From 5bf3953d7e9831ea22b0bc017ce97409b801ccf1 Mon Sep 17 00:00:00 2001 From: slaren Date: Sun, 24 Dec 2023 14:34:22 +0100 Subject: cuda : improve cuda pool efficiency using virtual memory (#4606) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit * cuda : improve cuda pool efficiency using virtual memory * fix mixtral * fix cmake build * check for vmm support, disable for hip ggml-ci * fix hip build * clarify granularity * move all caps to g_device_caps * refactor error checking * add cuda_pool_alloc, refactor most pool allocations ggml-ci * fix hip build * CUBLAS_TF32_TENSOR_OP_MATH is not a macro * more hip crap * llama : fix msvc warnings * ggml : fix msvc warnings * minor * minor * cuda : fallback to CPU on host buffer alloc fail * Update ggml-cuda.cu Co-authored-by: Johannes Gäßler * Update ggml-cuda.cu Co-authored-by: Johannes Gäßler * ensure allocations are always aligned * act_size -> actual_size --------- Co-authored-by: Johannes Gäßler --- tests/test-grad0.cpp | 3 --- 1 file changed, 3 deletions(-) (limited to 'tests') diff --git a/tests/test-grad0.cpp b/tests/test-grad0.cpp index 14914def..8ff76c89 100644 --- a/tests/test-grad0.cpp +++ b/tests/test-grad0.cpp @@ -883,9 +883,6 @@ int main(int argc, const char ** argv) { srand(seed); const int nargs = 1; - int64_t ne2[4]; - ne2[0] = 1; - for (int ndims = 1; ndims <= 2; ++ndims) { x[0] = get_random_tensor_f32(ctx0, ndims, ne, -1.0f, 1.0f); -- cgit v1.2.3