Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Reconfiguring for M2 Max #17

Open
MyLittleRunaway opened this issue Oct 29, 2023 · 8 comments
Open

Reconfiguring for M2 Max #17

MyLittleRunaway opened this issue Oct 29, 2023 · 8 comments

Comments

@MyLittleRunaway
Copy link

Hi,

Super quick one (and thanks for the god-tier code)

My vanity wallet randomly got drained last month and I'm trying to figure out if this was used to do it.
I noticed you have an M1 and 16GB of RAM - but I'd like to use my M2 with 96GB RAM and a better GPU to try and make it faster. I looked through the code and there didn't seem to be anything to set as a variable for the max memory.

I was wondering if this would be as simple as changing a var somewhere I didn't look hard enough for, or it would require a complete rewrite.
I've attached a picture of the current setup for you. Once again - cheers. I missed the news about this, but hopefully it'll shed some light on if it was used to drain me (but hopefully quicker!

Screenshot 2023-10-29 at 12 23 32 Screenshot 2023-10-29 at 12 39 53
@ytrezq
Copy link

ytrezq commented Nov 4, 2023

@MyLittleRunaway a similar problem as me. But you need to do it the deterministic instead of the probabilistic way in order to have a definite no answer. See : https://ethereum.stackexchange.com/q/156124 and #18.

@ytrezq
Copy link

ytrezq commented Nov 6, 2023

Hi,

Super quick one (and thanks for the god-tier code)

My vanity wallet randomly got drained last month and I'm trying to figure out if this was used to do it. I noticed you have an M1 and 16GB of RAM - but I'd like to use my M2 with 96GB RAM and a better GPU to try and make it faster. I looked through the code and there didn't seem to be anything to set as a variable for the max memory.

I was wondering if this would be as simple as changing a var somewhere I didn't look hard enough for, or it would require a complete rewrite. I've attached a picture of the current setup for you. Once again - cheers. I missed the news about this, but hopefully it'll shed some light on if it was used to drain me (but hopefully quicker!
Screenshot 2023-10-29 at 12 23 32 Screenshot 2023-10-29 at 12 39 53

Hi, first of all… Are you managing to get the example command line in the Readme working ? I’m meaning, to find something ?
Then the maximum you’ll need is 32Gb because of the number of keys. This is linked to the .extended variable in the code.

@MyLittleRunaway
Copy link
Author

Hey thanks for your reply, @ytrezq

Hi, first of all… Are you managing to get the example command line in the Readme working ? I’m meaning, to find something ?

I am, mate, yes.

Then the maximum you’ll need is 32Gb because of the number of keys. This is linked to the .extended variable in the code.

Any idea how I can set this?

@ytrezq
Copy link

ytrezq commented Nov 7, 2023

Hey thanks for your reply, @ytrezq

Hi, first of all… Are you managing to get the example command line in the Readme working ? I’m meaning, to find something ?

I am, mate, yes.

Then the maximum you’ll need is 32Gb because of the number of keys. This is linked to the .extended variable in the code.

Any idea how I can set this?

You increase and decrease the values of mode.extended when true by 2 accordingly in Dispatcher.cpp.
Does the program still produce results on mac if you replace CL_DEVICE_TYPE_GPU by CL_DEVICE_TYPE_ALL in

std::vector<cl_device_id> getAllDevices(cl_device_type deviceType = CL_DEVICE_TYPE_GPU)
?

@stevealexrs
Copy link

stevealexrs commented Nov 8, 2023

Any idea how I can set this?

try adding --extended command line argument?

It seems like this tool only work on Mac.

@ytrezq
Copy link

ytrezq commented Nov 8, 2023

try adding --extended command line argument?

It seems like this tool only work on Mac.

The extended argument only allow to take 16Gib. In order to use 32/60Gib, it s required to increase and decrease the values of mode.extended when true by 2 accordingly in Dispatcher.cpp.

The reason it only works on osx might be an initialization issue. Hence I suspect it doesn t works on Intel Mac nor does it works with CL_DEVICE_TYPE_ALL. Someone that owns mac has to test it.

@stevealexrs
Copy link

I think the extended parameter is only used to change the size of hash table and number of batch. I can't test it right now but you can try changing (mode.extended ? 2 : 1) to (mode.extended ? 8 : 1) , m_mode.extended ? 4 : 8 to m_mode.extended ? 1 : 8, m_mode.extended ? "16Gb" : "8Gb" to m_mode.extended ? "64Gb" : "8Gb" and add --extended command line arg, so it uses 64 GB hash table and run one batch only. (why the author use Gb, shouldnt it be GB?)

@ytrezq
Copy link

ytrezq commented Nov 8, 2023

I think the extended parameter is only used to change the size of hash table and number of batch. I can't test it right now but you can try changing (mode.extended ? 2 : 1) to (mode.extended ? 8 : 1) , m_mode.extended ? 4 : 8 to m_mode.extended ? 1 : 8, m_mode.extended ? "16Gb" : "8Gb" to m_mode.extended ? "64Gb" : "8Gb" and add --extended command line arg, so it uses 64 GB hash table and run one batch only. (why the author use Gb, shouldnt it be GB?)

No. You change the values of when mode.extended is true. This gives you 2 batches of 32Gb used (you can t go below 1 for 1 of the values which is why you can increase and decrease by 4 but only 2).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants