aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2016-07-04tests: OSX: cast st.Blksize to int64Jakob Unterwurzacher
On Linux, Blksize is already int64. But on OSX it is int32 and that caused a compile failure. https://github.com/rfjakob/gocryptfs/issues/15
2016-07-04prefer_openssl: OSX: get rid of /proc warningsJakob Unterwurzacher
OSX does not have /proc/cpuinfo, but let's not warn the user about it.
2016-07-04stupidgcm: use __builtin_trap()Jakob Unterwurzacher
[...]/stupidgcm/locking.go:16:2: warning: indirection of non-volatile null pointer will be deleted, not trap [-Wnull-dereference] [...]/stupidgcm/locking.go:16:2: note: consider using __builtin_trap() or qualifying pointer with 'volatile' https://github.com/rfjakob/gocryptfs/issues/15
2016-07-03XFSTESTS.md: add output from latest fuse-xfstestsJakob Unterwurzacher
fuse-xfstests is regularily rebased to xfstests master.
2016-07-03syscallcompat: OSX: add Mknodat wrapperJakob Unterwurzacher
Protip: find naked *at syscalls using: git grep "syscall." | grep "at(" | grep -v syscallcompat
2016-07-03syscallcompat: OSX: add Unlinkat wrapperJakob Unterwurzacher
Also, replace remaining naked syscall.Openat calls.
2016-07-03fusefrontend: downgrade Renameat log message to debugJakob Unterwurzacher
2016-07-03syscallcompat: OSX: add Renamat wrapperJakob Unterwurzacher
Adds a poor man's renameat implementation for OSX.
2016-07-03syscallcompat: OSX: add Fallocate and Openat wrappersJakob Unterwurzacher
...and convert all calls to syscall.{Fallocate,Openat} to syscallcompat . Both syscalls are not available on OSX. We emulate Openat and just return EOPNOTSUPP for Fallocate.
2016-07-03syscallcompat: move syscall wrapper to their own packageJakob Unterwurzacher
We will get more of them as OSX also lacks support for openat.
2016-07-03main: bake build date into version stringJakob Unterwurzacher
$ gocryptfs -version gocryptfs v0.12-36-ge021b9d-dirty; go-fuse a4c968c; 2016-07-03 go1.6.2
2016-07-03main: yellow text for masterkey and zerokey warningsJakob Unterwurzacher
Also, mention that the shell may save the command line into the history and reduce "ps -auxwww" to "ps ax" (that's all you need).
2016-07-03nametransform: hide detailed padding error behind the debug flagJakob Unterwurzacher
unPad16 returns detailed errors including the position of the incorrect bytes. Kill a possible padding oracle by lumping everything into a generic error. The detailed error is only logged if debug is active.
2016-07-02tests: call umount instead of fusermount on OSXJakob Unterwurzacher
Reported at https://github.com/rfjakob/gocryptfs/issues/15
2016-07-02fusefrontend: add fallocate supportJakob Unterwurzacher
Mode=0 (default) and mode=1 (keep size) are supported. The patch includes test cases and the whole thing passed xfstests. Fixes https://github.com/rfjakob/gocryptfs/issues/1 .
2016-07-02fusefronted: move Truncate() and Allocate() to their own fileJakob Unterwurzacher
These are large complicated implementations that will share some code.
2016-07-02contentenc: rename PlaintextRange and CiphertextRangeJakob Unterwurzacher
The name could be misunderstood and actually caused a bug: doWrite used to always preallocate 4128 instead of the actual data length.
2016-07-01fusefrontend: coalesce grows in Truncate()Jakob Unterwurzacher
We were growing the file block-by-block which was pretty inefficient. We now coalesce all the grows into a single Ftruncate. Also simplifies the code! Simplistic benchmark: Before: $ time truncate -s 1000M foo real 0m0.568s After: $ time truncate -s 1000M foo real 0m0.205s
2016-07-01fusefrontend: better comments for TruncateJakob Unterwurzacher
2016-06-30fusefrontend: handle dir-overwrites-dir on XFSJakob Unterwurzacher
XFS returns a different error code if you try to overwrite a non-empty directory with a directory: XFS: mv: cannot move ‘foo’ to ‘bar/foo’: File exists ext4: mv: cannot move 'foo' to 'bar/foo': Directory not empty So have EEXIST trigger the Rmdir logic as well. Fixes issue #20 Link: https://github.com/rfjakob/gocryptfs/issues/20
2016-06-30test: split up integration_testsJakob Unterwurzacher
...into "matrix" and "normal". Also: * Make running multiple packages in parallel safe, see http://stackoverflow.com/questions/23715302/go-how-to-run-tests-for-multiple-packages * Don't depent on test_helper.TmpDir and friends to have a terminating slash
2016-06-29build.bash: support colon-separated GOPATH variableJakob Unterwurzacher
Fixes issue #31 https://github.com/rfjakob/gocryptfs/issues/31
2016-06-27tests: run example_filesystems with "-openssl=false" AND "=true"Jakob Unterwurzacher
Run the tests twice, once with openssl and once with Go crypto.
2016-06-27tests: give "-plaintextnames" its own test packageJakob Unterwurzacher
...and add tests for checking that gocryptfs.diriv does not get created. The main "integration_tests" package has become quite big and convoluted over time. This small separate package should make writing tests for "-plaintextnames" easier. As seen in "fusefrontend: fix PlaintextNames versions of Mkdir, Rmdir", we need more of them.
2016-06-27tlog: clean up messages from the go-fuse libraryJakob Unterwurzacher
Drop the date and add the "go-fuse: " prefix so you can see where the message is coming from. Before: Jun 27 09:03:15 brikett gocryptfs[4150]: 2016/06/27 09:03:15 Unimplemented opcode INTERRUPT After: Jun 27 09:10:58 brikett gocryptfs[4961]: go-fuse: Unimplemented opcode INTERRUPT
2016-06-27fusefrontend: fix PlaintextNames versions of Mkdir, RmdirJakob Unterwurzacher
The "!fs.args.DirIV" special case was removed by b17f0465c7 but that, by accident, also removed the handling for PlaintextNames. Re-add it as an explicit PlaintextNames special case. Also adds support for removing directories that miss their gocryptfs.diriv file for some reason.
2016-06-26tlog: switch default logger to syslogJakob Unterwurzacher
...unless "-nosyslog" is passed. All gocryptfs messages already go to syslog, but the messages that the go-fuse lib emits were still printed to stdout. Fixes issue #13 ( https://github.com/rfjakob/gocryptfs/issues/13 )
2016-06-26MANPAGE: note that "-plaintextnames" disables symlink encryptionJakob Unterwurzacher
This is no change in behavoir, just a clarification in the man page.
2016-06-26main: add "-o" option to enable "suid" and "dev"Jakob Unterwurzacher
Device files and suid binaries are often not needed when running gocryptfs as root. As they are potentially dangerous, let the user enable them explicitely via the new "-o" option instead of always enabling them when running as root.
2016-06-26main, fusefrontend: enable suid functionalityJakob Unterwurzacher
FUSE filesystems are mounted with "nosuid" by default. If we run as root, we can use device files by passing the opposite mount option, "suid". Also we have to use syscall.Chmod instead of os.Chmod because the portability translation layer "syscallMode" messes up the sgid and suid bits. Fixes 70% of the failures in xfstests generic/193. The remaining are related to truncate, but we err on the safe side: $ diff -u tests/generic/193.out /home/jakob/src/fuse-xfstests/results//generic/193.out.bad [...] check that suid/sgid bits are cleared after successful truncate... with no exec perm before: -rwSr-Sr-- -after: -rw-r-Sr-- +after: -rw-r--r--
2016-06-26main: enable device files when running as rootJakob Unterwurzacher
FUSE filesystems are mounted with "nodev" by default. If we run as root, we can use device files by passing the opposite mount option, "dev". Fixes xfstests generic/184.
2016-06-26fusefronted: preserve owner if running as rootJakob Unterwurzacher
If allow_other is set and we run as root, try to give newly created files to the right user.
2016-06-26fuserfrontend: support truncate(2) by wrapping ftruncate(2)Jakob Unterwurzacher
Support truncate(2) by opening the file and calling ftruncate(2) While the glibc "truncate" wrapper seems to always use ftruncate, fsstress from xfstests uses this a lot by calling "truncate64" directly.
2016-06-26tests: add missing file "example_test_helpers.go"Jakob Unterwurzacher
This file was forgotten in commit "tests: make tests for unsupported FSs more compact".
2016-06-23README: fix copy-paste errorJakob Unterwurzacher
2016-06-23README: Note that v0.6 filesystems can no longer be mountedJakob Unterwurzacher
2016-06-23Drop deprecated "-gcmiv128" optionJakob Unterwurzacher
The GCMIV128 feature flag is already mandatory, dropping the command line option is the final step. Completes https://github.com/rfjakob/gocryptfs/issues/29 .
2016-06-23tests: make tests for unsupported FSs more compactJakob Unterwurzacher
There is no need to test that deprecated command-line options produce an error. I trust the flags package. Also split the example_filesystem helper functions into a separate file.
2016-06-23Drop deprecated "-emenames" optionJakob Unterwurzacher
The EMENames feature flag is already mandatory, dropping the command line option is the final step.
2016-06-23nametransform: drop unused noiv functionsJakob Unterwurzacher
As DirIV is now mandatory there is no user for the noiv functions.
2016-06-23Drop deprecated "-diriv" optionJakob Unterwurzacher
The DirIV feature flag is already mandatory, dropping the command line option is the final step.
2016-06-23tests: add v0.7-plaintextnames example filesystemJakob Unterwurzacher
The v0.6-plaintextnames example FS lacks the GCMIV128 feature flag, is no longer mountable and can no longer be used for testing. Add a new "-plaintextnames" filesystem created by gocryptfs v0.7. There have been no format changes to "-plaintextnames" since then.
2016-06-19build.bash: check if the go-fuse tree is dirtyJakob Unterwurzacher
If it is, append "-dirty" to the hash.
2016-06-19tests: expect mount failure for v0.6 and older example filesystemsJakob Unterwurzacher
2016-06-19Refuse mounting of v0.6 and older filesystemsJakob Unterwurzacher
2016-06-19main: drop "on-disk format" from -version output, add Go versionJakob Unterwurzacher
As v0.4 introduced ext4-style feature flags, the on-disk format version is unlinkely to change. Drop it from the version output to reduce clutter. Use "gocryptfs -version -debug" to see it. Add the Go version string because only Go 1.6 and newer have an optimized AES-GCM implementation. This will help users to understand the performance of their build.
2016-06-19README: abbreviate ticket #29 linkJakob Unterwurzacher
2016-06-19README: release v0.12v0.12Jakob Unterwurzacher
2016-06-19travis: skip tests on go1.3.3Jakob Unterwurzacher
Go 1.3.3 does not support testing.M, so skip the tests there.
2016-06-16README: mention stdin passwordsJakob Unterwurzacher