firewire: octlet AT payloads can be stack-allocated

We do not need slab allocations anymore in order to satisfy
streaming DMA mapping constraints, thanks to commit da28947e7e
"firewire: ohci: avoid separate DMA mapping for small AT payloads".

(Besides, the slab-allocated buffers that firewire-core, firewire-sbp2,
and firedtv used to provide for 8-byte write and lock requests were
still not fully portable since they crossed cacheline boundaries or
shared a cacheline with unrelated CPU-accessed data.  snd-firewire-lib
got this aspect right by using an extra kmalloc/ kfree just for the
8-byte transaction buffer.)

This change replaces kmalloc'ed lock transaction scratch buffers in
firewire-core, firedtv, and snd-firewire-lib by local stack allocations.
Perhaps the most notable result of the change is simpler locking because
there is no need to serialize usages of preallocated per-device buffers
anymore.  Also, allocations and deallocations are simpler.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Acked-by: Clemens Ladisch <clemens@ladisch.de>

drivers/firewire/core-transaction.c

@@ -326,8 +326,8 @@ static int allocate_tlabel(struct fw_card *card)
  * It will contain tag, channel, and sy data instead of a node ID then.
  *
  * The payload buffer at @data is going to be DMA-mapped except in case of
- * quadlet-sized payload or of local (loopback) requests.  Hence make sure that
- * the buffer complies with the restrictions for DMA-mapped memory.  The
+ * @length <= 8 or of local (loopback) requests.  Hence make sure that the
+ * buffer complies with the restrictions of the streaming DMA mapping API.
  * @payload must not be freed before the @callback is called.
  *
  * In case of request types without payload, @data is NULL and @length is 0.
@@ -411,7 +411,8 @@ static void transaction_callback(struct fw_card *card, int rcode,
  *
  * Returns the RCODE.  See fw_send_request() for parameter documentation.
  * Unlike fw_send_request(), @data points to the payload of the request or/and
- * to the payload of the response.
+ * to the payload of the response.  DMA mapping restrictions apply to outbound
+ * request payloads of >= 8 bytes but not to inbound response payloads.
  */
 int fw_run_transaction(struct fw_card *card, int tcode, int destination_id,
 		       int generation, int speed, unsigned long long offset,