From 899f9a5350937db8b80900a1604edec5997e4b01 Mon Sep 17 00:00:00 2001
From: Georgi Gerganov <ggerganov@gmail.com>
Date: Thu, 17 Aug 2023 19:49:21 +0300
Subject: [PATCH] llama : fix lambda capture

ggml-ci
---
 llama.cpp | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/llama.cpp b/llama.cpp
index 753d130d6..c8403e9c5 100644
--- a/llama.cpp
+++ b/llama.cpp
@@ -3576,7 +3576,7 @@ static void llama_model_quantize_internal(const std::string & fname_inp, const s
             new_data = work.data();
             std::vector<int64_t> hist_cur(1 << 4, 0);
 
-            const int chunk_size = 32 * 512;
+            static const int chunk_size = 32 * 512;
             const int nchunk = (nelements + chunk_size - 1)/chunk_size;
             const int nthread_use = nthread > 1 ? std::max(1, std::min(nthread, nchunk)) : 1;
             if (nthread_use < 2) {
@@ -3584,7 +3584,7 @@ static void llama_model_quantize_internal(const std::string & fname_inp, const s
             } else {
                 size_t counter = 0;
                 new_size = 0;
-                auto compute = [&mutex, &counter, &hist_cur, &new_size, new_type, f32_data, new_data, nelements, chunk_size]() { // NOLINT
+                auto compute = [&mutex, &counter, &hist_cur, &new_size, new_type, f32_data, new_data, nelements]() {
                     std::vector<int64_t> local_hist;
                     size_t local_size = 0;
                     while (true) {