kernel.git/arch/x86/kernel/fpu, branch linux-5.18.y

x86/fpu: KVM: Set the base guest FPU uABI size to sizeof(struct kvm_xsave)

2022-06-06T06:48:55Z

commit d187ba5312307d51818beafaad87d28a7d939adf upstream. Set the starting uABI size of KVM's guest FPU to 'struct kvm_xsave', i.e. to KVM's historical uABI size. When saving FPU state for usersapce, KVM (well, now the FPU) sets the FP+SSE bits in the XSAVE header even if the host doesn't support XSAVE. Setting the XSAVE header allows the VM to be migrated to a host that does support XSAVE without the new host having to handle FPU state that may or may not be compatible with XSAVE. Setting the uABI size to the host's default size results in out-of-bounds writes (setting the FP+SSE bits) and data corruption (that is thankfully caught by KASAN) when running on hosts without XSAVE, e.g. on Core2 CPUs. WARN if the default size is larger than KVM's historical uABI size; all features that can push the FPU size beyond the historical size must be opt-in. ================================================================== BUG: KASAN: slab-out-of-bounds in fpu_copy_uabi_to_guest_fpstate+0x86/0x130 Read of size 8 at addr ffff888011e33a00 by task qemu-build/681 CPU: 1 PID: 681 Comm: qemu-build Not tainted 5.18.0-rc5-KASAN-amd64 #1 Hardware name: /DG35EC, BIOS ECG3510M.86A.0118.2010.0113.1426 01/13/2010 Call Trace: dump_stack_lvl+0x34/0x45 print_report.cold+0x45/0x575 kasan_report+0x9b/0xd0 fpu_copy_uabi_to_guest_fpstate+0x86/0x130 kvm_arch_vcpu_ioctl+0x72a/0x1c50 [kvm] kvm_vcpu_ioctl+0x47f/0x7b0 [kvm] __x64_sys_ioctl+0x5de/0xc90 do_syscall_64+0x31/0x50 entry_SYSCALL_64_after_hwframe+0x44/0xae Allocated by task 0: (stack is not available) The buggy address belongs to the object at ffff888011e33800 which belongs to the cache kmalloc-512 of size 512 The buggy address is located 0 bytes to the right of 512-byte region [ffff888011e33800, ffff888011e33a00) The buggy address belongs to the physical page: page:0000000089cd4adb refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x11e30 head:0000000089cd4adb order:2 compound_mapcount:0 compound_pincount:0 flags: 0x4000000000010200(slab|head|zone=1) raw: 4000000000010200 dead000000000100 dead000000000122 ffff888001041c80 raw: 0000000000000000 0000000080100010 00000001ffffffff 0000000000000000 page dumped because: kasan: bad access detected Memory state around the buggy address: ffff888011e33900: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ffff888011e33980: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 >ffff888011e33a00: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc ^ ffff888011e33a80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc ffff888011e33b00: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc ================================================================== Disabling lock debugging due to kernel taint Fixes: be50b2065dfa ("kvm: x86: Add support for getting/setting expanded xstate buffer") Fixes: c60427dd50ba ("x86/fpu: Add uabi_size to guest_fpu") Reported-by: Zdenek Kaspar Cc: Maciej S. Szmigiero Cc: Paolo Bonzini Cc: kvm@vger.kernel.org Cc: stable@vger.kernel.org Signed-off-by: Sean Christopherson Tested-by: Zdenek Kaspar Message-Id: <20220504001219.983513-1-seanjc@google.com> Signed-off-by: Paolo Bonzini Signed-off-by: Greg Kroah-Hartman

x86/fpu: Prevent FPU state corruption

2022-05-05T00:40:19Z

The FPU usage related to task FPU management is either protected by disabling interrupts (switch_to, return to user) or via fpregs_lock() which is a wrapper around local_bh_disable(). When kernel code wants to use the FPU then it has to check whether it is possible by calling irq_fpu_usable(). But the condition in irq_fpu_usable() is wrong. It allows FPU to be used when: !in_interrupt() || interrupted_user_mode() || interrupted_kernel_fpu_idle() The latter is checking whether some other context already uses FPU in the kernel, but if that's not the case then it allows FPU to be used unconditionally even if the calling context interrupted a fpregs_lock() critical region. If that happens then the FPU state of the interrupted context becomes corrupted. Allow in kernel FPU usage only when no other context has in kernel FPU usage and either the calling context is not hard interrupt context or the hard interrupt did not interrupt a local bottomhalf disabled region. It's hard to find a proper Fixes tag as the condition was broken in one way or the other for a very long time and the eager/lazy FPU changes caused a lot of churn. Picked something remotely connected from the history. This survived undetected for quite some time as FPU usage in interrupt context is rare, but the recent changes to the random code unearthed it at least on a kernel which had FPU debugging enabled. There is probably a higher rate of silent corruption as not all issues can be detected by the FPU debugging code. This will be addressed in a subsequent change. Fixes: 5d2bd7009f30 ("x86, fpu: decouple non-lazy/eager fpu restore from xsave") Reported-by: Filipe Manana Signed-off-by: Thomas Gleixner Tested-by: Filipe Manana Reviewed-by: Borislav Petkov Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20220501193102.588689270@linutronix.de

x86/fpu/xstate: Consolidate size calculations

2022-03-30T10:04:09Z

Use the offset calculation to do the size calculation which avoids yet another series of CPUID instructions for each invocation. [ Fix the FP/SSE only case which missed to take the xstate header into account, as Reported-by: kernel test robot ] Signed-off-by: Thomas Gleixner Signed-off-by: Borislav Petkov Link: https://lore.kernel.org/r/87o81pgbp2.ffs@tglx

x86/fpu/xstate: Handle supervisor states in XSTATE permissions

2022-03-30T09:35:22Z

The size calculation in __xstate_request_perm() fails to take supervisor states into account because the permission bitmap is only relevant for user states. Up to 5.17 this does not matter because there are no supervisor states supported, but the (re-)enabling of ENQCMD makes them available. Fixes: 7c1ef59145f1 ("x86/cpufeatures: Re-enable ENQCMD") Signed-off-by: Thomas Gleixner Signed-off-by: Borislav Petkov Link: https://lore.kernel.org/r/20220324134623.681768598@linutronix.de

x86/fpu/xsave: Handle compacted offsets correctly with supervisor states

2022-03-30T09:20:36Z

So far the cached fixed compacted offsets worked, but with (re-)enabling of ENQCMD this does no longer work with KVM fpstate. KVM does not have supervisor features enabled for the guest FPU, which means that KVM has then a different XSAVE area layout than the host FPU state. This in turn breaks the copy from/to UABI functions when invoked for a guest state. Remove the pre-calculated compacted offsets and calculate the offset of each component at runtime based on the XCOMP_BV field in the XSAVE header. The runtime overhead is not interesting because these copy from/to UABI functions are not used in critical fast paths. KVM uses them to save and restore FPU state during migration. The host uses them for ptrace and for the slow path of 32bit signal handling. Fixes: 7c1ef59145f1 ("x86/cpufeatures: Re-enable ENQCMD") Signed-off-by: Thomas Gleixner Signed-off-by: Borislav Petkov Link: https://lore.kernel.org/r/20220324134623.627636809@linutronix.de

x86/fpu: Cache xfeature flags from CPUID

2022-03-30T09:05:32Z

In preparation for runtime calculation of XSAVE offsets cache the feature flags for each XSTATE component during feature enumeration via CPUID(0xD). EDX has two relevant bits: 0 Supervisor component 1 Feature storage must be 64 byte aligned These bits are currently only evaluated during init, but the alignment bit must be cached to make runtime calculation of XSAVE offsets efficient. Cache the full EDX content and use it for the existing alignment and supervisor checks. Signed-off-by: Thomas Gleixner Signed-off-by: Borislav Petkov Link: https://lore.kernel.org/r/20220324134623.573656209@linutronix.de

x86/fpu/xsave: Initialize offset/size cache early

2022-03-30T08:55:44Z

Reading XSTATE feature information from CPUID over and over does not make sense. The information has to be cached anyway, so it can be done early. Prepare for runtime calculation of XSTATE offsets and allow consolidation of the size calculation functions in a later step. Rename the function while at it as it does not setup any features. Signed-off-by: Thomas Gleixner Signed-off-by: Borislav Petkov Link: https://lore.kernel.org/r/20220324134623.519411939@linutronix.de

x86/fpu: Remove unused supervisor only offsets

2022-03-30T08:44:51Z

No users. Signed-off-by: Thomas Gleixner Signed-off-by: Borislav Petkov Link: https://lore.kernel.org/r/20220324134623.465066249@linutronix.de

x86/fpu: Remove redundant XCOMP_BV initialization

2022-03-29T18:57:18Z

fpu_copy_uabi_to_guest_fpstate() initializes the XCOMP_BV field in the XSAVE header. That's a leftover from the old KVM FPU buffer handling code. Since d69c1382e1b7 ("x86/kvm: Convert FPU handling to a single swap buffer") KVM uses the FPU core allocation code, which initializes the XCOMP_BV field already. Signed-off-by: Thomas Gleixner Signed-off-by: Borislav Petkov Link: https://lore.kernel.org/r/20220324134623.408932232@linutronix.de

x86/fpu/xstate: Fix the ARCH_REQ_XCOMP_PERM implementation

2022-03-23T20:28:34Z

ARCH_REQ_XCOMP_PERM is supposed to add the requested feature to the permission bitmap of thread_group_leader()->fpu. But the code overwrites the bitmap with the requested feature bit only rather than adding it. Fix the code to add the requested feature bit to the master bitmask. Fixes: db8268df0983 ("x86/arch_prctl: Add controls for dynamic XSTATE components") Signed-off-by: Yang Zhong Signed-off-by: Chang S. Bae Signed-off-by: Thomas Gleixner Cc: Paolo Bonzini Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20220129173647.27981-2-chang.seok.bae@intel.com