whisper : use vulkan as gpu backend when available (#2302)

* ggml: use vulkan as gpu backend when available

Signed-off-by: Matt Stephenson <mstephenson6@users.noreply.github.com>

* whisper: enable using vk as default buffer type

Signed-off-by: Matt Stephenson <mstephenson6@users.noreply.github.com>

---------

Signed-off-by: Matt Stephenson <mstephenson6@users.noreply.github.com>
This commit is contained in:
Matt Stephenson
2024-07-16 03:21:09 -04:00
committed by GitHub
parent 7ae885c1ef
commit f68298ce06
2 changed files with 19 additions and 1 deletions

View File

@@ -2042,9 +2042,9 @@ void ggml_vk_instance_init() {
}
static void ggml_vk_init(ggml_backend_vk_context * ctx, size_t idx) {
GGML_ASSERT(idx < vk_instance.device_indices.size());
VK_LOG_DEBUG("ggml_vk_init(" << ctx->name << ", " << idx << ")");
ggml_vk_instance_init();
GGML_ASSERT(idx < vk_instance.device_indices.size());
ctx->name = GGML_VK_NAME + std::to_string(idx);