Merge tag 'v3.9-rc5' into wq/for-3.10

Writeback conversion to workqueue will be based on top of wq/for-3.10
branch to take advantage of custom attrs and NUMA support for unbound
workqueues.  Mainline currently contains two commits which result in
non-trivial merge conflicts with wq/for-3.10 and because
block/for-3.10/core is based on v3.9-rc3 which contains one of the
conflicting commits, we need a pre-merge-window merge anyway.  Let's
pull v3.9-rc5 into wq/for-3.10 so that the block tree doesn't suffer
from workqueue merge conflicts.

The two conflicts and their resolutions:

* e68035fb65 ("workqueue: convert to idr_alloc()") in mainline changes
  worker_pool_assign_id() to use idr_alloc() instead of the old idr
  interface.  worker_pool_assign_id() goes through multiple locking
  changes in wq/for-3.10 causing the following conflict.

  static int worker_pool_assign_id(struct worker_pool *pool)
  {
	  int ret;

  <<<<<<< HEAD
	  lockdep_assert_held(&wq_pool_mutex);

	  do {
		  if (!idr_pre_get(&worker_pool_idr, GFP_KERNEL))
			  return -ENOMEM;
		  ret = idr_get_new(&worker_pool_idr, pool, &pool->id);
	  } while (ret == -EAGAIN);
  =======
	  mutex_lock(&worker_pool_idr_mutex);
	  ret = idr_alloc(&worker_pool_idr, pool, 0, 0, GFP_KERNEL);
	  if (ret >= 0)
		  pool->id = ret;
	  mutex_unlock(&worker_pool_idr_mutex);
  >>>>>>> c67bf5361e7e66a0ff1f4caf95f89347d55dfb89

	  return ret < 0 ? ret : 0;
  }

  We want locking from the former and idr_alloc() usage from the
  latter, which can be combined to the following.

  static int worker_pool_assign_id(struct worker_pool *pool)
  {
	  int ret;

	  lockdep_assert_held(&wq_pool_mutex);

	  ret = idr_alloc(&worker_pool_idr, pool, 0, 0, GFP_KERNEL);
	  if (ret >= 0) {
		  pool->id = ret;
		  return 0;
	  }
	  return ret;
   }

* eb2834285c ("workqueue: fix possible pool stall bug in
  wq_unbind_fn()") updated wq_unbind_fn() such that it has single
  larger for_each_std_worker_pool() loop instead of two separate loops
  with a schedule() call inbetween.  wq/for-3.10 renamed
  pool->assoc_mutex to pool->manager_mutex causing the following
  conflict (earlier function body and comments omitted for brevity).

  static void wq_unbind_fn(struct work_struct *work)
  {
  ...
		  spin_unlock_irq(&pool->lock);
  <<<<<<< HEAD
		  mutex_unlock(&pool->manager_mutex);
	  }
  =======
		  mutex_unlock(&pool->assoc_mutex);
  >>>>>>> c67bf5361e7e66a0ff1f4caf95f89347d55dfb89

		  schedule();

  <<<<<<< HEAD
	  for_each_cpu_worker_pool(pool, cpu)
  =======
  >>>>>>> c67bf5361e7e66a0ff1f4caf95f89347d55dfb89
		  atomic_set(&pool->nr_running, 0);

		  spin_lock_irq(&pool->lock);
		  wake_up_worker(pool);
		  spin_unlock_irq(&pool->lock);
	  }
  }

  The resolution is mostly trivial.  We want the control flow of the
  latter with the rename of the former.

  static void wq_unbind_fn(struct work_struct *work)
  {
  ...
		  spin_unlock_irq(&pool->lock);
		  mutex_unlock(&pool->manager_mutex);

		  schedule();

		  atomic_set(&pool->nr_running, 0);

		  spin_lock_irq(&pool->lock);
		  wake_up_worker(pool);
		  spin_unlock_irq(&pool->lock);
	  }
  }

Signed-off-by: Tejun Heo <tj@kernel.org>

1 files changed, 30 insertions, 5 deletions

diff --git a/drivers/net/wireless/iwlwifi/pcie/internal.h b/drivers/net/wireless/iwlwifi/pcie/internal.h
index aa2a39a637d..148843e7f34 100644
--- a/drivers/net/wireless/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/iwlwifi/pcie/internal.h
@@ -137,10 +137,6 @@ static inline int iwl_queue_dec_wrap(int index, int n_bd)
 struct iwl_cmd_meta {
 	/* only for SYNC commands, iff the reply skb is wanted */
 	struct iwl_host_cmd *source;
-
-	DEFINE_DMA_UNMAP_ADDR(mapping);
-	DEFINE_DMA_UNMAP_LEN(len);
-
 	u32 flags;
 };
 
@@ -182,19 +178,39 @@ struct iwl_queue {
 #define TFD_TX_CMD_SLOTS 256
 #define TFD_CMD_SLOTS 32
 
+/*
+ * The FH will write back to the first TB only, so we need
+ * to copy some data into the buffer regardless of whether
+ * it should be mapped or not. This indicates how big the
+ * first TB must be to include the scratch buffer. Since
+ * the scratch is 4 bytes at offset 12, it's 16 now. If we
+ * make it bigger then allocations will be bigger and copy
+ * slower, so that's probably not useful.
+ */
+#define IWL_HCMD_SCRATCHBUF_SIZE	16
+
 struct iwl_pcie_txq_entry {
 	struct iwl_device_cmd *cmd;
-	struct iwl_device_cmd *copy_cmd;
 	struct sk_buff *skb;
 	/* buffer to free after command completes */
 	const void *free_buf;
 	struct iwl_cmd_meta meta;
 };
 
+struct iwl_pcie_txq_scratch_buf {
+	struct iwl_cmd_header hdr;
+	u8 buf[8];
+	__le32 scratch;
+};
+
 /**
  * struct iwl_txq - Tx Queue for DMA
  * @q: generic Rx/Tx queue descriptor
  * @tfds: transmit frame descriptors (DMA memory)
+ * @scratchbufs: start of command headers, including scratch buffers, for
+ *	the writeback -- this is DMA memory and an array holding one buffer
+ *	for each command on the queue
+ * @scratchbufs_dma: DMA address for the scratchbufs start
  * @entries: transmit entries (driver state)
  * @lock: queue lock
  * @stuck_timer: timer that fires if queue gets stuck
@@ -208,6 +224,8 @@ struct iwl_pcie_txq_entry {
 struct iwl_txq {
 	struct iwl_queue q;
 	struct iwl_tfd *tfds;
+	struct iwl_pcie_txq_scratch_buf *scratchbufs;
+	dma_addr_t scratchbufs_dma;
 	struct iwl_pcie_txq_entry *entries;
 	spinlock_t lock;
 	struct timer_list stuck_timer;
@@ -216,6 +234,13 @@ struct iwl_txq {
 	u8 active;
 };
 
+static inline dma_addr_t
+iwl_pcie_get_scratchbuf_dma(struct iwl_txq *txq, int idx)
+{
+	return txq->scratchbufs_dma +
+	       sizeof(struct iwl_pcie_txq_scratch_buf) * idx;
+}
+
 /**
  * struct iwl_trans_pcie - PCIe transport specific data
  * @rxq: all the RX queue data