OpenCloudOS-Kernel/drivers/char/snsc.c

467 lines
11 KiB
C
Raw Normal View History

/*
* SN Platform system controller communication support
*
* This file is subject to the terms and conditions of the GNU General Public
* License. See the file "COPYING" in the main directory of this archive
* for more details.
*
* Copyright (C) 2004, 2006 Silicon Graphics, Inc. All rights reserved.
*/
/*
* System controller communication driver
*
* This driver allows a user process to communicate with the system
* controller (a.k.a. "IRouter") network in an SGI SN system.
*/
#include <linux/interrupt.h>
#include <linux/sched.h>
#include <linux/device.h>
#include <linux/poll.h>
#include <linux/module.h>
#include <linux/slab.h>
#include <linux/mutex.h>
#include <asm/sn/io.h>
#include <asm/sn/sn_sal.h>
#include <asm/sn/module.h>
#include <asm/sn/geo.h>
#include <asm/sn/nodepda.h>
#include "snsc.h"
#define SYSCTL_BASENAME "snsc"
#define SCDRV_BUFSZ 2048
#define SCDRV_TIMEOUT 1000
static DEFINE_MUTEX(scdrv_mutex);
static irqreturn_t
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 21:55:46 +08:00
scdrv_interrupt(int irq, void *subch_data)
{
struct subch_data_s *sd = subch_data;
unsigned long flags;
int status;
spin_lock_irqsave(&sd->sd_rlock, flags);
spin_lock(&sd->sd_wlock);
status = ia64_sn_irtr_intr(sd->sd_nasid, sd->sd_subch);
if (status > 0) {
if (status & SAL_IROUTER_INTR_RECV) {
wake_up(&sd->sd_rq);
}
if (status & SAL_IROUTER_INTR_XMIT) {
ia64_sn_irtr_intr_disable
(sd->sd_nasid, sd->sd_subch,
SAL_IROUTER_INTR_XMIT);
wake_up(&sd->sd_wq);
}
}
spin_unlock(&sd->sd_wlock);
spin_unlock_irqrestore(&sd->sd_rlock, flags);
return IRQ_HANDLED;
}
/*
* scdrv_open
*
* Reserve a subchannel for system controller communication.
*/
static int
scdrv_open(struct inode *inode, struct file *file)
{
struct sysctl_data_s *scd;
struct subch_data_s *sd;
int rv;
/* look up device info for this device file */
scd = container_of(inode->i_cdev, struct sysctl_data_s, scd_cdev);
/* allocate memory for subchannel data */
sd = kzalloc(sizeof (struct subch_data_s), GFP_KERNEL);
if (sd == NULL) {
printk("%s: couldn't allocate subchannel data\n",
__func__);
return -ENOMEM;
}
/* initialize subch_data_s fields */
sd->sd_nasid = scd->scd_nasid;
sd->sd_subch = ia64_sn_irtr_open(scd->scd_nasid);
if (sd->sd_subch < 0) {
kfree(sd);
printk("%s: couldn't allocate subchannel\n", __func__);
return -EBUSY;
}
spin_lock_init(&sd->sd_rlock);
spin_lock_init(&sd->sd_wlock);
init_waitqueue_head(&sd->sd_rq);
init_waitqueue_head(&sd->sd_wq);
sema_init(&sd->sd_rbs, 1);
sema_init(&sd->sd_wbs, 1);
file->private_data = sd;
/* hook this subchannel up to the system controller interrupt */
mutex_lock(&scdrv_mutex);
rv = request_irq(SGI_UART_VECTOR, scdrv_interrupt,
IRQF_SHARED | IRQF_DISABLED,
SYSCTL_BASENAME, sd);
if (rv) {
ia64_sn_irtr_close(sd->sd_nasid, sd->sd_subch);
kfree(sd);
printk("%s: irq request failed (%d)\n", __func__, rv);
mutex_unlock(&scdrv_mutex);
return -EBUSY;
}
mutex_unlock(&scdrv_mutex);
return 0;
}
/*
* scdrv_release
*
* Release a previously-reserved subchannel.
*/
static int
scdrv_release(struct inode *inode, struct file *file)
{
struct subch_data_s *sd = (struct subch_data_s *) file->private_data;
int rv;
/* free the interrupt */
free_irq(SGI_UART_VECTOR, sd);
/* ask SAL to close the subchannel */
rv = ia64_sn_irtr_close(sd->sd_nasid, sd->sd_subch);
kfree(sd);
return rv;
}
/*
* scdrv_read
*
* Called to read bytes from the open IRouter pipe.
*
*/
static inline int
read_status_check(struct subch_data_s *sd, int *len)
{
return ia64_sn_irtr_recv(sd->sd_nasid, sd->sd_subch, sd->sd_rb, len);
}
static ssize_t
scdrv_read(struct file *file, char __user *buf, size_t count, loff_t *f_pos)
{
int status;
int len;
unsigned long flags;
struct subch_data_s *sd = (struct subch_data_s *) file->private_data;
/* try to get control of the read buffer */
if (down_trylock(&sd->sd_rbs)) {
/* somebody else has it now;
* if we're non-blocking, then exit...
*/
if (file->f_flags & O_NONBLOCK) {
return -EAGAIN;
}
/* ...or if we want to block, then do so here */
if (down_interruptible(&sd->sd_rbs)) {
/* something went wrong with wait */
return -ERESTARTSYS;
}
}
/* anything to read? */
len = CHUNKSIZE;
spin_lock_irqsave(&sd->sd_rlock, flags);
status = read_status_check(sd, &len);
/* if not, and we're blocking I/O, loop */
while (status < 0) {
DECLARE_WAITQUEUE(wait, current);
if (file->f_flags & O_NONBLOCK) {
spin_unlock_irqrestore(&sd->sd_rlock, flags);
up(&sd->sd_rbs);
return -EAGAIN;
}
len = CHUNKSIZE;
set_current_state(TASK_INTERRUPTIBLE);
add_wait_queue(&sd->sd_rq, &wait);
spin_unlock_irqrestore(&sd->sd_rlock, flags);
schedule_timeout(SCDRV_TIMEOUT);
remove_wait_queue(&sd->sd_rq, &wait);
if (signal_pending(current)) {
/* wait was interrupted */
up(&sd->sd_rbs);
return -ERESTARTSYS;
}
spin_lock_irqsave(&sd->sd_rlock, flags);
status = read_status_check(sd, &len);
}
spin_unlock_irqrestore(&sd->sd_rlock, flags);
if (len > 0) {
/* we read something in the last read_status_check(); copy
* it out to user space
*/
if (count < len) {
pr_debug("%s: only accepting %d of %d bytes\n",
__func__, (int) count, len);
}
len = min((int) count, len);
if (copy_to_user(buf, sd->sd_rb, len))
len = -EFAULT;
}
/* release the read buffer and wake anyone who might be
* waiting for it
*/
up(&sd->sd_rbs);
/* return the number of characters read in */
return len;
}
/*
* scdrv_write
*
* Writes a chunk of an IRouter packet (or other system controller data)
* to the system controller.
*
*/
static inline int
write_status_check(struct subch_data_s *sd, int count)
{
return ia64_sn_irtr_send(sd->sd_nasid, sd->sd_subch, sd->sd_wb, count);
}
static ssize_t
scdrv_write(struct file *file, const char __user *buf,
size_t count, loff_t *f_pos)
{
unsigned long flags;
int status;
struct subch_data_s *sd = (struct subch_data_s *) file->private_data;
/* try to get control of the write buffer */
if (down_trylock(&sd->sd_wbs)) {
/* somebody else has it now;
* if we're non-blocking, then exit...
*/
if (file->f_flags & O_NONBLOCK) {
return -EAGAIN;
}
/* ...or if we want to block, then do so here */
if (down_interruptible(&sd->sd_wbs)) {
/* something went wrong with wait */
return -ERESTARTSYS;
}
}
count = min((int) count, CHUNKSIZE);
if (copy_from_user(sd->sd_wb, buf, count)) {
up(&sd->sd_wbs);
return -EFAULT;
}
/* try to send the buffer */
spin_lock_irqsave(&sd->sd_wlock, flags);
status = write_status_check(sd, count);
/* if we failed, and we want to block, then loop */
while (status <= 0) {
DECLARE_WAITQUEUE(wait, current);
if (file->f_flags & O_NONBLOCK) {
spin_unlock(&sd->sd_wlock);
up(&sd->sd_wbs);
return -EAGAIN;
}
set_current_state(TASK_INTERRUPTIBLE);
add_wait_queue(&sd->sd_wq, &wait);
spin_unlock_irqrestore(&sd->sd_wlock, flags);
schedule_timeout(SCDRV_TIMEOUT);
remove_wait_queue(&sd->sd_wq, &wait);
if (signal_pending(current)) {
/* wait was interrupted */
up(&sd->sd_wbs);
return -ERESTARTSYS;
}
spin_lock_irqsave(&sd->sd_wlock, flags);
status = write_status_check(sd, count);
}
spin_unlock_irqrestore(&sd->sd_wlock, flags);
/* release the write buffer and wake anyone who's waiting for it */
up(&sd->sd_wbs);
/* return the number of characters accepted (should be the complete
* "chunk" as requested)
*/
if ((status >= 0) && (status < count)) {
pr_debug("Didn't accept the full chunk; %d of %d\n",
status, (int) count);
}
return status;
}
static unsigned int
scdrv_poll(struct file *file, struct poll_table_struct *wait)
{
unsigned int mask = 0;
int status = 0;
struct subch_data_s *sd = (struct subch_data_s *) file->private_data;
unsigned long flags;
poll_wait(file, &sd->sd_rq, wait);
poll_wait(file, &sd->sd_wq, wait);
spin_lock_irqsave(&sd->sd_rlock, flags);
spin_lock(&sd->sd_wlock);
status = ia64_sn_irtr_intr(sd->sd_nasid, sd->sd_subch);
spin_unlock(&sd->sd_wlock);
spin_unlock_irqrestore(&sd->sd_rlock, flags);
if (status > 0) {
if (status & SAL_IROUTER_INTR_RECV) {
mask |= POLLIN | POLLRDNORM;
}
if (status & SAL_IROUTER_INTR_XMIT) {
mask |= POLLOUT | POLLWRNORM;
}
}
return mask;
}
static const struct file_operations scdrv_fops = {
.owner = THIS_MODULE,
.read = scdrv_read,
.write = scdrv_write,
.poll = scdrv_poll,
.open = scdrv_open,
.release = scdrv_release,
llseek: automatically add .llseek fop All file_operations should get a .llseek operation so we can make nonseekable_open the default for future file operations without a .llseek pointer. The three cases that we can automatically detect are no_llseek, seq_lseek and default_llseek. For cases where we can we can automatically prove that the file offset is always ignored, we use noop_llseek, which maintains the current behavior of not returning an error from a seek. New drivers should normally not use noop_llseek but instead use no_llseek and call nonseekable_open at open time. Existing drivers can be converted to do the same when the maintainer knows for certain that no user code relies on calling seek on the device file. The generated code is often incorrectly indented and right now contains comments that clarify for each added line why a specific variant was chosen. In the version that gets submitted upstream, the comments will be gone and I will manually fix the indentation, because there does not seem to be a way to do that using coccinelle. Some amount of new code is currently sitting in linux-next that should get the same modifications, which I will do at the end of the merge window. Many thanks to Julia Lawall for helping me learn to write a semantic patch that does all this. ===== begin semantic patch ===== // This adds an llseek= method to all file operations, // as a preparation for making no_llseek the default. // // The rules are // - use no_llseek explicitly if we do nonseekable_open // - use seq_lseek for sequential files // - use default_llseek if we know we access f_pos // - use noop_llseek if we know we don't access f_pos, // but we still want to allow users to call lseek // @ open1 exists @ identifier nested_open; @@ nested_open(...) { <+... nonseekable_open(...) ...+> } @ open exists@ identifier open_f; identifier i, f; identifier open1.nested_open; @@ int open_f(struct inode *i, struct file *f) { <+... ( nonseekable_open(...) | nested_open(...) ) ...+> } @ read disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ read_no_fpos disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { ... when != off } @ write @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ write_no_fpos @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { ... when != off } @ fops0 @ identifier fops; @@ struct file_operations fops = { ... }; @ has_llseek depends on fops0 @ identifier fops0.fops; identifier llseek_f; @@ struct file_operations fops = { ... .llseek = llseek_f, ... }; @ has_read depends on fops0 @ identifier fops0.fops; identifier read_f; @@ struct file_operations fops = { ... .read = read_f, ... }; @ has_write depends on fops0 @ identifier fops0.fops; identifier write_f; @@ struct file_operations fops = { ... .write = write_f, ... }; @ has_open depends on fops0 @ identifier fops0.fops; identifier open_f; @@ struct file_operations fops = { ... .open = open_f, ... }; // use no_llseek if we call nonseekable_open //////////////////////////////////////////// @ nonseekable1 depends on !has_llseek && has_open @ identifier fops0.fops; identifier nso ~= "nonseekable_open"; @@ struct file_operations fops = { ... .open = nso, ... +.llseek = no_llseek, /* nonseekable */ }; @ nonseekable2 depends on !has_llseek @ identifier fops0.fops; identifier open.open_f; @@ struct file_operations fops = { ... .open = open_f, ... +.llseek = no_llseek, /* open uses nonseekable */ }; // use seq_lseek for sequential files ///////////////////////////////////// @ seq depends on !has_llseek @ identifier fops0.fops; identifier sr ~= "seq_read"; @@ struct file_operations fops = { ... .read = sr, ... +.llseek = seq_lseek, /* we have seq_read */ }; // use default_llseek if there is a readdir /////////////////////////////////////////// @ fops1 depends on !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier readdir_e; @@ // any other fop is used that changes pos struct file_operations fops = { ... .readdir = readdir_e, ... +.llseek = default_llseek, /* readdir is present */ }; // use default_llseek if at least one of read/write touches f_pos ///////////////////////////////////////////////////////////////// @ fops2 depends on !fops1 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read.read_f; @@ // read fops use offset struct file_operations fops = { ... .read = read_f, ... +.llseek = default_llseek, /* read accesses f_pos */ }; @ fops3 depends on !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, ... + .llseek = default_llseek, /* write accesses f_pos */ }; // Use noop_llseek if neither read nor write accesses f_pos /////////////////////////////////////////////////////////// @ fops4 depends on !fops1 && !fops2 && !fops3 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; identifier write_no_fpos.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, .read = read_f, ... +.llseek = noop_llseek, /* read and write both use no f_pos */ }; @ depends on has_write && !has_read && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write_no_fpos.write_f; @@ struct file_operations fops = { ... .write = write_f, ... +.llseek = noop_llseek, /* write uses no f_pos */ }; @ depends on has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; @@ struct file_operations fops = { ... .read = read_f, ... +.llseek = noop_llseek, /* read uses no f_pos */ }; @ depends on !has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; @@ struct file_operations fops = { ... +.llseek = noop_llseek, /* no read or write fn */ }; ===== End semantic patch ===== Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Julia Lawall <julia@diku.dk> Cc: Christoph Hellwig <hch@infradead.org>
2010-08-16 00:52:59 +08:00
.llseek = noop_llseek,
};
static struct class *snsc_class;
/*
* scdrv_init
*
* Called at boot time to initialize the system controller communication
* facility.
*/
int __init
scdrv_init(void)
{
geoid_t geoid;
cnodeid_t cnode;
char devname[32];
char *devnamep;
struct sysctl_data_s *scd;
void *salbuf;
dev_t first_dev, dev;
nasid_t event_nasid;
if (!ia64_platform_is("sn2"))
return -ENODEV;
event_nasid = ia64_sn_get_console_nasid();
if (alloc_chrdev_region(&first_dev, 0, num_cnodes,
SYSCTL_BASENAME) < 0) {
printk("%s: failed to register SN system controller device\n",
__func__);
return -ENODEV;
}
snsc_class = class_create(THIS_MODULE, SYSCTL_BASENAME);
for (cnode = 0; cnode < num_cnodes; cnode++) {
geoid = cnodeid_get_geoid(cnode);
devnamep = devname;
format_module_id(devnamep, geo_module(geoid),
MODULE_FORMAT_BRIEF);
devnamep = devname + strlen(devname);
sprintf(devnamep, "^%d#%d", geo_slot(geoid),
geo_slab(geoid));
/* allocate sysctl device data */
scd = kzalloc(sizeof (struct sysctl_data_s),
GFP_KERNEL);
if (!scd) {
printk("%s: failed to allocate device info"
"for %s/%s\n", __func__,
SYSCTL_BASENAME, devname);
continue;
}
/* initialize sysctl device data fields */
scd->scd_nasid = cnodeid_to_nasid(cnode);
if (!(salbuf = kmalloc(SCDRV_BUFSZ, GFP_KERNEL))) {
printk("%s: failed to allocate driver buffer"
"(%s%s)\n", __func__,
SYSCTL_BASENAME, devname);
kfree(scd);
continue;
}
if (ia64_sn_irtr_init(scd->scd_nasid, salbuf,
SCDRV_BUFSZ) < 0) {
printk
("%s: failed to initialize SAL for"
" system controller communication"
" (%s/%s): outdated PROM?\n",
__func__, SYSCTL_BASENAME, devname);
kfree(scd);
kfree(salbuf);
continue;
}
dev = first_dev + cnode;
cdev_init(&scd->scd_cdev, &scdrv_fops);
if (cdev_add(&scd->scd_cdev, dev, 1)) {
printk("%s: failed to register system"
" controller device (%s%s)\n",
__func__, SYSCTL_BASENAME, devname);
kfree(scd);
kfree(salbuf);
continue;
}
device_create(snsc_class, NULL, dev, NULL,
"%s", devname);
ia64_sn_irtr_intr_enable(scd->scd_nasid,
0 /*ignored */ ,
SAL_IROUTER_INTR_RECV);
/* on the console nasid, prepare to receive
* system controller environmental events
*/
if(scd->scd_nasid == event_nasid) {
scdrv_event_init(scd);
}
}
return 0;
}
module_init(scdrv_init);