Allocator fragmentation diagnostics (dev tooling) - #1619
Open
dxqb wants to merge 2 commits into
Open
Conversation
…()/evict() API
Replaces the per-model `{part}_to(device)` methods across all model classes,
plus scattered call sites in dataLoader/modelSetup/modelSampler/GenericTrainer,
with generic BaseModel methods driven by the existing ModelType.model_parts()
registry: materialize(*parts), evict(*parts), and materialize_only(*parts)
(evict everything else, then materialize the given parts - the swap-in/swap-out
pattern used throughout the Samplers and text-caching setup). eval() and
adapters() are likewise made concrete on BaseModel instead of hand-written per
model. Models whose component names diverge (Wuerstchen) or that have
components outside model_parts() (SD's depth_estimator, Anima's
text_conditioner) override the relevant methods directly.
Also fixes multi-TE samplers (Flux/SD3/SDXL/HiDream/HunyuanVideo) that
previously evicted all but the first text encoder and ran encode_text with
the rest still on temp_device.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
print_fragmentation() reports live allocated/reserved/free/largest-free and external fragmentation (free memory not in the largest contiguous hole) for the current CUDA device. An OOM is about contiguity, not total free bytes, so ext_frag near 1 (free memory scattered across many small holes) is the fragmentation that blocks a large allocation even when free > request. Call sites are commented out by default (training stop, after unload, and end of BaseModel.evict()) since collecting a memory_snapshot walks every allocator segment/block; uncomment to use during investigation. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
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.
Summary
Adds debug info for vram fragmentation.
only ~ 30 code lines, but it's based on #1617 so it'll show a large diff initially
Test plan
pre-commit run --all-filespassesAI assistance