Merge mainline llama.cpp (#3)

* Merging mainline - WIP

* Merging mainline - WIP

AVX2 and CUDA appear to work.
CUDA performance seems slightly (~1-2%) lower as it is so often
the case with llama.cpp/ggml after some "improvements" have been made.

* Merging mainline - fix Metal

* Remove check

---------

Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>

1 files changed, 13 insertions, 13 deletions

diff --git a/scripts/check-requirements.sh b/scripts/check-requirements.sh
index 0c6afdd5..d3bbded1 100755
--- a/scripts/check-requirements.sh
+++ b/scripts/check-requirements.sh
@@ -97,9 +97,9 @@ check_requirements() {
 }
 
 check_convert_script() {
-    local py=$1             # e.g. ./convert-hf-to-gguf.py
-    local pyname=${py##*/}  # e.g. convert-hf-to-gguf.py
-    pyname=${pyname%.py}    # e.g. convert-hf-to-gguf
+    local py=$1             # e.g. ./convert_hf_to_gguf.py
+    local pyname=${py##*/}  # e.g. convert_hf_to_gguf.py
+    pyname=${pyname%.py}    # e.g. convert_hf_to_gguf
 
     info "$py: beginning check"
 
@@ -108,6 +108,11 @@ check_convert_script() {
         fatal "$py missing requirements. Expected: $reqs"
     fi
 
+    # Check that all sub-requirements are added to top-level requirements.txt
+    if ! grep -qF "$reqs" requirements.txt; then
+        fatal "$reqs needs to be added to requirements.txt"
+    fi
+
     local venv="$workdir/$pyname-venv"
     python3 -m venv "$venv"
 
@@ -134,12 +139,7 @@ EOF
 
 readonly ignore_eq_eq='check_requirements: ignore "=="'
 
-for req in "$reqs_dir"/*; do
-    # Check that all sub-requirements are added to top-level requirements.txt
-    if ! grep -qF "$req" requirements.txt; then
-        fatal "$req needs to be added to requirements.txt"
-    fi
-
+for req in */**/requirements*.txt; do
     # Make sure exact release versions aren't being pinned in the requirements
     # Filters out the ignore string
     if grep -vF "$ignore_eq_eq" "$req" | grep -q '=='; then
@@ -166,12 +166,12 @@ if (( do_cleanup )); then
     rm -rf -- "$all_venv"
 fi
 
-check_convert_script examples/convert-legacy-llama.py
-for py in convert-*.py; do
-    # skip convert-hf-to-gguf-update.py
+check_convert_script examples/convert_legacy_llama.py
+for py in convert_*.py; do
+    # skip convert_hf_to_gguf_update.py
     # TODO: the check is failing for some reason:
     #       https://github.com/ggerganov/llama.cpp/actions/runs/8875330981/job/24364557177?pr=6920
-    [[ $py == convert-hf-to-gguf-update.py ]] && continue
+    [[ $py == convert_hf_to_gguf_update.py ]] && continue
 
     check_convert_script "$py"
 done