Skip to content
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

Generate SVE for 80bit load/stores when possible #4166

Open
wants to merge 3 commits into
base: main
Choose a base branch
from

Conversation

pmatos
Copy link
Collaborator

@pmatos pmatos commented Nov 20, 2024

Fixes #4126

@pmatos
Copy link
Collaborator Author

pmatos commented Nov 20, 2024

Probably not a big win in practice. A single 80-bit store required 2 stores (64 + 16). Now, we require three instructions: mov + whilelt + st1b. You need three 80-bit stores in a block to get to a draw instruction-wise. The next step would be not to assemble the predicate register every time, which I will do next, but we'll still require in practice at least three stores per block for it to "win" instruction-wise.

@pmatos
Copy link
Collaborator Author

pmatos commented Nov 20, 2024

Converting to draft, so it's not merged by mistake.

@pmatos pmatos marked this pull request as draft November 20, 2024 09:55
pmatos added a commit to pmatos/FEX that referenced this pull request Nov 22, 2024
In preparation for FEX-Emu#4166 which should improve on these results.
@pmatos pmatos changed the title Generate SVE for 80bit stores when possible Generate SVE for 80bit load/stores when possible Dec 1, 2024
pmatos added a commit to pmatos/FEX that referenced this pull request Dec 1, 2024
In preparation for FEX-Emu#4166 which should improve on these results.
pmatos added a commit to pmatos/FEX that referenced this pull request Dec 1, 2024
pmatos added a commit to pmatos/FEX that referenced this pull request Dec 1, 2024
In preparation for FEX-Emu#4166 which should improve on these results.
pmatos added a commit to pmatos/FEX that referenced this pull request Dec 1, 2024
In preparation for FEX-Emu#4166 which should improve on these results.
pmatos added a commit to pmatos/FEX that referenced this pull request Dec 1, 2024
pmatos added a commit to pmatos/FEX that referenced this pull request Dec 1, 2024
In preparation for FEX-Emu#4166 which should improve on these results.
pmatos added a commit to pmatos/FEX that referenced this pull request Dec 2, 2024
In preparation for FEX-Emu#4166 which should improve on these results.
pmatos added a commit to pmatos/FEX that referenced this pull request Dec 2, 2024
In preparation for FEX-Emu#4166 which should improve on these results.
pmatos added a commit to pmatos/FEX that referenced this pull request Dec 2, 2024
In preparation for FEX-Emu#4166 which should improve on these results.
pmatos added a commit to pmatos/FEX that referenced this pull request Dec 2, 2024
@pmatos pmatos marked this pull request as ready for review December 2, 2024 17:50
@pmatos
Copy link
Collaborator Author

pmatos commented Dec 2, 2024

This is almost ready - the bit missing is really that the predicate register is not yet being cached. I thought RA would take care of this but... apparently there's still some magic missing.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

80-bit x87 Loadstores can use SVE masked loadstores
2 participants