mån 2003-05-26 klockan 10.05 skrev Evgeny Kotsuba:
> I don't understood what you wrote about :-/
> I see only bad change in 2.5s3 vs 2.5s2(or s1?)
>
> http_hdr_owner_type
> ho = hdr->owner;
> assert(hdr);
> ho = hdr->owner;
>
> If hdr is NULL, then what will be in first ho = ?
> And why double code for initialization of variable ho is better ?
What I am saying is that there is no changes. The code has been as bad
since when http_hdr_owner_type was introduced more than 4 years ago. If
you have a copy of the code looking better then this code is not the
official Squid source code but a modified copy.
This piece of code is already cleaned up in Squid-3. I see no reason why
touching it in Squid-2.5 as it in worst case is only a microscopic
performance issue and should be optimized away by the compiler when -O1
or higher is used.
Regards
Henrik
-- Henrik Nordstrom <hno@marasystems.com> MARA Systems ABReceived on Mon May 26 2003 - 03:09:37 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:19:57 MST