Move the extern declaration to a header file common to all users of the
variable. This fixes the following sparse warning:
symbol 'ldlm_cancel_unused_locks_before_replay' was not declared.
Should it be static?
Signed-off-by: Imre Deak <imre.deak@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The preferred indentation for cases and switches has the cases at
the same level as the switch.
Signed-off-by: Christopher H. Pezley <chris@pezley.net> Acked-by: Laura Abbott <labbott@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Staging: ks7010: remove unused function in ks_wlan_net.c
We get 1 warning when building kernel with W=1:
drivers/staging/ks7010/ks_wlan_net.c:3520:5: warning: no previous prototype for 'ks_wlan_reset' [-Wmissing-prototypes]
In fact, these functions are unused in
ks_wlan_net.c, but should be removed.
staging: rtl8192u: remove unused functions in r8192U_core.c
We get 2 warnings when building kernel with W=1:
drivers/staging/rtl8192u/r8192U_core.c:925:12: warning: no previous declaration for 'ieeerate2rtlrate' [-Wmissing-declarations]
drivers/staging/rtl8192u/r8192U_core.c:958:12: warning: no previous declaration for 'rtl8192_rate2rate' [-Wmissing-declarations]
drivers/staging/rtl8192u/r8192U_core.c:1322:11: warning: no previous declaration for 'rtl8192_IsWirelessBMode' [-Wmissing-declarations]
In fact, these functions are unused in
r8192U_core.c, but should be removed.
staging: rtl8192u: ieee80211: ieee80211_softmac: mark symbols static where possible
We get 5 warnings when building kernel with W=1:
drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c:287:13: warning: no previous declaration for 'softmac_ps_mgmt_xmit' [-Wmissing-declarations]
drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c:323:24: warning: no previous declaration for 'ieee80211_probe_req' [-Wmissing-declarations]
drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c:643:24: warning: no previous declaration for 'ieee80211_authentication_req' [-Wmissing-declarations]
drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c:981:24: warning: no previous declaration for 'ieee80211_association_req' [-Wmissing-declarations]
drivers/staging/rtl8192u/ieee80211/ieee80211_softmac.c:3094:24: warning: no previous declaration for 'ieee80211_disassociate_skb' [-Wmissing-declarations]
In fact, these functions are only used in the file in which they are
declared and don't need a declaration, but can be made static.
so this patch marks these functions with 'static'.
staging: android: ion: mark symbols static where possible
We get 4 warnings when building kernel with W=1:
drivers/staging/android/ion/ion_carveout_heap.c:36:17: warning: no previous prototype for 'ion_carveout_allocate' [-Wmissing-prototypes]
drivers/staging/android/ion/ion_carveout_heap.c:50:6: warning: no previous prototype for 'ion_carveout_free' [-Wmissing-prototypes]
drivers/staging/android/ion/ion_of.c:28:5: warning: no previous prototype for 'ion_parse_dt_heap_common' [-Wmissing-prototypes]
drivers/staging/android/ion/ion_of.c:54:5: warning: no previous prototype for 'ion_setup_heap_common' [-Wmissing-prototypes]
In fact, these functions are only used in the file in which they are
declared and don't need a declaration, but can be made static.
so this patch marks these functions with 'static'.
Christian Gromm [Fri, 23 Sep 2016 13:20:02 +0000 (15:20 +0200)]
staging: most: aim-cdev: make syscall write accept buffers of arbitrary size
This patch allows to call the write() function for synchronous and
isochronous channels with buffers of any size. The AIM simply waits for
data to fill up the MOST buffer object according to the network interface
controller specification for streaming channels, before it submits the
buffer to the HDM.
The new behavior is backward compatible to the old applications, since
all known applications needed to fill the buffer completely anyway.
Signed-off-by: Christian Gromm <christian.gromm@microchip.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: rtl8188eu: hal: rtl8188e_cmd: Use ether_addr_copy() instead of memcpy()
The checkpatch.pl found the warning:
WARNING: Prefer ether_addr_copy() over memcpy() if the Ethernet
addresses are __aligned(2)
Checked if the the Ethernet addresses are __aligned(2) by using pahole
tool. The type of pwlanhdr is struct ieee80211_hdr and pahole shows that
addr1, addr2, and addr3 are aligned to u16.
staging: greybus: arche-apb-ctrl: declare struct of_device_id as const
Declare the 'arche_apb_ctrl_of_match' array as const. This array is only stored
in the .of_match_table field of a device_driver structure, which is declared as
const. This addresses the checkpatch warning:
WARNING: struct of_device_id should be normally const
Signed-off-by: Eva Rachel Retuya <eraretuya@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: greybus: arche-platform: declare struct of_device_id as const
Declare the arrays 'arche_platform_of_match' and 'arche_combined_id' as const.
'arche_platform_of_match' is only stored in the .of_match_table field of a
device_driver structure, which is declared as const.
'arche_combined_id' is passed to MODULE_DEVICE_TABLE. This macro does not
modify the array and therefore the array can be declared as const. Checkpatch
pointed out both issues.
WARNING: struct of_device_id should be normally const
Signed-off-by: Eva Rachel Retuya <eraretuya@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: ks7010: Remove unnecessary variable used to store return value
This patch removes an unnecessary variable used to store return values
in order to reduce memory usage.
Done using coccinelle:
@@
type T;
constant C;
identifier ret;
@@
- T ret = C;
... when != ret
return
- ret
+ C
;
Remove 'x=a;' from a consecutive double assignment of the form 'x=a; x=b;'.
Issue detected by the semantic patch found here:
https://github.com/coccinelle/coccinellery/blob/master/write2/write2.cocci
staging: xgifb: Remove unused variable and if statement
Setting the variable 'refresh_rate = 60;' has no use to the rest of the
function. Hence the if statement setting it and the variable itself can be
removed. This was detected using the following Coccinelle semantic patch:
@@
type T;
identifier i;
constant C;
@@
(
extern T i;
|
- T i;
<+... when != i
- i = C;
...+>
)
staging: greybus: arche-platform: compress return logic into one line
Modify return statement to use the value being returned directly instead of
assigning it first to 'ret' and returning this variable. Coccinelle semantic
patch used:
@@
expression e;
local idexpression ret;
@@
-ret =
+return
e;
-return ret;
Signed-off-by: Eva Rachel Retuya <eraretuya@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Christian Gromm [Wed, 21 Sep 2016 12:49:09 +0000 (14:49 +0200)]
staging: most: clean up configuration strings
This patch adds the strings 'rx', 'tx' and 'isoc' to the list of accepted
identifiers when setting up a channel configuration. To keep consistency it
removes the prefix "dir_" from strings returned by the attributes
set_direction and available_directions and it removes the suffix "_avp"
from the string "isoc_avp" returned by the attributes set_datatype and
available_datatypes.
Signed-off-by: Christian Gromm <christian.gromm@microchip.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Christian Gromm [Wed, 21 Sep 2016 12:49:05 +0000 (14:49 +0200)]
staging: most: make function most_submit_mbo return void
Function most_submit_mbo() causes an exception only if either the pointer
mbo or mbo->context equals NULL. From the underlying architecture's point
of view both cases must _not_ come true and would happen only, if something
has tampered with the pointers. This would render runtime code unable to
recover anyway. So, instead trying to hide that things are already
critically out of control we're better off with a WARN_ON() assertion.
This patch replaces the return type of the function most_submit_mbo() with
'void' and adds a WARN_ONCE() assertion. Additionally, code calling the
function is adapted accordingly.
Signed-off-by: Christian Gromm <christian.gromm@microchip.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Being over zealous in removing unused EXPORT_SYMBOLs the function
lustre_swab_lov_mds_md exports were removed. They need to be
exported so this patch restores those EXPORT_SYMBOLS. Same
mistake was done when porting to the upstream client.
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: http://review.whamcloud.com/14545
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6486 Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com> Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: rts5208: Put constant on right side of comparison
Replaces position of constant from left to right side of a comparison.
Additionally, modifies logical continuations to be on the previous line and
fixes alignment to match open parenthesis. Issues found by checkpatch.
Instead of storing the return value into a variable and then returning it, we
can club the two into a single return statement. This change was made using
the following semantic patch by Coccinelle:
staging: media: cxd2099: replace printk with dev_info
Use dev_info() instead of printk(). Majority of the conversion was done
previously on this commit: 011b2aa [media] staging/media: Use dev_ printks in cxd2099/cxd2099.[ch]
Signed-off-by: Eva Rachel Retuya <eraretuya@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: media: bcm2048: Merge assignment with return
Instead of storing the return value into a variable and then returning it, we
can club the two into a single return statement. This change was made using
the following semantic patch by Coccinelle:
staging: media: davinci_vpfe: Merge assignment with return
Instead of storing the return value into a variable and then returning it, we
can club the two into a single return statement. This change was made using
the following semantic patch by Coccinelle: