-
Notifications
You must be signed in to change notification settings - Fork 334
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
binary intrinsic declarations #205
Conversation
Current coverage is 91.93% (diff: 100%)@@ master #205 diff @@
==========================================
Files 32 32
Lines 4574 4585 +11
Methods 0 0
Messages 0 0
Branches 319 322 +3
==========================================
+ Hits 4204 4215 +11
Misses 300 300
Partials 70 70
|
Any word on this? Is anyone aware of a workaround to use in the meantime? |
Oops, sorry. I hadn't noticed this PR. |
@@ -172,10 +178,13 @@ def _error(): | |||
fnty = types.FunctionType(tys[0], tys*2) |
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.
llvm.pow
is already handled here (note you can also pass the function type fnty
explicitly).
I understand better what's going on now and my patch isn't quite right. The solution for me is just to avoid the However, there's still something to fix here I feel:
crashes because the name mangling isn't correct:
Or is the intended way to call function just exactly what's needed to match the llvm intrinsic full name? I.e.:
In this case I can just update the docstring for |
Right. For |
Okay, thanks @pitrou. I understand the behaviour of |
Fixes binary intrinsic (e.g.
llvm.maxnum
) IR generation.Note: I haven't tested e.g.
llvm.convert.from.fp16
.