- 23 Aug, 2017 34 commits
-
-
James Simmons authored
Check if lcfg passed in is NULL and if it is just return NULL. Signed-off-by:
James Simmons <uja.ornl@yahoo.com> Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6245 Reviewed-on: https://review.whamcloud.com/22138Reviewed-by:
Dmitry Eremin <dmitry.eremin@intel.com> Reviewed-by:
Jinshan Xiong <jinshan.xiong@intel.com> Reviewed-by:
Fan Yong <fan.yong@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>
-
James Simmons authored
Change i from int to __u32 to match lcfg_bufcount field. Also this matches what the other functions also do. Signed-off-by:
Quentin Bouget <quentin.bouget@cea.fr> Reviewed-by:
Ben Evans <bevans@cray.com> Reviewed-by:
-
James Simmons authored
Several libcfs.h macros are present in lustre_cfg.h, but libcfs.h is not available so lets replace those macros with real code. The header libcfs.h also provides errno.h so with libcfs.h gone include errno.h directly. Signed-off-by:
-
James Simmons authored
Move lustre_cfg.h to its proper place. Adjust additonal headers included in lustre_cfg.h. Signed-off-by:
-
James Simmons authored
In class_config_llog_handler() a flag, inst, was used to determine if it was safe to call kfree. Its not needed so remove the flag. Signed-off-by:
-
James Simmons authored
To perserve the truncate warning move lustre_cfg_string() to obd_config.c. A identical function was created for userland. This function was a bit big for a inline function. Signed-off-by:
-
James Simmons authored
The inline function lustre_cfg_new() calls kzalloc() but this is a UAPI header. Remove kzalloc() and rename the function to lustre_cfg_init(). The lustre kernel code that was calling lustre_cfg_new() can doing the memory allocation and pass the new buffer to lustre_cfg_init() to fill in. Signed-off-by:
-
James Simmons authored
Replace all lustre_cfg_free() calls with direct kfree() call. Signed-off-by:
-
James Simmons authored
Remove the white space in the lustre_param.h header and align the code so its easier to read. Signed-off-by:
John L. Hammond <john.hammond@intel.com> Reviewed-by:
-
James Simmons authored
Change LUSTRE_PARAM_H_ to _UAPI_LUSTRE_PARAM_H_ now that is a proper UAPI header Signed-off-by:
-
James Simmons authored
Move the header lustre_param.h to proper uapi directory. Signed-off-by:
-
James Simmons authored
All the included headers in lustre_param.h are lustre kernel internal headers and they are not even needed. Just remove them. Signed-off-by:
-
James Simmons authored
Move all the kernel specific function prototypes from lustre_param.h into obd_config.h which is a kernel only header. Signed-off-by:
-
James Simmons authored
Change LUSTRE_IOCTL_H_ to _UAPI_LUSTRE_IOCTL_H_ now that is a proper UAPI header Signed-off-by:
Andreas Dilger <andreas.dilger@intel.com> Reviewed-by:
-
James Simmons authored
With the cleanups we no longer need to drag in tons of user land headers. Lets simplify the include headers. Signed-off-by:
-
James Simmons authored
Replace cfs_size_round() standard __ALIGN_KERNEL macro. This removes the dependency of libcfs.h which is a kernel only header. Signed-off-by:
-
James Simmons authored
Move the header lustre_ioctl.h to proper uapi directory. Signed-off-by:
-
James Simmons authored
The inline function obd_ioctl_is_invalid() is no longer needed by userland and also the function was pretty bug for a inline function. Since this is the case we can move this kernel only code to the linux-module.c which is the only place it is used. Signed-off-by:
-
James Simmons authored
Move obd_ioctl_getdata() from lustre_ioctl.h to obd_class.h since this function is only used by kernel space. Signed-off-by:
-
James Simmons authored
Replace obd_ioctl_popdata() with direct copy_to_user() call. Signed-off-by:
-
James Simmons authored
Replace obd_ioctl_freedata() with direct kvfree() call. Signed-off-by:
-
James Simmons authored
The header lustre_ioctl.h has both kernel and user space versions of obd_ioctl_*() functions. Lets remove the user land version which have been integrated into the lustre tools. Signed-off-by:
-
James Simmons authored
Delete code that is used by only the server or just in user land. Now this file is a normal kernel header. The UAPI version of this header is only needed for server side so no need to create a new header in uapi directory. Signed-off-by:
Lai Siyao <lai.siyao@intel.com> Reviewed-by:
-
Ben Evans authored
Move lu_fid functions into lustre/lustre_fid.h Move ost_id functions into lustre/lustre_ostid.h Fix indenting, include new headers as needed. Several inline functions in the header lustre_ostid.h are using debug macros instead of returning proper errors. Remove the debug macros and properly handle the returned error codes. Place both UAPI headers lustre_fid.h and lustre_ostid.h into the uapi directory. Signed-off-by:
Frank Zago <fzago@cray.com> Reviewed-by:
-
Ben Evans authored
The wiki has been revived so update the URL about the FID design in lustre_fid.h. Signed-off-by:
-
Ben Evans authored
Currently the function ostid_set_id() just logs an error and just keeps going. This function is also used by user land tools so having special lustre kernel debugging code is not desired. So instead just return an error code and have the calling kernel code print the error instead. Signed-off-by:
-
James Simmons authored
The inline function fid_is_namespace_visible and ost_fid_from_resid are only used for server code so we can remove it. Signed-off-by:
-
Ben Evans authored
Migrate functions set/get_mrc_cr_flags, ldlm_res_eq ldlm_extent_overlap, ldlm_extent_contain, ldlm_request_bufsize, and all the PTLRPC dump_* functions out of lustre_idl.h which is a UAPI header to the places in the kernel code they are actually used. Delete unused lmv_mds_md_stripe_count and agent_req_in_final_state. Signed-off-by:
-
Colin Ian King authored
Trivial fixes to spelling mistake in pr_err error messages: errorr -> error mismactch -> mismatch Signed-off-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
Colin Ian King authored
The function pointer tcpm_get_pwr_opmode is local to the source and does not need to be in global scope, so make it static. Cleans up sparse warning: symbol 'tcpm_get_pwr_opmode' was not declared. Should it be static? Signed-off-by:
-
Srishti Sharma authored
Fixed alignment so that it matched open parenthesis Signed-off-by:
Srishti Sharma <srishtishar@gmail.com> Signed-off-by:
-
Arvind Yadav authored
snd_soc_dai_ops are not supposed to change at runtime. All functions working with snd_soc_dai_ops provided by <sound/soc-dai.h> work with const snd_soc_dai_ops. So mark the non-const structs as const. Signed-off-by:
Arvind Yadav <arvind.yadav.cs@gmail.com> Acked-by:
Mark Greer <mgreer@animalcreek.com> Signed-off-by:
-
Eames Trinh authored
Fixed a spelling error. Signed-off-by:
Eames Trinh <eamestrinh@gmail.com> Signed-off-by:
-
Randy Dunlap authored
Fix many build errors when NETDEVICES is not enabled. Signed-off-by:
Randy Dunlap <rdunlap@infradead.org> Cc: Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by:
-
- 22 Aug, 2017 6 commits
-
-
David Kershner authored
Greg noticed that we were creating our own attribute groups, instead of using the ATTRIBUTE_GROUPS macro, this patch fixes our mistake. Reported-by:
David Kershner <david.kershner@unisys.com> Reviewed-by:
Tim Sell <timothy.sell@unisys.com> Signed-off-by:
-
Sameer Wadgaonkar authored
The function visorchipset_device_destroy just called remove_visor_device, we can just combine it with remove_visor_device. Signed-off-by:
Sameer Wadgaonkar <sameer.wadgaonkar@unisys.com> Signed-off-by:
-
Sameer Wadgaonkar authored
The function visorchipset_device_create just called create_visor_device, we can just combine it with create_visor_device. Signed-off-by:
-
Sameer Wadgaonkar authored
The function visorchipset_bus_destroy just called viosrbus_remove_instance, we can just combine it with visorbus_remove_instance. Signed-off-by:
-
David Binder authored
Adds the proper Unisys copyright statement to the top of iochannel.h. Signed-off-by:
David Binder <david.binder@unisys.com> Signed-off-by:
-
David Binder authored
The first line of the top file comment should begin on the line following the block comment opening, thereby following the convention used elsewhere in the driver set. Signed-off-by:
-
