kevin/linux
1
Fork 0
Code
a0ab36689a
linux/drivers/serial/sunsab.c

1150 lines
29 KiB
C
Raw Normal View History

Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
/* sunsab.c: ASYNC Driver for the SIEMENS SAB82532 DUSCC.
*
* Copyright (C) 1997 Eddie C. Dost (ecd@skynet.be)
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
* Copyright (C) 2002, 2006 David S. Miller (davem@davemloft.net)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
*
* Rewrote buffer handling to use CIRC(Circular Buffer) macros.
* Maxim Krasnyanskiy <maxk@qualcomm.com>
*
* Fixed to use tty_get_baud_rate, and to allow for arbitrary baud
* rates to be programmed into the UART. Also eliminated a lot of
* duplicated code in the console setup.
* Theodore Ts'o <tytso@mit.edu>, 2001-Oct-12
*
* Ported to new 2.5.x UART layer.
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
* David S. Miller <davem@davemloft.net>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
*/
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/tty.h>
#include <linux/tty_flip.h>
#include <linux/major.h>
#include <linux/string.h>
#include <linux/ptrace.h>
#include <linux/ioport.h>
#include <linux/circ_buf.h>
#include <linux/serial.h>
#include <linux/sysrq.h>
#include <linux/console.h>
#include <linux/spinlock.h>
#include <linux/slab.h>
#include <linux/delay.h>
#include <linux/init.h>
sparc/drivers: use linux/of_device.h instead of asm/of_device.h Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-11 14:30:53 -07:00
#include <linux/of_device.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
#include <asm/io.h>
#include <asm/irq.h>
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
#include <asm/prom.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SUNSAB]: Fix broken SYSRQ. Include SYSRQ support for SUNSAB if SUNSAB_CONSOLE is selected Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-25 22:49:37 -07:00
#if defined(CONFIG_SERIAL_SUNSAB_CONSOLE) && defined(CONFIG_MAGIC_SYSRQ)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
#define SUPPORT_SYSRQ
#endif
#include <linux/serial_core.h>
#include "suncore.h"
#include "sunsab.h"
struct uart_sunsab_port {
struct uart_port port; /* Generic UART port */
union sab82532_async_regs __iomem *regs; /* Chip registers */
unsigned long irqflags; /* IRQ state flags */
int dsr; /* Current DSR state */
unsigned int cec_timeout; /* Chip poll timeout... */
unsigned int tec_timeout; /* likewise */
unsigned char interrupt_mask0;/* ISR0 masking */
unsigned char interrupt_mask1;/* ISR1 masking */
unsigned char pvr_dtr_bit; /* Which PVR bit is DTR */
unsigned char pvr_dsr_bit; /* Which PVR bit is DSR */
[SUNSAB]: Fix several bugs. * don't register irq until ->startup() (and release in ->shutdown()). That avoids oopsen with the current tree when interrupt comes before we'd set up the data structures for ttyb. * handle console=ttyS... even when OBP talks to screen/keyboard * register irq handler for each port, let kernel/irq/handle.c call it for both if needed. Kills code duplication in sunsab_interrupt(). BTW, there'd been bitrot in it - ttya handling had stopped calling check_status() on BRK (correctly), ttyb copy of that code had kept the bogus call in that case. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-17 16:47:07 -07:00
unsigned int gis_shift;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
int type; /* SAB82532 version */
[SUNSAB]: Defer register updates until transmitter is idle. The chip can emit garbage characters if we touch the settings while characters are going out. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-11 11:34:32 -07:00
/* Setting configuration bits while the transmitter is active
* can cause garbage characters to get emitted by the chip.
* Therefore, we cache such writes here and do the real register
* write the next time the transmitter becomes idle.
*/
unsigned int cached_ebrg;
unsigned char cached_mode;
unsigned char cached_pvr;
unsigned char cached_dafo;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
};
/*
* This assumes you have a 29.4912 MHz clock for your UART.
*/
#define SAB_BASE_BAUD ( 29491200 / 16 )
static char *sab82532_version[16] = {
"V1.0", "V2.0", "V3.2", "V(0x03)",
"V(0x04)", "V(0x05)", "V(0x06)", "V(0x07)",
"V(0x08)", "V(0x09)", "V(0x0a)", "V(0x0b)",
"V(0x0c)", "V(0x0d)", "V(0x0e)", "V(0x0f)"
};
#define SAB82532_MAX_TEC_TIMEOUT 200000 /* 1 character time (at 50 baud) */
#define SAB82532_MAX_CEC_TIMEOUT 50000 /* 2.5 TX CLKs (at 50 baud) */
#define SAB82532_RECV_FIFO_SIZE 32 /* Standard async fifo sizes */
#define SAB82532_XMIT_FIFO_SIZE 32
static __inline__ void sunsab_tec_wait(struct uart_sunsab_port *up)
{
int timeout = up->tec_timeout;
while ((readb(&up->regs->r.star) & SAB82532_STAR_TEC) && --timeout)
udelay(1);
}
static __inline__ void sunsab_cec_wait(struct uart_sunsab_port *up)
{
int timeout = up->cec_timeout;
while ((readb(&up->regs->r.star) & SAB82532_STAR_CEC) && --timeout)
udelay(1);
}
static struct tty_struct *
receive_chars(struct uart_sunsab_port *up,
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 06:55:46 -07:00
union sab82532_irq_status *stat)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
struct tty_struct *tty = NULL;
unsigned char buf[32];
int saw_console_brk = 0;
int free_fifo = 0;
int count = 0;
int i;
serial: kill off uart_info We moved this into uart_state, now move the fields out of the separate structure and kill it off. Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2009-09-19 13:13:28 -07:00
if (up->port.state != NULL) /* Unopened serial console */
tty = up->port.state->port.tty;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
/* Read number of BYTES (Character + Status) available. */
if (stat->sreg.isr0 & SAB82532_ISR0_RPF) {
count = SAB82532_RECV_FIFO_SIZE;
free_fifo++;
}
if (stat->sreg.isr0 & SAB82532_ISR0_TCD) {
count = readb(&up->regs->r.rbcl) & (SAB82532_RECV_FIFO_SIZE - 1);
free_fifo++;
}
/* Issue a FIFO read command in case we where idle. */
if (stat->sreg.isr0 & SAB82532_ISR0_TIME) {
sunsab_cec_wait(up);
writeb(SAB82532_CMDR_RFRD, &up->regs->w.cmdr);
return tty;
}
if (stat->sreg.isr0 & SAB82532_ISR0_RFO)
free_fifo++;
/* Read the FIFO. */
for (i = 0; i < count; i++)
buf[i] = readb(&up->regs->r.rfifo[i]);
/* Issue Receive Message Complete command. */
if (free_fifo) {
sunsab_cec_wait(up);
writeb(SAB82532_CMDR_RMC, &up->regs->w.cmdr);
}
/* Count may be zero for BRK, so we check for it here */
if ((stat->sreg.isr1 & SAB82532_ISR1_BRK) &&
(up->port.line == up->port.cons->index))
saw_console_brk = 1;
for (i = 0; i < count; i++) {
[PATCH] TTY layer buffering revamp The API and code have been through various bits of initial review by serial driver people but they definitely need to live somewhere for a while so the unconverted drivers can get knocked into shape, existing drivers that have been updated can be better tuned and bugs whacked out. This replaces the tty flip buffers with kmalloc objects in rings. In the normal situation for an IRQ driven serial port at typical speeds the behaviour is pretty much the same, two buffers end up allocated and the kernel cycles between them as before. When there are delays or at high speed we now behave far better as the buffer pool can grow a bit rather than lose characters. This also means that we can operate at higher speeds reliably. For drivers that receive characters in blocks (DMA based, USB and especially virtualisation) the layer allows a lot of driver specific code that works around the tty layer with private secondary queues to be removed. The IBM folks need this sort of layer, the smart serial port people do, the virtualisers do (because a virtualised tty typically operates at infinite speed rather than emulating 9600 baud). Finally many drivers had invalid and unsafe attempts to avoid buffer overflows by directly invoking tty methods extracted out of the innards of work queue structs. These are no longer needed and all go away. That fixes various random hangs with serial ports on overflow. The other change in here is to optimise the receive_room path that is used by some callers. It turns out that only one ldisc uses receive room except asa constant and it updates it far far less than the value is read. We thus make it a variable not a function call. I expect the code to contain bugs due to the size alone but I'll be watching and squashing them and feeding out new patches as it goes. Because the buffers now dynamically expand you should only run out of buffering when the kernel runs out of memory for real. That means a lot of the horrible hacks high performance drivers used to do just aren't needed any more. Description: tty_insert_flip_char is an old API and continues to work as before, as does tty_flip_buffer_push() [this is why many drivers dont need modification]. It does now also return the number of chars inserted There are also tty_buffer_request_room(tty, len) which asks for a buffer block of the length requested and returns the space found. This improves efficiency with hardware that knows how much to transfer. and tty_insert_flip_string_flags(tty, str, flags, len) to insert a string of characters and flags For a smart interface the usual code is len = tty_request_buffer_room(tty, amount_hardware_says); tty_insert_flip_string(tty, buffer_from_card, len); More description! At the moment tty buffers are attached directly to the tty. This is causing a lot of the problems related to tty layer locking, also problems at high speed and also with bursty data (such as occurs in virtualised environments) I'm working on ripping out the flip buffers and replacing them with a pool of dynamically allocated buffers. This allows both for old style "byte I/O" devices and also helps virtualisation and smart devices where large blocks of data suddenely materialise and need storing. So far so good. Lots of drivers reference tty->flip.*. Several of them also call directly and unsafely into function pointers it provides. This will all break. Most drivers can use tty_insert_flip_char which can be kept as an API but others need more. At the moment I've added the following interfaces, if people think more will be needed now is a good time to say int tty_buffer_request_room(tty, size) Try and ensure at least size bytes are available, returns actual room (may be zero). At the moment it just uses the flipbuf space but that will change. Repeated calls without characters being added are not cumulative. (ie if you call it with 1, 1, 1, and then 4 you'll have four characters of space. The other functions will also try and grow buffers in future but this will be a more efficient way when you know block sizes. int tty_insert_flip_char(tty, ch, flag) As before insert a character if there is room. Now returns 1 for success, 0 for failure. int tty_insert_flip_string(tty, str, len) Insert a block of non error characters. Returns the number inserted. int tty_prepare_flip_string(tty, strptr, len) Adjust the buffer to allow len characters to be added. Returns a buffer pointer in strptr and the length available. This allows for hardware that needs to use functions like insl or mencpy_fromio. Signed-off-by: Alan Cox <alan@redhat.com> Cc: Paul Fulghum <paulkf@microgate.com> Signed-off-by: Hirokazu Takata <takata@linux-m32r.org> Signed-off-by: Serge Hallyn <serue@us.ibm.com> Signed-off-by: Jeff Dike <jdike@addtoit.com> Signed-off-by: John Hawkes <hawkes@sgi.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Adrian Bunk <bunk@stusta.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-09 21:54:13 -07:00
unsigned char ch = buf[i], flag;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
if (tty == NULL) {
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 06:55:46 -07:00
uart_handle_sysrq_char(&up->port, ch);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
continue;
}
[PATCH] TTY layer buffering revamp The API and code have been through various bits of initial review by serial driver people but they definitely need to live somewhere for a while so the unconverted drivers can get knocked into shape, existing drivers that have been updated can be better tuned and bugs whacked out. This replaces the tty flip buffers with kmalloc objects in rings. In the normal situation for an IRQ driven serial port at typical speeds the behaviour is pretty much the same, two buffers end up allocated and the kernel cycles between them as before. When there are delays or at high speed we now behave far better as the buffer pool can grow a bit rather than lose characters. This also means that we can operate at higher speeds reliably. For drivers that receive characters in blocks (DMA based, USB and especially virtualisation) the layer allows a lot of driver specific code that works around the tty layer with private secondary queues to be removed. The IBM folks need this sort of layer, the smart serial port people do, the virtualisers do (because a virtualised tty typically operates at infinite speed rather than emulating 9600 baud). Finally many drivers had invalid and unsafe attempts to avoid buffer overflows by directly invoking tty methods extracted out of the innards of work queue structs. These are no longer needed and all go away. That fixes various random hangs with serial ports on overflow. The other change in here is to optimise the receive_room path that is used by some callers. It turns out that only one ldisc uses receive room except asa constant and it updates it far far less than the value is read. We thus make it a variable not a function call. I expect the code to contain bugs due to the size alone but I'll be watching and squashing them and feeding out new patches as it goes. Because the buffers now dynamically expand you should only run out of buffering when the kernel runs out of memory for real. That means a lot of the horrible hacks high performance drivers used to do just aren't needed any more. Description: tty_insert_flip_char is an old API and continues to work as before, as does tty_flip_buffer_push() [this is why many drivers dont need modification]. It does now also return the number of chars inserted There are also tty_buffer_request_room(tty, len) which asks for a buffer block of the length requested and returns the space found. This improves efficiency with hardware that knows how much to transfer. and tty_insert_flip_string_flags(tty, str, flags, len) to insert a string of characters and flags For a smart interface the usual code is len = tty_request_buffer_room(tty, amount_hardware_says); tty_insert_flip_string(tty, buffer_from_card, len); More description! At the moment tty buffers are attached directly to the tty. This is causing a lot of the problems related to tty layer locking, also problems at high speed and also with bursty data (such as occurs in virtualised environments) I'm working on ripping out the flip buffers and replacing them with a pool of dynamically allocated buffers. This allows both for old style "byte I/O" devices and also helps virtualisation and smart devices where large blocks of data suddenely materialise and need storing. So far so good. Lots of drivers reference tty->flip.*. Several of them also call directly and unsafely into function pointers it provides. This will all break. Most drivers can use tty_insert_flip_char which can be kept as an API but others need more. At the moment I've added the following interfaces, if people think more will be needed now is a good time to say int tty_buffer_request_room(tty, size) Try and ensure at least size bytes are available, returns actual room (may be zero). At the moment it just uses the flipbuf space but that will change. Repeated calls without characters being added are not cumulative. (ie if you call it with 1, 1, 1, and then 4 you'll have four characters of space. The other functions will also try and grow buffers in future but this will be a more efficient way when you know block sizes. int tty_insert_flip_char(tty, ch, flag) As before insert a character if there is room. Now returns 1 for success, 0 for failure. int tty_insert_flip_string(tty, str, len) Insert a block of non error characters. Returns the number inserted. int tty_prepare_flip_string(tty, strptr, len) Adjust the buffer to allow len characters to be added. Returns a buffer pointer in strptr and the length available. This allows for hardware that needs to use functions like insl or mencpy_fromio. Signed-off-by: Alan Cox <alan@redhat.com> Cc: Paul Fulghum <paulkf@microgate.com> Signed-off-by: Hirokazu Takata <takata@linux-m32r.org> Signed-off-by: Serge Hallyn <serue@us.ibm.com> Signed-off-by: Jeff Dike <jdike@addtoit.com> Signed-off-by: John Hawkes <hawkes@sgi.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Adrian Bunk <bunk@stusta.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-09 21:54:13 -07:00
flag = TTY_NORMAL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
up->port.icount.rx++;
if (unlikely(stat->sreg.isr0 & (SAB82532_ISR0_PERR |
SAB82532_ISR0_FERR |
SAB82532_ISR0_RFO)) ||
unlikely(stat->sreg.isr1 & SAB82532_ISR1_BRK)) {
/*
* For statistics only
*/
if (stat->sreg.isr1 & SAB82532_ISR1_BRK) {
stat->sreg.isr0 &= ~(SAB82532_ISR0_PERR |
SAB82532_ISR0_FERR);
up->port.icount.brk++;
/*
* We do the SysRQ and SAK checking
* here because otherwise the break
* may get masked by ignore_status_mask
* or read_status_mask.
*/
if (uart_handle_break(&up->port))
continue;
} else if (stat->sreg.isr0 & SAB82532_ISR0_PERR)
up->port.icount.parity++;
else if (stat->sreg.isr0 & SAB82532_ISR0_FERR)
up->port.icount.frame++;
if (stat->sreg.isr0 & SAB82532_ISR0_RFO)
up->port.icount.overrun++;
/*
* Mask off conditions which should be ingored.
*/
stat->sreg.isr0 &= (up->port.read_status_mask & 0xff);
stat->sreg.isr1 &= ((up->port.read_status_mask >> 8) & 0xff);
if (stat->sreg.isr1 & SAB82532_ISR1_BRK) {
[PATCH] TTY layer buffering revamp The API and code have been through various bits of initial review by serial driver people but they definitely need to live somewhere for a while so the unconverted drivers can get knocked into shape, existing drivers that have been updated can be better tuned and bugs whacked out. This replaces the tty flip buffers with kmalloc objects in rings. In the normal situation for an IRQ driven serial port at typical speeds the behaviour is pretty much the same, two buffers end up allocated and the kernel cycles between them as before. When there are delays or at high speed we now behave far better as the buffer pool can grow a bit rather than lose characters. This also means that we can operate at higher speeds reliably. For drivers that receive characters in blocks (DMA based, USB and especially virtualisation) the layer allows a lot of driver specific code that works around the tty layer with private secondary queues to be removed. The IBM folks need this sort of layer, the smart serial port people do, the virtualisers do (because a virtualised tty typically operates at infinite speed rather than emulating 9600 baud). Finally many drivers had invalid and unsafe attempts to avoid buffer overflows by directly invoking tty methods extracted out of the innards of work queue structs. These are no longer needed and all go away. That fixes various random hangs with serial ports on overflow. The other change in here is to optimise the receive_room path that is used by some callers. It turns out that only one ldisc uses receive room except asa constant and it updates it far far less than the value is read. We thus make it a variable not a function call. I expect the code to contain bugs due to the size alone but I'll be watching and squashing them and feeding out new patches as it goes. Because the buffers now dynamically expand you should only run out of buffering when the kernel runs out of memory for real. That means a lot of the horrible hacks high performance drivers used to do just aren't needed any more. Description: tty_insert_flip_char is an old API and continues to work as before, as does tty_flip_buffer_push() [this is why many drivers dont need modification]. It does now also return the number of chars inserted There are also tty_buffer_request_room(tty, len) which asks for a buffer block of the length requested and returns the space found. This improves efficiency with hardware that knows how much to transfer. and tty_insert_flip_string_flags(tty, str, flags, len) to insert a string of characters and flags For a smart interface the usual code is len = tty_request_buffer_room(tty, amount_hardware_says); tty_insert_flip_string(tty, buffer_from_card, len); More description! At the moment tty buffers are attached directly to the tty. This is causing a lot of the problems related to tty layer locking, also problems at high speed and also with bursty data (such as occurs in virtualised environments) I'm working on ripping out the flip buffers and replacing them with a pool of dynamically allocated buffers. This allows both for old style "byte I/O" devices and also helps virtualisation and smart devices where large blocks of data suddenely materialise and need storing. So far so good. Lots of drivers reference tty->flip.*. Several of them also call directly and unsafely into function pointers it provides. This will all break. Most drivers can use tty_insert_flip_char which can be kept as an API but others need more. At the moment I've added the following interfaces, if people think more will be needed now is a good time to say int tty_buffer_request_room(tty, size) Try and ensure at least size bytes are available, returns actual room (may be zero). At the moment it just uses the flipbuf space but that will change. Repeated calls without characters being added are not cumulative. (ie if you call it with 1, 1, 1, and then 4 you'll have four characters of space. The other functions will also try and grow buffers in future but this will be a more efficient way when you know block sizes. int tty_insert_flip_char(tty, ch, flag) As before insert a character if there is room. Now returns 1 for success, 0 for failure. int tty_insert_flip_string(tty, str, len) Insert a block of non error characters. Returns the number inserted. int tty_prepare_flip_string(tty, strptr, len) Adjust the buffer to allow len characters to be added. Returns a buffer pointer in strptr and the length available. This allows for hardware that needs to use functions like insl or mencpy_fromio. Signed-off-by: Alan Cox <alan@redhat.com> Cc: Paul Fulghum <paulkf@microgate.com> Signed-off-by: Hirokazu Takata <takata@linux-m32r.org> Signed-off-by: Serge Hallyn <serue@us.ibm.com> Signed-off-by: Jeff Dike <jdike@addtoit.com> Signed-off-by: John Hawkes <hawkes@sgi.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Adrian Bunk <bunk@stusta.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-09 21:54:13 -07:00
flag = TTY_BREAK;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
} else if (stat->sreg.isr0 & SAB82532_ISR0_PERR)
[PATCH] TTY layer buffering revamp The API and code have been through various bits of initial review by serial driver people but they definitely need to live somewhere for a while so the unconverted drivers can get knocked into shape, existing drivers that have been updated can be better tuned and bugs whacked out. This replaces the tty flip buffers with kmalloc objects in rings. In the normal situation for an IRQ driven serial port at typical speeds the behaviour is pretty much the same, two buffers end up allocated and the kernel cycles between them as before. When there are delays or at high speed we now behave far better as the buffer pool can grow a bit rather than lose characters. This also means that we can operate at higher speeds reliably. For drivers that receive characters in blocks (DMA based, USB and especially virtualisation) the layer allows a lot of driver specific code that works around the tty layer with private secondary queues to be removed. The IBM folks need this sort of layer, the smart serial port people do, the virtualisers do (because a virtualised tty typically operates at infinite speed rather than emulating 9600 baud). Finally many drivers had invalid and unsafe attempts to avoid buffer overflows by directly invoking tty methods extracted out of the innards of work queue structs. These are no longer needed and all go away. That fixes various random hangs with serial ports on overflow. The other change in here is to optimise the receive_room path that is used by some callers. It turns out that only one ldisc uses receive room except asa constant and it updates it far far less than the value is read. We thus make it a variable not a function call. I expect the code to contain bugs due to the size alone but I'll be watching and squashing them and feeding out new patches as it goes. Because the buffers now dynamically expand you should only run out of buffering when the kernel runs out of memory for real. That means a lot of the horrible hacks high performance drivers used to do just aren't needed any more. Description: tty_insert_flip_char is an old API and continues to work as before, as does tty_flip_buffer_push() [this is why many drivers dont need modification]. It does now also return the number of chars inserted There are also tty_buffer_request_room(tty, len) which asks for a buffer block of the length requested and returns the space found. This improves efficiency with hardware that knows how much to transfer. and tty_insert_flip_string_flags(tty, str, flags, len) to insert a string of characters and flags For a smart interface the usual code is len = tty_request_buffer_room(tty, amount_hardware_says); tty_insert_flip_string(tty, buffer_from_card, len); More description! At the moment tty buffers are attached directly to the tty. This is causing a lot of the problems related to tty layer locking, also problems at high speed and also with bursty data (such as occurs in virtualised environments) I'm working on ripping out the flip buffers and replacing them with a pool of dynamically allocated buffers. This allows both for old style "byte I/O" devices and also helps virtualisation and smart devices where large blocks of data suddenely materialise and need storing. So far so good. Lots of drivers reference tty->flip.*. Several of them also call directly and unsafely into function pointers it provides. This will all break. Most drivers can use tty_insert_flip_char which can be kept as an API but others need more. At the moment I've added the following interfaces, if people think more will be needed now is a good time to say int tty_buffer_request_room(tty, size) Try and ensure at least size bytes are available, returns actual room (may be zero). At the moment it just uses the flipbuf space but that will change. Repeated calls without characters being added are not cumulative. (ie if you call it with 1, 1, 1, and then 4 you'll have four characters of space. The other functions will also try and grow buffers in future but this will be a more efficient way when you know block sizes. int tty_insert_flip_char(tty, ch, flag) As before insert a character if there is room. Now returns 1 for success, 0 for failure. int tty_insert_flip_string(tty, str, len) Insert a block of non error characters. Returns the number inserted. int tty_prepare_flip_string(tty, strptr, len) Adjust the buffer to allow len characters to be added. Returns a buffer pointer in strptr and the length available. This allows for hardware that needs to use functions like insl or mencpy_fromio. Signed-off-by: Alan Cox <alan@redhat.com> Cc: Paul Fulghum <paulkf@microgate.com> Signed-off-by: Hirokazu Takata <takata@linux-m32r.org> Signed-off-by: Serge Hallyn <serue@us.ibm.com> Signed-off-by: Jeff Dike <jdike@addtoit.com> Signed-off-by: John Hawkes <hawkes@sgi.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Adrian Bunk <bunk@stusta.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-09 21:54:13 -07:00
flag = TTY_PARITY;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
else if (stat->sreg.isr0 & SAB82532_ISR0_FERR)
[PATCH] TTY layer buffering revamp The API and code have been through various bits of initial review by serial driver people but they definitely need to live somewhere for a while so the unconverted drivers can get knocked into shape, existing drivers that have been updated can be better tuned and bugs whacked out. This replaces the tty flip buffers with kmalloc objects in rings. In the normal situation for an IRQ driven serial port at typical speeds the behaviour is pretty much the same, two buffers end up allocated and the kernel cycles between them as before. When there are delays or at high speed we now behave far better as the buffer pool can grow a bit rather than lose characters. This also means that we can operate at higher speeds reliably. For drivers that receive characters in blocks (DMA based, USB and especially virtualisation) the layer allows a lot of driver specific code that works around the tty layer with private secondary queues to be removed. The IBM folks need this sort of layer, the smart serial port people do, the virtualisers do (because a virtualised tty typically operates at infinite speed rather than emulating 9600 baud). Finally many drivers had invalid and unsafe attempts to avoid buffer overflows by directly invoking tty methods extracted out of the innards of work queue structs. These are no longer needed and all go away. That fixes various random hangs with serial ports on overflow. The other change in here is to optimise the receive_room path that is used by some callers. It turns out that only one ldisc uses receive room except asa constant and it updates it far far less than the value is read. We thus make it a variable not a function call. I expect the code to contain bugs due to the size alone but I'll be watching and squashing them and feeding out new patches as it goes. Because the buffers now dynamically expand you should only run out of buffering when the kernel runs out of memory for real. That means a lot of the horrible hacks high performance drivers used to do just aren't needed any more. Description: tty_insert_flip_char is an old API and continues to work as before, as does tty_flip_buffer_push() [this is why many drivers dont need modification]. It does now also return the number of chars inserted There are also tty_buffer_request_room(tty, len) which asks for a buffer block of the length requested and returns the space found. This improves efficiency with hardware that knows how much to transfer. and tty_insert_flip_string_flags(tty, str, flags, len) to insert a string of characters and flags For a smart interface the usual code is len = tty_request_buffer_room(tty, amount_hardware_says); tty_insert_flip_string(tty, buffer_from_card, len); More description! At the moment tty buffers are attached directly to the tty. This is causing a lot of the problems related to tty layer locking, also problems at high speed and also with bursty data (such as occurs in virtualised environments) I'm working on ripping out the flip buffers and replacing them with a pool of dynamically allocated buffers. This allows both for old style "byte I/O" devices and also helps virtualisation and smart devices where large blocks of data suddenely materialise and need storing. So far so good. Lots of drivers reference tty->flip.*. Several of them also call directly and unsafely into function pointers it provides. This will all break. Most drivers can use tty_insert_flip_char which can be kept as an API but others need more. At the moment I've added the following interfaces, if people think more will be needed now is a good time to say int tty_buffer_request_room(tty, size) Try and ensure at least size bytes are available, returns actual room (may be zero). At the moment it just uses the flipbuf space but that will change. Repeated calls without characters being added are not cumulative. (ie if you call it with 1, 1, 1, and then 4 you'll have four characters of space. The other functions will also try and grow buffers in future but this will be a more efficient way when you know block sizes. int tty_insert_flip_char(tty, ch, flag) As before insert a character if there is room. Now returns 1 for success, 0 for failure. int tty_insert_flip_string(tty, str, len) Insert a block of non error characters. Returns the number inserted. int tty_prepare_flip_string(tty, strptr, len) Adjust the buffer to allow len characters to be added. Returns a buffer pointer in strptr and the length available. This allows for hardware that needs to use functions like insl or mencpy_fromio. Signed-off-by: Alan Cox <alan@redhat.com> Cc: Paul Fulghum <paulkf@microgate.com> Signed-off-by: Hirokazu Takata <takata@linux-m32r.org> Signed-off-by: Serge Hallyn <serue@us.ibm.com> Signed-off-by: Jeff Dike <jdike@addtoit.com> Signed-off-by: John Hawkes <hawkes@sgi.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Adrian Bunk <bunk@stusta.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-09 21:54:13 -07:00
flag = TTY_FRAME;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 06:55:46 -07:00
if (uart_handle_sysrq_char(&up->port, ch))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
continue;
if ((stat->sreg.isr0 & (up->port.ignore_status_mask & 0xff)) == 0 &&
[PATCH] TTY layer buffering revamp The API and code have been through various bits of initial review by serial driver people but they definitely need to live somewhere for a while so the unconverted drivers can get knocked into shape, existing drivers that have been updated can be better tuned and bugs whacked out. This replaces the tty flip buffers with kmalloc objects in rings. In the normal situation for an IRQ driven serial port at typical speeds the behaviour is pretty much the same, two buffers end up allocated and the kernel cycles between them as before. When there are delays or at high speed we now behave far better as the buffer pool can grow a bit rather than lose characters. This also means that we can operate at higher speeds reliably. For drivers that receive characters in blocks (DMA based, USB and especially virtualisation) the layer allows a lot of driver specific code that works around the tty layer with private secondary queues to be removed. The IBM folks need this sort of layer, the smart serial port people do, the virtualisers do (because a virtualised tty typically operates at infinite speed rather than emulating 9600 baud). Finally many drivers had invalid and unsafe attempts to avoid buffer overflows by directly invoking tty methods extracted out of the innards of work queue structs. These are no longer needed and all go away. That fixes various random hangs with serial ports on overflow. The other change in here is to optimise the receive_room path that is used by some callers. It turns out that only one ldisc uses receive room except asa constant and it updates it far far less than the value is read. We thus make it a variable not a function call. I expect the code to contain bugs due to the size alone but I'll be watching and squashing them and feeding out new patches as it goes. Because the buffers now dynamically expand you should only run out of buffering when the kernel runs out of memory for real. That means a lot of the horrible hacks high performance drivers used to do just aren't needed any more. Description: tty_insert_flip_char is an old API and continues to work as before, as does tty_flip_buffer_push() [this is why many drivers dont need modification]. It does now also return the number of chars inserted There are also tty_buffer_request_room(tty, len) which asks for a buffer block of the length requested and returns the space found. This improves efficiency with hardware that knows how much to transfer. and tty_insert_flip_string_flags(tty, str, flags, len) to insert a string of characters and flags For a smart interface the usual code is len = tty_request_buffer_room(tty, amount_hardware_says); tty_insert_flip_string(tty, buffer_from_card, len); More description! At the moment tty buffers are attached directly to the tty. This is causing a lot of the problems related to tty layer locking, also problems at high speed and also with bursty data (such as occurs in virtualised environments) I'm working on ripping out the flip buffers and replacing them with a pool of dynamically allocated buffers. This allows both for old style "byte I/O" devices and also helps virtualisation and smart devices where large blocks of data suddenely materialise and need storing. So far so good. Lots of drivers reference tty->flip.*. Several of them also call directly and unsafely into function pointers it provides. This will all break. Most drivers can use tty_insert_flip_char which can be kept as an API but others need more. At the moment I've added the following interfaces, if people think more will be needed now is a good time to say int tty_buffer_request_room(tty, size) Try and ensure at least size bytes are available, returns actual room (may be zero). At the moment it just uses the flipbuf space but that will change. Repeated calls without characters being added are not cumulative. (ie if you call it with 1, 1, 1, and then 4 you'll have four characters of space. The other functions will also try and grow buffers in future but this will be a more efficient way when you know block sizes. int tty_insert_flip_char(tty, ch, flag) As before insert a character if there is room. Now returns 1 for success, 0 for failure. int tty_insert_flip_string(tty, str, len) Insert a block of non error characters. Returns the number inserted. int tty_prepare_flip_string(tty, strptr, len) Adjust the buffer to allow len characters to be added. Returns a buffer pointer in strptr and the length available. This allows for hardware that needs to use functions like insl or mencpy_fromio. Signed-off-by: Alan Cox <alan@redhat.com> Cc: Paul Fulghum <paulkf@microgate.com> Signed-off-by: Hirokazu Takata <takata@linux-m32r.org> Signed-off-by: Serge Hallyn <serue@us.ibm.com> Signed-off-by: Jeff Dike <jdike@addtoit.com> Signed-off-by: John Hawkes <hawkes@sgi.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Adrian Bunk <bunk@stusta.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-09 21:54:13 -07:00
(stat->sreg.isr1 & ((up->port.ignore_status_mask >> 8) & 0xff)) == 0)
tty_insert_flip_char(tty, ch, flag);
if (stat->sreg.isr0 & SAB82532_ISR0_RFO)
tty_insert_flip_char(tty, 0, TTY_OVERRUN);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
if (saw_console_brk)
sun_do_break();
return tty;
}
[SERIAL] Clean up and fix tty transmission start/stoping The start_tx and stop_tx methods were passed a flag to indicate whether the start/stop was from the tty start/stop callbacks, and some drivers used this flag to decide whether to ask the UART to immediately stop transmission (where the UART supports such a feature.) There are other cases when we wish this to occur - when CTS is lowered, or if we change from soft to hard flow control and CTS is inactive. In these cases, this flag was false, and we would allow the transmitter to drain before stopping. There is really only one case where we want to let the transmitter drain before disabling, and that's when we run out of characters to send. Hence, re-jig the start_tx and stop_tx methods to eliminate this flag, and introduce new functions for the special "disable and allow transmitter to drain" case. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-08-31 02:12:14 -07:00
static void sunsab_stop_tx(struct uart_port *);
[SUNSAB]: Defer register updates until transmitter is idle. The chip can emit garbage characters if we touch the settings while characters are going out. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-11 11:34:32 -07:00
static void sunsab_tx_idle(struct uart_sunsab_port *);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
static void transmit_chars(struct uart_sunsab_port *up,
union sab82532_irq_status *stat)
{
serial: kill off uart_info We moved this into uart_state, now move the fields out of the separate structure and kill it off. Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2009-09-19 13:13:28 -07:00
struct circ_buf *xmit = &up->port.state->xmit;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
int i;
if (stat->sreg.isr1 & SAB82532_ISR1_ALLS) {
up->interrupt_mask1 |= SAB82532_IMR1_ALLS;
writeb(up->interrupt_mask1, &up->regs->w.imr1);
set_bit(SAB82532_ALLS, &up->irqflags);
}
#if 0 /* bde@nwlink.com says this check causes problems */
if (!(stat->sreg.isr1 & SAB82532_ISR1_XPR))
return;
#endif
if (!(readb(&up->regs->r.star) & SAB82532_STAR_XFW))
return;
set_bit(SAB82532_XPR, &up->irqflags);
[SUNSAB]: Defer register updates until transmitter is idle. The chip can emit garbage characters if we touch the settings while characters are going out. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-11 11:34:32 -07:00
sunsab_tx_idle(up);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
if (uart_circ_empty(xmit) || uart_tx_stopped(&up->port)) {
up->interrupt_mask1 |= SAB82532_IMR1_XPR;
writeb(up->interrupt_mask1, &up->regs->w.imr1);
return;
}
up->interrupt_mask1 &= ~(SAB82532_IMR1_ALLS|SAB82532_IMR1_XPR);
writeb(up->interrupt_mask1, &up->regs->w.imr1);
clear_bit(SAB82532_ALLS, &up->irqflags);
/* Stuff 32 bytes into Transmit FIFO. */
clear_bit(SAB82532_XPR, &up->irqflags);
for (i = 0; i < up->port.fifosize; i++) {
writeb(xmit->buf[xmit->tail],
&up->regs->w.xfifo[i]);
xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
up->port.icount.tx++;
if (uart_circ_empty(xmit))
break;
}
/* Issue a Transmit Frame command. */
sunsab_cec_wait(up);
writeb(SAB82532_CMDR_XF, &up->regs->w.cmdr);
if (uart_circ_chars_pending(xmit) < WAKEUP_CHARS)
uart_write_wakeup(&up->port);
if (uart_circ_empty(xmit))
[SERIAL] Clean up and fix tty transmission start/stoping The start_tx and stop_tx methods were passed a flag to indicate whether the start/stop was from the tty start/stop callbacks, and some drivers used this flag to decide whether to ask the UART to immediately stop transmission (where the UART supports such a feature.) There are other cases when we wish this to occur - when CTS is lowered, or if we change from soft to hard flow control and CTS is inactive. In these cases, this flag was false, and we would allow the transmitter to drain before stopping. There is really only one case where we want to let the transmitter drain before disabling, and that's when we run out of characters to send. Hence, re-jig the start_tx and stop_tx methods to eliminate this flag, and introduce new functions for the special "disable and allow transmitter to drain" case. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-08-31 02:12:14 -07:00
sunsab_stop_tx(&up->port);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
static void check_status(struct uart_sunsab_port *up,
union sab82532_irq_status *stat)
{
if (stat->sreg.isr0 & SAB82532_ISR0_CDSC)
uart_handle_dcd_change(&up->port,
!(readb(&up->regs->r.vstr) & SAB82532_VSTR_CD));
if (stat->sreg.isr1 & SAB82532_ISR1_CSC)
uart_handle_cts_change(&up->port,
(readb(&up->regs->r.star) & SAB82532_STAR_CTS));
if ((readb(&up->regs->r.pvr) & up->pvr_dsr_bit) ^ up->dsr) {
up->dsr = (readb(&up->regs->r.pvr) & up->pvr_dsr_bit) ? 0 : 1;
up->port.icount.dsr++;
}
serial: move delta_msr_wait into the tty_port This is used by various drivers not just serial and can be extracted as commonality Signed-off-by: Alan Cox <alan@linux.intel.com>
2009-09-19 13:13:31 -07:00
wake_up_interruptible(&up->port.state->port.delta_msr_wait);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 06:55:46 -07:00
static irqreturn_t sunsab_interrupt(int irq, void *dev_id)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
struct uart_sunsab_port *up = dev_id;
struct tty_struct *tty;
union sab82532_irq_status status;
unsigned long flags;
[SUNSAB]: Fix several bugs. * don't register irq until ->startup() (and release in ->shutdown()). That avoids oopsen with the current tree when interrupt comes before we'd set up the data structures for ttyb. * handle console=ttyS... even when OBP talks to screen/keyboard * register irq handler for each port, let kernel/irq/handle.c call it for both if needed. Kills code duplication in sunsab_interrupt(). BTW, there'd been bitrot in it - ttya handling had stopped calling check_status() on BRK (correctly), ttyb copy of that code had kept the bogus call in that case. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-17 16:47:07 -07:00
unsigned char gis;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
spin_lock_irqsave(&up->port.lock, flags);
status.stat = 0;
[SUNSAB]: Fix several bugs. * don't register irq until ->startup() (and release in ->shutdown()). That avoids oopsen with the current tree when interrupt comes before we'd set up the data structures for ttyb. * handle console=ttyS... even when OBP talks to screen/keyboard * register irq handler for each port, let kernel/irq/handle.c call it for both if needed. Kills code duplication in sunsab_interrupt(). BTW, there'd been bitrot in it - ttya handling had stopped calling check_status() on BRK (correctly), ttyb copy of that code had kept the bogus call in that case. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-17 16:47:07 -07:00
gis = readb(&up->regs->r.gis) >> up->gis_shift;
if (gis & 1)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
status.sreg.isr0 = readb(&up->regs->r.isr0);
[SUNSAB]: Fix several bugs. * don't register irq until ->startup() (and release in ->shutdown()). That avoids oopsen with the current tree when interrupt comes before we'd set up the data structures for ttyb. * handle console=ttyS... even when OBP talks to screen/keyboard * register irq handler for each port, let kernel/irq/handle.c call it for both if needed. Kills code duplication in sunsab_interrupt(). BTW, there'd been bitrot in it - ttya handling had stopped calling check_status() on BRK (correctly), ttyb copy of that code had kept the bogus call in that case. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-17 16:47:07 -07:00
if (gis & 2)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
status.sreg.isr1 = readb(&up->regs->r.isr1);
tty = NULL;
if (status.stat) {
if ((status.sreg.isr0 & (SAB82532_ISR0_TCD | SAB82532_ISR0_TIME |
SAB82532_ISR0_RFO | SAB82532_ISR0_RPF)) ||
(status.sreg.isr1 & SAB82532_ISR1_BRK))
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 06:55:46 -07:00
tty = receive_chars(up, &status);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
if ((status.sreg.isr0 & SAB82532_ISR0_CDSC) ||
(status.sreg.isr1 & SAB82532_ISR1_CSC))
check_status(up, &status);
if (status.sreg.isr1 & (SAB82532_ISR1_ALLS | SAB82532_ISR1_XPR))
transmit_chars(up, &status);
}
spin_unlock_irqrestore(&up->port.lock, flags);
if (tty)
tty_flip_buffer_push(tty);
return IRQ_HANDLED;
}
/* port->lock is not held. */
static unsigned int sunsab_tx_empty(struct uart_port *port)
{
struct uart_sunsab_port *up = (struct uart_sunsab_port *) port;
int ret;
/* Do not need a lock for a state test like this. */
if (test_bit(SAB82532_ALLS, &up->irqflags))
ret = TIOCSER_TEMT;
else
ret = 0;
return ret;
}
/* port->lock held by caller. */
static void sunsab_set_mctrl(struct uart_port *port, unsigned int mctrl)
{
struct uart_sunsab_port *up = (struct uart_sunsab_port *) port;
if (mctrl & TIOCM_RTS) {
[SUNSAB]: Defer register updates until transmitter is idle. The chip can emit garbage characters if we touch the settings while characters are going out. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-11 11:34:32 -07:00
up->cached_mode &= ~SAB82532_MODE_FRTS;
up->cached_mode |= SAB82532_MODE_RTS;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
} else {
[SUNSAB]: Defer register updates until transmitter is idle. The chip can emit garbage characters if we touch the settings while characters are going out. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-11 11:34:32 -07:00
up->cached_mode |= (SAB82532_MODE_FRTS |
SAB82532_MODE_RTS);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
if (mctrl & TIOCM_DTR) {
[SUNSAB]: Defer register updates until transmitter is idle. The chip can emit garbage characters if we touch the settings while characters are going out. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-11 11:34:32 -07:00
up->cached_pvr &= ~(up->pvr_dtr_bit);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
} else {
[SUNSAB]: Defer register updates until transmitter is idle. The chip can emit garbage characters if we touch the settings while characters are going out. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-11 11:34:32 -07:00
up->cached_pvr |= up->pvr_dtr_bit;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
[SUNSAB]: Defer register updates until transmitter is idle. The chip can emit garbage characters if we touch the settings while characters are going out. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-11 11:34:32 -07:00
set_bit(SAB82532_REGS_PENDING, &up->irqflags);
if (test_bit(SAB82532_XPR, &up->irqflags))
sunsab_tx_idle(up);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
[PATCH] Serial: Adjust serial locking This patch changes the way serial ports are locked when getting modem status. This change is necessary because we will need to atomically read the modem status and take action depending on the CTS status. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-29 01:42:38 -07:00
/* port->lock is held by caller and interrupts are disabled. */
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
static unsigned int sunsab_get_mctrl(struct uart_port *port)
{
struct uart_sunsab_port *up = (struct uart_sunsab_port *) port;
unsigned char val;
unsigned int result;
result = 0;
val = readb(&up->regs->r.pvr);
result |= (val & up->pvr_dsr_bit) ? 0 : TIOCM_DSR;
val = readb(&up->regs->r.vstr);
result |= (val & SAB82532_VSTR_CD) ? 0 : TIOCM_CAR;
val = readb(&up->regs->r.star);
result |= (val & SAB82532_STAR_CTS) ? TIOCM_CTS : 0;
return result;
}
/* port->lock held by caller. */
[SERIAL] Clean up and fix tty transmission start/stoping The start_tx and stop_tx methods were passed a flag to indicate whether the start/stop was from the tty start/stop callbacks, and some drivers used this flag to decide whether to ask the UART to immediately stop transmission (where the UART supports such a feature.) There are other cases when we wish this to occur - when CTS is lowered, or if we change from soft to hard flow control and CTS is inactive. In these cases, this flag was false, and we would allow the transmitter to drain before stopping. There is really only one case where we want to let the transmitter drain before disabling, and that's when we run out of characters to send. Hence, re-jig the start_tx and stop_tx methods to eliminate this flag, and introduce new functions for the special "disable and allow transmitter to drain" case. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-08-31 02:12:14 -07:00
static void sunsab_stop_tx(struct uart_port *port)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
struct uart_sunsab_port *up = (struct uart_sunsab_port *) port;
up->interrupt_mask1 |= SAB82532_IMR1_XPR;
writeb(up->interrupt_mask1, &up->regs->w.imr1);
}
[SUNSAB]: Defer register updates until transmitter is idle. The chip can emit garbage characters if we touch the settings while characters are going out. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-11 11:34:32 -07:00
/* port->lock held by caller. */
static void sunsab_tx_idle(struct uart_sunsab_port *up)
{
if (test_bit(SAB82532_REGS_PENDING, &up->irqflags)) {
u8 tmp;
clear_bit(SAB82532_REGS_PENDING, &up->irqflags);
writeb(up->cached_mode, &up->regs->rw.mode);
writeb(up->cached_pvr, &up->regs->rw.pvr);
writeb(up->cached_dafo, &up->regs->w.dafo);
writeb(up->cached_ebrg & 0xff, &up->regs->w.bgr);
tmp = readb(&up->regs->rw.ccr2);
tmp &= ~0xc0;
tmp |= (up->cached_ebrg >> 2) & 0xc0;
writeb(tmp, &up->regs->rw.ccr2);
}
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
/* port->lock held by caller. */
[SERIAL] Clean up and fix tty transmission start/stoping The start_tx and stop_tx methods were passed a flag to indicate whether the start/stop was from the tty start/stop callbacks, and some drivers used this flag to decide whether to ask the UART to immediately stop transmission (where the UART supports such a feature.) There are other cases when we wish this to occur - when CTS is lowered, or if we change from soft to hard flow control and CTS is inactive. In these cases, this flag was false, and we would allow the transmitter to drain before stopping. There is really only one case where we want to let the transmitter drain before disabling, and that's when we run out of characters to send. Hence, re-jig the start_tx and stop_tx methods to eliminate this flag, and introduce new functions for the special "disable and allow transmitter to drain" case. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-08-31 02:12:14 -07:00
static void sunsab_start_tx(struct uart_port *port)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
struct uart_sunsab_port *up = (struct uart_sunsab_port *) port;
serial: kill off uart_info We moved this into uart_state, now move the fields out of the separate structure and kill it off. Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2009-09-19 13:13:28 -07:00
struct circ_buf *xmit = &up->port.state->xmit;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
int i;
up->interrupt_mask1 &= ~(SAB82532_IMR1_ALLS|SAB82532_IMR1_XPR);
writeb(up->interrupt_mask1, &up->regs->w.imr1);
if (!test_bit(SAB82532_XPR, &up->irqflags))
return;
clear_bit(SAB82532_ALLS, &up->irqflags);
clear_bit(SAB82532_XPR, &up->irqflags);
for (i = 0; i < up->port.fifosize; i++) {
writeb(xmit->buf[xmit->tail],
&up->regs->w.xfifo[i]);
xmit->tail = (xmit->tail + 1) & (UART_XMIT_SIZE - 1);
up->port.icount.tx++;
if (uart_circ_empty(xmit))
break;
}
/* Issue a Transmit Frame command. */
sunsab_cec_wait(up);
writeb(SAB82532_CMDR_XF, &up->regs->w.cmdr);
}
/* port->lock is not held. */
static void sunsab_send_xchar(struct uart_port *port, char ch)
{
struct uart_sunsab_port *up = (struct uart_sunsab_port *) port;
unsigned long flags;
spin_lock_irqsave(&up->port.lock, flags);
sunsab_tec_wait(up);
writeb(ch, &up->regs->w.tic);
spin_unlock_irqrestore(&up->port.lock, flags);
}
/* port->lock held by caller. */
static void sunsab_stop_rx(struct uart_port *port)
{
struct uart_sunsab_port *up = (struct uart_sunsab_port *) port;
up->interrupt_mask0 |= SAB82532_ISR0_TCD;
writeb(up->interrupt_mask1, &up->regs->w.imr0);
}
/* port->lock held by caller. */
static void sunsab_enable_ms(struct uart_port *port)
{
/* For now we always receive these interrupts. */
}
/* port->lock is not held. */
static void sunsab_break_ctl(struct uart_port *port, int break_state)
{
struct uart_sunsab_port *up = (struct uart_sunsab_port *) port;
unsigned long flags;
unsigned char val;
spin_lock_irqsave(&up->port.lock, flags);
[SUNSAB]: Defer register updates until transmitter is idle. The chip can emit garbage characters if we touch the settings while characters are going out. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-11 11:34:32 -07:00
val = up->cached_dafo;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
if (break_state)
val |= SAB82532_DAFO_XBRK;
else
val &= ~SAB82532_DAFO_XBRK;
[SUNSAB]: Defer register updates until transmitter is idle. The chip can emit garbage characters if we touch the settings while characters are going out. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-11 11:34:32 -07:00
up->cached_dafo = val;
set_bit(SAB82532_REGS_PENDING, &up->irqflags);
if (test_bit(SAB82532_XPR, &up->irqflags))
sunsab_tx_idle(up);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
spin_unlock_irqrestore(&up->port.lock, flags);
}
/* port->lock is not held. */
static int sunsab_startup(struct uart_port *port)
{
struct uart_sunsab_port *up = (struct uart_sunsab_port *) port;
unsigned long flags;
unsigned char tmp;
[SUNSAB]: Fix several bugs. * don't register irq until ->startup() (and release in ->shutdown()). That avoids oopsen with the current tree when interrupt comes before we'd set up the data structures for ttyb. * handle console=ttyS... even when OBP talks to screen/keyboard * register irq handler for each port, let kernel/irq/handle.c call it for both if needed. Kills code duplication in sunsab_interrupt(). BTW, there'd been bitrot in it - ttya handling had stopped calling check_status() on BRK (correctly), ttyb copy of that code had kept the bogus call in that case. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-17 16:47:07 -07:00
int err = request_irq(up->port.irq, sunsab_interrupt,
IRQF_SHARED, "sab", up);
if (err)
return err;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
spin_lock_irqsave(&up->port.lock, flags);
/*
* Wait for any commands or immediate characters
*/
sunsab_cec_wait(up);
sunsab_tec_wait(up);
/*
* Clear the FIFO buffers.
*/
writeb(SAB82532_CMDR_RRES, &up->regs->w.cmdr);
sunsab_cec_wait(up);
writeb(SAB82532_CMDR_XRES, &up->regs->w.cmdr);
/*
* Clear the interrupt registers.
*/
(void) readb(&up->regs->r.isr0);
(void) readb(&up->regs->r.isr1);
/*
* Now, initialize the UART
*/
writeb(0, &up->regs->w.ccr0); /* power-down */
writeb(SAB82532_CCR0_MCE | SAB82532_CCR0_SC_NRZ |
SAB82532_CCR0_SM_ASYNC, &up->regs->w.ccr0);
writeb(SAB82532_CCR1_ODS | SAB82532_CCR1_BCR | 7, &up->regs->w.ccr1);
writeb(SAB82532_CCR2_BDF | SAB82532_CCR2_SSEL |
SAB82532_CCR2_TOE, &up->regs->w.ccr2);
writeb(0, &up->regs->w.ccr3);
writeb(SAB82532_CCR4_MCK4 | SAB82532_CCR4_EBRG, &up->regs->w.ccr4);
[SUNSAB]: Defer register updates until transmitter is idle. The chip can emit garbage characters if we touch the settings while characters are going out. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-11 11:34:32 -07:00
up->cached_mode = (SAB82532_MODE_RTS | SAB82532_MODE_FCTS |
SAB82532_MODE_RAC);
writeb(up->cached_mode, &up->regs->w.mode);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
writeb(SAB82532_RFC_DPS|SAB82532_RFC_RFTH_32, &up->regs->w.rfc);
tmp = readb(&up->regs->rw.ccr0);
tmp |= SAB82532_CCR0_PU; /* power-up */
writeb(tmp, &up->regs->rw.ccr0);
/*
* Finally, enable interrupts
*/
up->interrupt_mask0 = (SAB82532_IMR0_PERR | SAB82532_IMR0_FERR |
SAB82532_IMR0_PLLA);
writeb(up->interrupt_mask0, &up->regs->w.imr0);
up->interrupt_mask1 = (SAB82532_IMR1_BRKT | SAB82532_IMR1_ALLS |
SAB82532_IMR1_XOFF | SAB82532_IMR1_TIN |
SAB82532_IMR1_CSC | SAB82532_IMR1_XON |
SAB82532_IMR1_XPR);
writeb(up->interrupt_mask1, &up->regs->w.imr1);
set_bit(SAB82532_ALLS, &up->irqflags);
set_bit(SAB82532_XPR, &up->irqflags);
spin_unlock_irqrestore(&up->port.lock, flags);
return 0;
}
/* port->lock is not held. */
static void sunsab_shutdown(struct uart_port *port)
{
struct uart_sunsab_port *up = (struct uart_sunsab_port *) port;
unsigned long flags;
spin_lock_irqsave(&up->port.lock, flags);
/* Disable Interrupts */
up->interrupt_mask0 = 0xff;
writeb(up->interrupt_mask0, &up->regs->w.imr0);
up->interrupt_mask1 = 0xff;
writeb(up->interrupt_mask1, &up->regs->w.imr1);
/* Disable break condition */
[SUNSAB]: Defer register updates until transmitter is idle. The chip can emit garbage characters if we touch the settings while characters are going out. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-11 11:34:32 -07:00
up->cached_dafo = readb(&up->regs->rw.dafo);
up->cached_dafo &= ~SAB82532_DAFO_XBRK;
writeb(up->cached_dafo, &up->regs->rw.dafo);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
/* Disable Receiver */
[SUNSAB]: Defer register updates until transmitter is idle. The chip can emit garbage characters if we touch the settings while characters are going out. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-11 11:34:32 -07:00
up->cached_mode &= ~SAB82532_MODE_RAC;
writeb(up->cached_mode, &up->regs->rw.mode);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
/*
* XXX FIXME
*
* If the chip is powered down here the system hangs/crashes during
* reboot or shutdown. This needs to be investigated further,
* similar behaviour occurs in 2.4 when the driver is configured
* as a module only. One hint may be that data is sometimes
* transmitted at 9600 baud during shutdown (regardless of the
* speed the chip was configured for when the port was open).
*/
#if 0
/* Power Down */
tmp = readb(&up->regs->rw.ccr0);
tmp &= ~SAB82532_CCR0_PU;
writeb(tmp, &up->regs->rw.ccr0);
#endif
spin_unlock_irqrestore(&up->port.lock, flags);
[SUNSAB]: Fix several bugs. * don't register irq until ->startup() (and release in ->shutdown()). That avoids oopsen with the current tree when interrupt comes before we'd set up the data structures for ttyb. * handle console=ttyS... even when OBP talks to screen/keyboard * register irq handler for each port, let kernel/irq/handle.c call it for both if needed. Kills code duplication in sunsab_interrupt(). BTW, there'd been bitrot in it - ttya handling had stopped calling check_status() on BRK (correctly), ttyb copy of that code had kept the bogus call in that case. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-17 16:47:07 -07:00
free_irq(up->port.irq, up);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
/*
* This is used to figure out the divisor speeds.
*
* The formula is: Baud = SAB_BASE_BAUD / ((N + 1) * (1 << M)),
*
* with 0 <= N < 64 and 0 <= M < 16
*/
static void calc_ebrg(int baud, int *n_ret, int *m_ret)
{
int n, m;
if (baud == 0) {
*n_ret = 0;
*m_ret = 0;
return;
}
/*
* We scale numbers by 10 so that we get better accuracy
* without having to use floating point. Here we increment m
* until n is within the valid range.
*/
n = (SAB_BASE_BAUD * 10) / baud;
m = 0;
while (n >= 640) {
n = n / 2;
m++;
}
n = (n+5) / 10;
/*
* We try very hard to avoid speeds with M == 0 since they may
* not work correctly for XTAL frequences above 10 MHz.
*/
if ((m == 0) && ((n & 1) == 0)) {
n = n / 2;
m++;
}
*n_ret = n - 1;
*m_ret = m;
}
/* Internal routine, port->lock is held and local interrupts are disabled. */
static void sunsab_convert_to_sab(struct uart_sunsab_port *up, unsigned int cflag,
[SPARC64]: In sunsab driver, make sure to set the uart timeout. This breaks serial consoles badly. Thanks to Eric Brower for tracking down the problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-04-21 22:18:03 -07:00
unsigned int iflag, unsigned int baud,
unsigned int quot)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
unsigned char dafo;
int bits, n, m;
/* Byte size and parity */
switch (cflag & CSIZE) {
case CS5: dafo = SAB82532_DAFO_CHL5; bits = 7; break;
case CS6: dafo = SAB82532_DAFO_CHL6; bits = 8; break;
case CS7: dafo = SAB82532_DAFO_CHL7; bits = 9; break;
case CS8: dafo = SAB82532_DAFO_CHL8; bits = 10; break;
/* Never happens, but GCC is too dumb to figure it out */
default: dafo = SAB82532_DAFO_CHL5; bits = 7; break;
}
if (cflag & CSTOPB) {
dafo |= SAB82532_DAFO_STOP;
bits++;
}
if (cflag & PARENB) {
dafo |= SAB82532_DAFO_PARE;
bits++;
}
if (cflag & PARODD) {
dafo |= SAB82532_DAFO_PAR_ODD;
} else {
dafo |= SAB82532_DAFO_PAR_EVEN;
}
[SUNSAB]: Defer register updates until transmitter is idle. The chip can emit garbage characters if we touch the settings while characters are going out. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-11 11:34:32 -07:00
up->cached_dafo = dafo;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
calc_ebrg(baud, &n, &m);
[SUNSAB]: Defer register updates until transmitter is idle. The chip can emit garbage characters if we touch the settings while characters are going out. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-11 11:34:32 -07:00
up->cached_ebrg = n | (m << 6);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
up->tec_timeout = (10 * 1000000) / baud;
up->cec_timeout = up->tec_timeout >> 2;
/* CTS flow control flags */
/* We encode read_status_mask and ignore_status_mask like so:
*
* ---------------------
* | ... | ISR1 | ISR0 |
* ---------------------
* .. 15 8 7 0
*/
up->port.read_status_mask = (SAB82532_ISR0_TCD | SAB82532_ISR0_TIME |
SAB82532_ISR0_RFO | SAB82532_ISR0_RPF |
SAB82532_ISR0_CDSC);
up->port.read_status_mask |= (SAB82532_ISR1_CSC |
SAB82532_ISR1_ALLS |
SAB82532_ISR1_XPR) << 8;
if (iflag & INPCK)
up->port.read_status_mask |= (SAB82532_ISR0_PERR |
SAB82532_ISR0_FERR);
if (iflag & (BRKINT | PARMRK))
up->port.read_status_mask |= (SAB82532_ISR1_BRK << 8);
/*
* Characteres to ignore
*/
up->port.ignore_status_mask = 0;
if (iflag & IGNPAR)
up->port.ignore_status_mask |= (SAB82532_ISR0_PERR |
SAB82532_ISR0_FERR);
if (iflag & IGNBRK) {
up->port.ignore_status_mask |= (SAB82532_ISR1_BRK << 8);
/*
* If we're ignoring parity and break indicators,
* ignore overruns too (for real raw support).
*/
if (iflag & IGNPAR)
up->port.ignore_status_mask |= SAB82532_ISR0_RFO;
}
/*
* ignore all characters if CREAD is not set
*/
if ((cflag & CREAD) == 0)
up->port.ignore_status_mask |= (SAB82532_ISR0_RPF |
SAB82532_ISR0_TCD);
[SPARC64]: In sunsab driver, make sure to set the uart timeout. This breaks serial consoles badly. Thanks to Eric Brower for tracking down the problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-04-21 22:18:03 -07:00
uart_update_timeout(&up->port, cflag,
(up->port.uartclk / (16 * quot)));
[SUNSAB]: Defer register updates until transmitter is idle. The chip can emit garbage characters if we touch the settings while characters are going out. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-05-11 11:34:32 -07:00
/* Now schedule a register update when the chip's
* transmitter is idle.
*/
up->cached_mode |= SAB82532_MODE_RAC;
set_bit(SAB82532_REGS_PENDING, &up->irqflags);
if (test_bit(SAB82532_XPR, &up->irqflags))
sunsab_tx_idle(up);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
/* port->lock is not held. */
[PATCH] tty: switch to ktermios This is the grungy swap all the occurrences in the right places patch that goes with the updates. At this point we have the same functionality as before (except that sgttyb() returns speeds not zero) and are ready to begin turning new stuff on providing nobody reports lots of bugs If you are a tty driver author converting an out of tree driver the only impact should be termios->ktermios name changes for the speed/property setting functions from your upper layers. If you are implementing your own TCGETS function before then your driver was broken already and its about to get a whole lot more painful for you so please fix it 8) Also fill in c_ispeed/ospeed on init for most devices, although the current code will do this for you anyway but I'd like eventually to lose that extra paranoia [akpm@osdl.org: bluetooth fix] [mp3@de.ibm.com: sclp fix] [mp3@de.ibm.com: warning fix for tty3270] [hugh@veritas.com: fix tty_ioctl powerpc build] [jdike@addtoit.com: uml: fix ->set_termios declaration] Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Martin Peschke <mp3@de.ibm.com> Acked-by: Peter Oberparleiter <oberpar@de.ibm.com> Cc: Cornelia Huck <cornelia.huck@de.ibm.com> Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Jeff Dike <jdike@addtoit.com> Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-08 03:38:45 -07:00
static void sunsab_set_termios(struct uart_port *port, struct ktermios *termios,
struct ktermios *old)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
struct uart_sunsab_port *up = (struct uart_sunsab_port *) port;
unsigned long flags;
[SPARC64]: In sunsab driver, make sure to set the uart timeout. This breaks serial consoles badly. Thanks to Eric Brower for tracking down the problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-04-21 22:18:03 -07:00
unsigned int baud = uart_get_baud_rate(port, termios, old, 0, 4000000);
unsigned int quot = uart_get_divisor(port, baud);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
spin_lock_irqsave(&up->port.lock, flags);
[SPARC64]: In sunsab driver, make sure to set the uart timeout. This breaks serial consoles badly. Thanks to Eric Brower for tracking down the problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-04-21 22:18:03 -07:00
sunsab_convert_to_sab(up, termios->c_cflag, termios->c_iflag, baud, quot);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
spin_unlock_irqrestore(&up->port.lock, flags);
}
static const char *sunsab_type(struct uart_port *port)
{
struct uart_sunsab_port *up = (void *)port;
static char buf[36];
sprintf(buf, "SAB82532 %s", sab82532_version[up->type]);
return buf;
}
static void sunsab_release_port(struct uart_port *port)
{
}
static int sunsab_request_port(struct uart_port *port)
{
return 0;
}
static void sunsab_config_port(struct uart_port *port, int flags)
{
}
static int sunsab_verify_port(struct uart_port *port, struct serial_struct *ser)
{
return -EINVAL;
}
static struct uart_ops sunsab_pops = {
.tx_empty = sunsab_tx_empty,
.set_mctrl = sunsab_set_mctrl,
.get_mctrl = sunsab_get_mctrl,
.stop_tx = sunsab_stop_tx,
.start_tx = sunsab_start_tx,
.send_xchar = sunsab_send_xchar,
.stop_rx = sunsab_stop_rx,
.enable_ms = sunsab_enable_ms,
.break_ctl = sunsab_break_ctl,
.startup = sunsab_startup,
.shutdown = sunsab_shutdown,
.set_termios = sunsab_set_termios,
.type = sunsab_type,
.release_port = sunsab_release_port,
.request_port = sunsab_request_port,
.config_port = sunsab_config_port,
.verify_port = sunsab_verify_port,
};
static struct uart_driver sunsab_reg = {
.owner = THIS_MODULE,
serial: Fix sparc driver name strings. They were all "serial" so if multiple of these drivers registered, we'd trigger sysfs directory creation errors: [ 1.695793] proc_dir_entry 'serial' already registered [ 1.695839] Call Trace: [ 1.831891] [00000000004f2534] create_proc_entry+0x7c/0x98 [ 1.833608] [00000000004f3a58] proc_tty_register_driver+0x40/0x70 [ 1.833663] [0000000000594700] tty_register_driver+0x1fc/0x208 [ 1.835371] [00000000005aade4] uart_register_driver+0x134/0x16c [ 1.841762] [00000000005ac274] sunserial_register_minors+0x34/0x68 [ 1.841818] [00000000007db2a4] sunsu_init+0xf8/0x150 [ 1.867697] [00000000007c62a4] kernel_init+0x190/0x330 [ 1.939147] [0000000000426cf8] kernel_thread+0x38/0x48 [ 1.939198] [00000000006a0d90] rest_init+0x18/0x5c Signed-off-by: David S. Miller <davem@davemloft.net>
2008-05-01 01:14:27 -07:00
.driver_name = "sunsab",
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
.dev_name = "ttyS",
.major = TTY_MAJOR,
};
static struct uart_sunsab_port *sunsab_ports;
#ifdef CONFIG_SERIAL_SUNSAB_CONSOLE
[SERIAL] kernel console should send CRLF not LFCR Glen Turner reported that writing LFCR rather than the more traditional CRLF causes issues with some terminals. Since this aflicts many serial drivers, extract the common code to a library function (uart_console_write) and arrange for each driver to supply a "putchar" function. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2006-03-20 13:00:09 -07:00
static void sunsab_console_putchar(struct uart_port *port, int c)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
[SERIAL] kernel console should send CRLF not LFCR Glen Turner reported that writing LFCR rather than the more traditional CRLF causes issues with some terminals. Since this aflicts many serial drivers, extract the common code to a library function (uart_console_write) and arrange for each driver to supply a "putchar" function. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2006-03-20 13:00:09 -07:00
struct uart_sunsab_port *up = (struct uart_sunsab_port *)port;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
sunsab_tec_wait(up);
writeb(c, &up->regs->w.tic);
}
static void sunsab_console_write(struct console *con, const char *s, unsigned n)
{
struct uart_sunsab_port *up = &sunsab_ports[con->index];
[SERIAL]: Fix console write locking in sparc drivers. Mirror the logic in 8250 for proper console write locking when SYSRQ is triggered or an OOPS is in progress. Signed-off-by: David S. Miller <davem@davemloft.net>
2007-07-15 23:53:32 -07:00
unsigned long flags;
int locked = 1;
local_irq_save(flags);
if (up->port.sysrq) {
locked = 0;
} else if (oops_in_progress) {
locked = spin_trylock(&up->port.lock);
} else
spin_lock(&up->port.lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SERIAL] kernel console should send CRLF not LFCR Glen Turner reported that writing LFCR rather than the more traditional CRLF causes issues with some terminals. Since this aflicts many serial drivers, extract the common code to a library function (uart_console_write) and arrange for each driver to supply a "putchar" function. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2006-03-20 13:00:09 -07:00
uart_console_write(&up->port, s, n, sunsab_console_putchar);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
sunsab_tec_wait(up);
[SERIAL]: Fix console write locking in sparc drivers. Mirror the logic in 8250 for proper console write locking when SYSRQ is triggered or an OOPS is in progress. Signed-off-by: David S. Miller <davem@davemloft.net>
2007-07-15 23:53:32 -07:00
if (locked)
spin_unlock(&up->port.lock);
local_irq_restore(flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
static int sunsab_console_setup(struct console *con, char *options)
{
struct uart_sunsab_port *up = &sunsab_ports[con->index];
unsigned long flags;
[SPARC64]: In sunsab driver, make sure to set the uart timeout. This breaks serial consoles badly. Thanks to Eric Brower for tracking down the problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-04-21 22:18:03 -07:00
unsigned int baud, quot;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SERIAL] sunsab: Fix E250 console with RSC. This fixes yet another sunsab problem, when console is set to anything but the first port. The console framework calls sunsab_console_setup for each port, and we end up setting up a console on a not yet discovered port, which leads to an Oops. Instead, defer console setup until the requested port is properly initialized. Tested on an E250 through an RSC console. Reported by Daniel Smolik <marvin@mydatex.cz> Signed-off-by: Marc Zyngier <maz@misterjones.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2006-08-23 15:50:57 -07:00
/*
* The console framework calls us for each and every port
* registered. Defer the console setup until the requested
* port has been properly discovered. A bit of a hack,
* though...
*/
if (up->port.type != PORT_SUNSAB)
return -1;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
printk("Console: ttyS%d (SAB82532)\n",
(sunsab_reg.minor - 64) + con->index);
serial: suncore: Fix RSC/LOM handling in sunserial_console_termios(). RSC and LOM devices have fixed speed settings. We already had some code to match and handle "rsc" named devices on E250 systems, but we also have to handle 'rsc-console', 'rsc-control', and 'lom-console'. Also, in order to get this right regardless of what 'output-device' happens to be, explicitly pass the UART device node pointer to this routine. Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-24 15:09:56 -07:00
sunserial_console_termios(con, to_of_device(up->port.dev)->node);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
switch (con->cflag & CBAUD) {
case B150: baud = 150; break;
case B300: baud = 300; break;
case B600: baud = 600; break;
case B1200: baud = 1200; break;
case B2400: baud = 2400; break;
case B4800: baud = 4800; break;
default: case B9600: baud = 9600; break;
case B19200: baud = 19200; break;
case B38400: baud = 38400; break;
[SPARC64]: Serial Console for E250 Patch From: Eddie C. Dost <ecd@brainaid.de> I have the following patch for serial console over the RSC (remote system controller) on my E250 machine. It basically adds support for input-device=rsc and output-device=rsc from OBP, and allows 115200,8,n,1,- serial mode setting. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-01-18 15:54:31 -07:00
case B57600: baud = 57600; break;
case B115200: baud = 115200; break;
case B230400: baud = 230400; break;
case B460800: baud = 460800; break;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
};
/*
* Temporary fix.
*/
spin_lock_init(&up->port.lock);
/*
* Initialize the hardware
*/
sunsab_startup(&up->port);
spin_lock_irqsave(&up->port.lock, flags);
/*
* Finally, enable interrupts
*/
up->interrupt_mask0 = SAB82532_IMR0_PERR | SAB82532_IMR0_FERR |
SAB82532_IMR0_PLLA | SAB82532_IMR0_CDSC;
writeb(up->interrupt_mask0, &up->regs->w.imr0);
up->interrupt_mask1 = SAB82532_IMR1_BRKT | SAB82532_IMR1_ALLS |
SAB82532_IMR1_XOFF | SAB82532_IMR1_TIN |
SAB82532_IMR1_CSC | SAB82532_IMR1_XON |
SAB82532_IMR1_XPR;
writeb(up->interrupt_mask1, &up->regs->w.imr1);
[SPARC64]: In sunsab driver, make sure to set the uart timeout. This breaks serial consoles badly. Thanks to Eric Brower for tracking down the problem. Signed-off-by: David S. Miller <davem@davemloft.net>
2005-04-21 22:18:03 -07:00
quot = uart_get_divisor(&up->port, baud);
sunsab_convert_to_sab(up, con->cflag, 0, baud, quot);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
sunsab_set_mctrl(&up->port, TIOCM_DTR | TIOCM_RTS);
spin_unlock_irqrestore(&up->port.lock, flags);
return 0;
}
static struct console sunsab_console = {
.name = "ttyS",
.write = sunsab_console_write,
.device = uart_console_device,
.setup = sunsab_console_setup,
.flags = CON_PRINTBUFFER,
.index = -1,
.data = &sunsab_reg,
};
[SPARC64]: Prevent registering wrong serial console. If the console is not for a particular Sun serial controller, set the drv->cons to NULL. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-13 21:09:10 -07:00
static inline struct console *SUNSAB_CONSOLE(void)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
[SPARC64]: Prevent registering wrong serial console. If the console is not for a particular Sun serial controller, set the drv->cons to NULL. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-13 21:09:10 -07:00
return &sunsab_console;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
#else
[SPARC64]: Prevent registering wrong serial console. If the console is not for a particular Sun serial controller, set the drv->cons to NULL. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-13 21:09:10 -07:00
#define SUNSAB_CONSOLE() (NULL)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
#define sunsab_console_init() do { } while (0)
#endif
[SERIAL] sunsab: Fix section mis-match errors. sunsab_init_one() needs to be __devinit, not __init Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:18:50 -07:00
static int __devinit sunsab_init_one(struct uart_sunsab_port *up,
struct of_device *op,
unsigned long offset,
int line)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
up->port.line = line;
up->port.dev = &op->dev;
up->port.mapbase = op->resource[0].start + offset;
up->port.membase = of_ioremap(&op->resource[0], offset,
sizeof(union sab82532_async_regs),
"sab");
if (!up->port.membase)
return -ENOMEM;
up->regs = (union sab82532_async_regs __iomem *) up->port.membase;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
up->port.irq = op->irqs[0];
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
up->port.fifosize = SAB82532_XMIT_FIFO_SIZE;
up->port.iotype = UPIO_MEM;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
writeb(SAB82532_IPC_IC_ACT_LOW, &up->regs->w.ipc);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
up->port.ops = &sunsab_pops;
up->port.type = PORT_SUNSAB;
up->port.uartclk = SAB_BASE_BAUD;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
up->type = readb(&up->regs->r.vstr) & 0x0f;
writeb(~((1 << 1) | (1 << 2) | (1 << 4)), &up->regs->w.pcr);
writeb(0xff, &up->regs->w.pim);
if ((up->port.line & 0x1) == 0) {
up->pvr_dsr_bit = (1 << 0);
up->pvr_dtr_bit = (1 << 1);
[SUNSAB]: Fix several bugs. * don't register irq until ->startup() (and release in ->shutdown()). That avoids oopsen with the current tree when interrupt comes before we'd set up the data structures for ttyb. * handle console=ttyS... even when OBP talks to screen/keyboard * register irq handler for each port, let kernel/irq/handle.c call it for both if needed. Kills code duplication in sunsab_interrupt(). BTW, there'd been bitrot in it - ttya handling had stopped calling check_status() on BRK (correctly), ttyb copy of that code had kept the bogus call in that case. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-17 16:47:07 -07:00
up->gis_shift = 2;
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
} else {
up->pvr_dsr_bit = (1 << 3);
up->pvr_dtr_bit = (1 << 2);
[SUNSAB]: Fix several bugs. * don't register irq until ->startup() (and release in ->shutdown()). That avoids oopsen with the current tree when interrupt comes before we'd set up the data structures for ttyb. * handle console=ttyS... even when OBP talks to screen/keyboard * register irq handler for each port, let kernel/irq/handle.c call it for both if needed. Kills code duplication in sunsab_interrupt(). BTW, there'd been bitrot in it - ttya handling had stopped calling check_status() on BRK (correctly), ttyb copy of that code had kept the bogus call in that case. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-17 16:47:07 -07:00
up->gis_shift = 0;
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
}
up->cached_pvr = (1 << 1) | (1 << 2) | (1 << 4);
writeb(up->cached_pvr, &up->regs->w.pvr);
up->cached_mode = readb(&up->regs->rw.mode);
up->cached_mode |= SAB82532_MODE_FRTS;
writeb(up->cached_mode, &up->regs->rw.mode);
up->cached_mode |= SAB82532_MODE_RTS;
writeb(up->cached_mode, &up->regs->rw.mode);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
up->tec_timeout = SAB82532_MAX_TEC_TIMEOUT;
up->cec_timeout = SAB82532_MAX_CEC_TIMEOUT;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
return 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
static int __devinit sab_probe(struct of_device *op, const struct of_device_id *match)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
static int inst;
struct uart_sunsab_port *up;
int err;
up = &sunsab_ports[inst * 2];
err = sunsab_init_one(&up[0], op,
[SERIAL] sunsab: Get line numbers and table sizing correct. Table sizing code should look for "se" not "su" nodes. The chip at the lower address should get the first index. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-07-17 16:40:26 -07:00
0,
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
(inst * 2) + 0);
if (err)
[SUNSAB]: Fix several bugs. * don't register irq until ->startup() (and release in ->shutdown()). That avoids oopsen with the current tree when interrupt comes before we'd set up the data structures for ttyb. * handle console=ttyS... even when OBP talks to screen/keyboard * register irq handler for each port, let kernel/irq/handle.c call it for both if needed. Kills code duplication in sunsab_interrupt(). BTW, there'd been bitrot in it - ttya handling had stopped calling check_status() on BRK (correctly), ttyb copy of that code had kept the bogus call in that case. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-17 16:47:07 -07:00
goto out;
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
[SERIAL] sunsab: Get line numbers and table sizing correct. Table sizing code should look for "se" not "su" nodes. The chip at the lower address should get the first index. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-07-17 16:40:26 -07:00
err = sunsab_init_one(&up[1], op,
sizeof(union sab82532_async_regs),
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
(inst * 2) + 1);
[SUNSAB]: Fix several bugs. * don't register irq until ->startup() (and release in ->shutdown()). That avoids oopsen with the current tree when interrupt comes before we'd set up the data structures for ttyb. * handle console=ttyS... even when OBP talks to screen/keyboard * register irq handler for each port, let kernel/irq/handle.c call it for both if needed. Kills code duplication in sunsab_interrupt(). BTW, there'd been bitrot in it - ttya handling had stopped calling check_status() on BRK (correctly), ttyb copy of that code had kept the bogus call in that case. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-17 16:47:07 -07:00
if (err)
goto out1;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SPARC]: Fix serial console device detection. The current scheme works on static interpretation of text names, which is wrong. The output-device setting, for example, must be resolved via an alias or similar to a full path name to the console device. Paths also contain an optional set of 'options', which starts with a colon at the end of the path. The option area is used to specify which of two serial ports ('a' or 'b') the path refers to when a device node drives multiple ports. 'a' is assumed if the option specification is missing. This was caught by the UltraSPARC-T1 simulator. The 'output-device' property was set to 'ttya' and we didn't pick upon the fact that this is an OBP alias set to '/virtual-devices/console'. Instead we saw it as the first serial console device, instead of the hypervisor console. The infrastructure is now there to take advantage of this to resolve the console correctly even in multi-head situations in fbcon too. Thanks to Greg Onufer for the bug report. Signed-off-by: David S. Miller <davem@davemloft.net>
2007-07-20 16:59:26 -07:00
sunserial_console_match(SUNSAB_CONSOLE(), op->node,
serial: suncore: Add 'ignore_line' argument to sunserial_console_match(). This tells the logic to ignore the line match when deciding whether the device is the OpenFirmware specified console device or not. This is going to be used in the SU driver for rsc-console detection. There is probably a better way to handle this, but this is the least intrusive solution for now which we can validate won't break any other cases. Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-24 15:03:34 -07:00
&sunsab_reg, up[0].port.line,
false);
[SPARC]: Fix serial console device detection. The current scheme works on static interpretation of text names, which is wrong. The output-device setting, for example, must be resolved via an alias or similar to a full path name to the console device. Paths also contain an optional set of 'options', which starts with a colon at the end of the path. The option area is used to specify which of two serial ports ('a' or 'b') the path refers to when a device node drives multiple ports. 'a' is assumed if the option specification is missing. This was caught by the UltraSPARC-T1 simulator. The 'output-device' property was set to 'ttya' and we didn't pick upon the fact that this is an OBP alias set to '/virtual-devices/console'. Instead we saw it as the first serial console device, instead of the hypervisor console. The infrastructure is now there to take advantage of this to resolve the console correctly even in multi-head situations in fbcon too. Thanks to Greg Onufer for the bug report. Signed-off-by: David S. Miller <davem@davemloft.net>
2007-07-20 16:59:26 -07:00
sunserial_console_match(SUNSAB_CONSOLE(), op->node,
serial: suncore: Add 'ignore_line' argument to sunserial_console_match(). This tells the logic to ignore the line match when deciding whether the device is the OpenFirmware specified console device or not. This is going to be used in the SU driver for rsc-console detection. There is probably a better way to handle this, but this is the least intrusive solution for now which we can validate won't break any other cases. Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-24 15:03:34 -07:00
&sunsab_reg, up[1].port.line,
false);
[SUNSAB]: Fix several bugs. * don't register irq until ->startup() (and release in ->shutdown()). That avoids oopsen with the current tree when interrupt comes before we'd set up the data structures for ttyb. * handle console=ttyS... even when OBP talks to screen/keyboard * register irq handler for each port, let kernel/irq/handle.c call it for both if needed. Kills code duplication in sunsab_interrupt(). BTW, there'd been bitrot in it - ttya handling had stopped calling check_status() on BRK (correctly), ttyb copy of that code had kept the bogus call in that case. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-17 16:47:07 -07:00
err = uart_add_one_port(&sunsab_reg, &up[0].port);
if (err)
goto out2;
err = uart_add_one_port(&sunsab_reg, &up[1].port);
if (err)
goto out3;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
dev_set_drvdata(&op->dev, &up[0]);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
inst++;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
return 0;
[SPARC64]: Fix of_iounmap() region release. We need to pass in the resource otherwise we cannot release the region properly. We must know whether it is an I/O or MEM resource. Spotted by Eric Brower. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-12-28 22:01:32 -07:00
[SUNSAB]: Fix several bugs. * don't register irq until ->startup() (and release in ->shutdown()). That avoids oopsen with the current tree when interrupt comes before we'd set up the data structures for ttyb. * handle console=ttyS... even when OBP talks to screen/keyboard * register irq handler for each port, let kernel/irq/handle.c call it for both if needed. Kills code duplication in sunsab_interrupt(). BTW, there'd been bitrot in it - ttya handling had stopped calling check_status() on BRK (correctly), ttyb copy of that code had kept the bogus call in that case. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-17 16:47:07 -07:00
out3:
uart_remove_one_port(&sunsab_reg, &up[0].port);
out2:
[SPARC64]: Fix of_iounmap() region release. We need to pass in the resource otherwise we cannot release the region properly. We must know whether it is an I/O or MEM resource. Spotted by Eric Brower. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-12-28 22:01:32 -07:00
of_iounmap(&op->resource[0],
[SUNSAB]: Fix several bugs. * don't register irq until ->startup() (and release in ->shutdown()). That avoids oopsen with the current tree when interrupt comes before we'd set up the data structures for ttyb. * handle console=ttyS... even when OBP talks to screen/keyboard * register irq handler for each port, let kernel/irq/handle.c call it for both if needed. Kills code duplication in sunsab_interrupt(). BTW, there'd been bitrot in it - ttya handling had stopped calling check_status() on BRK (correctly), ttyb copy of that code had kept the bogus call in that case. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-17 16:47:07 -07:00
up[1].port.membase,
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
sizeof(union sab82532_async_regs));
[SUNSAB]: Fix several bugs. * don't register irq until ->startup() (and release in ->shutdown()). That avoids oopsen with the current tree when interrupt comes before we'd set up the data structures for ttyb. * handle console=ttyS... even when OBP talks to screen/keyboard * register irq handler for each port, let kernel/irq/handle.c call it for both if needed. Kills code duplication in sunsab_interrupt(). BTW, there'd been bitrot in it - ttya handling had stopped calling check_status() on BRK (correctly), ttyb copy of that code had kept the bogus call in that case. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-17 16:47:07 -07:00
out1:
of_iounmap(&op->resource[0],
up[0].port.membase,
sizeof(union sab82532_async_regs));
out:
return err;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
static int __devexit sab_remove(struct of_device *op)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
{
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
struct uart_sunsab_port *up = dev_get_drvdata(&op->dev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SUNSAB]: Fix several bugs. * don't register irq until ->startup() (and release in ->shutdown()). That avoids oopsen with the current tree when interrupt comes before we'd set up the data structures for ttyb. * handle console=ttyS... even when OBP talks to screen/keyboard * register irq handler for each port, let kernel/irq/handle.c call it for both if needed. Kills code duplication in sunsab_interrupt(). BTW, there'd been bitrot in it - ttya handling had stopped calling check_status() on BRK (correctly), ttyb copy of that code had kept the bogus call in that case. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-17 16:47:07 -07:00
uart_remove_one_port(&sunsab_reg, &up[1].port);
uart_remove_one_port(&sunsab_reg, &up[0].port);
of_iounmap(&op->resource[0],
up[1].port.membase,
sizeof(union sab82532_async_regs));
of_iounmap(&op->resource[0],
up[0].port.membase,
sizeof(union sab82532_async_regs));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
dev_set_drvdata(&op->dev, NULL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
return 0;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
sparc: Annotate of_device_id arrays with const or __initdata. As suggested by Stephen Rothwell. Signed-off-by: David S. Miller <davem@davemloft.net>
2008-08-31 01:23:17 -07:00
static const struct of_device_id sab_match[] = {
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
{
.name = "se",
},
{
.name = "serial",
.compatible = "sab82532",
},
{},
};
MODULE_DEVICE_TABLE(of, sab_match);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
static struct of_platform_driver sab_driver = {
.name = "sab",
.match_table = sab_match,
.probe = sab_probe,
.remove = __devexit_p(sab_remove),
};
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
static int __init sunsab_init(void)
{
struct device_node *dp;
int err;
[SERIAL] sparc: Infrastructure to fix section mismatch bugs. This patch against 2.6.23 sparc-2.6.git contains a number of minor cleanups of the sparc serial drivers. Initially I fixed this build warning: WARNING: vmlinux.o(.text+0x107a2c): Section mismatch: reference to .init.text:add_preferred_console (between 'sunserial_console_match' and 'sunserial_console_termios') which is done by declaring sunserial_console_match() as __init. This resulted in build warnings on sunserial_current_minor. To resolve these the variable was changed so it is no longer global, and to hide operations on it inside 2 new functions. These functions handle the UART minor handling code that is common to all sparc serial drivers. These changes allowed to clean up the uart counters in all the sparc serial drivers, and the administration of minor device numbers. Lastly, sunserial_console_termios() does not need to be exported since it is only called from non-modular code. Sadly, the following build warning still exists: WARNING: vmlinux.o(__ksymtab+0x2910): Section mismatch: reference to .init.text:sunserial_console_match (between '__ksymtab_sunserial_console_match' and '__ksymtab_sunserial_unregister_minors') This could be resolved by not exporting sunserial_console_match(), but this is not possible at the moment because it is being called from modular code. On the other hand, this is a bogus warning since it comes from a ksymtab section. Signed-off-by: Martin Habets <errandir_news@mph.eclipse.co.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-11 04:37:04 -07:00
int num_channels = 0;
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
[SERIAL] sunsab: Get line numbers and table sizing correct. Table sizing code should look for "se" not "su" nodes. The chip at the lower address should get the first index. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-07-17 16:40:26 -07:00
for_each_node_by_name(dp, "se")
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
num_channels += 2;
for_each_node_by_name(dp, "serial") {
if (of_device_is_compatible(dp, "sab82532"))
num_channels += 2;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
if (num_channels) {
sunsab_ports = kzalloc(sizeof(struct uart_sunsab_port) *
num_channels, GFP_KERNEL);
if (!sunsab_ports)
return -ENOMEM;
[SPARC] serial: Make sure sysfs nodes get named correctly. Because we play this trick where we use ttyS? in increasing minor numbers for different sunfoo.c drivers, we have to inform the TTY layer of this. Do so by setting the tty->name_base appropriately. Probably there should be a generic way to do this in the serial core, but for now... Signed-off-by: David S. Miller <davem@davemloft.net>
2006-02-20 15:39:16 -07:00
[SERIAL] sparc: Infrastructure to fix section mismatch bugs. This patch against 2.6.23 sparc-2.6.git contains a number of minor cleanups of the sparc serial drivers. Initially I fixed this build warning: WARNING: vmlinux.o(.text+0x107a2c): Section mismatch: reference to .init.text:add_preferred_console (between 'sunserial_console_match' and 'sunserial_console_termios') which is done by declaring sunserial_console_match() as __init. This resulted in build warnings on sunserial_current_minor. To resolve these the variable was changed so it is no longer global, and to hide operations on it inside 2 new functions. These functions handle the UART minor handling code that is common to all sparc serial drivers. These changes allowed to clean up the uart counters in all the sparc serial drivers, and the administration of minor device numbers. Lastly, sunserial_console_termios() does not need to be exported since it is only called from non-modular code. Sadly, the following build warning still exists: WARNING: vmlinux.o(__ksymtab+0x2910): Section mismatch: reference to .init.text:sunserial_console_match (between '__ksymtab_sunserial_console_match' and '__ksymtab_sunserial_unregister_minors') This could be resolved by not exporting sunserial_console_match(), but this is not possible at the moment because it is being called from modular code. On the other hand, this is a bogus warning since it comes from a ksymtab section. Signed-off-by: Martin Habets <errandir_news@mph.eclipse.co.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-11 04:37:04 -07:00
err = sunserial_register_minors(&sunsab_reg, num_channels);
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
if (err) {
kfree(sunsab_ports);
sunsab_ports = NULL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
return err;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
return of_register_driver(&sab_driver, &of_bus_type);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
static void __exit sunsab_exit(void)
{
[SERIAL] sunsab: Convert to of_driver framework. Signed-off-by: David S. Miller <davem@davemloft.net>
2006-06-29 15:17:47 -07:00
of_unregister_driver(&sab_driver);
[SERIAL] sparc: Infrastructure to fix section mismatch bugs. This patch against 2.6.23 sparc-2.6.git contains a number of minor cleanups of the sparc serial drivers. Initially I fixed this build warning: WARNING: vmlinux.o(.text+0x107a2c): Section mismatch: reference to .init.text:add_preferred_console (between 'sunserial_console_match' and 'sunserial_console_termios') which is done by declaring sunserial_console_match() as __init. This resulted in build warnings on sunserial_current_minor. To resolve these the variable was changed so it is no longer global, and to hide operations on it inside 2 new functions. These functions handle the UART minor handling code that is common to all sparc serial drivers. These changes allowed to clean up the uart counters in all the sparc serial drivers, and the administration of minor device numbers. Lastly, sunserial_console_termios() does not need to be exported since it is only called from non-modular code. Sadly, the following build warning still exists: WARNING: vmlinux.o(__ksymtab+0x2910): Section mismatch: reference to .init.text:sunserial_console_match (between '__ksymtab_sunserial_console_match' and '__ksymtab_sunserial_unregister_minors') This could be resolved by not exporting sunserial_console_match(), but this is not possible at the moment because it is being called from modular code. On the other hand, this is a bogus warning since it comes from a ksymtab section. Signed-off-by: Martin Habets <errandir_news@mph.eclipse.co.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-11 04:37:04 -07:00
if (sunsab_reg.nr) {
sunserial_unregister_minors(&sunsab_reg, sunsab_reg.nr);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-16 15:20:36 -07:00
}
kfree(sunsab_ports);
sunsab_ports = NULL;
}
module_init(sunsab_init);
module_exit(sunsab_exit);
MODULE_AUTHOR("Eddie C. Dost and David S. Miller");
MODULE_DESCRIPTION("Sun SAB82532 serial port driver");
MODULE_LICENSE("GPL");
Copy Permalink