iwlwifi: DMA fixes

A few issues wrt DMA were uncovered when using the driver with swiotlb.
- driver should not use memory after it has been mapped
- iwl3945's RX queue management cannot use all of iwlagn because
  the size of the RX buffer is different. Revert back to using
  iwl3945 specific routines that map/unmap memory.
- no need to "dma_syn_single_range_for_cpu" followed by pci_unmap_single,
  we can just call pci_unmap_single initially
- only map the memory area that will be used by device. this is especially
  relevant to the mapping of iwl_cmd. we should not map the entire
  structure because the meta data at the beginning of structure contains
  the address to be used later for unmapping. If the address to be used for
  unmapping is stored in mapped data it creates a problem.
- ensure that _if_ memory needs to be modified after it is mapped that we
  call _sync_single_for_cpu first, and then release it back to device with
  _sync_single_for_device
- we mapped the wrong length of data for host commands, with mapped length
  differing with length provided to device, fix that.

Thanks to Jason Andryuk <jandryuk@gmail.com> for significant bisecting
help to find these issues.

This fixes http://www.intellinuxwireless.org/bugzilla/show_bug.cgi?id=1964

Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
Tested-by: Jason Andryuk <jandryuk@gmail.com>
Tested-by: Ben Gamari <bgamari@gmail.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

1 files changed, 53 insertions, 42 deletions

diff --git a/drivers/net/wireless/iwlwifi/iwl-tx.c b/drivers/net/wireless/iwlwifi/iwl-tx.c
index 8ef53e2..71d5b8a 100644
--- a/drivers/net/wireless/iwlwifi/iwl-tx.c
+++ b/drivers/net/wireless/iwlwifi/iwl-tx.c
@@ -799,6 +799,22 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 	/* Copy MAC header from skb into command buffer */
 	memcpy(tx_cmd->hdr, hdr, hdr_len);
 
+
+	/* Total # bytes to be transmitted */
+	len = (u16)skb->len;
+	tx_cmd->len = cpu_to_le16(len);
+
+	if (info->control.hw_key)
+		iwl_tx_cmd_build_hwcrypto(priv, info, tx_cmd, skb, sta_id);
+
+	/* TODO need this for burst mode later on */
+	iwl_tx_cmd_build_basic(priv, tx_cmd, info, hdr, sta_id);
+
+	/* set is_hcca to 0; it probably will never be implemented */
+	iwl_tx_cmd_build_rate(priv, tx_cmd, info, fc, sta_id, 0);
+
+	iwl_update_tx_stats(priv, le16_to_cpu(fc), len);
+
 	/*
 	 * Use the first empty entry in this queue's command buffer array
 	 * to contain the Tx command and MAC header concatenated together
@@ -819,21 +835,30 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 	else
 		len_org = 0;
 
+	/* Tell NIC about any 2-byte padding after MAC header */
+	if (len_org)
+		tx_cmd->tx_flags |= TX_CMD_FLG_MH_PAD_MSK;
+
 	/* Physical address of this Tx command's header (not MAC header!),
 	 * within command buffer array. */
 	txcmd_phys = pci_map_single(priv->pci_dev,
-				    out_cmd, sizeof(struct iwl_cmd),
+				    &out_cmd->hdr, len,
 				    PCI_DMA_BIDIRECTIONAL);
 	pci_unmap_addr_set(&out_cmd->meta, mapping, txcmd_phys);
-	pci_unmap_len_set(&out_cmd->meta, len, sizeof(struct iwl_cmd));
+	pci_unmap_len_set(&out_cmd->meta, len, len);
 	/* Add buffer containing Tx command and MAC(!) header to TFD's
 	 * first entry */
-	txcmd_phys += offsetof(struct iwl_cmd, hdr);
 	priv->cfg->ops->lib->txq_attach_buf_to_tfd(priv, txq,
 						   txcmd_phys, len, 1, 0);
 
-	if (info->control.hw_key)
-		iwl_tx_cmd_build_hwcrypto(priv, info, tx_cmd, skb, sta_id);
+	if (!ieee80211_has_morefrags(hdr->frame_control)) {
+		txq->need_update = 1;
+		if (qc)
+			priv->stations[sta_id].tid[tid].seq_number = seq_number;
+	} else {
+		wait_write_ptr = 1;
+		txq->need_update = 0;
+	}
 
 	/* Set up TFD's 2nd entry to point directly to remainder of skb,
 	 * if any (802.11 null frames have no payload). */
@@ -846,35 +871,17 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 							   0, 0);
 	}
 
-	/* Tell NIC about any 2-byte padding after MAC header */
-	if (len_org)
-		tx_cmd->tx_flags |= TX_CMD_FLG_MH_PAD_MSK;
-
-	/* Total # bytes to be transmitted */
-	len = (u16)skb->len;
-	tx_cmd->len = cpu_to_le16(len);
-	/* TODO need this for burst mode later on */
-	iwl_tx_cmd_build_basic(priv, tx_cmd, info, hdr, sta_id);
-
-	/* set is_hcca to 0; it probably will never be implemented */
-	iwl_tx_cmd_build_rate(priv, tx_cmd, info, fc, sta_id, 0);
-
-	iwl_update_tx_stats(priv, le16_to_cpu(fc), len);
-
 	scratch_phys = txcmd_phys + sizeof(struct iwl_cmd_header) +
-		offsetof(struct iwl_tx_cmd, scratch);
+				offsetof(struct iwl_tx_cmd, scratch);
+
+	len = sizeof(struct iwl_tx_cmd) +
+		sizeof(struct iwl_cmd_header) + hdr_len;
+	/* take back ownership of DMA buffer to enable update */
+	pci_dma_sync_single_for_cpu(priv->pci_dev, txcmd_phys,
+				    len, PCI_DMA_BIDIRECTIONAL);
 	tx_cmd->dram_lsb_ptr = cpu_to_le32(scratch_phys);
 	tx_cmd->dram_msb_ptr = iwl_get_dma_hi_addr(scratch_phys);
 
-	if (!ieee80211_has_morefrags(hdr->frame_control)) {
-		txq->need_update = 1;
-		if (qc)
-			priv->stations[sta_id].tid[tid].seq_number = seq_number;
-	} else {
-		wait_write_ptr = 1;
-		txq->need_update = 0;
-	}
-
 	IWL_DEBUG_TX(priv, "sequence nr = 0X%x \n",
 		     le16_to_cpu(out_cmd->hdr.sequence));
 	IWL_DEBUG_TX(priv, "tx_flags = 0X%x \n", le32_to_cpu(tx_cmd->tx_flags));
@@ -882,7 +889,11 @@ int iwl_tx_skb(struct iwl_priv *priv, struct sk_buff *skb)
 	iwl_print_hex_dump(priv, IWL_DL_TX, (u8 *)tx_cmd->hdr, hdr_len);
 
 	/* Set up entry for this TFD in Tx byte-count array */
-	priv->cfg->ops->lib->txq_update_byte_cnt_tbl(priv, txq, len);
+	priv->cfg->ops->lib->txq_update_byte_cnt_tbl(priv, txq,
+						     le16_to_cpu(tx_cmd->len));
+
+	pci_dma_sync_single_for_device(priv->pci_dev, txcmd_phys,
+				       len, PCI_DMA_BIDIRECTIONAL);
 
 	/* Tell device the write index *just past* this latest filled TFD */
 	q->write_ptr = iwl_queue_inc_wrap(q->write_ptr, q->n_bd);
@@ -970,18 +981,9 @@ int iwl_enqueue_hcmd(struct iwl_priv *priv, struct iwl_host_cmd *cmd)
 			INDEX_TO_SEQ(q->write_ptr));
 	if (out_cmd->meta.flags & CMD_SIZE_HUGE)
 		out_cmd->hdr.sequence |= SEQ_HUGE_FRAME;
-	len = (idx == TFD_CMD_SLOTS) ?
-			IWL_MAX_SCAN_SIZE : sizeof(struct iwl_cmd);
+	len = sizeof(struct iwl_cmd) - sizeof(struct iwl_cmd_meta);
+	len += (idx == TFD_CMD_SLOTS) ?  IWL_MAX_SCAN_SIZE : 0;
 
-	phys_addr = pci_map_single(priv->pci_dev, out_cmd,
-				   len, PCI_DMA_BIDIRECTIONAL);
-	pci_unmap_addr_set(&out_cmd->meta, mapping, phys_addr);
-	pci_unmap_len_set(&out_cmd->meta, len, len);
-	phys_addr += offsetof(struct iwl_cmd, hdr);
-
-	priv->cfg->ops->lib->txq_attach_buf_to_tfd(priv, txq,
-						   phys_addr, fix_size, 1,
-						   U32_PAD(cmd->len));
 
 #ifdef CONFIG_IWLWIFI_DEBUG
 	switch (out_cmd->hdr.cmd) {
@@ -1009,6 +1011,15 @@ int iwl_enqueue_hcmd(struct iwl_priv *priv, struct iwl_host_cmd *cmd)
 		/* Set up entry in queue's byte count circular buffer */
 		priv->cfg->ops->lib->txq_update_byte_cnt_tbl(priv, txq, 0);
 
+	phys_addr = pci_map_single(priv->pci_dev, &out_cmd->hdr,
+				   fix_size, PCI_DMA_BIDIRECTIONAL);
+	pci_unmap_addr_set(&out_cmd->meta, mapping, phys_addr);
+	pci_unmap_len_set(&out_cmd->meta, len, fix_size);
+
+	priv->cfg->ops->lib->txq_attach_buf_to_tfd(priv, txq,
+						   phys_addr, fix_size, 1,
+						   U32_PAD(cmd->len));
+
 	/* Increment and update queue's write index */
 	q->write_ptr = iwl_queue_inc_wrap(q->write_ptr, q->n_bd);
 	ret = iwl_txq_update_write_ptr(priv, txq);