diff options
author | Linus Torvalds <torvalds@g5.osdl.org> | 2006-03-25 19:41:09 +0300 |
---|---|---|
committer | Linus Torvalds <torvalds@g5.osdl.org> | 2006-03-25 19:41:09 +0300 |
commit | 1e8c573933fd7975679766850252ad08667e5ca4 (patch) | |
tree | 9600d0c7ee5ea8925f3c4dc30680c819e0363805 | |
parent | d71eecf3b8e893757cc3dec560c96a32ac090890 (diff) | |
parent | 232443e2c90cc2930624dec89df327615b002c55 (diff) | |
download | linux-1e8c573933fd7975679766850252ad08667e5ca4.tar.xz |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/bunk/trivial
* git://git.kernel.org/pub/scm/linux/kernel/git/bunk/trivial: (21 commits)
BUG_ON() Conversion in drivers/video/
BUG_ON() Conversion in drivers/parisc/
BUG_ON() Conversion in drivers/block/
BUG_ON() Conversion in sound/sparc/cs4231.c
BUG_ON() Conversion in drivers/s390/block/dasd.c
BUG_ON() Conversion in lib/swiotlb.c
BUG_ON() Conversion in kernel/cpu.c
BUG_ON() Conversion in ipc/msg.c
BUG_ON() Conversion in block/elevator.c
BUG_ON() Conversion in fs/coda/
BUG_ON() Conversion in fs/binfmt_elf_fdpic.c
BUG_ON() Conversion in input/serio/hil_mlc.c
BUG_ON() Conversion in md/dm-hw-handler.c
BUG_ON() Conversion in md/bitmap.c
The comment describing how MS_ASYNC works in msync.c is confusing
rcu: undeclared variable used in documentation
fix typos "wich" -> "which"
typo patch for fs/ufs/super.c
Fix simple typos
tabify drivers/char/Makefile
...
39 files changed, 116 insertions, 142 deletions
diff --git a/Documentation/RCU/whatisRCU.txt b/Documentation/RCU/whatisRCU.txt index 5ed85af88789..b4ea51ad3610 100644 --- a/Documentation/RCU/whatisRCU.txt +++ b/Documentation/RCU/whatisRCU.txt @@ -360,7 +360,7 @@ uses of RCU may be found in listRCU.txt, arrayRCU.txt, and NMI-RCU.txt. struct foo *new_fp; struct foo *old_fp; - new_fp = kmalloc(sizeof(*fp), GFP_KERNEL); + new_fp = kmalloc(sizeof(*new_fp), GFP_KERNEL); spin_lock(&foo_mutex); old_fp = gbl_foo; *new_fp = *old_fp; @@ -461,7 +461,7 @@ The foo_update_a() function might then be written as follows: struct foo *new_fp; struct foo *old_fp; - new_fp = kmalloc(sizeof(*fp), GFP_KERNEL); + new_fp = kmalloc(sizeof(*new_fp), GFP_KERNEL); spin_lock(&foo_mutex); old_fp = gbl_foo; *new_fp = *old_fp; diff --git a/Documentation/arm/Booting b/Documentation/arm/Booting index fad566bb02fc..76850295af8f 100644 --- a/Documentation/arm/Booting +++ b/Documentation/arm/Booting @@ -118,7 +118,7 @@ to store page tables. The recommended placement is 32KiB into RAM. In either case, the following conditions must be met: -- Quiesce all DMA capable devicess so that memory does not get +- Quiesce all DMA capable devices so that memory does not get corrupted by bogus network packets or disk data. This will save you many hours of debug. diff --git a/Documentation/arm/README b/Documentation/arm/README index 5ed6f3530b86..9b9c8226fdc4 100644 --- a/Documentation/arm/README +++ b/Documentation/arm/README @@ -89,7 +89,7 @@ Modules Although modularisation is supported (and required for the FP emulator), each module on an ARM2/ARM250/ARM3 machine when is loaded will take memory up to the next 32k boundary due to the size of the pages. - Therefore, modularisation on these machines really worth it? + Therefore, is modularisation on these machines really worth it? However, ARM6 and up machines allow modules to take multiples of 4k, and as such Acorn RiscPCs and other architectures using these processors can diff --git a/Documentation/arm/Setup b/Documentation/arm/Setup index 0abd0720d7ed..0cb1e64bde80 100644 --- a/Documentation/arm/Setup +++ b/Documentation/arm/Setup @@ -58,7 +58,7 @@ below: video_y This describes the character position of cursor on VGA console, and - is otherwise unused. (should not used for other console types, and + is otherwise unused. (should not be used for other console types, and should not be used for other purposes). memc_control_reg diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt index 944cf109a6f5..99902ae6804e 100644 --- a/Documentation/filesystems/proc.txt +++ b/Documentation/filesystems/proc.txt @@ -121,7 +121,7 @@ Table 1-1: Process specific entries in /proc .............................................................................. File Content cmdline Command line arguments - cpu Current and last cpu in wich it was executed (2.4)(smp) + cpu Current and last cpu in which it was executed (2.4)(smp) cwd Link to the current working directory environ Values of environment variables exe Link to the executable of this process @@ -309,13 +309,13 @@ is the same by default: > cat /proc/irq/0/smp_affinity ffffffff -It's a bitmask, in wich you can specify wich CPUs can handle the IRQ, you can +It's a bitmask, in which you can specify which CPUs can handle the IRQ, you can set it by doing: > echo 1 > /proc/irq/prof_cpu_mask This means that only the first CPU will handle the IRQ, but you can also echo 5 -wich means that only the first and fourth CPU can handle the IRQ. +which means that only the first and fourth CPU can handle the IRQ. The way IRQs are routed is handled by the IO-APIC, and it's Round Robin between all the CPUs which are allowed to handle it. As usual the kernel has diff --git a/Documentation/networking/packet_mmap.txt b/Documentation/networking/packet_mmap.txt index 8d4cf78258e4..4fc8e9874320 100644 --- a/Documentation/networking/packet_mmap.txt +++ b/Documentation/networking/packet_mmap.txt @@ -40,7 +40,7 @@ network interface card supports some sort of interrupt load mitigation or + How to use CONFIG_PACKET_MMAP -------------------------------------------------------------------------------- -From the user standpoint, you should use the higher level libpcap library, wich +From the user standpoint, you should use the higher level libpcap library, which is a de facto standard, portable across nearly all operating systems including Win32. @@ -217,8 +217,8 @@ called pg_vec, its size limits the number of blocks that can be allocated. kmalloc allocates any number of bytes of phisically contiguous memory from a pool of pre-determined sizes. This pool of memory is mantained by the slab -allocator wich is at the end the responsible for doing the allocation and -hence wich imposes the maximum memory that kmalloc can allocate. +allocator which is at the end the responsible for doing the allocation and +hence which imposes the maximum memory that kmalloc can allocate. In a 2.4/2.6 kernel and the i386 architecture, the limit is 131072 bytes. The predetermined sizes that kmalloc uses can be checked in the "size-<bytes>" @@ -254,7 +254,7 @@ and, the number of frames be <block number> * <block size> / <frame size> -Suposse the following parameters, wich apply for 2.6 kernel and an +Suposse the following parameters, which apply for 2.6 kernel and an i386 architecture: <size-max> = 131072 bytes @@ -360,7 +360,7 @@ TP_STATUS_LOSING : indicates there were packet drops from last time statistics where checked with getsockopt() and the PACKET_STATISTICS option. -TP_STATUS_CSUMNOTREADY: currently it's used for outgoing IP packets wich +TP_STATUS_CSUMNOTREADY: currently it's used for outgoing IP packets which it's checksum will be done in hardware. So while reading the packet we should not try to check the checksum. diff --git a/MAINTAINERS b/MAINTAINERS index fa92d0d3674b..4e8fbbc5566d 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2025,12 +2025,6 @@ L: parisc-linux@parisc-linux.org W: http://www.parisc-linux.org/ S: Maintained -PERSONALITY HANDLING -P: Christoph Hellwig -M: hch@infradead.org -L: linux-abi-devel@lists.sourceforge.net -S: Maintained - PCI ERROR RECOVERY P: Linas Vepstas M: linas@austin.ibm.com @@ -2081,6 +2075,12 @@ M: tsbogend@alpha.franken.de L: netdev@vger.kernel.org S: Maintained +PERSONALITY HANDLING +P: Christoph Hellwig +M: hch@infradead.org +L: linux-abi-devel@lists.sourceforge.net +S: Maintained + PHRAM MTD DRIVER P: Jörn Engel M: joern@wh.fh-wedel.de diff --git a/block/elevator.c b/block/elevator.c index 5e558c4689a4..56c2ed06a9e2 100644 --- a/block/elevator.c +++ b/block/elevator.c @@ -728,8 +728,7 @@ void elv_unregister_queue(struct request_queue *q) int elv_register(struct elevator_type *e) { spin_lock_irq(&elv_list_lock); - if (elevator_find(e->elevator_name)) - BUG(); + BUG_ON(elevator_find(e->elevator_name)); list_add_tail(&e->list, &elv_list); spin_unlock_irq(&elv_list_lock); diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig index 82710ae39228..5cb96300eb0f 100644 --- a/drivers/acpi/Kconfig +++ b/drivers/acpi/Kconfig @@ -256,7 +256,8 @@ config ACPI_CUSTOM_DSDT_FILE depends on ACPI_CUSTOM_DSDT default "" help - Enter the full path name to the file wich includes the AmlCode declaration. + Enter the full path name to the file which includes the AmlCode + declaration. config ACPI_BLACKLIST_YEAR int "Disable ACPI for systems before Jan 1st this year" if X86_32 diff --git a/drivers/block/DAC960.c b/drivers/block/DAC960.c index 37b8cda3e8bc..9bdea2a5cf0e 100644 --- a/drivers/block/DAC960.c +++ b/drivers/block/DAC960.c @@ -228,8 +228,7 @@ static void *slice_dma_loaf(struct dma_loaf *loaf, size_t len, void *cpu_end = loaf->cpu_free + len; void *cpu_addr = loaf->cpu_free; - if (cpu_end > loaf->cpu_base + loaf->length) - BUG(); + BUG_ON(cpu_end > loaf->cpu_base + loaf->length); *dma_handle = loaf->dma_free; loaf->cpu_free = cpu_end; loaf->dma_free += len; diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c index 1f2890989b56..71ec9e664383 100644 --- a/drivers/block/cciss.c +++ b/drivers/block/cciss.c @@ -2361,8 +2361,7 @@ queue: if (!creq) goto startio; - if (creq->nr_phys_segments > MAXSGENTRIES) - BUG(); + BUG_ON(creq->nr_phys_segments > MAXSGENTRIES); if (( c = cmd_alloc(h, 1)) == NULL) goto full; diff --git a/drivers/block/cciss_scsi.c b/drivers/block/cciss_scsi.c index 9e35de05d5c5..0e66e904bd8c 100644 --- a/drivers/block/cciss_scsi.c +++ b/drivers/block/cciss_scsi.c @@ -1316,7 +1316,7 @@ cciss_scsi_queue_command (struct scsi_cmnd *cmd, void (* done)(struct scsi_cmnd cp->Request.Timeout = 0; memset(cp->Request.CDB, 0, sizeof(cp->Request.CDB)); - if (cmd->cmd_len > sizeof(cp->Request.CDB)) BUG(); + BUG_ON(cmd->cmd_len > sizeof(cp->Request.CDB)); cp->Request.CDBLen = cmd->cmd_len; memcpy(cp->Request.CDB, cmd->cmnd, cmd->cmd_len); cp->Request.Type.Type = TYPE_CMD; diff --git a/drivers/block/cpqarray.c b/drivers/block/cpqarray.c index 862b9abac0ae..b6ea2f0c7276 100644 --- a/drivers/block/cpqarray.c +++ b/drivers/block/cpqarray.c @@ -906,8 +906,7 @@ queue_next: if (!creq) goto startio; - if (creq->nr_phys_segments > SG_MAX) - BUG(); + BUG_ON(creq->nr_phys_segments > SG_MAX); if ((c = cmd_alloc(h,1)) == NULL) goto startio; diff --git a/drivers/char/Makefile b/drivers/char/Makefile index 090d154098bb..b2a11245fa95 100644 --- a/drivers/char/Makefile +++ b/drivers/char/Makefile @@ -45,56 +45,57 @@ obj-$(CONFIG_HVC_CONSOLE) += hvc_console.o hvc_vio.o hvsi.o obj-$(CONFIG_RAW_DRIVER) += raw.o obj-$(CONFIG_SGI_SNSC) += snsc.o snsc_event.o obj-$(CONFIG_MMTIMER) += mmtimer.o -obj-$(CONFIG_VIOCONS) += viocons.o +obj-$(CONFIG_VIOCONS) += viocons.o obj-$(CONFIG_VIOTAPE) += viotape.o obj-$(CONFIG_HVCS) += hvcs.o obj-$(CONFIG_SGI_MBCS) += mbcs.o -obj-$(CONFIG_PRINTER) += lp.o -obj-$(CONFIG_TIPAR) += tipar.o - -obj-$(CONFIG_DTLK) += dtlk.o -obj-$(CONFIG_R3964) += n_r3964.o -obj-$(CONFIG_APPLICOM) += applicom.o -obj-$(CONFIG_SONYPI) += sonypi.o -obj-$(CONFIG_RTC) += rtc.o -obj-$(CONFIG_HPET) += hpet.o -obj-$(CONFIG_GEN_RTC) += genrtc.o -obj-$(CONFIG_EFI_RTC) += efirtc.o -obj-$(CONFIG_SGI_DS1286) += ds1286.o -obj-$(CONFIG_SGI_IP27_RTC) += ip27-rtc.o -obj-$(CONFIG_DS1302) += ds1302.o -obj-$(CONFIG_S3C2410_RTC) += s3c2410-rtc.o -obj-$(CONFIG_RTC_VR41XX) += vr41xx_rtc.o +obj-$(CONFIG_PRINTER) += lp.o +obj-$(CONFIG_TIPAR) += tipar.o + +obj-$(CONFIG_DTLK) += dtlk.o +obj-$(CONFIG_R3964) += n_r3964.o +obj-$(CONFIG_APPLICOM) += applicom.o +obj-$(CONFIG_SONYPI) += sonypi.o +obj-$(CONFIG_RTC) += rtc.o +obj-$(CONFIG_HPET) += hpet.o +obj-$(CONFIG_GEN_RTC) += genrtc.o +obj-$(CONFIG_EFI_RTC) += efirtc.o +obj-$(CONFIG_SGI_DS1286) += ds1286.o +obj-$(CONFIG_SGI_IP27_RTC) += ip27-rtc.o +obj-$(CONFIG_DS1302) += ds1302.o +obj-$(CONFIG_S3C2410_RTC) += s3c2410-rtc.o +obj-$(CONFIG_RTC_VR41XX) += vr41xx_rtc.o ifeq ($(CONFIG_GENERIC_NVRAM),y) - obj-$(CONFIG_NVRAM) += generic_nvram.o + obj-$(CONFIG_NVRAM) += generic_nvram.o else - obj-$(CONFIG_NVRAM) += nvram.o + obj-$(CONFIG_NVRAM) += nvram.o endif -obj-$(CONFIG_TOSHIBA) += toshiba.o -obj-$(CONFIG_I8K) += i8k.o -obj-$(CONFIG_DS1620) += ds1620.o -obj-$(CONFIG_HW_RANDOM) += hw_random.o -obj-$(CONFIG_FTAPE) += ftape/ -obj-$(CONFIG_COBALT_LCD) += lcd.o -obj-$(CONFIG_PPDEV) += ppdev.o -obj-$(CONFIG_NWBUTTON) += nwbutton.o -obj-$(CONFIG_NWFLASH) += nwflash.o -obj-$(CONFIG_SCx200_GPIO) += scx200_gpio.o -obj-$(CONFIG_CS5535_GPIO) += cs5535_gpio.o -obj-$(CONFIG_GPIO_VR41XX) += vr41xx_giu.o -obj-$(CONFIG_TANBAC_TB0219) += tb0219.o -obj-$(CONFIG_TELCLOCK) += tlclk.o - -obj-$(CONFIG_WATCHDOG) += watchdog/ -obj-$(CONFIG_MWAVE) += mwave/ -obj-$(CONFIG_AGP) += agp/ -obj-$(CONFIG_DRM) += drm/ -obj-$(CONFIG_PCMCIA) += pcmcia/ -obj-$(CONFIG_IPMI_HANDLER) += ipmi/ - -obj-$(CONFIG_HANGCHECK_TIMER) += hangcheck-timer.o -obj-$(CONFIG_TCG_TPM) += tpm/ +obj-$(CONFIG_TOSHIBA) += toshiba.o +obj-$(CONFIG_I8K) += i8k.o +obj-$(CONFIG_DS1620) += ds1620.o +obj-$(CONFIG_HW_RANDOM) += hw_random.o +obj-$(CONFIG_FTAPE) += ftape/ +obj-$(CONFIG_COBALT_LCD) += lcd.o +obj-$(CONFIG_PPDEV) += ppdev.o +obj-$(CONFIG_NWBUTTON) += nwbutton.o +obj-$(CONFIG_NWFLASH) += nwflash.o +obj-$(CONFIG_SCx200_GPIO) += scx200_gpio.o +obj-$(CONFIG_CS5535_GPIO) += cs5535_gpio.o +obj-$(CONFIG_GPIO_VR41XX) += vr41xx_giu.o +obj-$(CONFIG_TANBAC_TB0219) += tb0219.o +obj-$(CONFIG_TELCLOCK) += tlclk.o + +obj-$(CONFIG_WATCHDOG) += watchdog/ +obj-$(CONFIG_MWAVE) += mwave/ +obj-$(CONFIG_AGP) += agp/ +obj-$(CONFIG_DRM) += drm/ +obj-$(CONFIG_PCMCIA) += pcmcia/ +obj-$(CONFIG_IPMI_HANDLER) += ipmi/ + +obj-$(CONFIG_HANGCHECK_TIMER) += hangcheck-timer.o +obj-$(CONFIG_TCG_TPM) += tpm/ + # Files generated that shall be removed upon make clean clean-files := consolemap_deftbl.c defkeymap.c qtronixmap.c diff --git a/drivers/input/serio/hil_mlc.c b/drivers/input/serio/hil_mlc.c index 5704204964a3..ea499783fb12 100644 --- a/drivers/input/serio/hil_mlc.c +++ b/drivers/input/serio/hil_mlc.c @@ -556,7 +556,7 @@ static inline void hilse_setup_input(hil_mlc *mlc, struct hilse_node *node) { do_gettimeofday(&(mlc->instart)); mlc->icount = 15; memset(mlc->ipacket, 0, 16 * sizeof(hil_packet)); - if (down_trylock(&(mlc->isem))) BUG(); + BUG_ON(down_trylock(&(mlc->isem))); return; } diff --git a/drivers/isdn/i4l/isdn_x25iface.c b/drivers/isdn/i4l/isdn_x25iface.c index edf14a2aa3c8..743ac4077f35 100644 --- a/drivers/isdn/i4l/isdn_x25iface.c +++ b/drivers/isdn/i4l/isdn_x25iface.c @@ -7,7 +7,7 @@ * * stuff needed to support the Linux X.25 PLP code on top of devices that * can provide a lab_b service using the concap_proto mechanism. - * This module supports a network interface wich provides lapb_sematics + * This module supports a network interface which provides lapb_sematics * -- as defined in Documentation/networking/x25-iface.txt -- to * the upper layer and assumes that the lower layer provides a reliable * data link service by means of the concap_device_ops callbacks. diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index 670eb01a5a23..e1c18aa1d712 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -1309,7 +1309,7 @@ int bitmap_startwrite(struct bitmap *bitmap, sector_t offset, unsigned long sect case 1: *bmc = 2; } - if ((*bmc & COUNTER_MAX) == COUNTER_MAX) BUG(); + BUG_ON((*bmc & COUNTER_MAX) == COUNTER_MAX); (*bmc)++; spin_unlock_irq(&bitmap->lock); diff --git a/drivers/md/dm-hw-handler.c b/drivers/md/dm-hw-handler.c index 4cc0010e0156..baafaaba4d4b 100644 --- a/drivers/md/dm-hw-handler.c +++ b/drivers/md/dm-hw-handler.c @@ -83,8 +83,7 @@ void dm_put_hw_handler(struct hw_handler_type *hwht) if (--hwhi->use == 0) module_put(hwhi->hwht.module); - if (hwhi->use < 0) - BUG(); + BUG_ON(hwhi->use < 0); out: up_read(&_hwh_lock); diff --git a/drivers/parisc/sba_iommu.c b/drivers/parisc/sba_iommu.c index 5d47c5965c51..0821747e44cf 100644 --- a/drivers/parisc/sba_iommu.c +++ b/drivers/parisc/sba_iommu.c @@ -1724,9 +1724,7 @@ printk("sba_hw_init(): mem_boot 0x%x 0x%x 0x%x 0x%x\n", PAGE0->mem_boot.hpa, sba_dev->chip_resv.start = PCI_F_EXTEND | 0xfef00000UL; sba_dev->chip_resv.end = PCI_F_EXTEND | (0xff000000UL - 1) ; err = request_resource(&iomem_resource, &(sba_dev->chip_resv)); - if (err < 0) { - BUG(); - } + BUG_ON(err < 0); } else if (IS_PLUTO(sba_dev->iodc)) { int err; @@ -2185,8 +2183,7 @@ void sba_directed_lmmio(struct parisc_device *pci_hba, struct resource *r) int i; int rope = (pci_hba->hw_path & (ROPES_PER_IOC-1)); /* rope # */ - if ((t!=HPHW_IOA) && (t!=HPHW_BCPORT)) - BUG(); + BUG_ON((t!=HPHW_IOA) && (t!=HPHW_BCPORT)); r->start = r->end = 0; @@ -2228,8 +2225,7 @@ void sba_distributed_lmmio(struct parisc_device *pci_hba, struct resource *r ) int base, size; int rope = (pci_hba->hw_path & (ROPES_PER_IOC-1)); /* rope # */ - if ((t!=HPHW_IOA) && (t!=HPHW_BCPORT)) - BUG(); + BUG_ON((t!=HPHW_IOA) && (t!=HPHW_BCPORT)); r->start = r->end = 0; diff --git a/drivers/parisc/superio.c b/drivers/parisc/superio.c index ba971fecd0d8..ad6d3b28a3a6 100644 --- a/drivers/parisc/superio.c +++ b/drivers/parisc/superio.c @@ -157,8 +157,8 @@ superio_init(struct pci_dev *pcidev) if (sio->suckyio_irq_enabled) return; - if (!pdev) BUG(); - if (!sio->usb_pdev) BUG(); + BUG_ON(!pdev); + BUG_ON(!sio->usb_pdev); /* use the IRQ iosapic found for USB INT D... */ pdev->irq = sio->usb_pdev->irq; @@ -474,8 +474,7 @@ superio_probe(struct pci_dev *dev, const struct pci_device_id *id) dev->subsystem_vendor, dev->subsystem_device, dev->class); - if (!sio->suckyio_irq_enabled) - BUG(); /* Enabled by PCI_FIXUP_FINAL */ + BUG_ON(!sio->suckyio_irq_enabled); /* Enabled by PCI_FIXUP_FINAL */ if (dev->device == PCI_DEVICE_ID_NS_87560_LIO) { /* Function 1 */ superio_parport_init(); diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c index 7967916bda18..0a9f12c4e911 100644 --- a/drivers/s390/block/dasd.c +++ b/drivers/s390/block/dasd.c @@ -541,9 +541,8 @@ dasd_kmalloc_request(char *magic, int cplength, int datasize, struct dasd_ccw_req *cqr; /* Sanity checks */ - if ( magic == NULL || datasize > PAGE_SIZE || - (cplength*sizeof(struct ccw1)) > PAGE_SIZE) - BUG(); + BUG_ON( magic == NULL || datasize > PAGE_SIZE || + (cplength*sizeof(struct ccw1)) > PAGE_SIZE); cqr = kzalloc(sizeof(struct dasd_ccw_req), GFP_ATOMIC); if (cqr == NULL) @@ -583,9 +582,8 @@ dasd_smalloc_request(char *magic, int cplength, int datasize, int size; /* Sanity checks */ - if ( magic == NULL || datasize > PAGE_SIZE || - (cplength*sizeof(struct ccw1)) > PAGE_SIZE) - BUG(); + BUG_ON( magic == NULL || datasize > PAGE_SIZE || + (cplength*sizeof(struct ccw1)) > PAGE_SIZE); size = (sizeof(struct dasd_ccw_req) + 7L) & -8L; if (cplength > 0) diff --git a/drivers/serial/mpc52xx_uart.c b/drivers/serial/mpc52xx_uart.c index 928e6cf12dca..6459edc7f5c5 100644 --- a/drivers/serial/mpc52xx_uart.c +++ b/drivers/serial/mpc52xx_uart.c @@ -40,7 +40,7 @@ * and so on). So the PSC1 is mapped to /dev/ttyPSC0, PSC2 to /dev/ttyPSC1 and * so on. But be warned, it's an ABSOLUTE REQUIREMENT ! This is needed mainly * fpr the console code : without this 1:1 mapping, at early boot time, when we - * are parsing the kernel args console=ttyPSC?, we wouldn't know wich PSC it + * are parsing the kernel args console=ttyPSC?, we wouldn't know which PSC it * will be mapped to. */ diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c index a8455c9e79dd..495db5755df9 100644 --- a/drivers/usb/serial/option.c +++ b/drivers/usb/serial/option.c @@ -102,7 +102,7 @@ static struct usb_driver option_driver = { .no_dynamic_id = 1, }; -/* The card has three separate interfaces, wich the serial driver +/* The card has three separate interfaces, which the serial driver * recognizes separately, thus num_port=1. */ static struct usb_serial_driver option_3port_device = { diff --git a/drivers/video/bw2.c b/drivers/video/bw2.c index c029db4646f6..6577fdfdfc16 100644 --- a/drivers/video/bw2.c +++ b/drivers/video/bw2.c @@ -327,8 +327,7 @@ static void bw2_init_one(struct sbus_dev *sdev) } else #else { - if (!sdev) - BUG(); + BUG_ON(!sdev); all->par.physbase = sdev->reg_addrs[0].phys_addr; resp = &sdev->resource[0]; sbusfb_fill_var(&all->info.var, (sdev ? sdev->prom_node : 0), 1); diff --git a/drivers/video/ffb.c b/drivers/video/ffb.c index 9c9b21d469a1..7633e41adda1 100644 --- a/drivers/video/ffb.c +++ b/drivers/video/ffb.c @@ -466,8 +466,7 @@ static void ffb_fillrect(struct fb_info *info, const struct fb_fillrect *rect) unsigned long flags; u32 fg; - if (rect->rop != ROP_COPY && rect->rop != ROP_XOR) - BUG(); + BUG_ON(rect->rop != ROP_COPY && rect->rop != ROP_XOR); fg = ((u32 *)info->pseudo_palette)[rect->color]; diff --git a/drivers/video/sstfb.c b/drivers/video/sstfb.c index 99921df35474..8c1a8b5135c6 100644 --- a/drivers/video/sstfb.c +++ b/drivers/video/sstfb.c @@ -32,7 +32,7 @@ -TODO: at one time or another test that the mode is acceptable by the monitor -ASK: Can I choose different ordering for the color bitfields (rgba argb ...) - wich one should i use ? is there any preferred one ? It seems ARGB is + which one should i use ? is there any preferred one ? It seems ARGB is the one ... -TODO: in set_var check the validity of timings (hsync vsync)... -TODO: check and recheck the use of sst_wait_idle : we don't flush the fifo via diff --git a/drivers/w1/masters/matrox_w1.c b/drivers/w1/masters/matrox_w1.c index 591809cbbb97..2788b8ca9bb1 100644 --- a/drivers/w1/masters/matrox_w1.c +++ b/drivers/w1/masters/matrox_w1.c @@ -98,7 +98,7 @@ static void matrox_w1_write_ddc_bit(void *, u8); * * Using tristate pins, since i can't find any open-drain pin in whole motherboard. * Unfortunately we can't connect to Intel's 82801xx IO controller - * since we don't know motherboard schema, wich has pretty unused(may be not) GPIO. + * since we don't know motherboard schema, which has pretty unused(may be not) GPIO. * * I've heard that PIIX also has open drain pin. * diff --git a/fs/befs/datastream.c b/fs/befs/datastream.c index 785f6b2d5d10..b7d6b920f65f 100644 --- a/fs/befs/datastream.c +++ b/fs/befs/datastream.c @@ -118,7 +118,7 @@ befs_fblock2brun(struct super_block *sb, befs_data_stream * data, * befs_read_lsmylink - read long symlink from datastream. * @sb: Filesystem superblock * @ds: Datastrem to read from - * @buf: Buffer in wich to place long symlink data + * @buf: Buffer in which to place long symlink data * @len: Length of the long symlink in bytes * * Returns the number of bytes read diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index 5b3076e8ee90..a2e48c999c24 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -572,8 +572,7 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, csp -= sizeof(unsigned long); __put_user(bprm->argc, (unsigned long *) csp); - if (csp != sp) - BUG(); + BUG_ON(csp != sp); /* fill in the argv[] array */ #ifdef CONFIG_MMU diff --git a/fs/coda/cache.c b/fs/coda/cache.c index c607d923350a..5d0527133266 100644 --- a/fs/coda/cache.c +++ b/fs/coda/cache.c @@ -51,7 +51,7 @@ void coda_cache_clear_all(struct super_block *sb) struct coda_sb_info *sbi; sbi = coda_sbp(sb); - if (!sbi) BUG(); + BUG_ON(!sbi); atomic_inc(&permission_epoch); } diff --git a/fs/coda/cnode.c b/fs/coda/cnode.c index 23aeef5aa814..4c9fecbfa91f 100644 --- a/fs/coda/cnode.c +++ b/fs/coda/cnode.c @@ -120,8 +120,7 @@ void coda_replace_fid(struct inode *inode, struct CodaFid *oldfid, cii = ITOC(inode); - if (!coda_fideq(&cii->c_fid, oldfid)) - BUG(); + BUG_ON(!coda_fideq(&cii->c_fid, oldfid)); /* replace fid and rehash inode */ /* XXX we probably need to hold some lock here! */ diff --git a/fs/ufs/super.c b/fs/ufs/super.c index d257644a1aeb..db98a4c71e63 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -575,7 +575,7 @@ static int ufs_fill_super(struct super_block *sb, void *data, int silent) if (!silent) printk("You didn't specify the type of your ufs filesystem\n\n" "mount -t ufs -o ufstype=" - "sun|sunx86|44bsd|ufs2|5xbsd|old|hp|nextstep|netxstep-cd|openstep ...\n\n" + "sun|sunx86|44bsd|ufs2|5xbsd|old|hp|nextstep|nextstep-cd|openstep ...\n\n" ">>>WARNING<<< Wrong ufstype may corrupt your filesystem, " "default is ufstype=old\n"); ufs_set_opt (sbi->s_mount_opt, UFSTYPE_OLD); diff --git a/include/linux/timer.h b/include/linux/timer.h index 9b9877fd2505..ee5a09e806e8 100644 --- a/include/linux/timer.h +++ b/include/linux/timer.h @@ -69,13 +69,13 @@ extern unsigned long next_timer_interrupt(void); * @timer: the timer to be added * * The kernel will do a ->function(->data) callback from the - * timer interrupt at the ->expired point in the future. The + * timer interrupt at the ->expires point in the future. The * current time is 'jiffies'. * - * The timer's ->expired, ->function (and if the handler uses it, ->data) + * The timer's ->expires, ->function (and if the handler uses it, ->data) * fields must be set prior calling this function. * - * Timers with an ->expired field in the past will be executed in the next + * Timers with an ->expires field in the past will be executed in the next * timer tick. */ static inline void add_timer(struct timer_list *timer) diff --git a/ipc/msg.c b/ipc/msg.c index fbf757064a32..60c1e5c23418 100644 --- a/ipc/msg.c +++ b/ipc/msg.c @@ -220,8 +220,7 @@ asmlinkage long sys_msgget (key_t key, int msgflg) ret = -EEXIST; } else { msq = msg_lock(id); - if(msq==NULL) - BUG(); + BUG_ON(msq==NULL); if (ipcperms(&msq->q_perm, msgflg)) ret = -EACCES; else { diff --git a/kernel/cpu.c b/kernel/cpu.c index e882c6babf41..8be22bd80933 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -223,8 +223,7 @@ int __devinit cpu_up(unsigned int cpu) ret = __cpu_up(cpu); if (ret != 0) goto out_notify; - if (!cpu_online(cpu)) - BUG(); + BUG_ON(!cpu_online(cpu)); /* Now call notifier in preparation. */ notifier_call_chain(&cpu_chain, CPU_ONLINE, hcpu); diff --git a/lib/swiotlb.c b/lib/swiotlb.c index 0af497b6b9a8..10625785eefd 100644 --- a/lib/swiotlb.c +++ b/lib/swiotlb.c @@ -296,8 +296,7 @@ map_single(struct device *hwdev, char *buffer, size_t size, int dir) else stride = 1; - if (!nslots) - BUG(); + BUG_ON(!nslots); /* * Find suitable number of IO TLB entries size that will fit this @@ -416,14 +415,14 @@ sync_single(struct device *hwdev, char *dma_addr, size_t size, case SYNC_FOR_CPU: if (likely(dir == DMA_FROM_DEVICE || dir == DMA_BIDIRECTIONAL)) memcpy(buffer, dma_addr, size); - else if (dir != DMA_TO_DEVICE) - BUG(); + else + BUG_ON(dir != DMA_TO_DEVICE); break; case SYNC_FOR_DEVICE: if (likely(dir == DMA_TO_DEVICE || dir == DMA_BIDIRECTIONAL)) memcpy(dma_addr, buffer, size); - else if (dir != DMA_FROM_DEVICE) - BUG(); + else + BUG_ON(dir != DMA_FROM_DEVICE); break; default: BUG(); @@ -529,8 +528,7 @@ swiotlb_map_single(struct device *hwdev, void *ptr, size_t size, int dir) unsigned long dev_addr = virt_to_phys(ptr); void *map; - if (dir == DMA_NONE) - BUG(); + BUG_ON(dir == DMA_NONE); /* * If the pointer passed in happens to be in the device's DMA window, * we can safely return the device addr and not worry about bounce @@ -592,8 +590,7 @@ swiotlb_unmap_single(struct device *hwdev, dma_addr_t dev_addr, size_t size, { char *dma_addr = phys_to_virt(dev_addr); - if (dir == DMA_NONE) - BUG(); + BUG_ON(dir == DMA_NONE); if (dma_addr >= io_tlb_start && dma_addr < io_tlb_end) unmap_single(hwdev, dma_addr, size, dir); else if (dir == DMA_FROM_DEVICE) @@ -616,8 +613,7 @@ swiotlb_sync_single(struct device *hwdev, dma_addr_t dev_addr, { char *dma_addr = phys_to_virt(dev_addr); - if (dir == DMA_NONE) - BUG(); + BUG_ON(dir == DMA_NONE); if (dma_addr >= io_tlb_start && dma_addr < io_tlb_end) sync_single(hwdev, dma_addr, size, dir, target); else if (dir == DMA_FROM_DEVICE) @@ -648,8 +644,7 @@ swiotlb_sync_single_range(struct device *hwdev, dma_addr_t dev_addr, { char *dma_addr = phys_to_virt(dev_addr) + offset; - if (dir == DMA_NONE) - BUG(); + BUG_ON(dir == DMA_NONE); if (dma_addr >= io_tlb_start && dma_addr < io_tlb_end) sync_single(hwdev, dma_addr, size, dir, target); else if (dir == DMA_FROM_DEVICE) @@ -696,8 +691,7 @@ swiotlb_map_sg(struct device *hwdev, struct scatterlist *sg, int nelems, unsigned long dev_addr; int i; - if (dir == DMA_NONE) - BUG(); + BUG_ON(dir == DMA_NONE); for (i = 0; i < nelems; i++, sg++) { addr = SG_ENT_VIRT_ADDRESS(sg); @@ -730,8 +724,7 @@ swiotlb_unmap_sg(struct device *hwdev, struct scatterlist *sg, int nelems, { int i; - if (dir == DMA_NONE) - BUG(); + BUG_ON(dir == DMA_NONE); for (i = 0; i < nelems; i++, sg++) if (sg->dma_address != SG_ENT_PHYS_ADDRESS(sg)) @@ -753,8 +746,7 @@ swiotlb_sync_sg(struct device *hwdev, struct scatterlist *sg, { int i; - if (dir == DMA_NONE) - BUG(); + BUG_ON(dir == DMA_NONE); for (i = 0; i < nelems; i++, sg++) if (sg->dma_address != SG_ENT_PHYS_ADDRESS(sg)) diff --git a/mm/msync.c b/mm/msync.c index 2672b8dc3d89..bc6c95376366 100644 --- a/mm/msync.c +++ b/mm/msync.c @@ -126,7 +126,7 @@ static unsigned long msync_page_range(struct vm_area_struct *vma, * write out the dirty pages and wait on the writeout and check the result. * Or the application may run fadvise(FADV_DONTNEED) against the fd to start * async writeout immediately. - * So my _not_ starting I/O in MS_ASYNC we provide complete flexibility to + * So by _not_ starting I/O in MS_ASYNC we provide complete flexibility to * applications. */ static int msync_interval(struct vm_area_struct *vma, unsigned long addr, diff --git a/scripts/Makefile.modpost b/scripts/Makefile.modpost index bf96a61d4b86..9a3ec20b0be2 100644 --- a/scripts/Makefile.modpost +++ b/scripts/Makefile.modpost @@ -4,7 +4,7 @@ # # Stage one of module building created the following: # a) The individual .o files used for the module -# b) A <module>.o file wich is the .o files above linked together +# b) A <module>.o file which is the .o files above linked together # c) A <module>.mod file in $(MODVERDIR)/, listing the name of the # the preliminary <module>.o file, plus all .o files diff --git a/sound/sparc/cs4231.c b/sound/sparc/cs4231.c index 53a148b01b6b..8804f26ddb3a 100644 --- a/sound/sparc/cs4231.c +++ b/sound/sparc/cs4231.c @@ -611,8 +611,7 @@ static void snd_cs4231_advance_dma(struct cs4231_dma_control *dma_cont, unsigned int period_size = snd_pcm_lib_period_bytes(substream); unsigned int offset = period_size * (*periods_sent); - if (period_size >= (1 << 24)) - BUG(); + BUG_ON(period_size >= (1 << 24)); if (dma_cont->request(dma_cont, runtime->dma_addr + offset, period_size)) return; @@ -1079,8 +1078,7 @@ static int snd_cs4231_playback_prepare(struct snd_pcm_substream *substream) chip->image[CS4231_IFACE_CTRL] &= ~(CS4231_PLAYBACK_ENABLE | CS4231_PLAYBACK_PIO); - if (runtime->period_size > 0xffff + 1) - BUG(); + BUG_ON(runtime->period_size > 0xffff + 1); chip->p_periods_sent = 0; spin_unlock_irqrestore(&chip->lock, flags); |