]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 50e81bd)
kv-cache : log (debug) all streams in find_slot (#15176)
authorDaniel Bevenius <redacted>
Mon, 11 Aug 2025 09:21:19 +0000 (11:21 +0200)
committerGitHub <redacted>
Mon, 11 Aug 2025 09:21:19 +0000 (11:21 +0200)
commitcd3069dfcbeee8e0e96cffb93b0ebb9e595e273a
treea8d2c53d2da92bd9fad805f2e45029e5bcbc0d54tree
parent50e81bdf5db563ab57a9a722b08f96fa8a76c927commit | diff
kv-cache : log (debug) all streams in find_slot (#15176)

This commit updates `llama_kv_cache_unified::find_slot` to log
information for all streams when debug is enabled.

The motivation for this change is that currently if a non-unified
kv-cache is used, then only one stream will be logged because the
code was currently uses `seq_to_stream[1]`.
src/llama-kv-cache-unified.cpp diff | blob | history
Packaging of ggml-org/llama.cpp