diff options
Diffstat (limited to 'arch/powerpc/kvm/book3s_xics.c')
| -rw-r--r-- | arch/powerpc/kvm/book3s_xics.c | 556 |
1 files changed, 382 insertions, 174 deletions
diff --git a/arch/powerpc/kvm/book3s_xics.c b/arch/powerpc/kvm/book3s_xics.c index 94c1dd46b83d..589a8f257120 100644 --- a/arch/powerpc/kvm/book3s_xics.c +++ b/arch/powerpc/kvm/book3s_xics.c @@ -1,10 +1,7 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * Copyright 2012 Michael Ellerman, IBM Corporation. * Copyright 2012 Benjamin Herrenschmidt, IBM Corporation. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License, version 2, as - * published by the Free Software Foundation. */ #include <linux/kernel.h> @@ -12,15 +9,16 @@ #include <linux/err.h> #include <linux/gfp.h> #include <linux/anon_inodes.h> +#include <linux/spinlock.h> +#include <linux/debugfs.h> +#include <linux/uaccess.h> -#include <asm/uaccess.h> #include <asm/kvm_book3s.h> #include <asm/kvm_ppc.h> #include <asm/hvcall.h> #include <asm/xics.h> -#include <asm/debug.h> +#include <asm/time.h> -#include <linux/debugfs.h> #include <linux/seq_file.h> #include "book3s_xics.h" @@ -38,8 +36,8 @@ * LOCKING * ======= * - * Each ICS has a mutex protecting the information about the IRQ - * sources and avoiding simultaneous deliveries if the same interrupt. + * Each ICS has a spin lock protecting the information about the IRQ + * sources and avoiding simultaneous deliveries of the same interrupt. * * ICP operations are done via a single compare & swap transaction * (most ICP state fits in the union kvmppc_icp_state) @@ -61,14 +59,19 @@ /* -- ICS routines -- */ static void icp_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, - u32 new_irq); + u32 new_irq, bool check_resend); -static int ics_deliver_irq(struct kvmppc_xics *xics, u32 irq, u32 level, - bool report_status) +/* + * Return value ideally indicates how the interrupt was handled, but no + * callers look at it (given that we don't implement KVM_IRQ_LINE_STATUS), + * so just return 0. + */ +static int ics_deliver_irq(struct kvmppc_xics *xics, u32 irq, u32 level) { struct ics_irq_state *state; struct kvmppc_ics *ics; u16 src; + u32 pq_old, pq_new; XICS_DBG("ics deliver %#x (level: %d)\n", irq, level); @@ -81,25 +84,42 @@ static int ics_deliver_irq(struct kvmppc_xics *xics, u32 irq, u32 level, if (!state->exists) return -EINVAL; - if (report_status) - return state->asserted; - + if (level == KVM_INTERRUPT_SET_LEVEL || level == KVM_INTERRUPT_SET) + level = 1; + else if (level == KVM_INTERRUPT_UNSET) + level = 0; /* - * We set state->asserted locklessly. This should be fine as - * we are the only setter, thus concurrent access is undefined - * to begin with. + * Take other values the same as 1, consistent with original code. + * maybe WARN here? */ - if (level == KVM_INTERRUPT_SET_LEVEL) - state->asserted = 1; - else if (level == KVM_INTERRUPT_UNSET) { - state->asserted = 0; + + if (!state->lsi && level == 0) /* noop for MSI */ return 0; - } - /* Attempt delivery */ - icp_deliver_irq(xics, NULL, irq); + do { + pq_old = state->pq_state; + if (state->lsi) { + if (level) { + if (pq_old & PQ_PRESENTED) + /* Setting already set LSI ... */ + return 0; + + pq_new = PQ_PRESENTED; + } else + pq_new = 0; + } else + pq_new = ((pq_old << 1) & 3) | PQ_PRESENTED; + } while (cmpxchg(&state->pq_state, pq_old, pq_new) != pq_old); + + /* Test P=1, Q=0, this is the only case where we present */ + if (pq_new == PQ_PRESENTED) + icp_deliver_irq(xics, NULL, irq, false); + + /* Record which CPU this arrived on for passed-through interrupts */ + if (state->host_irq) + state->intr_cpu = raw_smp_processor_id(); - return state->asserted; + return 0; } static void ics_check_resend(struct kvmppc_xics *xics, struct kvmppc_ics *ics, @@ -107,23 +127,14 @@ static void ics_check_resend(struct kvmppc_xics *xics, struct kvmppc_ics *ics, { int i; - mutex_lock(&ics->lock); - for (i = 0; i < KVMPPC_XICS_IRQ_PER_ICS; i++) { struct ics_irq_state *state = &ics->irq_state[i]; - - if (!state->resend) - continue; - - XICS_DBG("resend %#x prio %#x\n", state->number, - state->priority); - - mutex_unlock(&ics->lock); - icp_deliver_irq(xics, icp, state->number); - mutex_lock(&ics->lock); + if (state->resend) { + XICS_DBG("resend %#x prio %#x\n", state->number, + state->priority); + icp_deliver_irq(xics, icp, state->number, true); + } } - - mutex_unlock(&ics->lock); } static bool write_xive(struct kvmppc_xics *xics, struct kvmppc_ics *ics, @@ -131,8 +142,10 @@ static bool write_xive(struct kvmppc_xics *xics, struct kvmppc_ics *ics, u32 server, u32 priority, u32 saved_priority) { bool deliver; + unsigned long flags; - mutex_lock(&ics->lock); + local_irq_save(flags); + arch_spin_lock(&ics->lock); state->server = server; state->priority = priority; @@ -140,10 +153,12 @@ static bool write_xive(struct kvmppc_xics *xics, struct kvmppc_ics *ics, deliver = false; if ((state->masked_pending || state->resend) && priority != MASKED) { state->masked_pending = 0; + state->resend = 0; deliver = true; } - mutex_unlock(&ics->lock); + arch_spin_unlock(&ics->lock); + local_irq_restore(flags); return deliver; } @@ -173,7 +188,7 @@ int kvmppc_xics_set_xive(struct kvm *kvm, u32 irq, u32 server, u32 priority) state->masked_pending, state->resend); if (write_xive(xics, ics, state, server, priority, priority)) - icp_deliver_irq(xics, icp, irq); + icp_deliver_irq(xics, icp, irq, false); return 0; } @@ -184,6 +199,7 @@ int kvmppc_xics_get_xive(struct kvm *kvm, u32 irq, u32 *server, u32 *priority) struct kvmppc_ics *ics; struct ics_irq_state *state; u16 src; + unsigned long flags; if (!xics) return -ENODEV; @@ -193,10 +209,12 @@ int kvmppc_xics_get_xive(struct kvm *kvm, u32 irq, u32 *server, u32 *priority) return -EINVAL; state = &ics->irq_state[src]; - mutex_lock(&ics->lock); + local_irq_save(flags); + arch_spin_lock(&ics->lock); *server = state->server; *priority = state->priority; - mutex_unlock(&ics->lock); + arch_spin_unlock(&ics->lock); + local_irq_restore(flags); return 0; } @@ -223,7 +241,7 @@ int kvmppc_xics_int_on(struct kvm *kvm, u32 irq) if (write_xive(xics, ics, state, state->server, state->saved_priority, state->saved_priority)) - icp_deliver_irq(xics, icp, irq); + icp_deliver_irq(xics, icp, irq, false); return 0; } @@ -265,7 +283,7 @@ static inline bool icp_try_update(struct kvmppc_icp *icp, if (!success) goto bail; - XICS_DBG("UPD [%04x] - C:%02x M:%02x PP: %02x PI:%06x R:%d O:%d\n", + XICS_DBG("UPD [%04lx] - C:%02x M:%02x PP: %02x PI:%06x R:%d O:%d\n", icp->server_num, old.cppr, old.mfrr, old.pending_pri, old.xisr, old.need_resend, old.out_ee); @@ -289,7 +307,7 @@ static inline bool icp_try_update(struct kvmppc_icp *icp, */ if (new.out_ee) { kvmppc_book3s_queue_irqprio(icp->vcpu, - BOOK3S_INTERRUPT_EXTERNAL_LEVEL); + BOOK3S_INTERRUPT_EXTERNAL); if (!change_self) kvmppc_fast_vcpu_kick(icp->vcpu); } @@ -321,11 +339,11 @@ static bool icp_try_to_deliver(struct kvmppc_icp *icp, u32 irq, u8 priority, union kvmppc_icp_state old_state, new_state; bool success; - XICS_DBG("try deliver %#x(P:%#x) to server %#x\n", irq, priority, + XICS_DBG("try deliver %#x(P:%#x) to server %#lx\n", irq, priority, icp->server_num); do { - old_state = new_state = ACCESS_ONCE(icp->state); + old_state = new_state = READ_ONCE(icp->state); *reject = 0; @@ -357,12 +375,13 @@ static bool icp_try_to_deliver(struct kvmppc_icp *icp, u32 irq, u8 priority, } static void icp_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, - u32 new_irq) + u32 new_irq, bool check_resend) { struct ics_irq_state *state; struct kvmppc_ics *ics; u32 reject; u16 src; + unsigned long flags; /* * This is used both for initial delivery of an interrupt and @@ -389,7 +408,8 @@ static void icp_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, state = &ics->irq_state[src]; /* Get a lock on the ICS */ - mutex_lock(&ics->lock); + local_irq_save(flags); + arch_spin_lock(&ics->lock); /* Get our server */ if (!icp || state->server != icp->server_num) { @@ -401,6 +421,10 @@ static void icp_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, } } + if (check_resend) + if (!state->resend) + goto out; + /* Clear the resend bit of that interrupt */ state->resend = 0; @@ -415,7 +439,7 @@ static void icp_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, * the whole masked_pending business which is about not * losing interrupts that occur while masked. * - * I don't differenciate normal deliveries and resends, this + * I don't differentiate normal deliveries and resends, this * implementation will differ from PAPR and not lose such * interrupts. */ @@ -432,13 +456,13 @@ static void icp_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, * * Note that if successful, the new delivery might have itself * rejected an interrupt that was "delivered" before we took the - * icp mutex. + * ics spin lock. * * In this case we do the whole sequence all over again for the * new guy. We cannot assume that the rejected interrupt is less * favored than the new one, and thus doesn't need to be delivered, * because by the time we exit icp_try_to_deliver() the target - * processor may well have alrady consumed & completed it, and thus + * processor may well have already consumed & completed it, and thus * the rejected interrupt might actually be already acceptable. */ if (icp_try_to_deliver(icp, new_irq, state->priority, &reject)) { @@ -446,8 +470,10 @@ static void icp_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, * Delivery was successful, did we reject somebody else ? */ if (reject && reject != XICS_IPI) { - mutex_unlock(&ics->lock); + arch_spin_unlock(&ics->lock); + local_irq_restore(flags); new_irq = reject; + check_resend = false; goto again; } } else { @@ -455,10 +481,16 @@ static void icp_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, * We failed to deliver the interrupt we need to set the * resend map bit and mark the ICS state as needing a resend */ - set_bit(ics->icsid, icp->resend_map); state->resend = 1; /* + * Make sure when checking resend, we don't miss the resend + * if resend_map bit is seen and cleared. + */ + smp_wmb(); + set_bit(ics->icsid, icp->resend_map); + + /* * If the need_resend flag got cleared in the ICP some time * between icp_try_to_deliver() atomic update and now, then * we know it might have missed the resend_map bit. So we @@ -466,12 +498,16 @@ static void icp_deliver_irq(struct kvmppc_xics *xics, struct kvmppc_icp *icp, */ smp_mb(); if (!icp->state.need_resend) { - mutex_unlock(&ics->lock); + state->resend = 0; + arch_spin_unlock(&ics->lock); + local_irq_restore(flags); + check_resend = false; goto again; } } out: - mutex_unlock(&ics->lock); + arch_spin_unlock(&ics->lock); + local_irq_restore(flags); } static void icp_down_cppr(struct kvmppc_xics *xics, struct kvmppc_icp *icp, @@ -510,7 +546,7 @@ static void icp_down_cppr(struct kvmppc_xics *xics, struct kvmppc_icp *icp, * in virtual mode. */ do { - old_state = new_state = ACCESS_ONCE(icp->state); + old_state = new_state = READ_ONCE(icp->state); /* Down_CPPR */ new_state.cppr = new_cppr; @@ -554,8 +590,7 @@ static noinline unsigned long kvmppc_h_xirr(struct kvm_vcpu *vcpu) u32 xirr; /* First, remove EE from the processor */ - kvmppc_book3s_dequeue_irqprio(icp->vcpu, - BOOK3S_INTERRUPT_EXTERNAL_LEVEL); + kvmppc_book3s_dequeue_irqprio(icp->vcpu, BOOK3S_INTERRUPT_EXTERNAL); /* * ICP State: Accept_Interrupt @@ -565,7 +600,7 @@ static noinline unsigned long kvmppc_h_xirr(struct kvm_vcpu *vcpu) * pending priority */ do { - old_state = new_state = ACCESS_ONCE(icp->state); + old_state = new_state = READ_ONCE(icp->state); xirr = old_state.xisr | (((u32)old_state.cppr) << 24); if (!old_state.xisr) @@ -611,13 +646,28 @@ static noinline int kvmppc_h_ipi(struct kvm_vcpu *vcpu, unsigned long server, * there might be a previously-rejected interrupt needing * to be resent. * + * ICP state: Check_IPI + * * If the CPPR is less favored, then we might be replacing - * an interrupt, and thus need to possibly reject it as in + * an interrupt, and thus need to possibly reject it. * - * ICP state: Check_IPI + * ICP State: IPI + * + * Besides rejecting any pending interrupts, we also + * update XISR and pending_pri to mark IPI as pending. + * + * PAPR does not describe this state, but if the MFRR is being + * made less favored than its earlier value, there might be + * a previously-rejected interrupt needing to be resent. + * Ideally, we would want to resend only if + * prio(pending_interrupt) < mfrr && + * prio(pending_interrupt) < cppr + * where pending interrupt is the one that was rejected. But + * we don't have that state, so we simply trigger a resend + * whenever the MFRR is made less favored. */ do { - old_state = new_state = ACCESS_ONCE(icp->state); + old_state = new_state = READ_ONCE(icp->state); /* Set_MFRR */ new_state.mfrr = mfrr; @@ -627,13 +677,14 @@ static noinline int kvmppc_h_ipi(struct kvm_vcpu *vcpu, unsigned long server, resend = false; if (mfrr < new_state.cppr) { /* Reject a pending interrupt if not an IPI */ - if (mfrr <= new_state.pending_pri) + if (mfrr <= new_state.pending_pri) { reject = new_state.xisr; - new_state.pending_pri = mfrr; - new_state.xisr = XICS_IPI; + new_state.pending_pri = mfrr; + new_state.xisr = XICS_IPI; + } } - if (mfrr > old_state.mfrr && mfrr > new_state.cppr) { + if (mfrr > old_state.mfrr) { resend = new_state.need_resend; new_state.need_resend = 0; } @@ -641,7 +692,7 @@ static noinline int kvmppc_h_ipi(struct kvm_vcpu *vcpu, unsigned long server, /* Handle reject */ if (reject && reject != XICS_IPI) - icp_deliver_irq(xics, icp, reject); + icp_deliver_irq(xics, icp, reject, false); /* Handle resend */ if (resend) @@ -661,7 +712,7 @@ static int kvmppc_h_ipoll(struct kvm_vcpu *vcpu, unsigned long server) if (!icp) return H_PARAMETER; } - state = ACCESS_ONCE(icp->state); + state = READ_ONCE(icp->state); kvmppc_set_gpr(vcpu, 4, ((u32)state.cppr << 24) | state.xisr); kvmppc_set_gpr(vcpu, 5, state.mfrr); return H_SUCCESS; @@ -699,11 +750,10 @@ static noinline void kvmppc_h_cppr(struct kvm_vcpu *vcpu, unsigned long cppr) * We can remove EE from the current processor, the update * transaction will set it again if needed */ - kvmppc_book3s_dequeue_irqprio(icp->vcpu, - BOOK3S_INTERRUPT_EXTERNAL_LEVEL); + kvmppc_book3s_dequeue_irqprio(icp->vcpu, BOOK3S_INTERRUPT_EXTERNAL); do { - old_state = new_state = ACCESS_ONCE(icp->state); + old_state = new_state = READ_ONCE(icp->state); reject = 0; new_state.cppr = cppr; @@ -721,17 +771,54 @@ static noinline void kvmppc_h_cppr(struct kvm_vcpu *vcpu, unsigned long cppr) * attempt (see comments in icp_deliver_irq). */ if (reject && reject != XICS_IPI) - icp_deliver_irq(xics, icp, reject); + icp_deliver_irq(xics, icp, reject, false); } -static noinline int kvmppc_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr) +static int ics_eoi(struct kvm_vcpu *vcpu, u32 irq) { struct kvmppc_xics *xics = vcpu->kvm->arch.xics; struct kvmppc_icp *icp = vcpu->arch.icp; struct kvmppc_ics *ics; struct ics_irq_state *state; - u32 irq = xirr & 0x00ffffff; u16 src; + u32 pq_old, pq_new; + + /* + * ICS EOI handling: For LSI, if P bit is still set, we need to + * resend it. + * + * For MSI, we move Q bit into P (and clear Q). If it is set, + * resend it. + */ + + ics = kvmppc_xics_find_ics(xics, irq, &src); + if (!ics) { + XICS_DBG("ios_eoi: IRQ 0x%06x not found !\n", irq); + return H_PARAMETER; + } + state = &ics->irq_state[src]; + + if (state->lsi) + pq_new = state->pq_state; + else + do { + pq_old = state->pq_state; + pq_new = pq_old >> 1; + } while (cmpxchg(&state->pq_state, pq_old, pq_new) != pq_old); + + if (pq_new & PQ_PRESENTED) + icp_deliver_irq(xics, icp, irq, false); + + kvm_notify_acked_irq(vcpu->kvm, 0, irq); + + return H_SUCCESS; +} + +static noinline int kvmppc_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr) +{ + struct kvmppc_xics *xics = vcpu->kvm->arch.xics; + struct kvmppc_icp *icp = vcpu->arch.icp; + u32 irq = xirr & 0x00ffffff; XICS_DBG("h_eoi vcpu %d eoi %#lx\n", vcpu->vcpu_id, xirr); @@ -740,7 +827,7 @@ static noinline int kvmppc_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr) * * Note: If EOI is incorrectly used by SW to lower the CPPR * value (ie more favored), we do not check for rejection of - * a pending interrupt, this is a SW error and PAPR sepcifies + * a pending interrupt, this is a SW error and PAPR specifies * that we don't have to deal with it. * * The sending of an EOI to the ICS is handled after the @@ -754,27 +841,11 @@ static noinline int kvmppc_h_eoi(struct kvm_vcpu *vcpu, unsigned long xirr) /* IPIs have no EOI */ if (irq == XICS_IPI) return H_SUCCESS; - /* - * EOI handling: If the interrupt is still asserted, we need to - * resend it. We can take a lockless "peek" at the ICS state here. - * - * "Message" interrupts will never have "asserted" set - */ - ics = kvmppc_xics_find_ics(xics, irq, &src); - if (!ics) { - XICS_DBG("h_eoi: IRQ 0x%06x not found !\n", irq); - return H_PARAMETER; - } - state = &ics->irq_state[src]; - /* Still asserted, resend it */ - if (state->asserted) - icp_deliver_irq(xics, icp, irq); - - return H_SUCCESS; + return ics_eoi(vcpu, irq); } -static noinline int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall) +int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall) { struct kvmppc_xics *xics = vcpu->kvm->arch.xics; struct kvmppc_icp *icp = vcpu->arch.icp; @@ -782,17 +853,24 @@ static noinline int kvmppc_xics_rm_complete(struct kvm_vcpu *vcpu, u32 hcall) XICS_DBG("XICS_RM: H_%x completing, act: %x state: %lx tgt: %p\n", hcall, icp->rm_action, icp->rm_dbgstate.raw, icp->rm_dbgtgt); - if (icp->rm_action & XICS_RM_KICK_VCPU) + if (icp->rm_action & XICS_RM_KICK_VCPU) { + icp->n_rm_kick_vcpu++; kvmppc_fast_vcpu_kick(icp->rm_kick_target); - if (icp->rm_action & XICS_RM_CHECK_RESEND) - icp_check_resend(xics, icp); - if (icp->rm_action & XICS_RM_REJECT) - icp_deliver_irq(xics, icp, icp->rm_reject); + } + if (icp->rm_action & XICS_RM_CHECK_RESEND) { + icp->n_rm_check_resend++; + icp_check_resend(xics, icp->rm_resend_icp); + } + if (icp->rm_action & XICS_RM_NOTIFY_EOI) { + icp->n_rm_notify_eoi++; + kvm_notify_acked_irq(vcpu->kvm, 0, icp->rm_eoied_irq); + } icp->rm_action = 0; return H_SUCCESS; } +EXPORT_SYMBOL_GPL(kvmppc_xics_rm_complete); int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 req) { @@ -817,7 +895,7 @@ int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 req) } /* Check for real mode returning too hard */ - if (xics->real_mode) + if (xics->real_mode && is_kvmppc_hv_enabled(vcpu->kvm)) return kvmppc_xics_rm_complete(vcpu, req); switch (req) { @@ -839,20 +917,48 @@ int kvmppc_xics_hcall(struct kvm_vcpu *vcpu, u32 req) return rc; } +EXPORT_SYMBOL_GPL(kvmppc_xics_hcall); /* -- Initialisation code etc. -- */ +static void xics_debugfs_irqmap(struct seq_file *m, + struct kvmppc_passthru_irqmap *pimap) +{ + int i; + + if (!pimap) + return; + seq_printf(m, "========\nPIRQ mappings: %d maps\n===========\n", + pimap->n_mapped); + for (i = 0; i < pimap->n_mapped; i++) { + seq_printf(m, "r_hwirq=%x, v_hwirq=%x\n", + pimap->mapped[i].r_hwirq, pimap->mapped[i].v_hwirq); + } +} + static int xics_debug_show(struct seq_file *m, void *private) { struct kvmppc_xics *xics = m->private; struct kvm *kvm = xics->kvm; struct kvm_vcpu *vcpu; - int icsid, i; + int icsid; + unsigned long flags, i; + unsigned long t_rm_kick_vcpu, t_rm_check_resend; + unsigned long t_rm_notify_eoi; + unsigned long t_reject, t_check_resend; if (!kvm) return 0; + t_rm_kick_vcpu = 0; + t_rm_notify_eoi = 0; + t_rm_check_resend = 0; + t_check_resend = 0; + t_reject = 0; + + xics_debugfs_irqmap(m, kvm->arch.pimap); + seq_printf(m, "=========\nICP state\n=========\n"); kvm_for_each_vcpu(i, vcpu, kvm) { @@ -862,13 +968,23 @@ static int xics_debug_show(struct seq_file *m, void *private) if (!icp) continue; - state.raw = ACCESS_ONCE(icp->state.raw); + state.raw = READ_ONCE(icp->state.raw); seq_printf(m, "cpu server %#lx XIRR:%#x PPRI:%#x CPPR:%#x MFRR:%#x OUT:%d NR:%d\n", icp->server_num, state.xisr, state.pending_pri, state.cppr, state.mfrr, state.out_ee, state.need_resend); + t_rm_kick_vcpu += icp->n_rm_kick_vcpu; + t_rm_notify_eoi += icp->n_rm_notify_eoi; + t_rm_check_resend += icp->n_rm_check_resend; + t_check_resend += icp->n_check_resend; + t_reject += icp->n_reject; } + seq_printf(m, "ICP Guest->Host totals: kick_vcpu=%lu check_resend=%lu notify_eoi=%lu\n", + t_rm_kick_vcpu, t_rm_check_resend, + t_rm_notify_eoi); + seq_printf(m, "ICP Real Mode totals: check_resend=%lu resend=%lu\n", + t_check_resend, t_reject); for (icsid = 0; icsid <= KVMPPC_XICS_MAX_ICS_ID; icsid++) { struct kvmppc_ics *ics = xics->ics[icsid]; @@ -878,49 +994,32 @@ static int xics_debug_show(struct seq_file *m, void *private) seq_printf(m, "=========\nICS state for ICS 0x%x\n=========\n", icsid); - mutex_lock(&ics->lock); + local_irq_save(flags); + arch_spin_lock(&ics->lock); for (i = 0; i < KVMPPC_XICS_IRQ_PER_ICS; i++) { struct ics_irq_state *irq = &ics->irq_state[i]; - seq_printf(m, "irq 0x%06x: server %#x prio %#x save prio %#x asserted %d resend %d masked pending %d\n", + seq_printf(m, "irq 0x%06x: server %#x prio %#x save prio %#x pq_state %d resend %d masked pending %d\n", irq->number, irq->server, irq->priority, - irq->saved_priority, irq->asserted, + irq->saved_priority, irq->pq_state, irq->resend, irq->masked_pending); } - mutex_unlock(&ics->lock); + arch_spin_unlock(&ics->lock); + local_irq_restore(flags); } return 0; } -static int xics_debug_open(struct inode *inode, struct file *file) -{ - return single_open(file, xics_debug_show, inode->i_private); -} - -static const struct file_operations xics_debug_fops = { - .open = xics_debug_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; +DEFINE_SHOW_ATTRIBUTE(xics_debug); static void xics_debugfs_init(struct kvmppc_xics *xics) { - char *name; - - name = kasprintf(GFP_KERNEL, "kvm-xics-%p", xics); - if (!name) { - pr_err("%s: no memory for name\n", __func__); - return; - } - - xics->dentry = debugfs_create_file(name, S_IRUGO, powerpc_debugfs_root, + xics->dentry = debugfs_create_file("xics", 0444, xics->kvm->debugfs_dentry, xics, &xics_debug_fops); - pr_debug("%s: created %s\n", __func__, name); - kfree(name); + pr_debug("%s: created\n", __func__); } static struct kvmppc_ics *kvmppc_xics_create_ics(struct kvm *kvm, @@ -942,7 +1041,6 @@ static struct kvmppc_ics *kvmppc_xics_create_ics(struct kvm *kvm, if (!ics) goto out; - mutex_init(&ics->lock); ics->icsid = icsid; for (i = 0; i < KVMPPC_XICS_IRQ_PER_ICS; i++) { @@ -961,7 +1059,7 @@ static struct kvmppc_ics *kvmppc_xics_create_ics(struct kvm *kvm, return xics->ics[icsid]; } -int kvmppc_xics_create_icp(struct kvm_vcpu *vcpu, unsigned long server_num) +static int kvmppc_xics_create_icp(struct kvm_vcpu *vcpu, unsigned long server_num) { struct kvmppc_icp *icp; @@ -1045,8 +1143,7 @@ int kvmppc_xics_set_icp(struct kvm_vcpu *vcpu, u64 icpval) * Deassert the CPU interrupt request. * icp_try_update will reassert it if necessary. */ - kvmppc_book3s_dequeue_irqprio(icp->vcpu, - BOOK3S_INTERRUPT_EXTERNAL_LEVEL); + kvmppc_book3s_dequeue_irqprio(icp->vcpu, BOOK3S_INTERRUPT_EXTERNAL); /* * Note that if we displace an interrupt from old_state.xisr, @@ -1059,7 +1156,7 @@ int kvmppc_xics_set_icp(struct kvm_vcpu *vcpu, u64 icpval) * the ICS states before the ICP states. */ do { - old_state = ACCESS_ONCE(icp->state); + old_state = READ_ONCE(icp->state); if (new_state.mfrr <= old_state.mfrr) { resend = false; @@ -1084,13 +1181,15 @@ static int xics_get_source(struct kvmppc_xics *xics, long irq, u64 addr) u64 __user *ubufp = (u64 __user *) addr; u16 idx; u64 val, prio; + unsigned long flags; ics = kvmppc_xics_find_ics(xics, irq, &idx); if (!ics) return -ENOENT; irqp = &ics->irq_state[idx]; - mutex_lock(&ics->lock); + local_irq_save(flags); + arch_spin_lock(&ics->lock); ret = -ENOENT; if (irqp->exists) { val = irqp->server; @@ -1100,13 +1199,23 @@ static int xics_get_source(struct kvmppc_xics *xics, long irq, u64 addr) prio = irqp->saved_priority; } val |= prio << KVM_XICS_PRIORITY_SHIFT; - if (irqp->asserted) - val |= KVM_XICS_LEVEL_SENSITIVE | KVM_XICS_PENDING; - else if (irqp->masked_pending || irqp->resend) + if (irqp->lsi) { + val |= KVM_XICS_LEVEL_SENSITIVE; + if (irqp->pq_state & PQ_PRESENTED) + val |= KVM_XICS_PENDING; + } else if (irqp->masked_pending || irqp->resend) val |= KVM_XICS_PENDING; + + if (irqp->pq_state & PQ_PRESENTED) + val |= KVM_XICS_PRESENTED; + + if (irqp->pq_state & PQ_QUEUED) + val |= KVM_XICS_QUEUED; + ret = 0; } - mutex_unlock(&ics->lock); + arch_spin_unlock(&ics->lock); + local_irq_restore(flags); if (!ret && put_user(val, ubufp)) ret = -EFAULT; @@ -1123,6 +1232,7 @@ static int xics_set_source(struct kvmppc_xics *xics, long irq, u64 addr) u64 val; u8 prio; u32 server; + unsigned long flags; if (irq < KVMPPC_XICS_FIRST_IRQ || irq >= KVMPPC_XICS_NR_IRQS) return -ENOENT; @@ -1143,7 +1253,8 @@ static int xics_set_source(struct kvmppc_xics *xics, long irq, u64 addr) kvmppc_xics_find_server(xics->kvm, server) == NULL) return -EINVAL; - mutex_lock(&ics->lock); + local_irq_save(flags); + arch_spin_lock(&ics->lock); irqp->server = server; irqp->saved_priority = prio; if (val & KVM_XICS_MASKED) @@ -1151,24 +1262,33 @@ static int xics_set_source(struct kvmppc_xics *xics, long irq, u64 addr) irqp->priority = prio; irqp->resend = 0; irqp->masked_pending = 0; - irqp->asserted = 0; - if ((val & KVM_XICS_PENDING) && (val & KVM_XICS_LEVEL_SENSITIVE)) - irqp->asserted = 1; + irqp->lsi = 0; + irqp->pq_state = 0; + if (val & KVM_XICS_LEVEL_SENSITIVE) + irqp->lsi = 1; + /* If PENDING, set P in case P is not saved because of old code */ + if (val & KVM_XICS_PRESENTED || val & KVM_XICS_PENDING) + irqp->pq_state |= PQ_PRESENTED; + if (val & KVM_XICS_QUEUED) + irqp->pq_state |= PQ_QUEUED; irqp->exists = 1; - mutex_unlock(&ics->lock); + arch_spin_unlock(&ics->lock); + local_irq_restore(flags); if (val & KVM_XICS_PENDING) - icp_deliver_irq(xics, NULL, irqp->number); + icp_deliver_irq(xics, NULL, irqp->number, false); return 0; } -int kvm_set_irq(struct kvm *kvm, int irq_source_id, u32 irq, int level, - bool line_status) +int kvmppc_xics_set_irq(struct kvm *kvm, int irq_source_id, u32 irq, int level, + bool line_status) { struct kvmppc_xics *xics = kvm->arch.xics; - return ics_deliver_irq(xics, irq, level, line_status); + if (!xics) + return -ENODEV; + return ics_deliver_irq(xics, irq, level); } static int xics_set_attr(struct kvm_device *dev, struct kvm_device_attr *attr) @@ -1205,65 +1325,122 @@ static int xics_has_attr(struct kvm_device *dev, struct kvm_device_attr *attr) return -ENXIO; } -static void kvmppc_xics_free(struct kvm_device *dev) +/* + * Called when device fd is closed. kvm->lock is held. + */ +static void kvmppc_xics_release(struct kvm_device *dev) { struct kvmppc_xics *xics = dev->private; - int i; + unsigned long i; struct kvm *kvm = xics->kvm; + struct kvm_vcpu *vcpu; + + pr_devel("Releasing xics device\n"); + + /* + * Since this is the device release function, we know that + * userspace does not have any open fd referring to the + * device. Therefore there can not be any of the device + * attribute set/get functions being executed concurrently, + * and similarly, the connect_vcpu and set/clr_mapped + * functions also cannot be being executed. + */ debugfs_remove(xics->dentry); + /* + * We should clean up the vCPU interrupt presenters first. + */ + kvm_for_each_vcpu(i, vcpu, kvm) { + /* + * Take vcpu->mutex to ensure that no one_reg get/set ioctl + * (i.e. kvmppc_xics_[gs]et_icp) can be done concurrently. + * Holding the vcpu->mutex also means that execution is + * excluded for the vcpu until the ICP was freed. When the vcpu + * can execute again, vcpu->arch.icp and vcpu->arch.irq_type + * have been cleared and the vcpu will not be going into the + * XICS code anymore. + */ + mutex_lock(&vcpu->mutex); + kvmppc_xics_free_icp(vcpu); + mutex_unlock(&vcpu->mutex); + } + if (kvm) kvm->arch.xics = NULL; - for (i = 0; i <= xics->max_icsid; i++) + for (i = 0; i <= xics->max_icsid; i++) { kfree(xics->ics[i]); - kfree(xics); + xics->ics[i] = NULL; + } + /* + * A reference of the kvmppc_xics pointer is now kept under + * the xics_device pointer of the machine for reuse. It is + * freed when the VM is destroyed for now until we fix all the + * execution paths. + */ kfree(dev); } +static struct kvmppc_xics *kvmppc_xics_get_device(struct kvm *kvm) +{ + struct kvmppc_xics **kvm_xics_device = &kvm->arch.xics_device; + struct kvmppc_xics *xics = *kvm_xics_device; + + if (!xics) { + xics = kzalloc(sizeof(*xics), GFP_KERNEL); + *kvm_xics_device = xics; + } else { + memset(xics, 0, sizeof(*xics)); + } + + return xics; +} + static int kvmppc_xics_create(struct kvm_device *dev, u32 type) { struct kvmppc_xics *xics; struct kvm *kvm = dev->kvm; - int ret = 0; - xics = kzalloc(sizeof(*xics), GFP_KERNEL); + pr_devel("Creating xics for partition\n"); + + /* Already there ? */ + if (kvm->arch.xics) + return -EEXIST; + + xics = kvmppc_xics_get_device(kvm); if (!xics) return -ENOMEM; dev->private = xics; xics->dev = dev; xics->kvm = kvm; + kvm->arch.xics = xics; - /* Already there ? */ - mutex_lock(&kvm->lock); - if (kvm->arch.xics) - ret = -EEXIST; - else - kvm->arch.xics = xics; - mutex_unlock(&kvm->lock); - - if (ret) - return ret; - - xics_debugfs_init(xics); - -#ifdef CONFIG_KVM_BOOK3S_64_HV - if (cpu_has_feature(CPU_FTR_ARCH_206)) { +#ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE + if (cpu_has_feature(CPU_FTR_ARCH_206) && + cpu_has_feature(CPU_FTR_HVMODE)) { /* Enable real mode support */ xics->real_mode = ENABLE_REALMODE; xics->real_mode_dbg = DEBUG_REALMODE; } -#endif /* CONFIG_KVM_BOOK3S_64_HV */ +#endif /* CONFIG_KVM_BOOK3S_HV_POSSIBLE */ return 0; } +static void kvmppc_xics_init(struct kvm_device *dev) +{ + struct kvmppc_xics *xics = dev->private; + + xics_debugfs_init(xics); +} + struct kvm_device_ops kvm_xics_ops = { .name = "kvm-xics", .create = kvmppc_xics_create, - .destroy = kvmppc_xics_free, + .init = kvmppc_xics_init, + .release = kvmppc_xics_release, .set_attr = xics_set_attr, .get_attr = xics_get_attr, .has_attr = xics_has_attr, @@ -1279,7 +1456,7 @@ int kvmppc_xics_connect_vcpu(struct kvm_device *dev, struct kvm_vcpu *vcpu, return -EPERM; if (xics->kvm != vcpu->kvm) return -EPERM; - if (vcpu->arch.irq_type) + if (vcpu->arch.irq_type != KVMPPC_IRQ_DEFAULT) return -EBUSY; r = kvmppc_xics_create_icp(vcpu, xcpu); @@ -1297,3 +1474,34 @@ void kvmppc_xics_free_icp(struct kvm_vcpu *vcpu) vcpu->arch.icp = NULL; vcpu->arch.irq_type = KVMPPC_IRQ_DEFAULT; } + +void kvmppc_xics_set_mapped(struct kvm *kvm, unsigned long irq, + unsigned long host_irq) +{ + struct kvmppc_xics *xics = kvm->arch.xics; + struct kvmppc_ics *ics; + u16 idx; + + ics = kvmppc_xics_find_ics(xics, irq, &idx); + if (!ics) + return; + + ics->irq_state[idx].host_irq = host_irq; + ics->irq_state[idx].intr_cpu = -1; +} +EXPORT_SYMBOL_GPL(kvmppc_xics_set_mapped); + +void kvmppc_xics_clr_mapped(struct kvm *kvm, unsigned long irq, + unsigned long host_irq) +{ + struct kvmppc_xics *xics = kvm->arch.xics; + struct kvmppc_ics *ics; + u16 idx; + + ics = kvmppc_xics_find_ics(xics, irq, &idx); + if (!ics) + return; + + ics->irq_state[idx].host_irq = 0; +} +EXPORT_SYMBOL_GPL(kvmppc_xics_clr_mapped); |