- 14 Jul, 2016 19 commits
-
-
Radim Krčmář authored
kzalloc was replaced with kvm_kvzalloc to allow non-contiguous areas and rcu had to be modified to cope with it. The practical limit for KVM_MAX_VCPU_ID right now is INT_MAX, but lower value was chosen in case there were bugs. 1023 is sufficient maximum APIC ID for 288 VCPUs. Signed-off-by:
Radim Krčmář <rkrcmar@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Radim Krčmář authored
288 is in high demand because of Knights Landing CPU. We cannot set the limit to 640k, because that would be wasting space. Reviewed-by:
Paolo Bonzini <pbonzini@redhat.com> Signed-off-by:
Radim Krčmář <rkrcmar@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Radim Krčmář authored
Add KVM_X2APIC_API_DISABLE_BROADCAST_QUIRK as a feature flag to KVM_CAP_X2APIC_API. The quirk made KVM interpret 0xff as a broadcast even in x2APIC mode. The enableable capability is needed in order to support standard x2APIC and remain backward compatible. Signed-off-by:
Radim Krčmář <rkrcmar@redhat.com> [Expand kvm_apic_mda comment. - Paolo] Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Radim Krčmář authored
KVM_CAP_X2APIC_API is a capability for features related to x2APIC enablement. KVM_X2APIC_API_32BIT_FORMAT feature can be enabled to extend APIC ID in get/set ioctl and MSI addresses to 32 bits. Both are needed to support x2APIC. The feature has to be enableable and disabled by default, because get/set ioctl shifted and truncated APIC ID to 8 bits by using a non-standard protocol inspired by xAPIC and the change is not backward-compatible. Changes to MSI addresses follow the format used by interrupt remapping unit. The upper address word, that used to be 0, contains upper 24 bits of the LAPIC address in its upper 24 bits. Lower 8 bits are reserved as 0. Using the upper address word is not backward-compatible either as we didn't check that userspace zeroed the word. Reserved bits are still not explicitly checked, but non-zero data will affect LAPIC addresses, which will cause a bug. Signed-off-by:
Radim Krčmář <rkrcmar@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Radim Krčmář authored
Arch-specific code will use it. Signed-off-by:
Radim Krčmář <rkrcmar@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Radim Krčmář authored
LAPIC is reset in xAPIC mode and the surrounding code expects that. KVM never resets after initialization. This patch is just for sanity. Reviewed-by:
Paolo Bonzini <pbonzini@redhat.com> Signed-off-by:
Radim Krčmář <rkrcmar@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Radim Krčmář authored
The register is in hardware-compatible format now, so there is not need to intercept. Reviewed-by:
Paolo Bonzini <pbonzini@redhat.com> Signed-off-by:
Radim Krčmář <rkrcmar@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Radim Krčmář authored
APIC ID should be set to the initial APIC ID when enabling LAPIC. This only matters if the guest changes APIC ID. No sane OS does that. Signed-off-by:
Radim Krčmář <rkrcmar@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Radim Krčmář authored
We currently always shift APIC ID as if APIC was in xAPIC mode. x2APIC mode wants to use more bits and storing a hardware-compabible value is the the sanest option. KVM API to set the lapic expects that bottom 8 bits of APIC ID are in top 8 bits of APIC_ID register, so the register needs to be shifted in x2APIC mode. Signed-off-by:
Radim Krčmář <rkrcmar@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Radim Krčmář authored
Reviewed-by:
Paolo Bonzini <pbonzini@redhat.com> Signed-off-by:
Radim Krčmář <rkrcmar@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Radim Krčmář authored
x2APIC supports up to 2^32-1 LAPICs, but most guest in coming years will probably has fewer VCPUs. Dynamic size saves memory at the cost of turning one constant into a variable. apic_map mutex had to be moved before allocation to avoid races with cpu hotplug. Signed-off-by:
Radim Krčmář <rkrcmar@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Radim Krčmář authored
Logical x2APIC IDs map injectively to physical x2APIC IDs, so we can reuse the physical array for them. This allows us to save space by sizing the logical maps according to the needs of xAPIC. Signed-off-by:
Radim Krčmář <rkrcmar@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Radim Krčmář authored
kvm_irq_delivery_to_apic_fast and kvm_intr_is_single_vcpu_fast both compute the interrupt destination. Factor the code. 'struct kvm_lapic **dst = NULL' had to be added to silence GCC. GCC might complain about potential NULL access in the future, because it missed conditions that avoided uninitialized uses of dst. Signed-off-by:
Radim Krčmář <rkrcmar@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Radim Krčmář authored
240 has been well tested by Red Hat. Reviewed-by:
Paolo Bonzini <pbonzini@redhat.com> Signed-off-by:
Radim Krčmář <rkrcmar@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Bandan Das authored
MMU now knows about execute only mappings, so advertise the feature to L1 hypervisors Signed-off-by:
Bandan Das <bsd@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Bandan Das authored
To support execute only mappings on behalf of L1 hypervisors, reuse ACC_USER_MASK to signify if the L1 hypervisor has the R bit set. For the nested EPT case, we assumed that the U bit was always set since there was no equivalent in EPT page tables. Strictly speaking, this was not necessary because handle_ept_violation never set PFERR_USER_MASK in the error code (uf=0 in the parlance of update_permission_bitmask). We now have to set both U and UF correctly, respectively in FNAME(gpte_access) and in handle_ept_violation. Also in handle_ept_violation bit 3 of the exit qualification is not enough to detect a present PTE; all three bits 3-5 have to be checked. Signed-off-by:
Bandan Das <bsd@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Bandan Das authored
To support execute only mappings on behalf of L1 hypervisors, we need to teach set_spte() to honor all three of L1's XWR bits. As a start, add a new variable "shadow_present_mask" that will be set for non-EPT shadow paging and clear for EPT. Signed-off-by:
Bandan Das <bsd@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Bandan Das authored
We have two versions of the above function. To prevent confusion and bugs in the future, remove the non-FNAME version entirely and replace all calls with the actual check. Signed-off-by:
Bandan Das <bsd@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Bandan Das authored
This is safe because this function is called on host controlled page table and non-present/non-MMIO sptes never use bits 1..31. For the EPT case, this ensures that cases where only the execute bit is set is marked valid. Signed-off-by:
Bandan Das <bsd@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
- 11 Jul, 2016 5 commits
-
-
Paolo Bonzini authored
Merge branch 'kvm-ppc-next' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc into HEAD
-
Paolo Bonzini authored
There is no reason to read the entry/exit control fields of the VMCS and immediately write back the same value. Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Paolo Bonzini authored
Because the vmcs12 preemption timer is emulated through a separate hrtimer, we can keep on using the preemption timer in the vmcs02 to emulare L1's TSC deadline timer. However, the corresponding bit in the pin-based execution control field must be kept consistent between vmcs01 and vmcs02. On vmentry we copy it into the vmcs02; on vmexit the preemption timer must be disabled in the vmcs01 if a preemption timer vmexit happened while in guest mode. The preemption timer value in the vmcs02 is set by vmx_vcpu_run, so it need not be considered in prepare_vmcs02. Cc: Yunhong Jiang <yunhong.jiang@intel.com> Cc: Haozhong Zhang <haozhong.zhang@intel.com> Tested-by:
Wanpeng Li <kernellwp@gmail.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Wanpeng Li authored
The preemption timer for nested VMX is emulated by hrtimer which is started on L2 entry, stopped on L2 exit and evaluated via the check_nested_events hook. However, nested_vmx_exit_handled is always returning true for preemption timer vmexit. Then, the L1 preemption timer vmexit is captured and be treated as a L2 preemption timer vmexit, causing NULL pointer dereferences or worse in the L1 guest's vmexit handler: BUG: unable to handle kernel NULL pointer dereference at (null) IP: [< (null)>] (null) PGD 0 Oops: 0010 [#1] SMP Call Trace: ? kvm_lapic_expired_hv_timer+0x47/0x90 [kvm] handle_preemption_timer+0xe/0x20 [kvm_intel] vmx_handle_exit+0x169/0x15a0 [kvm_intel] ? kvm_arch_vcpu_ioctl_run+0xd5d/0x19d0 [kvm] kvm_arch_vcpu_ioctl_run+0xdee/0x19d0 [kvm] ? kvm_arch_vcpu_ioctl_run+0xd5d/0x19d0 [kvm] ? vcpu_load+0x1c/0x60 [kvm] ? kvm_arch_vcpu_load+0x57/0x260 [kvm] kvm_vcpu_ioctl+0x2d3/0x7c0 [kvm] do_vfs_ioctl+0x96/0x6a0 ? __fget_light+0x2a/0x90 SyS_ioctl+0x79/0x90 do_syscall_64+0x68/0x180 entry_SYSCALL64_slow_path+0x25/0x25 Code: Bad RIP value. RIP [< (null)>] (null) RSP <ffff8800b5263c48> CR2: 0000000000000000 ---[ end trace 9c70c48b1a2bc66e ]--- This can be reproduced readily by preemption timer enabled on L0 and disabled on L1. Return false since preemption timer vmexits must never be reflected to L2. Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim Krčmář <rkrcmar@redhat.com> Cc: Yunhong Jiang <yunhong.jiang@intel.com> Cc: Jan Kiszka <jan.kiszka@siemens.com> Cc: Haozhong Zhang <haozhong.zhang@intel.com> Signed-off-by:
Wanpeng Li <wanpeng.li@hotmail.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Paolo Bonzini authored
Simplify cpu_has_vmx_preemption_timer. This is consistent with the rest of setup_vmcs_config and preparatory for the next patch. Tested-by:
Wanpeng Li <kernellwp@gmail.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
- 05 Jul, 2016 16 commits
-
-
James Hogan authored
Default the guest PRId register to represent a generic QEMU machine instead of a 24kc on MIPSr6. 24kc isn't supported by r6 Linux kernels. Signed-off-by:
James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim KrÄmáÅ
™ <rkrcmar@redhat.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Signed-off-by:Paolo Bonzini <pbonzini@redhat.com>
-
James Hogan authored
When KVM emulates the RDHWR instruction, decode the instruction more strictly. The rs field (bits 25:21) should be zero, as should bits 10:9. Bits 8:6 is the register select field in MIPSr6, so we aren't strict about those bits (no other operations should use that encoding space). Signed-off-by:
James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim KrÄmáÅ
™ <rkrcmar@redhat.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Signed-off-by:Paolo Bonzini <pbonzini@redhat.com>
-
James Hogan authored
Recognise the new MIPSr6 CACHE instruction encoding rather than the pre-r6 one when an r6 kernel is being built. A SPECIAL3 opcode is used and the immediate field is reduced to 9 bits wide since MIPSr6. Signed-off-by:
James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim KrÄmáÅ
™ <rkrcmar@redhat.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Signed-off-by:Paolo Bonzini <pbonzini@redhat.com>
-
James Hogan authored
Add support in KVM for emulation of instructions in the forbidden slot of MIPSr6 compact branches. If we hit an exception on the forbidden slot, then the branch must not have been taken, which makes calculation of the resume PC trivial. Signed-off-by:
James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim KrÄmáÅ
™ <rkrcmar@redhat.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Signed-off-by:Paolo Bonzini <pbonzini@redhat.com>
-
James Hogan authored
MIPSr6 doesn't have lo/hi registers, so don't bother saving or restoring them, and don't expose them to userland with the KVM ioctl interface either. In fact the lo/hi registers aren't callee saved in the MIPS ABIs anyway, so there is no need to preserve the host lo/hi values at all when transitioning to and from the guest (which happens via a function call). Signed-off-by:
James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim KrÄmáÅ
™ <rkrcmar@redhat.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Signed-off-by:Paolo Bonzini <pbonzini@redhat.com>
-
James Hogan authored
The atomic KVM register access macros in kvm_host.h (for the guest Cause register with KVM in trap & emulate mode) use ll/sc instructions, however they still .set mips3, which causes pre-MIPSr6 instruction encodings to be emitted, even for a MIPSr6 build. Fix it to use MIPS_ISA_ARCH_LEVEL as other parts of arch/mips already do. Signed-off-by:
James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim KrÄmáÅ
™ <rkrcmar@redhat.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Signed-off-by:Paolo Bonzini <pbonzini@redhat.com>
-
James Hogan authored
__kvm_save_fpu and __kvm_restore_fpu use .set mips64r2 so that they can access the odd FPU registers as well as the even, however this causes misassembly of the return instruction on MIPSr6. Fix by replacing .set mips64r2 with .set fp=64, which doesn't change the architecture revision. Signed-off-by:
James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim KrÄmáÅ
™ <rkrcmar@redhat.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Signed-off-by:Paolo Bonzini <pbonzini@redhat.com>
-
Paul Burton authored
The opcodes currently defined in inst.h as cbcond0_op & cbcond1_op are actually defined in the MIPS base instruction set manuals as pop10 & pop30 respectively. Rename them as such, for consistency with the documentation. Signed-off-by:
Paul Burton <paul.burton@imgtec.com> Signed-off-by:
James Hogan <james.hogan@imgtec.com> Acked-by:
Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Paul Burton authored
The opcodes currently defined in inst.h as beqzcjic_op & bnezcjialc_op are actually defined in the MIPS base instruction set manuals as pop66 & pop76 respectively. Rename them as such, for consistency with the documentation. Signed-off-by:
Paul Burton <paul.burton@imgtec.com> Signed-off-by:
James Hogan <james.hogan@imgtec.com> Acked-by:
Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
James Hogan authored
Currently on a guest exception the guest's k0 register is saved to the scratch temp register and the guest k1 saved to the exception base address + 0x3000 using k0 to extract the Exception Base field of the EBase register and as the base operand to the store. Both are then copied into the VCPU structure after the other general purpose registers have been saved there. This bouncing to exception base + 0x3000 is not actually necessary as the VCPU pointer can be determined and written through just as easily with only a single spare register. The VCPU pointer is already needed in k1 for saving the other GP registers, so lets save the guest k0 register straight into the VCPU structure through k1, first saving k1 into the scratch temp register instead of k0. This could potentially pave the way for having a single exception base area for use by all guests. The ehb after saving the k register to the scratch temp register is also delayed until just before it needs to be read back. Signed-off-by:
James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim KrÄmáÅ
™ <rkrcmar@redhat.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Signed-off-by:Paolo Bonzini <pbonzini@redhat.com>
-
James Hogan authored
Use a relative branch to get from the individual exception vectors to the common guest exit handler, rather than loading the address of the exit handler and jumping to it. This is made easier due to the fact we are now generating the entry code dynamically. This will also allow the exception code to be further reduced in future patches. Signed-off-by:
James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim KrÄmáÅ
™ <rkrcmar@redhat.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Signed-off-by:Paolo Bonzini <pbonzini@redhat.com>
-
James Hogan authored
Scratch cop0 registers are needed by KVM to be able to save/restore all the GPRs, including k0/k1, and for storing the VCPU pointer. However no registers are universally suitable for these purposes, so the decision should be made at runtime. Until now, we've used DDATA_LO to store the VCPU pointer, and ErrorEPC as a temporary. It could be argued that this is abuse of those registers, and DDATA_LO is known not to be usable on certain implementations (Cavium Octeon). If KScratch registers are present, use them instead. We save & restore the temporary register in addition to the VCPU pointer register when using a KScratch register for it, as it may be used for normal host TLB handling too. Signed-off-by:
James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim KrÄmáÅ
™ <rkrcmar@redhat.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Signed-off-by:Paolo Bonzini <pbonzini@redhat.com>
-
James Hogan authored
On return from the exit handler to the host (without re-entering the guest) we restore the saved value of the DDATA_LO register which we use as a scratch register. However we've already restored it ready for calling the exit handler so there is no need to do it again, so drop that code. Signed-off-by:
James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim KrÄmáÅ
™ <rkrcmar@redhat.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Signed-off-by:Paolo Bonzini <pbonzini@redhat.com>
-
James Hogan authored
Check for presence of MSA at uasm assembly time rather than at runtime in the generated KVM host entry code. This optimises the guest exit path by eliminating the MSA code entirely if not present, and eliminating the read of Config3.MSAP and conditional branch if MSA is present. Signed-off-by:
James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim KrÄmáÅ
™ <rkrcmar@redhat.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Signed-off-by:Paolo Bonzini <pbonzini@redhat.com>
-
James Hogan authored
The FPU handling code on entry from guest is unnecessary if no FPU is present, so allow it to be dropped at uasm assembly time. Signed-off-by:
James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim KrÄmáÅ
™ <rkrcmar@redhat.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Signed-off-by:Paolo Bonzini <pbonzini@redhat.com>
-
James Hogan authored
Now that locore.S is converted to uasm, remove a bunch of the assembly offset definitions created by asm-offsets.c, including the CPUINFO_ ones for reading the variable asid mask, and the non FPU/MSA related VCPU_ definitions. KVM's fpu.S and msa.S still use the remaining definitions. Signed-off-by:
James Hogan <james.hogan@imgtec.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Radim KrÄmáÅ
™ <rkrcmar@redhat.com> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: linux-mips@linux-mips.org Cc: kvm@vger.kernel.org Signed-off-by:Paolo Bonzini <pbonzini@redhat.com>
-