2.3.40 loop-device-behaviour

[Guest section DW]

On Wed, Jan 26, 2000 at 12:27:38PM -0800, Pete Zaitcev wrote:

> Consider the following script:
> 
> genromfs -a 512 -f flash.root.image -d proto
> mount -t romfs -o ro,loop=/dev/loop0 flash.root.image /mnt
> silo -C /mnt/etc/silo.conf -J flash.total -i /mnt/boot/ieee32 \
>  -b /mnt/boot/second.b
> umount /mnt
> 
> It works only once. On the second time silo gets stale data.
> My workaround is to add "cat /wad/mp3/*.mp3 > /dev/null"
> right before the mount.
> 
> I talked to Alan and he says that somebody needs to do
> ioctl(fd, BLKFLSBUF), presumably losetup. I have not tested
> his suggestion yet.

Yes. These days you can do that from the command line:

	blockdev --flushbufs device


---
"Linux doesn't have inconsistent caches"

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo en vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/