> Purpose of this patch is of course not to fill in the datablocks with zeroes.
> The purpose (as applied to reiserfs) is to fill indirect data pointers (that is - pointers to real data blocks)
> with zeroes (and to organize proper in-tree data structure for such pointers).
> As of now such organization and zero-filling is done on a lazy manner at disk-flushing time.
> Unfortunatelly this leads to races in the code.
> I do not know why parts of this code can be needed by other filesystem and why Al Viro put it in generic VFS
> code. (but he can comment on it, I think)
I could, if I would remember doing that... ;-/
Seriously, it looks like a half-arsed and very old attempt to do common
expanding truncate() for no-holes filesystems. BTW, these days rlimit checks
are done by vmtruncate().
-
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/