RE: [PATCH RFC] introduce ioctl to completely invalidate page cache

[Thanos Makatos]

Thanks, Jan, it's much clearer now what I need to do.

>   So most notably they want the ioctl to work not only for block devices but also for any regular file. That's easily doable - you just call
> filemap_write_and_wait() and invalidate_inode_pages2() in the ioctl handler for regular files.

All right, so I need to find out how I can direct the new ioctl to
file-systems as well. I thought that I could get away with it by simply looking
at which file-systems have the same block device as the one to which the ioctl
is directed, but IIUC this doesn't make sense as NFS doesn't use a block device
at all.

> Also they wanted to be able to specify a range of a mapping to invalidate -
> that's easily doable as well. Finally they wanted a 'flags' argument so you can
> additionally ask fs to invalidate also some metadata. How invalidation is done
> will be a fs specific thing and for now I guess we don't need to go into
> details. NFS guys can sort that out when they decide to implement it.

So after I've figured out how to direct this new ioctl to a file-system, I need
to understand out how to invalidate a specific range of data. I will gracefully
fail metadata invalidation operations with EOPNOTSUPP or ENOSYS.

> So in the beginning we can just have u64 flags argument and in it a single
> 'INVAL_DATA' flag meaning that invalidation of data in a given range is
> requested. Later NFS guys can add further flags.

OK that I can do.

I suppose we'll always fail metadata invalidation operations when the target
of the ioctl is a block device.

Thanks
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/