iwlwifi: use coherent DMA memory for command header

Recently in commit 8a964f44e01ad3bbc208c3e80d931ba91b9ea786
("iwlwifi: always copy first 16 bytes of commands") we fixed
the problem that the hardware writes back to the command and
that could overwrite parts of the data that was still needed
and would thus be corrupted.

Investigating this problem more closely we found that this
write-back isn't really ordered very well with respect to
other DMA traffic. Therefore, it sometimes happened that the
write-back occurred after unmapping the command again which
is clearly an issue and could corrupt the next allocation
that goes to that spot, or (better) cause IOMMU faults.

To fix this, allocate coherent memory for the first 16 bytes
of each command, containing the write-back part, and use it
for all queues. All the dynamic DMA mappings only need to be
TO_DEVICE then. This ensures that even when the write-back
happens "too late" it can't hit memory that has been freed
or a mapping that doesn't exist any more.

Since now the actual command is no longer modified, we can
also remove CMD_WANT_HCMD and get rid of the DMA sync that
was necessary to update the scratch pointer.

Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

1 files changed, 1 insertions, 7 deletions

diff --git a/drivers/net/wireless/iwlwifi/iwl-trans.h b/drivers/net/wireless/iwlwifi/iwl-trans.h
index 058947f..0cac2b7 100644
--- a/drivers/net/wireless/iwlwifi/iwl-trans.h
+++ b/drivers/net/wireless/iwlwifi/iwl-trans.h
@@ -186,19 +186,13 @@ struct iwl_rx_packet {
  * @CMD_ASYNC: Return right away and don't want for the response
  * @CMD_WANT_SKB: valid only with CMD_SYNC. The caller needs the buffer of the
  *	response. The caller needs to call iwl_free_resp when done.
- * @CMD_WANT_HCMD: The caller needs to get the HCMD that was sent in the
- *	response handler. Chunks flagged by %IWL_HCMD_DFL_NOCOPY won't be
- *	copied. The pointer passed to the response handler is in the transport
- *	ownership and don't need to be freed by the op_mode. This also means
- *	that the pointer is invalidated after the op_mode's handler returns.
  * @CMD_ON_DEMAND: This command is sent by the test mode pipe.
  */
 enum CMD_MODE {
 	CMD_SYNC		= 0,
 	CMD_ASYNC		= BIT(0),
 	CMD_WANT_SKB		= BIT(1),
-	CMD_WANT_HCMD		= BIT(2),
-	CMD_ON_DEMAND		= BIT(3),
+	CMD_ON_DEMAND		= BIT(2),
 };
 
 #define DEF_CMD_PAYLOAD_SIZE 320