Current poll transfer will read/write one word, then wait till the
hw is non-busy, it's not efficient. This patch will try to read/write
as many words as permitted by hardware FIFO depth.
Signed-off-by: Alek Du <alek...@intel.com>
Signed-off-by: Feng Tang <feng.t...@intel.com>
---
drivers/spi/dw_spi.c | 71 +++++++++++++++++++++++++++++++++----------------
1 files changed, 48 insertions(+), 23 deletions(-)
diff --git a/drivers/spi/dw_spi.c b/drivers/spi/dw_spi.c
index 35ac2b9..ba8b394 100644
--- a/drivers/spi/dw_spi.c
+++ b/drivers/spi/dw_spi.c
@@ -160,6 +160,37 @@ static inline void mrst_spi_debugfs_remove(struct dw_spi
*dws)
}
#endif /* CONFIG_DEBUG_FS */
+/* Return the max entries we can fill into tx fifo */
+static inline u32 tx_max(struct dw_spi *dws)
+{
+ u32 tx_left, tx_room, rxtx_gap;
+
+ tx_left = (dws->tx_end - dws->tx) / dws->n_bytes;
+ tx_room = dws->fifo_len - dw_readw(dws, txflr);
+
+ /*
+ * Another concern is about the tx/rx mismatch, we
+ * though to use (dws->fifo_len - rxflr - txflr) as
+ * one maximum value for tx, but it doesn't cover the
+ * data which is out of tx/rx fifo and inside the
+ * shift registers. So a control from sw point of
+ * view is taken.
+ */
+ rxtx_gap = ((dws->rx_end - dws->rx) - (dws->tx_end - dws->tx))
+ / dws->n_bytes;
+
+ return min3(tx_left, tx_room, (u32) (dws->fifo_len - rxtx_gap));
+}
+
+/* Return the max entries we should read out of rx fifo */
+static inline u32 rx_max(struct dw_spi *dws)
+{
+ u32 rx_left = (dws->rx_end - dws->rx) / dws->n_bytes;
+
+ return min(rx_left, (u32)dw_readw(dws, rxflr));
+}
+
+
static void wait_till_not_busy(struct dw_spi *dws)
{
unsigned long end = jiffies + 1 + usecs_to_jiffies(5000);
@@ -175,33 +206,30 @@ static void wait_till_not_busy(struct dw_spi *dws)
static int dw_writer(struct dw_spi *dws)
{
+ u32 max = tx_max(dws);
u16 txw = 0;
- if (!(dw_readw(dws, sr) & SR_TF_NOT_FULL)
- || (dws->tx == dws->tx_end))
- return 0;
-
- /* Set the tx word if the transfer's original "tx" is not null */
- if (dws->tx_end - dws->len) {
- if (dws->n_bytes == 8)
- txw = *(u8 *)(dws->tx);
- else
- txw = *(u16 *)(dws->tx);
+ while (max--) {
+ /* Set the tx word if the transfer's original "tx" is not null
*/
+ if (dws->tx_end - dws->len) {
+ if (dws->n_bytes == 8)
+ txw = *(u8 *)(dws->tx);
+ else
+ txw = *(u16 *)(dws->tx);
+ }
+ dw_writew(dws, dr, txw);
+ dws->tx += dws->n_bytes;
}
- dw_writew(dws, dr, txw);
- dws->tx += dws->n_bytes;
-
- wait_till_not_busy(dws);
return 1;
}
static int dw_reader(struct dw_spi *dws)
{
+ u32 max = rx_max(dws);
u16 rxw;
- while ((dw_readw(dws, sr) & SR_RF_NOT_EMPT)
- && (dws->rx < dws->rx_end)) {
+ while (max--) {
rxw = dw_readw(dws, dr);
/* Care rx only if the transfer's original "rx" is not null */
if (dws->rx_end - dws->len) {
@@ -213,7 +241,6 @@ static int dw_reader(struct dw_spi *dws)
dws->rx += dws->n_bytes;
}
- wait_till_not_busy(dws);
return dws->rx == dws->rx_end;
}
@@ -368,13 +395,11 @@ static irqreturn_t dw_spi_irq(int irq, void *dev_id)
/* Must be called inside pump_transfers() */
static void poll_transfer(struct dw_spi *dws)
{
- while (dw_writer(dws))
+ do {
+ dw_writer(dws);
dw_reader(dws);
- /*
- * There is a possibility that the last word of a transaction
- * will be lost if data is not ready. Re-read to solve this issue.
- */
- dw_reader(dws);
+ cpu_relax();
+ } while (dws->rx_end > dws->rx);
dw_spi_xfer_done(dws);
}
--
1.7.0.4
------------------------------------------------------------------------------
Colocation vs. Managed Hosting
A question and answer guide to determining the best fit
for your organization - today and in the future.
http://p.sf.net/sfu/internap-sfd2d
_______________________________________________
spi-devel-general mailing list
spi-devel-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/spi-devel-general
