Does UFS2 send BIO_FLUSH to GEOM when update metadata (with softupdates)?

[Kostik Belousov]

On Wed, Nov 23, 2011 at 11:00:26PM +0400, Lev Serebryakov wrote:
> Hello, Freebsd-fs.
> 
>   Does UFS2 with softupdates (without journal) issues BIO_FLUSH to
> GEOM layer when it need to ensure consistency on on-disk metadata?
No. Softupdates do not need flushes.

Read about SU in the design and implementation, or standalone article
about it.
> 
>   Does UFS2 with SU+J issues this command?
Again, no.

> 
>   geom_raid5 respects this command, but when write log is enabled, it
> produces tons of "Unexpected softupate inconsistency" errors on crash.
> It seems, that there is no BIO_FLUSH and writes, which should be
> synchronous (metadata updates) isn't :(
You are making wrong conclusions from the false assumptions.

The only requirement of the SU is that writes reported as done by disk
driver are indeed safely landed in the involatile storage.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 196 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-fs/attachments/20111123/6edaf058/attachment.pgp