-
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
Add Float16 support? #30
Comments
Would it be enough to add Lines 26 to 29 in 03f2429
|
That seems to work on A64FX (UPDATE: x86 via LLVM's half yields identical results) julia> @count_ops run_model(Float16,Ndays=5,scale=64)
Starting ShallowWaters on Wed, 14 Apr 2021 06:57:44 without output.
60% Integration done in 1min, 24s.
Flop Counter: 221063563 flop
┌────────┬──────────┬─────────┬─────────┐
│ │ Float16 │ Float32 │ Float64 │
├────────┼──────────┼─────────┼─────────┤
│ muladd │ 0 │ 0 │ 333040 │
│ add │ 91734565 │ 0 │ 72271 │
│ sub │ 22795383 │ 0 │ 166263 │
│ mul │ 92043896 │ 5575128 │ 244707 │
│ div │ 2580653 │ 0 │ 81624 │
│ rem │ 0 │ 0 │ 48 │
│ abs │ 0 │ 0 │ 59320 │
│ neg │ 4990000 │ 0 │ 48775 │
│ sqrt │ 0 │ 0 │ 4850 │
└────────┴──────────┴─────────┴─────────┘ PR coming! |
Closed
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Running the following on an A64FX with native Float16 support yields
The second executes every Float32 in Float16, but there are not counted. I'd be happy to test this on A64FX if there's interest to include Float16 support.
The text was updated successfully, but these errors were encountered: