iwlagn: support multiple TBs per command

The current "huge" command handling is a bit
confusing, and very limited since only one
command may be huge at a time. Additionally,
we often copy data around quite pointlessly
since we could instead map the existing scan
buffer for example and use it directly.

This patch makes that possible. The first
change is that multiple buffers may be given
to each command (this change was prepared
earlier so callsites don't need to change).
Each of those can be mapped attached to a TB
in the TFD, and the command header can use a
TB (the first one) in the TFD as well.

Doing this allows getting rid of huge commands
in favour of mapping existing buffers. The
beacon transmission is also optimised to not
copy the SKB at all but use multiple TBs.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com>

1 files changed, 28 insertions, 18 deletions

diff --git a/drivers/net/wireless/iwlwifi/iwl-dev.h b/drivers/net/wireless/iwlwifi/iwl-dev.h
index 38254bdfabbb..3e3b8b8939d6 100644
--- a/drivers/net/wireless/iwlwifi/iwl-dev.h
+++ b/drivers/net/wireless/iwlwifi/iwl-dev.h
@@ -110,8 +110,6 @@ struct iwl_cmd_meta {
                          struct iwl_device_cmd *cmd,
                          struct iwl_rx_packet *pkt);
 
-        /* The CMD_SIZE_HUGE flag bit indicates that the command
-         * structure is stored at the end of the shared queue memory. */
         u32 flags;
 
         DEFINE_DMA_UNMAP_ADDR(mapping);
@@ -121,7 +119,23 @@ struct iwl_cmd_meta {
 /*
  * Generic queue structure
  *
- * Contains common data for Rx and Tx queues
+ * Contains common data for Rx and Tx queues.
+ *
+ * Note the difference between n_bd and n_window: the hardware
+ * always assumes 256 descriptors, so n_bd is always 256 (unless
+ * there might be HW changes in the future). For the normal TX
+ * queues, n_window, which is the size of the software queue data
+ * is also 256; however, for the command queue, n_window is only
+ * 32 since we don't need so many commands pending. Since the HW
+ * still uses 256 BDs for DMA though, n_bd stays 256. As a result,
+ * the software buffers (in the variables @meta, @txb in struct
+ * iwl_tx_queue) only have 32 entries, while the HW buffers (@tfds
+ * in the same struct) have 256.
+ * This means that we end up with the following:
+ *  HW entries: | 0 | ... | N * 32 | ... | N * 32 + 31 | ... | 255 |
+ *  SW entries:           | 0      | ... | 31          |
+ * where N is a number between 0 and 7. This means that the SW
+ * data is a window overlayed over the HW queue.
  */
 struct iwl_queue {
         int n_bd;              /* number of BDs in this queue */
@@ -163,7 +177,7 @@ struct iwl_tx_info {
 
 struct iwl_tx_queue {
         struct iwl_queue q;
-        void *tfds;
+        struct iwl_tfd *tfds;
         struct iwl_device_cmd **cmd;
         struct iwl_cmd_meta *meta;
         struct iwl_tx_info *txb;
@@ -245,7 +259,6 @@ enum {
         CMD_SYNC = 0,
         CMD_SIZE_NORMAL = 0,
         CMD_NO_SKB = 0,
-        CMD_SIZE_HUGE = (1 << 0),
         CMD_ASYNC = (1 << 1),
         CMD_WANT_SKB = (1 << 2),
         CMD_MAPPED = (1 << 3),
@@ -257,8 +270,8 @@ enum {
  * struct iwl_device_cmd
  *
  * For allocation of the command and tx queues, this establishes the overall
- * size of the largest command we send to uCode, except for a scan command
- * (which is relatively huge; space is allocated separately).
+ * size of the largest command we send to uCode, except for commands that
+ * aren't fully copied and use other TFD space.
  */
 struct iwl_device_cmd {
         struct iwl_cmd_header hdr;      /* uCode API */
@@ -275,7 +288,11 @@ struct iwl_device_cmd {
 
 #define TFD_MAX_PAYLOAD_SIZE (sizeof(struct iwl_device_cmd))
 
-#define IWL_MAX_CMD_TFDS        1
+#define IWL_MAX_CMD_TFDS        2
+
+enum iwl_hcmd_dataflag {
+        IWL_HCMD_DFL_NOCOPY     = BIT(0),
+};
 
 struct iwl_host_cmd {
         const void *data[IWL_MAX_CMD_TFDS];
@@ -285,6 +302,7 @@ struct iwl_host_cmd {
                          struct iwl_rx_packet *pkt);
         u32 flags;
         u16 len[IWL_MAX_CMD_TFDS];
+        u8 dataflags[IWL_MAX_CMD_TFDS];
         u8 id;
 };
 
@@ -687,17 +705,8 @@ static inline int iwl_queue_used(const struct iwl_queue *q, int i)
 }
 
 
-static inline u8 get_cmd_index(struct iwl_queue *q, u32 index, int is_huge)
+static inline u8 get_cmd_index(struct iwl_queue *q, u32 index)
 {
-        /*
-         * This is for init calibration result and scan command which
-         * required buffer > TFD_MAX_PAYLOAD_SIZE,
-         * the big buffer at end of command array
-         */
-        if (is_huge)
-                return q->n_window;     /* must be power of 2 */
-
-        /* Otherwise, use normal size buffers */
         return index & (q->n_window - 1);
 }
 
@@ -1451,6 +1460,7 @@ struct iwl_priv {
         struct work_struct beacon_update;
         struct iwl_rxon_context *beacon_ctx;
         struct sk_buff *beacon_skb;
+        void *beacon_cmd;
 
         struct work_struct tt_work;
         struct work_struct ct_enter;