On 12/06/2013 12:23, Dominique Pellé wrote:
Bram Moolenaar wrote:
After the patch, starting nerdtree in VIM's src is ~35% faster (1.7s vs
2.6s) and the CRT's locale handling functions no longer appear in the
profile (and I warmed the file cache to remove disk reading from the
times). In general this should help speed up longer running VIML scripts.
Yeah, we've had problems with library functions before.
I think using a table is often slower than simple compares, because a
table lookup has to access memory, while a compare is local inside the
CPU. Especially for ASCII_ISLOWER and ASCII_ISUPPER, it's just two
compares. Changing VIM_ISDIGIT() this way is likely to make it slower.
ASCII_ISALNUM() requires six compares, perhaps a table is faster then?
You can half the number of comparisons when doing range checking.
Instead of doing 2 comparisons...
if (c >= '0' && c <= '9')
You can do it in one comparison with this trick:
if ((unsigned)c - '0' < 10)
Yup, works for digit, lower, and upper. alnum not so well.
And that can be faster than tables lookup because table access
thrashes the L1 cache.
That would be a "it depends" wouldn't it? How much other memory is
being accessed in the loop? How it maps through to the L1 cache, and
how well the CPU's read ahead can pre-empt the access, hopefully from L2
as well.
Either way, as long as I keep the large speed up processing VIML on
Windows I'll be happy.
Mike
--
Yoghurt the Great, Yoghurt the All Powerful? No just plain yoghurt.
--
--
You received this message from the "vim_dev" maillist.
Do not top-post! Type your reply below the text you are replying to.
For more information, visit http://www.vim.org/maillist.php
---
You received this message because you are subscribed to the Google Groups "vim_dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.
