Re: sdirno type?

On a second thought, no this is not acceptable. Changing it to only 4
bits would only leave room for max 8 swapdirs.

Why: sdirno is signed, wasting one whole bit only to be able to store
-1. And fixing sdirno to be unsigned is not a Squid-2.5 task I think.

Regards
Henrik

Henrik Nordstrom wrote:
>
> Fine by me.
>
> Note: in the long term we should be able to get rid of all those _status
> bits from the StoreEntry. In theory it should be possible to move all of
> them to the MemObject as active StoreObject's without a MemObject is
> pretty much all of the same kind (plain cached object).
>
> Regards
> Henrik
>

> > + sdirno swap_dirn:4;
> > mem_status_t mem_status:3;
> > ping_status_t ping_status:3;
> > store_status_t store_status:3;
> > swap_status_t swap_status:3;
> > };
> >
> > Such layout keeps us at 48 bytes/sentry, the only limitation is
> > max 16 sdirs. Is it acceptable compromise?
> > I wonder how many people need more than that number of sdirs?
> > Perhaps make this compiletime option. I understand that 32bit
> > sfileno is currently requirement for using coss best.
> >
> > ------------------------------------
> > Andres Kroonmaa <andre@online.ee>
> > CTO, Microlink Online
> > Tel: 6501 731, Fax: 6501 725
> > Pärnu mnt. 158, Tallinn,
> > 11317 Estonia
Received on Thu Oct 04 2001 - 13:50:48 MDT