X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=drivers%2Fdma%2Fdmaengine.c;h=9d31d5eb95c1ea47618b01ed61206eaf315a7925;hb=b3040e40675ec2c43542331cd30d4ee3dae797e8;hp=404cc7b6e70530066b5a91970d791c8ae8f978c0;hpb=7405f74badf46b5d023c5d2b670b4471525f6c91;p=safe%2Fjmp%2Flinux-2.6 diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c index 404cc7b..9d31d5e 100644 --- a/drivers/dma/dmaengine.c +++ b/drivers/dma/dmaengine.c @@ -31,30 +31,18 @@ * * LOCKING: * - * The subsystem keeps two global lists, dma_device_list and dma_client_list. - * Both of these are protected by a mutex, dma_list_mutex. + * The subsystem keeps a global list of dma_device structs it is protected by a + * mutex, dma_list_mutex. + * + * A subsystem can get access to a channel by calling dmaengine_get() followed + * by dma_find_channel(), or if it has need for an exclusive channel it can call + * dma_request_channel(). Once a channel is allocated a reference is taken + * against its corresponding driver to disable removal. * * Each device has a channels list, which runs unlocked but is never modified * once the device is registered, it's just setup by the driver. * - * Each client has a channels list, it's only modified under the client->lock - * and in an RCU callback, so it's safe to read under rcu_read_lock(). - * - * Each device has a kref, which is initialized to 1 when the device is - * registered. A kref_put is done for each class_device registered. When the - * class_device is released, the coresponding kref_put is done in the release - * method. Every time one of the device's channels is allocated to a client, - * a kref_get occurs. When the channel is freed, the coresponding kref_put - * happens. The device's release function does a completion, so - * unregister_device does a remove event, class_device_unregister, a kref_put - * for the first reference, then waits on the completion for all other - * references to finish. - * - * Each channel has an open-coded implementation of Rusty Russell's "bigref," - * with a kref and a per_cpu local_t. A single reference is set when on an - * ADDED event, and removed with a REMOVE event. Net DMA client takes an - * extra reference per outstanding transaction. The relase function does a - * kref_put on the device. -ChrisL + * See Documentation/dmaengine.txt for more details */ #include @@ -68,103 +56,199 @@ #include #include #include +#include +#include +#include static DEFINE_MUTEX(dma_list_mutex); static LIST_HEAD(dma_device_list); -static LIST_HEAD(dma_client_list); +static long dmaengine_ref_count; +static struct idr dma_idr; /* --- sysfs implementation --- */ -static ssize_t show_memcpy_count(struct class_device *cd, char *buf) +/** + * dev_to_dma_chan - convert a device pointer to the its sysfs container object + * @dev - device node + * + * Must be called under dma_list_mutex + */ +static struct dma_chan *dev_to_dma_chan(struct device *dev) { - struct dma_chan *chan = container_of(cd, struct dma_chan, class_dev); + struct dma_chan_dev *chan_dev; + + chan_dev = container_of(dev, typeof(*chan_dev), device); + return chan_dev->chan; +} + +static ssize_t show_memcpy_count(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct dma_chan *chan; unsigned long count = 0; int i; + int err; - for_each_possible_cpu(i) - count += per_cpu_ptr(chan->local, i)->memcpy_count; + mutex_lock(&dma_list_mutex); + chan = dev_to_dma_chan(dev); + if (chan) { + for_each_possible_cpu(i) + count += per_cpu_ptr(chan->local, i)->memcpy_count; + err = sprintf(buf, "%lu\n", count); + } else + err = -ENODEV; + mutex_unlock(&dma_list_mutex); - return sprintf(buf, "%lu\n", count); + return err; } -static ssize_t show_bytes_transferred(struct class_device *cd, char *buf) +static ssize_t show_bytes_transferred(struct device *dev, struct device_attribute *attr, + char *buf) { - struct dma_chan *chan = container_of(cd, struct dma_chan, class_dev); + struct dma_chan *chan; unsigned long count = 0; int i; + int err; - for_each_possible_cpu(i) - count += per_cpu_ptr(chan->local, i)->bytes_transferred; + mutex_lock(&dma_list_mutex); + chan = dev_to_dma_chan(dev); + if (chan) { + for_each_possible_cpu(i) + count += per_cpu_ptr(chan->local, i)->bytes_transferred; + err = sprintf(buf, "%lu\n", count); + } else + err = -ENODEV; + mutex_unlock(&dma_list_mutex); - return sprintf(buf, "%lu\n", count); + return err; } -static ssize_t show_in_use(struct class_device *cd, char *buf) +static ssize_t show_in_use(struct device *dev, struct device_attribute *attr, char *buf) { - struct dma_chan *chan = container_of(cd, struct dma_chan, class_dev); + struct dma_chan *chan; + int err; - return sprintf(buf, "%d\n", (chan->client ? 1 : 0)); + mutex_lock(&dma_list_mutex); + chan = dev_to_dma_chan(dev); + if (chan) + err = sprintf(buf, "%d\n", chan->client_count); + else + err = -ENODEV; + mutex_unlock(&dma_list_mutex); + + return err; } -static struct class_device_attribute dma_class_attrs[] = { +static struct device_attribute dma_attrs[] = { __ATTR(memcpy_count, S_IRUGO, show_memcpy_count, NULL), __ATTR(bytes_transferred, S_IRUGO, show_bytes_transferred, NULL), __ATTR(in_use, S_IRUGO, show_in_use, NULL), __ATTR_NULL }; -static void dma_async_device_cleanup(struct kref *kref); - -static void dma_class_dev_release(struct class_device *cd) +static void chan_dev_release(struct device *dev) { - struct dma_chan *chan = container_of(cd, struct dma_chan, class_dev); - kref_put(&chan->device->refcount, dma_async_device_cleanup); + struct dma_chan_dev *chan_dev; + + chan_dev = container_of(dev, typeof(*chan_dev), device); + if (atomic_dec_and_test(chan_dev->idr_ref)) { + mutex_lock(&dma_list_mutex); + idr_remove(&dma_idr, chan_dev->dev_id); + mutex_unlock(&dma_list_mutex); + kfree(chan_dev->idr_ref); + } + kfree(chan_dev); } static struct class dma_devclass = { - .name = "dma", - .class_dev_attrs = dma_class_attrs, - .release = dma_class_dev_release, + .name = "dma", + .dev_attrs = dma_attrs, + .dev_release = chan_dev_release, }; /* --- client and device registration --- */ +#define dma_device_satisfies_mask(device, mask) \ + __dma_device_satisfies_mask((device), &(mask)) +static int +__dma_device_satisfies_mask(struct dma_device *device, dma_cap_mask_t *want) +{ + dma_cap_mask_t has; + + bitmap_and(has.bits, want->bits, device->cap_mask.bits, + DMA_TX_TYPE_END); + return bitmap_equal(want->bits, has.bits, DMA_TX_TYPE_END); +} + +static struct module *dma_chan_to_owner(struct dma_chan *chan) +{ + return chan->device->dev->driver->owner; +} + /** - * dma_client_chan_alloc - try to allocate a channel to a client - * @client: &dma_client + * balance_ref_count - catch up the channel reference count + * @chan - channel to balance ->client_count versus dmaengine_ref_count * - * Called with dma_list_mutex held. + * balance_ref_count must be called under dma_list_mutex */ -static struct dma_chan *dma_client_chan_alloc(struct dma_client *client) +static void balance_ref_count(struct dma_chan *chan) { - struct dma_device *device; - struct dma_chan *chan; - unsigned long flags; - int desc; /* allocated descriptor count */ + struct module *owner = dma_chan_to_owner(chan); - /* Find a channel, any DMA engine will do */ - list_for_each_entry(device, &dma_device_list, global_node) { - list_for_each_entry(chan, &device->channels, device_node) { - if (chan->client) - continue; + while (chan->client_count < dmaengine_ref_count) { + __module_get(owner); + chan->client_count++; + } +} - desc = chan->device->device_alloc_chan_resources(chan); - if (desc >= 0) { - kref_get(&device->refcount); - kref_init(&chan->refcount); - chan->slow_ref = 0; - INIT_RCU_HEAD(&chan->rcu); - chan->client = client; - spin_lock_irqsave(&client->lock, flags); - list_add_tail_rcu(&chan->client_node, - &client->channels); - spin_unlock_irqrestore(&client->lock, flags); - return chan; - } - } +/** + * dma_chan_get - try to grab a dma channel's parent driver module + * @chan - channel to grab + * + * Must be called under dma_list_mutex + */ +static int dma_chan_get(struct dma_chan *chan) +{ + int err = -ENODEV; + struct module *owner = dma_chan_to_owner(chan); + + if (chan->client_count) { + __module_get(owner); + err = 0; + } else if (try_module_get(owner)) + err = 0; + + if (err == 0) + chan->client_count++; + + /* allocate upon first client reference */ + if (chan->client_count == 1 && err == 0) { + int desc_cnt = chan->device->device_alloc_chan_resources(chan); + + if (desc_cnt < 0) { + err = desc_cnt; + chan->client_count = 0; + module_put(owner); + } else if (!dma_has_cap(DMA_PRIVATE, chan->device->cap_mask)) + balance_ref_count(chan); } - return NULL; + return err; +} + +/** + * dma_chan_put - drop a reference to a dma channel's parent driver module + * @chan - channel to release + * + * Must be called under dma_list_mutex + */ +static void dma_chan_put(struct dma_chan *chan) +{ + if (!chan->client_count) + return; /* this channel failed alloc_chan_resources */ + chan->client_count--; + module_put(dma_chan_to_owner(chan)); + if (chan->client_count == 0) + chan->device->device_free_chan_resources(chan); } enum dma_status dma_sync_wait(struct dma_chan *chan, dma_cookie_t cookie) @@ -186,148 +270,397 @@ enum dma_status dma_sync_wait(struct dma_chan *chan, dma_cookie_t cookie) EXPORT_SYMBOL(dma_sync_wait); /** - * dma_chan_cleanup - release a DMA channel's resources - * @kref: kernel reference structure that contains the DMA channel device + * dma_cap_mask_all - enable iteration over all operation types + */ +static dma_cap_mask_t dma_cap_mask_all; + +/** + * dma_chan_tbl_ent - tracks channel allocations per core/operation + * @chan - associated channel for this entry + */ +struct dma_chan_tbl_ent { + struct dma_chan *chan; +}; + +/** + * channel_table - percpu lookup table for memory-to-memory offload providers */ -void dma_chan_cleanup(struct kref *kref) +static struct dma_chan_tbl_ent __percpu *channel_table[DMA_TX_TYPE_END]; + +static int __init dma_channel_table_init(void) { - struct dma_chan *chan = container_of(kref, struct dma_chan, refcount); - chan->device->device_free_chan_resources(chan); - chan->client = NULL; - kref_put(&chan->device->refcount, dma_async_device_cleanup); + enum dma_transaction_type cap; + int err = 0; + + bitmap_fill(dma_cap_mask_all.bits, DMA_TX_TYPE_END); + + /* 'interrupt', 'private', and 'slave' are channel capabilities, + * but are not associated with an operation so they do not need + * an entry in the channel_table + */ + clear_bit(DMA_INTERRUPT, dma_cap_mask_all.bits); + clear_bit(DMA_PRIVATE, dma_cap_mask_all.bits); + clear_bit(DMA_SLAVE, dma_cap_mask_all.bits); + + for_each_dma_cap_mask(cap, dma_cap_mask_all) { + channel_table[cap] = alloc_percpu(struct dma_chan_tbl_ent); + if (!channel_table[cap]) { + err = -ENOMEM; + break; + } + } + + if (err) { + pr_err("dmaengine: initialization failure\n"); + for_each_dma_cap_mask(cap, dma_cap_mask_all) + if (channel_table[cap]) + free_percpu(channel_table[cap]); + } + + return err; } -EXPORT_SYMBOL(dma_chan_cleanup); +arch_initcall(dma_channel_table_init); -static void dma_chan_free_rcu(struct rcu_head *rcu) +/** + * dma_find_channel - find a channel to carry out the operation + * @tx_type: transaction type + */ +struct dma_chan *dma_find_channel(enum dma_transaction_type tx_type) { - struct dma_chan *chan = container_of(rcu, struct dma_chan, rcu); - int bias = 0x7FFFFFFF; - int i; - for_each_possible_cpu(i) - bias -= local_read(&per_cpu_ptr(chan->local, i)->refcount); - atomic_sub(bias, &chan->refcount.refcount); - kref_put(&chan->refcount, dma_chan_cleanup); + return this_cpu_read(channel_table[tx_type]->chan); } +EXPORT_SYMBOL(dma_find_channel); -static void dma_client_chan_free(struct dma_chan *chan) +/** + * dma_issue_pending_all - flush all pending operations across all channels + */ +void dma_issue_pending_all(void) { - atomic_add(0x7FFFFFFF, &chan->refcount.refcount); - chan->slow_ref = 1; - call_rcu(&chan->rcu, dma_chan_free_rcu); + struct dma_device *device; + struct dma_chan *chan; + + rcu_read_lock(); + list_for_each_entry_rcu(device, &dma_device_list, global_node) { + if (dma_has_cap(DMA_PRIVATE, device->cap_mask)) + continue; + list_for_each_entry(chan, &device->channels, device_node) + if (chan->client_count) + device->device_issue_pending(chan); + } + rcu_read_unlock(); } +EXPORT_SYMBOL(dma_issue_pending_all); /** - * dma_chans_rebalance - reallocate channels to clients + * nth_chan - returns the nth channel of the given capability + * @cap: capability to match + * @n: nth channel desired * - * When the number of DMA channel in the system changes, - * channels need to be rebalanced among clients. + * Defaults to returning the channel with the desired capability and the + * lowest reference count when 'n' cannot be satisfied. Must be called + * under dma_list_mutex. */ -static void dma_chans_rebalance(void) +static struct dma_chan *nth_chan(enum dma_transaction_type cap, int n) { - struct dma_client *client; + struct dma_device *device; struct dma_chan *chan; - unsigned long flags; - - mutex_lock(&dma_list_mutex); + struct dma_chan *ret = NULL; + struct dma_chan *min = NULL; - list_for_each_entry(client, &dma_client_list, global_node) { - while (client->chans_desired > client->chan_count) { - chan = dma_client_chan_alloc(client); - if (!chan) - break; - client->chan_count++; - client->event_callback(client, - chan, - DMA_RESOURCE_ADDED); - } - while (client->chans_desired < client->chan_count) { - spin_lock_irqsave(&client->lock, flags); - chan = list_entry(client->channels.next, - struct dma_chan, - client_node); - list_del_rcu(&chan->client_node); - spin_unlock_irqrestore(&client->lock, flags); - client->chan_count--; - client->event_callback(client, - chan, - DMA_RESOURCE_REMOVED); - dma_client_chan_free(chan); + list_for_each_entry(device, &dma_device_list, global_node) { + if (!dma_has_cap(cap, device->cap_mask) || + dma_has_cap(DMA_PRIVATE, device->cap_mask)) + continue; + list_for_each_entry(chan, &device->channels, device_node) { + if (!chan->client_count) + continue; + if (!min) + min = chan; + else if (chan->table_count < min->table_count) + min = chan; + + if (n-- == 0) { + ret = chan; + break; /* done */ + } } + if (ret) + break; /* done */ } - mutex_unlock(&dma_list_mutex); + if (!ret) + ret = min; + + if (ret) + ret->table_count++; + + return ret; } /** - * dma_async_client_register - allocate and register a &dma_client - * @event_callback: callback for notification of channel addition/removal + * dma_channel_rebalance - redistribute the available channels + * + * Optimize for cpu isolation (each cpu gets a dedicated channel for an + * operation type) in the SMP case, and operation isolation (avoid + * multi-tasking channels) in the non-SMP case. Must be called under + * dma_list_mutex. */ -struct dma_client *dma_async_client_register(dma_event_callback event_callback) +static void dma_channel_rebalance(void) { - struct dma_client *client; + struct dma_chan *chan; + struct dma_device *device; + int cpu; + int cap; + int n; + + /* undo the last distribution */ + for_each_dma_cap_mask(cap, dma_cap_mask_all) + for_each_possible_cpu(cpu) + per_cpu_ptr(channel_table[cap], cpu)->chan = NULL; + + list_for_each_entry(device, &dma_device_list, global_node) { + if (dma_has_cap(DMA_PRIVATE, device->cap_mask)) + continue; + list_for_each_entry(chan, &device->channels, device_node) + chan->table_count = 0; + } + + /* don't populate the channel_table if no clients are available */ + if (!dmaengine_ref_count) + return; + + /* redistribute available channels */ + n = 0; + for_each_dma_cap_mask(cap, dma_cap_mask_all) + for_each_online_cpu(cpu) { + if (num_possible_cpus() > 1) + chan = nth_chan(cap, n++); + else + chan = nth_chan(cap, -1); - client = kzalloc(sizeof(*client), GFP_KERNEL); - if (!client) + per_cpu_ptr(channel_table[cap], cpu)->chan = chan; + } +} + +static struct dma_chan *private_candidate(dma_cap_mask_t *mask, struct dma_device *dev, + dma_filter_fn fn, void *fn_param) +{ + struct dma_chan *chan; + + if (!__dma_device_satisfies_mask(dev, mask)) { + pr_debug("%s: wrong capabilities\n", __func__); return NULL; + } + /* devices with multiple channels need special handling as we need to + * ensure that all channels are either private or public. + */ + if (dev->chancnt > 1 && !dma_has_cap(DMA_PRIVATE, dev->cap_mask)) + list_for_each_entry(chan, &dev->channels, device_node) { + /* some channels are already publicly allocated */ + if (chan->client_count) + return NULL; + } + + list_for_each_entry(chan, &dev->channels, device_node) { + if (chan->client_count) { + pr_debug("%s: %s busy\n", + __func__, dma_chan_name(chan)); + continue; + } + if (fn && !fn(chan, fn_param)) { + pr_debug("%s: %s filter said false\n", + __func__, dma_chan_name(chan)); + continue; + } + return chan; + } - INIT_LIST_HEAD(&client->channels); - spin_lock_init(&client->lock); - client->chans_desired = 0; - client->chan_count = 0; - client->event_callback = event_callback; + return NULL; +} + +/** + * dma_request_channel - try to allocate an exclusive channel + * @mask: capabilities that the channel must satisfy + * @fn: optional callback to disposition available channels + * @fn_param: opaque parameter to pass to dma_filter_fn + */ +struct dma_chan *__dma_request_channel(dma_cap_mask_t *mask, dma_filter_fn fn, void *fn_param) +{ + struct dma_device *device, *_d; + struct dma_chan *chan = NULL; + int err; + /* Find a channel */ mutex_lock(&dma_list_mutex); - list_add_tail(&client->global_node, &dma_client_list); + list_for_each_entry_safe(device, _d, &dma_device_list, global_node) { + chan = private_candidate(mask, device, fn, fn_param); + if (chan) { + /* Found a suitable channel, try to grab, prep, and + * return it. We first set DMA_PRIVATE to disable + * balance_ref_count as this channel will not be + * published in the general-purpose allocator + */ + dma_cap_set(DMA_PRIVATE, device->cap_mask); + device->privatecnt++; + err = dma_chan_get(chan); + + if (err == -ENODEV) { + pr_debug("%s: %s module removed\n", __func__, + dma_chan_name(chan)); + list_del_rcu(&device->global_node); + } else if (err) + pr_err("dmaengine: failed to get %s: (%d)\n", + dma_chan_name(chan), err); + else + break; + if (--device->privatecnt == 0) + dma_cap_clear(DMA_PRIVATE, device->cap_mask); + chan = NULL; + } + } mutex_unlock(&dma_list_mutex); - return client; + pr_debug("%s: %s (%s)\n", __func__, chan ? "success" : "fail", + chan ? dma_chan_name(chan) : NULL); + + return chan; +} +EXPORT_SYMBOL_GPL(__dma_request_channel); + +void dma_release_channel(struct dma_chan *chan) +{ + mutex_lock(&dma_list_mutex); + WARN_ONCE(chan->client_count != 1, + "chan reference count %d != 1\n", chan->client_count); + dma_chan_put(chan); + /* drop PRIVATE cap enabled by __dma_request_channel() */ + if (--chan->device->privatecnt == 0) + dma_cap_clear(DMA_PRIVATE, chan->device->cap_mask); + mutex_unlock(&dma_list_mutex); } -EXPORT_SYMBOL(dma_async_client_register); +EXPORT_SYMBOL_GPL(dma_release_channel); /** - * dma_async_client_unregister - unregister a client and free the &dma_client - * @client: &dma_client to free - * - * Force frees any allocated DMA channels, frees the &dma_client memory + * dmaengine_get - register interest in dma_channels */ -void dma_async_client_unregister(struct dma_client *client) +void dmaengine_get(void) { + struct dma_device *device, *_d; struct dma_chan *chan; + int err; - if (!client) - return; + mutex_lock(&dma_list_mutex); + dmaengine_ref_count++; - rcu_read_lock(); - list_for_each_entry_rcu(chan, &client->channels, client_node) - dma_client_chan_free(chan); - rcu_read_unlock(); + /* try to grab channels */ + list_for_each_entry_safe(device, _d, &dma_device_list, global_node) { + if (dma_has_cap(DMA_PRIVATE, device->cap_mask)) + continue; + list_for_each_entry(chan, &device->channels, device_node) { + err = dma_chan_get(chan); + if (err == -ENODEV) { + /* module removed before we could use it */ + list_del_rcu(&device->global_node); + break; + } else if (err) + pr_err("dmaengine: failed to get %s: (%d)\n", + dma_chan_name(chan), err); + } + } - mutex_lock(&dma_list_mutex); - list_del(&client->global_node); + /* if this is the first reference and there were channels + * waiting we need to rebalance to get those channels + * incorporated into the channel table + */ + if (dmaengine_ref_count == 1) + dma_channel_rebalance(); mutex_unlock(&dma_list_mutex); - - kfree(client); - dma_chans_rebalance(); } -EXPORT_SYMBOL(dma_async_client_unregister); +EXPORT_SYMBOL(dmaengine_get); /** - * dma_async_client_chan_request - request DMA channels - * @client: &dma_client - * @number: count of DMA channels requested - * - * Clients call dma_async_client_chan_request() to specify how many - * DMA channels they need, 0 to free all currently allocated. - * The resulting allocations/frees are indicated to the client via the - * event callback. + * dmaengine_put - let dma drivers be removed when ref_count == 0 */ -void dma_async_client_chan_request(struct dma_client *client, - unsigned int number) +void dmaengine_put(void) +{ + struct dma_device *device; + struct dma_chan *chan; + + mutex_lock(&dma_list_mutex); + dmaengine_ref_count--; + BUG_ON(dmaengine_ref_count < 0); + /* drop channel references */ + list_for_each_entry(device, &dma_device_list, global_node) { + if (dma_has_cap(DMA_PRIVATE, device->cap_mask)) + continue; + list_for_each_entry(chan, &device->channels, device_node) + dma_chan_put(chan); + } + mutex_unlock(&dma_list_mutex); +} +EXPORT_SYMBOL(dmaengine_put); + +static bool device_has_all_tx_types(struct dma_device *device) { - client->chans_desired = number; - dma_chans_rebalance(); + /* A device that satisfies this test has channels that will never cause + * an async_tx channel switch event as all possible operation types can + * be handled. + */ + #ifdef CONFIG_ASYNC_TX_DMA + if (!dma_has_cap(DMA_INTERRUPT, device->cap_mask)) + return false; + #endif + + #if defined(CONFIG_ASYNC_MEMCPY) || defined(CONFIG_ASYNC_MEMCPY_MODULE) + if (!dma_has_cap(DMA_MEMCPY, device->cap_mask)) + return false; + #endif + + #if defined(CONFIG_ASYNC_MEMSET) || defined(CONFIG_ASYNC_MEMSET_MODULE) + if (!dma_has_cap(DMA_MEMSET, device->cap_mask)) + return false; + #endif + + #if defined(CONFIG_ASYNC_XOR) || defined(CONFIG_ASYNC_XOR_MODULE) + if (!dma_has_cap(DMA_XOR, device->cap_mask)) + return false; + + #ifndef CONFIG_ASYNC_TX_DISABLE_XOR_VAL_DMA + if (!dma_has_cap(DMA_XOR_VAL, device->cap_mask)) + return false; + #endif + #endif + + #if defined(CONFIG_ASYNC_PQ) || defined(CONFIG_ASYNC_PQ_MODULE) + if (!dma_has_cap(DMA_PQ, device->cap_mask)) + return false; + + #ifndef CONFIG_ASYNC_TX_DISABLE_PQ_VAL_DMA + if (!dma_has_cap(DMA_PQ_VAL, device->cap_mask)) + return false; + #endif + #endif + + return true; +} + +static int get_dma_id(struct dma_device *device) +{ + int rc; + + idr_retry: + if (!idr_pre_get(&dma_idr, GFP_KERNEL)) + return -ENOMEM; + mutex_lock(&dma_list_mutex); + rc = idr_get_new(&dma_idr, NULL, &device->dev_id); + mutex_unlock(&dma_list_mutex); + if (rc == -EAGAIN) + goto idr_retry; + else if (rc != 0) + return rc; + + return 0; } -EXPORT_SYMBOL(dma_async_client_chan_request); /** * dma_async_device_register - registers DMA devices found @@ -335,9 +668,9 @@ EXPORT_SYMBOL(dma_async_client_chan_request); */ int dma_async_device_register(struct dma_device *device) { - static int id; int chancnt = 0, rc; struct dma_chan* chan; + atomic_t *idr_ref; if (!device) return -ENODEV; @@ -347,62 +680,121 @@ int dma_async_device_register(struct dma_device *device) !device->device_prep_dma_memcpy); BUG_ON(dma_has_cap(DMA_XOR, device->cap_mask) && !device->device_prep_dma_xor); - BUG_ON(dma_has_cap(DMA_ZERO_SUM, device->cap_mask) && - !device->device_prep_dma_zero_sum); + BUG_ON(dma_has_cap(DMA_XOR_VAL, device->cap_mask) && + !device->device_prep_dma_xor_val); + BUG_ON(dma_has_cap(DMA_PQ, device->cap_mask) && + !device->device_prep_dma_pq); + BUG_ON(dma_has_cap(DMA_PQ_VAL, device->cap_mask) && + !device->device_prep_dma_pq_val); BUG_ON(dma_has_cap(DMA_MEMSET, device->cap_mask) && !device->device_prep_dma_memset); - BUG_ON(dma_has_cap(DMA_ZERO_SUM, device->cap_mask) && + BUG_ON(dma_has_cap(DMA_INTERRUPT, device->cap_mask) && !device->device_prep_dma_interrupt); + BUG_ON(dma_has_cap(DMA_SLAVE, device->cap_mask) && + !device->device_prep_slave_sg); + BUG_ON(dma_has_cap(DMA_SLAVE, device->cap_mask) && + !device->device_control); BUG_ON(!device->device_alloc_chan_resources); BUG_ON(!device->device_free_chan_resources); - BUG_ON(!device->device_dependency_added); - BUG_ON(!device->device_is_tx_complete); + BUG_ON(!device->device_tx_status); BUG_ON(!device->device_issue_pending); BUG_ON(!device->dev); - init_completion(&device->done); - kref_init(&device->refcount); - device->dev_id = id++; + /* note: this only matters in the + * CONFIG_ASYNC_TX_DISABLE_CHANNEL_SWITCH=y case + */ + if (device_has_all_tx_types(device)) + dma_cap_set(DMA_ASYNC_TX, device->cap_mask); + + idr_ref = kmalloc(sizeof(*idr_ref), GFP_KERNEL); + if (!idr_ref) + return -ENOMEM; + rc = get_dma_id(device); + if (rc != 0) { + kfree(idr_ref); + return rc; + } + + atomic_set(idr_ref, 0); /* represent channels in sysfs. Probably want devs too */ list_for_each_entry(chan, &device->channels, device_node) { + rc = -ENOMEM; chan->local = alloc_percpu(typeof(*chan->local)); if (chan->local == NULL) - continue; + goto err_out; + chan->dev = kzalloc(sizeof(*chan->dev), GFP_KERNEL); + if (chan->dev == NULL) { + free_percpu(chan->local); + chan->local = NULL; + goto err_out; + } chan->chan_id = chancnt++; - chan->class_dev.class = &dma_devclass; - chan->class_dev.dev = NULL; - snprintf(chan->class_dev.class_id, BUS_ID_SIZE, "dma%dchan%d", - device->dev_id, chan->chan_id); - - rc = class_device_register(&chan->class_dev); + chan->dev->device.class = &dma_devclass; + chan->dev->device.parent = device->dev; + chan->dev->chan = chan; + chan->dev->idr_ref = idr_ref; + chan->dev->dev_id = device->dev_id; + atomic_inc(idr_ref); + dev_set_name(&chan->dev->device, "dma%dchan%d", + device->dev_id, chan->chan_id); + + rc = device_register(&chan->dev->device); if (rc) { - chancnt--; free_percpu(chan->local); chan->local = NULL; + kfree(chan->dev); + atomic_dec(idr_ref); goto err_out; } - - kref_get(&device->refcount); + chan->client_count = 0; } + device->chancnt = chancnt; mutex_lock(&dma_list_mutex); - list_add_tail(&device->global_node, &dma_device_list); + /* take references on public channels */ + if (dmaengine_ref_count && !dma_has_cap(DMA_PRIVATE, device->cap_mask)) + list_for_each_entry(chan, &device->channels, device_node) { + /* if clients are already waiting for channels we need + * to take references on their behalf + */ + if (dma_chan_get(chan) == -ENODEV) { + /* note we can only get here for the first + * channel as the remaining channels are + * guaranteed to get a reference + */ + rc = -ENODEV; + mutex_unlock(&dma_list_mutex); + goto err_out; + } + } + list_add_tail_rcu(&device->global_node, &dma_device_list); + if (dma_has_cap(DMA_PRIVATE, device->cap_mask)) + device->privatecnt++; /* Always private */ + dma_channel_rebalance(); mutex_unlock(&dma_list_mutex); - dma_chans_rebalance(); - return 0; err_out: + /* if we never registered a channel just release the idr */ + if (atomic_read(idr_ref) == 0) { + mutex_lock(&dma_list_mutex); + idr_remove(&dma_idr, device->dev_id); + mutex_unlock(&dma_list_mutex); + kfree(idr_ref); + return rc; + } + list_for_each_entry(chan, &device->channels, device_node) { if (chan->local == NULL) continue; - kref_put(&device->refcount, dma_async_device_cleanup); - class_device_unregister(&chan->class_dev); - chancnt--; + mutex_lock(&dma_list_mutex); + chan->dev->chan = NULL; + mutex_unlock(&dma_list_mutex); + device_unregister(&chan->dev->device); free_percpu(chan->local); } return rc; @@ -410,47 +802,31 @@ err_out: EXPORT_SYMBOL(dma_async_device_register); /** - * dma_async_device_cleanup - function called when all references are released - * @kref: kernel reference object - */ -static void dma_async_device_cleanup(struct kref *kref) -{ - struct dma_device *device; - - device = container_of(kref, struct dma_device, refcount); - complete(&device->done); -} - -/** - * dma_async_device_unregister - unregisters DMA devices + * dma_async_device_unregister - unregister a DMA device * @device: &dma_device + * + * This routine is called by dma driver exit routines, dmaengine holds module + * references to prevent it being called while channels are in use. */ void dma_async_device_unregister(struct dma_device *device) { struct dma_chan *chan; - unsigned long flags; mutex_lock(&dma_list_mutex); - list_del(&device->global_node); + list_del_rcu(&device->global_node); + dma_channel_rebalance(); mutex_unlock(&dma_list_mutex); list_for_each_entry(chan, &device->channels, device_node) { - if (chan->client) { - spin_lock_irqsave(&chan->client->lock, flags); - list_del(&chan->client_node); - chan->client->chan_count--; - spin_unlock_irqrestore(&chan->client->lock, flags); - chan->client->event_callback(chan->client, - chan, - DMA_RESOURCE_REMOVED); - dma_client_chan_free(chan); - } - class_device_unregister(&chan->class_dev); + WARN_ONCE(chan->client_count, + "%s called while %d clients hold a reference\n", + __func__, chan->client_count); + mutex_lock(&dma_list_mutex); + chan->dev->chan = NULL; + mutex_unlock(&dma_list_mutex); + device_unregister(&chan->dev->device); + free_percpu(chan->local); } - dma_chans_rebalance(); - - kref_put(&device->refcount, dma_async_device_cleanup); - wait_for_completion(&device->done); } EXPORT_SYMBOL(dma_async_device_unregister); @@ -472,26 +848,30 @@ dma_async_memcpy_buf_to_buf(struct dma_chan *chan, void *dest, { struct dma_device *dev = chan->device; struct dma_async_tx_descriptor *tx; - dma_addr_t addr; + dma_addr_t dma_dest, dma_src; dma_cookie_t cookie; - int cpu; + unsigned long flags; - tx = dev->device_prep_dma_memcpy(chan, len, 0); - if (!tx) + dma_src = dma_map_single(dev->dev, src, len, DMA_TO_DEVICE); + dma_dest = dma_map_single(dev->dev, dest, len, DMA_FROM_DEVICE); + flags = DMA_CTRL_ACK | + DMA_COMPL_SRC_UNMAP_SINGLE | + DMA_COMPL_DEST_UNMAP_SINGLE; + tx = dev->device_prep_dma_memcpy(chan, dma_dest, dma_src, len, flags); + + if (!tx) { + dma_unmap_single(dev->dev, dma_src, len, DMA_TO_DEVICE); + dma_unmap_single(dev->dev, dma_dest, len, DMA_FROM_DEVICE); return -ENOMEM; + } - tx->ack = 1; tx->callback = NULL; - addr = dma_map_single(dev->dev, src, len, DMA_TO_DEVICE); - tx->tx_set_src(addr, tx, 0); - addr = dma_map_single(dev->dev, dest, len, DMA_FROM_DEVICE); - tx->tx_set_dest(addr, tx, 0); cookie = tx->tx_submit(tx); - cpu = get_cpu(); - per_cpu_ptr(chan->local, cpu)->bytes_transferred += len; - per_cpu_ptr(chan->local, cpu)->memcpy_count++; - put_cpu(); + preempt_disable(); + __this_cpu_add(chan->local->bytes_transferred, len); + __this_cpu_inc(chan->local->memcpy_count); + preempt_enable(); return cookie; } @@ -516,26 +896,28 @@ dma_async_memcpy_buf_to_pg(struct dma_chan *chan, struct page *page, { struct dma_device *dev = chan->device; struct dma_async_tx_descriptor *tx; - dma_addr_t addr; + dma_addr_t dma_dest, dma_src; dma_cookie_t cookie; - int cpu; + unsigned long flags; - tx = dev->device_prep_dma_memcpy(chan, len, 0); - if (!tx) + dma_src = dma_map_single(dev->dev, kdata, len, DMA_TO_DEVICE); + dma_dest = dma_map_page(dev->dev, page, offset, len, DMA_FROM_DEVICE); + flags = DMA_CTRL_ACK | DMA_COMPL_SRC_UNMAP_SINGLE; + tx = dev->device_prep_dma_memcpy(chan, dma_dest, dma_src, len, flags); + + if (!tx) { + dma_unmap_single(dev->dev, dma_src, len, DMA_TO_DEVICE); + dma_unmap_page(dev->dev, dma_dest, len, DMA_FROM_DEVICE); return -ENOMEM; + } - tx->ack = 1; tx->callback = NULL; - addr = dma_map_single(dev->dev, kdata, len, DMA_TO_DEVICE); - tx->tx_set_src(addr, tx, 0); - addr = dma_map_page(dev->dev, page, offset, len, DMA_FROM_DEVICE); - tx->tx_set_dest(addr, tx, 0); cookie = tx->tx_submit(tx); - cpu = get_cpu(); - per_cpu_ptr(chan->local, cpu)->bytes_transferred += len; - per_cpu_ptr(chan->local, cpu)->memcpy_count++; - put_cpu(); + preempt_disable(); + __this_cpu_add(chan->local->bytes_transferred, len); + __this_cpu_inc(chan->local->memcpy_count); + preempt_enable(); return cookie; } @@ -562,26 +944,29 @@ dma_async_memcpy_pg_to_pg(struct dma_chan *chan, struct page *dest_pg, { struct dma_device *dev = chan->device; struct dma_async_tx_descriptor *tx; - dma_addr_t addr; + dma_addr_t dma_dest, dma_src; dma_cookie_t cookie; - int cpu; + unsigned long flags; - tx = dev->device_prep_dma_memcpy(chan, len, 0); - if (!tx) + dma_src = dma_map_page(dev->dev, src_pg, src_off, len, DMA_TO_DEVICE); + dma_dest = dma_map_page(dev->dev, dest_pg, dest_off, len, + DMA_FROM_DEVICE); + flags = DMA_CTRL_ACK; + tx = dev->device_prep_dma_memcpy(chan, dma_dest, dma_src, len, flags); + + if (!tx) { + dma_unmap_page(dev->dev, dma_src, len, DMA_TO_DEVICE); + dma_unmap_page(dev->dev, dma_dest, len, DMA_FROM_DEVICE); return -ENOMEM; + } - tx->ack = 1; tx->callback = NULL; - addr = dma_map_page(dev->dev, src_pg, src_off, len, DMA_TO_DEVICE); - tx->tx_set_src(addr, tx, 0); - addr = dma_map_page(dev->dev, dest_pg, dest_off, len, DMA_FROM_DEVICE); - tx->tx_set_dest(addr, tx, 0); cookie = tx->tx_submit(tx); - cpu = get_cpu(); - per_cpu_ptr(chan->local, cpu)->bytes_transferred += len; - per_cpu_ptr(chan->local, cpu)->memcpy_count++; - put_cpu(); + preempt_disable(); + __this_cpu_add(chan->local->bytes_transferred, len); + __this_cpu_inc(chan->local->memcpy_count); + preempt_enable(); return cookie; } @@ -591,16 +976,79 @@ void dma_async_tx_descriptor_init(struct dma_async_tx_descriptor *tx, struct dma_chan *chan) { tx->chan = chan; + #ifndef CONFIG_ASYNC_TX_DISABLE_CHANNEL_SWITCH spin_lock_init(&tx->lock); - INIT_LIST_HEAD(&tx->depend_node); - INIT_LIST_HEAD(&tx->depend_list); + #endif } EXPORT_SYMBOL(dma_async_tx_descriptor_init); +/* dma_wait_for_async_tx - spin wait for a transaction to complete + * @tx: in-flight transaction to wait on + */ +enum dma_status +dma_wait_for_async_tx(struct dma_async_tx_descriptor *tx) +{ + unsigned long dma_sync_wait_timeout = jiffies + msecs_to_jiffies(5000); + + if (!tx) + return DMA_SUCCESS; + + while (tx->cookie == -EBUSY) { + if (time_after_eq(jiffies, dma_sync_wait_timeout)) { + pr_err("%s timeout waiting for descriptor submission\n", + __func__); + return DMA_ERROR; + } + cpu_relax(); + } + return dma_sync_wait(tx->chan, tx->cookie); +} +EXPORT_SYMBOL_GPL(dma_wait_for_async_tx); + +/* dma_run_dependencies - helper routine for dma drivers to process + * (start) dependent operations on their target channel + * @tx: transaction with dependencies + */ +void dma_run_dependencies(struct dma_async_tx_descriptor *tx) +{ + struct dma_async_tx_descriptor *dep = txd_next(tx); + struct dma_async_tx_descriptor *dep_next; + struct dma_chan *chan; + + if (!dep) + return; + + /* we'll submit tx->next now, so clear the link */ + txd_clear_next(tx); + chan = dep->chan; + + /* keep submitting up until a channel switch is detected + * in that case we will be called again as a result of + * processing the interrupt from async_tx_channel_switch + */ + for (; dep; dep = dep_next) { + txd_lock(dep); + txd_clear_parent(dep); + dep_next = txd_next(dep); + if (dep_next && dep_next->chan == chan) + txd_clear_next(dep); /* ->next will be submitted */ + else + dep_next = NULL; /* submit current dep and terminate */ + txd_unlock(dep); + + dep->tx_submit(dep); + } + + chan->device->device_issue_pending(chan); +} +EXPORT_SYMBOL_GPL(dma_run_dependencies); + static int __init dma_bus_init(void) { + idr_init(&dma_idr); mutex_init(&dma_list_mutex); return class_register(&dma_devclass); } -subsys_initcall(dma_bus_init); +arch_initcall(dma_bus_init); +