Tony Nguyen c8bbf7c00b Improve write performance by using dmu_read_by_dnode()
In zfs_log_write(), we can use dmu_read_by_dnode() rather than
dmu_read() thus avoiding unnecessary dnode_hold() calls.

We get a 2-5% performance gain for large sequential_writes tests, >=128K
writes to files with recordsize=8K.

Testing done on Ubuntu 18.04 with 4.15 kernel, 8vCPUs and SSD storage on
VMware ESX.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Tony Nguyen <tony.nguyen@delphix.com>
Closes #9156
2019-08-15 17:36:24 -06:00
2019-07-16 17:22:31 -07:00
2019-07-30 10:06:09 -07:00
2019-06-24 09:32:47 -07:00
2017-11-13 09:18:18 -08:00
2018-05-29 16:00:33 -07:00
2019-04-30 10:58:45 -07:00
2018-05-29 16:00:33 -07:00
2018-05-29 16:00:33 -07:00
2019-06-24 09:32:47 -07:00
2019-05-21 11:11:41 -07:00
2018-09-18 12:03:47 -07:00
2018-05-29 16:00:33 -07:00
2018-05-29 16:00:33 -07:00

img

ZFS on Linux is an advanced file system and volume manager which was originally developed for Solaris and is now maintained by the OpenZFS community.

codecov coverity

Official Resources

Installation

Full documentation for installing ZoL on your favorite Linux distribution can be found at our site.

Contribute & Develop

We have a separate document with contribution guidelines.

Release

ZFS on Linux is released under a CDDL license.
For more details see the NOTICE, LICENSE and COPYRIGHT files; UCRL-CODE-235197

Supported Kernels

  • The META file contains the officially recognized supported kernel versions.
Description
freebsd with flexible iflib nic queues
Readme 2.6 GiB
Languages
C 60.1%
C++ 26.1%
Roff 4.9%
Shell 3%
Assembly 1.7%
Other 3.7%