/*
* Generic version of the affinity autoselector.
*/
static int
setup_affinity(unsigned int irq, struct irq_desc *desc, struct cpumask *mask)
{
struct cpumask *set = irq_default_affinity;
int node = desc->irq_data.node;
/* Excludes PER_CPU and NO_BALANCE interrupts */
if (!irq_can_set_affinity(irq))
return 0;
/*
* Preserve an userspace affinity setup, but make sure that
* one of the targets is online.
*/
if (irqd_has_set(&desc->irq_data, IRQD_AFFINITY_SET)) {
if (cpumask_intersects(desc->irq_data.affinity,
cpu_online_mask))
set = desc->irq_data.affinity;
else
irqd_clear(&desc->irq_data, IRQD_AFFINITY_SET);
}
cpumask_and(mask, cpu_online_mask, set);
if (node != NUMA_NO_NODE) {
const struct cpumask *nodemask = cpumask_of_node(node);
/* make sure at least one of the cpus in nodemask is online */
if (cpumask_intersects(mask, nodemask))
cpumask_and(mask, mask, nodemask);
}
irq_do_set_affinity(&desc->irq_data, mask, false);
return 0;
}
static ssize_t node_read_cpumap(struct device *dev, bool list, char *buf)
{
struct node *node_dev = to_node(dev);
const struct cpumask *mask = cpumask_of_node(node_dev->dev.id);
/* 2008/04/07: buf currently PAGE_SIZE, need 9 chars per 32 bits. */
BUILD_BUG_ON((NR_CPUS/32 * 9) > (PAGE_SIZE-1));
return cpumap_print_to_pagebuf(list, buf, mask);
}
static acpi_status __devinit
acpi_map_iosapic(acpi_handle handle, u32 depth, void *context, void **ret)
{
struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
union acpi_object *obj;
struct acpi_madt_io_sapic *iosapic;
unsigned int gsi_base;
int pxm, node;
/* Only care about objects w/ a method that returns the MADT */
if (ACPI_FAILURE(acpi_evaluate_object(handle, "_MAT", NULL, &buffer)))
return AE_OK;
if (!buffer.length || !buffer.pointer)
return AE_OK;
obj = buffer.pointer;
if (obj->type != ACPI_TYPE_BUFFER ||
obj->buffer.length < sizeof(*iosapic)) {
kfree(buffer.pointer);
return AE_OK;
}
iosapic = (struct acpi_madt_io_sapic *)obj->buffer.pointer;
if (iosapic->header.type != ACPI_MADT_TYPE_IO_SAPIC) {
kfree(buffer.pointer);
return AE_OK;
}
gsi_base = iosapic->global_irq_base;
kfree(buffer.pointer);
/*
* OK, it's an IOSAPIC MADT entry, look for a _PXM value to tell
* us which node to associate this with.
*/
pxm = acpi_get_pxm(handle);
if (pxm < 0)
return AE_OK;
node = pxm_to_node(pxm);
if (node >= MAX_NUMNODES || !node_online(node) ||
cpumask_empty(cpumask_of_node(node)))
return AE_OK;
/* We know a gsi to node mapping! */
map_iosapic_to_node(gsi_base, node);
return AE_OK;
}
/**
* _hfi1_schedule_send - schedule progress
* @qp: the QP
*
* This schedules qp progress w/o regard to the s_flags.
*
* It is only used in the post send, which doesn't hold
* the s_lock.
*/
void _hfi1_schedule_send(struct rvt_qp *qp)
{
struct hfi1_qp_priv *priv = qp->priv;
struct hfi1_ibport *ibp =
to_iport(qp->ibqp.device, qp->port_num);
struct hfi1_pportdata *ppd = ppd_from_ibp(ibp);
struct hfi1_devdata *dd = dd_from_ibdev(qp->ibqp.device);
iowait_schedule(&priv->s_iowait, ppd->hfi1_wq,
priv->s_sde ?
priv->s_sde->cpu :
cpumask_first(cpumask_of_node(dd->node)));
}
/* Update distances based on topology */
for_each_cpu(cpu, update_mask) {
if (cpu_rmap_copy_neigh(rmap, cpu,
topology_sibling_cpumask(cpu), 1))
continue;
if (cpu_rmap_copy_neigh(rmap, cpu,
topology_core_cpumask(cpu), 2))
continue;
if (cpu_rmap_copy_neigh(rmap, cpu,
cpumask_of_node(cpu_to_node(cpu)), 3))
continue;
/* We could continue into NUMA node distances, but for now
* we give up.
*/
}
static int local_cpus_show(int node)
{
const cpumask_t *mask;
int len = 1;
#ifdef CONFIG_NUMA
mask = cpumask_of_pcibus(node);
#else
mask = cpumask_of_node(node);
#endif
printf("mask: %ld\n", *mask);
return len;
}
static acpi_status __devinit
acpi_map_iosapic(acpi_handle handle, u32 depth, void *context, void **ret)
{
struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
union acpi_object *obj;
struct acpi_madt_io_sapic *iosapic;
unsigned int gsi_base;
int pxm, node;
if (ACPI_FAILURE(acpi_evaluate_object(handle, "_MAT", NULL, &buffer)))
return AE_OK;
if (!buffer.length || !buffer.pointer)
return AE_OK;
obj = buffer.pointer;
if (obj->type != ACPI_TYPE_BUFFER ||
obj->buffer.length < sizeof(*iosapic)) {
kfree(buffer.pointer);
return AE_OK;
}
iosapic = (struct acpi_madt_io_sapic *)obj->buffer.pointer;
if (iosapic->header.type != ACPI_MADT_TYPE_IO_SAPIC) {
kfree(buffer.pointer);
return AE_OK;
}
gsi_base = iosapic->global_irq_base;
kfree(buffer.pointer);
pxm = acpi_get_pxm(handle);
if (pxm < 0)
return AE_OK;
node = pxm_to_node(pxm);
if (node >= MAX_NUMNODES || !node_online(node) ||
cpumask_empty(cpumask_of_node(node)))
return AE_OK;
map_iosapic_to_node(gsi_base, node);
return AE_OK;
}
static ssize_t node_read_cpumap(struct device *dev, int type, char *buf)
{
struct node *node_dev = to_node(dev);
const struct cpumask *mask = cpumask_of_node(node_dev->dev.id);
int len;
/* 2008/04/07: buf currently PAGE_SIZE, need 9 chars per 32 bits. */
BUILD_BUG_ON((NR_CPUS/32 * 9) > (PAGE_SIZE-1));
len = type?
cpulist_scnprintf(buf, PAGE_SIZE-2, mask) :
cpumask_scnprintf(buf, PAGE_SIZE-2, mask);
buf[len++] = '\n';
buf[len] = '\0';
return len;
}
static acpi_status acpi_map_iosapic(acpi_handle handle, u32 depth,
void *context, void **ret)
{
struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
union acpi_object *obj;
struct acpi_madt_io_sapic *iosapic;
unsigned int gsi_base;
int node;
/* Only care about objects w/ a method that returns the MADT */
if (ACPI_FAILURE(acpi_evaluate_object(handle, "_MAT", NULL, &buffer)))
return AE_OK;
if (!buffer.length || !buffer.pointer)
return AE_OK;
obj = buffer.pointer;
if (obj->type != ACPI_TYPE_BUFFER ||
obj->buffer.length < sizeof(*iosapic)) {
kfree(buffer.pointer);
return AE_OK;
}
iosapic = (struct acpi_madt_io_sapic *)obj->buffer.pointer;
if (iosapic->header.type != ACPI_MADT_TYPE_IO_SAPIC) {
kfree(buffer.pointer);
return AE_OK;
}
gsi_base = iosapic->global_irq_base;
kfree(buffer.pointer);
/* OK, it's an IOSAPIC MADT entry; associate it with a node */
node = acpi_get_node(handle);
if (node == NUMA_NO_NODE || !node_online(node) ||
cpumask_empty(cpumask_of_node(node)))
return AE_OK;
/* We know a gsi to node mapping! */
map_iosapic_to_node(gsi_base, node);
return AE_OK;
}
static unsigned long max_pages(unsigned long min_pages)
{
unsigned long node_free_pages, max;
int node = numa_node_id();
struct zone *zones = NODE_DATA(node)->node_zones;
int num_cpus_on_node;
node_free_pages =
#ifdef CONFIG_ZONE_DMA
zone_page_state(&zones[ZONE_DMA], NR_FREE_PAGES) +
#endif
#ifdef CONFIG_ZONE_DMA32
zone_page_state(&zones[ZONE_DMA32], NR_FREE_PAGES) +
#endif
zone_page_state(&zones[ZONE_NORMAL], NR_FREE_PAGES);
max = node_free_pages / FRACTION_OF_NODE_MEM;
num_cpus_on_node = cpumask_weight(cpumask_of_node(node));
max /= num_cpus_on_node;
return max(max, min_pages);
}
static int
setup_affinity(unsigned int irq, struct irq_desc *desc, struct cpumask *mask)
{
struct irq_chip *chip = irq_desc_get_chip(desc);
struct cpumask *set = irq_default_affinity;
int ret, node = desc->irq_data.node;
if (!irq_can_set_affinity(irq))
return 0;
if (irqd_has_set(&desc->irq_data, IRQD_AFFINITY_SET)) {
if (cpumask_intersects(desc->irq_data.affinity,
cpu_online_mask))
set = desc->irq_data.affinity;
else
irqd_clear(&desc->irq_data, IRQD_AFFINITY_SET);
}
cpumask_and(mask, cpu_online_mask, set);
if (node != NUMA_NO_NODE) {
const struct cpumask *nodemask = cpumask_of_node(node);
if (cpumask_intersects(mask, nodemask))
cpumask_and(mask, mask, nodemask);
}
ret = chip->irq_set_affinity(&desc->irq_data, mask, false);
switch (ret) {
case IRQ_SET_MASK_OK:
cpumask_copy(desc->irq_data.affinity, mask);
case IRQ_SET_MASK_OK_NOCOPY:
irq_set_thread_affinity(desc);
}
return 0;
}
static const cpumask_t *
cpumask_of_pcibus(int node)
{
return (node == -1) ? &cpu_online_mask
: cpumask_of_node(node);
}
