aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
* Install init script in $DESTDIR/etc, not $prefix/$sysconfdirGravatar Nikolaus Rath2017-06-21
| | | | Fixes: #178.
* tst_readdir(): actually go through FUSEGravatar Nikolaus Rath2017-06-20
| | | | The previous code didn't actually go through the mountpoint at all.
* Clarify *_timeout semanticsGravatar Nikolaus Rath2017-06-08
| | | | | When changes always come through kernel, timeouts should be large.
* Describe supported platforms.Gravatar Nikolaus Rath2017-06-06
|
* getgroups(): clarify codeGravatar Nikolaus Rath2017-06-05
| | | | | | read() return value should always be positive or -1. However, since we cast to unsigned a little later, it's clearer to check for non-negativity.
* Fix comparison of integers of different signsGravatar Angelo G. Del Regno2017-06-05
| | | | | | | | Some variables of different size and sign were getting compared without any safe casting. The build system also throws warnings at this and, being this library used for filesystems, it's really important to ensure stability.
* examples/passthrough_ll: added support for create()Gravatar Nikolaus Rath2017-06-05
|
* example/passthrough_ll: added write supportGravatar Nikolaus Rath2017-06-05
|
* notify_store_retrieve(): fix race on unmountGravatar Nikolaus Rath2017-05-31
| | | | | | | update_fs_loop() is still running when the filesystem unmounts, but it that case calls to fuse_lowlevel_notify_* will fail. Fixes: #105.
* Clarify that destroy() is called without kernel connectionGravatar Nikolaus Rath2017-05-31
|
* example/notify_store_retrieve: add debugging code for issue #105.Gravatar Nikolaus Rath2017-05-31
|
* Revert "example/passthrough.c: add debugging code for issue #157."Gravatar Nikolaus Rath2017-05-31
| | | | This reverts commit 179fa13b40387645e722089873079488f9dbe3d8.
* tst_link(): wait for RELEASE requestGravatar Nikolaus Rath2017-05-31
| | | | | | | | | | Since RELEASE requests are asynchronous, it is possible that libfuse still considers the file to be open when userspace has closed it, so that a successive unlink() call from userspace actually triggers a rename(). We avoid the resulting test failure by re-trying a few times. Fixes: #157.
* Factored out C-based testsGravatar Nikolaus Rath2017-05-25
|
* Fix typo in comment.Gravatar Nikolaus Rath2017-05-25
|
* Document RENAME_EXCHANGE and RENAME_NOREPLACE flags.Gravatar Nikolaus Rath2017-05-25
|
* passthrough_ll: document that fuse_ino_t and uintptr_t sizes must matchGravatar Nikolaus Rath2017-05-25
| | | | | | | We can consider changing the code to drop the requirement if it turns out that there's a system where this isn't given. Fixes issue #167.
* example/passthrough.c: add debugging code for issue #157.Gravatar Nikolaus Rath2017-05-25
|
* tst_link(): add more assertionsGravatar Nikolaus Rath2017-05-25
| | | | Hopefully this helps debugging issue #157.
* Remove fuse_fs_fgetattr and fuse_fs_ftruncate from linker scriptGravatar pablomh2017-05-25
| | | They were removed from source here: https://github.com/libfuse/libfuse/commit/73b6ff4b75cf1228ea61262c293fcb2fda5dfeea
* Travis-CI: Don't abort on first failed testGravatar Nikolaus Rath2017-05-24
| | | | This should help debugging issue #157.
* Added tst_open_read()Gravatar Nikolaus Rath2017-05-24
| | | | | Slightly increases coverage of examples/passthrough_ll.c (which supports open for reading, but not for writing).
* lib/meson.build: don't crash if there's no libdlGravatar Nikolaus Rath2017-05-24
| | | | | | For example, FreeBSD doesn't have it. Fixes: #173.
* fuse_signals.c: use new do_nothing function instead of SIG_IGNGravatar Nikolaus Rath2017-05-24
| | | | Fixes: #160.
* Released libfuse 3.0.2Gravatar Nikolaus Rath2017-05-24
|
* Make tests build on bsdGravatar Brian Naylor2017-05-24
|
* make buffer size match kernel max transfer sizeGravatar Carlos Maiolino2017-04-20
| | | | | | | | | | | | | | | Currently libfuse has a hardcoded buffer limit to 128kib, while fuse kernel module has a limit up to 32 pages. This patch changes buffer limit to match the current page size, instead of assuming 4096 bytes pages, enabling architectures with bigger pages to use larger buffers, improving performance. Also, add a new macro (HEADER_SIZE) to specify the space needed to accommodate the header, making it easier to understand why those extra 4096 bytes are needed Signed-off-by: Carlos Maiolino <cmaiolino-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
* Set default options before parsingGravatar Tej Chajed2017-04-14
|
* Improved documentation of fuse_context.private_dataGravatar Nikolaus Rath2017-04-12
| | | | | | | | | | In particular, don't call it "user_data" in one place and "private_data" elsewhere. Changing the name of the variable in the prototype should not affect backwards compatibility. Fixes: #155.
* Released 3.0.1Gravatar Nikolaus Rath2017-04-10
|
* Fix travis build script. Broken in e372d.Gravatar Nikolaus Rath2017-04-10
|
* Added ChangeLog for commits c24cc to eb972.Gravatar Nikolaus Rath2017-04-10
|
* Remove 'tests' target in favor of calling py.test directly.Gravatar Nikolaus Rath2017-04-10
| | | | Together with the previous commit, this fixes #156.
* Build tests by default.Gravatar Nikolaus Rath2017-04-10
|
* Only build test/*.py if out of date.Gravatar Nikolaus Rath2017-04-10
|
* Actually test passthrough* examplesGravatar Nikolaus Rath2017-04-07
| | | | | | | | Since os.path.join() interprets leading slashes, we were actually never accessing the mountpoint and doing all the tests in the source directory. Fixes: #139
* passthrough: implemented create()Gravatar Nikolaus Rath2017-04-07
| | | | | | This allows calls like open(file, O_CREAT|O_RDONLY, 0200) which would otherwise fail because we cannot open the file after mknod() has created it with 0200 permissions.
* example/passthrough: use fi->fh for read, write, fallocateGravatar Nikolaus Rath2017-04-07
| | | | | No reason not to use it. May even be a little faster and will consume less resources :-).
* passthrough:truncate(): work on file descriptor when possibleGravatar Nikolaus Rath2017-04-07
| | | | | This allows truncating an open file even if write permission was removed after open() (which is the expected behavior).
* example/passthrough: close open files in release()Gravatar Nikolaus Rath2017-04-07
| | | | That way we can use the file descriptor for other operations.
* passthrough, passthrough_fh: disable attribute cachingGravatar Nikolaus Rath2017-04-07
| | | | Required for better hardlink handling, see comments in patch.
* test_examples(): test without debug messages firstGravatar Nikolaus Rath2017-04-07
| | | | | That way, we are not drowning in messages when a test would also fail without debugging enabled.
* test_examples(): avoid false positives from fuse debug outputGravatar Nikolaus Rath2017-04-07
|
* passthrough_ll: only test functions that are actually providedGravatar Nikolaus Rath2017-04-07
| | | | | | This appeared to work because of an unrelated bug that caused us to actually never access the mountpoint at all and do all tests on the lower filesystem. This issue will be fixed in a separate commit.
* Fix rst markup.Gravatar Nikolaus Rath2017-04-07
|
* Document true meaning of the 'use_ino' option.Gravatar Nikolaus Rath2017-04-07
|
* Remove checked_unlink()Gravatar Nikolaus Rath2017-04-07
| | | | | There is no reason why so many tests require the file system to support unlink() and/or rmdir().
* Turn tst_mknod() into tst_create()Gravatar Nikolaus Rath2017-04-07
| | | | | Ensure that we are really creating a new file. Don't attempt to write, we do that in tst_open_write().
* Renamed tst_write() to tst_open_write()Gravatar Nikolaus Rath2017-04-07
| | | | | We are actually testing both opening of an existing file and writing to it.
* Added tst_unlink()Gravatar Nikolaus Rath2017-04-07
| | | | To check for unlink() support without requiring create()/mknod().
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-25 22:31:53 +0000