1a90d1cc3c
Provide default implementation of waveInPrepareHeader and
...
waveInUnprepareHeader if driver doesn't support them.
2005-03-16 11:30:34 +00:00
8b55b76daf
Fix memory leak in error path.
2005-03-15 15:39:03 +00:00
cf4d0d9707
Fix memory leak when there are too many drivers.
2005-03-15 15:38:48 +00:00
17cff085fb
Avoid trap in mixerGetLineControlsA when cControls is uninitialized
...
and MIXER_GETLINECONTROLSSF_ONEBYTYPE requested.
2005-03-15 15:38:34 +00:00
001a936176
Assorted memleak fixes. Found on Michael Stefaniuc smatch list.
2005-03-11 12:50:27 +00:00
7b738b7f8c
Make sure to always pass valid QS_* flags to
...
MsgWaitForMultipleObjects.
2005-03-08 17:00:38 +00:00
d5749b454d
Make wineoss's mixer know about DST_HEADPHONES, DST_LINE and
...
DST_VOICEIN. Use proper constants instead of 0 and 1.
2005-03-07 10:59:28 +00:00
5727918424
Added a check for poll.h.
2005-03-04 12:38:36 +00:00
9e53994abe
Warn the user if he is running the test in interactive mode but the
...
sound is turned off.
2005-03-03 13:49:30 +00:00
fbb3343547
Assorted spelling fixes.
2005-03-02 13:53:50 +00:00
7105248ae6
Apply Jeremy White's SMPTE calculation fix to each audio driver.
2005-03-02 13:39:23 +00:00
d0ee9f9309
Avoid spaces before preprocessor directives, that's not portable.
2005-03-02 12:23:20 +00:00
36e845ce05
Aggressively round up to the proper alignment when reporting position
...
on streams where we are converting up.
2005-03-02 12:17:28 +00:00
cf23045791
More correctly establish cases where the OSS input buffer should not
...
be enabled.
2005-03-02 10:12:35 +00:00
18a9738943
Added some more cases to MCI_MapMsgAtoW.
2005-02-26 17:41:11 +00:00
2db20ef1f0
More correct fix for dsp empty case.
2005-02-23 21:00:02 +00:00
494a34bfda
An interactive test should have volume, else a rational user will
...
decide it is broken.
2005-02-23 20:28:28 +00:00
aa2046f643
In calculating the SMPTE time, we would round incorrectly (0:0:4.30
...
instead of 0:0:5.0 [the last column only goes to 29]).
2005-02-23 15:40:54 +00:00
920ef0fbc9
When we get down to the last fragment or two in our output to the DSP,
...
we should use the SND_DSP_POST ioctl to let the sound card know to
flush those fragments. This allows us to precisely know when the
buffer is empty, and fixes another set of winmm tests.
2005-02-23 12:43:04 +00:00
1b685779a4
When testing the WAVE_MAPPER device, it seems inappropriate (to me) to
...
demand WAVE_FORMAT_DIRECT; this fixes another slew of tests.
2005-02-23 12:42:43 +00:00
15ec2780ef
When we are converting from one sample rate to another, we do have to
...
adjust when calculating TIME_SAMPLES.
2005-02-23 12:42:32 +00:00
87844214ab
Fixed offsets calculations.
2005-02-18 12:53:26 +00:00
e3566d8260
Fix timer tests to pass on windows 98.
2005-02-18 12:51:54 +00:00
18bfe6a095
Change SUBLANG_DEFAULT to SUBLANG_NEUTRAL for LANG_SPANISH in all
...
resources, so that Spanish locales other than Spain also use Spanish
resources.
2005-02-16 21:10:59 +00:00
993fbb94e0
Fix improper type-conversions.
2005-02-16 17:52:35 +00:00
ecdf41ef2c
Add some missing \ns.
2005-02-14 20:54:12 +00:00
b33ef44105
Added parameter checking to timeGetDevCaps.
...
Added timer tests.
2005-02-14 11:00:59 +00:00
1eed552b4d
- rewrite MCI interfaces as Unicode interface (as it should be)
...
- made some winmm APIs rely on the Unicode version for the core
implementation (instead of Ansi flavor)
2005-02-09 22:26:25 +00:00
7d92b5d8b4
Fix structure member name to match windows headers and fix up all
...
uses.
Nameless struct fix for dsound/propset.c to compile on both windows
and wine.
2005-01-28 17:21:07 +00:00
d1685b00bf
Authors: Klemens Friedl <klemens_friedl@gmx.net>, Henning Gerhardt <henning.gerhardt@web.de>
...
Spelling fixes.
2005-01-17 13:43:31 +00:00
ed12a3defd
- Use Interlocked* instead of ++/-- in AddRef/Release.
...
- Use only stored result of Interlocked* in AddRef/Release.
- Expand TRACEs to display the ref count.
2005-01-12 19:55:24 +00:00
151015fa02
Only use Alsa if 1.0 is detected.
2005-01-10 13:26:33 +00:00
4a0f999859
Use CreateEventW in preference to CreateEventA for unnamed events.
2005-01-07 15:40:09 +00:00
9d589acc26
Assorted spelling fixes.
2005-01-04 20:39:54 +00:00
75f87dd43f
Get rid of the ThunkData stubs, these are not functions.
2005-01-03 20:47:48 +00:00
15a308a9d7
Do not check for non NULL pointer before HeapFree'ing it. It's
...
redundant.
2005-01-03 14:56:42 +00:00
5ad7d858e0
Do not check for non NULL pointer before HeapFree'ing it. It's
...
redundant.
2004-12-23 17:06:43 +00:00
ee0344a4a2
Remove unneeded NULL casts.
2004-12-22 15:13:20 +00:00
2d8e75dab5
A driver does not have to support all formats.
2004-12-20 19:08:18 +00:00
6ca75f8bb7
Return actual result on error.
2004-12-20 17:09:12 +00:00
7b53f635f7
Return flag for DRVM_MAPPER_PREFERRED_GET.
2004-12-17 19:04:28 +00:00
ebe3c529b1
Janitorial: C booleans must not be compared against TRUE.
2004-12-09 14:07:59 +00:00
c24ca67f58
Match PSDK definitions for LPCWAVEFORMATEX.
2004-12-08 13:57:55 +00:00
0e9f423b17
Fixed linking issues.
2004-12-08 13:44:41 +00:00
5f1ef6d09b
Fixed compile error.
2004-12-07 14:31:53 +00:00
b4e495574b
Authors: Eric Pouech <pouech-eric@wanadoo.fr>, Filip Navara <xnavara@volny.cz>
...
32 bit low level drivers now use a Unicode interface (used to be ANSI):
- Send the *_GETDEVCAPS messages in unicode format from the
*GetDevCapsW function and call the *GetDevCapsW from their respective
Ansi versions.
- Modify all low level drivers to correctly process the *_GETDEVCAPS
messages as unicode.
- *_GETDEVCAPS messages are now mapped from/to unicode for 16 bit code.
- Removed all SoundBlaster naming oldies.
Better use of some unicode functions (instead of Ansi) in winmm.
2004-12-06 20:55:25 +00:00
155c8acfbd
- Check the return value from *_GETNUMDEVS message as per KB90562.
...
- Fix typo in MMDRV_InstallMap.
2004-12-01 15:32:19 +00:00
0a258964c2
Const correctness fixes.
2004-11-30 21:38:57 +00:00
5c2a891c43
Made some functions and variables static.
2004-11-29 18:00:10 +00:00
8cd13776e2
Add some missing error values in WINMM_ErrorToString.
2004-11-21 15:37:04 +00:00
Robert Reif