spi: Get rid of SPI_ATOMIC_SEQUENCING

SPI_ATOMIC_SEQUENCING was added to accomodate spi flash controllers with
the ability to perform tx and rx of flash command and response at the
same time. Instead of introducing this notion at SPI flash driver layer,
clean up the interface to SPI used by flash.

Flash uses a command-response kind of communication. Thus, even though
SPI is duplex, flash command needs to be sent out on SPI bus and then
flash response should be received on the bus. Some specialized x86
flash controllers are capable of handling command and response in a
single transaction.

In order to support all the varied cases:
1. Add spi_xfer_vector that takes as input a vector of SPI operations
and calls back into SPI controller driver to process these operations.
2. In order to accomodate flash command-response model, use two vectors
while calling into spi_xfer_vector -- one with dout set to
non-NULL(command) and other with din set to non-NULL(response).
3. For specialized SPI flash controllers combine two successive vectors
if the transactions look like a command-response pair.
4. Provide helper functions for common cases like supporting only 2
vectors at a time, supporting n vectors at a time, default vector
operation to cycle through all SPI op vectors one by one.

BUG=chrome-os-partner:59832
BRANCH=None
TEST=Compiles successfully

Change-Id: I4c9e78c585ad95c40c0d5af078ff8251da286236
Signed-off-by: Furquan Shaikh <furquan@chromium.org>
Reviewed-on: https://review.coreboot.org/17681
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
diff --git a/src/drivers/spi/spi-generic.c b/src/drivers/spi/spi-generic.c
index 4fcd04c..805e17a 100644
--- a/src/drivers/spi/spi-generic.c
+++ b/src/drivers/spi/spi-generic.c
@@ -14,6 +14,7 @@
  * GNU General Public License for more details.
  */
 
+#include <assert.h>
 #include <spi-generic.h>
 #include <string.h>
 
@@ -32,10 +33,55 @@
 		ctrlr->release_bus(slave);
 }
 
+static int spi_xfer_single_op(const struct spi_slave *slave,
+			struct spi_op *op)
+{
+	const struct spi_ctrlr *ctrlr = slave->ctrlr;
+	int ret;
+
+	if (!ctrlr || !ctrlr->xfer)
+		return -1;
+
+	ret = ctrlr->xfer(slave, op->dout, op->bytesout, op->din, op->bytesin);
+	if (ret)
+		op->status = SPI_OP_FAILURE;
+	else
+		op->status = SPI_OP_SUCCESS;
+
+	return ret;
+}
+
+static int spi_xfer_vector_default(const struct spi_slave *slave,
+				struct spi_op vectors[], size_t count)
+{
+	size_t i;
+	int ret;
+
+	for (i = 0; i < count; i++) {
+		ret = spi_xfer_single_op(slave, &vectors[i]);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+int spi_xfer_vector(const struct spi_slave *slave,
+		struct spi_op vectors[], size_t count)
+{
+	const struct spi_ctrlr *ctrlr = slave->ctrlr;
+
+	if (ctrlr && ctrlr->xfer_vector)
+		return ctrlr->xfer_vector(slave, vectors, count);
+
+	return spi_xfer_vector_default(slave, vectors, count);
+}
+
 int spi_xfer(const struct spi_slave *slave, const void *dout, size_t bytesout,
 	     void *din, size_t bytesin)
 {
 	const struct spi_ctrlr *ctrlr = slave->ctrlr;
+
 	if (ctrlr && ctrlr->xfer)
 		return ctrlr->xfer(slave, dout, bytesout, din, bytesin);
 
@@ -76,3 +122,56 @@
 
 	return 0;
 }
+
+static int spi_xfer_combine_two_vectors(const struct spi_slave *slave,
+					struct spi_op *v1, struct spi_op *v2)
+{
+	struct spi_op op = {
+		.dout = v1->dout, .bytesout = v1->bytesout,
+		.din = v2->din, .bytesin = v2->bytesin,
+	};
+	int ret;
+
+	/*
+	 * Combine two vectors only if:
+	 * v1 has non-NULL dout and NULL din and
+	 * v2 has non-NULL din and NULL dout and
+	 *
+	 * In all other cases, do not combine the two vectors.
+	 */
+	if ((!v1->dout || v1->din) || (v2->dout || !v2->din))
+		return -1;
+
+	ret = spi_xfer_single_op(slave, &op);
+	v1->status = v2->status = op.status;
+
+	return ret;
+}
+
+/*
+ * Helper function to allow chipsets to combine two vectors if possible. This
+ * function can only handle upto 2 vectors.
+ *
+ * Two vectors are combined if first vector has a non-NULL dout and NULL din and
+ * second vector has a non-NULL din and NULL dout. Otherwise, each vector is
+ * operated upon one at a time.
+ *
+ * Returns 0 on success and non-zero on failure.
+ */
+int spi_xfer_two_vectors(const struct spi_slave *slave,
+			struct spi_op vectors[], size_t count)
+{
+	int ret;
+
+	assert (count <= 2);
+
+	if (count == 2) {
+		ret = spi_xfer_combine_two_vectors(slave, &vectors[0],
+						&vectors[1]);
+
+		if (!ret || (vectors[0].status != SPI_OP_NOT_EXECUTED))
+			return ret;
+	}
+
+	return spi_xfer_vector_default(slave, vectors, count);
+}