fix: rename sampling TEMP var to TEMPERATURE to avoid HIP runtime env collision

The ROCm HIP runtime (libamdhip64) reads the TEMP environment variable as
its JIT kernel-cache temp directory. Commit db3eb01 added TEMP=1.0 to the
env files as a sampling default; the serve scripts export it via set -a.
HIP then used the path "1.0" for its cache, failed, and segfaulted with a
NULL-refcount unref (offset 0x446153 in libamdhip64) during model tensor
upload (ggml_backend_cuda_buffer_set_tensor). This crashed frank-default-
large on every restart from Aug 13 16:22 onward; the earlier c984a3e
"fix" (KV/batch changes) was unrelated.

Rename the sampling variable TEMP -> TEMPERATURE in both env files and
serve scripts. llama-server still receives --temp from the renamed var, so
sampling defaults are unchanged. The HIP runtime no longer sees a bogus
TEMP value. Affects frank-default-large (dankclaw) and frank-default-small
(dankserver).
This commit is contained in:
Frank committed 2026-08-13 17:54:29 -07:00
1 parent 6663af1041
commit 227ce01417
4 files changed
+12 -6

No files matched your search

+2 -2
View File
@@ -37,7 +37,7 @@ fi
: "${CACHE_TYPE_V:=q8_0}"
: "${BATCH_SIZE:=2048}"
: "${UBATCH_SIZE:=512}"
: "${TEMP:=1.0}"
: "${TEMPERATURE:=1.0}"
: "${TOP_K:=20}"
: "${TOP_P:=0.95}"
: "${MIN_P:=0.0}"
@@ -72,7 +72,7 @@ exec "$LLAMA_SERVER" \
--parallel "$PARALLEL" \
--batch-size "$BATCH_SIZE" \
--ubatch-size "$UBATCH_SIZE" \
--temp "$TEMP" \
--temp "$TEMPERATURE" \
--top-k "$TOP_K" \
--top-p "$TOP_P" \
--min-p "$MIN_P" \