Skip to content

Instantly share code, notes, and snippets.

@matrix
matrix / AppleM1-Metal-v6.2.5-209-g431c46f43-FullBenchmark-arm64-Vec2
Created February 10, 2022 23:18
Hashcat v6.2.5-209-g431c46f43+, Full Benchmark on Apple M1 (arm64 binary) using Metal runtime, Vector size 2
% ./hashcat -I
hashcat (v6.2.5-209-g431c46f43+) starting in backend information mode
System Info:
============
OS.Name......: Darwin
OS.Release...: 21.3.0
HW.Model.....: Macmini9,1
HW.Platform..: arm64
@matrix
matrix / AppleM1-Metal-v6.2.5_dev-FullBenchmark-arm64
Created February 3, 2022 07:58
Hashcat v6.2.5-85-g0abdcb1ae+, Full Benchmark on Apple M1 (arm64 binary) using Metal runtime
matrix@matrixs-Mac-mini hashcat_metal % ./hashcat -I
hashcat (v6.2.5-85-g0abdcb1ae+) starting in backend information mode
Metal Info:
===========
Metal.Version.: 258.18
Backend Device ID #1 (Alias: #2)
Type...........: GPU
@matrix
matrix / GTX_1080_Ti-CUDA-v6.2.5_dev-FullBenchmark
Created January 24, 2022 15:57
Hashcat v6.2.5-178-gf7506c35f+, Full Benchmark on GTX 1080 Ti using CUDA runtime / Linux
Same configuration used for https://gist.github.com/matrix/54a3830db3f686ee9596e21c8f819384
PR's #3140, #3141 and #3142 was included, but on Linux with NVIDIA GeForce GTX 1080 Ti
$ ./hashcat -b --benchmark-all
hashcat (v6.2.5-178-gf7506c35f+) starting in benchmark mode
Benchmarking uses hand-optimized kernel code by default.
You can use it in your cracking session by setting the -O option.
Note: Using optimized kernel code limits the maximum supported password length.
To disable the optimized kernel code in benchmark mode, use the -w option.
@matrix
matrix / AppleM1-OpenCL-v6.2.5_dev-FullBenchmark-arm64
Created January 23, 2022 20:00
Hashcat v6.2.5-158-g01a28f80f+, Full Benchmark on Apple M1 (arm64 binary) using OpenCL runtime
PR's #3140, #3141 and #3142 was included.
matrix@matrixs-Mac-mini hashcat % ./hashcat -b --benchmark-all
hashcat (v6.2.5-158-g01a28f80f+) starting in benchmark mode
Benchmarking uses hand-optimized kernel code by default.
You can use it in your cracking session by setting the -O option.
Note: Using optimized kernel code limits the maximum supported password length.
To disable the optimized kernel code in benchmark mode, use the -w option.
@matrix
matrix / AppleM1-OpenCL-v6.2.5_dev-FullBenchmark-x86_64
Last active January 22, 2022 11:54
Hashcat v6.2.5-139-ge2eacb4dd+, Full Benchmark on Apple M1 (x86_64 binary / rosetta) using OpenCL runtime
To make an x86_64 binary on Apple M1 the following patch was applied (as well as installed rosetta)
[[
diff --git a/src/Makefile b/src/Makefile
index fc0afc4fb..3a4d61afd 100644
--- a/src/Makefile
+++ b/src/Makefile
@@ -324,6 +324,7 @@ ifeq ($(UNAME),Darwin)
export MACOSX_DEPLOYMENT_TARGET=10.15
CFLAGS_NATIVE := $(CFLAGS)
@matrix
matrix / AppleM1_Metal_v6.2.5-dev_fastBenchmark
Created January 8, 2022 23:56
Hashcat v6.2.5-dev, Fast Benchmark on Apple M1 with Metal runtime
bash-3.2$ ./hashcat -b
hashcat (v6.2.5-85-g0abdcb1ae+) starting in benchmark mode
Benchmarking uses hand-optimized kernel code by default.
You can use it in your cracking session by setting the -O option.
Note: Using optimized kernel code limits the maximum supported password length.
To disable the optimized kernel code in benchmark mode, use the -w option.
* Device #2: Apple's OpenCL drivers (GPU) are known to be unreliable.
You have been warned.