Received-SPF: pass (google.com: domain of linux-ext4+bounces-1441-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
From: Phillip Susi <phill@thesusis.net>
To: Theodore Ts'o <tytso@mit.edu>
Cc: linux-ext4@vger.kernel.org
Subject: Re: [PATCH] [RFC] Fix jbd2 to stop waking up sleeping disks on sync
In-Reply-To: <20240229145858.GE272762@mit.edu>
References: <20240227212546.110340-1-phill@thesusis.net>
 <20240229080759.GB57093@macsyma.local> <87edcv1h94.fsf@vps.thesusis.net>
 <20240229145858.GE272762@mit.edu>
Date: Thu, 29 Feb 2024 15:19:50 -0500
Message-ID: <877cin58gp.fsf@vps.thesusis.net>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain

"Theodore Ts'o" <tytso@mit.edu> writes:

> Because no metadata changed, jbd2 will not even *start* a (jbd2)
> transaction as a result of that write (overwrite) to an already
> allocated data block..  Since it didn't start a jbd2 transaction for
> this file system operation, there's no reason to force a jbd2
> transaction to close.
>
> (Note: this could because there *is* no currently existing open
> transaction, or there might be a currently open transaction, but it's
> not relevent to the activity associated with the file descriptor being
> fsync'ed.)
>
> This is a critical performance optimization, because for many database
> workloads, which are overwriting existing blocks and using
> fdatasync(2), there is no reason to force a jbd2 transaction commit
> for every single fdatasync(2) issued by the database.  However, we
> still need to send a cache flush operation so that the data block is
> safely persistend onto stable storage.

So maybe what ext4's sync_fs needs to know is whether ANY writes have
been done since the last transaction committed?  Is there a way to know
that?  As long as NOTHING has been written since the last commit, then
there is no need to issue a flush.