memory : migrate from llama_kv_cache to more generic llama_memory #14006
Conversation
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
1 task
ggerganov
force-pushed
the
gg/llama-memory-public
branch
from
fe4b1b3 to
bca2671
Compare
ggerganov
force-pushed
the
gg/llama-memory-public
branch
from
bca2671 to
f149a8e
Compare
ggerganov
commented
Jun 5, 2025
|
|
||
| // general concept of LLM memory | ||
| // the KV cache is a type of LLM memory, but there can be other types | ||
| struct llama_memory_i { |
There was a problem hiding this comment.
Changed this from class to struct to be compatible with the C-header declaration.
slaren
approved these changes
Jun 5, 2025
src/llama-context.cpp
Outdated
| llama_kv_cache * kv_self = static_cast<llama_kv_cache *>(memory.get()); | ||
| return kv_self; | ||
| llama_memory_t llama_context::get_memory() const { | ||
| return static_cast<llama_memory_t>(memory.get()); |
There was a problem hiding this comment.
This cast shouldn't be necessary.
src/llama-context.cpp
Outdated
| llama_kv_cache * llama_get_kv_self(llama_context * ctx) { | ||
| return ctx->get_kv_self(); | ||
| return static_cast<llama_kv_cache *>(ctx->get_memory()); |
There was a problem hiding this comment.
I think this is not a safe cast, so it should be a checked with dynamic_cast
ggml-ci
furyhawk
pushed a commit
to furyhawk/llama.cpp
that referenced
this pull request
Jun 6, 2025
…ml-org#14006) * memory : merge llama_kv_cache into llama_memory + new `llama_memory` API ggml-ci * context : fix casts ggml-ci
shefben
added a commit
to shefben/llama.cpp
that referenced
this pull request
Jun 6, 2025
…mory (ggml-org#14006)" This reverts commit 7f37b6c.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
cont #13988
llama_kv_cacheintollama_memory_illama_kv_cache_unifiednow implementsllama_memory_illama_kv_cache_recurrentnow implementsllama_memory_illama_memory_public API tolibllamallama_kv_self_*public API is now simply routing to the newllama_memory_API and it will be deprecated in the next PRTODO
llama_memory_public APINext PRs
llama_kv_self_*public API in favor of the newllama_memory_API