-
Notifications
You must be signed in to change notification settings - Fork 8.6k
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
Capture CUDA logging output #7298
Conversation
Most of the printfs there are during Debug builds - these are not a problem. |
Co-authored-by: slaren <[email protected]>
Co-authored-by: slaren <[email protected]>
Addresses logging capture issues reported here #5797.
This PR adds a CUDA logging callback similar to that in
ggml-metal.h
:llama.cpp/ggml-metal.h
Line 43 in 583fd6b
However, the issue OP has certainly experienced other uncaptured log messages. For example:
As evident methods in
ggml-alloc.c
, e.g.ggml_gallocr_reserve_n
, use directlystderr
. So that begs the question whether the internal logging inllama.cpp
should not be made commonly available, e.g. incommon.h
.