28 Commits

Author SHA1 Message Date
Thomas Lamprecht 020d1fc78c Revert "buildsys: switch upload dist over to buster"
this was intended for the master branch, argh..

This reverts commit bdbccd235a.

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2019-05-24 11:13:43 +02:00
Thomas Lamprecht bdbccd235a buildsys: switch upload dist over to buster
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2019-05-23 18:21:21 +02:00
Thomas Lamprecht 696f8e69e2 bump version to 0.7.13-pve1~bpo2
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2019-03-27 10:29:30 +01:00
Thomas Lamprecht f8d2e88078 remove conflict with insserv (<< 1.18)
we do not use or ship initscripts anyway and support only systemd

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2019-03-27 09:55:28 +01:00
Thomas Lamprecht 5e3e80687e fixup spl und zfs changelog entries
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2019-03-08 06:53:42 +01:00
Stoiko Ivanov d968500bc7 bump version to 0.7.13-pve1~bpo1
Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
2019-03-08 06:46:25 +01:00
Stoiko Ivanov 5440c7266a add vdev_id.conf.scsi.example to examples
(adapted from salsa.d.o 7ca194b9)

Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
2019-03-08 06:46:25 +01:00
Stoiko Ivanov 00430d087a rebase patches on top of zfs-0.7.13
rebase done with import-patchqueue/export-patchqueue

Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
2019-03-08 06:46:25 +01:00
Stoiko Ivanov c3ea9788ad update ZFS to 0.7.13
Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
2019-03-08 06:46:25 +01:00
Stoiko Ivanov d8740a92c1 spl: drop patch already contained in 0.7.13
Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
2019-03-08 06:46:25 +01:00
Stoiko Ivanov 256c6c0238 update SPL to 0.7.13
Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
2019-03-08 06:46:25 +01:00
Stoiko Ivanov f7305de3c0 build: use dpkg-parsechangelog
Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
2019-03-08 06:46:25 +01:00
Fabian Grünbichler fa456d6c4b build: add kernel target
to extract the patched sources for module building

Reviewed-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Tested-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2019-02-27 13:46:29 +01:00
Fabian Grünbichler 3d885de40c build: add dsc target
and rework build directory preparation accordingly.

Reviewed-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Tested-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2019-02-27 13:46:28 +01:00
Fabian Grünbichler 4a816df53e debian: switch to DH 10
Reviewed-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Tested-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2019-02-27 13:46:28 +01:00
Fabian Grünbichler f3f5ce5d7f submodules: remove old submodule references
Reviewed-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Tested-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2019-02-27 13:46:28 +01:00
Fabian Grünbichler 2a904e377e build: update Makefile for new structure
Reviewed-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Tested-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2019-02-27 13:46:27 +01:00
Fabian Grünbichler ac43a7be1f update README
to reference Debian's git repositories on salsa

Reviewed-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Tested-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2019-02-27 13:46:27 +01:00
Fabian Grünbichler dee2ef0e31 spl/debian: add packaging files
based on Debian's packaging work, but simplified:
- no DKMS

and remove old patches which were based on top of Debian's packaging.

Reviewed-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Tested-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2019-02-27 13:46:26 +01:00
Fabian Grünbichler 28f635be8c zfs/debian: add packaging files
based on Debian's packaging work, but simplified:
- no DKMS
- no dracut
- no udebs

and remove old patches which were based on top of Debian's packaging.

Reviewed-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Tested-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2019-02-27 13:46:26 +01:00
Fabian Grünbichler 2e2aafd11c submodules: add upstream mirrors
pointing to ZFS on Linux's github repositories

Reviewed-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Tested-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2019-02-27 13:46:25 +01:00
Fabian Grünbichler 4e389d452b spl: rebase patches on top of debian/0.7.12-1
and drop those contained in 0.7.12

Reviewed-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Tested-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2019-02-27 13:46:20 +01:00
Fabian Grünbichler 3e9c7d6dff spl: update submodule to debian/0.7.12-1
Reviewed-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Tested-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2019-02-27 13:46:19 +01:00
Fabian Grünbichler 2249b05575 zfs: rebase patches on top of 0.7.12-2
and drop those that are part of 0.7.12

Reviewed-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Tested-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2019-02-27 13:46:15 +01:00
Fabian Grünbichler 002d5f1b8c zfs: update submodule to debian/0.7.12-2
Reviewed-by: Stoiko Ivanov <s.ivanov@proxmox.com>
Tested-by: Stoiko Ivanov <s.ivanov@proxmox.com>
2019-02-27 13:44:49 +01:00
Stoiko Ivanov 65a9766fa0 bump version to 0.7.12-pve1~bpo1
Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
2018-11-14 18:27:04 +01:00
Stoiko Ivanov 04a710dd91 update/rebase to zfs-0.7.12 with patches from ZOL
Reorder patches, so that the upstream changeset comes last

Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
2018-11-14 18:27:04 +01:00
Stoiko Ivanov 2e5067b011 update/rebase to spl-0.7.12 with patches from ZOL
Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
2018-11-14 18:27:04 +01:00
82 changed files with 5099 additions and 2589 deletions
+6 -6
View File
@@ -1,6 +1,6 @@
[submodule "zfs-debian"]
path = zfs-debian
url = ../mirror_zfs-debian
[submodule "spl-debian"]
path = spl-debian
url = ../mirror_spl-debian
[submodule "zfs/upstream"]
path = zfs/upstream
url = ../mirror_zfs
[submodule "spl/upstream"]
path = spl/upstream
url = ../mirror_spl
+48 -27
View File
@@ -2,19 +2,21 @@ RELEASE=5.1
# source form https://github.com/zfsonlinux/
ZFSVER=0.7.11
ZFSPKGREL=pve2~bpo1
SPLPKGREL=pve2~bpo1
ZFSPKGVER=${ZFSVER}-${ZFSPKGREL}
SPLPKGVER=${ZFSVER}-${SPLPKGREL}
SPLDIR=spl-linux_${ZFSVER}
SPLSRC=spl/upstream
SPLPKG=spl/debian
ZFSDIR=zfs-linux_${ZFSVER}
ZFSSRC=zfs/upstream
ZFSPKG=zfs/debian
SPLDIR=spl-build
SPLSRC=spl-debian
ZFSDIR=zfs-build
ZFSSRC=zfs-debian
ZFSVER != dpkg-parsechangelog -l ${ZFSPKG}/changelog -Sversion | cut -d- -f1
ZFSPKGVER != dpkg-parsechangelog -l ${ZFSPKG}/changelog -Sversion
SPLPKGVER != dpkg-parsechangelog -l ${SPLPKG}/changelog -Sversion
SPL_DEB = \
spl_${SPLPKGVER}_amd64.deb
SPL_DSC = spl-linux_${SPLPKGVER}.dsc
ZFS_DEB1= libnvpair1linux_${ZFSPKGVER}_amd64.deb
ZFS_DEB2= \
@@ -28,11 +30,22 @@ zfs-initramfs_${ZFSPKGVER}_all.deb \
zfs-test_${ZFSPKGVER}_amd64.deb \
zfsutils-linux_${ZFSPKGVER}_amd64.deb
ZFS_DEBS= $(ZFS_DEB1) $(ZFS_DEB2)
ZFS_DSC = zfs-linux_${ZFSPKGVER}.dsc
DEBS=${SPL_DEB} ${ZFS_DEBS}
DSCS=${SPL_DSC} ${ZFS_DSC}
all: deb
.PHONY: deb
deb: ${DEBS}
.PHONY: dsc
dsc: ${DSCS}
# called from pve-kernel's Makefile to get patched sources
.PHONY: kernel
kernel: dsc
dpkg-source -x ${SPL_DSC} ../pkg-spl
dpkg-source -x ${ZFS_DSC} ../pkg-zfs
.PHONY: dinstall
dinstall: ${DEBS}
@@ -40,44 +53,52 @@ dinstall: ${DEBS}
.PHONY: submodule
submodule:
test -f "${ZFSSRC}/debian/changelog" || git submodule update --init
test -f "${SPLSRC}/debian/changelog" || git submodule update --init
test -f "${ZFSSRC}/README.markdown" || git submodule update --init
test -f "${SPLSRC}/README.markdown" || git submodule update --init
.PHONY: spl
spl: ${SPL_DEB}
${SPL_DEB}: ${SPLSRC}
${SPL_DEB}: ${SPLDIR}
cd ${SPLDIR}; dpkg-buildpackage -b -uc -us
lintian ${SPL_DEB}
${SPL_DSC}: ${SPLDIR}
tar czf spl-linux_${ZFSVER}.orig.tar.gz ${SPLDIR}
cd ${SPLDIR}; dpkg-buildpackage -S -uc -us -d
lintian $@
${SPLDIR}: ${SPLSRC} ${SPLPKG}
rm -rf ${SPLDIR}
mkdir ${SPLDIR}
cp -a ${SPLSRC}/* ${SPLDIR}/
mv ${SPLDIR}/debian/changelog ${SPLDIR}/debian/changelog.org
cat spl-changelog.Debian ${SPLDIR}/debian/changelog.org > ${SPLDIR}/debian/changelog
cd ${SPLDIR}; ln -s ../spl-patches patches
cd ${SPLDIR}; quilt push -a
cd ${SPLDIR}; rm -rf .pc ./patches
cd ${SPLDIR}; dpkg-buildpackage -b -uc -us
cp -a ${SPLPKG} ${SPLDIR}/debian
.PHONY: zfs
zfs: $(ZFS_DEBS)
$(ZFS_DEB2): $(ZFS_DEB1)
$(ZFS_DEB1): $(ZFSSRC)
$(ZFS_DEB1): ${ZFSDIR}
cd ${ZFSDIR}; dpkg-buildpackage -b -uc -us
lintian ${ZFS_DEBS}
${ZFS_DSC}: ${ZFSDIR}
tar czf zfs-linux_${ZFSVER}.orig.tar.gz ${ZFSDIR}
cd ${ZFSDIR}; dpkg-buildpackage -S -uc -us -d
lintian $@
${ZFSDIR}: $(ZFSSRC) ${ZFSPKG}
rm -rf ${ZFSDIR}
mkdir ${ZFSDIR}
cp -a ${ZFSSRC}/* ${ZFSDIR}/
mv ${ZFSDIR}/debian/changelog ${ZFSDIR}/debian/changelog.org
cat zfs-changelog.Debian ${ZFSDIR}/debian/changelog.org > ${ZFSDIR}/debian/changelog
cd ${ZFSDIR}; ln -s ../zfs-patches patches
cd ${ZFSDIR}; quilt push -a
cd ${ZFSDIR}; rm -rf .pc ./patches
cd ${ZFSDIR}; dpkg-buildpackage -b -uc -us
cp -a ${ZFSPKG} ${ZFSDIR}/debian
.PHONY: clean
clean:
rm -rf *~ *.deb *.changes *.buildinfo ${ZFSDIR} ${SPLDIR}
rm -rf *~ *.deb *.changes *.buildinfo *.dsc *.orig.tar.* *.debian.tar.* ${ZFSDIR} ${SPLDIR}
.PHONY: distclean
distclean: clean
.PHONY: upload
upload: ${DEBS}
tar -cf - ${DEBS} | ssh repoman@repo.proxmox.com -- upload --product pve,pmg --dist stretch --arch amd64
+2 -2
View File
@@ -2,8 +2,8 @@ Proxmox VE packaging for ZFS on Linux
This is based on code from:
https://github.com/zfsonlinux/pkg-spl.git
https://github.com/zfsonlinux/pkg-zfs.git
https://salsa.debian.org/zfsonlinux-team/zfs
https://salsa.debian.org/zfsonlinux-team/spl
We removed the dkms/modules related code, because we ship the
modules with the kernel.
-153
View File
@@ -1,153 +0,0 @@
spl-linux (0.7.11-pve2~bpo1) unstable; urgency=medium
* update SPL to debian/0.7.11
* Backport deadlock fix between mm_sem and tx assign in zfs_write() and page
fault
-- Proxmox Support Team <support@proxmox.com> Mon, 29 Oct 2018 13:47:39 +0100
spl-linux (0.7.11-pve1~bpo1) unstable; urgency=medium
* update SPL to 0.7.11
-- Proxmox Support Team <support@proxmox.com> Mon, 10 Sep 2018 15:47:06 +0200
spl-linux (0.7.9-pve1~bpo9) unstable; urgency=medium
* update SPL to 0.7.9
-- Proxmox Support Team <support@proxmox.com> Fri, 18 May 2018 13:49:09 +0200
spl-linux (0.7.8-pve1~bpo9) unstable; urgency=medium
* update SPL to 0.7.8 (no changes)
-- Proxmox Support Team <support@proxmox.com> Mon, 30 Apr 2018 13:47:22 +0200
spl-linux (0.7.7-pve1~bpo9) unstable; urgency=medium
* update SPL to 0.7.7
-- Proxmox Support Team <support@proxmox.com> Tue, 3 Apr 2018 14:28:35 +0200
spl-linux (0.7.6-pve1~bpo9) unstable; urgency=medium
* update SPL to 0.7.6
-- Proxmox Support Team <support@proxmox.com> Wed, 21 Feb 2018 09:47:54 +0100
spl-linux (0.7.4-pve1~bpo9) unstable; urgency=medium
* update SPL to 0.7.4
-- Proxmox Support Team <support@proxmox.com> Tue, 19 Dec 2017 12:52:42 +0100
spl-linux (0.7.3-pve1~bpo9) unstable; urgency=medium
* update spl to 0.7.3
-- Proxmox Support Team <support@proxmox.com> Mon, 6 Nov 2017 09:14:19 +0100
spl-linux (0.7.2-pve1~bpo90) unstable; urgency=medium
* update spl to 0.7.2
* clean up man pages
-- Proxmox Support Team <support@proxmox.com> Thu, 12 Oct 2017 13:33:04 +0200
spl-linux (0.6.5.11-pve10~bpo90) unstable; urgency=medium
* update spl to debian/0.6.5.11-1
-- Proxmox Support Team <support@proxmox.com> Thu, 27 Jul 2017 13:11:45 +0200
spl-linux (0.6.5.9-pve9~bpo90) unstable; urgency=medium
* rebuild for PVE 5.0 / Debian Stretch
-- Proxmox Support Team <support@proxmox.com> Fri, 10 Mar 2017 09:45:52 +0100
spl-linux (0.6.5.9-pve8~bpo80) unstable; urgency=medium
* update spl to debian/0.6.5.9-1
-- Proxmox Support Team <support@proxmox.com> Mon, 6 Feb 2017 12:39:35 +0100
spl-linux (0.6.5.8-pve7~bpo80) unstable; urgency=medium
* update spl to debian/0.6.5.8-2
* switch package upstream sources to Debian (Jessie)
-- Proxmox Support Team <support@proxmox.com> Wed, 12 Oct 2016 11:16:02 +0200
spl-linux (0.6.5.7-pve6~bpo80) unstable; urgency=medium
* update pkg-spl to jessie/0.6.5.7-5
* change package versioning to allow upgrades from PVE3/wheezy
-- Proxmox Support Team <support@proxmox.com> Tue, 28 Jun 2016 09:50:00 +0200
spl-linux (0.6.5-pve5~jessie) unstable; urgency=medium
* Update pkg-spl to jessie/0.6.5.6-3
-- Proxmox Support Team <support@proxmox.com> Fri, 15 Apr 2016 11:11:59 +0200
spl-linux (0.6.5-pve4~jessie) unstable; urgency=medium
* update to 0.6.5.6
-- Proxmox Support Team <support@proxmox.com> Wed, 30 Mar 2016 10:49:49 +0200
spl-linux (0.6.5-pve3~jessie) unstable; urgency=medium
* update to 0.6.5.4
-- Proxmox Support Team <support@proxmox.com> Thu, 14 Jan 2016 11:00:40 +0100
spl-linux (0.6.5-pve2~jessie) unstable; urgency=medium
* update to 0.6.5.3
-- Proxmox Support Team <support@proxmox.com> Wed, 04 Nov 2015 17:40:58 +0100
spl-linux (0.6.5-pve1~jessie) unstable; urgency=medium
* update to master/debian/jessie/0.6.5-1
-- Proxmox Support Team <support@proxmox.com> Mon, 21 Sep 2015 10:04:23 +0200
spl-linux (0.6.4-pve3~jessie) unstable; urgency=medium
* update to snapshot/debian/jessie/0.6.4-8-8ac6ff
-- Proxmox Support Team <support@proxmox.com> Mon, 24 Aug 2015 06:14:12 +0200
spl-linux (0.6.4-pve2~jessie) unstable; urgency=medium
* update to 0.6.4.2 (snapshot/debian/jessie/0.6.4-6-37d7cd)
-- Proxmox Support Team <support@proxmox.com> Tue, 21 Jul 2015 11:02:31 +0200
spl-linux (0.6.4-pve1~jessie) unstable; urgency=medium
* update to 0.6.4 (use upstream zol package definitions)
-- Proxmox Support Team <support@proxmox.com> Tue, 14 Apr 2015 10:17:26 +0200
spl-linux (0.6.3-pve1~jessie) unstable; urgency=medium
* recompile for jessie
-- Proxmox Support Team <support@proxmox.com> Sat, 28 Feb 2015 16:03:01 +0100
spl-linux (0.6.3-pve1~wheezy) unstable; urgency=low
* first version for Proxmox VE
-- Proxmox Support Team <support@proxmox.com> Sat, 10 Jan 2015 14:06:34 +0100
Submodule spl-debian deleted from 3ac1211d59
@@ -1,167 +0,0 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Fabian=20Gr=C3=BCnbichler?= <f.gruenbichler@proxmox.com>
Date: Wed, 12 Oct 2016 10:57:39 +0200
Subject: [PATCH] remove DKMS and module build
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
---
debian/control | 1 -
debian/control.in | 31 -------------------------
debian/rules | 67 +------------------------------------------------------
3 files changed, 1 insertion(+), 98 deletions(-)
diff --git a/debian/control b/debian/control
index d83f962..f780c80 100644
--- a/debian/control
+++ b/debian/control
@@ -9,7 +9,6 @@ Build-Depends: autogen,
autotools-dev,
debhelper (>= 9),
dh-autoreconf,
- dkms (>> 2.2.0.2-1~),
libtool
Standards-Version: 4.1.2
Homepage: http://www.zfsonlinux.org/
diff --git a/debian/control.in b/debian/control.in
index 0e6c3fc..c8fe170 100644
--- a/debian/control.in
+++ b/debian/control.in
@@ -9,47 +9,16 @@ Build-Depends: autogen,
autotools-dev,
debhelper (>= 9),
dh-autoreconf,
- dkms (>> 2.2.0.2-1~),
libtool
Standards-Version: 4.1.2
Homepage: http://www.zfsonlinux.org/
Vcs-Git: https://salsa.debian.org/zfsonlinux-team/spl.git
Vcs-Browser: https://salsa.debian.org/zfsonlinux-team/spl
-Package: spl-dkms
-Architecture: all
-Depends: dkms (>> 2.2.1.0),
- file,
- libc-dev,
- libelf-dev,
- lsb-release,
- ${misc:Depends}
-Suggests: spl (>= ${source:Upstream-Version}),
- @LINUX_COMPAT@
-Conflicts: spl (<< 0.7.9~)
-Provides: spl-modules
-Description: Solaris Porting Layer kernel modules for Linux
- The Solaris Porting Layer (SPL) is a Linux kernel module which provides
- many of the Solaris kernel APIs. This shim layer makes it possible to
- run Solaris kernel code in the Linux kernel with relatively minimal
- modification. The Solaris Porting LAyer Tests (SPLAT) is a Linux kernel
- module which provides a testing harness for the SPL module.
- .
- SPL can be particularly useful when you want to track upstream Illumos
- (or any other OpenSolaris fork) development closely and don't want the
- overhead of maintaining a large patch which converts Solaris primitives
- to Linux primitives.
- .
- This package contains the source code for the SPL and SPLAT Linux kernel
- modules, which can be used with DKMS, so that local kernel modules are
- automatically built and installed every time the kernel packages are
- upgraded.
-
Package: spl
Architecture: linux-any
Conflicts: spl-dev, splat
Depends: ${misc:Depends}, ${shlibs:Depends}
-Recommends: spl-modules | spl-dkms
Description: Solaris Porting Layer user-space utilities for Linux
The Solaris Porting Layer (SPL) is a Linux kernel module which provides
many of the Solaris kernel APIs. This shim layer makes it possible to
diff --git a/debian/rules b/debian/rules
index 72ded49..5abeebb 100755
--- a/debian/rules
+++ b/debian/rules
@@ -14,11 +14,8 @@ ifndef KVERS
KVERS=$(shell uname -r)
endif
-non_epoch_version=$(shell echo $(KVERS) | perl -pe 's/^\d+://')
-pmodules = $(NAME)-modules-$(non_epoch_version)
-
%:
- dh $@ --with dkms,autoreconf --parallel
+ dh $@ --with autoreconf --parallel
override_dh_auto_configure:
sed "s/@LINUX_COMPAT@/linux-libc-dev \(<< $(LINUX_NEXT)~\)/" debian/control.in > debian/control
@@ -44,69 +41,7 @@ override_dh_auto_install:
@# This creates the $(CURDIR)/$(NAME)-$(VERSION)/ tree.
$(MAKE) distdir
- @# This shunt allows DKMS to install the Module.symvers and spl_config.h
- @# files to the ${dkms_tree} area through the POST_INSTALL directive.
- echo '#!/bin/sh' >'$(CURDIR)/$(NAME)-$(VERSION)/cp'
- echo 'cp "$$@"' >>'$(CURDIR)/$(NAME)-$(VERSION)/cp'
- chmod 755 '$(CURDIR)/$(NAME)-$(VERSION)/cp'
-
- # Install the DKMS source.
- mkdir -p '$(CURDIR)/debian/tmp/usr/src/'
- mv '$(CURDIR)/$(NAME)-$(VERSION)' '$(CURDIR)/debian/tmp/usr/src/'
-
-override_dh_dkms:
- dh_dkms -V $(VERSION)
-
override_dh_auto_clean:
dh_auto_clean
@if test -e META.orig; then mv META.orig META; fi
sed "s/@LINUX_COMPAT@/linux-libc-dev \(<< $(LINUX_NEXT)~\)/" debian/control.in > debian/control
-
-# ------------
-
-override_dh_prep-deb-files:
- for templ in $(wildcard $(CURDIR)/debian/*_KVERS_*.in); do \
- sed -e 's/##KVERS##/$(KVERS)/g ; s/#KVERS#/$(KVERS)/g ; s/_KVERS_/$(KVERS)/g ; s/##KDREV##/$(KDREV)/g ; s/#KDREV#/$(KDREV)/g ; s/_KDREV_/$(KDREV)/g ; s/_ARCH_/$(DEB_HOST_ARCH)/g' \
- < $$templ > `echo $$templ | sed -e 's/_KVERS_/$(KVERS)/g ; s/_ARCH_/$(DEB_HOST_ARCH)/g ; s/\.in$$//'` ; \
- done
- sed -e 's/##KVERS##/$(KVERS)/g ; s/#KVERS#/$(KVERS)/g ; s/_KVERS_/$(KVERS)/g ; s/##KDREV##/$(KDREV)/g ; s/#KDREV#/$(KDREV)/g ; s/_KDREV_/$(KDREV)/g ; s/_ARCH_/$(DEB_HOST_ARCH)/g' \
- < debian/control.modules.in > debian/control
-
-override_dh_configure_modules: override_dh_configure_modules_stamp
-override_dh_configure_modules_stamp:
- ./configure --with-config=kernel --with-linux=$(KSRC) \
- --with-linux-obj=$(KOBJ) \
- --disable-debug-kmem
- touch override_dh_configure_modules_stamp
-
-override_dh_binary-modules-udeb: override_dh_prep-deb-files override_dh_configure_modules
- dh_testdir
- dh_testroot
- dh_prep
-
- $(MAKE) -C $(CURDIR)/module modules
-
- dh_installdirs -p${pmodules}-di
- dh_install -p${pmodules}-di
- dh_gencontrol -p${pmodules}-di
-
- dh_builddeb -p${pmodules}-di
-
-override_dh_binary-modules: override_dh_prep-deb-files override_dh_configure_modules
- dh_testdir
- dh_testroot
- dh_prep
-
- $(MAKE) -C $(CURDIR)/module modules
-
- dh_installdocs -p${pmodules}
- dh_install -p${pmodules}
- dh_installchangelogs -p${pmodules}
- dh_compress -p${pmodules}
- dh_strip -p${pmodules}
- dh_fixperms -p${pmodules}
- dh_installdeb -p${pmodules}
- dh_gencontrol -p${pmodules}
- dh_md5sums -p${pmodules}
-
- dh_builddeb -p${pmodules}
@@ -1,42 +0,0 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: ilbsmart <wgqimut@gmail.com>
Date: Wed, 17 Oct 2018 02:11:24 +0800
Subject: [PATCH] deadlock between mm_sem and tx assign in zfs_write() and page
fault
The bug time sequence:
1. thread #1, `zfs_write` assign a txg "n".
2. In a same process, thread #2, mmap page fault (which means the
`mm_sem` is hold) occurred, `zfs_dirty_inode` open a txg failed,
and wait previous txg "n" completed.
3. thread #1 call `uiomove` to write, however page fault is occurred
in `uiomove`, which means it need `mm_sem`, but `mm_sem` is hold by
thread #2, so it stuck and can't complete, then txg "n" will
not complete.
So thread #1 and thread #2 are deadlocked.
Reviewed-by: Chunwei Chen <tuxoko@gmail.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Signed-off-by: Grady Wong <grady.w@xtaotech.com>
Closes #7939
(backported from: zfs-upstream 779a6c0bf6df76e0dd92c1ccf81f48512b835bb0)
Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
---
include/sys/uio.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/include/sys/uio.h b/include/sys/uio.h
index 764beb9..2895690 100644
--- a/include/sys/uio.h
+++ b/include/sys/uio.h
@@ -53,6 +53,7 @@ typedef struct uio {
int uio_iovcnt;
offset_t uio_loffset;
uio_seg_t uio_segflg;
+ boolean_t uio_fault_disable;
uint16_t uio_fmode;
uint16_t uio_extflg;
offset_t uio_limit;
-2
View File
@@ -1,2 +0,0 @@
0001-remove-DKMS-and-module-build.patch
0002-deadlock-between-mm_sem-and-tx-assign-in-zfs_write-a.patch
+380
View File
@@ -0,0 +1,380 @@
spl-linux (0.7.13-pve1~bpo2) unstable; urgency=medium
* update SPL to 0.7.13
* base spl-linux on upstream ZOL instead of debian
-- Proxmox Support Team <support@proxmox.com> Fri, 08 Mar 2019 06:53:13 +0100
spl-linux (0.7.12-pve1~bpo1) unstable; urgency=medium
* update SPL to 0.7.12
-- Proxmox Support Team <support@proxmox.com> Wed, 14 Nov 2018 17:51:59 +0100
spl-linux (0.7.11-pve2~bpo1) unstable; urgency=medium
* update SPL to debian/0.7.11
* Backport deadlock fix between mm_sem and tx assign in zfs_write() and page
fault
-- Proxmox Support Team <support@proxmox.com> Mon, 29 Oct 2018 13:47:39 +0100
spl-linux (0.7.11-pve1~bpo1) unstable; urgency=medium
* update SPL to 0.7.11
-- Proxmox Support Team <support@proxmox.com> Mon, 10 Sep 2018 15:47:06 +0200
spl-linux (0.7.9-pve1~bpo9) unstable; urgency=medium
* update SPL to 0.7.9
-- Proxmox Support Team <support@proxmox.com> Fri, 18 May 2018 13:49:09 +0200
spl-linux (0.7.8-pve1~bpo9) unstable; urgency=medium
* update SPL to 0.7.8 (no changes)
-- Proxmox Support Team <support@proxmox.com> Mon, 30 Apr 2018 13:47:22 +0200
spl-linux (0.7.7-pve1~bpo9) unstable; urgency=medium
* update SPL to 0.7.7
-- Proxmox Support Team <support@proxmox.com> Tue, 3 Apr 2018 14:28:35 +0200
spl-linux (0.7.6-pve1~bpo9) unstable; urgency=medium
* update SPL to 0.7.6
-- Proxmox Support Team <support@proxmox.com> Wed, 21 Feb 2018 09:47:54 +0100
spl-linux (0.7.4-pve1~bpo9) unstable; urgency=medium
* update SPL to 0.7.4
-- Proxmox Support Team <support@proxmox.com> Tue, 19 Dec 2017 12:52:42 +0100
spl-linux (0.7.3-pve1~bpo9) unstable; urgency=medium
* update spl to 0.7.3
-- Proxmox Support Team <support@proxmox.com> Mon, 6 Nov 2017 09:14:19 +0100
spl-linux (0.7.2-pve1~bpo90) unstable; urgency=medium
* update spl to 0.7.2
* clean up man pages
-- Proxmox Support Team <support@proxmox.com> Thu, 12 Oct 2017 13:33:04 +0200
spl-linux (0.6.5.11-pve10~bpo90) unstable; urgency=medium
* update spl to debian/0.6.5.11-1
-- Proxmox Support Team <support@proxmox.com> Thu, 27 Jul 2017 13:11:45 +0200
spl-linux (0.6.5.9-pve9~bpo90) unstable; urgency=medium
* rebuild for PVE 5.0 / Debian Stretch
-- Proxmox Support Team <support@proxmox.com> Fri, 10 Mar 2017 09:45:52 +0100
spl-linux (0.6.5.9-pve8~bpo80) unstable; urgency=medium
* update spl to debian/0.6.5.9-1
-- Proxmox Support Team <support@proxmox.com> Mon, 6 Feb 2017 12:39:35 +0100
spl-linux (0.6.5.8-pve7~bpo80) unstable; urgency=medium
* update spl to debian/0.6.5.8-2
* switch package upstream sources to Debian (Jessie)
-- Proxmox Support Team <support@proxmox.com> Wed, 12 Oct 2016 11:16:02 +0200
spl-linux (0.6.5.7-pve6~bpo80) unstable; urgency=medium
* update pkg-spl to jessie/0.6.5.7-5
* change package versioning to allow upgrades from PVE3/wheezy
-- Proxmox Support Team <support@proxmox.com> Tue, 28 Jun 2016 09:50:00 +0200
spl-linux (0.6.5-pve5~jessie) unstable; urgency=medium
* Update pkg-spl to jessie/0.6.5.6-3
-- Proxmox Support Team <support@proxmox.com> Fri, 15 Apr 2016 11:11:59 +0200
spl-linux (0.6.5-pve4~jessie) unstable; urgency=medium
* update to 0.6.5.6
-- Proxmox Support Team <support@proxmox.com> Wed, 30 Mar 2016 10:49:49 +0200
spl-linux (0.6.5-pve3~jessie) unstable; urgency=medium
* update to 0.6.5.4
-- Proxmox Support Team <support@proxmox.com> Thu, 14 Jan 2016 11:00:40 +0100
spl-linux (0.6.5-pve2~jessie) unstable; urgency=medium
* update to 0.6.5.3
-- Proxmox Support Team <support@proxmox.com> Wed, 04 Nov 2015 17:40:58 +0100
spl-linux (0.6.5-pve1~jessie) unstable; urgency=medium
* update to master/debian/jessie/0.6.5-1
-- Proxmox Support Team <support@proxmox.com> Mon, 21 Sep 2015 10:04:23 +0200
spl-linux (0.6.4-pve3~jessie) unstable; urgency=medium
* update to snapshot/debian/jessie/0.6.4-8-8ac6ff
-- Proxmox Support Team <support@proxmox.com> Mon, 24 Aug 2015 06:14:12 +0200
spl-linux (0.6.4-pve2~jessie) unstable; urgency=medium
* update to 0.6.4.2 (snapshot/debian/jessie/0.6.4-6-37d7cd)
-- Proxmox Support Team <support@proxmox.com> Tue, 21 Jul 2015 11:02:31 +0200
spl-linux (0.6.4-pve1~jessie) unstable; urgency=medium
* update to 0.6.4 (use upstream zol package definitions)
-- Proxmox Support Team <support@proxmox.com> Tue, 14 Apr 2015 10:17:26 +0200
spl-linux (0.6.3-pve1~jessie) unstable; urgency=medium
* recompile for jessie
-- Proxmox Support Team <support@proxmox.com> Sat, 28 Feb 2015 16:03:01 +0100
spl-linux (0.6.3-pve1~wheezy) unstable; urgency=low
* first version for Proxmox VE
-- Proxmox Support Team <support@proxmox.com> Sat, 10 Jan 2015 14:06:34 +0100
spl-linux (0.7.12-1) unstable; urgency=medium
* New upstream version 0.7.12
* Bump linux compatibility to 4.19 .
-- Mo Zhou <cdluminate@gmail.com> Mon, 19 Nov 2018 11:25:53 +0000
spl-linux (0.7.11-1) unstable; urgency=medium
[ Nicolas Braud-Santoni ]
* control: Use a canonical URI for Vcs-Git
* copyright: Use HTTPs Format URI in copyright
* debian/rules: Enable build hardening
* debian/rules: Use DEB_VERSION* instead of manually parsing.
[ Mo Zhou ]
* New upstream version 0.7.11 (Closes: #908483)
* Append myself to Uploaders.
* Suggests linux-libc-dev (<< LINUX_NEXT~) instead of (<< LINUX_NEXT).
* Replace get_next.sh with one-liner awk script in rules.
* Bump linux_compat to 4.18 .
* Fix debian-rules-sets-dpkg-architecture-variable.
* Fix debian-watch-uses-insecure-uri.
-- Mo Zhou <cdluminate@gmail.com> Wed, 19 Sep 2018 04:13:01 +0000
spl-linux (0.7.9-3) unstable; urgency=medium
* d/control: migrate to alioth-lists (Closes: #899692)
-- Aron Xu <aron@debian.org> Mon, 28 May 2018 18:05:37 +0800
spl-linux (0.7.9-2) unstable; urgency=medium
[ Aron Xu ]
* New upstream version 0.7.9
* Fix lintian obsolete-relation-form-in-source
* Bump supported linux version to 4.16
[ Antonio Russo ]
* Reduce dependency on spl, and move documentation (Closes: #894608)
-- Aron Xu <aron@debian.org> Wed, 16 May 2018 21:36:17 +0800
spl-linux (0.7.6-1) unstable; urgency=medium
* New upstream release (Closes: #891235)
* Update VCS-* URL to salsa.debian.org
* Apply wrap-and-sort to control*
* Recommends: linux-libc-dev (< ${LINUX_NEXT}):
Tries to prevent unexpected upgrades of kernel that is not known
to be supported by the packaged version of ZFS/SPL.
-- Aron Xu <aron@debian.org> Mon, 26 Feb 2018 16:29:50 +0800
spl-linux (0.7.5-1) unstable; urgency=medium
* Add libelf-dev to Depends (Closes: #886453)
* New upstream version 0.7.5 (Closes: #886453)
-- Aron Xu <aron@debian.org> Fri, 19 Jan 2018 15:23:26 +0800
spl-linux (0.7.4-1) unstable; urgency=medium
* New upstream version 0.7.4 (Closes: #883830)
* Update stdver to 4.1.2, no change required
-- Aron Xu <aron@debian.org> Mon, 18 Dec 2017 22:05:32 +0800
spl-linux (0.7.3-1) unstable; urgency=medium
* New upstream version 0.7.3
-- Aron Xu <aron@debian.org> Tue, 31 Oct 2017 17:29:08 +0800
spl-linux (0.6.5.11-1) unstable; urgency=medium
* Imported Upstream version 0.6.5.11
-- Aron Xu <aron@debian.org> Fri, 14 Jul 2017 14:39:21 +0800
spl-linux (0.6.5.10-1) unstable; urgency=medium
* New upstream version 0.6.5.10
-- Aron Xu <aron@debian.org> Wed, 05 Jul 2017 18:00:58 +0800
spl-linux (0.6.5.9-1) unstable; urgency=medium
* Imported Upstream version 0.6.5.9
-- Aron Xu <aron@debian.org> Mon, 06 Feb 2017 15:52:47 +0800
spl-linux (0.6.5.8-3) unstable; urgency=medium
* Fix Linux 4.9 compatibility (Closes: #851352)
Thanks to Fabian Grünbichler!
-- Aron Xu <aron@debian.org> Tue, 17 Jan 2017 20:57:13 +0800
spl-linux (0.6.5.8-2) unstable; urgency=medium
[ Petter Reinholdtsen ]
* Added d/gbp.conf to enforce the use of pristine-tar.
* Added extra autopkgtest script check.sh from the Ubuntu package.
[ Aron Xu ]
* Make use of dh-autoreconf
* Deal with /etc/hostid in postinst of spl package
* Add lintian-overrides: extra-license-file
* Simplify rules by merging Ubuntu one
* Handle the case /etc/hostid is being wrongly marked as conffile
* Install man pages for spl
* Move command-with-path-in-maintainer-script override to spl
-- Aron Xu <aron@debian.org> Wed, 28 Sep 2016 05:02:06 +0800
spl-linux (0.6.5.8-1) unstable; urgency=medium
[ Petter Reinholdtsen ]
* Fix invalid command in dkms (Closes: #836578)
[ Zhou Mo ]
* control: bump standards version to 3.9.8
[ Aron Xu ]
* Imported Upstream version 0.6.5.8 (Closes: #835992)
* rules: allow parallel building
-- Aron Xu <aron@debian.org> Sun, 18 Sep 2016 01:06:04 +0800
spl-linux (0.6.5.7-1) unstable; urgency=medium
* Imported Upstream version 0.6.5.7
-- Aron Xu <aron@debian.org> Wed, 25 May 2016 12:17:05 +0800
spl-linux (0.6.5.6-2) unstable; urgency=medium
* Add license names to entries missing it.
* Added autopkgtest using code from Ubuntu using dkms >= 2.2.0.3-3~.
-- Petter Reinholdtsen <pere@hungry.com> Thu, 28 Apr 2016 20:21:29 +0800
spl-linux (0.6.5.6-1) unstable; urgency=medium
* New upstream release
-- Aron Xu <aron@debian.org> Thu, 24 Mar 2016 07:14:47 +0800
spl-linux (0.6.5.5-1) unstable; urgency=medium
* New upstream release
-- Aron Xu <aron@debian.org> Sun, 20 Mar 2016 22:50:26 +0800
spl-linux (0.6.5.4-1) unstable; urgency=medium
* New upstream release
-- Aron Xu <aron@debian.org> Sun, 10 Jan 2016 20:13:15 +0800
spl-linux (0.6.5.3-1) unstable; urgency=medium
* New upstream release
-- Aron Xu <aron@debian.org> Thu, 05 Nov 2015 21:08:25 +0800
spl-linux (0.6.5.2-1) unstable; urgency=medium
* Imported Upstream version 0.6.5.2
* Add conflicts to splat (Closes: #769147)
-- Aron Xu <aron@debian.org> Tue, 13 Oct 2015 15:26:33 +0100
spl-linux (0.6.4.2-1) experimental; urgency=medium
* New upstream release.
-- Aron Xu <aron@debian.org> Thu, 13 Aug 2015 19:58:26 +0800
spl-linux (0.6.4.1-1) experimental; urgency=medium
* New upstream release.
-- Aron Xu <aron@debian.org> Sun, 17 May 2015 05:47:51 +0800
spl-linux (0.6.2-2) experimental; urgency=low
* Add file and libc-dev to Depends of spl-dkms (Closes: #717565).
-- Aron Xu <aron@debian.org> Sun, 25 Aug 2013 02:02:15 +0800
spl-linux (0.6.2-1) experimental; urgency=low
* New upstream release.
-- Aron Xu <aron@debian.org> Sat, 24 Aug 2013 14:30:25 +0800
spl-linux (0.6.1-2) experimental; urgency=low
* Add license info for scripts/kmodtool
* Add linux 3.10 compatibility patch from upstream. (Closes: #717565)
* Add facility to build udeb and binary modules, thanks Turbo Fredriksson
for the work.
-- Aron Xu <aron@debian.org> Mon, 29 Jul 2013 16:04:24 +0800
spl-linux (0.6.1-1) experimental; urgency=low
* Initial Release (Closes: #686453)
-- Aron Xu <aron@debian.org> Tue, 02 Apr 2013 01:03:05 +0200
+1
View File
@@ -0,0 +1 @@
10
+32
View File
@@ -0,0 +1,32 @@
Source: spl-linux
Section: kernel
Priority: optional
Maintainer: Proxmox Support Team <support@proxmox.com>
Build-Depends: autogen,
autotools-dev,
debhelper (>= 10~),
dh-autoreconf,
libtool
Standards-Version: 4.1.2
Homepage: http://www.zfsonlinux.org/
Vcs-Git: https://git.proxmox.com/git/zfsonlinux.git
Vcs-Browser: https://git.proxmox.com/?p=zfsonlinux.git;a=summary
Package: spl
Architecture: linux-any
Conflicts: spl-dev, splat
Depends: ${misc:Depends}, ${shlibs:Depends}
Description: Solaris Porting Layer user-space utilities for Linux
The Solaris Porting Layer (SPL) is a Linux kernel module which provides
many of the Solaris kernel APIs. This shim layer makes it possible to
run Solaris kernel code in the Linux kernel with relatively minimal
modification. The Solaris Porting LAyer Tests (SPLAT) is a Linux kernel
module which provides a testing harness for the SPL module.
.
SPL can be particularly useful when you want to track upstream Illumos
(or any other OpenSolaris fork) development closely and don't want the
overhead of maintaining a large patch which converts Solaris primitives
to Linux primitives.
.
This package includes the user space utilities needed to test and debug
the SPL and SPLAT kernel modules.
+189
View File
@@ -0,0 +1,189 @@
Format: https://www.debian.org/doc/packaging-manuals/copyright-format/1.0/
Upstream-Name: Solaris Porting Layer for Linux
Upstream-Contact: Brian Behlendorf <behlendorf1@llnl.gov>
Source: https://github.com/zfsonlinux/spl/
Disclaimer:
This work was produced at the Lawrence Livermore National Laboratory
(LLNL) under Contract No. DE-AC52-07NA27344 (Contract 44) between
the U.S. Department of Energy (DOE) and Lawrence Livermore National
Security, LLC (LLNS) for the operation of LLNL.
.
This work was prepared as an account of work sponsored by an agency of
the United States Government. Neither the United States Government nor
Lawrence Livermore National Security, LLC nor any of their employees,
makes any warranty, express or implied, or assumes any liability or
responsibility for the accuracy, completeness, or usefulness of any
information, apparatus, product, or process disclosed, or represents
that its use would not infringe privately-owned rights.
.
Reference herein to any specific commercial products, process, or
services by trade name, trademark, manufacturer or otherwise does
not necessarily constitute or imply its endorsement, recommendation,
or favoring by the United States Government or Lawrence Livermore
National Security, LLC. The views and opinions of authors expressed
herein do not necessarily state or reflect those of the Untied States
Government or Lawrence Livermore National Security, LLC, and shall
not be used for advertising or product endorsement purposes.
Files: *
Copyright: Lawrence Livermore National Security, LLC.
The Regents of the University of California
License: GPL-2+
Files: config/config.guess
Copyright: Free Software Foundation, Inc.
License: GPL-2+
Files: config/config.sub
Copyright: Free Software Foundation, Inc.
License: GPL-2+
Files: config/deb.am
Copyright: Lawrence Livermore National Security, LLC.
License: GPL-2+
Files: config/depcomp
Copyright: Free Software Foundation, Inc.
License: GPL-2+
Files: config/install-sh
Copyright: 1994 X Consortium
License: X11-install-sh
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to
deal in the Software without restriction, including without limitation the
rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
sell copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
.
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
X CONSORTIUM BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN
AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNEC-
TION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
.
Except as contained in this notice, the name of the X Consortium shall not
be used in advertising or otherwise to promote the sale, use or other deal-
ings in this Software without prior written authorization from the X Consor-
tium.
Files: config/ltmain.sh
Copyright: Free Software Foundation, Inc.
License: GPL-2+
Files: config/missing
Copyright: Free Software Foundation, Inc.
License: GPL-2+
Files: config/tgz.am
Copyright: Lawrence Livermore National Security, LLC.
License: GPL-2+
Files: debian/*
Copyright: Darik Horn <dajhorn@vanadac.com>
Carlos Alberto Lopez Perez <clopez@igalia.com>
Aron Xu <aron@debian.org>
License: GPL-2+
Files: include/linux/zlib_compat.h
Copyright: Lawrence Livermore National Security, LLC.
License: GPL-2+
Files: include/rpc/xdr.h
Copyright: Sun Microsystems, Inc.
License: GPL-2+
Files: include/sys/extdirent.h
Copyright: Lawrence Livermore National Security, LLC.
License: GPL-2+
Files: include/sys/fcntl.h
Copyright: Lawrence Livermore National Security, LLC.
License: GPL-2+
Files: include/sys/idmap.h
Copyright: Lawrence Livermore National Security, LLC.
License: GPL-2+
Files: include/sys/tsd.h
Copyright: Lawrence Livermore National Security, LLC.
License: GPL-2+
Files: include/sys/zmod.h
Copyright: Jean-loup Gailly
Mark Adler
License: zmod
This software is provided 'as-is', without any express or implied
warranty. In no event will the authors be held liable for any damages
arising from the use of this software.
.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute it
freely, subject to the following restrictions:
.
1. The origin of this software must not be misrepresented; you must not
claim that you wrote the original software. If you use this software
in a product, an acknowledgment in the product documentation would be
appreciated but is not required.
.
2. Altered source versions must be plainly marked as such, and must not be
misrepresented as being the original software.
.
3. This notice may not be removed or altered from any source distribution.
Files: module/spl/spl-tsd.c
Copyright: Lawrence Livermore National Security, LLC.
License: GPL-2+
Files: module/spl/spl-xdr.c
Copyright: Sun Microsystems, Inc.
License: GPL-2+
Files: module/splat/splat-linux.c
Copyright: Lawrence Livermore National Security, LLC.
License: GPL-2+
Files: scripts/kmodtool
Copyright: 2003-2012 Ville Skyttä <ville.skytta@iki.fi>
Thorsten Leemhuis <fedora@leemhuis.info>
Nicolas Chauvet <kwizart@gmail.com>
License: BSD-like
Permission is hereby granted, free of charge, to any person obtaining
a copy of this software and associated documentation files (the
"Software"), to deal in the Software without restriction, including
without limitation the rights to use, copy, modify, merge, publish,
distribute, sublicense, and/or sell copies of the Software, and to
permit persons to whom the Software is furnished to do so, subject to
the following conditions:
.
The above copyright notice and this permission notice shall be
included in all copies or substantial portions of the Software.
.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
License: GPL-2+
This program is free software; you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free Software
Foundation; either version 2 of the License, or (at your option) any later
version.
.
This program is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
.
You should have received a copy of the GNU General Public License along with
this package; if not, write to the Free Software Foundation, Inc., 51 Franklin
St, Fifth Floor, Boston, MA 02110-1301 USA
.
On Debian systems, the full text of the GNU General Public License version 2
can be found in the file /usr/share/common-licenses/GPL-2.
+33
View File
@@ -0,0 +1,33 @@
#!/usr/bin/make -f
include /usr/share/dpkg/pkg-info.mk
VERSION := $(DEB_VERSION_UPSTREAM)
REVISION := $(shell echo $(DEB_VERSION) | cut -d- -f2)
export DEB_BUILD_MAINT_OPTIONS = hardening=+all
%:
dh $@ --with autoreconf --parallel
override_dh_auto_configure:
@# Embed the downstream version in the module.
@sed \
-e 's/^Version:.*/Version: $(VERSION)/' \
-e 's/^Release:.*/Release: $(REVISION)/' \
-i.orig META
@# Build the userland, but don't build the kernel modules.
dh_auto_configure -- --with-config=user --disable-debug-kmem
override_dh_auto_test:
# scripts/check.sh tries insmod and rmmod, so it cannot
# run in an unprivileged build environment.
override_dh_auto_install:
@# Install the utilities.
$(MAKE) install DESTDIR='$(CURDIR)/debian/tmp'
override_dh_auto_clean:
dh_auto_clean
@if test -e META.orig; then mv META.orig META; fi
+1
View File
@@ -0,0 +1 @@
3.0 (quilt)
+1
View File
@@ -0,0 +1 @@
extend-diff-ignore='.*'
+2
View File
@@ -0,0 +1,2 @@
usr/sbin
usr/share/man/man1/splat.1
+1
View File
@@ -0,0 +1 @@
spl: command-with-path-in-maintainer-script
+49
View File
@@ -0,0 +1,49 @@
#!/bin/sh
set -e
# The hostname and hostid of the last system to access a ZFS pool are stored in
# the ZFS pool itself. A pool is foreign if, during `zpool import`, the
# current hostname and hostid are different than the stored values thereof.
#
# The hostid on Solaris is intrinsic, but is not on Linux (see #595790), so the
# spl kernel module invokes /usr/bin/hostid from the userland in its initialization
# routine.
#
# /usr/bin/hostid will return the 4 first bytes of the file /etc/hostid.
# If this file is not present or contains less than 4 bytes, then /usr/bin/hostid
# will return the bytes of the IP address of $(hostname) flipped, or zero if
# such IP couldn't be obtained
#
# This means that things like a DHCP lease change can affect the hostid.
#
# Therefore the only way of having a stable hostid is to define it on /etc/hostid.
# This postinst helper will check if we already have the hostid stabilized by
# checking the existence of the file /etc/hostid to be 4 bytes at least.
# If this file don't already exists on our system or has less than 4 bytes, then
# we will stabilize our current hostid by writing its value to /etc/hostid
# Detect if /etc/hostid is a conffile of previous spl package, migrate if yes
# hostid file should preserve even when package is purged
if $(dpkg-query --showformat='${Conffiles}\n' --show spl >/dev/null 2>&1); then
dpkg-maintscript-helper rm_conffile /etc/hostid -- "$@"
fi
if [ ! -f /etc/hostid ] || [ $(stat -c %s /etc/hostid) -lt 4 ] ; then
# Write our current hostid to /etc/hostid
HOSTID=$(hostid)
AA=$(echo $HOSTID | cut -b 1,2)
BB=$(echo $HOSTID | cut -b 3,4)
CC=$(echo $HOSTID | cut -b 5,6)
DD=$(echo $HOSTID | cut -b 7,8)
# Big Endian
if [ $(echo -n I | od -to2 | awk 'FNR==1{ print substr($2,6,1)}' 2>/dev/null) = 0 ]; then
# Invoke the printf from coreutils. shell builtin lacks the byte format.
/usr/bin/printf "\x$AA\x$BB\x$CC\x$DD" >/etc/hostid
else
# Little Endian
/usr/bin/printf "\x$DD\x$CC\x$BB\x$AA" >/etc/hostid
fi
fi
#DEBHELPER#
+10
View File
@@ -0,0 +1,10 @@
#!/bin/sh
set -e
# Detect if /etc/hostid is a conffile of previous spl package, migrate if yes
# hostid file should preserve even when package is purged
if $(dpkg-query --showformat='${Conffiles}\n' --show spl >/dev/null 2>&1); then
dpkg-maintscript-helper rm_conffile /etc/hostid -- "$@"
fi
#DEBHELPER#
+10
View File
@@ -0,0 +1,10 @@
#!/bin/sh
set -e
# Detect if /etc/hostid is a conffile of previous spl package, migrate if yes
# hostid file should preserve even when package is purged
if $(dpkg-query --showformat='${Conffiles}\n' --show spl >/dev/null 2>&1); then
dpkg-maintscript-helper rm_conffile /etc/hostid -- "$@"
fi
#DEBHELPER#
+2
View File
@@ -0,0 +1,2 @@
version=3
https://zfsonlinux.org/ .*spl-([\d\.]+)\.tar\.gz$
Submodule
+1
Submodule spl/upstream added at d665632350
-270
View File
@@ -1,270 +0,0 @@
zfs-linux (0.7.11-pve2~bpo1) unstable; urgency=medium
* update ZFS to debian/0.7.11-3
* Cherry-pick two fixes planned for 0.7.12
* Backport deadlock fix between mm_sem and tx assign in zfs_write() and page
fault
* Fix missing Breaks/Replaces in zfs-initramfs
-- Proxmox Support Team <support@proxmox.com> Mon, 29 Oct 2018 13:45:49 +0100
zfs-linux (0.7.11-pve1~bpo1) unstable; urgency=medium
* update ZFS to 0.7.11
-- Proxmox Support Team <support@proxmox.com> Mon, 10 Sep 2018 15:47:31 +0200
zfs-linux (0.7.9-pve3~bpo9) unstable; urgency=medium
* Cherry-pick fix for deadlock umount/snapentry_expire
-- Proxmox Support Team <support@proxmox.com> Fri, 03 Aug 2018 11:41:11 +0200
zfs-linux (0.7.9-pve2~bpo9) unstable; urgency=medium
* Cherry-pick fix for zpl_mount deadlock
-- Proxmox Support Team <support@proxmox.com> Fri, 12 Jul 2018 12:37:50 +0200
zfs-linux (0.7.9-pve1~bpo9) unstable; urgency=medium
* update ZFS to 0.7.9
-- Proxmox Support Team <support@proxmox.com> Fri, 18 May 2018 13:49:09 +0200
zfs-linux (0.7.8-pve1~bpo9) unstable; urgency=medium
* update ZFS to 0.7.8 (no changes)
-- Proxmox Support Team <support@proxmox.com> Mon, 30 Apr 2018 13:47:22 +0200
zfs-linux (0.7.7-pve2~bpo9) unstable; urgency=medium
* (temporarily) revert likely cause of #7401
-- Proxmox Support Team <support@proxmox.com> Mon, 9 Apr 2018 09:49:27 +0200
zfs-linux (0.7.7-pve1~bpo9) unstable; urgency=medium
* update ZFS to 0.7.7
-- Proxmox Support Team <support@proxmox.com> Tue, 3 Apr 2018 14:28:35 +0200
zfs-linux (0.7.6-pve1~bpo9) unstable; urgency=medium
* update ZFS to 0.7.6
-- Proxmox Support Team <support@proxmox.com> Wed, 21 Feb 2018 09:48:29 +0100
zfs-linux (0.7.4-pve2~bpo9) unstable; urgency=medium
* cherry-pick ARC hit rate fix from 0.7.6
* always load ZFS modules on boot
-- Proxmox Support Team <support@proxmox.com> Fri, 19 Jan 2018 11:31:14 +0100
zfs-linux (0.7.4-pve1~bpo9) unstable; urgency=medium
* update ZFS to 0.7.4
-- Proxmox Support Team <support@proxmox.com> Tue, 19 Dec 2017 12:52:42 +0100
zfs-linux (0.7.3-pve1~bpo9) unstable; urgency=medium
* update ZFS to 0.7.3
* include fix for user namespace setgid issue (#6800 upstream)
-- Proxmox Support Team <support@proxmox.com> Mon, 6 Nov 2017 09:14:19 +0100
zfs-linux (0.7.2-pve1~bpo90) unstable; urgency=medium
* update zfs to 0.7.2
* add PR 6616 - zfs send/recv compatibility with 0.6.5
* add PR 6695 - zfs recv: don't skip over objects which should be freed
* various small fixes
* split test suite into zfs-test package
-- Proxmox Support Team <support@proxmox.com> Thu, 12 Oct 2017 13:34:05 +0200
zfs-linux (0.6.5.11-pve18~bpo90) unstable; urgency=medium
* fix #1509: arc_summary error with L2ARC
-- Proxmox Support Team <support@proxmox.com> Mon, 2 Oct 2017 09:30:41 +0200
zfs-linux (0.6.5.11-pve17~bpo90) unstable; urgency=medium
* update zfs to debian/0.6.5.11-1
-- Proxmox Support Team <support@proxmox.com> Thu, 27 Jul 2017 13:11:45 +0200
zfs-linux (0.6.5.9-pve16~bpo90) unstable; urgency=medium
* drop transitional packages
* convert python scripts to python 3, drop .py suffix
* rebuild for PVE 5.0 / Debian Stretch
-- Proxmox Support Team <support@proxmox.com> Fri, 10 Mar 2017 09:45:52 +0100
zfs-linux (0.6.5.9-pve15~bpo80) unstable; urgency=medium
* update zfs to debian/0.6.5.9-1
-- Proxmox Support Team <support@proxmox.com> Mon, 6 Feb 2017 12:39:10 +0100
zfs-linux (0.6.5.8-pve14~bpo80) unstable; urgency=medium
* update zfs to debian/0.6.5.8-3
-- Proxmox Support Team <support@proxmox.com> Wed, 11 Jan 2017 09:34:58 +0100
zfs-linux (0.6.5.8-pve13~bpo80) unstable; urgency=medium
* fix #1184: zfs-share.service has wrong path to 'rm' command
-- Proxmox Support Team <support@proxmox.com> Thu, 27 Oct 2016 11:25:47 +0200
zfs-linux (0.6.5.8-pve12~bpo80) unstable; urgency=medium
* import with "-d /dev/disk/by-id" in systemd service
-- Proxmox Support Team <support@proxmox.com> Mon, 24 Oct 2016 13:48:08 +0200
zfs-linux (0.6.5.8-pve11~bpo80) unstable; urgency=medium
* update zfs to debian/0.6.5.8-1
* switch package upstream sources to Debian (Jessie)
* add transitional packages for upgrades
-- Proxmox Support Team <support@proxmox.com> Wed, 12 Oct 2016 11:16:02 +0200
zfs-linux (0.6.5.7-pve10~bpo80) unstable; urgency=medium
* update to pkg-zfs jessie/0.6.5.7-8
* change package versioning to allow upgrades from PVE3/wheezy
-- Proxmox Support Team <support@proxmox.com> Tue, 28 Jun 2016 09:50:00 +0200
zfs-linux (0.6.5-pve9~jessie) unstable; urgency=medium
* Update to pkg-zfs jessie/0.6.5.6-3
-- Proxmox Support Team <support@proxmox.com> Tue, 12 Apr 2016 09:51:35 +0200
zfs-linux (0.6.5-pve8~jessie) unstable; urgency=medium
* update to zfs-0.6.5.6
-- Proxmox Support Team <support@proxmox.com> Wed, 30 Mar 2016 10:50:22 +0200
zfs-linux (0.6.5-pve7~jessie) unstable; urgency=medium
* update to zfs-0.6.5.4
-- Proxmox Support Team <support@proxmox.com> Thu, 14 Jan 2016 10:51:17 +0100
zfs-linux (0.6.5-pve6~jessie) unstable; urgency=medium
* Prepare to tag zfs-0.6.5.3
-- Proxmox Support Team <support@proxmox.com> Thu, 05 Nov 2015 07:21:44 +0100
zfs-linux (0.6.5-pve5~jessie) unstable; urgency=medium
* Illumos 6267 - dn_bonus evicted too early
* Fix use-after-free in vdev_disk_physio_completion
-- Proxmox Support Team <support@proxmox.com> Wed, 04 Nov 2015 17:42:42 +0100
zfs-linux (0.6.5-pve4~jessie) unstable; urgency=medium
* update to master/debian/jessie/0.6.5.2-2
-- Proxmox Support Team <support@proxmox.com> Mon, 05 Oct 2015 17:56:45 +0200
zfs-linux (0.6.5-pve3~jessie) unstable; urgency=medium
* do not install /etc/init.d/zfs-zed script to avoid double startup
with systemd
-- Proxmox Support Team <support@proxmox.com> Mon, 28 Sep 2015 10:14:40 +0200
zfs-linux (0.6.5-pve2~jessie) unstable; urgency=medium
* update to master/debian/jessie/0.6.5.1-4
-- Proxmox Support Team <support@proxmox.com> Thu, 24 Sep 2015 12:45:33 +0200
zfs-linux (0.6.5-pve1~jessie) unstable; urgency=medium
* update to master/debian/jessie/0.6.5.1-2
-- Proxmox Support Team <support@proxmox.com> Mon, 21 Sep 2015 10:02:41 +0200
zfs-linux (0.6.4-pve3~jessie) unstable; urgency=medium
* update to snapshot/debian/jessie/0.6.4-24-6bec43
-- Proxmox Support Team <support@proxmox.com> Mon, 24 Aug 2015 06:13:44 +0200
zfs-linux (0.6.4-pve2~jessie) unstable; urgency=medium
* update to 0.6.4.2 (snapshot/debian/jessie/0.6.4-21-53b1d9)
-- Proxmox Support Team <support@proxmox.com> Tue, 21 Jul 2015 11:03:21 +0200
zfs-linux (0.6.4-pve1~jessie) unstable; urgency=medium
* update to 0.6.4 (use upstream zol package definitions)
-- Proxmox Support Team <support@proxmox.com> Tue, 14 Apr 2015 10:19:22 +0200
zfs-linux (0.6.3-pve3~jessie) unstable; urgency=medium
* ignore zfs-import-scan errors
-- Proxmox Support Team <support@proxmox.com> Tue, 31 Mar 2015 17:52:40 +0200
zfs-linux (0.6.3-pve2~jessie) unstable; urgency=medium
* use systemd for service startup
* install zed configuration file /etc/zfs/zed.d/zed.rc
-- Proxmox Support Team <support@proxmox.com> Sun, 15 Mar 2015 15:52:54 +0100
zfs-linux (0.6.3-pve1~jessie) unstable; urgency=medium
* recompile on jessie
-- Proxmox Support Team <support@proxmox.com> Sat, 28 Feb 2015 16:02:08 +0100
zfs-linux (0.6.3-2~wheezy) unstable; urgency=low
* use /sbin/modprobe to avoid warning inside initrd
* fix warning about undefined values inside initrd
-- Proxmox Support Team <support@proxmox.com> Wed, 14 Jan 2015 19:03:04 +0100
zfs-linux (0.6.3-1~wheezy) unstable; urgency=low
* first version for Proxmox VE
-- Proxmox Support Team <support@proxmox.com> Sat, 10 Jan 2015 14:06:34 +0100
Submodule zfs-debian deleted from 5b01147829
@@ -1,289 +0,0 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Fabian=20Gr=C3=BCnbichler?= <f.gruenbichler@proxmox.com>
Date: Mon, 6 Feb 2017 11:03:10 +0100
Subject: [PATCH] remove DKMS, modules and dracut build
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
---
debian/control | 1 -
debian/control.in | 40 ++-----------------
debian/not-installed | 2 +
debian/rules | 108 +--------------------------------------------------
4 files changed, 7 insertions(+), 144 deletions(-)
diff --git a/debian/control b/debian/control
index 4d22ff50..f33008df 100644
--- a/debian/control
+++ b/debian/control
@@ -10,7 +10,6 @@ Build-Depends: autotools-dev,
debhelper (>= 10.2),
dh-autoreconf,
dh-python,
- dkms (>> 2.1.1.2-5),
libattr1-dev,
libblkid-dev,
libselinux1-dev,
diff --git a/debian/control.in b/debian/control.in
index 96154c5c..0a9ceef6 100644
--- a/debian/control.in
+++ b/debian/control.in
@@ -10,7 +10,6 @@ Build-Depends: autotools-dev,
debhelper (>= 10.2),
dh-autoreconf,
dh-python,
- dkms (>> 2.1.1.2-5),
libattr1-dev,
libblkid-dev,
libselinux1-dev,
@@ -95,26 +94,10 @@ Description: OpenZFS pool library for Linux
.
This zpool library provides support for managing zpools.
-Package: zfs-dkms
-Architecture: all
-Pre-Depends: spl-dkms (<<${source:Upstream-Version}.),
- spl-dkms (>= ${source:Upstream-Version})
-Depends: dkms (>> 2.1.1.2-5), lsb-release, ${misc:Depends}
-Recommends: zfs-zed, zfsutils-linux (>= ${binary:Version}), @LINUX_COMPAT@
-Provides: zfs-modules
-Description: OpenZFS filesystem kernel modules for Linux
- The Z file system is a pooled filesystem designed for maximum data
- integrity, supporting data snapshots, multiple copies, and data
- checksums.
- .
- This DKMS package includes the SPA, DMU, ZVOL, and ZPL components of
- OpenZFS.
-
Package: zfs-initramfs
Architecture: all
Depends: busybox-initramfs | busybox-static | busybox,
initramfs-tools,
- zfs-modules | zfs-dkms,
zfsutils-linux (>= ${binary:Version}),
${misc:Depends}
Description: OpenZFS root filesystem capabilities for Linux - initramfs
@@ -125,29 +108,14 @@ Description: OpenZFS root filesystem capabilities for Linux - initramfs
This package adds OpenZFS to the system initramfs with a hook
for the initramfs-tools infrastructure.
-Package: zfs-dracut
-Architecture: all
-Depends: dracut,
- zfs-modules | zfs-dkms,
- zfsutils-linux (>= ${binary:Version}),
- ${misc:Depends}
-Description: OpenZFS root filesystem capabilities for Linux - dracut
- The Z file system is a pooled filesystem designed for maximum data
- integrity, supporting data snapshots, multiple copies, and data
- checksums.
- .
- This package adds OpenZFS to the system initramfs with a hook
- for the dracut infrastructure.
-
Package: zfsutils-linux
Section: contrib/admin
Architecture: linux-any
Depends: python3, ${misc:Depends}, ${python3:Depends}, ${shlibs:Depends}
-Recommends: lsb-base, zfs-modules | zfs-dkms, zfs-zed
-Breaks: zfs-dkms (<< ${binary:Version}), zfs-dkms (>> ${binary:Version})
+Recommends: lsb-base, zfs-zed
Suggests: nfs-kernel-server,
samba-common-bin (>= 3.0.23),
- zfs-initramfs | zfs-dracut
+ zfs-initramfs
Conflicts: zfs, zfs-fuse
Provides: zfsutils
Description: command-line tools to manage OpenZFS filesystems
@@ -161,8 +129,7 @@ Description: command-line tools to manage OpenZFS filesystems
Package: zfs-zed
Section: contrib/admin
Architecture: linux-any
-Depends: zfs-modules | zfs-dkms,
- zfsutils-linux (>= ${binary:Version}),
+Depends: zfsutils-linux (>= ${binary:Version}),
${misc:Depends},
${shlibs:Depends}
Description: OpenZFS Event Daemon
@@ -185,7 +152,6 @@ Depends: acl,
python,
sudo,
sysstat,
- zfs-modules | zfs-dkms,
zfsutils-linux (>=${binary:Version}),
${misc:Depends},
${shlibs:Depends}
diff --git a/debian/not-installed b/debian/not-installed
index a008985c..c60b0016 100644
--- a/debian/not-installed
+++ b/debian/not-installed
@@ -6,3 +6,5 @@ etc/zfs/vdev_id.conf.alias.example
etc/zfs/vdev_id.conf.multipath.example
etc/zfs/vdev_id.conf.sas_direct.example
etc/zfs/vdev_id.conf.sas_switch.example
+usr/lib/dracut
+usr/share/zfs/enum-extract.pl
diff --git a/debian/rules b/debian/rules
index 21b8ccef..3ba4b99a 100755
--- a/debian/rules
+++ b/debian/rules
@@ -6,9 +6,6 @@ LSB_DISTRIBUTOR := $(shell lsb_release -is)
NAME := $(shell awk '$$1 == "Name:" { print $$2; }' META)
LINUX_NEXT := $(shell awk -F. '{print $$1 "." $$2+1}' debian/linux_compat)
-DKMSFILES := module include config zfs.release.in autogen.sh META AUTHORS \
- DISCLAIMER COPYRIGHT OPENSOLARIS.LICENSE README.markdown
-
ifndef BUILD_UDEB
BUILD_UDEB=false
endif
@@ -18,21 +15,16 @@ KVERS=$(shell uname -r)
endif
ifndef SPL
-SPL=/usr/src/spl-$(DEB_VERSION_UPSTREAM)
-endif
-
-ifndef SPLOBJ
-SPLOBJ=/var/lib/dkms/spl/$(DEB_VERSION_UPSTREAM)/$(KVERS)/$(DEB_TARGET_GNU_CPU)
+SPL=../spl-build
endif
non_epoch_version=$(shell echo $(KVERS) | perl -pe 's/^\d+://')
PACKAGE=zfs
-pmodules = $(PACKAGE)-modules-$(non_epoch_version)
export DEB_BUILD_MAINT_OPTIONS = hardening=+all
%:
- dh $@ --with autoreconf,dkms,python3,systemd --parallel
+ dh $@ --with autoreconf,python3,systemd --parallel
override_dh_autoreconf:
@# Embed the downstream version in the module.
@@ -57,14 +49,6 @@ endif
--with-systemdpresetdir=/lib/systemd/system-preset \
--with-config=user
-override_dh_auto_build:
- @# Get a bare copy of the source code for DKMS.
- @# This creates the $(CURDIR)/$(NAME)-$(DEB_VERSION_UPSTREAM)/ tree, which does not
- @# contain the userland sources. NB: Remove-userland-dist-rules.patch
- $(MAKE) distdir
-
- dh_auto_build
-
override_dh_auto_test:
# The dh_auto_test rule is disabled because
# `make check` cannot run in an unprivileged build environment.
@@ -88,29 +72,6 @@ override_dh_auto_install:
@# Zed has dependencies outside of the system root.
mv '$(CURDIR)/debian/tmp/sbin/zed' '$(CURDIR)/debian/tmp/usr/sbin/zed'
- @# Install the DKMS source.
- @# We only want the files needed to build the modules
- install -D -t '$(CURDIR)/debian/tmp/usr/src/$(NAME)-$(DEB_VERSION_UPSTREAM)/scripts' \
- '$(CURDIR)/scripts/enum-extract.pl' \
- '$(CURDIR)/scripts/dkms.postbuild'
- $(foreach file,$(DKMSFILES),mv '$(CURDIR)/$(NAME)-$(DEB_VERSION_UPSTREAM)/$(file)' '$(CURDIR)/debian/tmp/usr/src/$(NAME)-$(DEB_VERSION_UPSTREAM)' || exit 1;)
- @# Hellish awk line:
- @# * Deletes from configure.ac the parts not needed for building the kernel module
- @# * It deletes from inside AC_CONFIG_FILES([]) everything except:
- @# (Makefile$|include/|module/|*.release$)
- @# * Takes care of spaces and tabs
- @# * Remove reference to ZFS_AC_PACKAGE
- awk '/^AC_CONFIG_FILES\(\[/,/^\]\)/ { if ($$0 !~ /^(AC_CONFIG_FILES\(\[([ \t]+)?$$|\]\)([ \t]+)?$$|([ \t]+)?(include\/|module\/|Makefile([ \t]+)?$$|zfs\.release([ \t]+)?$$))/){next} } {print}' \
- '$(CURDIR)/$(NAME)-$(DEB_VERSION_UPSTREAM)/configure.ac' | sed '/ZFS_AC_PACKAGE/d' > '$(CURDIR)/debian/tmp/usr/src/$(NAME)-$(DEB_VERSION_UPSTREAM)/configure.ac'
- @# Set "SUBDIRS = module include" for CONFIG_KERNEL and remove SUBDIRS for all other configs.
- sed '1,/CONFIG_KERNEL/s/SUBDIRS.*=.*//g;s/SUBDIRS.*=.*/SUBDIRS = module include/g;' \
- '$(CURDIR)/$(NAME)-$(DEB_VERSION_UPSTREAM)/Makefile.am' > '$(CURDIR)/debian/tmp/usr/src/$(NAME)-$(DEB_VERSION_UPSTREAM)/Makefile.am'
- @# Sanity test
- grep -q 'SUBDIRS = module include' '$(CURDIR)/debian/tmp/usr/src/$(NAME)-$(DEB_VERSION_UPSTREAM)/Makefile.am'
- @# Run autogen on the stripped source tree
- cd '$(CURDIR)/debian/tmp/usr/src/$(NAME)-$(DEB_VERSION_UPSTREAM)'; ./autogen.sh
- rm -fr '$(CURDIR)/debian/tmp/usr/src/$(NAME)-$(DEB_VERSION_UPSTREAM)/autom4te.cache'
-
mkdir -p $(CURDIR)/debian/tmp/usr/lib
for i in `ls $(CURDIR)/debian/tmp/lib/*.so`; do \
ln -s /lib/`readlink $${i}` $(CURDIR)/debian/tmp/usr/lib/`basename $${i}`; \
@@ -121,11 +82,6 @@ override_dh_auto_install:
chmod a-x '$(CURDIR)/debian/tmp/etc/default/zfs'
chmod a-x '$(CURDIR)/debian/tmp/usr/share/bash-completion/completions/zfs'
-override_dh_dkms:
- '$(CURDIR)/scripts/dkms.mkconf' -n $(NAME) -v $(DEB_VERSION_UPSTREAM) -f '$(CURDIR)/scripts/zfs-dkms.dkms'
- dh_dkms
- rm -f '$(CURDIR)/scripts/zfs-dkms.dkms'
-
override_dh_makeshlibs:
dh_makeshlibs -a -V
ifeq ($(BUILD_UDEB), true)
@@ -171,65 +127,5 @@ override_dh_installinit:
# ------------
-override_dh_prep-deb-files:
- for templ in $(wildcard $(CURDIR)/debian/*_KVERS_*.in); do \
- sed -e 's/##KVERS##/$(KVERS)/g ; s/#KVERS#/$(KVERS)/g ; s/_KVERS_/$(KVERS)/g ; s/##KDREV##/$(KDREV)/g ; s/#KDREV#/$(KDREV)/g ; s/_KDREV_/$(KDREV)/g ; s/_ARCH_/$(DEB_HOST_ARCH)/' \
- < $$templ > `echo $$templ | sed -e 's/_KVERS_/$(KVERS)/g ; s/_ARCH_/$(DEB_HOST_ARCH)/g ; s/\.in$$//'` ; \
- done
- sed -e 's/##KVERS##/$(KVERS)/g ; s/#KVERS#/$(KVERS)/g ; s/_KVERS_/$(KVERS)/g ; s/##KDREV##/$(KDREV)/g ; s/#KDREV#/$(KDREV)/g ; s/_KDREV_/$(KDREV)/g ; s/_ARCH_/$(DEB_HOST_ARCH)/g' \
- < debian/control.modules.in > debian/control
-
-override_dh_configure_modules_udeb: override_dh_configure_modules_udeb_stamp
-override_dh_configure_modules_udeb_stamp:
- ./configure \
- --without-selinux \
- --with-config=kernel \
- --with-linux=$(KSRC) \
- --with-linux-obj=$(KOBJ) \
- --with-spl=$(SPL) \
- --with-spl-obj=$(SPLOBJ)
- touch override_dh_configure_modules_udeb_stamp
-
-override_dh_configure_modules: override_dh_configure_modules_stamp
-override_dh_configure_modules_stamp:
- ./configure \
- --with-config=kernel \
- --with-linux=$(KSRC) \
- --with-linux-obj=$(KOBJ) \
- --with-spl=$(SPL) \
- --with-spl-obj=$(SPLOBJ)
- touch override_dh_configure_modules_stamp
-
-override_dh_binary-modules-udeb: override_dh_prep-deb-files override_dh_configure_modules_udeb
- dh_testdir
- dh_testroot
- dh_prep
-
- $(MAKE) -C $(CURDIR)/module modules
-
- dh_installdirs -p${pmodules}-di
- dh_install -p${pmodules}-di
- dh_gencontrol -p${pmodules}-di
-
- dh_builddeb -p${pmodules}-di
-
-override_dh_binary-modules: override_dh_prep-deb-files override_dh_configure_modules
- dh_testdir
- dh_testroot
- dh_prep
-
- $(MAKE) -C $(CURDIR)/module modules
-
- dh_install -p${pmodules}
- dh_installdocs -p${pmodules}
- dh_installchangelogs -p${pmodules}
- dh_compress -p${pmodules}
- dh_strip -p${pmodules}
- dh_fixperms -p${pmodules}
- dh_installdeb -p${pmodules}
- dh_gencontrol -p${pmodules}
- dh_md5sums -p${pmodules}
- dh_builddeb -p${pmodules}
-
debian-copyright:
cme update dpkg-copyright -file debian/copyright.cme
@@ -1,256 +0,0 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Tom Caputi <tcaputi@datto.com>
Date: Tue, 10 Apr 2018 14:15:05 -0400
Subject: [PATCH] Fix race in dnode_check_slots_free()
Currently, dnode_check_slots_free() works by checking dn->dn_type
in the dnode to determine if the dnode is reclaimable. However,
there is a small window of time between dnode_free_sync() in the
first call to dsl_dataset_sync() and when the useraccounting code
is run when the type is set DMU_OT_NONE, but the dnode is not yet
evictable, leading to crashes. This patch adds the ability for
dnodes to track which txg they were last dirtied in and adds a
check for this before performing the reclaim.
This patch also corrects several instances when dn_dirty_link was
treated as a list_node_t when it is technically a multilist_node_t.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Tom Caputi <tcaputi@datto.com>
Requires-spl: spl-0.7-release
Issue #7147
Issue #7388
Issue #7997
(cherry-picked from behlendorf/issue-7997 4764f6f3be90be073d2700653dff286371e52583)
Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
---
include/sys/dmu_impl.h | 1 +
include/sys/dnode.h | 4 ++++
module/zfs/dbuf.c | 3 +++
module/zfs/dmu.c | 2 +-
module/zfs/dmu_objset.c | 15 +++++++++++++++
module/zfs/dnode.c | 29 +++++++++++++++++++----------
6 files changed, 43 insertions(+), 11 deletions(-)
diff --git a/include/sys/dmu_impl.h b/include/sys/dmu_impl.h
index 65e417e3..03a63077 100644
--- a/include/sys/dmu_impl.h
+++ b/include/sys/dmu_impl.h
@@ -161,6 +161,7 @@ extern "C" {
* dn_allocated_txg
* dn_free_txg
* dn_assigned_txg
+ * dn_dirty_txg
* dd_assigned_tx
* dn_notxholds
* dn_dirtyctx
diff --git a/include/sys/dnode.h b/include/sys/dnode.h
index ea7defe1..2dd087b3 100644
--- a/include/sys/dnode.h
+++ b/include/sys/dnode.h
@@ -260,6 +260,7 @@ struct dnode {
uint64_t dn_allocated_txg;
uint64_t dn_free_txg;
uint64_t dn_assigned_txg;
+ uint64_t dn_dirty_txg; /* txg dnode was last dirtied */
kcondvar_t dn_notxholds;
enum dnode_dirtycontext dn_dirtyctx;
uint8_t *dn_dirtyctx_firstset; /* dbg: contents meaningless */
@@ -362,6 +363,9 @@ void dnode_evict_dbufs(dnode_t *dn);
void dnode_evict_bonus(dnode_t *dn);
void dnode_free_interior_slots(dnode_t *dn);
+#define DNODE_IS_DIRTY(_dn) \
+ ((_dn)->dn_dirty_txg >= spa_syncing_txg((_dn)->dn_objset->os_spa))
+
#define DNODE_IS_CACHEABLE(_dn) \
((_dn)->dn_objset->os_primary_cache == ZFS_CACHE_ALL || \
(DMU_OT_IS_METADATA((_dn)->dn_type) && \
diff --git a/module/zfs/dbuf.c b/module/zfs/dbuf.c
index 4ee121f5..6edb39d6 100644
--- a/module/zfs/dbuf.c
+++ b/module/zfs/dbuf.c
@@ -1606,6 +1606,9 @@ dbuf_dirty(dmu_buf_impl_t *db, dmu_tx_t *tx)
FTAG);
}
}
+
+ if (tx->tx_txg > dn->dn_dirty_txg)
+ dn->dn_dirty_txg = tx->tx_txg;
mutex_exit(&dn->dn_mtx);
if (db->db_blkid == DMU_SPILL_BLKID)
diff --git a/module/zfs/dmu.c b/module/zfs/dmu.c
index 6f09aa2f..a09ac4f9 100644
--- a/module/zfs/dmu.c
+++ b/module/zfs/dmu.c
@@ -2044,7 +2044,7 @@ dmu_offset_next(objset_t *os, uint64_t object, boolean_t hole, uint64_t *off)
* Check if dnode is dirty
*/
for (i = 0; i < TXG_SIZE; i++) {
- if (list_link_active(&dn->dn_dirty_link[i])) {
+ if (multilist_link_active(&dn->dn_dirty_link[i])) {
clean = B_FALSE;
break;
}
diff --git a/module/zfs/dmu_objset.c b/module/zfs/dmu_objset.c
index 449ebedf..0bed2d3e 100644
--- a/module/zfs/dmu_objset.c
+++ b/module/zfs/dmu_objset.c
@@ -1213,10 +1213,23 @@ dmu_objset_sync_dnodes(multilist_sublist_t *list, dmu_tx_t *tx)
ASSERT3U(dn->dn_nlevels, <=, DN_MAX_LEVELS);
multilist_sublist_remove(list, dn);
+ /*
+ * If we are not doing useraccounting (os_synced_dnodes == NULL)
+ * we are done with this dnode for this txg. Unset dn_dirty_txg
+ * if later txgs aren't dirtying it so that future holders do
+ * not get a stale value. Otherwise, we will do this in
+ * userquota_updates_task() when processing has completely
+ * finished for this txg.
+ */
multilist_t *newlist = dn->dn_objset->os_synced_dnodes;
if (newlist != NULL) {
(void) dnode_add_ref(dn, newlist);
multilist_insert(newlist, dn);
+ } else {
+ mutex_enter(&dn->dn_mtx);
+ if (dn->dn_dirty_txg == tx->tx_txg)
+ dn->dn_dirty_txg = 0;
+ mutex_exit(&dn->dn_mtx);
}
dnode_sync(dn, tx);
@@ -1621,6 +1634,8 @@ userquota_updates_task(void *arg)
dn->dn_id_flags |= DN_ID_CHKED_BONUS;
}
dn->dn_id_flags &= ~(DN_ID_NEW_EXIST);
+ if (dn->dn_dirty_txg == spa_syncing_txg(os->os_spa))
+ dn->dn_dirty_txg = 0;
mutex_exit(&dn->dn_mtx);
multilist_sublist_remove(list, dn);
diff --git a/module/zfs/dnode.c b/module/zfs/dnode.c
index d465b545..4a169c49 100644
--- a/module/zfs/dnode.c
+++ b/module/zfs/dnode.c
@@ -137,7 +137,7 @@ dnode_cons(void *arg, void *unused, int kmflag)
bzero(&dn->dn_next_blksz[0], sizeof (dn->dn_next_blksz));
for (i = 0; i < TXG_SIZE; i++) {
- list_link_init(&dn->dn_dirty_link[i]);
+ multilist_link_init(&dn->dn_dirty_link[i]);
dn->dn_free_ranges[i] = NULL;
list_create(&dn->dn_dirty_records[i],
sizeof (dbuf_dirty_record_t),
@@ -147,6 +147,7 @@ dnode_cons(void *arg, void *unused, int kmflag)
dn->dn_allocated_txg = 0;
dn->dn_free_txg = 0;
dn->dn_assigned_txg = 0;
+ dn->dn_dirty_txg = 0;
dn->dn_dirtyctx = 0;
dn->dn_dirtyctx_firstset = NULL;
dn->dn_bonus = NULL;
@@ -184,7 +185,7 @@ dnode_dest(void *arg, void *unused)
ASSERT(!list_link_active(&dn->dn_link));
for (i = 0; i < TXG_SIZE; i++) {
- ASSERT(!list_link_active(&dn->dn_dirty_link[i]));
+ ASSERT(!multilist_link_active(&dn->dn_dirty_link[i]));
ASSERT3P(dn->dn_free_ranges[i], ==, NULL);
list_destroy(&dn->dn_dirty_records[i]);
ASSERT0(dn->dn_next_nblkptr[i]);
@@ -199,6 +200,7 @@ dnode_dest(void *arg, void *unused)
ASSERT0(dn->dn_allocated_txg);
ASSERT0(dn->dn_free_txg);
ASSERT0(dn->dn_assigned_txg);
+ ASSERT0(dn->dn_dirty_txg);
ASSERT0(dn->dn_dirtyctx);
ASSERT3P(dn->dn_dirtyctx_firstset, ==, NULL);
ASSERT3P(dn->dn_bonus, ==, NULL);
@@ -523,6 +525,7 @@ dnode_destroy(dnode_t *dn)
dn->dn_allocated_txg = 0;
dn->dn_free_txg = 0;
dn->dn_assigned_txg = 0;
+ dn->dn_dirty_txg = 0;
dn->dn_dirtyctx = 0;
if (dn->dn_dirtyctx_firstset != NULL) {
@@ -592,6 +595,7 @@ dnode_allocate(dnode_t *dn, dmu_object_type_t ot, int blocksize, int ibs,
ASSERT0(dn->dn_maxblkid);
ASSERT0(dn->dn_allocated_txg);
ASSERT0(dn->dn_assigned_txg);
+ ASSERT0(dn->dn_dirty_txg);
ASSERT(refcount_is_zero(&dn->dn_tx_holds));
ASSERT3U(refcount_count(&dn->dn_holds), <=, 1);
ASSERT(avl_is_empty(&dn->dn_dbufs));
@@ -604,7 +608,7 @@ dnode_allocate(dnode_t *dn, dmu_object_type_t ot, int blocksize, int ibs,
ASSERT0(dn->dn_next_bonustype[i]);
ASSERT0(dn->dn_rm_spillblk[i]);
ASSERT0(dn->dn_next_blksz[i]);
- ASSERT(!list_link_active(&dn->dn_dirty_link[i]));
+ ASSERT(!multilist_link_active(&dn->dn_dirty_link[i]));
ASSERT3P(list_head(&dn->dn_dirty_records[i]), ==, NULL);
ASSERT3P(dn->dn_free_ranges[i], ==, NULL);
}
@@ -779,6 +783,7 @@ dnode_move_impl(dnode_t *odn, dnode_t *ndn)
ndn->dn_allocated_txg = odn->dn_allocated_txg;
ndn->dn_free_txg = odn->dn_free_txg;
ndn->dn_assigned_txg = odn->dn_assigned_txg;
+ ndn->dn_dirty_txg = odn->dn_dirty_txg;
ndn->dn_dirtyctx = odn->dn_dirtyctx;
ndn->dn_dirtyctx_firstset = odn->dn_dirtyctx_firstset;
ASSERT(refcount_count(&odn->dn_tx_holds) == 0);
@@ -845,6 +850,7 @@ dnode_move_impl(dnode_t *odn, dnode_t *ndn)
odn->dn_allocated_txg = 0;
odn->dn_free_txg = 0;
odn->dn_assigned_txg = 0;
+ odn->dn_dirty_txg = 0;
odn->dn_dirtyctx = 0;
odn->dn_dirtyctx_firstset = NULL;
odn->dn_have_spill = B_FALSE;
@@ -1069,6 +1075,10 @@ dnode_check_slots_free(dnode_children_t *children, int idx, int slots)
{
ASSERT3S(idx + slots, <=, DNODES_PER_BLOCK);
+ /*
+ * If all dnode slots are either already free or
+ * evictable return B_TRUE.
+ */
for (int i = idx; i < idx + slots; i++) {
dnode_handle_t *dnh = &children->dnc_children[i];
dnode_t *dn = dnh->dnh_dnode;
@@ -1077,18 +1087,17 @@ dnode_check_slots_free(dnode_children_t *children, int idx, int slots)
continue;
} else if (DN_SLOT_IS_PTR(dn)) {
mutex_enter(&dn->dn_mtx);
- dmu_object_type_t type = dn->dn_type;
+ boolean_t can_free = (dn->dn_type == DMU_OT_NONE &&
+ !DNODE_IS_DIRTY(dn));
mutex_exit(&dn->dn_mtx);
- if (type != DMU_OT_NONE)
+ if (!can_free)
return (B_FALSE);
-
- continue;
+ else
+ continue;
} else {
return (B_FALSE);
}
-
- return (B_FALSE);
}
return (B_TRUE);
@@ -1594,7 +1603,7 @@ dnode_setdirty(dnode_t *dn, dmu_tx_t *tx)
/*
* If we are already marked dirty, we're done.
*/
- if (list_link_active(&dn->dn_dirty_link[txg & TXG_MASK])) {
+ if (multilist_link_active(&dn->dn_dirty_link[txg & TXG_MASK])) {
multilist_sublist_unlock(mls);
return;
}
@@ -1,861 +0,0 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Matthew Ahrens <mahrens@delphix.com>
Date: Thu, 2 Aug 2018 15:51:45 -0700
Subject: [PATCH] Reduce taskq and context-switch cost of zio pipe
When doing a read from disk, ZFS creates 3 ZIO's: a zio_null(), the
logical zio_read(), and then a physical zio. Currently, each of these
results in a separate taskq_dispatch(zio_execute).
On high-read-iops workloads, this causes a significant performance
impact. By processing all 3 ZIO's in a single taskq entry, we reduce the
overhead on taskq locking and context switching. We accomplish this by
allowing zio_done() to return a "next zio to execute" to zio_execute().
This results in a ~12% performance increase for random reads, from
96,000 iops to 108,000 iops (with recordsize=8k, on SSD's).
Reviewed by: Pavel Zakharov <pavel.zakharov@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed by: George Wilson <george.wilson@delphix.com>
Signed-off-by: Matthew Ahrens <mahrens@delphix.com>
External-issue: DLPX-59292
Requires-spl: spl-0.7-release
Closes #7736
(cherry-picked from behlendorf/issue-7736 496657ab3bcfeb638b1786e1759980ccfcacb08e)
Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
---
include/sys/zio.h | 4 +-
module/zfs/zio.c | 250 +++++++++++++++++++++++++++++-------------------------
2 files changed, 137 insertions(+), 117 deletions(-)
diff --git a/include/sys/zio.h b/include/sys/zio.h
index 4b0eecc2..3618912c 100644
--- a/include/sys/zio.h
+++ b/include/sys/zio.h
@@ -237,7 +237,7 @@ enum zio_child {
#define ZIO_CHILD_DDT_BIT ZIO_CHILD_BIT(ZIO_CHILD_DDT)
#define ZIO_CHILD_LOGICAL_BIT ZIO_CHILD_BIT(ZIO_CHILD_LOGICAL)
#define ZIO_CHILD_ALL_BITS \
- (ZIO_CHILD_VDEV_BIT | ZIO_CHILD_GANG_BIT | \
+ (ZIO_CHILD_VDEV_BIT | ZIO_CHILD_GANG_BIT | \
ZIO_CHILD_DDT_BIT | ZIO_CHILD_LOGICAL_BIT)
enum zio_wait_type {
@@ -375,7 +375,7 @@ typedef struct zio_transform {
struct zio_transform *zt_next;
} zio_transform_t;
-typedef int zio_pipe_stage_t(zio_t *zio);
+typedef zio_t *zio_pipe_stage_t(zio_t *zio);
/*
* The io_reexecute flags are distinct from io_flags because the child must
diff --git a/module/zfs/zio.c b/module/zfs/zio.c
index 9a465e1b..b08b4747 100644
--- a/module/zfs/zio.c
+++ b/module/zfs/zio.c
@@ -75,9 +75,6 @@ uint64_t zio_buf_cache_frees[SPA_MAXBLOCKSIZE >> SPA_MINBLOCKSHIFT];
int zio_delay_max = ZIO_DELAY_MAX;
-#define ZIO_PIPELINE_CONTINUE 0x100
-#define ZIO_PIPELINE_STOP 0x101
-
#define BP_SPANB(indblkshift, level) \
(((uint64_t)1) << ((level) * ((indblkshift) - SPA_BLKPTRSHIFT)))
#define COMPARE_META_LEVEL 0x80000000ul
@@ -516,7 +513,8 @@ zio_wait_for_children(zio_t *zio, uint8_t childbits, enum zio_wait_type wait)
__attribute__((always_inline))
static inline void
-zio_notify_parent(zio_t *pio, zio_t *zio, enum zio_wait_type wait)
+zio_notify_parent(zio_t *pio, zio_t *zio, enum zio_wait_type wait,
+ zio_t **next_to_executep)
{
uint64_t *countp = &pio->io_children[zio->io_child_type][wait];
int *errorp = &pio->io_child_error[zio->io_child_type];
@@ -535,13 +533,33 @@ zio_notify_parent(zio_t *pio, zio_t *zio, enum zio_wait_type wait)
ZIO_TASKQ_INTERRUPT;
pio->io_stall = NULL;
mutex_exit(&pio->io_lock);
+
/*
- * Dispatch the parent zio in its own taskq so that
- * the child can continue to make progress. This also
- * prevents overflowing the stack when we have deeply nested
- * parent-child relationships.
+ * If we can tell the caller to execute this parent next, do
+ * so. Otherwise dispatch the parent zio as its own task.
+ *
+ * Having the caller execute the parent when possible reduces
+ * locking on the zio taskq's, reduces context switch
+ * overhead, and has no recursion penalty. Note that one
+ * read from disk typically causes at least 3 zio's: a
+ * zio_null(), the logical zio_read(), and then a physical
+ * zio. When the physical ZIO completes, we are able to call
+ * zio_done() on all 3 of these zio's from one invocation of
+ * zio_execute() by returning the parent back to
+ * zio_execute(). Since the parent isn't executed until this
+ * thread returns back to zio_execute(), the caller should do
+ * so promptly.
+ *
+ * In other cases, dispatching the parent prevents
+ * overflowing the stack when we have deeply nested
+ * parent-child relationships, as we do with the "mega zio"
+ * of writes for spa_sync(), and the chain of ZIL blocks.
*/
- zio_taskq_dispatch(pio, type, B_FALSE);
+ if (next_to_executep != NULL && *next_to_executep == NULL) {
+ *next_to_executep = pio;
+ } else {
+ zio_taskq_dispatch(pio, type, B_FALSE);
+ }
} else {
mutex_exit(&pio->io_lock);
}
@@ -1187,7 +1205,7 @@ zio_shrink(zio_t *zio, uint64_t size)
* ==========================================================================
*/
-static int
+static zio_t *
zio_read_bp_init(zio_t *zio)
{
blkptr_t *bp = zio->io_bp;
@@ -1221,15 +1239,15 @@ zio_read_bp_init(zio_t *zio)
if (BP_GET_DEDUP(bp) && zio->io_child_type == ZIO_CHILD_LOGICAL)
zio->io_pipeline = ZIO_DDT_READ_PIPELINE;
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
-static int
+static zio_t *
zio_write_bp_init(zio_t *zio)
{
if (!IO_IS_ALLOCATING(zio))
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
ASSERT(zio->io_child_type != ZIO_CHILD_DDT);
@@ -1244,7 +1262,7 @@ zio_write_bp_init(zio_t *zio)
zio->io_pipeline = ZIO_INTERLOCK_PIPELINE;
if (BP_IS_EMBEDDED(bp))
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
/*
* If we've been overridden and nopwrite is set then
@@ -1255,13 +1273,13 @@ zio_write_bp_init(zio_t *zio)
ASSERT(!zp->zp_dedup);
ASSERT3U(BP_GET_CHECKSUM(bp), ==, zp->zp_checksum);
zio->io_flags |= ZIO_FLAG_NOPWRITE;
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
ASSERT(!zp->zp_nopwrite);
if (BP_IS_HOLE(bp) || !zp->zp_dedup)
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
ASSERT((zio_checksum_table[zp->zp_checksum].ci_flags &
ZCHECKSUM_FLAG_DEDUP) || zp->zp_dedup_verify);
@@ -1269,7 +1287,7 @@ zio_write_bp_init(zio_t *zio)
if (BP_GET_CHECKSUM(bp) == zp->zp_checksum) {
BP_SET_DEDUP(bp, 1);
zio->io_pipeline |= ZIO_STAGE_DDT_WRITE;
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
/*
@@ -1281,10 +1299,10 @@ zio_write_bp_init(zio_t *zio)
zio->io_pipeline = zio->io_orig_pipeline;
}
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
-static int
+static zio_t *
zio_write_compress(zio_t *zio)
{
spa_t *spa = zio->io_spa;
@@ -1303,11 +1321,11 @@ zio_write_compress(zio_t *zio)
*/
if (zio_wait_for_children(zio, ZIO_CHILD_LOGICAL_BIT |
ZIO_CHILD_GANG_BIT, ZIO_WAIT_READY)) {
- return (ZIO_PIPELINE_STOP);
+ return (NULL);
}
if (!IO_IS_ALLOCATING(zio))
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
if (zio->io_children_ready != NULL) {
/*
@@ -1366,7 +1384,7 @@ zio_write_compress(zio_t *zio)
zio->io_pipeline = ZIO_INTERLOCK_PIPELINE;
ASSERT(spa_feature_is_active(spa,
SPA_FEATURE_EMBEDDED_DATA));
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
} else {
/*
* Round up compressed size up to the ashift
@@ -1459,10 +1477,10 @@ zio_write_compress(zio_t *zio)
zio->io_pipeline |= ZIO_STAGE_NOP_WRITE;
}
}
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
-static int
+static zio_t *
zio_free_bp_init(zio_t *zio)
{
blkptr_t *bp = zio->io_bp;
@@ -1472,7 +1490,7 @@ zio_free_bp_init(zio_t *zio)
zio->io_pipeline = ZIO_DDT_FREE_PIPELINE;
}
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
/*
@@ -1541,12 +1559,12 @@ zio_taskq_member(zio_t *zio, zio_taskq_type_t q)
return (B_FALSE);
}
-static int
+static zio_t *
zio_issue_async(zio_t *zio)
{
zio_taskq_dispatch(zio, ZIO_TASKQ_ISSUE, B_FALSE);
- return (ZIO_PIPELINE_STOP);
+ return (NULL);
}
void
@@ -1687,14 +1705,13 @@ __attribute__((always_inline))
static inline void
__zio_execute(zio_t *zio)
{
- zio->io_executor = curthread;
-
ASSERT3U(zio->io_queued_timestamp, >, 0);
while (zio->io_stage < ZIO_STAGE_DONE) {
enum zio_stage pipeline = zio->io_pipeline;
enum zio_stage stage = zio->io_stage;
- int rv;
+
+ zio->io_executor = curthread;
ASSERT(!MUTEX_HELD(&zio->io_lock));
ASSERT(ISP2(stage));
@@ -1736,12 +1753,16 @@ __zio_execute(zio_t *zio)
zio->io_stage = stage;
zio->io_pipeline_trace |= zio->io_stage;
- rv = zio_pipeline[highbit64(stage) - 1](zio);
- if (rv == ZIO_PIPELINE_STOP)
- return;
+ /*
+ * The zio pipeline stage returns the next zio to execute
+ * (typically the same as this one), or NULL if we should
+ * stop.
+ */
+ zio = zio_pipeline[highbit64(stage) - 1](zio);
- ASSERT(rv == ZIO_PIPELINE_CONTINUE);
+ if (zio == NULL)
+ return;
}
}
@@ -2215,7 +2236,7 @@ zio_gang_tree_issue(zio_t *pio, zio_gang_node_t *gn, blkptr_t *bp, abd_t *data,
zio_nowait(zio);
}
-static int
+static zio_t *
zio_gang_assemble(zio_t *zio)
{
blkptr_t *bp = zio->io_bp;
@@ -2227,16 +2248,16 @@ zio_gang_assemble(zio_t *zio)
zio_gang_tree_assemble(zio, bp, &zio->io_gang_tree);
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
-static int
+static zio_t *
zio_gang_issue(zio_t *zio)
{
blkptr_t *bp = zio->io_bp;
if (zio_wait_for_children(zio, ZIO_CHILD_GANG_BIT, ZIO_WAIT_DONE)) {
- return (ZIO_PIPELINE_STOP);
+ return (NULL);
}
ASSERT(BP_IS_GANG(bp) && zio->io_gang_leader == zio);
@@ -2250,7 +2271,7 @@ zio_gang_issue(zio_t *zio)
zio->io_pipeline = ZIO_INTERLOCK_PIPELINE;
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
static void
@@ -2290,7 +2311,7 @@ zio_write_gang_done(zio_t *zio)
abd_put(zio->io_abd);
}
-static int
+static zio_t *
zio_write_gang_block(zio_t *pio)
{
spa_t *spa = pio->io_spa;
@@ -2349,7 +2370,7 @@ zio_write_gang_block(zio_t *pio)
}
pio->io_error = error;
- return (ZIO_PIPELINE_CONTINUE);
+ return (pio);
}
if (pio == gio) {
@@ -2423,7 +2444,7 @@ zio_write_gang_block(zio_t *pio)
zio_nowait(zio);
- return (ZIO_PIPELINE_CONTINUE);
+ return (pio);
}
/*
@@ -2444,7 +2465,7 @@ zio_write_gang_block(zio_t *pio)
* used for nopwrite, assuming that the salt and the checksums
* themselves remain secret.
*/
-static int
+static zio_t *
zio_nop_write(zio_t *zio)
{
blkptr_t *bp = zio->io_bp;
@@ -2471,7 +2492,7 @@ zio_nop_write(zio_t *zio)
BP_GET_COMPRESS(bp) != BP_GET_COMPRESS(bp_orig) ||
BP_GET_DEDUP(bp) != BP_GET_DEDUP(bp_orig) ||
zp->zp_copies != BP_GET_NDVAS(bp_orig))
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
/*
* If the checksums match then reset the pipeline so that we
@@ -2491,7 +2512,7 @@ zio_nop_write(zio_t *zio)
zio->io_flags |= ZIO_FLAG_NOPWRITE;
}
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
/*
@@ -2519,7 +2540,7 @@ zio_ddt_child_read_done(zio_t *zio)
mutex_exit(&pio->io_lock);
}
-static int
+static zio_t *
zio_ddt_read_start(zio_t *zio)
{
blkptr_t *bp = zio->io_bp;
@@ -2540,7 +2561,7 @@ zio_ddt_read_start(zio_t *zio)
zio->io_vsd = dde;
if (ddp_self == NULL)
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
for (p = 0; p < DDT_PHYS_TYPES; p++, ddp++) {
if (ddp->ddp_phys_birth == 0 || ddp == ddp_self)
@@ -2553,23 +2574,23 @@ zio_ddt_read_start(zio_t *zio)
zio->io_priority, ZIO_DDT_CHILD_FLAGS(zio) |
ZIO_FLAG_DONT_PROPAGATE, &zio->io_bookmark));
}
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
zio_nowait(zio_read(zio, zio->io_spa, bp,
zio->io_abd, zio->io_size, NULL, NULL, zio->io_priority,
ZIO_DDT_CHILD_FLAGS(zio), &zio->io_bookmark));
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
-static int
+static zio_t *
zio_ddt_read_done(zio_t *zio)
{
blkptr_t *bp = zio->io_bp;
if (zio_wait_for_children(zio, ZIO_CHILD_DDT_BIT, ZIO_WAIT_DONE)) {
- return (ZIO_PIPELINE_STOP);
+ return (NULL);
}
ASSERT(BP_GET_DEDUP(bp));
@@ -2581,12 +2602,12 @@ zio_ddt_read_done(zio_t *zio)
ddt_entry_t *dde = zio->io_vsd;
if (ddt == NULL) {
ASSERT(spa_load_state(zio->io_spa) != SPA_LOAD_NONE);
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
if (dde == NULL) {
zio->io_stage = ZIO_STAGE_DDT_READ_START >> 1;
zio_taskq_dispatch(zio, ZIO_TASKQ_ISSUE, B_FALSE);
- return (ZIO_PIPELINE_STOP);
+ return (NULL);
}
if (dde->dde_repair_abd != NULL) {
abd_copy(zio->io_abd, dde->dde_repair_abd,
@@ -2599,7 +2620,7 @@ zio_ddt_read_done(zio_t *zio)
ASSERT(zio->io_vsd == NULL);
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
static boolean_t
@@ -2780,7 +2801,7 @@ zio_ddt_ditto_write_done(zio_t *zio)
ddt_exit(ddt);
}
-static int
+static zio_t *
zio_ddt_write(zio_t *zio)
{
spa_t *spa = zio->io_spa;
@@ -2822,7 +2843,7 @@ zio_ddt_write(zio_t *zio)
}
zio->io_pipeline = ZIO_WRITE_PIPELINE;
ddt_exit(ddt);
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
ditto_copies = ddt_ditto_copies_needed(ddt, dde, ddp);
@@ -2848,7 +2869,7 @@ zio_ddt_write(zio_t *zio)
zio->io_bp_override = NULL;
BP_ZERO(bp);
ddt_exit(ddt);
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
dio = zio_write(zio, spa, txg, bp, zio->io_orig_abd,
@@ -2890,12 +2911,12 @@ zio_ddt_write(zio_t *zio)
if (dio)
zio_nowait(dio);
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
ddt_entry_t *freedde; /* for debugging */
-static int
+static zio_t *
zio_ddt_free(zio_t *zio)
{
spa_t *spa = zio->io_spa;
@@ -2916,7 +2937,7 @@ zio_ddt_free(zio_t *zio)
}
ddt_exit(ddt);
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
/*
@@ -2953,7 +2974,7 @@ zio_io_to_allocate(spa_t *spa)
return (zio);
}
-static int
+static zio_t *
zio_dva_throttle(zio_t *zio)
{
spa_t *spa = zio->io_spa;
@@ -2963,7 +2984,7 @@ zio_dva_throttle(zio_t *zio)
!spa_normal_class(zio->io_spa)->mc_alloc_throttle_enabled ||
zio->io_child_type == ZIO_CHILD_GANG ||
zio->io_flags & ZIO_FLAG_NODATA) {
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
ASSERT(zio->io_child_type > ZIO_CHILD_GANG);
@@ -2979,22 +3000,7 @@ zio_dva_throttle(zio_t *zio)
nio = zio_io_to_allocate(zio->io_spa);
mutex_exit(&spa->spa_alloc_lock);
- if (nio == zio)
- return (ZIO_PIPELINE_CONTINUE);
-
- if (nio != NULL) {
- ASSERT(nio->io_stage == ZIO_STAGE_DVA_THROTTLE);
- /*
- * We are passing control to a new zio so make sure that
- * it is processed by a different thread. We do this to
- * avoid stack overflows that can occur when parents are
- * throttled and children are making progress. We allow
- * it to go to the head of the taskq since it's already
- * been waiting.
- */
- zio_taskq_dispatch(nio, ZIO_TASKQ_ISSUE, B_TRUE);
- }
- return (ZIO_PIPELINE_STOP);
+ return (nio);
}
void
@@ -3013,7 +3019,7 @@ zio_allocate_dispatch(spa_t *spa)
zio_taskq_dispatch(zio, ZIO_TASKQ_ISSUE, B_TRUE);
}
-static int
+static zio_t *
zio_dva_allocate(zio_t *zio)
{
spa_t *spa = zio->io_spa;
@@ -3054,18 +3060,18 @@ zio_dva_allocate(zio_t *zio)
zio->io_error = error;
}
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
-static int
+static zio_t *
zio_dva_free(zio_t *zio)
{
metaslab_free(zio->io_spa, zio->io_bp, zio->io_txg, B_FALSE);
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
-static int
+static zio_t *
zio_dva_claim(zio_t *zio)
{
int error;
@@ -3074,7 +3080,7 @@ zio_dva_claim(zio_t *zio)
if (error)
zio->io_error = error;
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
/*
@@ -3172,7 +3178,7 @@ zio_free_zil(spa_t *spa, uint64_t txg, blkptr_t *bp)
* force the underlying vdev layers to call either zio_execute() or
* zio_interrupt() to ensure that the pipeline continues with the correct I/O.
*/
-static int
+static zio_t *
zio_vdev_io_start(zio_t *zio)
{
vdev_t *vd = zio->io_vd;
@@ -3192,7 +3198,7 @@ zio_vdev_io_start(zio_t *zio)
* The mirror_ops handle multiple DVAs in a single BP.
*/
vdev_mirror_ops.vdev_op_io_start(zio);
- return (ZIO_PIPELINE_STOP);
+ return (NULL);
}
ASSERT3P(zio->io_logical, !=, zio);
@@ -3269,31 +3275,31 @@ zio_vdev_io_start(zio_t *zio)
!vdev_dtl_contains(vd, DTL_PARTIAL, zio->io_txg, 1)) {
ASSERT(zio->io_type == ZIO_TYPE_WRITE);
zio_vdev_io_bypass(zio);
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
if (vd->vdev_ops->vdev_op_leaf &&
(zio->io_type == ZIO_TYPE_READ || zio->io_type == ZIO_TYPE_WRITE)) {
if (zio->io_type == ZIO_TYPE_READ && vdev_cache_read(zio))
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
if ((zio = vdev_queue_io(zio)) == NULL)
- return (ZIO_PIPELINE_STOP);
+ return (NULL);
if (!vdev_accessible(vd, zio)) {
zio->io_error = SET_ERROR(ENXIO);
zio_interrupt(zio);
- return (ZIO_PIPELINE_STOP);
+ return (NULL);
}
zio->io_delay = gethrtime();
}
vd->vdev_ops->vdev_op_io_start(zio);
- return (ZIO_PIPELINE_STOP);
+ return (NULL);
}
-static int
+static zio_t *
zio_vdev_io_done(zio_t *zio)
{
vdev_t *vd = zio->io_vd;
@@ -3301,7 +3307,7 @@ zio_vdev_io_done(zio_t *zio)
boolean_t unexpected_error = B_FALSE;
if (zio_wait_for_children(zio, ZIO_CHILD_VDEV_BIT, ZIO_WAIT_DONE)) {
- return (ZIO_PIPELINE_STOP);
+ return (NULL);
}
ASSERT(zio->io_type == ZIO_TYPE_READ || zio->io_type == ZIO_TYPE_WRITE);
@@ -3337,7 +3343,7 @@ zio_vdev_io_done(zio_t *zio)
if (unexpected_error)
VERIFY(vdev_probe(vd, zio) == NULL);
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
/*
@@ -3366,13 +3372,13 @@ zio_vsd_default_cksum_report(zio_t *zio, zio_cksum_report_t *zcr, void *ignored)
zcr->zcr_free = zio_abd_free;
}
-static int
+static zio_t *
zio_vdev_io_assess(zio_t *zio)
{
vdev_t *vd = zio->io_vd;
if (zio_wait_for_children(zio, ZIO_CHILD_VDEV_BIT, ZIO_WAIT_DONE)) {
- return (ZIO_PIPELINE_STOP);
+ return (NULL);
}
if (vd == NULL && !(zio->io_flags & ZIO_FLAG_CONFIG_WRITER))
@@ -3402,7 +3408,7 @@ zio_vdev_io_assess(zio_t *zio)
zio->io_stage = ZIO_STAGE_VDEV_IO_START >> 1;
zio_taskq_dispatch(zio, ZIO_TASKQ_ISSUE,
zio_requeue_io_start_cut_in_line);
- return (ZIO_PIPELINE_STOP);
+ return (NULL);
}
/*
@@ -3442,7 +3448,7 @@ zio_vdev_io_assess(zio_t *zio)
zio->io_physdone(zio->io_logical);
}
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
void
@@ -3477,7 +3483,7 @@ zio_vdev_io_bypass(zio_t *zio)
* Generate and verify checksums
* ==========================================================================
*/
-static int
+static zio_t *
zio_checksum_generate(zio_t *zio)
{
blkptr_t *bp = zio->io_bp;
@@ -3491,7 +3497,7 @@ zio_checksum_generate(zio_t *zio)
checksum = zio->io_prop.zp_checksum;
if (checksum == ZIO_CHECKSUM_OFF)
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
ASSERT(checksum == ZIO_CHECKSUM_LABEL);
} else {
@@ -3505,10 +3511,10 @@ zio_checksum_generate(zio_t *zio)
zio_checksum_compute(zio, checksum, zio->io_abd, zio->io_size);
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
-static int
+static zio_t *
zio_checksum_verify(zio_t *zio)
{
zio_bad_cksum_t info;
@@ -3523,7 +3529,7 @@ zio_checksum_verify(zio_t *zio)
* We're either verifying a label checksum, or nothing at all.
*/
if (zio->io_prop.zp_checksum == ZIO_CHECKSUM_OFF)
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
ASSERT(zio->io_prop.zp_checksum == ZIO_CHECKSUM_LABEL);
}
@@ -3538,7 +3544,7 @@ zio_checksum_verify(zio_t *zio)
}
}
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
/*
@@ -3581,7 +3587,7 @@ zio_worst_error(int e1, int e2)
* I/O completion
* ==========================================================================
*/
-static int
+static zio_t *
zio_ready(zio_t *zio)
{
blkptr_t *bp = zio->io_bp;
@@ -3590,7 +3596,7 @@ zio_ready(zio_t *zio)
if (zio_wait_for_children(zio, ZIO_CHILD_GANG_BIT | ZIO_CHILD_DDT_BIT,
ZIO_WAIT_READY)) {
- return (ZIO_PIPELINE_STOP);
+ return (NULL);
}
if (zio->io_ready) {
@@ -3636,7 +3642,7 @@ zio_ready(zio_t *zio)
*/
for (; pio != NULL; pio = pio_next) {
pio_next = zio_walk_parents(zio, &zl);
- zio_notify_parent(pio, zio, ZIO_WAIT_READY);
+ zio_notify_parent(pio, zio, ZIO_WAIT_READY, NULL);
}
if (zio->io_flags & ZIO_FLAG_NODATA) {
@@ -3652,7 +3658,7 @@ zio_ready(zio_t *zio)
zio->io_spa->spa_syncing_txg == zio->io_txg)
zio_handle_ignored_writes(zio);
- return (ZIO_PIPELINE_CONTINUE);
+ return (zio);
}
/*
@@ -3716,7 +3722,7 @@ zio_dva_throttle_done(zio_t *zio)
zio_allocate_dispatch(zio->io_spa);
}
-static int
+static zio_t *
zio_done(zio_t *zio)
{
/*
@@ -3733,7 +3739,7 @@ zio_done(zio_t *zio)
* wait for them and then repeat this pipeline stage.
*/
if (zio_wait_for_children(zio, ZIO_CHILD_ALL_BITS, ZIO_WAIT_DONE)) {
- return (ZIO_PIPELINE_STOP);
+ return (NULL);
}
/*
@@ -3957,7 +3963,12 @@ zio_done(zio_t *zio)
if ((pio->io_flags & ZIO_FLAG_GODFATHER) &&
(zio->io_reexecute & ZIO_REEXECUTE_SUSPEND)) {
zio_remove_child(pio, zio, remove_zl);
- zio_notify_parent(pio, zio, ZIO_WAIT_DONE);
+ /*
+ * This is a rare code path, so we don't
+ * bother with "next_to_execute".
+ */
+ zio_notify_parent(pio, zio, ZIO_WAIT_DONE,
+ NULL);
}
}
@@ -3969,7 +3980,11 @@ zio_done(zio_t *zio)
*/
ASSERT(!(zio->io_flags & ZIO_FLAG_GODFATHER));
zio->io_flags |= ZIO_FLAG_DONT_PROPAGATE;
- zio_notify_parent(pio, zio, ZIO_WAIT_DONE);
+ /*
+ * This is a rare code path, so we don't bother with
+ * "next_to_execute".
+ */
+ zio_notify_parent(pio, zio, ZIO_WAIT_DONE, NULL);
} else if (zio->io_reexecute & ZIO_REEXECUTE_SUSPEND) {
/*
* We'd fail again if we reexecuted now, so suspend
@@ -3987,7 +4002,7 @@ zio_done(zio_t *zio)
(task_func_t *)zio_reexecute, zio, 0,
&zio->io_tqent);
}
- return (ZIO_PIPELINE_STOP);
+ return (NULL);
}
ASSERT(zio->io_child_count == 0);
@@ -4023,12 +4038,17 @@ zio_done(zio_t *zio)
zio->io_state[ZIO_WAIT_DONE] = 1;
mutex_exit(&zio->io_lock);
+ /*
+ * We are done executing this zio. We may want to execute a parent
+ * next. See the comment in zio_notify_parent().
+ */
+ zio_t *next_to_execute = NULL;
zl = NULL;
for (pio = zio_walk_parents(zio, &zl); pio != NULL; pio = pio_next) {
zio_link_t *remove_zl = zl;
pio_next = zio_walk_parents(zio, &zl);
zio_remove_child(pio, zio, remove_zl);
- zio_notify_parent(pio, zio, ZIO_WAIT_DONE);
+ zio_notify_parent(pio, zio, ZIO_WAIT_DONE, &next_to_execute);
}
if (zio->io_waiter != NULL) {
@@ -4040,7 +4060,7 @@ zio_done(zio_t *zio)
zio_destroy(zio);
}
- return (ZIO_PIPELINE_STOP);
+ return (next_to_execute);
}
/*
@@ -1,376 +0,0 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: ilbsmart <wgqimut@gmail.com>
Date: Wed, 17 Oct 2018 02:11:24 +0800
Subject: [PATCH] deadlock between mm_sem and tx assign in zfs_write() and page
fault
The bug time sequence:
1. thread #1, `zfs_write` assign a txg "n".
2. In a same process, thread #2, mmap page fault (which means the
`mm_sem` is hold) occurred, `zfs_dirty_inode` open a txg failed,
and wait previous txg "n" completed.
3. thread #1 call `uiomove` to write, however page fault is occurred
in `uiomove`, which means it need `mm_sem`, but `mm_sem` is hold by
thread #2, so it stuck and can't complete, then txg "n" will
not complete.
So thread #1 and thread #2 are deadlocked.
Reviewed-by: Chunwei Chen <tuxoko@gmail.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Signed-off-by: Grady Wong <grady.w@xtaotech.com>
Closes #7939
(backported from: zfs-upstream 779a6c0bf6df76e0dd92c1ccf81f48512b835bb0)
Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
---
include/sys/uio_impl.h | 2 +-
module/zcommon/zfs_uio.c | 31 ++++-
module/zfs/zfs_vnops.c | 24 +++-
tests/zfs-tests/cmd/mmapwrite/mmapwrite.c | 140 +++++++++++++++------
.../tests/functional/mmap/mmap_write_001_pos.ksh | 8 +-
5 files changed, 151 insertions(+), 54 deletions(-)
diff --git a/include/sys/uio_impl.h b/include/sys/uio_impl.h
index 37e283da..cfef0b95 100644
--- a/include/sys/uio_impl.h
+++ b/include/sys/uio_impl.h
@@ -42,7 +42,7 @@
#include <sys/uio.h>
extern int uiomove(void *, size_t, enum uio_rw, uio_t *);
-extern void uio_prefaultpages(ssize_t, uio_t *);
+extern int uio_prefaultpages(ssize_t, uio_t *);
extern int uiocopy(void *, size_t, enum uio_rw, uio_t *, size_t *);
extern void uioskip(uio_t *, size_t);
diff --git a/module/zcommon/zfs_uio.c b/module/zcommon/zfs_uio.c
index 7b4175bb..8e969bbc 100644
--- a/module/zcommon/zfs_uio.c
+++ b/module/zcommon/zfs_uio.c
@@ -50,6 +50,7 @@
#include <sys/types.h>
#include <sys/uio_impl.h>
#include <linux/kmap_compat.h>
+#include <linux/uaccess.h>
/*
* Move "n" bytes at byte address "p"; "rw" indicates the direction
@@ -77,8 +78,24 @@ uiomove_iov(void *p, size_t n, enum uio_rw rw, struct uio *uio)
if (copy_to_user(iov->iov_base+skip, p, cnt))
return (EFAULT);
} else {
- if (copy_from_user(p, iov->iov_base+skip, cnt))
- return (EFAULT);
+ if (uio->uio_fault_disable) {
+ if (!access_ok(VERIFY_READ,
+ (iov->iov_base + skip), cnt)) {
+ return (EFAULT);
+ }
+
+ pagefault_disable();
+ if (__copy_from_user_inatomic(p,
+ (iov->iov_base + skip), cnt)) {
+ pagefault_enable();
+ return (EFAULT);
+ }
+ pagefault_enable();
+ } else {
+ if (copy_from_user(p,
+ (iov->iov_base + skip), cnt))
+ return (EFAULT);
+ }
}
break;
case UIO_SYSSPACE:
@@ -156,7 +173,7 @@ EXPORT_SYMBOL(uiomove);
* error will terminate the process as this is only a best attempt to get
* the pages resident.
*/
-void
+int
uio_prefaultpages(ssize_t n, struct uio *uio)
{
const struct iovec *iov;
@@ -170,7 +187,7 @@ uio_prefaultpages(ssize_t n, struct uio *uio)
switch (uio->uio_segflg) {
case UIO_SYSSPACE:
case UIO_BVEC:
- return;
+ return (0);
case UIO_USERSPACE:
case UIO_USERISPACE:
break;
@@ -194,7 +211,7 @@ uio_prefaultpages(ssize_t n, struct uio *uio)
p = iov->iov_base + skip;
while (cnt) {
if (fuword8((uint8_t *)p, &tmp))
- return;
+ return (EFAULT);
incr = MIN(cnt, PAGESIZE);
p += incr;
cnt -= incr;
@@ -204,8 +221,10 @@ uio_prefaultpages(ssize_t n, struct uio *uio)
*/
p--;
if (fuword8((uint8_t *)p, &tmp))
- return;
+ return (EFAULT);
}
+
+ return (0);
}
EXPORT_SYMBOL(uio_prefaultpages);
diff --git a/module/zfs/zfs_vnops.c b/module/zfs/zfs_vnops.c
index 5a2e55eb..c866352d 100644
--- a/module/zfs/zfs_vnops.c
+++ b/module/zfs/zfs_vnops.c
@@ -675,7 +675,10 @@ zfs_write(struct inode *ip, uio_t *uio, int ioflag, cred_t *cr)
xuio = (xuio_t *)uio;
else
#endif
- uio_prefaultpages(MIN(n, max_blksz), uio);
+ if (uio_prefaultpages(MIN(n, max_blksz), uio)) {
+ ZFS_EXIT(zfsvfs);
+ return (SET_ERROR(EFAULT));
+ }
/*
* If in append mode, set the io offset pointer to eof.
@@ -820,8 +823,19 @@ zfs_write(struct inode *ip, uio_t *uio, int ioflag, cred_t *cr)
if (abuf == NULL) {
tx_bytes = uio->uio_resid;
+ uio->uio_fault_disable = B_TRUE;
error = dmu_write_uio_dbuf(sa_get_db(zp->z_sa_hdl),
uio, nbytes, tx);
+ if (error == EFAULT) {
+ dmu_tx_commit(tx);
+ if (uio_prefaultpages(MIN(n, max_blksz), uio)) {
+ break;
+ }
+ continue;
+ } else if (error != 0) {
+ dmu_tx_commit(tx);
+ break;
+ }
tx_bytes -= uio->uio_resid;
} else {
tx_bytes = nbytes;
@@ -921,8 +935,12 @@ zfs_write(struct inode *ip, uio_t *uio, int ioflag, cred_t *cr)
ASSERT(tx_bytes == nbytes);
n -= nbytes;
- if (!xuio && n > 0)
- uio_prefaultpages(MIN(n, max_blksz), uio);
+ if (!xuio && n > 0) {
+ if (uio_prefaultpages(MIN(n, max_blksz), uio)) {
+ error = EFAULT;
+ break;
+ }
+ }
}
zfs_inode_update(zp);
diff --git a/tests/zfs-tests/cmd/mmapwrite/mmapwrite.c b/tests/zfs-tests/cmd/mmapwrite/mmapwrite.c
index 190d31af..b9915d5d 100644
--- a/tests/zfs-tests/cmd/mmapwrite/mmapwrite.c
+++ b/tests/zfs-tests/cmd/mmapwrite/mmapwrite.c
@@ -31,74 +31,132 @@
#include <string.h>
#include <sys/mman.h>
#include <pthread.h>
+#include <errno.h>
+#include <err.h>
/*
* --------------------------------------------------------------------
- * Bug Id: 5032643
+ * Bug Issue Id: #7512
+ * The bug time sequence:
+ * 1. context #1, zfs_write assign a txg "n".
+ * 2. In the same process, context #2, mmap page fault (which means the mm_sem
+ * is hold) occurred, zfs_dirty_inode open a txg failed, and wait previous
+ * txg "n" completed.
+ * 3. context #1 call uiomove to write, however page fault is occurred in
+ * uiomove, which means it need mm_sem, but mm_sem is hold by
+ * context #2, so it stuck and can't complete, then txg "n" will not
+ * complete.
*
- * Simply writing to a file and mmaping that file at the same time can
- * result in deadlock. Nothing perverse like writing from the file's
- * own mapping is required.
+ * So context #1 and context #2 trap into the "dead lock".
* --------------------------------------------------------------------
*/
+#define NORMAL_WRITE_TH_NUM 2
+
static void *
-mapper(void *fdp)
+normal_writer(void *filename)
{
- void *addr;
- int fd = *(int *)fdp;
+ char *file_path = filename;
+ int fd = -1;
+ ssize_t write_num = 0;
+ int page_size = getpagesize();
- if ((addr =
- mmap(0, 8192, PROT_READ, MAP_SHARED, fd, 0)) == MAP_FAILED) {
- perror("mmap");
- exit(1);
+ fd = open(file_path, O_RDWR | O_CREAT, 0777);
+ if (fd == -1) {
+ err(1, "failed to open %s", file_path);
}
- for (;;) {
- if (mmap(addr, 8192, PROT_READ,
- MAP_SHARED|MAP_FIXED, fd, 0) == MAP_FAILED) {
- perror("mmap");
- exit(1);
+
+ char *buf = malloc(1);
+ while (1) {
+ write_num = write(fd, buf, 1);
+ if (write_num == 0) {
+ err(1, "write failed!");
+ break;
}
+ lseek(fd, page_size, SEEK_CUR);
+ }
+
+ if (buf) {
+ free(buf);
}
- /* NOTREACHED */
- return ((void *)1);
}
-int
-main(int argc, char **argv)
+static void *
+map_writer(void *filename)
{
- int fd;
- char buf[1024];
- pthread_t tid;
+ int fd = -1;
+ int ret = 0;
+ char *buf = NULL;
+ int page_size = getpagesize();
+ int op_errno = 0;
+ char *file_path = filename;
- memset(buf, 'a', sizeof (buf));
+ while (1) {
+ ret = access(file_path, F_OK);
+ if (ret) {
+ op_errno = errno;
+ if (op_errno == ENOENT) {
+ fd = open(file_path, O_RDWR | O_CREAT, 0777);
+ if (fd == -1) {
+ err(1, "open file failed");
+ }
- if (argc != 2) {
- (void) printf("usage: %s <file name>\n", argv[0]);
- exit(1);
- }
+ ret = ftruncate(fd, page_size);
+ if (ret == -1) {
+ err(1, "truncate file failed");
+ }
+ } else {
+ err(1, "access file failed!");
+ }
+ } else {
+ fd = open(file_path, O_RDWR, 0777);
+ if (fd == -1) {
+ err(1, "open file failed");
+ }
+ }
- if ((fd = open(argv[1], O_RDWR|O_CREAT|O_TRUNC, 0666)) == -1) {
- perror("open");
- exit(1);
+ if ((buf = mmap(NULL, page_size, PROT_READ | PROT_WRITE,
+ MAP_SHARED, fd, 0)) == MAP_FAILED) {
+ err(1, "map file failed");
+ }
+
+ if (fd != -1)
+ close(fd);
+
+ char s[10] = {0, };
+ memcpy(buf, s, 10);
+ ret = munmap(buf, page_size);
+ if (ret != 0) {
+ err(1, "unmap file failed");
+ }
}
+}
- (void) pthread_setconcurrency(2);
- if (pthread_create(&tid, NULL, mapper, &fd) != 0) {
- perror("pthread_create");
- close(fd);
+int
+main(int argc, char **argv)
+{
+ pthread_t map_write_tid;
+ pthread_t normal_write_tid[NORMAL_WRITE_TH_NUM];
+ int i = 0;
+
+ if (argc != 3) {
+ (void) printf("usage: %s <normal write file name>"
+ "<map write file name>\n", argv[0]);
exit(1);
}
- for (;;) {
- if (write(fd, buf, sizeof (buf)) == -1) {
- perror("write");
- close(fd);
- exit(1);
+
+ for (i = 0; i < NORMAL_WRITE_TH_NUM; i++) {
+ if (pthread_create(&normal_write_tid[i], NULL, normal_writer,
+ argv[1])) {
+ err(1, "pthread_create normal_writer failed.");
}
}
- close(fd);
+ if (pthread_create(&map_write_tid, NULL, map_writer, argv[2])) {
+ err(1, "pthread_create map_writer failed.");
+ }
/* NOTREACHED */
+ pthread_join(map_write_tid, NULL);
return (0);
}
diff --git a/tests/zfs-tests/tests/functional/mmap/mmap_write_001_pos.ksh b/tests/zfs-tests/tests/functional/mmap/mmap_write_001_pos.ksh
index 1eda9710..24150b82 100755
--- a/tests/zfs-tests/tests/functional/mmap/mmap_write_001_pos.ksh
+++ b/tests/zfs-tests/tests/functional/mmap/mmap_write_001_pos.ksh
@@ -53,12 +53,14 @@ if ! is_mp; then
fi
log_must chmod 777 $TESTDIR
-mmapwrite $TESTDIR/test-write-file &
+mmapwrite $TESTDIR/normal_write_file $TESTDIR/map_write_file &
PID_MMAPWRITE=$!
-log_note "mmapwrite $TESTDIR/test-write-file pid: $PID_MMAPWRITE"
+log_note "mmapwrite $TESTDIR/normal_write_file $TESTDIR/map_write_file"\
+ "pid: $PID_MMAPWRITE"
log_must sleep 30
log_must kill -9 $PID_MMAPWRITE
-log_must ls -l $TESTDIR/test-write-file
+log_must ls -l $TESTDIR/normal_write_file
+log_must ls -l $TESTDIR/map_write_file
log_pass "write(2) a mmap(2)'ing file succeeded."
@@ -1,43 +0,0 @@
From 5ac80068e911d3b0935903f713c5f492d518da91 Mon Sep 17 00:00:00 2001
From: Stoiko Ivanov <s.ivanov@proxmox.com>
Date: Mon, 29 Oct 2018 15:49:20 +0100
Subject: [PATCH] Add Breaks/Replaces to zfs-initramfs
addressing the move of zdev initramfs hook from zfsutils-linux to
zfs-initramfs
Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
---
debian/control | 2 ++
debian/control.in | 2 ++
2 files changed, 4 insertions(+)
diff --git a/debian/control b/debian/control
index 4d22ff50..a414e449 100644
--- a/debian/control
+++ b/debian/control
@@ -117,6 +117,8 @@ Depends: busybox-initramfs | busybox-static | busybox,
zfs-modules | zfs-dkms,
zfsutils-linux (>= ${binary:Version}),
${misc:Depends}
+Breaks: zfsutils-linux (<= 0.7.11-pve1~bpo1)
+Replaces: zfsutils-linux (<= 0.7.11-pve1~bpo1)
Description: OpenZFS root filesystem capabilities for Linux - initramfs
The Z file system is a pooled filesystem designed for maximum data
integrity, supporting data snapshots, multiple copies, and data
diff --git a/debian/control.in b/debian/control.in
index 96154c5c..b9c34331 100644
--- a/debian/control.in
+++ b/debian/control.in
@@ -117,6 +117,8 @@ Depends: busybox-initramfs | busybox-static | busybox,
zfs-modules | zfs-dkms,
zfsutils-linux (>= ${binary:Version}),
${misc:Depends}
+Breaks: zfsutils-linux (<= 0.7.11-pve1~bpo1)
+Replaces: zfsutils-linux (<= 0.7.11-pve1~bpo1)
Description: OpenZFS root filesystem capabilities for Linux - initramfs
The Z file system is a pooled filesystem designed for maximum data
integrity, supporting data snapshots, multiple copies, and data
--
2.11.0
@@ -1,78 +0,0 @@
From c37a6401469cae16519db06d623d49c6f606f312 Mon Sep 17 00:00:00 2001
From: Thomas Lamprecht <t.lamprecht@proxmox.com>
Date: Wed, 31 Oct 2018 15:37:39 +0100
Subject: [PATCH] Revert "Install init scripts to support non-systemd setups.
(Closes: #826994)"
This reverts commit 2ad6c7f9e150e9c2fa146891785b09465aa0c7e5.
Breaks on systems with sysv-rc installed (possible on Jessie upgraded
systems)
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
---
debian/rules | 8 --------
debian/zfs-zed.zfs-zed.init | 1 -
debian/zfsutils-linux.zfs-import.init | 1 -
debian/zfsutils-linux.zfs-mount.init | 1 -
debian/zfsutils-linux.zfs-share.init | 1 -
5 files changed, 12 deletions(-)
delete mode 120000 debian/zfs-zed.zfs-zed.init
delete mode 120000 debian/zfsutils-linux.zfs-import.init
delete mode 120000 debian/zfsutils-linux.zfs-mount.init
delete mode 120000 debian/zfsutils-linux.zfs-share.init
diff --git a/debian/rules b/debian/rules
index 5fba58ff..81c301e4 100644
--- a/debian/rules
+++ b/debian/rules
@@ -161,14 +153,6 @@ override_dh_install:
find . -name lib*.la -delete
dh_install --fail-missing
-override_dh_installinit:
- dh_installinit -r --no-start --name zfs-import
- dh_installinit -r --no-start --name zfs-mount
- dh_installinit -r --no-start --name zfs-share
- dh_installinit -R --no-start --name zfs-zed
- mkdir -p debian/zfsutils-linux/lib/systemd/system
- ln -sr /dev/null debian/zfsutils-linux/lib/systemd/system/zfs-import.service
-
# ------------
override_dh_prep-deb-files:
diff --git a/debian/zfs-zed.zfs-zed.init b/debian/zfs-zed.zfs-zed.init
deleted file mode 120000
index 3f41f681..00000000
--- a/debian/zfs-zed.zfs-zed.init
+++ /dev/null
@@ -1 +0,0 @@
-../etc/init.d/zfs-zed
\ No newline at end of file
diff --git a/debian/zfsutils-linux.zfs-import.init b/debian/zfsutils-linux.zfs-import.init
deleted file mode 120000
index bfc368fd..00000000
--- a/debian/zfsutils-linux.zfs-import.init
+++ /dev/null
@@ -1 +0,0 @@
-../etc/init.d/zfs-import
\ No newline at end of file
diff --git a/debian/zfsutils-linux.zfs-mount.init b/debian/zfsutils-linux.zfs-mount.init
deleted file mode 120000
index 62a54433..00000000
--- a/debian/zfsutils-linux.zfs-mount.init
+++ /dev/null
@@ -1 +0,0 @@
-../etc/init.d/zfs-mount
\ No newline at end of file
diff --git a/debian/zfsutils-linux.zfs-share.init b/debian/zfsutils-linux.zfs-share.init
deleted file mode 120000
index 3f069f9b..00000000
--- a/debian/zfsutils-linux.zfs-share.init
+++ /dev/null
@@ -1 +0,0 @@
-../etc/init.d/zfs-share
\ No newline at end of file
--
2.19.1
-9
View File
@@ -1,9 +0,0 @@
0001-remove-DKMS-modules-and-dracut-build.patch
0002-import-with-d-dev-disk-by-id-in-scan-service.patch
0003-always-load-ZFS-module-on-boot.patch
0004-Fix-deadlock-between-zfs-umount-snapentry_expire.patch
0005-Fix-race-in-dnode_check_slots_free.patch
0006-Reduce-taskq-and-context-switch-cost-of-zio-pipe.patch
0007-deadlock-between-mm_sem-and-tx-assign-in-zfs_write-a.patch
0008-Add-Breaks-Replaces-to-zfs-initramfs.patch
0009-Revert-Install-init-scripts-to-support-non-systemd-s.patch
+691
View File
@@ -0,0 +1,691 @@
zfs-linux (0.7.13-pve1~bpo2) unstable; urgency=medium
* remove conflict with insserv (<< 1.18)
-- Proxmox Support Team <support@proxmox.com> Wed, 27 Mar 2019 09:56:15 +0100
zfs-linux (0.7.13-pve1~bpo1) unstable; urgency=medium
* update ZFS to 0.7.13
* base zfs-linux on upstream ZOL instead of debian
-- Proxmox Support Team <support@proxmox.com> Fri, 08 Mar 2019 06:53:00 +0100
zfs-linux (0.7.12-pve1~bpo1) unstable; urgency=medium
* update ZFS to 0.7.12
-- Proxmox Support Team <support@proxmox.com> Wed, 14 Nov 2018 17:51:59 +0100
zfs-linux (0.7.11-pve2~bpo1) unstable; urgency=medium
* update ZFS to debian/0.7.11-3
* Cherry-pick two fixes planned for 0.7.12
* Backport deadlock fix between mm_sem and tx assign in zfs_write() and page
fault
* Fix missing Breaks/Replaces in zfs-initramfs
-- Proxmox Support Team <support@proxmox.com> Mon, 29 Oct 2018 13:45:49 +0100
zfs-linux (0.7.11-pve1~bpo1) unstable; urgency=medium
* update ZFS to 0.7.11
-- Proxmox Support Team <support@proxmox.com> Mon, 10 Sep 2018 15:47:31 +0200
zfs-linux (0.7.9-pve3~bpo9) unstable; urgency=medium
* Cherry-pick fix for deadlock umount/snapentry_expire
-- Proxmox Support Team <support@proxmox.com> Fri, 03 Aug 2018 11:41:11 +0200
zfs-linux (0.7.9-pve2~bpo9) unstable; urgency=medium
* Cherry-pick fix for zpl_mount deadlock
-- Proxmox Support Team <support@proxmox.com> Fri, 12 Jul 2018 12:37:50 +0200
zfs-linux (0.7.9-pve1~bpo9) unstable; urgency=medium
* update ZFS to 0.7.9
-- Proxmox Support Team <support@proxmox.com> Fri, 18 May 2018 13:49:09 +0200
zfs-linux (0.7.8-pve1~bpo9) unstable; urgency=medium
* update ZFS to 0.7.8 (no changes)
-- Proxmox Support Team <support@proxmox.com> Mon, 30 Apr 2018 13:47:22 +0200
zfs-linux (0.7.7-pve2~bpo9) unstable; urgency=medium
* (temporarily) revert likely cause of #7401
-- Proxmox Support Team <support@proxmox.com> Mon, 9 Apr 2018 09:49:27 +0200
zfs-linux (0.7.7-pve1~bpo9) unstable; urgency=medium
* update ZFS to 0.7.7
-- Proxmox Support Team <support@proxmox.com> Tue, 3 Apr 2018 14:28:35 +0200
zfs-linux (0.7.6-pve1~bpo9) unstable; urgency=medium
* update ZFS to 0.7.6
-- Proxmox Support Team <support@proxmox.com> Wed, 21 Feb 2018 09:48:29 +0100
zfs-linux (0.7.4-pve2~bpo9) unstable; urgency=medium
* cherry-pick ARC hit rate fix from 0.7.6
* always load ZFS modules on boot
-- Proxmox Support Team <support@proxmox.com> Fri, 19 Jan 2018 11:31:14 +0100
zfs-linux (0.7.4-pve1~bpo9) unstable; urgency=medium
* update ZFS to 0.7.4
-- Proxmox Support Team <support@proxmox.com> Tue, 19 Dec 2017 12:52:42 +0100
zfs-linux (0.7.3-pve1~bpo9) unstable; urgency=medium
* update ZFS to 0.7.3
* include fix for user namespace setgid issue (#6800 upstream)
-- Proxmox Support Team <support@proxmox.com> Mon, 6 Nov 2017 09:14:19 +0100
zfs-linux (0.7.2-pve1~bpo90) unstable; urgency=medium
* update zfs to 0.7.2
* add PR 6616 - zfs send/recv compatibility with 0.6.5
* add PR 6695 - zfs recv: don't skip over objects which should be freed
* various small fixes
* split test suite into zfs-test package
-- Proxmox Support Team <support@proxmox.com> Thu, 12 Oct 2017 13:34:05 +0200
zfs-linux (0.6.5.11-pve18~bpo90) unstable; urgency=medium
* fix #1509: arc_summary error with L2ARC
-- Proxmox Support Team <support@proxmox.com> Mon, 2 Oct 2017 09:30:41 +0200
zfs-linux (0.6.5.11-pve17~bpo90) unstable; urgency=medium
* update zfs to debian/0.6.5.11-1
-- Proxmox Support Team <support@proxmox.com> Thu, 27 Jul 2017 13:11:45 +0200
zfs-linux (0.6.5.9-pve16~bpo90) unstable; urgency=medium
* drop transitional packages
* convert python scripts to python 3, drop .py suffix
* rebuild for PVE 5.0 / Debian Stretch
-- Proxmox Support Team <support@proxmox.com> Fri, 10 Mar 2017 09:45:52 +0100
zfs-linux (0.6.5.9-pve15~bpo80) unstable; urgency=medium
* update zfs to debian/0.6.5.9-1
-- Proxmox Support Team <support@proxmox.com> Mon, 6 Feb 2017 12:39:10 +0100
zfs-linux (0.6.5.8-pve14~bpo80) unstable; urgency=medium
* update zfs to debian/0.6.5.8-3
-- Proxmox Support Team <support@proxmox.com> Wed, 11 Jan 2017 09:34:58 +0100
zfs-linux (0.6.5.8-pve13~bpo80) unstable; urgency=medium
* fix #1184: zfs-share.service has wrong path to 'rm' command
-- Proxmox Support Team <support@proxmox.com> Thu, 27 Oct 2016 11:25:47 +0200
zfs-linux (0.6.5.8-pve12~bpo80) unstable; urgency=medium
* import with "-d /dev/disk/by-id" in systemd service
-- Proxmox Support Team <support@proxmox.com> Mon, 24 Oct 2016 13:48:08 +0200
zfs-linux (0.6.5.8-pve11~bpo80) unstable; urgency=medium
* update zfs to debian/0.6.5.8-1
* switch package upstream sources to Debian (Jessie)
* add transitional packages for upgrades
-- Proxmox Support Team <support@proxmox.com> Wed, 12 Oct 2016 11:16:02 +0200
zfs-linux (0.6.5.7-pve10~bpo80) unstable; urgency=medium
* update to pkg-zfs jessie/0.6.5.7-8
* change package versioning to allow upgrades from PVE3/wheezy
-- Proxmox Support Team <support@proxmox.com> Tue, 28 Jun 2016 09:50:00 +0200
zfs-linux (0.6.5-pve9~jessie) unstable; urgency=medium
* Update to pkg-zfs jessie/0.6.5.6-3
-- Proxmox Support Team <support@proxmox.com> Tue, 12 Apr 2016 09:51:35 +0200
zfs-linux (0.6.5-pve8~jessie) unstable; urgency=medium
* update to zfs-0.6.5.6
-- Proxmox Support Team <support@proxmox.com> Wed, 30 Mar 2016 10:50:22 +0200
zfs-linux (0.6.5-pve7~jessie) unstable; urgency=medium
* update to zfs-0.6.5.4
-- Proxmox Support Team <support@proxmox.com> Thu, 14 Jan 2016 10:51:17 +0100
zfs-linux (0.6.5-pve6~jessie) unstable; urgency=medium
* Prepare to tag zfs-0.6.5.3
-- Proxmox Support Team <support@proxmox.com> Thu, 05 Nov 2015 07:21:44 +0100
zfs-linux (0.6.5-pve5~jessie) unstable; urgency=medium
* Illumos 6267 - dn_bonus evicted too early
* Fix use-after-free in vdev_disk_physio_completion
-- Proxmox Support Team <support@proxmox.com> Wed, 04 Nov 2015 17:42:42 +0100
zfs-linux (0.6.5-pve4~jessie) unstable; urgency=medium
* update to master/debian/jessie/0.6.5.2-2
-- Proxmox Support Team <support@proxmox.com> Mon, 05 Oct 2015 17:56:45 +0200
zfs-linux (0.6.5-pve3~jessie) unstable; urgency=medium
* do not install /etc/init.d/zfs-zed script to avoid double startup
with systemd
-- Proxmox Support Team <support@proxmox.com> Mon, 28 Sep 2015 10:14:40 +0200
zfs-linux (0.6.5-pve2~jessie) unstable; urgency=medium
* update to master/debian/jessie/0.6.5.1-4
-- Proxmox Support Team <support@proxmox.com> Thu, 24 Sep 2015 12:45:33 +0200
zfs-linux (0.6.5-pve1~jessie) unstable; urgency=medium
* update to master/debian/jessie/0.6.5.1-2
-- Proxmox Support Team <support@proxmox.com> Mon, 21 Sep 2015 10:02:41 +0200
zfs-linux (0.6.4-pve3~jessie) unstable; urgency=medium
* update to snapshot/debian/jessie/0.6.4-24-6bec43
-- Proxmox Support Team <support@proxmox.com> Mon, 24 Aug 2015 06:13:44 +0200
zfs-linux (0.6.4-pve2~jessie) unstable; urgency=medium
* update to 0.6.4.2 (snapshot/debian/jessie/0.6.4-21-53b1d9)
-- Proxmox Support Team <support@proxmox.com> Tue, 21 Jul 2015 11:03:21 +0200
zfs-linux (0.6.4-pve1~jessie) unstable; urgency=medium
* update to 0.6.4 (use upstream zol package definitions)
-- Proxmox Support Team <support@proxmox.com> Tue, 14 Apr 2015 10:19:22 +0200
zfs-linux (0.6.3-pve3~jessie) unstable; urgency=medium
* ignore zfs-import-scan errors
-- Proxmox Support Team <support@proxmox.com> Tue, 31 Mar 2015 17:52:40 +0200
zfs-linux (0.6.3-pve2~jessie) unstable; urgency=medium
* use systemd for service startup
* install zed configuration file /etc/zfs/zed.d/zed.rc
-- Proxmox Support Team <support@proxmox.com> Sun, 15 Mar 2015 15:52:54 +0100
zfs-linux (0.6.3-pve1~jessie) unstable; urgency=medium
* recompile on jessie
-- Proxmox Support Team <support@proxmox.com> Sat, 28 Feb 2015 16:02:08 +0100
zfs-linux (0.6.3-2~wheezy) unstable; urgency=low
* use /sbin/modprobe to avoid warning inside initrd
* fix warning about undefined values inside initrd
-- Proxmox Support Team <support@proxmox.com> Wed, 14 Jan 2015 19:03:04 +0100
zfs-linux (0.6.3-1~wheezy) unstable; urgency=low
* first version for Proxmox VE
-- Proxmox Support Team <support@proxmox.com> Sat, 10 Jan 2015 14:06:34 +0100
zfs-linux (0.7.12-2) unstable; urgency=medium
[ Colin Ian King ]
* Only run autopkgtests for amd64, arm64, ppc64el and s390x (LP#1805627).
[ Martin Bagge / brother ]
* [INTL:sv] Swedish strings for zfs-linux debconf (Closes: #918020)
[ Anders Jonsson ]
* sv.po: typo fix
[ Mo Zhou ]
* Change init script's behaviour to default during postinst.
* Add ${perl:Depends} to zfs-dkms's Depends.
* Add autopkgtest script to test zfs-dkms build.
* autopkgtest: minor fix
[ Aron Xu ]
* Add XS-Autobuild: yes to d/control
* Conflicts with insserv << 1.18 (Closes: #915831)
-- Aron Xu <aron@debian.org> Fri, 11 Jan 2019 21:32:06 +0800
zfs-linux (0.7.12-1) unstable; urgency=medium
[ Stoiko Ivanov ]
* Add Breaks/Replaces to zfs-initramfs
[ Mo Zhou ]
* New upstream version 0.7.12
* Drop unnecessary patch init-start-stop-dep-on-local-fs.patch .
* Override init.d-script-missing-dependency-on-local_fs for zfs-zed.
* Bump linux compatibility to 4.19 .
-- Mo Zhou <cdluminate@gmail.com> Mon, 19 Nov 2018 11:32:44 +0000
zfs-linux (0.7.11-3) unstable; urgency=medium
[ Antonio Russo ]
+ https://salsa.debian.org/zfsonlinux-team/zfs/merge_requests/9
* Break/Replace upstream .deb packages (Closes: #839921)
* Install upstream bash completion file instead of embedded one.
* Modify META before autoreconf.
* Make dkms distdir before build to avoid including build artifacts.
* Remove ZFS_AC_PACKAGE macros from DKMS sources.
This removes dpkg-dev dependency from zfs-dkms package.
* Use upstream's dkms.mkconf script to produce dkms.conf .
* Ship initramfs zdev hook in zfs-initramfs (Closes: #902052)
[ Nicolas Braud-Santoni ]
* Update debian/copyright, removing unused wildcards.
[ Nicholas D Steeves ]
* Change -dbg package's priority from extra to optional.
[ Mo Zhou ]
* Fix FTBFS on architecture=all due to FileNotFound. (Closes: #911937)
* Add isolation-machine restriction to autopkgtest because the tests
needs to interact with the kernel, i.e. loading kernel module.
-- Mo Zhou <cdluminate@gmail.com> Sun, 28 Oct 2018 10:28:52 +0000
zfs-linux (0.7.11-2) unstable; urgency=medium
* Support Devuan in dkms script. (Closes: #900089)
Thanks to Chris Dos <chris@chrisdos.com>
* Install init scripts to support non-systemd setups. (Closes: #826994)
Thanks to Chris Dos <chris@chrisdos.com>
* Override init.d-script-does-not-source-init-functions for
zfsutils-linux and zfs-zed.
* Patch upstream init scripts to make them work for Debian+OpenRC setup.
* Patch upstream init script to fix missing dependency on local_fs.
-- Mo Zhou <cdluminate@gmail.com> Fri, 26 Oct 2018 09:32:06 +0000
zfs-linux (0.7.11-1) unstable; urgency=medium
[ Aron Xu ]
* Add dpkg-dev to Depends of zfs-dkms (Closes: #900714)
[ Nicolas Braud-Santoni ]
* Use canonical HTTPS format URL for Vcs-Git (Closes: #895873)
[ Mo Zhou ]
* New upstream version 0.7.11 (Closes: #908290)
* Bump linux_compat to 4.18 .
* Replace get_next.sh with one-liner awk script in rules.
* Append myself to Uploaders and refresh auto-generated control.
* Use HTTPS format URI in watch file.
* Recommends linux-libc-dev (<< LINUX_NEXT~) instead of (<< LINUX_NEXT).
-- Mo Zhou <cdluminate@gmail.com> Wed, 19 Sep 2018 08:45:18 +0000
zfs-linux (0.7.9-3) unstable; urgency=medium
[ Antonio Russo ]
* Expand zfs-test and add Breaks/Conflicts (Closes: #899047)
[ Aron Xu ]
* d/control: migrate to alioth-lists (Closes: #899756)
-- Aron Xu <aron@debian.org> Mon, 28 May 2018 18:22:02 +0800
zfs-linux (0.7.9-2) unstable; urgency=medium
[ Aron Xu ]
* Move more zfs test tools to zfs-test package (Closes: #868653)
* New upstream version 0.7.9
* d/rules: add --enable-systemd
* Fix lintian obsolete-relation-form-in-source
* Bump supported linux version to 4.16
[ Antonio Russo ]
* Install enum-extract.pl with dkms
* Handle /proc/kallsym obfuscation (Closes: #891936)
-- Aron Xu <aron@debian.org> Thu, 17 May 2018 23:47:29 +0800
zfs-linux (0.7.6-1) unstable; urgency=medium
[ Lev Lamberov ]
* [INTL:ru] Updated Russian translation of debconf (Closes: #885990)
[ Aron Xu]
* New upstream release (Closes: #889795, #890576)
* 0001-Fix-bug-in-distclean-which-removes-needed-files.patch:
removed, applied upstream
* Update VCS-* URL to salsa.debian.org
* Apply wrap-and-sort
* Recommends: linux-libc-dev (< ${LINUX_NEXT}):
Tries to prevent unexpected upgrades of kernel that is not known to
be supported by the packaged version of ZFS/SPL. (Closes: #849420)
-- Aron Xu <aron@debian.org> Mon, 26 Feb 2018 16:32:29 +0800
zfs-linux (0.7.5-1) unstable; urgency=medium
[ Aron Xu ]
* New upstream version 0.7.5 (Closes: #884812)
[ Antonio Russo ]
* Add version dependency on zfsutils-linux package (Closes: #880889)
-- Aron Xu <aron@debian.org> Fri, 19 Jan 2018 15:39:23 +0800
zfs-linux (0.7.4-1) unstable; urgency=medium
* New upstream version 0.7.4 (Closes: #884287, #883832)
* Require debhelper >= 10.2
* cherry-pick: fix distclean which removes needed files (Closes: #884706)
* Refresh patches
* Update stdver to 4.1.2, no change required
* Install zfs-import.target
-- Aron Xu <aron@debian.org> Mon, 18 Dec 2017 22:48:59 +0800
zfs-linux (0.7.3-3) unstable; urgency=medium
[ Antonio Russo ]
* Add maximum version dependency on spl-dkms (Closes: #883008)
-- Aron Xu <aron@debian.org> Thu, 30 Nov 2017 00:34:30 +0800
zfs-linux (0.7.3-2) unstable; urgency=medium
[ Fabian Grünbichler ]
* d/rules: remove obsolete calls to dpkg-architecture (Closes: #882209)
* zfs-test: add proper Breaks+Replaces (Closes: #880902)
* build: add implicit version to dh_makeshlibs (Closes: #880709)
[ Aron Xu ]
* Depend on matching version of spl-dkms (Closes: ##881013)
-- Aron Xu <aron@debian.org> Tue, 28 Nov 2017 16:16:34 +0800
zfs-linux (0.7.3-1) unstable; urgency=medium
[ Antonio Russo ]
* Refresh manual builds DKMS prevention patch
[ Fabian Grünbichler ]
* zfs-test package
* add files to debian/not-installed
* dh_install: switch to --fail-missing
* add new files from 0.7 to install
* dkms: build icp module as well
[ Antonio Russo ]
* dracut: make module-setup.sh shebang explicit
* add man page reference to systemd units
* Fix install path of zpool.d scripts
* Incorporate DebianPT.org Portuguese translation
* Fix typo in debconf templates
* Drop dependency on dh-systemd
[ Aron Xu ]
* Drop merged patches, update remainders
* Update std-ver to 4.1.1
* New upstream version 0.7.3
* Update debconf pot file
* Update control.in for dh-systemd deprecation
* Add lintian override for zfs-test
[ Colin King ]
* Improve cloning performance for large numbers of clones (LP: #1567557)
Bump zcmd buffer from 16K to 256K.
-- Aron Xu <aron@debian.org> Tue, 31 Oct 2017 18:52:01 +0800
zfs-linux (0.6.5.11-1) unstable; urgency=medium
[ Aron Xu ]
* Imported Upstream version 0.6.5.11
[ Fabian Grünbichler ]
* fix rm path in zfs-share.service
-- Aron Xu <aron@debian.org> Fri, 14 Jul 2017 16:33:23 +0800
zfs-linux (0.6.5.10-1) unstable; urgency=medium
* Add kernel version to depmod cmd (Closes: #860958)
* New upstream version 0.6.5.10
-- Aron Xu <aron@debian.org> Wed, 05 Jul 2017 18:11:39 +0800
zfs-linux (0.6.5.9-5) unstable; urgency=medium
* Add zfs initramfs conf for root pool setup
(Closes: #848157, LP: #1673197)
-- Aron Xu <aron@debian.org> Sun, 19 Mar 2017 18:14:57 +0800
zfs-linux (0.6.5.9-4) unstable; urgency=medium
* autopkgtest: load zfs module before running tests
-- Aron Xu <aron@debian.org> Tue, 14 Mar 2017 11:38:08 +0800
zfs-linux (0.6.5.9-3) unstable; urgency=medium
[ Petter Reinholdtsen ]
* Updated German debconf translation by Helge Kreutzmann. (Closes: #857528)
* Updated metadata on a few patches.
[ Aron Xu ]
* Cherry-pick upstream fix for merged /usr/lib and /lib
* Manually maintain adt test Depends
-- Aron Xu <aron@debian.org> Mon, 13 Mar 2017 12:23:32 +0800
zfs-linux (0.6.5.9-2) unstable; urgency=medium
[ Fabian Grünbichler ]
* fix zed-service-bindir patch
-- Aron Xu <aron@debian.org> Tue, 07 Feb 2017 17:22:02 +0800
zfs-linux (0.6.5.9-1) unstable; urgency=medium
[ Aron Xu ]
* Imported Upstream version 0.6.5.9 (Closes: #851513)
[ Lukas Wunner ]
* Cherry picks for root zpool with dracut (Closes: #849969)
* Fix installation path of systemd files
* Fix build breakage caused by nonstandard umask
[ Fabian Grünbichler ]
* fix python script install path (Closes: #842237)
-- Aron Xu <aron@debian.org> Mon, 06 Feb 2017 15:57:50 +0800
zfs-linux (0.6.5.8-3) unstable; urgency=medium
* Fix the path on the zfs-zed unit file (Closes: #849813)
-- Carlos Alberto Lopez Perez <clopez@igalia.com> Thu, 05 Jan 2017 16:23:16 +0100
zfs-linux (0.6.5.8-2) unstable; urgency=medium
[ Richard Laager ]
* Remove .py extension from utilities in /usr/sbin as per policy
10.4 Scripts (LP: #1628279)
[ Colin Ian King ]
* Use python3 for arcstat.py, arc_summary.py & dbufstat.py (LP: #1627909)
[ Richard Laager ]
* Set PATH in cron.d job to fix monthly scrubs. (LP: #1548009)
[ Aron Xu ]
* Install zed into /usr/sbin
* Rename zfsutils path to follow the package name
* Add missing part in python3 move
* Install zed to /usr/sbin
[ Petter Reinholdtsen ]
* Updated Italian debconf translation by Beatrice Torracca.
(Closes: #846928)
* Added patch 1003-linux-4.9-compat.patch from upstream to build with
Linux kernel 4.9. (Closes: #847018)
-- Aron Xu <aron@debian.org> Sat, 17 Dec 2016 17:42:21 +0800
zfs-linux (0.6.5.8-1) unstable; urgency=medium
[ Carlos Alberto Lopez Perez ]
* Reflow changelog from last upload to avoid lintian warning.
[ Aron Xu ]
* Imported Upstream version 0.6.5.8 (Closes: #838192)
* Conflicts with zutils (Closes: #836853)
[ Zhou Mo ]
* Patch: remove merged patches.
* Upstream renamed zed.service to zfs-zed.service .
* Avoid installing zfs-zed.service twice.
-- Aron Xu <aron@debian.org> Tue, 20 Sep 2016 15:20:21 +0800
zfs-linux (0.6.5.7-2) unstable; urgency=medium
[ Aron Xu ]
* Add busybox to zfs-initramfs list of dependencies. (Closes: #824976)
[ Petter Reinholdtsen ]
* Updated Danish debconf translation by Joe Hansen. (Closes: #830652)
* Added Dutch (nl) debconf translation by Frans Spiesschaert.
(Closes: #832280)
* Norwegian Bokmål (nb) debconf template translation by Petter Reinholdtsen.
[ Eric Desrochers ]
* Change utilities path (bindir) to /usr/sbin. (Closes: #832938)
[ Carlos Alberto Lopez Perez]
* Add tunable to ignore hole_birth, and enable it by default.
(Closes: #830824)
-- Carlos Alberto Lopez Perez <clopez@igalia.com> Tue, 16 Aug 2016 17:43:48 +0200
zfs-linux (0.6.5.7-1) unstable; urgency=medium
[ YunQiang Su ]
* 1002-fix-mips-build.patch: fix builds on mips* archs
[ Aron Xu ]
* New upstream release.
* 1001-Fix-aarch64-compilation.patch: dropped, not needed anymore
* Merge patches from Ubuntu:
- 0002-Check-for-META-and-DCH-consistency-in-autoconf.patch
- 0003-Add-libuutil-to-LIBADD-for-libzfs-and-libzfs_core.patch
- enable-zed.patch
-- Aron Xu <aron@debian.org> Tue, 31 May 2016 14:10:49 +0800
zfs-linux (0.6.5.6-2) unstable; urgency=medium
[ Aron Xu ]
* Adding smoke testing scripts from Ubuntu
* Fix binary module builds
* Add libblkid-dev, libattr1-dev to build-dep
* Re-sync source tree
* Add dh-python to b-d
* Remove .gitignore files and clean build tree
* Scrub all healthy pools monthly from Richard Laager
[ Petter Reinholdtsen ]
* Copied 1000-ppc64el-endian-support.patch from Ubuntu to fix endian
build problem on ppc64el
* Copied 1001-Fix-aarch64-compilation.patch from Ubuntu to fix build
problem on arm64.
* Copied 0001-Prevent-manual-builds-in-the-DKMS-source.patch from
Ubuntu to block manual building in the DKMS source tree.
* Updated Standards-Version from 3.9.7 to 3.9.8.
* Bring some files back to the upstream tarball content to get gbp
buildpackage working.
-- Petter Reinholdtsen <pere@debian.org> Thu, 12 May 2016 12:19:55 +0200
zfs-linux (0.6.5.6-1) unstable; urgency=medium
[ Aron Xu ]
* New upstream version 0.6.5.6.
[ Petter Reinholdtsen ]
* Generated new copyright.cme file based on version 0.6.5.6.
* Updated d/copyright file, add missing BSD licensed init.d scripts, and
new copyright holders in the new upstream version.
* Updated Standards-Version from 3.9.6 to 3.9.7.
* Added myself as uploader.
* Updated debconf po files based on newest pot file.
* Correct URL to git repo in d/control.
-- Petter Reinholdtsen <pere@debian.org> Sat, 26 Mar 2016 07:08:11 +0000
zfs-linux (0.6.5.5-1) unstable; urgency=medium
* Initial Release (Closes: #686447)
-- Aron Xu <aron@debian.org> Sun, 20 Mar 2016 22:57:06 +0800
+1
View File
@@ -0,0 +1 @@
10
+184
View File
@@ -0,0 +1,184 @@
Source: zfs-linux
Section: contrib/kernel
Priority: optional
Maintainer: Proxmox Support Team <support@proxmox.com>
Build-Depends: autotools-dev,
debhelper (>= 10~),
dh-autoreconf,
dh-python,
libattr1-dev,
libblkid-dev,
libselinux1-dev,
libtool,
lsb-release,
python3,
uuid-dev,
zlib1g-dev
Standards-Version: 4.1.2
Homepage: http://www.zfsonlinux.org/
Vcs-Git: https://git.proxmox.com/git/zfsonlinux.git
Vcs-Browser: https://git.proxmox.com/?p=zfsonlinux.git;a=summary
Package: libnvpair1linux
Section: contrib/libs
Architecture: linux-any
Depends: ${misc:Depends}, ${shlibs:Depends}
Breaks: libnvpair1
Replaces: libnvpair1
Description: Solaris name-value library for Linux
This library provides routines for packing and unpacking nv pairs for
transporting data across process boundaries, transporting between
kernel and userland, and possibly saving onto disk files.
Package: libuutil1linux
Section: contrib/libs
Architecture: linux-any
Depends: ${misc:Depends}, ${shlibs:Depends}
Breaks: libuutil1
Replaces: libuutil1
Description: Solaris userland utility library for Linux
This library provides a variety of glue functions for ZFS on Linux:
* libspl: The Solaris Porting Layer userland library, which provides APIs
that make it possible to run Solaris user code in a Linux environment
with relatively minimal modification.
* libavl: The Adelson-Velskii Landis balanced binary tree manipulation
library.
* libefi: The Extensible Firmware Interface library for GUID disk
partitioning.
* libshare: NFS, SMB, and iSCSI service integration for ZFS.
Package: libzfslinux-dev
Section: contrib/libdevel
Architecture: linux-any
Depends: libnvpair1linux (= ${binary:Version}),
libuutil1linux (= ${binary:Version}),
libzfs2linux (= ${binary:Version}),
libzpool2linux (= ${binary:Version}),
${misc:Depends}
Provides: libnvpair-dev, libuutil-dev
Description: OpenZFS filesystem development files for Linux
Header files and static libraries for compiling software against
libraries of OpenZFS filesystem.
.
This package includes the development files of libnvpair1, libuutil1,
libzpool2 and libzfs2.
Package: libzfs2linux
Section: contrib/libs
Architecture: linux-any
Depends: ${misc:Depends}, ${shlibs:Depends}
Breaks: libzfs2
Replaces: libzfs2
Description: OpenZFS filesystem library for Linux
The Z file system is a pooled filesystem designed for maximum data
integrity, supporting data snapshots, multiple copies, and data
checksums.
.
The OpenZFS library provides support for managing OpenZFS filesystems.
Package: libzpool2linux
Section: contrib/libs
Architecture: linux-any
Depends: ${misc:Depends}, ${shlibs:Depends}
Breaks: libzpool2
Replaces: libzpool2
Description: OpenZFS pool library for Linux
The Z file system is a pooled filesystem designed for maximum data
integrity, supporting data snapshots, multiple copies, and data
checksums.
.
This zpool library provides support for managing zpools.
Package: zfs-initramfs
Architecture: all
Depends: busybox-initramfs | busybox-static | busybox,
initramfs-tools,
zfsutils-linux (>= ${binary:Version}),
${misc:Depends}
Breaks: zfsutils-linux (<= 0.7.11-pve1~bpo1)
Replaces: zfsutils-linux (<= 0.7.11-pve1~bpo1)
Description: OpenZFS root filesystem capabilities for Linux - initramfs
The Z file system is a pooled filesystem designed for maximum data
integrity, supporting data snapshots, multiple copies, and data
checksums.
.
This package adds OpenZFS to the system initramfs with a hook
for the initramfs-tools infrastructure.
Package: zfsutils-linux
Section: contrib/admin
Architecture: linux-any
Depends: python3, ${misc:Depends}, ${python3:Depends}, ${shlibs:Depends}
Recommends: lsb-base, zfs-zed
Suggests: nfs-kernel-server,
samba-common-bin (>= 3.0.23),
zfs-initramfs
Conflicts: zfs, zfs-fuse
Provides: zfsutils
Description: command-line tools to manage OpenZFS filesystems
The Z file system is a pooled filesystem designed for maximum data
integrity, supporting data snapshots, multiple copies, and data
checksums.
.
This package provides the zfs and zpool commands to create and administer
OpenZFS filesystems.
Package: zfs-zed
Section: contrib/admin
Architecture: linux-any
Depends: zfsutils-linux (>= ${binary:Version}),
${misc:Depends},
${shlibs:Depends}
Description: OpenZFS Event Daemon
The Z file system is a pooled filesystem designed for maximum data
integrity, supporting data snapshots, multiple copies, and data
checksums.
.
This package provides the OpenZFS Event Daemon (zed).
Package: zfs-test
Section: contrib/admin
Architecture: linux-any
Depends: acl,
bc,
fio,
ksh,
lsscsi,
mdadm,
parted,
python,
sudo,
sysstat,
zfsutils-linux (>=${binary:Version}),
${misc:Depends},
${shlibs:Depends}
Breaks: zfsutils-linux (<= 0.7.9-2)
Replaces: zfsutils-linux (<= 0.7.9-2)
Conflicts: zutils
Description: OpenZFS test infrastructure an support scripts
The Z file system is a pooled filesystem designed for maximum data
integrity, supporting data snapshots, multiple copies, and data
checksums.
.
This package provides the OpenZFS test infrastructure for destructively
testing and validating a system using OpenZFS. It is entirely optional
and should only be installed and used in test environments.
Package: zfs-dbg
Section: contrib/debug
Priority: optional
Architecture: linux-any
Depends: libnvpair1linux (= ${binary:Version}),
libuutil1linux (= ${binary:Version}),
libzfs2linux (= ${binary:Version}),
libzpool2linux (= ${binary:Version}),
zfs-zed (= ${binary:Version}),
zfsutils-linux (= ${binary:Version}),
${misc:Depends}
Description: Debugging symbols for OpenZFS userland libraries and tools
The Z file system is a pooled filesystem designed for maximum data
integrity, supporting data snapshots, multiple copies, and data
checksums.
.
This package contains the debugging symbols for libzpool2linux, libzfs2linux,
libnvpair1linux, libuutil1linux, zfs-zed and zfsutils-linux.
+1579
View File
File diff suppressed because it is too large Load Diff
File diff suppressed because it is too large Load Diff
+2
View File
@@ -0,0 +1,2 @@
COPYRIGHT
OPENSOLARIS.LICENSE
+1
View File
@@ -0,0 +1 @@
lib/libnvpair.so.*
@@ -0,0 +1 @@
libnvpair1linux: package-name-doesnt-match-sonames
+2
View File
@@ -0,0 +1,2 @@
COPYRIGHT
OPENSOLARIS.LICENSE
+1
View File
@@ -0,0 +1 @@
lib/libuutil.so.*
@@ -0,0 +1 @@
libuutil1linux: package-name-doesnt-match-sonames
+2
View File
@@ -0,0 +1,2 @@
COPYRIGHT
OPENSOLARIS.LICENSE
+2
View File
@@ -0,0 +1,2 @@
lib/libzfs.so.*
lib/libzfs_core.so.*
@@ -0,0 +1 @@
libzfs2linux: package-name-doesnt-match-sonames
+2
View File
@@ -0,0 +1,2 @@
COPYRIGHT
OPENSOLARIS.LICENSE
+4
View File
@@ -0,0 +1,4 @@
lib/*.a usr/lib/
usr/include
usr/lib/*.so
usr/share/pkgconfig/*.pc
+2
View File
@@ -0,0 +1,2 @@
COPYRIGHT
OPENSOLARIS.LICENSE
+1
View File
@@ -0,0 +1 @@
lib/libzpool.so.*
@@ -0,0 +1 @@
libzpool2linux: package-name-doesnt-match-sonames
+11
View File
@@ -0,0 +1,11 @@
usr/share/zfs/enum-extract.pl
usr/share/zfs/zfs-helpers.sh
etc/init.d
etc/sudoers.d
etc/zfs/vdev_id.conf.alias.example
etc/zfs/vdev_id.conf.multipath.example
etc/zfs/vdev_id.conf.sas_direct.example
etc/zfs/vdev_id.conf.sas_switch.example
etc/zfs/vdev_id.conf.scsi.example
usr/lib/dracut
usr/share/zfs/enum-extract.pl
@@ -0,0 +1,69 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Debian ZFS on Linux maintainers
<pkg-zfsonlinux-devel@alioth-lists.debian.net>
Date: Wed, 30 Jan 2019 15:12:04 +0100
Subject: [PATCH] Check-for-META-and-DCH-consistency-in-autoconf
---
config/zfs-meta.m4 | 34 +++++++++++++++++++++++++++++-----
1 file changed, 29 insertions(+), 5 deletions(-)
diff --git a/config/zfs-meta.m4 b/config/zfs-meta.m4
index 3e1429d99..b8e26c492 100644
--- a/config/zfs-meta.m4
+++ b/config/zfs-meta.m4
@@ -1,9 +1,10 @@
dnl #
dnl # DESCRIPTION:
-dnl # Read meta data from the META file. When building from a git repository
-dnl # the ZFS_META_RELEASE field will be overwritten if there is an annotated
-dnl # tag matching the form ZFS_META_NAME-ZFS_META_VERSION-*. This allows
-dnl # for working builds to be uniquely identified using the git commit hash.
+dnl # Read meta data from the META file or the debian/changelog file if it
+dnl # exists. When building from a git repository the ZFS_META_RELEASE field
+dnl # will be overwritten if there is an annotated tag matching the form
+dnl # ZFS_META_NAME-ZFS_META_VERSION-*. This allows for working builds to be
+dnl # uniquely identified using the git commit hash.
dnl #
dnl # The META file format is as follows:
dnl # ^[ ]*KEY:[ \t]+VALUE$
@@ -49,6 +50,7 @@ AC_DEFUN([ZFS_AC_META], [
_zfs_ac_meta_type="none"
if test -f "$META"; then
_zfs_ac_meta_type="META file"
+ _dpkg_parsechangelog=$(dpkg-parsechangelog 2>/dev/null)
ZFS_META_NAME=_ZFS_AC_META_GETVAL([(Name|Project|Package)]);
if test -n "$ZFS_META_NAME"; then
@@ -66,8 +68,30 @@ AC_DEFUN([ZFS_AC_META], [
AC_SUBST([ZFS_META_VERSION])
fi
+ if test -n "${_dpkg_parsechangelog}"; then
+ _dpkg_version=$(echo "${_dpkg_parsechangelog}" \
+ | $AWK '$[]1 == "Version:" { print $[]2; }' \
+ | cut -d- -f1)
+ if test "${_dpkg_version}" != "$ZFS_META_VERSION"; then
+ AC_MSG_ERROR([
+ *** Version $ZFS_META_VERSION in the META file is different than
+ *** version $_dpkg_version in the debian/changelog file. DKMS and DEB
+ *** packaging require that these files have the same version.
+ ])
+ fi
+ fi
+
ZFS_META_RELEASE=_ZFS_AC_META_GETVAL([Release]);
- if test ! -f ".nogitrelease" && git rev-parse --git-dir > /dev/null 2>&1; then
+
+ if test -n "${_dpkg_parsechangelog}"; then
+ _dpkg_release=$(echo "${_dpkg_parsechangelog}" \
+ | $AWK '$[]1 == "Version:" { print $[]2; }' \
+ | cut -d- -f2-)
+ if test -n "${_dpkg_release}"; then
+ ZFS_META_RELEASE=${_dpkg_release}
+ _zfs_ac_meta_type="dpkg-parsechangelog"
+ fi
+ elif test ! -f ".nogitrelease" && git rev-parse --git-dir > /dev/null 2>&1; then
_match="${ZFS_META_NAME}-${ZFS_META_VERSION}"
_alias=$(git describe --match=${_match} 2>/dev/null)
_release=$(echo ${_alias}|cut -f3- -d'-'|sed 's/-/_/g')
@@ -0,0 +1,37 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Debian ZFS on Linux maintainers
<pkg-zfsonlinux-devel@alioth-lists.debian.net>
Date: Wed, 30 Jan 2019 15:12:04 +0100
Subject: [PATCH] Add-libuutil-to-LIBADD-for-libzfs-and-libzfs_core
---
lib/libzfs/Makefile.am | 1 +
lib/libzfs_core/Makefile.am | 3 ++-
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/lib/libzfs/Makefile.am b/lib/libzfs/Makefile.am
index f1260ea71..5b07b9d9d 100644
--- a/lib/libzfs/Makefile.am
+++ b/lib/libzfs/Makefile.am
@@ -31,6 +31,7 @@ nodist_libzfs_la_SOURCES = \
libzfs_la_LIBADD = \
$(top_builddir)/lib/libzfs_core/libzfs_core.la \
+ $(top_builddir)/lib/libuutil/libuutil.la \
$(top_builddir)/lib/libshare/libshare.la \
$(top_builddir)/lib/libnvpair/libnvpair.la \
$(top_builddir)/lib/libzpool/libzpool.la
diff --git a/lib/libzfs_core/Makefile.am b/lib/libzfs_core/Makefile.am
index 5eafc25c0..10d6de3d6 100644
--- a/lib/libzfs_core/Makefile.am
+++ b/lib/libzfs_core/Makefile.am
@@ -16,7 +16,8 @@ nodist_libzfs_core_la_SOURCES = \
$(KERNEL_C)
libzfs_core_la_LIBADD = \
- $(top_builddir)/lib/libnvpair/libnvpair.la
+ $(top_builddir)/lib/libnvpair/libnvpair.la \
+ $(top_builddir)/lib/libuutil/libuutil.la
libzfs_core_la_LDFLAGS = -version-info 1:0:0
@@ -0,0 +1,60 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Fabian=20Gr=C3=BCnbichler?= <f.gruenbichler@proxmox.com>
Date: Mon, 4 Sep 2017 10:59:32 +0200
Subject: [PATCH] add man page reference to systemd units
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
---
etc/systemd/system/zfs-import-cache.service.in | 1 +
etc/systemd/system/zfs-import-scan.service.in | 1 +
etc/systemd/system/zfs-mount.service.in | 1 +
etc/systemd/system/zfs-share.service.in | 1 +
4 files changed, 4 insertions(+)
diff --git a/etc/systemd/system/zfs-import-cache.service.in b/etc/systemd/system/zfs-import-cache.service.in
index 97808dee7..cb4358fb4 100644
--- a/etc/systemd/system/zfs-import-cache.service.in
+++ b/etc/systemd/system/zfs-import-cache.service.in
@@ -1,5 +1,6 @@
[Unit]
Description=Import ZFS pools by cache file
+Documentation=man:zpool(8)
DefaultDependencies=no
Requires=systemd-udev-settle.service
After=systemd-udev-settle.service
diff --git a/etc/systemd/system/zfs-import-scan.service.in b/etc/systemd/system/zfs-import-scan.service.in
index c097e3852..e7b1d8862 100644
--- a/etc/systemd/system/zfs-import-scan.service.in
+++ b/etc/systemd/system/zfs-import-scan.service.in
@@ -1,5 +1,6 @@
[Unit]
Description=Import ZFS pools by device scanning
+Documentation=man:zpool(8)
DefaultDependencies=no
Requires=systemd-udev-settle.service
After=systemd-udev-settle.service
diff --git a/etc/systemd/system/zfs-mount.service.in b/etc/systemd/system/zfs-mount.service.in
index 728fc630c..8a73716f5 100644
--- a/etc/systemd/system/zfs-mount.service.in
+++ b/etc/systemd/system/zfs-mount.service.in
@@ -1,5 +1,6 @@
[Unit]
Description=Mount ZFS filesystems
+Documentation=man:zfs(8)
DefaultDependencies=no
After=systemd-udev-settle.service
After=zfs-import.target
diff --git a/etc/systemd/system/zfs-share.service.in b/etc/systemd/system/zfs-share.service.in
index 494f5cbef..d0c93a3e6 100644
--- a/etc/systemd/system/zfs-share.service.in
+++ b/etc/systemd/system/zfs-share.service.in
@@ -1,5 +1,6 @@
[Unit]
Description=ZFS file system shares
+Documentation=man:zfs(8)
After=nfs-server.service nfs-kernel-server.service
After=smb.service
After=zfs-mount.service
@@ -18,11 +18,11 @@ Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/etc/modules-load.d/zfs.conf.in b/etc/modules-load.d/zfs.conf.in
index 8b41baa3..59b058c9 100644
index 44e1bb3ed..7509b03cb 100644
--- a/etc/modules-load.d/zfs.conf.in
+++ b/etc/modules-load.d/zfs.conf.in
@@ -1,3 +1,3 @@
# Always load kernel modules at boot. The default behavior is to load the
# kernel modules in the zfs-import-*.service or when blkid(8) detects a pool.
# The default behavior is to allow udev to load the kernel modules on demand.
# Uncomment the following line to unconditionally load them at boot.
-#zfs
+zfs
@@ -0,0 +1,38 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Fabian=20Gr=C3=BCnbichler?= <f.gruenbichler@proxmox.com>
Date: Thu, 12 Oct 2017 08:57:48 +0200
Subject: [PATCH] fix install path of zpool.d scripts
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
---
cmd/zpool/Makefile.am | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/cmd/zpool/Makefile.am b/cmd/zpool/Makefile.am
index d07f8d616..5d26f326e 100644
--- a/cmd/zpool/Makefile.am
+++ b/cmd/zpool/Makefile.am
@@ -22,11 +22,11 @@ zpool_LDADD = \
-lm $(LIBBLKID)
zpoolconfdir = $(sysconfdir)/zfs/zpool.d
-zpoolexecdir = $(libexecdir)/zfs/zpool.d
+zpoollibdir = /usr/lib/zfs-linux/zpool.d
EXTRA_DIST = zpool.d/README
-dist_zpoolexec_SCRIPTS = \
+dist_zpoollib_SCRIPTS = \
zpool.d/enc \
zpool.d/encdev \
zpool.d/fault_led \
@@ -119,5 +119,5 @@ install-data-hook:
for f in $(zpoolconfdefaults); do \
test -f "$(DESTDIR)$(zpoolconfdir)/$${f}" -o \
-L "$(DESTDIR)$(zpoolconfdir)/$${f}" || \
- ln -s "$(zpoolexecdir)/$${f}" "$(DESTDIR)$(zpoolconfdir)"; \
+ ln -s "$(zpoollibdir)/$${f}" "$(DESTDIR)$(zpoolconfdir)"; \
done
@@ -0,0 +1,41 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Mon, 17 Oct 2016 14:30:56 +0800
Subject: [PATCH] Use python3 for arcstat.py, arc_summary.py & dbufstat.py
---
cmd/arc_summary/arc_summary.py | 2 +-
cmd/arcstat/arcstat.py | 2 +-
cmd/dbufstat/dbufstat.py | 2 +-
3 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/cmd/arc_summary/arc_summary.py b/cmd/arc_summary/arc_summary.py
index f6dbb9bfb..723c2e5a6 100755
--- a/cmd/arc_summary/arc_summary.py
+++ b/cmd/arc_summary/arc_summary.py
@@ -1,4 +1,4 @@
-#!/usr/bin/python
+#!/usr/bin/python3
#
# $Id: arc_summary.pl,v 388:e27800740aa2 2011-07-08 02:53:29Z jhell $
#
diff --git a/cmd/arcstat/arcstat.py b/cmd/arcstat/arcstat.py
index d7d3e9b73..3a7a47d67 100755
--- a/cmd/arcstat/arcstat.py
+++ b/cmd/arcstat/arcstat.py
@@ -1,4 +1,4 @@
-#!/usr/bin/python
+#!/usr/bin/python3
#
# Print out ZFS ARC Statistics exported via kstat(1)
# For a definition of fields, or usage, use arctstat.pl -v
diff --git a/cmd/dbufstat/dbufstat.py b/cmd/dbufstat/dbufstat.py
index 42bb0c7c7..73e02ca11 100755
--- a/cmd/dbufstat/dbufstat.py
+++ b/cmd/dbufstat/dbufstat.py
@@ -1,4 +1,4 @@
-#!/usr/bin/python
+#!/usr/bin/python3
#
# Print out statistics for all cached dmu buffers. This information
# is available through the dbufs kstat and may be post-processed as
@@ -0,0 +1,25 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Fabian=20Gr=C3=BCnbichler?= <f.gruenbichler@proxmox.com>
Date: Mon, 6 Feb 2017 12:04:35 +0100
Subject: [PATCH] Fix the path to the zed binary on the systemd unit.
We install zed into /usr/sbin manually meanwhile the upstream default is
installing it into /sbin. Ubuntu packages also install zed to /usr/sbin, but
they ship their own zfs-zed unit.
---
etc/systemd/system/zfs-zed.service.in | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/etc/systemd/system/zfs-zed.service.in b/etc/systemd/system/zfs-zed.service.in
index e3dec3dca..0b85f169a 100644
--- a/etc/systemd/system/zfs-zed.service.in
+++ b/etc/systemd/system/zfs-zed.service.in
@@ -5,7 +5,7 @@ After=zfs-import-cache.service
After=zfs-import-scan.service
[Service]
-ExecStart=@sbindir@/zed -F
+ExecStart=/usr/sbin/zed -F
Restart=on-abort
[Install]
@@ -0,0 +1,75 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.king@canonical.com>
Date: Tue, 31 Oct 2017 19:12:42 +0800
Subject: [PATCH] increase-default-zcmd-allocation-to-256K
Increase default zcmd allocation to 256K (LP: #567557)
When creating hundreds of clones (for example using containers with
LXD) cloning slows down as the number of clones increases over time.
The reason for this is that the fetching of the clone information
using a small zcmd buffer requires two ioctl calls, one to determine
the size and a second to return the data. However, this requires
gathering the data twice, once to determine the size and again to
populate the zcmd buffer to return it to userspace.
These are expensive ioctl() calls, so instead, make the default buffer
size much larger: 256K. This may sound large, but on 64 bit systems
running ZFS this is not a huge chunk of memory for the speed
improvement we gains for large sets of clones:
16K zcmd 256K zcmd
Clones Time Clones Time Clone % improvement
(secs) per sec (secs) per sec
100 7 14.29 5 20.00 28.57
200 10 20.00 9 22.22 10.00
300 19 15.79 18 16.67 5.26
400 22 18.18 22 18.18 0.00
500 29 17.24 29 17.24 0.00
600 39 15.38 39 15.38 0.00
700 46 15.22 45 15.56 2.17
800 58 13.79 51 15.69 12.07
900 74 12.16 61 14.75 17.57
1000 90 11.11 74 13.51 17.78
1100 98 11.22 87 12.64 11.22
1200 102 11.76 95 12.63 6.86
1300 113 11.50 104 12.50 7.96
1400 143 9.79 109 12.84 23.78
1500 145 10.34 132 11.36 8.97
1600 165 9.70 145 11.03 12.12
1700 187 9.09 156 10.90 16.58
1800 210 8.57 166 10.84 20.95
1900 226 8.41 183 10.38 19.03
2000 256 7.81 198 10.10 22.66
2200 311 7.07 238 9.24 23.47
2400 373 6.43 271 8.86 27.35
2600 487 5.34 316 8.23 35.11
3000 619 4.85 426 7.04 31.18
3400 915 3.72 549 6.19 40.00
4000 1332 3.00 923 4.33 30.71
As one can see, with > 2000 clones we get 25-40% speed
improvement.
This patch was originally suggested by Brian Behlendorf
(see https://github.com/zfsonlinux/zfs/issues/6372), however
this fix is a more generic fix to cover all zcmd cases.
Signed-off-by: Colin Ian King <colin.king@canonical.com>
---
lib/libzfs/libzfs_util.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/libzfs/libzfs_util.c b/lib/libzfs/libzfs_util.c
index 9703f844f..0cfcf9c2f 100644
--- a/lib/libzfs/libzfs_util.c
+++ b/lib/libzfs/libzfs_util.c
@@ -1355,7 +1355,7 @@ int
zcmd_alloc_dst_nvlist(libzfs_handle_t *hdl, zfs_cmd_t *zc, size_t len)
{
if (len == 0)
- len = 16 * 1024;
+ len = 256 * 1024;
zc->zc_nvlist_dst_size = len;
zc->zc_nvlist_dst =
(uint64_t)(uintptr_t)zfs_alloc(hdl, zc->zc_nvlist_dst_size);
@@ -13,13 +13,13 @@ Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/etc/systemd/system/zfs-import-scan.service.in b/etc/systemd/system/zfs-import-scan.service.in
index abc8e8e6..8fe2c107 100644
index e7b1d8862..060540f63 100644
--- a/etc/systemd/system/zfs-import-scan.service.in
+++ b/etc/systemd/system/zfs-import-scan.service.in
@@ -12,7 +12,7 @@ ConditionPathExists=!@sysconfdir@/zfs/zpool.cache
[Service]
Type=oneshot
RemainAfterExit=yes
ExecStartPre=-/sbin/modprobe zfs
-ExecStart=@sbindir@/zpool import -aN -o cachefile=none
+ExecStart=@sbindir@/zpool import -aN -d /dev/disk/by-id -o cachefile=none
@@ -28,7 +28,7 @@ Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/module/zfs/zfs_ctldir.c b/module/zfs/zfs_ctldir.c
index 98897fe6..0ab5b4f0 100644
index bf5a1d059..2964b65a6 100644
--- a/module/zfs/zfs_ctldir.c
+++ b/module/zfs/zfs_ctldir.c
@@ -358,8 +358,6 @@ snapentry_expire(void *data)
@@ -0,0 +1,34 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Richard Laager <rlaager@wiktel.com>
Date: Wed, 30 Jan 2019 15:12:04 +0100
Subject: [PATCH] Enable zed emails
The OpenZFS event daemon monitors pools. This patch enables the email sending
function by default (if zed is installed). This is consistent with the default
behavior of mdadm.
---
cmd/zed/zed.d/zed.rc | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/cmd/zed/zed.d/zed.rc b/cmd/zed/zed.d/zed.rc
index 35a4d1275..c6c106b06 100644
--- a/cmd/zed/zed.d/zed.rc
+++ b/cmd/zed/zed.d/zed.rc
@@ -15,7 +15,7 @@
# Email will only be sent if ZED_EMAIL_ADDR is defined.
# Disabled by default; uncomment to enable.
#
-#ZED_EMAIL_ADDR="root"
+ZED_EMAIL_ADDR="root"
##
# Name or path of executable responsible for sending notifications via email;
@@ -41,7 +41,7 @@
##
# Minimum number of seconds between notifications for a similar event.
#
-#ZED_NOTIFY_INTERVAL_SECS=3600
+ZED_NOTIFY_INTERVAL_SECS=3600
##
# Notification verbosity.
+11
View File
@@ -0,0 +1,11 @@
0001-Check-for-META-and-DCH-consistency-in-autoconf.patch
0002-Add-libuutil-to-LIBADD-for-libzfs-and-libzfs_core.patch
0003-add-man-page-reference-to-systemd-units.patch
0004-always-load-ZFS-module-on-boot.patch
0005-fix-install-path-of-zpool.d-scripts.patch
0006-Use-python3-for-arcstat.py-arc_summary.py-dbufstat.p.patch
0007-Fix-the-path-to-the-zed-binary-on-the-systemd-unit.patch
0008-increase-default-zcmd-allocation-to-256K.patch
0009-import-with-d-dev-disk-by-id-in-scan-service.patch
0010-Fix-deadlock-between-zfs-umount-snapentry_expire.patch
0011-Enable-zed-emails.patch
+87
View File
@@ -0,0 +1,87 @@
#!/usr/bin/make -f
include /usr/share/dpkg/default.mk
export DEB_BUILD_MAINT_OPTIONS = hardening=+all
%:
dh $@ --with autoreconf,python3,systemd --parallel
override_dh_autoreconf:
@# Embed the downstream version in the module.
@sed -e 's/^Version:.*/Version: $(DEB_VERSION_UPSTREAM)/' -i.orig META
dh_autoreconf
override_dh_auto_configure:
@# Build the userland, but don't build the kernel modules.
dh_auto_configure -- \
--bindir=/usr/bin \
--sbindir=/sbin \
--libdir=/lib \
--with-udevdir=/lib/udev \
--enable-systemd \
--with-systemdunitdir=/lib/systemd/system \
--with-systemdpresetdir=/lib/systemd/system-preset \
--with-config=user
override_dh_auto_test:
# The dh_auto_test rule is disabled because
# `make check` cannot run in an unprivileged build environment.
override_dh_auto_install:
@# Install the utilities.
$(MAKE) install DESTDIR='$(CURDIR)/debian/tmp'
# Use upstream's bash completion
install -D -t '$(CURDIR)/debian/tmp/usr/share/bash-completion/completions/' \
'$(CURDIR)/contrib/bash_completion.d/zfs'
# Move from bin_dir to /usr/sbin
# Remove suffix (.py) as per policy 10.4 - Scripts
# https://www.debian.org/doc/debian-policy/ch-files.html#s-scripts
mkdir -p '$(CURDIR)/debian/tmp/usr/sbin/'
mv '$(CURDIR)/debian/tmp/usr/bin/arc_summary.py' '$(CURDIR)/debian/tmp/usr/sbin/arc_summary'
mv '$(CURDIR)/debian/tmp/usr/bin/arcstat.py' '$(CURDIR)/debian/tmp/usr/sbin/arcstat'
mv '$(CURDIR)/debian/tmp/usr/bin/dbufstat.py' '$(CURDIR)/debian/tmp/usr/sbin/dbufstat'
@# Zed has dependencies outside of the system root.
mv '$(CURDIR)/debian/tmp/sbin/zed' '$(CURDIR)/debian/tmp/usr/sbin/zed'
mkdir -p $(CURDIR)/debian/tmp/usr/lib
for i in `ls $(CURDIR)/debian/tmp/lib/*.so`; do \
ln -s /lib/`readlink $${i}` $(CURDIR)/debian/tmp/usr/lib/`basename $${i}`; \
rm $${i}; \
done
chmod a-x '$(CURDIR)/debian/tmp/etc/zfs/zfs-functions'
chmod a-x '$(CURDIR)/debian/tmp/etc/default/zfs'
chmod a-x '$(CURDIR)/debian/tmp/usr/share/bash-completion/completions/zfs'
override_dh_makeshlibs:
dh_makeshlibs -a -V
ifeq ($(BUILD_UDEB), true)
dh_makeshlibs -V -plibnvpair1linux
dh_makeshlibs -V -plibuutil1linux
dh_makeshlibs -V -plibzfs2linux
dh_makeshlibs -V -plibzpool2linux
dh_makeshlibs -V -pzfsutils-linux
endif
override_dh_strip:
dh_strip --dbg-package=zfs-dbg
override_dh_auto_clean:
find . -name .gitignore -delete
rm -rf zfs-$(DEB_VERSION_UPSTREAM)
dh_auto_clean
@if test -e META.orig; then mv META.orig META; fi
override_dh_install:
find . -name lib*.la -delete
dh_install --fail-missing
# ------------
debian-copyright:
cme update dpkg-copyright -file debian/copyright.cme
+1
View File
@@ -0,0 +1 @@
3.0 (quilt)
@@ -0,0 +1,8 @@
for x in $(cat /proc/cmdline)
do
case $x in
root=ZFS=*)
BOOT=zfs
;;
esac
done
@@ -0,0 +1,67 @@
#!/bin/sh
#
# Add udev rules for ZoL to the initrd.
#
PREREQ="udev"
PREREQ_UDEV_RULES="60-zvol.rules 69-vdev.rules"
COPY_EXEC_LIST="/lib/udev/zvol_id /lib/udev/vdev_id"
# Generic result code.
RC=0
case $1 in
prereqs)
echo "$PREREQ"
exit 0
;;
esac
for ii in $COPY_EXEC_LIST
do
if [ ! -x "$ii" ]
then
echo "Error: $ii is not executable."
RC=2
fi
done
if [ "$RC" -ne 0 ]
then
exit "$RC"
fi
. /usr/share/initramfs-tools/hook-functions
mkdir -p "$DESTDIR/lib/udev/rules.d/"
for ii in $PREREQ_UDEV_RULES
do
if [ -e "/etc/udev/rules.d/$ii" ]
then
cp -p "/etc/udev/rules.d/$ii" "$DESTDIR/lib/udev/rules.d/"
elif [ -e "/lib/udev/rules.d/$ii" ]
then
cp -p "/lib/udev/rules.d/$ii" "$DESTDIR/lib/udev/rules.d/"
else
echo "Error: Missing udev rule: $ii"
echo " This file must be in the /etc/udev/rules.d or /lib/udev/rules.d directory."
exit 1
fi
done
for ii in $COPY_EXEC_LIST
do
copy_exec "$ii"
done
if [ -f '/etc/default/zfs' -a -r '/etc/default/zfs' ]
then
mkdir -p "$DESTDIR/etc/default"
cp -a '/etc/default/zfs' "$DESTDIR/etc/default/"
fi
if [ -d '/etc/zfs' -a -r '/etc/zfs' ]
then
mkdir -p "$DESTDIR/etc"
cp -a '/etc/zfs' "$DESTDIR/etc/"
fi
@@ -0,0 +1,18 @@
## Allow read-only ZoL commands to be called through sudo
## without a password. Remove the first '#' column to enable.
##
## CAUTION: Any syntax error introduced here will break sudo.
##
## Cmnd alias specification
#Cmnd_Alias C_ZFS = \
# /sbin/zfs "", /sbin/zfs help *, \
# /sbin/zfs get, /sbin/zfs get *, \
# /sbin/zfs list, /sbin/zfs list *, \
# /sbin/zpool "", /sbin/zpool help *, \
# /sbin/zpool iostat, /sbin/zpool iostat *, \
# /sbin/zpool list, /sbin/zpool list *, \
# /sbin/zpool status, /sbin/zpool status *, \
# /sbin/zpool upgrade, /sbin/zpool upgrade -v
#
## allow any user to use basic read-only ZFS commands
#ALL ALL = (root) NOPASSWD: C_ZFS
@@ -0,0 +1,14 @@
#!/bin/sh
# Sub-test to exclude ZVOLs
set -e
partition="$1"
. /usr/share/os-prober/common.sh
if [ "$(stat -L -c %t "$partition")" = "e6" ] ; then
debug "$1 is a ZVOL; skipping"
exit 0
fi
# No ZVOLs found
exit 1
+9
View File
@@ -0,0 +1,9 @@
#!/bin/sh -eu
# Scrub all healthy pools.
zpool list -H -o health,name 2>&1 | \
awk 'BEGIN {FS="\t"} {if ($1 ~ /^ONLINE/) print $2}' | \
while read pool
do
zpool scrub "$pool"
done
+2
View File
@@ -0,0 +1,2 @@
version=3
https://zfsonlinux.org/ .*zfs-([\d\.]+)\.tar\.gz$
+2
View File
@@ -0,0 +1,2 @@
../tree/zfs-initramfs/* /
usr/share/initramfs-tools/*
+1
View File
@@ -0,0 +1 @@
activate update-initramfs
+21
View File
@@ -0,0 +1,21 @@
sbin/zinject
sbin/zpios
sbin/ztest
usr/bin/raidz_test
usr/share/man/man1/raidz_test.1
usr/share/man/man1/test-runner.1
usr/share/man/man1/zpios.1
usr/share/man/man1/ztest.1
usr/share/man/man8/zinject.8
usr/share/zfs/common.sh
usr/share/zfs/runfiles/
usr/share/zfs/smb.sh
usr/share/zfs/test-runner
usr/share/zfs/zfs-tests.sh
usr/share/zfs/zfs-tests/
usr/share/zfs/zfs.sh
usr/share/zfs/zimport.sh
usr/share/zfs/zloop.sh
usr/share/zfs/zpios*
usr/share/zfs/zpool-config/
usr/share/zfs/zpool-create.sh
+1
View File
@@ -0,0 +1 @@
zfs-test: arch-dependent-file-in-usr-share
+5
View File
@@ -0,0 +1,5 @@
etc/zfs/zed.d/*
lib/systemd/system/zfs-zed.service
usr/lib/*/zfs/zed.d/*
usr/sbin/zed
usr/share/man/man8/zed.8
+4
View File
@@ -0,0 +1,4 @@
PATH=/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin
# Scrub the second Sunday of every month.
24 0 8-14 * * root [ $(date +\%w) -eq 0 ] && [ -x /usr/lib/zfs-linux/scrub ] && /usr/lib/zfs-linux/scrub
+2
View File
@@ -0,0 +1,2 @@
COPYRIGHT
OPENSOLARIS.LICENSE
+5
View File
@@ -0,0 +1,5 @@
etc/zfs/vdev_id.conf.alias.example
etc/zfs/vdev_id.conf.multipath.example
etc/zfs/vdev_id.conf.sas_direct.example
etc/zfs/vdev_id.conf.sas_switch.example
etc/zfs/vdev_id.conf.scsi.example
+36
View File
@@ -0,0 +1,36 @@
../tree/zfsutils-linux/* /
etc/default/zfs
etc/zfs/zfs-functions
etc/zfs/zpool.d/
lib/systemd/system-preset/
lib/systemd/system/zfs-import-cache.service
lib/systemd/system/zfs-import-scan.service
lib/systemd/system/zfs-import.target
lib/systemd/system/zfs-mount.service
lib/systemd/system/zfs-share.service
lib/systemd/system/zfs.target
lib/udev/
sbin/fsck.zfs
sbin/mount.zfs
sbin/zdb
sbin/zfs
sbin/zhack
sbin/zpool
sbin/zstreamdump
usr/bin/zgenhostid
usr/lib/modules-load.d/ lib/
usr/lib/zfs-linux/zpool.d/
usr/sbin/arc_summary
usr/sbin/arcstat
usr/sbin/dbufstat
usr/share/bash-completion/completions
usr/share/man/man1/zhack.1
usr/share/man/man5/
usr/share/man/man8/fsck.zfs.8
usr/share/man/man8/mount.zfs.8
usr/share/man/man8/vdev_id.8
usr/share/man/man8/zdb.8
usr/share/man/man8/zfs.8
usr/share/man/man8/zgenhostid.8
usr/share/man/man8/zpool.8
usr/share/man/man8/zstreamdump.8
@@ -0,0 +1 @@
systemd-service-file-refers-to-unusual-wantedby-target
Submodule
+1
Submodule zfs/upstream added at a8c2b7ebc6