-
Notifications
You must be signed in to change notification settings - Fork 4.9k
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
[JIT] ARM64 - Optimize x < 0
and x >= 0
#83176
Conversation
Tagging subscribers to this area: @JulieLeeMSFT, @jakobbotsch, @kunalspathak Issue DetailsAddresses the first case of #43440 As for handling
|
x < 0
and x >= 0
Co-authored-by: Jakob Botsch Nielsen <Jakob.botsch.nielsen@gmail.com>
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.
LGTM with one fix
I'm going to do another round for this optimization, filed #83562 to track it. Therefore, I will go ahead and merge this one. |
Description
#43440
As for handling
GT_GE
, I do not think it is worth doing the optimization in the non-conditional case. It would requireneg
and alsr
which I believe would be slower than just thecmp
andcset
. According to A78, the pair ofcmp
andcset
instructions are accelerated through instruction fusion - so it's probably only worth replacing them if we can do it in just one instruction.Acceptance Criteria
x < 0
andx >= 0
#83562