Skip to contentSkip to conversation

drivers: PL011: add SBSA subset support

RFC PATCH New Andre Przywara 3 participants 14 messages Sep 19, 2014 – Sep 30, 2014

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

cover drivers: PL011: add SBSA subset support

(This version has still issues, but I'd like to have feedback on the general feasibility of this approach before wasting even more time on debugging).

The ARM Server Base System Architecture[1] document describes a generic UART which is a subset of the PL011 UART.
It lacks DMA support, baud rate control and modem status line control, among other things.
The idea is to move the UART initialization and setup into the firmware (which does this job today already) and let the kernel just use the UART for sending and receiving characters.

This patchset is an attempt to integrate support for this UART subset into the existing PL011 driver - basically by refactoring some functions and providing a new uart_ops structure for it. It also has a separate probe function to be not dependent on AMBA/PrimeCell. Beside the obvious effect of code sharing it has the advantage of not introducing another serial device prefix, so it can go with ttyAMA, which seems to be pretty common.

There is still one issue in bringing up the UART. On the PL011 we send a dummy character via the loopback facility with the FIFOs disabled to get things going, but both FIFO control and loopback are outside of the SBSA spec. I consider this loopback approach a kludge anyway, but am struggling to find a better way of fixing this due to my lack of understanding of the TTY/serial layer.
If someone with more wisdom could enlighten me, I would be grateful.

Patch 1/10 contains a bug fix which applies to the PL011 part also, it should be considered regardless of the rest of the series. Patch 2-7 refactor some PL011 functions by splitting them up into smaller pieces, so that most the code can be reused later by the SBSA part.
Patch 8 and 9 introduce two new properties for the vendor structure, this is for SBSA functionality which cannot be controlled by separate uart_ops members only.
Patch 10 then finally drops in the SBSA specific code, by providing a new uart_ops, vendor struct and probe function for it.

Please have a look and tell me whether this SBSA driver approach has any future.

Regards,
Andre

Andre Przywara (10):
drivers: PL011: avoid potential unregister_driver call
drivers: PL011: refactor pl011_startup()
drivers: PL011: refactor pl011_shutdown()
drivers: PL011: refactor pl011_set_termios()
drivers: PL011: refactor pl011_probe()
drivers: PL011: replace UART_MIS reading with _RIS & _IMSC
drivers: PL011: move cts_event workaround into separate function
drivers: PL011: allow avoiding UART enabling/disabling
drivers: PL011: allow to supply fixed option string
drivers: PL011: add support for the ARM SBSA generic UART

 .../devicetree/bindings/serial/arm_sbsa_uart.txt   |    9 +
 drivers/tty/serial/amba-pl011.c                    |  509 ++++++++++++++------
 2 files changed, 380 insertions(+), 138 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/serial/arm_sbsa_uart.txt
Signature · 1 line
1.7.9.5
#1 lore raw

RFC PATCH 1/10 drivers: PL011: avoid potential unregister_driver call

Although we care about not unregistering the driver if there are still ports connected during the .remove callback, we do miss this check in the pl011_probe function. So if the current port allocation fails, but there are other ports already registered, we will kill those.
So factor out the port removal into a separate function and use that in the probe function, too.

---
 drivers/tty/serial/amba-pl011.c |   38 +++++++++++++++++++++-----------------
 1 file changed, 21 insertions(+), 17 deletions(-)
drivers/tty/serial/amba-pl011.c +21 -17
index 8572f2a..8fab985 100644
@@ -2118,6 +2118,24 @@ static int pl011_probe_dt_alias(int index, struct device *dev)
 	return ret;
 }
 
+/* unregisters the driver also if no more ports are left */
+static void pl011_unregister_port(struct uart_amba_port *uap)
+{
+	int i;
+	bool busy = false;
+
+	for (i = 0; i < ARRAY_SIZE(amba_ports); i++) {
+		if (amba_ports[i] == uap)
+			amba_ports[i] = NULL;
+		else if (amba_ports[i])
+			busy = true;
+	}
+	pl011_dma_remove(uap);
+	if (!busy)
+		uart_unregister_driver(&amba_reg);
+}
+
+
 static int pl011_probe(struct amba_device *dev, const struct amba_id *id)
 {
 	struct uart_amba_port *uap;
@@ -2183,11 +2201,8 @@ static int pl011_probe(struct amba_device *dev, const struct amba_id *id)
 	}
 
 	ret = uart_add_one_port(&amba_reg, &uap->port);
-	if (ret) {
-		amba_ports[i] = NULL;
-		uart_unregister_driver(&amba_reg);
-		pl011_dma_remove(uap);
-	}
+	if (ret)
+		pl011_unregister_port(uap);
 
 	return ret;
 }
@@ -2195,20 +2210,9 @@ static int pl011_probe(struct amba_device *dev, const struct amba_id *id)
 static int pl011_remove(struct amba_device *dev)
 {
 	struct uart_amba_port *uap = amba_get_drvdata(dev);
-	bool busy = false;
-	int i;
 
 	uart_remove_one_port(&amba_reg, &uap->port);
-
-	for (i = 0; i < ARRAY_SIZE(amba_ports); i++)
-		if (amba_ports[i] == uap)
-			amba_ports[i] = NULL;
-		else if (amba_ports[i])
-			busy = true;
-
-	pl011_dma_remove(uap);
-	if (!busy)
-		uart_unregister_driver(&amba_reg);
+	pl011_unregister_port(uap);
 	return 0;
 }
 
Signature · 1 line
1.7.9.5
#2 lore raw

RFC PATCH 2/10 drivers: PL011: refactor pl011_startup()

In reply to Andre Przywara

Split the pl011_startup() function into smaller chunks to allow easier reuse later when adding SBSA support.

---
 drivers/tty/serial/amba-pl011.c |   51 +++++++++++++++++++++++----------------
 1 file changed, 30 insertions(+), 21 deletions(-)
drivers/tty/serial/amba-pl011.c +30 -21
index 8fab985..22f113c 100644
@@ -1524,6 +1524,32 @@ static void pl011_write_lcr_h(struct uart_amba_port *uap, unsigned int lcr_h)
 	}
 }
 
+static int pl011_allocate_irq(struct uart_amba_port *uap)
+{
+	writew(uap->im, uap->port.membase + UART011_IMSC);
+
+	return request_irq(uap->port.irq, pl011_int, 0, "uart-pl011", uap);
+}
+
+/*
+ * Enable interrupts, only timeouts when using DMA
+ * if initial RX DMA job failed, start in interrupt mode
+ * as well.
+ */
+static void pl011_enable_interrupts(struct uart_amba_port *uap)
+{
+	spin_lock_irq(&uap->port.lock);
+
+	/* Clear out any spuriously appearing RX interrupts */
+	writew(UART011_RTIS | UART011_RXIS,
+		uap->port.membase + UART011_ICR);
+	uap->im = UART011_RTIM;
+	if (!pl011_dma_rx_running(uap))
+		uap->im |= UART011_RXIM;
+	writew(uap->im, uap->port.membase + UART011_IMSC);
+	spin_unlock_irq(&uap->port.lock);
+}
+
 static int pl011_startup(struct uart_port *port)
 {
 	struct uart_amba_port *uap = (struct uart_amba_port *)port;
@@ -1534,19 +1560,15 @@ static int pl011_startup(struct uart_port *port)
 	if (retval)
 		goto clk_dis;
 
-	writew(uap->im, uap->port.membase + UART011_IMSC);
-
-	/*
-	 * Allocate the IRQ
-	 */
-	retval = request_irq(uap->port.irq, pl011_int, 0, "uart-pl011", uap);
+	retval = pl011_allocate_irq(uap);
 	if (retval)
 		goto clk_dis;
 
 	writew(uap->vendor->ifls, uap->port.membase + UART011_IFLS);
 
 	/*
-	 * Provoke TX FIFO interrupt into asserting. Taking care to preserve
+	 * Provoke TX FIFO interrupt into asserting by enabling loopback and
+	 * sending a dummy character. Taking care to preserve
 	 * baud rate and data format specified by FBRD, IBRD and LCRH as the
 	 * UART may already be in use as a console.
 	 */
@@ -1584,20 +1606,7 @@ static int pl011_startup(struct uart_port *port)
 	/* Startup DMA */
 	pl011_dma_startup(uap);
 
-	/*
-	 * Finally, enable interrupts, only timeouts when using DMA
-	 * if initial RX DMA job failed, start in interrupt mode
-	 * as well.
-	 */
-	spin_lock_irq(&uap->port.lock);
-	/* Clear out any spuriously appearing RX interrupts */
-	 writew(UART011_RTIS | UART011_RXIS,
-		uap->port.membase + UART011_ICR);
-	uap->im = UART011_RTIM;
-	if (!pl011_dma_rx_running(uap))
-		uap->im |= UART011_RXIM;
-	writew(uap->im, uap->port.membase + UART011_IMSC);
-	spin_unlock_irq(&uap->port.lock);
+	pl011_enable_interrupts(uap);
 
 	return 0;
 
Signature · 1 line
1.7.9.5
#3 lore raw

RFC PATCH 3/10 drivers: PL011: refactor pl011_shutdown()

In reply to Andre Przywara

Split the pl011_shutdown() function into smaller chunks to allow easier reuse later when adding SBSA support.

---
 drivers/tty/serial/amba-pl011.c |   57 ++++++++++++++++++++++-----------------
 1 file changed, 33 insertions(+), 24 deletions(-)
drivers/tty/serial/amba-pl011.c +33 -24
index 22f113c..17ba997 100644
@@ -1625,33 +1625,15 @@ static void pl011_shutdown_channel(struct uart_amba_port *uap,
       writew(val, uap->port.membase + lcrh);
 }
 
-static void pl011_shutdown(struct uart_port *port)
+/*
+ * disable the port. It should not disable RTS and DTR.
+ * Also RTS and DTR state should be preserved to restore
+ * it during startup().
+ */
+static void pl011_disable_uart(struct uart_amba_port *uap)
 {
-	struct uart_amba_port *uap = (struct uart_amba_port *)port;
 	unsigned int cr;
 
-	/*
-	 * disable all interrupts
-	 */
-	spin_lock_irq(&uap->port.lock);
-	uap->im = 0;
-	writew(uap->im, uap->port.membase + UART011_IMSC);
-	writew(0xffff, uap->port.membase + UART011_ICR);
-	spin_unlock_irq(&uap->port.lock);
-
-	pl011_dma_shutdown(uap);
-
-	/*
-	 * Free the interrupt
-	 */
-	free_irq(uap->port.irq, uap);
-
-	/*
-	 * disable the port
-	 * disable the port. It should not disable RTS and DTR.
-	 * Also RTS and DTR state should be preserved to restore
-	 * it during startup().
-	 */
 	uap->autorts = false;
 	spin_lock_irq(&uap->port.lock);
 	cr = readw(uap->port.membase + UART011_CR);
@@ -1667,6 +1649,33 @@ static void pl011_shutdown(struct uart_port *port)
 	pl011_shutdown_channel(uap, uap->lcrh_rx);
 	if (uap->lcrh_rx != uap->lcrh_tx)
 		pl011_shutdown_channel(uap, uap->lcrh_tx);
+}
+
+static void pl011_disable_interrupts(struct uart_amba_port *uap)
+{
+	spin_lock_irq(&uap->port.lock);
+
+	/*
+	 * mask all interrupts and clear all pending ones
+	 */
+	uap->im = 0;
+	writew(uap->im, uap->port.membase + UART011_IMSC);
+	writew(0xffff, uap->port.membase + UART011_ICR);
+
+	spin_unlock_irq(&uap->port.lock);
+}
+
+static void pl011_shutdown(struct uart_port *port)
+{
+	struct uart_amba_port *uap = (struct uart_amba_port *)port;
+
+	pl011_disable_interrupts(uap);
+
+	pl011_dma_shutdown(uap);
+
+	free_irq(uap->port.irq, uap);
+
+	pl011_disable_uart(uap);
 
 	/*
 	 * Shut down the clock producer
Signature · 1 line
1.7.9.5
#4 lore raw

RFC PATCH 4/10 drivers: PL011: refactor pl011_set_termios()

In reply to Andre Przywara

Split the pl011_set_termios() function into smaller chunks to allow easier reuse later when adding SBSA support.

---
 drivers/tty/serial/amba-pl011.c |   60 +++++++++++++++++++++------------------
 1 file changed, 33 insertions(+), 27 deletions(-)
drivers/tty/serial/amba-pl011.c +33 -27
index 17ba997..9219127 100644
@@ -1695,6 +1695,38 @@ static void pl011_shutdown(struct uart_port *port)
 }
 
 static void
+pl011_setup_status_masks(struct uart_port *port, struct ktermios *termios)
+{
+	port->read_status_mask = UART011_DR_OE | 255;
+	if (termios->c_iflag & INPCK)
+		port->read_status_mask |= UART011_DR_FE | UART011_DR_PE;
+	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
+		port->read_status_mask |= UART011_DR_BE;
+
+	/*
+	 * Characters to ignore
+	 */
+	port->ignore_status_mask = 0;
+	if (termios->c_iflag & IGNPAR)
+		port->ignore_status_mask |= UART011_DR_FE | UART011_DR_PE;
+	if (termios->c_iflag & IGNBRK) {
+		port->ignore_status_mask |= UART011_DR_BE;
+		/*
+		 * If we're ignoring parity and break indicators,
+		 * ignore overruns too (for real raw support).
+		 */
+		if (termios->c_iflag & IGNPAR)
+			port->ignore_status_mask |= UART011_DR_OE;
+	}
+
+	/*
+	 * Ignore all characters if CREAD is not set.
+	 */
+	if ((termios->c_cflag & CREAD) == 0)
+		port->ignore_status_mask |= UART_DUMMY_DR_RX;
+}
+
+static void
 pl011_set_termios(struct uart_port *port, struct ktermios *termios,
 		     struct ktermios *old)
 {
@@ -1757,33 +1789,7 @@ pl011_set_termios(struct uart_port *port, struct ktermios *termios,
 	 */
 	uart_update_timeout(port, termios->c_cflag, baud);
 
-	port->read_status_mask = UART011_DR_OE | 255;
-	if (termios->c_iflag & INPCK)
-		port->read_status_mask |= UART011_DR_FE | UART011_DR_PE;
-	if (termios->c_iflag & (IGNBRK | BRKINT | PARMRK))
-		port->read_status_mask |= UART011_DR_BE;
-
-	/*
-	 * Characters to ignore
-	 */
-	port->ignore_status_mask = 0;
-	if (termios->c_iflag & IGNPAR)
-		port->ignore_status_mask |= UART011_DR_FE | UART011_DR_PE;
-	if (termios->c_iflag & IGNBRK) {
-		port->ignore_status_mask |= UART011_DR_BE;
-		/*
-		 * If we're ignoring parity and break indicators,
-		 * ignore overruns too (for real raw support).
-		 */
-		if (termios->c_iflag & IGNPAR)
-			port->ignore_status_mask |= UART011_DR_OE;
-	}
-
-	/*
-	 * Ignore all characters if CREAD is not set.
-	 */
-	if ((termios->c_cflag & CREAD) == 0)
-		port->ignore_status_mask |= UART_DUMMY_DR_RX;
+	pl011_setup_status_masks(port, termios);
 
 	if (UART_ENABLE_MS(port, termios->c_cflag))
 		pl011_enable_ms(port);
Signature · 1 line
1.7.9.5
#5 lore raw

RFC PATCH 5/10 drivers: PL011: refactor pl011_probe()

In reply to Andre Przywara

Currently the pl011_probe() function is relying on some AMBA IDs and a device tree node to initialize the driver and a port. Both features are not necessarily required for the driver:
- we lack AMBA IDs in the ARM SBSA generic UART and
- we lack a DT node in ACPI systems.
So lets refactor the function to ease later reuse.

---
 drivers/tty/serial/amba-pl011.c |   90 ++++++++++++++++++++++++++-------------
 1 file changed, 61 insertions(+), 29 deletions(-)
drivers/tty/serial/amba-pl011.c +61 -29
index 9219127..457a8d90 100644
@@ -2159,13 +2159,10 @@ static void pl011_unregister_port(struct uart_amba_port *uap)
 		uart_unregister_driver(&amba_reg);
 }
 
-
-static int pl011_probe(struct amba_device *dev, const struct amba_id *id)
+static int pl011_allocate_port(struct device *dev, struct uart_amba_port **_uap)
 {
 	struct uart_amba_port *uap;
-	struct vendor_data *vendor = id->data;
-	void __iomem *base;
-	int i, ret;
+	int i;
 
 	for (i = 0; i < ARRAY_SIZE(amba_ports); i++)
 		if (amba_ports[i] == NULL)
@@ -2174,47 +2171,50 @@ static int pl011_probe(struct amba_device *dev, const struct amba_id *id)
 	if (i == ARRAY_SIZE(amba_ports))
 		return -EBUSY;
 
-	uap = devm_kzalloc(&dev->dev, sizeof(struct uart_amba_port),
-			   GFP_KERNEL);
+	uap = devm_kzalloc(dev, sizeof(struct uart_amba_port), GFP_KERNEL);
 	if (uap == NULL)
 		return -ENOMEM;
 
-	i = pl011_probe_dt_alias(i, &dev->dev);
+	if (_uap)
+		*_uap = uap;
+
+	return i;
+}
 
-	base = devm_ioremap(&dev->dev, dev->res.start,
-			    resource_size(&dev->res));
+static int pl011_setup_port(struct device *dev, struct uart_amba_port *uap,
+			    struct resource *mmiobase, int index)
+{
+	void __iomem *base;
+
+	base = devm_ioremap_resource(dev, mmiobase);
 	if (!base)
 		return -ENOMEM;
 
-	uap->clk = devm_clk_get(&dev->dev, NULL);
-	if (IS_ERR(uap->clk))
-		return PTR_ERR(uap->clk);
+	index = pl011_probe_dt_alias(index, dev);
 
-	uap->vendor = vendor;
-	uap->lcrh_rx = vendor->lcrh_rx;
-	uap->lcrh_tx = vendor->lcrh_tx;
 	uap->old_cr = 0;
-	uap->fifosize = vendor->get_fifosize(dev);
-	uap->port.dev = &dev->dev;
-	uap->port.mapbase = dev->res.start;
+	uap->port.dev = dev;
+	uap->port.mapbase = mmiobase->start;
 	uap->port.membase = base;
 	uap->port.iotype = UPIO_MEM;
-	uap->port.irq = dev->irq[0];
 	uap->port.fifosize = uap->fifosize;
-	uap->port.ops = &amba_pl011_pops;
 	uap->port.flags = UPF_BOOT_AUTOCONF;
-	uap->port.line = i;
-	pl011_dma_probe(&dev->dev, uap);
+	uap->port.line = index;
 
-	/* Ensure interrupts from this UART are masked and cleared */
-	writew(0, uap->port.membase + UART011_IMSC);
-	writew(0xffff, uap->port.membase + UART011_ICR);
+	pl011_dma_probe(dev, uap);
 
-	snprintf(uap->type, sizeof(uap->type), "PL011 rev%u", amba_rev(dev));
+	amba_ports[index] = uap;
 
-	amba_ports[i] = uap;
+	return 0;
+}
 
-	amba_set_drvdata(dev, uap);
+static int pl011_register_port(struct uart_amba_port *uap)
+{
+	int ret;
+
+	/* Ensure interrupts from this UART are masked and cleared */
+	writew(0, uap->port.membase + UART011_IMSC);
+	writew(0xffff, uap->port.membase + UART011_ICR);
 
 	if (!amba_reg.state) {
 		ret = uart_register_driver(&amba_reg);
@@ -2231,6 +2231,38 @@ static int pl011_probe(struct amba_device *dev, const struct amba_id *id)
 	return ret;
 }
 
+static int pl011_probe(struct amba_device *dev, const struct amba_id *id)
+{
+	struct uart_amba_port *uap;
+	struct vendor_data *vendor = id->data;
+	int portnr, ret;
+
+	portnr = pl011_allocate_port(&dev->dev, &uap);
+	if (portnr < 0)
+		return portnr;
+
+	uap->clk = devm_clk_get(&dev->dev, NULL);
+	if (IS_ERR(uap->clk))
+		return PTR_ERR(uap->clk);
+
+	uap->vendor = vendor;
+	uap->lcrh_rx = vendor->lcrh_rx;
+	uap->lcrh_tx = vendor->lcrh_tx;
+	uap->fifosize = vendor->get_fifosize(dev);
+	uap->port.irq = dev->irq[0];
+	uap->port.ops = &amba_pl011_pops;
+
+	snprintf(uap->type, sizeof(uap->type), "PL011 rev%u", amba_rev(dev));
+
+	ret = pl011_setup_port(&dev->dev, uap, &dev->res, portnr);
+	if (ret)
+		return ret;
+
+	amba_set_drvdata(dev, uap);
+
+	return pl011_register_port(uap);
+}
+
 static int pl011_remove(struct amba_device *dev)
 {
 	struct uart_amba_port *uap = amba_get_drvdata(dev);
Signature · 1 line
1.7.9.5
#6 lore raw

RFC PATCH 5/10 drivers: PL011: refactor pl011_probe()

Quoted 6 lines · Andre Przywara

Currently the pl011_probe() function is relying on some AMBA IDs and a device tree node to initialize the driver and a port. Both features are not necessarily required for the driver:
- we lack AMBA IDs in the ARM SBSA generic UART and
- we lack a DT node in ACPI systems.
So lets refactor the function to ease later reuse.

Both features are not "necessary" for any driver, but that doesn't mean we work around it by throwing a platform driver at it. We already have this same problem with PL010 UARTs, and the DT support for AMBA has the ability to specify the AMBA ID.

Moreover, there is nothing which prevents you from declaring an AMBA device with an appropriate AMBA ID for the driver to bind to - which is presumably something very similar to what you're already doing with a platform device.

Signature · 2 lines
FTTC broadband for 0.8mile line: currently at 9.5Mbps down 400kbps up
according to speedtest.net.
#7 lore raw

RFC PATCH 6/10 drivers: PL011: replace UART_MIS reading with _RIS & _IMSC

In reply to Andre Przywara

The PL011 register UART_MIS is actually a bitwise AND of the UART_RIS and the UART_MISC register.
Since the SBSA UART does not include the _MIS register, use the two separate registers to get the same behaviour. Since we are inside the spinlock and we read the _IMSC register only once, there should be no race issue.

---
 drivers/tty/serial/amba-pl011.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/tty/serial/amba-pl011.c +4 -2
index 457a8d90..e7566ed 100644
@@ -1298,11 +1298,13 @@ static irqreturn_t pl011_int(int irq, void *dev_id)
 	struct uart_amba_port *uap = dev_id;
 	unsigned long flags;
 	unsigned int status, pass_counter = AMBA_ISR_PASS_LIMIT;
+	u16 imsc;
 	int handled = 0;
 	unsigned int dummy_read;
 
 	spin_lock_irqsave(&uap->port.lock, flags);
-	status = readw(uap->port.membase + UART011_MIS);
+	imsc = readw(uap->port.membase + UART011_IMSC);
+	status = readw(uap->port.membase + UART011_RIS) & imsc;
 	if (status) {
 		do {
 			if (uap->vendor->cts_event_workaround) {
@@ -1337,7 +1339,7 @@ static irqreturn_t pl011_int(int irq, void *dev_id)
 			if (pass_counter-- == 0)
 				break;
 
-			status = readw(uap->port.membase + UART011_MIS);
+			status = readw(uap->port.membase + UART011_RIS) & imsc;
 		} while (status != 0);
 		handled = 1;
 	}
Signature · 1 line
1.7.9.5
#8 lore raw

RFC PATCH 7/10 drivers: PL011: move cts_event workaround into separate function

In reply to Andre Przywara

To avoid lines with more than 80 characters and to make the pl011_int() function more readable, move the workaround out into a separate function.

---
 drivers/tty/serial/amba-pl011.c |   33 ++++++++++++++++++++-------------
 1 file changed, 20 insertions(+), 13 deletions(-)
drivers/tty/serial/amba-pl011.c +20 -13
index e7566ed..65a0411 100644
@@ -1293,6 +1293,25 @@ static void pl011_modem_status(struct uart_amba_port *uap)
 	wake_up_interruptible(&uap->port.state->port.delta_msr_wait);
 }
 
+static void check_apply_cts_event_workaround(struct uart_amba_port *uap)
+{
+	unsigned int dummy_read;
+
+	if (!uap->vendor->cts_event_workaround)
+		return;
+
+	/* workaround to make sure that all bits are unlocked.. */
+	writew(0x00, uap->port.membase + UART011_ICR);
+
+	/*
+	 * WA: introduce 26ns(1 uart clk) delay before W1C;
+	 * single apb access will incur 2 pclk(133.12Mhz) delay,
+	 * so add 2 dummy reads
+	 */
+	dummy_read = readw(uap->port.membase + UART011_ICR);
+	dummy_read = readw(uap->port.membase + UART011_ICR);
+}
+
 static irqreturn_t pl011_int(int irq, void *dev_id)
 {
 	struct uart_amba_port *uap = dev_id;
@@ -1300,25 +1319,13 @@ static irqreturn_t pl011_int(int irq, void *dev_id)
 	unsigned int status, pass_counter = AMBA_ISR_PASS_LIMIT;
 	u16 imsc;
 	int handled = 0;
-	unsigned int dummy_read;
 
 	spin_lock_irqsave(&uap->port.lock, flags);
 	imsc = readw(uap->port.membase + UART011_IMSC);
 	status = readw(uap->port.membase + UART011_RIS) & imsc;
 	if (status) {
 		do {
-			if (uap->vendor->cts_event_workaround) {
-				/* workaround to make sure that all bits are unlocked.. */
-				writew(0x00, uap->port.membase + UART011_ICR);
-
-				/*
-				 * WA: introduce 26ns(1 uart clk) delay before W1C;
-				 * single apb access will incur 2 pclk(133.12Mhz) delay,
-				 * so add 2 dummy reads
-				 */
-				dummy_read = readw(uap->port.membase + UART011_ICR);
-				dummy_read = readw(uap->port.membase + UART011_ICR);
-			}
+			check_apply_cts_event_workaround(uap);
 
 			writew(status & ~(UART011_TXIS|UART011_RTIS|
 					  UART011_RXIS),
Signature · 1 line
1.7.9.5
#9 lore raw

RFC PATCH 8/10 drivers: PL011: allow avoiding UART enabling/disabling

In reply to Andre Przywara

The SBSA UART should not be enabled or disabled (it is always on), and consequently the spec lacks the UART_CR register. Add a vendor specific property to skip disabling or enabling of the UART. This will be used later by the SBSA UART support.

---
 drivers/tty/serial/amba-pl011.c |   18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)
drivers/tty/serial/amba-pl011.c +12 -6
index 65a0411..70af210 100644
@@ -78,6 +78,7 @@ struct vendor_data {
 	bool			oversampling;
 	bool			dma_threshold;
 	bool			cts_event_workaround;
+	bool			always_enabled;
 
 	unsigned int (*get_fifosize)(struct amba_device *dev);
 };
@@ -94,6 +95,7 @@ static struct vendor_data vendor_arm = {
 	.oversampling		= false,
 	.dma_threshold		= false,
 	.cts_event_workaround	= false,
+	.always_enabled		= false,
 	.get_fifosize		= get_fifosize_arm,
 };
 
@@ -109,6 +111,7 @@ static struct vendor_data vendor_st = {
 	.oversampling		= true,
 	.dma_threshold		= true,
 	.cts_event_workaround	= true,
+	.always_enabled		= false,
 	.get_fifosize		= get_fifosize_st,
 };
 
@@ -1944,7 +1947,7 @@ static void
 pl011_console_write(struct console *co, const char *s, unsigned int count)
 {
 	struct uart_amba_port *uap = amba_ports[co->index];
-	unsigned int status, old_cr, new_cr;
+	unsigned int status, old_cr = 0, new_cr;
 	unsigned long flags;
 	int locked = 1;
 
@@ -1961,10 +1964,12 @@ pl011_console_write(struct console *co, const char *s, unsigned int count)
 	/*
 	 *	First save the CR then disable the interrupts
 	 */
-	old_cr = readw(uap->port.membase + UART011_CR);
-	new_cr = old_cr & ~UART011_CR_CTSEN;
-	new_cr |= UART01x_CR_UARTEN | UART011_CR_TXE;
-	writew(new_cr, uap->port.membase + UART011_CR);
+	if (!uap->vendor->always_enabled) {
+		old_cr = readw(uap->port.membase + UART011_CR);
+		new_cr = old_cr & ~UART011_CR_CTSEN;
+		new_cr |= UART01x_CR_UARTEN | UART011_CR_TXE;
+		writew(new_cr, uap->port.membase + UART011_CR);
+	}
 
 	uart_console_write(&uap->port, s, count, pl011_console_putchar);
 
@@ -1975,7 +1980,8 @@ pl011_console_write(struct console *co, const char *s, unsigned int count)
 	do {
 		status = readw(uap->port.membase + UART01x_FR);
 	} while (status & UART01x_FR_BUSY);
-	writew(old_cr, uap->port.membase + UART011_CR);
+	if (!uap->vendor->always_enabled)
+		writew(old_cr, uap->port.membase + UART011_CR);
 
 	if (locked)
 		spin_unlock(&uap->port.lock);
Signature · 1 line
1.7.9.5
#10 lore raw

RFC PATCH 9/10 drivers: PL011: allow to supply fixed option string

In reply to Andre Przywara

The SBSA UART has a fixed baud rate and flow control setting, which cannot be changed or queried by software.
Add a vendor specific property to always return fixed values when trying to read the console options.

---
 drivers/tty/serial/amba-pl011.c |    6 ++++++
 1 file changed, 6 insertions(+)
drivers/tty/serial/amba-pl011.c +6 -0
index 70af210..12f1a5f 100644
@@ -79,6 +79,7 @@ struct vendor_data {
 	bool			dma_threshold;
 	bool			cts_event_workaround;
 	bool			always_enabled;
+	char			*fixed_options;
 
 	unsigned int (*get_fifosize)(struct amba_device *dev);
 };
@@ -96,6 +97,7 @@ static struct vendor_data vendor_arm = {
 	.dma_threshold		= false,
 	.cts_event_workaround	= false,
 	.always_enabled		= false,
+	.fixed_options		= NULL,
 	.get_fifosize		= get_fifosize_arm,
 };
 
@@ -112,6 +114,7 @@ static struct vendor_data vendor_st = {
 	.dma_threshold		= true,
 	.cts_event_workaround	= true,
 	.always_enabled		= false,
+	.fixed_options		= NULL,
 	.get_fifosize		= get_fifosize_st,
 };
 
@@ -2062,6 +2065,9 @@ static int __init pl011_console_setup(struct console *co, char *options)
 
 	uap->port.uartclk = clk_get_rate(uap->clk);
 
+	if (!options && uap->vendor->fixed_options)
+		options = uap->vendor->fixed_options;
+
 	if (options)
 		uart_parse_options(options, &baud, &parity, &bits, &flow);
 	else
Signature · 1 line
1.7.9.5
#11 lore raw

RFC PATCH 10/10 drivers: PL011: add support for the ARM SBSA generic UART

In reply to Andre Przywara

The ARM Server Base System Architecture[1] document describes a generic UART which is a subset of the PL011 UART.
It lacks DMA support, baud rate control and modem status line control, among other things.
The idea is to move the UART initialization and setup into the firmware (which does this job today already) and let the kernel just use the UART for sending and receiving characters.
We use the recent refactoring the build a new struct uart_ops variable which points to some new functions avoiding access to the missing registers. We reuse as much existing PL011 code as possible.

In contrast to the PL011 the SBSA UART does not define any AMBA or PrimeCell relations, so we go a pretty generic probe function which only uses platform device functions.
A DT binding is provided, but other systems can easily attach to it, too (hint, hint!).

---
 .../devicetree/bindings/serial/arm_sbsa_uart.txt   |    9 ++
 drivers/tty/serial/amba-pl011.c                    |  152 ++++++++++++++++++++
 2 files changed, 161 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/serial/arm_sbsa_uart.txt
Documentation/devicetree/bindings/serial/arm_sbsa_uart.txt new file +9 -0
new file mode 100644
index 0000000..21d211f
@@ -0,0 +1,9 @@
+* ARM SBSA defined generic UART
+This UART uses a subset of the PL011 registers and consequently lives
+in the PL011 driver. It's baudrate and other communication parameters
+cannot be adjusted at runtime, so it lacks a clock specifier here.
+
+Required properties:
+- compatible: must be "arm,sbsa-uart"
+- reg: exactly one register range
+- interrupts: exactly one interrupt specifier
drivers/tty/serial/amba-pl011.c +152 -0

189 lines not on this page · raw message

Signature · 1 line
1.7.9.5
#12 lore raw

cover drivers: PL011: add SBSA subset support

In reply to Andre Przywara

Hi,

(ping)

Quoted 12 lines · Andre Przywara

(This version has still issues, but I'd like to have feedback on the general feasibility of this approach before wasting even more time on debugging).

The ARM Server Base System Architecture[1] document describes a generic UART which is a subset of the PL011 UART.
It lacks DMA support, baud rate control and modem status line control, among other things.
The idea is to move the UART initialization and setup into the firmware (which does this job today already) and let the kernel just use the UART for sending and receiving characters.

Quoted 7 lines

This patchset is an attempt to integrate support for this UART subset into the existing PL011 driver - basically by refactoring some functions and providing a new uart_ops structure for it. It also has a separate probe function to be not dependent on AMBA/PrimeCell. Beside the obvious effect of code sharing it has the advantage of not introducing another serial device prefix, so it can go with ttyAMA, which seems to be pretty common.

Is there any opinion on this patchset? Is this approach of integrating the SBSA UART support in the PL011 driver a promising way? Or do we look for a separate driver - which could not go with ttyAMA?

Grateful for any comments,
Andre

Quoted 44 lines

There is still one issue in bringing up the UART. On the PL011 we send a dummy character via the loopback facility with the FIFOs disabled to get things going, but both FIFO control and loopback are outside of the SBSA spec. I consider this loopback approach a kludge anyway, but am struggling to find a better way of fixing this due to my lack of understanding of the TTY/serial layer.
If someone with more wisdom could enlighten me, I would be grateful.

Patch 1/10 contains a bug fix which applies to the PL011 part also, it should be considered regardless of the rest of the series. Patch 2-7 refactor some PL011 functions by splitting them up into smaller pieces, so that most the code can be reused later by the SBSA part.
Patch 8 and 9 introduce two new properties for the vendor structure, this is for SBSA functionality which cannot be controlled by separate uart_ops members only.
Patch 10 then finally drops in the SBSA specific code, by providing a new uart_ops, vendor struct and probe function for it.

Please have a look and tell me whether this SBSA driver approach has any future.

Regards,
Andre

Andre Przywara (10):
drivers: PL011: avoid potential unregister_driver call
drivers: PL011: refactor pl011_startup()
drivers: PL011: refactor pl011_shutdown()
drivers: PL011: refactor pl011_set_termios()
drivers: PL011: refactor pl011_probe()
drivers: PL011: replace UART_MIS reading with _RIS & _IMSC
drivers: PL011: move cts_event workaround into separate function
drivers: PL011: allow avoiding UART enabling/disabling
drivers: PL011: allow to supply fixed option string
drivers: PL011: add support for the ARM SBSA generic UART

 .../devicetree/bindings/serial/arm_sbsa_uart.txt   |    9 +
 drivers/tty/serial/amba-pl011.c                    |  509 ++++++++++++++------
 2 files changed, 380 insertions(+), 138 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/serial/arm_sbsa_uart.txt
#13 lore raw

cover drivers: PL011: add SBSA subset support

In reply to Andre Przywara
Quoted 20 lines · Andre Przywara

(This version has still issues, but I'd like to have feedback on the general feasibility of this approach before wasting even more time on debugging).

The ARM Server Base System Architecture[1] document describes a generic UART which is a subset of the PL011 UART.
It lacks DMA support, baud rate control and modem status line control, among other things.
The idea is to move the UART initialization and setup into the firmware (which does this job today already) and let the kernel just use the UART for sending and receiving characters.

This patchset is an attempt to integrate support for this UART subset into the existing PL011 driver - basically by refactoring some functions and providing a new uart_ops structure for it. It also has a separate probe function to be not dependent on AMBA/PrimeCell. Beside the obvious effect of code sharing it has the advantage of not introducing another serial device prefix, so it can go with ttyAMA, which seems to be pretty common.

Quoted 8 lines

There is still one issue in bringing up the UART. On the PL011 we send a dummy character via the loopback facility with the FIFOs disabled to get things going, but both FIFO control and loopback are outside of the SBSA spec. I consider this loopback approach a kludge anyway, but am struggling to find a better way of fixing this due to my lack of understanding of the TTY/serial layer.
If someone with more wisdom could enlighten me, I would be grateful.

This is the killer issue at the moment it seems, once we get to usespace the tty does not work.

Quoted 12 lines

Patch 1/10 contains a bug fix which applies to the PL011 part also, it should be considered regardless of the rest of the series. Patch 2-7 refactor some PL011 functions by splitting them up into smaller pieces, so that most the code can be reused later by the SBSA part.
Patch 8 and 9 introduce two new properties for the vendor structure, this is for SBSA functionality which cannot be controlled by separate uart_ops members only.
Patch 10 then finally drops in the SBSA specific code, by providing a new uart_ops, vendor struct and probe function for it.

Quoted 3 lines

Please have a look and tell me whether this SBSA driver approach has any future.

The patch series looks reasonable to me based on my very limited knowledge of the tty layer and the pl011 UART.

Graeme

Quoted 28 lines

Regards,
Andre

Andre Przywara (10):
drivers: PL011: avoid potential unregister_driver call
drivers: PL011: refactor pl011_startup()
drivers: PL011: refactor pl011_shutdown()
drivers: PL011: refactor pl011_set_termios()
drivers: PL011: refactor pl011_probe()
drivers: PL011: replace UART_MIS reading with _RIS & _IMSC
drivers: PL011: move cts_event workaround into separate function
drivers: PL011: allow avoiding UART enabling/disabling
drivers: PL011: allow to supply fixed option string
drivers: PL011: add support for the ARM SBSA generic UART

 .../devicetree/bindings/serial/arm_sbsa_uart.txt   |    9 +
 drivers/tty/serial/amba-pl011.c                    |  509 ++++++++++++++------
 2 files changed, 380 insertions(+), 138 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/serial/arm_sbsa_uart.txt
Signature · 7 lines
1.7.9.5


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel at lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel