- 15 Aug, 2016 40 commits
-
-
Tim Sell authored
A kernel timer is now used as the vehicle to periodically call the channel_interrupt function of registered visor drivers, instead of a workqueue. This simplifies a lot of things by making periodic_work.c and periodic_work.h no longer necessary. This change also means that the channel_interrupt() callbacks registered by visor drivers (via visorbus_register_visor_driver()) will now be called in atomic context (i.e., canNOT sleep) rather than kernel thread context (CAN sleep). Fortunately this did NOT necessitate any change to the existing channel_interrupt() callbacks, because none of them ever perform any operations that would be invalid in atomic context. Signed-off-by:
Tim Sell <Timothy.Sell@unisys.com> Signed-off-by:
David Kershner <david.kershner@unisys.com> Acked-By:
Neil Horman <nhorman@tuxdriver.com> Reviewed-by:
Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Tim Sell authored
Locking in the _interrupt() function is NOT necessary so long as we ensure that interrupts have been stopped whenever we need to pause or resume the device, which we now do. While a device is paused, we ensure that interrupts stay disabled, i.e. that the _interrupt() function will NOT be called, yet remember the desired state in devdata->interrupts_enabled if open() or close() are called are called while the device is paused. Then when the device is resumed, we restore the actual state of interrupts (i.e., whether _interrupt() is going to be called or not) to the desired state in devdata->interrupts_enabled. Signed-off-by:
-
Tim Sell authored
periodic_test_workqueue was an unused relic from the past, and was removed. Signed-off-by:
-
David Binder authored
Code relating to ktheads was previously removed from s-Par driver code. This patch cleans up lingering remnants of kthreads by removing thread- related enum types. Signed-off-by:
David Binder <david.binder@unisys.com> Signed-off-by:
-
David Binder authored
Fixes a comment spelling mistake in visornic. Signed-off-by:
-
David Binder authored
Modifies the format string of snprintf to expect an unsigned int instead of a signed one, per the supplied argument. Signed-off-by:
-
David Binder authored
Removes unused struct definition, channel_size_info, in response to findings by SonarQube. Signed-off-by:
-
David Binder authored
Removes unused module parameters from visorbus_main.c, in response to findings by SonarQube. Signed-off-by:
-
Erik Arfvidson authored
This patch modifies the caller of parahotplug_request_complete() to check the return value and return appropriate result. Signed-off-by:
Erik Arfvidson <erik.arfvidson@unisys.com> Signed-off-by:
-
Erik Arfvidson authored
This patch removes the return values from the write_vbus_* channel functions. Nobody was checking the return values and the value stored into the vbus info is not critical to the functioning of the device. Signed-off-by:
-
Christian Gromm authored
This patch reduces context of the list_del mutex for the function hdm_enqueue(). Signed-off-by:
Andrey Shvetsov <andrey.shvetsov@k2l.de> Signed-off-by:
Christian Gromm <christian.gromm@microchip.com> Signed-off-by:
-
Christian Gromm authored
This patch gives the struct device variable of the core a more meaningful name. Signed-off-by:
-
Christian Gromm authored
This patch removes the debug print in case an HDM ran out of rx buffers. This is not needed since the status is reflected in the sysfs channel_starving flag. Additionally, it prevents the system form slowing down. Signed-off-by:
-
Christian Gromm authored
If a DIM2 channel is being closed while it is busy, the channels's status flag could possibly stay active for this very channel. This causes the kernel to freeze by the time the channel is opened again. This patch fixes the problem. Signed-off-by:
-
Christian Gromm authored
This patch adds possibility to configure the DIM2-IP parameter representing the number of frames per sub-buffer for synchronous channels. Signed-off-by:
-
Christian Gromm authored
This patch fixes race between the function poison_channel that clears the state is_initialized and the tasklet function dim2_tasklet_fn that checks the state is_initialized. Signed-off-by:
-
Christian Gromm authored
This patch removes the unnecessary spaces. Signed-off-by:
-
Christian Gromm authored
For optimization purposes this patch removes the label err_vbi_dev. Signed-off-by:
-
Christian Gromm authored
The function aim_register_videodev() uses the variables 'ret' and 'retval' to represent the same value. This patch removes 'retval' and replaces it with 'ret'. Further, it replaces the constant return value '-ENODEV' with the result returned by function video_register_device() in the event something went wrong. Signed-off-by:
-
Christian Gromm authored
This patch replaces pr_info, pr_err, etc. function calls by the v4l2_... ones to get the device dependent logs. Signed-off-by:
-
Christian Gromm authored
The current V4L2 AIM implementation assigns to all video devices and to all V4L devices the same names. This patch makes use of hardware dependent names for the video devices and allows the user to choose the names for the V4L devices. Signed-off-by:
-
Christian Gromm authored
As the function get_aim_dev() does not delete elements of the list, the use of macro list_for_each_entry_safe is not necessary. This patch replaces the macro list_for_each_entry_safe with the macro list_for_each_entry. Signed-off-by:
-
Christian Gromm authored
The functions get_aim_dev() and aim_rx_data() are using interrupt unsafe spinlocks even though they may be called from an interrupt context. This patch fixes the described problem. Signed-off-by:
-
Colin Ian King authored
synth_add allows one to add MAXSYNTHS synths to the synths array; however it always NULLifies the next synth in the array which means that on the MAXSYNTHS synth we get an out-of-bounds write of the NULL to the synths array. Make the synths array MAXSYNTHS + 1 elements in size to allow for the final NULL sentinal to avoid the out-of-bounds write. Issue found wit CoverityScan, CID#744671 Signed-off-by:
Colin Ian King <colin.king@canonical.com> Reviewed-by:
Samuel Thibault <samuel.thibault@ens-lyon.org> Signed-off-by:
-
Laurence Rochfort authored
Fix the checkpatch.pl warning "No space is necessary after a cast". Signed-off-by:
Laurence Rochfort <laurence.rochfort@gmail.com> Signed-off-by:
-
Daeseok Youn authored
The 'ch_intr_rx' variable was used only for increasing. So the 'ch_intr_rx' variable is not useful for this driver. Signed-off-by:
Daeseok Youn <daeseok.youn@gmail.com> Signed-off-by:
-
Daeseok Youn authored
The 'ch_intr_tx' variable was used only for increasing. So the 'ch_intr_tx' variable is not useful for this driver. Signed-off-by:
-
Daeseok Youn authored
The 'ch_intr_modem' variable was used only for increasing. So the 'ch_intr_modem' variable is not useful for this driver. Signed-off-by:
-
Daeseok Youn authored
The 'intr_rx' variable was used only for increasing. So the 'intr_rx' variable is not useful for this driver. Signed-off-by:
-
Daeseok Youn authored
The 'intr_tx' variable was used only for increasing. So the 'intr_tx' variable is not useful for this driver. Signed-off-by:
-
Daeseok Youn authored
The 'intr_modem' variable was used only for increasing. So the 'intr_modem' variable is not useful for this driver. Signed-off-by:
-
Daeseok Youn authored
the "brd" was already checked for NULL before calling dgnc_do_remap(). the dgnc_do_remap() function was called only from the dgnc_found_board() and the DGNC_BOARD_MAGIC value was assigned to "brd->magic" in dgcn_found_board(). So it doesn't need to check about magic value. Signed-off-by:
-
Daeseok Youn authored
The 'intr_count' variable was used only for increasing. So the 'intr_count' variable is not useful for this driver. Signed-off-by:
-
Daeseok Youn authored
the "brd" value cannot be NULL in dgnc_finalize_board_init(). Because "brd" as a parameter of this function was already checked for NULL. the dgnc_finalize_board_init() as a static function was called only from dgnc_found_board() function and brd->magic value was assigned once in dgnc_found_board(). So it doesn't need to check for DGNC_BOARD_MAGIC value. Signed-off-by:
-
Daeseok Youn authored
Re-arrange the functions for removing forward declarations in dgnc_cls.c file. Signed-off-by:
-
Daeseok Youn authored
The dgnc_set_modem_info() used only channel_t variable. Any other variables were used only for checking NULL. So fist parameter changed from "tty_struct" to "channel_t" and useless NULL checks and variables are removed. Signed-off-by:
-
Daeseok Youn authored
The unit struct(un_t) was not used in dgnc_tty_hangup(). Signed-off-by:
-
Daeseok Youn authored
The local variable "bd" was not used in dgnc_carrier() function. Signed-off-by:
-
Daeseok Youn authored
The "result" variable in dgnc_get_mstat() was initialized with "-EIO". But if the "ch" is not null, "result" will be set to zero and if the "ch" is null, dgnc_get_mstat() will return "-ENXIO" as an error. So "-EIO" error value was useless in dgnc_get_mstat(). Signed-off-by:
-
Daeseok Youn authored
The dgnc_block_til_ready() is only used in dgnc_tty_open(). The unit data(struct un_t) was stored into tty->driver_data in dgnc_tty_open(). And also tty and un were tested about NULL so these variables doesn't need to check for NULL in dgnc_block_til_ready(). Signed-off-by:
-
