- 15 Aug, 2016 40 commits
-
-
Al Viro authored
Signed-off-by:
Al Viro <viro@zeniv.linux.org.uk> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Al Viro authored
... by using ITER_BVEC recvmsg Signed-off-by:
-
Al Viro authored
Signed-off-by:
-
wang di authored
Increase the post-work for the statahead thread in the readdir case since it can become very busy. Signed-off-by:
wang di <di.wang@intel.com> Reviewed-on: http://review.whamcloud.com/10761 Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4906Reviewed-by:
John L. Hammond <john.hammond@intel.com> Reviewed-by:
Andreas Dilger <andreas.dilger@intel.com> Reviewed-by:
Oleg Drokin <oleg.drokin@intel.com> Signed-off-by:
James Simmons <jsimmons@infradead.org> Signed-off-by:
-
wang di authored
Some cases we want to preserve the ctx->pos value or use a different value altogther. So allow the passing in of a position offset to ll_dir_read. Signed-off-by:
-
wang di authored
We have if (1) conditionals which is pointless so remove it and the next code block is never called so remove that as well. Signed-off-by:
Jinshan Xiong <jinshan.xiong@intel.com> Reviewed-by:
-
wang di authored
Make some of the error reporting more clear for the statahead thread startup and is_first_dirent() function. Signed-off-by:
-
wang di authored
The variables api32 and hash64 was renamed to is_api32 and is_hash64. Signed-off-by:
-
wang di authored
Change the done flag from integer to bool. Signed-off-by:
-
wang di authored
Invert the done flag test to reduce the code indentation. If done is true release the page and break out of the while loop. Signed-off-by:
-
wang di authored
The variable next needs only to be set when done is false. Signed-off-by:
-
wang di authored
Instead of making a large chunk of code conditional based on if a page is valid we reverse the page validity test and exit the loop if the page is invalid instead. This allows a section of code to reduce its indentation one level. Signed-off-by:
-
wang di authored
Remove debug message and struct ll_inode_info. Signed-off-by:
-
wang di authored
Add struct md_op_data as a parameter to ll_dir_read. Signed-off-by:
-
wang di authored
We report the inode size but never print in the debug message that the number reported is the size. Signed-off-by:
-
wang di authored
The readdir operations lock was incomplete. This patch fills in the missing pieces. Signed-off-by:
-
wang di authored
Newer lustre version on metadata servers support different version of lmv magic. This add the new data structures to handle these new lmv magic versions. Signed-off-by:
-
wang di authored
Cache directory striping information that the clients receive from the metadata servers. Signed-off-by:
-
wang di authored
Remove the wiretest check for this data structure. Signed-off-by:
-
wang di authored
Rename struct lmv_stripe_md in struct lustre_md from mea to lmv. Signed-off-by:
-
wang di authored
Add new flags for metadata handling. These flags are related to DNE2 handling. Signed-off-by:
-
James Simmons authored
Port the xattr functionality to the new xattr_handler API. This is smallest changes needed to move to this new API. The function ll_removexattr can be replaced by generic_removexattr as well since it also uses the xattr_handler set xattr backend. To tell the difference between the two cases we test the flag passed in for XATTR_REPLACE. The ll_getxattr function is replaced by the generic_getxattr function. Signed-off-by:
-
Ben Evans authored
These #if 0 blocks have been in place for years. Assume they are not used and remove them Signed-off-by:
Ben Evans <bevans@cray.com> Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8058 Reviewed-on: http://review.whamcloud.com/20414Reviewed-by:
James Simmons <uja.ornl@yahoo.com> Reviewed-by:
-
James Simmons authored
Split the function ll_getxattr_common into two functions. The code used for listing xattrs and ll_getxattr_common is placed into a new function ll_getxattr_list. This allows ll_listxattr to call directly ll_getxattr_list instead of going through ll_getxattr_common. This change is needed for the upcoming VFS move xattr_handler from [s|g]etxattr. Signed-off-by:
-
Bhaktipriya Shridhar authored
The workqueue is being used to run deferred work for the android binder. The "binder_deferred_workqueue" queues only a single work item and hence does not require ordering. Also, this workqueue is not being used on a memory recliam path. Hence, the singlethreaded workqueue has been replaced with the use of system_wq. System workqueues have been able to handle high level of concurrency for a long time now and hence it's not required to have a singlethreaded workqueue just to gain concurrency. Unlike a dedicated per-cpu workqueue created with create_singlethread_workqueue(), system_wq allows multiple work items to overlap executions even on the same CPU; however, a per-cpu workqueue doesn't have any CPU locality or global ordering guarantee unless the target CPU is explicitly specified and thus the increase of local concurrency shouldn't make any difference. Signed-off-by:
Bhaktipriya Shridhar <bhaktipriya96@gmail.com> Signed-off-by:
-
Didik Setiawan authored
fix checkpatch.pl warning about 'line over 80 characters'. Signed-off-by:
Didik Setiawan <didik.swn@gmail.com> Signed-off-by:
-
Gustavo Padovan authored
SW_SYNC allows to run tests on the sync_file framework via debugfs on <debugfs>/sync/sw_sync Opening and closing the file triggers creation and release of a sync timeline. To create fences on this timeline the SW_SYNC_IOC_CREATE_FENCE ioctl should be used. To increment the timeline value use SW_SYNC_IOC_INC. Also it exports Sync information on <debugfs>/sync/info Signed-off-by:
Gustavo Padovan <gustavo.padovan@collabora.co.uk> Reviewed-by:
Eric Engestrom <eric.engestrom@imgtec.com> Signed-off-by:
-
Gustavo Padovan authored
This interface is hidden from kernel headers and it is intended for use only for testing. So testers would have to add the ioctl information internally. This is to prevent misuse of this feature. v2: take in Eric suggestions for the Documentation v3: really take in Eric suggestions Signed-off-by:
-
Gustavo Padovan authored
remove file paths in the comments and add short description about each file. v2: remove file paths instead of just change them. v3: improve header description as sugggested by Eric Signed-off-by:
-
Gustavo Padovan authored
The common behaviour for trace headers is to have them in the same folder they are used, instead of creating a special trace/ directory. Signed-off-by:
-
Gustavo Padovan authored
Closing the timeline without waiting all fences to signal is not a critical failure, it is just bad usage from userspace so avoid calling WARN_ON in this case. Signed-off-by:
-
Laura Abbott authored
ion_reserve was supposed to be used to reserve memory in board files. These days, board files are no more and there are other more controlled mechanisms for reserving memory. Get rid of this function. Signed-off-by:
Laura Abbott <labbott@redhat.com> Reviewed-by:
Benjamin Gaignard <benjamin.gaignard@linaro.org> Signed-off-by:
-
Laura Abbott authored
ion_carveout_allocate and ion_carveout_free aren't used outside of the carveout heap. Get rid of the definitions. Signed-off-by:
-
Laura Abbott authored
The map_dma API interface was designed to generate an sg_table. Currently, every client just creates the table at allocation time and then returns the one table. Nothing happens on unmap_dma either. Just get rid of the API and assign the sg_table directly. Signed-off-by:
-
Laura Abbott authored
ion_phys was an interface used for older legacy behavior. sg_tables are the standard now. Get rid of it. Signed-off-by:
-
Laura Abbott authored
The ion_sg_table interface is mostly a reimplementation of what dma_buf is doing. Clients should be using dma_buf APIs instead. Signed-off-by:
-
Colin Ian King authored
priv is being dereferenced before a check for it being null is made, so there is a possibililty a null pointer deference can occur. Instead, only dereference priv if it is non-null. Signed-off-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
Bing Sun authored
Fixed coding style issue: Enclose multiple statements macros definition in a do while loop. Use one space around binary operators. Signed-off-by:
Bing Sun <sunbing@redflag-linux.com> Reviewed-by:
Wolfram Sang <wsa@the-dreams.de> Signed-off-by:
-
Markus Elfring authored
The local variable "rc" was assigned a zero at one place. But it was not read within this function. Thus delete it. Signed-off-by:
Markus Elfring <elfring@users.sourceforge.net> Reviewed-by:
-
Markus Elfring authored
A few variables were assigned a null pointer despite of the detail that they were immediately reassigned by the following statement. Thus remove such unnecessary assignments. Signed-off-by:
-
