean to control whether to perform WBINVD when calling the relocate_kernel(), that comment isn't needed anymore. But it is still a useful comment, so expand the comment around load_segments() to mention that due to depth tracking no function call can be made after load_segments(). [*] The "race" in native_stop_other_cpus() Commit 1f5e7eb: ("x86/smp: Make stop_other_cpus() more robust") introduced a new 'cpus_stop_mask' to resolve an "intermittent lockups on poweroff" issue which was caused by the WBINVD in stop_this_cpu(). Specifically, the new cpumask resolved the below problem mentioned in that commit: CPU0 CPU1 stop_other_cpus() send_IPIs(REBOOT); stop_this_cpu() while (num_online_cpus() > 1); set_online(false); proceed... -> hang wbinvd() While it fixed the reported issue, that commit explained the new cpumask "cannot plug all holes either". This is because it doesn't address the "race" mentioned in the #3 in the comment in native_stop_other_cpus(): /* * 1) Send an IPI on the reboot vector to all other CPUs. * * The other CPUs should react on it after leaving critical * sections and re-enabling interrupts. They might still hold * locks, but there is nothing which can be done about that. * * 2) Wait for all other CPUs to report that they reached the * HLT loop in stop_this_cpu() * * 3) If #2 timed out send an NMI to the CPUs which did not * yet report * * 4) Wait for all other CPUs to report that they reached the * HLT loop in stop_this_cpu() * * #3 can obviously race against a CPU reaching the HLT loop late. * That CPU will have reported already and the "have all CPUs * reached HLT" condition will be true despite the fact that the * other CPU is still handling the NMI. Again, there is no * protection against that as "disabled" APICs still respond to * NMIs. */ Consider below case: CPU 0 CPU 1 native_stop_other_cpus() stop_this_cpu() // sends REBOOT IPI to stop remote CPUs ... wbinvd(); // wait timesout, try NMI if (!cpumask_empty(&cpus_stop_mask)) { for_each_cpu(cpu, &cpus_stop_mask) { ... cpumask_clear_cpu(cpu, &cpus_stop_mask); hlt; // send NMI ---> wakeup from hlt and run stop_this_cpu(): // WAIT CPUs TO STOP while (!cpumask_empty( &cpus_stop_mask) && ...) } ... proceed ... wbinvd(); ... hlt; The "WAIT CPUs TO STOP" is supposed to wait until all remote CPUs are stopped, but actually it quits immediately because the remote CPUs have been cleared in cpus_stop_mask when stop_this_cpu() is called from the REBOOT IPI. Doing WBINVD in stop_this_cpu() could potentially increase the chance to trigger the above "race" despite it's still rare to happen. Signed-off-by: Kai Huang <kai.huang@intel.com>

yamahata and others added 30 commits June 27, 2022 07:49

[MARKER] The start of TDX KVM patch series: KVM MMU GPA shared bits

214d885

This empty commit is to mark the start of patch series of KVM MMU GPA shared bits. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

[MARKER] The start of TDX KVM patch series: KVM TDP refactoring for TDX

a08806c

This empty commit is to mark the start of patch series of KVM TDP refactoring for TDX. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

[MARKER] The start of TDX KVM patch series: KVM TDP MMU hooks

082bd1d

This empty commit is to mark the start of patch series of KVM TDP MMU hooks. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

KVM: x86/mmu: Focibly use TDP MMU for TDX

c198d2b

In this patch series, TDX supports only TDP MMU and doesn't support legacy MMU. Forcibly use TDP MMU for TDX irrelevant of kernel parameter to disable TDP MMU. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

[MARKER] The start of TDX KVM patch series: TDX EPT violation

9d0cbb0

This empty commit is to mark the start of patch series of TDX EPT violation. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

[MARKER] The start of TDX KVM patch series: KVM TDP MMU MapGPA

874326b

This empty commit is to mark the start of patch series of KVM TDP MMU MapGPA. Signed-off-by: Isaku Yamahata <isaku.yamahata@intel.com>

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Update KVM-TDX.README.md with BKC reference wiki link #2

Update KVM-TDX.README.md with BKC reference wiki link #2

Uh oh!

hongyuni commented Jun 29, 2022

Uh oh!

Uh oh!

Update KVM-TDX.README.md with BKC reference wiki link #2

Update KVM-TDX.README.md with BKC reference wiki link #2

Uh oh!

Conversation

hongyuni commented Jun 29, 2022

Uh oh!

Uh oh!