the code for credit scheduler
-----------------------------------------------------------------------------------------------------------------------
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 4d76bb6..9e88ff0 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -284,7 +284,7 @@ int kvm_apic_match_dest(struct kvm_vcpu *vcpu,
struct kvm_lapic *source,
"dest_mode 0x%x, short_hand 0x%x\n",
target, source, dest, dest_mode, short_hand);
- ASSERT(!target);
+ ASSERT(target);
switch (short_hand) {
case APIC_DEST_NOSHORT:
if (dest_mode == 0)
diff --git a/arch/x86/kvm/sched_credit.c b/arch/x86/kvm/sched_credit.c
new file mode 100644
index 0000000..77c726a
--- /dev/null
+++ b/arch/x86/kvm/sched_credit.c
@@ -0,0 +1,1604 @@
+/****************************************************************************
+ * (C) 2005-2006 - Emmanuel Ackaouy - XenSource Inc.
+ * (C) 2009 - Xiaojian Liu
+ ****************************************************************************
+ *
+ * File: common/csched_credit.c
+ * Author: Emmanuel Ackaouy
+ * Updated by Xiaojian Liu to make it runnable under KVM
+ *
+ * Description: Credit-based SMP CPU scheduler
+ */
+#include <linux/cpumask.h>
+#include <linux/kvm_host.h>
+#include <linux/trace.h>
+#include <linux/sched-if.h>
+#include <linux/ipi.h>
+#include <linux/hardirq.h>
+
+#ifdef DEBUG
+#define ASSERT(x) \
+do { \
+ if (!(x)) { \
+ printk(KERN_EMERG "assertion failed %s: %d: %s\n", \
+ __FILE__, __LINE__, #x); \
+ BUG(); \
+ } \
+} while (0)
+#else
+#define ASSERT(x) do { } while (0)
+#endif
+/*
+ * CSCHED_STATS
+ *
+ * Manage very basic counters and stats.
+ *
+ * Useful for debugging live systems. The stats are displayed
+ * with runq dumps ('r' on the Xen console).
+ */
+#define CSCHED_STATS
+
+#define MILLISECS(_ms) ((s_time_t)((_ms) * 1000000ULL))
+
+/*
+ * Basic constants
+ */
+#define CSCHED_DEFAULT_WEIGHT 256
+#define CSCHED_TICKS_PER_TSLICE 3
+#define CSCHED_TICKS_PER_ACCT 3
+#define CSCHED_MSECS_PER_TICK 10
+#define CSCHED_MSECS_PER_TSLICE \
+ (CSCHED_MSECS_PER_TICK * CSCHED_TICKS_PER_TSLICE)
+#define CSCHED_CREDITS_PER_TICK 100
+#define CSCHED_CREDITS_PER_TSLICE \
+ (CSCHED_CREDITS_PER_TICK * CSCHED_TICKS_PER_TSLICE)
+#define CSCHED_CREDITS_PER_ACCT \
+ (CSCHED_CREDITS_PER_TICK * CSCHED_TICKS_PER_ACCT)
+
+
+/*
+ * Priorities
+ */
+#define CSCHED_PRI_TS_BOOST 0 /* time-share waking up */
+#define CSCHED_PRI_TS_UNDER -1 /* time-share w/ credits */
+#define CSCHED_PRI_TS_OVER -2 /* time-share w/o credits */
+#define CSCHED_PRI_IDLE -64 /* idle */
+
+
+/*
+ * Flags
+ */
+#define CSCHED_FLAG_VCPU_PARKED 0x0001 /* VCPU over capped credits */
+
+
+/*
+ * Useful macros
+ */
+#define CSCHED_PCPU(_c) \
+ ((struct csched_pcpu *)per_cpu(schedule_data, _c).sched_priv)
+#define CSCHED_VCPU(_vcpu) ((struct csched_vcpu *) (_vcpu)->sched_priv)
+#define CSCHED_VM(_dom) ((struct csched_vm *) (_dom)->sched_priv)
+#define RUNQ(_cpu) (&(CSCHED_PCPU(_cpu)->runq))
+
+
+/*
+ * Stats
+ */
+#ifdef CSCHED_STATS
+
+#define CSCHED_STAT(_X) (csched_priv.stats._X)
+#define CSCHED_STAT_DEFINE(_X) uint32_t _X;
+#define CSCHED_STAT_PRINTK(_X) \
+ do \
+ { \
+ printk("\t%-30s = %u\n", #_X, CSCHED_STAT(_X)); \
+ } while ( 0 );
+
+/*
+ * Try and keep often cranked stats on top so they'll fit on one
+ * cache line.
+ */
+#define CSCHED_STATS_EXPAND_SCHED(_MACRO) \
+ _MACRO(schedule) \
+ _MACRO(acct_run) \
+ _MACRO(acct_no_work) \
+ _MACRO(acct_balance) \
+ _MACRO(acct_reorder) \
+ _MACRO(acct_min_credit) \
+ _MACRO(acct_vcpu_active) \
+ _MACRO(acct_vcpu_idle) \
+ _MACRO(vcpu_sleep) \
+ _MACRO(vcpu_wake_running) \
+ _MACRO(vcpu_wake_onrunq) \
+ _MACRO(vcpu_wake_runnable) \
+ _MACRO(vcpu_wake_not_runnable) \
+ _MACRO(vcpu_park) \
+ _MACRO(vcpu_unpark) \
+ _MACRO(tickle_local_idler) \
+ _MACRO(tickle_local_over) \
+ _MACRO(tickle_local_under) \
+ _MACRO(tickle_local_other) \
+ _MACRO(tickle_idlers_none) \
+ _MACRO(tickle_idlers_some) \
+ _MACRO(load_balance_idle) \
+ _MACRO(load_balance_over) \
+ _MACRO(load_balance_other) \
+ _MACRO(steal_trylock_failed) \
+ _MACRO(steal_peer_idle) \
+ _MACRO(migrate_queued) \
+ _MACRO(migrate_running) \
+ _MACRO(dom_init) \
+ _MACRO(dom_destroy) \
+ _MACRO(vcpu_init) \
+ _MACRO(vcpu_destroy)
+
+#ifndef NDEBUG
+#define CSCHED_STATS_EXPAND_CHECKS(_MACRO) \
+ _MACRO(vcpu_check)
+#else
+#define CSCHED_STATS_EXPAND_CHECKS(_MACRO)
+#endif
+
+#define CSCHED_STATS_EXPAND(_MACRO) \
+ CSCHED_STATS_EXPAND_CHECKS(_MACRO) \
+ CSCHED_STATS_EXPAND_SCHED(_MACRO)
+
+#define CSCHED_STATS_RESET() \
+ do \
+ { \
+ memset(&csched_priv.stats, 0, sizeof(csched_priv.stats)); \
+ } while ( 0 )
+
+#define CSCHED_STATS_DEFINE() \
+ struct \
+ { \
+ CSCHED_STATS_EXPAND(CSCHED_STAT_DEFINE) \
+ } stats;
+
+#define CSCHED_STATS_PRINTK() \
+ do \
+ { \
+ printk("stats:\n"); \
+ CSCHED_STATS_EXPAND(CSCHED_STAT_PRINTK) \
+ } while ( 0 )
+
+#define CSCHED_STAT_CRANK(_X) (CSCHED_STAT(_X)++)
+
+#define CSCHED_VCPU_STATS_RESET(_V) \
+ do \
+ { \
+ memset(&(_V)->stats, 0, sizeof((_V)->stats)); \
+ } while ( 0 )
+
+#define CSCHED_VCPU_STAT_CRANK(_V, _X) (((_V)->stats._X)++)
+
+#define CSCHED_VCPU_STAT_SET(_V, _X, _Y) (((_V)->stats._X) = (_Y))
+
+#else /* CSCHED_STATS */
+
+#define CSCHED_STATS_RESET() do {} while ( 0 )
+#define CSCHED_STATS_DEFINE()
+#define CSCHED_STATS_PRINTK() do {} while ( 0 )
+#define CSCHED_STAT_CRANK(_X) do {} while ( 0 )
+#define CSCHED_VCPU_STATS_RESET(_V) do {} while ( 0 )
+#define CSCHED_VCPU_STAT_CRANK(_V, _X) do {} while ( 0 )
+#define CSCHED_VCPU_STAT_SET(_V, _X, _Y) do {} while ( 0 )
+
+#endif /* CSCHED_STATS */
+
+
+/*
+ * Physical CPU
+ */
+struct csched_pcpu {
+ struct list_head runq;
+ uint32_t runq_sort_last;
+ struct hrtimer ticker;
+ unsigned int tick;
+ int cpu;
+ volatile bool in_use;
+};
+
+/*
+ * Virtual CPU
+ */
+struct csched_vcpu {
+ struct list_head runq_elem;
+ struct list_head active_vcpu_elem;
+ struct csched_vm *sdom;
+ struct kvm_vcpu *vcpu;
+ atomic_t credit;
+ uint16_t flags;
+ int16_t pri;
+#ifdef CSCHED_STATS
+ struct {
+ int credit_last;
+ uint32_t credit_incr;
+ uint32_t state_active;
+ uint32_t state_idle;
+ uint32_t migrate_q;
+ uint32_t migrate_r;
+ } stats;
+#endif
+};
+
+/*
+ * Domain
+ */
+struct csched_vm{
+ struct list_head active_vcpu;
+ struct list_head active_sdom_elem;
+ struct kvm *vm;
+ uint16_t active_vcpu_count;
+ uint16_t weight;
+ uint16_t cap;
+};
+
+/*
+ * System-wide private data
+ */
+struct csched_private {
+ spinlock_t lock;
+ struct list_head active_sdom;
+ uint32_t ncpus;
+ unsigned int master;
+ cpumask_t idlers;
+ uint32_t weight;
+ uint32_t credit;
+ int credit_balance;
+ uint32_t runq_sort;
+ CSCHED_STATS_DEFINE()
+};
+
+/*
+ * Global variables
+ */
+static struct csched_private csched_priv;
+
+static enum hrtimer_restart csched_tick_stub(struct hrtimer *data);
+DECLARE_PER_CPU(cpumask_t, cpu_core_map);
+DECLARE_PER_CPU(cpumask_t, cpu_sibling_map);
+
+extern void vcpu_pause(struct kvm_vcpu *v);
+extern void vcpu_unpause(struct kvm_vcpu *v);
+extern void vcpu_pause_nosync(struct kvm_vcpu *v);
+
+static void csched_disable(int cpu)
+{
+ struct csched_pcpu *spc = CSCHED_PCPU(cpu);
+ tasklet_kill(&per_cpu(schedule_data, cpu).tick_tasklet);
+}
+void kvm_try_tasklet_schedule(void *_unused)
+{
+ int cpu = get_cpu();
+ TRACE_2D(TRC_TASKLET_SCHEDULE, cpu, __LINE__);
+ if (!spin_is_locked(&per_cpu(schedule_data, cpu).schedule_lock)){
+ tasklet_schedule(&per_cpu(schedule_data, cpu).sched_tasklet);
+ TRACE_2D(TRC_TASKLET_SCHEDULE, cpu, __LINE__);
+ }
+ put_cpu();
+}
+static inline int
+__cycle_cpu(int cpu, const cpumask_t *mask)
+{
+ int nxt = next_cpu(cpu, *mask);
+ if (nxt == NR_CPUS)
+ nxt = first_cpu(*mask);
+ return nxt;
+}
+
+static inline int
+__vcpu_on_runq(struct csched_vcpu *svc)
+{
+ return !list_empty(&svc->runq_elem);
+}
+
+static inline struct csched_vcpu *
+__runq_elem(struct list_head *elem)
+{
+ return list_entry(elem, struct csched_vcpu, runq_elem);
+}
+
+static inline void
+__runq_insert(unsigned int cpu, struct csched_vcpu *svc)
+{
+ const struct list_head * const runq = RUNQ(cpu);
+ struct list_head *iter;
+
+ BUG_ON( __vcpu_on_runq(svc) );
+ BUG_ON( cpu != svc->vcpu->processor);
+
+ list_for_each( iter, runq )
+ {
+ const struct csched_vcpu * const iter_svc = __runq_elem(iter);
+ if ( svc->pri > iter_svc->pri )
+ break;
+ }
+
+ list_add_tail(&svc->runq_elem, iter);
+}
+
+static inline void
+__runq_remove(struct csched_vcpu *svc)
+{
+ BUG_ON( !__vcpu_on_runq(svc) );
+ list_del_init(&svc->runq_elem);
+}
+
+static inline void
+__runq_tickle(unsigned int cpu, struct csched_vcpu *new)
+{
+ struct csched_vcpu * const cur =
+ CSCHED_VCPU(per_cpu(schedule_data, cpu).curr);
+ cpumask_var_t cpus;
+
+ int my_cpu = get_cpu();
+
+ ASSERT(cur);
+ if(alloc_cpumask_var(&cpus, GFP_ATOMIC))
+ cpumask_clear(cpus);
+
+ /* If strictly higher priority than current VCPU, signal the CPU */
+ if ( new->pri > cur->pri )
+ {
+ if ( cur->pri == CSCHED_PRI_IDLE )
+ CSCHED_STAT_CRANK(tickle_local_idler);
+ else if ( cur->pri == CSCHED_PRI_TS_OVER )
+ CSCHED_STAT_CRANK(tickle_local_over);
+ else if ( cur->pri == CSCHED_PRI_TS_UNDER )
+ CSCHED_STAT_CRANK(tickle_local_under);
+ else
+ CSCHED_STAT_CRANK(tickle_local_other);
+
+ cpumask_set_cpu(cpu, cpus);
+ }
+
+ /*
+ * If this CPU has at least two runnable VCPUs, we tickle any idlers to
+ * let them know there is runnable work in the system...
+ */
+ if ( cur->pri > CSCHED_PRI_IDLE )
+ {
+ if ( cpumask_empty(&csched_priv.idlers) )
+ {
+ CSCHED_STAT_CRANK(tickle_idlers_none);
+ }
+ else
+ {
+ CSCHED_STAT_CRANK(tickle_idlers_some);
+ cpumask_or(cpus, cpus, &csched_priv.idlers);
+ cpumask_and(cpus, cpus, &new->vcpu->cpu_affinity);
+ }
+ }
+
+ /* Send scheduler interrupts to designated CPUs */
+ if (cpu_isset(my_cpu, *cpus)){
+ kvm_try_tasklet_schedule(NULL);
+ cpu_clear(my_cpu, cpus);
+ }
+
+ if ( !cpumask_empty(cpus) ){
+ int r;
+ r = insert_pending_ipi(my_cpu, *cpus,
kvm_try_tasklet_schedule, NULL, 1);
+ if (r) {
+ dump_traces(NULL);
+ BUG_ON(1);
+ }
+ wake_up(&per_cpu(schedule_data, my_cpu).ipi_wq);
+ }
+ free_cpumask_var(cpus);
+ put_cpu();
+}
+
+static int
+csched_pcpu_init(int cpu)
+{
+ struct csched_pcpu *spc;
+ unsigned long flags;
+ printk("cpu %d exec func %s\n", cpu, __FUNCTION__);
+
+ /* Allocate per-PCPU info */
+ spc = kmalloc(sizeof(struct csched_pcpu), GFP_KERNEL);
+ if ( spc == NULL )
+ return -1;
+
+ spin_lock_irqsave(&csched_priv.lock, flags);
+
+ /* Initialize/update system-wide config */
+ csched_priv.credit += CSCHED_CREDITS_PER_ACCT;
+ if ( csched_priv.ncpus <= cpu )
+ csched_priv.ncpus = cpu + 1;
+ if ( csched_priv.master >= csched_priv.ncpus )
+ csched_priv.master = cpu;
+
+ spc->cpu = cpu;
+ hrtimer_init(&spc->ticker, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
+ spc->ticker.function = csched_tick_stub;
+ // hrtimer_data_pointer(&spc->ticker);
+
+ INIT_LIST_HEAD(&spc->runq);
+ spc->runq_sort_last = csched_priv.runq_sort;
+ per_cpu(schedule_data, cpu).sched_priv = spc;
+
+ spc->in_use = false;
+
+ /* Start off idling... */
+ BUG_ON(!is_idle_vcpu(per_cpu(schedule_data, cpu).curr));
+ cpumask_set_cpu(cpu, &csched_priv.idlers);
+
+ spin_unlock_irqrestore(&csched_priv.lock, flags);
+ return 0;
+}
+
+#ifndef NDEBUG
+static inline void
+__csched_vcpu_check(struct kvm_vcpu *vc)
+{
+ struct csched_vcpu * const svc = CSCHED_VCPU(vc);
+ struct csched_vm * const sdom = svc->sdom;
+
+ BUG_ON( svc->vcpu != vc );
+ BUG_ON( sdom != CSCHED_VM(vc->kvm) );
+ if ( sdom )
+ {
+ BUG_ON( is_idle_vcpu(vc) );
+ BUG_ON( sdom->vm != vc->kvm);
+ }
+ else
+ {
+ BUG_ON( !is_idle_vcpu(vc) );
+ }
+
+ CSCHED_STAT_CRANK(vcpu_check);
+}
+#define CSCHED_VCPU_CHECK(_vc) (__csched_vcpu_check(_vc))
+#else
+#define CSCHED_VCPU_CHECK(_vc)
+#endif
+
+static inline int
+__csched_vcpu_is_migrateable(struct kvm_vcpu *vc, int dest_cpu)
+{
+ /*
+ * Don't pick up work that's in the peer's scheduling tail. Also only pick
+ * up work that's allowed to run on our CPU.
+ */
+
+ if(!per_cpu(schedule_data, raw_smp_processor_id()).can_migrate) return 0;
+ return !vc->is_running && cpu_isset(dest_cpu, vc->cpu_affinity);
+}
+
+static int
+csched_cpu_pick(struct kvm_vcpu *vc)
+{
+ cpumask_t cpus;
+ cpumask_t idlers;
+ int cpu;
+
+ /*
+ * Pick from online CPUs in VCPU's affinity mask, giving a
+ * preference to its current processor if it's in there.
+ */
+ cpus_and(cpus, cpu_online_map, vc->cpu_affinity);
+ cpu = cpu_isset(vc->processor, cpus)
+ ? vc->processor
+ : __cycle_cpu(vc->processor, &cpus);
+ ASSERT( !cpus_empty(cpus) && cpu_isset(cpu, cpus) );
+
+ /*
+ * Try to find an idle processor within the above constraints.
+ *
+ * In multi-core and multi-threaded CPUs, not all idle execution
+ * vehicles are equal!
+ *
+ * We give preference to the idle execution vehicle with the most
+ * idling neighbours in its grouping. This distributes work across
+ * distinct cores first and guarantees we don't do something stupid
+ * like run two VCPUs on co-hyperthreads while there are idle cores
+ * or sockets.
+ */
+ idlers = csched_priv.idlers;
+ cpu_set(cpu, idlers);
+ cpus_and(cpus, cpus, idlers);
+ cpu_clear(cpu, cpus);
+
+ while ( !cpus_empty(cpus) )
+ {
+ cpumask_t cpu_idlers;
+ cpumask_t nxt_idlers;
+ int nxt;
+
+ nxt = __cycle_cpu(cpu, &cpus);
+
+ if ( cpu_isset(cpu, per_cpu(cpu_core_map,nxt)) )
+ {
+ ASSERT( cpu_isset(nxt, per_cpu(cpu_core_map,cpu)) );
+ cpus_and(cpu_idlers, idlers, per_cpu(cpu_sibling_map,cpu));
+ cpus_and(nxt_idlers, idlers, per_cpu(cpu_sibling_map,nxt));
+ }
+ else
+ {
+ ASSERT( !cpu_isset(nxt, per_cpu(cpu_core_map,cpu)) );
+ cpus_and(cpu_idlers, idlers, per_cpu(cpu_core_map,cpu));
+ cpus_and(nxt_idlers, idlers, per_cpu(cpu_core_map,nxt));
+ }
+
+ if ( cpus_weight(cpu_idlers) < cpus_weight(nxt_idlers) )
+ {
+ cpu = nxt;
+ cpu_clear(cpu, cpus);
+ }
+ else
+ {
+ cpus_andnot(cpus, cpus, nxt_idlers);
+ }
+ }
+
+ return cpu;
+}
+
+static inline void
+__csched_vcpu_acct_start(struct csched_vcpu *svc)
+{
+ struct csched_vm * const sdom = svc->sdom;
+ unsigned long flags;
+
+ spin_lock_irqsave(&csched_priv.lock, flags);
+
+ if ( list_empty(&svc->active_vcpu_elem) )
+ {
+ CSCHED_VCPU_STAT_CRANK(svc, state_active);
+ CSCHED_STAT_CRANK(acct_vcpu_active);
+
+ sdom->active_vcpu_count++;
+ list_add(&svc->active_vcpu_elem, &sdom->active_vcpu);
+ if ( list_empty(&sdom->active_sdom_elem) )
+ {
+ list_add(&sdom->active_sdom_elem, &csched_priv.active_sdom);
+ csched_priv.weight += sdom->weight;
+ }
+ }
+
+ spin_unlock_irqrestore(&csched_priv.lock, flags);
+}
+
+static inline void
+__csched_vcpu_acct_stop_locked(struct csched_vcpu *svc)
+{
+ struct csched_vm * const sdom = svc->sdom;
+
+ BUG_ON( list_empty(&svc->active_vcpu_elem) );
+
+ CSCHED_VCPU_STAT_CRANK(svc, state_idle);
+ CSCHED_STAT_CRANK(acct_vcpu_idle);
+
+ sdom->active_vcpu_count--;
+ list_del_init(&svc->active_vcpu_elem);
+ if ( list_empty(&sdom->active_vcpu) )
+ {
+ BUG_ON( csched_priv.weight < sdom->weight );
+ list_del_init(&sdom->active_sdom_elem);
+ csched_priv.weight -= sdom->weight;
+ }
+}
+
+static void
+csched_vcpu_acct(unsigned int cpu)
+{
+ struct csched_vcpu * const svc = CSCHED_VCPU(current_vcpu);
+
+ if ( current_vcpu->processor != cpu ) {
+ if(is_host_vcpu(current_vcpu)) printk("host_vcpu runing ");
+ if(is_idle_vcpu(current_vcpu)) printk("idle_vcpu runing ");
+ printk(" on cpu %d me is cpu %d\n", current_vcpu->processor, cpu);
+ BUG_ON(1);
+ }
+ ASSERT( svc->sdom != NULL );
+
+ /*
+ * If this VCPU's priority was boosted when it last awoke, reset it.
+ * If the VCPU is found here, then it's consuming a non-negligeable
+ * amount of CPU resources and should no longer be boosted.
+ */
+ if ( svc->pri == CSCHED_PRI_TS_BOOST )
+ svc->pri = CSCHED_PRI_TS_UNDER;
+
+ /*
+ * Update credits
+ */
+ atomic_sub(CSCHED_CREDITS_PER_TICK, &svc->credit);
+
+ /*
+ * Put this VCPU and domain back on the active list if it was
+ * idling.
+ *
+ * If it's been active a while, check if we'd be better off
+ * migrating it to run elsewhere (see multi-core and multi-thread
+ * support in csched_cpu_pick()).
+ */
+ if ( list_empty(&svc->active_vcpu_elem) )
+ {
+ __csched_vcpu_acct_start(svc);
+ }
+ else if ( csched_cpu_pick(current_vcpu) != cpu )
+ {
+ CSCHED_VCPU_STAT_CRANK(svc, migrate_r);
+ CSCHED_STAT_CRANK(migrate_running);
+ set_bit(_VPF_migrating, ¤t_vcpu->pause_flags);
+ kvm_try_tasklet_schedule(NULL);
+ }
+}
+
+static int
+csched_vcpu_init(struct kvm_vcpu *vc)
+{
+ struct kvm* const kvm= vc->kvm;
+ struct csched_vm *sdom = CSCHED_VM(kvm);
+ struct csched_vcpu *svc;
+
+
+ CSCHED_STAT_CRANK(vcpu_init);
+
+ /* Allocate per-VCPU info */
+ svc = kmalloc(sizeof(struct csched_vcpu), GFP_KERNEL);
+ if ( svc == NULL )
+ return -1;
+
+ INIT_LIST_HEAD(&svc->runq_elem);
+ INIT_LIST_HEAD(&svc->active_vcpu_elem);
+ svc->sdom = sdom;
+ svc->vcpu = vc;
+ atomic_set(&svc->credit, 0);
+ svc->flags = 0U;
+ svc->pri = is_idle_vm(kvm) ? CSCHED_PRI_IDLE : CSCHED_PRI_TS_UNDER;
+ CSCHED_VCPU_STATS_RESET(svc);
+ vc->sched_priv = svc;
+
+ /* Allocate per-PCPU info */
+ if ( unlikely(!CSCHED_PCPU(vc->processor)) )
+ {
+ if ( csched_pcpu_init(vc->processor) != 0 )
+ return -1;
+ }
+
+ CSCHED_VCPU_CHECK(vc);
+ return 0;
+}
+
+static void
+csched_vcpu_destroy(struct kvm_vcpu *vc)
+{
+ struct csched_vcpu * const svc = CSCHED_VCPU(vc);
+ struct csched_vm * const sdom = svc->sdom;
+ unsigned long flags;
+
+ CSCHED_STAT_CRANK(vcpu_destroy);
+
+ if(is_idle_vcpu(vc) || is_host_vcpu(vc)) {
+ kfree(svc);
+ return;
+ }
+
+ BUG_ON( sdom == NULL );
+ BUG_ON( !list_empty(&svc->runq_elem) );
+
+ spin_lock_irqsave(&csched_priv.lock, flags);
+
+ if ( !list_empty(&svc->active_vcpu_elem) )
+ __csched_vcpu_acct_stop_locked(svc);
+
+ spin_unlock_irqrestore(&csched_priv.lock, flags);
+
+ kfree(svc);
+}
+
+static void
+csched_vcpu_sleep(struct kvm_vcpu *vc)
+{
+ struct csched_vcpu * const svc = CSCHED_VCPU(vc);
+ int my_cpu = get_cpu();
+
+ CSCHED_STAT_CRANK(vcpu_sleep);
+
+ BUG_ON( is_idle_vcpu(vc) );
+
+ if ( per_cpu(schedule_data, vc->processor).curr == vc ){
+ if(vc->processor== my_cpu)
+ kvm_try_tasklet_schedule(NULL);
+ else{
+ int r;
+ r = insert_pending_ipi(my_cpu, cpumask_of_cpu(vc->processor),
kvm_try_tasklet_schedule, NULL, 1);
+ if(r) {
+ dump_traces(NULL);
+ BUG_ON(1);
+ }
+ wake_up(&per_cpu(schedule_data, my_cpu).ipi_wq);
+ }
+ }
+ else if ( __vcpu_on_runq(svc) )
+ __runq_remove(svc);
+ put_cpu();
+}
+
+static void
+csched_vcpu_wake(struct kvm_vcpu *vc)
+{
+ struct csched_vcpu * const svc = CSCHED_VCPU(vc);
+ const unsigned int cpu = vc->processor;
+ BUG_ON( is_idle_vcpu(vc) );
+ TRACE_2D(TRC_CSCHED_WAKE, __LINE__, (unsigned long)vc);
+
+ if ( unlikely(per_cpu(schedule_data, cpu).curr == vc) )
+ {
+ CSCHED_STAT_CRANK(vcpu_wake_running);
+ return;
+ }
+ if ( unlikely(__vcpu_on_runq(svc)) )
+ {
+ CSCHED_STAT_CRANK(vcpu_wake_onrunq);
+ return;
+ }
+
+ if ( likely(vcpu_runnable(vc)) )
+ CSCHED_STAT_CRANK(vcpu_wake_runnable);
+ else
+ CSCHED_STAT_CRANK(vcpu_wake_not_runnable);
+
+ /*
+ * We temporarly boost the priority of awaking VCPUs!
+ *
+ * If this VCPU consumes a non negligeable amount of CPU, it
+ * will eventually find itself in the credit accounting code
+ * path where its priority will be reset to normal.
+ *
+ * If on the other hand the VCPU consumes little CPU and is
+ * blocking and awoken a lot (doing I/O for example), its
+ * priority will remain boosted, optimizing it's wake-to-run
+ * latencies.
+ *
+ * This allows wake-to-run latency sensitive VCPUs to preempt
+ * more CPU resource intensive VCPUs without impacting overall
+ * system fairness.
+ *
+ * The one exception is for VCPUs of capped domains unpausing
+ * after earning credits they had overspent. We don't boost
+ * those.
+ */
+ if ( svc->pri == CSCHED_PRI_TS_UNDER &&
+ !(svc->flags & CSCHED_FLAG_VCPU_PARKED) )
+ {
+ svc->pri = CSCHED_PRI_TS_BOOST;
+ }
+
+ /* Put the VCPU on the runq and tickle CPUs */
+ TRACE_2D(TRC_CSCHED_WAKE, __LINE__, vc);
+ __runq_insert(cpu, svc);
+ TRACE_2D(TRC_CSCHED_WAKE, __LINE__, vc);
+ __runq_tickle(cpu, svc);
+ TRACE_2D(TRC_CSCHED_WAKE, __LINE__, vc);
+}
+static void get_param(char* param, char **endp, int *weight, int *cap)
+{
+ char *opt_1;
+ char *opt_2;
+ /* skip heading spaces */
+ while((*param == ' ') || (*param == '\t') || (*param == '\n'))
+ param++;
+
+ opt_1 = strchr(param, '=');
+ if (opt_1 != NULL){
+ char *end = opt_1-1;
+ *opt_1++ = '\0';
+ while((*end == ' ') || (*end == '\t') || (*end == '\n'))
+ *end-- = '\0';
+ }
+ if(strcmp(param, "weight") == 0)
+ *weight = simple_strtol(opt_1, endp, 0);
+ else if(strcmp(param, "cap") == 0)
+ *cap = simple_strtol(opt_1, endp, 0);
+}
+static int csched_read_param(struct kvm *kvm, char *buf, int size)
+{
+ struct csched_vm * const sdom = CSCHED_VM(kvm);
+ unsigned long flags;
+ int r;
+
+ r = snprintf(buf, size, "Credit Schedule parameters for VM %d\n"
+ "Weight: %d Cap: %d\n", kvm->vmid, sdom->weight, sdom->cap);
+ if (r < 0) printk("the buf for sched_read_param is too small!\n");
+ return (r>0);
+}
+static int csched_write_param(struct kvm *kvm, char *param)
+{
+ char *tmp;
+ int weight = 256;
+ int cap = 0;
+ struct csched_vm * const sdom = CSCHED_VM(kvm);
+ unsigned long flags;
+
+ get_param(param, &tmp, &weight, &cap);
+ get_param(tmp, &tmp, &weight, &cap);
+ spin_lock_irqsave(&csched_priv.lock, flags);
+
+ if ( !list_empty(&sdom->active_sdom_elem) )
+ {
+ csched_priv.weight -= sdom->weight;
+ csched_priv.weight += weight;
+ }
+ sdom->weight = weight;
+ sdom->cap = cap;
+ spin_unlock_irqrestore(&csched_priv.lock, flags);
+ printk("the weight is %d cap is %d\n", weight, cap);
+ return 1;
+}
+
+static int
+csched_vm_init(struct kvm *kvm)
+{
+ struct csched_vm *sdom;
+
+ CSCHED_STAT_CRANK(dom_init);
+ if ( is_idle_vm(kvm) )
+ return 0;
+
+ sdom = kmalloc(sizeof(struct csched_vm), GFP_KERNEL);
+ if ( sdom == NULL )
+ return -ENOMEM;
+
+ /* Initialize credit and weight */
+ INIT_LIST_HEAD(&sdom->active_vcpu);
+ sdom->active_vcpu_count = 0;
+ INIT_LIST_HEAD(&sdom->active_sdom_elem);
+ sdom->vm = kvm;
+ sdom->weight = CSCHED_DEFAULT_WEIGHT;
+ sdom->cap = 0U;
+ kvm->sched_priv = sdom;
+
+ return 0;
+}
+
+static void
+csched_vm_destroy(struct kvm *kvm)
+{
+ CSCHED_STAT_CRANK(dom_destroy);
+ kfree(CSCHED_VM(kvm));
+}
+
+/*
+ * This is a O(n) optimized sort of the runq.
+ *
+ * Time-share VCPUs can only be one of two priorities, UNDER or OVER. We walk
+ * through the runq and move up any UNDERs that are preceded by OVERS. We
+ * remember the last UNDER to make the move up operation O(1).
+ */
+static void
+csched_runq_sort(unsigned int cpu)
+{
+ struct csched_pcpu * const spc = CSCHED_PCPU(cpu);
+ struct list_head *runq, *elem, *next, *last_under;
+ struct csched_vcpu *svc_elem;
+ unsigned long flags;
+ int sort_epoch;
+
+ sort_epoch = csched_priv.runq_sort;
+ if ( sort_epoch == spc->runq_sort_last )
+ return;
+
+ spc->runq_sort_last = sort_epoch;
+
+ spin_lock_irqsave(&per_cpu(schedule_data, cpu).schedule_lock, flags);
+
+ runq = &spc->runq;
+ elem = runq->next;
+ last_under = runq;
+
+ while ( elem != runq )
+ {
+ next = elem->next;
+ svc_elem = __runq_elem(elem);
+
+ if ( svc_elem->pri >= CSCHED_PRI_TS_UNDER )
+ {
+ /* does elem need to move up the runq? */
+ if ( elem->prev != last_under )
+ {
+ list_del(elem);
+ list_add(elem, last_under);
+ }
+ last_under = elem;
+ }
+
+ elem = next;
+ }
+ spin_unlock_irqrestore(&per_cpu(schedule_data, cpu).schedule_lock, flags);
+}
+
+static void
+csched_acct(void)
+{
+ unsigned long flags;
+ struct list_head *iter_vcpu, *next_vcpu;
+ struct list_head *iter_sdom, *next_sdom;
+ struct csched_vcpu *svc;
+ struct csched_vm *sdom;
+ uint32_t credit_total;
+ uint32_t weight_total;
+ uint32_t weight_left;
+ uint32_t credit_fair;
+ uint32_t credit_peak;
+ uint32_t credit_cap;
+ int credit_balance;
+ int credit_xtra;
+ int credit;
+
+
+ spin_lock_irqsave(&csched_priv.lock, flags);
+
+ weight_total = csched_priv.weight;
+ credit_total = csched_priv.credit;
+
+ /* Converge balance towards 0 when it drops negative */
+ if ( csched_priv.credit_balance < 0 )
+ {
+ credit_total -= csched_priv.credit_balance;
+ CSCHED_STAT_CRANK(acct_balance);
+ }
+
+ if ( unlikely(weight_total == 0) )
+ {
+ csched_priv.credit_balance = 0;
+ spin_unlock_irqrestore(&csched_priv.lock, flags);
+ CSCHED_STAT_CRANK(acct_no_work);
+ return;
+ }
+
+ CSCHED_STAT_CRANK(acct_run);
+
+ weight_left = weight_total;
+ credit_balance = 0;
+ credit_xtra = 0;
+ credit_cap = 0U;
+
+ list_for_each_safe( iter_sdom, next_sdom, &csched_priv.active_sdom )
+ {
+ sdom = list_entry(iter_sdom, struct csched_vm, active_sdom_elem);
+
+ BUG_ON( is_idle_vm(sdom->vm) );
+ BUG_ON( sdom->active_vcpu_count == 0 );
+ BUG_ON( sdom->weight == 0 );
+ BUG_ON( sdom->weight > weight_left );
+
+ weight_left -= sdom->weight;
+
+ /*
+ * A domain's fair share is computed using its weight in competition
+ * with that of all other active domains.
+ *
+ * At most, a domain can use credits to run all its active VCPUs
+ * for one full accounting period. We allow a domain to earn more
+ * only when the system-wide credit balance is negative.
+ */
+ credit_peak = sdom->active_vcpu_count * CSCHED_CREDITS_PER_ACCT;
+ if ( csched_priv.credit_balance < 0 )
+ {
+ credit_peak += ( ( -csched_priv.credit_balance * sdom->weight) +
+ (weight_total - 1)
+ ) / weight_total;
+ }
+
+ if ( sdom->cap != 0U )
+ {
+ credit_cap = ((sdom->cap * CSCHED_CREDITS_PER_ACCT) + 99) / 100;
+ if ( credit_cap < credit_peak )
+ credit_peak = credit_cap;
+
+ credit_cap = ( credit_cap + ( sdom->active_vcpu_count - 1 )
+ ) / sdom->active_vcpu_count;
+ }
+
+ credit_fair = ( ( credit_total * sdom->weight) + (weight_total - 1)
+ ) / weight_total;
+
+ if ( credit_fair < credit_peak )
+ {
+ credit_xtra = 1;
+ }
+ else
+ {
+ if ( weight_left != 0U )
+ {
+ /* Give other domains a chance at unused credits */
+ credit_total += ( ( ( credit_fair - credit_peak
+ ) * weight_total
+ ) + ( weight_left - 1 )
+ ) / weight_left;
+ }
+
+ if ( credit_xtra )
+ {
+ /*
+ * Lazily keep domains with extra credits at the head of
+ * the queue to give others a chance at them in future
+ * accounting periods.
+ */
+ CSCHED_STAT_CRANK(acct_reorder);
+ list_del(&sdom->active_sdom_elem);
+ list_add(&sdom->active_sdom_elem, &csched_priv.active_sdom);
+ }
+
+ credit_fair = credit_peak;
+ }
+
+ /* Compute fair share per VCPU */
+ credit_fair = ( credit_fair + ( sdom->active_vcpu_count - 1 )
+ ) / sdom->active_vcpu_count;
+
+
+ list_for_each_safe( iter_vcpu, next_vcpu, &sdom->active_vcpu )
+ {
+ svc = list_entry(iter_vcpu, struct csched_vcpu, active_vcpu_elem);
+ BUG_ON( sdom != svc->sdom );
+
+ /* Increment credit */
+ atomic_add(credit_fair, &svc->credit);
+ credit = atomic_read(&svc->credit);
+
+ /*
+ * Recompute priority or, if VCPU is idling, remove it from
+ * the active list.
+ */
+ if ( credit < 0 )
+ {
+ svc->pri = CSCHED_PRI_TS_OVER;
+
+ /* Park running VCPUs of capped-out domains */
+ if ( sdom->cap != 0U &&
+ credit < -credit_cap &&
+ !(svc->flags & CSCHED_FLAG_VCPU_PARKED) )
+ {
+ CSCHED_STAT_CRANK(vcpu_park);
+ vcpu_pause_nosync(svc->vcpu);
+ svc->flags |= CSCHED_FLAG_VCPU_PARKED;
+ }
+
+ /* Lower bound on credits */
+ if ( credit < -CSCHED_CREDITS_PER_TSLICE )
+ {
+ CSCHED_STAT_CRANK(acct_min_credit);
+ credit = -CSCHED_CREDITS_PER_TSLICE;
+ atomic_set(&svc->credit, credit);
+ }
+ }
+ else
+ {
+ svc->pri = CSCHED_PRI_TS_UNDER;
+
+ /* Unpark any capped domains whose credits go positive */
+ if ( svc->flags & CSCHED_FLAG_VCPU_PARKED)
+ {
+ /*
+ * It's important to unset the flag AFTER the unpause()
+ * call to make sure the VCPU's priority is not boosted
+ * if it is woken up here.
+ */
+ CSCHED_STAT_CRANK(vcpu_unpark);
+ vcpu_unpause(svc->vcpu);
+ svc->flags &= ~CSCHED_FLAG_VCPU_PARKED;
+ }
+
+ /* Upper bound on credits means VCPU stops earning */
+ if ( credit > CSCHED_CREDITS_PER_TSLICE )
+ {
+ __csched_vcpu_acct_stop_locked(svc);
+ credit = 0;
+ atomic_set(&svc->credit, credit);
+ }
+ }
+
+ CSCHED_VCPU_STAT_SET(svc, credit_last, credit);
+ CSCHED_VCPU_STAT_SET(svc, credit_incr, credit_fair);
+ credit_balance += credit;
+ }
+ }
+
+ csched_priv.credit_balance = credit_balance;
+
+ spin_unlock_irqrestore(&csched_priv.lock, flags);
+
+ /* Inform each CPU that its runq needs to be sorted */
+ csched_priv.runq_sort++;
+}
+
+/* the return value indicates whether this routine has enabled the pseudo_irq
+ * Return Value:
+ * 0
+ */
+static void _csched_tick(unsigned long _unused)
+{
+ int my_cpu = get_cpu();
+ struct csched_pcpu *spc = CSCHED_PCPU(my_cpu);
+ unsigned int cpu = spc->cpu;
+ ktime_t now;
+
+ /* if we fail to lock, it indicates the KVM modules are going to
+ * be unloaded
+ */
+ if(cmpxchg(&spc->in_use, false, true) != false) {
+ put_cpu();
+ return;
+ }
+
+ BUG_ON(thread_preemptible());
+ BUG_ON(cpu != my_cpu);
+
+ TRACE_2D(TRC_INTERNAL_CSCHED_TICK, my_cpu, __LINE__);
+ // tasklet_disable(&per_cpu(schedule_data, my_cpu).sched_tasklet);
+try_again:
+ while(per_cpu(schedule_data, my_cpu).sched_state == SCHEDULER_USER)
+ ;
+ if(per_cpu(schedule_data, my_cpu).sched_state == SCHEDULER_KERNEL) {
+ /* this indicates that someone else has entered the
critical region now
+ * we should not continue
+ */
+ TRACE_2D(TRC_INTERNAL_CSCHED_TICK, my_cpu, __LINE__);
+ if(!shutting_down) {
+ hrtimer_cancel(&spc->ticker);
+ now = spc->ticker.base->get_time();
+ hrtimer_start(&spc->ticker,
+ ktime_add_ns(now, 10000),
+ HRTIMER_MODE_ABS);
+ }
+ // tasklet_enable(&per_cpu(schedule_data, my_cpu).sched_tasklet);
+ spc->in_use = false;
+ put_cpu();
+ return;
+ }
+ if (cmpxchg(&per_cpu(schedule_data, my_cpu).sched_state,
+ SCHEDULER_FREE, SCHEDULER_KERNEL) != SCHEDULER_FREE)
+ goto try_again;
+
+ TRACE_2D(TRC_INTERNAL_CSCHED_TICK, my_cpu, __LINE__);
+
+ spc->tick++;
+
+ /*
+ * Accounting for running VCPU
+ */
+ if ( !is_idle_vcpu(current_vcpu) )
+ csched_vcpu_acct(cpu);
+
+ TRACE_2D(TRC_INTERNAL_CSCHED_TICK, my_cpu, __LINE__);
+ /*
+ * Host-wide accounting duty
+ *
+ * Note: Currently, this is always done by the master boot
CPU. Eventually,
+ * we could distribute or at the very least cycle the duty.
+ */
+ if ( (csched_priv.master == cpu) &&
+ (spc->tick % CSCHED_TICKS_PER_ACCT) == 0 )
+ {
+ csched_acct();
+ }
+
+ TRACE_2D(TRC_INTERNAL_CSCHED_TICK, my_cpu, __LINE__);
+ /*
+ * Check if runq needs to be sorted
+ *
+ * Every physical CPU resorts the runq after the accounting master has
+ * modified priorities. This is a special O(n) sort and runs at most
+ * once per accounting period (currently 30 milliseconds).
+ */
+ csched_runq_sort(cpu);
+ TRACE_2D(TRC_INTERNAL_CSCHED_TICK, my_cpu, __LINE__);
+
+ if (!shutting_down) {
+ now = spc->ticker.base->get_time();
+ hrtimer_start(&spc->ticker,
+ ktime_add_ns(now, MILLISECS(CSCHED_MSECS_PER_TICK)),
+ HRTIMER_MODE_ABS);
+ }
+ per_cpu(schedule_data, my_cpu).sched_state = SCHEDULER_FREE;
+ // tasklet_enable(&per_cpu(schedule_data, my_cpu).sched_tasklet);
+ TRACE_2D(TRC_INTERNAL_CSCHED_TICK, my_cpu, __LINE__);
+ spc->in_use = false;
+ put_cpu();
+}
+
+/* NOTICE:
+ * Linux can not guarrentee the timer set on one cpu will be destined
+ * to be triggered on that cpu. That is to say: one cpu might receive
+ * other cpu's timer expire message. In this case, the message should
+ * be relayed to the corresponding cpu. This is done via an IPI-tasklet
+ * --Alex
+ */
+static enum hrtimer_restart csched_tick_stub(struct hrtimer *data)
+{
+ u64 delta;
+ int cur_cpu = get_cpu();
+ struct csched_pcpu *spc = container_of(data, struct
csched_pcpu, ticker);
+ BUG_ON(!data);
+ BUG_ON(!spc);
+
+ /* if the ticker does not belong to this cpu, relay it */
+ if(spc->cpu != cur_cpu){
+ int r;
+ r = insert_pending_ipi(cur_cpu, cpumask_of_cpu(spc->cpu),
csched_tick_stub, (void*)data, 1);
+ if (r) {
+ dump_traces(NULL);
+ BUG_ON(1);
+ }
+ TRACE_3D(TRC_CSCHED_TICK,cur_cpu, spc->cpu, __LINE__);
+ wake_up(&per_cpu(schedule_data, cur_cpu).ipi_wq);
+ }else {
+ TRACE_3D(TRC_CSCHED_TICK,cur_cpu, spc->cpu, __LINE__);
+ tasklet_schedule(&per_cpu(schedule_data, cur_cpu).tick_tasklet);
+ }
+
+ /* the timer will be restarted by the tick_tasklet */
+ /* to add hrtimer_cancel() here!*/
+ put_cpu();
+ TRACE_2D(TRC_CSCHED_TICK, cur_cpu, __LINE__);
+ return HRTIMER_NORESTART;
+}
+
+static struct csched_vcpu *
+csched_runq_steal(int peer_cpu, int cpu, int pri)
+{
+ const struct csched_pcpu * const peer_pcpu = CSCHED_PCPU(peer_cpu);
+ const struct kvm_vcpu * const peer_vcpu = per_cpu(schedule_data,
peer_cpu).curr;
+ struct csched_vcpu *speer;
+ struct list_head *iter;
+ struct kvm_vcpu *vc;
+
+ /*
+ * Don't steal from an idle CPU's runq because it's about to
+ * pick up work from it itself.
+ */
+ if ( peer_pcpu != NULL && !is_idle_vcpu(peer_vcpu) )
+ {
+ list_for_each( iter, &peer_pcpu->runq )
+ {
+ speer = __runq_elem(iter);
+
+ /*
+ * If next available VCPU here is not of strictly higher
+ * priority than ours, this PCPU is useless to us.
+ */
+ if ( speer->pri <= pri )
+ break;
+
+ /* Is this VCPU is runnable on our PCPU? */
+ vc = speer->vcpu;
+ BUG_ON( is_idle_vcpu(vc) );
+
+ if (__csched_vcpu_is_migrateable(vc, cpu))
+ {
+ /* We got a candidate. Grab it! */
+ CSCHED_VCPU_STAT_CRANK(speer, migrate_q);
+ CSCHED_STAT_CRANK(migrate_queued);
+ __runq_remove(speer);
+ printk("setting thread %d's cpuaffinity to %d!\n",
+ vc->thread->pid, cpu);
+ kvm_sched_setaffinity(vc->thread->pid, cpumask_of_cpu(cpu));
+ vc->processor= cpu;
+ return speer;
+ }
+ }
+ }
+
+ CSCHED_STAT_CRANK(steal_peer_idle);
+ return NULL;
+}
+
+static struct csched_vcpu *
+csched_load_balance(int cpu, struct csched_vcpu *snext)
+{
+ struct csched_vcpu *speer;
+ cpumask_t workers;
+ int peer_cpu;
+
+ BUG_ON( cpu != snext->vcpu->processor);
+
+ if ( snext->pri == CSCHED_PRI_IDLE )
+ CSCHED_STAT_CRANK(load_balance_idle);
+ else if ( snext->pri == CSCHED_PRI_TS_OVER )
+ CSCHED_STAT_CRANK(load_balance_over);
+ else
+ CSCHED_STAT_CRANK(load_balance_other);
+
+ /*
+ * Peek at non-idling CPUs in the system, starting with our
+ * immediate neighbour.
+ */
+ cpus_andnot(workers, cpu_online_map, csched_priv.idlers);
+ cpu_clear(cpu, workers);
+ peer_cpu = cpu;
+
+ while ( !cpus_empty(workers) )
+ {
+ peer_cpu = __cycle_cpu(peer_cpu, &workers);
+ cpu_clear(peer_cpu, workers);
+
+ /*
+ * Get ahold of the scheduler lock for this peer CPU.
+ *
+ * Note: We don't spin on this lock but simply try it. Spinning could
+ * cause a deadlock if the peer CPU is also load balancing and trying
+ * to lock this CPU.
+ */
+ if ( !spin_trylock(&per_cpu(schedule_data, peer_cpu).schedule_lock) )
+ {
+ CSCHED_STAT_CRANK(steal_trylock_failed);
+ continue;
+ }
+ TRACE_3D(TRC_LOAD_BALANCE, peer_cpu, raw_smp_processor_id(), __LINE__);
+
+
+ /*
+ * Any work over there to steal?
+ */
+ speer = csched_runq_steal(peer_cpu, cpu, snext->pri);
+ spin_unlock(&per_cpu(schedule_data, peer_cpu).schedule_lock);
+ TRACE_3D(TRC_LOAD_BALANCE, cpu, peer_cpu, __LINE__);
+ if ( speer != NULL )
+ return speer;
+ }
+
+ TRACE_3D(TRC_LOAD_BALANCE, cpu, peer_cpu, __LINE__);
+ /* Failed to find more important work elsewhere... */
+ __runq_remove(snext);
+ TRACE_3D(TRC_LOAD_BALANCE, cpu, peer_cpu, __LINE__);
+ return snext;
+}
+
+/*
+ * This function is in the critical path. It is designed to be simple and
+ * fast for the common case.
+ */
+static struct task_slice
+csched_schedule(s_time_t now)
+{
+ const int cpu = smp_processor_id();
+ struct list_head * const runq = RUNQ(cpu);
+ struct csched_vcpu * const scurr = CSCHED_VCPU(current_vcpu);
+ struct csched_vcpu *snext;
+ struct task_slice ret;
+
+ if(shutting_down) {
+ ret.time = MILLISECS(CSCHED_MSECS_PER_TSLICE);
+ ret.task = idle_vm_kvm->vcpus[cpu];
+ return ret;
+ }
+
+ CSCHED_STAT_CRANK(schedule);
+ CSCHED_VCPU_CHECK(current_vcpu);
+
+ /*
+ * Select next runnable local VCPU (ie top of local runq)
+ */
+ if ( vcpu_runnable(current_vcpu) )
+ __runq_insert(cpu, scurr);
+ else
+ BUG_ON( is_idle_vcpu(current_vcpu) || list_empty(runq) );
+
+ snext = __runq_elem(runq->next);
+
+ /*
+ * SMP Load balance:
+ *
+ * If the next highest priority local runnable VCPU has already eaten
+ * through its credits, look on other PCPUs to see if we have more
+ * urgent work... If not, csched_load_balance() will return snext, but
+ * already removed from the runq.
+ */
+ if ( snext->pri > CSCHED_PRI_TS_OVER )
+ __runq_remove(snext);
+ else
+ snext = csched_load_balance(cpu, snext);
+ TRACE_3D(TRC_CSCHED_SCHEDULE, cpu, snext, __LINE__);
+
+ /*
+ * Update idlers mask if necessary. When we're idling, other CPUs
+ * will tickle us when they get extra work.
+ */
+ if ( snext->pri == CSCHED_PRI_IDLE )
+ {
+ if ( !cpu_isset(cpu, csched_priv.idlers) )
+ cpu_set(cpu, csched_priv.idlers);
+ }
+ else if ( cpu_isset(cpu, csched_priv.idlers) )
+ {
+ cpu_clear(cpu, csched_priv.idlers);
+ }
+
+ /*
+ * Return task to run next...
+ */
+ ret.time = MILLISECS(CSCHED_MSECS_PER_TSLICE);
+ ret.task = snext->vcpu;
+
+ CSCHED_VCPU_CHECK(ret.task);
+ return ret;
+}
+
+static void
+csched_dump_vcpu(struct csched_vcpu *svc)
+{
+ printk("func %s unimplemented\n", __FUNCTION__);
+}
+
+static void
+csched_dump_pcpu(int cpu)
+{
+ struct list_head *runq, *iter;
+ struct csched_pcpu *spc;
+ struct csched_vcpu *svc;
+ int loop;
+
+ spc = CSCHED_PCPU(cpu);
+ runq = &spc->runq;
+
+ printk(" sort=%d, sibling=0x%lx, core=0x%lx\n",
+ spc->runq_sort_last,
+ per_cpu(cpu_sibling_map,cpu).bits[0],
+ per_cpu(cpu_core_map,cpu).bits[0]);
+
+ /* current VCPU */
+ svc = CSCHED_VCPU(per_cpu(schedule_data, cpu).curr);
+ if ( svc )
+ {
+ printk("\trun: ");
+ csched_dump_vcpu(svc);
+ }
+
+ loop = 0;
+ list_for_each( iter, runq )
+ {
+ svc = __runq_elem(iter);
+ if ( svc )
+ {
+ printk("\t%3d: ", ++loop);
+ csched_dump_vcpu(svc);
+ }
+ }
+}
+
+static void
+csched_dump(void)
+{
+ struct list_head *iter_sdom, *iter_svc;
+ int loop;
+
+ printk("info:\n"
+ "\tncpus = %u\n"
+ "\tmaster = %u\n"
+ "\tcredit = %u\n"
+ "\tcredit balance = %d\n"
+ "\tweight = %u\n"
+ "\trunq_sort = %u\n"
+ "\tdefault-weight = %d\n"
+ "\tmsecs per tick = %dms\n"
+ "\tcredits per tick = %d\n"
+ "\tticks per tslice = %d\n"
+ "\tticks per acct = %d\n",
+ csched_priv.ncpus,
+ csched_priv.master,
+ csched_priv.credit,
+ csched_priv.credit_balance,
+ csched_priv.weight,
+ csched_priv.runq_sort,
+ CSCHED_DEFAULT_WEIGHT,
+ CSCHED_MSECS_PER_TICK,
+ CSCHED_CREDITS_PER_TICK,
+ CSCHED_TICKS_PER_TSLICE,
+ CSCHED_TICKS_PER_ACCT);
+
+ printk("idlers: 0x%lx\n", csched_priv.idlers.bits[0]);
+
+ CSCHED_STATS_PRINTK();
+
+ printk("active vcpus:\n");
+ loop = 0;
+ list_for_each( iter_sdom, &csched_priv.active_sdom )
+ {
+ struct csched_vm *sdom;
+ sdom = list_entry(iter_sdom, struct csched_vm, active_sdom_elem);
+
+ list_for_each( iter_svc, &sdom->active_vcpu )
+ {
+ struct csched_vcpu *svc;
+ svc = list_entry(iter_svc, struct csched_vcpu, active_vcpu_elem);
+
+ printk("\t%3d: ", ++loop);
+ csched_dump_vcpu(svc);
+ }
+ }
+}
+static void csched_stop_schedule(int cpu)
+{
+ struct csched_pcpu *spc;
+ spc = CSCHED_PCPU(cpu);
+ if(!spc) {
+ printk("func %s line %d cpu %d\n",
+ __FUNCTION__, __LINE__, cpu);
+ BUG_ON(1);
+ }
+ hrtimer_cancel(&spc->ticker);
+}
+
+
+/* Tickers cannot be kicked until SMP subsystem is alive. */
+int csched_start(int cpu)
+{
+ struct csched_pcpu *spc;
+ ktime_t now;
+
+ /* Is the credit scheduler initialised? */
+ if ( csched_priv.ncpus == 0 ){
+ printk("(ERROR) You should not see func %s line %d\n",
+ __FUNCTION__, __LINE__);
+ BUG_ON(1);
+ return 1;
+ }
+
+ tasklet_init(&per_cpu(schedule_data, cpu).tick_tasklet,
_csched_tick, (unsigned long)NULL);
+ spc = CSCHED_PCPU(cpu);
+ if(!spc) {
+ printk("func %s line %d cpu %d\n",
+ __FUNCTION__, __LINE__, cpu);
+ BUG_ON(1);
+ return 1;
+ }
+
+ now = spc->ticker.base->get_time();
+
+ hrtimer_start(&spc->ticker,
+ ktime_add_ns(now, MILLISECS(CSCHED_MSECS_PER_TICK)),
+ HRTIMER_MODE_ABS);
+
+ return 0;
+}
+static void
+csched_init(void)
+{
+ spin_lock_init(&csched_priv.lock);
+ INIT_LIST_HEAD(&csched_priv.active_sdom);
+ csched_priv.ncpus = 0;
+ csched_priv.master = UINT_MAX;
+ cpus_clear(csched_priv.idlers);
+ csched_priv.weight = 0U;
+ csched_priv.credit = 0U;
+ csched_priv.credit_balance = 0;
+ csched_priv.runq_sort = 0U;
+ CSCHED_STATS_RESET();
+}
+
+#define SCHEDULER_CREDIT 1
+struct scheduler sched_credit_def = {
+ .name = "SMP Credit Scheduler",
+ .opt_name = "credit",
+ .sched_id = SCHEDULER_CREDIT,
+
+ .init_vm = csched_vm_init,
+ .destroy_vm = csched_vm_destroy,
+
+ .init_vcpu = csched_vcpu_init,
+ .destroy_vcpu = csched_vcpu_destroy,
+
+ .sleep = csched_vcpu_sleep,
+ .wake = csched_vcpu_wake,
+
+ .read_schedule_info = csched_read_param,
+ .write_schedule_info = csched_write_param,
+
+ .pick_cpu = csched_cpu_pick,
+ .do_schedule = csched_schedule,
+
+ .disable_scheduler = csched_disable,
+ .start_scheduler = csched_start,
+ .stop_schedule = csched_stop_schedule,
+
+ .dump_cpu_state = csched_dump_pcpu,
+ .dump_settings = csched_dump,
+ .init = csched_init,
+};
diff --git a/arch/x86/kvm/schedule.c b/arch/x86/kvm/schedule.c
new file mode 100644
index 0000000..d6938d5
--- /dev/null
+++ b/arch/x86/kvm/schedule.c
@@ -0,0 +1,959 @@
+/****************************************************************************
+ * (C) 2002-2003 - Rolf Neugebauer - Intel Research Cambridge
+ * (C) 2002-2003 University of Cambridge
+ * (C) 2004 - Mark Williamson - Intel Research Cambridge
+ * (C) 2009 - Xiaojian Liu
+ ****************************************************************************
+ *
+ * File: common/schedule.c
+ * Author: Rolf Neugebauer & Keir Fraser
+ * Updated for generic API by Mark Williamson
+ * Updated for KVM by Xiaojian Liu
+ *
+ * Description: Generic CPU scheduling code
+ * implements support functionality for the Xen scheduler API.
+ *
+ */
+#include <linux/kvm_host.h>
+#include <linux/spinlock.h>
+#include <linux/trace.h>
+#include <linux/sched-if.h>
+#include <linux/ipi.h>
+#include <linux/proc_fs.h>
+#include <linux/uaccess.h>
+#include <linux/kthread.h>
+
+#ifdef DEBUG
+#define ASSERT(x) \
+do { \
+ if (!(x)) { \
+ printk(KERN_EMERG "assertion failed %s: %d: %s\n", \
+ __FILE__, __LINE__, #x); \
+ BUG(); \
+ } \
+} while (0)
+#else
+#define ASSERT(x) do { } while (0)
+#endif
+
+#if 0
+long (*sched_setaffinity_p)(pid_t pid, cpumask_t new_mask);
+#else
+extern long (*sched_setaffinity_p)(pid_t pid, cpumask_t* in_mask);
+#endif
+// EXPORT_SYMBOL_GPL(sched_setaffinity_p);
+
+DEFINE_PER_CPU(struct schedule_data, schedule_data);
+DEFINE_PER_CPU(rwlock_t, pseudo_cli);
+
+static struct list_head sched_vm_list;
+static spinlock_t sched_vm_list_lock;
+
+#define KVM_SCHEDULER "kvm"
+
+struct proc_dir_entry *kvm_scheduler_root;
+
+#ifndef COMPAT
+
+/* Various timer handlers. */
+static enum hrtimer_restart s_timer_fn(struct hrtimer* timer);
+static void vcpu_singleshot_timer_fn(void *data);
+static void poll_timer_fn(void *data);
+
+/* This is global for now so that private implementations can reach it */
+
+extern struct scheduler sched_credit_def;
+static struct scheduler *schedulers[] = {
+ &sched_credit_def,
+ NULL
+};
+
+static struct scheduler ops;
+
+#define SCHED_OP(fn, ...) \
+ (( ops.fn != NULL ) ? ops.fn( __VA_ARGS__ ) \
+ : (typeof(ops.fn(__VA_ARGS__)))0 )
+
+void vcpu_pause(struct kvm_vcpu *v)
+{
+ ASSERT(v != current_vcpu);
+
+ atomic_inc(&v->pause_count);
+ vcpu_sleep_sync(v);
+}
+void vcpu_pause_nosync(struct kvm_vcpu *v)
+{
+ atomic_inc(&v->pause_count);
+ vcpu_sleep_nosync(v);
+}
+void vcpu_unpause(struct kvm_vcpu *v)
+{
+ if(atomic_dec_and_test(&v->pause_count))
+ vcpu_wake(v);
+}
+
+void vm_pause(struct kvm *kvm)
+{
+ int late_wait = -1;
+ int i;
+
+ for(i=0; i< KVM_MAX_VCPUS; i++) {
+ if(kvm->vcpus[i]) {
+ if(current_vcpu == kvm->vcpus[i]) {
+ atomic_inc(&kvm->vcpus[i]->pause_count);
+ late_wait = i;
+ continue;
+ }
+ vcpu_sleep_sync(kvm->vcpus[i]);
+ }
+ }
+ if (late_wait != -1) {
+ if(current_vcpu == kvm->vcpus[late_wait]) {
+ if(!is_host_vm(kvm) && !is_idle_vm(kvm)) {
+ tasklet_schedule(&per_cpu(schedule_data,
raw_smp_processor_id()).sched_tasklet);
+ while(current_vcpu->kvm == kvm) {
+ printk("waiting\n");
+ schedule();
+ }
+ }
+ }
+ }
+}
+
+void vm_unpause(struct kvm *kvm)
+{
+ int i;
+ struct kvm_vcpu* v;
+ int cpu = get_cpu();
+
+ if ( atomic_dec_and_test(&kvm->pause_count) ){
+ for(i = 0; i < KVM_MAX_VCPUS; ++i) {
+ if(unlikely(i == cpu)) continue;
+ v = kvm->vcpus[i];
+ if (v) {
+ printk("me is %d waking vcpu %d\n", cpu, i);
+ vcpu_wake(v);
+ }
+ }
+ if(kvm->vcpus[cpu]){
+ printk("waking myself %d now\n", cpu);
+ vcpu_wake(kvm->vcpus[cpu]);
+ }
+ }
+ put_cpu();
+}
+void vm_pause_by_systemcontroller(struct kvm* kvm)
+{
+ vm_pause(kvm);
+ if(test_and_set_bool(kvm->is_paused_by_controller))
+ vm_unpause(kvm);
+}
+
+void vm_unpause_by_systemcontroller(struct kvm* kvm)
+{
+ if(test_and_clear_bool(kvm->is_paused_by_controller))
+ vm_unpause(kvm);
+
+}
+
+static inline void vcpu_runstate_change(
+ struct kvm_vcpu *v, int new_state, s_time_t new_entry_time)
+{
+ ASSERT(v->runstate.state != new_state);
+ ASSERT(spin_is_locked(&per_cpu(schedule_data,v->processor).schedule_lock));
+
+ v->runstate.time[v->runstate.state] +=
+ new_entry_time - v->runstate.state_entry_time;
+ v->runstate.state_entry_time = new_entry_time;
+ v->runstate.state = new_state;
+}
+
+void vcpu_runstate_get(struct kvm_vcpu *v, struct vcpu_runstate_info *runstate)
+{
+ if ( likely(v == current_vcpu) )
+ {
+ /* Fast lock-free path. */
+ memcpy(runstate, &v->runstate, sizeof(*runstate));
+ ASSERT(runstate->state == RUNSTATE_running);
+ runstate->time[RUNSTATE_running] += NOW() - runstate->state_entry_time;
+ }
+ else
+ {
+ unsigned long flags;
+ get_cpu();
+ local_irq_save(flags);
+ vcpu_schedule_lock_irq(v);
+ memcpy(runstate, &v->runstate, sizeof(*runstate));
+ runstate->time[runstate->state] += NOW() - runstate->state_entry_time;
+ vcpu_schedule_unlock_irq(v);
+ local_irq_restore(flags);
+ put_cpu();
+ }
+}
+
+int sched_init_vcpu(struct kvm_vcpu *v, unsigned int processor)
+{
+ int r;
+ struct kvm *kvm = v->kvm;
+
+ /*
+ * Initialize processor and affinity settings. The idler, and
potentially
+ * domain-0 VCPUs, are pinned onto their respective physical CPUs.
+ */
+ v->processor = processor;
+ printk("sched_setaffinity_p is %p\n", sched_setaffinity_p);
+ if (!is_host_vcpu(v) && !is_idle_vcpu(v))
+ kvm_sched_setaffinity(v->thread->pid, cpumask_of_cpu(processor));
+
+ if ( is_idle_vm(kvm) || is_host_vm(kvm))
+ v->cpu_affinity = cpumask_of_cpu(processor);
+ else
+ cpus_setall(v->cpu_affinity);
+
+ /* Idle VCPUs are scheduled immediately. */
+ if ( is_idle_vm(kvm) )
+ {
+ per_cpu(schedule_data, v->processor).curr = v;
+ per_cpu(schedule_data, v->processor).idle = v;
+ v->is_running = 1;
+ }
+
+ TRACE_1D(TRC_SCHED_VM_ADD, v->vcpu_id);
+ r = SCHED_OP(init_vcpu, v);
+ return r;
+}
+
+void sched_destroy_vcpu(struct kvm_vcpu *v)
+{
+ int cpu = get_cpu();
+ SCHED_OP(destroy_vcpu, v);
+ put_cpu();
+}
+struct kvm* get_kvm_by_id(int id)
+{
+
+ struct kvm *kvm, *n;
+ spin_lock(&sched_vm_list_lock);
+ list_for_each_entry_safe(kvm, n, &sched_vm_list, vm_link) {
+ if (kvm->vmid == id) {
+ spin_unlock(&sched_vm_list_lock);
+ return kvm;
+ }
+ }
+ return NULL;
+}
+static inline struct kvm* get_proc_kvm(struct file *file)
+{
+ struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode);
+ vmid_t id;
+
+ if(strcmp(pde->name, "host") == 0)
+ return host_vm_kvm;
+
+ sscanf(pde->name, "%d", &id);
+ return get_kvm_by_id(id);
+}
+
+#define MAX_SCHEDULE_PARAMS 50
+static int kvm_scheduler_write(struct file *file, const char __user *buffer,
+ unsigned long count, void *data)
+{
+ char scheduler_param[MAX_SCHEDULE_PARAMS] = { '\0'};
+ struct kvm* kvm = get_proc_kvm(file);
+ int i;
+ int r = count;
+ unsigned long flags;
+
+ get_cpu();
+
+ if(!kvm) {
+ r = -EINVAL;
+ goto quit;
+ }
+
+ if (count > MAX_SCHEDULE_PARAMS - 1 ) {
+ r = -EINVAL;
+ goto quit;
+ }
+ if(copy_from_user(scheduler_param, buffer, count)) {
+ r = -EFAULT;
+ goto quit;
+ }
+ scheduler_param[count] = '\0';
+
+ for(i=0; i < KVM_MAX_VCPUS; i++){
+ struct kvm_vcpu *v = kvm->vcpus[i];
+ if(v && v != current_vcpu)
+ vcpu_pause(v);
+ }
+
+ if ( kvm == current_vcpu->kvm) {
+ local_irq_save(flags);
+ if(!vcpu_schedule_try_lock(current_vcpu)) {
+ local_irq_restore(flags);
+ r = -EAGAIN;
+ goto quit_with_unpause;
+ }
+ }
+
+ if ( (SCHED_OP(write_schedule_info, kvm, scheduler_param)) == 0 )
+ printk("failed write schedule info!\n");
+
+ if ( kvm == current_vcpu->kvm) {
+ vcpu_schedule_unlock(current_vcpu);
+ local_irq_restore(flags);
+ }
+
+quit_with_unpause:
+ for(i=0; i < KVM_MAX_VCPUS; i++){
+ struct kvm_vcpu *v = kvm->vcpus[i];
+ if(v && v != current_vcpu)
+ vcpu_unpause(v);
+ }
+quit:
+ put_cpu();
+ return r;
+}
+
+static ssize_t kvm_scheduler_read(struct file *file,
+ char __user * buffer, size_t count, loff_t *ppos)
+{
+ int res;
+ struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode);
+ struct kvm *kvm;
+ int i;
+ char *msg;
+ unsigned long flags;
+
+ get_cpu();
+
+ msg = kmalloc(256, GFP_KERNEL);
+ if (!msg) {
+ res = -ENOMEM;
+ goto quit;
+ }
+
+ kvm = get_proc_kvm(file);
+ if(!kvm) {
+ sprintf(msg, "file /proc/%s/%s does not have a corresponding kvm!\n",
+ KVM_SCHEDULER, pde->name);
+ goto quit_with_msg;
+ }
+
+ for(i=0; i < KVM_MAX_VCPUS; i++){
+ struct kvm_vcpu *v = kvm->vcpus[i];
+ if(v && v != current_vcpu)
+ vcpu_pause(v);
+ }
+
+ if ( kvm == current_vcpu->kvm){
+ local_irq_save(flags);
+ if(!vcpu_schedule_try_lock(current_vcpu)) {
+ local_irq_restore(flags);
+ snprintf(msg, 256,
+ "scheduler is locked by current domain. Try again!\n");
+ goto quit_with_unpause;
+ }
+ }
+
+ if ( (SCHED_OP(read_schedule_info, kvm, msg, 256)) == 0 )
+ printk("failed read schedule info!\n");
+
+ if ( kvm == current_vcpu->kvm) {
+ vcpu_schedule_unlock(current_vcpu);
+ local_irq_restore(flags);
+ }
+
+quit_with_unpause:
+ for(i=0; i < KVM_MAX_VCPUS; i++){
+ struct kvm_vcpu *v = kvm->vcpus[i];
+ if(v && v != current_vcpu)
+ vcpu_unpause(v);
+ }
+
+quit_with_msg:
+ res = simple_read_from_buffer(buffer, count, ppos, msg, strlen(msg));
+ kfree(msg);
+quit:
+ put_cpu();
+ return res;
+
+}
+static const struct file_operations kvm_scheduler_ops = {
+ .read = kvm_scheduler_read,
+ .write = kvm_scheduler_write,
+};
+int sched_init_vm(struct kvm *kvm)
+{
+ struct proc_dir_entry *entry;
+ int error = 0;
+ char name[MAX_PROCESSID_LEN];
+ int r;
+ int cpu = get_cpu();
+
+ BUG_ON(kvm->vmid);
+
+ if(unlikely(is_idle_vm(kvm)))
+ kvm->vmid = IDLE_VM_ID;
+ else if(unlikely(is_host_vm(kvm)))
+ kvm->vmid = HOST_VM_ID;
+ else {
+ kvm->vmid = (unsigned long)current->pid;
+ if(kvm->vmid > MAX_PROCESSID) {
+ printk("process id %lx too big. only the lower 32bits are used!\n",
+ kvm->vmid);
+ kvm->vmid &= 0x0ffffffff;
+ }
+ }
+ spin_lock(&sched_vm_list_lock);
+ list_add(&kvm->vm_link, &sched_vm_list);
+ spin_unlock(&sched_vm_list_lock);
+
+ if(!is_idle_vm(kvm) && (kvm_scheduler_root != NULL)) {
+ if(!is_host_vm(kvm))
+ sprintf(name, "%d", kvm->vmid);
+ else
+ sprintf(name, "host");
+ entry = create_proc_entry(name, S_IRUGO | S_IWUGO | S_IFREG,
+ kvm_scheduler_root);
+ if (entry)
+ entry->proc_fops = &kvm_scheduler_ops;
+ else {
+ printk("failed to create vm %x's corresponding procfs\n",
kvm->vmid);
+ printk("user will not be able to control its scheduling!\n");
+ kvm->vmid = ANONY_VM_ID;
+ }
+ }
+
+ r = SCHED_OP(init_vm, kvm);
+ put_cpu();
+ return r;
+}
+
+void sched_destroy_vm(struct kvm *d)
+{
+ get_cpu();
+
+ char name[MAX_PROCESSID_LEN];
+
+ spin_lock(&sched_vm_list_lock);
+ list_del(&d->vm_link);
+ spin_unlock(&sched_vm_list_lock);
+
+ if(d->vmid){
+ if(is_host_vm(d))
+ sprintf(name, "host");
+ else
+ sprintf(name, "%d", d->vmid);
+ remove_proc_entry(name, kvm_scheduler_root);
+ }
+
+ SCHED_OP(destroy_vm, d);
+ put_cpu();
+}
+
+void vcpu_sleep_nosync(struct kvm_vcpu *v)
+{
+ unsigned long flags;
+ unsigned long my_flags;
+
+ get_cpu();
+ local_irq_save(my_flags);
+ vcpu_schedule_lock_irqsave(v, flags);
+
+ if ( likely(!vcpu_runnable(v)) )
+ {
+ if ( v->runstate.state == RUNSTATE_runnable )
+ vcpu_runstate_change(v, RUNSTATE_offline, NOW());
+
+ SCHED_OP(sleep, v);
+ }
+
+ vcpu_schedule_unlock_irqrestore(v, flags);
+ local_irq_restore(my_flags);
+ put_cpu();
+
+ TRACE_1D(TRC_SCHED_SLEEP, v->vcpu_id);
+}
+void sync_vcpu_execstate(struct kvm_vcpu *v)
+{
+ printk("TODO: func %s called. \n", __FUNCTION__);
+}
+void vcpu_sleep_sync(struct kvm_vcpu *v)
+{
+ vcpu_sleep_nosync(v);
+
+ while ( !vcpu_runnable(v) && v->is_running )
+ cpu_relax();
+
+ sync_vcpu_execstate(v);
+}
+
+void vcpu_wake(struct kvm_vcpu *v)
+{
+ unsigned long flags;
+ unsigned long my_flags;
+ int tmp_cpu = get_cpu();;
+ local_irq_save(my_flags);
+ vcpu_schedule_lock_irqsave(v, flags);
+ TRACE_2D(TRC_SCHED_WAKE, tmp_cpu, v);
+
+ if ( likely(vcpu_runnable(v)) )
+ {
+ if ( v->runstate.state >= RUNSTATE_blocked )
+ vcpu_runstate_change(v, RUNSTATE_runnable, NOW());
+ SCHED_OP(wake, v);
+ }
+ else if ( !test_bit(_VPF_blocked, &v->pause_flags) )
+ {
+ if ( v->runstate.state == RUNSTATE_blocked )
+ vcpu_runstate_change(v, RUNSTATE_offline, NOW());
+ }
+
+ vcpu_schedule_unlock_irqrestore(v, flags);
+ local_irq_restore(my_flags);
+
+ TRACE_3D(TRC_SCHED_WAKE, tmp_cpu, v->vcpu_id, __LINE__);
+ put_cpu();
+}
+static void vcpu_migrate(struct kvm_vcpu *v)
+{
+ unsigned long flags;
+ unsigned long my_flags;
+ int old_cpu;
+ int tmp_cpu = get_cpu();;
+
+ local_irq_save(my_flags);
+ vcpu_schedule_lock_irqsave(v, flags);
+
+ /*
+ * NB. Check of v->running happens /after/ setting migration flag
+ * because they both happen in (different) spinlock regions, and those
+ * regions are strictly serialised.
+ */
+ if ( v->is_running ||
+ !test_and_clear_bit(_VPF_migrating, &v->pause_flags) )
+ {
+ vcpu_schedule_unlock_irqrestore(v, flags);
+ local_irq_restore(my_flags);
+ put_cpu();
+ return;
+ }
+
+ /* Switch to new CPU, then unlock old CPU. */
+ old_cpu = v->processor;
+ v->processor = SCHED_OP(pick_cpu, v);
+
+ /* because v->cpu is changed,
+ *we should not use vcpu_schedule_unlock_restore()
+ */
+ BUG_ON(old_cpu != tmp_cpu);
+ spin_unlock(&per_cpu(schedule_data, old_cpu).schedule_lock);
+ pseudo_irq_restore(flags);
+ local_irq_restore(my_flags);
+ /* Wake on new CPU. */
+ vcpu_wake(v);
+ put_cpu();
+}
+
+/*
+ * Force a VCPU through a deschedule/reschedule path.
+ * For example, using this when setting the periodic timer period means that
+ * most periodic-timer state need only be touched from within the scheduler
+ * which can thus be done without need for synchronisation.
+ */
+void vcpu_force_reschedule(struct kvm_vcpu *v)
+{
+ printk("TODO: func %s called\n", __FUNCTION__);
+}
+
+int vcpu_set_affinity(struct kvm_vcpu *v, cpumask_t *affinity)
+{
+ printk("TODO: func %s called\n", __FUNCTION__);
+ return 0;
+}
+
+/* Block the currently-executing domain until a pertinent event occurs. */
+static long do_block(void)
+{
+ printk("TODO: func %s called\n", __FUNCTION__);
+ return 0;
+}
+
+/* Voluntarily yield the processor for this allocation. */
+static long do_yield(void)
+{
+ TRACE_1D(TRC_SCHED_YIELD, current_vcpu->vcpu_id);
+ tasklet_schedule(&per_cpu(schedule_data,
raw_smp_processor_id()).sched_tasklet);
+ return 0;
+}
+long do_sched_op_compat(int cmd, unsigned long arg)
+{
+ long ret = 0;
+ printk("TODO: unimplemented func %s \n", __FUNCTION__);
+ return ret;
+}
+
+typedef long ret_t;
+
+#endif /* !COMPAT */
+
+#ifndef COMPAT
+
+/* sched_id - fetch ID of current scheduler */
+int sched_id(void)
+{
+ return ops.sched_id;
+}
+
+
+void continue_running(struct kvm_vcpu *vcpu)
+{
+}
+void context_saved(struct kvm_vcpu *prev)
+{
+ prev->is_running = 0;
+ cmpxchg(&prev->status, VCPU_RUNNING, VCPU_YIELD);
+
+ if(unlikely(test_bit(_VPF_migrating, &prev->pause_flags))){
+ vcpu_migrate(prev);
+ }
+}
+void vm_context_switch(struct kvm_vcpu *prev, struct kvm_vcpu *next)
+{
+ context_saved(prev);
+ cmpxchg(&next->status, VCPU_YIELD, VCPU_RUNNING);
+ wake_up(&next->wq);
+}
+/* Return Value: 0
+ * meaning: upon completion, the pseudo_irq remains disabled
+ */
+static int __vcpu_schedule(int my_cpu)
+{
+ struct kvm_vcpu *prev = current_vcpu, *next = NULL;
+ s_time_t now;
+ ktime_t ktime_now;
+ struct schedule_data *sd;
+ struct task_slice next_slice;
+ s32 r_time; /* time for new dom to run */
+ static int cnt=0;
+ unsigned long tmp_flags;
+
+ BUG_ON(thread_preemptible());
+
+ ASSERT(!in_irq());
+
+ sd = &per_cpu(schedule_data, my_cpu);
+
+ /* if the kvm module is to be unloaded, the lock might fail */
+ if(cmpxchg(&sd->in_use, false, true) != false)
+ return 0;
+
+ spin_lock(&sd->schedule_lock);
+
+ hrtimer_cancel(&sd->s_timer);
+ ktime_now = sd->s_timer.base->get_time();
+ now = ktime_to_ns(ktime_now);
+
+ /* get policy-specific decision on scheduling... */
+ next_slice = ops.do_schedule(now);
+
+ r_time = next_slice.time;
+ next = next_slice.task;
+
+ sd->curr = next;
+
+ if(unlikely(shutting_down)) {
+ spin_unlock(&sd->schedule_lock);
+ goto finished;
+ }
+
+ if ( unlikely(prev == next) )
+ {
+ spin_unlock(&sd->schedule_lock);
+
+ continue_running(prev);
+ goto finished;
+ }
+
+ TRACE_1D(TRC_SCHED_SWITCH_INFPREV,
+ // prev->domain->domain_id,
+ now - prev->runstate.state_entry_time);
+ TRACE_2D(TRC_SCHED_SWITCH_INFNEXT,
+ // next->domain->domain_id,
+ (next->runstate.state == RUNSTATE_runnable) ?
+ (now - next->runstate.state_entry_time) : 0,
+ r_time);
+
+ ASSERT(prev->runstate.state == RUNSTATE_running);
+ vcpu_runstate_change(
+ prev,
+ (test_bit(_VPF_blocked, &prev->pause_flags) ? RUNSTATE_blocked :
+ (vcpu_runnable(prev) ? RUNSTATE_runnable : RUNSTATE_offline)),
+ now);
+
+ ASSERT(next->runstate.state != RUNSTATE_running);
+ vcpu_runstate_change(next, RUNSTATE_running, now);
+
+ ASSERT(!next->is_running);
+ next->is_running = 1;
+
+ spin_unlock(&sd->schedule_lock);
+
+ TRACE_3D(TRC_SCHED_SWITCH, __LINE__, prev->vcpu_id, next->vcpu_id);
+ vm_context_switch(prev, next);
+finished:
+ hrtimer_cancel(&sd->watchdog);
+ if (!shutting_down) {
+ hrtimer_start(&sd->s_timer, ktime_add_ns(ktime_now, r_time),
HRTIMER_MODE_ABS);
+
+ /* restart the watchdog */
+ ktime_now = sd->watchdog.base->get_time();
+ hrtimer_start(&sd->watchdog, ktime_add_ns(ktime_now, WATCHDOG_NS),
HRTIMER_MODE_ABS);
+ }
+ sd->in_use = false;
+ return 0;
+}
+/*
+ * The main function
+ * - deschedule the current domain (scheduler independent).
+ * - pick a new domain (scheduler dependent).
+ */
+static void vcpu_schedule(unsigned long _unused)
+{
+
+ int my_cpu = raw_smp_processor_id();
+ // tasklet_disable(&per_cpu(schedule_data, my_cpu).tick_tasklet);
+try_again:
+ while(per_cpu(schedule_data, my_cpu).sched_state == SCHEDULER_USER)
+ ;
+ if(per_cpu(schedule_data, my_cpu).sched_state == SCHEDULER_KERNEL) {
+ /* for the case that others has entered the critical section
+ * we abort scheduling the vcpus this time
+ */
+ struct schedule_data *sd = &per_cpu(schedule_data, my_cpu);
+ ktime_t now = sd->s_timer.base->get_time();
+
+ hrtimer_cancel(&sd->s_timer);
+ if(!shutting_down)
+ hrtimer_start(&sd->s_timer, ktime_add_ns(now, 10000),
HRTIMER_MODE_ABS);
+ // tasklet_enable(&per_cpu(schedule_data, my_cpu).tick_tasklet);
+ return;
+ }
+ if(cmpxchg(&per_cpu(schedule_data, my_cpu).sched_state,
+ SCHEDULER_FREE, SCHEDULER_KERNEL) != SCHEDULER_FREE)
+ goto try_again;
+
+ preempt_disable();
+ __vcpu_schedule(my_cpu);
+ per_cpu(schedule_data, my_cpu).sched_state = SCHEDULER_FREE;
+
+ // tasklet_enable(&per_cpu(schedule_data, my_cpu).tick_tasklet);
+ preempt_enable();
+ // put_cpu();
+
+ return;
+}
+
+void kvm_force_tasklet_schedule(void *_unused)
+{
+ get_cpu();
+ tasklet_schedule(&per_cpu(schedule_data,
raw_smp_processor_id()).sched_tasklet);
+ put_cpu();
+}
+/* The scheduler timer: force a run through the scheduler */
+static enum hrtimer_restart s_timer_fn(struct hrtimer* timer)
+{
+ int cpu = raw_smp_processor_id();
+ struct schedule_data *sd = container_of(timer, struct
schedule_data, s_timer);
+ if(cpu != sd->id){
+ int r;
+ r = insert_pending_ipi(cpu, cpumask_of_cpu(sd->id),
kvm_force_tasklet_schedule, NULL, 1);
+ // tasklet_schedule(&sd->ipi_tasklet);
+ wake_up(&sd->ipi_wq);
+ if(r){
+ dump_traces(NULL);
+ BUG_ON(1);
+ }
+ }else{
+ per_cpu(schedule_data, raw_smp_processor_id()).can_migrate =
!in_atomic_preempt_off();
+ tasklet_schedule(&per_cpu(schedule_data, cpu).sched_tasklet);
+ }
+ /* no need to restart the timer. the vcpu_schedule() will do it */
+ return HRTIMER_NORESTART;
+}
+static void per_cpu_stop_sched(int cpu)
+{
+ struct schedule_data* sd = &per_cpu(schedule_data, cpu);
+ hrtimer_cancel(&sd->s_timer);
+ hrtimer_cancel(&sd->watchdog);
+}
+static void per_cpu_kill_scheduler(int cpu)
+{
+ struct schedule_data* sd = &per_cpu(schedule_data, cpu);
+ tasklet_kill(&sd->sched_tasklet);
+
+}
+
+bool shutting_down = false;
+void stop_auto_schedule(void)
+{
+ int i;
+
+ shutting_down = true;
+ barrier();
+
+ for_each_online_cpu(i) {
+ struct schedule_data *sd = &per_cpu(schedule_data, i);
+
+ sd->ipi_quit = true;
+ wake_up(&sd->ipi_wq);
+ printk("waiting ipi helper stop!\n");
+ while(sd->ipi_quit) schedule();
+ printk("ipi helper stopped!\n");
+ SCHED_OP(stop_schedule, i);
+ per_cpu_stop_sched(i);
+
+ }
+}
+static inline void scheduler_disabled(int cpu)
+{
+ struct schedule_data *sd = &per_cpu(schedule_data, cpu);
+ per_cpu_kill_scheduler(cpu);
+}
+void wait_scheduler_stops(void)
+{
+ int i;
+ for_each_online_cpu(i) {
+ SCHED_OP(disable_scheduler, i);
+ scheduler_disabled(i);
+ }
+}
+void scheduler_start(void)
+{
+ int i;
+
+ for_each_online_cpu(i) {
+ if(SCHED_OP(start_scheduler, i)) {
+ printk("error start scheduler %d!\n", i);
+ BUG_ON(1);
+ }
+ }
+}
+void scheduler_destroy(void)
+{
+ int i;
+
+ if(kvm_scheduler_root)
+ remove_proc_entry("kvm", NULL);
+
+ for_each_online_cpu(i) {
+ destroy_pending_ipi_buf(i);
+ }
+}
+
+extern void inject_pending_ipi(void);
+static void per_cpu_init_sched(int cpu)
+{
+ struct pending_work * pd_work;
+ ktime_t now;
+ struct schedule_data *sd = &per_cpu(schedule_data, cpu);
+ struct task_struct *task;
+ char name[]="ipi_helper_0";
+
+ sd->sched_state = SCHEDULER_FREE;
+ tasklet_init(&sd->sched_tasklet, vcpu_schedule, (unsigned long)NULL);
+
+ name[strlen(name)-1] += cpu;
+ init_waitqueue_head(&sd->ipi_wq);
+ sd->ipi_quit = false;
+ task = kthread_run(inject_pending_ipi, cpu, name);
+ if(IS_ERR(task)){
+ printk("error creating help thread %d\n", cpu);
+ BUG_ON(1);
+ }
+ kvm_sched_setaffinity(task->pid, cpumask_of_cpu(cpu));
+
+ rwlock_init(&per_cpu(pseudo_cli, cpu));
+ spin_lock_init(&sd->schedule_lock);
+ sd->in_use = false;
+ sd->can_migrate = false;
+ sd->id = cpu;
+
+ /* the scheduler timer */
+ hrtimer_init(&sd->s_timer,
+ CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
+ sd->s_timer.function = s_timer_fn;
+ // hrtimer_data_pointer(&sd->s_timer);
+
+ /* the watchdog timer */
+ hrtimer_init(&sd->watchdog,
+ CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
+ sd->watchdog.function = dump_cpu_trace;
+ // hrtimer_data_pointer(&sd->watchdog);
+
+ /* start the watchdog */
+ now = sd->watchdog.base->get_time();
+ if(hrtimer_start(&sd->watchdog,
+ ktime_add_ns(now, WATCHDOG_NS), HRTIMER_MODE_ABS)) {
+ printk("start watchdog timer failed!\n");
+ BUG_ON(1);
+ }
+}
+
+/* Initialise the data structures. */
+void scheduler_init(void)
+{
+ int i;
+
+ INIT_LIST_HEAD(&sched_vm_list);
+ spin_lock_init(&sched_vm_list_lock);
+
+ kvm_scheduler_root = proc_mkdir(KVM_SCHEDULER, NULL);
+ if (!kvm_scheduler_root){
+ printk("fail to create the /proc/kvm procfs!\n");
+ printk("the user scheduler control function will be disabled!\n");
+ }
+
+
+ for_each_online_cpu(i)
+ per_cpu_init_sched(i);
+
+ ops = *schedulers[0];
+ SCHED_OP(init);
+}
+
+void dump_runq(unsigned char key)
+{
+ s_time_t now = NOW();
+ int i;
+ unsigned long flags;
+
+ local_irq_save(flags);
+
+ printk("Scheduler: %s (%s)\n", ops.name, ops.opt_name);
+ SCHED_OP(dump_settings);
+ printk("NOW=0x%08X%08X\n", (u32)(now>>32), (u32)now);
+
+ for_each_online_cpu ( i )
+ {
+ spin_lock(&per_cpu(schedule_data, i).schedule_lock);
+ printk("CPU[%02d] ", i);
+ SCHED_OP(dump_cpu_state, i);
+ spin_unlock(&per_cpu(schedule_data, i).schedule_lock);
+ }
+
+ local_irq_restore(flags);
+}
+
+#endif /* !COMPAT */
+
+/*
+ * Local variables:
+ * mode: C
+ * c-set-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
