K. Y. Srinivasan [Tue, 10 May 2011 14:53:48 +0000 (07:53 -0700)]
Staging: hv: blkvsc_drv: Cleanup spin lock usage
This spin lock is potentially acquired from interrupt context.
Ensure that the interrupts are blocked whenever the lock is held.
The current code was not consistent with regards to blocking
interrupts - the same lock would be acquired without blocking
interrupts in some instance while the interrupts would be blocked
in other instances. Fix this potential deadlock problem.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: Abhishek Kane <v-abkane@microsoft.com> Signed-off-by: Hank Janssen <hjanssen@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Andy CH Lin [Tue, 3 May 2011 16:35:51 +0000 (17:35 +0100)]
intel_sst: intelmid_v2_control: correct jack event type
Correct event type of audio jack while receiving long press event.
Signed-off-by: Andy CH Lin <andy_ch_lin@wistron.com> Signed-off-by: Jekyll Lai <jekyll_lai@wistron.com> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Wu Fengguang [Tue, 3 May 2011 16:35:31 +0000 (17:35 +0100)]
intel_sst: MRST can only do 16bit recording
Fix bug
$ arecord -D dsnoop -r 48000 -f S16_LE 1.wav
ALSA lib pcm_direct.c:980:(snd1_pcm_direct_initialize_slave) unable to install hw params
ALSA lib pcm_dsnoop.c:604:(snd_pcm_dsnoop_open) unable to initialize slave
arecord: main:654: audio open error: Interrupted system call
Root cause is, the driver is reporting 32-bit capture capability
that is not supported by the MRST hardware. So the dsnoop plugin
end up requesting 32-bit capture which fails.
Root cause is, the driver is reporting 2-channel capture capability
that is not supported by the MRST hardware. So the plughw plugin
end up requesting 2-channel capture which fails.
Feng Tang [Tue, 3 May 2011 16:34:55 +0000 (17:34 +0100)]
intel_sst: make sure the sst_drop_stream() get called when needed
When "aplay test.wav", if we "ctrl+z" to suspend it, the last piece of
sound will be played endlessly. So we need make sure the drop_stream
is called in such a case.
Lu Guanqun [Tue, 3 May 2011 16:34:40 +0000 (17:34 +0100)]
intel_sst: return correct output/input device id
Without this patch, the change of output/input device can't be reflected
correctly. The mismatch reflects insane design, and should be fixed when
switching to the ASoC framework.
Signed-off-by: Lu Guanqun <guanqun.lu@intel.com> Acked-by: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Lu Guanqun [Tue, 3 May 2011 16:34:27 +0000 (17:34 +0100)]
intel_sst: move jack detection related configs to init time
The old policy will not enable MIC2BIAS on Moorstown platform
by default, it's only enabled when the user selects HS_MIC as input source.
Therefore when user selects DMIC and then inserts the jack,
no interrupt will be generated to notify the driver about the
jack insertion event and to take action to auto mute the speaker.
The new policy will enable this bit by default,
then no matter what the output device is, an interrupt will be generated.
This behaviour is more likely what a user expects.
Signed-off-by: Lu Guanqun <guanqun.lu@intel.com> Reviewed-by: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Ramesh Babu K V [Tue, 3 May 2011 16:33:13 +0000 (17:33 +0100)]
intel_sst: rework jack implementation
This patch fixes the below issues w.r.t jack implementation
a) The current jack implementation in driver is implemented
in intelmid.c. It has moved to vendor files for better managebility
b) Cleaned up jack reporting per upstream comments
c) Implemented jack for msic, added code to read adc and deduce jack
type based on mic bias
d) Support detection of american headset
Signed-off-by: Dharageswari R <dharageswari.r@intel.com> Signed-off-by: Ramesh Babu K V <ramesh.babu@intel.com>
[Corrections] Signed-off-by: Lu Guanqun <guanqun.lu@intel.com> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Dharageswari R [Tue, 3 May 2011 16:32:38 +0000 (17:32 +0100)]
intel_sst: Line out support
This patch adds the support for lineout. The
lineout input can be selected as any input channel
by using a new alsa mixer kcontrol.
Signed-off-by: Dharageswari R <dharageswari.r@intel.com> Signed-off-by: Ramesh Babu K V <ramesh.babu@intel.com> Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Lu Guanqun [Fri, 8 Apr 2011 00:41:38 +0000 (08:41 +0800)]
sst: fix compile error
Add several include files to fix the below compile error.
drivers/staging/intel_sst/intelmid.c: In function ‘snd_intelmad_sst_register’:
drivers/staging/intel_sst/intelmid.c:805:2: error: ‘sst_drv_ctx’ undeclared (first use in this function)
drivers/staging/intel_sst/intelmid.c:805:2: note: each undeclared identifier is reported only once for each function it appears in
Signed-off-by: Lu Guanqun <guanqun.lu@intel.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Lu Guanqun [Wed, 6 Apr 2011 02:20:42 +0000 (10:20 +0800)]
sst: make register_sst_card more self-contained
register_sst_card is used in ASoC code with field `scard_ops` being NULL.
Without this patch, there will be NULL dereference.
Signed-off-by: Lu Guanqun <guanqun.lu@intel.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Randy Dunlap [Mon, 9 May 2011 17:47:14 +0000 (10:47 -0700)]
staging/keucr: eliminate sparse warnings in ms.c
Eliminate sparse warnings in ms.c:
drivers/staging/keucr/ms.c:28:58: warning: right shift by bigger than source value
drivers/staging/keucr/ms.c:31:58: warning: right shift by bigger than source value
drivers/staging/keucr/ms.c:789:59: warning: cast truncates bits from constant value (ffffff7f becomes 7f)
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Alan Cox [Mon, 9 May 2011 15:04:45 +0000 (16:04 +0100)]
gma500: Fix dumb create crash
The error path from gtt_alloc returns NULL not a ptr error. The underlying
fail is caused by a bug in the size calculation. With these two fixed it
passes kmstest, although it's not really doing anything useful yet.
Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
staging: brcm80211: make use of fls bit operation in wlc_phy_compute_dB
wlc_phy_compute_dB converts absolute power value to dB implementing
a fixed point calculation for 10*log10(x). It does this by determining
the most significant bit for value x. This can be done using the fls()
bit operation, which has arch specific and possibly more efficient
implementation.
Cc: devel@linuxdriverproject.org Cc: linux-wireless@vger.kernel.org Reviewed-by: Roland Vossen <rvossen@broadcoom.com> Reviewed-by: Henry Ptasinski <henryp@broadcom.com> Reviewed-by: Brett Rudley <brudley@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
staging: brcm80211: use int_sqrt kernel function iso driver implementation
The driver code had its own implementation for fixed point square root
calculation, but such a function is already available in the linux
kernel so the driver implementation has been removed using the kernel
function.
Cc: devel@linuxdriverproject.org Cc: linux-wireless@vger.kernel.org Reviewed-by: Roland Vossen <rvossen@broadcoom.com> Reviewed-by: Henry Ptasinski <henryp@broadcom.com> Reviewed-by: Brett Rudley <brudley@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
staging: brcm80211: wlc_aggregatable function prototype moved
The function wlc_aggregatable is used by wl_mac80211.c. The prototype
of the function was placed in include file wlc_scb.h but the interface
for wl_mac80211.c to other driver functions is to be placed in wlc_pub.h.
Cc: devel@linuxdriverproject.org Cc: linux-wireless@vger.kernel.org Reviewed-by: Roland Vossen <rvossen@broadcoom.com> Reviewed-by: Henry Ptasinski <henryp@broadcom.com> Reviewed-by: Brett Rudley <brudley@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Henry Ptasinski [Mon, 9 May 2011 14:33:09 +0000 (16:33 +0200)]
staging: brcm80211: move functions only used by brcmsmac out of utils
The functions gitvar() and getintvar() are only used by brcmsmac, so move them
out of the bcmutils.c file, which is shared by both brcmsmac and brcmfmac.
Henry Ptasinski [Mon, 9 May 2011 14:33:03 +0000 (16:33 +0200)]
staging: brcm80211: move files only used by brcmsmac into appropriate dir
A number of files in the util subdir are only used by the brcmsmac driver.
Move those files into the brcmsmac subdir instead, and do the necessary
Makefile updates to get the files from the new location.
staging: brcm80211: fix perimeter locking in wl_ops_bss_info_changed
Several code path in the wl_ops_bss_info_changed() callback were not
using the perimeter lock. This gives potential of raise conditions
so lock/unlock calls have been added.
Cc: devel@linuxdriverproject.org Cc: linux-wireless@vger.kernel.org Reviewed-by: Roland Vossen <rvossen@broadcoom.com> Reviewed-by: Henry Ptasinski <henryp@broadcom.com> Reviewed-by: Brett Rudley <brudley@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
staging: sm7xx: Use kernel framebuffer mode setting
This patch implements dynamic framebuffer mode setting.
Previous code works with mode setting in a hard code way. Previous hard
code configuration is used as default configuration if dynamic mode
setting or boot mode setting (via sm712vga_setup) is not used.
Tested with SM712 supporting 1024x600x16 as default hardware resolution.
Changes:
- Implement fb_check_var and fb_set_par callbacks
- Remove __maybe_unused decorator in function being used (sm712vga_setup)
- Minor cleanup on initialization structs related with mode settings
- Updated author copyright
- Updated TODO file
Signed-off-by: Javier M. Mellid <jmunhoz@igalia.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
matt mooney [Fri, 6 May 2011 10:47:50 +0000 (03:47 -0700)]
staging: usbip: vhci_hcd.c: coding style cleanup
Fix alignment for consistency and remove extraneous lines.
Move MODULE_ macros to the end of file.
Make bit_desc[] an array of const pointers.
Add KERN_ level to a couple of printks.
Signed-off-by: matt mooney <mfm@muteddisk.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
matt mooney [Fri, 6 May 2011 10:47:44 +0000 (03:47 -0700)]
staging: usbip: stub_tx.c: coding style cleanup
Fix alignment for consistency, checkpatch.pl warnings for lines
over 80 characters, remove extraneous lines, and change conversion
specifier within a format string to remove warning.
Signed-off-by: matt mooney <mfm@muteddisk.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Alan Cox [Fri, 6 May 2011 15:48:27 +0000 (16:48 +0100)]
gma500: The MID devices have the register offset different
This is another small step towards getting Moorestown/Oaktrail support to
work but for Moorestown at least we still need to sort out GEM backed base
framebuffer, which means figuring out why GEM explodes early on at the
moment.
Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Larry Finger [Thu, 5 May 2011 23:52:56 +0000 (18:52 -0500)]
staging: vt6655: Fix warnings if CONFIG_PM is not defined
If staging:vt6655 is built without CONFIG_DM being defined, there are large
numbers of warnings of the following form due to use of #if instead of #ifdef:
In file included from drivers/staging/vt6655/upc.h:32,
from drivers/staging/vt6655/mac.h:39,
from drivers/staging/vt6655/wroute.c:34:
drivers/staging/vt6655/device.h:399:5: warning: "CONFIG_PM" is not defined
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Grant Grundler [Wed, 4 May 2011 16:59:47 +0000 (09:59 -0700)]
staging: brcm80211: SDIO/MMC cleanups
misc coding style cleanups to dhd_sdio/sdmmc
o replace PKTFREE2 macro with static dhdsdio_pktfree2()
o drop "delta" local var
o drop GSPI_PR55150_BAILOUT
o reformat some of the comments (white space changes)
o drop dhd_bcmsdh_recv_buf wrapper and directly call bcmsdh_recv_buf
Signed-off-by: Grant Grundler <grundler@chromium.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
staging: hv: make HYPERV_UTILS depend on NLS to prevent a build error
When HYPERV_UTILS=y and NLS=n the build fails with this error:
MODPOST 3 modules
ERROR: "utf8s_to_utf16s" [drivers/staging/hv/hv_utils.ko] undefined!
This patch fixes this by adding a dependency on NLS to HYPERV_UTILS.
Signed-off-by: Peter Foley <pefoley2@verizon.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
K. Y. Srinivasan [Fri, 29 Apr 2011 20:45:15 +0000 (13:45 -0700)]
Staging: hv: VMBUS is a acpi enumerated device; get irq value from bios
On some Windows hosts, the Linux PCI sub-system is not
allocating irq resources to the
vmbus driver. It looks like VMBUS is an ACPI enumerated device.
Retrieve the irq information from DSDT.
Currently we use this bios specified irq, if the PCI
sub-system fails to allocate the irq.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: Abhishek Kane <v-abkane@microsoft.com> Signed-off-by: Hank Janssen <hjanssen@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
projects / linux-beck.git / log