- 06 Nov, 2014 40 commits
-
-
Konrad Zapalowicz authored
This commit enables support for the Digi Classic adapters line in the jsm driver. This means changes in: - device probing code - device cleanup code - driver description (Kconfig) The init/cleanup code is based on the staging/dgnc driver. Signed-off-by:
Konrad Zapalowicz <bergo.torino@gmail.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Konrad Zapalowicz authored
This commit adds the Digi Classic board implementation to the staging/jsm driver. The code here is taken from the staging/dgnc driver and modified to match the serial/jsm state. This work is mostly based on the changes that has been done to the code handling the Digi Neo cards with the inspiration coming from the diff between staging/dgnc and serial/jsm as well as the LKML history for the jsm_neo.c The code compiles now and has no sparse and checkpatch errors or warnings. Signed-off-by:
-
Frans Klaver authored
In serial8250_rx_chars(), max_count is set to 256. Due to the post-decrement operator used in the while() condition, the maximum number of iterations actually 257. This is not a problem, but it is mildly surprising if you're debugging. Use pre-decrement instead. Signed-off-by:
Frans Klaver <frans.klaver@xsens.com> Signed-off-by:
-
Jiri Slaby authored
When somebody calls TIOCSSERIAL ioctl with serial flags to set one of * ASYNC_SESSION_LOCKOUT * ASYNC_PGRP_LOCKOUT * ASYNC_CALLOUT_NOHUP * ASYNC_AUTOPROBE nothing happens. We actually ignore the flags for over a decade at least (I checked 2.6.0). So start yelling at users who use those flags, that they shouldn't. Signed-off-by:
Jiri Slaby <jslaby@suse.cz> Cc: Peter Hurley <peter@hurleysoftware.com> Cc: Alan Cox <gnomes@lxorguk.ukuu.org.uk> Signed-off-by:
-
Konrad Zapalowicz authored
This commit adds the UART structure for the Digi Classic cards. This code comes from the staging/dgnc driver. Signed-off-by:
-
Sudip Mukherjee authored
port->membase was allocated using devm_ioremap_nocache, so ideally we should unmap it using devm_iounmap. but it was using iounmap. Signed-off-by:
Sudip Mukherjee <sudip@vectorindia.org> Signed-off-by:
-
Andy Shevchenko authored
In the serial8250_tx_dma() the tx_err flag is set in case of error. Thus, there is no need to repeat this in __dma_tx_complete(). Signed-off-by:
Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by:
-
Andy Shevchenko authored
Since we have the same check inside the function we may drop it away in __dma_tx_complete(). Signed-off-by:
-
Ricardo Ribalda Delgado authored
Introduce an homogeneous lock system between setting and using the rs485 data of the uart_port. This patch should not be split into multiple ones in order to avoid leaving the tree in an unstable state. Acked-by:
Nicolas Ferre <nicolas.ferre@atmel.com> Reviewed-by:
Alan Cox <alan@linux.intel.com> Suggested-by:
Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com> Signed-off-by:
-
Ricardo Ribalda Delgado authored
Once there is no more handlers for TIOC[GS]RS485 there is no need to call the driver specific ioctl when the generic implementation is missing. Reviewed-by:
-
Ricardo Ribalda Delgado authored
In order to unify all the rs485 ioctl handling Use the implementation of TIOC[GS]RS485 ioctl handling on serial_core. Reviewed-by:
-
Ricardo Ribalda Delgado authored
In order to unify all the rs485 ioctl handling Use the implementation of TIOC[GS]RS485 ioctl handling on serial_core. Reviewed-by:
-
Ricardo Ribalda Delgado authored
In order to unify all the rs485 ioctl handling. Use the implementation of TIOC[GS]RS485 ioctl handling on serial_core. Reviewed-by:
-
Ricardo Ribalda Delgado authored
In order to unify all the rs485 ioctl handling. Use the implementation of TIOC[GS]RS485 ioctl handling on serial_core. Reviewed-by:
-
Ricardo Ribalda Delgado authored
In order to unify all the rs485 ioctl handling. Use the implementation of TIOC[GS]RS485 ioctl handling on serial_core. Reviewed-by:
-
Ricardo Ribalda Delgado authored
There is no more users for this functions. All the 8250 drivers are using the rs485 handler on serial_core instead. Reviewed-by:
-
Ricardo Ribalda Delgado authored
In order to remove the handler for rs485 ioctls on serial_8250, all the drivers must use the implementation on serial_core. Reviewed-by:
-
Ricardo Ribalda Delgado authored
Initialize recently added rs485 fields on serial_core Reviewed-by:
-
Ricardo Ribalda Delgado authored
The following drivers: 8250_core, atmel_serial, max310x, mcf, omap-serial and sci16is7xx implement code to handle RS485 ioctls. In order to avoid code duplication, we implement a simple ioctl handler on the serial_core layer. This handler can be used by all the other drivers instead of duplicating code. Until this is the only RS485 ioctl handler, it will try first the rs485_config callback and if it is not present it will call the driver specific ioctl. Reviewed-by:
-
Sudip Mukherjee authored
fixed a sparse warning in 8250_core.c : incorrect type in assignment (different address spaces) the warning was because an unsigned char pointer was being assigned to a pointer of unsigned char __iomem type . Signed-off-by:
-
Sergej Pupykin authored
WCH382 is a PCI-E card with 1 LPT and 2 DB9 COM ports detected as Serial controller: Device 1c00:3250 (rev 10) (prog-if 05 [16850]) Signed-off-by:
Sergej Pupykin <ml@sergej.pp.ru> Signed-off-by:
-
Peter Hurley authored
Switching to the N_PPS line discipline may require enabling modem status interrupts; conversely switching from N_PPS may require disabling modem status interrupts. Affected drivers: 8250 amba-pl010 atmel Cc: Nicolas Ferre <nicolas.ferre@atmel.com> Cc: Russell King <linux@arm.linux.org.uk> Signed-off-by:
Peter Hurley <peter@hurleysoftware.com> Signed-off-by:
-
Peter Hurley authored
Three UART drivers (8250, atmel & amba-pl010) directly call their enable_ms() method; the uart port lock must be acquired before any h/w programming. Signed-off-by:
-
Peter Hurley authored
UART drivers which enable modem status interrupts when switching to N_PPS line discipline need to determine if modem status interrupts should be disabled when switching from N_PPS. Specifically, the set_ldisc() notification needs to evaluate UART_ENABLE_MS() which requires termios->c_cflag. Convert in-tree UART drivers to new interface. Signed-off-by:
-
Peter Hurley authored
Three UART drivers (8250, atmel & amba-pl010) enable modem status interrupts if the line discipline is changed to N_PPS. However, the uart port flags may only be safely modified while holding the port mutex. Signed-off-by:
-
Peter Hurley authored
Allow a tty driver to safely access termios settings while handling the set_ldisc() notification. UART drivers use the set_ldisc() notification to check if the N_PPS line discipline is being enabled; if so, modem status interrupts may also need to be enabled. Conversely, modem status interrupts may need to be disabled if switching away from the N_PPS line discipline. Signed-off-by:
-
Peter Hurley authored
The line discipline buffer and the tty buffers must be flushed again after hardware shutdown; otherwise, a brief window exists between the ldisc flush in tty_port_close_start() and the subsequent tty_port_shutdown(), during which more data could be received into the tty buffers. A racing open might then be able to receive data from the previous session. Signed-off-by:
-
Peter Hurley authored
tty->closing is a bitfield member; prevent corruption from non-atomic update by assigning a unique memory location. Signed-off-by:
-
Peter Hurley authored
The port->lock does not protect the filp->f_op field; move the tty_hung_up_p() test outside the port->lock critical section in tty_port_close_start(). Signed-off-by:
-
Lars-Peter Clausen authored
Add support for setting the state of the DTR and RTS signals. Acked-by:
Soren Brinkmann <soren.brinkmann@xilinx.com> Signed-off-by:
Lars-Peter Clausen <lars@metafoo.de> Signed-off-by:
-
Julien CHAUVEAU authored
Get index of serial line from device tree using function of_alias_get_id(). If no alias is found, the 8250 core takes care of incrementing the line number. Signed-off-by:
Julien CHAUVEAU <julien.chauveau@neo-technologies.fr> Signed-off-by:
-
Lucas Stach authored
The general agreed way to specify a fixed line number for a serial console is to provide a "serial" alias in the devicetree. Start parsing this property in of_serial. Signed-off-by:
Lucas Stach <dev@lynxeye.de> Signed-off-by:
-
Aaron Sierra authored
Check the return value of ioremap_nocache to make sure we got a valid mapping. Signed-off-by:
Aaron Sierra <asierra@xes-inc.com> Signed-off-by:
-
Aaron Sierra authored
Several init/setup functions passed the PCI BAR resource start address to ioremap_nocache() via an unsigned long. This caused address truncation for a 32-bit device mapped above 4 GiB (i.e. the CPU interacts with the device via a translated address), which resulted in a kernel panic. This patch replaces all of the instances of intermediate variable use with pci_ioremap_bar() to ensure the full resource_size_t start address is used and that ioremap_nocache() is still called. The kernel panic (Exar XR17V358 PCIe device on a Freescale P2020 SBC): Machine check in kernel mode. Caused by (from MCSR=10008): Bus - Read Data Bus Error Oops: Machine check, sig: 7 [#1] SMP NR_CPUS=2 X-ES P2020 Modules linked in: CPU: 1 PID: 1 Comm: swapper/0 Not tainted 3.14.15-xes_r2-00002-g560e401 #978 task: bf850000 ti: bffee000 task.ti: bf84c000 NIP: 80318e10 LR: 80319ecc CTR: 80318dfc REGS: bffeff10 TRAP: 0204 Not tainted (3.14.15-xes_r2-00002-g560e401) MSR: 00021000 <CE,ME> CR: 20adbe42 XER: 00000000 DEAR: c1058001 ESR: 00000000 GPR00: 00000000 bf84db30 bf850000 80cb4af8 00000001 00000000 80000007 80000000 GPR08: bf837c9c c1058001 00000001 00000000 80000007 00000000 80002a10 00000000 GPR16: 00000000 00000000 00000000 00000000 00000000 00000000 80cb0000 80c72dc4 GPR24: 80cb4900 fffffffe 00029000 00000001 bf8c11e8 ffffffea 80c72ce4 80cb4af8 NIP [80318e10] mem_serial_in+0x14/0x28 LR [80319ecc] serial8250_config_port+0x160/0xe38 Call Trace: [bf84db30] [80319d94] serial8250_config_port+0x28/0xe38 (unreliable) [bf84db60] [80315e3c] uart_add_one_port+0x148/0x3a4 [bf84dbf0] [8031bf40] serial8250_register_8250_port+0x2dc/0x3c8 [bf84dc20] [8032111c] pciserial_init_ports+0xd4/0x1c0 [bf84dd50] [803212f8] pciserial_init_one+0xf0/0x224 [bf84dd90] [802d8ff4] local_pci_probe+0x34/0x8c [bf84dda0] [802d92c8] pci_device_probe+0x84/0xa0 [bf84ddc0] [80329ee0] driver_probe_device+0xac/0x26c [bf84dde0] [8032a15c] __driver_attach+0xbc/0xc0 [bf84de00] [80328388] bus_for_each_dev+0x90/0xcc [bf84de30] [80329cd0] driver_attach+0x24/0x34 [bf84de40] [80328e28] bus_add_driver+0x104/0x1fc [bf84de60] [8032a8c8] driver_register+0x70/0x138 [bf84de70] [802d93c0] __pci_register_driver+0x48/0x58 [bf84de80] [8077e0e4] serial_pci_driver_init+0x24/0x34 [bf84de90] [80002228] do_one_initcall+0x34/0x1b0 [bf84df00] [80764294] kernel_init_freeable+0x138/0x1e8 [bf84df30] [80002a24] kernel_init+0x14/0x108 [bf84df40] [8000ef94] ret_from_kernel_thread+0x5c/0x64 Instruction dump: 800800c4 7d290214 39290001 7c0004ac 7ca049ae 7c0004ac 4e800020 88030035 81230008 7c840030 7d292214 7c0004ac <88690000> 0c030000 4c00012c 5463063e ---[ end trace e3c16443b5d573c6 ]--- Signed-off-by:
-
Jingchang Lu authored
This adds 32-bit register lpuart32 power management support, this also updates the 8-bit register lpuart resume function. Signed-off-by:
Jingchang Lu <jingchang.lu@freescale.com> Signed-off-by:
-
Maxime Ripard authored
We are removing the dmaengine_device_control API, that shouldn't even have been exposed in the first place. Change the callers to use the proper API. Signed-off-by:
Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by:
-
Eddie Huang authored
In mtk8250_set_termios function, calculating quot value can not be zero, otherwise, using DIV_ROUND_CLOSEST(port->uartclk, quot * baud) will fail due to divisor is zero. Signed-off-by:
Eddie Huang <eddie.huang@mediatek.com> Signed-off-by:
-
Kevin Cernekee authored
I wrote this driver and use it daily on several machines for work, so why not. Signed-off-by:
Kevin Cernekee <cernekee@gmail.com> Acked-by:
Florian Fainelli <f.fainelli@gmail.com> Signed-off-by:
-
Kevin Cernekee authored
We don't really need to perform the ioremap "on demand" so it's simpler just to do it from the probe function. This also lets us eliminate the UART_REG_SIZE constant and rely on the resource information passed in from the DT or platform code. Signed-off-by:
-
Kevin Cernekee authored
This enables early console output if there is a chosen/stdout-path property referencing a UART node with the "brcm,bcm6345-uart" compatible string. The bootloader sets up the pinmux and baud/parity/etc. Tested on bcm3384 (MIPS, DT). Signed-off-by:
-
