Re: ZSH performance regression in 5.8.1.2-test

[Stephane Chazelas <stephane@xxxxxxxxxxxx>]

2022-04-26 09:31:55 +0100, Peter Stephenson:
[...]
> Not sure how much difference it makes but to avoid additional system calls
> we could cache the result, invalidating it when we open a new file.  (We
> could even cache whether a seek unexpectedly failed, but that might be
> getting too paranoid.)
[...]

FYI, see https://github.com/att/ast/issues/15 for the kind of
issue you can find with ksh93 caching read data.

IMO, if the point is make sure we leave the position within the
input past the delimiter of the line we have just read, so that
other things can read it if they want, then it's wrong to cache,
as those other things could just as well overwrite things.

It's also probably not going to get you that much of a benefit
as the OS will have cached the data as well.

-- 
Stephane