- 09 Mar, 2018 12 commits
-
-
Uwe Kleine-König authored
serial_core might call the .start_tx callback without any data being available to send. In this case return early instead of going through all the setup needed for sending which might include disabling RX in RS485 half-duplex mode. Signed-off-by:
Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Uwe Kleine-König authored
When using RS485 half duplex the Transmitter Complete irq is needed to determine the moment when the transmitter can be disabled. When using DMA this irq must only be enabled when DMA has completed to transfer all data. Otherwise the CPU might busily trigger this irq which is not properly handled and so the also pending irq for the DMA transfer cannot trigger. Signed-off-by:
-
Uwe Kleine-König authored
This allows to increase the RX waterlevel which allows to delay the RRDY irq. The desired effect is that less irqs are needed to handle characters and so reduce irq count of the system. Signed-off-by:
-
Uwe Kleine-König authored
Make sure that UCR1.RXDMAEN and UCR1.ATDMAEN (for the DMA case) and UCR1.RRDYEN (for the PIO case) are off iff UCR1.RXEN is disabled. This ensures that the fifo isn't read with RX disabled which results in an exception. Signed-off-by:
-
Uwe Kleine-König authored
When the UART is used in DMA mode, .stop_rx() does nothing if the port isn't suspended. This is wrong as .stop_rx() should stop receiving characters unconditionally. When the port is about to be closed the DMA channel is stopped in .shutdown(), so this isn't necessary to be in .stop_rx() here, too. Signed-off-by:
-
Uwe Kleine-König authored
The aging timer fires if there are characters in the RX fifo but the water level isn't reached yet. Make sure that the waterlevel is configured before the aging timer is enabled to trigger a DMA request (UCR1_ATDMAEN). Signed-off-by:
-
Uwe Kleine-König authored
The serial/imx driver is full of inconsistently named and typed variables that hold different register values. Consistently use u32 as type (matching what readl and writel use) and name the variables after the register whose value they are holding. This makes it easier to notice when UCR2_RTSEN is written to UCR1. The only difference introduced by this commit in the compiled driver is that twice the second argument to warn_slowpath_null() changed because the two WARN_ON in dma_rx_callback() pass __LINE__ to warn_slowpath_null(). Signed-off-by:
-
Uwe Kleine-König authored
The original code looks as follows: if (sport->dma_is_enabled) { ... make sure TX DMA is running, i.e. .dma_is_txing = 1 } if (sport->dma_is_txing) return; As .dma_is_txing can only be true if .dma_is_enabled is, the return can go at the end of the first if body without an additional check. Signed-off-by: -
Uwe Kleine-König authored
Neither .dma_is_txing nor .dma_is_rxing can evaluate to true if .dma_is_enabled evaluates to false: The only function that sets .dma_is_txing to a non-zero value is imx_dma_tx() which is only called if .dma_is_enabled is true. Same for .dma_is_rxing and start_rx_dma(). And before .dma_is_enabled is set to 0 when imx_shutdown calls imx_disable_dma(), .dma_is_rxing and .dma_is_txing are reset to zero before, too. For this reason sport->dma_is_enabled && sport->dma_is_rxing has the same value as sport->dma_is_rxing which allows to simplify three if conditions. Signed-off-by:
-
Uwe Kleine-König authored
This reduces the amount of read accesses to the register space by shadowing the values for five registers that only change on writing them. There is a single bit in UCR2 that might change without being written to it, this is handled accordingly. Signed-off-by:
-
Uwe Kleine-König authored
This prepares implementing shadow copies for the control registers and additionally provides a good place to hook in debug code to trace register usage. Most of this patch was done using pattern substitution: perl -p -i -e ' s/\breadl(?:_relaxed)?\((?:sport->port\.|port->)membase \+/imx_uart_readl(sport,/; s/\bwritel(?:_relaxed)?\(([^,]*), (sport->port\.|port->)membase \+/imx_uart_writel(sport, $1,/; ' drivers/tty/serial/imx.c Signed-off-by:
-
Palmer Dabbelt authored
The RISC-V ISA defines a simple console that is availiable via SBI calls on all systems. The SBI console is designed to be availiable at all times, so while it's most natural to use this as an early printk target it's also possible to use this as the system console when there isn't a better one availiable. This patch adds support for the RISC-V SBI console via the HVC infastructure. It's entirely independent from our early printk support, which results in early boot messages appearing twice over the SBI console. As far as I can tell that's the fault of our early printk support (we should support earlycon) as opposed to this driver. There is one checkpatch.pl warning here: to check the MAINTAINERS file. They're all matched by the "K: riscv" line. Signed-off-by:
Palmer Dabbelt <palmer@sifive.com> Signed-off-by:
-
- 28 Feb, 2018 28 commits
-
-
Uwe Kleine-König authored
Consistently indicate being called with irqs off and the port lock taken for all functions that this applies to. Signed-off-by:
-
Uwe Kleine-König authored
Only the reference manual for the i.MX1 (I have MC9328MX1RM/D Rev 5 from 2004) uses TDMAEN and RDMAEN for these. All reference manuals for the newer chips use TXDMAEN and RXDMAEN. Update to the newer name with the assumption that most imx users don't use an imx1 any more. Signed-off-by:
-
Andy Shevchenko authored
PCI ID database is for IDs used across several drivers. Here is the case for SUNIX combo cards. No functional change intended. Signed-off-by:
Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by:
-
Geert Uytterhoeven authored
The cdns_uart_port[] array is indexed using a value derived from the "serialN" alias in DT, which may lead to an out-of-bounds access. Fix this by adding a range check. Fixes: 928e9263 ("tty: xuartps: Initialize ports according to aliases") Signed-off-by:
Geert Uytterhoeven <geert+renesas@glider.be> Reviewed-by:
Michal Simek <michal.simek@xilinx.com> Signed-off-by:
-
Geert Uytterhoeven authored
The sirf_ports[] array is indexed using a value derived from the "serialN" alias in DT, which may lead to an out-of-bounds access. Fix this by adding a range check. Fixes: a6ffe896 ("serial: sirf: use dynamic method allocate uart structure") Signed-off-by:
-
Geert Uytterhoeven authored
The sci_ports[] array is indexed using a value derived from the "serialN" alias in DT, which may lead to an out-of-bounds access. Fix this by adding a range check. Note that the array size is defined by a Kconfig symbol (CONFIG_SERIAL_SH_SCI_NR_UARTS), so this can even be triggered using a legitimate DTB. Fixes: 97ed9790 ("serial: sh-sci: Remove unused platform data capabilities field") Signed-off-by:
-
Geert Uytterhoeven authored
The s3c24xx_serial_ports[] array is indexed using a value derived from the "serialN" alias in DT, or from an incrementing probe index, which may lead to an out-of-bounds access. Fix this by adding a range check. Note that the array size is defined by a Kconfig symbol (CONFIG_SERIAL_SAMSUNG_UARTS), so this can even be triggered using a legitimate DTB or legitimate board code. Fixes: 13a9f6c6 ("serial: samsung: Consider DT alias when probing ports") Signed-off-by:
-
Geert Uytterhoeven authored
The serial_pxa_ports[] array is indexed using a value derived from the "serialN" alias in DT, or from platform data, which may lead to an out-of-bounds access. Fix this by adding a range check. Fixes: 699c20f3 ("serial: pxa: add OF support") Signed-off-by:
-
Geert Uytterhoeven authored
The auart_port[] array is indexed using a value derived from the "serialN" alias in DT, or from platform data, which may lead to an out-of-bounds access. Fix this by adding a range check. Fixes: 1ea6607d ("serial: mxs-auart: Allow device tree probing") Signed-off-by:
-
Geert Uytterhoeven authored
The imx_ports[] array is indexed using a value derived from the "serialN" alias in DT, or from platform data, which may lead to an out-of-bounds access. Fix this by adding a range check. Fixes: ff05967a ("serial/imx: add of_alias_get_id() reference back") Signed-off-by:
-
Geert Uytterhoeven authored
The lpuart_ports[] array is indexed using a value derived from the "serialN" alias in DT, which may lead to an out-of-bounds access. Fix this by adding a range check. Fixes: c9e2e946 ("tty: serial: add Freescale lpuart driver support") Signed-off-by:
-
Geert Uytterhoeven authored
The arc_uart_ports[] array is indexed using a value derived from the "serialN" alias in DT, which may lead to an out-of-bounds access. Fix this by adding a range check. Note that the array size is defined by a Kconfig symbol (CONFIG_SERIAL_ARC_NR_PORTS), so this can even be triggered using a legitimate DTB. Fixes: ea28fd56 ("serial/arc-uart: switch to devicetree based probing") Signed-off-by:
-
Uwe Kleine-König authored
imx_shutdown() calls imx_disable_dma if .dma_is_enabled. So after imx_shudown() completes, .dma_is_enabled is zero. For this reason .dma_is_enabled is also zero when imx_startup() is called. So the check for this variable being zero can be dropped. Signed-off-by:
Shawn Guo <shawnguo@kernel.org> Signed-off-by:
-
Uwe Kleine-König authored
The .dma_is_inited member is only set to a value != 0 when the port's startup function calls imx_uart_dma_init(). On shutdown of the port imx_uart_dma_exit is called which sets the value back to 0. So .dma_is_inited is always 0 when imx_startup() is called (assuming .startup() and .shutdown() are correctly balanced) and the check for !sport->dma_is_inited can go away. This allows to replace .dma_is_inited by a variable local to imx_startup. Signed-off-by:
-
Uwe Kleine-König authored
Handling an irq that isn't enabled can have some undesired side effects. Some of these are mentioned in the newly introduced code comment. Some of the irq sources already had their handling right, some don't. Handle them all in the same consistent way. The change for USR1_RRDY and USR1_AGTIM drops the check for dma_is_enabled. This is correct as UCR1_RRDYEN and UCR2_ATEN are always off if dma is enabled. Signed-off-by:
-
Uwe Kleine-König authored
Now the variable holding the value of register USR1 is called usr1 instead of sts which is more straight forward. The same is also done for sts2 which is called usr2 now. Signed-off-by:
-
Uwe Kleine-König authored
The check sts & USR1_DTRD was just evaluated to true two lines above. So this change doesn't have any effect on the semantic of the driver. Fixes: 27e16501 ("serial: imx: implement DSR irq handling for DTE mode") Signed-off-by:
-
Colin Ian King authored
Bit pattern STAT_FRM_ERR is being bit-wise or'd twice; remove the redundant 2nd STAT_FRM_ERR Signed-off-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
Troy Kisky authored
This allows me to login after sending a break when service serial-getty@ttymxc0.service is running The "tty_insert_flip_char(port, 0, TTY_BREAK)" in clear_rx_errors fixes this by allowing the higher layers to see a break. Also, call uart_handle_break to handle possible "secure attention key." FYI: Martin said the ROM sdma firmware works with this patch, but external sdma firmware still does not send breaks on a i.mx6UL Signed-off-by:
Troy Kisky <troy.kisky@boundarydevices.com> Tested-by:
Martin Hicks <mort@bork.org> Reviewed-by:
Fabio Estevam <fabio.estevam@nxp.com> Acked-by:
-
Joshua Scott authored
An issue has been observed on the Marvell Armada 38x serial port. Writes to UART_LCR can result in characters that are currently held in the TX FIFO being lost rather than sent, even if the userspace process has attempted to flush them. This is most visible when using the "resize" command (tested on Busybox), where we have observed the escape code for restoring cursor position becoming mangled. Signed-off-by:
Joshua Scott <joshua.scott@alliedtelesis.co.nz> Signed-off-by:
-
Ulrich Hecht authored
High latencies of classic timers cause performance issues for high- speed serial transmissions. This patch transforms rx_timer into an hrtimer to reduce the minimum latency. Signed-off-by:
Ulrich Hecht <ulrich.hecht+renesas@gmail.com> Signed-off-by:
-
Vignesh R authored
omap_8250_throttle() is called when tty RX buffer is about to overflow and can no longer keep up with the rate at which UART is receiving data. So, the expectation of this callback, is that UART stops RX and asserts HW flow control to signal the sender to stop sending more data. omap_8250_throttle() disables RX FIFO interrupts thus FIFO is no longer serviced, leading to assertion of flow control once RX FIFO is full. But, this does not work when DMA is enabled as driver keeps queuing new RX DMA request in completion handler without brothering about throttling request made by the higher layer. This patch introduces a flag that can be used to determine whether or not to queue next RX DMA request based on throttling request. Without this patch, tty buffer overflows are reported at higher baudrates. Signed-off-by:
Vignesh R <vigneshr@ti.com> Signed-off-by:
-
Vignesh R authored
Currently, data in RX FIFO is read based on UART_LSR register state even if RDI and RLSI interrupts are disabled in UART_IER register. This is because when IRQ handler is called due to TX FIFO empty event, RX FIFO is serviced based on UART_LSR register status instead of UART_IIR status. This defeats the purpose of disabling UART RX FIFO interrupts during throttling(see, omap_8250_throttle()) as IRQ handler continues to drain UART RX FIFO resulting in overflow of buffer at tty layer. Fix this by making sure that driver drains UART RX FIFO only when UART_IIR_RDI is set along with UART_LSR_BI or UART_LSR_DR bits. Signed-off-by:
-
Sergei Shtylyov authored
R-Car V3H (R8A77980) SoC has the R-Car gen3 compatible SCIF and HSCIF ports, so document the SoC specific bindings. Signed-off-by:
Sergei Shtylyov <sergei.shtylyov@cogentembedded.com> Reviewed-by:
Rob Herring <robh@kernel.org> Reviewed-by:
Simon Horman <horms+renesas@verge.net.au> Reviewed-by:
-
Jan Kundrát authored
Some debugging tools (/sys/kernel/debug/gpio, `lsgpio`) use the gpio_chip's label for displaying an additional context. Right now, the information duplicates stuff which is already available from the parent's device. This is how e.g. `lsgpio`'s output looks like: GPIO chip: gpiochip2, "spi1.2", 16 GPIO lines Comparing the output of other GPIO expanders that I have available: gpiochip4: GPIOs 464-479, parent: spi/spi1.1, mcp23s17, can sleep: gpiochip5: GPIOs 448-463, parent: i2c/0-0020, pca9555, can sleep: gpiochip2: GPIOs 496-511, parent: spi/spi1.2, spi1.2, can sleep: This patch ensures that the type of the real HW device is shown instead of duplicating the SPI path: gpiochip2: GPIOs 496-511, parent: spi/spi1.2, MAX14830, can sleep: Signed-off-by:
Jan Kundrát <jan.kundrat@cesnet.cz> Signed-off-by:
-
Uwe Kleine-König authored
The UART can be operated without an irq. In this case a timer is setup that regularily calls altera_uart_interrupt(). The receiving part depends on pp->imr having the bit ALTERA_UART_STATUS_RRDY_MSK set, otherwise altera_uart_rx_chars() is never called. So ensure that the bit gets set (disguised as ALTERA_UART_CONTROL_RRDY_MSK) by not returning early from altera_uart_startup() if port->irq is 0. This doesn't affect the hardware as the ALTERA_UART_CONTROL_RRDY_MSK bit isn't actually written to the control register. Signed-off-by:
Tobias Klauser <tklauser@distanz.ch> Signed-off-by:
-
Uwe Kleine-König authored
If the irq line of an altera UART device isn't used to report interrupts for this device the driver better ensures that this device doesn't pull this line to active state and so disturb the whatever might be connected to this line. Signed-off-by:
-
Uwe Kleine-König authored
Most register accesses in the altera driver honor port->regshift by using altera_uart_writel(). There are a few accesses however that were missed when the driver was converted to use port->regshift and some others were added later in commit 4d9d7d89 ("serial: altera_uart: add earlycon support"). Fixes: 2780ad42 ("tty: serial: altera_uart: Use port->regshift to store bus shift") Signed-off-by:
-
