Re: [PATCH v3] xfs: allow read IO and FICLONE to run concurrently

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, Oct 12, 2023 at 01:34:34AM -0700, Christoph Hellwig wrote:
> On Wed, Oct 11, 2023 at 06:08:45PM -0700, Catherine Hoang wrote:
> > Clone operations and read IO do not change any data in the source file, so they
> > should be able to run concurrently. Demote the exclusive locks taken by FICLONE
> > to shared locks to allow reads while cloning. While a clone is in progress,
> > writes will take the IOLOCK_EXCL, so they block until the clone completes.
> 
> FYI, the first two lines are too long for the normal commit log format.
> 
> Can you provide a justification for the change, i.e. worksloads for
> which this matters, and what metrics are improved?

Catherine started with this,
https://lore.kernel.org/linux-xfs/8911B94D-DD29-4D6E-B5BC-32EAF1866245@xxxxxxxxxx/

and the rest of us whittled it down to the single patch you see here.
Sections 1-2 are still relevant; S3 was the path not taken.

The commit message should Link: to that.

--D



[Index of Archives]     [XFS Filesystem Development (older mail)]     [Linux Filesystem Development]     [Linux Audio Users]     [Yosemite Trails]     [Linux Kernel]     [Linux RAID]     [Linux SCSI]


  Powered by Linux