[autotuner] Triton reduction seed heuristic (generalizable core)#2762
Open
calebmkim wants to merge 1 commit into
Open
[autotuner] Triton reduction seed heuristic (generalizable core)#2762calebmkim wants to merge 1 commit into
calebmkim wants to merge 1 commit into
Conversation
calebmkim
pushed a commit
that referenced
this pull request
Jun 11, 2026
Add the Triton inner-reduction seed heuristic, reading the reduction facts from the previous PR. - TritonReductionTileHeuristic (T1: rollable rdim — sum, rms_norm, layer_norm, softmax-row, cross_entropy) and TritonReductionUserTileHeuristic (T2: user-tiled — softmax_two_pass, kl_div, jsd, welford/groupnorm). Persistent-vs-looped element/byte spill caps, the rnumel num_warps ramp, Band-B R_BLOCK + Band-C combine footprint caps, the narrow-row w1 occupancy gate, the M_BLOCK>1 apply-loop stream cap, per-slot stream/re-read eviction, off-sm90 conservative fallback. This is the pruned generalizable core — the over-fit dtype tail is intentionally deferred. - helion/_compiler/autotuner_heuristics/__init__.py: register the two heuristics. - test/test_autotuner_heuristics.py, test/test_best_available.py: heuristic decisions + reduction-loop config round-trip coverage. stack-info: PR: #2762, branch: calebmkim/stack/3
calebmkim
force-pushed
the
calebmkim/stack/2
branch
from
5041cf6 to
124d450
Compare
calebmkim
force-pushed
the
calebmkim/stack/3
branch
from
8c3bb63 to
4a766eb
Compare
meta-cla
Bot
added
the
CLA Signed
calebmkim
pushed a commit
that referenced
this pull request
Jun 11, 2026
Add the Triton inner-reduction seed heuristic, reading the reduction facts from the previous PR. - TritonReductionTileHeuristic (T1: rollable rdim — sum, rms_norm, layer_norm, softmax-row, cross_entropy) and TritonReductionUserTileHeuristic (T2: user-tiled — softmax_two_pass, kl_div, jsd, welford/groupnorm). Persistent-vs-looped element/byte spill caps, the rnumel num_warps ramp, Band-B R_BLOCK + Band-C combine footprint caps, the narrow-row w1 occupancy gate, the M_BLOCK>1 apply-loop stream cap, per-slot stream/re-read eviction, off-sm90 conservative fallback. This is the pruned generalizable core — the over-fit dtype tail is intentionally deferred. - helion/_compiler/autotuner_heuristics/__init__.py: register the two heuristics. - test/test_autotuner_heuristics.py, test/test_best_available.py: heuristic decisions + reduction-loop config round-trip coverage. stack-info: PR: #2762, branch: calebmkim/stack/3
calebmkim
force-pushed
the
calebmkim/stack/3
branch
from
4a766eb to
4a1c32d
Compare
calebmkim
force-pushed
the
calebmkim/stack/2
branch
from
124d450 to
0e1b911
Compare
calebmkim
pushed a commit
that referenced
this pull request
Jun 11, 2026
Add the Triton inner-reduction seed heuristic, reading the reduction facts from the previous PR. - TritonReductionTileHeuristic (T1: rollable rdim — sum, rms_norm, layer_norm, softmax-row, cross_entropy) and TritonReductionUserTileHeuristic (T2: user-tiled — softmax_two_pass, kl_div, jsd, welford/groupnorm). Persistent-vs-looped element/byte spill caps, the rnumel num_warps ramp, Band-B R_BLOCK + Band-C combine footprint caps, the narrow-row w1 occupancy gate, the M_BLOCK>1 apply-loop stream cap, per-slot stream/re-read eviction, off-sm90 conservative fallback. This is the pruned generalizable core — the over-fit dtype tail is intentionally deferred. - helion/_compiler/autotuner_heuristics/__init__.py: register the two heuristics. - test/test_autotuner_heuristics.py, test/test_best_available.py: heuristic decisions + reduction-loop config round-trip coverage. stack-info: PR: #2762, branch: calebmkim/stack/3
calebmkim
force-pushed
the
calebmkim/stack/3
branch
from
4a1c32d to
bbfa3cd
Compare
calebmkim
pushed a commit
that referenced
this pull request
Jun 12, 2026
Add the Triton inner-reduction seed heuristic, reading the reduction facts from the previous PR. - TritonReductionTileHeuristic (T1: rollable rdim — sum, rms_norm, layer_norm, softmax-row, cross_entropy) and TritonReductionUserTileHeuristic (T2: user-tiled — softmax_two_pass, kl_div, jsd, welford/groupnorm). Persistent-vs-looped element/byte spill caps, the rnumel num_warps ramp, Band-B R_BLOCK + Band-C combine footprint caps, the narrow-row w1 occupancy gate, the M_BLOCK>1 apply-loop stream cap, per-slot stream/re-read eviction, off-sm90 conservative fallback. This is the pruned generalizable core — the over-fit dtype tail is intentionally deferred. - helion/_compiler/autotuner_heuristics/__init__.py: register the two heuristics. - test/test_autotuner_heuristics.py, test/test_best_available.py: heuristic decisions + reduction-loop config round-trip coverage. stack-info: PR: #2762, branch: calebmkim/stack/3
calebmkim
force-pushed
the
calebmkim/stack/3
branch
from
bbfa3cd to
d6ad115
Compare
Add the Triton inner-reduction seed heuristic, reading the reduction facts from the previous PR. - TritonReductionTileHeuristic (T1: rollable rdim — sum, rms_norm, layer_norm, softmax-row, cross_entropy) and TritonReductionUserTileHeuristic (T2: user-tiled — softmax_two_pass, kl_div, jsd, welford/groupnorm). Persistent-vs-looped element/byte spill caps, the rnumel num_warps ramp, Band-B R_BLOCK + Band-C combine footprint caps, the narrow-row w1 occupancy gate, the M_BLOCK>1 apply-loop stream cap, per-slot stream/re-read eviction, off-sm90 conservative fallback. This is the pruned generalizable core — the over-fit dtype tail is intentionally deferred. - helion/_compiler/autotuner_heuristics/__init__.py: register the two heuristics. - test/test_autotuner_heuristics.py, test/test_best_available.py: heuristic decisions + reduction-loop config round-trip coverage. stack-info: PR: #2762, branch: calebmkim/stack/3
calebmkim
force-pushed
the
calebmkim/stack/3
branch
from
d6ad115 to
4a7c335
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Stacked PRs:
[autotuner] Triton reduction seed heuristic (generalizable core)
Add the Triton inner-reduction seed heuristic, reading the reduction facts from the
previous PR.
softmax-row, cross_entropy) and TritonReductionUserTileHeuristic (T2: user-tiled —
softmax_two_pass, kl_div, jsd, welford/groupnorm). Persistent-vs-looped element/byte
spill caps, the rnumel num_warps ramp, Band-B R_BLOCK + Band-C combine footprint
caps, the narrow-row w1 occupancy gate, the M_BLOCK>1 apply-loop stream cap,
per-slot stream/re-read eviction, off-sm90 conservative fallback. This is the pruned
generalizable core — the over-fit dtype tail is intentionally deferred.