Re: HFS Patch 15 out

[Szakacsits Szabolcs]

On Mon, 27 Sep 2004, K.G. wrote:

> I eventually identified the bottleneck, which stands in linux.c, flush_cache,
> called but the sync function of Parted:
> There's a loop which does an BLKFLSBUF ioctl 

Nowadays BLKFLSBUF is useless, it doesn't do anything (should flush and
invalidate buffers). To be honest, I even can't guess why parted uses it
instead of sync. It just seems wrong.

> on every umounted partitions of the disk, and that take (umount test
> and ioctl) an amount of time that is just too high for my high use of
> syncs.

Umount test? Ouch. Every partitions? Ouch again. I can't get this part
either. Perheps you're using a wrong API?
 
> So here this is a 29x real time speedup and a 930x CPU time speedup on
> my test system :p
> 
> I would like to know what BLKFLSBUF does exactly (though I don't think this
> is what take the more time, I think it's the mount test, but I havn't bench
> it separatly yet) ?

I also think the mount test slows things done. I even can't get why you
need it.

> (and as a more general question, does someone knows a good documentation
> about any ioctl...)

Kernel source ;-)
 
        Szaka