| Age | Commit message (Collapse) | Author | 
|---|
|  | Same thing like contrib/getdents, but written in C. | 
|  | Small tool to try to debug unix.Getdents problems on CIFS mounts
https://github.com/rfjakob/gocryptfs/issues/483 | 
|  | And also, stop using the wrong directory for sshfs git init.
sshfs-benchmark.bash:    sshfs  gocryptfs-on-sshfs
git init                  4.35                7.82
rsync                     7.72               11.66
rm -R                     2.71               11.04
mkdir                     1.33                4.15
rmdir                     0.47                3.97
touch                     2.32                2.85
rm                        0.45                0.45 | 
|  | When filename encryption is on, we do know when we
overwrite a directory, and can clear only in this case.
sshfs-benchmark.bash:    sshfs  gocryptfs-on-sshfs
git init                  1.74                7.80
rsync                     6.19               11.63 | 
|  | Mkdir can not cause existing entries in the cache to go
stale. So don't clear it. Benchmark results:
sshfs-benchmark.bash:    sshfs  gocryptfs-on-sshfs
git init                  1.65                8.74
rsync                     6.09               17.54 | 
|  | Let's get some reproducible numbers for
https://github.com/rfjakob/gocryptfs/issues/481
and
https://github.com/rfjakob/gocryptfs/issues/410
Example run:
$ ./sshfs-benchmark.bash nuetzlich.net
working directory: /tmp/sshfs-benchmark.bash.vu4
sshfs mounted: nuetzlich.net:/tmp -> sshfs.mnt
gocryptfs mounted: sshfs.mnt/sshfs-benchmark.bash.KM9/gocryptfs.crypt -> gocryptfs.mnt
                         sshfs  gocryptfs-on-sshfs
git init                  1.68               11.23
rsync                     6.07               20.35 | 
|  | Fixes: https://github.com/rfjakob/gocryptfs/issues/483
Related: https://github.com/golang/go/issues/38836 | 
|  | Looking at the dircache debug output, we see
that a "git status" workload has a very bad
cache hit rate because the entries expire or
get evicted before they can be reused.
Increase both cache size and lifetime for
a 4x speedup:
Before: 75s
After:  17s
https://github.com/rfjakob/gocryptfs/issues/410 | 
|  | Before:
Lookup "errno.html/1/2/3/4/5": miss
Store: "errno.html/1/2/3/4/5" fd=26 iv=21be6e083d60dcabfe7368264d5082b7
Lookup "errno.html": hit 25 6d68a16d217978915036a3bd55428ae7
Lookup "errno.html/1": hit 25 932a464c299b3430c5e55c924f98ac4d
Lookup "errno.html/1/2": hit 25 7d53348b1692d537f017bf86b3cf5feb
Lookup "errno.html/1/2/3": hit 25 2aef1c9d1ab2b55b163215053fefe703
Lookup "errno.html/1/2/3/4": hit 25 cb802be53721c46a46247c5e4e0f4ce6
Lookup "errno.html/1/2/3/4": hit 25 cb802be53721c46a46247c5e4e0f4ce6
Lookup "errno.html": hit 25 6d68a16d217978915036a3bd55428ae7
After:
Lookup "earlyoom/.git/refs"                     hit fd=10 dup=17 iv=6ae2cecd269a25e8d946aff6afe9b8b8
Lookup "earlyoom/.git/refs/remotes"             hit fd=19 dup=17 iv=f04c2d2a5bcc33ebdeaca664859c980d
Lookup "earlyoom/.git/refs/remotes/origin"      miss
Store  "earlyoom/.git/refs/remotes/origin"      fd=17 iv=834a64a1697c9f5705455ba6dbed22b5
Lookup "earlyoom"                               hit fd=7 dup=25 iv=2303a892d6e2357c483574a8070b7679
Lookup "earlyoom/.git"                          hit fd=11 dup=25 iv=d43ca4aff23720c57789c9f62f0aee00
Lookup "earlyoom/.git"                          hit fd=11 dup=25 iv=d43ca4aff23720c57789c9f62f0aee00
Lookup "earlyoom/.git/refs"                     hit fd=10 dup=25 iv=6ae2cecd269a25e8d946aff6afe9b8b8
Lookup "earlyoom/.git/refs/heads"               hit fd=13 dup=25 iv=f9245e7c066b9adc768a1a666da9fbc8 | 
|  |  | 
|  | Each file will be read and then concatenated
for the effictive password. This can be used as a
kind of multi-factor authenticiton.
Fixes https://github.com/rfjakob/gocryptfs/issues/288 | 
|  | The go-fuse v1 dependency is spurious. Will be fixed by
https://github.com/hanwen/go-fuse/pull/360 | 
|  | We need
https://github.com/hanwen/go-fuse/commit/fd7328faf9fdf75709f7ba7df7072aaf4eeb18b3
to fix a crash reported in https://github.com/rfjakob/gocryptfs/issues/430 :
  2019/10/30 17:14:16 Unknown opcode 2016
  panic: runtime error: invalid memory address or nil pointer dereference
  [signal SIGSEGV: segmentation violation code=0x1 addr=0x20 pc=0x508d38]
This patch is only in the v2.x.x branch. Upgrade to v2, as the
old API is also supported there.
Running
  git grep hanwen/go-fuse | grep -v hanwen/go-fuse/v2
to check for forgotten references comes back clean. | 
|  | https://github.com/client9/misspell | 
|  | Closes https://github.com/rfjakob/gocryptfs/issues/416 | 
|  |  | 
|  | The -0 flags works like xargs -0. | 
|  | Should show up on https://pkg.go.dev/github.com/rfjakob/gocryptfs?tab=doc
which currently reads "No documentation available for this package!" | 
|  | Implementation seems to work ok, but is missing tests and
documentation for now.
I will only delete ctlsock-encrypt.bash when both are
done.
https://github.com/rfjakob/gocryptfs/issues/416 | 
|  |  | 
|  |  | 
|  | lsof may get stuck when gocryptfs itself is stuck. | 
|  | The former interal ctlsock server package is renamed
to ctlsocksrv. | 
|  | Tests that `gocryptfs -passwd -masterkey=stdin` works.
This was fixed by ff04b1d83ab1201.
Fixes https://github.com/rfjakob/gocryptfs/issues/461 | 
|  | This was handled both in getMasterKey(). Split it apart. | 
|  | Make it clear that function does NOT parse the "-masterkey"
command line argument, it just unhexes the payload. | 
|  | We did not use t.Name() as it was not available
before Go 1.8. Now the oldest Go version we support is
Go 1.11, so we can use it. | 
|  | The command line option is now called `-badname`,
so adjust the test name to match. | 
|  |  | 
|  | Bisecting shows that the performance drop is caused by
this commit:
commit ca9e912a28b901387e1dbb85f6c531119f2d5ef2 (refs/bisect/bad)
Author: Jakob Unterwurzacher <jakobunt@gmail.com>
Date:   Sat Feb 29 19:58:08 2020 +0100
    fusefrontend: drop xattr user namespace restriction | 
|  | Updated using
    go get -t -u ./... | 
|  | We redirected the wrong ldd fd to /dev/null. Fix it. | 
|  | Wrong bit operator was used. | 
|  | Also add a test for this.
Thanks @slackner for the comment. | 
|  | This was committed by accident. | 
|  | Gets rid of static inode number value limitations.
Fixes https://github.com/rfjakob/gocryptfs/issues/457 | 
|  | Verify that virtual files get assigned inode numbers
we expect. | 
|  |  | 
|  | Adding flags allows to use inomap in reverse mode,
replacing the clunky inoBaseDirIV/inoBaseNameFile
logic that causes problems with high underlying
inode numbers ( https://github.com/rfjakob/gocryptfs/issues/457 )
Microbenchmarks (values below) show that the "SingleDev"
case is now much slower due to an extra map lookup,
but this has no visible effects in ./test.bash results,
so there was no time spent optimizing the case further.
$ go test -bench=.
goos: linux
goarch: amd64
pkg: github.com/rfjakob/gocryptfs/internal/inomap
BenchmarkTranslateSingleDev-4   	18757510	        61.5 ns/op
BenchmarkTranslateManyDevs-4    	18061515	        64.5 ns/op
PASS
ok  	github.com/rfjakob/gocryptfs/internal/inomap	2.467s | 
|  | $ go test -bench=.
goos: linux
goarch: amd64
pkg: github.com/rfjakob/gocryptfs/internal/inomap
BenchmarkTranslateSingleDev-4   	202479382	         5.88 ns/op
BenchmarkTranslateManyDevs-4    	16095795	        71.9 ns/op
PASS
ok  	github.com/rfjakob/gocryptfs/internal/inomap	3.039s | 
|  | The case of a git repo without any tags used to fail
with:
  fatal: No names found, cannot describe anything.
Now we continue, using "[no_tags_found]" as the
version string. | 
|  | Causes warnings:
  $ ./build-without-openssl.bash
  # github.com/rfjakob/gocryptfs
  loadinternal: cannot find runtime/cgo
  # github.com/rfjakob/gocryptfs/gocryptfs-xray
  loadinternal: cannot find runtime/cgo
  # github.com/rfjakob/gocryptfs/contrib/statfs
  loadinternal: cannot find runtime/cgo
  gocryptfs v1.7.1-48-gf6b1c68 without_openssl; go-fuse v1.0.1-0.20190319092520-161a16484456; 2020-04-18 go1.13.6 linux/amd64
https://github.com/golang/go/issues/30986 | 
|  | The comment still mentioned CBC, which has been removed
a long time ago.
The test definition can be rewritten using slice literals,
saving sume stuttering. | 
|  | We used to prefer openssl in this situation, which
used to make sense, but now Go gained an optimized
assembly implementation for aes-gcm on arm64 with
aes instructions:
  root@q1:~/go/src/github.com/rfjakob/gocryptfs# ./gocryptfs -speed
  gocryptfs v1.7.1-46-g73436d9; go-fuse v1.0.1-0.20190319092520-161a16484456; 2020-04-13 go1.14.2 linux/arm64
  AES-GCM-256-OpenSSL      212.30 MB/s    (selected in auto mode)
  AES-GCM-256-Go           452.30 MB/s
  AES-SIV-512-Go           100.25 MB/s
  XChaCha20-Poly1305-Go    137.35 MB/s
https://github.com/rfjakob/gocryptfs/issues/452 | 
|  |  | 
|  | GOFLAGS exists since Go 1.11: https://golang.org/doc/go1.11
https://github.com/rfjakob/gocryptfs/pull/460 | 
|  |  | 
|  | Following https://blog.golang.org/migrating-to-go-modules | 
|  | And run shellcheck in test.bash. | 
|  |  |