vgvassilev wrote:

> > Can you summarize the before and after file stat counts in the description 
> > of this PR?
> 
> Done so in the "Filesystem access in unchanged" section.
> 
> > Also, 40% is the max empirical finding, however, what's our theoretical 
> > expectation? Nominally one would think that if a pcm is dominated by copied 
> > content the reuse will yield much higher numbers... That is, is there any 
> > performance gains still left on the table beyond what we see currently?
> 
> Clarified this in the "How high can this go section".
> 
> @vgvassilev

Thank you. I am surprised that the peak memory went up. Do we understand why 
and can we minimize it?

https://github.com/llvm/llvm-project/pull/209795
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Reply via email to