Skip to content

unify profiling, planning and inference - #5

Open
SiddhanthRaja-Nvidia wants to merge 8 commits into
aukarande:pshard/unifiedfrom
SiddhanthRaja-Nvidia:master
Open

unify profiling, planning and inference#5
SiddhanthRaja-Nvidia wants to merge 8 commits into
aukarande:pshard/unifiedfrom
SiddhanthRaja-Nvidia:master

Conversation

@SiddhanthRaja-Nvidia

Copy link
Copy Markdown

Overview

Merge branches - profiling, planning and inference. Retain 3-staged usage.

Additional information

02-unify-notes.md
03-unify-master-plan.md

Requirements

aukarande and others added 8 commits July 27, 2026 08:23
# Conflicts:
#	common/arg.cpp
#	common/common.cpp
#	common/common.h
#	common/fit.cpp
#	common/fit.h
#	ggml/include/ggml-backend.h
#	ggml/src/ggml-backend.cpp
#	src/CMakeLists.txt
#	src/llama-context-pshard.cpp
#	src/llama-context.cpp
#	src/llama-context.h
#	src/llama-ext.h
#	src/llama-kv-cache.cpp
#	src/llama-memory-pshard.cpp
#	src/llama-memory-pshard.h
#	src/llama-memory-recurrent.cpp
#	src/llama-model.cpp
#	src/llama-model.h
#	src/llama-pipe-shard.h
#	src/llama-pshard-plan.h
The planning and inference stages diverged in llama_memory_pshard::assign_tensors:
planning dispatches on the plan's backend_id map, inference dispatches on
l.is_pinned and requires GPU addresses that only pshard_pack_cache_region()
assigns. Taking only the inference variant broke the planner, which never calls
pack_cache_region.

Keep both verbatim as assign_tensors_{planning,inference} and dispatch on
external_buf, which is already the runtime-vs-probe discriminator in init().
Also pass no external buffer during planner probes (no_alloc).
@github-actions github-actions Bot added documentation Improvements or additions to documentation CUDA ggml examples labels Jul 30, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

CUDA documentation Improvements or additions to documentation examples ggml

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants