From e041c683412d5bf44dc2b109053e3b837b71742d Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Mon, 27 Mar 2006 01:16:30 -0800 Subject: [PATCH] Notifier chain update: API changes The kernel's implementation of notifier chains is unsafe. There is no protection against entries being added to or removed from a chain while the chain is in use. The issues were discussed in this thread: http://marc.theaimsgroup.com/?l=linux-kernel&m=113018709002036&w=2 We noticed that notifier chains in the kernel fall into two basic usage classes: "Blocking" chains are always called from a process context and the callout routines are allowed to sleep; "Atomic" chains can be called from an atomic context and the callout routines are not allowed to sleep. We decided to codify this distinction and make it part of the API. Therefore this set of patches introduces three new, parallel APIs: one for blocking notifiers, one for atomic notifiers, and one for "raw" notifiers (which is really just the old API under a new name). New kinds of data structures are used for the heads of the chains, and new routines are defined for registration, unregistration, and calling a chain. The three APIs are explained in include/linux/notifier.h and their implementation is in kernel/sys.c. With atomic and blocking chains, the implementation guarantees that the chain links will not be corrupted and that chain callers will not get messed up by entries being added or removed. For raw chains the implementation provides no guarantees at all; users of this API must provide their own protections. (The idea was that situations may come up where the assumptions of the atomic and blocking APIs are not appropriate, so it should be possible for users to handle these things in their own way.) There are some limitations, which should not be too hard to live with. For atomic/blocking chains, registration and unregistration must always be done in a process context since the chain is protected by a mutex/rwsem. Also, a callout routine for a non-raw chain must not try to register or unregister entries on its own chain. (This did happen in a couple of places and the code had to be changed to avoid it.) Since atomic chains may be called from within an NMI handler, they cannot use spinlocks for synchronization. Instead we use RCU. The overhead falls almost entirely in the unregister routine, which is okay since unregistration is much less frequent that calling a chain. Here is the list of chains that we adjusted and their classifications. None of them use the raw API, so for the moment it is only a placeholder. ATOMIC CHAINS ------------- arch/i386/kernel/traps.c: i386die_chain arch/ia64/kernel/traps.c: ia64die_chain arch/powerpc/kernel/traps.c: powerpc_die_chain arch/sparc64/kernel/traps.c: sparc64die_chain arch/x86_64/kernel/traps.c: die_chain drivers/char/ipmi/ipmi_si_intf.c: xaction_notifier_list kernel/panic.c: panic_notifier_list kernel/profile.c: task_free_notifier net/bluetooth/hci_core.c: hci_notifier net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_chain net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_expect_chain net/ipv6/addrconf.c: inet6addr_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_expect_chain net/netlink/af_netlink.c: netlink_chain BLOCKING CHAINS --------------- arch/powerpc/platforms/pseries/reconfig.c: pSeries_reconfig_chain arch/s390/kernel/process.c: idle_chain arch/x86_64/kernel/process.c idle_notifier drivers/base/memory.c: memory_chain drivers/cpufreq/cpufreq.c cpufreq_policy_notifier_list drivers/cpufreq/cpufreq.c cpufreq_transition_notifier_list drivers/macintosh/adb.c: adb_client_list drivers/macintosh/via-pmu.c sleep_notifier_list drivers/macintosh/via-pmu68k.c sleep_notifier_list drivers/macintosh/windfarm_core.c wf_client_list drivers/usb/core/notify.c usb_notifier_list drivers/video/fbmem.c fb_notifier_list kernel/cpu.c cpu_chain kernel/module.c module_notify_list kernel/profile.c munmap_notifier kernel/profile.c task_exit_notifier kernel/sys.c reboot_notifier_list net/core/dev.c netdev_chain net/decnet/dn_dev.c: dnaddr_chain net/ipv4/devinet.c: inetaddr_chain It's possible that some of these classifications are wrong. If they are, please let us know or submit a patch to fix them. Note that any chain that gets called very frequently should be atomic, because the rwsem read-locking used for blocking chains is very likely to incur cache misses on SMP systems. (However, if the chain's callout routines may sleep then the chain cannot be atomic.) The patch set was written by Alan Stern and Chandra Seetharaman, incorporating material written by Keith Owens and suggestions from Paul McKenney and Andrew Morton. [jes@sgi.com: restructure the notifier chain initialization macros] Signed-off-by: Alan Stern Signed-off-by: Chandra Seetharaman Signed-off-by: Jes Sorensen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/ipmi/ipmi_msghandler.c | 4 ++-- drivers/char/ipmi/ipmi_si_intf.c | 7 ++++--- drivers/char/ipmi/ipmi_watchdog.c | 6 ++++-- 3 files changed, 10 insertions(+), 7 deletions(-) (limited to 'drivers/char') diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index b8fb87c6c29f..40eb005b9d77 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -3744,7 +3744,7 @@ static int ipmi_init_msghandler(void) ipmi_timer.expires = jiffies + IPMI_TIMEOUT_JIFFIES; add_timer(&ipmi_timer); - notifier_chain_register(&panic_notifier_list, &panic_block); + atomic_notifier_chain_register(&panic_notifier_list, &panic_block); initialized = 1; @@ -3764,7 +3764,7 @@ static __exit void cleanup_ipmi(void) if (!initialized) return; - notifier_chain_unregister(&panic_notifier_list, &panic_block); + atomic_notifier_chain_unregister(&panic_notifier_list, &panic_block); /* This can't be called if any interfaces exist, so no worry about shutting down the interfaces. */ diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index 12f858dc9994..35fbd4d8ed4b 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -237,10 +237,10 @@ struct smi_info static int try_smi_init(struct smi_info *smi); -static struct notifier_block *xaction_notifier_list; +static ATOMIC_NOTIFIER_HEAD(xaction_notifier_list); static int register_xaction_notifier(struct notifier_block * nb) { - return notifier_chain_register(&xaction_notifier_list, nb); + return atomic_notifier_chain_register(&xaction_notifier_list, nb); } static void si_restart_short_timer(struct smi_info *smi_info); @@ -302,7 +302,8 @@ static enum si_sm_result start_next_msg(struct smi_info *smi_info) do_gettimeofday(&t); printk("**Start2: %d.%9.9d\n", t.tv_sec, t.tv_usec); #endif - err = notifier_call_chain(&xaction_notifier_list, 0, smi_info); + err = atomic_notifier_call_chain(&xaction_notifier_list, + 0, smi_info); if (err & NOTIFY_STOP_MASK) { rv = SI_SM_CALL_WITHOUT_DELAY; goto out; diff --git a/drivers/char/ipmi/ipmi_watchdog.c b/drivers/char/ipmi/ipmi_watchdog.c index 616539310d9a..7ece9f3c8f70 100644 --- a/drivers/char/ipmi/ipmi_watchdog.c +++ b/drivers/char/ipmi/ipmi_watchdog.c @@ -1158,7 +1158,8 @@ static int __init ipmi_wdog_init(void) } register_reboot_notifier(&wdog_reboot_notifier); - notifier_chain_register(&panic_notifier_list, &wdog_panic_notifier); + atomic_notifier_chain_register(&panic_notifier_list, + &wdog_panic_notifier); printk(KERN_INFO PFX "driver initialized\n"); @@ -1176,7 +1177,8 @@ static __exit void ipmi_unregister_watchdog(void) release_nmi(&ipmi_nmi_handler); #endif - notifier_chain_unregister(&panic_notifier_list, &wdog_panic_notifier); + atomic_notifier_chain_unregister(&panic_notifier_list, + &wdog_panic_notifier); unregister_reboot_notifier(&wdog_reboot_notifier); if (! watchdog_user) -- cgit v1.2.1 From 12b824fb15a37cdcdfa3e92c9e912a07cc6f7a24 Mon Sep 17 00:00:00 2001 From: Alessandro Zummo Date: Mon, 27 Mar 2006 01:16:35 -0800 Subject: [PATCH] RTC subsystem: ARM cleanup This patch removes from the ARM subsytem some of the rtc-related functions that have been included in the RTC subsystem. It also fixes some naming collisions. Signed-off-by: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/char') diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig index facc3f1d9e37..73d30bf01582 100644 --- a/drivers/char/Kconfig +++ b/drivers/char/Kconfig @@ -696,7 +696,7 @@ config NVRAM config RTC tristate "Enhanced Real Time Clock Support" - depends on !PPC && !PARISC && !IA64 && !M68K && (!SPARC || PCI) && !FRV + depends on !PPC && !PARISC && !IA64 && !M68K && (!SPARC || PCI) && !FRV && !ARM ---help--- If you say Y here and create a character special file /dev/rtc with major number 10 and minor number 135 using mknod ("man mknod"), you -- cgit v1.2.1 From 86a34147d1f1c94e94500e63e83f9fa42548a088 Mon Sep 17 00:00:00 2001 From: Paul Fulghum Date: Tue, 28 Mar 2006 01:56:14 -0800 Subject: [PATCH] synclink: remove dead code Remove dead code from synclink driver. This was used previously when the write method had a from_user flag, which has been removed. Signed-off-by: Paul Fulghum Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/synclink.c | 31 +------------------------------ 1 file changed, 1 insertion(+), 30 deletions(-) (limited to 'drivers/char') diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c index d68be61f0a49..abb03e52fe12 100644 --- a/drivers/char/synclink.c +++ b/drivers/char/synclink.c @@ -941,17 +941,6 @@ static void* mgsl_get_text_ptr(void) return mgsl_get_text_ptr; } -/* - * tmp_buf is used as a temporary buffer by mgsl_write. We need to - * lock it in case the COPY_FROM_USER blocks while swapping in a page, - * and some other program tries to do a serial write at the same time. - * Since the lock will only come under contention when the system is - * swapping and available memory is low, it makes sense to share one - * buffer across all the serial ioports, since it significantly saves - * memory if large numbers of serial ports are open. - */ -static unsigned char *tmp_buf; - static inline int mgsl_paranoia_check(struct mgsl_struct *info, char *name, const char *routine) { @@ -2150,7 +2139,7 @@ static int mgsl_write(struct tty_struct * tty, if (mgsl_paranoia_check(info, tty->name, "mgsl_write")) goto cleanup; - if (!tty || !info->xmit_buf || !tmp_buf) + if (!tty || !info->xmit_buf) goto cleanup; if ( info->params.mode == MGSL_MODE_HDLC || @@ -3438,7 +3427,6 @@ static int mgsl_open(struct tty_struct *tty, struct file * filp) { struct mgsl_struct *info; int retval, line; - unsigned long page; unsigned long flags; /* verify range of specified line number */ @@ -3472,18 +3460,6 @@ static int mgsl_open(struct tty_struct *tty, struct file * filp) goto cleanup; } - if (!tmp_buf) { - page = get_zeroed_page(GFP_KERNEL); - if (!page) { - retval = -ENOMEM; - goto cleanup; - } - if (tmp_buf) - free_page(page); - else - tmp_buf = (unsigned char *) page; - } - info->tty->low_latency = (info->flags & ASYNC_LOW_LATENCY) ? 1 : 0; spin_lock_irqsave(&info->netlock, flags); @@ -4502,11 +4478,6 @@ static void synclink_cleanup(void) kfree(tmp); } - if (tmp_buf) { - free_page((unsigned long) tmp_buf); - tmp_buf = NULL; - } - if (pci_registered) pci_unregister_driver(&synclink_pci_driver); } -- cgit v1.2.1 From 0080b7aae88c75e2a6b38dfcb228b0f239e18e3c Mon Sep 17 00:00:00 2001 From: Paul Fulghum Date: Tue, 28 Mar 2006 01:56:15 -0800 Subject: [PATCH] synclink_gt add gpio feature Add driver support for general purpose I/O feature of the Synclink GT adapters. Signed-off-by: Paul Fulghum Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/synclink_gt.c | 246 ++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 242 insertions(+), 4 deletions(-) (limited to 'drivers/char') diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c index 738ec2f4e563..8818042bad7c 100644 --- a/drivers/char/synclink_gt.c +++ b/drivers/char/synclink_gt.c @@ -1,5 +1,5 @@ /* - * $Id: synclink_gt.c,v 4.22 2006/01/09 20:16:06 paulkf Exp $ + * $Id: synclink_gt.c,v 4.25 2006/02/06 21:20:33 paulkf Exp $ * * Device driver for Microgate SyncLink GT serial adapters. * @@ -92,7 +92,7 @@ * module identification */ static char *driver_name = "SyncLink GT"; -static char *driver_version = "$Revision: 4.22 $"; +static char *driver_version = "$Revision: 4.25 $"; static char *tty_driver_name = "synclink_gt"; static char *tty_dev_prefix = "ttySLG"; MODULE_LICENSE("GPL"); @@ -187,6 +187,20 @@ static void hdlcdev_exit(struct slgt_info *info); #define SLGT_MAX_PORTS 4 #define SLGT_REG_SIZE 256 +/* + * conditional wait facility + */ +struct cond_wait { + struct cond_wait *next; + wait_queue_head_t q; + wait_queue_t wait; + unsigned int data; +}; +static void init_cond_wait(struct cond_wait *w, unsigned int data); +static void add_cond_wait(struct cond_wait **head, struct cond_wait *w); +static void remove_cond_wait(struct cond_wait **head, struct cond_wait *w); +static void flush_cond_wait(struct cond_wait **head); + /* * DMA buffer descriptor and access macros */ @@ -269,6 +283,9 @@ struct slgt_info { struct timer_list tx_timer; struct timer_list rx_timer; + unsigned int gpio_present; + struct cond_wait *gpio_wait_q; + spinlock_t lock; /* spinlock for synchronizing with ISR */ struct work_struct task; @@ -379,6 +396,11 @@ static MGSL_PARAMS default_params = { #define MASK_OVERRUN BIT4 #define GSR 0x00 /* global status */ +#define JCR 0x04 /* JTAG control */ +#define IODR 0x08 /* GPIO direction */ +#define IOER 0x0c /* GPIO interrupt enable */ +#define IOVR 0x10 /* GPIO value */ +#define IOSR 0x14 /* GPIO interrupt status */ #define TDR 0x80 /* tx data */ #define RDR 0x80 /* rx data */ #define TCR 0x82 /* tx control */ @@ -503,6 +525,9 @@ static int tiocmset(struct tty_struct *tty, struct file *file, static void set_break(struct tty_struct *tty, int break_state); static int get_interface(struct slgt_info *info, int __user *if_mode); static int set_interface(struct slgt_info *info, int if_mode); +static int set_gpio(struct slgt_info *info, struct gpio_desc __user *gpio); +static int get_gpio(struct slgt_info *info, struct gpio_desc __user *gpio); +static int wait_gpio(struct slgt_info *info, struct gpio_desc __user *gpio); /* * driver functions @@ -1112,6 +1137,12 @@ static int ioctl(struct tty_struct *tty, struct file *file, return get_interface(info, argp); case MGSL_IOCSIF: return set_interface(info,(int)arg); + case MGSL_IOCSGPIO: + return set_gpio(info, argp); + case MGSL_IOCGGPIO: + return get_gpio(info, argp); + case MGSL_IOCWAITGPIO: + return wait_gpio(info, argp); case TIOCGICOUNT: spin_lock_irqsave(&info->lock,flags); cnow = info->icount; @@ -2158,6 +2189,24 @@ static void isr_txeom(struct slgt_info *info, unsigned short status) } } +static void isr_gpio(struct slgt_info *info, unsigned int changed, unsigned int state) +{ + struct cond_wait *w, *prev; + + /* wake processes waiting for specific transitions */ + for (w = info->gpio_wait_q, prev = NULL ; w != NULL ; w = w->next) { + if (w->data & changed) { + w->data = state; + wake_up_interruptible(&w->q); + if (prev != NULL) + prev->next = w->next; + else + info->gpio_wait_q = w->next; + } else + prev = w; + } +} + /* interrupt service routine * * irq interrupt number @@ -2193,6 +2242,22 @@ static irqreturn_t slgt_interrupt(int irq, void *dev_id, struct pt_regs * regs) } } + if (info->gpio_present) { + unsigned int state; + unsigned int changed; + while ((changed = rd_reg32(info, IOSR)) != 0) { + DBGISR(("%s iosr=%08x\n", info->device_name, changed)); + /* read latched state of GPIO signals */ + state = rd_reg32(info, IOVR); + /* clear pending GPIO interrupt bits */ + wr_reg32(info, IOSR, changed); + for (i=0 ; i < info->port_count ; i++) { + if (info->port_array[i] != NULL) + isr_gpio(info->port_array[i], changed, state); + } + } + } + for(i=0; i < info->port_count ; i++) { struct slgt_info *port = info->port_array[i]; @@ -2276,6 +2341,8 @@ static void shutdown(struct slgt_info *info) set_signals(info); } + flush_cond_wait(&info->gpio_wait_q); + spin_unlock_irqrestore(&info->lock,flags); if (info->tty) @@ -2650,6 +2717,175 @@ static int set_interface(struct slgt_info *info, int if_mode) return 0; } +/* + * set general purpose IO pin state and direction + * + * user_gpio fields: + * state each bit indicates a pin state + * smask set bit indicates pin state to set + * dir each bit indicates a pin direction (0=input, 1=output) + * dmask set bit indicates pin direction to set + */ +static int set_gpio(struct slgt_info *info, struct gpio_desc __user *user_gpio) +{ + unsigned long flags; + struct gpio_desc gpio; + __u32 data; + + if (!info->gpio_present) + return -EINVAL; + if (copy_from_user(&gpio, user_gpio, sizeof(gpio))) + return -EFAULT; + DBGINFO(("%s set_gpio state=%08x smask=%08x dir=%08x dmask=%08x\n", + info->device_name, gpio.state, gpio.smask, + gpio.dir, gpio.dmask)); + + spin_lock_irqsave(&info->lock,flags); + if (gpio.dmask) { + data = rd_reg32(info, IODR); + data |= gpio.dmask & gpio.dir; + data &= ~(gpio.dmask & ~gpio.dir); + wr_reg32(info, IODR, data); + } + if (gpio.smask) { + data = rd_reg32(info, IOVR); + data |= gpio.smask & gpio.state; + data &= ~(gpio.smask & ~gpio.state); + wr_reg32(info, IOVR, data); + } + spin_unlock_irqrestore(&info->lock,flags); + + return 0; +} + +/* + * get general purpose IO pin state and direction + */ +static int get_gpio(struct slgt_info *info, struct gpio_desc __user *user_gpio) +{ + struct gpio_desc gpio; + if (!info->gpio_present) + return -EINVAL; + gpio.state = rd_reg32(info, IOVR); + gpio.smask = 0xffffffff; + gpio.dir = rd_reg32(info, IODR); + gpio.dmask = 0xffffffff; + if (copy_to_user(user_gpio, &gpio, sizeof(gpio))) + return -EFAULT; + DBGINFO(("%s get_gpio state=%08x dir=%08x\n", + info->device_name, gpio.state, gpio.dir)); + return 0; +} + +/* + * conditional wait facility + */ +static void init_cond_wait(struct cond_wait *w, unsigned int data) +{ + init_waitqueue_head(&w->q); + init_waitqueue_entry(&w->wait, current); + w->data = data; +} + +static void add_cond_wait(struct cond_wait **head, struct cond_wait *w) +{ + set_current_state(TASK_INTERRUPTIBLE); + add_wait_queue(&w->q, &w->wait); + w->next = *head; + *head = w; +} + +static void remove_cond_wait(struct cond_wait **head, struct cond_wait *cw) +{ + struct cond_wait *w, *prev; + remove_wait_queue(&cw->q, &cw->wait); + set_current_state(TASK_RUNNING); + for (w = *head, prev = NULL ; w != NULL ; prev = w, w = w->next) { + if (w == cw) { + if (prev != NULL) + prev->next = w->next; + else + *head = w->next; + break; + } + } +} + +static void flush_cond_wait(struct cond_wait **head) +{ + while (*head != NULL) { + wake_up_interruptible(&(*head)->q); + *head = (*head)->next; + } +} + +/* + * wait for general purpose I/O pin(s) to enter specified state + * + * user_gpio fields: + * state - bit indicates target pin state + * smask - set bit indicates watched pin + * + * The wait ends when at least one watched pin enters the specified + * state. When 0 (no error) is returned, user_gpio->state is set to the + * state of all GPIO pins when the wait ends. + * + * Note: Each pin may be a dedicated input, dedicated output, or + * configurable input/output. The number and configuration of pins + * varies with the specific adapter model. Only input pins (dedicated + * or configured) can be monitored with this function. + */ +static int wait_gpio(struct slgt_info *info, struct gpio_desc __user *user_gpio) +{ + unsigned long flags; + int rc = 0; + struct gpio_desc gpio; + struct cond_wait wait; + u32 state; + + if (!info->gpio_present) + return -EINVAL; + if (copy_from_user(&gpio, user_gpio, sizeof(gpio))) + return -EFAULT; + DBGINFO(("%s wait_gpio() state=%08x smask=%08x\n", + info->device_name, gpio.state, gpio.smask)); + /* ignore output pins identified by set IODR bit */ + if ((gpio.smask &= ~rd_reg32(info, IODR)) == 0) + return -EINVAL; + init_cond_wait(&wait, gpio.smask); + + spin_lock_irqsave(&info->lock, flags); + /* enable interrupts for watched pins */ + wr_reg32(info, IOER, rd_reg32(info, IOER) | gpio.smask); + /* get current pin states */ + state = rd_reg32(info, IOVR); + + if (gpio.smask & ~(state ^ gpio.state)) { + /* already in target state */ + gpio.state = state; + } else { + /* wait for target state */ + add_cond_wait(&info->gpio_wait_q, &wait); + spin_unlock_irqrestore(&info->lock, flags); + schedule(); + if (signal_pending(current)) + rc = -ERESTARTSYS; + else + gpio.state = wait.data; + spin_lock_irqsave(&info->lock, flags); + remove_cond_wait(&info->gpio_wait_q, &wait); + } + + /* disable all GPIO interrupts if no waiting processes */ + if (info->gpio_wait_q == NULL) + wr_reg32(info, IOER, 0); + spin_unlock_irqrestore(&info->lock,flags); + + if ((rc == 0) && copy_to_user(user_gpio, &gpio, sizeof(gpio))) + rc = -EFAULT; + return rc; +} + static int modem_input_wait(struct slgt_info *info,int arg) { unsigned long flags; @@ -3166,8 +3402,10 @@ static void device_init(int adapter_num, struct pci_dev *pdev) } else { port_array[0]->irq_requested = 1; adapter_test(port_array[0]); - for (i=1 ; i < port_count ; i++) + for (i=1 ; i < port_count ; i++) { port_array[i]->init_error = port_array[0]->init_error; + port_array[i]->gpio_present = port_array[0]->gpio_present; + } } } } @@ -4301,7 +4539,7 @@ static int register_test(struct slgt_info *info) break; } } - + info->gpio_present = (rd_reg32(info, JCR) & BIT5) ? 1 : 0; info->init_error = rc ? 0 : DiagStatus_AddressFailure; return rc; } -- cgit v1.2.1 From be4676e61f969b624185f14d819adef23d45ffc2 Mon Sep 17 00:00:00 2001 From: Paul Fulghum Date: Tue, 28 Mar 2006 01:56:16 -0800 Subject: [PATCH] synclink_gt: remove uneeded async code Remove code in async receive handling that serves no purpose with new tty receive buffering. Previously this code tried to free up receive buffer space, but now does nothing useful while making expensive calls. Signed-off-by: Paul Fulghum Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/synclink_gt.c | 4 ---- 1 file changed, 4 deletions(-) (limited to 'drivers/char') diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c index 8818042bad7c..b4d1f4eea435 100644 --- a/drivers/char/synclink_gt.c +++ b/drivers/char/synclink_gt.c @@ -1793,10 +1793,6 @@ static void rx_async(struct slgt_info *info) DBGDATA(info, p, count, "rx"); for(i=0 ; i < count; i+=2, p+=2) { - if (tty && chars) { - tty_flip_buffer_push(tty); - chars = 0; - } ch = *p; icount->rx++; -- cgit v1.2.1 From 99ac48f54a91d02140c497edc31dc57d4bc5c85d Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Tue, 28 Mar 2006 01:56:41 -0800 Subject: [PATCH] mark f_ops const in the inode Mark the f_ops members of inodes as const, as well as fix the ripple-through this causes by places that copy this f_ops and then "do stuff" with it. Signed-off-by: Arjan van de Ven Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/drm/drm_fops.c | 2 +- drivers/char/drm/i810_dma.c | 2 +- drivers/char/drm/i830_dma.c | 2 +- drivers/char/mem.c | 2 +- drivers/char/misc.c | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-) (limited to 'drivers/char') diff --git a/drivers/char/drm/drm_fops.c b/drivers/char/drm/drm_fops.c index 641f7633878c..b7f7951c4587 100644 --- a/drivers/char/drm/drm_fops.c +++ b/drivers/char/drm/drm_fops.c @@ -175,7 +175,7 @@ int drm_stub_open(struct inode *inode, struct file *filp) drm_device_t *dev = NULL; int minor = iminor(inode); int err = -ENODEV; - struct file_operations *old_fops; + const struct file_operations *old_fops; DRM_DEBUG("\n"); diff --git a/drivers/char/drm/i810_dma.c b/drivers/char/drm/i810_dma.c index ae0aa6d7e0bb..c658dde3633b 100644 --- a/drivers/char/drm/i810_dma.c +++ b/drivers/char/drm/i810_dma.c @@ -126,7 +126,7 @@ static int i810_map_buffer(drm_buf_t * buf, struct file *filp) drm_device_t *dev = priv->head->dev; drm_i810_buf_priv_t *buf_priv = buf->dev_private; drm_i810_private_t *dev_priv = dev->dev_private; - struct file_operations *old_fops; + const struct file_operations *old_fops; int retcode = 0; if (buf_priv->currently_mapped == I810_BUF_MAPPED) diff --git a/drivers/char/drm/i830_dma.c b/drivers/char/drm/i830_dma.c index 163f2cbfe60d..b0f815d8cea8 100644 --- a/drivers/char/drm/i830_dma.c +++ b/drivers/char/drm/i830_dma.c @@ -128,7 +128,7 @@ static int i830_map_buffer(drm_buf_t * buf, struct file *filp) drm_device_t *dev = priv->head->dev; drm_i830_buf_priv_t *buf_priv = buf->dev_private; drm_i830_private_t *dev_priv = dev->dev_private; - struct file_operations *old_fops; + const struct file_operations *old_fops; unsigned long virtual; int retcode = 0; diff --git a/drivers/char/mem.c b/drivers/char/mem.c index 5245ba1649ed..66719f9d294c 100644 --- a/drivers/char/mem.c +++ b/drivers/char/mem.c @@ -899,7 +899,7 @@ static const struct { unsigned int minor; char *name; umode_t mode; - struct file_operations *fops; + const struct file_operations *fops; } devlist[] = { /* list of minor devices */ {1, "mem", S_IRUSR | S_IWUSR | S_IRGRP, &mem_fops}, {2, "kmem", S_IRUSR | S_IWUSR | S_IRGRP, &kmem_fops}, diff --git a/drivers/char/misc.c b/drivers/char/misc.c index 3e4c0414a01a..96eb2a709e21 100644 --- a/drivers/char/misc.c +++ b/drivers/char/misc.c @@ -129,7 +129,7 @@ static int misc_open(struct inode * inode, struct file * file) int minor = iminor(inode); struct miscdevice *c; int err = -ENODEV; - struct file_operations *old_fops, *new_fops = NULL; + const struct file_operations *old_fops, *new_fops = NULL; down(&misc_sem); -- cgit v1.2.1 From 05613bdd8647bfc4535b0dcc8f1b95c8b39be394 Mon Sep 17 00:00:00 2001 From: Marcelo Feitoza Parisi Date: Tue, 28 Mar 2006 01:56:45 -0800 Subject: [PATCH] nvidia-agp: use time_before_eq() It deals with wrapping correctly and is nicer to read. Signed-off-by: Marcelo Feitoza Parisi Signed-off-by: Alexey Dobriyan Cc: Dave Jones Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/agp/nvidia-agp.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'drivers/char') diff --git a/drivers/char/agp/nvidia-agp.c b/drivers/char/agp/nvidia-agp.c index 70b8ed9cd172..4c67135c12d8 100644 --- a/drivers/char/agp/nvidia-agp.c +++ b/drivers/char/agp/nvidia-agp.c @@ -11,6 +11,7 @@ #include #include #include +#include #include "agp.h" /* NVIDIA registers */ @@ -256,7 +257,7 @@ static void nvidia_tlbflush(struct agp_memory *mem) do { pci_read_config_dword(nvidia_private.dev_1, NVIDIA_1_WBC, &wbc_reg); - if ((signed)(end - jiffies) <= 0) { + if (time_before_eq(end, jiffies)) { printk(KERN_ERR PFX "TLB flush took more than 3 seconds.\n"); } -- cgit v1.2.1 From 7f927fcc2fd1575d01efb4b76665975007945690 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 28 Mar 2006 01:56:53 -0800 Subject: [PATCH] Typo fixes Fix a lot of typos. Eyeballed by jmc@ in OpenBSD. Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/mxser.h | 2 +- drivers/char/synclink.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/char') diff --git a/drivers/char/mxser.h b/drivers/char/mxser.h index e7fd0b08e0b7..7e188a4d602a 100644 --- a/drivers/char/mxser.h +++ b/drivers/char/mxser.h @@ -118,7 +118,7 @@ // enable CTS interrupt #define MOXA_MUST_IER_ECTSI 0x80 -// eanble RTS interrupt +// enable RTS interrupt #define MOXA_MUST_IER_ERTSI 0x40 // enable Xon/Xoff interrupt #define MOXA_MUST_IER_XINT 0x20 diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c index abb03e52fe12..fee2aca3f6a5 100644 --- a/drivers/char/synclink.c +++ b/drivers/char/synclink.c @@ -5996,7 +5996,7 @@ static void usc_set_async_mode( struct mgsl_struct *info ) * <15..8> ? RxFIFO IRQ Request Level * * Note: For async mode the receive FIFO level must be set - * to 0 to aviod the situation where the FIFO contains fewer bytes + * to 0 to avoid the situation where the FIFO contains fewer bytes * than the trigger level and no more data is expected. * * <7> 0 Exited Hunt IA (Interrupt Arm) -- cgit v1.2.1