> > I have a script (attached). At one point it tries to do sync... That
> > sync take a long time, with disk mostly unused.
>
> When doing (say) ext2-on-loop-on-ext2 you should always ensure
> that the blocksize for the topmost filesystem is the same as
> the one underneath. So probably you wanted `mkfs.ext2 -b 4096'.
>
> If you have a 1k blocksize filesystem loop-mounted on a 4k blocksize
> filesystem, every write of a 1k block requires a read of the underlying
> 4k block. Which is excrutiatingly slow.
Oh, but I *want* to do 1k filesystem test!
Performance seems okay in 2.4.19-pre?aa?.... And even 2.4.18 is
slightly faster. [BTW it is fully cached (100MB test on 256MB machine)
so 4k reads should not be such a big problem.]
Pavel
-- Casualities in World Trade Center: ~3k dead inside the building, cryptography in U.S.A. and free speech in Czech Republic.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/