aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
* Unify handling of fuse_conn_info optionsGravatar Nikolaus Rath2016-10-15
| | | | | | Instead of using command line options to modify struct fuse_conn_info before and after calling the init() handler, we now give the file system explicit control over this.
* Merge branch 'fixup-lock-options'Gravatar Nikolaus Rath2016-10-15
|\
| * Re-activated lost no_remote_*lock options.Gravatar Nikolaus Rath2016-10-15
| |
* | tests: use freshly-build fusermount (instead of system version)Gravatar Nikolaus Rath2016-10-13
| | | | | | | | When running tests as non-root, make fusermount setuid root.
* | Travis: use "make test" instead of calling pytest directlyGravatar Nikolaus Rath2016-10-13
| |
* | Only test writeback cache if kernel is recent enough.Gravatar Nikolaus Rath2016-10-13
| |
* | Added lost ChangeLog entries for -o writeback_cache and -o async_dio.Gravatar Nikolaus Rath2016-10-13
| |
* | Make -o clone_fd into a parameter of session_loop_mt().Gravatar Nikolaus Rath2016-10-13
| | | | | | | | | | | | This option really affects the behavior of the session loop, not the low-level interface. Therefore, it does not belong in the fuse_session object.
* | Move session options into sub-structGravatar Nikolaus Rath2016-10-13
| | | | | | | | | | | | | | The session options are used only once to determine the proper conn->want flags. It is nice to have them clearly separated from the other struct fuse_session members that are used throughout the life of the file system.
* | Mention atomic_o_trunc capability in description of open() handler.Gravatar Nikolaus Rath2016-10-13
| |
* | Use NULL for option processing function where possible.Gravatar Nikolaus Rath2016-10-13
| |
* | Activate splice_read by default if write_buf is implemented.Gravatar Nikolaus Rath2016-10-13
|/
* do_init(): treat command line options consistentlyGravatar Nikolaus Rath2016-10-13
| | | | | | | Previously, some command line options would change the FUSE defaults but leave the final value to the file systems `init` handler while others would override any changes made by `init`. Now, command line options do both: they modify the default, *and* take precedence.
* Document when fuse_reply_data will use splice()Gravatar Nikolaus Rath2016-10-10
|
* Factored out LL_OPTIONS macroGravatar Nikolaus Rath2016-10-10
|
* Removed 'async_read' field in fuse_conn_infoGravatar Nikolaus Rath2016-10-10
| | | | This is redundant with the capability flags in `wants` and `capable`.
* Make several -o NN options work againGravatar Nikolaus Rath2016-10-10
| | | | | In commit 2ed7af, we accidentally set the default values *after* parsing the command line arguments.
* Fix race condition in notify_* examplesGravatar Nikolaus Rath2016-10-10
| | | | | | | The fix in commit cf4159156b was incomplete. While some false positives are caused by sleep() in the file system taking longer than expected, there was also a race condition where the file system would run before the contents are initialized properly.
* Added write cache tests.Gravatar Nikolaus Rath2016-10-10
|
* Use "se" instead of "f" for fuse_session pointer where possible.Gravatar Nikolaus Rath2016-10-10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | These changes were generated with the following Coccinelle semantic patch: @@ symbol f, se; // avoid unneeded warnings from Coccinelle @@ struct fuse_session * -f +se ; <... -f +se ...> @@ identifier fn; @@ fn(...,struct fuse_session * -f +se ,...) { <... -f +se ...> }
* fuse_main(): extend support for printing helpGravatar Nikolaus Rath2016-10-10
| | | | | | There's now a way to inhibit the "usage" line (which actually got lost in commit 225c12aebf2d), which makes it easier for simply file-systems to generate good-looking --help output.
* Use NULL as option processor where possible.Gravatar Nikolaus Rath2016-10-09
|
* fuse_main_real(): use fuse_parse_cmdline().Gravatar Nikolaus Rath2016-10-09
|
* Added ChangeLog entry for commit 225c12aebf2d2f27e1d0.Gravatar Nikolaus Rath2016-10-09
|
* Added cuse unit test.Gravatar Nikolaus Rath2016-10-09
|
* Renamed cuses example and added test programGravatar Nikolaus Rath2016-10-09
| | | | | | | | An earlier version of the fioclient.c example was intended to be used together with cusexmp.c. The former has since evolved into ioctl_client.c and no longer has the function necessary to test CUSE. Therefore, we've added a new cuse_client.c that is clearly associated with the cuse.c example file system.
* Increase timeouts in notify_* testsGravatar Nikolaus Rath2016-10-09
| | | | When running under Valgrind, we otherwise get sporadic test failures.
* Don't confuse lookup count for mountpoint and fileGravatar Nikolaus Rath2016-10-09
| | | | | I think this is the reason for a sporadic test failure, where fuse_lowlevel_notify_store() fails.
* Renamed timefsN examples to fuse_notify_*Gravatar Nikolaus Rath2016-10-09
| | | | | This should make it more obvious at first glance what the different examples do.
* Include testfsN examples in Doxygen documentation.Gravatar Nikolaus Rath2016-10-09
|
* Renamed ioctl and poll examplesGravatar Nikolaus Rath2016-10-09
| | | | | The new names should make it more obvious at first glance what each example demonstrates.
* Renamed some examples to make their function more obviousGravatar Nikolaus Rath2016-10-09
| | | | Also, added more comments for the same purpose.
* fuse_parse_cmdline(): do not print help/version textGravatar Nikolaus Rath2016-10-09
| | | | | The current behavior makes it difficult to add help for additional options. With the change, this becomes a lot easier.
* Added missing export of fuse_pkgversion.Gravatar Nikolaus Rath2016-10-09
|
* Merge pull request #80 from libfuse/master-proposedGravatar Nikolaus Rath2016-10-09
|\
| * Whitespace fix.Gravatar Nikolaus Rath2016-10-09
| |
| * Fix valgrind warningGravatar Nikolaus Rath2016-10-09
| | | | | | | | | | If we don't assign a value to padding, we get a warning about reading uninitialized data when sending the iovec to the kernel.
| * Fix documentation of fuse_parse_cmdline().Gravatar Nikolaus Rath2016-10-09
| | | | | | | | For --help and --version, it returns -1.
| * Dropped example/null.cGravatar Nikolaus Rath2016-10-09
|/ | | | | This does not seem to be working. Maybe because it tries to treat the mountpoint as a file rather than a directory?
* Added timefs3 to test notify_inval_entry.Gravatar Nikolaus Rath2016-10-08
| | | | Fixes #32.
* Add background and multithreading support to hello_ll and fuse_lo-plusGravatar Nikolaus Rath2016-10-08
|
* Update list of requests that can be answered with fuse_reply_none().Gravatar Nikolaus Rath2016-10-08
|
* testfs2: Also test fuse_notify_retrieve().Gravatar Nikolaus Rath2016-10-08
|
* Added timefs[12] examples.Gravatar Nikolaus Rath2016-10-08
| | | | | These examplesdemonstrate the use of the `fuse_lowlevel_notify_store` and `fuse_lowlevel_notify_inval_inode` functions.
* Added safe_sleep()Gravatar Nikolaus Rath2016-10-08
|
* Clarify that readdir() *may* report . and .. entries.Gravatar Nikolaus Rath2016-10-08
|
* Removed ``-o big_writes`` optionGravatar Nikolaus Rath2016-10-08
| | | | | | This option is obsolete and should always be enabled. File systems that want to limit the size of write requests should use the ``-o max_write=<N>`` option instead.
* Test more combinations of low-level options.Gravatar Nikolaus Rath2016-10-05
|
* Remove pointless aliasing of fuse_session *se to *fGravatar Nikolaus Rath2016-10-05
| | | | This is an overlooked artifact of the fuse_ll-fuse_session merge.
* Removed obsolete, unused *f member from struct fuse_session.Gravatar Nikolaus Rath2016-10-05
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-28 19:45:43 +0000