4a6ed42d6b
Merge pull request 'readme: 4090 at 727 MH/s on the loop kernel, how to select it' (#30) from docs/readme-lair2 into main
v4.0.2-lair.2
grenade2026-09-14 16:39:23 +00:00
bbbaf815ac
readme: 4090 at 727 MH/s on the loop kernel, how to select it
rob thijssen2026-09-14 19:30:43 +03:00
ad77d99dfa
Merge pull request 'engine-cuda: loop-structured mining_loop kernel, opt-in via MINER_CUDA_KERNEL=loop' (#29) from perf/loop-kernel into main
grenade2026-09-14 13:44:55 +00:00
12bd36aee2
deploy: per-host CUDA kernel choice; benjy runs the loop kernel
rob thijssen2026-09-14 16:36:02 +03:00
8b01387e97
Merge pull request 'release: tag-triggered binaries and a README that states the measured rates' (#28) from release/binaries-and-readme into main
v4.0.2-lair.1
grenade2026-09-13 22:24:05 +00:00
7831323362
release: tag-triggered binaries, README section with measured rates
rob thijssen2026-09-14 01:13:49 +03:00
17caeabc10
Merge pull request 'perf: whole-grid batches, early reject on element 0, precomputed nonce direction (+2.9% on beast)' (#26) from perf/scheduling into main
grenade2026-09-13 21:06:31 +00:00
2e6fce162e
engine-cuda: batches of whole grids, early reject on element 0, nonce direction
rob thijssen2026-09-13 23:54:17 +03:00
dc6c58ceb6
Merge pull request 'perf: fused-PTX field arithmetic, +54..66% on every card' (#23) from perf/fused-field-ptx into main
grenade2026-09-13 19:40:26 +00:00
3763996ae6
engine-cuda: fused-PTX field arithmetic, +54..66% on every card
rob thijssen2026-09-13 22:29:46 +03:00
78dee04ff6
Merge pull request 'perf: parameter-bank uniforms and per-arch carry path (+20.5% sm_120)' (#22) from perf/sm120-spills into main
grenade2026-09-13 17:20:53 +00:00
206aa0d876
engine-cuda: record why gf_sqr uses the general product
rob thijssen2026-09-13 18:42:48 +03:00
7804d9d442
engine-cuda: launch uniforms in the parameter bank, no spills
rob thijssen2026-09-13 18:32:20 +03:00
2cfb2aae1d
Merge pull request 'deploy: miner credentials from the mainnet chain' (#21) from deploy/mainnet-chain into main
grenade2026-09-09 09:04:30 +00:00
b1c6eb2ba7
deploy: miner credentials from the mainnet chain
rob thijssen2026-09-09 12:03:21 +03:00
fd965ea312
Merge pull request 'deploy: beast joins the miner fleet for mainnet' (#20) from deploy/enable-beast into main
grenade2026-09-09 06:15:02 +00:00
7ac9608b3d
deploy: beast joins the miner fleet for mainnet
rob thijssen2026-09-09 09:14:32 +03:00
03b9c4239f
Merge pull request 'deploy: chain id as a dispatch input for the miner credential path' (#19) from deploy/chain-input into main
grenade2026-09-09 05:05:36 +00:00
9f41e0f3b9
deploy: chain id as a dispatch input for the miner credential path
rob thijssen2026-09-09 08:05:04 +03:00
f061ccfeea
Merge pull request 'bench: record the memory clock alongside the SM clock (schema 2)' (#18) from bench/record-memory-clock into main
grenade2026-09-03 15:54:47 +00:00
a7a8ff0532
bench: record the memory clock alongside the SM clock (schema 2)
rob thijssen2026-09-03 18:47:17 +03:00
ad66f1334f
Merge pull request 'engine-cuda: deferred carries, carry-flag arithmetic, one nonce per thread' (#17) from cuda/tuning into main
grenade2026-09-03 13:01:46 +00:00
b640d3a711
engine-cuda: fold round constants into the preceding linear layer; bench on cuda paths
rob thijssen2026-09-03 15:56:23 +03:00
2bf3246634
engine-cuda: deferred carries, carry-flag arithmetic, one nonce per thread
rob thijssen2026-09-03 15:51:11 +03:00
d5e72d6a15
Merge pull request 'engine-cuda: native CUDA mining engine behind MinerEngine' (#16) from cuda/engine-skeleton into main
grenade2026-09-03 11:57:42 +00:00
6ecc5ee6c7
engine-cuda: native CUDA mining engine behind MinerEngine
rob thijssen2026-09-03 14:52:10 +03:00
48ab3933ba
Merge pull request 'metrics: stale work is the batch in flight at cancellation, not the whole search' (#15) from metrics/stale-semantics into main
grenade2026-09-03 11:28:43 +00:00
20749c622b
metrics: stale work is the batch in flight at cancellation, not the whole search
rob thijssen2026-09-03 14:23:37 +03:00
1c8174951c
Merge pull request 'metrics: attribute hashrate to build, device, kernel and job outcomes' (#14) from metrics/build-device-job into main
grenade2026-09-03 11:16:25 +00:00
3ba996dcdf
metrics: attribute hashrate to build, device, kernel and job outcomes
rob thijssen2026-09-03 14:09:43 +03:00
f1064345e5
Merge pull request 'deploy: wait for the miner to connect before judging its metrics' (#13) from deploy/validate-readiness into main
grenade2026-09-03 10:38:21 +00:00
40b4ea32ca
deploy: restore the previous binary with install, not cp
rob thijssen2026-09-03 13:38:09 +03:00
b10f2f0099
deploy: wait for the miner to connect before judging its metrics
rob thijssen2026-09-03 13:37:23 +03:00
95b6420793
Merge pull request 'deploy: build on push to main and land the binary on the mining hosts' (#12) from deploy/on-main into main
grenade2026-09-03 10:33:17 +00:00
b9e1d21ec9
deploy: build on push to main and land the binary on the mining hosts
rob thijssen2026-09-03 13:28:52 +03:00
ad846af984
Merge pull request 'bench: quantus-bench harness and a workflow that measures on a mining host' (#11) from bench/harness into main
grenade2026-09-03 10:19:33 +00:00
ca6642960c
bench: serialise measurements per host with a flock, per-run file names
rob thijssen2026-09-03 13:06:45 +03:00
5f3c51c5cb
bench: allow more worker threads than cards to measure submission overlap
rob thijssen2026-09-03 13:03:21 +03:00
36ffbbadbf
Merge pull request 'ci: gitea workflow, drop inherited github workflows, embed commit in --version' (#10) from ci/gitea-workflows-and-build-sha into main
grenade2026-09-03 10:02:39 +00:00
bad02fbc33
bench: quantus-bench harness and a workflow that measures on a mining host
rob thijssen2026-09-03 12:58:48 +03:00
db6316713a
ci: add rustfmt and clippy components explicitly
rob thijssen2026-09-03 12:54:29 +03:00
094536a341
ci: gitea workflow, drop inherited github workflows, embed commit in --version
rob thijssen2026-09-03 12:50:45 +03:00
grenade
perf/minblocks-knob
Nikolaus Heger