vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

llama : fix data units (#4101)

Browse source 
* llama : fix data units

ggml-ci

* Revert "llama : fix data units"

This reverts commit f5feac831f.

* llama : disambiguate data units

ggml-ci
This commit is contained in:
Georgi Gerganov 2023-11-17 10:00:15 +02:00 • committed by GitHub
parent 91f6499393
commit 4f447a4833
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
3 changed files with 25 additions and 25 deletions
Download patch file Download diff file Expand all files Collapse all files

4
ggml-cuda.cu
View file

@ -5840,7 +5840,7 @@ static void * ggml_cuda_pool_malloc(size_t size, size_t * actual_size) {
return ptr;
}
#ifdef DEBUG_CUDA_MALLOC
fprintf(stderr, "%s: %d buffers, max_size = %u MB, tot_size = %u MB, requested %u MB\n", __func__, nnz,
fprintf(stderr, "%s: %d buffers, max_size = %u MiB, tot_size = %u MiB, requested %u MiB\n", __func__, nnz,
(uint32_t)(max_size/1024/1024), (uint32_t)(tot_size/1024/1024), (uint32_t)(size/1024/1024));
#endif
void * ptr;
@ -5978,7 +5978,7 @@ void * ggml_cuda_host_malloc(size_t size) {
// The allocation error can be bypassed. A null ptr will assigned out of this function.
// This can fixed the OOM error in WSL.
cudaGetLastError();
fprintf(stderr, "WARNING: failed to allocate %.2f MB of pinned memory: %s\n",
fprintf(stderr, "WARNING: failed to allocate %.2f MiB of pinned memory: %s\n",
size/1024.0/1024.0, cudaGetErrorString(err));
return nullptr;
}