[fpc-devel] Experimentation: "Branch stitching"

[Martin Frb]

On 28/11/2022 16:19, J. Gareth Moreton via fpc-devel wrote:
> I admit I can be disorganised sometimes and lose documents, so I 
> apologise if you have sent them already and I mislaid them in my mess 
> of a directory tree.  Believe me though, I want to swallow all of this 
> up if it means squeezing out every cycle I can out of the generated 
> machine code!
>
> Curious to know... at which point did it become favourable to do a 
> 32-byte align rather than a 16-byte align on x86 processors? Should 
> the compiler start favouring 32-byte aligns for loops, say?

https://www.agner.org/optimize/optimizing_assembly.pdf

"11.3μop cache"

I couldn't find the 32byte align in that doc though. I must have picked 
that up elsewhere. (I think).
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freepascal.org/pipermail/fpc-devel/attachments/20221128/c7efa6ff/attachment.htm>