Commit Graph

21 Commits

Author SHA1 Message Date
Jakob Borg
e82ed6e3d3
style: gofumpt all the things (#9829)
Literally `gofumpt -w .` from the top level dir. Guaranteed to be minor
style changes only and nothing else.

@imsodin per request?
2024-11-19 11:32:56 +01:00
Gusted
356c5055ad
lib/sha256: Remove it (#9643)
### Purpose

Remove the `lib/sha256` package, because it's no longer necessary. Go's
standard library now has the same performance and is on par with
`sha256-simd` since [Since Go
1.21](1a64574f42).
Therefore using `sha256-simd` has no benefits anymore.

ARM already has optimized sha256 assembly code since
7b8a7f8272,
`sha256-simd` published their results before that optimized assembly was
implemented,
f941fedda8.
The assembly looks very similar and the benchmarks in the Go commit
match that of `sha256-simd`.

This patch removes all of the related code of `lib/sha256` and makes
`crypto/sha256` the 'default'.

Benchmark of `sha256-simd` and `crypto/sha256`:
<details>

```
cpu: AMD Ryzen 5 3600X 6-Core Processor
                │  simd.txt   │               go.txt                │
                │   sec/op    │    sec/op     vs base               │
Hash/8Bytes-12    63.25n ± 1%    73.38n ± 1%  +16.02% (p=0.002 n=6)
Hash/64Bytes-12   98.73n ± 1%   105.30n ± 1%   +6.65% (p=0.002 n=6)
Hash/1K-12        567.2n ± 1%    572.8n ± 1%   +0.99% (p=0.002 n=6)
Hash/8K-12        4.062µ ± 1%    4.062µ ± 1%        ~ (p=0.396 n=6)
Hash/1M-12        512.1µ ± 0%    510.6µ ± 1%        ~ (p=0.485 n=6)
Hash/5M-12        2.556m ± 1%    2.564m ± 0%        ~ (p=0.093 n=6)
Hash/10M-12       5.112m ± 0%    5.127m ± 0%        ~ (p=0.093 n=6)
geomean           13.82µ         14.27µ        +3.28%

                │   simd.txt   │               go.txt                │
                │     B/s      │     B/s       vs base               │
Hash/8Bytes-12    120.6Mi ± 1%   104.0Mi ± 1%  -13.81% (p=0.002 n=6)
Hash/64Bytes-12   618.2Mi ± 1%   579.8Mi ± 1%   -6.22% (p=0.002 n=6)
Hash/1K-12        1.682Gi ± 1%   1.665Gi ± 1%   -0.98% (p=0.002 n=6)
Hash/8K-12        1.878Gi ± 1%   1.878Gi ± 1%        ~ (p=0.310 n=6)
Hash/1M-12        1.907Gi ± 0%   1.913Gi ± 1%        ~ (p=0.485 n=6)
Hash/5M-12        1.911Gi ± 1%   1.904Gi ± 0%        ~ (p=0.093 n=6)
Hash/10M-12       1.910Gi ± 0%   1.905Gi ± 0%        ~ (p=0.093 n=6)
geomean           1.066Gi        1.032Gi        -3.18%
```

</details>


### Testing

Compiled and tested on Linux.

### Documentation

https://github.com/syncthing/docs/pull/874
2024-08-10 12:58:20 +01:00
deepsource-autofix[bot]
0930bccf88
cmd/ursrv, lib/scanner: Remove unnecessary slicing of slices (#7491)
Co-authored-by: deepsource-autofix[bot] <62050782+deepsource-autofix[bot]@users.noreply.github.com>
2021-03-17 21:04:36 +01:00
Simon Frei
90e248615f
lib/scanner: Test weak hash consistency (ref #5556) (#6794)
Relevant much earlier changes:
  9b1c592fb7
  bd1c29ee32

Make sure vanilla and rolling adler are consistent. And that they match
with scanner.Validate.
2020-06-25 14:47:35 +02:00
greatroar
9c0825c0d9
lib/scanner: Simplify, optimize and document Validate (#6674) (#6688) 2020-05-27 22:23:12 +02:00
Jakob Borg
c2ddc83509 all: Revert the underscore sillyness 2019-02-02 12:16:27 +01:00
Jakob Borg
0b2cabbc31
all: Even more boring linter fixes (#5501) 2019-02-02 11:45:17 +01:00
Jakob Borg
c0a26c918a
lib/scanner, vendor: Update github.com/chmduquesne/rollinghash (fixes #5334) (#5335)
Updates the package and fixes a test that depended on the old behavior
of Write() being equivalent to Reset()+Write() which is no longer the
case. The scanner already did resets after each block write, so this is
fine.
2018-11-22 08:50:06 +01:00
Jakob Borg
19c7cd99f5 all: Implement variable sized blocks (fixes #4807) 2018-04-16 19:08:50 +01:00
Lars K.W. Gohlke
89a021609b lib/scanner: Refactoring
GitHub-Pull-Request: https://github.com/syncthing/syncthing/pull/4642
LGTM: imsodin, AudriusButkevicius
2018-01-14 14:30:11 +00:00
Jakob Borg
d6fbfc3545 lib/fs, lib/model, lib/scanner: Make scans cancellable (fixes #3965)
The folder already knew how to stop properly, but the fs.Walk() didn't
and can potentially take a very long time. This adds context support to
Walk and the underlying scanning stuff, and passes in an appropriate
context from above. The stop channel in model.folder is replaced with a
context for this purpose.

To test I added an infiniteFS that represents a large amount of data
(not actually infinite, but close) and verify that walking it is
properly stopped. For that to be implemented smoothly I moved out the
Walk function to it's own type, as typically the implementer of a new
filesystem type might not need or want to reimplement Walk.

It's somewhat tricky to test that this actually works properly on the
actual sendReceiveFolder and so on, as those are started from inside the
model and the filesystem isn't easily pluggable etc. Instead I've tested
that part manually by adding a huge folder and verifying that pause,
resume and reconfig do the right things by looking at debug output.

GitHub-Pull-Request: https://github.com/syncthing/syncthing/pull/4117
2017-04-26 00:15:23 +00:00
Jakob Borg
f7fc0c1d3e all: Update license url to https (ref #3976) 2017-02-09 08:04:16 +01:00
Audrius Butkevicius
dd78177ae0 scanner: Allow disabling weak hash in scanning (fixes #3891)
GitHub-Pull-Request: https://github.com/syncthing/syncthing/pull/3905
2017-01-23 13:50:32 +00:00
Jakob Borg
bd1c29ee32 lib/scanner, vendor: Fix previous commit
Can't do what I did, as the rolling function is not the same as the
non-rolling one. Instead this uses an improved version of the rolling
adler32 to accomplish the same thing. (PR filed on upstream, so should
be able to use that directly in the future.)
2017-01-18 11:57:01 +01:00
Jakob Borg
9b1c592fb7 lib/scanner: Speed up weak hash
The rolling version of adler32 is just a wrapper around the standard
hash/adler32 when used in a non-rolling fashion, but it's inefficient as
it allocates a new hash instance for every Write(). This uses the
default version instead in the block hasher, and adds a test to verify
the result is the same as they were before. It reduces allocations by
88% and increases speed about 5%.

	benchmark               old ns/op     new ns/op     delta
	BenchmarkHashFile-8     64434698      61303647      -4.86%

	benchmark               old MB/s     new MB/s     speedup
	BenchmarkHashFile-8     276.65       290.78       1.05x

	benchmark               old allocs     new allocs     delta
	BenchmarkHashFile-8     1238           150            -87.88%

	benchmark               old bytes     new bytes     delta
	BenchmarkHashFile-8     17877363      49292         -99.72%
2017-01-18 10:33:17 +01:00
Audrius Butkevicius
29d010ec0e lib/model, lib/weakhash: Hash using adler32, add heuristic in puller
Adler32 is much faster, and the heuristic avoid the obvious cases where it
will not help.

GitHub-Pull-Request: https://github.com/syncthing/syncthing/pull/3872
2017-01-04 21:04:13 +00:00
Jakob Borg
7aaa1dd8a3 lib/scanner: Recheck file size and modification time after hashing (ref #3440)
To catch the case where the file changed. Also make sure we never let a
size-vs-blocklist mismatch slip through.

GitHub-Pull-Request: https://github.com/syncthing/syncthing/pull/3443
2016-07-26 08:51:39 +00:00
Jakob Borg
f343c8ba36 lib/model, lib/scanner: Silence vet warnings
GitHub-Pull-Request: https://github.com/syncthing/syncthing/pull/3333
2016-06-20 21:00:39 +00:00
Jakob Borg
4581c57478 Fix import paths 2015-09-22 19:38:46 +02:00
AudriusButkevicius
94c52e3a77 Add scan percentages (fixes #1030) 2015-08-27 19:20:43 +01:00
Jakob Borg
7705a6c1f1 mv internal lib 2015-08-09 09:35:26 +02:00