Re: still more 2.5.11 comments

[Bruce Lilly]

John Millaway wrote:

> > And you have apparently failed to account for the presence of the
> > generated header files in the disk buffer cache.
>
> The point was to ballpark the addition of one thousand headers. All 57MB of
> headers certainly did not reside in the disk cache.

Why not?  Your disk statistics seem to show a 128 MB buffer:
> /dev/hda1:
>  Timing buffer-cache reads:   128 MB in  2.51 seconds =51.00 MB/sec


[...]
> In any case, this debate is now moot. You'll be relieved to know that the
> latest revision of flex omits the user actions, dfa tables, and function
> declarations from the generated headers. The resulting headers should be of
> constant size, approximately 23KB.  -John

Umm, *some* function declarations are obviously desired; that's what
started this thread.  Anyway, that's all the 2.5.11 comments that I have
-- I look forward to checking out the relevant changes in 2.5.12.

Best regards,
  Bruce Lilly