-
Notifications
You must be signed in to change notification settings - Fork 3.7k
[microTVM] Refactor uTVM to microTVM #8283
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
Conversation
|
Hi @mehrdadh Since I joined the project I was inclined to use (and tried to stick with) the form "uTVM". I have no strong feelings other than really sticking with only one form, hence I think that PR makes sense indeed (ppl might even write mTVM sometimes, which doesn't look good at all). I'm just wondering if I missed any discussion or thread about favoring microTVM. Either way I think using microTVM is the right choice, because uTVM is often written also using the mu symbol (Greek), which is confusing sometimes and might be difficult to type on some consoles. On Linux I also prefer typing "micro" instead of CTRL-Shift-u-b-5 :) |
|
@gromero thanks for the feedback! |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I was thinking about something similar to this PR when I read @areusch's PR with the MLF documentation. I agree with @gromero that sticking with one form is the important bit here.
microTVM looks great to me, because it is plain and clear on what it means. It is also already in use in many parts of the docs.
There are these places also that come to mind that need to be changed (from µTVM to microTVM):
python/tvm/micro/contrib/zephyr.pypython/tvm/driver/tvmc/runner.pypython/tvm/driver/tvmc/compiler.py
In terms of the need or not for an RFC, I think the microTVM spelling is already in so many parts of the live documentation (e.g. https://tvm.apache.org/docs/microtvm/index.html), that makes this PR basically a fix PR, rather than an introductory PR. Based on that, I don't think we need an RFC for this.
Thanks for doing this work @mehrdadh!
Yeah, thanks for proposing that change.
I asked just out of curiosity. I thought of a change like that just after I joined the project so I was wondering if I missed anything recently in the open so I could check it out curiosity. I really don't think it's necessary a RFC for that change. |
|
This is a great improvement @mehrdadh, I agree One small thing that follows on from @leandron's comment is that there are also some |
|
@leandron @Mousius thanks for pointing out those files. I fixed all of them.
We could also refactor these if someone with contributor access could help in renaming these PRs. |
|
cc @hogepodge |
areusch
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
thanks @mehrdadh i agree this makes more sense.
* refactor * rename utvm_rpc_server.h * rename file * rename * rename file * rename * variables * directories * format * trigger * more refactor * one more * format
* refactor * rename utvm_rpc_server.h * rename file * rename * rename file * rename * variables * directories * format * trigger * more refactor * one more * format
This PR: