diff options
Diffstat (limited to 'drivers/base')
| -rw-r--r-- | drivers/base/base.h | 8 | ||||
| -rw-r--r-- | drivers/base/devres.c | 21 | ||||
| -rw-r--r-- | drivers/base/memory.c | 53 | ||||
| -rw-r--r-- | drivers/base/node.c | 52 | ||||
| -rw-r--r-- | drivers/base/power/main.c | 32 | ||||
| -rw-r--r-- | drivers/base/power/runtime.c | 3 | ||||
| -rw-r--r-- | drivers/base/property.c | 2 |
7 files changed, 120 insertions, 51 deletions
diff --git a/drivers/base/base.h b/drivers/base/base.h index 123031a757d9..700aecd22fd3 100644 --- a/drivers/base/base.h +++ b/drivers/base/base.h @@ -251,6 +251,14 @@ void device_links_unbind_consumers(struct device *dev); void fw_devlink_drivers_done(void); void fw_devlink_probing_done(void); +#define dev_for_each_link_to_supplier(__link, __dev) \ + list_for_each_entry_srcu(__link, &(__dev)->links.suppliers, c_node, \ + device_links_read_lock_held()) + +#define dev_for_each_link_to_consumer(__link, __dev) \ + list_for_each_entry_srcu(__link, &(__dev)->links.consumers, s_node, \ + device_links_read_lock_held()) + /* device pm support */ void device_pm_move_to_tail(struct device *dev); diff --git a/drivers/base/devres.c b/drivers/base/devres.c index ff55e1bcfa30..c948c88d3956 100644 --- a/drivers/base/devres.c +++ b/drivers/base/devres.c @@ -1117,6 +1117,27 @@ void *devm_kmemdup(struct device *dev, const void *src, size_t len, gfp_t gfp) } EXPORT_SYMBOL_GPL(devm_kmemdup); +/** + * devm_kmemdup_const - conditionally duplicate and manage a region of memory + * + * @dev: Device this memory belongs to + * @src: memory region to duplicate + * @len: memory region length, + * @gfp: GFP mask to use + * + * Return: source address if it is in .rodata or the return value of kmemdup() + * to which the function falls back otherwise. + */ +const void * +devm_kmemdup_const(struct device *dev, const void *src, size_t len, gfp_t gfp) +{ + if (is_kernel_rodata((unsigned long)src)) + return src; + + return devm_kmemdup(dev, src, len, gfp); +} +EXPORT_SYMBOL_GPL(devm_kmemdup_const); + struct pages_devres { unsigned long addr; unsigned int order; diff --git a/drivers/base/memory.c b/drivers/base/memory.c index 5c6c1d6bb59f..6d84a02cfa5d 100644 --- a/drivers/base/memory.c +++ b/drivers/base/memory.c @@ -769,21 +769,22 @@ static struct zone *early_node_zone_for_memory_block(struct memory_block *mem, #ifdef CONFIG_NUMA /** - * memory_block_add_nid() - Indicate that system RAM falling into this memory - * block device (partially) belongs to the given node. + * memory_block_add_nid_early() - Indicate that early system RAM falling into + * this memory block device (partially) belongs + * to the given node. * @mem: The memory block device. * @nid: The node id. - * @context: The memory initialization context. * - * Indicate that system RAM falling into this memory block (partially) belongs - * to the given node. If the context indicates ("early") that we are adding the - * node during node device subsystem initialization, this will also properly - * set/adjust mem->zone based on the zone ranges of the given node. + * Indicate that early system RAM falling into this memory block (partially) + * belongs to the given node. This will also properly set/adjust mem->zone based + * on the zone ranges of the given node. + * + * Memory hotplug handles this on memory block creation, where we can only have + * a single nid span a memory block. */ -void memory_block_add_nid(struct memory_block *mem, int nid, - enum meminit_context context) +void memory_block_add_nid_early(struct memory_block *mem, int nid) { - if (context == MEMINIT_EARLY && mem->nid != nid) { + if (mem->nid != nid) { /* * For early memory we have to determine the zone when setting * the node id and handle multiple nodes spanning a single @@ -797,19 +798,18 @@ void memory_block_add_nid(struct memory_block *mem, int nid, mem->zone = early_node_zone_for_memory_block(mem, nid); else mem->zone = NULL; + /* + * If this memory block spans multiple nodes, we only indicate + * the last processed node. If we span multiple nodes (not applicable + * to hotplugged memory), zone == NULL will prohibit memory offlining + * and consequently unplug. + */ + mem->nid = nid; } - - /* - * If this memory block spans multiple nodes, we only indicate - * the last processed node. If we span multiple nodes (not applicable - * to hotplugged memory), zone == NULL will prohibit memory offlining - * and consequently unplug. - */ - mem->nid = nid; } #endif -static int add_memory_block(unsigned long block_id, unsigned long state, +static int add_memory_block(unsigned long block_id, int nid, unsigned long state, struct vmem_altmap *altmap, struct memory_group *group) { @@ -827,7 +827,7 @@ static int add_memory_block(unsigned long block_id, unsigned long state, mem->start_section_nr = block_id * sections_per_block; mem->state = state; - mem->nid = NUMA_NO_NODE; + mem->nid = nid; mem->altmap = altmap; INIT_LIST_HEAD(&mem->group_next); @@ -854,13 +854,6 @@ static int add_memory_block(unsigned long block_id, unsigned long state, return 0; } -static int add_hotplug_memory_block(unsigned long block_id, - struct vmem_altmap *altmap, - struct memory_group *group) -{ - return add_memory_block(block_id, MEM_OFFLINE, altmap, group); -} - static void remove_memory_block(struct memory_block *memory) { if (WARN_ON_ONCE(memory->dev.bus != &memory_subsys)) @@ -886,7 +879,7 @@ static void remove_memory_block(struct memory_block *memory) * Called under device_hotplug_lock. */ int create_memory_block_devices(unsigned long start, unsigned long size, - struct vmem_altmap *altmap, + int nid, struct vmem_altmap *altmap, struct memory_group *group) { const unsigned long start_block_id = pfn_to_block_id(PFN_DOWN(start)); @@ -900,7 +893,7 @@ int create_memory_block_devices(unsigned long start, unsigned long size, return -EINVAL; for (block_id = start_block_id; block_id != end_block_id; block_id++) { - ret = add_hotplug_memory_block(block_id, altmap, group); + ret = add_memory_block(block_id, nid, MEM_OFFLINE, altmap, group); if (ret) break; } @@ -1005,7 +998,7 @@ void __init memory_dev_init(void) continue; block_id = memory_block_id(nr); - ret = add_memory_block(block_id, MEM_ONLINE, NULL, NULL); + ret = add_memory_block(block_id, NUMA_NO_NODE, MEM_ONLINE, NULL, NULL); if (ret) { panic("%s() failed to add memory block: %d\n", __func__, ret); diff --git a/drivers/base/node.c b/drivers/base/node.c index 3399594136b2..83aeb0518e1d 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -249,6 +249,44 @@ void node_set_perf_attrs(unsigned int nid, struct access_coordinate *coord, EXPORT_SYMBOL_GPL(node_set_perf_attrs); /** + * node_update_perf_attrs - Update the performance values for given access class + * @nid: Node identifier to be updated + * @coord: Heterogeneous memory performance coordinates + * @access: The access class for the given attributes + */ +void node_update_perf_attrs(unsigned int nid, struct access_coordinate *coord, + enum access_coordinate_class access) +{ + struct node_access_nodes *access_node; + struct node *node; + int i; + + if (WARN_ON_ONCE(!node_online(nid))) + return; + + node = node_devices[nid]; + list_for_each_entry(access_node, &node->access_list, list_node) { + if (access_node->access != access) + continue; + + access_node->coord = *coord; + for (i = 0; access_attrs[i]; i++) { + sysfs_notify(&access_node->dev.kobj, + NULL, access_attrs[i]->name); + } + break; + } + + /* When setting CPU access coordinates, update mempolicy */ + if (access != ACCESS_COORDINATE_CPU) + return; + + if (mempolicy_set_node_perf(nid, coord)) + pr_info("failed to set mempolicy attrs for node %d\n", nid); +} +EXPORT_SYMBOL_GPL(node_update_perf_attrs); + +/** * struct node_cache_info - Internal tracking for memory node caches * @dev: Device represeting the cache level * @node: List element for tracking in the node @@ -781,13 +819,10 @@ int unregister_cpu_under_node(unsigned int cpu, unsigned int nid) #ifdef CONFIG_MEMORY_HOTPLUG static void do_register_memory_block_under_node(int nid, - struct memory_block *mem_blk, - enum meminit_context context) + struct memory_block *mem_blk) { int ret; - memory_block_add_nid(mem_blk, nid, context); - ret = sysfs_create_link_nowarn(&node_devices[nid]->dev.kobj, &mem_blk->dev.kobj, kobject_name(&mem_blk->dev.kobj)); @@ -815,7 +850,7 @@ static int register_mem_block_under_node_hotplug(struct memory_block *mem_blk, { int nid = *(int *)arg; - do_register_memory_block_under_node(nid, mem_blk, MEMINIT_HOTPLUG); + do_register_memory_block_under_node(nid, mem_blk); return 0; } @@ -855,7 +890,8 @@ static void register_memory_blocks_under_nodes(void) if (!mem) continue; - do_register_memory_block_under_node(nid, mem, MEMINIT_EARLY); + memory_block_add_nid_early(mem, nid); + do_register_memory_block_under_node(nid, mem); put_device(&mem->dev); } @@ -885,6 +921,10 @@ int register_one_node(int nid) node_devices[nid] = node; error = register_node(node_devices[nid], nid); + if (error) { + node_devices[nid] = NULL; + return error; + } /* link cpu under this node */ for_each_present_cpu(cpu) { diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 2ea6e05e6ec9..b9a34c3425ec 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -40,10 +40,6 @@ typedef int (*pm_callback_t)(struct device *); -#define list_for_each_entry_rcu_locked(pos, head, member) \ - list_for_each_entry_rcu(pos, head, member, \ - device_links_read_lock_held()) - /* * The entries in the dpm_list list are in a depth first order, simply * because children are guaranteed to be discovered after parents, and @@ -281,7 +277,7 @@ static void dpm_wait_for_suppliers(struct device *dev, bool async) * callbacks freeing the link objects for the links in the list we're * walking. */ - list_for_each_entry_rcu_locked(link, &dev->links.suppliers, c_node) + dev_for_each_link_to_supplier(link, dev) if (READ_ONCE(link->status) != DL_STATE_DORMANT) dpm_wait(link->supplier, async); @@ -338,7 +334,7 @@ static void dpm_wait_for_consumers(struct device *dev, bool async) * continue instead of trying to continue in parallel with its * unregistration). */ - list_for_each_entry_rcu_locked(link, &dev->links.consumers, s_node) + dev_for_each_link_to_consumer(link, dev) if (READ_ONCE(link->status) != DL_STATE_DORMANT) dpm_wait(link->consumer, async); @@ -675,7 +671,7 @@ static void dpm_async_resume_subordinate(struct device *dev, async_func_t func) idx = device_links_read_lock(); /* Start processing the device's "async" consumers. */ - list_for_each_entry_rcu_locked(link, &dev->links.consumers, s_node) + dev_for_each_link_to_consumer(link, dev) if (READ_ONCE(link->status) != DL_STATE_DORMANT) dpm_async_with_cleanup(link->consumer, func); @@ -724,8 +720,20 @@ static void device_resume_noirq(struct device *dev, pm_message_t state, bool asy if (dev->power.syscore || dev->power.direct_complete) goto Out; - if (!dev->power.is_noirq_suspended) + if (!dev->power.is_noirq_suspended) { + /* + * This means that system suspend has been aborted in the noirq + * phase before invoking the noirq suspend callback for the + * device, so if device_suspend_late() has left it in suspend, + * device_resume_early() should leave it in suspend either in + * case the early resume of it depends on the noirq resume that + * has not run. + */ + if (dev_pm_skip_suspend(dev)) + dev->power.must_resume = false; + goto Out; + } if (!dpm_wait_for_superior(dev, async)) goto Out; @@ -1330,7 +1338,7 @@ static void dpm_async_suspend_superior(struct device *dev, async_func_t func) idx = device_links_read_lock(); /* Start processing the device's "async" suppliers. */ - list_for_each_entry_rcu_locked(link, &dev->links.suppliers, c_node) + dev_for_each_link_to_supplier(link, dev) if (READ_ONCE(link->status) != DL_STATE_DORMANT) dpm_async_with_cleanup(link->supplier, func); @@ -1384,7 +1392,7 @@ static void dpm_superior_set_must_resume(struct device *dev) idx = device_links_read_lock(); - list_for_each_entry_rcu_locked(link, &dev->links.suppliers, c_node) + dev_for_each_link_to_supplier(link, dev) link->supplier->power.must_resume = true; device_links_read_unlock(idx); @@ -1813,7 +1821,7 @@ static void dpm_clear_superiors_direct_complete(struct device *dev) idx = device_links_read_lock(); - list_for_each_entry_rcu_locked(link, &dev->links.suppliers, c_node) { + dev_for_each_link_to_supplier(link, dev) { spin_lock_irq(&link->supplier->power.lock); link->supplier->power.direct_complete = false; spin_unlock_irq(&link->supplier->power.lock); @@ -2065,7 +2073,7 @@ static bool device_prepare_smart_suspend(struct device *dev) idx = device_links_read_lock(); - list_for_each_entry_rcu_locked(link, &dev->links.suppliers, c_node) { + dev_for_each_link_to_supplier(link, dev) { if (!device_link_test(link, DL_FLAG_PM_RUNTIME)) continue; diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c index 3e84dc4122de..7420b9851fe0 100644 --- a/drivers/base/power/runtime.c +++ b/drivers/base/power/runtime.c @@ -1903,8 +1903,7 @@ void pm_runtime_get_suppliers(struct device *dev) idx = device_links_read_lock(); - list_for_each_entry_rcu(link, &dev->links.suppliers, c_node, - device_links_read_lock_held()) + dev_for_each_link_to_supplier(link, dev) if (device_link_test(link, DL_FLAG_PM_RUNTIME)) { link->supplier_preactivated = true; pm_runtime_get_sync(link->supplier); diff --git a/drivers/base/property.c b/drivers/base/property.c index f626d5bbe806..6a63860579dd 100644 --- a/drivers/base/property.c +++ b/drivers/base/property.c @@ -578,7 +578,7 @@ EXPORT_SYMBOL_GPL(fwnode_property_match_property_string); * @prop: The name of the property * @nargs_prop: The name of the property telling the number of * arguments in the referred node. NULL if @nargs is known, - * otherwise @nargs is ignored. Only relevant on OF. + * otherwise @nargs is ignored. * @nargs: Number of arguments. Ignored if @nargs_prop is non-NULL. * @index: Index of the reference, from zero onwards. * @args: Result structure with reference and integer arguments. |