Pull sound fixes from Takashi Iwai:
"Slightly a high amount of commits come from Adrian Knoth's HDSPM
driver fixes. Other than that, all small trival fixes or quirks that
are pretty driver-specific."
* tag 'sound-3.7' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ASoC: wm8994: Only enable extra BCLK cycles when required
ALSA: als3000: check for the kzalloc return value
ALSA: sound/isa/opti9xx/miro.c: eliminate possible double free
ALSA: hda - Fix silent headphone output from Toshiba P200
ALSA: hdspm - Fix coding style in CTL_ELEM macros
ALSA: hdspm - Fix typo in kcontrol element on RME MADI cards
ALSA: hdspm - Fix sync_in detection on AES/AES32
ALSA: hdspm - Fix sync_in reporting on RME MADI cards
ALSA: hdspm - Also report autosync_sample_rate on MADI and MADIface
ALSA: hdspm - Fix reported autosync_sample_rate
ALSA: hdspm - Fix sync check reporting on all RME HDSPM cards
ALSA: hdspm - Report external rate in slave mode on PCI MADI
ALSA: hdspm - Allow DDS/Varispeed to be set from userspace
ALSA: hda - add dock support for Thinkpad T430
ASoC: ux500_msp_i2s: Fix devm_* and return code merge error
ASoC: Ux500: Dispose of device nodes correctly
Pull DMA-mapping revert from Marek Szyprowski:
"Due to my mistake, my previous pull request (merged as commit
cff7b8ba60: "Merge branch 'fixes_for_linus' ..") contained a patch
which is aimed for v3.8 and lacks its dependences. This pull request
reverts it and fixes build break of ARM architecture."
* 'fixes_for_linus' of git://git.linaro.org/people/mszyprowski/linux-dma-mapping:
Revert "ARM: dma-mapping: support debug_dma_mapping_error"
slic_card_locate does a kzalloc of physcard, and if it fails,
in my previous patch i returned -ENOMEM, but left the driver leak
the memory if card_hostid == SLIC_HOSTID_DEFAULT,
fix this memory leak if the above condition is true
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This seems like a cut and paste bug. Smatch complains that we don't
allocate enough memory for a set_stakey_rsp struct. In fact this is
used as a set_assocsta_rsp struct throughout and that also matches the
name of the function.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Pull x86 fixes from Ingo Molnar:
"This fixes a couple of nasty page table initialization bugs which were
causing kdump regressions. A clean rearchitecturing of the code is in
the works - meanwhile these are reverts that restore the
best-known-working state of the kernel.
There's also EFI fixes and other small fixes."
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86, mm: Undo incorrect revert in arch/x86/mm/init.c
x86: efi: Turn off efi_enabled after setup on mixed fw/kernel
x86, mm: Find_early_table_space based on ranges that are actually being mapped
x86, mm: Use memblock memory loop instead of e820_RAM
x86, mm: Trim memory in memblock to be page aligned
x86/irq/ioapic: Check for valid irq_cfg pointer in smp_irq_move_cleanup_interrupt
x86/efi: Fix oops caused by incorrect set_memory_uc() usage
x86-64: Fix page table accounting
Revert "x86/mm: Fix the size calculation of mapping tables"
MAINTAINERS: Add EFI git repository location
Pull perf fixes from Ingo Molnar:
"Most of the kernel diffstat relates to a group of Intel P6 and KNC
(Xeon-Phi Knights Corner) PMU driver fixes, neither of which is in
heavy use, so we took the fixes.
The rest is diverse smallish fixes to the tooling and kernel side."
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/x86: Remove unused variable in nhmex_rbox_alter_er()
perf/x86: Enable overflow on Intel KNC with a custom knc_pmu_handle_irq()
perf/x86: Remove cpuc->enable check on Intl KNC event enable/disable
perf/x86: Make Intel KNC use full 40-bit width of counters
perf/x86/uncore: Handle pci_read_config_dword() errors
perf/x86: Remove P6 cpuc->enabled check
perf/x86: Update/fix generic events on P6 PMU
perf/x86: Fix P6 FP_ASSIST event constraint
perf, cpu hotplug: Use cached value of smp_processor_id()
perf, cpu hotplug: Run CPU_STARTING notifiers with irqs disabled
x86/perf: Fix virtualization sanity check
perf test: Fix exclude_guest parse events tests
perf tools: do not flush maps on COMM for perf report
perf help: Fix --help for builtins
perf trace: Check if sample raw_data field is set
perf trace: Validate syscall id before growing syscall table
Pull btrfs fixes from Chris Mason:
"This has our series of fixes for the next rc. The biggest batch is
from Jan Schmidt, fixing up some problems in our subvolume quota code
and fixing btrfs send/receive to work with the new extended inode
refs."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs: do not bug when we fail to commit the transaction
Btrfs: fix memory leak when cloning root's node
Btrfs: Use btrfs_update_inode_fallback when creating a snapshot
Btrfs: Send: preserve ownership (uid and gid) also for symlinks.
Btrfs: fix deadlock caused by the nested chunk allocation
btrfs: Return EINVAL when length to trim is less than FSB
Btrfs: fix memory leak in btrfs_quota_enable()
Btrfs: send correct rdev and mode in btrfs-send
Btrfs: extended inode refs support for send mechanism
Btrfs: Fix wrong error handling code
Fix a sign bug causing invalid memory access in the ino_paths ioctl.
Btrfs: comment for loop in tree_mod_log_insert_move
Btrfs: fix extent buffer reference for tree mod log roots
Btrfs: determine level of old roots
Btrfs: tree mod log's old roots could still be part of the tree
Btrfs: fix a tree mod logging issue for root replacement operations
Btrfs: don't put removals from push_node_left into tree mod log twice
Fix the use of insn->chanspec with the CR_CHAN, CR_RANGE, and CR_AREF
macros. insn->chanspec is an unsigned int not an array.
Cleanup the comments when creating the 'val' to write to the channel
list fifo so that the code is a bit more readable.
Use the 'val' variable when getting the value from the ADC fifo and
then munging the data. This cleans up the goofy line breaks.
This function still does not follow the expectations of the comedi core.
It is supposed to read insn->n values from the channel and return those
values. Currently it only reads a single value.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This function only needs to update the output channel state when
the 'mask' (data[0]) indicates that the state is changing. It's
also supposed to return the state for all the input and output
channels in data[1].
This function appears to work correctly but it's coded quite
differently form the other comedi drivers. Rework the function so
it follows the normal form. Also, use a couple local variables
for the mmio addresses used to read/write the dio ports.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently this function does not work like the comedi code expects.
Fix the function so that it checks the instruction, data[0], and
does the correct action based on it.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The boards supported by this driver either have analog outputs or don't
have them. Add a new boardinfo value, 'has_ao', to indicate this.
The boards that have analog outputs always have 4, 12-bit channels. Remove
the unnecessary boardinfo and just open-code the values in the subdevice
initialization.
The boards with analog outputs also share the same output range capabilities.
Rename the comedi_lrange to make it common for all board types.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
All the boards supported by this driver have the same analog input
specifications. Remove the unnecessary boardinfo and just open-code
the values in the subdevice initialization.
The boards also share the same analog input range capabilities. Remove
one of the duplicate tables and rename the other to make it common for
all board types.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reformat the copyright and driver description comments to follow the
preferred block comment style. Reword the driver description to follow
comedi style described in the skel driver.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The analog input subdevice functions me_ai_do_cmd_test() and me_ai_do_cmd()
are only stubbed in. They both just return 0.
The me_ai_cancel() function does have code to stop any running conversions
but the me_ai_insn_read() function does that also.
Just remove the incomplete functions and remove SDF_CMD_READ from the
subdevice_flags.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This function is supposed to write to a single analog output channel.
The channel number is packed in insn->chanspec, which is an unsigned
int, and unpacked using the CR_CHAN() macro.
Currently this function is trying to use the chanspec as an array.
This works only if a single value is written.
Fix the function so that the desired channel is determined and all
the data is written to that channel.
Also, fix the return. The comedi core expects insn_read functions to
return the number of data values (insn->n).
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This function is supposed to return the analog output value for
a specified channel. The channel number is packed in insn->chanspec,
which is an unsigned int, and unpacked using the CR_CHAN() macro.
Currently this function is trying to use the chanspec as an array.
This works only if a single data value is read.
Fix the function so that the desired channel is determined and all
the data returned is from that channel.
Also, fix the return. The comedi core expects insn_read functions to
return the number of data values (insn->n).
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This driver uses the comedi auto config mechanism and does not
walk the pci bus to find the pci_dev. It should not be calling
pci_dev_put() to decrement the ref count.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Rename the boardinfo variables so they are a bit more concise.
Remove the unnecessary comments in the boardinfo.
Add some whitespace to make the code more readable.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
All the boards supported by this driver have 32 dio channels.
Remove the boardinfo and just open code the value in the attach.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Factor out the code in me_attach_pci() that handles the PLX-Bug
workaround to a separate function.
This looks odd. It appears that the bug workaround either swaps
PCI bars 0 and 5 or it modifies PCI bar 0. Shouldn't this happen
before PCI bar 0 is ioremap'ed?
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Instead of returning -EIO when comedi_pci_enable() fails, pass on
the actual error code. Also, remove the kernel noise when the
function fails.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Two local variables are used in me_attach_pci() to check for errors
when calling other functions, 'result' and 'error'. Remove both of
them and just use a common 'ret' variable.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
PCI bar 2 is the main i/o address for this driver. Currently the
pci_resource_len of the bar is stored in the private data but
it is never used.
Remove 'me_regbase_size' from the private data. Also, remove the
local variables used for the pci_resource_start and _len. Just
pass the values directly to the ioremap().
Remove the kernel noise when the ioremap fails.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
As pointed out by Dan Carpenter, dev_private could be NULL when it
is checked before calling comedi_pci_disable(). The private data
variable 'plx_regbase_size' is only used to see if the pci device
has been enabled.
Remove that variable from the private data and just use the unused
dev->iobase variable to let the detach know that the pci device
is enabled.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
None of the callers check the return value. Just make it a void.
Flip the status check at the end of the function to check for an
error before outputing the dev_dbg(). Use __func__ for the function
name in the message.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove most of the dev_dbg() output from this driver. Most of it
is just noise.
For now I have left the dev_dbg() in the dt3k_send_cmd() function.
I still have not quite figured out how the low-level i/o works in
this driver...
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The PCI device ids supported by this driver are used multiple
places in the code. To improve maintainability, create #define's
for them.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Convert this driver to use the comedi PCI auto config mechanism
by adding an attach_pci callback.
Since this driver requires no extra configuration options, and
the attach callback is now optional, remove it.
Remove the kernel noise that is output when the attach fails
for various reasons.
Use the found 'dev->board_name' for the resource name passed to
comedi_pci_enable() and request_irq().
Pass on the error returned from request_irq() instead of returning
-EINVAL on error.
Since this driver no longer walks the pci bus to find the pci_dev,
remove the pci_dev_put() in the detach.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This macro relies on a local variable having a specific name.
Remove it and use the comedi_board() helper to get the pointer.
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
fixed below checkpatch warning.
- WARNING: Prefer netdev_warn(netdev, ... then dev_warn(dev, ... then pr_warn(... to printk(KERN_WARNING ...
some of them have been replaced by dev_dbg.
Signed-off-by: YAMANE Toshiaki <yamanetoshi@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
these are not called at anypoint or enabled at anyplace in the code,
remove them
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
