22 May, 2007
1 commit
-
First thing mm.h does is including sched.h solely for can_do_mlock() inline
function which has "current" dereference inside. By dealing with can_do_mlock()
mm.h can be detached from sched.h which is good. See below, why.This patch
a) removes unconditional inclusion of sched.h from mm.h
b) makes can_do_mlock() normal function in mm/mlock.c
c) exports can_do_mlock() to not break compilation
d) adds sched.h inclusions back to files that were getting it indirectly.
e) adds less bloated headers to some files (asm/signal.h, jiffies.h) that were
getting them indirectlyNet result is:
a) mm.h users would get less code to open, read, preprocess, parse, ... if
they don't need sched.h
b) sched.h stops being dependency for significant number of files:
on x86_64 allmodconfig touching sched.h results in recompile of 4083 files,
after patch it's only 3744 (-8.3%).Cross-compile tested on
all arm defconfigs, all mips defconfigs, all powerpc defconfigs,
alpha alpha-up
arm
i386 i386-up i386-defconfig i386-allnoconfig
ia64 ia64-up
m68k
mips
parisc parisc-up
powerpc powerpc-up
s390 s390-up
sparc sparc-up
sparc64 sparc64-up
um-x86_64
x86_64 x86_64-up x86_64-defconfig x86_64-allnoconfigas well as my two usual configs.
Signed-off-by: Alexey Dobriyan
Signed-off-by: Linus Torvalds
16 May, 2007
8 commits
-
This patch makes sure that short USB replies are treated as an
error when requesting the value of a certain mixer control.Signed-off-by: Thomas Reitmayr
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
Included is a patch against current alsa-kernel hg tip adding
quirks that Ubuntu Dapper/Edgy/Feisty contains.Signed-off-by: Daniel T Chen
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
isa_register_driver() returns an error if no device is found
and it's no fatal error for the drivers with pnp support.Signed-off-by: Rene Herman
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
Fixed some platform no sound and Add JD Function for below platform list:
Sony(0x104d) 0x9015, 0x900e, 0x1f00
Toshiba(0x1179) DALLAS 0xff00, 0xff01
Targa(0x1462) 0x28fb
Asus(0x1043) A7J 0x060d
Lenovo(0x17aa) 0x3bfd, 0x2085
MEDION(0x17c0) MD2 0x4071Signed-off-by: Kailang Yang
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
This patch fixes a bug whereby AC97 bus device data was being clobbered
when AC97 codecs using the generic ac97_codec.c driver were being
registered. Codecs that didn't use the generic driver were unaffected
(e.g. WM9712, WM9713).
Changes:-
o Add new AC97 codec class for custom (or need bus dev registration)
AC97 codecs.
o Only register/deregister this custom codec device with the AC97 bus.
The generic AC97 driver already does this for generic codec devices.
This may be related to bug #3038 :-
https://bugtrack.alsa-project.org/alsa-bug/view.php?id=3038Signed-off-by: Liam Girdwood
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
This patch fixes a bug whereby the GPL'ed symbol ac97_dai was being
declared as static.Signed-off-by: Liam Girdwood
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
In some cases, mixer elements return -EINVAL because it couldn't
obtain proper amp_cap bits. The patch improves the robustness,
trying the amp_cap query again in such a case.Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
This patch adds the ID for another quickcam microphone, reported by freqmod
on ALSA ticket #0003040
I'm going to submit a USB patch separately to provide a macro to simplify these
entries, as suggested by Alan Stern. We could switch to using that in future.Signed-off-by: Daniel Drake
Signed-off-by: Clemens Ladisch
Signed-off-by: Jaroslav Kysela
11 May, 2007
31 commits
-
Some verb tables (such as an Asus VT sent by IDT) contain only speaker
outs in the default pin configs, and no line-outs. In such a case the
speaker sequence numbers have to be used to order the speaker out
pins, just as is being done for line-out pins. Then, when speaker-outs
are copied to line-outs, the line-outs will be ordered properly.Signed-off-by: Steve Longerbeam
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
- Added ALC861VD Lenovo support (17aa:3802, 17aa:2066)
- Modify alc_subsystem_idSigned-off-by: Kailang Yang
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
Fix the retrival of widget connection list in the generic parser.
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
Add a workaround for the ESI M4U that claims to support 32-byte packets
but ignores the remaining bytes of packets bigger than four bytes.Signed-off-by: Clemens Ladisch
Signed-off-by: Jaroslav Kysela -
The firmware of the M-Audio USB Uno MIDI Interface has, at least in
hardware revision 1.25, a bug that garbles its USB output. When it
receives a Note On MIDI message that uses running status, the resulting
USB MIDI packet has a wrong CIN (4 instead of 9) and a wrong length
(2 bytes, the status byte is still missing).
This patch adds a workaround to track the CINs and the MIDI messages of
received USB MIDI packets to detect whether a packet with CIN 4 is a
correct SysEx packet or a buggy running status packet.Signed-off-by: Clemens Ladisch
Signed-off-by: Jaroslav Kysela -
Commit 93c8bf45e083b89dffe3a708363c15c1b220c723 modified the USB device
matching behaviour to ignore interface class matches if the device class
is vendor-specific.
This patch adds explicit ID matches for Logitech QuickCam devices, which
have a vendor specific device class (but standards-compliant audio
interfaces).
This fixes a 2.6.20 regression where the audio component of these
devices was no longer usable.
http://bugs.gentoo.org/show_bug.cgi?id=175715
https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.20/+bug/93822
https://bugtrack.alsa-project.org/alsa-bug/view.php?id=3040
Based on a patch from sergiomSigned-off-by: Daniel Drake
Signed-off-by: Clemens Ladisch
Signed-off-by: Jaroslav Kysela -
The AMP mute bit is bit 7. No real influence since no one uses this
definition yet, though...Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
Fix the auto-mute controls of ALC880 uniwill model. Split to two
individual functions to handle HP and front-mic mutes. For front-mic
mute, use snd_hda_codec_amp_update() to be consistent with mixer.Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
Fix AD1988 SPDIF playback route control for selecting ADC1-3.
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
I quuestion the testing status of that patch!
Signed-off-by: Andrew Morton
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
From: David Rientjes
isapnp[] is only used for CONFIG_PNP. If this configuration option is
not set, do not declare the array.
Cc: Adam Belay
Cc: Takashi IwaiSigned-off-by: David Rientjes
Signed-off-by: Andrew Morton
Signed-off-by: Jaroslav Kysela -
* fixed surround playback on stac922x. Pin direction control bits were
not being set correctly in stac92xx_set_pinctl(). Specifically it
would refuse to set the port as an output if the port was already
configured as an input. Last hunk (#8).
* fixed an input mux bug on 92xx codecs. When there is more than one
possible input calculated for the muxes, the actual mux widget never
gets set from its reset default, which is index 0, in the stac9221
case that is port E. So alsamixer/amixer/gnome-mixer report the Mic
as being the selected input source, but in fact is something else
(line-in port E in stac9221 case). Another problem with this is that
if you actually try to set the mux input to 'Mic', nothing happens
because *cur_val == idx (see snd_hda_input_mux_put). You have to
actually toggle input source to line-in then back to mic to actually
set the mux widget. Hunk #7.
* fixed some typos in patch_sigmatel.c. Hunk #6.
* fix to stac92xx_add_dyn_out_pins() that fixes surround playback on
codecs with less that 4 DACs (stac9205 for example). It reads the widget
caps cache created by hda_codec to count the total number of analog DACs
found. It then uses that to determine whether there will be enough
independent DACs available for line/mic switch controls. Hunk #1, #2,
and #3.
* improvements to stac92xx_auto_fill_dac_nids() to make it more general.
This fixes surround playback on some codecs in combination with the
fix to stac92xx_add_dyn_out_pins() above. It reads the full connection
list now, instead of just the first entry, and then locates an analog
DAC in the list. If one is found and it's free, assign it to that line-out.
If no free DAC is found for the line-out, return -ENODEV. It also makes
sure to actually select the chosen DAC if more than one DAC is input to
the pin. Hunks #4, #5.Signed-off-by: Steve Longerbeam
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
Add MODULE_FIRMWARE() entries, where appropriate.
Signed-off-by: Clemens Ladisch
Signed-off-by: Jaroslav Kysela -
Since request_firmware() is no longer used when the internal firmware
images are used, it is no longer necessary to depend on FW_LOADER in
this case.Signed-off-by: Clemens Ladisch
Signed-off-by: Jaroslav Kysela -
Added a missing call to resume mixer controls for STAC92xx codecs.
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
The last patch didn't really work (false report).
Although the hardware supports 125us minimum period, the current
usb-audio driver code assumes the 1ms period in many places.
Rollback the change.Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
This patch adds support for 3 new HP laptops to the Conexant
'Venice' driver.Signed-off-by: Tobin Davis
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
The minimal period size is 125us for high-speed mode while
1ms for full-speed mode.Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
It seems noone ever tried to use this driver with more than one device.
Signed-off-by: Adrian Bunk
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
E-Mu 1010 is not currently supported yet. Needs development work.
Signed-off-by: James Courtier-Dutton
Signed-off-by: Jaroslav Kysela -
Add handling of Intel-iMac-specific pinconfig of the sound card.
Intel-iMac now handled as a separated subsystem.Signed-off-by: Sylvain FORET
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
This patch is for ATI RS780 and R600 HDMI support.
Signed-off-by: Wolke Liu
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
Prefer the name of audio codecs as the mixer name even if
modem codecs are probed before the audio codecs.Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
Don't use request_firmware() if the internal firmwares are defined
via Kconfig. Otherwise it results in a significant delay at loading
time (minutes).Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
Fixed the model (asus-laptop) for ASUS A9rp with ALC660 codec.
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
This patch fixes the headphone squeal and noise on Conexant CX20551
(Waikiki) audio.Signed-off-by: Tobin Davis
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
- always set 256fs in SPDIF master clock mode
- disable deemphasis filter in AK4114 for Prodigy192Signed-off-by: Pavel Hofman
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
Some boards have the audio codec on slot #3 while the modem codec
on slot #0. The driver should continue to probe the slots when
no audio codec is found.
This fixes the problem of no device on Toshiba A100 (and some other
ATI SB450 devices).Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
Use PIN_HP output type for HP pin widgets accordingly, instead of
always applying PIN_OUT.Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
Added line_out_type field to auto_pin_cfg struct to provide the
pin type of default line_outs.Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela -
This patch adds support for the Asus A8JN Laptop.
Other modes were tested, this one worked best.Signed-off-by: Tobin Davis
Signed-off-by: Takashi Iwai
Signed-off-by: Jaroslav Kysela
