summaryrefslogtreecommitdiff
path: root/kernel/irq
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2019-09-17 11:42:15 -0700
committerLinus Torvalds <torvalds@linux-foundation.org>2019-09-17 11:42:15 -0700
commita572ba63298d04e2c5178e2abd82d6bd6e5677e7 (patch)
treefd0eacfc05ccedb988505d28539d03919ac03349 /kernel/irq
parent258b16ec9a542d57e78f82e0af0e600bb4aec7fa (diff)
parent9cc5b7fba57919a7d96f89ce79ba99d1dd1af965 (diff)
downloadlwn-a572ba63298d04e2c5178e2abd82d6bd6e5677e7.tar.gz
lwn-a572ba63298d04e2c5178e2abd82d6bd6e5677e7.zip
Merge branch 'irq-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull core irq updates from Thomas Gleixner: "Updates from the irq departement: - Update the interrupt spreading code so it handles numa node with different CPU counts properly. - A large overhaul of the ARM GiCv3 driver to support new PPI and SPI ranges. - Conversion of all alloc_fwnode() users to use physical addresses instead of virtual addresses so the virtual addresses are not leaked. The physical address is sufficient to identify the associated interrupt chip. - Add support for Marvel MMP3, Amlogic Meson SM1 interrupt chips. - Enforce interrupt threading at compile time if RT is enabled. - Small updates and improvements all over the place" * 'irq-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (37 commits) irqchip/gic-v3-its: Fix LPI release for Multi-MSI devices irqchip/uniphier-aidet: Use devm_platform_ioremap_resource() irqdomain: Add the missing assignment of domain->fwnode for named fwnode irqchip/mmp: Coexist with GIC root IRQ controller irqchip/mmp: Mask off interrupts from other cores irqchip/mmp: Add missing chained_irq_{enter,exit}() irqchip/mmp: Do not use of_address_to_resource() to get mux regs irqchip/meson-gpio: Add support for meson sm1 SoCs dt-bindings: interrupt-controller: New binding for the meson sm1 SoCs genirq/affinity: Remove const qualifier from node_to_cpumask argument genirq/affinity: Spread vectors on node according to nr_cpu ratio genirq/affinity: Improve __irq_build_affinity_masks() irqchip: Remove dev_err() usage after platform_get_irq() irqchip: Add include guard to irq-partition-percpu.h irqchip/mmp: Do not call irq_set_default_host() on DT platforms irqchip/gic-v3-its: Remove the redundant set_bit for lpi_map irqchip/gic-v3: Add quirks for HIP06/07 invalid GICD_TYPER erratum 161010803 irqchip/gic: Skip DT quirks when evaluating IIDR-based quirks irqchip/gic-v3: Warn about inconsistent implementations of extended ranges irqchip/gic-v3: Add EPPI range support ...
Diffstat (limited to 'kernel/irq')
-rw-r--r--kernel/irq/affinity.c231
-rw-r--r--kernel/irq/irqdomain.c10
-rw-r--r--kernel/irq/manage.c2
3 files changed, 208 insertions, 35 deletions
diff --git a/kernel/irq/affinity.c b/kernel/irq/affinity.c
index 6fef48033f96..4d89ad4fae3b 100644
--- a/kernel/irq/affinity.c
+++ b/kernel/irq/affinity.c
@@ -7,6 +7,7 @@
#include <linux/kernel.h>
#include <linux/slab.h>
#include <linux/cpu.h>
+#include <linux/sort.h>
static void irq_spread_init_one(struct cpumask *irqmsk, struct cpumask *nmsk,
unsigned int cpus_per_vec)
@@ -94,6 +95,155 @@ static int get_nodes_in_cpumask(cpumask_var_t *node_to_cpumask,
return nodes;
}
+struct node_vectors {
+ unsigned id;
+
+ union {
+ unsigned nvectors;
+ unsigned ncpus;
+ };
+};
+
+static int ncpus_cmp_func(const void *l, const void *r)
+{
+ const struct node_vectors *ln = l;
+ const struct node_vectors *rn = r;
+
+ return ln->ncpus - rn->ncpus;
+}
+
+/*
+ * Allocate vector number for each node, so that for each node:
+ *
+ * 1) the allocated number is >= 1
+ *
+ * 2) the allocated numbver is <= active CPU number of this node
+ *
+ * The actual allocated total vectors may be less than @numvecs when
+ * active total CPU number is less than @numvecs.
+ *
+ * Active CPUs means the CPUs in '@cpu_mask AND @node_to_cpumask[]'
+ * for each node.
+ */
+static void alloc_nodes_vectors(unsigned int numvecs,
+ cpumask_var_t *node_to_cpumask,
+ const struct cpumask *cpu_mask,
+ const nodemask_t nodemsk,
+ struct cpumask *nmsk,
+ struct node_vectors *node_vectors)
+{
+ unsigned n, remaining_ncpus = 0;
+
+ for (n = 0; n < nr_node_ids; n++) {
+ node_vectors[n].id = n;
+ node_vectors[n].ncpus = UINT_MAX;
+ }
+
+ for_each_node_mask(n, nodemsk) {
+ unsigned ncpus;
+
+ cpumask_and(nmsk, cpu_mask, node_to_cpumask[n]);
+ ncpus = cpumask_weight(nmsk);
+
+ if (!ncpus)
+ continue;
+ remaining_ncpus += ncpus;
+ node_vectors[n].ncpus = ncpus;
+ }
+
+ numvecs = min_t(unsigned, remaining_ncpus, numvecs);
+
+ sort(node_vectors, nr_node_ids, sizeof(node_vectors[0]),
+ ncpus_cmp_func, NULL);
+
+ /*
+ * Allocate vectors for each node according to the ratio of this
+ * node's nr_cpus to remaining un-assigned ncpus. 'numvecs' is
+ * bigger than number of active numa nodes. Always start the
+ * allocation from the node with minimized nr_cpus.
+ *
+ * This way guarantees that each active node gets allocated at
+ * least one vector, and the theory is simple: over-allocation
+ * is only done when this node is assigned by one vector, so
+ * other nodes will be allocated >= 1 vector, since 'numvecs' is
+ * bigger than number of numa nodes.
+ *
+ * One perfect invariant is that number of allocated vectors for
+ * each node is <= CPU count of this node:
+ *
+ * 1) suppose there are two nodes: A and B
+ * ncpu(X) is CPU count of node X
+ * vecs(X) is the vector count allocated to node X via this
+ * algorithm
+ *
+ * ncpu(A) <= ncpu(B)
+ * ncpu(A) + ncpu(B) = N
+ * vecs(A) + vecs(B) = V
+ *
+ * vecs(A) = max(1, round_down(V * ncpu(A) / N))
+ * vecs(B) = V - vecs(A)
+ *
+ * both N and V are integer, and 2 <= V <= N, suppose
+ * V = N - delta, and 0 <= delta <= N - 2
+ *
+ * 2) obviously vecs(A) <= ncpu(A) because:
+ *
+ * if vecs(A) is 1, then vecs(A) <= ncpu(A) given
+ * ncpu(A) >= 1
+ *
+ * otherwise,
+ * vecs(A) <= V * ncpu(A) / N <= ncpu(A), given V <= N
+ *
+ * 3) prove how vecs(B) <= ncpu(B):
+ *
+ * if round_down(V * ncpu(A) / N) == 0, vecs(B) won't be
+ * over-allocated, so vecs(B) <= ncpu(B),
+ *
+ * otherwise:
+ *
+ * vecs(A) =
+ * round_down(V * ncpu(A) / N) =
+ * round_down((N - delta) * ncpu(A) / N) =
+ * round_down((N * ncpu(A) - delta * ncpu(A)) / N) >=
+ * round_down((N * ncpu(A) - delta * N) / N) =
+ * cpu(A) - delta
+ *
+ * then:
+ *
+ * vecs(A) - V >= ncpu(A) - delta - V
+ * =>
+ * V - vecs(A) <= V + delta - ncpu(A)
+ * =>
+ * vecs(B) <= N - ncpu(A)
+ * =>
+ * vecs(B) <= cpu(B)
+ *
+ * For nodes >= 3, it can be thought as one node and another big
+ * node given that is exactly what this algorithm is implemented,
+ * and we always re-calculate 'remaining_ncpus' & 'numvecs', and
+ * finally for each node X: vecs(X) <= ncpu(X).
+ *
+ */
+ for (n = 0; n < nr_node_ids; n++) {
+ unsigned nvectors, ncpus;
+
+ if (node_vectors[n].ncpus == UINT_MAX)
+ continue;
+
+ WARN_ON_ONCE(numvecs == 0);
+
+ ncpus = node_vectors[n].ncpus;
+ nvectors = max_t(unsigned, 1,
+ numvecs * ncpus / remaining_ncpus);
+ WARN_ON_ONCE(nvectors > ncpus);
+
+ node_vectors[n].nvectors = nvectors;
+
+ remaining_ncpus -= ncpus;
+ numvecs -= nvectors;
+ }
+}
+
static int __irq_build_affinity_masks(unsigned int startvec,
unsigned int numvecs,
unsigned int firstvec,
@@ -102,10 +252,11 @@ static int __irq_build_affinity_masks(unsigned int startvec,
struct cpumask *nmsk,
struct irq_affinity_desc *masks)
{
- unsigned int n, nodes, cpus_per_vec, extra_vecs, done = 0;
+ unsigned int i, n, nodes, cpus_per_vec, extra_vecs, done = 0;
unsigned int last_affv = firstvec + numvecs;
unsigned int curvec = startvec;
nodemask_t nodemsk = NODE_MASK_NONE;
+ struct node_vectors *node_vectors;
if (!cpumask_weight(cpu_mask))
return 0;
@@ -126,42 +277,56 @@ static int __irq_build_affinity_masks(unsigned int startvec,
return numvecs;
}
- for_each_node_mask(n, nodemsk) {
- unsigned int ncpus, v, vecs_to_assign, vecs_per_node;
+ node_vectors = kcalloc(nr_node_ids,
+ sizeof(struct node_vectors),
+ GFP_KERNEL);
+ if (!node_vectors)
+ return -ENOMEM;
- /* Spread the vectors per node */
- vecs_per_node = (numvecs - (curvec - firstvec)) / nodes;
+ /* allocate vector number for each node */
+ alloc_nodes_vectors(numvecs, node_to_cpumask, cpu_mask,
+ nodemsk, nmsk, node_vectors);
- /* Get the cpus on this node which are in the mask */
- cpumask_and(nmsk, cpu_mask, node_to_cpumask[n]);
+ for (i = 0; i < nr_node_ids; i++) {
+ unsigned int ncpus, v;
+ struct node_vectors *nv = &node_vectors[i];
+
+ if (nv->nvectors == UINT_MAX)
+ continue;
- /* Calculate the number of cpus per vector */
+ /* Get the cpus on this node which are in the mask */
+ cpumask_and(nmsk, cpu_mask, node_to_cpumask[nv->id]);
ncpus = cpumask_weight(nmsk);
- vecs_to_assign = min(vecs_per_node, ncpus);
+ if (!ncpus)
+ continue;
+
+ WARN_ON_ONCE(nv->nvectors > ncpus);
/* Account for rounding errors */
- extra_vecs = ncpus - vecs_to_assign * (ncpus / vecs_to_assign);
+ extra_vecs = ncpus - nv->nvectors * (ncpus / nv->nvectors);
- for (v = 0; curvec < last_affv && v < vecs_to_assign;
- curvec++, v++) {
- cpus_per_vec = ncpus / vecs_to_assign;
+ /* Spread allocated vectors on CPUs of the current node */
+ for (v = 0; v < nv->nvectors; v++, curvec++) {
+ cpus_per_vec = ncpus / nv->nvectors;
/* Account for extra vectors to compensate rounding errors */
if (extra_vecs) {
cpus_per_vec++;
--extra_vecs;
}
+
+ /*
+ * wrapping has to be considered given 'startvec'
+ * may start anywhere
+ */
+ if (curvec >= last_affv)
+ curvec = firstvec;
irq_spread_init_one(&masks[curvec].mask, nmsk,
cpus_per_vec);
}
-
- done += v;
- if (done >= numvecs)
- break;
- if (curvec >= last_affv)
- curvec = firstvec;
- --nodes;
+ done += nv->nvectors;
}
+ kfree(node_vectors);
return done;
}
@@ -174,7 +339,7 @@ static int irq_build_affinity_masks(unsigned int startvec, unsigned int numvecs,
unsigned int firstvec,
struct irq_affinity_desc *masks)
{
- unsigned int curvec = startvec, nr_present, nr_others;
+ unsigned int curvec = startvec, nr_present = 0, nr_others = 0;
cpumask_var_t *node_to_cpumask;
cpumask_var_t nmsk, npresmsk;
int ret = -ENOMEM;
@@ -189,15 +354,17 @@ static int irq_build_affinity_masks(unsigned int startvec, unsigned int numvecs,
if (!node_to_cpumask)
goto fail_npresmsk;
- ret = 0;
/* Stabilize the cpumasks */
get_online_cpus();
build_node_to_cpumask(node_to_cpumask);
/* Spread on present CPUs starting from affd->pre_vectors */
- nr_present = __irq_build_affinity_masks(curvec, numvecs,
- firstvec, node_to_cpumask,
- cpu_present_mask, nmsk, masks);
+ ret = __irq_build_affinity_masks(curvec, numvecs, firstvec,
+ node_to_cpumask, cpu_present_mask,
+ nmsk, masks);
+ if (ret < 0)
+ goto fail_build_affinity;
+ nr_present = ret;
/*
* Spread on non present CPUs starting from the next vector to be
@@ -210,12 +377,16 @@ static int irq_build_affinity_masks(unsigned int startvec, unsigned int numvecs,
else
curvec = firstvec + nr_present;
cpumask_andnot(npresmsk, cpu_possible_mask, cpu_present_mask);
- nr_others = __irq_build_affinity_masks(curvec, numvecs,
- firstvec, node_to_cpumask,
- npresmsk, nmsk, masks);
+ ret = __irq_build_affinity_masks(curvec, numvecs, firstvec,
+ node_to_cpumask, npresmsk, nmsk,
+ masks);
+ if (ret >= 0)
+ nr_others = ret;
+
+ fail_build_affinity:
put_online_cpus();
- if (nr_present < numvecs)
+ if (ret >= 0)
WARN_ON(nr_present + nr_others < numvecs);
free_node_to_cpumask(node_to_cpumask);
@@ -225,7 +396,7 @@ static int irq_build_affinity_masks(unsigned int startvec, unsigned int numvecs,
fail_nmsk:
free_cpumask_var(nmsk);
- return ret;
+ return ret < 0 ? ret : 0;
}
static void default_calc_sets(struct irq_affinity *affd, unsigned int affvecs)
diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c
index 3078d0e48bba..132672b74e4b 100644
--- a/kernel/irq/irqdomain.c
+++ b/kernel/irq/irqdomain.c
@@ -31,7 +31,7 @@ struct irqchip_fwid {
struct fwnode_handle fwnode;
unsigned int type;
char *name;
- void *data;
+ phys_addr_t *pa;
};
#ifdef CONFIG_GENERIC_IRQ_DEBUGFS
@@ -62,7 +62,8 @@ EXPORT_SYMBOL_GPL(irqchip_fwnode_ops);
* domain struct.
*/
struct fwnode_handle *__irq_domain_alloc_fwnode(unsigned int type, int id,
- const char *name, void *data)
+ const char *name,
+ phys_addr_t *pa)
{
struct irqchip_fwid *fwid;
char *n;
@@ -77,7 +78,7 @@ struct fwnode_handle *__irq_domain_alloc_fwnode(unsigned int type, int id,
n = kasprintf(GFP_KERNEL, "%s-%d", name, id);
break;
default:
- n = kasprintf(GFP_KERNEL, "irqchip@%p", data);
+ n = kasprintf(GFP_KERNEL, "irqchip@%pa", pa);
break;
}
@@ -89,7 +90,7 @@ struct fwnode_handle *__irq_domain_alloc_fwnode(unsigned int type, int id,
fwid->type = type;
fwid->name = n;
- fwid->data = data;
+ fwid->pa = pa;
fwid->fwnode.ops = &irqchip_fwnode_ops;
return &fwid->fwnode;
}
@@ -148,6 +149,7 @@ struct irq_domain *__irq_domain_add(struct fwnode_handle *fwnode, int size,
switch (fwid->type) {
case IRQCHIP_FWNODE_NAMED:
case IRQCHIP_FWNODE_NAMED_ID:
+ domain->fwnode = fwnode;
domain->name = kstrdup(fwid->name, GFP_KERNEL);
if (!domain->name) {
kfree(domain);
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 9d50fbe5531a..1753486b440c 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -23,7 +23,7 @@
#include "internals.h"
-#ifdef CONFIG_IRQ_FORCED_THREADING
+#if defined(CONFIG_IRQ_FORCED_THREADING) && !defined(CONFIG_PREEMPT_RT)
__read_mostly bool force_irqthreads;
EXPORT_SYMBOL_GPL(force_irqthreads);