Dmitry Torokhov [Wed, 22 Mar 2006 08:07:56 +0000 (00:07 -0800)]
[PATCH] dcdbas: convert to the new platform device interface
Do not use platform_device_register_simple() as it is going away, define
dcdbas_driver and implement ->probe() and ->remove() functions so manual
binding and unbinding will work with this driver.
Also switch to using attribute_group when creating sysfs attributes and
make sure to check and handle errors; explicitely remove attributes when
detaching driver.
Dmitry Torokhov [Wed, 22 Mar 2006 08:07:53 +0000 (00:07 -0800)]
[PATCH] vr41xx: convert to the new platform device interface
The patch does the following for v441xx seris drivers:
- stop using platform_device_register_simple() as it is going away
- mark ->probe() and ->remove() methods as __devinit and __devexit
respectively
- initialize "owner" field in driver structure so there is a link
from /sys/modules to the driver
- mark *_init() and *_exit() functions as __init and __exit
Atsushi Nemoto [Wed, 22 Mar 2006 08:07:45 +0000 (00:07 -0800)]
[PATCH] serial: serial_txx9 driver update
Update the serial_txx9 driver.
* More strict check in verify_port. Cleanup.
* Do not insert a char caused previous overrun.
* Fix some spin_locks.
* Do not call uart_add_one_port for absent ports.
Also, this patch removes a BROKEN tag from Kconfig. This driver has been
marked as BROKEN by removal of uart_register_port, but it has been solved
already on Sep 2005.
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp> Cc: Russell King <rmk@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The code that handles bios that span table target boundaries by breaking
them up into smaller bios will not split an individual struct bio_vec into
more than two pieces. Sometimes more than that are required.
This patch adds a loop to break the second piece up into as many pieces as
are necessary.
Cc: "Abhishek Gupta" <abhishekgupt@gmail.com> Cc: Dan Smith <danms@us.ibm.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com> Cc: <stable@kernel.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
[PATCH] unshare: Error if passed unsupported flags
A bare bones trivial patch to ensure we always get -EINVAL on the
unsupported cases for sys_unshare. If this goes in before 2.6.16 it allows
us to forward compatible with future applications using sys_unshare.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: JANAK DESAI <janak@us.ibm.com> Cc: <stable@kerenl.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Andrew Morton [Wed, 22 Mar 2006 08:07:39 +0000 (00:07 -0800)]
[PATCH] __get_page_state() cpumask cleanup and fix
__get_page_state() has an open-coded for_each_cpu_mask() loop in it.
Tidy that up, then notice that the code was buggy:
while (cpu < NR_CPUS) {
unsigned long *in, *out, off;
if (!cpu_isset(cpu, *cpumask))
continue;
an obvious infinite loop. I guess we just never call it with a holey cpu
mask.
Even after my cpumask size-reduction work, this patch increases code size :(
Cc: Paul Jackson <pj@sgi.com> Cc: Christoph Lameter <clameter@engr.sgi.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Latchesar Ionkov [Wed, 22 Mar 2006 08:07:37 +0000 (00:07 -0800)]
[PATCH] v9fs: assign dentry ops to negative dentries
If a file is not found in v9fs_vfs_lookup, the function creates negative
dentry, but doesn't assign any dentry ops. This leaves the negative entry
in the cache (there is no d_delete to mark it for removal). If the file is
created outside of the mounted v9fs filesystem, the file shows up in the
directory with weird permissions.
This patch assigns the default v9fs dentry ops to the negative dentry.
Signed-off-by: Latchesar Ionkov <lucho@ionkov.net> Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com> Cc: <stable@kernel.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Herbert Poetzl [Wed, 22 Mar 2006 08:07:34 +0000 (00:07 -0800)]
[PATCH] don't call check_acpi_pci() on x86 with ACPI disabled
check_acpi_pci() is called from arch/i386/kernel/setup.c even if
CONFIG_ACPI is not defined, but the code in include/asm/acpi.h doesn't
provide it in this case.
Signed-off-by: Herbert Pƶtzl <herbert@13thfloor.at> Cc: "Brown, Len" <len.brown@intel.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Mike Galbraith [Wed, 22 Mar 2006 08:07:33 +0000 (00:07 -0800)]
[PATCH] sched: remove sleep_avg multiplier
Remove the sleep_avg multiplier. This multiplier was necessary back when
we had 10 seconds of dynamic range in sleep_avg, but now that we only have
one second, it causes that one second to be compressed down to 100ms in
some cases. This is particularly noticeable when compiling a kernel in a
slow NFS mount, and I believe it to be a very likely candidate for other
recently reported network related interactivity problems.
In testing, I can detect no negative impact of this removal.
Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Sergei Shtylyov [Tue, 21 Mar 2006 10:58:48 +0000 (11:58 +0100)]
[ALSA] AMD Au1x00: fix DMA init/cleanup
Modules: MIPS AU1x00 driver
AMD Au1x00 ALSA driver causes kernel oops in au1000_init() by trying
to set DMA channel to -1 in yet unallocated audio streams. Here's the
patch that staightens up DMA init/cleanup code.
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Sergei Shtylylov [Mon, 20 Mar 2006 17:38:21 +0000 (18:38 +0100)]
[ALSA] AMD Au1x00: make driver build after cleanup
Modules: MIPS AU1x00 driver
AMD Au1x00 ALSA driver doesn't build after the recent code cleanup:
sound/mips/au1x00.c: In function 'au1000_setup_dma_link':
sound/mips/au1x00.c:173: error: 'pointer' undeclared (first use in this function)
sound/mips/au1x00.c:173: error: (Each undeclared identifier is reported only once
sound/mips/au1x00.c:173: error: for each function it appears in.)
sound/mips/au1x00.c: In function 'snd_au1000_hw_params':
sound/mips/au1x00.c:339: warning: implicit declaration of function 'snd_mask_min'
Thibault LE MEUR [Tue, 14 Mar 2006 10:44:53 +0000 (11:44 +0100)]
[ALSA] Fixes audiophile usb analog capture with the new device_setup parameter
Modules: Documentation,USB generic driver
The patch adds the 'device_setup' module parameter and a specific
quirk to correctly initialize the audiophile usb device: this fixes
the distorted sound bug on the Analog capture port. Backward
compatibility is achieved by simply omitting the new parameter.
Signed-off-by: Thibault LE MEUR <Thibault.LeMeur@supelec.fr> Signed-off-by: Takashi Iwai <tiwai@suse.de>
Doug McLain [Tue, 28 Feb 2006 10:53:06 +0000 (11:53 +0100)]
[ALSA] ice1712 - Fix wordclock status on Delta1010LT
Modules: ICE1712 driver
I posted this patch to bug 1806 a while back, and have been awaiting a
reply or commit. It currently reports the Locked/No Signal text
backwards in envy24control, since i was using an older version of
envy24control when I wrote the original code. The Locked/No Signal test
was recently reversed in envy24control cvs, so the test in my code needs
to be reversed as well. Here is the patch, once again.
Jonathan Woithe [Tue, 28 Feb 2006 10:47:47 +0000 (11:47 +0100)]
[ALSA] HDA/ALC260: 7/7 - add SPDIF enable to test model
Modules: HDA Codec driver
This patch adds mixer controls to the 'test' ALC260 model which allow the
user to selectively enable or disable the SPDIF output pins. This should
assist people identify digital outputs on machines which bring them to the
outside world.
Note that while the patch *should* work, I cannot personally verify it since
my laptop doesn't bring the SPDIF lines out.
As for the GPIO switches added in patch 4, these controls are currently
only compiled in if debug mode is selected.
Jonathan Woithe [Tue, 28 Feb 2006 10:46:14 +0000 (11:46 +0100)]
[ALSA] HDA/ALC260: 6/7 - Fujitsu/test model tweaks
Modules: HDA Codec driver
This patch is janitorial - it cleans up a number of cosmetic issues with
the 'fujitsu' and 'test' models. Issues addressed:
* spaces instead of tabs used for some indents,
* clarified/corrected selected comments.
Unlike a patch from earlier this week, this keeps both ADCs connected to
the mic1 pin widget by default. I believe this is the better default, since
some laptops don't bring the line1 widget to the outside world. Most (if
not all) do bring mic1 out though, so it seems to me that this default
should remain in place.
Jonathan Woithe [Tue, 28 Feb 2006 10:45:11 +0000 (11:45 +0100)]
[ALSA] HDA/ALC260: 5/7 - add 'acer' model
Modules: HDA Codec driver
This patch adds a new 'acer' model. It is based on work by Rimas and many
other dedicated Acer owners over the past few weeks (see bug number
1618) - kudos to them. This adds support for a variety of Acer laptops,
although we are still in the process of collecting pci/subsystem IDs to add.
There are still some potentially outstanding issues: there are reports that
the CD control might not yet be functional for example. However, the time
is probably right to get a 'first cut' into the kernel which can be refined
as more test reports come in.
From the reports by various Acer owners, this code allows them to record
from internal mics and external jacks. Playback also seems to work to all
external jacks and the internal speaker. The 'beep' control doesn't appear
functional at the moment.
This patch depends on the changes made in patch 3/7.
Jonathan Woithe [Tue, 28 Feb 2006 10:43:27 +0000 (11:43 +0100)]
[ALSA] HDA/ALC260: 4/7 - add GPIO switches to test model
Modules: HDA Codec driver
This patch adds 'test' model mixer switches for the GPIO controls found on
the ALC260. It has been found that some laptops (eg: Acer) can use these to
enable particular controls, so it would be useful to have access to these
via the 'test' model. It will make testing new models easy, especially if
certain outputs cannot be made to work any other way.
This patch *should* work, but because the GPIO pins don't do anything in
my laptop I cannot personally verify that all this works as expected.
Jonathan Woithe [Tue, 28 Feb 2006 10:38:35 +0000 (11:38 +0100)]
[ALSA] HDA/ALC260: 3/7 - generalise some structures
Modules: HDA Codec driver
This patch generalises a structure added for the 'fujitsu' model but which
is potentially useful for other models as well. It turns the
'alc260_fujitsu_adc_nids' array into 'alc260_dual_adc_nids'; for other
models which decide to utilise the dual ADC functionality there's really
no reason why they need to define their own list of ADC nids.
The 'fujitsu' model preset is adjusted accordingly.
This patch adds functionality which switches the input/output buffer enables
of retasking pins when the user changes their mode with the mode control.
This probably reduces noise *slightly* for recording compared to the case
where both input and output buffers were enabled simultaneously.
Jonathan Woithe [Tue, 28 Feb 2006 10:35:18 +0000 (11:35 +0100)]
[ALSA] HDA/ALC260: 1/7 - Fix test model input mux label
Modules: HDA Codec driver
This patch fixes input mux labels used for the ALC260 'test' model. This is
needed to avoid confusion which comes about because the two ADCs in the
ALC260 have slightly different mappings for their input selectors. Since
this is just the test model it's sufficient to simply report both options
where they exist. If a model comes along for which this becomes an issue,
the ALC260 input mux code will have to be extended to allow different mux
layouts for different ADCs.
Takashi Iwai [Wed, 22 Feb 2006 16:14:34 +0000 (17:14 +0100)]
[ALSA] Fix sleep in atomic in virmidi driver
Modules: ALSA sequencer
Debug: sleeping function called from invalid context at /usr/src/linux/include/linux/rwsem.h:43
in_atomic():1, irqs_disabled():0
[<f999d15e>] snd_seq_deliver_event+0xb4/0x1a8 [snd_seq]
[<f999d2be>] snd_seq_kernel_client_dispatch+0x6c/0x7c [snd_seq]
[<f93321fc>] snd_virmidi_output_trigger+0xca/0xe5 [snd_seq_virmidi]
Takashi Iwai [Thu, 16 Feb 2006 18:51:10 +0000 (19:51 +0100)]
[ALSA] hda-intel - Auto-correction of the DMA position mode
Modules: HDA Intel driver
Switch the method to measure the current DMA position automatically
from position-buffer mode to LPIB-read mode with a sanity check.
Some hardwares seems to have problem with the position buffer.
Takashi Iwai [Thu, 16 Feb 2006 17:17:58 +0000 (18:17 +0100)]
[ALSA] hda-intel - Automatic correction to single_cmd mode
Modules: HDA Codec driver,HDA Intel driver
Switch to single_cmd mode automatically as a fallback when CORB/RIRB
communication doesn't work well. It may make the driver working on
some devices with broken BIOS/ACPI support.
Jonathan Woithe [Thu, 9 Feb 2006 11:01:26 +0000 (12:01 +0100)]
[ALSA] hda: ALC260 test model implementation
Modules: HDA Codec driver
The following patch adds a 'test' ALC260 model specification to the
patch_realtek.c driver if CONFIG_SND_DEBUG is set. This is similar to the
'test' ALC880 model in that it sets up mixer controls for almost everything
to make it easier for people to test their laptop/soundcard when working out
what pin widgets are connected to which real-world devices.
This patch assumes my previous patch (adding the second PCM to the ALC260)
has previously been applied since it uses infrastructure added by that
patch.
In developing this patch it was found that not all retasking pins accept all
the modes - in particular, some ignore the VREFxx variants. The pin mode
control has therefore been tweaked to prevent this becoming a problem in
mixer applications.
Takashi Iwai [Thu, 9 Feb 2006 10:57:01 +0000 (11:57 +0100)]
[ALSA] hda: minor correction to fujitsu ALC260 initverbs
Modules: HDA Codec driver
Jonathan Woithe <jwoithe@physics.adelaide.edu.au>
The following patch changes the initverbs associated with the output sum
widgets 0x08, 0x09 and 0x0a used with the fujitsu model in patch_realtek.c.
It is against alsa 1.0.11rc3.
Currently these widgets have their output muted, but a close reading of the
datasheet suggests that they don't actually have an output mute. They have
an *input* mute for each of their two inputs and a single output gain. This
patch therefore activates the input mutes and zeros the output gain. It has
not yet been tested but the above interpretation of the ALC260 datasheet
seems correct.
Jonathan Woithe [Thu, 9 Feb 2006 10:53:48 +0000 (11:53 +0100)]
[ALSA] hda: add PCM for 2nd ADC on ALC260
Modules: HDA Codec driver
The following patch against alsa 1.0.11rc3 creates a PCM device (pcm1c) for
the second ADC present on the ALC260 codec used by the hda driver. It also
defines a new mixer control allowing the mode of retasking pins to be set;
this means a user can (for example) designate the headphone jack to be a
second input. With this patch in place it is possible to do 4 channel
recording on laptops equipped with an ALC260 codec assuming both a stereo
line-in jack is provided in addition to a headphone jack.
Mixer controls are provided to allow the headphone jack to be switched as
an input. In addition, an (input only) mode control is configured for
the line-in jack to allow a bias voltage to be requested (VREF80 or VREF50)
so headsets based on condensor microphones have a chance of working.
This patch has been tested on a Fujitsu S7020 laptop and as such these
features are currently only configured for the 'fujitsu' model.