On Thu, 19 Feb 2004, Xia Hongtao wrote:

> So in current 2.5.STABLE4 version, just the first 4KB fs I/O was done by 
> asyncronous threads? If the content is larger then 4KB, the main squid thread
> will do the fs I/O remained ? 

The functions in disk.c is not at all used for aufs disk I/O. These are 
currently only used for log writes.

> I had view Squid-2.2, this version was actually async-io operated.Why you 
> changed this feature in next versions? Which version do you want to add this 
> feature back?

In Squid-2.3 the way cache I/O is performed is totally redone, adding
support for multiple different types of cache I/O in the same Squid
binary. Because of this the async-io layer was moved from disk.c to the
aufs store where it is needed.

Any I/O in aufs cache_dir type still uses async-io threads like in the old
days.

Any I/O in diskd cache_dir type uses the diskd slave process.

Regards
Henrik

Reply via email to