I have both a 1080 Ti and RX480 running in my system.
Due to this "metric" determining speed:

...the program selects my RX480 over the 1080 Ti.
A command line option to select the compute device number would probably work.

Can anyone help me out with this part? Im trying to install a custom firmware on my 3ds. Im at the Seedminer part where i need to type these two commands for the movable files to pop up

py -3 seedminer_launcher3.py update-db
py -3 seedminer_launcher3.py gpu

But it keeps showing "out of resource" which i am guessing my gpu isn't strong enough to gather the movable files. Can anyone give any advice to resolve this issue?

Here's the full "text" on the windowspowershell

Old3DS msed - this can happen on a New3DS
LFCS : 0x6632e2c
msed3 est : 0x146f9e2
Error est : 3930
ID0 hash 0: c7934b018b7a0a392a1e46ac5db68b7a
Hash total: 1
movable_part2.sed generation success
bfcl msky 2c2e63060000000000000000e2f94601 c7934b018b7a0a392a1e46ac5db68b7a 00000000
selected device GeForce 840M on platform NVIDIA CUDA
0.004 seconds for OpenCL compiling
local work size: 1024
ocl_assert: ocl_brute.c, function ocl_brute_msky, line 383
clEnqueueReadBuffer(command_queue, mem_out, CL_TRUE, 0, sizeof(cl_uint), &out, 0, NULL, NULL)
error: out of resources

I am attempting to compile this program to run on ARM chips with a Mali GPU, but am running into compiler errors. Has anyone tested ARM chips running Linux? If not, how much of the program relies on Windows/x86 specific compiler options?

This project's README does not explicitly state whether or not the repository is licensed, and there is also no sort of licensing file. Apache 2.0 is not a copyleft license, thus this project is not subject to having to be licensed under it as well so I am wondering if you would consider licensing this project, as in its current state, it's fully copyrighted. Thank you.

Related:

zoogie/seedminer#16
https://gbatemp.net/posts/7851408/
https://gbatemp.net/posts/7879961/
https://gbatemp.net/posts/7826386/
https://gbatemp.net/posts/7851868/
https://gbatemp.net/posts/7884725/
https://gbatemp.net/posts/7881547/

I'll need testers and reports, including these info:

your GPU model, GPU RAM size, OS version, driver version.
bfcl info output
does seedminer's GPU mode throw a "out of resources" error for you? (yes I also need successful reports)
if the former is true, try the following build with two test commands, does it also say "out of resources"?

Test build:
bfCL-test-reduced-work-size-msky-lfcs-20.zip

Two test commands:
bfcl lfcs 00000007 0000 17f5c00d8b581e5e
bfcl msky c27164f2e0994db8000000007dd5c901 afcb0cc132bd2aeb8e0a6b6a841c51c0

Techy stuff

Despite what it looks like, this doesn't mean your GPU is not powerful/big enough, this program works on Intel IGPU and uses about several KB of GPU RAM, it's more like a OpenCL runtime bug from nvidia to me.

A reduced work size(from a little above 100,000,000 to 1,000,000) helped this guy with a GTX 980, so I guess this is the problem.

from OpenCL SDK document:

global_work_size
Points to an array of work_dim unsigned values that describe the number of global work-items in work_dim dimensions that will execute the kernel function. The total number of global work-items is computed as global_work_size[0] ... global_work_size[work_dim - 1].

The values specified in global_work_size cannot exceed the range given by the sizeof(size_t) for the device on which the kernel execution will be enqueued. The sizeof(size_t) for a device can be determined using CL_DEVICE_ADDRESS_BITS in the table of OpenCL Device Queries for clGetDeviceInfo. If, for example, CL_DEVICE_ADDRESS_BITS = 32, i.e. the device uses a 32-bit address space, size_t is a 32-bit unsigned integer and global_work_size values must be in the range 1 .. 2^32 - 1. Values outside this range return a CL_OUT_OF_RESOURCES error.

nvidia runtime announces GTX 980's address bits = 64, and 100,000,000 is no where near that.

./bfcl msky a888d30500000000000000008f432a01 bd1c2425900e71f1ef8b81f8a257739d 00000000 00400000
selected device AMD Radeon R9 380 Series (tonga, LLVM 14.0.6, DRM 3.47, 5.19.12-arch1-1) on platform Clover
0.846 seconds for OpenCL compiling
local work size: 256
22.01 seconds, 3197043.68 M/s
Max offset reached! Brute-forcing will now terminate!

is an example, I'm using the mesa drivers in Linux. Windows with the drivers from amd's website works fine

i want to known how to use this thing help me please