sync : llama.cpp #981
Merged
sync : llama.cpp #981
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
* vulkan : do not use tensor->extra This patch allows using the Vulkan backend with the RPC backend as tensor->extra is no longer used. Ref: #8536 * Adapt GGML_VULKAN_CHECK_RESULTS to extra removal (llama/2) --------- Co-authored-by: 0cc4m <picard12@live.de>
sycl: initial cmake support of SYCL for AMD GPUs
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
Also: - metal : fix compute pass descriptor autorelease crash - ggml-backend : add device description to CPU backend - ggml: unify backend logging mechanism
slaren
reviewed
Oct 3, 2024
examples/mnist/mnist-common.h
Outdated
Comment on lines
59
to
71
| mnist_model(const std::string & backend_name) { | ||
| const size_t backend_index = ggml_backend_reg_find_by_name(backend_name.c_str()); | ||
| if (backend_index == SIZE_MAX) { | ||
| const ggml_backend_reg_t reg = ggml_backend_reg_by_name(backend_name.c_str()); | ||
| if (reg == nullptr) { | ||
| fprintf(stderr, "%s: ERROR: backend %s not found, available:\n", __func__, backend_name.c_str()); | ||
| for (size_t i = 0; i < ggml_backend_reg_get_count(); ++i) { | ||
| fprintf(stderr, " - %s\n", ggml_backend_reg_get_name(i)); | ||
| for (size_t i = 0; i < ggml_backend_reg_count(); ++i) { | ||
| fprintf(stderr, " - %s\n", ggml_backend_reg_name(ggml_backend_reg_get(i))); | ||
| } | ||
| exit(1); | ||
| } | ||
|
|
||
| fprintf(stderr, "%s: using %s backend\n", __func__, backend_name.c_str()); | ||
| backend = ggml_backend_reg_init_backend(backend_index, nullptr); | ||
|
|
||
| ggml_backend_t backend = ggml_backend_init_by_name(backend_name.c_str(), NULL); |
There was a problem hiding this comment.
The equivalent in the current API would be to find a device instead. Something like this:
mnist_model(const std::string & backend_name) {
ggml_backend_dev_t dev = ggml_backend_dev_by_name(backend_name.c_str());
if (reg == nullptr) {
fprintf(stderr, "%s: ERROR: backend %s not found, available:\n", __func__, backend_name.c_str());
for (size_t i = 0; i < ggml_backend_dev_count(); ++i) {
fprintf(stderr, " - %s\n", ggml_backend_dev_name(ggml_backend_dev_get(i)));
}
exit(1);
}
fprintf(stderr, "%s: using %s backend\n", __func__, backend_name.c_str());
ggml_backend_t backend = ggml_backend_dev_init(dev, NULL);There was a problem hiding this comment.
Thanks, can you also verify the changes in test-backend-buffer.cpp
There was a problem hiding this comment.
test-backend-buffer should also test all the devices, however the test is very outdated and doesn't really test anything useful at this point. It may be better to remove it entirely. I may try to create a more comprehensive test to ensure that backends are implementing the API correctly in the future.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
TODO: update examples