Experience sharing of hashcat using GPU acceleration to improve d

After some attempts, I recorded some of my experiences as follows.

First of all, we know that GUP needs to meet two conditions to accelerate in hashcat: 1. An accelerable graphics card (GPU);2. Cuda is installed.

My computer Win10 system, n-card 3060

1. Summary of failure

At first, I wanted to complete the acceleration of hashcat in kali linux, but the graphics card in the virtual machine was algorithmically generated and could not be used for acceleration. When I saw that the n-card could directly connect to the virtual machine, I searched for it with hope, but I couldn't find a graphics card direct connect tutorial suitable for kali linux. Moreover, there was only a linux host computer on the Internet that had direct connect to the windows virtual machine. It was the opposite of me that I wanted to connect to linux from the windows system, so I had no suitable way to complete this step. Even if it is completed, I have to switch graphics cards back and forth. I mainly use the win10 system. If I have the ability, I can write a program to switch graphics cards, but it is not suitable for me at the moment, so it is over. I hope this step can be improved in the future.

2. Compromise of the host computer. The host

computer chose to complete the acceleration of hashcat on the host computer. The default graphics card of the host computer is 3060, so as long as cuda and hashcat are installed, GPU acceleration can be used normally. Cuda can be downloaded and installed directly, and you can install it in the next step (link). You should pay attention to hashcat. Hashcat recommends directly downloading the official website binary file (link) and you can use it after decompression. Note that hashcat is a command-line application. Double-clicking to open hashcat.exe will flash back. You should use the command line to open it in the system terminal cmd.

3. Avoidable problems

(problem with installing hashcat on msys2: CUDA SDK toolkit is not installed or incorrectly installed)

If I install hashcat using the installation instructions on github, I have some problems (CUDA SDK toolkit is not installed or incorrectly installed).

This problem can be seen that cuda couldn't be detected, but I have already installed cuda. Because I didn't find cuda, the system default the api to opencl. I personally don't know what the problem is. I visited some English websites and tried some operations, but it didn't work, so I gave up. But at the same time, because I downloaded binary hashcat, I tried it, but it worked. I don't know why.

The same problem will appear in the hashcat installed with msys2 above whether it is in the msys2 shell or the system shell

. This problem does not occur when using the system shell to open the directly downloaded binary hashcat

The api automatically detected cuda and reported no error.

4. Summary and understanding of hashcat

Through hashcat -b, you can know the speed at which hashcat hits various hashes. Some are fast, some are not fast. For different types of hashes, the difference in hashes per second is very large. By comparing with or without GPU acceleration, hashcat is 60 times faster with GPU acceleration than without acceleration.

The article is reproduced from the Internet! Non-document password recovery master website original