KVM: PPC: Bookehv: Get vcpu's last instruction for emulation
On book3e, KVM uses load external pid (lwepx) dedicated instruction to read
guest last instruction on the exit path. lwepx exceptions (DTLB_MISS, DSI
and LRAT), generated by loading a guest address, needs to be handled by KVM.
These exceptions are generated in a substituted guest translation context
(EPLC[EGS] = 1) from host context (MSR[GS] = 0).
Currently, KVM hooks only interrupts generated from guest context (MSR[GS] = 1),
doing minimal checks on the fast path to avoid host performance degradation.
lwepx exceptions originate from host state (MSR[GS] = 0) which implies
additional checks in DO_KVM macro (beside the current MSR[GS] = 1) by looking
at the Exception Syndrome Register (ESR[EPID]) and the External PID Load Context
Register (EPLC[EGS]). Doing this on each Data TLB miss exception is obvious
too intrusive for the host.
Read guest last instruction from kvmppc_load_last_inst() by searching for the
physical address and kmap it. This address the TODO for TLB eviction and
execute-but-not-read entries, and allow us to get rid of lwepx until we are
able to handle failures.
A simple stress benchmark shows a 1% sys performance degradation compared with
previous approach (lwepx without failure handling):
time for i in `seq 1 10000`; do /bin/echo > /dev/null; done
real 0m 8.85s
user 0m 4.34s
sys 0m 4.48s
vs
real 0m 8.84s
user 0m 4.36s
sys 0m 4.44s
A solution to use lwepx and to handle its exceptions in KVM would be to temporary
highjack the interrupt vector from host. This imposes additional synchronizations
for cores like FSL e6500 that shares host IVOR registers between hardware threads.
This optimized solution can be later developed on top of this patch.
Signed-off-by: Mihai Caraman <mihai.caraman@freescale.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
diff --git a/arch/powerpc/kvm/e500_mmu_host.c b/arch/powerpc/kvm/e500_mmu_host.c
index 4385c14..4150826 100644
--- a/arch/powerpc/kvm/e500_mmu_host.c
+++ b/arch/powerpc/kvm/e500_mmu_host.c
@@ -610,11 +610,103 @@
}
}
+#ifdef CONFIG_KVM_BOOKE_HV
+int kvmppc_load_last_inst(struct kvm_vcpu *vcpu, enum instruction_type type,
+ u32 *instr)
+{
+ gva_t geaddr;
+ hpa_t addr;
+ hfn_t pfn;
+ hva_t eaddr;
+ u32 mas1, mas2, mas3;
+ u64 mas7_mas3;
+ struct page *page;
+ unsigned int addr_space, psize_shift;
+ bool pr;
+ unsigned long flags;
+
+ /* Search TLB for guest pc to get the real address */
+ geaddr = kvmppc_get_pc(vcpu);
+
+ addr_space = (vcpu->arch.shared->msr & MSR_IS) >> MSR_IR_LG;
+
+ local_irq_save(flags);
+ mtspr(SPRN_MAS6, (vcpu->arch.pid << MAS6_SPID_SHIFT) | addr_space);
+ mtspr(SPRN_MAS5, MAS5_SGS | vcpu->kvm->arch.lpid);
+ asm volatile("tlbsx 0, %[geaddr]\n" : :
+ [geaddr] "r" (geaddr));
+ mtspr(SPRN_MAS5, 0);
+ mtspr(SPRN_MAS8, 0);
+ mas1 = mfspr(SPRN_MAS1);
+ mas2 = mfspr(SPRN_MAS2);
+ mas3 = mfspr(SPRN_MAS3);
+#ifdef CONFIG_64BIT
+ mas7_mas3 = mfspr(SPRN_MAS7_MAS3);
+#else
+ mas7_mas3 = ((u64)mfspr(SPRN_MAS7) << 32) | mas3;
+#endif
+ local_irq_restore(flags);
+
+ /*
+ * If the TLB entry for guest pc was evicted, return to the guest.
+ * There are high chances to find a valid TLB entry next time.
+ */
+ if (!(mas1 & MAS1_VALID))
+ return EMULATE_AGAIN;
+
+ /*
+ * Another thread may rewrite the TLB entry in parallel, don't
+ * execute from the address if the execute permission is not set
+ */
+ pr = vcpu->arch.shared->msr & MSR_PR;
+ if (unlikely((pr && !(mas3 & MAS3_UX)) ||
+ (!pr && !(mas3 & MAS3_SX)))) {
+ pr_err_ratelimited(
+ "%s: Instuction emulation from guest addres %08lx without execute permission\n",
+ __func__, geaddr);
+ return EMULATE_AGAIN;
+ }
+
+ /*
+ * The real address will be mapped by a cacheable, memory coherent,
+ * write-back page. Check for mismatches when LRAT is used.
+ */
+ if (has_feature(vcpu, VCPU_FTR_MMU_V2) &&
+ unlikely((mas2 & MAS2_I) || (mas2 & MAS2_W) || !(mas2 & MAS2_M))) {
+ pr_err_ratelimited(
+ "%s: Instuction emulation from guest addres %08lx mismatches storage attributes\n",
+ __func__, geaddr);
+ return EMULATE_AGAIN;
+ }
+
+ /* Get pfn */
+ psize_shift = MAS1_GET_TSIZE(mas1) + 10;
+ addr = (mas7_mas3 & (~0ULL << psize_shift)) |
+ (geaddr & ((1ULL << psize_shift) - 1ULL));
+ pfn = addr >> PAGE_SHIFT;
+
+ /* Guard against emulation from devices area */
+ if (unlikely(!page_is_ram(pfn))) {
+ pr_err_ratelimited("%s: Instruction emulation from non-RAM host addres %08llx is not supported\n",
+ __func__, addr);
+ return EMULATE_AGAIN;
+ }
+
+ /* Map a page and get guest's instruction */
+ page = pfn_to_page(pfn);
+ eaddr = (unsigned long)kmap_atomic(page);
+ *instr = *(u32 *)(eaddr | (unsigned long)(addr & ~PAGE_MASK));
+ kunmap_atomic((u32 *)eaddr);
+
+ return EMULATE_DONE;
+}
+#else
int kvmppc_load_last_inst(struct kvm_vcpu *vcpu, enum instruction_type type,
u32 *instr)
{
return EMULATE_AGAIN;
}
+#endif
/************* MMU Notifiers *************/