Commit Graph

50 Commits

Author SHA1 Message Date
Takashi Iwai 0914f7961b ALSA: Avoid endless sleep after disconnect
When disconnect callback is called, each component should wake up
sleepers and check card->shutdown flag for avoiding the endless sleep
blocking the proper resource release.

Cc: <stable@vger.kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2012-10-30 11:07:15 +01:00
Takashi Iwai a0830dbd4e ALSA: Add a reference counter to card instance
For more strict protection for wild disconnections, a refcount is
introduced to the card instance, and let it up/down when an object is
referred via snd_lookup_*() in the open ops.

The free-after-last-close check is also changed to check this refcount
instead of the empty list, too.

Reported-by: Matthieu CASTET <matthieu.castet@parrot.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2012-10-30 11:07:10 +01:00
Paul Gortmaker 65a772172b sound: fix drivers needing module.h not moduleparam.h
The implicit presence of module.h lured several users into
incorrectly thinking that they only needed/used modparam.h
but once we clean up the module.h presence, these will show
up as build failures, so fix 'em now.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-31 19:31:19 -04:00
Takashi Iwai 3101ba035c ALSA: Use krealloc() in possible places
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2011-07-12 08:05:16 +02:00
Takashi Iwai b3c705aa9e ALSA: rawmidi - Use workq for event handling
Kill tasklet usage in rawmidi core code.  Use workq for the event callback
instead of tasklet (which is used only in core/seq/seq_midi.c).

Signed-off-by: Takashi Iwai <tiwai@suse.de>
2011-06-14 14:37:06 +02:00
Clemens Ladisch aa73aec6c3 ALSA: rawmidi: fix oops (use after free) when unloading a driver module
When a driver module is unloaded and the last still open file is a raw
MIDI device, the card and its devices will be actually freed in the
snd_card_file_remove() call when that file is closed.  Afterwards, rmidi
and rmidi->card point into freed memory, so the module pointer is likely
to be garbage.
(This was introduced by commit 9a1b64caac82aa02cb74587ffc798e6f42c6170a.)

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Reported-by: Krzysztof Foltman <wdev@foltman.com>
Cc: 2.6.30-2.6.35 <stable@kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2010-10-17 10:11:40 +02:00
Dan Carpenter a7a13d0676 ALSA: rawmidi: fix the get next midi device ioctl
If we pass in a device which is higher than SNDRV_RAWMIDI_DEVICES then
the "next device" should be -1.  This function just returns device + 1.

But the main thing is that "device + 1" can lead to a (harmless) integer
overflow and that annoys static analysis tools.

[fix the case for device == SNDRV_RAWMIDI_DEVICE by tiwai]

Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2010-09-09 09:05:21 +02:00
Takashi Iwai 02f4865fa4 ALSA: core - Define llseek fops
Set no_llseek to llseek file ops of each sound component (but for hwdep).
This avoids the implicit BKL invocation via generic_file_llseek() used
as default when fops.llseek is NULL.

Also call nonseekable_open() at each open ops to ensure the file flags
have no seek bit.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
2010-04-13 12:01:21 +02:00
Christoph Hellwig 6b2f3d1f76 vfs: Implement proper O_SYNC semantics
While Linux provided an O_SYNC flag basically since day 1, it took until
Linux 2.4.0-test12pre2 to actually get it implemented for filesystems,
since that day we had generic_osync_around with only minor changes and the
great "For now, when the user asks for O_SYNC, we'll actually give
O_DSYNC" comment.  This patch intends to actually give us real O_SYNC
semantics in addition to the O_DSYNC semantics.  After Jan's O_SYNC
patches which are required before this patch it's actually surprisingly
simple, we just need to figure out when to set the datasync flag to
vfs_fsync_range and when not.

This patch renames the existing O_SYNC flag to O_DSYNC while keeping it's
numerical value to keep binary compatibility, and adds a new real O_SYNC
flag.  To guarantee backwards compatiblity it is defined as expanding to
both the O_DSYNC and the new additional binary flag (__O_SYNC) to make
sure we are backwards-compatible when compiled against the new headers.

This also means that all places that don't care about the differences can
just check O_DSYNC and get the right behaviour for O_SYNC, too - only
places that actuall care need to check __O_SYNC in addition.  Drivers and
network filesystems have been updated in a fail safe way to always do the
full sync magic if O_DSYNC is set.  The few places setting O_SYNC for
lower layers are kept that way for now to stay failsafe.

We enforce that O_DSYNC is set when __O_SYNC is set early in the open path
to make sure we always get these sane options.

Note that parisc really screwed up their headers as they already define a
O_DSYNC that has always been a no-op.  We try to repair it by using it for
the new O_DSYNC and redefinining O_SYNC to send both the traditional
O_SYNC numerical value _and_ the O_DSYNC one.

Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Grant Grundler <grundler@parisc-linux.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andreas Dilger <adilger@sun.com>
Acked-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Acked-by: Kyle McMartin <kyle@mcmartin.ca>
Acked-by: Ulrich Drepper <drepper@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jan Kara <jack@suse.cz>
2009-12-10 15:02:50 +01:00
Clemens Ladisch 7584af10cf sound: rawmidi: record a substream's owner process
Record the pid of the task that opened a RawMIDI substream.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2009-11-10 16:32:38 +01:00
Clemens Ladisch 91d12c485b sound: rawmidi: fix opened substreams count
The substream_opened field is to count the number of opened substreams,
not the number of times that any substreams have been opened.

Furthermore, all substreams being opened does not imply that the next
open would fail, due to the possibility of O_APPEND.  With this wrong
check, opening a substream multiple times would succeed only if the
device had more unopened substreams.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2009-11-10 16:32:10 +01:00
Takashi Iwai 3f225c07c7 Merge branch 'topic/ctl-pid-lock' into topic/core-change 2009-11-10 16:30:03 +01:00
Clemens Ladisch b7fe750fcc sound: rawmidi: fix MIDI device O_APPEND error handling
Commit 9a1b64caac in 2.6.30 broke the
error handling code in rawmidi_open_priv().

If only the output substream of a RawMIDI device has been opened and
if this device is then opened with O_RDWR | O_APPEND and if the
initialization of the input substream fails (either because of low
memory or because the device driver's open callback fails), then the
runtime structure of the already open output substream will be freed
and all following writes through the first handle will cause
snd_rawmidi_write() to use the NULL runtime pointer.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Cc: <stable@kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2009-11-10 16:22:59 +01:00
Clemens Ladisch 16fb109644 sound: rawmidi: fix checking of O_APPEND when opening MIDI device
Commit 9a1b64caac in 2.6.30 dropped the
check that a substream must already have been opened with O_APPEND to be
able to open it a second time.

This would make it possible for a substream to be switched to append
mode, which would mean that non-atomic writes would fail unexpectedly.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Cc: <stable@kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2009-11-10 16:21:30 +01:00
Clemens Ladisch 8579d2d777 sound: rawmidi: fix double init when opening MIDI device with O_APPEND
Commit 9a1b64caac in 2.6.30 moved the
substream initialization code to where it would be executed every time
the substream is opened.

This had the consequence that any further opening would drop and leak
the data in the existing buffer, and that the device driver's open
callback would be called multiple times, unexpectedly.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Cc: <stable@kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2009-11-10 16:20:43 +01:00
Clemens Ladisch 25d27eded1 control: use reference-counted pid
Instead of storing the PID number, take a reference to the task's pid
structure.  This protects against duplicates due to PID overflows, and
using pid_vnr() ensures that the PID returned by snd_ctl_elem_info() is
correct as seen from the current namespace.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2009-11-06 14:32:06 +01:00
Clemens Ladisch 2d4b842014 sound: rawmidi: disable active-sensing-on-close by default
Sending an Active Sensing message when closing a port can interfere with
the following data if the port is reopened and a note-on is sent before
the device's timeout has elapsed.  Therefore, it is better to disable
this setting by default.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2009-07-15 11:57:20 +02:00
Takashi Iwai 9a1b64caac ALSA: rawmidi - Refactor rawmidi open/close codes
Refactor rawmidi open/close code messes.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
2009-03-09 15:17:23 +01:00
Takashi Iwai f9d202833d ALSA: rawmidi - Fix possible race in open
The module refcount should be handled in the register_mutex to avoid
possible races with module unloading.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
2009-03-09 15:17:21 +01:00
Takashi Iwai 1f04128a3d ALSA: hda - Convert from takslet_hi_schedule() to tasklet_schedule()
Replace all tasklet_hi_schedule() callers with the normal
tasklet_schedule().  The former often causes troubles with
RT-kernels, and has actually no merit.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
2008-12-18 12:17:55 +01:00
Takashi Iwai 219df32fae ALSA: rawmidi - Add open check in rawmidi callbacks
The drivers (e.g. mtpav) may call rawmidi functions in irq handlers
even though the streams are not opened.  This results in Oops or panic.

This patch adds the rawmidi state check before actually operating the
rawmidi buffers.

Tested-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2008-11-03 08:57:12 +01:00
Takashi Iwai 7eaa943c8e ALSA: Kill snd_assert() in sound/core/*
Kill snd_assert() in sound/core/*, either removed or replaced with
if () with snd_BUG_ON().

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
2008-08-13 11:46:35 +02:00
Takashi Iwai 399ccdc1cd ALSA: fix locking in snd_pcm_open*() and snd_rawmidi_open*()
The PCM and rawmidi open callbacks have a lock against card->controls_list
but it takes a wrong one, card->controls_rwsem, instead of a right one
card->ctl_files_rwlock.  This patch fixes them.

This change also fixes automatically the potential deadlocks due to
mm->mmap_sem in munmap and copy_from/to_user, reported by Sitsofe
Wheeler:

  A: snd_ctl_elem_user_tlv(): card->controls_rwsem => mm->mmap_sem
  B: snd_pcm_open(): card->open_mutex => card->controls_rwsem
  C: munmap: mm->mmap_sem => snd_pcm_release(): card->open_mutex

The patch breaks the chain.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
2008-09-25 14:51:03 +02:00
Marcin Ślusarz 17596a80d3 [ALSA] rawmidi: let sparse know what is going on _for real_
snd_rawmidi_kernel_read1/write1 weren't annotated but used
copy_to_user/copy_from_user when one of parameters (kernel) was equal to 0
remove it and add properly annotated parameter

Signed-off-by: Marcin Ślusarz <marcin.slusarz@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
2008-01-31 17:29:49 +01:00
Takashi Iwai 9004acc70e [ALSA] Remove sound/driver.h
This header file exists only for some hacks to adapt alsa-driver
tree.  It's useless for building in the kernel.  Let's move a few
lines in it to sound/core.h and remove it.
With this patch, sound/driver.h isn't removed but has just a single
compile warning to include it.  This should be really killed in
future.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
2008-01-31 17:29:48 +01:00
Jaroslav Kysela c1017a4cdb [ALSA] Changed Jaroslav Kysela's e-mail from perex@suse.cz to perex@perex.cz
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
2007-10-16 16:51:18 +02:00
Jesper Juhl 25765c4e5d [ALSA] Clean up duplicate includes in sound/core/
This patch cleans up duplicate includes in
	sound/core/

Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
2007-10-16 15:03:19 +02:00
Randy Dunlap e63340ae6b header cleaning: don't include smp_lock.h when not used
Remove includes of <linux/smp_lock.h> where it is not used/needed.
Suggested by Al Viro.

Builds cleanly on x86_64, i386, alpha, ia64, powerpc, sparc,
sparc64, and arm (all 59 defconfigs).

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-05-08 11:15:07 -07:00
Arjan van de Ven 9c2e08c592 [PATCH] mark struct file_operations const 9
Many struct file_operations in the kernel can be "const".  Marking them const
moves these to the .rodata section, which avoids false sharing with potential
dirty data.  In addition it'll catch accidental writes at compile time to
these shared resources.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-02-12 09:48:46 -08:00
Johannes Berg 9244b2c307 [ALSA] alsa core: convert to list_for_each_entry*
This patch converts most uses of list_for_each to list_for_each_entry all
across alsa. In some place apparently an item can be on a list with
different pointers so of course that isn't compatible with list_for_each, I
therefore didn't touch those places.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
2007-02-09 09:00:10 +01:00
Akinobu Mita c13893d7be [ALSA] sound: initialize rawmidi substream list
If snd_rawmidi_new() failed to allocate substreams for input
(snd_rawmidi_alloc_substreams() failed to populate a
&rmidi->streams[SNDRV_RAWMIDI_STREAM_INPUT]), it will try to
free rawmidi instance by snd_rawmidi_free().
But it will cause oops because snd_rawmidi_free() tries to free
both of substreams list but list for output
(&rmidi->streams[SNDRV_RAWMIDI_STREAM_OUTPUT]) is not initialized yet.

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
2006-12-20 08:55:59 +01:00
Takashi Iwai 2529bba760 [ALSA] Fix substream selection in PCM and rawmidi
The PCM and rawmidi substreams can be selected explicitly by opening
control handle and set via *_PREFER_SUBDEVICE ioctl.  But, when
multiple controls are opened, the driver gets confused.
The patch fixes the initialization of prefer_*_subdevice and the
check of multiple controls.  The first set subdevice is picked up
as the valid one.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
2006-09-23 10:39:59 +02:00
Takashi Iwai c461482c80 [ALSA] Unregister device files at disconnection
Orignally proposed by Sam Revitch <sam.revitch@gmail.com>.
Unregister device files at disconnection to avoid the futher accesses.
Also, the dev_unregister callback is removed and replaced with the
combination of disconnect + free.
A new function snd_card_free_when_closed() is introduced, which is
used in USB disconnect callback.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
2006-09-23 10:36:58 +02:00
Takashi Iwai 746d4a02e6 [ALSA] Fix disconnection of proc interface
- Add the linked list to each proc entry to enable a single-shot
  disconnection (unregister)
- Deprecate snd_info_unregister(), use snd_info_free_entry()
- Removed NULL checks of snd_info_free_entry()

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jaroslav Kysela <perex@suse.cz>
2006-09-23 10:36:48 +02:00
Takashi Iwai 6581f4e74d [ALSA] Remove zero-initialization of static variables
Removed zero-initializations of static variables.
A tiny optimization.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
2006-06-22 21:33:45 +02:00
Takashi Iwai bf850204a7 [ALSA] Remove unneeded read/write_size fields in proc text ops
Remove unneeded read/write_size fields in proc text ops.
snd_info_set_text_ops() is fixed, too.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
2006-06-22 21:33:09 +02:00
Jesper Juhl 4fa95ef639 [PATCH] sound: Remove unneeded kmalloc() return value casts
Get rid of unnessesary casts of kmalloc() return value in sound/

Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Cc: Jaroslav Kysela <perex@perex.cz>
Cc: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-28 09:16:09 -08:00
Ingo Molnar 1a60d4c5a0 [ALSA] semaphore -> mutex (core part)
Semaphore to mutex conversion.

The conversion was generated via scripts, and the result was validated
automatically via a script as well.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2006-03-22 10:24:50 +01:00
Adrian Bunk 18612048b3 [ALSA] sound/: possible cleanups
Modules: RawMidi Midlevel,HDA generic driver

This patch contains the following possible cleanups:
- pci/hda/hda_proc.c should #include 'hda_local.h' for including the
  prototype of it's global function snd_hda_codec_proc_new()
- core/rawmidi.c: make the needlessly global and EXPORT_SYMBOL'ed
  function snd_rawmidi_info() static

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
2006-01-03 12:29:40 +01:00
Clemens Ladisch 1b98ea4791 [ALSA] rawmidi: adjust runtime->avail when changing output buffer size
Modules: RawMidi Midlevel

When the output buffer size is changed, runtime->avail must be
adjusted to the new size; otherwise, draining of a bigger buffer would
appear to fail because avail does not reach buffer_size.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
2006-01-03 12:29:25 +01:00
Clemens Ladisch f87135f56c [ALSA] dynamic minors (3/6): store device-specific object pointers dynamically
Instead of storing the pointers to the device-specific structures in an
array, put them into the struct snd_minor, and look them up dynamically.

This makes the device type modules independent of the minor number
encoding.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
2006-01-03 12:29:17 +01:00
Clemens Ladisch 2af677fc88 [ALSA] dynamic minors (1/6): store device type in struct snd_minor
Instead of a comment string, store the device type in the snd_minor
structure.  This makes snd_minor more flexible, and has the nice side
effect that we don't need anymore to create a separate snd_minor
template for registering a device but can pass the file_operations
directly to snd_register_device().

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
2006-01-03 12:29:14 +01:00
Clemens Ladisch a106cd3d9e [ALSA] rawmidi: protect against invalid device number in snd_rawmidi_info_select()
Modules: RawMidi Midlevel

In snd_rawmidi_info_select(), check that the device identified by the
passed device number actually exists.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
2006-01-03 12:29:12 +01:00
Takashi Iwai 73e77ba023 [ALSA] Add error messages
Add error messages in the critial error path to be more verbose.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
2006-01-03 12:28:58 +01:00
Takashi Iwai 48c9d417d7 [ALSA] Remove xxx_t typedefs: Raw MIDI
Modules: RawMidi Midlevel

Remove xxx_t typedefs from the core raw MIDI codes.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
2006-01-03 12:17:35 +01:00
Clemens Ladisch f190286016 [ALSA] fix improper CONFIG_SND_MAJOR usage
Modules: HWDEP Midlevel,PCM Midlevel,RawMidi Midlevel,ALSA Core

Replace usage of CONFIG_SND_MAJOR with snd_major, where appropriate.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
2005-11-04 13:20:08 +01:00
Takashi Iwai ca2c096656 [ALSA] Replace with kzalloc() - core stuff
Control Midlevel,ALSA Core,HWDEP Midlevel,PCM Midlevel,RawMidi Midlevel
Timer Midlevel,ALSA<-OSS emulation
Replace kcalloc(1,..) with kzalloc().

Signed-off-by: Takashi Iwai <tiwai@suse.de>
2005-09-12 10:47:59 +02:00
Takashi Iwai df8db936e5 [ALSA] Fix DocBook warnings
PCM Midlevel,RawMidi Midlevel
Fix DocBook warnings.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
2005-09-12 10:42:22 +02:00
Clemens Ladisch 4d23359b7e [ALSA] sparse address space annotations
ALSA Core,RawMidi Midlevel,ALSA<-OSS emulation,ALSA sequencer
RME32 driver,RME96 driver,EMU10K1/EMU10K2 driver,NM256 driver
Add sparse annotations where we do strange this with __iomem/__user
pointers.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
2005-09-12 10:41:16 +02:00
Linus Torvalds 1da177e4c3 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