- 12 Feb, 2020 8 commits
-
-
Greg Kroah-Hartman authored
It's been over 6 months, and no one has noticed that these drivers are deleted, probably because no one actually has this hardware. As no one has volunteered to maintain the code, let's drop it for good. Link: https://lore.kernel.org/r/20200210231417.GA1736729@kroah.comSigned-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Ajay Singh authored
Refactor handling of P2P specific action frames. Make use of 'struct' to handle the P2P frames instead of manipulating using 'buf' pointer. Signed-off-by:
Ajay Singh <ajay.kathat@microchip.com> Link: https://lore.kernel.org/r/20200212154503.8835-1-ajay.kathat@microchip.comSigned-off-by:
-
Marcelo Diop-Gonzalez authored
There are a few places where a service's reference count is incremented, something quick is done, and the refcount is dropped. This can be made a little simpler/faster by not grabbing a reference in these cases. Signed-off-by:
Marcelo Diop-Gonzalez <marcgonzalez@google.com> Link: https://lore.kernel.org/r/ac6186ac888f1acf489b5b504efcba8b0d6a8b25.1581532523.git.marcgonzalez@google.comSigned-off-by:
-
Marcelo Diop-Gonzalez authored
Currently reference counts are implemented by locking service_spinlock and then incrementing the service's ->ref_count field, calling kfree() when the last reference has been dropped. But at the same time, there's code in multiple places that dereferences pointers to services without having a reference, so there could be a race there. It should be possible to avoid taking any lock in unlock_service() or service_release() because we are setting a single array element to NULL, and on service creation, a mutex is locked before looking for a NULL spot to put the new service in. Using a struct kref and RCU-delaying the freeing of services fixes this race condition while still making it possible to skip grabbing a reference in many places. Also it avoids the need to acquire a single spinlock when e.g. taking a reference on state->services[i] when somebody else is in the middle of taking a reference on state->services[j]. Signed-off-by:
-
Kaaira Gupta authored
Fix three checkpatch.pl warnings of 'quoted string split across lines' in gasket_core.c by merging the strings in one line. Though some strings are over 80 characters long, fixing this warning is necessary to ease grep-ing the source for printk. Signed-off-by:
Kaaira Gupta <kgupta@es.iitr.ac.in> Link: https://lore.kernel.org/r/20200212175826.GA5967@kaaira-HP-Pavilion-NotebookSigned-off-by:
-
Marcelo Diop-Gonzalez authored
This fixes some checkpatch warnings about incorrect indentation levels Signed-off-by:
-
Marcelo Diop-Gonzalez authored
there are extra parentheses around many conditional statements that make things a little harder to read Signed-off-by:
-
Marcelo Diop-Gonzalez authored
vchiq_get_service_fourcc() doesn't seem to be used anywhere Signed-off-by:
-
- 11 Feb, 2020 26 commits
-
-
Gustavo A. R. Silva authored
The current codebase makes use of the zero-length array language extension to the C90 standard, but the preferred mechanism to declare variable-length types such as these ones is a flexible array member[1][2], introduced in C99: struct foo { int stuff; struct boo array[]; }; By making use of the mechanism above, we will get a compiler warning in case the flexible array does not occur last in the structure, which will help us prevent some kind of undefined behavior bugs from being inadvertenly introduced[3] to the codebase from now on. This issue was found with the help of Coccinelle. [1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html [2] https://github.com/KSPP/linux/issues/21 [3] commit 76497732 ("cxgb3/l2t: Fix undefined behaviour") Signed-off-by:Gustavo A. R. Silva <gustavo@embeddedor.com> Link: https://lore.kernel.org/r/20200211211219.GA673@embeddedorSigned-off-by:
-
Gustavo A. R. Silva authored
The current codebase makes use of the zero-length array language extension to the C90 standard, but the preferred mechanism to declare variable-length types such as these ones is a flexible array member[1][2], introduced in C99: struct foo { int stuff; struct boo array[]; }; By making use of the mechanism above, we will get a compiler warning in case the flexible array does not occur last in the structure, which will help us prevent some kind of undefined behavior bugs from being inadvertenly introduced[3] to the codebase from now on. This issue was found with the help of Coccinelle. [1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html [2] https://github.com/KSPP/linux/issues/21 [3] commit 76497732 ("cxgb3/l2t: Fix undefined behaviour") Signed-off-by: -
Pragat Pandya authored
Fix checkpatch warning: Avoid CamelCase Change all occurrences of identifier "AccessTimestamp" to "access_timestamp" Signed-off-by:
Pragat Pandya <pragat.pandya@gmail.com> Link: https://lore.kernel.org/r/20200210183558.11836-20-pragat.pandya@gmail.comSigned-off-by:
-
Pragat Pandya authored
Fix checkpatch warning: Avoid CamelCase Change all occurrences of identifier "ModifyTimestamp" to "modify_timestamp" Signed-off-by:
-
Pragat Pandya authored
Fix checkpatch warning: Avoid CamelCase Change all occurrences of identifier "CreateTimestamp" to "create_timestamp" Signed-off-by:
-
Pragat Pandya authored
Fix checkpatch warning: Avoid CamelCase Change all occurrences of identifier "NumSubdirs" to "num_subdirs" Signed-off-by:
-
Pragat Pandya authored
Fix checkpatch warning: Avoid CamelCase Change all occurrences of identifier "Attr" to "attr" Signed-off-by:
-
Pragat Pandya authored
Fix checkpatch warning: Avoid CamelCase Change all occurrences of identifier "ShortName" to "short_name" Signed-off-by:
-
Pragat Pandya authored
Fix checkpatch warning: Avoid CamelCase Change all occurrences of identifier "Name" to "name" Signed-off-by:
-
Pragat Pandya authored
Fix checkpatch warning: Avoid CamelCase Change all occurrences of identifier "MilliSecond" to "millisecond" Signed-off-by:
-
Pragat Pandya authored
Fix checkpatch warning: Avoid CamelCase Change all occurrences of identifier "Second" to "second" Signed-off-by:
-
Pragat Pandya authored
Fix checkpatch warning: Avoid CamelCase Change all occurrences of identifier "Minute" to "minute" Signed-off-by:
-
Pragat Pandya authored
Fix checkpatch warning: Avoid CamelCase Change all occurrences of identifier "Hour" to "hour" Signed-off-by:
-
Pragat Pandya authored
Fix checkpatch warning: Avoid CamelCase Change all occurrences of identifier "Day" to "day" Signed-off-by:
-
Pragat Pandya authored
Fix checkpatch warning: Avoid CamelCase Change all occurrences of identifier "Month" to "month" Signed-off-by:
-
Pragat Pandya authored
Fix checkpatch warning: Avoid CamelCase Change all occurrences of identifier "Year" to "year" Signed-off-by:
-
Mohana Datta Yelugoti authored
This patch fixes "WARNING: Missing a blank line after declarations" generated from checkpatch.pl by adding a blank line after declarations. Signed-off-by:
Mohana Datta Yelugoti <ymdatta.work@gmail.com> Link: https://lore.kernel.org/r/20200211095813.11426-1-ymdatta.work@gmail.comSigned-off-by:
-
Ajay Singh authored
The 'priv' handler is already present in 'vif' struct so directly fetch its value from vif handler in wilc_handle_roc_expired(). Signed-off-by:
-
Ajay Singh authored
Remove the use of vendor specific IE parameter to decide p2p_GO/p2p_Client roles between two 'wilc' device. Previously p2p group formation between two 'wilc' device make use of vendor IE for roles decision. The role is decided based on the 'go_intent' value. Signed-off-by:
-
Michał Mirosław authored
Replace udelay() with usleep_range() as all uses are in a sleepable context. Signed-off-by:
Michał Mirosław <mirq-linux@rere.qmqm.pl> Link: https://lore.kernel.org/r/59e1e4e5bd80c1879ef36eaa59916e47005dbb04.1581416843.git.mirq-linux@rere.qmqm.plSigned-off-by:
-
Michał Mirosław authored
Driver calls GPIO get/set only from non-atomic context and so can use any GPIOs. Signed-off-by:
-
Michał Mirosław authored
As for SPI, follow "vendor,chip" format 'compatible' string also for SDIO bus. Fixes: 0096214a ("staging: wfx: add support for I/O access") Signed-off-by:
-
Michał Mirosław authored
Add "compatible" string matching "vendor,chip" template and proper GPIO flags handling. Keep support for old name and reset polarity for older devicetrees. Cc: stable@vger.kernel.org # d3a5bcb4 ("gpio: add gpiod_toggle_active_low()") Cc: stable@vger.kernel.org Fixes: 0096214a ("staging: wfx: add support for I/O access") Signed-off-by:
-
Michał Mirosław authored
Lockdep is complaining about recursive locking, because it can't make a difference between locked skb_queues. Annotate nested locks and avoid double bh_disable/enable. [...] insmod/815 is trying to acquire lock: cb7d6418 (&(&list->lock)->rlock){+...}, at: wfx_tx_queues_clear+0xfc/0x198 [wfx] but task is already holding lock: cb7d61f4 (&(&list->lock)->rlock){+...}, at: wfx_tx_queues_clear+0xa0/0x198 [wfx] [...] Possible unsafe locking scenario: CPU0 ---- lock(&(&list->lock)->rlock); lock(&(&list->lock)->rlock); Cc: stable@vger.kernel.org Fixes: 9bca45f3 ("staging: wfx: allow to send 802.11 frames") Signed-off-by: -
Michał Mirosław authored
Current code races in init/exit with interrupt handlers. This is noticed by the warning below. Fix it by using devres for ordering allocations and IRQ de/registration. WARNING: CPU: 0 PID: 827 at drivers/staging/wfx/bus_spi.c:142 wfx_spi_irq_handler+0x5c/0x64 [wfx] race condition in driver init/deinit Cc: stable@vger.kernel.org Fixes: 0096214a ("staging: wfx: add support for I/O access") Signed-off-by:
Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/f0c66cbb3110c2736cd4357c753fba8c14ee3aee.1581416843.git.mirq-linux@rere.qmqm.plSigned-off-by:
-
YueHaibing authored
drivers/staging/wfx/queue.c: In function wfx_tx_queues_get: drivers/staging/wfx/queue.c:484:28: warning: variable tx_priv set but not used [-Wunused-but-set-variable] commit 2e57865e ("staging: wfx: pspoll_mask make no sense") left behind this unused variable. Reported-by:
Hulk Robot <hulkci@huawei.com> Signed-off-by:
YueHaibing <yuehaibing@huawei.com> Reviewed-by:
-
- 10 Feb, 2020 6 commits
-
-
Michael Straube authored
Rename the local variable pnetdev in rtw_alloc_etherdev_with_old_priv to avoid hungarian notation and clear the last checkpatch warning in the file osdep_service.c. rtl8188eu/os_dep/osdep_service.c:32: WARNING: line over 80 characters Signed-off-by:
Michael Straube <straube.linux@gmail.com> Link: https://lore.kernel.org/r/20200210200830.22868-1-straube.linux@gmail.comSigned-off-by:
-
JieunKim authored
pr_info is preferred to use than printk. pr_info calls printk with KERN_INFO macros by itself. Signed-off-by:
JieunKim <jieun.kim4758@gmail.com> Link: https://lore.kernel.org/r/20200210091421.12335-1-jieun.kim4758@gmail.comSigned-off-by:
-
Ian Abbott authored
We do not currently provide routing information for NI PCIe-6251 and PCI-6259 boards, but they are functionally identical to the PCI-6251 and PCI-6259 boards and can share their routing information. (This has been confirmed for the PCIe-6251 by Éric Piel, using the "NI MAX" software for Windows. It is hoped that it applies to PCIe-6259, but has not yet been checked due to lack of hardware.) Initialize the `alt_route_name` member of the board information for PCIe-6251 and PCIe-6259 to allow them to make use of the routing information provided for PCI-6251 and PCI-6259 respectively. Cc: Éric Piel <piel@delmic.com> Cc: Spencer E. Olson <olsonse@umich.edu> Signed-off-by:
Ian Abbott <abbotti@mev.co.uk> Link: https://lore.kernel.org/r/20200207151400.272678-5-abbotti@mev.co.ukSigned-off-by:
-
Ian Abbott authored
We do not have or do not provide routing information for all supported boards. Some of the boards for which we do not provide routing information actually have routes that are identical to a similar board for which we already provide routing information. To allow boards to share identical routing information, add an `alt_route_name` member to `struct ni_board_struct`. This will be initialized to `NULL` for all boards except those that will use make use of the identical routing information that has been provided for a similar board, in which case it will name that board. Pass the `alt_route_name` member value to `ni_assign_device_routes()` as the `alt_board_name` parameter, which it will use if no routing information could be found for the actual board name. Cc: Éric Piel <piel@delmic.com> Cc: Spencer E. Olson <olsonse@umich.edu> Signed-off-by:
-
Ian Abbott authored
We do not have or provide routing information available for all supported boards. Some of the boards for which we do not currently provide routing information actually have identical routes to a similar board for which we do provide routing information. To avoid having to provide duplicate routing information, add an "alternate board name" parameter (possibly `NULl`) to `ni_assign_device_routes()` and `ni_find_device_routes()`. If the routing information cannot be found for the actual board name, try finding it using the alternate board name. Cc: Éric Piel <piel@delmic.com> Cc: Spencer E. Olson <olsonse@umich.edu> Signed-off-by:
-
Ian Abbott authored
Split out the loops in `ni_find_valid_routes()` into separate functions: * ni_find_route_values(device_family) to find the list of route values for a device family (e.g "ni-mseries"); and * ni_find_valid_routes(board_name) to find the set of valid routes for a board name. The functions above return `NULL` if the information is not found (as we do not currently have the routing information available for all supported boards). Cc: Éric Piel <piel@delmic.com> Cc: Spencer E. Olson <olsonse@umich.edu> Signed-off-by:
-
