- 10 Feb, 2012 19 commits
-
-
Arve Hjønnevåg authored
If an alarm was restarted with a value that moved it away from the head of a queue, the hrtimer would not be updated. This would cause unnecessary wakeups. CC: Arve Hjønnevåg <arve@android.com> CC: Android Kernel Team <kernel-team@android.com> Change-Id: If379f8dd92b0bdb3173bd8d057adfe0dc1d15259 Signed-off-by:
Arve Hjønnevåg <arve@android.com> Signed-off-by:
John Stultz <john.stultz@linaro.org> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Praneeth Kumar Bajjuri authored
Do not enable Android alarm driver by default CC: Arve Hjønnevåg <arve@android.com> CC: Android Kernel Team <kernel-team@android.com> CC: Praneeth Bajjuri <praneeth@ti.com> Change-Id: Iff8f7a65c4eceecfd084074937c72824697b5e7f Signed-off-by:
Praneeth Bajjuri <praneeth@ti.com> [jstultz: tweaked commit subject & msg] Signed-off-by:
-
John Stultz authored
Now that it builds, re-enable android alarm driver in the makefile and kconfig CC: Arve Hjønnevåg <arve@android.com> CC: Android Kernel Team <kernel-team@android.com> Signed-off-by:
-
John Stultz authored
Allow Android alarmtimer device to build while wakelocks are still out of tree. CC: Arve Hjønnevåg <arve@android.com> CC: Android Kernel Team <kernel-team@android.com> Signed-off-by:
-
John Stultz authored
The upstreamed alarmtimers are similar but not quite 100% API compatibile with the android in-kernel alarm api. To aid the transition, prefix the the android in-kernel api with android_ CC: Arve Hjønnevåg <arve@android.com> CC: Android Kernel Team <kernel-team@android.com> Signed-off-by:
-
John Stultz authored
The file asm/mach/time.h doesn't exist on all arches, so include <linux/time.h>. Also linux/sysdev.h is gone so kill it. CC: Arve Hjønnevåg <arve@android.com> CC: Android Kernel Team <kernel-team@android.com> Signed-off-by:
-
Andy Green authored
Add module.h includes required to build CC: Arve Hjønnevåg <arve@android.com> CC: Android Kernel Team <kernel-team@android.com> CC: Andy Green <andy.green@linaro.org> Signed-off-by:
Andy Green <andy.green@linaro.org> [jstultz: Tweaked commit subject, folded two patches into one] Signed-off-by:
-
Arve Hjønnevåg authored
Remove references to non-existant save_time_delta. Change-Id: Iaefeca497de02fe36b7f5d79075912f6e349ec53 CC: Arve Hjønnevåg <arve@android.com> CC: Android Kernel Team <kernel-team@android.com> Signed-off-by:
-
Arve Hjønnevåg authored
Drivers can now create alarms that will use an hrtimer while the system is running and the rtc to wake up from suspend. CC: Arve Hjønnevåg <arve@android.com> CC: Android Kernel Team <kernel-team@android.com> Signed-off-by:
-
Gerard Ryan authored
This is a patch to the led_control.h file that fixes numerous warnings and errors reported by the checkpatch.pl tool. There still remain a few more, but as this is my first attempt at a commit, I'm not going to be too adventurous! Signed-off-by:
Gerard Ryan <gerard@ryan.lt> Signed-off-by:
-
Larry Finger authored
When this driver was upgraded to the vendor 20100831 version in commit 93c55dda et al,, one listhead initialization was missed. This broke complete operation of the driver whenever AP mode was enabled. This fixes https://bugs.archlinux.org/task/27996. The configuration parameter R8712_AP is misleading as the driver cannot function as an AP without a heavily hacked version of hostapd. Thus, it makes sense to remove the parameter; however the code and data configured for the option is left in. Signed-off-by:
Larry Finger <Larry.Finger@lwfinger.net> Cc: Stable <stable@vger.kernel.org> Signed-off-by:
-
Peter Huewe authored
This patch removes the remaining defines that are already defined identically in the sis initdef.h header. Signed-off-by:
Peter Huewe <peterhuewe@gmx.de> Signed-off-by:
-
Peter Huewe authored
Since the defines TVVCLKDIV2, TVVCLK, HiTVVCLKDIV2, HiTVVCLK, HiTVSimuVCLK and HiTVTextVCLK are now defined as relative values, we have to use TVCLKBASE_315 (0x31) as a base address to get the same values as before the merge. The old and now duplicated defines were removed Signed-off-by:
-
Peter Huewe authored
This patch renames some of the defines that exist in the sis initdef.h but seem to have a different value. In order to preserve the functionality of the driver, we simply prepend these defines with XGI_ (for now) to resolve conflicts and review them later on. Renames: SetCRT2ToLCDA -> XGI_SetCRT2ToLCDA LCDVESATiming -> XGI_LCDVESATiming EnableLVDSDDA -> XGI_EnableLVDSDDA LCDDualLink -> XGI_LCDDualLink ModeSwitchStatus -> XGI_ModeSwitchStatus YPbPr750pVCLK -> XGI_YPbPr750pVCLK Signed-off-by:
-
Peter Huewe authored
This patch renames the remaining duplicate defines and their usage to the naming convention of the sis initdef.h and removes the now duplicated defines. Renames: CRT2DisplayFlag -> DisableCRT2Display ModeInfoFlag -> ModeTypeMask Support16Bpp -> Mode16Bpp Support32Bpp -> Mode32Bpp SupportHiVisionTV -> SupportHiVision SupportYPbPr -> SupportYPbPr750p SwitchToCRT2 -> SwitchCRT2 VB_XGI301 -> VB_SIS301 VB_XGI301B -> VB_SIS301B VB_XGI301LV -> VB_SIS301LV VB_XGI302B -> VB_SIS302B VB_XGI302LV -> VB_SIS302LV VB_YPbPr525p -> YPbPr525p VB_YPbPr750p -> YPbPr750p VCLK108_2 -> VCLK108_2_315 VCLK65 -> VCLK65_315 XGI_CRT2_PORT_04 -> SIS_CRT2_PORT_04 XGI_CRT2_PORT_10 -> SIS_CRT2_PORT_10 XGI_CRT2_PORT_12 -> SIS_CRT2_PORT_12 XGI_CRT2_PORT_14 -> SIS_CRT2_PORT_14 Signed-off-by:
-
Peter Huewe authored
This patch renames the Set* defines and their usage to the naming convention of the sis initdef.h and removes the now duplicated defines. Renames: SetCRT2ToHiVisionTV -> SetCRT2ToHiVision SetCRT2ToYPbPr -> SetCRT2ToYPbPr525750 SetNTSCJ -> TVSetNTSCJ SetPALMTV -> TVSetPALM SetPALNTV -> TVSetPALN SetPALTV -> TVSetPAL SetYPbPrMode1080i -> TVSetHiVision SetYPbPrMode525i -> TVSetYPbPr525i SetYPbPrMode525p -> TVSetYPbPr525p SetYPbPrMode750p -> TVSetYPbPr750p Signed-off-by:
-
Peter Huewe authored
This patch renames the Panel* defines and their usage to the naming convention of the sis initdef.h and removes the now duplicated defines. Renames: Panel320x480 -> Panel_320x480 Panel800x600 -> Panel_800x600 Panel1024x768 -> Panel_1024x768 Panel1024x768x75 -> Panel_1024x768x75 Panel1280x1024 -> Panel_1280x1024 Panel1280x1024x75 -> Panel_1280x1024x75 Panel1280x960 -> Panel_1280x960 Panel1400x1050 -> Panel_1400x1050 Panel1600x1200 -> Panel_1600x1200 Signed-off-by:
-
Peter Huewe authored
This patch includes the initdef.h header from the sis driver. Since the xgi driver used to redefine a lot of stuff from the sis driver, we can simply include the headers of the sis driver itself, so we can remove duplicated stuff later on. In order to include the initdef.h we have to rename the header guards. Signed-off-by:
-
Greg Kroah-Hartman authored
Turns out it's not quite ready to be included, thanks to some other work done in the zcache and zram code, which breaks this driver. So, delete it for now, per the recommendation of Dan. Acked-by:
Dan Magenheimer <dan.magenheimer@oracle.com> Signed-off-by:
-
- 09 Feb, 2012 21 commits
-
-
Seth Jennings authored
Replaces xvmalloc with zsmalloc as the persistent memory allocator for zcache Signed-off-by:
Seth Jennings <sjenning@linux.vnet.ibm.com> Signed-off-by:
-
Seth Jennings authored
In a multithreaded workload, the zv_curr_dist_counts and zv_cumul_dist_counts statistics are being corrupted because the increments and decrements in zv_create and zv_free are not atomic. This patch converts these statistics and their corresponding increments/decrements/reads to atomic operations. Signed-off-by:
-
Jonathan Cameron authored
Neither of these drivers has ever been anywhere near the iio abi. Probably as a result of this the fact they had two event groups each was not picked up when we restricted IIO to having only 1 event line per device (as part of the chrdev merge set). As such these definitely didn't work before. This patch squishes the only element from the 'comparator' event line that isn't in the 'interrupt' one into it and kills off the 'comparator' one. Ultimately both of these drivers belong in hwmon not IIO and are just waiting here because I don't want to kill off a driver that may prove useful to someone. (Ultimately I will ask Greg to scrap these two if no one steps up to deal with them.) Signed-off-by:
Jonathan Cameron <jic23@kernel.org> Acked-by:
Lars-Peter Clausen <lars@metafoo.de> Signed-off-by:
-
Jonathan Cameron authored
This is the most controversial of this set of is_visible removals. There are two conditions controlling availability of attrs resulting in 4 different attribute groups. Still for a few more lines things are clearer to read to my mind. Signed-off-by:
-
Jonathan Cameron authored
Trivial case where no attributes are valid for some parts. Better handled using two iio_info structures and selecting the right one at probe time. Signed-off-by:
-
Jonathan Cameron authored
Trivial usecase in which just having two different attr groups covers all options. Slightly more code, but a simpler to follow result. Signed-off-by:
-
Jonathan Cameron authored
It saves a couple of lines of code but reduces simplicity of code. I generally wish to discourage use of is_visible throughout IIO. Signed-off-by:
-
Lars-Peter Clausen authored
We always hold the waitqueue lock when modifying the flags field. So it is safe to use the non-atomic bitops here instead of the atomic versions. The lock has to be held, because we need to clear the busy flag and flush the event FIFO in one atomic operation when closing the event file descriptor. Signed-off-by:
-
Lars-Peter Clausen authored
Add poll support to the event queue. This will allow us to check for pending events in a application's event loop using poll() or similar. Since we already have support for blocking reads adding poll support as well is trivial. Signed-off-by:
-
Lars-Peter Clausen authored
Use the waitqueue lock to protect the event queue instead of a custom mutex. This has the advantage that we can call the waitqueue operations with the lock held, which simplifies the code flow a bit. Signed-off-by:
-
Lars-Peter Clausen authored
The current IIO event code uses a list to emulate FIFO like behavior. Just use a kfifo directly instead to implement the event queue. As part of this patch the maximum of events in the queue is increased from 10 to 16 since kfifo requires a power of two for the number of FIFO elements. Signed-off-by:
-
Lars-Peter Clausen authored
The core iio file has gotten quite cluttered over time. This patch moves the event handling code into its own file. Since the event handling code is largely independent from the core code the only code changes necessary for this are to make the moved iio_device_register_eventset, iio_device_unregister_eventset and iio_event_getfd functions non static. This has also the advantage that industrialio-core.c is now closer again to its counterpart in the outofstaging branch. Signed-off-by:
-
Lars-Peter Clausen authored
The documentation for the iio_event_interface does not match the actual struct anymore. This patch removes the documentation for non-existing fields and adds documentation for missing fields. Signed-off-by:
-
Lars-Peter Clausen authored
Setup the buffer access functions in the buffer allocate function. There is no need to let each driver handle this on its own. Signed-off-by:
-
Olof Johansson authored
Just like isl29018; trivial addition. Using both asahi-kasei,ak8975 and the non-prefixed version (I couldn't figure out if Asahi Kasei had a stock symbol to use, I only found numerical indexes for their stock info). Signed-off-by:
Olof Johansson <olof@lixom.net> Acked-by:
-
Olof Johansson authored
As simple as can be right now; just one ID and no custom properties to parse. Signed-off-by:
Grant Likely <grant.likely@secretlab.ca> Acked-by:
-
Dan Carpenter authored
These are bitfields and the intend was to OR them together. A logical OR here is simply 1. Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Acked-by:
Ian Abbott <abbotti@mev.co.uk> Signed-off-by:
-
Ian Abbott authored
This patch is ported over by me (Ian Abbott) from the out-of-tree Comedi git repository at "git://comedi.org/git/comedi/comedi.git". The original patch is by Nicholas Nell. The patch adds support for trailing (falling) edge external triggers for scans in asynchronous command support in the ni_pcidio driver. This is supported at least on the PCI-DIO_32HS and PCI-6533 boards; not sure about the other boards. Signed-off-by:
-
Ian Abbott authored
This patch is ported over by me (Ian Abbott) from the out-of-tree Comedi git repository at "git://comedi.org/git/comedi/comedi.git". The original patch is by Nicholas Nell. The patch adds support for the COMEDI_POLL ioctl to the ni_pcidio driver. Signed-off-by:
-
Ian Abbott authored
When setting up the DMA for 'read' streaming acquisition command, grab the MITE channel spinlock before preparing and arming the DMA. Change inspired by ni_ai_setup_MITE_dma() in ni_mio_common.c. Signed-off-by:
-
Ian Abbott authored
When setting up the DMA for 'read' streaming acquisition on the DIO subdevice, mark the whole buffer as writable before starting the DMA. This prevents a spurious detection of a DMA overwrite of good data during the first interrupt. Problem reported by Nicholas Nell. Fix suggested by Frank Mori Hess. Signed-off-by:
-
