| 2020-02-11 18:40:11 |
Matthew Ahrens |
bug |
|
|
added bug |
| 2020-02-11 18:40:11 |
Matthew Ahrens |
attachment added |
|
lspci-vnvn.log https://bugs.launchpad.net/bugs/1862822/+attachment/5327397/+files/lspci-vnvn.log |
|
| 2020-02-11 19:00:07 |
Ubuntu Kernel Bot |
linux (Ubuntu): status |
New |
Incomplete |
|
| 2020-02-11 19:26:34 |
Matthew Ahrens |
linux (Ubuntu): status |
Incomplete |
Confirmed |
|
| 2020-02-12 12:36:57 |
Mauricio Faria de Oliveira |
bug |
|
|
added subscriber Mauricio Faria de Oliveira |
| 2020-02-12 12:59:27 |
Mauricio Faria de Oliveira |
removed subscriber Mauricio Faria de Oliveira |
|
|
|
| 2020-02-12 12:59:29 |
Mauricio Faria de Oliveira |
bug |
|
|
added subscriber Mauricio Faria de Oliveira |
| 2020-02-12 13:32:56 |
Heitor Alves de Siqueira |
bug |
|
|
added subscriber Heitor Alves de Siqueira |
| 2020-02-12 16:42:34 |
Seth Forshee |
bug |
|
|
added subscriber Seth Forshee |
| 2020-02-13 03:25:36 |
Terry Rudd |
bug |
|
|
added subscriber Terry Rudd |
| 2020-02-13 05:44:18 |
Matthew Ahrens |
tags |
|
apport-collected bionic uec-images |
|
| 2020-02-13 05:44:19 |
Matthew Ahrens |
description |
Apologies if this is not the right forum to raise this concern. If not please point me to a better place to have this discussion.
The 5.3 linux kernel added a new feature which allows pages to be zeroed when allocating or freeing them: init_on_alloc and init_on_free. The init_on_alloc flag is enabled by default on the Ubuntu 18.04 HWE kernel.
Linux kernel commit: https://github.com/torvalds/linux/commit/6471384af2a6530696fc0203bafe4de41a23c9ef
ZFS allocates and frees pages frequently (via the ABD structure), e.g. for every disk access. The additional overhead of zeroing these pages is significant. For example, I measured a ~40% regression in performance of an uncached "zfs send ... >/dev/null". While this is a specific workload, this is a general problem because whenever we read from disk we are going to allocate pages, unnecessarily zero them, and then immediately fill them with the data from disk.
I don't see any mechanism for ZFS to request that pages *not* be zeroed when they are allocated.
This performance regression has been noted by users of ZFS on Ubuntu: https://github.com/zfsonlinux/zfs/issues/9910
I'd like to request that the default in Ubuntu be changed to init_on_alloc=0, so that users of ZFS on Ubuntu do not experience this performance regression.
(FYI, my employer, Delphix, has worked around this by changing the default in our product, but my concern is that the vast majority of Ubuntu users will simply experience this as "Ubuntu HWE is much slower than before".)
$ cat /proc/version_signature
Ubuntu 5.3.0-28.30~18.04.1-generic 5.3.13 |
Apologies if this is not the right forum to raise this concern. If not please point me to a better place to have this discussion.
The 5.3 linux kernel added a new feature which allows pages to be zeroed when allocating or freeing them: init_on_alloc and init_on_free. The init_on_alloc flag is enabled by default on the Ubuntu 18.04 HWE kernel.
Linux kernel commit: https://github.com/torvalds/linux/commit/6471384af2a6530696fc0203bafe4de41a23c9ef
ZFS allocates and frees pages frequently (via the ABD structure), e.g. for every disk access. The additional overhead of zeroing these pages is significant. For example, I measured a ~40% regression in performance of an uncached "zfs send ... >/dev/null". While this is a specific workload, this is a general problem because whenever we read from disk we are going to allocate pages, unnecessarily zero them, and then immediately fill them with the data from disk.
I don't see any mechanism for ZFS to request that pages *not* be zeroed when they are allocated.
This performance regression has been noted by users of ZFS on Ubuntu: https://github.com/zfsonlinux/zfs/issues/9910
I'd like to request that the default in Ubuntu be changed to init_on_alloc=0, so that users of ZFS on Ubuntu do not experience this performance regression.
(FYI, my employer, Delphix, has worked around this by changing the default in our product, but my concern is that the vast majority of Ubuntu users will simply experience this as "Ubuntu HWE is much slower than before".)
$ cat /proc/version_signature
Ubuntu 5.3.0-28.30~18.04.1-generic 5.3.13
---
ProblemType: Bug
AlsaDevices:
total 0
crw-rw---- 1 root audio 116, 1 Feb 11 18:34 seq
crw-rw---- 1 root audio 116, 33 Feb 11 18:34 timer
AplayDevices: Error: [Errno 2] No such file or directory: 'aplay': 'aplay'
ApportVersion: 2.20.9-0ubuntu7.9
Architecture: amd64
ArecordDevices: Error: [Errno 2] No such file or directory: 'arecord': 'arecord'
AudioDevicesInUse: Error: command ['fuser', '-v', '/dev/snd/seq', '/dev/snd/timer'] failed with exit code 1:
DistroRelease: Ubuntu 18.04
IwConfig: Error: [Errno 2] No such file or directory: 'iwconfig': 'iwconfig'
Lsusb: Error: [Errno 2] No such file or directory: 'lsusb': 'lsusb'
MachineType: VMware, Inc. VMware Virtual Platform
NonfreeKernelModules: zfs zunicode zavl icp zcommon znvpair
Package: linux (not installed)
PciMultimedia:
ProcEnviron:
TERM=xterm
PATH=(custom, no user)
LANG=en_US.UTF-8
SHELL=/bin/bash
ProcFB: 0 svgadrmfb
ProcKernelCmdLine: BOOT_IMAGE=/ROOT/username.90dy59N/root@/boot/vmlinuz-5.3.0-28-generic root=ZFS=rpool/ROOT/username.90dy59N/root ro console=tty0 console=ttyS0,38400n8 mitigations=off ipv6.disable=1 elevator=noop crashkernel=256M,high crashkernel=256M,low init_on_alloc=0
ProcVersionSignature: Ubuntu 5.3.0-28.30~18.04.1-generic 5.3.13
RelatedPackageVersions:
linux-restricted-modules-5.3.0-28-generic N/A
linux-backports-modules-5.3.0-28-generic N/A
linux-firmware 1.173.14
RfKill: Error: [Errno 2] No such file or directory: 'rfkill': 'rfkill'
Tags: bionic uec-images
Uname: Linux 5.3.0-28-generic x86_64
UnreportableReason: This report is about a package that is not installed.
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: linux-staff root
WifiSyslog:
_MarkForUpload: False
dmi.bios.date: 12/12/2018
dmi.bios.vendor: Phoenix Technologies LTD
dmi.bios.version: 6.00
dmi.board.name: 440BX Desktop Reference Platform
dmi.board.vendor: Intel Corporation
dmi.board.version: None
dmi.chassis.asset.tag: No Asset Tag
dmi.chassis.type: 1
dmi.chassis.vendor: No Enclosure
dmi.chassis.version: N/A
dmi.modalias: dmi:bvnPhoenixTechnologiesLTD:bvr6.00:bd12/12/2018:svnVMware,Inc.:pnVMwareVirtualPlatform:pvrNone:rvnIntelCorporation:rn440BXDesktopReferencePlatform:rvrNone:cvnNoEnclosure:ct1:cvrN/A:
dmi.product.name: VMware Virtual Platform
dmi.product.version: None
dmi.sys.vendor: VMware, Inc. |
|
| 2020-02-13 05:44:19 |
Matthew Ahrens |
attachment added |
|
CRDA.txt https://bugs.launchpad.net/bugs/1862822/+attachment/5327846/+files/CRDA.txt |
|
| 2020-02-13 05:44:21 |
Matthew Ahrens |
attachment added |
|
CurrentDmesg.txt https://bugs.launchpad.net/bugs/1862822/+attachment/5327847/+files/CurrentDmesg.txt |
|
| 2020-02-13 05:44:22 |
Matthew Ahrens |
attachment added |
|
Lspci.txt https://bugs.launchpad.net/bugs/1862822/+attachment/5327848/+files/Lspci.txt |
|
| 2020-02-13 05:44:24 |
Matthew Ahrens |
attachment added |
|
ProcCpuinfoMinimal.txt https://bugs.launchpad.net/bugs/1862822/+attachment/5327849/+files/ProcCpuinfoMinimal.txt |
|
| 2020-02-13 05:44:25 |
Matthew Ahrens |
attachment added |
|
ProcInterrupts.txt https://bugs.launchpad.net/bugs/1862822/+attachment/5327850/+files/ProcInterrupts.txt |
|
| 2020-02-13 05:44:26 |
Matthew Ahrens |
attachment added |
|
ProcModules.txt https://bugs.launchpad.net/bugs/1862822/+attachment/5327851/+files/ProcModules.txt |
|
| 2020-02-13 05:44:29 |
Matthew Ahrens |
attachment added |
|
UdevDb.txt https://bugs.launchpad.net/bugs/1862822/+attachment/5327852/+files/UdevDb.txt |
|
| 2020-02-13 17:35:28 |
Jay Vosburgh |
bug |
|
|
added subscriber Jay Vosburgh |
| 2020-05-05 14:58:13 |
Richard Laager |
bug |
|
|
added subscriber Colin Ian King |
| 2020-05-25 22:45:30 |
Nate Crawford |
bug |
|
|
added subscriber Nate Crawford |
| 2020-12-25 02:22:36 |
Andrew Berry |
bug |
|
|
added subscriber Andrew Berry |
| 2021-01-04 23:20:22 |
Ryan C. Underwood |
bug |
|
|
added subscriber Ryan C. Underwood |
| 2021-02-21 21:25:52 |
Christopher Teutsch |
bug |
|
|
added subscriber Christopher Teutsch |
| 2022-03-06 07:29:33 |
Joel Low |
bug |
|
|
added subscriber Joel Low |
| 2022-06-06 14:21:36 |
James |
bug |
|
|
added subscriber James |
| 2023-11-10 07:20:52 |
Steve Beattie |
bug |
|
|
added subscriber Steve Beattie |
