Age | Commit message (Collapse) | Author | Lines | |
---|---|---|---|---|
2017-06-05 | examples/passthrough_ll: added support for create() | Nikolaus Rath | -5/+27 | |
2017-06-05 | example/passthrough_ll: added write support | Nikolaus Rath | -2/+24 | |
2017-05-31 | notify_store_retrieve(): fix race on unmount | Nikolaus Rath | -4/+11 | |
update_fs_loop() is still running when the filesystem unmounts, but it that case calls to fuse_lowlevel_notify_* will fail. Fixes: #105. | ||||
2017-05-31 | Clarify that destroy() is called without kernel connection | Nikolaus Rath | -2/+4 | |
2017-05-31 | example/notify_store_retrieve: add debugging code for issue #105. | Nikolaus Rath | -2/+10 | |
2017-05-31 | Revert "example/passthrough.c: add debugging code for issue #157." | Nikolaus Rath | -9/+2 | |
This reverts commit 179fa13b40387645e722089873079488f9dbe3d8. | ||||
2017-05-31 | tst_link(): wait for RELEASE request | Nikolaus Rath | -9/+16 | |
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. | ||||
2017-05-25 | Factored out C-based tests | Nikolaus Rath | -48/+65 | |
2017-05-25 | Fix typo in comment. | Nikolaus Rath | -1/+1 | |
2017-05-25 | Document RENAME_EXCHANGE and RENAME_NOREPLACE flags. | Nikolaus Rath | -2/+21 | |
2017-05-25 | passthrough_ll: document that fuse_ino_t and uintptr_t sizes must match | Nikolaus Rath | -0/+13 | |
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. | ||||
2017-05-25 | example/passthrough.c: add debugging code for issue #157. | Nikolaus Rath | -2/+9 | |
2017-05-25 | tst_link(): add more assertions | Nikolaus Rath | -3/+17 | |
Hopefully this helps debugging issue #157. | ||||
2017-05-25 | Remove fuse_fs_fgetattr and fuse_fs_ftruncate from linker script | pablomh | -2/+0 | |
They were removed from source here: https://github.com/libfuse/libfuse/commit/73b6ff4b75cf1228ea61262c293fcb2fda5dfeea | ||||
2017-05-24 | Travis-CI: Don't abort on first failed test | Nikolaus Rath | -4/+6 | |
This should help debugging issue #157. | ||||
2017-05-24 | Added tst_open_read() | Nikolaus Rath | -0/+9 | |
Slightly increases coverage of examples/passthrough_ll.c (which supports open for reading, but not for writing). | ||||
2017-05-24 | lib/meson.build: don't crash if there's no libdl | Nikolaus Rath | -1/+1 | |
For example, FreeBSD doesn't have it. Fixes: #173. | ||||
2017-05-24 | fuse_signals.c: use new do_nothing function instead of SIG_IGN | Nikolaus Rath | -2/+18 | |
Fixes: #160. | ||||
2017-05-24 | Released libfuse 3.0.2 | Nikolaus Rath | -3/+17 | |
2017-05-24 | Make tests build on bsd | Brian Naylor | -0/+10 | |
2017-04-20 | make buffer size match kernel max transfer size | Carlos Maiolino | -3/+5 | |
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> | ||||
2017-04-14 | Set default options before parsing | Tej Chajed | -5/+5 | |
2017-04-12 | Improved documentation of fuse_context.private_data | Nikolaus Rath | -13/+20 | |
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. | ||||
2017-04-10 | Released 3.0.1 | Nikolaus Rath | -3/+8 | |
2017-04-10 | Fix travis build script. Broken in e372d. | Nikolaus Rath | -3/+3 | |
2017-04-10 | Added ChangeLog for commits c24cc to eb972. | Nikolaus Rath | -1/+4 | |
2017-04-10 | Remove 'tests' target in favor of calling py.test directly. | Nikolaus Rath | -14/+6 | |
Together with the previous commit, this fixes #156. | ||||
2017-04-10 | Build tests by default. | Nikolaus Rath | -9/+8 | |
2017-04-10 | Only build test/*.py if out of date. | Nikolaus Rath | -1/+1 | |
2017-04-07 | Actually test passthrough* examples | Nikolaus Rath | -1/+1 | |
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 | ||||
2017-04-07 | passthrough: implemented create() | Nikolaus Rath | -0/+14 | |
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. | ||||
2017-04-07 | example/passthrough: use fi->fh for read, write, fallocate | Nikolaus Rath | -7/+21 | |
No reason not to use it. May even be a little faster and will consume less resources :-). | ||||
2017-04-07 | passthrough:truncate(): work on file descriptor when possible | Nikolaus Rath | -2/+4 | |
This allows truncating an open file even if write permission was removed after open() (which is the expected behavior). | ||||
2017-04-07 | example/passthrough: close open files in release() | Nikolaus Rath | -5/+2 | |
That way we can use the file descriptor for other operations. | ||||
2017-04-07 | passthrough, passthrough_fh: disable attribute caching | Nikolaus Rath | -0/+24 | |
Required for better hardlink handling, see comments in patch. | ||||
2017-04-07 | test_examples(): test without debug messages first | Nikolaus Rath | -1/+1 | |
That way, we are not drowning in messages when a test would also fail without debugging enabled. | ||||
2017-04-07 | test_examples(): avoid false positives from fuse debug output | Nikolaus Rath | -1/+7 | |
2017-04-07 | passthrough_ll: only test functions that are actually provided | Nikolaus Rath | -19/+23 | |
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. | ||||
2017-04-07 | Fix rst markup. | Nikolaus Rath | -2/+1 | |
2017-04-07 | Document true meaning of the 'use_ino' option. | Nikolaus Rath | -2/+10 | |
2017-04-07 | Remove checked_unlink() | Nikolaus Rath | -18/+7 | |
There is no reason why so many tests require the file system to support unlink() and/or rmdir(). | ||||
2017-04-07 | Turn tst_mknod() into tst_create() | Nikolaus Rath | -8/+15 | |
Ensure that we are really creating a new file. Don't attempt to write, we do that in tst_open_write(). | ||||
2017-04-07 | Renamed tst_write() to tst_open_write() | Nikolaus Rath | -7/+12 | |
We are actually testing both opening of an existing file and writing to it. | ||||
2017-04-07 | Added tst_unlink() | Nikolaus Rath | -0/+13 | |
To check for unlink() support without requiring create()/mknod(). | ||||
2017-04-07 | tst_mkdir(): factor out tst_rmdir() | Nikolaus Rath | -3/+14 | |
This allows testing a filesystem that offers mkdir(), but no rmdir() (and vice versa). | ||||
2017-04-07 | Rename tst_unlink() to tst_open_unlink() | Nikolaus Rath | -4/+4 | |
This makes more sense, since we are specifically checking unlinking of an open file. | ||||
2017-04-07 | tst_readdir(): don't require create/mkdir support | Nikolaus Rath | -3/+3 | |
By creating the files in the lower filesystem, we can test readdir() even for filesystems that don't implement create() or mkdir(). | ||||
2017-04-07 | passthrough_ll: document that functionality is restricted | Nikolaus Rath | -1/+5 | |
2017-03-28 | hello.c: don't use constant instead of magic number | guraga | -1/+1 | |
2017-03-16 | Document that -o auto_unmount implies -o nodev,nosuid | Nikolaus Rath | -0/+5 | |
See also issue #148. |