mirror of
https://github.com/ollama/ollama.git
synced 2025-11-11 22:37:48 +01:00
d5a0d8d904
This changes the memory allocation strategy from upfront estimation to tracking actual allocations done by the engine and reacting to that. The goal is avoid issues caused by both under-estimation (crashing) and over-estimation (low performance due to under-utilized GPUs). It is currently opt-in and can be enabled for models running on the Ollama engine by setting OLLAMA_NEW_ESTIMATES=1. Behavior in other cases is unchanged and will continue to use the existing estimates.
26 lines
1.0 KiB
Diff
26 lines
1.0 KiB
Diff
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
|
|
From: Daniel Hiltgen <daniel@ollama.com>
|
|
Date: Sun, 3 Aug 2025 10:00:20 -0700
|
|
Subject: [PATCH] Disable ggml-blas on macos v13 and older
|
|
|
|
---
|
|
ggml/src/ggml-blas/ggml-blas.cpp | 5 +++++
|
|
1 file changed, 5 insertions(+)
|
|
|
|
diff --git a/ggml/src/ggml-blas/ggml-blas.cpp b/ggml/src/ggml-blas/ggml-blas.cpp
|
|
index aeac2e57..40738d5b 100644
|
|
--- a/ggml/src/ggml-blas/ggml-blas.cpp
|
|
+++ b/ggml/src/ggml-blas/ggml-blas.cpp
|
|
@@ -505,6 +505,11 @@ static const struct ggml_backend_reg_i ggml_backend_blas_reg_i = {
|
|
};
|
|
|
|
ggml_backend_reg_t ggml_backend_blas_reg(void) {
|
|
+ // MacOS prior to v14 does not include cblas_sgemm - disable this backend if it isn't available
|
|
+ if (&cblas_sgemm == NULL) {
|
|
+ GGML_LOG_INFO("Disabling ggml-blas backend on old MacOS version\n");
|
|
+ return NULL;
|
|
+ }
|
|
static struct ggml_backend_reg ggml_backend_blas_reg = {
|
|
/* .api_version = */ GGML_BACKEND_API_VERSION,
|
|
/* .iface = */ ggml_backend_blas_reg_i,
|