blackbeard/miner
1
0
Fork 0
Code Issues 2 Pull Requests Actions Packages Projects Releases 2 Wiki Activity
77 Commits 4 Branches 20 Tags
92f58d1b7645dbfe1a0239da8c0874ef0de60f7c
Commit Graph
77 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
rob thijssen 92f58d1b76 immutable bindings for is_g2 and d_distance 2025-09-13 00:10:40 +03:00
rob thijssen 4ca906ac1b corrected host-side constant upload to pass CStrings 2025-09-12 23:55:04 +03:00
rob thijssen 943d35100f Fix conditional use of n0_inv in qpow Montgomery kernels 2025-09-12 18:35:01 +03:00
rob thijssen 398c9f9bc9 Use __constant__ memory for per-job constants in CUDA kernels 
Add optional __constant__ symbols for G2 kernel constants and prefer
these over parameter pointers when available. Update host code to
attempt populating these constant symbols for faster access.
 2025-09-12 18:14:26 +03:00
rob thijssen b7f56fb05a Add CUDA miner environment presets for various NVIDIA GPUs 
Provide example .env files targeting typical block/thread configurations
for RTX 3060/3080/3090/4080/4090/5090 and A5000/A6000.

Include guidance on tuning iterations, pinned buffers, and kernel modes.
Document usage with systemd and manual export for ease of deployment.
 2025-09-12 17:19:23 +03:00
rob thijssen 69cbb7fa99 remove 1024 thread cap in CUDA kernel launches 2025-09-12 16:24:45 +03:00
rob thijssen 1d956c499e Allocate and copy-back CUDA output buffer for G1 path 
Add optional pinned host buffer support and parallel host SHA3
validation over GPU results, updating hash count and early exit on
success.
 2025-09-12 16:00:52 +03:00
rob thijssen 297a74181a Refactor CUDA kernel launch to conditionally allocate output buffer 
Allocate d_y_out only for G1 kernel instead of always allocating before
launch, and track kernel type with is_g2 flag for clearer control flow.
 2025-09-12 15:52:21 +03:00
rob thijssen 9167b8e48d Refactor CUDA kernel launch to separate G1 and G2 modes 
Reorder code to launch G2 kernel with on-device SHA3 and early-exit
handling separately from G1 kernel that returns y values. Rename index
variables for clarity and adjust logging to measure kernel execution
time in both branches.
 2025-09-12 15:09:32 +03:00
rob thijssen baf15fbb75 Add G2 kernel with device SHA3-512, early-exit, and threshold compare 
- Implemented qpow_montgomery_g2_kernel for on-device SHA3-512 hashing -
Added early-exit flag and linear index tracking in G2 kernel -
Integrated G2 kernel launch and result handling in Rust CUDA engine -
Introduced MINER_CUDA_MODE=g2 config option to select G2 kernel - Added
MINER_CUDA_PINNED option for pinned host buffers and async copies -
Updated README with new runtime config knobs and G2 usage instructions
 2025-09-12 14:38:13 +03:00
rob thijssen 5b44ef20df Use optional pinned host buffer and async copy for CUDA results 
Enable using a page-locked pinned buffer with asynchronous copy-back
when MINER_CUDA_PINNED is set, otherwise fall back to synchronous copy.
 2025-09-12 14:00:53 +03:00
rob thijssen c0fa1d15aa Limit SHA3 host hashing to remaining range and use scoped threads 2025-09-12 12:56:58 +03:00
rob thijssen cec209170f Fix paths-ignore order in CUDA workflow triggers 2025-09-12 12:47:29 +03:00
rob thijssen 99af459363 Reorder disk action step in CUDA build workflow 2025-09-12 12:44:04 +03:00
rob thijssen 910e614cfb Add MINER_CUDA_IMAGE override and parallel host SHA3 workers 
- Allow forcing PTX or CUBIN image via MINER_CUDA_IMAGE env var - Enable
selecting G2 kernel with fallback to G1 via MINER_CUDA_MODE=g2 -
Allocate device constant buffers once outside loop - Implement parallel
host SHA3 over GPU results using MINER_CUDA_HASH_THREADS - Update README
with new env vars and usage notes
 2025-09-12 12:40:47 +03:00
rob thijssen abd8c878f7 Add disk space cleanup action and integrate in CUDA workflow 
Ignore unrelated file changes in CUDA workflow triggers to reduce
unnecessary runs
 2025-09-12 11:29:06 +03:00
rob thijssen 96875e12b2 Use self-hosted runner for quality-control job in CI workflow 2025-09-12 11:17:36 +03:00
rob thijssen 9bb6dc840d Log SHA3 host timing after GPU result consumption 2025-09-12 11:15:42 +03:00
rob thijssen 14115a12c1 Add timing logs for CUDA kernel and copy-back operations 2025-09-12 11:09:28 +03:00
rob thijssen 39d9eed005 Add CARGO_TERM_COLOR=always environment variable to Containerfile 2025-09-12 09:35:37 +03:00
rob thijssen 60d25d0d8e Export CUDA_TAG and SM args as environment variables in Containerfile 2025-09-12 09:34:02 +03:00
rob thijssen bfeb0328cd Refactor CUDA build workflow to use repo Containerfile 
Replace inline Dockerfile with repo Containerfile and export built
binary via build output instead of docker image copy.
 2025-09-12 09:14:55 +03:00
rob thijssen 6e089208ab Add fallback CUDA include paths in build script and deduplicate dirs 2025-09-12 08:57:28 +03:00
rob thijssen 48e99f9db5 reduce artifact retention to 3 days and add CUDA_HOME env variable 2025-09-12 08:49:26 +03:00
rob thijssen 70485d3d91 Remove optional rsync to CDN from CUDA build workflow 2025-09-12 08:39:29 +03:00
rob thijssen c703e7d273 Fix condition to check SSH key presence correctly in workflow 2025-09-12 08:38:46 +03:00
rob thijssen aef8c865cb clippy cleanup 2025-09-12 08:34:54 +03:00
rob thijssen d2c39bbdcb ci: build cuda binaries 2025-09-12 08:29:09 +03:00
rob thijssen 8b02b790a4 wip: cubin embed 2025-09-12 04:22:15 +03:00
rob thijssen 222890dc92 cuda scaffold 
- Add engine-gpu-cuda crate with basic CUDA kernel (qpow G1) - Implement
CUDA build.rs compiling .cu to PTX when "cuda" feature enabled -
Integrate GPU engine into miner-service with runtime CUDA availability
check - Add fallback to CPU engine in gpu-cuda until full GPU kernel
support is ready - Enable optional "cuda" feature in miner-cli and
miner-service crates
 2025-09-11 18:51:52 +08:00
rob thijssen e107f03c5a Merge pull request #3 from Quantus-Network/feat/montgomery 
fast modular arithmetic (montgomery multiplication)
 2025-09-11 18:40:48 +08:00
rob thijssen 50a620b9b4 fix clippy warning and simplify cpu feature detection in mont_portable 
module
 2025-09-11 10:13:35 +03:00
rob thijssen 113a2ad78c refactor montgomery normalization and fix mul target detection 2025-09-11 10:06:07 +03:00
rob thijssen 80af3ce052 test triage 2025-09-11 09:38:30 +03:00
rob thijssen 68c5835fc6 clippy cleanup 2025-09-11 09:19:07 +03:00
rob thijssen 60cc91ddd6 taplo tidy 2025-09-11 09:07:20 +03:00
rob thijssen f8e2378045 chore: replace references to deprecated --num-cores flag with --workers 2025-09-11 08:24:23 +03:00
rob thijssen 63f271519d Add aarch64 UMULH/ADCS Montgomery backend and update docs 2025-09-11 01:42:06 +08:00
rob thijssen 317734ad0e Add aarch64 UMULH optimized Montgomery multiplication backend 2025-09-11 01:32:31 +08:00
rob thijssen ef2797d295 Add BMI2+ADX optimized Montgomery multiplication with inline asm 
Implement CIOS multiplication using MULX, ADCX, and ADOX instructions
for x86_64 with BMI2 and ADX features detected at runtime. Fall back to
other implementations when features are unavailable.
 2025-09-11 01:21:15 +08:00
rob thijssen 11839c07b2 feat: conditional metrics for engine-montgomery in miner-service cargo 
features
 2025-09-10 23:36:56 +08:00
rob thijssen 8b758492c1 fix: optional metrics feature to engine-montgomery crate 2025-09-10 23:17:01 +08:00
rob thijssen 7ec802ee24 feat: montgomery-optimized cpu mining engine 
- introduce new `engine-montgomery` crate implementing a 512-bit
fixed-width montgomery multiplication backend for qpow mining. - mirror
`cpu-fast` engine behavior and metrics for direct comparison. - add
runtime backend selection with portable and x86_64 bmi2/adx kernels. -
expose `--engine cpu-montgomery` cli option in miner-cli. - integrate
engine and metrics support in miner-service with feature gating. - add
engine backend info metric for observability and a/b testing.
 2025-09-10 22:52:04 +08:00
rob thijssen 604f61fd9d Merge pull request #2 from Quantus-Network/workers 
chore: replace references to deprecated --num-cores flag with --workers
 2025-09-10 21:31:29 +08:00
rob thijssen b035a497fc chore: replace references to deprecated --num-cores flag with --workers 2025-09-10 19:45:06 +08:00
rob thijssen fc6a6b1014 Merge pull request #1 from Quantus-Network/feat/optimisation 
2500× hashrate boost: 20k → up to 50m ops/s (cpu-fast, metrics, throttling, stable)
 2025-09-10 17:35:40 +08:00
rob thijssen 1b4771ebc3 pr-cleanup: scrubbed references to implementation docs 2025-09-10 11:34:32 +08:00
rob thijssen 7cbeb43b39 pr-cleanup 2025-09-10 11:08:39 +08:00
rob thijssen b5cfa06ffe 09: unit tests 
- Verify Baseline and Fast CPU engines produce matching results - Test
Exhausted and Cancelled engine states with precise hash counts - Cover
miner-service request validation error branches - Test miner-service
HTTP endpoints for expected status codes - Confirm chunked mining sends
progress and completion messages
 2025-09-10 10:46:36 +08:00
rob thijssen cff7a3284b 08: cpuset metric and systemd units 
- Detect effective CPUs with cpuset awareness (cgroup v2/v1 fallback) -
Log detected cpuset mask or fallback info at startup (debug) - Expose
miner_effective_cpus Prometheus gauge (metrics feature) - Add
production-ready systemd unit and example CPU affinity overrides -
Include documentation for installation, overrides, and validation
 2025-09-10 09:57:42 +08:00