KVM: PPC: Book3S HV: Change dec_expires to be relative to guest timebase

Change dec_expires to be relative to the guest timebase, and allow
it to be moved into low level P9 guest entry functions, to improve
SPR access scheduling.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211123095231.1036501-23-npiggin@gmail.com

arch/powerpc/include/asm/kvm_book3s.h

@@ -406,6 +406,12 @@ static inline ulong kvmppc_get_fault_dar(struct kvm_vcpu *vcpu)
 	return vcpu->arch.fault_dar;
 }
 
+/* Expiry time of vcpu DEC relative to host TB */
+static inline u64 kvmppc_dec_expires_host_tb(struct kvm_vcpu *vcpu)
+{
+	return vcpu->arch.dec_expires - vcpu->arch.vcore->tb_offset;
+}
+
 static inline bool is_kvmppc_resume_guest(int r)
 {
 	return (r == RESUME_GUEST || r == RESUME_GUEST_NV);

arch/powerpc/include/asm/kvm_host.h

@@ -741,7 +741,7 @@ struct kvm_vcpu_arch {
 
 	struct hrtimer dec_timer;
 	u64 dec_jiffies;
-	u64 dec_expires;
+	u64 dec_expires;	/* Relative to guest timebase. */
 	unsigned long pending_exceptions;
 	u8 ceded;
 	u8 prodded;

arch/powerpc/kvm/book3s_hv.c

@@ -2261,8 +2261,7 @@ static int kvmppc_get_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
 		*val = get_reg_val(id, vcpu->arch.vcore->arch_compat);
 		break;
 	case KVM_REG_PPC_DEC_EXPIRY:
-		*val = get_reg_val(id, vcpu->arch.dec_expires +
-				   vcpu->arch.vcore->tb_offset);
+		*val = get_reg_val(id, vcpu->arch.dec_expires);
 		break;
 	case KVM_REG_PPC_ONLINE:
 		*val = get_reg_val(id, vcpu->arch.online);
@@ -2514,8 +2513,7 @@ static int kvmppc_set_one_reg_hv(struct kvm_vcpu *vcpu, u64 id,
 		r = kvmppc_set_arch_compat(vcpu, set_reg_val(id, *val));
 		break;
 	case KVM_REG_PPC_DEC_EXPIRY:
-		vcpu->arch.dec_expires = set_reg_val(id, *val) -
-			vcpu->arch.vcore->tb_offset;
+		vcpu->arch.dec_expires = set_reg_val(id, *val);
 		break;
 	case KVM_REG_PPC_ONLINE:
 		i = set_reg_val(id, *val);
@@ -2902,13 +2900,13 @@ static void kvmppc_set_timer(struct kvm_vcpu *vcpu)
 	unsigned long dec_nsec, now;
 
 	now = get_tb();
-	if (now > vcpu->arch.dec_expires) {
+	if (now > kvmppc_dec_expires_host_tb(vcpu)) {
 		/* decrementer has already gone negative */
 		kvmppc_core_queue_dec(vcpu);
 		kvmppc_core_prepare_to_enter(vcpu);
 		return;
 	}
-	dec_nsec = tb_to_ns(vcpu->arch.dec_expires - now);
+	dec_nsec = tb_to_ns(kvmppc_dec_expires_host_tb(vcpu) - now);
 	hrtimer_start(&vcpu->arch.dec_timer, dec_nsec, HRTIMER_MODE_REL);
 	vcpu->arch.timer_running = 1;
 }
@@ -3380,7 +3378,7 @@ static void post_guest_process(struct kvmppc_vcore *vc, bool is_master)
 		 */
 		spin_unlock(&vc->lock);
 		/* cancel pending dec exception if dec is positive */
-		if (now < vcpu->arch.dec_expires &&
+		if (now < kvmppc_dec_expires_host_tb(vcpu) &&
 		    kvmppc_core_pending_dec(vcpu))
 			kvmppc_core_dequeue_dec(vcpu);
 
@@ -4224,20 +4222,6 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
 
 	load_spr_state(vcpu);
 
-	/*
-	 * When setting DEC, we must always deal with irq_work_raise via NMI vs
-	 * setting DEC. The problem occurs right as we switch into guest mode
-	 * if a NMI hits and sets pending work and sets DEC, then that will
-	 * apply to the guest and not bring us back to the host.
-	 *
-	 * irq_work_raise could check a flag (or possibly LPCR[HDICE] for
-	 * example) and set HDEC to 1? That wouldn't solve the nested hv
-	 * case which needs to abort the hcall or zero the time limit.
-	 *
-	 * XXX: Another day's problem.
-	 */
-	mtspr(SPRN_DEC, vcpu->arch.dec_expires - tb);
-
 	if (kvmhv_on_pseries()) {
 		/*
 		 * We need to save and restore the guest visible part of the
@@ -4263,6 +4247,23 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
 			hvregs.vcpu_token = vcpu->vcpu_id;
 		}
 		hvregs.hdec_expiry = time_limit;
+
+		/*
+		 * When setting DEC, we must always deal with irq_work_raise
+		 * via NMI vs setting DEC. The problem occurs right as we
+		 * switch into guest mode if a NMI hits and sets pending work
+		 * and sets DEC, then that will apply to the guest and not
+		 * bring us back to the host.
+		 *
+		 * irq_work_raise could check a flag (or possibly LPCR[HDICE]
+		 * for example) and set HDEC to 1? That wouldn't solve the
+		 * nested hv case which needs to abort the hcall or zero the
+		 * time limit.
+		 *
+		 * XXX: Another day's problem.
+		 */
+		mtspr(SPRN_DEC, kvmppc_dec_expires_host_tb(vcpu) - tb);
+
 		mtspr(SPRN_DAR, vcpu->arch.shregs.dar);
 		mtspr(SPRN_DSISR, vcpu->arch.shregs.dsisr);
 		trap = plpar_hcall_norets(H_ENTER_NESTED, __pa(&hvregs),
@@ -4274,6 +4275,12 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
 		vcpu->arch.psscr = mfspr(SPRN_PSSCR_PR);
 		mtspr(SPRN_PSSCR_PR, host_psscr);
 
+		dec = mfspr(SPRN_DEC);
+		if (!(lpcr & LPCR_LD)) /* Sign extend if not using large decrementer */
+			dec = (s32) dec;
+		tb = mftb();
+		vcpu->arch.dec_expires = dec + (tb + vc->tb_offset);
+
 		/* H_CEDE has to be handled now, not later */
 		if (trap == BOOK3S_INTERRUPT_SYSCALL && !vcpu->arch.nested &&
 		    kvmppc_get_gpr(vcpu, 3) == H_CEDE) {
@@ -4281,6 +4288,7 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
 			kvmppc_set_gpr(vcpu, 3, 0);
 			trap = 0;
 		}
+
 	} else {
 		kvmppc_xive_push_vcpu(vcpu);
 		trap = kvmhv_vcpu_entry_p9(vcpu, time_limit, lpcr);
@@ -4312,12 +4320,6 @@ static int kvmhv_p9_guest_entry(struct kvm_vcpu *vcpu, u64 time_limit,
 			vcpu->arch.slb_max = 0;
 	}
 
-	dec = mfspr(SPRN_DEC);
-	if (!(lpcr & LPCR_LD)) /* Sign extend if not using large decrementer */
-		dec = (s32) dec;
-	tb = mftb();
-	vcpu->arch.dec_expires = dec + tb;
-
 	store_spr_state(vcpu);
 
 	restore_p9_host_os_sprs(vcpu, &host_os_sprs);
@@ -4827,7 +4829,7 @@ int kvmhv_run_single_vcpu(struct kvm_vcpu *vcpu, u64 time_limit,
 	 * by L2 and the L1 decrementer is provided in hdec_expires
 	 */
 	if (kvmppc_core_pending_dec(vcpu) &&
-			((get_tb() < vcpu->arch.dec_expires) ||
+			((get_tb() < kvmppc_dec_expires_host_tb(vcpu)) ||
 			 (trap == BOOK3S_INTERRUPT_SYSCALL &&
 			  kvmppc_get_gpr(vcpu, 3) == H_ENTER_NESTED)))
 		kvmppc_core_dequeue_dec(vcpu);

arch/powerpc/kvm/book3s_hv_nested.c

@@ -358,6 +358,7 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
 	/* convert TB values/offsets to host (L0) values */
 	hdec_exp = l2_hv.hdec_expiry - vc->tb_offset;
 	vc->tb_offset += l2_hv.tb_offset;
+	vcpu->arch.dec_expires += l2_hv.tb_offset;
 
 	/* set L1 state to L2 state */
 	vcpu->arch.nested = l2;
@@ -399,6 +400,8 @@ long kvmhv_enter_nested_guest(struct kvm_vcpu *vcpu)
 	if (l2_regs.msr & MSR_TS_MASK)
 		vcpu->arch.shregs.msr |= MSR_TS_S;
 	vc->tb_offset = saved_l1_hv.tb_offset;
+	/* XXX: is this always the same delta as saved_l1_hv.tb_offset? */
+	vcpu->arch.dec_expires -= l2_hv.tb_offset;
 	restore_hv_regs(vcpu, &saved_l1_hv);
 	vcpu->arch.purr += delta_purr;
 	vcpu->arch.spurr += delta_spurr;

arch/powerpc/kvm/book3s_hv_p9_entry.c

@@ -188,7 +188,7 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
 	struct kvm *kvm = vcpu->kvm;
 	struct kvm_nested_guest *nested = vcpu->arch.nested;
 	struct kvmppc_vcore *vc = vcpu->arch.vcore;
-	s64 hdec;
+	s64 hdec, dec;
 	u64 tb, purr, spurr;
 	u64 *exsave;
 	bool ri_set;
@@ -317,6 +317,8 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
 	 */
 	mtspr(SPRN_HDEC, hdec);
 
+	mtspr(SPRN_DEC, vcpu->arch.dec_expires - tb);
+
 #ifdef CONFIG_PPC_TRANSACTIONAL_MEM
 tm_return_to_guest:
 #endif
@@ -461,6 +463,12 @@ int kvmhv_vcpu_entry_p9(struct kvm_vcpu *vcpu, u64 time_limit, unsigned long lpc
 	vcpu->arch.shregs.sprg2 = mfspr(SPRN_SPRG2);
 	vcpu->arch.shregs.sprg3 = mfspr(SPRN_SPRG3);
 
+	dec = mfspr(SPRN_DEC);
+	if (!(lpcr & LPCR_LD)) /* Sign extend if not using large decrementer */
+		dec = (s32) dec;
+	tb = mftb();
+	vcpu->arch.dec_expires = dec + tb;
+
 	/* Preserve PSSCR[FAKE_SUSPEND] until we've called kvmppc_save_tm_hv */
 	mtspr(SPRN_PSSCR, host_psscr |
 	      (local_paca->kvm_hstate.fake_suspend << PSSCR_FAKE_SUSPEND_LG));

arch/powerpc/kvm/book3s_hv_rmhandlers.S

@@ -814,10 +814,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S)
 	 * Set the decrementer to the guest decrementer.
 	 */
 	ld	r8,VCPU_DEC_EXPIRES(r4)
-	/* r8 is a host timebase value here, convert to guest TB */
-	ld	r5,HSTATE_KVM_VCORE(r13)
-	ld	r6,VCORE_TB_OFFSET_APPL(r5)
-	add	r8,r8,r6
 	mftb	r7
 	subf	r3,r7,r8
 	mtspr	SPRN_DEC,r3
@@ -1192,9 +1188,6 @@ guest_bypass:
 	mftb	r6
 	extsw	r5,r5
 16:	add	r5,r5,r6
-	/* r5 is a guest timebase value here, convert to host TB */
-	ld	r4,VCORE_TB_OFFSET_APPL(r3)
-	subf	r5,r4,r5
 	std	r5,VCPU_DEC_EXPIRES(r9)
 
 	/* Increment exit count, poke other threads to exit */
@@ -2160,9 +2153,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_TM)
 	/* save expiry time of guest decrementer */
 	add	r3, r3, r5
 	ld	r4, HSTATE_KVM_VCPU(r13)
-	ld	r5, HSTATE_KVM_VCORE(r13)
-	ld	r6, VCORE_TB_OFFSET_APPL(r5)
-	subf	r3, r6, r3	/* convert to host TB value */
 	std	r3, VCPU_DEC_EXPIRES(r4)
 
 #ifdef CONFIG_KVM_BOOK3S_HV_EXIT_TIMING
@@ -2259,9 +2249,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_TM)
 
 	/* Restore guest decrementer */
 	ld	r3, VCPU_DEC_EXPIRES(r4)
-	ld	r5, HSTATE_KVM_VCORE(r13)
-	ld	r6, VCORE_TB_OFFSET_APPL(r5)
-	add	r3, r3, r6	/* convert host TB to guest TB value */
 	mftb	r7
 	subf	r3, r7, r3
 	mtspr	SPRN_DEC, r3