[v4,19/19] irqdomain: Switch to per-domain locking

Message ID 20230116135044.14998-20-johan+linaro@kernel.org
State New
Headers
Series irqdomain: fix mapping race and clean up locking |

Commit Message

Johan Hovold Jan. 16, 2023, 1:50 p.m. UTC
  The IRQ domain structures are currently protected by the global
irq_domain_mutex. Switch to using more fine-grained per-domain locking,
which may potentially speed up parallel probing somewhat.

Note that the domain lock of the root domain (innermost domain) must be
used for hierarchical domains. For non-hierarchical domain (as for root
domains), the new root pointer is set to the domain itself so that
domain->root->mutex can be used in shared code paths.

Also note that hierarchical domains should be constructed using
irq_domain_create_hierarchy() (or irq_domain_add_hierarchy()) to avoid
poking at irqdomain internals. As a safeguard, the lockdep assertion in
irq_domain_set_mapping() will catch any offenders that fail to set the
root domain pointer.

Tested-by: Hsin-Yi Wang <hsinyi@chromium.org>
Tested-by: Mark-PK Tsai <mark-pk.tsai@mediatek.com>
Signed-off-by: Johan Hovold <johan+linaro@kernel.org>
---
 include/linux/irqdomain.h |  4 +++
 kernel/irq/irqdomain.c    | 61 +++++++++++++++++++++++++--------------
 2 files changed, 44 insertions(+), 21 deletions(-)
  

Comments

Thomas Gleixner Jan. 17, 2023, 9:50 p.m. UTC | #1
On Mon, Jan 16 2023 at 14:50, Johan Hovold wrote:
> The IRQ domain structures are currently protected by the global
> irq_domain_mutex. Switch to using more fine-grained per-domain locking,
> which may potentially speed up parallel probing somewhat.

Does it or not?

If not then why adding all this churn for no real value?

Thanks,

        tglx
  
Johan Hovold Jan. 18, 2023, 9:51 a.m. UTC | #2
On Tue, Jan 17, 2023 at 10:50:39PM +0100, Thomas Gleixner wrote:
> On Mon, Jan 16 2023 at 14:50, Johan Hovold wrote:
> > The IRQ domain structures are currently protected by the global
> > irq_domain_mutex. Switch to using more fine-grained per-domain locking,
> > which may potentially speed up parallel probing somewhat.
> 
> Does it or not?
> 
> If not then why adding all this churn for no real value?

It probably doesn't make much difference, but Marc wanted per-domain
locking:

    > I'd really like to avoid a global mutex. At the very least this should
    > be a per-domain mutex, otherwise this will serialise a lot more than
    > what is needed.
    
    Yeah, I considered that too, but wanted to get your comments on this
    first.
    
    Also note that the likewise global irq_domain_mutex (and
    sparse_irq_lock) are taken in some of these paths so perhaps using finer
    locking won't actually matter that much as this is mostly for parallel
    probing.

    https://lore.kernel.org/lkml/YuKHiZuNvN+K9NCc@hovoldconsulting.com/

As part of fixing the races, this series has now replaced the per-domain
revmap mutexes with the global irq_domain_mutex, which could possibly be
perceived as a step in the wrong direction in this respect.

This patch restores per-domain locking for non-hierarchical domains and
extends it to hierarchical domains. This leaves the irq_domain_mutex to
only be used for things that actually need a global lock such as the
domain list.

I consider this mostly a clean up, and I did intentionally place it last
in order to not have the fixes depend on it.

Johan
  
Thomas Gleixner Jan. 18, 2023, 1:09 p.m. UTC | #3
On Wed, Jan 18 2023 at 10:51, Johan Hovold wrote:
> On Tue, Jan 17, 2023 at 10:50:39PM +0100, Thomas Gleixner wrote:
>> On Mon, Jan 16 2023 at 14:50, Johan Hovold wrote:
>> > The IRQ domain structures are currently protected by the global
>> > irq_domain_mutex. Switch to using more fine-grained per-domain locking,
>> > which may potentially speed up parallel probing somewhat.
>> 
>> Does it or not?
>> 
>> If not then why adding all this churn for no real value?
>
> It probably doesn't make much difference, but Marc wanted per-domain
> locking:
>
>     > I'd really like to avoid a global mutex. At the very least this should
>     > be a per-domain mutex, otherwise this will serialise a lot more than
>     > what is needed.

Sure it serializes more than what is needed, but the real question is
whether it matters. If it does not matter then I prefer KISS over a just
because we can optimization.

Thanks,

        tglx
  

Patch

diff --git a/include/linux/irqdomain.h b/include/linux/irqdomain.h
index 7fd3939328c2..b1b06d75d31a 100644
--- a/include/linux/irqdomain.h
+++ b/include/linux/irqdomain.h
@@ -125,6 +125,8 @@  struct irq_domain_chip_generic;
  *		core code.
  * @flags:	Per irq_domain flags
  * @mapcount:	The number of mapped interrupts
+ * @mutex:	Domain lock, hierarhical domains use root domain's lock
+ * @root:	Pointer to root domain, or containing structure if non-hierarchical
  *
  * Optional elements:
  * @fwnode:	Pointer to firmware node associated with the irq_domain. Pretty easy
@@ -152,6 +154,8 @@  struct irq_domain {
 	void				*host_data;
 	unsigned int			flags;
 	unsigned int			mapcount;
+	struct mutex			mutex;
+	struct irq_domain		*root;
 
 	/* Optional data */
 	struct fwnode_handle		*fwnode;
diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c
index 6f2b8a1248e1..77c31b89740d 100644
--- a/kernel/irq/irqdomain.c
+++ b/kernel/irq/irqdomain.c
@@ -226,6 +226,17 @@  struct irq_domain *__irq_domain_add(struct fwnode_handle *fwnode, unsigned int s
 
 	domain->revmap_size = size;
 
+	/*
+	 * Hierarchical domains use the domain lock of the root domain
+	 * (innermost domain).
+	 *
+	 * For non-hierarchical domains (as for root domains), the root
+	 * pointer is set to the domain itself so that domain->root->mutex
+	 * can be used in shared code paths.
+	 */
+	mutex_init(&domain->mutex);
+	domain->root = domain;
+
 	irq_domain_check_hierarchy(domain);
 
 	mutex_lock(&irq_domain_mutex);
@@ -503,7 +514,7 @@  static bool irq_domain_is_nomap(struct irq_domain *domain)
 static void irq_domain_clear_mapping(struct irq_domain *domain,
 				     irq_hw_number_t hwirq)
 {
-	lockdep_assert_held(&irq_domain_mutex);
+	lockdep_assert_held(&domain->root->mutex);
 
 	if (irq_domain_is_nomap(domain))
 		return;
@@ -518,7 +529,11 @@  static void irq_domain_set_mapping(struct irq_domain *domain,
 				   irq_hw_number_t hwirq,
 				   struct irq_data *irq_data)
 {
-	lockdep_assert_held(&irq_domain_mutex);
+	/*
+	 * This also makes sure that all domains point to the same root when
+	 * called from irq_domain_insert_irq() for each domain in a hierarchy.
+	 */
+	lockdep_assert_held(&domain->root->mutex);
 
 	if (irq_domain_is_nomap(domain))
 		return;
@@ -540,7 +555,7 @@  static void irq_domain_disassociate(struct irq_domain *domain, unsigned int irq)
 
 	hwirq = irq_data->hwirq;
 
-	mutex_lock(&irq_domain_mutex);
+	mutex_lock(&domain->mutex);
 
 	irq_set_status_flags(irq, IRQ_NOREQUEST);
 
@@ -562,7 +577,7 @@  static void irq_domain_disassociate(struct irq_domain *domain, unsigned int irq)
 	/* Clear reverse map for this hwirq */
 	irq_domain_clear_mapping(domain, hwirq);
 
-	mutex_unlock(&irq_domain_mutex);
+	mutex_unlock(&domain->mutex);
 }
 
 static int __irq_domain_associate(struct irq_domain *domain, unsigned int virq,
@@ -612,9 +627,9 @@  int irq_domain_associate(struct irq_domain *domain, unsigned int virq,
 {
 	int ret;
 
-	mutex_lock(&irq_domain_mutex);
+	mutex_lock(&domain->mutex);
 	ret = __irq_domain_associate(domain, virq, hwirq);
-	mutex_unlock(&irq_domain_mutex);
+	mutex_unlock(&domain->mutex);
 
 	return ret;
 }
@@ -731,7 +746,7 @@  unsigned int irq_create_mapping_affinity(struct irq_domain *domain,
 		return 0;
 	}
 
-	mutex_lock(&irq_domain_mutex);
+	mutex_lock(&domain->mutex);
 
 	/* Check if mapping already exists */
 	virq = irq_find_mapping(domain, hwirq);
@@ -742,7 +757,7 @@  unsigned int irq_create_mapping_affinity(struct irq_domain *domain,
 
 	virq = __irq_create_mapping_affinity(domain, hwirq, affinity);
 out:
-	mutex_unlock(&irq_domain_mutex);
+	mutex_unlock(&domain->mutex);
 
 	return virq;
 }
@@ -811,7 +826,7 @@  unsigned int irq_create_fwspec_mapping(struct irq_fwspec *fwspec)
 	if (WARN_ON(type & ~IRQ_TYPE_SENSE_MASK))
 		type &= IRQ_TYPE_SENSE_MASK;
 
-	mutex_lock(&irq_domain_mutex);
+	mutex_lock(&domain->root->mutex);
 
 	/*
 	 * If we've already configured this interrupt,
@@ -864,11 +879,11 @@  unsigned int irq_create_fwspec_mapping(struct irq_fwspec *fwspec)
 	/* Store trigger type */
 	irqd_set_trigger_type(irq_data, type);
 out:
-	mutex_unlock(&irq_domain_mutex);
+	mutex_unlock(&domain->root->mutex);
 
 	return virq;
 err:
-	mutex_unlock(&irq_domain_mutex);
+	mutex_unlock(&domain->root->mutex);
 
 	return 0;
 }
@@ -1132,6 +1147,7 @@  struct irq_domain *irq_domain_create_hierarchy(struct irq_domain *parent,
 	else
 		domain = irq_domain_create_tree(fwnode, ops, host_data);
 	if (domain) {
+		domain->root = parent->root;
 		domain->parent = parent;
 		domain->flags |= flags;
 	}
@@ -1528,10 +1544,10 @@  int __irq_domain_alloc_irqs(struct irq_domain *domain, int irq_base,
 			return -EINVAL;
 	}
 
-	mutex_lock(&irq_domain_mutex);
+	mutex_lock(&domain->root->mutex);
 	ret = ___irq_domain_alloc_irqs(domain, irq_base, nr_irqs, node, arg,
 				       realloc, affinity);
-	mutex_unlock(&irq_domain_mutex);
+	mutex_unlock(&domain->root->mutex);
 
 	return ret;
 }
@@ -1542,7 +1558,7 @@  static void irq_domain_fix_revmap(struct irq_data *d)
 {
 	void __rcu **slot;
 
-	lockdep_assert_held(&irq_domain_mutex);
+	lockdep_assert_held(&d->domain->root->mutex);
 
 	if (irq_domain_is_nomap(d->domain))
 		return;
@@ -1608,7 +1624,7 @@  int irq_domain_push_irq(struct irq_domain *domain, int virq, void *arg)
 	if (!parent_irq_data)
 		return -ENOMEM;
 
-	mutex_lock(&irq_domain_mutex);
+	mutex_lock(&domain->root->mutex);
 
 	/* Copy the original irq_data. */
 	*parent_irq_data = *irq_data;
@@ -1636,7 +1652,7 @@  int irq_domain_push_irq(struct irq_domain *domain, int virq, void *arg)
 	irq_domain_fix_revmap(parent_irq_data);
 	irq_domain_set_mapping(domain, irq_data->hwirq, irq_data);
 error:
-	mutex_unlock(&irq_domain_mutex);
+	mutex_unlock(&domain->root->mutex);
 
 	return rv;
 }
@@ -1691,7 +1707,7 @@  int irq_domain_pop_irq(struct irq_domain *domain, int virq)
 	if (WARN_ON(!parent_irq_data))
 		return -EINVAL;
 
-	mutex_lock(&irq_domain_mutex);
+	mutex_lock(&domain->root->mutex);
 
 	irq_data->parent_data = NULL;
 
@@ -1703,7 +1719,7 @@  int irq_domain_pop_irq(struct irq_domain *domain, int virq)
 
 	irq_domain_fix_revmap(irq_data);
 
-	mutex_unlock(&irq_domain_mutex);
+	mutex_unlock(&domain->root->mutex);
 
 	kfree(parent_irq_data);
 
@@ -1719,17 +1735,20 @@  EXPORT_SYMBOL_GPL(irq_domain_pop_irq);
 void irq_domain_free_irqs(unsigned int virq, unsigned int nr_irqs)
 {
 	struct irq_data *data = irq_get_irq_data(virq);
+	struct irq_domain *domain;
 	int i;
 
 	if (WARN(!data || !data->domain || !data->domain->ops->free,
 		 "NULL pointer, cannot free irq\n"))
 		return;
 
-	mutex_lock(&irq_domain_mutex);
+	domain = data->domain;
+
+	mutex_lock(&domain->root->mutex);
 	for (i = 0; i < nr_irqs; i++)
 		irq_domain_remove_irq(virq + i);
-	irq_domain_free_irqs_hierarchy(data->domain, virq, nr_irqs);
-	mutex_unlock(&irq_domain_mutex);
+	irq_domain_free_irqs_hierarchy(domain, virq, nr_irqs);
+	mutex_unlock(&domain->root->mutex);
 
 	irq_domain_free_irq_data(virq, nr_irqs);
 	irq_free_descs(virq, nr_irqs);