- 25 Apr, 2017 1 commit
-
-
Ian Abbott authored
If a process that has mmap'd a COMEDI buffer is being run under a debugger such as GDB, the buffer contents are inaccessible from the debugger. Support the `access()` VM operation to allow the buffer contents to be accessed by another process. Signed-off-by:
Ian Abbott <abbotti@mev.co.uk> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
- 18 Apr, 2017 39 commits
-
-
Laura Abbott authored
Most of the items have been taken care of by a clean up series. Remove the completed items and add a few new ones. Signed-off-by:
Laura Abbott <labbott@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Laura Abbott authored
This never got set in the ioctl. Properly set a return value of 0 on success. Signed-off-by:
Laura Abbott <labbott@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Laura Abbott authored
ion_handle was introduced as an abstraction to represent a reference to a buffer via an ion_client. As frameworks outside of Ion evolved, the dmabuf emerged as the preferred standard for use in the kernel. This has made the ion_handle an unnecessary abstraction and prone to race conditions. ion_client is also now only used internally. We have enough mechanisms for race conditions and leaks already so just drop ion_handle and ion_client. This also includes ripping out most of the debugfs infrastructure since much of that was tied to clients and handles. The debugfs infrastructure was prone to give confusing data (orphaned allocations) so it can be replaced with something better if people actually want it. Signed-off-by:
Laura Abbott <labbott@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Laura Abbott authored
Nobody uses this interface externally. Drop it. Signed-off-by:
Laura Abbott <labbott@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Laura Abbott authored
The current model of Ion heap registration is based on the outdated model of board files. The replacement for board files (devicetree) isn't a good replacement for what Ion wants to do. In actuality, Ion wants to show what memory is available in the system for something else to figure out what to use. Switch to a model where Ion creates its device unconditionally and heaps are registed as available regions. Currently, only system and CMA heaps are converted over to the new model. Carveout and chunk heaps can be converted over when someone wants to figure out how. Signed-off-by:
Laura Abbott <labbott@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Laura Abbott authored
Ion current has ion_priv.h and ion.h as header files. ion.h was intended to be used for public APIs but Ion never ended up really having anything public. Combine the two headers so there is only one internal header. Signed-off-by:
Laura Abbott <labbott@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Laura Abbott authored
Once upon a time, phys_addr_t was not everywhere in the kernel. These days it is used enough places that having a separate Ion type doesn't make sense. Remove the extra type and just use phys_addr_t directly. Signed-off-by:
Laura Abbott <labbott@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Laura Abbott authored
Several of the Ion ioctls were designed in such a way that they necessitate compat ioctls. We're breaking a bunch of other ABIs and cleaning stuff up anyway so let's follow the ioctl guidelines and clean things up while everyone is busy converting things over anyway. As part of this, also remove the useless alignment field from the allocation structure. Signed-off-by:
Laura Abbott <labbott@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Laura Abbott authored
Now that we have proper caching, stop setting the DMA address manually. It should be set after properly calling dma_map. Signed-off-by:
Laura Abbott <labbott@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Laura Abbott authored
When CMA was first introduced, its primary use was for DMA allocation and the only way to get CMA memory was to call dma_alloc_coherent. This put Ion in an awkward position since there was no device structure readily available and setting one up messed up the coherency model. These days, CMA can be allocated directly from the APIs. Switch to using this model to avoid needing a dummy device. This also mitigates some of the caching problems (e.g. dma_alloc_coherent only returning uncached memory). Signed-off-by:
Laura Abbott <labbott@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Laura Abbott authored
Frameworks (e.g. Ion) may want to iterate over each possible CMA area to allow for enumeration. Introduce a function to allow a callback. Signed-off-by:
Laura Abbott <labbott@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Laura Abbott authored
Frameworks that may want to enumerate CMA heaps (e.g. Ion) will find it useful to have an explicit name attached to each region. Store the name in each CMA structure. Signed-off-by:
Laura Abbott <labbott@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Greg Kroah-Hartman authored
Merge tag 'iio-fixes-for-4.11e' of git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next Jonathan writes: Fifth set of IIO fixes for the 4.11 cycle. As these are rather late in the cycle, they may sneak over into 4.12. There is a fix for a regression caused by another fix (hid sensors hardware seems to vary a lot in how various corner cases are handled). * ad7303 - fix channel description. Numeric values were being passed as characters presumably leading to garbage from the userspace interface. * as3935 - the write data macro was wrong so fix it. * bmp280 - incorrect handling of negative values as being unsigned broke humidity calculation. * hid-sensor - Restore the poll and hysteresis values after resume as some hardware doesn't do it. * stm32-trigger - buglet in reading the sampling frequency
-
Greg Kroah-Hartman authored
Merge tag 'iio-for-4.12d' of git://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into staging-next Jonathan writes: Fourth set of IIO new device support, features and cleanups for the 4.12 cycle New device support * max1117, 1118 and 1119 - new ADC driver * max9611 - new ADC driver * pm8xxx hk/xoadc - new driver with some shared features broken out from the SPMI vadc. * sun4i-gpadc - A33 thermal sensor support (with associated rework) * stm32-dac - new driver and bindings * stm32 trigger - enable support of quadrature encoder device and counter modes Features * apds9960 - use the runtime pm for normal suspend * stm32-adc - add opition to sest resolution via devicetree * xoadc - augment DT bindings to deal with some weird mux cases Cleanups * ad5933 - protect direct mode using claim and release helpers * ade7759 - S_IRUGO and friends to octal in two goes * adis16203 - drop unnecessary brackets * hid-sensor - fix unbalanced pm_runtieme_enable error when probing after remove * lsm6dsx - use actual part numbers for device name when known - simplify data read pin parsing * mpu3050 - avoid double reporting errors
-
Martin Karamihov authored
This patch fixes the following sparse warning: ieee80211_tx.c:174:36: incorrect type in assignment (different base types) ieee80211_tx.c:174:36: expected unsigned short [unsigned] [short] [usertype] <noident> ieee80211_tx.c:174:36: got restricted __be16 [usertype] <noident> by adding left side cast to __be16. Signed-off-by:
Martin Karamihov <martinowar@gmail.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Aishwarya Pant authored
Changed the type of wpa_ie_len from (int *) to (unsigned int *) in the function rtw_get_wpa_ie(..) to suppress signedness mismatch warnings in rtw_generate_ie of the type- drivers/staging/rtl8188eu//core/rtw_ieee80211.c:1009:60: warning: incorrect type in argument 2 (different signedness) drivers/staging/rtl8188eu//core/rtw_ieee80211.c:1009:60: expected int *wpa_ie_len drivers/staging/rtl8188eu//core/rtw_ieee80211.c:1009:60: got unsigned int *<noident> Signed-off-by:
Aishwarya Pant <aishpant@gmail.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Aishwarya Pant authored
Changed the type of sz from (int) to (unsigned int) to suppress signedness mismatch warnings of the type- drivers/staging/rtl8188eu//core/rtw_ieee80211.c:258:97: warning: incorrect type in argument 5 (different signedness) drivers/staging/rtl8188eu//core/rtw_ieee80211.c:258:97: expected unsigned int [usertype] *frlen drivers/staging/rtl8188eu//core/rtw_ieee80211.c:258:97: got int *<noident> Signed-off-by:
Aishwarya Pant <aishpant@gmail.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Aishwarya Pant authored
Changed the type of len from (int *) to (unsigned int *) in the function rtw_get_ie(..) and wherever this function is called to suppress signedness mismatch warnings of the type- drivers/staging/rtl8188eu//core/rtw_ap.c:78:60: warning: incorrect type in argument 3 (different signedness) drivers/staging/rtl8188eu//core/rtw_ap.c:78:60: expected int *len drivers/staging/rtl8188eu//core/rtw_ap.c:78:60: got unsigned int *<noident> Signed-off-by:
Aishwarya Pant <aishpant@gmail.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Colin Ian King authored
The two different paths for an if statement are identical and hence we can just replace it with the single statement. Detected by CoverityScan, CID#1428443 ("Identical code for different branches") Signed-off-by:
Colin Ian King <colin.king@canonical.com> Reviewed-by:
Hans de Goede <hdegoede@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Colin Ian King authored
The comparison of mode >= 0 is redundant as mode is a u32 and this is always true. Remove this redundant code. Detected with CoverityScan ("Unsigned compared against 0") Signed-off-by:
Colin Ian King <colin.king@canonical.com> Reviewed-by:
Hans de Goede <hdegoede@redhat.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Kevin Wern authored
Gstreamer's v4l2src reacted poorly to certain outputs from the bcm2835 video driver's ioctl ops function vidioc_enum_framesizes, so a workaround was created that could be activated by user input. This workaround would replace the driver's ioctl ops struct with another, similar struct--only with no function pointed to by vidioc_enum_framesizes. With no response, gstreamer would attempt to continue with some default settings that happened to work better. However, this bug has been fixed in gstreamer since 2014, so we shouldn't include this workaround in the stable version of the driver. Signed-off-by:
Kevin Wern <kevin.m.wern@gmail.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Tobin C. Harding authored
Checkpatch emits WARNING: quoted string split across lines. Concatenate string onto single line. Signed-off-by:
Tobin C. Harding <me@tobin.cc> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Tobin C. Harding authored
Driver SDIO code uses helper functions to do IO to the SDIO device. Current helpers handle IO of a single byte as well as multi-byte. Driver predominately uses single byte IO. If the common case is made simple it simplifies the whole driver. The common case can be made simple by splitting the multi-byte and single byte calls into separate functions, i.e 4 functions in total, read single byte, read multi-byte, write single byte, write multi-byte. Also, we need to handle the debug code. Currently debug calls after read/write fail access the IO buffer. This buffer, at best, does not hold useful data on the error path, at worst is uninitialized and holds garbage. Split read/write helper functions into two functions each, one for single byte IO and one for multi-byte IO. Fix all call sites. Do not change the program logic. Signed-off-by:
Tobin C. Harding <me@tobin.cc> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Tobin C. Harding authored
Checkpatch emits CHECK: Please don't use multiple blank lines. Remove multiple blank lines. Signed-off-by:
Tobin C. Harding <me@tobin.cc> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Tobin C. Harding authored
Checkpatch emits WARNING: please, no space before tabs. Remove space before tabs. Signed-off-by:
Tobin C. Harding <me@tobin.cc> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Tobin C. Harding authored
goto label includes 'err_' suffix but is executed on non-error paths. Remove err_ suffix from goto label. Signed-off-by:
Tobin C. Harding <me@tobin.cc> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Tobin C. Harding authored
SDIO code currently has a number of unneeded comments. Following kernel coding style we do not need extraneous comments, especially on code where it is clear what is being done. Spelling typos can be fixed. Remove unnecessary comments, fix typos in comments. Signed-off-by:
Tobin C. Harding <me@tobin.cc> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Tobin C. Harding authored
ks_sdio_card structure description does not have a kernel doc format comment. Add kernel doc format comment to struct ks_sdio_card. Signed-off-by:
Tobin C. Harding <me@tobin.cc> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Tobin C. Harding authored
Currently driver uses a hardware information struct description to group some SDIO related functionality (work, work queue, sdio private data pointer). This structure is then embedded in the device private data structure. Having nested structures described in different header files means that to view the device private data programmers must open two header files. This structure could be embedded anonymously in the device private data and achieve the same result (grouping of function specific to SDIO) without the need to open multiple headers. However, the SDIO private data structure already has various different data and pointers, adding the embedded structure adds little extra meaning and lengthens all the dereferences throughout the driver, often meaning addition line breaks and braces. We can increase readability and reduce code complexity by moving the hardware information data and pointers to directly be within the device private data structure description. While preparing for this refactoring it was noted that the identifier currently used for the delayed work is 'rw_wq', this is confusing since the 'wq' suffix typically means 'work queue'. This identifier would be more meaningful if it used the suffix 'dwork' as does the declaration of queue_delayed_work() (include/linux/workqueue.h). The identifier for the work queue is currently 'ks7010sdio_wq'. This identifier can be shortened without loss of meaning because there is only one work queue within the driver. Identifier 'wq' is typical within in-tree driver code and aptly describes the pointer. Current pointer to the SDIO private data is identified by 'sdio_card', this is sufficiently meaningful from within the hw_info structure but once the hw_info_t structure is removed the pointer would be better to have a prefix appended to it to retain the prior level of meaning. Move members from struct hw_info_t to struct ks_wlan_private. Rename identifiers; struct delayed_work pointer 'rw_wq' to 'rw_dwork'. struct workqueue_struct pointer 'ks7010sdio_wq' to 'wq'. struct ks_sdio_card pointer 'sdio_card' to 'ks_sdio_card'. Remove structure description hw_info_t. Fix init/destroy calls. Fix all call sites, SDIO private data access calls, and queuing calls. Signed-off-by:
Tobin C. Harding <me@tobin.cc> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Tobin C. Harding authored
Currently a pointer to the tasklet_struct used for bottom half processing on the receive path is within the hw_info_t structure. This structure is then embedded in the device private data structure. Having the tasklet_struct nested does not add meaning to the device private data, device private data already (and typically) has various data relating to the device, there is no real need to separate the tasklet_struct to a SDIO specific structure. While not adding allot of extra meaning having the nested structure means the programmer must open two header files to read the description of the device private data, the code would be easier to read if the device private data struct description was not spread over two files. Move tasklet_struct out of sdio header file and into the device private data structure description. Signed-off-by:
Tobin C. Harding <me@tobin.cc> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Tobin C. Harding authored
struct work_struct uses identifier ks_wlan_wakeup_task, this is confusing because the 'task' suffix implies that this is a tasklet_struct instead of a work struct. Suffix 'work' would be more clear. The code would be easier to read if it followed the principle of least surprise and used the 'work' suffix for a work_struct identifier. Rename work_struct structure 'ks_wlan_wakeup_task' to 'wakeup_work'. Signed-off-by:
Tobin C. Harding <me@tobin.cc> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Tobin C. Harding authored
SDIO header file does not use kernel doc format struct comments. Adding them aids readability and enables documentation to be built from the source code. Other comments may be tidied up as we do this. Add kernel format struct comments. Tidy up comments. Signed-off-by:
Tobin C. Harding <me@tobin.cc> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Tobin C. Harding authored
complete_handler() takes void * types as parameters. void * parameters are then cast to struct types. Call sites for this function either pass in NULL or pointers to the struct types cast to void *. This casting is unnecessary and can be removed. Struct tx_device_buffer (which contains a pointer member to the complete_handler() function) has as member 'ks_wlan_priv *priv' this is unnecessary, we always have a pointer to this struct there is no need to store it here. The complete_handler can be more clearly defined by using struct pointer types instead of void * types. The code is currently unnecessarily complex, storing and passing extraneous pointer parameters. Remove unnecessary parameters, unnecessary casting to/from 'void *'. Fix all call sites involving complete_handler(). Signed-off-by:
Tobin C. Harding <me@tobin.cc> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Tobin C. Harding authored
Header has multiple constants defined using preprocessor directive. In the cases where these are an integer progression an enumeration type can be used. Doing so adds documentation to the code and makes the usage explicit. Maintain original constant value, this value is returned by the device. Replace (integer progression) preprocessor constants with enumeration type. Signed-off-by:
Tobin C. Harding <me@tobin.cc> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Tobin C. Harding authored
SDIO header currently defines unused constants READ_STATUS_BUSY and WRITE_STATUS_IDLE. There are reciprocal constants that are used READ_STATUS_IDLE and WRITE_STATUS_BUSY. We can roll these into a single enumeration type and remove the two that are unused. Add enumeration type containing IDLE/BUSY pair that are currently used within the SDIO source. Change source to use new enum types. Signed-off-by:
Tobin C. Harding <me@tobin.cc> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Darryl T. Agostinelli authored
$ make includecheck | grep staging ./drivers/staging/greybus/uart.c: linux/serial.h is included more than once. Signed-off-by:
Darryl T. Agostinelli <dagostinelli@gmail.com> Reviewed-by:
Alex Elder <elder@linaro.org> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Darryl T. Agostinelli authored
Fixes: $ make versioncheck | grep staging ./drivers/staging/greybus/light.c: 15 linux/version.h not needed. Signed-off-by:
Darryl T. Agostinelli <dagostinelli@gmail.com> Reviewed-by:
Rui Miguel Silva <rmfrfs@gmail.com> Reviewed-by:
Alex Elder <elder@linaro.org> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Skanda Guruanand authored
The struct lu_dirpage elements in lustre_idl.h file are modified to __le64 and __le32 types since the elements are always converted from litte endian to processor native format in mdc_request.c file. Following warnings are removed by this fix. drivers/staging/lustre/lustre/mdc/mdc_request.c:958:42: warning: cast to restricted __le64 drivers/staging/lustre/lustre/mdc/mdc_request.c:959:42: warning: cast to restricted __le64 drivers/staging/lustre/lustre/mdc/mdc_request.c:962:42: warning: cast to restricted __le64 drivers/staging/lustre/lustre/mdc/mdc_request.c:963:42: warning: cast to restricted __le64 drivers/staging/lustre/lustre/mdc/mdc_request.c:985:50: warning: cast to restricted __le32 drivers/staging/lustre/lustre/mdc/mdc_request.c:1193:24: warning: cast to restricted __le64 drivers/staging/lustre/lustre/mdc/mdc_request.c:1328:25: warning: cast to restricted __le64 drivers/staging/lustre/lustre/mdc/mdc_request.c:1329:23: warning: cast to restricted __le64 drivers/staging/lustre/lustre/mdc/mdc_request.c:1332:25: warning: cast to restricted __le64 drivers/staging/lustre/lustre/mdc/mdc_request.c:1333:23: warning: cast to restricted __le64 Signed-off-by:
Skanda Guruanand <skanda.kashyap@gmail.com> Reviewed-by:
Andreas Dilger <andreas.dilger@intel.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Craig Inches authored
This resolves a checkpatch warning that "Single statement macros should not use a do {} while (0) loop" by removing the loop and adjusting line length accordingly. Signed-off-by:
Craig Inches <Craig@craiginches.com> Reviewed-by:
Andreas Dilger <andreas.dilger@intel.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-