-
Notifications
You must be signed in to change notification settings - Fork 9
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Bump jinja2 from 3.1.3 to 3.1.4 #17
base: master
Are you sure you want to change the base?
Commits on Jul 18, 2024
-
Configuration menu - View commit details
-
Copy full SHA for 506d2fe - Browse repository at this point
Copy the full SHA 506d2feView commit details -
Configuration menu - View commit details
-
Copy full SHA for 88289b5 - Browse repository at this point
Copy the full SHA 88289b5View commit details -
Configuration menu - View commit details
-
Copy full SHA for 3f7c4b9 - Browse repository at this point
Copy the full SHA 3f7c4b9View commit details -
Configuration menu - View commit details
-
Copy full SHA for 9d5207b - Browse repository at this point
Copy the full SHA 9d5207bView commit details -
llama : wrap llama_new_context_with_model in try/catch
This fixes a crash where ggml_vk_allocate fails in llama_kv_cache_init, but the exception is never caught.
Configuration menu - View commit details
-
Copy full SHA for b6891bc - Browse repository at this point
Copy the full SHA b6891bcView commit details -
Configuration menu - View commit details
-
Copy full SHA for b80287e - Browse repository at this point
Copy the full SHA b80287eView commit details -
kompute : fix ggml_vk_allocate failure control flow
The correct way to indicate an OOM condition is for alloc_buffer to return NULL. This fixes undefined behavior caused by passing an exception over the C boundary. The rest of the changes help fix VRAM leaks in GPT4All when model loading fails on GPU.
Configuration menu - View commit details
-
Copy full SHA for dc7a50b - Browse repository at this point
Copy the full SHA dc7a50bView commit details -
kompute : disable GPU offload for Mixtral
We haven't implemented the necessary GPU kernels yet. Fixes this crash: ggml_vk_graph_compute: error: unsupported op 'ARGSORT' GGML_ASSERT: /home/jared/src/forks/gpt4all/gpt4all-backend/llama.cpp-mainline/ggml-kompute.cpp:1508: !"unsupported op"
Configuration menu - View commit details
-
Copy full SHA for c5014f6 - Browse repository at this point
Copy the full SHA c5014f6View commit details -
Configuration menu - View commit details
-
Copy full SHA for c76f5c3 - Browse repository at this point
Copy the full SHA c76f5c3View commit details -
Configuration menu - View commit details
-
Copy full SHA for 6ff4387 - Browse repository at this point
Copy the full SHA 6ff4387View commit details -
Configuration menu - View commit details
-
Copy full SHA for 12dcddc - Browse repository at this point
Copy the full SHA 12dcddcView commit details -
Configuration menu - View commit details
-
Copy full SHA for 82b50e5 - Browse repository at this point
Copy the full SHA 82b50e5View commit details -
kompute : enable GPU support for 10 more model architectures
These are Baichuan, Bert and Nomic Bert, CodeShell, GPT-2, InternLM, MiniCPM, Orion, Qwen, and StarCoder.
Configuration menu - View commit details
-
Copy full SHA for a76f5f4 - Browse repository at this point
Copy the full SHA a76f5f4View commit details -
Configuration menu - View commit details
-
Copy full SHA for 877851b - Browse repository at this point
Copy the full SHA 877851bView commit details -
Configuration menu - View commit details
-
Copy full SHA for 729d661 - Browse repository at this point
Copy the full SHA 729d661View commit details -
Configuration menu - View commit details
-
Copy full SHA for 2b8cb26 - Browse repository at this point
Copy the full SHA 2b8cb26View commit details -
Configuration menu - View commit details
-
Copy full SHA for 6e0b5d9 - Browse repository at this point
Copy the full SHA 6e0b5d9View commit details -
Configuration menu - View commit details
-
Copy full SHA for aea0abe - Browse repository at this point
Copy the full SHA aea0abeView commit details -
Configuration menu - View commit details
-
Copy full SHA for 535c7b1 - Browse repository at this point
Copy the full SHA 535c7b1View commit details -
vulkan : guard against multiple initialization
This trades a late heap-use-after-free for an early abort, which feels more correct.
Configuration menu - View commit details
-
Copy full SHA for 2a91dbf - Browse repository at this point
Copy the full SHA 2a91dbfView commit details -
Configuration menu - View commit details
-
Copy full SHA for ad1ab57 - Browse repository at this point
Copy the full SHA ad1ab57View commit details -
Configuration menu - View commit details
-
Copy full SHA for 09058b1 - Browse repository at this point
Copy the full SHA 09058b1View commit details -
Configuration menu - View commit details
-
Copy full SHA for b0ccbe1 - Browse repository at this point
Copy the full SHA b0ccbe1View commit details -
Configuration menu - View commit details
-
Copy full SHA for 74a41c6 - Browse repository at this point
Copy the full SHA 74a41c6View commit details -
Configuration menu - View commit details
-
Copy full SHA for f10326c - Browse repository at this point
Copy the full SHA f10326cView commit details -
Configuration menu - View commit details
-
Copy full SHA for e5c0df7 - Browse repository at this point
Copy the full SHA e5c0df7View commit details -
llama : use the correct buffer type when we choose not to load on GPU
This fixes a regression in commit b2db03a ("llama: replace ngl=0 hack with llama_model_using_gpu").
Configuration menu - View commit details
-
Copy full SHA for 159235e - Browse repository at this point
Copy the full SHA 159235eView commit details -
Configuration menu - View commit details
-
Copy full SHA for c301b42 - Browse repository at this point
Copy the full SHA c301b42View commit details -
Configuration menu - View commit details
-
Copy full SHA for 7d402b3 - Browse repository at this point
Copy the full SHA 7d402b3View commit details -
Configuration menu - View commit details
-
Copy full SHA for 48a830c - Browse repository at this point
Copy the full SHA 48a830cView commit details -
kompute : fix dangling references in ggml_vk_graph_kompute
ggml_vk_get_tensor_aligned() returns a shared_ptr, not a reference, so we must copy the value.
Configuration menu - View commit details
-
Copy full SHA for 6e0ad3c - Browse repository at this point
Copy the full SHA 6e0ad3cView commit details -
kompute : avoid freeing device/instance until absolutely necessary
Eagerly freeing the instance when we are done with it is simple, but incurs an overhead, and more importantly, causes test-backend-ops crashes on the current proprietary NVIDIA driver. Instead, we now only cleanup device resources without freeing the device unless we actually need to change devices. And even when we free the device, we do not free the instance. We only free the instance when both the backend and all buffers have been unreferenced.
Configuration menu - View commit details
-
Copy full SHA for c3d5264 - Browse repository at this point
Copy the full SHA c3d5264View commit details -
kompute : update ggml_vk_supports_op to fix false pos/neg
test-backend-ops hit assertion failures in ggml_vk_graph_compute because of ops we do not yet support. Some of the checks have to be made more restrictive because of features that were added to llama.cpp. We also claimed to not support no-op operations on certain data types, even though they are actually supported on all data types. There are now 243 passsing tests, instead of 150 without the fixes for false negatives. This also fixes complaints during LLM inference about unsupported NONE operations for the output tensor.
Configuration menu - View commit details
-
Copy full SHA for 561d0ce - Browse repository at this point
Copy the full SHA 561d0ceView commit details -
Configuration menu - View commit details
-
Copy full SHA for cd13f44 - Browse repository at this point
Copy the full SHA cd13f44View commit details -
Bump jinja2 from 3.1.3 to 3.1.4
Bumps [jinja2](https://github.com/pallets/jinja) from 3.1.3 to 3.1.4. - [Release notes](https://github.com/pallets/jinja/releases) - [Changelog](https://github.com/pallets/jinja/blob/main/CHANGES.rst) - [Commits](pallets/jinja@3.1.3...3.1.4) --- updated-dependencies: - dependency-name: jinja2 dependency-type: indirect ... Signed-off-by: dependabot[bot] <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for 84c5f62 - Browse repository at this point
Copy the full SHA 84c5f62View commit details