This series introduces DEFINE_UART_PORT_TX_HELPER +
DEFINE_UART_PORT_TX_HELPER_LIMITED TX helpers. See PATCH 2/4 for the
details. Comments welcome.
Then it switches drivers to use them. First, to
DEFINE_UART_PORT_TX_HELPER() in 3/4 and then
DEFINE_UART_PORT_TX_HELPER_LIMITED() in 4/4.
The diffstat of patches 3+4 is as follows:
26 files changed, 191 insertions(+), 823 deletions(-)
which appears to be nice.
Cc: Tobias Klauser <[email protected]>
Cc: Richard Genoud <[email protected]>
Cc: Nicolas Ferre <[email protected]>
Cc: Alexandre Belloni <[email protected]>
Cc: Claudiu Beznea <[email protected]>
Cc: Vladimir Zapolskiy <[email protected]>
Cc: Liviu Dudau <[email protected]>
Cc: Sudeep Holla <[email protected]>
Cc: Lorenzo Pieralisi <[email protected]>
Cc: Shawn Guo <[email protected]>
Cc: Sascha Hauer <[email protected]>
Cc: Pengutronix Kernel Team <[email protected]>
Cc: Fabio Estevam <[email protected]>
Cc: NXP Linux Team <[email protected]>
Cc: "Andreas Färber" <[email protected]>
Cc: Manivannan Sadhasivam <[email protected]>
Cc: Russell King <[email protected]>
Cc: Florian Fainelli <[email protected]>
Cc: [email protected]
Cc: "Pali Rohár" <[email protected]>
Cc: Kevin Cernekee <[email protected]>
Cc: Palmer Dabbelt <[email protected]>
Cc: Paul Walmsley <[email protected]>
Cc: Orson Zhai <[email protected]>
Cc: Baolin Wang <[email protected]>
Cc: Chunyan Zhang <[email protected]>
Cc: Patrice Chotard <[email protected]>
Cc: [email protected]
Jiri Slaby (4):
tty: serial: move and cleanup vt8500_tx_empty()
tty: serial: introduce transmit helper generators
tty: serial: use DEFINE_UART_PORT_TX_HELPER()
tty: serial: use DEFINE_UART_PORT_TX_HELPER_LIMITED()
Documentation/driver-api/serial/driver.rst | 3 +
drivers/tty/serial/21285.c | 33 ++-------
drivers/tty/serial/altera_jtaguart.c | 42 +++--------
drivers/tty/serial/altera_uart.c | 37 ++--------
drivers/tty/serial/amba-pl010.c | 37 ++--------
drivers/tty/serial/apbuart.c | 36 ++--------
drivers/tty/serial/atmel_serial.c | 29 ++------
drivers/tty/serial/bcm63xx_uart.c | 47 +++----------
drivers/tty/serial/fsl_lpuart.c | 38 +++-------
drivers/tty/serial/lantiq.c | 44 ++++--------
drivers/tty/serial/lpc32xx_hs.c | 38 ++--------
drivers/tty/serial/mcf.c | 27 ++------
drivers/tty/serial/mpc52xx_uart.c | 43 +-----------
drivers/tty/serial/mps2-uart.c | 29 +-------
drivers/tty/serial/mux.c | 46 ++++--------
drivers/tty/serial/mvebu-uart.c | 40 ++---------
drivers/tty/serial/mxs-auart.c | 31 ++-------
drivers/tty/serial/omap-serial.c | 47 ++++---------
drivers/tty/serial/owl-uart.c | 35 +---------
drivers/tty/serial/pxa.c | 39 +++--------
drivers/tty/serial/rp2.c | 36 ++--------
drivers/tty/serial/sa1100.c | 49 +++++--------
drivers/tty/serial/serial_txx9.c | 37 ++--------
drivers/tty/serial/sifive.c | 45 ++----------
drivers/tty/serial/sprd_serial.c | 38 ++--------
drivers/tty/serial/st-asc.c | 50 ++-----------
drivers/tty/serial/vt8500_serial.c | 40 ++---------
include/linux/serial_core.h | 81 ++++++++++++++++++++++
28 files changed, 275 insertions(+), 822 deletions(-)
--
2.37.3
Many serial drivers do the same thing:
* send x_char if set
* keep sending from the xmit circular buffer until either
- the loop reaches the end of the xmit buffer
- TX is stopped
- HW fifo is full
* check for pending characters and:
- wake up tty writers to fill for more data into xmit buffer
- stop TX if there is nothing in the xmit buffer
The only differences are:
* how to write the character to the HW fifo
* the check of the end condition:
- is the HW fifo full?
- is limit of the written characters reached?
So unify the above into two helper generators:
* DEFINE_UART_PORT_TX_HELPER_LIMITED() -- it performs the above taking
the written characters limit into account, and
* DEFINE_UART_PORT_TX_HELPER() -- the same as above, except it only
checks the HW readiness, not the characters limit.
The HW specific operations (as stated as "differences" above) are passed
as arguments to the macros. They are:
* tx_ready -- returns true if HW can accept more data.
* put_char -- write a character to the device.
* tx_done -- when the write loop is done, perform arbitrary action
before potential invocation of ops->stop_tx() happens.
Note that the above macros are generators. This means the code is
generated in place and the above 3 arguments are "inlined". I.e. no
added penalty by generating call instructions for every single
character. Nor any indirect calls. (As in previous versions of this
patchset.)
Signed-off-by: Jiri Slaby <[email protected]>
---
Notes:
[v3] remove port + ch parameters of the macros (Greg)
[v2] instead of a function (uart_port_tx_limit()) in serial_core,
generate these in-place using macros. Thus eliminating "call"
penalty.
Documentation/driver-api/serial/driver.rst | 3 +
include/linux/serial_core.h | 81 ++++++++++++++++++++++
2 files changed, 84 insertions(+)
diff --git a/Documentation/driver-api/serial/driver.rst b/Documentation/driver-api/serial/driver.rst
index 23c6b956cd90..25775bf1fcc6 100644
--- a/Documentation/driver-api/serial/driver.rst
+++ b/Documentation/driver-api/serial/driver.rst
@@ -78,6 +78,9 @@ Other functions
uart_get_lsr_info uart_handle_dcd_change uart_handle_cts_change
uart_try_toggle_sysrq uart_get_console
+.. kernel-doc:: include/linux/serial_core.h
+ :identifiers: DEFINE_UART_PORT_TX_HELPER_LIMITED DEFINE_UART_PORT_TX_HELPER
+
Other notes
-----------
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index 6e4f4765d209..69abcc05fcfb 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -646,6 +646,87 @@ struct uart_driver {
void uart_write_wakeup(struct uart_port *port);
+#define __DEFINE_UART_PORT_TX_HELPER(name, tx_ready, put_char, tx_done, \
+ for_test, for_post, ...) \
+unsigned int name(struct uart_port *port __VA_OPT__(,) __VA_ARGS__) \
+{ \
+ struct circ_buf *xmit = &port->state->xmit; \
+ unsigned int pending; \
+ u8 ch; \
+ \
+ for (; (for_test) && (tx_ready); (for_post), port->icount.tx++) { \
+ if (port->x_char) { \
+ ch = port->x_char; \
+ (put_char); \
+ port->x_char = 0; \
+ continue; \
+ } \
+ \
+ if (uart_circ_empty(xmit) || uart_tx_stopped(port)) \
+ break; \
+ \
+ ch = xmit->buf[xmit->tail]; \
+ (put_char); \
+ xmit->tail = (xmit->tail + 1) % UART_XMIT_SIZE; \
+ } \
+ \
+ (tx_done); \
+ \
+ pending = uart_circ_chars_pending(xmit); \
+ if (pending < WAKEUP_CHARS) { \
+ uart_write_wakeup(port); \
+ \
+ if (pending == 0) \
+ port->ops->stop_tx(port); \
+ } \
+ \
+ return pending; \
+}
+
+/**
+ * DEFINE_UART_PORT_TX_HELPER_LIMITED -- generate transmit helper for uart_port
+ * with count limiting
+ * @name: name of the helper to generate
+ * @tx_ready: can HW accept more data function
+ * @put_char: function to write a character
+ * @tx_done: function to call after the loop is done
+ *
+ * This macro generates a function @name. The generated function is meant as a
+ * helper to transmit characters from the xmit buffer to the hardware using
+ * @put_char(). It does so until count (passed to @name) characters are sent
+ * and while @tx_ready evaluates to true.
+ *
+ * The generated function returns the number of characters in the xmit buffer
+ * when done.
+ *
+ * The expression in macro parameters are free to use @port (uart port) and @ch
+ * (character to send) variables and shall be designed as follows:
+ * * **tx_ready:** should evaluate to true if the HW can accept more data to
+ * be sent. This parameter can be %true, which means the HW is always ready.
+ * * **put_char:** shall write @ch to the device of @port.
+ * * **tx_done:** when the write loop is done, this can perform arbitrary
+ * action before potential invocation of ops->stop_tx() happens. If the
+ * driver does not need to do anything, use e.g. ({}).
+ *
+ * For all of them, @port->lock is held, interrupts are locally disabled and
+ * the expressions must not sleep.
+ */
+#define DEFINE_UART_PORT_TX_HELPER_LIMITED(name, tx_ready, put_char, tx_done) \
+ __DEFINE_UART_PORT_TX_HELPER(name, tx_ready, put_char, tx_done, \
+ count, count--, unsigned int count)
+
+/**
+ * DEFINE_UART_PORT_TX_HELPER -- generate transmit helper for uart_port
+ * @name: name of the helper to generate
+ * @tx_ready: can HW accept more data function
+ * @put_char: function to write a character
+ *
+ * See DEFINE_UART_PORT_TX_HELPER_LIMITED() for more details.
+ */
+#define DEFINE_UART_PORT_TX_HELPER(name, tx_ready, put_char) \
+ __DEFINE_UART_PORT_TX_HELPER(name, tx_ready, put_char, ({}), \
+ true, ({}))
+
/*
* Baud rate helpers.
*/
--
2.37.3
DEFINE_UART_PORT_TX_HELPER() is a new helper to send characters to the
device. Use it in these drivers.
Cc: Tobias Klauser <[email protected]>
Cc: Richard Genoud <[email protected]>
Cc: Nicolas Ferre <[email protected]>
Cc: Alexandre Belloni <[email protected]>
Cc: Claudiu Beznea <[email protected]>
Cc: Vladimir Zapolskiy <[email protected]>
Cc: Liviu Dudau <[email protected]>
Cc: Sudeep Holla <[email protected]>
Cc: Lorenzo Pieralisi <[email protected]>
Cc: Shawn Guo <[email protected]>
Cc: Sascha Hauer <[email protected]>
Cc: Pengutronix Kernel Team <[email protected]>
Cc: Fabio Estevam <[email protected]>
Cc: NXP Linux Team <[email protected]>
Cc: "Andreas Färber" <[email protected]>
Cc: Manivannan Sadhasivam <[email protected]>
Signed-off-by: Jiri Slaby <[email protected]>
---
Notes:
[v3] remove stray \n removal (Ilpo)
drivers/tty/serial/altera_uart.c | 37 +++-------------------
drivers/tty/serial/atmel_serial.c | 29 +++++-------------
drivers/tty/serial/fsl_lpuart.c | 38 ++++++-----------------
drivers/tty/serial/lantiq.c | 44 ++++++++-------------------
drivers/tty/serial/lpc32xx_hs.c | 38 +++++------------------
drivers/tty/serial/mcf.c | 27 +++-------------
drivers/tty/serial/mpc52xx_uart.c | 43 ++------------------------
drivers/tty/serial/mps2-uart.c | 29 ++----------------
drivers/tty/serial/mxs-auart.c | 31 +++++--------------
drivers/tty/serial/owl-uart.c | 35 ++-------------------
drivers/tty/serial/sa1100.c | 49 ++++++++++++------------------
drivers/tty/serial/vt8500_serial.c | 33 ++------------------
12 files changed, 84 insertions(+), 349 deletions(-)
diff --git a/drivers/tty/serial/altera_uart.c b/drivers/tty/serial/altera_uart.c
index 8b749ed557c6..70c0ad431cf9 100644
--- a/drivers/tty/serial/altera_uart.c
+++ b/drivers/tty/serial/altera_uart.c
@@ -246,37 +246,10 @@ static void altera_uart_rx_chars(struct altera_uart *pp)
tty_flip_buffer_push(&port->state->port);
}
-static void altera_uart_tx_chars(struct altera_uart *pp)
-{
- struct uart_port *port = &pp->port;
- struct circ_buf *xmit = &port->state->xmit;
-
- if (port->x_char) {
- /* Send special char - probably flow control */
- altera_uart_writel(port, port->x_char, ALTERA_UART_TXDATA_REG);
- port->x_char = 0;
- port->icount.tx++;
- return;
- }
-
- while (altera_uart_readl(port, ALTERA_UART_STATUS_REG) &
- ALTERA_UART_STATUS_TRDY_MSK) {
- if (xmit->head == xmit->tail)
- break;
- altera_uart_writel(port, xmit->buf[xmit->tail],
- ALTERA_UART_TXDATA_REG);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- port->icount.tx++;
- }
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(port);
-
- if (xmit->head == xmit->tail) {
- pp->imr &= ~ALTERA_UART_CONTROL_TRDY_MSK;
- altera_uart_update_ctrl_reg(pp);
- }
-}
+static DEFINE_UART_PORT_TX_HELPER(altera_uart_tx_chars,
+ altera_uart_readl(port, ALTERA_UART_STATUS_REG) &
+ ALTERA_UART_STATUS_TRDY_MSK,
+ altera_uart_writel(port, ch, ALTERA_UART_TXDATA_REG));
static irqreturn_t altera_uart_interrupt(int irq, void *data)
{
@@ -290,7 +263,7 @@ static irqreturn_t altera_uart_interrupt(int irq, void *data)
if (isr & ALTERA_UART_STATUS_RRDY_MSK)
altera_uart_rx_chars(pp);
if (isr & ALTERA_UART_STATUS_TRDY_MSK)
- altera_uart_tx_chars(pp);
+ altera_uart_tx_chars(&pp->port);
spin_unlock(&port->lock);
return IRQ_RETVAL(isr);
diff --git a/drivers/tty/serial/atmel_serial.c b/drivers/tty/serial/atmel_serial.c
index 30ba9eef7b39..98c338df1e99 100644
--- a/drivers/tty/serial/atmel_serial.c
+++ b/drivers/tty/serial/atmel_serial.c
@@ -812,36 +812,21 @@ static void atmel_rx_chars(struct uart_port *port)
atmel_tasklet_schedule(atmel_port, &atmel_port->tasklet_rx);
}
+static DEFINE_UART_PORT_TX_HELPER(atmel_do_tx_chars,
+ atmel_uart_readl(port, ATMEL_US_CSR) & ATMEL_US_TXRDY,
+ atmel_uart_write_char(port, ch));
+
/*
* Transmit characters (called from tasklet with TXRDY interrupt
* disabled)
*/
static void atmel_tx_chars(struct uart_port *port)
{
- struct circ_buf *xmit = &port->state->xmit;
struct atmel_uart_port *atmel_port = to_atmel_uart_port(port);
+ unsigned int pending;
- if (port->x_char &&
- (atmel_uart_readl(port, ATMEL_US_CSR) & ATMEL_US_TXRDY)) {
- atmel_uart_write_char(port, port->x_char);
- port->icount.tx++;
- port->x_char = 0;
- }
- if (uart_circ_empty(xmit) || uart_tx_stopped(port))
- return;
-
- while (atmel_uart_readl(port, ATMEL_US_CSR) & ATMEL_US_TXRDY) {
- atmel_uart_write_char(port, xmit->buf[xmit->tail]);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- port->icount.tx++;
- if (uart_circ_empty(xmit))
- break;
- }
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(port);
-
- if (!uart_circ_empty(xmit)) {
+ pending = atmel_do_tx_chars(port);
+ if (pending) {
/* we still have characters to transmit, so we should continue
* transmitting them when TX is ready, regardless of
* mode or duplexity
diff --git a/drivers/tty/serial/fsl_lpuart.c b/drivers/tty/serial/fsl_lpuart.c
index f6c33cd228c8..7c64cadcfbf3 100644
--- a/drivers/tty/serial/fsl_lpuart.c
+++ b/drivers/tty/serial/fsl_lpuart.c
@@ -740,36 +740,18 @@ static int lpuart32_poll_get_char(struct uart_port *port)
}
#endif
-static inline void lpuart_transmit_buffer(struct lpuart_port *sport)
+static bool lpuart_tx_ready(struct uart_port *port)
{
- struct circ_buf *xmit = &sport->port.state->xmit;
-
- if (sport->port.x_char) {
- writeb(sport->port.x_char, sport->port.membase + UARTDR);
- sport->port.icount.tx++;
- sport->port.x_char = 0;
- return;
- }
-
- if (lpuart_stopped_or_empty(&sport->port)) {
- lpuart_stop_tx(&sport->port);
- return;
- }
-
- while (!uart_circ_empty(xmit) &&
- (readb(sport->port.membase + UARTTCFIFO) < sport->txfifo_size)) {
- writeb(xmit->buf[xmit->tail], sport->port.membase + UARTDR);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- sport->port.icount.tx++;
- }
+ struct lpuart_port *sport = container_of(port, struct lpuart_port,
+ port);
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(&sport->port);
-
- if (uart_circ_empty(xmit))
- lpuart_stop_tx(&sport->port);
+ return readb(port->membase + UARTTCFIFO) < sport->txfifo_size;
}
+static DEFINE_UART_PORT_TX_HELPER(lpuart_transmit_buffer,
+ lpuart_tx_ready(port),
+ writeb(ch, port->membase + UARTDR));
+
static inline void lpuart32_transmit_buffer(struct lpuart_port *sport)
{
struct circ_buf *xmit = &sport->port.state->xmit;
@@ -820,7 +802,7 @@ static void lpuart_start_tx(struct uart_port *port)
lpuart_dma_tx(sport);
} else {
if (readb(port->membase + UARTSR1) & UARTSR1_TDRE)
- lpuart_transmit_buffer(sport);
+ lpuart_transmit_buffer(&sport->port);
}
}
@@ -877,7 +859,7 @@ static unsigned int lpuart32_tx_empty(struct uart_port *port)
static void lpuart_txint(struct lpuart_port *sport)
{
spin_lock(&sport->port.lock);
- lpuart_transmit_buffer(sport);
+ lpuart_transmit_buffer(&sport->port);
spin_unlock(&sport->port.lock);
}
diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c
index a3120c3347dd..874568f417bf 100644
--- a/drivers/tty/serial/lantiq.c
+++ b/drivers/tty/serial/lantiq.c
@@ -95,7 +95,6 @@
#define ASCFSTAT_TXFREEMASK 0x3F000000
#define ASCFSTAT_TXFREEOFF 24
-static void lqasc_tx_chars(struct uart_port *port);
static struct ltq_uart_port *lqasc_port[MAXPORTS];
static struct uart_driver lqasc_reg;
@@ -139,6 +138,18 @@ lqasc_stop_tx(struct uart_port *port)
return;
}
+static bool
+lqasc_tx_ready(struct uart_port *port)
+{
+ u32 fstat = __raw_readl(port->membase + LTQ_ASC_FSTAT);
+
+ return (fstat & ASCFSTAT_TXFREEMASK) >> ASCFSTAT_TXFREEOFF;
+}
+
+static DEFINE_UART_PORT_TX_HELPER(lqasc_tx_chars,
+ lqasc_tx_ready(port),
+ writeb(ch, port->membase + LTQ_ASC_TBUF));
+
static void
lqasc_start_tx(struct uart_port *port)
{
@@ -219,37 +230,6 @@ lqasc_rx_chars(struct uart_port *port)
return 0;
}
-static void
-lqasc_tx_chars(struct uart_port *port)
-{
- struct circ_buf *xmit = &port->state->xmit;
- if (uart_tx_stopped(port)) {
- lqasc_stop_tx(port);
- return;
- }
-
- while (((__raw_readl(port->membase + LTQ_ASC_FSTAT) &
- ASCFSTAT_TXFREEMASK) >> ASCFSTAT_TXFREEOFF) != 0) {
- if (port->x_char) {
- writeb(port->x_char, port->membase + LTQ_ASC_TBUF);
- port->icount.tx++;
- port->x_char = 0;
- continue;
- }
-
- if (uart_circ_empty(xmit))
- break;
-
- writeb(port->state->xmit.buf[port->state->xmit.tail],
- port->membase + LTQ_ASC_TBUF);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- port->icount.tx++;
- }
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(port);
-}
-
static irqreturn_t
lqasc_tx_int(int irq, void *_port)
{
diff --git a/drivers/tty/serial/lpc32xx_hs.c b/drivers/tty/serial/lpc32xx_hs.c
index 93140cac1ca1..f0693e591864 100644
--- a/drivers/tty/serial/lpc32xx_hs.c
+++ b/drivers/tty/serial/lpc32xx_hs.c
@@ -276,41 +276,17 @@ static void __serial_lpc32xx_rx(struct uart_port *port)
tty_flip_buffer_push(tport);
}
-static void serial_lpc32xx_stop_tx(struct uart_port *port);
-
-static void __serial_lpc32xx_tx(struct uart_port *port)
+static bool serial_lpc32xx_tx_ready(struct uart_port *port)
{
- struct circ_buf *xmit = &port->state->xmit;
+ u32 level = readl(LPC32XX_HSUART_LEVEL(port->membase));
- if (port->x_char) {
- writel((u32)port->x_char, LPC32XX_HSUART_FIFO(port->membase));
- port->icount.tx++;
- port->x_char = 0;
- return;
- }
-
- if (uart_circ_empty(xmit) || uart_tx_stopped(port))
- goto exit_tx;
-
- /* Transfer data */
- while (LPC32XX_HSU_TX_LEV(readl(
- LPC32XX_HSUART_LEVEL(port->membase))) < 64) {
- writel((u32) xmit->buf[xmit->tail],
- LPC32XX_HSUART_FIFO(port->membase));
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- port->icount.tx++;
- if (uart_circ_empty(xmit))
- break;
- }
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(port);
-
-exit_tx:
- if (uart_circ_empty(xmit))
- serial_lpc32xx_stop_tx(port);
+ return LPC32XX_HSU_TX_LEV(level) < 64;
}
+static DEFINE_UART_PORT_TX_HELPER(__serial_lpc32xx_tx,
+ serial_lpc32xx_tx_ready(port),
+ writel(ch, LPC32XX_HSUART_FIFO(port->membase)));
+
static irqreturn_t serial_lpc32xx_interrupt(int irq, void *dev_id)
{
struct uart_port *port = dev_id;
diff --git a/drivers/tty/serial/mcf.c b/drivers/tty/serial/mcf.c
index f4aaaadd0742..735e6f261b64 100644
--- a/drivers/tty/serial/mcf.c
+++ b/drivers/tty/serial/mcf.c
@@ -324,32 +324,15 @@ static void mcf_rx_chars(struct mcf_uart *pp)
/****************************************************************************/
+static DEFINE_UART_PORT_TX_HELPER(mcf_do_tx_chars,
+ readb(port->membase + MCFUART_USR) & MCFUART_USR_TXREADY,
+ writeb(ch, port->membase + MCFUART_UTB));
+
static void mcf_tx_chars(struct mcf_uart *pp)
{
struct uart_port *port = &pp->port;
- struct circ_buf *xmit = &port->state->xmit;
-
- if (port->x_char) {
- /* Send special char - probably flow control */
- writeb(port->x_char, port->membase + MCFUART_UTB);
- port->x_char = 0;
- port->icount.tx++;
- return;
- }
-
- while (readb(port->membase + MCFUART_USR) & MCFUART_USR_TXREADY) {
- if (uart_circ_empty(xmit))
- break;
- writeb(xmit->buf[xmit->tail], port->membase + MCFUART_UTB);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE -1);
- port->icount.tx++;
- }
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(port);
- if (uart_circ_empty(xmit)) {
- mcf_stop_tx(port);
+ if (!mcf_do_tx_chars(port)) {
/* Disable TX to negate RTS automatically */
if (port->rs485.flags & SER_RS485_ENABLED)
writeb(MCFUART_UCR_TXDISABLE,
diff --git a/drivers/tty/serial/mpc52xx_uart.c b/drivers/tty/serial/mpc52xx_uart.c
index 3f1986c89694..16edf80284bd 100644
--- a/drivers/tty/serial/mpc52xx_uart.c
+++ b/drivers/tty/serial/mpc52xx_uart.c
@@ -1425,46 +1425,9 @@ mpc52xx_uart_int_rx_chars(struct uart_port *port)
return psc_ops->raw_rx_rdy(port);
}
-static inline int
-mpc52xx_uart_int_tx_chars(struct uart_port *port)
-{
- struct circ_buf *xmit = &port->state->xmit;
-
- /* Process out of band chars */
- if (port->x_char) {
- psc_ops->write_char(port, port->x_char);
- port->icount.tx++;
- port->x_char = 0;
- return 1;
- }
-
- /* Nothing to do ? */
- if (uart_circ_empty(xmit) || uart_tx_stopped(port)) {
- mpc52xx_uart_stop_tx(port);
- return 0;
- }
-
- /* Send chars */
- while (psc_ops->raw_tx_rdy(port)) {
- psc_ops->write_char(port, xmit->buf[xmit->tail]);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- port->icount.tx++;
- if (uart_circ_empty(xmit))
- break;
- }
-
- /* Wake up */
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(port);
-
- /* Maybe we're done after all */
- if (uart_circ_empty(xmit)) {
- mpc52xx_uart_stop_tx(port);
- return 0;
- }
-
- return 1;
-}
+static DEFINE_UART_PORT_TX_HELPER(mpc52xx_uart_int_tx_chars,
+ psc_ops->raw_tx_rdy(port),
+ psc_ops->write_char(port, ch));
static irqreturn_t
mpc5xxx_uart_process_int(struct uart_port *port)
diff --git a/drivers/tty/serial/mps2-uart.c b/drivers/tty/serial/mps2-uart.c
index 5e9429dcc51f..771598bf5b38 100644
--- a/drivers/tty/serial/mps2-uart.c
+++ b/drivers/tty/serial/mps2-uart.c
@@ -127,32 +127,9 @@ static void mps2_uart_stop_tx(struct uart_port *port)
mps2_uart_write8(port, control, UARTn_CTRL);
}
-static void mps2_uart_tx_chars(struct uart_port *port)
-{
- struct circ_buf *xmit = &port->state->xmit;
-
- while (!(mps2_uart_read8(port, UARTn_STATE) & UARTn_STATE_TX_FULL)) {
- if (port->x_char) {
- mps2_uart_write8(port, port->x_char, UARTn_DATA);
- port->x_char = 0;
- port->icount.tx++;
- continue;
- }
-
- if (uart_circ_empty(xmit) || uart_tx_stopped(port))
- break;
-
- mps2_uart_write8(port, xmit->buf[xmit->tail], UARTn_DATA);
- xmit->tail = (xmit->tail + 1) % UART_XMIT_SIZE;
- port->icount.tx++;
- }
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(port);
-
- if (uart_circ_empty(xmit))
- mps2_uart_stop_tx(port);
-}
+static DEFINE_UART_PORT_TX_HELPER(mps2_uart_tx_chars,
+ mps2_uart_tx_empty(port),
+ mps2_uart_write8(port, ch, UARTn_DATA));
static void mps2_uart_start_tx(struct uart_port *port)
{
diff --git a/drivers/tty/serial/mxs-auart.c b/drivers/tty/serial/mxs-auart.c
index 1944daf8593a..063c50090d13 100644
--- a/drivers/tty/serial/mxs-auart.c
+++ b/drivers/tty/serial/mxs-auart.c
@@ -566,6 +566,10 @@ static int mxs_auart_dma_tx(struct mxs_auart_port *s, int size)
return 0;
}
+static DEFINE_UART_PORT_TX_HELPER(mxs_auart_do_tx_chars,
+ !(mxs_read(to_auart_port(port), REG_STAT) & AUART_STAT_TXFF),
+ mxs_write(ch, to_auart_port(port), REG_DATA));
+
static void mxs_auart_tx_chars(struct mxs_auart_port *s)
{
struct circ_buf *xmit = &s->port.state->xmit;
@@ -603,31 +607,10 @@ static void mxs_auart_tx_chars(struct mxs_auart_port *s)
return;
}
-
- while (!(mxs_read(s, REG_STAT) & AUART_STAT_TXFF)) {
- if (s->port.x_char) {
- s->port.icount.tx++;
- mxs_write(s->port.x_char, s, REG_DATA);
- s->port.x_char = 0;
- continue;
- }
- if (!uart_circ_empty(xmit) && !uart_tx_stopped(&s->port)) {
- s->port.icount.tx++;
- mxs_write(xmit->buf[xmit->tail], s, REG_DATA);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- } else
- break;
- }
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(&s->port);
-
- if (uart_circ_empty(&(s->port.state->xmit)))
- mxs_clr(AUART_INTR_TXIEN, s, REG_INTR);
- else
+ if (mxs_auart_do_tx_chars(&s->port))
mxs_set(AUART_INTR_TXIEN, s, REG_INTR);
-
- if (uart_tx_stopped(&s->port))
- mxs_auart_stop_tx(&s->port);
+ else
+ mxs_clr(AUART_INTR_TXIEN, s, REG_INTR);
}
static void mxs_auart_rx_char(struct mxs_auart_port *s)
diff --git a/drivers/tty/serial/owl-uart.c b/drivers/tty/serial/owl-uart.c
index 888e17e3f25f..cc3abdc02f19 100644
--- a/drivers/tty/serial/owl-uart.c
+++ b/drivers/tty/serial/owl-uart.c
@@ -179,38 +179,9 @@ static void owl_uart_start_tx(struct uart_port *port)
owl_uart_write(port, val, OWL_UART_CTL);
}
-static void owl_uart_send_chars(struct uart_port *port)
-{
- struct circ_buf *xmit = &port->state->xmit;
- unsigned int ch;
-
- if (port->x_char) {
- while (!(owl_uart_read(port, OWL_UART_STAT) & OWL_UART_STAT_TFFU))
- cpu_relax();
- owl_uart_write(port, port->x_char, OWL_UART_TXDAT);
- port->icount.tx++;
- port->x_char = 0;
- }
-
- if (uart_tx_stopped(port))
- return;
-
- while (!(owl_uart_read(port, OWL_UART_STAT) & OWL_UART_STAT_TFFU)) {
- if (uart_circ_empty(xmit))
- break;
-
- ch = xmit->buf[xmit->tail];
- owl_uart_write(port, ch, OWL_UART_TXDAT);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- port->icount.tx++;
- }
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(port);
-
- if (uart_circ_empty(xmit))
- owl_uart_stop_tx(port);
-}
+static DEFINE_UART_PORT_TX_HELPER(owl_uart_send_chars,
+ !(owl_uart_read(port, OWL_UART_STAT) & OWL_UART_STAT_TFFU),
+ owl_uart_write(port, ch, OWL_UART_TXDAT));
static void owl_uart_receive_chars(struct uart_port *port)
{
diff --git a/drivers/tty/serial/sa1100.c b/drivers/tty/serial/sa1100.c
index e64e42a19d1a..66ff452700e5 100644
--- a/drivers/tty/serial/sa1100.c
+++ b/drivers/tty/serial/sa1100.c
@@ -226,45 +226,34 @@ sa1100_rx_chars(struct sa1100_port *sport)
tty_flip_buffer_push(&sport->port.state->port);
}
-static void sa1100_tx_chars(struct sa1100_port *sport)
+static bool sa1100_tx_ready(struct uart_port *port)
{
- struct circ_buf *xmit = &sport->port.state->xmit;
+ struct sa1100_port *sport =
+ container_of(port, struct sa1100_port, port);
- if (sport->port.x_char) {
- UART_PUT_CHAR(sport, sport->port.x_char);
- sport->port.icount.tx++;
- sport->port.x_char = 0;
- return;
- }
+ return UART_GET_UTSR1(sport) & UTSR1_TNF;
+}
+
+static void sa1100_put_char(struct uart_port *port, unsigned char ch)
+{
+ struct sa1100_port *sport =
+ container_of(port, struct sa1100_port, port);
+
+ UART_PUT_CHAR(sport, ch);
+}
+static DEFINE_UART_PORT_TX_HELPER(sa1100_do_tx_chars,
+ sa1100_tx_ready(port), sa1100_put_char(port, ch));
+
+static void sa1100_tx_chars(struct sa1100_port *sport)
+{
/*
* Check the modem control lines before
* transmitting anything.
*/
sa1100_mctrl_check(sport);
- if (uart_circ_empty(xmit) || uart_tx_stopped(&sport->port)) {
- sa1100_stop_tx(&sport->port);
- return;
- }
-
- /*
- * Tried using FIFO (not checking TNF) for fifo fill:
- * still had the '4 bytes repeated' problem.
- */
- while (UART_GET_UTSR1(sport) & UTSR1_TNF) {
- UART_PUT_CHAR(sport, xmit->buf[xmit->tail]);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- sport->port.icount.tx++;
- if (uart_circ_empty(xmit))
- break;
- }
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(&sport->port);
-
- if (uart_circ_empty(xmit))
- sa1100_stop_tx(&sport->port);
+ sa1100_do_tx_chars(&sport->port);
}
static irqreturn_t sa1100_int(int irq, void *dev_id)
diff --git a/drivers/tty/serial/vt8500_serial.c b/drivers/tty/serial/vt8500_serial.c
index 9e2c745b4535..bd755254d59d 100644
--- a/drivers/tty/serial/vt8500_serial.c
+++ b/drivers/tty/serial/vt8500_serial.c
@@ -194,36 +194,9 @@ static unsigned int vt8500_tx_empty(struct uart_port *port)
return idx < 16 ? TIOCSER_TEMT : 0;
}
-static void handle_tx(struct uart_port *port)
-{
- struct circ_buf *xmit = &port->state->xmit;
-
- if (port->x_char) {
- writeb(port->x_char, port->membase + VT8500_TXFIFO);
- port->icount.tx++;
- port->x_char = 0;
- }
- if (uart_circ_empty(xmit) || uart_tx_stopped(port)) {
- vt8500_stop_tx(port);
- return;
- }
-
- while (vt8500_tx_empty(port)) {
- if (uart_circ_empty(xmit))
- break;
-
- writeb(xmit->buf[xmit->tail], port->membase + VT8500_TXFIFO);
-
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- port->icount.tx++;
- }
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(port);
-
- if (uart_circ_empty(xmit))
- vt8500_stop_tx(port);
-}
+static DEFINE_UART_PORT_TX_HELPER(handle_tx,
+ vt8500_tx_empty(port),
+ writeb(ch, port->membase + VT8500_TXFIFO));
static void vt8500_start_tx(struct uart_port *port)
{
--
2.37.3
Make vt8500_tx_empty() more readable by introducing a new local variable
and move the function before handle_tx(). That way we can reuse it in
there too.
Cc: <[email protected]>
Signed-off-by: Jiri Slaby <[email protected]>
---
Notes:
[v3] this is new in v3 -- extracted as a separate change from later
patches
drivers/tty/serial/vt8500_serial.c | 15 ++++++++-------
1 file changed, 8 insertions(+), 7 deletions(-)
diff --git a/drivers/tty/serial/vt8500_serial.c b/drivers/tty/serial/vt8500_serial.c
index 6f08136ce78a..9e2c745b4535 100644
--- a/drivers/tty/serial/vt8500_serial.c
+++ b/drivers/tty/serial/vt8500_serial.c
@@ -187,6 +187,13 @@ static void handle_rx(struct uart_port *port)
tty_flip_buffer_push(tport);
}
+static unsigned int vt8500_tx_empty(struct uart_port *port)
+{
+ unsigned int idx = vt8500_read(port, VT8500_URFIDX) & 0x1f;
+
+ return idx < 16 ? TIOCSER_TEMT : 0;
+}
+
static void handle_tx(struct uart_port *port)
{
struct circ_buf *xmit = &port->state->xmit;
@@ -201,7 +208,7 @@ static void handle_tx(struct uart_port *port)
return;
}
- while ((vt8500_read(port, VT8500_URFIDX) & 0x1f) < 16) {
+ while (vt8500_tx_empty(port)) {
if (uart_circ_empty(xmit))
break;
@@ -260,12 +267,6 @@ static irqreturn_t vt8500_irq(int irq, void *dev_id)
return IRQ_HANDLED;
}
-static unsigned int vt8500_tx_empty(struct uart_port *port)
-{
- return (vt8500_read(port, VT8500_URFIDX) & 0x1f) < 16 ?
- TIOCSER_TEMT : 0;
-}
-
static unsigned int vt8500_get_mctrl(struct uart_port *port)
{
unsigned int usr;
--
2.37.3
DEFINE_UART_PORT_TX_HELPER_LIMITED() is a new helper to send characters
to the device. Use it in these drivers.
mux.c also needs to define tx_done(). But I'm not sure if the driver
really wants to wait for all the characters to dismiss from the HW fifo
at this code point. Hence I marked this as FIXME.
Signed-off-by: Jiri Slaby <[email protected]>
Cc: Russell King <[email protected]>
Cc: Florian Fainelli <[email protected]>
Cc: [email protected]
Cc: "Pali Rohár" <[email protected]>
Cc: Kevin Cernekee <[email protected]>
Cc: Palmer Dabbelt <[email protected]>
Cc: Paul Walmsley <[email protected]>
Cc: Orson Zhai <[email protected]>
Cc: Baolin Wang <[email protected]>
Cc: Chunyan Zhang <[email protected]>
Cc: Patrice Chotard <[email protected]>
Cc: [email protected]
---
drivers/tty/serial/21285.c | 33 ++++--------------
drivers/tty/serial/altera_jtaguart.c | 42 ++++++-----------------
drivers/tty/serial/amba-pl010.c | 37 +++-----------------
drivers/tty/serial/apbuart.c | 36 ++++----------------
drivers/tty/serial/bcm63xx_uart.c | 47 +++++---------------------
drivers/tty/serial/mux.c | 46 ++++++++-----------------
drivers/tty/serial/mvebu-uart.c | 40 ++++------------------
drivers/tty/serial/omap-serial.c | 47 +++++++-------------------
drivers/tty/serial/pxa.c | 39 +++++-----------------
drivers/tty/serial/rp2.c | 36 ++++----------------
drivers/tty/serial/serial_txx9.c | 37 +++-----------------
drivers/tty/serial/sifive.c | 45 +++----------------------
drivers/tty/serial/sprd_serial.c | 38 +++------------------
drivers/tty/serial/st-asc.c | 50 ++++------------------------
14 files changed, 103 insertions(+), 470 deletions(-)
diff --git a/drivers/tty/serial/21285.c b/drivers/tty/serial/21285.c
index 7520cc02fd4d..c057ae367b09 100644
--- a/drivers/tty/serial/21285.c
+++ b/drivers/tty/serial/21285.c
@@ -151,38 +151,17 @@ static irqreturn_t serial21285_rx_chars(int irq, void *dev_id)
return IRQ_HANDLED;
}
+static DEFINE_UART_PORT_TX_HELPER_LIMITED(serial21285_do_tx_chars,
+ !(*CSR_UARTFLG & 0x20),
+ *CSR_UARTDR = ch,
+ ({}));
+
static irqreturn_t serial21285_tx_chars(int irq, void *dev_id)
{
struct uart_port *port = dev_id;
- struct circ_buf *xmit = &port->state->xmit;
- int count = 256;
-
- if (port->x_char) {
- *CSR_UARTDR = port->x_char;
- port->icount.tx++;
- port->x_char = 0;
- goto out;
- }
- if (uart_circ_empty(xmit) || uart_tx_stopped(port)) {
- serial21285_stop_tx(port);
- goto out;
- }
-
- do {
- *CSR_UARTDR = xmit->buf[xmit->tail];
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- port->icount.tx++;
- if (uart_circ_empty(xmit))
- break;
- } while (--count > 0 && !(*CSR_UARTFLG & 0x20));
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(port);
- if (uart_circ_empty(xmit))
- serial21285_stop_tx(port);
+ serial21285_do_tx_chars(port, 256);
- out:
return IRQ_HANDLED;
}
diff --git a/drivers/tty/serial/altera_jtaguart.c b/drivers/tty/serial/altera_jtaguart.c
index cb791c5149a3..a8b7a6f58100 100644
--- a/drivers/tty/serial/altera_jtaguart.c
+++ b/drivers/tty/serial/altera_jtaguart.c
@@ -134,42 +134,20 @@ static void altera_jtaguart_rx_chars(struct altera_jtaguart *pp)
tty_flip_buffer_push(&port->state->port);
}
+static DEFINE_UART_PORT_TX_HELPER_LIMITED(altera_jtaguart_do_tx_chars,
+ true,
+ writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG),
+ ({}));
+
static void altera_jtaguart_tx_chars(struct altera_jtaguart *pp)
{
struct uart_port *port = &pp->port;
- struct circ_buf *xmit = &port->state->xmit;
- unsigned int pending, count;
-
- if (port->x_char) {
- /* Send special char - probably flow control */
- writel(port->x_char, port->membase + ALTERA_JTAGUART_DATA_REG);
- port->x_char = 0;
- port->icount.tx++;
- return;
- }
-
- pending = uart_circ_chars_pending(xmit);
- if (pending > 0) {
- count = (readl(port->membase + ALTERA_JTAGUART_CONTROL_REG) &
- ALTERA_JTAGUART_CONTROL_WSPACE_MSK) >>
- ALTERA_JTAGUART_CONTROL_WSPACE_OFF;
- if (count > pending)
- count = pending;
- if (count > 0) {
- pending -= count;
- while (count--) {
- writel(xmit->buf[xmit->tail],
- port->membase + ALTERA_JTAGUART_DATA_REG);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- port->icount.tx++;
- }
- if (pending < WAKEUP_CHARS)
- uart_write_wakeup(port);
- }
- }
+ unsigned int space;
- if (pending == 0)
- altera_jtaguart_stop_tx(port);
+ space = (readl(port->membase + ALTERA_JTAGUART_CONTROL_REG) &
+ ALTERA_JTAGUART_CONTROL_WSPACE_MSK) >>
+ ALTERA_JTAGUART_CONTROL_WSPACE_OFF;
+ altera_jtaguart_do_tx_chars(port, space);
}
static irqreturn_t altera_jtaguart_interrupt(int irq, void *data)
diff --git a/drivers/tty/serial/amba-pl010.c b/drivers/tty/serial/amba-pl010.c
index fae0b581ff42..a7b558a3bc15 100644
--- a/drivers/tty/serial/amba-pl010.c
+++ b/drivers/tty/serial/amba-pl010.c
@@ -162,37 +162,10 @@ static void pl010_rx_chars(struct uart_port *port)
tty_flip_buffer_push(&port->state->port);
}
-static void pl010_tx_chars(struct uart_port *port)
-{
- struct circ_buf *xmit = &port->state->xmit;
- int count;
-
- if (port->x_char) {
- writel(port->x_char, port->membase + UART01x_DR);
- port->icount.tx++;
- port->x_char = 0;
- return;
- }
- if (uart_circ_empty(xmit) || uart_tx_stopped(port)) {
- pl010_stop_tx(port);
- return;
- }
-
- count = port->fifosize >> 1;
- do {
- writel(xmit->buf[xmit->tail], port->membase + UART01x_DR);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- port->icount.tx++;
- if (uart_circ_empty(xmit))
- break;
- } while (--count > 0);
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(port);
-
- if (uart_circ_empty(xmit))
- pl010_stop_tx(port);
-}
+static DEFINE_UART_PORT_TX_HELPER_LIMITED(pl010_tx_chars,
+ true,
+ writel(ch, port->membase + UART01x_DR),
+ ({}));
static void pl010_modem_status(struct uart_amba_port *uap)
{
@@ -238,7 +211,7 @@ static irqreturn_t pl010_int(int irq, void *dev_id)
if (status & UART010_IIR_MIS)
pl010_modem_status(uap);
if (status & UART010_IIR_TIS)
- pl010_tx_chars(port);
+ pl010_tx_chars(port, port->fifosize >> 1);
if (pass_counter-- == 0)
break;
diff --git a/drivers/tty/serial/apbuart.c b/drivers/tty/serial/apbuart.c
index 9ef82d870ff2..e810b4a12d13 100644
--- a/drivers/tty/serial/apbuart.c
+++ b/drivers/tty/serial/apbuart.c
@@ -120,38 +120,14 @@ static void apbuart_rx_chars(struct uart_port *port)
tty_flip_buffer_push(&port->state->port);
}
+static DEFINE_UART_PORT_TX_HELPER_LIMITED(apbuart_do_tx_chars,
+ true,
+ UART_PUT_CHAR(port, ch),
+ ({}));
+
static void apbuart_tx_chars(struct uart_port *port)
{
- struct circ_buf *xmit = &port->state->xmit;
- int count;
-
- if (port->x_char) {
- UART_PUT_CHAR(port, port->x_char);
- port->icount.tx++;
- port->x_char = 0;
- return;
- }
-
- if (uart_circ_empty(xmit) || uart_tx_stopped(port)) {
- apbuart_stop_tx(port);
- return;
- }
-
- /* amba: fill FIFO */
- count = port->fifosize >> 1;
- do {
- UART_PUT_CHAR(port, xmit->buf[xmit->tail]);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- port->icount.tx++;
- if (uart_circ_empty(xmit))
- break;
- } while (--count > 0);
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(port);
-
- if (uart_circ_empty(xmit))
- apbuart_stop_tx(port);
+ apbuart_do_tx_chars(port, port->fifosize >> 1);
}
static irqreturn_t apbuart_int(int irq, void *dev_id)
diff --git a/drivers/tty/serial/bcm63xx_uart.c b/drivers/tty/serial/bcm63xx_uart.c
index 53b43174aa40..e29b7290ab66 100644
--- a/drivers/tty/serial/bcm63xx_uart.c
+++ b/drivers/tty/serial/bcm63xx_uart.c
@@ -297,59 +297,30 @@ static void bcm_uart_do_rx(struct uart_port *port)
tty_flip_buffer_push(tty_port);
}
+static DEFINE_UART_PORT_TX_HELPER_LIMITED(bcm_uart_tx,
+ true,
+ bcm_uart_writel(port, ch, UART_FIFO_REG),
+ ({}));
+
/*
* fill tx fifo with chars to send, stop when fifo is about to be full
* or when all chars have been sent.
*/
static void bcm_uart_do_tx(struct uart_port *port)
{
- struct circ_buf *xmit;
- unsigned int val, max_count;
-
- if (port->x_char) {
- bcm_uart_writel(port, port->x_char, UART_FIFO_REG);
- port->icount.tx++;
- port->x_char = 0;
- return;
- }
-
- if (uart_tx_stopped(port)) {
- bcm_uart_stop_tx(port);
- return;
- }
-
- xmit = &port->state->xmit;
- if (uart_circ_empty(xmit))
- goto txq_empty;
+ unsigned int val, pending;
val = bcm_uart_readl(port, UART_MCTL_REG);
val = (val & UART_MCTL_TXFIFOFILL_MASK) >> UART_MCTL_TXFIFOFILL_SHIFT;
- max_count = port->fifosize - val;
-
- while (max_count--) {
- unsigned int c;
- c = xmit->buf[xmit->tail];
- bcm_uart_writel(port, c, UART_FIFO_REG);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- port->icount.tx++;
- if (uart_circ_empty(xmit))
- break;
- }
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(port);
-
- if (uart_circ_empty(xmit))
- goto txq_empty;
- return;
+ pending = bcm_uart_tx(port, port->fifosize - val);
+ if (pending)
+ return;
-txq_empty:
/* nothing to send, disable transmit interrupt */
val = bcm_uart_readl(port, UART_IR_REG);
val &= ~UART_TX_INT_MASK;
bcm_uart_writel(port, val, UART_IR_REG);
- return;
}
/*
diff --git a/drivers/tty/serial/mux.c b/drivers/tty/serial/mux.c
index 0ba0f4d9459d..de928d5def77 100644
--- a/drivers/tty/serial/mux.c
+++ b/drivers/tty/serial/mux.c
@@ -171,6 +171,18 @@ static void mux_break_ctl(struct uart_port *port, int break_state)
{
}
+static void mux_tx_done(struct uart_port *port)
+{
+ /* FIXME js: really needs to wait? */
+ while (UART_GET_FIFO_CNT(port))
+ udelay(1);
+}
+
+static DEFINE_UART_PORT_TX_HELPER_LIMITED(mux_transmit,
+ true,
+ UART_PUT_CHAR(port, ch),
+ mux_tx_done(port));
+
/**
* mux_write - Write chars to the mux fifo.
* @port: Ptr to the uart_port.
@@ -180,39 +192,7 @@ static void mux_break_ctl(struct uart_port *port, int break_state)
*/
static void mux_write(struct uart_port *port)
{
- int count;
- struct circ_buf *xmit = &port->state->xmit;
-
- if(port->x_char) {
- UART_PUT_CHAR(port, port->x_char);
- port->icount.tx++;
- port->x_char = 0;
- return;
- }
-
- if(uart_circ_empty(xmit) || uart_tx_stopped(port)) {
- mux_stop_tx(port);
- return;
- }
-
- count = (port->fifosize) - UART_GET_FIFO_CNT(port);
- do {
- UART_PUT_CHAR(port, xmit->buf[xmit->tail]);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- port->icount.tx++;
- if(uart_circ_empty(xmit))
- break;
-
- } while(--count > 0);
-
- while(UART_GET_FIFO_CNT(port))
- udelay(1);
-
- if(uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(port);
-
- if (uart_circ_empty(xmit))
- mux_stop_tx(port);
+ mux_transmit(port, port->fifosize - UART_GET_FIFO_CNT(port));
}
/**
diff --git a/drivers/tty/serial/mvebu-uart.c b/drivers/tty/serial/mvebu-uart.c
index 65eaecd10b7c..bc015b63a402 100644
--- a/drivers/tty/serial/mvebu-uart.c
+++ b/drivers/tty/serial/mvebu-uart.c
@@ -333,42 +333,14 @@ static void mvebu_uart_rx_chars(struct uart_port *port, unsigned int status)
tty_flip_buffer_push(tport);
}
+static DEFINE_UART_PORT_TX_HELPER_LIMITED(mvebu_uart_do_tx_chars,
+ !(readl(port->membase + UART_STAT) & STAT_TX_FIFO_FUL),
+ writel(ch, port->membase + UART_TSH(port)),
+ ({}));
+
static void mvebu_uart_tx_chars(struct uart_port *port, unsigned int status)
{
- struct circ_buf *xmit = &port->state->xmit;
- unsigned int count;
- unsigned int st;
-
- if (port->x_char) {
- writel(port->x_char, port->membase + UART_TSH(port));
- port->icount.tx++;
- port->x_char = 0;
- return;
- }
-
- if (uart_circ_empty(xmit) || uart_tx_stopped(port)) {
- mvebu_uart_stop_tx(port);
- return;
- }
-
- for (count = 0; count < port->fifosize; count++) {
- writel(xmit->buf[xmit->tail], port->membase + UART_TSH(port));
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- port->icount.tx++;
-
- if (uart_circ_empty(xmit))
- break;
-
- st = readl(port->membase + UART_STAT);
- if (st & STAT_TX_FIFO_FUL)
- break;
- }
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(port);
-
- if (uart_circ_empty(xmit))
- mvebu_uart_stop_tx(port);
+ mvebu_uart_do_tx_chars(port, port->fifosize);
}
static irqreturn_t mvebu_uart_isr(int irq, void *dev_id)
diff --git a/drivers/tty/serial/omap-serial.c b/drivers/tty/serial/omap-serial.c
index 0aa666e247d5..8c5202f88f45 100644
--- a/drivers/tty/serial/omap-serial.c
+++ b/drivers/tty/serial/omap-serial.c
@@ -337,45 +337,22 @@ static void serial_omap_stop_rx(struct uart_port *port)
serial_out(up, UART_IER, up->ier);
}
-static void transmit_chars(struct uart_omap_port *up, unsigned int lsr)
+static void serial_omap_put_char(struct uart_port *port, unsigned char ch)
{
- struct circ_buf *xmit = &up->port.state->xmit;
- int count;
-
- if (up->port.x_char) {
- serial_out(up, UART_TX, up->port.x_char);
- up->port.icount.tx++;
- up->port.x_char = 0;
- if ((up->port.rs485.flags & SER_RS485_ENABLED) &&
- !(up->port.rs485.flags & SER_RS485_RX_DURING_TX))
- up->rs485_tx_filter_count++;
-
- return;
- }
- if (uart_circ_empty(xmit) || uart_tx_stopped(&up->port)) {
- serial_omap_stop_tx(&up->port);
- return;
- }
- count = up->port.fifosize / 4;
- do {
- serial_out(up, UART_TX, xmit->buf[xmit->tail]);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- up->port.icount.tx++;
- if ((up->port.rs485.flags & SER_RS485_ENABLED) &&
- !(up->port.rs485.flags & SER_RS485_RX_DURING_TX))
- up->rs485_tx_filter_count++;
-
- if (uart_circ_empty(xmit))
- break;
- } while (--count > 0);
+ struct uart_omap_port *up = to_uart_omap_port(port);
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(&up->port);
+ serial_out(up, UART_TX, ch);
- if (uart_circ_empty(xmit))
- serial_omap_stop_tx(&up->port);
+ if ((up->port.rs485.flags & SER_RS485_ENABLED) &&
+ !(up->port.rs485.flags & SER_RS485_RX_DURING_TX))
+ up->rs485_tx_filter_count++;
}
+static DEFINE_UART_PORT_TX_HELPER_LIMITED(transmit_chars,
+ true,
+ serial_omap_put_char(port, ch),
+ ({}));
+
static inline void serial_omap_enable_ier_thri(struct uart_omap_port *up)
{
if (!(up->ier & UART_IER_THRI)) {
@@ -573,7 +550,7 @@ static irqreturn_t serial_omap_irq(int irq, void *dev_id)
check_modem_status(up);
break;
case UART_IIR_THRI:
- transmit_chars(up, lsr);
+ transmit_chars(&up->port, up->port.fifosize / 4);
break;
case UART_IIR_RX_TIMEOUT:
case UART_IIR_RDI:
diff --git a/drivers/tty/serial/pxa.c b/drivers/tty/serial/pxa.c
index 9309ffd87c8e..45778be46f3c 100644
--- a/drivers/tty/serial/pxa.c
+++ b/drivers/tty/serial/pxa.c
@@ -172,39 +172,18 @@ static inline void receive_chars(struct uart_pxa_port *up, int *status)
serial_out(up, UART_IER, up->ier);
}
-static void transmit_chars(struct uart_pxa_port *up)
+static void serial_pxa_put_char(struct uart_port *port, u8 ch)
{
- struct circ_buf *xmit = &up->port.state->xmit;
- int count;
-
- if (up->port.x_char) {
- serial_out(up, UART_TX, up->port.x_char);
- up->port.icount.tx++;
- up->port.x_char = 0;
- return;
- }
- if (uart_circ_empty(xmit) || uart_tx_stopped(&up->port)) {
- serial_pxa_stop_tx(&up->port);
- return;
- }
-
- count = up->port.fifosize / 2;
- do {
- serial_out(up, UART_TX, xmit->buf[xmit->tail]);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- up->port.icount.tx++;
- if (uart_circ_empty(xmit))
- break;
- } while (--count > 0);
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(&up->port);
-
+ struct uart_pxa_port *up = (struct uart_pxa_port *)port;
- if (uart_circ_empty(xmit))
- serial_pxa_stop_tx(&up->port);
+ serial_out(up, UART_TX, ch);
}
+static DEFINE_UART_PORT_TX_HELPER_LIMITED(transmit_chars,
+ true,
+ serial_pxa_put_char(port, ch),
+ ({}));
+
static void serial_pxa_start_tx(struct uart_port *port)
{
struct uart_pxa_port *up = (struct uart_pxa_port *)port;
@@ -254,7 +233,7 @@ static inline irqreturn_t serial_pxa_irq(int irq, void *dev_id)
receive_chars(up, &lsr);
check_modem_status(up);
if (lsr & UART_LSR_THRE)
- transmit_chars(up);
+ transmit_chars(&up->port, up->port.fifosize / 2);
spin_unlock(&up->port.lock);
return IRQ_HANDLED;
}
diff --git a/drivers/tty/serial/rp2.c b/drivers/tty/serial/rp2.c
index 6689d8add8f7..4f59c883539a 100644
--- a/drivers/tty/serial/rp2.c
+++ b/drivers/tty/serial/rp2.c
@@ -426,35 +426,10 @@ static void rp2_rx_chars(struct rp2_uart_port *up)
tty_flip_buffer_push(port);
}
-static void rp2_tx_chars(struct rp2_uart_port *up)
-{
- u16 max_tx = FIFO_SIZE - readw(up->base + RP2_TX_FIFO_COUNT);
- struct circ_buf *xmit = &up->port.state->xmit;
-
- if (uart_tx_stopped(&up->port)) {
- rp2_uart_stop_tx(&up->port);
- return;
- }
-
- for (; max_tx != 0; max_tx--) {
- if (up->port.x_char) {
- writeb(up->port.x_char, up->base + RP2_DATA_BYTE);
- up->port.x_char = 0;
- up->port.icount.tx++;
- continue;
- }
- if (uart_circ_empty(xmit)) {
- rp2_uart_stop_tx(&up->port);
- break;
- }
- writeb(xmit->buf[xmit->tail], up->base + RP2_DATA_BYTE);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- up->port.icount.tx++;
- }
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(&up->port);
-}
+static DEFINE_UART_PORT_TX_HELPER_LIMITED(rp2_tx_chars,
+ true,
+ writeb(ch, port_to_up(port)->base + RP2_DATA_BYTE),
+ ({}));
static void rp2_ch_interrupt(struct rp2_uart_port *up)
{
@@ -472,7 +447,8 @@ static void rp2_ch_interrupt(struct rp2_uart_port *up)
if (status & RP2_CHAN_STAT_RXDATA_m)
rp2_rx_chars(up);
if (status & RP2_CHAN_STAT_TXEMPTY_m)
- rp2_tx_chars(up);
+ rp2_tx_chars(&up->port,
+ FIFO_SIZE - readw(up->base + RP2_TX_FIFO_COUNT));
if (status & RP2_CHAN_STAT_MS_CHANGED_MASK)
wake_up_interruptible(&up->port.state->port.delta_msr_wait);
diff --git a/drivers/tty/serial/serial_txx9.c b/drivers/tty/serial/serial_txx9.c
index 228e380db080..6e7a71e9a917 100644
--- a/drivers/tty/serial/serial_txx9.c
+++ b/drivers/tty/serial/serial_txx9.c
@@ -319,37 +319,10 @@ receive_chars(struct uart_port *up, unsigned int *status)
*status = disr;
}
-static inline void transmit_chars(struct uart_port *up)
-{
- struct circ_buf *xmit = &up->state->xmit;
- int count;
-
- if (up->x_char) {
- sio_out(up, TXX9_SITFIFO, up->x_char);
- up->icount.tx++;
- up->x_char = 0;
- return;
- }
- if (uart_circ_empty(xmit) || uart_tx_stopped(up)) {
- serial_txx9_stop_tx(up);
- return;
- }
-
- count = TXX9_SIO_TX_FIFO;
- do {
- sio_out(up, TXX9_SITFIFO, xmit->buf[xmit->tail]);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- up->icount.tx++;
- if (uart_circ_empty(xmit))
- break;
- } while (--count > 0);
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(up);
-
- if (uart_circ_empty(xmit))
- serial_txx9_stop_tx(up);
-}
+static DEFINE_UART_PORT_TX_HELPER_LIMITED(transmit_chars,
+ true,
+ sio_out(port, TXX9_SITFIFO, ch),
+ ({}));
static irqreturn_t serial_txx9_interrupt(int irq, void *dev_id)
{
@@ -371,7 +344,7 @@ static irqreturn_t serial_txx9_interrupt(int irq, void *dev_id)
if (status & TXX9_SIDISR_RDIS)
receive_chars(up, &status);
if (status & TXX9_SIDISR_TDIS)
- transmit_chars(up);
+ transmit_chars(up, TXX9_SIO_TX_FIFO);
/* Clear TX/RX Int. Status */
sio_mask(up, TXX9_SIDISR,
TXX9_SIDISR_TDIS | TXX9_SIDISR_RDIS |
diff --git a/drivers/tty/serial/sifive.c b/drivers/tty/serial/sifive.c
index 5c3a07546a58..1440f7d55fbd 100644
--- a/drivers/tty/serial/sifive.c
+++ b/drivers/tty/serial/sifive.c
@@ -277,45 +277,10 @@ static void __ssp_transmit_char(struct sifive_serial_port *ssp, int ch)
__ssp_writel(ch, SIFIVE_SERIAL_TXDATA_OFFS, ssp);
}
-/**
- * __ssp_transmit_chars() - enqueue multiple bytes onto the TX FIFO
- * @ssp: pointer to a struct sifive_serial_port
- *
- * Transfer up to a TX FIFO size's worth of characters from the Linux serial
- * transmit buffer to the SiFive UART TX FIFO.
- *
- * Context: Any context. Expects @ssp->port.lock to be held by caller.
- */
-static void __ssp_transmit_chars(struct sifive_serial_port *ssp)
-{
- struct circ_buf *xmit = &ssp->port.state->xmit;
- int count;
-
- if (ssp->port.x_char) {
- __ssp_transmit_char(ssp, ssp->port.x_char);
- ssp->port.icount.tx++;
- ssp->port.x_char = 0;
- return;
- }
- if (uart_circ_empty(xmit) || uart_tx_stopped(&ssp->port)) {
- sifive_serial_stop_tx(&ssp->port);
- return;
- }
- count = SIFIVE_TX_FIFO_DEPTH;
- do {
- __ssp_transmit_char(ssp, xmit->buf[xmit->tail]);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- ssp->port.icount.tx++;
- if (uart_circ_empty(xmit))
- break;
- } while (--count > 0);
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(&ssp->port);
-
- if (uart_circ_empty(xmit))
- sifive_serial_stop_tx(&ssp->port);
-}
+static DEFINE_UART_PORT_TX_HELPER_LIMITED(__ssp_transmit_chars,
+ true,
+ __ssp_transmit_char(port_to_sifive_serial_port(port), ch),
+ ({}));
/**
* __ssp_enable_txwm() - enable transmit watermark interrupts
@@ -553,7 +518,7 @@ static irqreturn_t sifive_serial_irq(int irq, void *dev_id)
if (ip & SIFIVE_SERIAL_IP_RXWM_MASK)
__ssp_receive_chars(ssp);
if (ip & SIFIVE_SERIAL_IP_TXWM_MASK)
- __ssp_transmit_chars(ssp);
+ __ssp_transmit_chars(&ssp->port, SIFIVE_TX_FIFO_DEPTH);
spin_unlock(&ssp->port.lock);
diff --git a/drivers/tty/serial/sprd_serial.c b/drivers/tty/serial/sprd_serial.c
index 4329b9c9cbf0..1616b63281bc 100644
--- a/drivers/tty/serial/sprd_serial.c
+++ b/drivers/tty/serial/sprd_serial.c
@@ -624,38 +624,10 @@ static inline void sprd_rx(struct uart_port *port)
tty_flip_buffer_push(tty);
}
-static inline void sprd_tx(struct uart_port *port)
-{
- struct circ_buf *xmit = &port->state->xmit;
- int count;
-
- if (port->x_char) {
- serial_out(port, SPRD_TXD, port->x_char);
- port->icount.tx++;
- port->x_char = 0;
- return;
- }
-
- if (uart_circ_empty(xmit) || uart_tx_stopped(port)) {
- sprd_stop_tx(port);
- return;
- }
-
- count = THLD_TX_EMPTY;
- do {
- serial_out(port, SPRD_TXD, xmit->buf[xmit->tail]);
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- port->icount.tx++;
- if (uart_circ_empty(xmit))
- break;
- } while (--count > 0);
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(port);
-
- if (uart_circ_empty(xmit))
- sprd_stop_tx(port);
-}
+static DEFINE_UART_PORT_TX_HELPER_LIMITED(sprd_tx,
+ true,
+ serial_out(port, SPRD_TXD, ch),
+ ({}));
/* this handles the interrupt from one port */
static irqreturn_t sprd_handle_irq(int irq, void *dev_id)
@@ -683,7 +655,7 @@ static irqreturn_t sprd_handle_irq(int irq, void *dev_id)
sprd_rx(port);
if (ims & SPRD_IMSR_TX_FIFO_EMPTY)
- sprd_tx(port);
+ sprd_tx(port, THLD_TX_EMPTY);
spin_unlock(&port->lock);
diff --git a/drivers/tty/serial/st-asc.c b/drivers/tty/serial/st-asc.c
index cce42f4c9bc2..8157fb52ceeb 100644
--- a/drivers/tty/serial/st-asc.c
+++ b/drivers/tty/serial/st-asc.c
@@ -230,6 +230,11 @@ static inline unsigned asc_hw_txroom(struct uart_port *port)
return 0;
}
+static DEFINE_UART_PORT_TX_HELPER_LIMITED(asc_do_transmit_chars,
+ true,
+ asc_out(port, ASC_TXBUF, ch),
+ ({}));
+
/*
* Start transmitting chars.
* This is called from both interrupt and task level.
@@ -237,50 +242,7 @@ static inline unsigned asc_hw_txroom(struct uart_port *port)
*/
static void asc_transmit_chars(struct uart_port *port)
{
- struct circ_buf *xmit = &port->state->xmit;
- int txroom;
- unsigned char c;
-
- txroom = asc_hw_txroom(port);
-
- if ((txroom != 0) && port->x_char) {
- c = port->x_char;
- port->x_char = 0;
- asc_out(port, ASC_TXBUF, c);
- port->icount.tx++;
- txroom = asc_hw_txroom(port);
- }
-
- if (uart_tx_stopped(port)) {
- /*
- * We should try and stop the hardware here, but I
- * don't think the ASC has any way to do that.
- */
- asc_disable_tx_interrupts(port);
- return;
- }
-
- if (uart_circ_empty(xmit)) {
- asc_disable_tx_interrupts(port);
- return;
- }
-
- if (txroom == 0)
- return;
-
- do {
- c = xmit->buf[xmit->tail];
- xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
- asc_out(port, ASC_TXBUF, c);
- port->icount.tx++;
- txroom--;
- } while ((txroom > 0) && (!uart_circ_empty(xmit)));
-
- if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
- uart_write_wakeup(port);
-
- if (uart_circ_empty(xmit))
- asc_disable_tx_interrupts(port);
+ asc_do_transmit_chars(port, asc_hw_txroom(port));
}
static void asc_receive_chars(struct uart_port *port)
--
2.37.3
On Tue, Sep 06, 2022 at 12:48:01PM +0200, Jiri Slaby wrote:
> This series introduces DEFINE_UART_PORT_TX_HELPER +
> DEFINE_UART_PORT_TX_HELPER_LIMITED TX helpers. See PATCH 2/4 for the
> details. Comments welcome.
>
> Then it switches drivers to use them. First, to
> DEFINE_UART_PORT_TX_HELPER() in 3/4 and then
> DEFINE_UART_PORT_TX_HELPER_LIMITED() in 4/4.
>
> The diffstat of patches 3+4 is as follows:
> 26 files changed, 191 insertions(+), 823 deletions(-)
> which appears to be nice.
Not really. This is horrid. Quality can't be measured in LoC (only).
The resulting code is unreadable. And for no good reason.
[ And note that you're "saving" something like 20 lines per driver:
12 files changed, 84 insertions(+), 349 deletions(-)
]
NAK
Johan
On 06. 09. 22, 13:30, Johan Hovold wrote:
> On Tue, Sep 06, 2022 at 12:48:01PM +0200, Jiri Slaby wrote:
>> This series introduces DEFINE_UART_PORT_TX_HELPER +
>> DEFINE_UART_PORT_TX_HELPER_LIMITED TX helpers. See PATCH 2/4 for the
>> details. Comments welcome.
>>
>> Then it switches drivers to use them. First, to
>> DEFINE_UART_PORT_TX_HELPER() in 3/4 and then
>> DEFINE_UART_PORT_TX_HELPER_LIMITED() in 4/4.
>>
>> The diffstat of patches 3+4 is as follows:
>> 26 files changed, 191 insertions(+), 823 deletions(-)
>> which appears to be nice.
>
> Not really. This is horrid. Quality can't be measured in LoC (only).
>
> The resulting code is unreadable. And for no good reason.
IMO, it's much more readable than the original ~ 30 various (and buggy
-- see Ilpo's fixes) copies of this code. Apart from that, it makes
further rework much easier (I have switch to kfifo in my mind for example).
> [ And note that you're "saving" something like 20 lines per driver:
It's not about saving, it's about deduplicating and unifying.
> 12 files changed, 84 insertions(+), 349 deletions(-)
> ]
>
> NAK
I'd love to come up with something nicer. That would be a function in
serial-core calling hooks like I had [1] for example. But provided all
those CPU workarounds/thunks, it'd be quite expensive to call two
functions per character.
Or creating a static inline (having ± the macro content) and the hooks
as parameters and hope for optimizations to eliminate thunks (also
suggested in the past [1]).
[1] https://lore.kernel.org/all/[email protected]/
thanks,
--
js
suse labs
On Tue, 6 Sep 2022, Jiri Slaby wrote:
> This series introduces DEFINE_UART_PORT_TX_HELPER +
> DEFINE_UART_PORT_TX_HELPER_LIMITED TX helpers. See PATCH 2/4 for the
> details. Comments welcome.
>
> Then it switches drivers to use them. First, to
> DEFINE_UART_PORT_TX_HELPER() in 3/4 and then
> DEFINE_UART_PORT_TX_HELPER_LIMITED() in 4/4.
>
> The diffstat of patches 3+4 is as follows:
> 26 files changed, 191 insertions(+), 823 deletions(-)
> which appears to be nice.
>
> Cc: Tobias Klauser <[email protected]>
> Cc: Richard Genoud <[email protected]>
> Cc: Nicolas Ferre <[email protected]>
> Cc: Alexandre Belloni <[email protected]>
> Cc: Claudiu Beznea <[email protected]>
> Cc: Vladimir Zapolskiy <[email protected]>
> Cc: Liviu Dudau <[email protected]>
> Cc: Sudeep Holla <[email protected]>
> Cc: Lorenzo Pieralisi <[email protected]>
> Cc: Shawn Guo <[email protected]>
> Cc: Sascha Hauer <[email protected]>
> Cc: Pengutronix Kernel Team <[email protected]>
> Cc: Fabio Estevam <[email protected]>
> Cc: NXP Linux Team <[email protected]>
> Cc: "Andreas Färber" <[email protected]>
> Cc: Manivannan Sadhasivam <[email protected]>
> Cc: Russell King <[email protected]>
> Cc: Florian Fainelli <[email protected]>
> Cc: [email protected]
> Cc: "Pali Rohár" <[email protected]>
> Cc: Kevin Cernekee <[email protected]>
> Cc: Palmer Dabbelt <[email protected]>
> Cc: Paul Walmsley <[email protected]>
> Cc: Orson Zhai <[email protected]>
> Cc: Baolin Wang <[email protected]>
> Cc: Chunyan Zhang <[email protected]>
> Cc: Patrice Chotard <[email protected]>
> Cc: [email protected]
>
> Jiri Slaby (4):
> tty: serial: move and cleanup vt8500_tx_empty()
> tty: serial: introduce transmit helper generators
> tty: serial: use DEFINE_UART_PORT_TX_HELPER()
> tty: serial: use DEFINE_UART_PORT_TX_HELPER_LIMITED()
>
> Documentation/driver-api/serial/driver.rst | 3 +
> drivers/tty/serial/21285.c | 33 ++-------
> drivers/tty/serial/altera_jtaguart.c | 42 +++--------
> drivers/tty/serial/altera_uart.c | 37 ++--------
> drivers/tty/serial/amba-pl010.c | 37 ++--------
> drivers/tty/serial/apbuart.c | 36 ++--------
> drivers/tty/serial/atmel_serial.c | 29 ++------
> drivers/tty/serial/bcm63xx_uart.c | 47 +++----------
> drivers/tty/serial/fsl_lpuart.c | 38 +++-------
> drivers/tty/serial/lantiq.c | 44 ++++--------
> drivers/tty/serial/lpc32xx_hs.c | 38 ++--------
> drivers/tty/serial/mcf.c | 27 ++------
> drivers/tty/serial/mpc52xx_uart.c | 43 +-----------
> drivers/tty/serial/mps2-uart.c | 29 +-------
> drivers/tty/serial/mux.c | 46 ++++--------
> drivers/tty/serial/mvebu-uart.c | 40 ++---------
> drivers/tty/serial/mxs-auart.c | 31 ++-------
> drivers/tty/serial/omap-serial.c | 47 ++++---------
> drivers/tty/serial/owl-uart.c | 35 +---------
> drivers/tty/serial/pxa.c | 39 +++--------
> drivers/tty/serial/rp2.c | 36 ++--------
> drivers/tty/serial/sa1100.c | 49 +++++--------
> drivers/tty/serial/serial_txx9.c | 37 ++--------
> drivers/tty/serial/sifive.c | 45 ++----------
> drivers/tty/serial/sprd_serial.c | 38 ++--------
> drivers/tty/serial/st-asc.c | 50 ++-----------
> drivers/tty/serial/vt8500_serial.c | 40 ++---------
> include/linux/serial_core.h | 81 ++++++++++++++++++++++
> 28 files changed, 275 insertions(+), 822 deletions(-)
For the whole series:
Reviewed-by: Ilpo Järvinen <[email protected]>
--
i.
On Wed, 7 Sep 2022, Jiri Slaby wrote:
> On 06. 09. 22, 13:30, Johan Hovold wrote:
> > On Tue, Sep 06, 2022 at 12:48:01PM +0200, Jiri Slaby wrote:
> > > This series introduces DEFINE_UART_PORT_TX_HELPER +
> > > DEFINE_UART_PORT_TX_HELPER_LIMITED TX helpers. See PATCH 2/4 for the
> > > details. Comments welcome.
> > >
> > > Then it switches drivers to use them. First, to
> > > DEFINE_UART_PORT_TX_HELPER() in 3/4 and then
> > > DEFINE_UART_PORT_TX_HELPER_LIMITED() in 4/4.
> > >
> > > The diffstat of patches 3+4 is as follows:
> > > 26 files changed, 191 insertions(+), 823 deletions(-)
> > > which appears to be nice.
> >
> > Not really. This is horrid. Quality can't be measured in LoC (only).
> >
> > The resulting code is unreadable. And for no good reason.
>
> IMO, it's much more readable than the original ~ 30 various (and buggy -- see
> Ilpo's fixes) copies of this code. Apart from that, it makes further rework
> much easier (I have switch to kfifo in my mind for example).
>
> > [ And note that you're "saving" something like 20 lines per driver:
>
> It's not about saving, it's about deduplicating and unifying.
>
> > 12 files changed, 84 insertions(+), 349 deletions(-)
> > ]
> >
> > NAK
>
> I'd love to come up with something nicer. That would be a function in
> serial-core calling hooks like I had [1] for example. But provided all those
> CPU workarounds/thunks, it'd be quite expensive to call two functions per
> character.
>
> Or creating a static inline (having ± the macro content) and the hooks as
> parameters and hope for optimizations to eliminate thunks (also suggested in
> the past [1]).
>
> [1] https://lore.kernel.org/all/[email protected]/
I second Jiri here.
Saving lines in drivers is not that important compared with all removing
all the variants of the same thing that have crept there over the years.
I suspect the main reason for the variants is that everybody just used
other drivers as examples and therefore we've a few "main" variant
branches depending on which of the drivers was used as an example for the
other. That is hardly a good enough reason to keep them different and as
long as each driver keeps its own function for this, it will eventually
lead to similar differentiation so e.g. a one-time band-aid similarization
would not help in the long run.
Also, I don't understand why you see it unreadable when the actual code is
out in the open in that macro. It's formatted much better than e.g.
read_poll_timeout() if you want an example of something that is hardly
readable ;-). I agree though there's a learning-curve, albeit small, that
it actually creates a function but that doesn't seem to me as big of an
obstacle you seem to think.
--
i.
On Wed, Sep 7, 2022, at 12:16 PM, Ilpo Järvinen wrote:
> On Wed, 7 Sep 2022, Jiri Slaby wrote:
>> On 06. 09. 22, 13:30, Johan Hovold wrote:
>> > On Tue, Sep 06, 2022 at 12:48:01PM +0200, Jiri Slaby wrote:
>> > NAK
>>
>> I'd love to come up with something nicer. That would be a function in
>> serial-core calling hooks like I had [1] for example. But provided all those
>> CPU workarounds/thunks, it'd be quite expensive to call two functions per
>> character.
>>
>> Or creating a static inline (having ± the macro content) and the hooks as
>> parameters and hope for optimizations to eliminate thunks (also suggested in
>> the past [1]).
>>
>> [1] https://lore.kernel.org/all/[email protected]/
>
> I second Jiri here.
>
> Saving lines in drivers is not that important compared with all removing
> all the variants of the same thing that have crept there over the years.
>
> I suspect the main reason for the variants is that everybody just used
> other drivers as examples and therefore we've a few "main" variant
> branches depending on which of the drivers was used as an example for the
> other. That is hardly a good enough reason to keep them different and as
> long as each driver keeps its own function for this, it will eventually
> lead to similar differentiation so e.g. a one-time band-aid similarization
> would not help in the long run.
>
> Also, I don't understand why you see it unreadable when the actual code is
> out in the open in that macro. It's formatted much better than e.g.
> read_poll_timeout() if you want an example of something that is hardly
> readable ;-). I agree though there's a learning-curve, albeit small, that
> it actually creates a function but that doesn't seem to me as big of an
> obstacle you seem to think.
I think it would help to replace the macro that defines
the function with a set of macros that can be used in
function bodies. This would avoid the __VA_ARGS__ stuff
and allow readers that are unfamiliar with tty drivers to
treat it as a function call.
So e.g. instead of
static DEFINE_UART_PORT_TX_HELPER_LIMITED(altera_jtaguart_do_tx_chars,
true,
writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG),
({}));
the altera_jtaguart driver would contain a function like
static int altera_jtaguart_do_tx_chars(struct uart_port *port,
unsigned int count)
{
char ch;
return uart_port_tx_helper_limited(port, ch, count, true,
writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG),
({}));
}
or some variation of that. It's a few more lines, but those
extra lines would help me understand what is actually going on
while still avoiding the usual bugs and duplication.
If the caller of that function is itself trivial (like
serial21285_tx_chars), then the intermediate function can
be omitted in order to save some of the extra complexity.
Arnd
On Wed, Sep 07, 2022 at 03:32:44PM +0300, Ilpo J?rvinen wrote:
> On Wed, 7 Sep 2022, Greg Kroah-Hartman wrote:
>
> > On Wed, Sep 07, 2022 at 03:21:28PM +0300, Ilpo J?rvinen wrote:
> > > On Wed, 7 Sep 2022, Arnd Bergmann wrote:
> > >
> > > > On Wed, Sep 7, 2022, at 12:16 PM, Ilpo J?rvinen wrote:
> > > > > On Wed, 7 Sep 2022, Jiri Slaby wrote:
> > > > >> On 06. 09. 22, 13:30, Johan Hovold wrote:
> > > > >> > On Tue, Sep 06, 2022 at 12:48:01PM +0200, Jiri Slaby wrote:
> > > > >> > NAK
> > > > >>
> > > > >> I'd love to come up with something nicer. That would be a function in
> > > > >> serial-core calling hooks like I had [1] for example. But provided all those
> > > > >> CPU workarounds/thunks, it'd be quite expensive to call two functions per
> > > > >> character.
> > > > >>
> > > > >> Or creating a static inline (having ? the macro content) and the hooks as
> > > > >> parameters and hope for optimizations to eliminate thunks (also suggested in
> > > > >> the past [1]).
> > > > >>
> > > > >> [1] https://lore.kernel.org/all/[email protected]/
> > > > >
> > > > > I second Jiri here.
> > > > >
> > > > > Saving lines in drivers is not that important compared with all removing
> > > > > all the variants of the same thing that have crept there over the years.
> > > > >
> > > > > I suspect the main reason for the variants is that everybody just used
> > > > > other drivers as examples and therefore we've a few "main" variant
> > > > > branches depending on which of the drivers was used as an example for the
> > > > > other. That is hardly a good enough reason to keep them different and as
> > > > > long as each driver keeps its own function for this, it will eventually
> > > > > lead to similar differentiation so e.g. a one-time band-aid similarization
> > > > > would not help in the long run.
> > > > >
> > > > > Also, I don't understand why you see it unreadable when the actual code is
> > > > > out in the open in that macro. It's formatted much better than e.g.
> > > > > read_poll_timeout() if you want an example of something that is hardly
> > > > > readable ;-). I agree though there's a learning-curve, albeit small, that
> > > > > it actually creates a function but that doesn't seem to me as big of an
> > > > > obstacle you seem to think.
> > > >
> > > > I think it would help to replace the macro that defines
> > > > the function with a set of macros that can be used in
> > > > function bodies. This would avoid the __VA_ARGS__ stuff
> > > > and allow readers that are unfamiliar with tty drivers to
> > > > treat it as a function call.
> > > >
> > > > So e.g. instead of
> > > >
> > > > static DEFINE_UART_PORT_TX_HELPER_LIMITED(altera_jtaguart_do_tx_chars,
> > > > true,
> > > > writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG),
> > > > ({}));
> > > >
> > > > the altera_jtaguart driver would contain a function like
> > > >
> > > > static int altera_jtaguart_do_tx_chars(struct uart_port *port,
> > > > unsigned int count)
> > > > {
> > > > char ch;
> > > >
> > > > return uart_port_tx_helper_limited(port, ch, count, true,
> > > > writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG),
> > > > ({}));
> > > > }
> > > >
> > > > or some variation of that. It's a few more lines, but those
> > > > extra lines would help me understand what is actually going on
> > > > while still avoiding the usual bugs and duplication.
> > > >
> > > > If the caller of that function is itself trivial (like
> > > > serial21285_tx_chars), then the intermediate function can
> > > > be omitted in order to save some of the extra complexity.
> > >
> > > I'd be ok with that. There's still a small startle factor associated to
> > > passing that writel(...) as an argument to a "function" but it's the same
> > > for other things such as read_poll_timeout() so not an end of the world.
> >
> > That's going to incure the function-pointer-indirection-call for every
> > character that Jiri's original submission had, so I don't think this is
> > a very viable solution, sorry.
>
> I don't think you got what Arnd meant. It must still be technically a
> #define because you cannot pass
> writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG)
> as an argument to a real function like he did in the example above.
> It's similar to how read_poll_timeout() and friends are #defines despite
> being lowercased.
Ok, no, I don't understand what Arnd meant here then :(
confused,
greg k-h
On Wed, Sep 07, 2022 at 03:21:28PM +0300, Ilpo J?rvinen wrote:
> On Wed, 7 Sep 2022, Arnd Bergmann wrote:
>
> > On Wed, Sep 7, 2022, at 12:16 PM, Ilpo J?rvinen wrote:
> > > On Wed, 7 Sep 2022, Jiri Slaby wrote:
> > >> On 06. 09. 22, 13:30, Johan Hovold wrote:
> > >> > On Tue, Sep 06, 2022 at 12:48:01PM +0200, Jiri Slaby wrote:
> > >> > NAK
> > >>
> > >> I'd love to come up with something nicer. That would be a function in
> > >> serial-core calling hooks like I had [1] for example. But provided all those
> > >> CPU workarounds/thunks, it'd be quite expensive to call two functions per
> > >> character.
> > >>
> > >> Or creating a static inline (having ? the macro content) and the hooks as
> > >> parameters and hope for optimizations to eliminate thunks (also suggested in
> > >> the past [1]).
> > >>
> > >> [1] https://lore.kernel.org/all/[email protected]/
> > >
> > > I second Jiri here.
> > >
> > > Saving lines in drivers is not that important compared with all removing
> > > all the variants of the same thing that have crept there over the years.
> > >
> > > I suspect the main reason for the variants is that everybody just used
> > > other drivers as examples and therefore we've a few "main" variant
> > > branches depending on which of the drivers was used as an example for the
> > > other. That is hardly a good enough reason to keep them different and as
> > > long as each driver keeps its own function for this, it will eventually
> > > lead to similar differentiation so e.g. a one-time band-aid similarization
> > > would not help in the long run.
> > >
> > > Also, I don't understand why you see it unreadable when the actual code is
> > > out in the open in that macro. It's formatted much better than e.g.
> > > read_poll_timeout() if you want an example of something that is hardly
> > > readable ;-). I agree though there's a learning-curve, albeit small, that
> > > it actually creates a function but that doesn't seem to me as big of an
> > > obstacle you seem to think.
> >
> > I think it would help to replace the macro that defines
> > the function with a set of macros that can be used in
> > function bodies. This would avoid the __VA_ARGS__ stuff
> > and allow readers that are unfamiliar with tty drivers to
> > treat it as a function call.
> >
> > So e.g. instead of
> >
> > static DEFINE_UART_PORT_TX_HELPER_LIMITED(altera_jtaguart_do_tx_chars,
> > true,
> > writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG),
> > ({}));
> >
> > the altera_jtaguart driver would contain a function like
> >
> > static int altera_jtaguart_do_tx_chars(struct uart_port *port,
> > unsigned int count)
> > {
> > char ch;
> >
> > return uart_port_tx_helper_limited(port, ch, count, true,
> > writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG),
> > ({}));
> > }
> >
> > or some variation of that. It's a few more lines, but those
> > extra lines would help me understand what is actually going on
> > while still avoiding the usual bugs and duplication.
> >
> > If the caller of that function is itself trivial (like
> > serial21285_tx_chars), then the intermediate function can
> > be omitted in order to save some of the extra complexity.
>
> I'd be ok with that. There's still a small startle factor associated to
> passing that writel(...) as an argument to a "function" but it's the same
> for other things such as read_poll_timeout() so not an end of the world.
That's going to incure the function-pointer-indirection-call for every
character that Jiri's original submission had, so I don't think this is
a very viable solution, sorry.
thanks,
greg k-h
On Wed, 7 Sep 2022, Greg Kroah-Hartman wrote:
> On Wed, Sep 07, 2022 at 03:21:28PM +0300, Ilpo J?rvinen wrote:
> > On Wed, 7 Sep 2022, Arnd Bergmann wrote:
> >
> > > On Wed, Sep 7, 2022, at 12:16 PM, Ilpo J?rvinen wrote:
> > > > On Wed, 7 Sep 2022, Jiri Slaby wrote:
> > > >> On 06. 09. 22, 13:30, Johan Hovold wrote:
> > > >> > On Tue, Sep 06, 2022 at 12:48:01PM +0200, Jiri Slaby wrote:
> > > >> > NAK
> > > >>
> > > >> I'd love to come up with something nicer. That would be a function in
> > > >> serial-core calling hooks like I had [1] for example. But provided all those
> > > >> CPU workarounds/thunks, it'd be quite expensive to call two functions per
> > > >> character.
> > > >>
> > > >> Or creating a static inline (having ? the macro content) and the hooks as
> > > >> parameters and hope for optimizations to eliminate thunks (also suggested in
> > > >> the past [1]).
> > > >>
> > > >> [1] https://lore.kernel.org/all/[email protected]/
> > > >
> > > > I second Jiri here.
> > > >
> > > > Saving lines in drivers is not that important compared with all removing
> > > > all the variants of the same thing that have crept there over the years.
> > > >
> > > > I suspect the main reason for the variants is that everybody just used
> > > > other drivers as examples and therefore we've a few "main" variant
> > > > branches depending on which of the drivers was used as an example for the
> > > > other. That is hardly a good enough reason to keep them different and as
> > > > long as each driver keeps its own function for this, it will eventually
> > > > lead to similar differentiation so e.g. a one-time band-aid similarization
> > > > would not help in the long run.
> > > >
> > > > Also, I don't understand why you see it unreadable when the actual code is
> > > > out in the open in that macro. It's formatted much better than e.g.
> > > > read_poll_timeout() if you want an example of something that is hardly
> > > > readable ;-). I agree though there's a learning-curve, albeit small, that
> > > > it actually creates a function but that doesn't seem to me as big of an
> > > > obstacle you seem to think.
> > >
> > > I think it would help to replace the macro that defines
> > > the function with a set of macros that can be used in
> > > function bodies. This would avoid the __VA_ARGS__ stuff
> > > and allow readers that are unfamiliar with tty drivers to
> > > treat it as a function call.
> > >
> > > So e.g. instead of
> > >
> > > static DEFINE_UART_PORT_TX_HELPER_LIMITED(altera_jtaguart_do_tx_chars,
> > > true,
> > > writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG),
> > > ({}));
> > >
> > > the altera_jtaguart driver would contain a function like
> > >
> > > static int altera_jtaguart_do_tx_chars(struct uart_port *port,
> > > unsigned int count)
> > > {
> > > char ch;
> > >
> > > return uart_port_tx_helper_limited(port, ch, count, true,
> > > writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG),
> > > ({}));
> > > }
> > >
> > > or some variation of that. It's a few more lines, but those
> > > extra lines would help me understand what is actually going on
> > > while still avoiding the usual bugs and duplication.
> > >
> > > If the caller of that function is itself trivial (like
> > > serial21285_tx_chars), then the intermediate function can
> > > be omitted in order to save some of the extra complexity.
> >
> > I'd be ok with that. There's still a small startle factor associated to
> > passing that writel(...) as an argument to a "function" but it's the same
> > for other things such as read_poll_timeout() so not an end of the world.
>
> That's going to incure the function-pointer-indirection-call for every
> character that Jiri's original submission had, so I don't think this is
> a very viable solution, sorry.
I don't think you got what Arnd meant. It must still be technically a
#define because you cannot pass
writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG)
as an argument to a real function like he did in the example above.
It's similar to how read_poll_timeout() and friends are #defines despite
being lowercased.
--
i.
On Wed, 7 Sep 2022, Arnd Bergmann wrote:
> On Wed, Sep 7, 2022, at 12:16 PM, Ilpo Järvinen wrote:
> > On Wed, 7 Sep 2022, Jiri Slaby wrote:
> >> On 06. 09. 22, 13:30, Johan Hovold wrote:
> >> > On Tue, Sep 06, 2022 at 12:48:01PM +0200, Jiri Slaby wrote:
> >> > NAK
> >>
> >> I'd love to come up with something nicer. That would be a function in
> >> serial-core calling hooks like I had [1] for example. But provided all those
> >> CPU workarounds/thunks, it'd be quite expensive to call two functions per
> >> character.
> >>
> >> Or creating a static inline (having ± the macro content) and the hooks as
> >> parameters and hope for optimizations to eliminate thunks (also suggested in
> >> the past [1]).
> >>
> >> [1] https://lore.kernel.org/all/[email protected]/
> >
> > I second Jiri here.
> >
> > Saving lines in drivers is not that important compared with all removing
> > all the variants of the same thing that have crept there over the years.
> >
> > I suspect the main reason for the variants is that everybody just used
> > other drivers as examples and therefore we've a few "main" variant
> > branches depending on which of the drivers was used as an example for the
> > other. That is hardly a good enough reason to keep them different and as
> > long as each driver keeps its own function for this, it will eventually
> > lead to similar differentiation so e.g. a one-time band-aid similarization
> > would not help in the long run.
> >
> > Also, I don't understand why you see it unreadable when the actual code is
> > out in the open in that macro. It's formatted much better than e.g.
> > read_poll_timeout() if you want an example of something that is hardly
> > readable ;-). I agree though there's a learning-curve, albeit small, that
> > it actually creates a function but that doesn't seem to me as big of an
> > obstacle you seem to think.
>
> I think it would help to replace the macro that defines
> the function with a set of macros that can be used in
> function bodies. This would avoid the __VA_ARGS__ stuff
> and allow readers that are unfamiliar with tty drivers to
> treat it as a function call.
>
> So e.g. instead of
>
> static DEFINE_UART_PORT_TX_HELPER_LIMITED(altera_jtaguart_do_tx_chars,
> true,
> writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG),
> ({}));
>
> the altera_jtaguart driver would contain a function like
>
> static int altera_jtaguart_do_tx_chars(struct uart_port *port,
> unsigned int count)
> {
> char ch;
>
> return uart_port_tx_helper_limited(port, ch, count, true,
> writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG),
> ({}));
> }
>
> or some variation of that. It's a few more lines, but those
> extra lines would help me understand what is actually going on
> while still avoiding the usual bugs and duplication.
>
> If the caller of that function is itself trivial (like
> serial21285_tx_chars), then the intermediate function can
> be omitted in order to save some of the extra complexity.
I'd be ok with that. There's still a small startle factor associated to
passing that writel(...) as an argument to a "function" but it's the same
for other things such as read_poll_timeout() so not an end of the world.
--
i.
On Wed, 7 Sep 2022, Greg Kroah-Hartman wrote:
> On Wed, Sep 07, 2022 at 03:32:44PM +0300, Ilpo J?rvinen wrote:
> > On Wed, 7 Sep 2022, Greg Kroah-Hartman wrote:
> >
> > > On Wed, Sep 07, 2022 at 03:21:28PM +0300, Ilpo J?rvinen wrote:
> > > > On Wed, 7 Sep 2022, Arnd Bergmann wrote:
> > > >
> > > > > On Wed, Sep 7, 2022, at 12:16 PM, Ilpo J?rvinen wrote:
> > > > > > On Wed, 7 Sep 2022, Jiri Slaby wrote:
> > > > > >> On 06. 09. 22, 13:30, Johan Hovold wrote:
> > > > > >> > On Tue, Sep 06, 2022 at 12:48:01PM +0200, Jiri Slaby wrote:
> > > > > >> > NAK
> > > > > >>
> > > > > >> I'd love to come up with something nicer. That would be a function in
> > > > > >> serial-core calling hooks like I had [1] for example. But provided all those
> > > > > >> CPU workarounds/thunks, it'd be quite expensive to call two functions per
> > > > > >> character.
> > > > > >>
> > > > > >> Or creating a static inline (having ? the macro content) and the hooks as
> > > > > >> parameters and hope for optimizations to eliminate thunks (also suggested in
> > > > > >> the past [1]).
> > > > > >>
> > > > > >> [1] https://lore.kernel.org/all/[email protected]/
> > > > > >
> > > > > > I second Jiri here.
> > > > > >
> > > > > > Saving lines in drivers is not that important compared with all removing
> > > > > > all the variants of the same thing that have crept there over the years.
> > > > > >
> > > > > > I suspect the main reason for the variants is that everybody just used
> > > > > > other drivers as examples and therefore we've a few "main" variant
> > > > > > branches depending on which of the drivers was used as an example for the
> > > > > > other. That is hardly a good enough reason to keep them different and as
> > > > > > long as each driver keeps its own function for this, it will eventually
> > > > > > lead to similar differentiation so e.g. a one-time band-aid similarization
> > > > > > would not help in the long run.
> > > > > >
> > > > > > Also, I don't understand why you see it unreadable when the actual code is
> > > > > > out in the open in that macro. It's formatted much better than e.g.
> > > > > > read_poll_timeout() if you want an example of something that is hardly
> > > > > > readable ;-). I agree though there's a learning-curve, albeit small, that
> > > > > > it actually creates a function but that doesn't seem to me as big of an
> > > > > > obstacle you seem to think.
> > > > >
> > > > > I think it would help to replace the macro that defines
> > > > > the function with a set of macros that can be used in
> > > > > function bodies. This would avoid the __VA_ARGS__ stuff
> > > > > and allow readers that are unfamiliar with tty drivers to
> > > > > treat it as a function call.
> > > > >
> > > > > So e.g. instead of
> > > > >
> > > > > static DEFINE_UART_PORT_TX_HELPER_LIMITED(altera_jtaguart_do_tx_chars,
> > > > > true,
> > > > > writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG),
> > > > > ({}));
> > > > >
> > > > > the altera_jtaguart driver would contain a function like
> > > > >
> > > > > static int altera_jtaguart_do_tx_chars(struct uart_port *port,
> > > > > unsigned int count)
> > > > > {
> > > > > char ch;
> > > > >
> > > > > return uart_port_tx_helper_limited(port, ch, count, true,
> > > > > writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG),
> > > > > ({}));
> > > > > }
> > > > >
> > > > > or some variation of that. It's a few more lines, but those
> > > > > extra lines would help me understand what is actually going on
> > > > > while still avoiding the usual bugs and duplication.
> > > > >
> > > > > If the caller of that function is itself trivial (like
> > > > > serial21285_tx_chars), then the intermediate function can
> > > > > be omitted in order to save some of the extra complexity.
> > > >
> > > > I'd be ok with that. There's still a small startle factor associated to
> > > > passing that writel(...) as an argument to a "function" but it's the same
> > > > for other things such as read_poll_timeout() so not an end of the world.
> > >
> > > That's going to incure the function-pointer-indirection-call for every
> > > character that Jiri's original submission had, so I don't think this is
> > > a very viable solution, sorry.
> >
> > I don't think you got what Arnd meant. It must still be technically a
> > #define because you cannot pass
> > writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG)
> > as an argument to a real function like he did in the example above.
> > It's similar to how read_poll_timeout() and friends are #defines despite
> > being lowercased.
>
> Ok, no, I don't understand what Arnd meant here then :(
I think he did just 2 things (compared with 2/4 of this series):
- __DEFINE_UART_PORT_TX_HELPER() doesn't create a function but just wraps
the macro body with ({ }). Therefore, a driver must use it inside
a function rather than the macro creating a new function with name.
- Use lowercase instead of uppercase (this is a trivial change)
When a driver is using these "function" that are just made look functions
but are macros for real, it will not add function-pointer-indirection-call
per character but pulls the whole loop from the macro into the function
in the driver and injects those putchar, etc. directly into that loop.
...It is just like read_poll_timeout() constructs the loop.
--
i.
On Wed, Sep 07, 2022 at 02:36:37PM +0200, Greg Kroah-Hartman wrote:
> On Wed, Sep 07, 2022 at 03:32:44PM +0300, Ilpo J?rvinen wrote:
> > On Wed, 7 Sep 2022, Greg Kroah-Hartman wrote:
> >
> > > On Wed, Sep 07, 2022 at 03:21:28PM +0300, Ilpo J?rvinen wrote:
> > > > On Wed, 7 Sep 2022, Arnd Bergmann wrote:
> > > >
> > > > > On Wed, Sep 7, 2022, at 12:16 PM, Ilpo J?rvinen wrote:
> > > > > > On Wed, 7 Sep 2022, Jiri Slaby wrote:
> > > > > >> On 06. 09. 22, 13:30, Johan Hovold wrote:
> > > > > >> > On Tue, Sep 06, 2022 at 12:48:01PM +0200, Jiri Slaby wrote:
> > > > > >> > NAK
> > > > > >>
> > > > > >> I'd love to come up with something nicer. That would be a function in
> > > > > >> serial-core calling hooks like I had [1] for example. But provided all those
> > > > > >> CPU workarounds/thunks, it'd be quite expensive to call two functions per
> > > > > >> character.
> > > > > >>
> > > > > >> Or creating a static inline (having ? the macro content) and the hooks as
> > > > > >> parameters and hope for optimizations to eliminate thunks (also suggested in
> > > > > >> the past [1]).
> > > > > >>
> > > > > >> [1] https://lore.kernel.org/all/[email protected]/
> > > > > >
> > > > > > I second Jiri here.
> > > > > >
> > > > > > Saving lines in drivers is not that important compared with all removing
> > > > > > all the variants of the same thing that have crept there over the years.
> > > > > >
> > > > > > I suspect the main reason for the variants is that everybody just used
> > > > > > other drivers as examples and therefore we've a few "main" variant
> > > > > > branches depending on which of the drivers was used as an example for the
> > > > > > other. That is hardly a good enough reason to keep them different and as
> > > > > > long as each driver keeps its own function for this, it will eventually
> > > > > > lead to similar differentiation so e.g. a one-time band-aid similarization
> > > > > > would not help in the long run.
> > > > > >
> > > > > > Also, I don't understand why you see it unreadable when the actual code is
> > > > > > out in the open in that macro. It's formatted much better than e.g.
> > > > > > read_poll_timeout() if you want an example of something that is hardly
> > > > > > readable ;-). I agree though there's a learning-curve, albeit small, that
> > > > > > it actually creates a function but that doesn't seem to me as big of an
> > > > > > obstacle you seem to think.
> > > > >
> > > > > I think it would help to replace the macro that defines
> > > > > the function with a set of macros that can be used in
> > > > > function bodies. This would avoid the __VA_ARGS__ stuff
> > > > > and allow readers that are unfamiliar with tty drivers to
> > > > > treat it as a function call.
> > > > >
> > > > > So e.g. instead of
> > > > >
> > > > > static DEFINE_UART_PORT_TX_HELPER_LIMITED(altera_jtaguart_do_tx_chars,
> > > > > true,
> > > > > writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG),
> > > > > ({}));
> > > > >
> > > > > the altera_jtaguart driver would contain a function like
> > > > >
> > > > > static int altera_jtaguart_do_tx_chars(struct uart_port *port,
> > > > > unsigned int count)
> > > > > {
> > > > > char ch;
> > > > >
> > > > > return uart_port_tx_helper_limited(port, ch, count, true,
> > > > > writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG),
> > > > > ({}));
> > > > > }
> > > > >
> > > > > or some variation of that. It's a few more lines, but those
> > > > > extra lines would help me understand what is actually going on
> > > > > while still avoiding the usual bugs and duplication.
> > > > >
> > > > > If the caller of that function is itself trivial (like
> > > > > serial21285_tx_chars), then the intermediate function can
> > > > > be omitted in order to save some of the extra complexity.
> > > >
> > > > I'd be ok with that. There's still a small startle factor associated to
> > > > passing that writel(...) as an argument to a "function" but it's the same
> > > > for other things such as read_poll_timeout() so not an end of the world.
> > >
> > > That's going to incure the function-pointer-indirection-call for every
> > > character that Jiri's original submission had, so I don't think this is
> > > a very viable solution, sorry.
> >
> > I don't think you got what Arnd meant. It must still be technically a
> > #define because you cannot pass
> > writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG)
> > as an argument to a real function like he did in the example above.
> > It's similar to how read_poll_timeout() and friends are #defines despite
> > being lowercased.
>
> Ok, no, I don't understand what Arnd meant here then :(
Arnd's opening sentence when proposing the new thing was:
"I think it would help to replace the macro that defines
the function with a set of macros that can be used in
function bodies."
Given "a set of macros" and the quoted example, I take
"uart_port_tx_helper_limited" to be one of those macros - it has to be
a macro because - as others have pointed out - you can't pass
"writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG)" to a function.
Of course, it would have been nicer to see the definition of this
macro, because then we can understand what the "ch" argument is to
this macro, and how that relates to the macro argument that is
shown in the example as a writel().
Maybe a more complete example would help clear up the confusion?
Arnd?
--
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!
On Wed, Sep 7, 2022, at 2:56 PM, Ilpo Järvinen wrote:
> On Wed, 7 Sep 2022, Greg Kroah-Hartman wrote:
>> Ok, no, I don't understand what Arnd meant here then :(
>
> I think he did just 2 things (compared with 2/4 of this series):
>
> - __DEFINE_UART_PORT_TX_HELPER() doesn't create a function but just wraps
> the macro body with ({ }). Therefore, a driver must use it inside
> a function rather than the macro creating a new function with name.
> - Use lowercase instead of uppercase (this is a trivial change)
>
> When a driver is using these "function" that are just made look functions
> but are macros for real, it will not add function-pointer-indirection-call
> per character but pulls the whole loop from the macro into the function
> in the driver and injects those putchar, etc. directly into that loop.
> ...It is just like read_poll_timeout() constructs the loop.
Right, the example I was thinking of was wait_event(), which is
another common macro that takes expressions as arguments but gets
used like a function call. Same thing really.
Note that even with an always_inline function that takes a
function pointer argument, anything from gcc-4.7 up has no
problem inlining through that, so it should not be any less
efficient though it might not help readability either.
See https://godbolt.org/z/o8GGG8TzM for an example of that.
Arnd
On Wed, Sep 7, 2022, at 3:52 PM, Russell King (Oracle) wrote:
> On Wed, Sep 07, 2022 at 02:36:37PM +0200, Greg Kroah-Hartman wrote:
>
> Of course, it would have been nicer to see the definition of this
> macro, because then we can understand what the "ch" argument is to
> this macro, and how that relates to the macro argument that is
> shown in the example as a writel().
I pulled out the 'ch' variable from the macro to avoid having
the macro define local variables that are then passed to the
inner expressions.
> Maybe a more complete example would help clear up the confusion?
> Arnd?
Here is a patch on top of the series that would implement the
uart_port_tx_helper_limited() and uart_port_tx_helper()
macros that can be used directly from drivers in place of defining
local functions, with the (alphabetically) first two drivers
converted to that.
I left the (now trivial) DEFINE_UART_PORT_TX_HELPER_LIMITED() and
DEFINE_UART_PORT_TX_HELPER() macros in place to keep it building,
but they would get removed if we decide to use something like
my suggested approach for all drivers.
Arnd
diff --git a/drivers/tty/serial/21285.c b/drivers/tty/serial/21285.c
index 40cf1bb534f3..a0f5c59d6128 100644
--- a/drivers/tty/serial/21285.c
+++ b/drivers/tty/serial/21285.c
@@ -151,16 +151,14 @@ static irqreturn_t serial21285_rx_chars(int irq, void *dev_id)
return IRQ_HANDLED;
}
-static DEFINE_UART_PORT_TX_HELPER_LIMITED(serial21285_do_tx_chars,
- !(*CSR_UARTFLG & 0x20),
- *CSR_UARTDR = ch,
- ({}));
-
static irqreturn_t serial21285_tx_chars(int irq, void *dev_id)
{
struct uart_port *port = dev_id;
+ unsigned int count = 256;
+ unsigned char ch;
- serial21285_do_tx_chars(port, 256);
+ uart_port_tx_helper_limited(port, !(*CSR_UARTFLG & 0x20),
+ *CSR_UARTDR = ch, ({}), count, ch);
return IRQ_HANDLED;
}
diff --git a/drivers/tty/serial/altera_uart.c b/drivers/tty/serial/altera_uart.c
index 70c0ad431cf9..f81dd950cd39 100644
--- a/drivers/tty/serial/altera_uart.c
+++ b/drivers/tty/serial/altera_uart.c
@@ -246,10 +246,14 @@ static void altera_uart_rx_chars(struct altera_uart *pp)
tty_flip_buffer_push(&port->state->port);
}
-static DEFINE_UART_PORT_TX_HELPER(altera_uart_tx_chars,
- altera_uart_readl(port, ALTERA_UART_STATUS_REG) &
- ALTERA_UART_STATUS_TRDY_MSK,
- altera_uart_writel(port, ch, ALTERA_UART_TXDATA_REG));
+static int altera_uart_tx_chars(struct uart_port *port)
+{
+ u8 ch;
+
+ return uart_port_tx_helper(port,
+ altera_uart_readl(port, ALTERA_UART_STATUS_REG) & ALTERA_UART_STATUS_TRDY_MSK,
+ altera_uart_writel(port, ch, ALTERA_UART_TXDATA_REG), ch);
+}
static irqreturn_t altera_uart_interrupt(int irq, void *data)
{
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index 7236fc76ba22..d48d2301d1b7 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -638,13 +638,11 @@ struct uart_driver {
void uart_write_wakeup(struct uart_port *port);
-#define __DEFINE_UART_PORT_TX_HELPER(name, tx_ready, put_char, tx_done, \
- for_test, for_post, ...) \
-unsigned int name(struct uart_port *port __VA_OPT__(,) __VA_ARGS__) \
-{ \
+#define __uart_port_tx_helper(port, tx_ready, put_char, tx_done, \
+ for_test, for_post, ch) \
+({ \
struct circ_buf *xmit = &port->state->xmit; \
unsigned int pending; \
- u8 ch; \
\
for (; (for_test) && (tx_ready); (for_post), port->icount.tx++) { \
if (port->x_char) { \
@@ -672,8 +670,15 @@ unsigned int name(struct uart_port *port __VA_OPT__(,) __VA_ARGS__) \
port->ops->stop_tx(port); \
} \
\
- return pending; \
-}
+ pending; \
+})
+
+#define uart_port_tx_helper_limited(port, tx_ready, put_char, tx_done, count, ch) \
+ __uart_port_tx_helper(port, tx_ready, put_char, tx_done, count, count--, ch)
+
+#define uart_port_tx_helper(port, tx_ready, put_char, ch) \
+ __uart_port_tx_helper(port, tx_ready, put_char, ({}), true, ({}), ch)
+
/**
* DEFINE_UART_PORT_TX_HELPER_LIMITED -- generate transmit helper for uart_port
@@ -703,9 +708,13 @@ unsigned int name(struct uart_port *port __VA_OPT__(,) __VA_ARGS__) \
* For all of them, @port->lock is held, interrupts are locally disabled and
* the expressions must not sleep.
*/
-#define DEFINE_UART_PORT_TX_HELPER_LIMITED(name, tx_ready, put_char, tx_done) \
- __DEFINE_UART_PORT_TX_HELPER(name, tx_ready, put_char, tx_done, \
- count, count--, unsigned int count)
+#define DEFINE_UART_PORT_TX_HELPER_LIMITED(name, tx_ready, put_char, tx_done) \
+unsigned int name(struct uart_port *port, unsigned int count) \
+{ \
+ u8 ch; \
+ return uart_port_tx_helper_limited(port, tx_ready, put_char, tx_done, \
+ count, ch); \
+}
/**
* DEFINE_UART_PORT_TX_HELPER -- generate transmit helper for uart_port
@@ -715,9 +724,12 @@ unsigned int name(struct uart_port *port __VA_OPT__(,) __VA_ARGS__) \
*
* See DEFINE_UART_PORT_TX_HELPER_LIMITED() for more details.
*/
-#define DEFINE_UART_PORT_TX_HELPER(name, tx_ready, put_char) \
- __DEFINE_UART_PORT_TX_HELPER(name, tx_ready, put_char, ({}), \
- true, ({}))
+#define DEFINE_UART_PORT_TX_HELPER(name, tx_ready, put_char, ...) \
+unsigned int name(struct uart_port *port __VA_OPT__(,) __VA_ARGS__) \
+{ \
+ u8 ch; \
+ return uart_port_tx_helper(port, tx_ready, put_char, ch); \
+}
/*
* Baud rate helpers.
On 07. 09. 22, 16:56, Arnd Bergmann wrote:
> On Wed, Sep 7, 2022, at 3:52 PM, Russell King (Oracle) wrote:
>> On Wed, Sep 07, 2022 at 02:36:37PM +0200, Greg Kroah-Hartman wrote:
>>
>> Of course, it would have been nicer to see the definition of this
>> macro, because then we can understand what the "ch" argument is to
>> this macro, and how that relates to the macro argument that is
>> shown in the example as a writel().
>
> I pulled out the 'ch' variable from the macro to avoid having
> the macro define local variables that are then passed to the
> inner expressions.
Note that I had "port" and "ch" as a part of the macro parameters in
[v2], but it didn't help the situation much.
>> Maybe a more complete example would help clear up the confusion?
>> Arnd?
>
> Here is a patch on top of the series that would implement the
> uart_port_tx_helper_limited() and uart_port_tx_helper()
> macros that can be used directly from drivers in place of defining
> local functions, with the (alphabetically) first two drivers
> converted to that.
If there are no objections, I will push the patches this directorin. I
like this more than [v2] or [v3] (the helper macros). Actually, I
mentioned this wait_event() style in [v1], but I perhaps simplified the
concept too much to completely eliminate the need of a wrapper function.
And that made it too complicated/too hard to understand.
Except I'd drop the "_helper" part from the name. Originally (in [v1]),
I had uart_port_tx() and uart_port_tx_limited() functions. In [v2+v3], I
added _helper to avoid confusion as we were generating a helpers using
the macros. Yes, technically, uart_port_tx() is still a helper, but I
think it's superfluous to have it in the name now.
thanks,
--
js
suse labs
On Fri, Sep 09, 2022 at 12:53:04PM +0200, Jiri Slaby wrote:
> On 07. 09. 22, 16:56, Arnd Bergmann wrote:
> > On Wed, Sep 7, 2022, at 3:52 PM, Russell King (Oracle) wrote:
> > > On Wed, Sep 07, 2022 at 02:36:37PM +0200, Greg Kroah-Hartman wrote:
> > >
> > > Of course, it would have been nicer to see the definition of this
> > > macro, because then we can understand what the "ch" argument is to
> > > this macro, and how that relates to the macro argument that is
> > > shown in the example as a writel().
> >
> > I pulled out the 'ch' variable from the macro to avoid having
> > the macro define local variables that are then passed to the
> > inner expressions.
>
> Note that I had "port" and "ch" as a part of the macro parameters in [v2],
> but it didn't help the situation much.
> > > Maybe a more complete example would help clear up the confusion?
> > > Arnd?
> >
> > Here is a patch on top of the series that would implement the
> > uart_port_tx_helper_limited() and uart_port_tx_helper()
> > macros that can be used directly from drivers in place of defining
> > local functions, with the (alphabetically) first two drivers
> > converted to that.
>
> If there are no objections, I will push the patches this directorin. I like
> this more than [v2] or [v3] (the helper macros). Actually, I mentioned this
> wait_event() style in [v1], but I perhaps simplified the concept too much to
> completely eliminate the need of a wrapper function. And that made it too
> complicated/too hard to understand.
>
> Except I'd drop the "_helper" part from the name. Originally (in [v1]), I
> had uart_port_tx() and uart_port_tx_limited() functions. In [v2+v3], I added
> _helper to avoid confusion as we were generating a helpers using the macros.
> Yes, technically, uart_port_tx() is still a helper, but I think it's
> superfluous to have it in the name now.
No objection from me, thanks for doing this work!
greg k-h
On Wed, Sep 07, 2022 at 01:16:23PM +0300, Ilpo Järvinen wrote:
> Also, I don't understand why you see it unreadable when the actual code is
> out in the open in that macro. It's formatted much better than e.g.
> read_poll_timeout() if you want an example of something that is hardly
> readable ;-). I agree though there's a learning-curve, albeit small, that
> it actually creates a function but that doesn't seem to me as big of an
> obstacle you seem to think.
There's a huge difference between the proposed macro with its
I-lost-count-of-how-many arguments and levels of indirection and
something like wait_event() which take one condition and has a
descriptive name:
wait_event(device->misc_wait, atomic_read(&device->ap_pending_cnt) == 0);
vs
static DEFINE_UART_PORT_TX_HELPER_LIMITED(altera_jtaguart_do_tx_chars,
true,
writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG),
({}));
In the former case, you just need to look at the code to understand what
is going on, very much unlike in the latter case.
Johan
On Wed, Sep 07, 2022 at 09:19:25AM +0200, Jiri Slaby wrote:
> On 06. 09. 22, 13:30, Johan Hovold wrote:
> > On Tue, Sep 06, 2022 at 12:48:01PM +0200, Jiri Slaby wrote:
> >> This series introduces DEFINE_UART_PORT_TX_HELPER +
> >> DEFINE_UART_PORT_TX_HELPER_LIMITED TX helpers. See PATCH 2/4 for the
> >> details. Comments welcome.
> >>
> >> Then it switches drivers to use them. First, to
> >> DEFINE_UART_PORT_TX_HELPER() in 3/4 and then
> >> DEFINE_UART_PORT_TX_HELPER_LIMITED() in 4/4.
> >>
> >> The diffstat of patches 3+4 is as follows:
> >> 26 files changed, 191 insertions(+), 823 deletions(-)
> >> which appears to be nice.
> >
> > Not really. This is horrid. Quality can't be measured in LoC (only).
> >
> > The resulting code is unreadable. And for no good reason.
>
> IMO, it's much more readable than the original ~ 30 various (and buggy
> -- see Ilpo's fixes) copies of this code. Apart from that, it makes
> further rework much easier (I have switch to kfifo in my mind for example).
Sure, but you can't have that at the cost of something that is
unreadable. Arnd's suggestions seems to improve things somewhat in this
respect, though.
> > [ And note that you're "saving" something like 20 lines per driver:
>
> It's not about saving, it's about deduplicating and unifying.
It was you who brought the diff stat that up and I put this in
parentheses for a reason.
> > 12 files changed, 84 insertions(+), 349 deletions(-)
> > ]
> >
> > NAK
>
> I'd love to come up with something nicer. That would be a function in
> serial-core calling hooks like I had [1] for example. But provided all
> those CPU workarounds/thunks, it'd be quite expensive to call two
> functions per character.
Yeah, that was even worse so no need to reiterate that.
> Or creating a static inline (having ± the macro content) and the hooks
> as parameters and hope for optimizations to eliminate thunks (also
> suggested in the past [1]).
>
> [1] https://lore.kernel.org/all/[email protected]/
If that works then that should improve readability as well.
Johan
On Fri, Sep 09, 2022 at 12:53:04PM +0200, Jiri Slaby wrote:
> On 07. 09. 22, 16:56, Arnd Bergmann wrote:
> > On Wed, Sep 7, 2022, at 3:52 PM, Russell King (Oracle) wrote:
> >> On Wed, Sep 07, 2022 at 02:36:37PM +0200, Greg Kroah-Hartman wrote:
> >>
> >> Of course, it would have been nicer to see the definition of this
> >> macro, because then we can understand what the "ch" argument is to
> >> this macro, and how that relates to the macro argument that is
> >> shown in the example as a writel().
> >
> > I pulled out the 'ch' variable from the macro to avoid having
> > the macro define local variables that are then passed to the
> > inner expressions.
>
> Note that I had "port" and "ch" as a part of the macro parameters in
> [v2], but it didn't help the situation much.
> >> Maybe a more complete example would help clear up the confusion?
> >> Arnd?
> >
> > Here is a patch on top of the series that would implement the
> > uart_port_tx_helper_limited() and uart_port_tx_helper()
> > macros that can be used directly from drivers in place of defining
> > local functions, with the (alphabetically) first two drivers
> > converted to that.
>
> If there are no objections, I will push the patches this directorin. I
> like this more than [v2] or [v3] (the helper macros). Actually, I
> mentioned this wait_event() style in [v1], but I perhaps simplified the
> concept too much to completely eliminate the need of a wrapper function.
> And that made it too complicated/too hard to understand.
This sounds much better. You also had some users that needed some
preamble which could now go in the same function (e.g.
altera_jtaguart_tx_chars()).
> Except I'd drop the "_helper" part from the name. Originally (in [v1]),
> I had uart_port_tx() and uart_port_tx_limited() functions. In [v2+v3], I
> added _helper to avoid confusion as we were generating a helpers using
> the macros. Yes, technically, uart_port_tx() is still a helper, but I
> think it's superfluous to have it in the name now.
That would also be an in improvement. For the altera example you could
end up with something like:
static void altera_jtaguart_tx_chars(struct altera_jtaguart *pp)
{
struct uart_port *port = &pp->port;
unsigned int space;
space = (readl(port->membase + ALTERA_JTAGUART_CONTROL_REG) &
ALTERA_JTAGUART_CONTROL_WSPACE_MSK) >>
ALTERA_JTAGUART_CONTROL_WSPACE_OFF;
uart_port_tx_chars(port, space,
writel(ch, port->membase + ALTERA_JTAGUART_DATA_REG));
}
which would be understandable.
If there are too many arguments to be passed in, then perhaps you can
explore Arnd's (and you own) suggestion to use inline helpers so that
the arguments are named.
Johan