Merge branch 'master' into issue-8478

.editorconfig

@@ -0,0 +1,10 @@
+root = true
+
+[*]
+end_of_line = lf
+insert_final_newline = true
+trim_trailing_whitespace = true
+
+[*.{c,h}]
+tab_width = 8
+indent_style = tab

.github/CONTRIBUTING.md

@@ -1,10 +1,12 @@
-# Contributing to ZFS on Linux
+# Contributing to OpenZFS
-<p align="center"><img src="http://zfsonlinux.org/images/zfs-linux.png"/></p>
+<p align="center">
+  <img alt="OpenZFS Logo"
+    src="https://openzfs.github.io/openzfs-docs/_static/img/logo/480px-Open-ZFS-Secondary-Logo-Colour-halfsize.png"/>
+</p>
 
 *First of all, thank you for taking the time to contribute!*
 
-By using the following guidelines, you can help us make ZFS on Linux even
+By using the following guidelines, you can help us make OpenZFS even better.
-better.
 
 ## Table Of Contents
 [What should I know before I get
@@ -32,17 +34,17 @@ started?](#what-should-i-know-before-i-get-started)
 
 Helpful resources
 
-  * [ZFS on Linux wiki](https://github.com/zfsonlinux/zfs/wiki)
+  * [OpenZFS Documentation](https://openzfs.github.io/openzfs-docs/)
-  * [OpenZFS Documentation](http://open-zfs.org/wiki/Developer_resources)
+  * [OpenZFS Developer Resources](http://open-zfs.org/wiki/Developer_resources)
-  * [Git and GitHub for beginners](https://github.com/zfsonlinux/zfs/wiki/Git-and-GitHub-for-beginners)
+  * [Git and GitHub for beginners](https://openzfs.github.io/openzfs-docs/Developer%20Resources/Git%20and%20GitHub%20for%20beginners.html)
 
 ## What should I know before I get started?
 
 ### Get ZFS
 You can build zfs packages by following [these
-instructions](https://github.com/zfsonlinux/zfs/wiki/Building-ZFS),
+instructions](https://openzfs.github.io/openzfs-docs/Developer%20Resources/Building%20ZFS.html),
 or install stable packages from [your distribution's
-repository](https://github.com/zfsonlinux/zfs/wiki/Getting-Started).
+repository](https://openzfs.github.io/openzfs-docs/Getting%20Started/index.html).
 
 ### Debug ZFS
 A variety of methods and tools are available to aid ZFS developers.
@@ -51,29 +53,30 @@ configure option should be set. This will enable additional correctness
 checks and all the ASSERTs to help quickly catch potential issues.
 
 In addition, there are numerous utilities and debugging files which
-provide visibility in to the inner workings of ZFS.  The most useful
+provide visibility into the inner workings of ZFS.  The most useful
-of these tools are discussed in detail on the [debugging ZFS wiki
+of these tools are discussed in detail on the [Troubleshooting
-page](https://github.com/zfsonlinux/zfs/wiki/Debugging).
+page](https://openzfs.github.io/openzfs-docs/Basic%20Concepts/Troubleshooting.html).
 
 ### Where can I ask for help?
-[The zfs-discuss mailing list or IRC](http://list.zfsonlinux.org)
+The [zfs-discuss mailing
-are the best places to ask for help. Please do not file support requests
+list](https://openzfs.github.io/openzfs-docs/Project%20and%20Community/Mailing%20Lists.html)
-on the GitHub issue tracker.
+or IRC are the best places to ask for help. Please do not file
+support requests on the GitHub issue tracker.
 
 ## How Can I Contribute?
 
 ### Reporting Bugs
 *Please* contact us via the [zfs-discuss mailing
-list or IRC](http://list.zfsonlinux.org) if you aren't
+list](https://openzfs.github.io/openzfs-docs/Project%20and%20Community/Mailing%20Lists.html)
-certain that you are experiencing a bug.
+or IRC if you aren't certain that you are experiencing a bug.
 
 If you run into an issue, please search our [issue
-tracker](https://github.com/zfsonlinux/zfs/issues) *first* to ensure the
+tracker](https://github.com/openzfs/zfs/issues) *first* to ensure the
 issue hasn't been reported before. Open a new issue only if you haven't
 found anything similar to your issue.
 
 You can open a new issue and search existing issues using the public [issue
-tracker](https://github.com/zfsonlinux/zfs/issues).
+tracker](https://github.com/openzfs/zfs/issues).
 
 #### When opening a new issue, please include the following information at the top of the issue:
 * What distribution (with version) you are using.
@@ -105,13 +108,13 @@ information like:
 * Stack traces which may be logged to `dmesg`.
 
 ### Suggesting Enhancements
-ZFS on Linux is a widely deployed production filesystem which is under
+OpenZFS is a widely deployed production filesystem which is under active
-active development. The team's primary focus is on fixing known issues,
+development. The team's primary focus is on fixing known issues, improving
-improving performance, and adding compelling new features.
+performance, and adding compelling new features.
 
 You can view the list of proposed features
-by filtering the issue tracker by the ["Feature"
+by filtering the issue tracker by the ["Type: Feature"
-label](https://github.com/zfsonlinux/zfs/issues?q=is%3Aopen+is%3Aissue+label%3AFeature).
+label](https://github.com/openzfs/zfs/issues?q=is%3Aopen+is%3Aissue+label%3A%22Type%3A+Feature%22).
 If you have an idea for a feature first check this list. If your idea already
 appears then add a +1 to the top most comment, this helps us gauge interest
 in that feature.
@@ -120,8 +123,11 @@ Otherwise, open a new issue and describe your proposed feature.  Why is this
 feature needed?  What problem does it solve?
 
 ### Pull Requests
-* All pull requests must be based on the current master branch and apply
+
-without conflicts.
+#### General
+
+* All pull requests, except backports and releases, must be based on the current master branch 
+and should apply without conflicts.
 * Please attempt to limit pull requests to a single commit which resolves
 one specific issue.
 * Make sure your commit messages are in the correct format. See the
@@ -133,16 +139,28 @@ logically independent patches which build on each other.  This makes large
 changes easier to review and approve which speeds up the merging process.
 * Try to keep pull requests simple. Simple code with comments is much easier
 to review and approve.
+* All proposed changes must be approved by an OpenZFS organization member.
+* If you have an idea you'd like to discuss or which requires additional testing, consider opening it as a draft pull request.
+Once everything is in good shape and the details have been worked out you can remove its draft status.
+Any required reviews can then be finalized and the pull request merged.
+
+#### Tests and Benchmarks
+* Every pull request will by tested by the buildbot on multiple platforms by running the [zfs-tests.sh and zloop.sh](
+https://openzfs.github.io/openzfs-docs/Developer%20Resources/Building%20ZFS.html#running-zloop-sh-and-zfs-tests-sh) test suites.
+* To verify your changes conform to the [style guidelines](
+https://github.com/openzfs/zfs/blob/master/.github/CONTRIBUTING.md#style-guides
+), please run `make checkstyle` and resolve any warnings.
+* Static code analysis of each pull request is performed by the buildbot; run `make lint` to check your changes.
 * Test cases should be provided when appropriate.
+This includes making sure new features have adequate code coverage.
 * If your pull request improves performance, please include some benchmarks.
 * The pull request must pass all required [ZFS
 Buildbot](http://build.zfsonlinux.org/) builders before
 being accepted. If you are experiencing intermittent TEST
 builder failures, you may be experiencing a [test suite
-issue](https://github.com/zfsonlinux/zfs/issues?q=is%3Aissue+is%3Aopen+label%3A%22Test+Suite%22).
+issue](https://github.com/openzfs/zfs/issues?q=is%3Aissue+is%3Aopen+label%3A%22Type%3A+Test+Suite%22).
-There are also various [buildbot options](https://github.com/zfsonlinux/zfs/wiki/Buildbot-Options)
+There are also various [buildbot options](https://openzfs.github.io/openzfs-docs/Developer%20Resources/Buildbot%20Options.html)
 to control how changes are tested.
-* All proposed changes must be approved by a ZFS on Linux organization member.
 
 ### Testing
 All help is appreciated! If you're in a position to run the latest code
@@ -152,16 +170,41 @@ range of realistic workloads, configurations and architectures we're better
 able quickly identify and resolve potential issues.
 
 Users can also run the [ZFS Test
-Suite](https://github.com/zfsonlinux/zfs/tree/master/tests) on their systems
+Suite](https://github.com/openzfs/zfs/tree/master/tests) on their systems
 to verify ZFS is behaving as intended.
 
 ## Style Guides
 
+### Repository Structure
+
+OpenZFS uses a standardised branching structure.
+- The "development and main branch", is the branch all development should be based on.
+- "Release branches" contain the latest released code for said version.
+- "Staging branches" contain selected commits prior to being released.
+
+**Branch Names:**
+- Development and Main branch: `master`
+- Release branches: `zfs-$VERSION-release`
+- Staging branches: `zfs-$VERSION-staging`
+
+`$VERSION` should be replaced with the `major.minor` version number.  
+_(This is the version number without the `.patch` version at the end)_
+
 ### Coding Conventions
 We currently use [C  Style  and  Coding  Standards  for
 SunOS](http://www.cis.upenn.edu/%7Elee/06cse480/data/cstyle.ms.pdf) as our
 coding convention.
 
+This repository has an `.editorconfig` file. If your editor [supports
+editorconfig](https://editorconfig.org/#download), it will
+automatically respect most of this project's whitespace preferences.
+
+Additionally, Git can help warn on whitespace problems as well:
+
+```
+git config --local core.whitespace trailing-space,space-before-tab,indent-with-non-tab,-tab-in-indent
+```
+
 ### Commit Message Formats
 #### New Changes
 Commit messages for new changes must meet the following guidelines:
@@ -187,70 +230,6 @@ attempting to solve.
 Signed-off-by: Contributor <contributor@email.com>
 ```
 
-#### OpenZFS Patch Ports
-If you are porting OpenZFS patches, the commit message must meet
-the following guidelines:
-* The first line must be the summary line from the most important OpenZFS commit being ported.
-It must begin with `OpenZFS dddd, dddd - ` where `dddd` are OpenZFS issue numbers.
-* Provides a `Authored by:` line to attribute each patch for each original author.
-* Provides the `Reviewed by:` and `Approved by:` lines from each original
-OpenZFS commit.
-* Provides a `Ported-by:` line with the developer's name followed by
-their email for each OpenZFS commit.
-* Provides a `OpenZFS-issue:` line with link for each original illumos
-issue.
-* Provides a `OpenZFS-commit:` line with link for each original OpenZFS commit.
-* If necessary, provide some porting notes to describe any deviations from
-the original OpenZFS commits.
-
-An example OpenZFS patch port commit message for a single patch is provided
-below.
-```
-OpenZFS 1234 - Summary from the original OpenZFS commit
-
-Authored by: Original Author <original@email.com>
-Reviewed by: Reviewer One <reviewer1@email.com>
-Reviewed by: Reviewer Two <reviewer2@email.com>
-Approved by: Approver One <approver1@email.com>
-Ported-by: ZFS Contributor <contributor@email.com>
-
-Provide some porting notes here if necessary.
-
-OpenZFS-issue: https://www.illumos.org/issues/1234
-OpenZFS-commit: https://github.com/openzfs/openzfs/commit/abcd1234
-```
-
-If necessary, multiple OpenZFS patches can be combined in a single port.
-This is useful when you are porting a new patch and its subsequent bug
-fixes. An example commit message is provided below.
-```
-OpenZFS 1234, 5678 - Summary of most important OpenZFS commit
-
-1234 Summary from original OpenZFS commit for 1234
-
-Authored by: Original Author <original@email.com>
-Reviewed by: Reviewer Two <reviewer2@email.com>
-Approved by: Approver One <approver1@email.com>
-Ported-by: ZFS Contributor <contributor@email.com>
-
-Provide some porting notes here for 1234 if necessary.
-
-OpenZFS-issue: https://www.illumos.org/issues/1234
-OpenZFS-commit: https://github.com/openzfs/openzfs/commit/abcd1234
-
-5678 Summary from original OpenZFS commit for 5678
-
-Authored by: Original Author2 <original2@email.com>
-Reviewed by: Reviewer One <reviewer1@email.com>
-Approved by: Approver Two <approver2@email.com>
-Ported-by: ZFS Contributor <contributor@email.com>
-
-Provide some porting notes here for 5678 if necessary.
-
-OpenZFS-issue: https://www.illumos.org/issues/5678
-OpenZFS-commit: https://github.com/openzfs/openzfs/commit/efgh5678
-```
-
 #### Coverity Defect Fixes
 If you are submitting a fix to a
 [Coverity defect](https://scan.coverity.com/projects/zfsonlinux-zfs),
@@ -290,3 +269,13 @@ Git can append the `Signed-off-by` line to your commit messages. Simply
 provide the `-s` or `--signoff` option when performing a `git commit`.
 For more information about writing commit messages, visit [How to Write
 a Git Commit Message](https://chris.beams.io/posts/git-commit/).
+
+#### Co-authored By
+If someone else had part in your pull request, please add the following to the commit:
+`Co-authored-by: Name <gitregistered@email.address>`
+This is useful if their authorship was lost during squashing, rebasing, etc.,
+but may be used in any situation where there are co-authors.
+
+The email address used here should be the same as on the GitHub profile of said user.
+If said user does not have their email address public, please use the following instead:
+`Co-authored-by: Name <[username]@users.noreply.github.com>`

.github/ISSUE_TEMPLATE.md

@@ -1,48 +0,0 @@
-<!-- Please fill out the following template, which will help other contributors address your issue. -->
-
-<!--
-Thank you for reporting an issue.
-
-*IMPORTANT* - Please search our issue tracker *before* making a new issue.
-If you cannot find a similar issue, then create a new issue.
-https://github.com/zfsonlinux/zfs/issues 
-
-*IMPORTANT* - This issue tracker is for *bugs* and *issues* only.
-Please search the wiki and the mailing list archives before asking 
-questions on the mailing list.
-https://github.com/zfsonlinux/zfs/wiki/Mailing-Lists
-
-Please fill in as much of the template as possible.
--->
-
-### System information
-<!--  add version after "|" character -->
-Type | Version/Name
- --- | --- 
-Distribution Name	| 
-Distribution Version	| 
-Linux Kernel	| 
-Architecture	| 
-ZFS Version	| 
-SPL Version	| 
-<!-- 
-Commands to find ZFS/SPL versions:
-modinfo zfs | grep -iw version
-modinfo spl | grep -iw version 
--->
-
-### Describe the problem you're observing
-
-### Describe how to reproduce the problem
-
-### Include any warning/errors/backtraces from the system logs
-<!-- 
-*IMPORTANT* - Please mark logs and text output from terminal commands 
-or else Github will not display them correctly. 
-An example is provided below.
-
-Example:
-```
-this is an example how log text should be marked (wrap it with ```)
-```
--->

.github/ISSUE_TEMPLATE/bug_report.md

@@ -0,0 +1,55 @@
+---
+name: Bug report
+about: Create a report to help us improve OpenZFS
+title: ''
+labels: 'Type: Defect'
+assignees: ''
+
+---
+
+<!-- Please fill out the following template, which will help other contributors address your issue. -->
+
+<!--
+Thank you for reporting an issue.
+
+*IMPORTANT* - Please check our issue tracker before opening a new issue.
+Additional valuable information can be found in the OpenZFS documentation
+and mailing list archives.
+
+Please fill in as much of the template as possible.
+-->
+
+### System information
+<!--  add version after "|" character -->
+Type | Version/Name
+ --- | ---
+Distribution Name	|
+Distribution Version	|
+Kernel Version	|
+Architecture	|
+OpenZFS Version	|
+<!--
+Command to find OpenZFS version:
+zfs version
+
+Commands to find kernel version:
+uname -r		# Linux
+freebsd-version -r	# FreeBSD
+-->
+
+### Describe the problem you're observing
+
+### Describe how to reproduce the problem
+
+### Include any warning/errors/backtraces from the system logs
+<!--
+*IMPORTANT* - Please mark logs and text output from terminal commands
+or else Github will not display them correctly.
+An example is provided below.
+
+Example:
+```
+this is an example how log text should be marked (wrap it with ```)
+```
+-->
+

.github/ISSUE_TEMPLATE/config.yml

@@ -0,0 +1,14 @@
+blank_issues_enabled: false
+contact_links:
+  - name: OpenZFS Questions
+    url: https://github.com/openzfs/zfs/discussions/new
+    about: Ask the community for help
+  - name: OpenZFS Community Support Mailing list (Linux)
+    url: https://zfsonlinux.topicbox.com/groups/zfs-discuss
+    about: Get community support for OpenZFS on Linux
+  - name: FreeBSD Community Support Mailing list
+    url: https://lists.freebsd.org/mailman/listinfo/freebsd-fs
+    about: Get community support for OpenZFS on FreeBSD
+  - name: OpenZFS on IRC
+    url: https://web.libera.chat/#openzfs
+    about: Use IRC to get community support for OpenZFS

.github/ISSUE_TEMPLATE/feature_request.md

@@ -0,0 +1,33 @@
+---
+name: Feature request
+about: Suggest a feature for OpenZFS
+title: ''
+labels: 'Type: Feature'
+assignees: ''
+
+---
+
+<!--
+Thank you for suggesting a feature.
+
+Please check our issue tracker before opening a new feature request.
+Filling out the following template will help other contributors better understand your proposed feature.
+-->
+
+### Describe the feature would like to see added to OpenZFS
+
+<!--
+Provide a clear and concise description of the feature.
+-->
+
+### How will this feature improve OpenZFS?
+
+<!--
+What problem does this feature solve?
+-->
+
+### Additional context
+
+<!--
+Any additional information you can add about the proposal?
+-->

.github/PULL_REQUEST_TEMPLATE.md

@@ -4,7 +4,7 @@
 
 <!---
 Documentation on ZFS Buildbot options can be found at
-https://github.com/zfsonlinux/zfs/wiki/Buildbot-Options
+https://openzfs.github.io/openzfs-docs/Developer%20Resources/Buildbot%20Options.html
 -->
 
 ### Motivation and Context
@@ -19,6 +19,7 @@ https://github.com/zfsonlinux/zfs/wiki/Buildbot-Options
 <!--- Include details of your testing environment, and the tests you ran to -->
 <!--- see how your change affects other areas of the code, etc. -->
 <!--- If your change is a performance enhancement, please provide benchmarks here. -->
+<!--- Please think about using the draft PR feature if appropriate -->
 
 ### Types of changes
 <!--- What types of changes does your code introduce? Put an `x` in all the boxes that apply: -->
@@ -27,14 +28,15 @@ https://github.com/zfsonlinux/zfs/wiki/Buildbot-Options
 - [ ] Performance enhancement (non-breaking change which improves efficiency)
 - [ ] Code cleanup (non-breaking change which makes code smaller or more readable)
 - [ ] Breaking change (fix or feature that would cause existing functionality to change)
+- [ ] Library ABI change (libzfs, libzfs\_core, libnvpair, libuutil and libzfsbootenv)
 - [ ] Documentation (a change to man pages or other documentation)
 
 ### Checklist:
 <!--- Go over all the following points, and put an `x` in all the boxes that apply. -->
 <!--- If you're unsure about any of these, don't hesitate to ask. We're here to help! -->
-- [ ] My code follows the ZFS on Linux [code style requirements](https://github.com/zfsonlinux/zfs/blob/master/.github/CONTRIBUTING.md#coding-conventions).
+- [ ] My code follows the OpenZFS [code style requirements](https://github.com/openzfs/zfs/blob/master/.github/CONTRIBUTING.md#coding-conventions).
 - [ ] I have updated the documentation accordingly.
-- [ ] I have read the [**contributing** document](https://github.com/zfsonlinux/zfs/blob/master/.github/CONTRIBUTING.md).
+- [ ] I have read the [**contributing** document](https://github.com/openzfs/zfs/blob/master/.github/CONTRIBUTING.md).
-- [ ] I have added [tests](https://github.com/zfsonlinux/zfs/tree/master/tests) to cover my changes.
+- [ ] I have added [tests](https://github.com/openzfs/zfs/tree/master/tests) to cover my changes.
-- [ ] All new and existing tests passed.
+- [ ] I have run the ZFS Test Suite with this change applied.
-- [ ] All commit messages are properly formatted and contain [`Signed-off-by`](https://github.com/zfsonlinux/zfs/blob/master/.github/CONTRIBUTING.md#signed-off-by).
+- [ ] All commit messages are properly formatted and contain [`Signed-off-by`](https://github.com/openzfs/zfs/blob/master/.github/CONTRIBUTING.md#signed-off-by).

.github/codecov.yml

@@ -4,7 +4,8 @@ codecov:
     after_n_builds: 2		# user and kernel
 
 coverage:
-  precision: 2			# 2 digits of precision
+  precision: 0			# 0 decimals of precision
+  round: nearest        # Round to nearest precision point
   range: "50...90"		# red -> yellow -> green
 
   status:
@@ -20,3 +21,5 @@ comment:
   layout: "reach, diff, flags, footer"
   behavior: once		# update if exists; post new; skip if deleted
   require_changes: yes		# only post when coverage changes
+
+# ignore: Please place any ignores in config/ax_code_coverage.m4 instead

.github/no-response.yml

@@ -0,0 +1,13 @@
+# Configuration for probot-no-response - https://github.com/probot/no-response
+
+# Number of days of inactivity before an Issue is closed for lack of response
+daysUntilClose: 31
+# Label requiring a response
+responseRequiredLabel: "Status: Feedback requested"
+# Comment to post when closing an Issue for lack of response. Set to `false` to disable
+closeComment: >
+  This issue has been automatically closed because there has been no response
+  to our request for more information from the original author. With only the
+  information that is currently in the issue, we don't have enough information
+  to take action. Please reach out if you have or find the answers we need so
+  that we can investigate further.

.github/stale.yml

@@ -0,0 +1,26 @@
+# Number of days of inactivity before an issue becomes stale
+daysUntilStale: 365
+# Number of days of inactivity before a stale issue is closed
+daysUntilClose: 90
+# Limit to only `issues` or `pulls`
+only: issues
+# Issues with these labels will never be considered stale
+exemptLabels:
+  - "Type: Feature"
+  - "Bot: Not Stale"
+  - "Status: Work in Progress"
+# Set to true to ignore issues in a project (defaults to false)
+exemptProjects: true
+# Set to true to ignore issues in a milestone (defaults to false)
+exemptMilestones: true
+# Set to true to ignore issues with an assignee (defaults to false)
+exemptAssignees: true
+# Label to use when marking an issue as stale
+staleLabel: "Status: Stale"
+# Comment to post when marking an issue as stale. Set to `false` to disable
+markComment: >
+  This issue has been automatically marked as "stale" because it has not had
+  any activity for a while. It will be closed in 90 days if no further activity occurs. 
+  Thank you for your contributions.
+# Limit the number of actions per hour, from 1-30. Default is 30
+limitPerRun: 6

.github/suppressions.txt

@@ -1,3 +0,0 @@
-preprocessorErrorDirective:./module/zfs/vdev_raidz_math_avx512f.c:243
-preprocessorErrorDirective:./module/zfs/vdev_raidz_math_sse2.c:266
-

.github/workflows/checkstyle.yaml

@@ -0,0 +1,50 @@
+name: checkstyle
+
+on:
+  push:
+  pull_request:
+
+jobs:
+  checkstyle:
+    runs-on: ubuntu-20.04
+    steps:
+    - uses: actions/checkout@v2
+      with:
+        ref: ${{ github.event.pull_request.head.sha }}
+    - name: Install dependencies
+      run: |
+        sudo apt-get update
+        sudo apt-get install --yes -qq build-essential autoconf libtool gawk alien fakeroot linux-headers-$(uname -r)
+        sudo apt-get install --yes -qq zlib1g-dev uuid-dev libattr1-dev libblkid-dev libselinux-dev libudev-dev libssl-dev python-dev python-setuptools python-cffi python3 python3-dev python3-setuptools python3-cffi
+        # packages for tests
+        sudo apt-get install --yes -qq parted lsscsi ksh attr acl nfs-kernel-server fio
+        sudo apt-get install --yes -qq mandoc cppcheck pax-utils devscripts
+        sudo -E pip --quiet install flake8
+    - name: Prepare
+      run: |
+        sh ./autogen.sh
+        ./configure
+        make -j$(nproc)
+    - name: Checkstyle
+      run: |
+        make checkstyle
+    - name: Lint
+      run: |
+        make lint
+    - name: CheckABI
+      id: CheckABI
+      run: |
+        sudo docker run -v $(pwd):/source ghcr.io/openzfs/libabigail make checkabi
+    - name: StoreABI
+      if: failure() && steps.CheckABI.outcome == 'failure'
+      run: |
+        sudo docker run -v $(pwd):/source ghcr.io/openzfs/libabigail make storeabi
+    - name: Prepare artifacts
+      if: failure() && steps.CheckABI.outcome == 'failure'
+      run: |
+        find -name *.abi | tar -cf abi_files.tar -T -
+    - uses: actions/upload-artifact@v2
+      if: failure() && steps.CheckABI.outcome == 'failure'
+      with:
+        name: New ABI files (use only if you're sure about interface changes)
+        path: abi_files.tar

.github/workflows/zfs-tests-functional.yml

@@ -0,0 +1,82 @@
+name: zfs-tests-functional
+
+on:
+  push:
+  pull_request:
+
+jobs:
+  tests-functional-ubuntu:
+    strategy:
+      fail-fast: false
+      matrix:
+        os: [18.04, 20.04]
+    runs-on: ubuntu-${{ matrix.os }}
+    steps:
+    - uses: actions/checkout@v2
+      with:
+        ref: ${{ github.event.pull_request.head.sha }}
+    - name: Install dependencies
+      run: |
+        sudo apt-get update
+        sudo apt-get install --yes -qq build-essential autoconf libtool gdb lcov \
+          git alien fakeroot wget curl bc fio acl \
+          sysstat mdadm lsscsi parted gdebi attr dbench watchdog ksh \
+          nfs-kernel-server samba rng-tools xz-utils \
+          zlib1g-dev uuid-dev libblkid-dev libselinux-dev \
+          xfslibs-dev libattr1-dev libacl1-dev libudev-dev libdevmapper-dev \
+          libssl-dev libffi-dev libaio-dev libelf-dev libmount-dev \
+          libpam0g-dev pamtester python-dev python-setuptools python-cffi \
+          python-packaging python3 python3-dev python3-setuptools python3-cffi \
+          libcurl4-openssl-dev python3-packaging
+    - name: Autogen.sh
+      run: |
+        sh autogen.sh
+    - name: Configure
+      run: |
+        ./configure --enable-debug --enable-debuginfo
+    - name: Make
+      run: |
+        make --no-print-directory -s pkg-utils pkg-kmod
+    - name: Install
+      run: |
+        sudo dpkg -i *.deb
+        # Update order of directories to search for modules, otherwise
+        #   Ubuntu will load kernel-shipped ones.
+        sudo sed -i.bak 's/updates/extra updates/' /etc/depmod.d/ubuntu.conf
+        sudo depmod
+        sudo modprobe zfs
+        # Workaround for cloud-init bug
+        # see https://github.com/openzfs/zfs/issues/12644
+        FILE=/lib/udev/rules.d/10-cloud-init-hook-hotplug.rules
+        if [ -r "${FILE}" ]; then
+          HASH=$(md5sum "${FILE}" | awk '{ print $1 }')
+          if [ "${HASH}" = "121ff0ef1936cd2ef65aec0458a35772" ]; then
+            # Just shove a zd* exclusion right above the hotplug hook...
+            sudo sed -i -e s/'LABEL="cloudinit_hook"'/'KERNEL=="zd*", GOTO="cloudinit_end"\n&'/ "${FILE}"
+            sudo udevadm control --reload-rules
+          fi
+        fi
+        # Workaround to provide additional free space for testing.
+        #   https://github.com/actions/virtual-environments/issues/2840
+        sudo rm -rf /usr/share/dotnet
+        sudo rm -rf /opt/ghc
+        sudo rm -rf "/usr/local/share/boost"
+        sudo rm -rf "$AGENT_TOOLSDIRECTORY"
+    - name: Tests
+      run: |
+        /usr/share/zfs/zfs-tests.sh -v -s 3G
+    - name: Prepare artifacts
+      if: failure()
+      run: |
+        RESULTS_PATH=$(readlink -f /var/tmp/test_results/current)
+        sudo dmesg > $RESULTS_PATH/dmesg
+        sudo cp /var/log/syslog $RESULTS_PATH/
+        sudo chmod +r $RESULTS_PATH/*
+        # Replace ':' in dir names, actions/upload-artifact doesn't support it
+        for f in $(find $RESULTS_PATH -name '*:*'); do mv "$f" "${f//:/__}"; done
+    - uses: actions/upload-artifact@v2
+      if: failure()
+      with:
+        name: Test logs Ubuntu-${{ matrix.os }}
+        path: /var/tmp/test_results/20*/
+        if-no-files-found: ignore

.github/workflows/zfs-tests-sanity.yml

@@ -0,0 +1,78 @@
+name: zfs-tests-sanity
+
+on:
+  push:
+  pull_request:
+
+jobs:
+  tests:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v2
+      with:
+        ref: ${{ github.event.pull_request.head.sha }}
+    - name: Install dependencies
+      run: |
+        sudo apt-get update
+        sudo apt-get install --yes -qq build-essential autoconf libtool gdb lcov \
+          git alien fakeroot wget curl bc fio acl \
+          sysstat mdadm lsscsi parted gdebi attr dbench watchdog ksh \
+          nfs-kernel-server samba rng-tools xz-utils \
+          zlib1g-dev uuid-dev libblkid-dev libselinux-dev \
+          xfslibs-dev libattr1-dev libacl1-dev libudev-dev libdevmapper-dev \
+          libssl-dev libffi-dev libaio-dev libelf-dev libmount-dev \
+          libpam0g-dev pamtester python-dev python-setuptools python-cffi \
+          python-packaging python3 python3-dev python3-setuptools python3-cffi \
+          python3-packaging libcurl4-openssl-dev
+    - name: Autogen.sh
+      run: |
+        sh autogen.sh
+    - name: Configure
+      run: |
+        ./configure --enable-debug --enable-debuginfo
+    - name: Make
+      run: |
+        make --no-print-directory -s pkg-utils pkg-kmod
+    - name: Install
+      run: |
+        sudo dpkg -i *.deb
+        # Update order of directories to search for modules, otherwise
+        #   Ubuntu will load kernel-shipped ones.
+        sudo sed -i.bak 's/updates/extra updates/' /etc/depmod.d/ubuntu.conf
+        sudo depmod
+        sudo modprobe zfs
+        # Workaround for cloud-init bug
+        # see https://github.com/openzfs/zfs/issues/12644
+        FILE=/lib/udev/rules.d/10-cloud-init-hook-hotplug.rules
+        if [ -r "${FILE}" ]; then
+          HASH=$(md5sum "${FILE}" | awk '{ print $1 }')
+          if [ "${HASH}" = "121ff0ef1936cd2ef65aec0458a35772" ]; then
+            # Just shove a zd* exclusion right above the hotplug hook...
+            sudo sed -i -e s/'LABEL="cloudinit_hook"'/'KERNEL=="zd*", GOTO="cloudinit_end"\n&'/ "${FILE}"
+            sudo udevadm control --reload-rules
+          fi
+        fi
+        # Workaround to provide additional free space for testing.
+        #   https://github.com/actions/virtual-environments/issues/2840
+        sudo rm -rf /usr/share/dotnet
+        sudo rm -rf /opt/ghc
+        sudo rm -rf "/usr/local/share/boost"
+        sudo rm -rf "$AGENT_TOOLSDIRECTORY"
+    - name: Tests
+      run: |
+        /usr/share/zfs/zfs-tests.sh -v -s 3G -r sanity
+    - name: Prepare artifacts
+      if: failure()
+      run: |
+        RESULTS_PATH=$(readlink -f /var/tmp/test_results/current)
+        sudo dmesg > $RESULTS_PATH/dmesg
+        sudo cp /var/log/syslog $RESULTS_PATH/
+        sudo chmod +r $RESULTS_PATH/*
+        # Replace ':' in dir names, actions/upload-artifact doesn't support it
+        for f in $(find $RESULTS_PATH -name '*:*'); do mv "$f" "${f//:/__}"; done
+    - uses: actions/upload-artifact@v2
+      if: failure()
+      with:
+        name: Test logs
+        path: /var/tmp/test_results/20*/
+        if-no-files-found: ignore

.github/workflows/zloop.yml

@@ -0,0 +1,67 @@
+name: zloop
+
+on:
+  push:
+  pull_request:
+
+jobs:
+  tests:
+    runs-on: ubuntu-latest
+    env:
+      TEST_DIR: /var/tmp/zloop
+    steps:
+    - uses: actions/checkout@v2
+      with:
+        ref: ${{ github.event.pull_request.head.sha }}
+    - name: Install dependencies
+      run: |
+        sudo apt-get update
+        sudo apt-get install --yes -qq build-essential autoconf libtool gdb \
+          git alien fakeroot \
+          zlib1g-dev uuid-dev libblkid-dev libselinux-dev \
+          xfslibs-dev libattr1-dev libacl1-dev libudev-dev libdevmapper-dev \
+          libssl-dev libffi-dev libaio-dev libelf-dev libmount-dev \
+          libpam0g-dev \
+          python-dev python-setuptools python-cffi python-packaging \
+          python3 python3-dev python3-setuptools python3-cffi python3-packaging
+    - name: Autogen.sh
+      run: |
+        sh autogen.sh
+    - name: Configure
+      run: |
+        ./configure --enable-debug --enable-debuginfo
+    - name: Make
+      run: |
+        make --no-print-directory -s pkg-utils pkg-kmod
+    - name: Install
+      run: |
+        sudo dpkg -i *.deb
+        # Update order of directories to search for modules, otherwise
+        #   Ubuntu will load kernel-shipped ones.
+        sudo sed -i.bak 's/updates/extra updates/' /etc/depmod.d/ubuntu.conf
+        sudo depmod
+        sudo modprobe zfs
+    - name: Tests
+      run: |
+        sudo mkdir -p $TEST_DIR
+        # run for 20 minutes to have a total runner time of 30 minutes
+        sudo /usr/share/zfs/zloop.sh -t 1200 -l -m1 -- -T 120 -P 60
+    - name: Prepare artifacts
+      if: failure()
+      run: |
+        sudo chmod +r -R $TEST_DIR/
+    - uses: actions/upload-artifact@v2
+      if: failure()
+      with:
+        name: Logs
+        path: |
+          /var/tmp/zloop/*/
+          !/var/tmp/zloop/*/vdev/
+        if-no-files-found: ignore
+    - uses: actions/upload-artifact@v2
+      if: failure()
+      with:
+        name: Pool files
+        path: |
+          /var/tmp/zloop/*/vdev/
+        if-no-files-found: ignore

.gitignore

@@ -36,6 +36,7 @@ Makefile.in
 # Top level generated files specific to this top level dir
 #
 /bin
+/build
 /configure
 /config.log
 /config.status
@@ -61,5 +62,9 @@ cscope.*
 *.patch
 *.orig
 *.log
+*.tmp
 venv
 
+*.so
+*.so.debug
+*.so.full

.gitmodules

@@ -1,3 +1,3 @@
 [submodule "scripts/zfs-images"]
 	path = scripts/zfs-images
-	url = https://github.com/zfsonlinux/zfs-images
+	url = https://github.com/openzfs/zfs-images

.travis.yml

@@ -1,38 +0,0 @@
-language: c
-sudo: required
-env:
-  global:
-    # Travis limits maximum log size, we have to cut tests output 
-    - ZFS_TEST_TRAVIS_LOG_MAX_LENGTH=800
-  matrix:
-    # tags are mainly in ascending order
-    - ZFS_TEST_TAGS='acl,atime,bootfs,cachefile,casenorm,chattr,checksum,clean_mirror,compression,ctime,delegate,devices,events,exec,fault,features,grow_pool,zdb,zfs,zfs_bookmark,zfs_change-key,zfs_clone,zfs_copies,zfs_create,zfs_diff,zfs_get,zfs_inherit,zfs_load-key,zfs_rename'
-    - ZFS_TEST_TAGS='cache,history,hkdf,inuse,zfs_property,zfs_receive,zfs_reservation,zfs_send,zfs_set,zfs_share,zfs_snapshot,zfs_unload-key,zfs_unmount,zfs_unshare,zfs_upgrade,zpool,zpool_add,zpool_attach,zpool_clear,zpool_create,zpool_destroy,zpool_detach'
-    - ZFS_TEST_TAGS='grow_replicas,mv_files,cli_user,zfs_mount,zfs_promote,zfs_rollback,zpool_events,zpool_expand,zpool_export,zpool_get,zpool_history,zpool_import,zpool_labelclear,zpool_offline,zpool_online,zpool_remove,zpool_reopen,zpool_replace,zpool_scrub,zpool_set,zpool_status,zpool_sync,zpool_upgrade'
-    - ZFS_TEST_TAGS='zfs_destroy,large_files,largest_pool,link_count,migration,mmap,mmp,mount,nestedfs,no_space,nopwrite,online_offline,pool_names,poolversion,privilege,quota,raidz,redundancy,rsend'
-    - ZFS_TEST_TAGS='inheritance,refquota,refreserv,rename_dirs,replacement,reservation,rootpool,scrub_mirror,slog,snapshot,snapused,sparse,threadsappend,tmpfile,truncate,upgrade,userquota,vdev_zaps,write_dirs,xattr,zvol,libzfs'
-before_install:
-    - sudo apt-get -qq update
-    - sudo apt-get install --yes -qq build-essential autoconf libtool gawk alien fakeroot linux-headers-$(uname -r)
-    - sudo apt-get install --yes -qq zlib1g-dev uuid-dev libattr1-dev libblkid-dev libselinux-dev libudev-dev libssl-dev
-    # packages for tests
-    - sudo apt-get install --yes -qq parted lsscsi ksh attr acl nfs-kernel-server fio
-install:
-    - git clone --depth=1 https://github.com/zfsonlinux/spl
-    - cd spl
-    - git checkout master
-    - sh autogen.sh
-    - ./configure
-    - make --no-print-directory -s pkg-utils pkg-kmod
-    - sudo dpkg -i *.deb
-    - cd ..
-    - sh autogen.sh
-    - ./configure
-    - make --no-print-directory -s pkg-utils pkg-kmod
-    - sudo dpkg -i *.deb
-script:
-    - travis_wait 50 /usr/share/zfs/zfs-tests.sh -v -T $ZFS_TEST_TAGS
-after_failure:
-    - find /var/tmp/test_results/current/log -type f -name '*' -printf "%f\n" -exec cut -c -$ZFS_TEST_TRAVIS_LOG_MAX_LENGTH {} \;
-after_success:
-    - find /var/tmp/test_results/current/log -type f -name '*' -printf "%f\n" -exec cut -c -$ZFS_TEST_TRAVIS_LOG_MAX_LENGTH {} \;

AUTHORS

@@ -83,6 +83,7 @@ CONTRIBUTORS:
     Christopher Voltz <cjunk@voltz.ws>
     Chunwei Chen <david.chen@nutanix.com>
     Clemens Fruhwirth <clemens@endorphin.org>
+    Coleman Kane <ckane@colemankane.org>
     Colin Ian King <colin.king@canonical.com>
     Craig Loomis <cloomis@astro.princeton.edu>
     Craig Sanders <github@taz.net.au>
@@ -181,6 +182,7 @@ CONTRIBUTORS:
     Keith M Wesolowski <wesolows@foobazco.org>
     Kevin Tanguy <kevin.tanguy@ovh.net>
     KireinaHoro <i@jsteward.moe>
+    Kjeld Schouten-Lebbing <kjeld@schouten-lebbing.nl>
     Kohsuke Kawaguchi <kk@kohsuke.org>
     Kyle Blatter <kyleblatter@llnl.gov>
     Kyle Fuller <inbox@kylefuller.co.uk>
@@ -209,6 +211,7 @@ CONTRIBUTORS:
     Michael Gebetsroither <michael@mgeb.org>
     Michael Kjorling <michael@kjorling.se>
     Michael Martin <mgmartin.mgm@gmail.com>
+    Michael Niewöhner <foss@mniewoehner.de>
     Mike Gerdts <mike.gerdts@joyent.com>
     Mike Harsch <mike@harschsystems.com>
     Mike Leddy <mike.leddy@gmail.com>
@@ -257,6 +260,7 @@ CONTRIBUTORS:
     Saso Kiselkov <saso.kiselkov@nexenta.com>
     Scot W. Stevenson <scot.stevenson@gmail.com>
     Sean Eric Fagan <sef@ixsystems.com>
+    Sebastian Gottschall <s.gottschall@dd-wrt.com>
     Sen Haerens <sen@senhaerens.be>
     Serapheim Dimitropoulos <serapheim@delphix.com>
     Seth Forshee <seth.forshee@canonical.com>

CODE_OF_CONDUCT.md

@@ -1,2 +1,2 @@
 The [OpenZFS Code of Conduct](http://www.open-zfs.org/wiki/Code_of_Conduct)
-applies to spaces associated with the ZFS on Linux project, including GitHub.
+applies to spaces associated with the OpenZFS project, including GitHub.

COPYRIGHT

@@ -19,7 +19,11 @@ notable exceptions and their respective licenses include:
   * AES Implementation: module/icp/asm-x86_64/aes/THIRDPARTYLICENSE.gladman
   * AES Implementation: module/icp/asm-x86_64/aes/THIRDPARTYLICENSE.openssl
   * PBKDF2 Implementation: lib/libzfs/THIRDPARTYLICENSE.openssl
-  * SPL Implementation: module/spl/THIRDPARTYLICENSE.gplv2
+  * SPL Implementation: module/os/linux/spl/THIRDPARTYLICENSE.gplv2
+  * GCM Implementation: module/icp/asm-x86_64/modes/THIRDPARTYLICENSE.cryptogams
+  * GCM Implementation: module/icp/asm-x86_64/modes/THIRDPARTYLICENSE.openssl
+  * GHASH Implementation: module/icp/asm-x86_64/modes/THIRDPARTYLICENSE.cryptogams
+  * GHASH Implementation: module/icp/asm-x86_64/modes/THIRDPARTYLICENSE.openssl
 
 This product includes software developed by the OpenSSL Project for use
 in the OpenSSL Toolkit (http://www.openssl.org/)

META

@@ -1,10 +1,10 @@
 Meta:          1
 Name:          zfs
 Branch:        1.0
-Version:       0.8.0
+Version:       2.1.99
 Release:       1
 Release-Tags:  relext
 License:       CDDL
-Author:        OpenZFS on Linux
+Author:        OpenZFS
-Linux-Maximum: 5.1
+Linux-Maximum: 5.14
-Linux-Minimum: 2.6.32
+Linux-Minimum: 3.10

Makefile.am

@@ -1,12 +1,17 @@
+include $(top_srcdir)/config/Shellcheck.am
+
 ACLOCAL_AMFLAGS = -I config
 
-include config/rpm.am
+SUBDIRS = include
-include config/deb.am
+if BUILD_LINUX
-include config/tgz.am
+SUBDIRS += rpm
+endif
 
-SUBDIRS = include rpm
 if CONFIG_USER
-SUBDIRS += udev etc man scripts lib tests cmd contrib
+SUBDIRS += man scripts lib tests cmd etc contrib
+if BUILD_LINUX
+SUBDIRS += udev
+endif
 endif
 if CONFIG_KERNEL
 SUBDIRS += module
@@ -14,33 +19,51 @@ SUBDIRS += module
 extradir = $(prefix)/src/zfs-$(VERSION)
 extra_HEADERS = zfs.release.in zfs_config.h.in
 
+if BUILD_LINUX
 kerneldir = $(prefix)/src/zfs-$(VERSION)/$(LINUX_VERSION)
 nodist_kernel_HEADERS = zfs.release zfs_config.h module/$(LINUX_SYMBOLS)
 endif
+endif
 
 AUTOMAKE_OPTIONS = foreign
 EXTRA_DIST  = autogen.sh copy-builtin
 EXTRA_DIST += config/config.awk config/rpm.am config/deb.am config/tgz.am
-EXTRA_DIST += META AUTHORS COPYRIGHT LICENSE NEWS NOTICE README.md
+EXTRA_DIST += AUTHORS CODE_OF_CONDUCT.md COPYRIGHT LICENSE META NEWS NOTICE
-EXTRA_DIST += CODE_OF_CONDUCT.md
+EXTRA_DIST += README.md RELEASES.md
+EXTRA_DIST += module/lua/README.zfs module/os/linux/spl/README.md
 
 # Include all the extra licensing information for modules
-EXTRA_DIST += module/icp/algs/skein/THIRDPARTYLICENSE module/icp/algs/skein/THIRDPARTYLICENSE.descrip
+EXTRA_DIST += module/icp/algs/skein/THIRDPARTYLICENSE
-EXTRA_DIST += module/icp/asm-x86_64/aes/THIRDPARTYLICENSE.gladman module/icp/asm-x86_64/aes/THIRDPARTYLICENSE.gladman.descrip
+EXTRA_DIST += module/icp/algs/skein/THIRDPARTYLICENSE.descrip
-EXTRA_DIST += module/icp/asm-x86_64/aes/THIRDPARTYLICENSE.openssl module/icp/asm-x86_64/aes/THIRDPARTYLICENSE.openssl.descrip
+EXTRA_DIST += module/icp/asm-x86_64/aes/THIRDPARTYLICENSE.gladman
-EXTRA_DIST += module/spl/THIRDPARTYLICENSE.gplv2 module/spl/THIRDPARTYLICENSE.gplv2.descrip
+EXTRA_DIST += module/icp/asm-x86_64/aes/THIRDPARTYLICENSE.gladman.descrip
-EXTRA_DIST += module/zfs/THIRDPARTYLICENSE.cityhash module/zfs/THIRDPARTYLICENSE.cityhash.descrip
+EXTRA_DIST += module/icp/asm-x86_64/aes/THIRDPARTYLICENSE.openssl
+EXTRA_DIST += module/icp/asm-x86_64/aes/THIRDPARTYLICENSE.openssl.descrip
+EXTRA_DIST += module/icp/asm-x86_64/modes/THIRDPARTYLICENSE.cryptogams
+EXTRA_DIST += module/icp/asm-x86_64/modes/THIRDPARTYLICENSE.cryptogams.descrip
+EXTRA_DIST += module/icp/asm-x86_64/modes/THIRDPARTYLICENSE.openssl
+EXTRA_DIST += module/icp/asm-x86_64/modes/THIRDPARTYLICENSE.openssl.descrip
+EXTRA_DIST += module/os/linux/spl/THIRDPARTYLICENSE.gplv2
+EXTRA_DIST += module/os/linux/spl/THIRDPARTYLICENSE.gplv2.descrip
+EXTRA_DIST += module/zfs/THIRDPARTYLICENSE.cityhash
+EXTRA_DIST += module/zfs/THIRDPARTYLICENSE.cityhash.descrip
 
 @CODE_COVERAGE_RULES@
 
-.PHONY: gitrev
+GITREV = include/zfs_gitrev.h
-gitrev:
-	-${top_srcdir}/scripts/make_gitrev.sh
 
-BUILT_SOURCES = gitrev
+PHONY = gitrev
+gitrev:
+	$(AM_V_GEN)$(top_srcdir)/scripts/make_gitrev.sh $(GITREV)
+
+all: gitrev
+
+# Double-colon rules are allowed; there are multiple independent definitions.
+maintainer-clean-local::
+	-$(RM) $(GITREV)
 
 distclean-local::
-	-$(RM) -R autom4te*.cache
+	-$(RM) -R autom4te*.cache build
 	-find . \( -name SCCS -o -name BitKeeper -o -name .svn -o -name CVS \
 		-o -name .pc -o -name .hg -o -name .git \) -prune -o \
 		\( -name '*.orig' -o -name '*.rej' -o -name '*~' \
@@ -52,13 +75,15 @@ distclean-local::
 		-type f -print | xargs $(RM)
 
 all-local:
-	-${top_srcdir}/scripts/zfs-tests.sh -c
+	-[ -x ${top_builddir}/scripts/zfs-tests.sh ] && \
+	    ${top_builddir}/scripts/zfs-tests.sh -c
 
-dist-hook: gitrev
+dist-hook:
-	cp ${top_srcdir}/include/zfs_gitrev.h $(distdir)/include; \
+	$(AM_V_GEN)$(top_srcdir)/scripts/make_gitrev.sh -D $(distdir) $(GITREV)
-	sed -i 's/Release:[[:print:]]*/Release:      $(RELEASE)/' \
+	$(SED) ${ac_inplace} -e 's/Release:[[:print:]]*/Release:      $(RELEASE)/' \
 		$(distdir)/META
 
+if BUILD_LINUX
 # For compatibility, create a matching spl-x.y.z directly which contains
 # symlinks to the updated header and object file locations.  These
 # compatibility links will be removed in the next major release.
@@ -75,75 +100,102 @@ install-data-hook:
 	ln -fs zfs_config.h spl_config.h && \
 	ln -fs zfs.release spl.release
 endif
+endif
 
-codecheck: cstyle shellcheck flake8 mancheck testscheck vcscheck
+PHONY += codecheck
+codecheck: cstyle shellcheck checkbashisms flake8 mancheck testscheck vcscheck
 
+PHONY += checkstyle
 checkstyle: codecheck commitcheck
 
+PHONY += commitcheck
 commitcheck:
 	@if git rev-parse --git-dir > /dev/null 2>&1; then \
 		${top_srcdir}/scripts/commitcheck.sh; \
 	fi
 
+PHONY += cstyle
 cstyle:
-	@find ${top_srcdir} -name '*.[hc]' ! -name 'zfs_config.*' \
+	@find ${top_srcdir} -name build -prune \
-		! -name '*.mod.c' -type f \
+		-o -type f -name '*.[hc]' \
+		! -name 'zfs_config.*' ! -name '*.mod.c' \
+		! -name 'opt_global.h' ! -name '*_if*.h' \
+		! -path './module/zstd/lib/*' \
 		-exec ${top_srcdir}/scripts/cstyle.pl -cpP {} \+
 
-shellcheck:
+filter_executable = -exec test -x '{}' \; -print
-	@if type shellcheck > /dev/null 2>&1; then \
-		shellcheck --exclude=SC1090 --format=gcc \
-			$$(find ${top_srcdir}/scripts/*.sh -type f) \
-			$$(find ${top_srcdir}/cmd/zed/zed.d/*.sh -type f) \
-			$$(find ${top_srcdir}/cmd/zpool/zpool.d/* -executable); \
-	else \
-		echo "skipping shellcheck because shellcheck is not installed"; \
-	fi
 
+SHELLCHECKDIRS = cmd contrib etc scripts tests
+SHELLCHECKSCRIPTS = autogen.sh
+
+PHONY += checkabi storeabi
+
+checklibabiversion:
+	libabiversion=`abidw -v | $(SED) 's/[^0-9]//g'`; \
+	if test $$libabiversion -lt "200"; then \
+        /bin/echo -e "\n" \
+        "*** Please use libabigail 2.0.0 version or newer;\n" \
+        "*** otherwise results are not consistent!\n" \
+        "(or see https://github.com/openzfs/libabigail-docker )\n"; \
+        exit 1; \
+    fi;
+
+checkabi: checklibabiversion lib
+	$(MAKE) -C lib checkabi
+
+storeabi: checklibabiversion lib
+	$(MAKE) -C lib storeabi
+
+PHONY += mancheck
 mancheck:
-	@if type mandoc > /dev/null 2>&1; then \
+	${top_srcdir}/scripts/mancheck.sh ${top_srcdir}/man ${top_srcdir}/tests/test-runner/man
-		find ${top_srcdir}/man/man8 -type f -name 'zfs.8' \
-			-o -name 'zpool.8' -o -name 'zdb.8' \
-			-o -name 'zgenhostid.8' | \
-			xargs mandoc -Tlint -Werror; \
-	else \
-		echo "skipping mancheck because mandoc is not installed"; \
-	fi
 
+if BUILD_LINUX
+stat_fmt = -c '%A %n'
+else
+stat_fmt = -f '%Sp %N'
+endif
+
+PHONY += testscheck
 testscheck:
 	@find ${top_srcdir}/tests/zfs-tests -type f \
-		\( -name '*.ksh' -not -executable \) -o \
+		\( -name '*.ksh' -not ${filter_executable} \) -o \
-		\( -name '*.kshlib' -executable \) -o \
+		\( -name '*.kshlib' ${filter_executable} \) -o \
-		\( -name '*.shlib' -executable \) -o \
+		\( -name '*.shlib' ${filter_executable} \) -o \
-		\( -name '*.cfg' -executable \) | \
+		\( -name '*.cfg' ${filter_executable} \) | \
-		xargs -r stat -c '%A %n' | \
+		xargs -r stat ${stat_fmt} | \
 		awk '{c++; print} END {if(c>0) exit 1}'
 
+PHONY += vcscheck
 vcscheck:
 	@if git rev-parse --git-dir > /dev/null 2>&1; then \
 		git ls-files . --exclude-standard --others | \
 		awk '{c++; print} END {if(c>0) exit 1}' ; \
 	fi
 
+PHONY += lint
 lint: cppcheck paxcheck
 
-cppcheck:
+CPPCHECKDIRS = cmd lib module
-	@if type cppcheck > /dev/null 2>&1; then \
+PHONY += cppcheck
-		cppcheck --quiet --force --error-exitcode=2 --inline-suppr \
+cppcheck: $(CPPCHECKDIRS)
-			--suppressions-list=.github/suppressions.txt \
+	@if test -n "$(CPPCHECK)"; then \
-			-UHAVE_SSE2 -UHAVE_AVX512F -UHAVE_UIO_ZEROCOPY \
+		set -e ; for dir in $(CPPCHECKDIRS) ; do \
-			${top_srcdir}; \
+			$(MAKE) -C $$dir cppcheck ; \
+		done \
 	else \
 		echo "skipping cppcheck because cppcheck is not installed"; \
 	fi
 
+PHONY += paxcheck
 paxcheck:
 	@if type scanelf > /dev/null 2>&1; then \
-		${top_srcdir}/scripts/paxcheck.sh ${top_srcdir}; \
+		${top_srcdir}/scripts/paxcheck.sh ${top_builddir}; \
 	else \
 		echo "skipping paxcheck because scanelf is not installed"; \
 	fi
 
+PHONY += flake8
 flake8:
 	@if type flake8 > /dev/null 2>&1; then \
 		flake8 ${top_srcdir}; \
@@ -151,17 +203,34 @@ flake8:
 		echo "skipping flake8 because flake8 is not installed"; \
 	fi
 
+PHONY += ctags
 ctags:
 	$(RM) tags
-	find $(top_srcdir) -name .git -prune -o -name '*.[hc]' | xargs ctags
+	find $(top_srcdir) -name '.?*' -prune \
+		-o -type f -name '*.[hcS]' -print | xargs ctags -a
 
+PHONY += etags
 etags:
 	$(RM) TAGS
-	find $(top_srcdir) -name .pc -prune -o -name '*.[hc]' | xargs etags -a
+	find $(top_srcdir) -name '.?*' -prune \
+		-o -type f -name '*.[hcS]' -print | xargs etags -a
 
+PHONY += cscopelist
+cscopelist:
+	find $(top_srcdir) -name '.?*' -prune \
+		-o -type f -name '*.[hc]' -print >cscope.files
+
+PHONY += tags
 tags: ctags etags
 
+PHONY += pkg pkg-dkms pkg-kmod pkg-utils
 pkg: @DEFAULT_PACKAGE@
 pkg-dkms: @DEFAULT_PACKAGE@-dkms
 pkg-kmod: @DEFAULT_PACKAGE@-kmod
 pkg-utils: @DEFAULT_PACKAGE@-utils
+
+include config/rpm.am
+include config/deb.am
+include config/tgz.am
+
+.PHONY: $(PHONY)

NEWS

@@ -1,3 +1,3 @@
 Descriptions of all releases can be found on github:
 
-https://github.com/zfsonlinux/zfs/releases
+https://github.com/openzfs/zfs/releases

README.md

@@ -1,31 +1,35 @@
-![img](http://zfsonlinux.org/images/zfs-linux.png)
+![img](https://openzfs.github.io/openzfs-docs/_static/img/logo/480px-Open-ZFS-Secondary-Logo-Colour-halfsize.png)
 
-ZFS on Linux is an advanced file system and volume manager which was originally
+OpenZFS is an advanced file system and volume manager which was originally
 developed for Solaris and is now maintained by the OpenZFS community.
+This repository contains the code for running OpenZFS on Linux and FreeBSD.
 
-[![codecov](https://codecov.io/gh/zfsonlinux/zfs/branch/master/graph/badge.svg)](https://codecov.io/gh/zfsonlinux/zfs)
+[![codecov](https://codecov.io/gh/openzfs/zfs/branch/master/graph/badge.svg)](https://codecov.io/gh/openzfs/zfs)
-[![coverity](https://scan.coverity.com/projects/1973/badge.svg)](https://scan.coverity.com/projects/zfsonlinux-zfs)
+[![coverity](https://scan.coverity.com/projects/1973/badge.svg)](https://scan.coverity.com/projects/openzfs-zfs)
 
 # Official Resources
 
-  * [Site](http://zfsonlinux.org)
+  * [Documentation](https://openzfs.github.io/openzfs-docs/) - for using and developing this repo
-  * [Wiki](https://github.com/zfsonlinux/zfs/wiki)
+  * [ZoL Site](https://zfsonlinux.org) - Linux release info & links
-  * [Mailing lists](https://github.com/zfsonlinux/zfs/wiki/Mailing-Lists)
+  * [Mailing lists](https://openzfs.github.io/openzfs-docs/Project%20and%20Community/Mailing%20Lists.html)
-  * [OpenZFS site](http://open-zfs.org/)
+  * [OpenZFS site](http://open-zfs.org/) - for conference videos and info on other platforms (illumos, OSX, Windows, etc)
 
 # Installation
 
-Full documentation for installing ZoL on your favorite Linux distribution can
+Full documentation for installing OpenZFS on your favorite operating system can
-be found at [our site](http://zfsonlinux.org/).
+be found at the [Getting Started Page](https://openzfs.github.io/openzfs-docs/Getting%20Started/index.html).
 
 # Contribute & Develop
 
 We have a separate document with [contribution guidelines](./.github/CONTRIBUTING.md).
 
+We have a [Code of Conduct](./CODE_OF_CONDUCT.md).
+
 # Release
 
-ZFS on Linux is released under a CDDL license.  
+OpenZFS is released under a CDDL license.
 For more details see the NOTICE, LICENSE and COPYRIGHT files; `UCRL-CODE-235197`
 
 # Supported Kernels
-  * The `META` file contains the officially recognized supported kernel versions.
+  * The `META` file contains the officially recognized supported Linux kernel versions.
+  * Supported FreeBSD versions are any supported branches and releases starting from 12.2-RELEASE.

RELEASES.md

@@ -0,0 +1,37 @@
+OpenZFS uses the MAJOR.MINOR.PATCH versioning scheme described here:
+
+  * MAJOR - Incremented at the discretion of the OpenZFS developers to indicate
+    a particularly noteworthy feature or change. An increase in MAJOR number
+    does not indicate any incompatible on-disk format change. The ability
+    to import a ZFS pool is controlled by the feature flags enabled on the
+    pool and the feature flags supported by the installed OpenZFS version.
+    Increasing the MAJOR version is expected to be an infrequent occurrence.
+
+  * MINOR - Incremented to indicate new functionality such as a new feature
+    flag, pool/dataset property, zfs/zpool sub-command, new user/kernel
+    interface, etc. MINOR releases may introduce incompatible changes to the
+    user space library APIs (libzfs.so). Existing user/kernel interfaces are
+    considered to be stable to maximize compatibility between OpenZFS releases.
+    Additions to the user/kernel interface are backwards compatible.
+
+  * PATCH - Incremented when applying documentation updates, important bug
+    fixes, minor performance improvements, and kernel compatibility patches.
+    The user space library APIs and user/kernel interface are considered to
+    be stable. PATCH releases for a MAJOR.MINOR are published as needed.
+
+Two release branches are maintained for OpenZFS, they are:
+
+  * OpenZFS LTS - A designated MAJOR.MINOR release with periodic PATCH
+    releases that incorporate important changes backported from newer OpenZFS
+    releases. This branch is intended for use in environments using an
+    LTS, enterprise, or similarly managed kernel (RHEL, Ubuntu LTS, Debian).
+    Minor changes to support these distribution kernels will be applied as
+    needed. New kernel versions released after the OpenZFS LTS release are
+    not supported. LTS releases will receive patches for at least 2 years.
+    The current LTS release is OpenZFS 2.1.
+
+  * OpenZFS current - Tracks the newest MAJOR.MINOR release. This branch
+    includes support for the latest OpenZFS features and recently releases
+    kernels.  When a new MINOR release is tagged the previous MINOR release
+    will no longer be maintained (unless it is an LTS release). New MINOR
+    releases are planned to occur roughly annually.

TEST

@@ -48,64 +48,3 @@
 #TEST_ZFSSTRESS_VDEV="/var/tmp/vdev"
 #TEST_ZFSSTRESS_DIR="/$TEST_ZFSSTRESS_POOL/$TEST_ZFSSTRESS_FS"
 #TEST_ZFSSTRESS_OPTIONS=""
-
-### per-builder customization
-#
-# BB_NAME=builder-name <distribution-version-architecture-type>
-# - distribution=Amazon,Debian,Fedora,RHEL,SUSE,Ubuntu
-# - version=x.y
-# - architecture=x86_64,i686,arm,aarch64
-# - type=build,test
-#
-case "$BB_NAME" in
-Amazon*)
-    # ZFS enabled xfstests fails to build
-    TEST_XFSTESTS_SKIP="yes"
-    ;;
-CentOS-7*)
-    # ZFS enabled xfstests fails to build
-    TEST_XFSTESTS_SKIP="yes"
-    ;;
-CentOS-6*)
-    ;;
-Debian*)
-    ;;
-Fedora*)
-    ;;
-RHEL*)
-    ;;
-SUSE*)
-    ;;
-Ubuntu-16.04*)
-    # ZFS enabled xfstests fails to build
-    TEST_XFSTESTS_SKIP="yes"
-    ;;
-Ubuntu*)
-    ;;
-*)
-    ;;
-esac
-
-###
-#
-# Run ztest longer on the "coverage" builders to gain more code coverage
-# data out of ztest, libzpool, etc.
-#
-case "$BB_NAME" in
-*coverage*)
-    TEST_ZTEST_TIMEOUT=3600
-    ;;
-*)
-    TEST_ZTEST_TIMEOUT=900
-    ;;
-esac
-
-###
-#
-# Disable the following test suites on 32-bit systems.
-#
-if [ $(getconf LONG_BIT) = "32" ]; then
-    TEST_ZTEST_SKIP="yes"
-    TEST_XFSTESTS_SKIP="yes"
-    TEST_ZFSSTRESS_SKIP="yes"
-fi

cmd/Makefile.am

@@ -1,3 +1,27 @@
-SUBDIRS  = zfs zpool zdb zhack zinject zstreamdump ztest
+include $(top_srcdir)/config/Shellcheck.am
-SUBDIRS += mount_zfs fsck_zfs zvol_id vdev_id arcstat dbufstat zed
+
-SUBDIRS += arc_summary raidz_test zgenhostid
+SUBDIRS  = zfs zpool zdb zhack zinject zstream ztest
+SUBDIRS += fsck_zfs vdev_id raidz_test zfs_ids_to_path
+SUBDIRS += zpool_influxdb
+
+CPPCHECKDIRS  = zfs zpool zdb zhack zinject zstream ztest
+CPPCHECKDIRS += raidz_test zfs_ids_to_path zpool_influxdb
+
+# TODO: #12084: SHELLCHECKDIRS = fsck_zfs vdev_id zpool
+SHELLCHECKDIRS = fsck_zfs zpool
+
+if USING_PYTHON
+SUBDIRS += arcstat arc_summary dbufstat
+endif
+
+if BUILD_LINUX
+SUBDIRS += mount_zfs zed zgenhostid zvol_id zvol_wait
+CPPCHECKDIRS += mount_zfs zed zgenhostid zvol_id
+SHELLCHECKDIRS += zed
+endif
+
+PHONY = cppcheck
+cppcheck: $(CPPCHECKDIRS)
+	set -e ; for dir in $(CPPCHECKDIRS) ; do \
+		$(MAKE) -C $$dir cppcheck ; \
+	done

cmd/arc_summary/.gitignore

@@ -0,0 +1 @@
+arc_summary

cmd/arc_summary/Makefile.am

@@ -1,13 +1,13 @@
+bin_SCRIPTS = arc_summary
+
+CLEANFILES = arc_summary
 EXTRA_DIST = arc_summary2 arc_summary3
 
 if USING_PYTHON_2
-dist_bin_SCRIPTS = arc_summary2
+SCRIPT = arc_summary2
-install-exec-hook:
+else
-	mv $(DESTDIR)$(bindir)/arc_summary2 $(DESTDIR)$(bindir)/arc_summary
+SCRIPT = arc_summary3
 endif
 
-if USING_PYTHON_3
+arc_summary: $(SCRIPT)
-dist_bin_SCRIPTS = arc_summary3
+	cp $< $@
-install-exec-hook:
-	mv $(DESTDIR)$(bindir)/arc_summary3 $(DESTDIR)$(bindir)/arc_summary
-endif

cmd/arc_summary/arc_summary2

@@ -1,4 +1,4 @@
-#!/usr/bin/python2
+#!/usr/bin/env python2
 #
 # $Id: arc_summary.pl,v 388:e27800740aa2 2011-07-08 02:53:29Z jhell $
 #
@@ -42,7 +42,7 @@
 Provides basic information on the ARC, its efficiency, the L2ARC (if present),
 the Data Management Unit (DMU), Virtual Devices (VDEVs), and tunables. See the
 in-source documentation and code at
-https://github.com/zfsonlinux/zfs/blob/master/module/zfs/arc.c for details.
+https://github.com/openzfs/zfs/blob/master/module/zfs/arc.c for details.
 """
 
 import getopt
@@ -54,46 +54,64 @@ import errno
 from subprocess import Popen, PIPE
 from decimal import Decimal as D
 
+
+if sys.platform.startswith('freebsd'):
+    # Requires py27-sysctl on FreeBSD
+    import sysctl
+
+    def is_value(ctl):
+        return ctl.type != sysctl.CTLTYPE_NODE
+
+    def load_kstats(namespace):
+        """Collect information on a specific subsystem of the ARC"""
+
+        base = 'kstat.zfs.misc.%s.' % namespace
+        fmt = lambda kstat: (kstat.name, D(kstat.value))
+        kstats = sysctl.filter(base)
+        return [fmt(kstat) for kstat in kstats if is_value(kstat)]
+
+    def load_tunables():
+        ctls = sysctl.filter('vfs.zfs')
+        return dict((ctl.name, ctl.value) for ctl in ctls if is_value(ctl))
+
+elif sys.platform.startswith('linux'):
+
+    def load_kstats(namespace):
+        """Collect information on a specific subsystem of the ARC"""
+
+        kstat = 'kstat.zfs.misc.%s.%%s' % namespace
+        path = '/proc/spl/kstat/zfs/%s' % namespace
+        with open(path) as f:
+            entries = [line.strip().split() for line in f][2:] # Skip header
+        return [(kstat % name, D(value)) for name, _, value in entries]
+
+    def load_tunables():
+        basepath = '/sys/module/zfs/parameters'
+        tunables = {}
+        for name in os.listdir(basepath):
+            if not name:
+                continue
+            path = '%s/%s' % (basepath, name)
+            with open(path) as f:
+                value = f.read()
+            tunables[name] = value.strip()
+        return tunables
+
+
 show_tunable_descriptions = False
 alternate_tunable_layout = False
 
 
-def handle_Exception(ex_cls, ex, tb):
-    if ex is IOError:
-        if ex.errno == errno.EPIPE:
-            sys.exit()
-
-    if ex is KeyboardInterrupt:
-        sys.exit()
-
-
-sys.excepthook = handle_Exception
-
-
 def get_Kstat():
     """Collect information on the ZFS subsystem from the /proc virtual
     file system. The name "kstat" is a holdover from the Solaris utility
     of the same name.
     """
 
-    def load_proc_kstats(fn, namespace):
-        """Collect information on a specific subsystem of the ARC"""
-
-        kstats = [line.strip() for line in open(fn)]
-        del kstats[0:2]
-        for kstat in kstats:
-            kstat = kstat.strip()
-            name, _, value = kstat.split()
-            Kstat[namespace + name] = D(value)
-
     Kstat = {}
-    load_proc_kstats('/proc/spl/kstat/zfs/arcstats',
+    Kstat.update(load_kstats('arcstats'))
-                     'kstat.zfs.misc.arcstats.')
+    Kstat.update(load_kstats('zfetchstats'))
-    load_proc_kstats('/proc/spl/kstat/zfs/zfetchstats',
+    Kstat.update(load_kstats('vdev_cache_stats'))
-                     'kstat.zfs.misc.zfetchstats.')
-    load_proc_kstats('/proc/spl/kstat/zfs/vdev_cache_stats',
-                     'kstat.zfs.misc.vdev_cache_stats.')
-
     return Kstat
 
 
@@ -195,12 +213,30 @@ def get_arc_summary(Kstat):
     deleted = Kstat["kstat.zfs.misc.arcstats.deleted"]
     mutex_miss = Kstat["kstat.zfs.misc.arcstats.mutex_miss"]
     evict_skip = Kstat["kstat.zfs.misc.arcstats.evict_skip"]
+    evict_l2_cached = Kstat["kstat.zfs.misc.arcstats.evict_l2_cached"]
+    evict_l2_eligible = Kstat["kstat.zfs.misc.arcstats.evict_l2_eligible"]
+    evict_l2_eligible_mfu = Kstat["kstat.zfs.misc.arcstats.evict_l2_eligible_mfu"]
+    evict_l2_eligible_mru = Kstat["kstat.zfs.misc.arcstats.evict_l2_eligible_mru"]
+    evict_l2_ineligible = Kstat["kstat.zfs.misc.arcstats.evict_l2_ineligible"]
+    evict_l2_skip = Kstat["kstat.zfs.misc.arcstats.evict_l2_skip"]
 
     # ARC Misc.
     output["arc_misc"] = {}
     output["arc_misc"]["deleted"] = fHits(deleted)
-    output["arc_misc"]['mutex_miss'] = fHits(mutex_miss)
+    output["arc_misc"]["mutex_miss"] = fHits(mutex_miss)
-    output["arc_misc"]['evict_skips'] = fHits(evict_skip)
+    output["arc_misc"]["evict_skips"] = fHits(evict_skip)
+    output["arc_misc"]["evict_l2_skip"] = fHits(evict_l2_skip)
+    output["arc_misc"]["evict_l2_cached"] = fBytes(evict_l2_cached)
+    output["arc_misc"]["evict_l2_eligible"] = fBytes(evict_l2_eligible)
+    output["arc_misc"]["evict_l2_eligible_mfu"] = {
+            'per': fPerc(evict_l2_eligible_mfu, evict_l2_eligible),
+            'num': fBytes(evict_l2_eligible_mfu),
+    }
+    output["arc_misc"]["evict_l2_eligible_mru"] = {
+            'per': fPerc(evict_l2_eligible_mru, evict_l2_eligible),
+            'num': fBytes(evict_l2_eligible_mru),
+    }
+    output["arc_misc"]["evict_l2_ineligible"] = fBytes(evict_l2_ineligible)
 
     # ARC Sizing
     arc_size = Kstat["kstat.zfs.misc.arcstats.size"]
@@ -316,8 +352,26 @@ def _arc_summary(Kstat):
     sys.stdout.write("\tDeleted:\t\t\t\t%s\n" % arc['arc_misc']['deleted'])
     sys.stdout.write("\tMutex Misses:\t\t\t\t%s\n" %
                      arc['arc_misc']['mutex_miss'])
-    sys.stdout.write("\tEvict Skips:\t\t\t\t%s\n" %
+    sys.stdout.write("\tEviction Skips:\t\t\t\t%s\n" %
                      arc['arc_misc']['evict_skips'])
+    sys.stdout.write("\tEviction Skips Due to L2 Writes:\t%s\n" %
+                     arc['arc_misc']['evict_l2_skip'])
+    sys.stdout.write("\tL2 Cached Evictions:\t\t\t%s\n" %
+                     arc['arc_misc']['evict_l2_cached'])
+    sys.stdout.write("\tL2 Eligible Evictions:\t\t\t%s\n" %
+                     arc['arc_misc']['evict_l2_eligible'])
+    sys.stdout.write("\tL2 Eligible MFU Evictions:\t%s\t%s\n" % (
+                     arc['arc_misc']['evict_l2_eligible_mfu']['per'],
+                     arc['arc_misc']['evict_l2_eligible_mfu']['num'],
+                     )
+    )
+    sys.stdout.write("\tL2 Eligible MRU Evictions:\t%s\t%s\n" % (
+                     arc['arc_misc']['evict_l2_eligible_mru']['per'],
+                     arc['arc_misc']['evict_l2_eligible_mru']['num'],
+                     )
+    )
+    sys.stdout.write("\tL2 Ineligible Evictions:\t\t%s\n" %
+                     arc['arc_misc']['evict_l2_ineligible'])
     sys.stdout.write("\n")
 
     # ARC Sizing
@@ -653,6 +707,11 @@ def get_l2arc_summary(Kstat):
     l2_writes_done = Kstat["kstat.zfs.misc.arcstats.l2_writes_done"]
     l2_writes_error = Kstat["kstat.zfs.misc.arcstats.l2_writes_error"]
     l2_writes_sent = Kstat["kstat.zfs.misc.arcstats.l2_writes_sent"]
+    l2_mfu_asize = Kstat["kstat.zfs.misc.arcstats.l2_mfu_asize"]
+    l2_mru_asize = Kstat["kstat.zfs.misc.arcstats.l2_mru_asize"]
+    l2_prefetch_asize = Kstat["kstat.zfs.misc.arcstats.l2_prefetch_asize"]
+    l2_bufc_data_asize = Kstat["kstat.zfs.misc.arcstats.l2_bufc_data_asize"]
+    l2_bufc_metadata_asize = Kstat["kstat.zfs.misc.arcstats.l2_bufc_metadata_asize"]
 
     l2_access_total = (l2_hits + l2_misses)
     output['l2_health_count'] = (l2_writes_error + l2_cksum_bad + l2_io_error)
@@ -675,7 +734,7 @@ def get_l2arc_summary(Kstat):
         output["io_errors"] = fHits(l2_io_error)
 
         output["l2_arc_size"] = {}
-        output["l2_arc_size"]["adative"] = fBytes(l2_size)
+        output["l2_arc_size"]["adaptive"] = fBytes(l2_size)
         output["l2_arc_size"]["actual"] = {
             'per': fPerc(l2_asize, l2_size),
             'num': fBytes(l2_asize)
@@ -684,6 +743,26 @@ def get_l2arc_summary(Kstat):
             'per': fPerc(l2_hdr_size, l2_size),
             'num': fBytes(l2_hdr_size),
         }
+        output["l2_arc_size"]["mfu_asize"] = {
+            'per': fPerc(l2_mfu_asize, l2_asize),
+            'num': fBytes(l2_mfu_asize),
+        }
+        output["l2_arc_size"]["mru_asize"] = {
+            'per': fPerc(l2_mru_asize, l2_asize),
+            'num': fBytes(l2_mru_asize),
+        }
+        output["l2_arc_size"]["prefetch_asize"] = {
+            'per': fPerc(l2_prefetch_asize, l2_asize),
+            'num': fBytes(l2_prefetch_asize),
+        }
+        output["l2_arc_size"]["bufc_data_asize"] = {
+            'per': fPerc(l2_bufc_data_asize, l2_asize),
+            'num': fBytes(l2_bufc_data_asize),
+        }
+        output["l2_arc_size"]["bufc_metadata_asize"] = {
+            'per': fPerc(l2_bufc_metadata_asize, l2_asize),
+            'num': fBytes(l2_bufc_metadata_asize),
+        }
 
         output["l2_arc_evicts"] = {}
         output["l2_arc_evicts"]['lock_retries'] = fHits(l2_evict_lock_retry)
@@ -748,7 +827,7 @@ def _l2arc_summary(Kstat):
         sys.stdout.write("\n")
 
         sys.stdout.write("L2 ARC Size: (Adaptive)\t\t\t\t%s\n" %
-                         arc["l2_arc_size"]["adative"])
+                         arc["l2_arc_size"]["adaptive"])
         sys.stdout.write("\tCompressed:\t\t\t%s\t%s\n" % (
             arc["l2_arc_size"]["actual"]["per"],
             arc["l2_arc_size"]["actual"]["num"],
@@ -759,11 +838,36 @@ def _l2arc_summary(Kstat):
             arc["l2_arc_size"]["head_size"]["num"],
             )
         )
+        sys.stdout.write("\tMFU Alloc. Size:\t\t%s\t%s\n" % (
+            arc["l2_arc_size"]["mfu_asize"]["per"],
+            arc["l2_arc_size"]["mfu_asize"]["num"],
+            )
+        )
+        sys.stdout.write("\tMRU Alloc. Size:\t\t%s\t%s\n" % (
+            arc["l2_arc_size"]["mru_asize"]["per"],
+            arc["l2_arc_size"]["mru_asize"]["num"],
+            )
+        )
+        sys.stdout.write("\tPrefetch Alloc. Size:\t\t%s\t%s\n" % (
+            arc["l2_arc_size"]["prefetch_asize"]["per"],
+            arc["l2_arc_size"]["prefetch_asize"]["num"],
+            )
+        )
+        sys.stdout.write("\tData (buf content) Alloc. Size:\t%s\t%s\n" % (
+            arc["l2_arc_size"]["bufc_data_asize"]["per"],
+            arc["l2_arc_size"]["bufc_data_asize"]["num"],
+            )
+        )
+        sys.stdout.write("\tMetadata (buf content) Size:\t%s\t%s\n" % (
+            arc["l2_arc_size"]["bufc_metadata_asize"]["per"],
+            arc["l2_arc_size"]["bufc_metadata_asize"]["num"],
+            )
+        )
         sys.stdout.write("\n")
 
         if arc["l2_arc_evicts"]['lock_retries'] != '0' or \
            arc["l2_arc_evicts"]["reading"] != '0':
-            sys.stdout.write("L2 ARC Evicts:\n")
+            sys.stdout.write("L2 ARC Evictions:\n")
             sys.stdout.write("\tLock Retries:\t\t\t\t%s\n" %
                              arc["l2_arc_evicts"]['lock_retries'])
             sys.stdout.write("\tUpon Reading:\t\t\t\t%s\n" %
@@ -921,14 +1025,7 @@ def _tunable_summary(Kstat):
     global show_tunable_descriptions
     global alternate_tunable_layout
 
-    names = os.listdir("/sys/module/zfs/parameters/")
+    tunables = load_tunables()
-
-    values = {}
-    for name in names:
-        with open("/sys/module/zfs/parameters/" + name) as f:
-            value = f.read()
-        values[name] = value.strip()
-
     descriptions = {}
 
     if show_tunable_descriptions:
@@ -966,22 +1063,17 @@ def _tunable_summary(Kstat):
             sys.stderr.write("Tunable descriptions will be disabled.\n")
 
     sys.stdout.write("ZFS Tunables:\n")
-    names.sort()
 
     if alternate_tunable_layout:
         fmt = "\t%s=%s\n"
     else:
         fmt = "\t%-50s%s\n"
 
-    for name in names:
+    for name in sorted(tunables.keys()):
-
-        if not name:
-            continue
-
         if show_tunable_descriptions and name in descriptions:
             sys.stdout.write("\t# %s\n" % descriptions[name])
 
-        sys.stdout.write(fmt % (name, values[name]))
+        sys.stdout.write(fmt % (name, tunables[name]))
 
 
 unSub = [
@@ -1033,48 +1125,55 @@ def main():
     global alternate_tunable_layout
 
     try:
-        opts, args = getopt.getopt(
-            sys.argv[1:],
-            "adp:h", ["alternate", "description", "page=", "help"]
-        )
-    except getopt.error as e:
-        sys.stderr.write("Error: %s\n" % e.msg)
-        usage()
-        sys.exit(1)
-
-    args = {}
-    for opt, arg in opts:
-        if opt in ('-a', '--alternate'):
-            args['a'] = True
-        if opt in ('-d', '--description'):
-            args['d'] = True
-        if opt in ('-p', '--page'):
-            args['p'] = arg
-        if opt in ('-h', '--help'):
-            usage()
-            sys.exit(0)
-
-    Kstat = get_Kstat()
-
-    alternate_tunable_layout = 'a' in args
-    show_tunable_descriptions = 'd' in args
-
-    pages = []
-
-    if 'p' in args:
         try:
-            pages.append(unSub[int(args['p']) - 1])
+            opts, args = getopt.getopt(
-        except IndexError:
+                sys.argv[1:],
-            sys.stderr.write('the argument to -p must be between 1 and ' +
+                "adp:h", ["alternate", "description", "page=", "help"]
-                             str(len(unSub)) + '\n')
+            )
+        except getopt.error as e:
+            sys.stderr.write("Error: %s\n" % e.msg)
+            usage()
             sys.exit(1)
-    else:
-        pages = unSub
 
-    zfs_header()
+        args = {}
-    for page in pages:
+        for opt, arg in opts:
-        page(Kstat)
+            if opt in ('-a', '--alternate'):
-        sys.stdout.write("\n")
+                args['a'] = True
+            if opt in ('-d', '--description'):
+                args['d'] = True
+            if opt in ('-p', '--page'):
+                args['p'] = arg
+            if opt in ('-h', '--help'):
+                usage()
+                sys.exit(0)
+
+        Kstat = get_Kstat()
+
+        alternate_tunable_layout = 'a' in args
+        show_tunable_descriptions = 'd' in args
+
+        pages = []
+
+        if 'p' in args:
+            try:
+                pages.append(unSub[int(args['p']) - 1])
+            except IndexError:
+                sys.stderr.write('the argument to -p must be between 1 and ' +
+                                 str(len(unSub)) + '\n')
+                sys.exit(1)
+        else:
+            pages = unSub
+
+        zfs_header()
+        for page in pages:
+            page(Kstat)
+            sys.stdout.write("\n")
+    except IOError as ex:
+        if (ex.errno == errno.EPIPE):
+            sys.exit(0)
+        raise
+    except KeyboardInterrupt:
+        sys.exit(0)
 
 
 if __name__ == '__main__':

cmd/arc_summary/arc_summary3

@@ -1,4 +1,4 @@
-#!/usr/bin/python3
+#!/usr/bin/env python3
 #
 # Copyright (c) 2008 Ben Rockwood <benr@cuddletech.com>,
 # Copyright (c) 2010 Martin Matuska <mm@FreeBSD.org>,
@@ -32,7 +32,7 @@
 Provides basic information on the ARC, its efficiency, the L2ARC (if present),
 the Data Management Unit (DMU), Virtual Devices (VDEVs), and tunables. See
 the in-source documentation and code at
-https://github.com/zfsonlinux/zfs/blob/master/module/zfs/arc.c for details.
+https://github.com/openzfs/zfs/blob/master/module/zfs/arc.c for details.
 The original introduction to arc_summary can be found at
 http://cuddletech.com/?p=454
 """
@@ -42,13 +42,17 @@ import os
 import subprocess
 import sys
 import time
+import errno
 
-DECRIPTION = 'Print ARC and other statistics for ZFS on Linux'
+# We can't use env -S portably, and we need python3 -u to handle pipes in
+# the shell abruptly closing the way we want to, so...
+import io
+if isinstance(sys.__stderr__.buffer, io.BufferedWriter):
+    os.execv(sys.executable, [sys.executable, "-u"] + sys.argv)
+
+DESCRIPTION = 'Print ARC and other statistics for OpenZFS'
 INDENT = ' '*8
 LINE_LENGTH = 72
-PROC_PATH = '/proc/spl/kstat/zfs/'
-SPL_PATH = '/sys/module/spl/parameters/'
-TUNABLES_PATH = '/sys/module/zfs/parameters/'
 DATE_FORMAT = '%a %b %d %H:%M:%S %Y'
 TITLE = 'ZFS Subsystem Report'
 
@@ -61,11 +65,10 @@ SECTION_PATHS = {'arc': 'arcstats',
                  'dmu': 'dmu_tx',
                  'l2arc': 'arcstats',  # L2ARC stuff lives in arcstats
                  'vdev': 'vdev_cache_stats',
-                 'xuio': 'xuio_stats',
                  'zfetch': 'zfetchstats',
                  'zil': 'zil'}
 
-parser = argparse.ArgumentParser(description=DECRIPTION)
+parser = argparse.ArgumentParser(description=DESCRIPTION)
 parser.add_argument('-a', '--alternate', action='store_true', default=False,
                     help='use alternate formatting for tunables and SPL',
                     dest='alt')
@@ -83,6 +86,172 @@ parser.add_argument('-s', '--section', dest='section', help=SECTION_HELP)
 ARGS = parser.parse_args()
 
 
+if sys.platform.startswith('freebsd'):
+    # Requires py36-sysctl on FreeBSD
+    import sysctl
+
+    VDEV_CACHE_SIZE = 'vdev.cache_size'
+
+    def is_value(ctl):
+        return ctl.type != sysctl.CTLTYPE_NODE
+
+    def namefmt(ctl, base='vfs.zfs.'):
+        # base is removed from the name
+        cut = len(base)
+        return ctl.name[cut:]
+
+    def load_kstats(section):
+        base = 'kstat.zfs.misc.{section}.'.format(section=section)
+        fmt = lambda kstat: '{name} : {value}'.format(name=namefmt(kstat, base),
+                                                      value=kstat.value)
+        kstats = sysctl.filter(base)
+        return [fmt(kstat) for kstat in kstats if is_value(kstat)]
+
+    def get_params(base):
+        ctls = sysctl.filter(base)
+        return {namefmt(ctl): str(ctl.value) for ctl in ctls if is_value(ctl)}
+
+    def get_tunable_params():
+        return get_params('vfs.zfs')
+
+    def get_vdev_params():
+        return get_params('vfs.zfs.vdev')
+
+    def get_version_impl(request):
+        # FreeBSD reports versions for zpl and spa instead of zfs and spl.
+        name = {'zfs': 'zpl',
+                'spl': 'spa'}[request]
+        mib = 'vfs.zfs.version.{}'.format(name)
+        version = sysctl.filter(mib)[0].value
+        return '{} version {}'.format(name, version)
+
+    def get_descriptions(_request):
+        ctls = sysctl.filter('vfs.zfs')
+        return {namefmt(ctl): ctl.description for ctl in ctls if is_value(ctl)}
+
+
+elif sys.platform.startswith('linux'):
+    KSTAT_PATH = '/proc/spl/kstat/zfs'
+    SPL_PATH = '/sys/module/spl/parameters'
+    TUNABLES_PATH = '/sys/module/zfs/parameters'
+
+    VDEV_CACHE_SIZE = 'zfs_vdev_cache_size'
+
+    def load_kstats(section):
+        path = os.path.join(KSTAT_PATH, section)
+        with open(path) as f:
+            return list(f)[2:] # Get rid of header
+
+    def get_params(basepath):
+        """Collect information on the Solaris Porting Layer (SPL) or the
+        tunables, depending on the PATH given. Does not check if PATH is
+        legal.
+        """
+        result = {}
+        for name in os.listdir(basepath):
+            path = os.path.join(basepath, name)
+            with open(path) as f:
+                value = f.read()
+                result[name] = value.strip()
+        return result
+
+    def get_spl_params():
+        return get_params(SPL_PATH)
+
+    def get_tunable_params():
+        return get_params(TUNABLES_PATH)
+
+    def get_vdev_params():
+        return get_params(TUNABLES_PATH)
+
+    def get_version_impl(request):
+        # The original arc_summary called /sbin/modinfo/{spl,zfs} to get
+        # the version information. We switch to /sys/module/{spl,zfs}/version
+        # to make sure we get what is really loaded in the kernel
+        try:
+            with open("/sys/module/{}/version".format(request)) as f:
+                return f.read().strip()
+        except:
+            return "(unknown)"
+
+    def get_descriptions(request):
+        """Get the descriptions of the Solaris Porting Layer (SPL) or the
+        tunables, return with minimal formatting.
+        """
+
+        if request not in ('spl', 'zfs'):
+            print('ERROR: description of "{0}" requested)'.format(request))
+            sys.exit(1)
+
+        descs = {}
+        target_prefix = 'parm:'
+
+        # We would prefer to do this with /sys/modules -- see the discussion at
+        # get_version() -- but there isn't a way to get the descriptions from
+        # there, so we fall back on modinfo
+        command = ["/sbin/modinfo", request, "-0"]
+
+        # The recommended way to do this is with subprocess.run(). However,
+        # some installed versions of Python are < 3.5, so we offer them
+        # the option of doing it the old way (for now)
+        info = ''
+
+        try:
+
+            if 'run' in dir(subprocess):
+                info = subprocess.run(command, stdout=subprocess.PIPE,
+                                      universal_newlines=True)
+                raw_output = info.stdout.split('\0')
+            else:
+                info = subprocess.check_output(command,
+                                               universal_newlines=True)
+                raw_output = info.split('\0')
+
+        except subprocess.CalledProcessError:
+            print("Error: Descriptions not available",
+                  "(can't access kernel module)")
+            sys.exit(1)
+
+        for line in raw_output:
+
+            if not line.startswith(target_prefix):
+                continue
+
+            line = line[len(target_prefix):].strip()
+            name, raw_desc = line.split(':', 1)
+            desc = raw_desc.rsplit('(', 1)[0]
+
+            if desc == '':
+                desc = '(No description found)'
+
+            descs[name.strip()] = desc.strip()
+
+        return descs
+
+def handle_unraisableException(exc_type, exc_value=None, exc_traceback=None,
+                               err_msg=None, object=None):
+   handle_Exception(exc_type, object, exc_traceback)
+
+def handle_Exception(ex_cls, ex, tb):
+    if ex_cls is KeyboardInterrupt:
+        sys.exit()
+
+    if ex_cls is BrokenPipeError:
+        # It turns out that while sys.exit() triggers an exception
+        # not handled message on Python 3.8+, os._exit() does not.
+        os._exit(0)
+
+    if ex_cls is OSError:
+      if ex.errno == errno.ENOTCONN:
+        sys.exit()
+
+    raise ex
+
+if hasattr(sys,'unraisablehook'): # Python 3.8+
+    sys.unraisablehook = handle_unraisableException
+sys.excepthook = handle_Exception
+
+
 def cleanup_line(single_line):
     """Format a raw line of data from /proc and isolate the name value
     part, returning a tuple with each. Currently, this gets rid of the
@@ -238,139 +407,48 @@ def format_raw_line(name, value):
     if ARGS.alt:
         result = '{0}{1}={2}'.format(INDENT, name, value)
     else:
-        spc = LINE_LENGTH-(len(INDENT)+len(value))
+        # Right-align the value within the line length if it fits,
-        result = '{0}{1:<{spc}}{2}'.format(INDENT, name, value, spc=spc)
+        # otherwise just separate it from the name by a single space.
+        fit = LINE_LENGTH - len(INDENT) - len(name)
+        overflow = len(value) + 1
+        w = max(fit, overflow)
+        result = '{0}{1}{2:>{w}}'.format(INDENT, name, value, w=w)
 
     return result
 
 
 def get_kstats():
-    """Collect information on the ZFS subsystem from the /proc Linux virtual
+    """Collect information on the ZFS subsystem. The step does not perform any
-    file system. The step does not perform any further processing, giving us
+    further processing, giving us the option to only work on what is actually
-    the option to only work on what is actually needed. The name "kstat" is a
+    needed. The name "kstat" is a holdover from the Solaris utility of the same
-    holdover from the Solaris utility of the same name.
+    name.
     """
 
     result = {}
-    secs = SECTION_PATHS.values()
 
-    for section in secs:
+    for section in SECTION_PATHS.values():
-
+        if section not in result:
-        with open(PROC_PATH+section, 'r') as proc_location:
+            result[section] = load_kstats(section)
-            lines = [line for line in proc_location]
-
-        del lines[0:2]  # Get rid of header
-        result[section] = lines
 
     return result
 
 
-def get_spl_tunables(PATH):
-    """Collect information on the Solaris Porting Layer (SPL) or the
-    tunables, depending on the PATH given. Does not check if PATH is
-    legal.
-    """
-
-    result = {}
-    parameters = os.listdir(PATH)
-
-    for name in parameters:
-
-        with open(PATH+name, 'r') as para_file:
-            value = para_file.read()
-            result[name] = value.strip()
-
-    return result
-
-
-def get_descriptions(request):
-    """Get the decriptions of the Solaris Porting Layer (SPL) or the
-    tunables, return with minimal formatting.
-    """
-
-    if request not in ('spl', 'zfs'):
-        print('ERROR: description of "{0}" requested)'.format(request))
-        sys.exit(1)
-
-    descs = {}
-    target_prefix = 'parm:'
-
-    # We would prefer to do this with /sys/modules -- see the discussion at
-    # get_version() -- but there isn't a way to get the descriptions from
-    # there, so we fall back on modinfo
-    command = ["/sbin/modinfo", request, "-0"]
-
-    # The recommended way to do this is with subprocess.run(). However,
-    # some installed versions of Python are < 3.5, so we offer them
-    # the option of doing it the old way (for now)
-    info = ''
-
-    try:
-
-        if 'run' in dir(subprocess):
-            info = subprocess.run(command, stdout=subprocess.PIPE,
-                                  universal_newlines=True)
-            raw_output = info.stdout.split('\0')
-        else:
-            info = subprocess.check_output(command, universal_newlines=True)
-            raw_output = info.split('\0')
-
-    except subprocess.CalledProcessError:
-        print("Error: Descriptions not available (can't access kernel module)")
-        sys.exit(1)
-
-    for line in raw_output:
-
-        if not line.startswith(target_prefix):
-            continue
-
-        line = line[len(target_prefix):].strip()
-        name, raw_desc = line.split(':', 1)
-        desc = raw_desc.rsplit('(', 1)[0]
-
-        if desc == '':
-            desc = '(No description found)'
-
-        descs[name.strip()] = desc.strip()
-
-    return descs
-
-
 def get_version(request):
     """Get the version number of ZFS or SPL on this machine for header.
     Returns an error string, but does not raise an error, if we can't
-    get the ZFS/SPL version via modinfo.
+    get the ZFS/SPL version.
     """
 
     if request not in ('spl', 'zfs'):
         error_msg = '(ERROR: "{0}" requested)'.format(request)
         return error_msg
 
-    # The original arc_summary called /sbin/modinfo/{spl,zfs} to get
+    return get_version_impl(request)
-    # the version information. We switch to /sys/module/{spl,zfs}/version
-    # to make sure we get what is really loaded in the kernel
-    command = ["cat", "/sys/module/{0}/version".format(request)]
-    req = request.upper()
-    version = "(Can't get {0} version)".format(req)
-
-    # The recommended way to do this is with subprocess.run(). However,
-    # some installed versions of Python are < 3.5, so we offer them
-    # the option of doing it the old way (for now)
-    info = ''
-    if 'run' in dir(subprocess):
-        info = subprocess.run(command, stdout=subprocess.PIPE,
-                              universal_newlines=True)
-        version = info.stdout.strip()
-    else:
-        info = subprocess.check_output(command, universal_newlines=True)
-        version = info.strip()
-
-    return version
 
 
 def print_header():
     """Print the initial heading with date and time as well as info on the
-    Linux and ZFS versions. This is not called for the graph.
+    kernel and ZFS versions. This is not called for the graph.
     """
 
     # datetime is now recommended over time but we keep the exact formatting
@@ -534,6 +612,20 @@ def section_arc(kstats_dict):
     prt_i1('Deleted:', f_hits(arc_stats['deleted']))
     prt_i1('Mutex misses:', f_hits(arc_stats['mutex_miss']))
     prt_i1('Eviction skips:', f_hits(arc_stats['evict_skip']))
+    prt_i1('Eviction skips due to L2 writes:',
+           f_hits(arc_stats['evict_l2_skip']))
+    prt_i1('L2 cached evictions:', f_bytes(arc_stats['evict_l2_cached']))
+    prt_i1('L2 eligible evictions:', f_bytes(arc_stats['evict_l2_eligible']))
+    prt_i2('L2 eligible MFU evictions:',
+           f_perc(arc_stats['evict_l2_eligible_mfu'],
+           arc_stats['evict_l2_eligible']),
+           f_bytes(arc_stats['evict_l2_eligible_mfu']))
+    prt_i2('L2 eligible MRU evictions:',
+           f_perc(arc_stats['evict_l2_eligible_mru'],
+           arc_stats['evict_l2_eligible']),
+           f_bytes(arc_stats['evict_l2_eligible_mru']))
+    prt_i1('L2 ineligible evictions:',
+           f_bytes(arc_stats['evict_l2_ineligible']))
     print()
 
 
@@ -672,15 +764,30 @@ def section_l2arc(kstats_dict):
     prt_i2('Header size:',
            f_perc(arc_stats['l2_hdr_size'], arc_stats['l2_size']),
            f_bytes(arc_stats['l2_hdr_size']))
+    prt_i2('MFU allocated size:',
+           f_perc(arc_stats['l2_mfu_asize'], arc_stats['l2_asize']),
+           f_bytes(arc_stats['l2_mfu_asize']))
+    prt_i2('MRU allocated size:',
+           f_perc(arc_stats['l2_mru_asize'], arc_stats['l2_asize']),
+           f_bytes(arc_stats['l2_mru_asize']))
+    prt_i2('Prefetch allocated size:',
+           f_perc(arc_stats['l2_prefetch_asize'], arc_stats['l2_asize']),
+           f_bytes(arc_stats['l2_prefetch_asize']))
+    prt_i2('Data (buffer content) allocated size:',
+           f_perc(arc_stats['l2_bufc_data_asize'], arc_stats['l2_asize']),
+           f_bytes(arc_stats['l2_bufc_data_asize']))
+    prt_i2('Metadata (buffer content) allocated size:',
+           f_perc(arc_stats['l2_bufc_metadata_asize'], arc_stats['l2_asize']),
+           f_bytes(arc_stats['l2_bufc_metadata_asize']))
 
     print()
     prt_1('L2ARC breakdown:', f_hits(l2_access_total))
     prt_i2('Hit ratio:',
            f_perc(arc_stats['l2_hits'], l2_access_total),
-           f_bytes(arc_stats['l2_hits']))
+           f_hits(arc_stats['l2_hits']))
     prt_i2('Miss ratio:',
            f_perc(arc_stats['l2_misses'], l2_access_total),
-           f_bytes(arc_stats['l2_misses']))
+           f_hits(arc_stats['l2_misses']))
     prt_i1('Feeds:', f_hits(arc_stats['l2_feeds']))
 
     print()
@@ -691,13 +798,13 @@ def section_l2arc(kstats_dict):
         prt_i2('Done ratio:',
                f_perc(arc_stats['l2_writes_done'],
                       arc_stats['l2_writes_sent']),
-               f_bytes(arc_stats['l2_writes_done']))
+               f_hits(arc_stats['l2_writes_done']))
         prt_i2('Error ratio:',
                f_perc(arc_stats['l2_writes_error'],
                       arc_stats['l2_writes_sent']),
-               f_bytes(arc_stats['l2_writes_error']))
+               f_hits(arc_stats['l2_writes_error']))
     else:
-        prt_i2('Writes sent:', '100 %', f_bytes(arc_stats['l2_writes_sent']))
+        prt_i2('Writes sent:', '100 %', f_hits(arc_stats['l2_writes_sent']))
 
     print()
     print('L2ARC evicts:')
@@ -708,10 +815,14 @@ def section_l2arc(kstats_dict):
 
 def section_spl(*_):
     """Print the SPL parameters, if requested with alternative format
-    and/or decriptions. This does not use kstats.
+    and/or descriptions. This does not use kstats.
     """
 
-    spls = get_spl_tunables(SPL_PATH)
+    if sys.platform.startswith('freebsd'):
+        # No SPL support in FreeBSD
+        return
+
+    spls = get_spl_params()
     keylist = sorted(spls.keys())
     print('Solaris Porting Layer (SPL):')
 
@@ -725,7 +836,7 @@ def section_spl(*_):
             try:
                 print(INDENT+'#', descriptions[key])
             except KeyError:
-                print(INDENT+'# (No decription found)')  # paranoid
+                print(INDENT+'# (No description found)')  # paranoid
 
         print(format_raw_line(key, value))
 
@@ -734,10 +845,10 @@ def section_spl(*_):
 
 def section_tunables(*_):
     """Print the tunables, if requested with alternative format and/or
-    decriptions. This does not use kstasts.
+    descriptions. This does not use kstasts.
     """
 
-    tunables = get_spl_tunables(TUNABLES_PATH)
+    tunables = get_tunable_params()
     keylist = sorted(tunables.keys())
     print('Tunables:')
 
@@ -751,7 +862,7 @@ def section_tunables(*_):
             try:
                 print(INDENT+'#', descriptions[key])
             except KeyError:
-                print(INDENT+'# (No decription found)')  # paranoid
+                print(INDENT+'# (No description found)')  # paranoid
 
         print(format_raw_line(key, value))
 
@@ -763,11 +874,11 @@ def section_vdev(kstats_dict):
 
     # Currently [Nov 2017] the VDEV cache is disabled, because it is actually
     # harmful. When this is the case, we just skip the whole entry. See
-    # https://github.com/zfsonlinux/zfs/blob/master/module/zfs/vdev_cache.c
+    # https://github.com/openzfs/zfs/blob/master/module/zfs/vdev_cache.c
     # for details
-    tunables = get_spl_tunables(TUNABLES_PATH)
+    tunables = get_vdev_params()
 
-    if tunables['zfs_vdev_cache_size'] == '0':
+    if tunables[VDEV_CACHE_SIZE] == '0':
         print('VDEV cache disabled, skipping section\n')
         return
 
@@ -789,7 +900,7 @@ def section_vdev(kstats_dict):
 
 def section_zil(kstats_dict):
     """Collect information on the ZFS Intent Log. Some of the information
-    taken from https://github.com/zfsonlinux/zfs/blob/master/include/sys/zil.h
+    taken from https://github.com/openzfs/zfs/blob/master/include/sys/zil.h
     """
 
     zil_stats = isolate_section('zil', kstats_dict)

cmd/arcstat/.gitignore

@@ -0,0 +1 @@
+arcstat

cmd/arcstat/Makefile.am

@@ -1,13 +1,5 @@
-dist_bin_SCRIPTS = arcstat
+include $(top_srcdir)/config/Substfiles.am
 
-#
+bin_SCRIPTS = arcstat
-# The arcstat script is compatibile with both Python 2.6 and 3.4.
+
-# As such the python 3 shebang can be replaced at install time when
+SUBSTFILES += $(bin_SCRIPTS)
-# targeting a python 2 system.  This allows us to maintain a single
-# version of the source.
-#
-if USING_PYTHON_2
-install-exec-hook:
-	sed --in-place 's|^#!/usr/bin/python3|#!/usr/bin/python2|' \
-	    $(DESTDIR)$(bindir)/arcstat
-endif

cmd/arcstat/arcstat.in

@@ -1,20 +1,25 @@
-#!/usr/bin/python3
+#!/usr/bin/env @PYTHON_SHEBANG@
 #
 # Print out ZFS ARC Statistics exported via kstat(1)
-# For a definition of fields, or usage, use arctstat.pl -v
+# For a definition of fields, or usage, use arcstat -v
 #
-# This script is a fork of the original arcstat.pl (0.1) by
+# This script was originally a fork of the original arcstat.pl (0.1)
-# Neelakanth Nadgir, originally published on his Sun blog on
+# by Neelakanth Nadgir, originally published on his Sun blog on
 # 09/18/2007
 #     http://blogs.sun.com/realneel/entry/zfs_arc_statistics
 #
-# This version aims to improve upon the original by adding features
+# A new version aimed to improve upon the original by adding features
-# and fixing bugs as needed.  This version is maintained by
+# and fixing bugs as needed.  This version was maintained by Mike
-# Mike Harsch and is hosted in a public open source repository:
+# Harsch and was hosted in a public open source repository:
 #    http://github.com/mharsch/arcstat
 #
-# Comments, Questions, or Suggestions are always welcome.
+# but has since moved to the illumos-gate repository.
-# Contact the maintainer at ( mike at harschsystems dot com )
+#
+# This Python port was written by John Hixson for FreeNAS, introduced
+# in commit e2c29f:
+#    https://github.com/freenas/freenas
+#
+# and has been improved by many people since.
 #
 # CDDL HEADER START
 #
@@ -51,16 +56,16 @@ import getopt
 import re
 import copy
 
-from decimal import Decimal
 from signal import signal, SIGINT, SIGWINCH, SIG_DFL
 
+
 cols = {
     # HDR:        [Size, Scale, Description]
     "time":       [8, -1, "Time"],
     "hits":       [4, 1000, "ARC reads per second"],
     "miss":       [4, 1000, "ARC misses per second"],
     "read":       [4, 1000, "Total ARC accesses per second"],
-    "hit%":       [4, 100, "ARC Hit percentage"],
+    "hit%":       [4, 100, "ARC hit percentage"],
     "miss%":      [5, 100, "ARC miss percentage"],
     "dhit":       [4, 1000, "Demand hits per second"],
     "dmis":       [4, 1000, "Demand misses per second"],
@@ -75,13 +80,20 @@ cols = {
     "mread":      [5, 1000, "Metadata accesses per second"],
     "mh%":        [3, 100, "Metadata hit percentage"],
     "mm%":        [3, 100, "Metadata miss percentage"],
-    "arcsz":      [5, 1024, "ARC Size"],
+    "arcsz":      [5, 1024, "ARC size"],
-    "c":          [4, 1024, "ARC Target Size"],
+    "size":       [4, 1024, "ARC size"],
-    "mfu":        [4, 1000, "MFU List hits per second"],
+    "c":          [4, 1024, "ARC target size"],
-    "mru":        [4, 1000, "MRU List hits per second"],
+    "mfu":        [4, 1000, "MFU list hits per second"],
-    "mfug":       [4, 1000, "MFU Ghost List hits per second"],
+    "mru":        [4, 1000, "MRU list hits per second"],
-    "mrug":       [4, 1000, "MRU Ghost List hits per second"],
+    "mfug":       [4, 1000, "MFU ghost list hits per second"],
+    "mrug":       [4, 1000, "MRU ghost list hits per second"],
     "eskip":      [5, 1000, "evict_skip per second"],
+    "el2skip":    [7, 1000, "evict skip, due to l2 writes, per second"],
+    "el2cach":    [7, 1024, "Size of L2 cached evictions per second"],
+    "el2el":      [5, 1024, "Size of L2 eligible evictions per second"],
+    "el2mfu":     [6, 1024, "Size of L2 eligible MFU evictions per second"],
+    "el2mru":     [6, 1024, "Size of L2 eligible MRU evictions per second"],
+    "el2inel":    [7, 1024, "Size of L2 ineligible evictions per second"],
     "mtxmis":     [6, 1000, "mutex_miss per second"],
     "dread":      [5, 1000, "Demand accesses per second"],
     "pread":      [5, 1000, "Prefetch accesses per second"],
@@ -90,17 +102,29 @@ cols = {
     "l2read":     [6, 1000, "Total L2ARC accesses per second"],
     "l2hit%":     [6, 100, "L2ARC access hit percentage"],
     "l2miss%":    [7, 100, "L2ARC access miss percentage"],
+    "l2pref":     [6, 1024, "L2ARC prefetch allocated size"],
+    "l2mfu":      [5, 1024, "L2ARC MFU allocated size"],
+    "l2mru":      [5, 1024, "L2ARC MRU allocated size"],
+    "l2data":     [6, 1024, "L2ARC data allocated size"],
+    "l2meta":     [6, 1024, "L2ARC metadata allocated size"],
+    "l2pref%":    [7, 100, "L2ARC prefetch percentage"],
+    "l2mfu%":     [6, 100, "L2ARC MFU percentage"],
+    "l2mru%":     [6, 100, "L2ARC MRU percentage"],
+    "l2data%":    [7, 100, "L2ARC data percentage"],
+    "l2meta%":    [7, 100, "L2ARC metadata percentage"],
     "l2asize":    [7, 1024, "Actual (compressed) size of the L2ARC"],
     "l2size":     [6, 1024, "Size of the L2ARC"],
-    "l2bytes":    [7, 1024, "bytes read per second from the L2ARC"],
+    "l2bytes":    [7, 1024, "Bytes read per second from the L2ARC"],
-    "grow":       [4, 1000, "ARC Grow disabled"],
+    "grow":       [4, 1000, "ARC grow disabled"],
-    "need":       [4, 1024, "ARC Reclaim need"],
+    "need":       [4, 1024, "ARC reclaim need"],
-    "free":       [4, 1024, "ARC Free memory"],
+    "free":       [4, 1024, "ARC free memory"],
+    "avail":      [5, 1024, "ARC available memory"],
+    "waste":      [5, 1024, "Wasted memory due to round up to pagesize"],
 }
 
 v = {}
 hdr = ["time", "read", "miss", "miss%", "dmis", "dm%", "pmis", "pm%", "mmis",
-       "mm%", "arcsz", "c"]
+       "mm%", "size", "c", "avail"]
 xhdr = ["time", "mfu", "mru", "mfug", "mrug", "eskip", "mtxmis", "dread",
         "pread", "read"]
 sint = 1               # Default interval is 1 second
@@ -110,12 +134,56 @@ opfile = None
 sep = "  "              # Default separator is 2 spaces
 version = "0.4"
 l2exist = False
-cmd = ("Usage: arcstat [-hvx] [-f fields] [-o file] [-s string] [interval "
+cmd = ("Usage: arcstat [-havxp] [-f fields] [-o file] [-s string] [interval "
        "[count]]\n")
 cur = {}
 d = {}
 out = None
 kstat = None
+pretty_print = True
+
+
+if sys.platform.startswith('freebsd'):
+    # Requires py-sysctl on FreeBSD
+    import sysctl
+
+    def kstat_update():
+        global kstat
+
+        k = [ctl for ctl in sysctl.filter('kstat.zfs.misc.arcstats')
+             if ctl.type != sysctl.CTLTYPE_NODE]
+
+        if not k:
+            sys.exit(1)
+
+        kstat = {}
+
+        for s in k:
+            if not s:
+                continue
+
+            name, value = s.name, s.value
+            # Trims 'kstat.zfs.misc.arcstats' from the name
+            kstat[name[24:]] = int(value)
+
+elif sys.platform.startswith('linux'):
+    def kstat_update():
+        global kstat
+
+        k = [line.strip() for line in open('/proc/spl/kstat/zfs/arcstats')]
+
+        if not k:
+            sys.exit(1)
+
+        del k[0:2]
+        kstat = {}
+
+        for s in k:
+            if not s:
+                continue
+
+            name, unused, value = s.split()
+            kstat[name] = int(value)
 
 
 def detailed_usage():
@@ -131,6 +199,7 @@ def detailed_usage():
 def usage():
     sys.stderr.write("%s\n" % cmd)
     sys.stderr.write("\t -h : Print this help message\n")
+    sys.stderr.write("\t -a : Print all possible stats\n")
     sys.stderr.write("\t -v : List all possible field headers and definitions"
                      "\n")
     sys.stderr.write("\t -x : Print extended stats\n")
@@ -138,6 +207,7 @@ def usage():
     sys.stderr.write("\t -o : Redirect output to the specified file\n")
     sys.stderr.write("\t -s : Override default field separator with custom "
                      "character or string\n")
+    sys.stderr.write("\t -p : Disable auto-scaling of numerical fields\n")
     sys.stderr.write("\nExamples:\n")
     sys.stderr.write("\tarcstat -o /tmp/a.log 2 10\n")
     sys.stderr.write("\tarcstat -s \",\" -o /tmp/a.log 2 10\n")
@@ -148,25 +218,6 @@ def usage():
     sys.exit(1)
 
 
-def kstat_update():
-    global kstat
-
-    k = [line.strip() for line in open('/proc/spl/kstat/zfs/arcstats')]
-
-    if not k:
-        sys.exit(1)
-
-    del k[0:2]
-    kstat = {}
-
-    for s in k:
-        if not s:
-            continue
-
-        name, unused, value = s.split()
-        kstat[name] = Decimal(value)
-
-
 def snap_stats():
     global cur
     global kstat
@@ -197,7 +248,7 @@ def prettynum(sz, scale, num=0):
     elif 0 < num < 1:
         num = 0
 
-    while num > scale and index < 5:
+    while abs(num) > scale and index < 5:
         save = num
         num = num / scale
         index += 1
@@ -205,7 +256,7 @@ def prettynum(sz, scale, num=0):
     if index == 0:
         return "%*d" % (sz, num)
 
-    if (save / scale) < 10:
+    if abs(save / scale) < 10:
         return "%*.1f%s" % (sz - 1, num, suffix[index])
     else:
         return "%*d%s" % (sz - 1, num, suffix[index])
@@ -215,12 +266,14 @@ def print_values():
     global hdr
     global sep
     global v
+    global pretty_print
 
-    for col in hdr:
+    if pretty_print:
-        sys.stdout.write("%s%s" % (
+        fmt = lambda col: prettynum(cols[col][0], cols[col][1], v[col])
-            prettynum(cols[col][0], cols[col][1], v[col]),
+    else:
-            sep
+        fmt = lambda col: v[col]
-        ))
+
+    sys.stdout.write(sep.join(fmt(col) for col in hdr))
     sys.stdout.write("\n")
     sys.stdout.flush()
 
@@ -228,9 +281,14 @@ def print_values():
 def print_header():
     global hdr
     global sep
+    global pretty_print
 
-    for col in hdr:
+    if pretty_print:
-        sys.stdout.write("%*s%s" % (cols[col][0], col, sep))
+        fmt = lambda col: "%*s" % (cols[col][0], col)
+    else:
+        fmt = lambda col: col
+
+    sys.stdout.write(sep.join(fmt(col) for col in hdr))
     sys.stdout.write("\n")
 
 
@@ -267,8 +325,10 @@ def init():
     global sep
     global out
     global l2exist
+    global pretty_print
 
     desired_cols = None
+    aflag = False
     xflag = False
     hflag = False
     vflag = False
@@ -277,14 +337,16 @@ def init():
     try:
         opts, args = getopt.getopt(
             sys.argv[1:],
-            "xo:hvs:f:",
+            "axo:hvs:f:p",
             [
+                "all",
                 "extended",
                 "outfile",
                 "help",
                 "verbose",
                 "separator",
-                "columns"
+                "columns",
+                "parsable"
             ]
         )
     except getopt.error as msg:
@@ -293,6 +355,8 @@ def init():
         opts = None
 
     for opt, arg in opts:
+        if opt in ('-a', '--all'):
+            aflag = True
         if opt in ('-x', '--extended'):
             xflag = True
         if opt in ('-o', '--outfile'):
@@ -308,19 +372,13 @@ def init():
         if opt in ('-f', '--columns'):
             desired_cols = arg
             i += 1
+        if opt in ('-p', '--parsable'):
+            pretty_print = False
         i += 1
 
     argv = sys.argv[i:]
-    sint = Decimal(argv[0]) if argv else sint
+    sint = int(argv[0]) if argv else sint
-    count = int(argv[1]) if len(argv) > 1 else count
+    count = int(argv[1]) if len(argv) > 1 else (0 if len(argv) > 0 else 1)
-
-    if len(argv) > 1:
-        sint = Decimal(argv[0])
-        count = int(argv[1])
-
-    elif len(argv) > 0:
-        sint = Decimal(argv[0])
-        count = 0
 
     if hflag or (xflag and desired_cols):
         usage()
@@ -360,6 +418,12 @@ def init():
                              incompat)
             usage()
 
+    if aflag:
+        if l2exist:
+            hdr = cols.keys()
+        else:
+            hdr = [col for col in cols.keys() if not col.startswith("l2")]
+
     if opfile:
         try:
             out = open(opfile, "w")
@@ -377,59 +441,79 @@ def calculate():
 
     v = dict()
     v["time"] = time.strftime("%H:%M:%S", time.localtime())
-    v["hits"] = d["hits"] / sint
+    v["hits"] = d["hits"] // sint
-    v["miss"] = d["misses"] / sint
+    v["miss"] = d["misses"] // sint
     v["read"] = v["hits"] + v["miss"]
-    v["hit%"] = 100 * v["hits"] / v["read"] if v["read"] > 0 else 0
+    v["hit%"] = 100 * v["hits"] // v["read"] if v["read"] > 0 else 0
     v["miss%"] = 100 - v["hit%"] if v["read"] > 0 else 0
 
-    v["dhit"] = (d["demand_data_hits"] + d["demand_metadata_hits"]) / sint
+    v["dhit"] = (d["demand_data_hits"] + d["demand_metadata_hits"]) // sint
-    v["dmis"] = (d["demand_data_misses"] + d["demand_metadata_misses"]) / sint
+    v["dmis"] = (d["demand_data_misses"] + d["demand_metadata_misses"]) // sint
 
     v["dread"] = v["dhit"] + v["dmis"]
-    v["dh%"] = 100 * v["dhit"] / v["dread"] if v["dread"] > 0 else 0
+    v["dh%"] = 100 * v["dhit"] // v["dread"] if v["dread"] > 0 else 0
     v["dm%"] = 100 - v["dh%"] if v["dread"] > 0 else 0
 
-    v["phit"] = (d["prefetch_data_hits"] + d["prefetch_metadata_hits"]) / sint
+    v["phit"] = (d["prefetch_data_hits"] + d["prefetch_metadata_hits"]) // sint
     v["pmis"] = (d["prefetch_data_misses"] +
-                 d["prefetch_metadata_misses"]) / sint
+                 d["prefetch_metadata_misses"]) // sint
 
     v["pread"] = v["phit"] + v["pmis"]
-    v["ph%"] = 100 * v["phit"] / v["pread"] if v["pread"] > 0 else 0
+    v["ph%"] = 100 * v["phit"] // v["pread"] if v["pread"] > 0 else 0
     v["pm%"] = 100 - v["ph%"] if v["pread"] > 0 else 0
 
     v["mhit"] = (d["prefetch_metadata_hits"] +
-                 d["demand_metadata_hits"]) / sint
+                 d["demand_metadata_hits"]) // sint
     v["mmis"] = (d["prefetch_metadata_misses"] +
-                 d["demand_metadata_misses"]) / sint
+                 d["demand_metadata_misses"]) // sint
 
     v["mread"] = v["mhit"] + v["mmis"]
-    v["mh%"] = 100 * v["mhit"] / v["mread"] if v["mread"] > 0 else 0
+    v["mh%"] = 100 * v["mhit"] // v["mread"] if v["mread"] > 0 else 0
     v["mm%"] = 100 - v["mh%"] if v["mread"] > 0 else 0
 
     v["arcsz"] = cur["size"]
+    v["size"] = cur["size"]
     v["c"] = cur["c"]
-    v["mfu"] = d["mfu_hits"] / sint
+    v["mfu"] = d["mfu_hits"] // sint
-    v["mru"] = d["mru_hits"] / sint
+    v["mru"] = d["mru_hits"] // sint
-    v["mrug"] = d["mru_ghost_hits"] / sint
+    v["mrug"] = d["mru_ghost_hits"] // sint
-    v["mfug"] = d["mfu_ghost_hits"] / sint
+    v["mfug"] = d["mfu_ghost_hits"] // sint
-    v["eskip"] = d["evict_skip"] / sint
+    v["eskip"] = d["evict_skip"] // sint
-    v["mtxmis"] = d["mutex_miss"] / sint
+    v["el2skip"] = d["evict_l2_skip"] // sint
+    v["el2cach"] = d["evict_l2_cached"] // sint
+    v["el2el"] = d["evict_l2_eligible"] // sint
+    v["el2mfu"] = d["evict_l2_eligible_mfu"] // sint
+    v["el2mru"] = d["evict_l2_eligible_mru"] // sint
+    v["el2inel"] = d["evict_l2_ineligible"] // sint
+    v["mtxmis"] = d["mutex_miss"] // sint
 
     if l2exist:
-        v["l2hits"] = d["l2_hits"] / sint
+        v["l2hits"] = d["l2_hits"] // sint
-        v["l2miss"] = d["l2_misses"] / sint
+        v["l2miss"] = d["l2_misses"] // sint
         v["l2read"] = v["l2hits"] + v["l2miss"]
-        v["l2hit%"] = 100 * v["l2hits"] / v["l2read"] if v["l2read"] > 0 else 0
+        v["l2hit%"] = 100 * v["l2hits"] // v["l2read"] if v["l2read"] > 0 else 0
 
         v["l2miss%"] = 100 - v["l2hit%"] if v["l2read"] > 0 else 0
         v["l2asize"] = cur["l2_asize"]
         v["l2size"] = cur["l2_size"]
-        v["l2bytes"] = d["l2_read_bytes"] / sint
+        v["l2bytes"] = d["l2_read_bytes"] // sint
+
+        v["l2pref"] = cur["l2_prefetch_asize"]
+        v["l2mfu"] = cur["l2_mfu_asize"]
+        v["l2mru"] = cur["l2_mru_asize"]
+        v["l2data"] = cur["l2_bufc_data_asize"]
+        v["l2meta"] = cur["l2_bufc_metadata_asize"]
+        v["l2pref%"] = 100 * v["l2pref"] // v["l2asize"]
+        v["l2mfu%"] = 100 * v["l2mfu"] // v["l2asize"]
+        v["l2mru%"] = 100 * v["l2mru"] // v["l2asize"]
+        v["l2data%"] = 100 * v["l2data"] // v["l2asize"]
+        v["l2meta%"] = 100 * v["l2meta"] // v["l2asize"]
 
     v["grow"] = 0 if cur["arc_no_grow"] else 1
     v["need"] = cur["arc_need_free"]
-    v["free"] = cur["arc_sys_free"]
+    v["free"] = cur["memory_free_bytes"]
+    v["avail"] = cur["memory_available_bytes"]
+    v["waste"] = cur["abd_chunk_waste_size"]
 
 
 def main():

cmd/dbufstat/.gitignore

@@ -0,0 +1 @@
+dbufstat

cmd/dbufstat/Makefile.am

@@ -1,13 +1,5 @@
-dist_bin_SCRIPTS = dbufstat
+include $(top_srcdir)/config/Substfiles.am
 
-#
+bin_SCRIPTS = dbufstat
-# The dbufstat script is compatibile with both Python 2.6 and 3.4.
+
-# As such the python 3 shebang can be replaced at install time when
+SUBSTFILES += $(bin_SCRIPTS)
-# targeting a python 2 system.  This allows us to maintain a single
-# version of the source.
-#
-if USING_PYTHON_2
-install-exec-hook:
-	sed --in-place 's|^#!/usr/bin/python3|#!/usr/bin/python2|' \
-	    $(DESTDIR)$(bindir)/dbufstat
-endif

cmd/dbufstat/dbufstat.in

@@ -1,4 +1,4 @@
-#!/usr/bin/python3
+#!/usr/bin/env @PYTHON_SHEBANG@
 #
 # Print out statistics for all cached dmu buffers.  This information
 # is available through the dbufs kstat and may be post-processed as
@@ -113,10 +113,25 @@ cmd = ("Usage: dbufstat [-bdhnrtvx] [-i file] [-f fields] [-o file] "
 raw = 0
 
 
+if sys.platform.startswith("freebsd"):
+    import io
+    # Requires py-sysctl on FreeBSD
+    import sysctl
+
+    def default_ifile():
+        dbufs = sysctl.filter("kstat.zfs.misc.dbufs")[0].value
+        sys.stdin = io.StringIO(dbufs)
+        return "-"
+
+elif sys.platform.startswith("linux"):
+    def default_ifile():
+        return "/proc/spl/kstat/zfs/dbufs"
+
+
 def print_incompat_helper(incompat):
     cnt = 0
     for key in sorted(incompat):
-        if cnt is 0:
+        if cnt == 0:
             sys.stderr.write("\t")
         elif cnt > 8:
             sys.stderr.write(",\n\t")
@@ -343,7 +358,7 @@ def get_compstring(c):
                     "ZIO_COMPRESS_GZIP_6",  "ZIO_COMPRESS_GZIP_7",
                     "ZIO_COMPRESS_GZIP_8",  "ZIO_COMPRESS_GZIP_9",
                     "ZIO_COMPRESS_ZLE",     "ZIO_COMPRESS_LZ4",
-                    "ZIO_COMPRESS_FUNCTION"]
+                    "ZIO_COMPRESS_ZSTD",    "ZIO_COMPRESS_FUNCTION"]
 
     # If "-rr" option is used, don't convert to string representation
     if raw > 1:
@@ -645,9 +660,9 @@ def main():
             sys.exit(1)
 
     if not ifile:
-        ifile = '/proc/spl/kstat/zfs/dbufs'
+        ifile = default_ifile()
 
-    if ifile is not "-":
+    if ifile != "-":
         try:
             tmp = open(ifile, "r")
             sys.stdin = tmp

cmd/fsck_zfs/.gitignore

@@ -0,0 +1 @@
+/fsck.zfs

cmd/fsck_zfs/Makefile.am

@@ -1 +1,6 @@
+include $(top_srcdir)/config/Substfiles.am
+include $(top_srcdir)/config/Shellcheck.am
+
 dist_sbin_SCRIPTS = fsck.zfs
+
+SUBSTFILES += $(dist_sbin_SCRIPTS)

cmd/fsck_zfs/fsck.zfs

@@ -1,9 +0,0 @@
-#!/bin/sh
-#
-# fsck.zfs: A fsck helper to accomidate distributions that expect
-# to be able to execute a fsck on all filesystem types.  Currently
-# this script does nothing but it could be extended to act as a
-# compatibility wrapper for 'zpool scrub'.
-#
-
-exit 0

cmd/fsck_zfs/fsck.zfs.in

@@ -0,0 +1,44 @@
+#!/bin/sh
+#
+# fsck.zfs: A fsck helper to accommodate distributions that expect
+# to be able to execute a fsck on all filesystem types.
+#
+# This script simply bubbles up some already-known-about errors,
+# see fsck.zfs(8)
+#
+
+if [ "$#" = "0" ]; then
+	echo "Usage: $0 [options] dataset…" >&2
+	exit 16
+fi
+
+ret=0
+for dataset in "$@"; do
+	case "$dataset" in
+		-*)
+			continue
+			;;
+		*)
+			;;
+	esac
+
+	pool="${dataset%%/*}"
+
+	case "$(@sbindir@/zpool list -Ho health "$pool")" in
+		DEGRADED)
+			ret=$(( ret | 4 ))
+			;;
+		FAULTED)
+			awk '!/^([[:space:]]*#.*)?$/ && $1 == "'"$dataset"'" && $3 == "zfs" {exit 1}' /etc/fstab || \
+				ret=$(( ret | 8 ))
+			;;
+		"")
+			# Pool not found, error printed by zpool(8)
+			ret=$(( ret | 8 ))
+			;;
+		*)
+			;;
+	esac
+done
+
+exit "$ret"

cmd/mount_zfs/Makefile.am

@@ -1,9 +1,5 @@
 include $(top_srcdir)/config/Rules.am
 
-DEFAULT_INCLUDES += \
-	-I$(top_srcdir)/include \
-	-I$(top_srcdir)/lib/libspl/include
-
 #
 # Ignore the prefix for the mount helper.  It must be installed in /sbin/
 # because this path is hardcoded in the mount(8) for security reasons.
@@ -17,5 +13,10 @@ mount_zfs_SOURCES = \
 	mount_zfs.c
 
 mount_zfs_LDADD = \
-	$(top_builddir)/lib/libnvpair/libnvpair.la \
+	$(abs_top_builddir)/lib/libzfs/libzfs.la \
-	$(top_builddir)/lib/libzfs/libzfs.la
+	$(abs_top_builddir)/lib/libzfs_core/libzfs_core.la \
+	$(abs_top_builddir)/lib/libnvpair/libnvpair.la
+
+mount_zfs_LDADD += $(LTLIBINTL)
+
+include $(top_srcdir)/config/CppCheck.am

cmd/mount_zfs/mount_zfs.c

@@ -42,247 +42,46 @@
 
 libzfs_handle_t *g_zfs;
 
-typedef struct option_map {
-	const char *name;
-	unsigned long mntmask;
-	unsigned long zfsmask;
-} option_map_t;
-
-static const option_map_t option_map[] = {
-	/* Canonicalized filesystem independent options from mount(8) */
-	{ MNTOPT_NOAUTO,	MS_COMMENT,	ZS_COMMENT	},
-	{ MNTOPT_DEFAULTS,	MS_COMMENT,	ZS_COMMENT	},
-	{ MNTOPT_NODEVICES,	MS_NODEV,	ZS_COMMENT	},
-	{ MNTOPT_DIRSYNC,	MS_DIRSYNC,	ZS_COMMENT	},
-	{ MNTOPT_NOEXEC,	MS_NOEXEC,	ZS_COMMENT	},
-	{ MNTOPT_GROUP,		MS_GROUP,	ZS_COMMENT	},
-	{ MNTOPT_NETDEV,	MS_COMMENT,	ZS_COMMENT	},
-	{ MNTOPT_NOFAIL,	MS_COMMENT,	ZS_COMMENT	},
-	{ MNTOPT_NOSUID,	MS_NOSUID,	ZS_COMMENT	},
-	{ MNTOPT_OWNER,		MS_OWNER,	ZS_COMMENT	},
-	{ MNTOPT_REMOUNT,	MS_REMOUNT,	ZS_COMMENT	},
-	{ MNTOPT_RO,		MS_RDONLY,	ZS_COMMENT	},
-	{ MNTOPT_RW,		MS_COMMENT,	ZS_COMMENT	},
-	{ MNTOPT_SYNC,		MS_SYNCHRONOUS,	ZS_COMMENT	},
-	{ MNTOPT_USER,		MS_USERS,	ZS_COMMENT	},
-	{ MNTOPT_USERS,		MS_USERS,	ZS_COMMENT	},
-	/* acl flags passed with util-linux-2.24 mount command */
-	{ MNTOPT_ACL,		MS_POSIXACL,	ZS_COMMENT	},
-	{ MNTOPT_NOACL,		MS_COMMENT,	ZS_COMMENT	},
-	{ MNTOPT_POSIXACL,	MS_POSIXACL,	ZS_COMMENT	},
-#ifdef MS_NOATIME
-	{ MNTOPT_NOATIME,	MS_NOATIME,	ZS_COMMENT	},
-#endif
-#ifdef MS_NODIRATIME
-	{ MNTOPT_NODIRATIME,	MS_NODIRATIME,	ZS_COMMENT	},
-#endif
-#ifdef MS_RELATIME
-	{ MNTOPT_RELATIME,	MS_RELATIME,	ZS_COMMENT	},
-#endif
-#ifdef MS_STRICTATIME
-	{ MNTOPT_STRICTATIME,	MS_STRICTATIME,	ZS_COMMENT	},
-#endif
-#ifdef MS_LAZYTIME
-	{ MNTOPT_LAZYTIME,	MS_LAZYTIME,	ZS_COMMENT	},
-#endif
-	{ MNTOPT_CONTEXT,	MS_COMMENT,	ZS_COMMENT	},
-	{ MNTOPT_FSCONTEXT,	MS_COMMENT,	ZS_COMMENT	},
-	{ MNTOPT_DEFCONTEXT,	MS_COMMENT,	ZS_COMMENT	},
-	{ MNTOPT_ROOTCONTEXT,	MS_COMMENT,	ZS_COMMENT	},
-#ifdef MS_I_VERSION
-	{ MNTOPT_IVERSION,	MS_I_VERSION,	ZS_COMMENT	},
-#endif
-#ifdef MS_MANDLOCK
-	{ MNTOPT_NBMAND,	MS_MANDLOCK,	ZS_COMMENT	},
-#endif
-	/* Valid options not found in mount(8) */
-	{ MNTOPT_BIND,		MS_BIND,	ZS_COMMENT	},
-#ifdef MS_REC
-	{ MNTOPT_RBIND,		MS_BIND|MS_REC,	ZS_COMMENT	},
-#endif
-	{ MNTOPT_COMMENT,	MS_COMMENT,	ZS_COMMENT	},
-#ifdef MS_NOSUB
-	{ MNTOPT_NOSUB,		MS_NOSUB,	ZS_COMMENT	},
-#endif
-#ifdef MS_SILENT
-	{ MNTOPT_QUIET,		MS_SILENT,	ZS_COMMENT	},
-#endif
-	/* Custom zfs options */
-	{ MNTOPT_XATTR,		MS_COMMENT,	ZS_COMMENT	},
-	{ MNTOPT_NOXATTR,	MS_COMMENT,	ZS_COMMENT	},
-	{ MNTOPT_ZFSUTIL,	MS_COMMENT,	ZS_ZFSUTIL	},
-	{ NULL,			0,		0		} };
-
 /*
- * Break the mount option in to a name/value pair.  The name is
+ * Opportunistically convert a target string into a pool name. If the
- * validated against the option map and mount flags set accordingly.
+ * string does not represent a block device with a valid zfs label
+ * then it is passed through without modification.
  */
-static int
+static void
-parse_option(char *mntopt, unsigned long *mntflags,
+parse_dataset(const char *target, char **dataset)
-    unsigned long *zfsflags, int sloppy)
 {
-	const option_map_t *opt;
-	char *ptr, *name, *value = NULL;
-	int error = 0;
-
-	name = strdup(mntopt);
-	if (name == NULL)
-		return (ENOMEM);
-
-	for (ptr = name; ptr && *ptr; ptr++) {
-		if (*ptr == '=') {
-			*ptr = '\0';
-			value = ptr+1;
-			VERIFY3P(value, !=, NULL);
-			break;
-		}
-	}
-
-	for (opt = option_map; opt->name != NULL; opt++) {
-		if (strncmp(name, opt->name, strlen(name)) == 0) {
-			*mntflags |= opt->mntmask;
-			*zfsflags |= opt->zfsmask;
-			error = 0;
-			goto out;
-		}
-	}
-
-	if (!sloppy)
-		error = ENOENT;
-out:
-	/* If required further process on the value may be done here */
-	free(name);
-	return (error);
-}
-
-/*
- * Translate the mount option string in to MS_* mount flags for the
- * kernel vfs.  When sloppy is non-zero unknown options will be ignored
- * otherwise they are considered fatal are copied in to badopt.
- */
-static int
-parse_options(char *mntopts, unsigned long *mntflags, unsigned long *zfsflags,
-    int sloppy, char *badopt, char *mtabopt)
-{
-	int error = 0, quote = 0, flag = 0, count = 0;
-	char *ptr, *opt, *opts;
-
-	opts = strdup(mntopts);
-	if (opts == NULL)
-		return (ENOMEM);
-
-	*mntflags = 0;
-	opt = NULL;
-
 	/*
-	 * Scan through all mount options which must be comma delimited.
+	 * Prior to util-linux 2.36.2, if a file or directory in the
-	 * We must be careful to notice regions which are double quoted
+	 * current working directory was named 'dataset' then mount(8)
-	 * and skip commas in these regions.  Each option is then checked
+	 * would prepend the current working directory to the dataset.
-	 * to determine if it is a known option.
+	 * Check for it and strip the prepended path when it is added.
 	 */
-	for (ptr = opts; ptr && !flag; ptr++) {
-		if (opt == NULL)
-			opt = ptr;
-
-		if (*ptr == '"')
-			quote = !quote;
-
-		if (quote)
-			continue;
-
-		if (*ptr == '\0')
-			flag = 1;
-
-		if ((*ptr == ',') || (*ptr == '\0')) {
-			*ptr = '\0';
-
-			error = parse_option(opt, mntflags, zfsflags, sloppy);
-			if (error) {
-				strcpy(badopt, opt);
-				goto out;
-
-			}
-
-			if (!(*mntflags & MS_REMOUNT) &&
-			    !(*zfsflags & ZS_ZFSUTIL)) {
-				if (count > 0)
-					strlcat(mtabopt, ",", MNT_LINE_MAX);
-
-				strlcat(mtabopt, opt, MNT_LINE_MAX);
-				count++;
-			}
-
-			opt = NULL;
-		}
-	}
-
-out:
-	free(opts);
-	return (error);
-}
-
-/*
- * Return the pool/dataset to mount given the name passed to mount.  This
- * is expected to be of the form pool/dataset, however may also refer to
- * a block device if that device contains a valid zfs label.
- */
-static char *
-parse_dataset(char *dataset)
-{
 	char cwd[PATH_MAX];
-	struct stat64 statbuf;
+	if (getcwd(cwd, PATH_MAX) == NULL) {
-	int error;
+		perror("getcwd");
-	int len;
+		return;
-
-	/*
-	 * We expect a pool/dataset to be provided, however if we're
-	 * given a device which is a member of a zpool we attempt to
-	 * extract the pool name stored in the label.  Given the pool
-	 * name we can mount the root dataset.
-	 */
-	error = stat64(dataset, &statbuf);
-	if (error == 0) {
-		nvlist_t *config;
-		char *name;
-		int fd;
-
-		fd = open(dataset, O_RDONLY);
-		if (fd < 0)
-			goto out;
-
-		error = zpool_read_label(fd, &config, NULL);
-		(void) close(fd);
-		if (error)
-			goto out;
-
-		error = nvlist_lookup_string(config,
-		    ZPOOL_CONFIG_POOL_NAME, &name);
-		if (error) {
-			nvlist_free(config);
-		} else {
-			dataset = strdup(name);
-			nvlist_free(config);
-			return (dataset);
-		}
 	}
-out:
+	int len = strlen(cwd);
-	/*
+	if (strncmp(cwd, target, len) == 0)
-	 * If a file or directory in your current working directory is
+		target += len;
-	 * named 'dataset' then mount(8) will prepend your current working
-	 * directory to the dataset.  There is no way to prevent this
-	 * behavior so we simply check for it and strip the prepended
-	 * patch when it is added.
-	 */
-	if (getcwd(cwd, PATH_MAX) == NULL)
-		return (dataset);
 
-	len = strlen(cwd);
+	/* Assume pool/dataset is more likely */
+	strlcpy(*dataset, target, PATH_MAX);
 
-	/* Do not add one when cwd already ends in a trailing '/' */
+	int fd = open(target, O_RDONLY | O_CLOEXEC);
-	if (strncmp(cwd, dataset, len) == 0)
+	if (fd < 0)
-		return (dataset + len + (cwd[len-1] != '/'));
+		return;
 
-	return (dataset);
+	nvlist_t *cfg = NULL;
+	if (zpool_read_label(fd, &cfg, NULL) == 0) {
+		char *nm = NULL;
+		if (!nvlist_lookup_string(cfg, ZPOOL_CONFIG_POOL_NAME, &nm))
+			strlcpy(*dataset, nm, PATH_MAX);
+		nvlist_free(cfg);
+	}
+
+	if (close(fd))
+		perror("close");
 }
 
 /*
@@ -326,8 +125,8 @@ mtab_update(char *dataset, char *mntpoint, char *type, char *mntopts)
 	if (!fp) {
 		(void) fprintf(stderr, gettext(
 		    "filesystem '%s' was mounted, but /etc/mtab "
-		    "could not be opened due to error %d\n"),
+		    "could not be opened due to error: %s\n"),
-		    dataset, errno);
+		    dataset, strerror(errno));
 		return (MOUNT_FILEIO);
 	}
 
@@ -335,8 +134,8 @@ mtab_update(char *dataset, char *mntpoint, char *type, char *mntopts)
 	if (error) {
 		(void) fprintf(stderr, gettext(
 		    "filesystem '%s' was mounted, but /etc/mtab "
-		    "could not be updated due to error %d\n"),
+		    "could not be updated due to error: %s\n"),
-		    dataset, errno);
+		    dataset, strerror(errno));
 		return (MOUNT_FILEIO);
 	}
 
@@ -345,34 +144,6 @@ mtab_update(char *dataset, char *mntpoint, char *type, char *mntopts)
 	return (MOUNT_SUCCESS);
 }
 
-static void
-append_mntopt(const char *name, const char *val, char *mntopts,
-    char *mtabopt, boolean_t quote)
-{
-	char tmp[MNT_LINE_MAX];
-
-	snprintf(tmp, MNT_LINE_MAX, quote ? ",%s=\"%s\"" : ",%s=%s", name, val);
-
-	if (mntopts)
-		strlcat(mntopts, tmp, MNT_LINE_MAX);
-
-	if (mtabopt)
-		strlcat(mtabopt, tmp, MNT_LINE_MAX);
-}
-
-static void
-zfs_selinux_setcontext(zfs_handle_t *zhp, zfs_prop_t zpt, const char *name,
-    char *mntopts, char *mtabopt)
-{
-	char context[ZFS_MAXPROPLEN];
-
-	if (zfs_prop_get(zhp, zpt, context, sizeof (context),
-	    NULL, NULL, 0, B_FALSE) == 0) {
-		if (strcmp(context, "none") != 0)
-			append_mntopt(name, context, mntopts, mtabopt, B_TRUE);
-	}
-}
-
 int
 main(int argc, char **argv)
 {
@@ -383,12 +154,13 @@ main(int argc, char **argv)
 	char badopt[MNT_LINE_MAX] = { '\0' };
 	char mtabopt[MNT_LINE_MAX] = { '\0' };
 	char mntpoint[PATH_MAX];
-	char *dataset;
+	char dataset[PATH_MAX], *pdataset = dataset;
 	unsigned long mntflags = 0, zfsflags = 0, remount = 0;
 	int sloppy = 0, fake = 0, verbose = 0, nomtab = 0, zfsutil = 0;
 	int error, c;
 
 	(void) setlocale(LC_ALL, "");
+	(void) setlocale(LC_NUMERIC, "C");
 	(void) textdomain(TEXT_DOMAIN);
 
 	opterr = 0;
@@ -413,10 +185,11 @@ main(int argc, char **argv)
 			break;
 		case 'h':
 		case '?':
-			(void) fprintf(stderr, gettext("Invalid option '%c'\n"),
+			if (optopt)
-			    optopt);
+				(void) fprintf(stderr,
+				    gettext("Invalid option '%c'\n"), optopt);
 			(void) fprintf(stderr, gettext("Usage: mount.zfs "
-			    "[-sfnv] [-o options] <dataset> <mountpoint>\n"));
+			    "[-sfnvh] [-o options] <dataset> <mountpoint>\n"));
 			return (MOUNT_USAGE);
 		}
 	}
@@ -438,18 +211,18 @@ main(int argc, char **argv)
 		return (MOUNT_USAGE);
 	}
 
-	dataset = parse_dataset(argv[0]);
+	parse_dataset(argv[0], &pdataset);
 
 	/* canonicalize the mount point */
 	if (realpath(argv[1], mntpoint) == NULL) {
 		(void) fprintf(stderr, gettext("filesystem '%s' cannot be "
-		    "mounted at '%s' due to canonicalization error %d.\n"),
+		    "mounted at '%s' due to canonicalization error: %s\n"),
-		    dataset, argv[1], errno);
+		    dataset, argv[1], strerror(errno));
 		return (MOUNT_SYSERR);
 	}
 
 	/* validate mount options and set mntflags */
-	error = parse_options(mntopts, &mntflags, &zfsflags, sloppy,
+	error = zfs_parse_mount_options(mntopts, &mntflags, &zfsflags, sloppy,
 	    badopt, mtabopt);
 	if (error) {
 		switch (error) {
@@ -489,7 +262,7 @@ main(int argc, char **argv)
 		zfsutil = 1;
 
 	if ((g_zfs = libzfs_init()) == NULL) {
-		(void) fprintf(stderr, "%s", libzfs_error_init(errno));
+		(void) fprintf(stderr, "%s\n", libzfs_error_init(errno));
 		return (MOUNT_SYSERR);
 	}
 
@@ -502,32 +275,7 @@ main(int argc, char **argv)
 		return (MOUNT_USAGE);
 	}
 
-	/*
+	zfs_adjust_mount_options(zhp, mntpoint, mntopts, mtabopt);
-	 * Checks to see if the ZFS_PROP_SELINUX_CONTEXT exists
-	 * if it does, create a tmp variable in case it's needed
-	 * checks to see if the selinux context is set to the default
-	 * if it is, allow the setting of the other context properties
-	 * this is needed because the 'context' property overrides others
-	 * if it is not the default, set the 'context' property
-	 */
-	if (zfs_prop_get(zhp, ZFS_PROP_SELINUX_CONTEXT, prop, sizeof (prop),
-	    NULL, NULL, 0, B_FALSE) == 0) {
-		if (strcmp(prop, "none") == 0) {
-			zfs_selinux_setcontext(zhp, ZFS_PROP_SELINUX_FSCONTEXT,
-			    MNTOPT_FSCONTEXT, mntopts, mtabopt);
-			zfs_selinux_setcontext(zhp, ZFS_PROP_SELINUX_DEFCONTEXT,
-			    MNTOPT_DEFCONTEXT, mntopts, mtabopt);
-			zfs_selinux_setcontext(zhp,
-			    ZFS_PROP_SELINUX_ROOTCONTEXT, MNTOPT_ROOTCONTEXT,
-			    mntopts, mtabopt);
-		} else {
-			append_mntopt(MNTOPT_CONTEXT, prop,
-			    mntopts, mtabopt, B_TRUE);
-		}
-	}
-
-	/* A hint used to determine an auto-mounted snapshot mount point */
-	append_mntopt(MNTOPT_MNTPOINT, mntpoint, mntopts, NULL, B_FALSE);
 
 	/* treat all snapshots as legacy mount points */
 	if (zfs_get_type(zhp) == ZFS_TYPE_SNAPSHOT)
@@ -620,8 +368,8 @@ main(int argc, char **argv)
 				    "mount the filesystem again.\n"), dataset);
 				return (MOUNT_SYSERR);
 			}
-			/* fallthru */
 #endif
+			fallthrough;
 		default:
 			(void) fprintf(stderr, gettext("filesystem "
 			    "'%s' can not be mounted: %s\n"), dataset,

cmd/raidz_test/Makefile.am

@@ -4,11 +4,7 @@ include $(top_srcdir)/config/Rules.am
 AM_CFLAGS += $(FRAME_LARGER_THAN)
 
 # Unconditionally enable ASSERTs
-AM_CPPFLAGS += -DDEBUG -UNDEBUG
+AM_CPPFLAGS += -DDEBUG -UNDEBUG -DZFS_DEBUG
-
-DEFAULT_INCLUDES += \
-	-I$(top_srcdir)/include \
-	-I$(top_srcdir)/lib/libspl/include
 
 bin_PROGRAMS = raidz_test
 
@@ -18,6 +14,9 @@ raidz_test_SOURCES = \
 	raidz_bench.c
 
 raidz_test_LDADD = \
-	$(top_builddir)/lib/libzpool/libzpool.la
+	$(abs_top_builddir)/lib/libzpool/libzpool.la \
+	$(abs_top_builddir)/lib/libzfs_core/libzfs_core.la
 
-raidz_test_LDADD += -lm -ldl
+raidz_test_LDADD += -lm
+
+include $(top_srcdir)/config/CppCheck.am

cmd/raidz_test/raidz_bench.c

@@ -31,8 +31,6 @@
 #include <sys/vdev_raidz_impl.h>
 #include <stdio.h>
 
-#include <sys/time.h>
-
 #include "raidz_test.h"
 
 #define	GEN_BENCH_MEMORY	(((uint64_t)1ULL)<<32)
@@ -83,8 +81,17 @@ run_gen_bench_impl(const char *impl)
 			/* create suitable raidz_map */
 			ncols = rto_opts.rto_dcols + fn + 1;
 			zio_bench.io_size = 1ULL << ds;
-			rm_bench = vdev_raidz_map_alloc(&zio_bench,
+
-			    BENCH_ASHIFT, ncols, fn+1);
+			if (rto_opts.rto_expand) {
+				rm_bench = vdev_raidz_map_alloc_expanded(
+				    zio_bench.io_abd,
+				    zio_bench.io_size, zio_bench.io_offset,
+				    rto_opts.rto_ashift, ncols+1, ncols,
+				    fn+1, rto_opts.rto_expand_offset);
+			} else {
+				rm_bench = vdev_raidz_map_alloc(&zio_bench,
+				    BENCH_ASHIFT, ncols, fn+1);
+			}
 
 			/* estimate iteration count */
 			iter_cnt = GEN_BENCH_MEMORY;
@@ -113,7 +120,7 @@ run_gen_bench_impl(const char *impl)
 	}
 }
 
-void
+static void
 run_gen_bench(void)
 {
 	char **impl_name;
@@ -163,8 +170,16 @@ run_rec_bench_impl(const char *impl)
 			    (1ULL << BENCH_ASHIFT))
 				continue;
 
-			rm_bench = vdev_raidz_map_alloc(&zio_bench,
+			if (rto_opts.rto_expand) {
-			    BENCH_ASHIFT, ncols, PARITY_PQR);
+				rm_bench = vdev_raidz_map_alloc_expanded(
+				    zio_bench.io_abd,
+				    zio_bench.io_size, zio_bench.io_offset,
+				    BENCH_ASHIFT, ncols+1, ncols,
+				    PARITY_PQR, rto_opts.rto_expand_offset);
+			} else {
+				rm_bench = vdev_raidz_map_alloc(&zio_bench,
+				    BENCH_ASHIFT, ncols, PARITY_PQR);
+			}
 
 			/* estimate iteration count */
 			iter_cnt = (REC_BENCH_MEMORY);
@@ -197,7 +212,7 @@ run_rec_bench_impl(const char *impl)
 	}
 }
 
-void
+static void
 run_rec_bench(void)
 {
 	char **impl_name;

cmd/raidz_test/raidz_test.c

@@ -37,11 +37,11 @@
 static int *rand_data;
 raidz_test_opts_t rto_opts;
 
-static char gdb[256];
+static char pid_s[16];
-static const char gdb_tmpl[] = "gdb -ex \"set pagination 0\" -p %d";
 
 static void sig_handler(int signo)
 {
+	int old_errno = errno;
 	struct sigaction action;
 	/*
 	 * Restore default action and re-raise signal so SIGSEGV and
@@ -52,10 +52,19 @@ static void sig_handler(int signo)
 	action.sa_flags = 0;
 	(void) sigaction(signo, &action, NULL);
 
-	if (rto_opts.rto_gdb)
+	if (rto_opts.rto_gdb) {
-		if (system(gdb)) { }
+		pid_t pid = fork();
+		if (pid == 0) {
+			execlp("gdb", "gdb", "-ex", "set pagination 0",
+			    "-p", pid_s, NULL);
+			_exit(-1);
+		} else if (pid > 0)
+			while (waitpid(pid, NULL, 0) == -1 && errno == EINTR)
+				;
+	}
 
 	raise(signo);
+	errno = old_errno;
 }
 
 static void print_opts(raidz_test_opts_t *opts, boolean_t force)
@@ -77,16 +86,20 @@ static void print_opts(raidz_test_opts_t *opts, boolean_t force)
 		(void) fprintf(stdout, DBLSEP "Running with options:\n"
 		    "  (-a) zio ashift                   : %zu\n"
 		    "  (-o) zio offset                   : 1 << %zu\n"
+		    "  (-e) expanded map                 : %s\n"
+		    "  (-r) reflow offset                : %llx\n"
 		    "  (-d) number of raidz data columns : %zu\n"
 		    "  (-s) size of DATA                 : 1 << %zu\n"
 		    "  (-S) sweep parameters             : %s \n"
 		    "  (-v) verbose                      : %s \n\n",
-		    opts->rto_ashift,			/* -a */
+		    opts->rto_ashift,				/* -a */
-		    ilog2(opts->rto_offset),		/* -o */
+		    ilog2(opts->rto_offset),			/* -o */
-		    opts->rto_dcols,			/* -d */
+		    opts->rto_expand ? "yes" : "no",		/* -e */
-		    ilog2(opts->rto_dsize),		/* -s */
+		    (u_longlong_t)opts->rto_expand_offset,	/* -r */
-		    opts->rto_sweep ? "yes" : "no",	/* -S */
+		    opts->rto_dcols,				/* -d */
-		    verbose);				/* -v */
+		    ilog2(opts->rto_dsize),			/* -s */
+		    opts->rto_sweep ? "yes" : "no",		/* -S */
+		    verbose);					/* -v */
 	}
 }
 
@@ -104,6 +117,8 @@ static void usage(boolean_t requested)
 	    "\t[-S parameter sweep (default: %s)]\n"
 	    "\t[-t timeout for parameter sweep test]\n"
 	    "\t[-B benchmark all raidz implementations]\n"
+	    "\t[-e use expanded raidz map (default: %s)]\n"
+	    "\t[-r expanded raidz map reflow offset (default: %llx)]\n"
 	    "\t[-v increase verbosity (default: %zu)]\n"
 	    "\t[-h (print help)]\n"
 	    "\t[-T test the test, see if failure would be detected]\n"
@@ -114,6 +129,8 @@ static void usage(boolean_t requested)
 	    o->rto_dcols,				/* -d */
 	    ilog2(o->rto_dsize),			/* -s */
 	    rto_opts.rto_sweep ? "yes" : "no",		/* -S */
+	    rto_opts.rto_expand ? "yes" : "no",		/* -e */
+	    (u_longlong_t)o->rto_expand_offset,		/* -r */
 	    o->rto_v);					/* -d */
 
 	exit(requested ? 0 : 1);
@@ -128,7 +145,7 @@ static void process_options(int argc, char **argv)
 
 	bcopy(&rto_opts_defaults, o, sizeof (*o));
 
-	while ((opt = getopt(argc, argv, "TDBSvha:o:d:s:t:")) != -1) {
+	while ((opt = getopt(argc, argv, "TDBSvha:er:o:d:s:t:")) != -1) {
 		value = 0;
 
 		switch (opt) {
@@ -136,6 +153,12 @@ static void process_options(int argc, char **argv)
 			value = strtoull(optarg, NULL, 0);
 			o->rto_ashift = MIN(13, MAX(9, value));
 			break;
+		case 'e':
+			o->rto_expand = 1;
+			break;
+		case 'r':
+			o->rto_expand_offset = strtoull(optarg, NULL, 0);
+			break;
 		case 'o':
 			value = strtoull(optarg, NULL, 0);
 			o->rto_offset = ((1ULL << MIN(12, value)) >> 9) << 9;
@@ -179,25 +202,34 @@ static void process_options(int argc, char **argv)
 	}
 }
 
-#define	DATA_COL(rm, i) ((rm)->rm_col[raidz_parity(rm) + (i)].rc_abd)
+#define	DATA_COL(rr, i) ((rr)->rr_col[rr->rr_firstdatacol + (i)].rc_abd)
-#define	DATA_COL_SIZE(rm, i) ((rm)->rm_col[raidz_parity(rm) + (i)].rc_size)
+#define	DATA_COL_SIZE(rr, i) ((rr)->rr_col[rr->rr_firstdatacol + (i)].rc_size)
 
-#define	CODE_COL(rm, i) ((rm)->rm_col[(i)].rc_abd)
+#define	CODE_COL(rr, i) ((rr)->rr_col[(i)].rc_abd)
-#define	CODE_COL_SIZE(rm, i) ((rm)->rm_col[(i)].rc_size)
+#define	CODE_COL_SIZE(rr, i) ((rr)->rr_col[(i)].rc_size)
 
 static int
 cmp_code(raidz_test_opts_t *opts, const raidz_map_t *rm, const int parity)
 {
-	int i, ret = 0;
+	int r, i, ret = 0;
 
 	VERIFY(parity >= 1 && parity <= 3);
 
-	for (i = 0; i < parity; i++) {
+	for (r = 0; r < rm->rm_nrows; r++) {
-		if (abd_cmp(CODE_COL(rm, i), CODE_COL(opts->rm_golden, i))
+		raidz_row_t * const rr = rm->rm_row[r];
-		    != 0) {
+		raidz_row_t * const rrg = opts->rm_golden->rm_row[r];
-			ret++;
+		for (i = 0; i < parity; i++) {
-			LOG_OPT(D_DEBUG, opts,
+			if (CODE_COL_SIZE(rrg, i) == 0) {
-			    "\nParity block [%d] different!\n", i);
+				VERIFY0(CODE_COL_SIZE(rr, i));
+				continue;
+			}
+
+			if (abd_cmp(CODE_COL(rr, i),
+			    CODE_COL(rrg, i)) != 0) {
+				ret++;
+				LOG_OPT(D_DEBUG, opts,
+				    "\nParity block [%d] different!\n", i);
+			}
 		}
 	}
 	return (ret);
@@ -206,16 +238,26 @@ cmp_code(raidz_test_opts_t *opts, const raidz_map_t *rm, const int parity)
 static int
 cmp_data(raidz_test_opts_t *opts, raidz_map_t *rm)
 {
-	int i, ret = 0;
+	int r, i, dcols, ret = 0;
-	int dcols = opts->rm_golden->rm_cols - raidz_parity(opts->rm_golden);
 
-	for (i = 0; i < dcols; i++) {
+	for (r = 0; r < rm->rm_nrows; r++) {
-		if (abd_cmp(DATA_COL(opts->rm_golden, i), DATA_COL(rm, i))
+		raidz_row_t *rr = rm->rm_row[r];
-		    != 0) {
+		raidz_row_t *rrg = opts->rm_golden->rm_row[r];
-			ret++;
+		dcols = opts->rm_golden->rm_row[0]->rr_cols -
+		    raidz_parity(opts->rm_golden);
+		for (i = 0; i < dcols; i++) {
+			if (DATA_COL_SIZE(rrg, i) == 0) {
+				VERIFY0(DATA_COL_SIZE(rr, i));
+				continue;
+			}
 
-			LOG_OPT(D_DEBUG, opts,
+			if (abd_cmp(DATA_COL(rrg, i),
-			    "\nData block [%d] different!\n", i);
+			    DATA_COL(rr, i)) != 0) {
+				ret++;
+
+				LOG_OPT(D_DEBUG, opts,
+				    "\nData block [%d] different!\n", i);
+			}
 		}
 	}
 	return (ret);
@@ -236,12 +278,13 @@ init_rand(void *data, size_t size, void *private)
 static void
 corrupt_colums(raidz_map_t *rm, const int *tgts, const int cnt)
 {
-	int i;
+	for (int r = 0; r < rm->rm_nrows; r++) {
-	raidz_col_t *col;
+		raidz_row_t *rr = rm->rm_row[r];
-
+		for (int i = 0; i < cnt; i++) {
-	for (i = 0; i < cnt; i++) {
+			raidz_col_t *col = &rr->rr_col[tgts[i]];
-		col = &rm->rm_col[tgts[i]];
+			abd_iterate_func(col->rc_abd, 0, col->rc_size,
-		abd_iterate_func(col->rc_abd, 0, col->rc_size, init_rand, NULL);
+			    init_rand, NULL);
+		}
 	}
 }
 
@@ -288,10 +331,22 @@ init_raidz_golden_map(raidz_test_opts_t *opts, const int parity)
 
 	VERIFY0(vdev_raidz_impl_set("original"));
 
-	opts->rm_golden = vdev_raidz_map_alloc(opts->zio_golden,
+	if (opts->rto_expand) {
-	    opts->rto_ashift, total_ncols, parity);
+		opts->rm_golden =
-	rm_test = vdev_raidz_map_alloc(zio_test,
+		    vdev_raidz_map_alloc_expanded(opts->zio_golden->io_abd,
-	    opts->rto_ashift, total_ncols, parity);
+		    opts->zio_golden->io_size, opts->zio_golden->io_offset,
+		    opts->rto_ashift, total_ncols+1, total_ncols,
+		    parity, opts->rto_expand_offset);
+		rm_test = vdev_raidz_map_alloc_expanded(zio_test->io_abd,
+		    zio_test->io_size, zio_test->io_offset,
+		    opts->rto_ashift, total_ncols+1, total_ncols,
+		    parity, opts->rto_expand_offset);
+	} else {
+		opts->rm_golden = vdev_raidz_map_alloc(opts->zio_golden,
+		    opts->rto_ashift, total_ncols, parity);
+		rm_test = vdev_raidz_map_alloc(zio_test,
+		    opts->rto_ashift, total_ncols, parity);
+	}
 
 	VERIFY(opts->zio_golden);
 	VERIFY(opts->rm_golden);
@@ -312,6 +367,187 @@ init_raidz_golden_map(raidz_test_opts_t *opts, const int parity)
 	return (err);
 }
 
+/*
+ * If reflow is not in progress, reflow_offset should be UINT64_MAX.
+ * For each row, if the row is entirely before reflow_offset, it will
+ * come from the new location.  Otherwise this row will come from the
+ * old location.  Therefore, rows that straddle the reflow_offset will
+ * come from the old location.
+ *
+ * NOTE: Until raidz expansion is implemented this function is only
+ * needed by raidz_test.c to the multi-row raid_map_t functionality.
+ */
+raidz_map_t *
+vdev_raidz_map_alloc_expanded(abd_t *abd, uint64_t size, uint64_t offset,
+    uint64_t ashift, uint64_t physical_cols, uint64_t logical_cols,
+    uint64_t nparity, uint64_t reflow_offset)
+{
+	/* The zio's size in units of the vdev's minimum sector size. */
+	uint64_t s = size >> ashift;
+	uint64_t q, r, bc, devidx, asize = 0, tot;
+
+	/*
+	 * "Quotient": The number of data sectors for this stripe on all but
+	 * the "big column" child vdevs that also contain "remainder" data.
+	 * AKA "full rows"
+	 */
+	q = s / (logical_cols - nparity);
+
+	/*
+	 * "Remainder": The number of partial stripe data sectors in this I/O.
+	 * This will add a sector to some, but not all, child vdevs.
+	 */
+	r = s - q * (logical_cols - nparity);
+
+	/* The number of "big columns" - those which contain remainder data. */
+	bc = (r == 0 ? 0 : r + nparity);
+
+	/*
+	 * The total number of data and parity sectors associated with
+	 * this I/O.
+	 */
+	tot = s + nparity * (q + (r == 0 ? 0 : 1));
+
+	/* How many rows contain data (not skip) */
+	uint64_t rows = howmany(tot, logical_cols);
+	int cols = MIN(tot, logical_cols);
+
+	raidz_map_t *rm = kmem_zalloc(offsetof(raidz_map_t, rm_row[rows]),
+	    KM_SLEEP);
+	rm->rm_nrows = rows;
+
+	for (uint64_t row = 0; row < rows; row++) {
+		raidz_row_t *rr = kmem_alloc(offsetof(raidz_row_t,
+		    rr_col[cols]), KM_SLEEP);
+		rm->rm_row[row] = rr;
+
+		/* The starting RAIDZ (parent) vdev sector of the row. */
+		uint64_t b = (offset >> ashift) + row * logical_cols;
+
+		/*
+		 * If we are in the middle of a reflow, and any part of this
+		 * row has not been copied, then use the old location of
+		 * this row.
+		 */
+		int row_phys_cols = physical_cols;
+		if (b + (logical_cols - nparity) > reflow_offset >> ashift)
+			row_phys_cols--;
+
+		/* starting child of this row */
+		uint64_t child_id = b % row_phys_cols;
+		/* The starting byte offset on each child vdev. */
+		uint64_t child_offset = (b / row_phys_cols) << ashift;
+
+		/*
+		 * We set cols to the entire width of the block, even
+		 * if this row is shorter.  This is needed because parity
+		 * generation (for Q and R) needs to know the entire width,
+		 * because it treats the short row as though it was
+		 * full-width (and the "phantom" sectors were zero-filled).
+		 *
+		 * Another approach to this would be to set cols shorter
+		 * (to just the number of columns that we might do i/o to)
+		 * and have another mechanism to tell the parity generation
+		 * about the "entire width".  Reconstruction (at least
+		 * vdev_raidz_reconstruct_general()) would also need to
+		 * know about the "entire width".
+		 */
+		rr->rr_cols = cols;
+		rr->rr_bigcols = bc;
+		rr->rr_missingdata = 0;
+		rr->rr_missingparity = 0;
+		rr->rr_firstdatacol = nparity;
+		rr->rr_abd_empty = NULL;
+		rr->rr_nempty = 0;
+
+		for (int c = 0; c < rr->rr_cols; c++, child_id++) {
+			if (child_id >= row_phys_cols) {
+				child_id -= row_phys_cols;
+				child_offset += 1ULL << ashift;
+			}
+			rr->rr_col[c].rc_devidx = child_id;
+			rr->rr_col[c].rc_offset = child_offset;
+			rr->rr_col[c].rc_orig_data = NULL;
+			rr->rr_col[c].rc_error = 0;
+			rr->rr_col[c].rc_tried = 0;
+			rr->rr_col[c].rc_skipped = 0;
+			rr->rr_col[c].rc_need_orig_restore = B_FALSE;
+
+			uint64_t dc = c - rr->rr_firstdatacol;
+			if (c < rr->rr_firstdatacol) {
+				rr->rr_col[c].rc_size = 1ULL << ashift;
+				rr->rr_col[c].rc_abd =
+				    abd_alloc_linear(rr->rr_col[c].rc_size,
+				    B_TRUE);
+			} else if (row == rows - 1 && bc != 0 && c >= bc) {
+				/*
+				 * Past the end, this for parity generation.
+				 */
+				rr->rr_col[c].rc_size = 0;
+				rr->rr_col[c].rc_abd = NULL;
+			} else {
+				/*
+				 * "data column" (col excluding parity)
+				 * Add an ASCII art diagram here
+				 */
+				uint64_t off;
+
+				if (c < bc || r == 0) {
+					off = dc * rows + row;
+				} else {
+					off = r * rows +
+					    (dc - r) * (rows - 1) + row;
+				}
+				rr->rr_col[c].rc_size = 1ULL << ashift;
+				rr->rr_col[c].rc_abd = abd_get_offset_struct(
+				    &rr->rr_col[c].rc_abdstruct,
+				    abd, off << ashift, 1 << ashift);
+			}
+
+			asize += rr->rr_col[c].rc_size;
+		}
+		/*
+		 * If all data stored spans all columns, there's a danger that
+		 * parity will always be on the same device and, since parity
+		 * isn't read during normal operation, that that device's I/O
+		 * bandwidth won't be used effectively. We therefore switch
+		 * the parity every 1MB.
+		 *
+		 * ...at least that was, ostensibly, the theory. As a practical
+		 * matter unless we juggle the parity between all devices
+		 * evenly, we won't see any benefit. Further, occasional writes
+		 * that aren't a multiple of the LCM of the number of children
+		 * and the minimum stripe width are sufficient to avoid pessimal
+		 * behavior. Unfortunately, this decision created an implicit
+		 * on-disk format requirement that we need to support for all
+		 * eternity, but only for single-parity RAID-Z.
+		 *
+		 * If we intend to skip a sector in the zeroth column for
+		 * padding we must make sure to note this swap. We will never
+		 * intend to skip the first column since at least one data and
+		 * one parity column must appear in each row.
+		 */
+		if (rr->rr_firstdatacol == 1 && rr->rr_cols > 1 &&
+		    (offset & (1ULL << 20))) {
+			ASSERT(rr->rr_cols >= 2);
+			ASSERT(rr->rr_col[0].rc_size == rr->rr_col[1].rc_size);
+			devidx = rr->rr_col[0].rc_devidx;
+			uint64_t o = rr->rr_col[0].rc_offset;
+			rr->rr_col[0].rc_devidx = rr->rr_col[1].rc_devidx;
+			rr->rr_col[0].rc_offset = rr->rr_col[1].rc_offset;
+			rr->rr_col[1].rc_devidx = devidx;
+			rr->rr_col[1].rc_offset = o;
+		}
+
+	}
+	ASSERT3U(asize, ==, tot << ashift);
+
+	/* init RAIDZ parity ops */
+	rm->rm_ops = vdev_raidz_math_get_ops();
+
+	return (rm);
+}
+
 static raidz_map_t *
 init_raidz_map(raidz_test_opts_t *opts, zio_t **zio, const int parity)
 {
@@ -330,8 +566,15 @@ init_raidz_map(raidz_test_opts_t *opts, zio_t **zio, const int parity)
 	(*zio)->io_abd = raidz_alloc(alloc_dsize);
 	init_zio_abd(*zio);
 
-	rm = vdev_raidz_map_alloc(*zio, opts->rto_ashift,
+	if (opts->rto_expand) {
-	    total_ncols, parity);
+		rm = vdev_raidz_map_alloc_expanded((*zio)->io_abd,
+		    (*zio)->io_size, (*zio)->io_offset,
+		    opts->rto_ashift, total_ncols+1, total_ncols,
+		    parity, opts->rto_expand_offset);
+	} else {
+		rm = vdev_raidz_map_alloc(*zio, opts->rto_ashift,
+		    total_ncols, parity);
+	}
 	VERIFY(rm);
 
 	/* Make sure code columns are destroyed */
@@ -420,7 +663,7 @@ run_rec_check_impl(raidz_test_opts_t *opts, raidz_map_t *rm, const int fn)
 	if (fn < RAIDZ_REC_PQ) {
 		/* can reconstruct 1 failed data disk */
 		for (x0 = 0; x0 < opts->rto_dcols; x0++) {
-			if (x0 >= rm->rm_cols - raidz_parity(rm))
+			if (x0 >= rm->rm_row[0]->rr_cols - raidz_parity(rm))
 				continue;
 
 			/* Check if should stop */
@@ -445,10 +688,11 @@ run_rec_check_impl(raidz_test_opts_t *opts, raidz_map_t *rm, const int fn)
 	} else if (fn < RAIDZ_REC_PQR) {
 		/* can reconstruct 2 failed data disk */
 		for (x0 = 0; x0 < opts->rto_dcols; x0++) {
-			if (x0 >= rm->rm_cols - raidz_parity(rm))
+			if (x0 >= rm->rm_row[0]->rr_cols - raidz_parity(rm))
 				continue;
 			for (x1 = x0 + 1; x1 < opts->rto_dcols; x1++) {
-				if (x1 >= rm->rm_cols - raidz_parity(rm))
+				if (x1 >= rm->rm_row[0]->rr_cols -
+				    raidz_parity(rm))
 					continue;
 
 				/* Check if should stop */
@@ -475,14 +719,15 @@ run_rec_check_impl(raidz_test_opts_t *opts, raidz_map_t *rm, const int fn)
 	} else {
 		/* can reconstruct 3 failed data disk */
 		for (x0 = 0; x0 < opts->rto_dcols; x0++) {
-			if (x0 >= rm->rm_cols - raidz_parity(rm))
+			if (x0 >= rm->rm_row[0]->rr_cols - raidz_parity(rm))
 				continue;
 			for (x1 = x0 + 1; x1 < opts->rto_dcols; x1++) {
-				if (x1 >= rm->rm_cols - raidz_parity(rm))
+				if (x1 >= rm->rm_row[0]->rr_cols -
+				    raidz_parity(rm))
 					continue;
 				for (x2 = x1 + 1; x2 < opts->rto_dcols; x2++) {
-					if (x2 >=
+					if (x2 >= rm->rm_row[0]->rr_cols -
-					    rm->rm_cols - raidz_parity(rm))
+					    raidz_parity(rm))
 						continue;
 
 					/* Check if should stop */
@@ -700,6 +945,8 @@ run_sweep(void)
 		opts->rto_dcols = dcols_v[d];
 		opts->rto_offset = (1 << ashift_v[a]) * rand();
 		opts->rto_dsize = size_v[s];
+		opts->rto_expand = rto_opts.rto_expand;
+		opts->rto_expand_offset = rto_opts.rto_expand_offset;
 		opts->rto_v = 0; /* be quiet */
 
 		VERIFY3P(thread_create(NULL, 0, sweep_thread, (void *) opts,
@@ -732,6 +979,7 @@ exit:
 	return (sweep_state == SWEEP_ERROR ? SWEEP_ERROR : 0);
 }
 
+
 int
 main(int argc, char **argv)
 {
@@ -739,8 +987,8 @@ main(int argc, char **argv)
 	struct sigaction action;
 	int err = 0;
 
-	/* init gdb string early */
+	/* init gdb pid string early */
-	(void) sprintf(gdb, gdb_tmpl, getpid());
+	(void) sprintf(pid_s, "%d", getpid());
 
 	action.sa_handler = sig_handler;
 	sigemptyset(&action.sa_mask);
@@ -757,7 +1005,7 @@ main(int argc, char **argv)
 
 	process_options(argc, argv);
 
-	kernel_init(FREAD);
+	kernel_init(SPA_MODE_READ);
 
 	/* setup random data because rand() is not reentrant */
 	rand_data = (int *)umem_alloc(SPA_MAXBLOCKSIZE, UMEM_NOFAIL);

cmd/raidz_test/raidz_test.h

@@ -38,18 +38,21 @@ static const char *raidz_impl_names[] = {
 	"avx512bw",
 	"aarch64_neon",
 	"aarch64_neonx2",
+	"powerpc_altivec",
 	NULL
 };
 
 typedef struct raidz_test_opts {
 	size_t rto_ashift;
-	size_t rto_offset;
+	uint64_t rto_offset;
 	size_t rto_dcols;
 	size_t rto_dsize;
 	size_t rto_v;
 	size_t rto_sweep;
 	size_t rto_sweep_timeout;
 	size_t rto_benchmark;
+	size_t rto_expand;
+	uint64_t rto_expand_offset;
 	size_t rto_sanity;
 	size_t rto_gdb;
 
@@ -68,6 +71,8 @@ static const raidz_test_opts_t rto_opts_defaults = {
 	.rto_v = 0,
 	.rto_sweep = 0,
 	.rto_benchmark = 0,
+	.rto_expand = 0,
+	.rto_expand_offset = -1ULL,
 	.rto_sanity = 0,
 	.rto_gdb = 0,
 	.rto_should_stop = B_FALSE
@@ -112,4 +117,7 @@ void init_zio_abd(zio_t *zio);
 
 void run_raidz_benchmark(void);
 
+struct raidz_map *vdev_raidz_map_alloc_expanded(abd_t *, uint64_t, uint64_t,
+    uint64_t, uint64_t, uint64_t, uint64_t, uint64_t);
+
 #endif /* RAIDZ_TEST_H */

cmd/vdev_id/Makefile.am

@@ -1 +1,3 @@
+include $(top_srcdir)/config/Shellcheck.am
+
 dist_udev_SCRIPTS = vdev_id

cmd/vdev_id/vdev_id

@@ -79,6 +79,34 @@
 # channel 86:00.0 1         A
 # channel 86:00.0 0         B
 
+# #
+# # Example vdev_id.conf - multipath / multijbod-daisychaining
+# #
+#
+# multipath yes
+# multijbod yes
+#
+# #       PCI_ID  HBA PORT  CHANNEL NAME
+# channel 85:00.0 1         A
+# channel 85:00.0 0         B
+# channel 86:00.0 1         A
+# channel 86:00.0 0         B
+
+# #
+# # Example vdev_id.conf - multipath / mixed
+# #
+#
+# multipath yes
+# slot mix
+#
+# #       PCI_ID  HBA PORT  CHANNEL NAME
+# channel 85:00.0 3         A
+# channel 85:00.0 2         B
+# channel 86:00.0 3         A
+# channel 86:00.0 2         B
+# channel af:00.0 0         C
+# channel af:00.0 1         C
+
 # #
 # # Example vdev_id.conf - alias
 # #
@@ -92,9 +120,10 @@ PATH=/bin:/sbin:/usr/bin:/usr/sbin
 CONFIG=/etc/zfs/vdev_id.conf
 PHYS_PER_PORT=
 DEV=
-MULTIPATH=
 TOPOLOGY=
 BAY=
+ENCL_ID=""
+UNIQ_ENCL_ID=""
 
 usage() {
 	cat << EOF
@@ -102,71 +131,153 @@ Usage: vdev_id [-h]
        vdev_id <-d device> [-c config_file] [-p phys_per_port]
                [-g sas_direct|sas_switch|scsi] [-m]
 
-  -c    specify name of alernate config file [default=$CONFIG]
+  -c    specify name of an alternative config file [default=$CONFIG]
   -d    specify basename of device (i.e. sda)
   -e    Create enclose device symlinks only (/dev/by-enclosure)
   -g    Storage network topology [default="$TOPOLOGY"]
   -m    Run in multipath mode
+  -j    Run in multijbod mode
   -p    number of phy's per switch port [default=$PHYS_PER_PORT]
   -h    show this summary
 EOF
-	exit 0
+	exit 1
+	# exit with error to avoid processing usage message by a udev rule
 }
 
 map_slot() {
-	local LINUX_SLOT=$1
+	LINUX_SLOT=$1
-	local CHANNEL=$2
+	CHANNEL=$2
-	local MAPPED_SLOT=
 
-	MAPPED_SLOT=`awk "\\$1 == \"slot\" && \\$2 == ${LINUX_SLOT} && \
+	MAPPED_SLOT=$(awk -v linux_slot="$LINUX_SLOT" -v channel="$CHANNEL" \
-			\\$4 ~ /^${CHANNEL}$|^$/ { print \\$3; exit }" $CONFIG`
+			'$1 == "slot" && $2 == linux_slot && \
+			($4 ~ "^"channel"$" || $4 ~ /^$/) { print $3; exit}' $CONFIG)
 	if [ -z "$MAPPED_SLOT" ] ; then
 		MAPPED_SLOT=$LINUX_SLOT
 	fi
-	printf "%d" ${MAPPED_SLOT}
+	printf "%d" "${MAPPED_SLOT}"
 }
 
 map_channel() {
-	local MAPPED_CHAN=
+	MAPPED_CHAN=
-	local PCI_ID=$1
+	PCI_ID=$1
-	local PORT=$2
+	PORT=$2
 
 	case $TOPOLOGY in
 		"sas_switch")
-		MAPPED_CHAN=`awk "\\$1 == \"channel\" && \\$2 == ${PORT} \
+		MAPPED_CHAN=$(awk -v port="$PORT" \
-			{ print \\$3; exit }" $CONFIG`
+			'$1 == "channel" && $2 == port \
+			{ print $3; exit }' $CONFIG)
 		;;
 		"sas_direct"|"scsi")
-		MAPPED_CHAN=`awk "\\$1 == \"channel\" && \
+		MAPPED_CHAN=$(awk -v pciID="$PCI_ID" -v port="$PORT" \
-			\\$2 == \"${PCI_ID}\" && \\$3 == ${PORT} \
+			'$1 == "channel" && $2 == pciID && $3 == port \
-			{ print \\$4; exit }" $CONFIG`
+			{print $4}' $CONFIG)
 		;;
 	esac
-	printf "%s" ${MAPPED_CHAN}
+	printf "%s" "${MAPPED_CHAN}"
+}
+
+get_encl_id() {
+	set -- $(echo $1)
+	count=$#
+
+	i=1
+	while [ $i -le $count ] ; do
+		d=$(eval echo '$'{$i})
+		id=$(cat "/sys/class/enclosure/${d}/id")
+		ENCL_ID="${ENCL_ID} $id"
+		i=$((i + 1))
+	done
+}
+
+get_uniq_encl_id() {
+	for uuid in ${ENCL_ID}; do
+		found=0
+
+		for count in ${UNIQ_ENCL_ID}; do
+			if [ $count = $uuid ]; then
+				found=1
+				break
+			fi
+		done
+
+		if [ $found -eq 0 ]; then
+			UNIQ_ENCL_ID="${UNIQ_ENCL_ID} $uuid"
+		fi
+	done
+}
+
+# map_jbod explainer: The bsg driver knows the difference between a SAS
+# expander and fanout expander. Use hostX instance along with top-level
+# (whole enclosure) expander instances in /sys/class/enclosure and
+# matching a field in an array of expanders, using the index of the
+# matched array field as the enclosure instance, thereby making jbod IDs
+# dynamic. Avoids reliance on high overhead userspace commands like
+# multipath and lsscsi and instead uses existing sysfs data.  $HOSTCHAN
+# variable derived from devpath gymnastics in sas_handler() function.
+map_jbod() {
+	DEVEXP=$(ls -l "/sys/block/$DEV/device/" | grep enclos | awk -F/ '{print $(NF-1) }')
+	DEV=$1
+
+	# Use "set --" to create index values (Arrays)
+	set -- $(ls -l /sys/class/enclosure | grep -v "^total" | awk '{print $9}')
+	# Get count of total elements
+	JBOD_COUNT=$#
+	JBOD_ITEM=$*
+
+	# Build JBODs (enclosure)  id from sys/class/enclosure/<dev>/id
+	get_encl_id "$JBOD_ITEM"
+	# Different expander instances for each paths.
+	# Filter out and keep only unique id.
+	get_uniq_encl_id
+
+	# Identify final 'mapped jbod'
+	j=0
+	for count in ${UNIQ_ENCL_ID}; do
+		i=1
+		j=$((j + 1))
+		while [ $i -le $JBOD_COUNT ] ; do
+			d=$(eval echo '$'{$i})
+			id=$(cat "/sys/class/enclosure/${d}/id")
+			if [ "$d" = "$DEVEXP" ] && [ $id = $count ] ; then
+				MAPPED_JBOD=$j
+				break
+			fi
+			i=$((i + 1))
+		done
+	done
+
+	printf "%d" "${MAPPED_JBOD}"
 }
 
 sas_handler() {
 	if [ -z "$PHYS_PER_PORT" ] ; then
-		PHYS_PER_PORT=`awk "\\$1 == \"phys_per_port\" \
+		PHYS_PER_PORT=$(awk '$1 == "phys_per_port" \
-			{print \\$2; exit}" $CONFIG`
+			{print $2; exit}' $CONFIG)
 	fi
 	PHYS_PER_PORT=${PHYS_PER_PORT:-4}
-	if ! echo $PHYS_PER_PORT | grep -q -E '^[0-9]+$' ; then
+
+	if ! echo "$PHYS_PER_PORT" | grep -q -E '^[0-9]+$' ; then
 		echo "Error: phys_per_port value $PHYS_PER_PORT is non-numeric"
 		exit 1
 	fi
 
 	if [ -z "$MULTIPATH_MODE" ] ; then
-		MULTIPATH_MODE=`awk "\\$1 == \"multipath\" \
+		MULTIPATH_MODE=$(awk '$1 == "multipath" \
-			{print \\$2; exit}" $CONFIG`
+			{print $2; exit}' $CONFIG)
+	fi
+
+	if [ -z "$MULTIJBOD_MODE" ] ; then
+		MULTIJBOD_MODE=$(awk '$1 == "multijbod" \
+			{print $2; exit}' $CONFIG)
 	fi
 
 	# Use first running component device if we're handling a dm-mpath device
 	if [ "$MULTIPATH_MODE" = "yes" ] ; then
 		# If udev didn't tell us the UUID via DM_NAME, check /dev/mapper
 		if [ -z "$DM_NAME" ] ; then
-			DM_NAME=`ls -l --full-time /dev/mapper |
+			DM_NAME=$(ls -l --full-time /dev/mapper |
-				awk "/\/$DEV$/{print \\$9}"`
+				grep "$DEV"$ | awk '{print $9}')
 		fi
 
 		# For raw disks udev exports DEVTYPE=partition when
@@ -176,28 +287,50 @@ sas_handler() {
 		# we have to append the -part suffix directly in the
 		# helper.
 		if [ "$DEVTYPE" != "partition" ] ; then
-			PART=`echo $DM_NAME | awk -Fp '/p/{print "-part"$2}'`
+			# Match p[number], remove the 'p' and prepend "-part"
+			PART=$(echo "$DM_NAME" |
+				awk 'match($0,/p[0-9]+$/) {print "-part"substr($0,RSTART+1,RLENGTH-1)}')
 		fi
 
 		# Strip off partition information.
-		DM_NAME=`echo $DM_NAME | sed 's/p[0-9][0-9]*$//'`
+		DM_NAME=$(echo "$DM_NAME" | sed 's/p[0-9][0-9]*$//')
 		if [ -z "$DM_NAME" ] ; then
 			return
 		fi
 
-		# Get the raw scsi device name from multipath -ll. Strip off
+		# Utilize DM device name to gather subordinate block devices
-		# leading pipe symbols to make field numbering consistent.
+		# using sysfs to avoid userspace utilities
-		DEV=`multipath -ll $DM_NAME |
+
-			awk '/running/{gsub("^[|]"," "); print $3 ; exit}'`
+		# If our DEVNAME is something like /dev/dm-177, then we may be
+		# able to get our DMDEV from it.
+		DMDEV=$(echo $DEVNAME | sed 's;/dev/;;g')
+		if [ ! -e /sys/block/$DMDEV/slaves/* ] ; then
+			# It's not there, try looking in /dev/mapper
+			DMDEV=$(ls -l --full-time /dev/mapper | grep $DM_NAME |
+			awk '{gsub("../", " "); print $NF}')
+		fi
+
+		# Use sysfs pointers in /sys/block/dm-X/slaves because using
+		# userspace tools creates lots of overhead and should be avoided
+		# whenever possible. Use awk to isolate lowest instance of
+		# sd device member in dm device group regardless of string
+		# length.
+		DEV=$(ls "/sys/block/$DMDEV/slaves" | awk '
+			{ len=sprintf ("%20s",length($0)); gsub(/ /,0,str); a[NR]=len "_" $0; }
+			END {
+				asort(a)
+				print substr(a[1],22)
+			}')
+
 		if [ -z "$DEV" ] ; then
 			return
 		fi
 	fi
 
-	if echo $DEV | grep -q ^/devices/ ; then
+	if echo "$DEV" | grep -q ^/devices/ ; then
 		sys_path=$DEV
 	else
-		sys_path=`udevadm info -q path -p /sys/block/$DEV 2>/dev/null`
+		sys_path=$(udevadm info -q path -p "/sys/block/$DEV" 2>/dev/null)
 	fi
 
 	# Use positional parameters as an ad-hoc array
@@ -207,84 +340,104 @@ sas_handler() {
 
 	# Get path up to /sys/.../hostX
 	i=1
-	while [ $i -le $num_dirs ] ; do
+
-		d=$(eval echo \${$i})
+	while [ $i -le "$num_dirs" ] ; do
+		d=$(eval echo '$'{$i})
 		scsi_host_dir="$scsi_host_dir/$d"
-		echo $d | grep -q -E '^host[0-9]+$' && break
+		echo "$d" | grep -q -E '^host[0-9]+$' && break
-		i=$(($i + 1))
+		i=$((i + 1))
 	done
 
-	if [ $i = $num_dirs ] ; then
+	# Lets grab the SAS host channel number and save it for JBOD sorting later
+	HOSTCHAN=$(echo "$d" | awk -F/ '{ gsub("host","",$NF); print $NF}')
+
+	if [ $i = "$num_dirs" ] ; then
 		return
 	fi
 
-	PCI_ID=$(eval echo \${$(($i -1))} | awk -F: '{print $2":"$3}')
+	PCI_ID=$(eval echo '$'{$((i -1))} | awk -F: '{print $2":"$3}')
 
 	# In sas_switch mode, the directory four levels beneath
 	# /sys/.../hostX contains symlinks to phy devices that reveal
 	# the switch port number.  In sas_direct mode, the phy links one
 	# directory down reveal the HBA port.
 	port_dir=$scsi_host_dir
+
 	case $TOPOLOGY in
-		"sas_switch") j=$(($i + 4)) ;;
+		"sas_switch") j=$((i + 4)) ;;
-		"sas_direct") j=$(($i + 1)) ;;
+		"sas_direct") j=$((i + 1)) ;;
 	esac
 
-	i=$(($i + 1))
+	i=$((i + 1))
+
 	while [ $i -le $j ] ; do
-		port_dir="$port_dir/$(eval echo \${$i})"
+		port_dir="$port_dir/$(eval echo '$'{$i})"
-		i=$(($i + 1))
+		i=$((i + 1))
 	done
 
-	PHY=`ls -d $port_dir/phy* 2>/dev/null | head -1 | awk -F: '{print $NF}'`
+	PHY=$(ls -vd "$port_dir"/phy* 2>/dev/null | head -1 | awk -F: '{print $NF}')
 	if [ -z "$PHY" ] ; then
 		PHY=0
 	fi
-	PORT=$(( $PHY / $PHYS_PER_PORT ))
+	PORT=$((PHY / PHYS_PER_PORT))
 
 	# Look in /sys/.../sas_device/end_device-X for the bay_identifier
 	# attribute.
 	end_device_dir=$port_dir
-	while [ $i -lt $num_dirs ] ; do
+
-		d=$(eval echo \${$i})
+	while [ $i -lt "$num_dirs" ] ; do
+		d=$(eval echo '$'{$i})
 		end_device_dir="$end_device_dir/$d"
-		if echo $d | grep -q '^end_device' ; then
+		if echo "$d" | grep -q '^end_device' ; then
 			end_device_dir="$end_device_dir/sas_device/$d"
 			break
 		fi
-		i=$(($i + 1))
+		i=$((i + 1))
 	done
 
+	# Add 'mix' slot type for environments where dm-multipath devices
+	# include end-devices connected via SAS expanders or direct connection
+	# to SAS HBA. A mixed connectivity environment such as pool devices
+	# contained in a SAS JBOD and spare drives or log devices directly
+	# connected in a server backplane without expanders in the I/O path.
 	SLOT=
+
 	case $BAY in
 	"bay")
-		SLOT=`cat $end_device_dir/bay_identifier 2>/dev/null`
+		SLOT=$(cat "$end_device_dir/bay_identifier" 2>/dev/null)
+		;;
+	"mix")
+		if [ $(cat "$end_device_dir/bay_identifier" 2>/dev/null) ] ; then
+			SLOT=$(cat "$end_device_dir/bay_identifier" 2>/dev/null)
+		else
+			SLOT=$(cat "$end_device_dir/phy_identifier" 2>/dev/null)
+		fi
 		;;
 	"phy")
-		SLOT=`cat $end_device_dir/phy_identifier 2>/dev/null`
+		SLOT=$(cat "$end_device_dir/phy_identifier" 2>/dev/null)
 		;;
 	"port")
-		d=$(eval echo \${$i})
+		d=$(eval echo '$'{$i})
-		SLOT=`echo $d | sed -e 's/^.*://'`
+		SLOT=$(echo "$d" | sed -e 's/^.*://')
 		;;
 	"id")
-		i=$(($i + 1))
+		i=$((i + 1))
-		d=$(eval echo \${$i})
+		d=$(eval echo '$'{$i})
-		SLOT=`echo $d | sed -e 's/^.*://'`
+		SLOT=$(echo "$d" | sed -e 's/^.*://')
 		;;
 	"lun")
-		i=$(($i + 2))
+		i=$((i + 2))
-		d=$(eval echo \${$i})
+		d=$(eval echo '$'{$i})
-		SLOT=`echo $d | sed -e 's/^.*://'`
+		SLOT=$(echo "$d" | sed -e 's/^.*://')
 		;;
 	"ses")
 		# look for this SAS path in all SCSI Enclosure Services
 		# (SES) enclosures
-		sas_address=`cat $end_device_dir/sas_address 2>/dev/null`
+		sas_address=$(cat "$end_device_dir/sas_address" 2>/dev/null)
-		enclosures=`lsscsi -g | \
+		enclosures=$(lsscsi -g | \
-			sed -n -e '/enclosu/s/^.* \([^ ][^ ]*\) *$/\1/p'`
+			sed -n -e '/enclosu/s/^.* \([^ ][^ ]*\) *$/\1/p')
 		for enclosure in $enclosures; do
-			set -- $(sg_ses -p aes $enclosure | \
+			set -- $(sg_ses -p aes "$enclosure" | \
 				awk "/device slot number:/{slot=\$12} \
 					/SAS address: $sas_address/\
 					{print slot}")
@@ -299,42 +452,55 @@ sas_handler() {
 		return
 	fi
 
-	CHAN=`map_channel $PCI_ID $PORT`
+	if [ "$MULTIJBOD_MODE" = "yes" ] ; then
-	SLOT=`map_slot $SLOT $CHAN`
+		CHAN=$(map_channel "$PCI_ID" "$PORT")
-	if [ -z "$CHAN" ] ; then
+		SLOT=$(map_slot "$SLOT" "$CHAN")
-		return
+		JBOD=$(map_jbod "$DEV")
+
+		if [ -z "$CHAN" ] ; then
+			return
+		fi
+		echo "${CHAN}"-"${JBOD}"-"${SLOT}${PART}"
+	else
+		CHAN=$(map_channel "$PCI_ID" "$PORT")
+		SLOT=$(map_slot "$SLOT" "$CHAN")
+
+		if [ -z "$CHAN" ] ; then
+			return
+		fi
+		echo "${CHAN}${SLOT}${PART}"
 	fi
-	echo ${CHAN}${SLOT}${PART}
 }
 
 scsi_handler() {
 	if [ -z "$FIRST_BAY_NUMBER" ] ; then
-		FIRST_BAY_NUMBER=`awk "\\$1 == \"first_bay_number\" \
+		FIRST_BAY_NUMBER=$(awk '$1 == "first_bay_number" \
-			{print \\$2; exit}" $CONFIG`
+			{print $2; exit}' $CONFIG)
 	fi
 	FIRST_BAY_NUMBER=${FIRST_BAY_NUMBER:-0}
 
 	if [ -z "$PHYS_PER_PORT" ] ; then
-		PHYS_PER_PORT=`awk "\\$1 == \"phys_per_port\" \
+		PHYS_PER_PORT=$(awk '$1 == "phys_per_port" \
-			{print \\$2; exit}" $CONFIG`
+			{print $2; exit}' $CONFIG)
 	fi
 	PHYS_PER_PORT=${PHYS_PER_PORT:-4}
-	if ! echo $PHYS_PER_PORT | grep -q -E '^[0-9]+$' ; then
+
+	if ! echo "$PHYS_PER_PORT" | grep -q -E '^[0-9]+$' ; then
 		echo "Error: phys_per_port value $PHYS_PER_PORT is non-numeric"
 		exit 1
 	fi
 
 	if [ -z "$MULTIPATH_MODE" ] ; then
-		MULTIPATH_MODE=`awk "\\$1 == \"multipath\" \
+		MULTIPATH_MODE=$(awk '$1 == "multipath" \
-			{print \\$2; exit}" $CONFIG`
+			{print $2; exit}' $CONFIG)
 	fi
 
 	# Use first running component device if we're handling a dm-mpath device
 	if [ "$MULTIPATH_MODE" = "yes" ] ; then
 		# If udev didn't tell us the UUID via DM_NAME, check /dev/mapper
 		if [ -z "$DM_NAME" ] ; then
-			DM_NAME=`ls -l --full-time /dev/mapper |
+			DM_NAME=$(ls -l --full-time /dev/mapper |
-				awk "/\/$DEV$/{print \\$9}"`
+				grep "$DEV"$ | awk '{print $9}')
 		fi
 
 		# For raw disks udev exports DEVTYPE=partition when
@@ -344,28 +510,30 @@ scsi_handler() {
 		# we have to append the -part suffix directly in the
 		# helper.
 		if [ "$DEVTYPE" != "partition" ] ; then
-			PART=`echo $DM_NAME | awk -Fp '/p/{print "-part"$2}'`
+			# Match p[number], remove the 'p' and prepend "-part"
+			PART=$(echo "$DM_NAME" |
+			    awk 'match($0,/p[0-9]+$/) {print "-part"substr($0,RSTART+1,RLENGTH-1)}')
 		fi
 
 		# Strip off partition information.
-		DM_NAME=`echo $DM_NAME | sed 's/p[0-9][0-9]*$//'`
+		DM_NAME=$(echo "$DM_NAME" | sed 's/p[0-9][0-9]*$//')
 		if [ -z "$DM_NAME" ] ; then
 			return
 		fi
 
 		# Get the raw scsi device name from multipath -ll. Strip off
 		# leading pipe symbols to make field numbering consistent.
-		DEV=`multipath -ll $DM_NAME |
+		DEV=$(multipath -ll "$DM_NAME" |
-			awk '/running/{gsub("^[|]"," "); print $3 ; exit}'`
+			awk '/running/{gsub("^[|]"," "); print $3 ; exit}')
 		if [ -z "$DEV" ] ; then
 			return
 		fi
 	fi
 
-	if echo $DEV | grep -q ^/devices/ ; then
+	if echo "$DEV" | grep -q ^/devices/ ; then
 		sys_path=$DEV
 	else
-		sys_path=`udevadm info -q path -p /sys/block/$DEV 2>/dev/null`
+		sys_path=$(udevadm info -q path -p "/sys/block/$DEV" 2>/dev/null)
 	fi
 
 	# expect sys_path like this, for example:
@@ -378,44 +546,47 @@ scsi_handler() {
 
 	# Get path up to /sys/.../hostX
 	i=1
-	while [ $i -le $num_dirs ] ; do
+
-		d=$(eval echo \${$i})
+	while [ $i -le "$num_dirs" ] ; do
+		d=$(eval echo '$'{$i})
 		scsi_host_dir="$scsi_host_dir/$d"
-		echo $d | grep -q -E '^host[0-9]+$' && break
+
-		i=$(($i + 1))
+		echo "$d" | grep -q -E '^host[0-9]+$' && break
+		i=$((i + 1))
 	done
 
-	if [ $i = $num_dirs ] ; then
+	if [ $i = "$num_dirs" ] ; then
 		return
 	fi
 
-	PCI_ID=$(eval echo \${$(($i -1))} | awk -F: '{print $2":"$3}')
+	PCI_ID=$(eval echo '$'{$((i -1))} | awk -F: '{print $2":"$3}')
 
 	# In scsi mode, the directory two levels beneath
 	# /sys/.../hostX reveals the port and slot.
 	port_dir=$scsi_host_dir
-	j=$(($i + 2))
+	j=$((i + 2))
 
-	i=$(($i + 1))
+	i=$((i + 1))
 	while [ $i -le $j ] ; do
-		port_dir="$port_dir/$(eval echo \${$i})"
+		port_dir="$port_dir/$(eval echo '$'{$i})"
-		i=$(($i + 1))
+		i=$((i + 1))
 	done
 
-	set -- $(echo $port_dir | sed -e 's/^.*:\([^:]*\):\([^:]*\)$/\1 \2/')
+	set -- $(echo "$port_dir" | sed -e 's/^.*:\([^:]*\):\([^:]*\)$/\1 \2/')
 	PORT=$1
-	SLOT=$(($2 + $FIRST_BAY_NUMBER))
+	SLOT=$(($2 + FIRST_BAY_NUMBER))
 
 	if [ -z "$SLOT" ] ; then
 		return
 	fi
 
-	CHAN=`map_channel $PCI_ID $PORT`
+	CHAN=$(map_channel "$PCI_ID" "$PORT")
-	SLOT=`map_slot $SLOT $CHAN`
+	SLOT=$(map_slot "$SLOT" "$CHAN")
+
 	if [ -z "$CHAN" ] ; then
 		return
 	fi
-	echo ${CHAN}${SLOT}${PART}
+	echo "${CHAN}${SLOT}${PART}"
 }
 
 # Figure out the name for the enclosure symlink
@@ -426,7 +597,7 @@ enclosure_handler () {
 
 	# Get the enclosure ID ("0:0:0:0")
 	ENC=$(basename $(readlink -m "/sys/$DEVPATH/../.."))
-	if [ ! -d /sys/class/enclosure/$ENC ] ; then
+	if [ ! -d "/sys/class/enclosure/$ENC" ] ; then
 		# Not an enclosure, bail out
 		return
 	fi
@@ -434,14 +605,14 @@ enclosure_handler () {
 	# Get the long sysfs device path to our enclosure. Looks like:
 	# /devices/pci0000:00/0000:00:03.0/0000:05:00.0/host0/port-0:0/ ... /enclosure/0:0:0:0
 
-	ENC_DEVICE=$(readlink /sys/class/enclosure/$ENC)
+	ENC_DEVICE=$(readlink "/sys/class/enclosure/$ENC")
 
 	# Grab the full path to the hosts port dir:
 	# /devices/pci0000:00/0000:00:03.0/0000:05:00.0/host0/port-0:0
-	PORT_DIR=$(echo $ENC_DEVICE | grep -Eo '.+host[0-9]+/port-[0-9]+:[0-9]+')
+	PORT_DIR=$(echo "$ENC_DEVICE" | grep -Eo '.+host[0-9]+/port-[0-9]+:[0-9]+')
 
 	# Get the port number
-	PORT_ID=$(echo $PORT_DIR | grep -Eo "[0-9]+$")
+	PORT_ID=$(echo "$PORT_DIR" | grep -Eo "[0-9]+$")
 
 	# The PCI directory is two directories up from the port directory
 	# /sys/devices/pci0000:00/0000:00:03.0/0000:05:00.0
@@ -452,7 +623,7 @@ enclosure_handler () {
 
 	# Name our device according to vdev_id.conf (like "L0" or "U1").
 	NAME=$(awk "/channel/{if (\$1 == \"channel\" && \$2 == \"$PCI_ID\" && \
-		\$3 == \"$PORT_ID\") {print \$4int(count[\$4])}; count[\$4]++}" $CONFIG)
+		\$3 == \"$PORT_ID\") {print \$4\$3}}" $CONFIG)
 
 	echo "${NAME}"
 }
@@ -487,10 +658,12 @@ alias_handler () {
 	#          digits as partitions, causing alias creation to fail. This
 	#          ambiguity seems unavoidable, so devices using this facility
 	#          must not use such names.
-	local DM_PART=
+	DM_PART=
-	if echo $DM_NAME | grep -q -E 'p[0-9][0-9]*$' ; then
+	if echo "$DM_NAME" | grep -q -E 'p[0-9][0-9]*$' ; then
 		if [ "$DEVTYPE" != "partition" ] ; then
-			DM_PART=`echo $DM_NAME | awk -Fp '/p/{print "-part"$2}'`
+			# Match p[number], remove the 'p' and prepend "-part"
+			DM_PART=$(echo "$DM_NAME" |
+			    awk 'match($0,/p[0-9]+$/) {print "-part"substr($0,RSTART+1,RLENGTH-1)}')
 		fi
 	fi
 
@@ -498,21 +671,25 @@ alias_handler () {
 	for link in $DEVLINKS ; do
 		# Remove partition information to match key of top-level device.
 		if [ -n "$DM_PART" ] ; then
-			link=`echo $link | sed 's/p[0-9][0-9]*$//'`
+			link=$(echo "$link" | sed 's/p[0-9][0-9]*$//')
 		fi
 		# Check both the fully qualified and the base name of link.
-		for l in $link `basename $link` ; do
+		for l in $link $(basename "$link") ; do
-			alias=`awk "\\$1 == \"alias\" && \\$3 == \"${l}\" \
+			if [ ! -z "$l" ]; then
-					{ print \\$2; exit }" $CONFIG`
+				alias=$(awk -v var="$l" '($1 == "alias") && \
-			if [ -n "$alias" ] ; then
+					($3 == var) \
-				echo ${alias}${DM_PART}
+					{ print $2; exit }' $CONFIG)
-				return
+				if [ -n "$alias" ] ; then
+					echo "${alias}${DM_PART}"
+					return
+				fi
 			fi
 		done
 	done
 }
 
-while getopts 'c:d:eg:mp:h' OPTION; do
+# main
+while getopts 'c:d:eg:jmp:h' OPTION; do
 	case ${OPTION} in
 	c)
 		CONFIG=${OPTARG}
@@ -525,7 +702,9 @@ while getopts 'c:d:eg:mp:h' OPTION; do
 	# create the enclosure device symlinks only.  We also need
 	# "enclosure_symlinks yes" set in vdev_id.config to actually create the
 	# symlink.
-	ENCLOSURE_MODE=$(awk '{if ($1 == "enclosure_symlinks") print $2}' $CONFIG)
+	ENCLOSURE_MODE=$(awk '{if ($1 == "enclosure_symlinks") \
+		print $2}' "$CONFIG")
+
 	if [ "$ENCLOSURE_MODE" != "yes" ] ; then
 		exit 0
 	fi
@@ -536,6 +715,9 @@ while getopts 'c:d:eg:mp:h' OPTION; do
 	p)
 		PHYS_PER_PORT=${OPTARG}
 		;;
+	j)
+		MULTIJBOD_MODE=yes
+		;;
 	m)
 		MULTIPATH_MODE=yes
 		;;
@@ -545,34 +727,35 @@ while getopts 'c:d:eg:mp:h' OPTION; do
 	esac
 done
 
-if [ ! -r $CONFIG ] ; then
+if [ ! -r "$CONFIG" ] ; then
-	exit 0
+	echo "Error: Config file \"$CONFIG\" not found"
+	exit 1
 fi
 
-if [ -z "$DEV" -a -z "$ENCLOSURE_MODE" ] ; then
+if [ -z "$DEV" ] && [ -z "$ENCLOSURE_MODE" ] ; then
 	echo "Error: missing required option -d"
 	exit 1
 fi
 
 if [ -z "$TOPOLOGY" ] ; then
-	TOPOLOGY=`awk "\\$1 == \"topology\" {print \\$2; exit}" $CONFIG`
+	TOPOLOGY=$(awk '($1 == "topology") {print $2; exit}' "$CONFIG")
 fi
 
 if [ -z "$BAY" ] ; then
-	BAY=`awk "\\$1 == \"slot\" {print \\$2; exit}" $CONFIG`
+	BAY=$(awk '($1 == "slot") {print $2; exit}' "$CONFIG")
 fi
 
 TOPOLOGY=${TOPOLOGY:-sas_direct}
 
 # Should we create /dev/by-enclosure symlinks?
-if [ "$ENCLOSURE_MODE" = "yes" -a "$TOPOLOGY" = "sas_direct" ] ; then
+if [ "$ENCLOSURE_MODE" = "yes" ] && [ "$TOPOLOGY" = "sas_direct" ] ; then
 	ID_ENCLOSURE=$(enclosure_handler)
 	if [ -z "$ID_ENCLOSURE" ] ; then
 		exit 0
 	fi
 
 	# Just create the symlinks to the enclosure devices and then exit.
-	ENCLOSURE_PREFIX=$(awk '/enclosure_symlinks_prefix/{print $2}' $CONFIG)
+	ENCLOSURE_PREFIX=$(awk '/enclosure_symlinks_prefix/{print $2}' "$CONFIG")
 	if [ -z "$ENCLOSURE_PREFIX" ] ; then
 		ENCLOSURE_PREFIX="enc"
 	fi
@@ -582,16 +765,16 @@ if [ "$ENCLOSURE_MODE" = "yes" -a "$TOPOLOGY" = "sas_direct" ] ; then
 fi
 
 # First check if an alias was defined for this device.
-ID_VDEV=`alias_handler`
+ID_VDEV=$(alias_handler)
 
 if [ -z "$ID_VDEV" ] ; then
 	BAY=${BAY:-bay}
 	case $TOPOLOGY in
 		sas_direct|sas_switch)
-			ID_VDEV=`sas_handler`
+			ID_VDEV=$(sas_handler)
 			;;
 		scsi)
-			ID_VDEV=`scsi_handler`
+			ID_VDEV=$(scsi_handler)
 			;;
 		*)
 			echo "Error: unknown topology $TOPOLOGY"

cmd/zdb/Makefile.am

@@ -1,11 +1,7 @@
 include $(top_srcdir)/config/Rules.am
 
 # Unconditionally enable debugging for zdb
-AM_CPPFLAGS += -DDEBUG -UNDEBUG
+AM_CPPFLAGS += -DDEBUG -UNDEBUG -DZFS_DEBUG
-
-DEFAULT_INCLUDES += \
-	-I$(top_srcdir)/include \
-	-I$(top_srcdir)/lib/libspl/include
 
 sbin_PROGRAMS = zdb
 
@@ -15,5 +11,8 @@ zdb_SOURCES = \
 	zdb.h
 
 zdb_LDADD = \
-	$(top_builddir)/lib/libnvpair/libnvpair.la \
+	$(abs_top_builddir)/lib/libzpool/libzpool.la \
-	$(top_builddir)/lib/libzpool/libzpool.la
+	$(abs_top_builddir)/lib/libzfs_core/libzfs_core.la \
+	$(abs_top_builddir)/lib/libnvpair/libnvpair.la
+
+include $(top_srcdir)/config/CppCheck.am

cmd/zdb/zdb_il.c

@@ -62,9 +62,9 @@ print_log_bp(const blkptr_t *bp, const char *prefix)
 
 /* ARGSUSED */
 static void
-zil_prt_rec_create(zilog_t *zilog, int txtype, void *arg)
+zil_prt_rec_create(zilog_t *zilog, int txtype, const void *arg)
 {
-	lr_create_t *lr = arg;
+	const lr_create_t *lr = arg;
 	time_t crtime = lr->lr_crtime[0];
 	char *name, *link;
 	lr_attr_t *lrattr;
@@ -98,9 +98,9 @@ zil_prt_rec_create(zilog_t *zilog, int txtype, void *arg)
 
 /* ARGSUSED */
 static void
-zil_prt_rec_remove(zilog_t *zilog, int txtype, void *arg)
+zil_prt_rec_remove(zilog_t *zilog, int txtype, const void *arg)
 {
-	lr_remove_t *lr = arg;
+	const lr_remove_t *lr = arg;
 
 	(void) printf("%sdoid %llu, name %s\n", tab_prefix,
 	    (u_longlong_t)lr->lr_doid, (char *)(lr + 1));
@@ -108,9 +108,9 @@ zil_prt_rec_remove(zilog_t *zilog, int txtype, void *arg)
 
 /* ARGSUSED */
 static void
-zil_prt_rec_link(zilog_t *zilog, int txtype, void *arg)
+zil_prt_rec_link(zilog_t *zilog, int txtype, const void *arg)
 {
-	lr_link_t *lr = arg;
+	const lr_link_t *lr = arg;
 
 	(void) printf("%sdoid %llu, link_obj %llu, name %s\n", tab_prefix,
 	    (u_longlong_t)lr->lr_doid, (u_longlong_t)lr->lr_link_obj,
@@ -119,9 +119,9 @@ zil_prt_rec_link(zilog_t *zilog, int txtype, void *arg)
 
 /* ARGSUSED */
 static void
-zil_prt_rec_rename(zilog_t *zilog, int txtype, void *arg)
+zil_prt_rec_rename(zilog_t *zilog, int txtype, const void *arg)
 {
-	lr_rename_t *lr = arg;
+	const lr_rename_t *lr = arg;
 	char *snm = (char *)(lr + 1);
 	char *tnm = snm + strlen(snm) + 1;
 
@@ -148,11 +148,11 @@ zil_prt_rec_write_cb(void *data, size_t len, void *unused)
 
 /* ARGSUSED */
 static void
-zil_prt_rec_write(zilog_t *zilog, int txtype, void *arg)
+zil_prt_rec_write(zilog_t *zilog, int txtype, const void *arg)
 {
-	lr_write_t *lr = arg;
+	const lr_write_t *lr = arg;
 	abd_t *data;
-	blkptr_t *bp = &lr->lr_blkptr;
+	const blkptr_t *bp = &lr->lr_blkptr;
 	zbookmark_phys_t zb;
 	int verbose = MAX(dump_opt['d'], dump_opt['i']);
 	int error;
@@ -211,9 +211,9 @@ out:
 
 /* ARGSUSED */
 static void
-zil_prt_rec_truncate(zilog_t *zilog, int txtype, void *arg)
+zil_prt_rec_truncate(zilog_t *zilog, int txtype, const void *arg)
 {
-	lr_truncate_t *lr = arg;
+	const lr_truncate_t *lr = arg;
 
 	(void) printf("%sfoid %llu, offset 0x%llx, length 0x%llx\n", tab_prefix,
 	    (u_longlong_t)lr->lr_foid, (longlong_t)lr->lr_offset,
@@ -222,9 +222,9 @@ zil_prt_rec_truncate(zilog_t *zilog, int txtype, void *arg)
 
 /* ARGSUSED */
 static void
-zil_prt_rec_setattr(zilog_t *zilog, int txtype, void *arg)
+zil_prt_rec_setattr(zilog_t *zilog, int txtype, const void *arg)
 {
-	lr_setattr_t *lr = arg;
+	const lr_setattr_t *lr = arg;
 	time_t atime = (time_t)lr->lr_atime[0];
 	time_t mtime = (time_t)lr->lr_mtime[0];
 
@@ -268,15 +268,15 @@ zil_prt_rec_setattr(zilog_t *zilog, int txtype, void *arg)
 
 /* ARGSUSED */
 static void
-zil_prt_rec_acl(zilog_t *zilog, int txtype, void *arg)
+zil_prt_rec_acl(zilog_t *zilog, int txtype, const void *arg)
 {
-	lr_acl_t *lr = arg;
+	const lr_acl_t *lr = arg;
 
 	(void) printf("%sfoid %llu, aclcnt %llu\n", tab_prefix,
 	    (u_longlong_t)lr->lr_foid, (u_longlong_t)lr->lr_aclcnt);
 }
 
-typedef void (*zil_prt_rec_func_t)(zilog_t *, int, void *);
+typedef void (*zil_prt_rec_func_t)(zilog_t *, int, const void *);
 typedef struct zil_rec_info {
 	zil_prt_rec_func_t	zri_print;
 	const char		*zri_name;
@@ -309,7 +309,7 @@ static zil_rec_info_t zil_rec_info[TX_MAX_TYPE] = {
 
 /* ARGSUSED */
 static int
-print_log_record(zilog_t *zilog, lr_t *lr, void *arg, uint64_t claim_txg)
+print_log_record(zilog_t *zilog, const lr_t *lr, void *arg, uint64_t claim_txg)
 {
 	int txtype;
 	int verbose = MAX(dump_opt['d'], dump_opt['i']);
@@ -343,7 +343,8 @@ print_log_record(zilog_t *zilog, lr_t *lr, void *arg, uint64_t claim_txg)
 
 /* ARGSUSED */
 static int
-print_log_block(zilog_t *zilog, blkptr_t *bp, void *arg, uint64_t claim_txg)
+print_log_block(zilog_t *zilog, const blkptr_t *bp, void *arg,
+    uint64_t claim_txg)
 {
 	char blkbuf[BP_SPRINTF_LEN + 10];
 	int verbose = MAX(dump_opt['d'], dump_opt['i']);

cmd/zed/Makefile.am

@@ -1,11 +1,10 @@
 include $(top_srcdir)/config/Rules.am
+include $(top_srcdir)/config/Shellcheck.am
 
-DEFAULT_INCLUDES += \
+AM_CFLAGS += $(LIBUDEV_CFLAGS) $(LIBUUID_CFLAGS)
-	-I$(top_srcdir)/include \
-	-I$(top_srcdir)/lib/libspl/include
 
-EXTRA_DIST = zed.d/README \
+SUBDIRS = zed.d
-	zed.d/history_event-zfs-list-cacher.sh.in
+SHELLCHECKDIRS = $(SUBDIRS)
 
 sbin_PROGRAMS = zed
 
@@ -41,61 +40,14 @@ FMA_SRC = \
 zed_SOURCES = $(ZED_SRC) $(FMA_SRC)
 
 zed_LDADD = \
-	$(top_builddir)/lib/libnvpair/libnvpair.la \
+	$(abs_top_builddir)/lib/libzfs/libzfs.la \
-	$(top_builddir)/lib/libuutil/libuutil.la \
+	$(abs_top_builddir)/lib/libzfs_core/libzfs_core.la \
-	$(top_builddir)/lib/libzfs/libzfs.la
+	$(abs_top_builddir)/lib/libnvpair/libnvpair.la \
+	$(abs_top_builddir)/lib/libuutil/libuutil.la
 
-zed_LDADD += -lrt
+zed_LDADD += -lrt $(LIBATOMIC_LIBS) $(LIBUDEV_LIBS) $(LIBUUID_LIBS)
 zed_LDFLAGS = -pthread
 
-zedconfdir = $(sysconfdir)/zfs/zed.d
+EXTRA_DIST = agents/README.md
 
-dist_zedconf_DATA = \
+include $(top_srcdir)/config/CppCheck.am
-	zed.d/zed-functions.sh \
-	zed.d/zed.rc
-
-zedexecdir = $(zfsexecdir)/zed.d
-
-dist_zedexec_SCRIPTS = \
-	zed.d/all-debug.sh \
-	zed.d/all-syslog.sh \
-	zed.d/data-notify.sh \
-	zed.d/generic-notify.sh \
-	zed.d/resilver_finish-notify.sh \
-	zed.d/scrub_finish-notify.sh \
-	zed.d/statechange-led.sh \
-	zed.d/statechange-notify.sh \
-	zed.d/vdev_clear-led.sh \
-	zed.d/vdev_attach-led.sh \
-	zed.d/pool_import-led.sh \
-	zed.d/resilver_finish-start-scrub.sh
-
-nodist_zedexec_SCRIPTS = zed.d/history_event-zfs-list-cacher.sh
-
-$(nodist_zedexec_SCRIPTS): %: %.in
-	-$(SED) -e 's,@bindir\@,$(bindir),g' \
-		-e 's,@runstatedir\@,$(runstatedir),g' \
-		-e 's,@sbindir\@,$(sbindir),g' \
-		-e 's,@sysconfdir\@,$(sysconfdir),g' \
-		$< >'$@'
-
-zedconfdefaults = \
-	all-syslog.sh \
-	data-notify.sh \
-	resilver_finish-notify.sh \
-	scrub_finish-notify.sh \
-	statechange-led.sh \
-	statechange-notify.sh \
-	vdev_clear-led.sh \
-	vdev_attach-led.sh \
-	pool_import-led.sh \
-	resilver_finish-start-scrub.sh
-
-install-data-hook:
-	$(MKDIR_P) "$(DESTDIR)$(zedconfdir)"
-	for f in $(zedconfdefaults); do \
-	  test -f "$(DESTDIR)$(zedconfdir)/$${f}" -o \
-	       -L "$(DESTDIR)$(zedconfdir)/$${f}" || \
-	    ln -s "$(zedexecdir)/$${f}" "$(DESTDIR)$(zedconfdir)"; \
-	done
-	chmod 0600 "$(DESTDIR)$(zedconfdir)/zed.rc"

cmd/zed/agents/fmd_api.c

@@ -25,7 +25,7 @@
  */
 
 /*
- * This file imlements the minimal FMD module API required to support the
+ * This file implements the minimal FMD module API required to support the
  * fault logic modules in ZED. This support includes module registration,
  * memory allocation, module property accessors, basic case management,
  * one-shot timers and SERD engines.

cmd/zed/agents/fmd_serd.c

@@ -281,7 +281,7 @@ fmd_serd_eng_empty(fmd_serd_eng_t *sgp)
 void
 fmd_serd_eng_reset(fmd_serd_eng_t *sgp)
 {
-	serd_log_msg("  SERD Engine: reseting %s", sgp->sg_name);
+	serd_log_msg("  SERD Engine: resetting %s", sgp->sg_name);
 
 	while (sgp->sg_count != 0)
 		fmd_serd_eng_discard(sgp, list_head(&sgp->sg_list));

cmd/zed/agents/zfs_agents.c

@@ -13,6 +13,7 @@
 /*
  * Copyright (c) 2016, Intel Corporation.
  * Copyright (c) 2018, loli10K <ezomori.nozomu@gmail.com>
+ * Copyright (c) 2021 Hewlett Packard Enterprise Development LP
  */
 
 #include <libnvpair.h>
@@ -116,7 +117,8 @@ zfs_agent_iter_vdev(zpool_handle_t *zhp, nvlist_t *nvl, void *arg)
 	/*
 	 * On a devid match, grab the vdev guid and expansion time, if any.
 	 */
-	if ((nvlist_lookup_string(nvl, ZPOOL_CONFIG_DEVID, &path) == 0) &&
+	if (gsp->gs_devid != NULL &&
+	    (nvlist_lookup_string(nvl, ZPOOL_CONFIG_DEVID, &path) == 0) &&
 	    (strcmp(gsp->gs_devid, path) == 0)) {
 		(void) nvlist_lookup_uint64(nvl, ZPOOL_CONFIG_GUID,
 		    &gsp->gs_vdev_guid);
@@ -176,10 +178,12 @@ zfs_agent_post_event(const char *class, const char *subclass, nvlist_t *nvl)
 	}
 
 	/*
-	 * On ZFS on Linux, we don't get the expected FM_RESOURCE_REMOVED
+	 * On Linux, we don't get the expected FM_RESOURCE_REMOVED ereport
-	 * ereport from vdev_disk layer after a hot unplug. Fortunately we
+	 * from the vdev_disk layer after a hot unplug. Fortunately we do
-	 * get a EC_DEV_REMOVE from our disk monitor and it is a suitable
+	 * get an EC_DEV_REMOVE from our disk monitor and it is a suitable
 	 * proxy so we remap it here for the benefit of the diagnosis engine.
+	 * Starting in OpenZFS 2.0, we do get FM_RESOURCE_REMOVED from the spa
+	 * layer. Processing multiple FM_RESOURCE_REMOVED events is not harmful.
 	 */
 	if ((strcmp(class, EC_DEV_REMOVE) == 0) &&
 	    (strcmp(subclass, ESC_DISK) == 0) &&
@@ -208,12 +212,18 @@ zfs_agent_post_event(const char *class, const char *subclass, nvlist_t *nvl)
 		 * For multipath, spare and l2arc devices ZFS_EV_VDEV_GUID or
 		 * ZFS_EV_POOL_GUID may be missing so find them.
 		 */
-		(void) nvlist_lookup_string(nvl, DEV_IDENTIFIER,
+		if (pool_guid == 0 || vdev_guid == 0) {
-		    &search.gs_devid);
+			if ((nvlist_lookup_string(nvl, DEV_IDENTIFIER,
-		(void) zpool_iter(g_zfs_hdl, zfs_agent_iter_pool, &search);
+			    &search.gs_devid) == 0) &&
-		pool_guid = search.gs_pool_guid;
+			    (zpool_iter(g_zfs_hdl, zfs_agent_iter_pool, &search)
-		vdev_guid = search.gs_vdev_guid;
+			    == 1)) {
-		devtype = search.gs_vdev_type;
+				if (pool_guid == 0)
+					pool_guid = search.gs_pool_guid;
+				if (vdev_guid == 0)
+					vdev_guid = search.gs_vdev_guid;
+				devtype = search.gs_vdev_type;
+			}
+		}
 
 		/*
 		 * We want to avoid reporting "remove" events coming from
@@ -382,6 +392,7 @@ zfs_agent_init(libzfs_handle_t *zfs_hdl)
 		list_destroy(&agent_events);
 		zed_log_die("Failed to initialize agents");
 	}
+	pthread_setname_np(g_agents_tid, "agents");
 }
 
 void

cmd/zed/agents/zfs_mod.c

@@ -63,13 +63,10 @@
  * If the device could not be replaced, then the second online attempt will
  * trigger the FMA fault that we skipped earlier.
  *
- * ZFS on Linux porting notes:
+ * On Linux udev provides a disk insert for both the disk and the partition.
- *	Linux udev provides a disk insert for both the disk and the partition
- *
  */
 
 #include <ctype.h>
-#include <devid.h>
 #include <fcntl.h>
 #include <libnvpair.h>
 #include <libzfs.h>
@@ -157,7 +154,7 @@ zfs_unavail_pool(zpool_handle_t *zhp, void *data)
  * 1. physical match with no fs, no partition
  *	tag it top, partition disk
  *
- * 2. physical match again, see partion and tag
+ * 2. physical match again, see partition and tag
  *
  */
 
@@ -192,8 +189,8 @@ zfs_process_add(zpool_handle_t *zhp, nvlist_t *vdev, boolean_t labeled)
 	char rawpath[PATH_MAX], fullpath[PATH_MAX];
 	char devpath[PATH_MAX];
 	int ret;
-	int is_dm = 0;
+	boolean_t is_dm = B_FALSE;
-	int is_sd = 0;
+	boolean_t is_sd = B_FALSE;
 	uint_t c;
 	vdev_stat_t *vs;
 
@@ -221,8 +218,8 @@ zfs_process_add(zpool_handle_t *zhp, nvlist_t *vdev, boolean_t labeled)
 
 	is_dm = zfs_dev_is_dm(path);
 	zed_log_msg(LOG_INFO, "zfs_process_add: pool '%s' vdev '%s', phys '%s'"
-	    " wholedisk %d, dm %d (%llu)", zpool_get_name(zhp), path,
+	    " wholedisk %d, %s dm (guid %llu)", zpool_get_name(zhp), path,
-	    physpath ? physpath : "NULL", wholedisk, is_dm,
+	    physpath ? physpath : "NULL", wholedisk, is_dm ? "is" : "not",
 	    (long long unsigned int)guid);
 
 	/*
@@ -267,7 +264,7 @@ zfs_process_add(zpool_handle_t *zhp, nvlist_t *vdev, boolean_t labeled)
 	 * testing)
 	 */
 	if (physpath != NULL && strcmp("scsidebug", physpath) == 0)
-		is_sd = 1;
+		is_sd = B_TRUE;
 
 	/*
 	 * If the pool doesn't have the autoreplace property set, then use
@@ -438,7 +435,15 @@ zfs_process_add(zpool_handle_t *zhp, nvlist_t *vdev, boolean_t labeled)
 		return;
 	}
 
-	ret = zpool_vdev_attach(zhp, fullpath, path, nvroot, B_TRUE);
+	/*
+	 * Prefer sequential resilvering when supported (mirrors and dRAID),
+	 * otherwise fallback to a traditional healing resilver.
+	 */
+	ret = zpool_vdev_attach(zhp, fullpath, path, nvroot, B_TRUE, B_TRUE);
+	if (ret != 0) {
+		ret = zpool_vdev_attach(zhp, fullpath, path, nvroot,
+		    B_TRUE, B_FALSE);
+	}
 
 	zed_log_msg(LOG_INFO, "  zpool_vdev_replace: %s with %s (%s)",
 	    fullpath, path, (ret == 0) ? "no errors" :
@@ -534,7 +539,7 @@ zfs_iter_vdev(zpool_handle_t *zhp, nvlist_t *nvl, void *data)
 	(dp->dd_func)(zhp, nvl, dp->dd_islabeled);
 }
 
-void
+static void
 zfs_enable_ds(void *arg)
 {
 	unavailpool_t *pool = (unavailpool_t *)arg;
@@ -635,6 +640,27 @@ devid_iter(const char *devid, zfs_process_func_t func, boolean_t is_slice)
 	return (data.dd_found);
 }
 
+/*
+ * Given a device guid, find any vdevs with a matching guid.
+ */
+static boolean_t
+guid_iter(uint64_t pool_guid, uint64_t vdev_guid, const char *devid,
+    zfs_process_func_t func, boolean_t is_slice)
+{
+	dev_data_t data = { 0 };
+
+	data.dd_func = func;
+	data.dd_found = B_FALSE;
+	data.dd_pool_guid = pool_guid;
+	data.dd_vdev_guid = vdev_guid;
+	data.dd_islabeled = is_slice;
+	data.dd_new_devid = devid;
+
+	(void) zpool_iter(g_zfshdl, zfs_iter_pool, &data);
+
+	return (data.dd_found);
+}
+
 /*
  * Handle a EC_DEV_ADD.ESC_DISK event.
  *
@@ -658,15 +684,18 @@ static int
 zfs_deliver_add(nvlist_t *nvl, boolean_t is_lofi)
 {
 	char *devpath = NULL, *devid;
+	uint64_t pool_guid = 0, vdev_guid = 0;
 	boolean_t is_slice;
 
 	/*
-	 * Expecting a devid string and an optional physical location
+	 * Expecting a devid string and an optional physical location and guid
 	 */
 	if (nvlist_lookup_string(nvl, DEV_IDENTIFIER, &devid) != 0)
 		return (-1);
 
 	(void) nvlist_lookup_string(nvl, DEV_PHYS_PATH, &devpath);
+	(void) nvlist_lookup_uint64(nvl, ZFS_EV_POOL_GUID, &pool_guid);
+	(void) nvlist_lookup_uint64(nvl, ZFS_EV_VDEV_GUID, &vdev_guid);
 
 	is_slice = (nvlist_lookup_boolean(nvl, DEV_IS_PART) == 0);
 
@@ -674,15 +703,19 @@ zfs_deliver_add(nvlist_t *nvl, boolean_t is_lofi)
 	    devid, devpath ? devpath : "NULL", is_slice);
 
 	/*
-	 * Iterate over all vdevs looking for a match in the folllowing order:
+	 * Iterate over all vdevs looking for a match in the following order:
 	 * 1. ZPOOL_CONFIG_DEVID (identifies the unique disk)
 	 * 2. ZPOOL_CONFIG_PHYS_PATH (identifies disk physical location).
-	 *
+	 * 3. ZPOOL_CONFIG_GUID (identifies unique vdev).
-	 * For disks, we only want to pay attention to vdevs marked as whole
-	 * disks or are a multipath device.
 	 */
-	if (!devid_iter(devid, zfs_process_add, is_slice) && devpath != NULL)
+	if (devid_iter(devid, zfs_process_add, is_slice))
-		(void) devphys_iter(devpath, devid, zfs_process_add, is_slice);
+		return (0);
+	if (devpath != NULL && devphys_iter(devpath, devid, zfs_process_add,
+	    is_slice))
+		return (0);
+	if (vdev_guid != 0)
+		(void) guid_iter(pool_guid, vdev_guid, devid, zfs_process_add,
+		    is_slice);
 
 	return (0);
 }
@@ -892,7 +925,7 @@ zfs_enum_pools(void *arg)
  *
  * sent messages from zevents or udev monitor
  *
- * For now, each agent has it's own libzfs instance
+ * For now, each agent has its own libzfs instance
  */
 int
 zfs_slm_init()
@@ -913,6 +946,7 @@ zfs_slm_init()
 		return (-1);
 	}
 
+	pthread_setname_np(g_zfs_tid, "enum-pools");
 	list_create(&g_device_list, sizeof (struct pendingdev),
 	    offsetof(struct pendingdev, pd_node));
 

cmd/zed/agents/zfs_retire.c

@@ -38,8 +38,10 @@
 #include <sys/fs/zfs.h>
 #include <sys/fm/protocol.h>
 #include <sys/fm/fs/zfs.h>
+#include <libzutil.h>
 #include <libzfs.h>
 #include <string.h>
+#include <libgen.h>
 
 #include "zfs_agents.h"
 #include "fmd_api.h"
@@ -219,12 +221,18 @@ replace_with_spare(fmd_hdl_t *hdl, zpool_handle_t *zhp, nvlist_t *vdev)
 	 * replace it.
 	 */
 	for (s = 0; s < nspares; s++) {
-		char *spare_name;
+		boolean_t rebuild = B_FALSE;
+		char *spare_name, *type;
 
 		if (nvlist_lookup_string(spares[s], ZPOOL_CONFIG_PATH,
 		    &spare_name) != 0)
 			continue;
 
+		/* prefer sequential resilvering for distributed spares */
+		if ((nvlist_lookup_string(spares[s], ZPOOL_CONFIG_TYPE,
+		    &type) == 0) && strcmp(type, VDEV_TYPE_DRAID_SPARE) == 0)
+			rebuild = B_TRUE;
+
 		/* if set, add the "ashift" pool property to the spare nvlist */
 		if (source != ZPROP_SRC_DEFAULT)
 			(void) nvlist_add_uint64(spares[s],
@@ -234,10 +242,10 @@ replace_with_spare(fmd_hdl_t *hdl, zpool_handle_t *zhp, nvlist_t *vdev)
 		    ZPOOL_CONFIG_CHILDREN, &spares[s], 1);
 
 		fmd_hdl_debug(hdl, "zpool_vdev_replace '%s' with spare '%s'",
-		    dev_name, basename(spare_name));
+		    dev_name, zfs_basename(spare_name));
 
 		if (zpool_vdev_attach(zhp, dev_name, spare_name,
-		    replacement, B_TRUE) == 0) {
+		    replacement, B_TRUE, rebuild) == 0) {
 			free(dev_name);
 			nvlist_free(replacement);
 			return (B_TRUE);
@@ -319,12 +327,16 @@ zfs_retire_recv(fmd_hdl_t *hdl, fmd_event_t *ep, nvlist_t *nvl,
 
 	fmd_hdl_debug(hdl, "zfs_retire_recv: '%s'", class);
 
+	nvlist_lookup_uint64(nvl, FM_EREPORT_PAYLOAD_ZFS_VDEV_STATE, &state);
+
 	/*
 	 * If this is a resource notifying us of device removal then simply
 	 * check for an available spare and continue unless the device is a
 	 * l2arc vdev, in which case we just offline it.
 	 */
-	if (strcmp(class, "resource.fs.zfs.removed") == 0) {
+	if (strcmp(class, "resource.fs.zfs.removed") == 0 ||
+	    (strcmp(class, "resource.fs.zfs.statechange") == 0 &&
+	    (state == VDEV_STATE_REMOVED || state == VDEV_STATE_FAULTED))) {
 		char *devtype;
 		char *devname;
 
@@ -347,9 +359,8 @@ zfs_retire_recv(fmd_hdl_t *hdl, fmd_event_t *ep, nvlist_t *nvl,
 			zpool_vdev_offline(zhp, devname, B_TRUE);
 		} else if (!fmd_prop_get_int32(hdl, "spare_on_remove") ||
 		    replace_with_spare(hdl, zhp, vdev) == B_FALSE) {
-			/* Could not handle with spare: offline the device */
+			/* Could not handle with spare */
-			fmd_hdl_debug(hdl, "zpool_vdev_offline '%s'", devname);
+			fmd_hdl_debug(hdl, "no spare for '%s'", devname);
-			zpool_vdev_offline(zhp, devname, B_TRUE);
 		}
 
 		free(devname);
@@ -361,12 +372,11 @@ zfs_retire_recv(fmd_hdl_t *hdl, fmd_event_t *ep, nvlist_t *nvl,
 		return;
 
 	/*
-	 * Note: on zfsonlinux statechange events are more than just
+	 * Note: on Linux statechange events are more than just
 	 * healthy ones so we need to confirm the actual state value.
 	 */
 	if (strcmp(class, "resource.fs.zfs.statechange") == 0 &&
-	    nvlist_lookup_uint64(nvl, FM_EREPORT_PAYLOAD_ZFS_VDEV_STATE,
+	    state == VDEV_STATE_HEALTHY) {
-	    &state) == 0 && state == VDEV_STATE_HEALTHY) {
 		zfs_vdev_repair(hdl, nvl);
 		return;
 	}
@@ -497,6 +507,7 @@ zfs_retire_recv(fmd_hdl_t *hdl, fmd_event_t *ep, nvlist_t *nvl,
 		 * Attempt to substitute a hot spare.
 		 */
 		(void) replace_with_spare(hdl, zhp, vdev);
+
 		zpool_close(zhp);
 	}
 

cmd/zed/zed.c

@@ -1,9 +1,9 @@
 /*
- * This file is part of the ZFS Event Daemon (ZED)
+ * This file is part of the ZFS Event Daemon (ZED).
- * for ZFS on Linux (ZoL) <http://zfsonlinux.org/>.
+ *
  * Developed at Lawrence Livermore National Laboratory (LLNL-CODE-403049).
  * Copyright (C) 2013-2014 Lawrence Livermore National Security, LLC.
- * Refer to the ZoL git commit log for authoritative copyright attribution.
+ * Refer to the OpenZFS git commit log for authoritative copyright attribution.
  *
  * The contents of this file are subject to the terms of the
  * Common Development and Distribution License Version 1.0 (CDDL-1.0).
@@ -60,8 +60,8 @@ _setup_sig_handlers(void)
 		zed_log_die("Failed to initialize sigset");
 
 	sa.sa_flags = SA_RESTART;
-	sa.sa_handler = SIG_IGN;
 
+	sa.sa_handler = SIG_IGN;
 	if (sigaction(SIGPIPE, &sa, NULL) < 0)
 		zed_log_die("Failed to ignore SIGPIPE");
 
@@ -75,6 +75,10 @@ _setup_sig_handlers(void)
 	sa.sa_handler = _hup_handler;
 	if (sigaction(SIGHUP, &sa, NULL) < 0)
 		zed_log_die("Failed to register SIGHUP handler");
+
+	(void) sigaddset(&sa.sa_mask, SIGCHLD);
+	if (pthread_sigmask(SIG_BLOCK, &sa.sa_mask, NULL) < 0)
+		zed_log_die("Failed to block SIGCHLD");
 }
 
 /*
@@ -212,22 +216,20 @@ _finish_daemonize(void)
 int
 main(int argc, char *argv[])
 {
-	struct zed_conf *zcp;
+	struct zed_conf zcp;
 	uint64_t saved_eid;
 	int64_t saved_etime[2];
 
 	zed_log_init(argv[0]);
 	zed_log_stderr_open(LOG_NOTICE);
-	zcp = zed_conf_create();
+	zed_conf_init(&zcp);
-	zed_conf_parse_opts(zcp, argc, argv);
+	zed_conf_parse_opts(&zcp, argc, argv);
-	if (zcp->do_verbose)
+	if (zcp.do_verbose)
 		zed_log_stderr_open(LOG_INFO);
 
 	if (geteuid() != 0)
 		zed_log_die("Must be run as root");
 
-	zed_conf_parse_file(zcp);
-
 	zed_file_close_from(STDERR_FILENO + 1);
 
 	(void) umask(0);
@@ -235,47 +237,72 @@ main(int argc, char *argv[])
 	if (chdir("/") < 0)
 		zed_log_die("Failed to change to root directory");
 
-	if (zed_conf_scan_dir(zcp) < 0)
+	if (zed_conf_scan_dir(&zcp) < 0)
 		exit(EXIT_FAILURE);
 
-	if (!zcp->do_foreground) {
+	if (!zcp.do_foreground) {
 		_start_daemonize();
 		zed_log_syslog_open(LOG_DAEMON);
 	}
 	_setup_sig_handlers();
 
-	if (zcp->do_memlock)
+	if (zcp.do_memlock)
 		_lock_memory();
 
-	if ((zed_conf_write_pid(zcp) < 0) && (!zcp->do_force))
+	if ((zed_conf_write_pid(&zcp) < 0) && (!zcp.do_force))
 		exit(EXIT_FAILURE);
 
-	if (!zcp->do_foreground)
+	if (!zcp.do_foreground)
 		_finish_daemonize();
 
 	zed_log_msg(LOG_NOTICE,
 	    "ZFS Event Daemon %s-%s (PID %d)",
 	    ZFS_META_VERSION, ZFS_META_RELEASE, (int)getpid());
 
-	if (zed_conf_open_state(zcp) < 0)
+	if (zed_conf_open_state(&zcp) < 0)
 		exit(EXIT_FAILURE);
 
-	if (zed_conf_read_state(zcp, &saved_eid, saved_etime) < 0)
+	if (zed_conf_read_state(&zcp, &saved_eid, saved_etime) < 0)
 		exit(EXIT_FAILURE);
 
-	zed_event_init(zcp);
+idle:
-	zed_event_seek(zcp, saved_eid, saved_etime);
+	/*
+	 * If -I is specified, attempt to open /dev/zfs repeatedly until
+	 * successful.
+	 */
+	do {
+		if (!zed_event_init(&zcp))
+			break;
+		/* Wait for some time and try again. tunable? */
+		sleep(30);
+	} while (!_got_exit && zcp.do_idle);
+
+	if (_got_exit)
+		goto out;
+
+	zed_event_seek(&zcp, saved_eid, saved_etime);
 
 	while (!_got_exit) {
+		int rv;
 		if (_got_hup) {
 			_got_hup = 0;
-			(void) zed_conf_scan_dir(zcp);
+			(void) zed_conf_scan_dir(&zcp);
 		}
-		zed_event_service(zcp);
+		rv = zed_event_service(&zcp);
+
+		/* ENODEV: When kernel module is unloaded (osx) */
+		if (rv != 0)
+			break;
 	}
+
 	zed_log_msg(LOG_NOTICE, "Exiting");
-	zed_event_fini(zcp);
+	zed_event_fini(&zcp);
-	zed_conf_destroy(zcp);
+
+	if (zcp.do_idle && !_got_exit)
+		goto idle;
+
+out:
+	zed_conf_destroy(&zcp);
 	zed_log_fini();
 	exit(EXIT_SUCCESS);
 }

cmd/zed/zed.d/Makefile.am

@@ -0,0 +1,57 @@
+include $(top_srcdir)/config/Rules.am
+include $(top_srcdir)/config/Substfiles.am
+include $(top_srcdir)/config/Shellcheck.am
+
+EXTRA_DIST += README
+
+zedconfdir = $(sysconfdir)/zfs/zed.d
+
+dist_zedconf_DATA = \
+	zed-functions.sh \
+	zed.rc
+
+zedexecdir = $(zfsexecdir)/zed.d
+
+dist_zedexec_SCRIPTS = \
+	all-debug.sh \
+	all-syslog.sh \
+	data-notify.sh \
+	generic-notify.sh \
+	resilver_finish-notify.sh \
+	scrub_finish-notify.sh \
+	statechange-led.sh \
+	statechange-notify.sh \
+	vdev_clear-led.sh \
+	vdev_attach-led.sh \
+	pool_import-led.sh \
+	resilver_finish-start-scrub.sh \
+	trim_finish-notify.sh
+
+nodist_zedexec_SCRIPTS = history_event-zfs-list-cacher.sh
+
+SUBSTFILES += $(nodist_zedexec_SCRIPTS)
+
+zedconfdefaults = \
+	all-syslog.sh \
+	data-notify.sh \
+	history_event-zfs-list-cacher.sh \
+	resilver_finish-notify.sh \
+	scrub_finish-notify.sh \
+	statechange-led.sh \
+	statechange-notify.sh \
+	vdev_clear-led.sh \
+	vdev_attach-led.sh \
+	pool_import-led.sh \
+	resilver_finish-start-scrub.sh
+
+install-data-hook:
+	$(MKDIR_P) "$(DESTDIR)$(zedconfdir)"
+	for f in $(zedconfdefaults); do \
+	  test -f "$(DESTDIR)$(zedconfdir)/$${f}" -o \
+	       -L "$(DESTDIR)$(zedconfdir)/$${f}" || \
+	    ln -s "$(zedexecdir)/$${f}" "$(DESTDIR)$(zedconfdir)"; \
+	done
+	chmod 0600 "$(DESTDIR)$(zedconfdir)/zed.rc"
+
+# False positive: 1>&"${ZED_FLOCK_FD}" looks suspiciously similar to a >&filename bash extension
+CHECKBASHISMS_IGNORE = -e 'should be >word 2>&1' -e '&"$${ZED_FLOCK_FD}"'

cmd/zed/zed.d/all-debug.sh

@@ -12,15 +12,11 @@
 
 zed_exit_if_ignoring_this_event
 
-lockfile="$(basename -- "${ZED_DEBUG_LOG}").lock"
+zed_lock "${ZED_DEBUG_LOG}"
+{
+	printenv | sort
+	echo
+} 1>&"${ZED_FLOCK_FD}"
+zed_unlock "${ZED_DEBUG_LOG}"
 
-umask 077
-zed_lock "${lockfile}"
-exec >> "${ZED_DEBUG_LOG}"
-
-printenv | sort
-echo
-
-exec >&-
-zed_unlock "${lockfile}"
 exit 0

cmd/zed/zed.d/all-syslog.sh

@@ -1,14 +1,51 @@
 #!/bin/sh
+#
+# Copyright (C) 2013-2014 Lawrence Livermore National Security, LLC.
+# Copyright (c) 2020 by Delphix. All rights reserved.
+#
+
 #
 # Log the zevent via syslog.
+#
 
 [ -f "${ZED_ZEDLET_DIR}/zed.rc" ] && . "${ZED_ZEDLET_DIR}/zed.rc"
 . "${ZED_ZEDLET_DIR}/zed-functions.sh"
 
 zed_exit_if_ignoring_this_event
 
-zed_log_msg "eid=${ZEVENT_EID}" "class=${ZEVENT_SUBCLASS}" \
+# build a string of name=value pairs for this event
-    "${ZEVENT_POOL_GUID:+"pool_guid=${ZEVENT_POOL_GUID}"}" \
+msg="eid=${ZEVENT_EID} class=${ZEVENT_SUBCLASS}"
-    "${ZEVENT_VDEV_PATH:+"vdev_path=${ZEVENT_VDEV_PATH}"}" \
+
-    "${ZEVENT_VDEV_STATE_STR:+"vdev_state=${ZEVENT_VDEV_STATE_STR}"}"
+if [ "${ZED_SYSLOG_DISPLAY_GUIDS}" = "1" ]; then
+    [ -n "${ZEVENT_POOL_GUID}" ] && msg="${msg} pool_guid=${ZEVENT_POOL_GUID}"
+    [ -n "${ZEVENT_VDEV_GUID}" ] && msg="${msg} vdev_guid=${ZEVENT_VDEV_GUID}"
+else
+    [ -n "${ZEVENT_POOL}" ] && msg="${msg} pool='${ZEVENT_POOL}'"
+    [ -n "${ZEVENT_VDEV_PATH}" ] && msg="${msg} vdev=$(basename "${ZEVENT_VDEV_PATH}")"
+fi
+
+# log pool state if state is anything other than 'ACTIVE'
+[ -n "${ZEVENT_POOL_STATE_STR}" ] && [ "$ZEVENT_POOL_STATE" -ne 0 ] && \
+    msg="${msg} pool_state=${ZEVENT_POOL_STATE_STR}"
+
+# Log the following payload nvpairs if they are present
+[ -n "${ZEVENT_VDEV_STATE_STR}" ]  && msg="${msg} vdev_state=${ZEVENT_VDEV_STATE_STR}"
+[ -n "${ZEVENT_CKSUM_ALGORITHM}" ] && msg="${msg} algorithm=${ZEVENT_CKSUM_ALGORITHM}"
+[ -n "${ZEVENT_ZIO_SIZE}" ]        && msg="${msg} size=${ZEVENT_ZIO_SIZE}"
+[ -n "${ZEVENT_ZIO_OFFSET}" ]      && msg="${msg} offset=${ZEVENT_ZIO_OFFSET}"
+[ -n "${ZEVENT_ZIO_PRIORITY}" ]    && msg="${msg} priority=${ZEVENT_ZIO_PRIORITY}"
+[ -n "${ZEVENT_ZIO_ERR}" ]         && msg="${msg} err=${ZEVENT_ZIO_ERR}"
+[ -n "${ZEVENT_ZIO_FLAGS}" ]       && msg="${msg} flags=$(printf '0x%x' "${ZEVENT_ZIO_FLAGS}")"
+
+# log delays that are >= 10 milisec
+[ -n "${ZEVENT_ZIO_DELAY}" ] && [ "$ZEVENT_ZIO_DELAY" -gt 10000000 ] && \
+    msg="${msg} delay=$((ZEVENT_ZIO_DELAY / 1000000))ms"
+
+# list the bookmark data together
+# shellcheck disable=SC2153
+[ -n "${ZEVENT_ZIO_OBJSET}" ] && \
+    msg="${msg} bookmark=${ZEVENT_ZIO_OBJSET}:${ZEVENT_ZIO_OBJECT}:${ZEVENT_ZIO_LEVEL}:${ZEVENT_ZIO_BLKID}"
+
+zed_log_msg "${msg}"
+
 exit 0

cmd/zed/zed.d/data-notify.sh

@@ -25,7 +25,7 @@ zed_rate_limit "${rate_limit_tag}" || exit 3
 
 umask 077
 note_subject="ZFS ${ZEVENT_SUBCLASS} error for ${ZEVENT_POOL} on $(hostname)"
-note_pathname="${TMPDIR:="/tmp"}/$(basename -- "$0").${ZEVENT_EID}.$$"
+note_pathname="$(mktemp)"
 {
     echo "ZFS has detected a data error:"
     echo

cmd/zed/zed.d/generic-notify.sh

@@ -31,7 +31,7 @@ umask 077
 pool_str="${ZEVENT_POOL:+" for ${ZEVENT_POOL}"}"
 host_str=" on $(hostname)"
 note_subject="ZFS ${ZEVENT_SUBCLASS} event${pool_str}${host_str}"
-note_pathname="${TMPDIR:="/tmp"}/$(basename -- "$0").${ZEVENT_EID}.$$"
+note_pathname="$(mktemp)"
 {
     echo "ZFS has posted the following event:"
     echo

cmd/zed/zed.d/history_event-zfs-list-cacher.sh.in

@@ -3,9 +3,8 @@
 # Track changes to enumerated pools for use in early-boot
 set -ef
 
-FSLIST_DIR="@sysconfdir@/zfs/zfs-list.cache"
+FSLIST="@sysconfdir@/zfs/zfs-list.cache/${ZEVENT_POOL}"
-FSLIST_TMP="@runstatedir@/zfs-list.cache.new"
+FSLIST_TMP="@runstatedir@/zfs-list.cache@${ZEVENT_POOL}"
-FSLIST="${FSLIST_DIR}/${ZEVENT_POOL}"
 
 # If the pool specific cache file is not writeable, abort
 [ -w "${FSLIST}" ] || exit 0
@@ -13,21 +12,21 @@ FSLIST="${FSLIST_DIR}/${ZEVENT_POOL}"
 [ -f "${ZED_ZEDLET_DIR}/zed.rc" ] && . "${ZED_ZEDLET_DIR}/zed.rc"
 . "${ZED_ZEDLET_DIR}/zed-functions.sh"
 
-zed_exit_if_ignoring_this_event
+[ "$ZEVENT_SUBCLASS" != "history_event" ] && exit 0
-zed_check_cmd "${ZFS}" sort diff grep
+zed_check_cmd "${ZFS}" sort diff
 
 # If we are acting on a snapshot, we have nothing to do
-printf '%s' "${ZEVENT_HISTORY_DSNAME}" | grep '@' && exit 0
+[ "${ZEVENT_HISTORY_DSNAME%@*}" = "${ZEVENT_HISTORY_DSNAME}" ] || exit 0
 
-# We obtain a lock on zfs-list to avoid any simultaneous writes.
+# We lock the output file to avoid simultaneous writes.
 # If we run into trouble, log and drop the lock
 abort_alter() {
-  zed_log_msg "Error updating zfs-list.cache!"
+  zed_log_msg "Error updating zfs-list.cache for ${ZEVENT_POOL}!"
-  zed_unlock zfs-list
+  zed_unlock "${FSLIST}"
 }
 
 finished() {
-  zed_unlock zfs-list
+  zed_unlock "${FSLIST}"
   trap - EXIT
   exit 0
 }
@@ -37,7 +36,7 @@ case "${ZEVENT_HISTORY_INTERNAL_NAME}" in
       ;;
 
     export)
-        zed_lock zfs-list
+        zed_lock "${FSLIST}"
         trap abort_alter EXIT
         echo > "${FSLIST}"
         finished
@@ -46,8 +45,13 @@ case "${ZEVENT_HISTORY_INTERNAL_NAME}" in
     set|inherit)
         # Only act if one of the tracked properties is altered.
         case "${ZEVENT_HISTORY_INTERNAL_STR%%=*}" in
-            canmount|mountpoint|atime|relatime|devices|exec| \
+            canmount|mountpoint|atime|relatime|devices|exec|readonly| \
-                readonly|setuid|nbmand) ;;
+              setuid|nbmand|encroot|keylocation|org.openzfs.systemd:requires| \
+              org.openzfs.systemd:requires-mounts-for| \
+              org.openzfs.systemd:before|org.openzfs.systemd:after| \
+              org.openzfs.systemd:wanted-by|org.openzfs.systemd:required-by| \
+              org.openzfs.systemd:nofail|org.openzfs.systemd:ignore \
+            ) ;;
             *) exit 0 ;;
         esac
       ;;
@@ -58,11 +62,15 @@ case "${ZEVENT_HISTORY_INTERNAL_NAME}" in
       ;;
 esac
 
-zed_lock zfs-list
+zed_lock "${FSLIST}"
 trap abort_alter EXIT
 
-PROPS="name,mountpoint,canmount,atime,relatime,devices,exec,readonly"
+PROPS="name,mountpoint,canmount,atime,relatime,devices,exec\
-PROPS="${PROPS},setuid,nbmand"
+,readonly,setuid,nbmand,encroot,keylocation\
+,org.openzfs.systemd:requires,org.openzfs.systemd:requires-mounts-for\
+,org.openzfs.systemd:before,org.openzfs.systemd:after\
+,org.openzfs.systemd:wanted-by,org.openzfs.systemd:required-by\
+,org.openzfs.systemd:nofail,org.openzfs.systemd:ignore"
 
 "${ZFS}" list -H -t filesystem -o $PROPS -r "${ZEVENT_POOL}" > "${FSLIST_TMP}"
 
@@ -70,7 +78,7 @@ PROPS="${PROPS},setuid,nbmand"
 sort "${FSLIST_TMP}" -o "${FSLIST_TMP}"
 
 # Don't modify the file if it hasn't changed
-diff -q "${FSLIST_TMP}" "${FSLIST}" || mv "${FSLIST_TMP}" "${FSLIST}"
+diff -q "${FSLIST_TMP}" "${FSLIST}" || cat "${FSLIST_TMP}" > "${FSLIST}"
 rm -f "${FSLIST_TMP}"
 
 finished

cmd/zed/zed.d/resilver_finish-start-scrub.sh

@@ -5,10 +5,12 @@
 # Exit codes:
 # 1: Internal error
 # 2: Script wasn't enabled in zed.rc
+# 3: Scrubs are automatically started for sequential resilvers
 [ -f "${ZED_ZEDLET_DIR}/zed.rc" ] && . "${ZED_ZEDLET_DIR}/zed.rc"
 . "${ZED_ZEDLET_DIR}/zed-functions.sh"
 
 [ "${ZED_SCRUB_AFTER_RESILVER}" = "1" ] || exit 2
+[ "${ZEVENT_RESILVER_TYPE}" != "sequential" ] || exit 3
 [ -n "${ZEVENT_POOL}" ] || exit 1
 [ -n "${ZEVENT_SUBCLASS}" ] || exit 1
 zed_check_cmd "${ZPOOL}" || exit 1

cmd/zed/zed.d/scrub_finish-notify.sh

@@ -41,7 +41,7 @@ fi
 
 umask 077
 note_subject="ZFS ${ZEVENT_SUBCLASS} event for ${ZEVENT_POOL} on $(hostname)"
-note_pathname="${TMPDIR:="/tmp"}/$(basename -- "$0").${ZEVENT_EID}.$$"
+note_pathname="$(mktemp)"
 {
     echo "ZFS has finished a ${action}:"
     echo

cmd/zed/zed.d/statechange-led.sh

@@ -1,26 +1,26 @@
 #!/bin/sh
 #
-# Turn off/on the VDEV's enclosure fault LEDs when the pool's state changes.
+# Turn off/on vdevs' enclosure fault LEDs when their pool's state changes.
 #
-# Turn the VDEV's fault LED on if it becomes FAULTED, DEGRADED or UNAVAIL.
+# Turn a vdev's fault LED on if it becomes FAULTED, DEGRADED or UNAVAIL.
-# Turn the LED off when it's back ONLINE again.
+# Turn its LED off when it's back ONLINE again.
 #
 # This script run in two basic modes:
 #
 # 1. If $ZEVENT_VDEV_ENC_SYSFS_PATH and $ZEVENT_VDEV_STATE_STR are set, then
-# only set the LED for that particular VDEV. This is the case for statechange
+# only set the LED for that particular vdev. This is the case for statechange
 # events and some vdev_* events.
 #
-# 2. If those vars are not set, then check the state of all VDEVs in the pool
+# 2. If those vars are not set, then check the state of all vdevs in the pool
 # and set the LEDs accordingly.  This is the case for pool_import events.
 #
 # Note that this script requires that your enclosure be supported by the
-# Linux SCSI enclosure services (ses) driver.  The script will do nothing
+# Linux SCSI Enclosure services (SES) driver.  The script will do nothing
 # if you have no enclosure, or if your enclosure isn't supported.
 #
 # Exit codes:
 #   0: enclosure led successfully set
-#   1: enclosure leds not not available
+#   1: enclosure leds not available
 #   2: enclosure leds administratively disabled
 #   3: The led sysfs path passed from ZFS does not exist
 #   4: $ZPOOL not set
@@ -29,7 +29,8 @@
 [ -f "${ZED_ZEDLET_DIR}/zed.rc" ] && . "${ZED_ZEDLET_DIR}/zed.rc"
 . "${ZED_ZEDLET_DIR}/zed-functions.sh"
 
-if [ ! -d /sys/class/enclosure ] ; then
+if [ ! -d /sys/class/enclosure ] && [ ! -d /sys/bus/pci/slots ] ; then
+	# No JBOD enclosure or NVMe slots
 	exit 1
 fi
 
@@ -59,6 +60,10 @@ check_and_set_led()
 	file="$1"
 	val="$2"
 
+	if [ -z "$val" ]; then
+		return 0
+	fi
+
 	if [ ! -e "$file" ] ; then
 		return 3
 	fi
@@ -66,11 +71,11 @@ check_and_set_led()
 	# If another process is accessing the LED when we attempt to update it,
 	# the update will be lost so retry until the LED actually changes or we
 	# timeout.
-	for _ in $(seq 1 5); do
+	for _ in 1 2 3 4 5; do
 		# We want to check the current state first, since writing to the
-		# 'fault' entry always always causes a SES command, even if the
+		# 'fault' entry always causes a SES command, even if the
 		# current state is already what you want.
-		current=$(cat "${file}")
+		read -r current < "${file}"
 
 		# On some enclosures if you write 1 to fault, and read it back,
 		# it will return 2.  Treat all non-zero values as 1 for
@@ -85,27 +90,84 @@ check_and_set_led()
 		else
 			break
 		fi
-        done
+	done
+}
+
+# Fault LEDs for JBODs and NVMe drives are handled a little differently.
+#
+# On JBODs the fault LED is called 'fault' and on a path like this:
+#
+#   /sys/class/enclosure/0:0:1:0/SLOT 10/fault
+#
+# On NVMe it's called 'attention' and on a path like this:
+#
+#   /sys/bus/pci/slot/0/attention
+#
+# This function returns the full path to the fault LED file for a given
+# enclosure/slot directory.
+#
+path_to_led()
+{
+	dir=$1
+	if [ -f "$dir/fault" ] ; then
+		echo "$dir/fault"
+	elif [ -f "$dir/attention" ] ; then
+		echo "$dir/attention"
+	fi
 }
 
 state_to_val()
 {
 	state="$1"
-	if [ "$state" = "FAULTED" ] || [ "$state" = "DEGRADED" ] || \
+	case "$state" in
-	   [ "$state" = "UNAVAIL" ] ; then
+		FAULTED|DEGRADED|UNAVAIL)
-		echo 1
+			echo 1
-	elif [ "$state" = "ONLINE" ] ; then
+			;;
-		echo 0
+		ONLINE)
-	fi
+			echo 0
+			;;
+	esac
 }
 
-# process_pool ([pool])
 #
-# Iterate through a pool (or pools) and set the VDEV's enclosure slot LEDs to
+# Given a nvme name like 'nvme0n1', pass back its slot directory
-# the VDEV's state.
+# like "/sys/bus/pci/slots/0"
+#
+nvme_dev_to_slot()
+{
+	dev="$1"
+
+	# Get the address "0000:01:00.0"
+	address=$(cat "/sys/class/block/$dev/device/address")
+
+	# For each /sys/bus/pci/slots subdir that is an actual number
+	# (rather than weird directories like "1-3/").
+	# shellcheck disable=SC2010
+	for i in $(ls /sys/bus/pci/slots/ | grep -E "^[0-9]+$") ; do
+		this_address=$(cat "/sys/bus/pci/slots/$i/address")
+
+		# The format of address is a little different between
+		# /sys/class/block/$dev/device/address and
+		# /sys/bus/pci/slots/
+		#
+		# address=           "0000:01:00.0"
+		# this_address =     "0000:01:00"
+		#
+		if echo "$address" | grep -Eq ^"$this_address" ; then
+			echo "/sys/bus/pci/slots/$i"
+			break
+		fi
+	done
+}
+
+
+# process_pool (pool)
+#
+# Iterate through a pool and set the vdevs' enclosure slot LEDs to
+# those vdevs' state.
 #
 # Arguments
-#   pool:	Optional pool name.  If not specified, iterate though all pools.
+#   pool:	Pool name.
 #
 # Return
 #  0 on success, 3 on missing sysfs path
@@ -113,19 +175,27 @@ state_to_val()
 process_pool()
 {
 	pool="$1"
+
+	# The output will be the vdevs only (from "grep '/dev/'"):
+	#
+	#    U45     ONLINE       0     0     0   /dev/sdk          0
+	#    U46     ONLINE       0     0     0   /dev/sdm          0
+	#    U47     ONLINE       0     0     0   /dev/sdn          0
+	#    U50     ONLINE       0     0     0  /dev/sdbn          0
+	#
+	ZPOOL_SCRIPTS_AS_ROOT=1 $ZPOOL status -c upath,fault_led "$pool" | grep '/dev/' | (
 	rc=0
-
+	while read -r vdev state _ _ _ therest; do
-	# Lookup all the current LED values and paths in parallel
-	#shellcheck disable=SC2016
-	cmd='echo led_token=$(cat "$VDEV_ENC_SYSFS_PATH/fault"),"$VDEV_ENC_SYSFS_PATH",'
-	out=$($ZPOOL status -vc "$cmd" "$pool" | grep 'led_token=')
-
-	#shellcheck disable=SC2034
-	echo "$out" | while read -r vdev state read write chksum therest; do
 		# Read out current LED value and path
-		tmp=$(echo "$therest" | sed 's/^.*led_token=//g')
+		# Get dev name (like 'sda')
-		vdev_enc_sysfs_path=$(echo "$tmp" | awk -F ',' '{print $2}')
+		dev=$(basename "$(echo "$therest" | awk '{print $(NF-1)}')")
-		current_val=$(echo "$tmp" | awk -F ',' '{print $1}')
+		vdev_enc_sysfs_path=$(realpath "/sys/class/block/$dev/device/enclosure_device"*)
+		if [ ! -d "$vdev_enc_sysfs_path" ] ; then
+			# This is not a JBOD disk, but it could be a PCI NVMe drive
+			vdev_enc_sysfs_path=$(nvme_dev_to_slot "$dev")
+		fi
+
+		current_val=$(echo "$therest" | awk '{print $NF}')
 
 		if [ "$current_val" != "0" ] ; then
 			current_val=1
@@ -136,40 +206,33 @@ process_pool()
 			continue
 		fi
 
-		if [ ! -e "$vdev_enc_sysfs_path/fault" ] ; then
+		led_path=$(path_to_led "$vdev_enc_sysfs_path")
-			#shellcheck disable=SC2030
+		if [ ! -e "$led_path" ] ; then
-			rc=1
+			rc=3
-			zed_log_msg "vdev $vdev '$file/fault' doesn't exist"
+			zed_log_msg "vdev $vdev '$led_path' doesn't exist"
-			continue;
+			continue
 		fi
 
 		val=$(state_to_val "$state")
 
 		if [ "$current_val" = "$val" ] ; then
 			# LED is already set correctly
-			continue;
+			continue
 		fi
 
-		if ! check_and_set_led "$vdev_enc_sysfs_path/fault" "$val"; then
+		if ! check_and_set_led "$led_path" "$val"; then
-			rc=1
+			rc=3
 		fi
-
 	done
-
+	exit "$rc"; )
-	#shellcheck disable=SC2031
-	if [ "$rc" = "0" ] ; then
-		return 0
-	else
-		# We didn't see a sysfs entry that we wanted to set
-		return 3
-	fi
 }
 
 if [ -n "$ZEVENT_VDEV_ENC_SYSFS_PATH" ] && [ -n "$ZEVENT_VDEV_STATE_STR" ] ; then
-	# Got a statechange for an individual VDEV
+	# Got a statechange for an individual vdev
 	val=$(state_to_val "$ZEVENT_VDEV_STATE_STR")
 	vdev=$(basename "$ZEVENT_VDEV_PATH")
-	check_and_set_led "$ZEVENT_VDEV_ENC_SYSFS_PATH/fault" "$val"
+	ledpath=$(path_to_led "$ZEVENT_VDEV_ENC_SYSFS_PATH")
+	check_and_set_led "$ledpath" "$val"
 else
 	# Process the entire pool
 	poolname=$(zed_guid_to_pool "$ZEVENT_POOL_GUID")

cmd/zed/zed.d/statechange-notify.sh

@@ -15,7 +15,7 @@
 # Send notification in response to a fault induced statechange
 #
 # ZEVENT_SUBCLASS: 'statechange'
-# ZEVENT_VDEV_STATE_STR: 'DEGRADED', 'FAULTED' or 'REMOVED'
+# ZEVENT_VDEV_STATE_STR: 'DEGRADED', 'FAULTED', 'REMOVED', or 'UNAVAIL'
 #
 # Exit codes:
 #   0: notification sent
@@ -31,13 +31,14 @@
 
 if [ "${ZEVENT_VDEV_STATE_STR}" != "FAULTED" ] \
         && [ "${ZEVENT_VDEV_STATE_STR}" != "DEGRADED" ] \
-        && [ "${ZEVENT_VDEV_STATE_STR}" != "REMOVED" ]; then
+        && [ "${ZEVENT_VDEV_STATE_STR}" != "REMOVED" ] \
+        && [ "${ZEVENT_VDEV_STATE_STR}" != "UNAVAIL" ]; then
     exit 3
 fi
 
 umask 077
 note_subject="ZFS device fault for pool ${ZEVENT_POOL_GUID} on $(hostname)"
-note_pathname="${TMPDIR:="/tmp"}/$(basename -- "$0").${ZEVENT_EID}.$$"
+note_pathname="$(mktemp)"
 {
     if [ "${ZEVENT_VDEV_STATE_STR}" = "FAULTED" ] ; then
         echo "The number of I/O errors associated with a ZFS device exceeded"

cmd/zed/zed.d/trim_finish-notify.sh

@@ -0,0 +1,37 @@
+#!/bin/sh
+#
+# Send notification in response to a TRIM_FINISH. The event
+# will be received for each vdev in the pool which was trimmed.
+#
+# Exit codes:
+#   0: notification sent
+#   1: notification failed
+#   2: notification not configured
+#   9: internal error
+
+[ -f "${ZED_ZEDLET_DIR}/zed.rc" ] && . "${ZED_ZEDLET_DIR}/zed.rc"
+. "${ZED_ZEDLET_DIR}/zed-functions.sh"
+
+[ -n "${ZEVENT_POOL}" ] || exit 9
+[ -n "${ZEVENT_SUBCLASS}" ] || exit 9
+
+zed_check_cmd "${ZPOOL}" || exit 9
+
+umask 077
+note_subject="ZFS ${ZEVENT_SUBCLASS} event for ${ZEVENT_POOL} on $(hostname)"
+note_pathname="$(mktemp)"
+{
+    echo "ZFS has finished a trim:"
+    echo
+    echo "   eid: ${ZEVENT_EID}"
+    echo " class: ${ZEVENT_SUBCLASS}"
+    echo "  host: $(hostname)"
+    echo "  time: ${ZEVENT_TIME_STRING}"
+
+    "${ZPOOL}" status -t "${ZEVENT_POOL}"
+
+} > "${note_pathname}"
+
+zed_notify "${note_subject}" "${note_pathname}"; rv=$?
+rm -f "${note_pathname}"
+exit "${rv}"

cmd/zed/zed.d/zed-functions.sh

@@ -126,10 +126,8 @@ zed_lock()
 
     # Obtain a lock on the file bound to the given file descriptor.
     #
-    eval "exec ${fd}> '${lockfile}'"
+    eval "exec ${fd}>> '${lockfile}'"
-    err="$(flock --exclusive "${fd}" 2>&1)"
+    if ! err="$(flock --exclusive "${fd}" 2>&1)"; then
-    # shellcheck disable=SC2181
-    if [ $? -ne 0 ]; then
         zed_log_err "failed to lock \"${lockfile}\": ${err}"
     fi
 
@@ -165,9 +163,7 @@ zed_unlock()
     fi
 
     # Release the lock and close the file descriptor.
-    err="$(flock --unlock "${fd}" 2>&1)"
+    if ! err="$(flock --unlock "${fd}" 2>&1)"; then
-    # shellcheck disable=SC2181
-    if [ $? -ne 0 ]; then
         zed_log_err "failed to unlock \"${lockfile}\": ${err}"
     fi
     eval "exec ${fd}>&-"
@@ -202,6 +198,14 @@ zed_notify()
     [ "${rv}" -eq 0 ] && num_success=$((num_success + 1))
     [ "${rv}" -eq 1 ] && num_failure=$((num_failure + 1))
 
+    zed_notify_slack_webhook "${subject}" "${pathname}"; rv=$?
+    [ "${rv}" -eq 0 ] && num_success=$((num_success + 1))
+    [ "${rv}" -eq 1 ] && num_failure=$((num_failure + 1))
+
+    zed_notify_pushover "${subject}" "${pathname}"; rv=$?
+    [ "${rv}" -eq 0 ] && num_success=$((num_success + 1))
+    [ "${rv}" -eq 1 ] && num_failure=$((num_failure + 1))
+
     [ "${num_success}" -gt 0 ] && return 0
     [ "${num_failure}" -gt 0 ] && return 1
     return 2
@@ -263,7 +267,7 @@ zed_notify_email()
                 -e "s/@SUBJECT@/${subject}/g")"
 
     # shellcheck disable=SC2086
-    eval "${ZED_EMAIL_PROG}" ${ZED_EMAIL_OPTS} < "${pathname}" >/dev/null 2>&1
+    eval ${ZED_EMAIL_PROG} ${ZED_EMAIL_OPTS} < "${pathname}" >/dev/null 2>&1
     rv=$?
     if [ "${rv}" -ne 0 ]; then
         zed_log_err "$(basename "${ZED_EMAIL_PROG}") exit=${rv}"
@@ -359,6 +363,158 @@ zed_notify_pushbullet()
 }
 
 
+# zed_notify_slack_webhook (subject, pathname)
+#
+# Notification via Slack Webhook <https://api.slack.com/incoming-webhooks>.
+# The Webhook URL (ZED_SLACK_WEBHOOK_URL) identifies this client to the
+# Slack channel.
+#
+# Requires awk, curl, and sed executables to be installed in the standard PATH.
+#
+# References
+#   https://api.slack.com/incoming-webhooks
+#
+# Arguments
+#   subject: notification subject
+#   pathname: pathname containing the notification message (OPTIONAL)
+#
+# Globals
+#   ZED_SLACK_WEBHOOK_URL
+#
+# Return
+#   0: notification sent
+#   1: notification failed
+#   2: not configured
+#
+zed_notify_slack_webhook()
+{
+    [ -n "${ZED_SLACK_WEBHOOK_URL}" ] || return 2
+
+    local subject="$1"
+    local pathname="${2:-"/dev/null"}"
+    local msg_body
+    local msg_tag
+    local msg_json
+    local msg_out
+    local msg_err
+    local url="${ZED_SLACK_WEBHOOK_URL}"
+
+    [ -n "${subject}" ] || return 1
+    if [ ! -r "${pathname}" ]; then
+        zed_log_err "slack webhook cannot read \"${pathname}\""
+        return 1
+    fi
+
+    zed_check_cmd "awk" "curl" "sed" || return 1
+
+    # Escape the following characters in the message body for JSON:
+    # newline, backslash, double quote, horizontal tab, vertical tab,
+    # and carriage return.
+    #
+    msg_body="$(awk '{ ORS="\\n" } { gsub(/\\/, "\\\\"); gsub(/"/, "\\\"");
+        gsub(/\t/, "\\t"); gsub(/\f/, "\\f"); gsub(/\r/, "\\r"); print }' \
+        "${pathname}")"
+
+    # Construct the JSON message for posting.
+    #
+    msg_json="$(printf '{"text": "*%s*\n%s"}' "${subject}" "${msg_body}" )"
+
+    # Send the POST request and check for errors.
+    #
+    msg_out="$(curl -X POST "${url}" \
+        --header "Content-Type: application/json" --data-binary "${msg_json}" \
+        2>/dev/null)"; rv=$?
+    if [ "${rv}" -ne 0 ]; then
+        zed_log_err "curl exit=${rv}"
+        return 1
+    fi
+    msg_err="$(echo "${msg_out}" \
+        | sed -n -e 's/.*"error" *:.*"message" *: *"\([^"]*\)".*/\1/p')"
+    if [ -n "${msg_err}" ]; then
+        zed_log_err "slack webhook \"${msg_err}"\"
+        return 1
+    fi
+    return 0
+}
+
+# zed_notify_pushover (subject, pathname)
+#
+# Send a notification via Pushover <https://pushover.net/>.
+# The access token (ZED_PUSHOVER_TOKEN) identifies this client to the
+# Pushover server. The user token (ZED_PUSHOVER_USER) defines the user or
+# group to which the notification will be sent.
+#
+# Requires curl and sed executables to be installed in the standard PATH.
+#
+# References
+#   https://pushover.net/api
+#
+# Arguments
+#   subject: notification subject
+#   pathname: pathname containing the notification message (OPTIONAL)
+#
+# Globals
+#   ZED_PUSHOVER_TOKEN
+#   ZED_PUSHOVER_USER
+#
+# Return
+#   0: notification sent
+#   1: notification failed
+#   2: not configured
+#
+zed_notify_pushover()
+{
+    local subject="$1"
+    local pathname="${2:-"/dev/null"}"
+    local msg_body
+    local msg_out
+    local msg_err
+    local url="https://api.pushover.net/1/messages.json"
+
+    [ -n "${ZED_PUSHOVER_TOKEN}" ] && [ -n "${ZED_PUSHOVER_USER}" ] || return 2
+
+    if [ ! -r "${pathname}" ]; then
+        zed_log_err "pushover cannot read \"${pathname}\""
+        return 1
+    fi
+
+    zed_check_cmd "curl" "sed" || return 1
+
+    # Read the message body in.
+    #
+    msg_body="$(cat "${pathname}")"
+
+    if [ -z "${msg_body}" ]
+    then
+        msg_body=$subject
+        subject=""
+    fi
+
+    # Send the POST request and check for errors.
+    #
+    msg_out="$( \
+        curl \
+        --form-string "token=${ZED_PUSHOVER_TOKEN}" \
+        --form-string "user=${ZED_PUSHOVER_USER}" \
+        --form-string "message=${msg_body}" \
+        --form-string "title=${subject}" \
+        "${url}" \
+        2>/dev/null \
+        )"; rv=$?
+    if [ "${rv}" -ne 0 ]; then
+        zed_log_err "curl exit=${rv}"
+        return 1
+    fi
+    msg_err="$(echo "${msg_out}" \
+        | sed -n -e 's/.*"errors" *:.*\[\(.*\)\].*/\1/p')"
+    if [ -n "${msg_err}" ]; then
+        zed_log_err "pushover \"${msg_err}"\"
+        return 1
+    fi
+    return 0
+}
+
+
 # zed_rate_limit (tag, [interval])
 #
 # Check whether an event of a given type [tag] has already occurred within the
@@ -433,10 +589,8 @@ zed_guid_to_pool()
 		return
 	fi
 
-	guid=$(printf "%llu" "$1")
+	guid="$(printf "%u" "$1")"
-	if [ -n "$guid" ] ; then
+	$ZPOOL get -H -ovalue,name guid | awk '$1 == '"$guid"' {print $2; exit}'
-		$ZPOOL get -H -ovalue,name guid | awk '$1=='"$guid"' {print $2}'
-	fi
 }
 
 # zed_exit_if_ignoring_this_event

cmd/zed/zed.d/zed.rc

@@ -74,6 +74,31 @@
 #
 #ZED_PUSHBULLET_CHANNEL_TAG=""
 
+##
+# Slack Webhook URL.
+# This allows posting to the given channel and includes an access token.
+#   <https://api.slack.com/incoming-webhooks>
+# Disabled by default; uncomment to enable.
+#
+#ZED_SLACK_WEBHOOK_URL=""
+
+##
+# Pushover token.
+# This defines the application from which the notification will be sent.
+#   <https://pushover.net/api#registration>
+# Disabled by default; uncomment to enable.
+# ZED_PUSHOVER_USER, below, must also be configured.
+#
+#ZED_PUSHOVER_TOKEN=""
+
+##
+# Pushover user key.
+# This defines which user or group will receive Pushover notifications.
+#  <https://pushover.net/api#identifiers>
+# Disabled by default; uncomment to enable.
+# ZED_PUSHOVER_TOKEN, above, must also be configured.
+#ZED_PUSHOVER_USER=""
+
 ##
 # Default directory for zed state files.
 #
@@ -81,8 +106,8 @@
 
 ##
 # Turn on/off enclosure LEDs when drives get DEGRADED/FAULTED.  This works for
-# device mapper and multipath devices as well.  Your enclosure must be
+# device mapper and multipath devices as well.  This works with JBOD enclosures
-# supported by the Linux SES driver for this to work.
+# and NVMe PCI drives (assuming they're supported by Linux in sysfs).
 #
 ZED_USE_ENCLOSURE_LEDS=1
 
@@ -110,5 +135,10 @@ ZED_USE_ENCLOSURE_LEDS=1
 # Otherwise, if ZED_SYSLOG_SUBCLASS_EXCLUDE is set, the
 # matching subclasses are excluded from logging.
 #ZED_SYSLOG_SUBCLASS_INCLUDE="checksum|scrub_*|vdev.*"
-#ZED_SYSLOG_SUBCLASS_EXCLUDE="statechange|config_*|history_event"
+ZED_SYSLOG_SUBCLASS_EXCLUDE="history_event"
+
+##
+# Use GUIDs instead of names when logging pool and vdevs
+# Disabled by default, 1 to enable and 0 to disable.
+#ZED_SYSLOG_DISPLAY_GUIDS=1
 

cmd/zed/zed.h

@@ -1,9 +1,9 @@
 /*
- * This file is part of the ZFS Event Daemon (ZED)
+ * This file is part of the ZFS Event Daemon (ZED).
- * for ZFS on Linux (ZoL) <http://zfsonlinux.org/>.
+ *
  * Developed at Lawrence Livermore National Laboratory (LLNL-CODE-403049).
  * Copyright (C) 2013-2014 Lawrence Livermore National Security, LLC.
- * Refer to the ZoL git commit log for authoritative copyright attribution.
+ * Refer to the OpenZFS git commit log for authoritative copyright attribution.
  *
  * The contents of this file are subject to the terms of the
  * Common Development and Distribution License Version 1.0 (CDDL-1.0).
@@ -15,11 +15,6 @@
 #ifndef	ZED_H
 #define	ZED_H
 
-/*
- * Absolute path for the default zed configuration file.
- */
-#define	ZED_CONF_FILE		SYSCONFDIR "/zfs/zed.conf"
-
 /*
  * Absolute path for the default zed pid file.
  */
@@ -35,16 +30,6 @@
  */
 #define	ZED_ZEDLET_DIR		SYSCONFDIR "/zfs/zed.d"
 
-/*
- * Reserved for future use.
- */
-#define	ZED_MAX_EVENTS		0
-
-/*
- * Reserved for future use.
- */
-#define	ZED_MIN_EVENTS		0
-
 /*
  * String prefix for ZED variables passed via environment variables.
  */

cmd/zed/zed_conf.c

@@ -1,9 +1,9 @@
 /*
- * This file is part of the ZFS Event Daemon (ZED)
+ * This file is part of the ZFS Event Daemon (ZED).
- * for ZFS on Linux (ZoL) <http://zfsonlinux.org/>.
+ *
  * Developed at Lawrence Livermore National Laboratory (LLNL-CODE-403049).
  * Copyright (C) 2013-2014 Lawrence Livermore National Security, LLC.
- * Refer to the ZoL git commit log for authoritative copyright attribution.
+ * Refer to the OpenZFS git commit log for authoritative copyright attribution.
  *
  * The contents of this file are subject to the terms of the
  * Common Development and Distribution License Version 1.0 (CDDL-1.0).
@@ -22,6 +22,7 @@
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
+#include <sys/types.h>
 #include <sys/stat.h>
 #include <sys/uio.h>
 #include <unistd.h>
@@ -32,43 +33,26 @@
 #include "zed_strings.h"
 
 /*
- * Return a new configuration with default values.
+ * Initialise the configuration with default values.
  */
-struct zed_conf *
+void
-zed_conf_create(void)
+zed_conf_init(struct zed_conf *zcp)
 {
-	struct zed_conf *zcp;
+	memset(zcp, 0, sizeof (*zcp));
 
-	zcp = calloc(1, sizeof (*zcp));
+	/* zcp->zfs_hdl opened in zed_event_init() */
-	if (!zcp)
+	/* zcp->zedlets created in zed_conf_scan_dir() */
-		goto nomem;
 
-	zcp->syslog_facility = LOG_DAEMON;
+	zcp->pid_fd = -1;		/* opened in zed_conf_write_pid() */
-	zcp->min_events = ZED_MIN_EVENTS;
+	zcp->state_fd = -1;		/* opened in zed_conf_open_state() */
-	zcp->max_events = ZED_MAX_EVENTS;
+	zcp->zevent_fd = -1;		/* opened in zed_event_init() */
-	zcp->pid_fd = -1;
-	zcp->zedlets = NULL;		/* created via zed_conf_scan_dir() */
-	zcp->state_fd = -1;		/* opened via zed_conf_open_state() */
-	zcp->zfs_hdl = NULL;		/* opened via zed_event_init() */
-	zcp->zevent_fd = -1;		/* opened via zed_event_init() */
 
-	if (!(zcp->conf_file = strdup(ZED_CONF_FILE)))
+	zcp->max_jobs = 16;
-		goto nomem;
 
-	if (!(zcp->pid_file = strdup(ZED_PID_FILE)))
+	if (!(zcp->pid_file = strdup(ZED_PID_FILE)) ||
-		goto nomem;
+	    !(zcp->zedlet_dir = strdup(ZED_ZEDLET_DIR)) ||
-
+	    !(zcp->state_file = strdup(ZED_STATE_FILE)))
-	if (!(zcp->zedlet_dir = strdup(ZED_ZEDLET_DIR)))
+		zed_log_die("Failed to create conf: %s", strerror(errno));
-		goto nomem;
-
-	if (!(zcp->state_file = strdup(ZED_STATE_FILE)))
-		goto nomem;
-
-	return (zcp);
-
-nomem:
-	zed_log_die("Failed to create conf: %s", strerror(errno));
-	return (NULL);
 }
 
 /*
@@ -79,9 +63,6 @@ nomem:
 void
 zed_conf_destroy(struct zed_conf *zcp)
 {
-	if (!zcp)
-		return;
-
 	if (zcp->state_fd >= 0) {
 		if (close(zcp->state_fd) < 0)
 			zed_log_msg(LOG_WARNING,
@@ -102,10 +83,6 @@ zed_conf_destroy(struct zed_conf *zcp)
 			    zcp->pid_file, strerror(errno));
 		zcp->pid_fd = -1;
 	}
-	if (zcp->conf_file) {
-		free(zcp->conf_file);
-		zcp->conf_file = NULL;
-	}
 	if (zcp->pid_file) {
 		free(zcp->pid_file);
 		zcp->pid_file = NULL;
@@ -122,7 +99,6 @@ zed_conf_destroy(struct zed_conf *zcp)
 		zed_strings_destroy(zcp->zedlets);
 		zcp->zedlets = NULL;
 	}
-	free(zcp);
 }
 
 /*
@@ -132,44 +108,52 @@ zed_conf_destroy(struct zed_conf *zcp)
  * otherwise, output to stderr and exit with a failure status.
  */
 static void
-_zed_conf_display_help(const char *prog, int got_err)
+_zed_conf_display_help(const char *prog, boolean_t got_err)
 {
+	struct opt { const char *o, *d, *v; };
+
 	FILE *fp = got_err ? stderr : stdout;
-	int w1 = 4;			/* width of leading whitespace */
+
-	int w2 = 8;			/* width of L-justified option field */
+	struct opt *oo;
+	struct opt iopts[] = {
+		{ .o = "-h", .d = "Display help" },
+		{ .o = "-L", .d = "Display license information" },
+		{ .o = "-V", .d = "Display version information" },
+		{},
+	};
+	struct opt nopts[] = {
+		{ .o = "-v", .d = "Be verbose" },
+		{ .o = "-f", .d = "Force daemon to run" },
+		{ .o = "-F", .d = "Run daemon in the foreground" },
+		{ .o = "-I",
+		    .d = "Idle daemon until kernel module is (re)loaded" },
+		{ .o = "-M", .d = "Lock all pages in memory" },
+		{ .o = "-P", .d = "$PATH for ZED to use (only used by ZTS)" },
+		{ .o = "-Z", .d = "Zero state file" },
+		{},
+	};
+	struct opt vopts[] = {
+		{ .o = "-d DIR", .d = "Read enabled ZEDLETs from DIR.",
+		    .v = ZED_ZEDLET_DIR },
+		{ .o = "-p FILE", .d = "Write daemon's PID to FILE.",
+		    .v = ZED_PID_FILE },
+		{ .o = "-s FILE", .d = "Write daemon's state to FILE.",
+		    .v = ZED_STATE_FILE },
+		{ .o = "-j JOBS", .d = "Start at most JOBS at once.",
+		    .v = "16" },
+		{},
+	};
 
 	fprintf(fp, "Usage: %s [OPTION]...\n", (prog ? prog : "zed"));
 	fprintf(fp, "\n");
-	fprintf(fp, "%*c%*s %s\n", w1, 0x20, -w2, "-h",
+	for (oo = iopts; oo->o; ++oo)
-	    "Display help.");
+		fprintf(fp, "    %*s %s\n", -8, oo->o, oo->d);
-	fprintf(fp, "%*c%*s %s\n", w1, 0x20, -w2, "-L",
-	    "Display license information.");
-	fprintf(fp, "%*c%*s %s\n", w1, 0x20, -w2, "-V",
-	    "Display version information.");
 	fprintf(fp, "\n");
-	fprintf(fp, "%*c%*s %s\n", w1, 0x20, -w2, "-v",
+	for (oo = nopts; oo->o; ++oo)
-	    "Be verbose.");
+		fprintf(fp, "    %*s %s\n", -8, oo->o, oo->d);
-	fprintf(fp, "%*c%*s %s\n", w1, 0x20, -w2, "-f",
-	    "Force daemon to run.");
-	fprintf(fp, "%*c%*s %s\n", w1, 0x20, -w2, "-F",
-	    "Run daemon in the foreground.");
-	fprintf(fp, "%*c%*s %s\n", w1, 0x20, -w2, "-M",
-	    "Lock all pages in memory.");
-	fprintf(fp, "%*c%*s %s\n", w1, 0x20, -w2, "-P",
-	    "$PATH for ZED to use (only used by ZTS).");
-	fprintf(fp, "%*c%*s %s\n", w1, 0x20, -w2, "-Z",
-	    "Zero state file.");
 	fprintf(fp, "\n");
-#if 0
+	for (oo = vopts; oo->o; ++oo)
-	fprintf(fp, "%*c%*s %s [%s]\n", w1, 0x20, -w2, "-c FILE",
+		fprintf(fp, "    %*s %s [%s]\n", -8, oo->o, oo->d, oo->v);
-	    "Read configuration from FILE.", ZED_CONF_FILE);
-#endif
-	fprintf(fp, "%*c%*s %s [%s]\n", w1, 0x20, -w2, "-d DIR",
-	    "Read enabled ZEDLETs from DIR.", ZED_ZEDLET_DIR);
-	fprintf(fp, "%*c%*s %s [%s]\n", w1, 0x20, -w2, "-p FILE",
-	    "Write daemon's PID to FILE.", ZED_PID_FILE);
-	fprintf(fp, "%*c%*s %s [%s]\n", w1, 0x20, -w2, "-s FILE",
-	    "Write daemon's state to FILE.", ZED_STATE_FILE);
 	fprintf(fp, "\n");
 
 	exit(got_err ? EXIT_FAILURE : EXIT_SUCCESS);
@@ -181,20 +165,14 @@ _zed_conf_display_help(const char *prog, int got_err)
 static void
 _zed_conf_display_license(void)
 {
-	const char **pp;
+	printf(
-	const char *text[] = {
+	    "The ZFS Event Daemon (ZED) is distributed under the terms of the\n"
-	    "The ZFS Event Daemon (ZED) is distributed under the terms of the",
+	    "  Common Development and Distribution License (CDDL-1.0)\n"
-	    "  Common Development and Distribution License (CDDL-1.0)",
+	    "  <http://opensource.org/licenses/CDDL-1.0>.\n"
-	    "  <http://opensource.org/licenses/CDDL-1.0>.",
+	    "\n"
-	    "",
 	    "Developed at Lawrence Livermore National Laboratory"
-	    " (LLNL-CODE-403049).",
+	    " (LLNL-CODE-403049).\n"
-	    "",
+	    "\n");
-	    NULL
-	};
-
-	for (pp = text; *pp; pp++)
-		printf("%s\n", *pp);
 
 	exit(EXIT_SUCCESS);
 }
@@ -229,16 +207,19 @@ _zed_conf_parse_path(char **resultp, const char *path)
 
 	if (path[0] == '/') {
 		*resultp = strdup(path);
-	} else if (!getcwd(buf, sizeof (buf))) {
-		zed_log_die("Failed to get current working dir: %s",
-		    strerror(errno));
-	} else if (strlcat(buf, "/", sizeof (buf)) >= sizeof (buf)) {
-		zed_log_die("Failed to copy path: %s", strerror(ENAMETOOLONG));
-	} else if (strlcat(buf, path, sizeof (buf)) >= sizeof (buf)) {
-		zed_log_die("Failed to copy path: %s", strerror(ENAMETOOLONG));
 	} else {
+		if (!getcwd(buf, sizeof (buf)))
+			zed_log_die("Failed to get current working dir: %s",
+			    strerror(errno));
+
+		if (strlcat(buf, "/", sizeof (buf)) >= sizeof (buf) ||
+		    strlcat(buf, path, sizeof (buf)) >= sizeof (buf))
+			zed_log_die("Failed to copy path: %s",
+			    strerror(ENAMETOOLONG));
+
 		*resultp = strdup(buf);
 	}
+
 	if (!*resultp)
 		zed_log_die("Failed to copy path: %s", strerror(ENOMEM));
 }
@@ -249,8 +230,9 @@ _zed_conf_parse_path(char **resultp, const char *path)
 void
 zed_conf_parse_opts(struct zed_conf *zcp, int argc, char **argv)
 {
-	const char * const opts = ":hLVc:d:p:P:s:vfFMZ";
+	const char * const opts = ":hLVd:p:P:s:vfFMZIj:";
 	int opt;
+	unsigned long raw;
 
 	if (!zcp || !argv || !argv[0])
 		zed_log_die("Failed to parse options: Internal error");
@@ -260,7 +242,7 @@ zed_conf_parse_opts(struct zed_conf *zcp, int argc, char **argv)
 	while ((opt = getopt(argc, argv, opts)) != -1) {
 		switch (opt) {
 		case 'h':
-			_zed_conf_display_help(argv[0], EXIT_SUCCESS);
+			_zed_conf_display_help(argv[0], B_FALSE);
 			break;
 		case 'L':
 			_zed_conf_display_license();
@@ -268,12 +250,12 @@ zed_conf_parse_opts(struct zed_conf *zcp, int argc, char **argv)
 		case 'V':
 			_zed_conf_display_version();
 			break;
-		case 'c':
-			_zed_conf_parse_path(&zcp->conf_file, optarg);
-			break;
 		case 'd':
 			_zed_conf_parse_path(&zcp->zedlet_dir, optarg);
 			break;
+		case 'I':
+			zcp->do_idle = 1;
+			break;
 		case 'p':
 			_zed_conf_parse_path(&zcp->pid_file, optarg);
 			break;
@@ -298,31 +280,30 @@ zed_conf_parse_opts(struct zed_conf *zcp, int argc, char **argv)
 		case 'Z':
 			zcp->do_zero = 1;
 			break;
+		case 'j':
+			errno = 0;
+			raw = strtoul(optarg, NULL, 0);
+			if (errno == ERANGE || raw > INT16_MAX) {
+				zed_log_die("%lu is too many jobs", raw);
+			} if (raw == 0) {
+				zed_log_die("0 jobs makes no sense");
+			} else {
+				zcp->max_jobs = raw;
+			}
+			break;
 		case '?':
 		default:
 			if (optopt == '?')
-				_zed_conf_display_help(argv[0], EXIT_SUCCESS);
+				_zed_conf_display_help(argv[0], B_FALSE);
 
-			fprintf(stderr, "%s: %s '-%c'\n\n", argv[0],
+			fprintf(stderr, "%s: Invalid option '-%c'\n\n",
-			    "Invalid option", optopt);
+			    argv[0], optopt);
-			_zed_conf_display_help(argv[0], EXIT_FAILURE);
+			_zed_conf_display_help(argv[0], B_TRUE);
 			break;
 		}
 	}
 }
 
-/*
- * Parse the configuration file into the configuration [zcp].
- *
- * FIXME: Not yet implemented.
- */
-void
-zed_conf_parse_file(struct zed_conf *zcp)
-{
-	if (!zcp)
-		zed_log_die("Failed to parse config: %s", strerror(EINVAL));
-}
-
 /*
  * Scan the [zcp] zedlet_dir for files to exec based on the event class.
  * Files must be executable by user, but not writable by group or other.
@@ -330,8 +311,6 @@ zed_conf_parse_file(struct zed_conf *zcp)
  *
  * Return 0 on success with an updated set of zedlets,
  * or -1 on error with errno set.
- *
- * FIXME: Check if zedlet_dir and all parent dirs are secure.
  */
 int
 zed_conf_scan_dir(struct zed_conf *zcp)
@@ -447,8 +426,6 @@ zed_conf_scan_dir(struct zed_conf *zcp)
 int
 zed_conf_write_pid(struct zed_conf *zcp)
 {
-	const mode_t dirmode = S_IRWXU | S_IRGRP | S_IXGRP | S_IROTH | S_IXOTH;
-	const mode_t filemode = S_IRUSR | S_IWUSR | S_IRGRP | S_IROTH;
 	char buf[PATH_MAX];
 	int n;
 	char *p;
@@ -476,7 +453,7 @@ zed_conf_write_pid(struct zed_conf *zcp)
 	if (p)
 		*p = '\0';
 
-	if ((mkdirp(buf, dirmode) < 0) && (errno != EEXIST)) {
+	if ((mkdirp(buf, 0755) < 0) && (errno != EEXIST)) {
 		zed_log_msg(LOG_ERR, "Failed to create directory \"%s\": %s",
 		    buf, strerror(errno));
 		goto err;
@@ -486,7 +463,7 @@ zed_conf_write_pid(struct zed_conf *zcp)
 	 */
 	mask = umask(0);
 	umask(mask | 022);
-	zcp->pid_fd = open(zcp->pid_file, (O_RDWR | O_CREAT), filemode);
+	zcp->pid_fd = open(zcp->pid_file, O_RDWR | O_CREAT | O_CLOEXEC, 0644);
 	umask(mask);
 	if (zcp->pid_fd < 0) {
 		zed_log_msg(LOG_ERR, "Failed to open PID file \"%s\": %s",
@@ -523,7 +500,7 @@ zed_conf_write_pid(struct zed_conf *zcp)
 		errno = ERANGE;
 		zed_log_msg(LOG_ERR, "Failed to write PID file \"%s\": %s",
 		    zcp->pid_file, strerror(errno));
-	} else if (zed_file_write_n(zcp->pid_fd, buf, n) != n) {
+	} else if (write(zcp->pid_fd, buf, n) != n) {
 		zed_log_msg(LOG_ERR, "Failed to write PID file \"%s\": %s",
 		    zcp->pid_file, strerror(errno));
 	} else if (fdatasync(zcp->pid_fd) < 0) {
@@ -551,7 +528,6 @@ int
 zed_conf_open_state(struct zed_conf *zcp)
 {
 	char dirbuf[PATH_MAX];
-	mode_t dirmode = S_IRWXU | S_IRGRP | S_IXGRP | S_IROTH | S_IXOTH;
 	int n;
 	char *p;
 	int rv;
@@ -573,7 +549,7 @@ zed_conf_open_state(struct zed_conf *zcp)
 	if (p)
 		*p = '\0';
 
-	if ((mkdirp(dirbuf, dirmode) < 0) && (errno != EEXIST)) {
+	if ((mkdirp(dirbuf, 0755) < 0) && (errno != EEXIST)) {
 		zed_log_msg(LOG_WARNING,
 		    "Failed to create directory \"%s\": %s",
 		    dirbuf, strerror(errno));
@@ -591,7 +567,7 @@ zed_conf_open_state(struct zed_conf *zcp)
 		(void) unlink(zcp->state_file);
 
 	zcp->state_fd = open(zcp->state_file,
-	    (O_RDWR | O_CREAT), (S_IRUSR | S_IWUSR | S_IRGRP | S_IROTH));
+	    O_RDWR | O_CREAT | O_CLOEXEC, 0644);
 	if (zcp->state_fd < 0) {
 		zed_log_msg(LOG_WARNING, "Failed to open state file \"%s\": %s",
 		    zcp->state_file, strerror(errno));

cmd/zed/zed_conf.h

@@ -1,9 +1,9 @@
 /*
- * This file is part of the ZFS Event Daemon (ZED)
+ * This file is part of the ZFS Event Daemon (ZED).
- * for ZFS on Linux (ZoL) <http://zfsonlinux.org/>.
+ *
  * Developed at Lawrence Livermore National Laboratory (LLNL-CODE-403049).
  * Copyright (C) 2013-2014 Lawrence Livermore National Security, LLC.
- * Refer to the ZoL git commit log for authoritative copyright attribution.
+ * Refer to the OpenZFS git commit log for authoritative copyright attribution.
  *
  * The contents of this file are subject to the terms of the
  * Common Development and Distribution License Version 1.0 (CDDL-1.0).
@@ -20,42 +20,39 @@
 #include "zed_strings.h"
 
 struct zed_conf {
-	unsigned	do_force:1;		/* true if force enabled */
-	unsigned	do_foreground:1;	/* true if run in foreground */
-	unsigned	do_memlock:1;		/* true if locking memory */
-	unsigned	do_verbose:1;		/* true if verbosity enabled */
-	unsigned	do_zero:1;		/* true if zeroing state */
-	int		syslog_facility;	/* syslog facility value */
-	int		min_events;		/* RESERVED FOR FUTURE USE */
-	int		max_events;		/* RESERVED FOR FUTURE USE */
-	char		*conf_file;		/* abs path to config file */
 	char		*pid_file;		/* abs path to pid file */
-	int		pid_fd;			/* fd to pid file for lock */
 	char		*zedlet_dir;		/* abs path to zedlet dir */
-	zed_strings_t	*zedlets;		/* names of enabled zedlets */
 	char		*state_file;		/* abs path to state file */
-	int		state_fd;		/* fd to state file */
+
 	libzfs_handle_t	*zfs_hdl;		/* handle to libzfs */
-	int		zevent_fd;		/* fd for access to zevents */
+	zed_strings_t	*zedlets;		/* names of enabled zedlets */
 	char		*path;		/* custom $PATH for zedlets to use */
+
+	int		pid_fd;			/* fd to pid file for lock */
+	int		state_fd;		/* fd to state file */
+	int		zevent_fd;		/* fd for access to zevents */
+
+	int16_t max_jobs;		/* max zedlets to run at one time */
+
+	boolean_t	do_force:1;		/* true if force enabled */
+	boolean_t	do_foreground:1;	/* true if run in foreground */
+	boolean_t	do_memlock:1;		/* true if locking memory */
+	boolean_t	do_verbose:1;		/* true if verbosity enabled */
+	boolean_t	do_zero:1;		/* true if zeroing state */
+	boolean_t	do_idle:1;		/* true if idle enabled */
 };
 
-struct zed_conf *zed_conf_create(void);
+void zed_conf_init(struct zed_conf *zcp);
-
 void zed_conf_destroy(struct zed_conf *zcp);
 
 void zed_conf_parse_opts(struct zed_conf *zcp, int argc, char **argv);
 
-void zed_conf_parse_file(struct zed_conf *zcp);
-
 int zed_conf_scan_dir(struct zed_conf *zcp);
 
 int zed_conf_write_pid(struct zed_conf *zcp);
 
 int zed_conf_open_state(struct zed_conf *zcp);
-
 int zed_conf_read_state(struct zed_conf *zcp, uint64_t *eidp, int64_t etime[]);
-
 int zed_conf_write_state(struct zed_conf *zcp, uint64_t eid, int64_t etime[]);
 
 #endif	/* !ZED_CONF_H */

cmd/zed/zed_disk_event.c

@@ -72,6 +72,8 @@ zed_udev_event(const char *class, const char *subclass, nvlist_t *nvl)
 		zed_log_msg(LOG_INFO, "\t%s: %s", DEV_PATH, strval);
 	if (nvlist_lookup_string(nvl, DEV_IDENTIFIER, &strval) == 0)
 		zed_log_msg(LOG_INFO, "\t%s: %s", DEV_IDENTIFIER, strval);
+	if (nvlist_lookup_boolean(nvl, DEV_IS_PART) == B_TRUE)
+		zed_log_msg(LOG_INFO, "\t%s: B_TRUE", DEV_IS_PART);
 	if (nvlist_lookup_string(nvl, DEV_PHYS_PATH, &strval) == 0)
 		zed_log_msg(LOG_INFO, "\t%s: %s", DEV_PHYS_PATH, strval);
 	if (nvlist_lookup_uint64(nvl, DEV_SIZE, &numval) == 0)
@@ -379,6 +381,7 @@ zed_disk_event_init()
 		return (-1);
 	}
 
+	pthread_setname_np(g_mon_tid, "udev monitor");
 	zed_log_msg(LOG_INFO, "zed_disk_event_init");
 
 	return (0);

cmd/zed/zed_event.c

@@ -1,9 +1,9 @@
 /*
- * This file is part of the ZFS Event Daemon (ZED)
+ * This file is part of the ZFS Event Daemon (ZED).
- * for ZFS on Linux (ZoL) <http://zfsonlinux.org/>.
+ *
  * Developed at Lawrence Livermore National Laboratory (LLNL-CODE-403049).
  * Copyright (C) 2013-2014 Lawrence Livermore National Security, LLC.
- * Refer to the ZoL git commit log for authoritative copyright attribution.
+ * Refer to the OpenZFS git commit log for authoritative copyright attribution.
  *
  * The contents of this file are subject to the terms of the
  * Common Development and Distribution License Version 1.0 (CDDL-1.0).
@@ -15,7 +15,7 @@
 #include <ctype.h>
 #include <errno.h>
 #include <fcntl.h>
-#include <libzfs.h>			/* FIXME: Replace with libzfs_core. */
+#include <libzfs_core.h>
 #include <paths.h>
 #include <stdarg.h>
 #include <stdio.h>
@@ -28,6 +28,7 @@
 #include "zed.h"
 #include "zed_conf.h"
 #include "zed_disk_event.h"
+#include "zed_event.h"
 #include "zed_exec.h"
 #include "zed_file.h"
 #include "zed_log.h"
@@ -40,25 +41,36 @@
 /*
  * Open the libzfs interface.
  */
-void
+int
 zed_event_init(struct zed_conf *zcp)
 {
 	if (!zcp)
 		zed_log_die("Failed zed_event_init: %s", strerror(EINVAL));
 
 	zcp->zfs_hdl = libzfs_init();
-	if (!zcp->zfs_hdl)
+	if (!zcp->zfs_hdl) {
+		if (zcp->do_idle)
+			return (-1);
 		zed_log_die("Failed to initialize libzfs");
+	}
 
-	zcp->zevent_fd = open(ZFS_DEV, O_RDWR);
+	zcp->zevent_fd = open(ZFS_DEV, O_RDWR | O_CLOEXEC);
-	if (zcp->zevent_fd < 0)
+	if (zcp->zevent_fd < 0) {
+		if (zcp->do_idle)
+			return (-1);
 		zed_log_die("Failed to open \"%s\": %s",
 		    ZFS_DEV, strerror(errno));
+	}
 
 	zfs_agent_init(zcp->zfs_hdl);
 
-	if (zed_disk_event_init() != 0)
+	if (zed_disk_event_init() != 0) {
+		if (zcp->do_idle)
+			return (-1);
 		zed_log_die("Failed to initialize disk events");
+	}
+
+	return (0);
 }
 
 /*
@@ -84,6 +96,47 @@ zed_event_fini(struct zed_conf *zcp)
 		libzfs_fini(zcp->zfs_hdl);
 		zcp->zfs_hdl = NULL;
 	}
+
+	zed_exec_fini();
+}
+
+static void
+_bump_event_queue_length(void)
+{
+	int zzlm = -1, wr;
+	char qlen_buf[12] = {0}; /* parameter is int => max "-2147483647\n" */
+	long int qlen;
+
+	zzlm = open("/sys/module/zfs/parameters/zfs_zevent_len_max", O_RDWR);
+	if (zzlm < 0)
+		goto done;
+
+	if (read(zzlm, qlen_buf, sizeof (qlen_buf)) < 0)
+		goto done;
+	qlen_buf[sizeof (qlen_buf) - 1] = '\0';
+
+	errno = 0;
+	qlen = strtol(qlen_buf, NULL, 10);
+	if (errno == ERANGE)
+		goto done;
+
+	if (qlen <= 0)
+		qlen = 512; /* default zfs_zevent_len_max value */
+	else
+		qlen *= 2;
+
+	if (qlen > INT_MAX)
+		qlen = INT_MAX;
+	wr = snprintf(qlen_buf, sizeof (qlen_buf), "%ld", qlen);
+
+	if (pwrite(zzlm, qlen_buf, wr, 0) < 0)
+		goto done;
+
+	zed_log_msg(LOG_WARNING, "Bumping queue length to %ld", qlen);
+
+done:
+	if (zzlm > -1)
+		(void) close(zzlm);
 }
 
 /*
@@ -124,10 +177,7 @@ zed_event_seek(struct zed_conf *zcp, uint64_t saved_eid, int64_t saved_etime[])
 
 		if (n_dropped > 0) {
 			zed_log_msg(LOG_WARNING, "Missed %d events", n_dropped);
-			/*
+			_bump_event_queue_length();
-			 * FIXME: Increase max size of event nvlist in
-			 *   /sys/module/zfs/parameters/zfs_zevent_len_max ?
-			 */
 		}
 		if (nvlist_lookup_uint64(nvl, "eid", &eid) != 0) {
 			zed_log_msg(LOG_WARNING, "Failed to lookup zevent eid");
@@ -199,7 +249,7 @@ _zed_event_value_is_hex(const char *name)
  *
  * All environment variables in [zsp] should be added through this function.
  */
-static int
+static __attribute__((format(printf, 5, 6))) int
 _zed_event_add_var(uint64_t eid, zed_strings_t *zsp,
     const char *prefix, const char *name, const char *fmt, ...)
 {
@@ -574,8 +624,6 @@ _zed_event_add_string_array(uint64_t eid, zed_strings_t *zsp,
  * Convert the nvpair [nvp] to a string which is added to the environment
  * of the child process.
  * Return 0 on success, -1 on error.
- *
- * FIXME: Refactor with cmd/zpool/zpool_main.c:zpool_do_events_nvprint()?
  */
 static void
 _zed_event_add_nvpair(uint64_t eid, zed_strings_t *zsp, nvpair_t *nvp)
@@ -674,23 +722,11 @@ _zed_event_add_nvpair(uint64_t eid, zed_strings_t *zsp, nvpair_t *nvp)
 		_zed_event_add_var(eid, zsp, prefix, name,
 		    "%llu", (u_longlong_t)i64);
 		break;
-	case DATA_TYPE_NVLIST:
-		_zed_event_add_var(eid, zsp, prefix, name,
-		    "%s", "_NOT_IMPLEMENTED_");			/* FIXME */
-		break;
 	case DATA_TYPE_STRING:
 		(void) nvpair_value_string(nvp, &str);
 		_zed_event_add_var(eid, zsp, prefix, name,
 		    "%s", (str ? str : "<NULL>"));
 		break;
-	case DATA_TYPE_BOOLEAN_ARRAY:
-		_zed_event_add_var(eid, zsp, prefix, name,
-		    "%s", "_NOT_IMPLEMENTED_");			/* FIXME */
-		break;
-	case DATA_TYPE_BYTE_ARRAY:
-		_zed_event_add_var(eid, zsp, prefix, name,
-		    "%s", "_NOT_IMPLEMENTED_");			/* FIXME */
-		break;
 	case DATA_TYPE_INT8_ARRAY:
 		_zed_event_add_int8_array(eid, zsp, prefix, nvp);
 		break;
@@ -718,9 +754,11 @@ _zed_event_add_nvpair(uint64_t eid, zed_strings_t *zsp, nvpair_t *nvp)
 	case DATA_TYPE_STRING_ARRAY:
 		_zed_event_add_string_array(eid, zsp, prefix, nvp);
 		break;
+	case DATA_TYPE_NVLIST:
+	case DATA_TYPE_BOOLEAN_ARRAY:
+	case DATA_TYPE_BYTE_ARRAY:
 	case DATA_TYPE_NVLIST_ARRAY:
-		_zed_event_add_var(eid, zsp, prefix, name,
+		_zed_event_add_var(eid, zsp, prefix, name, "_NOT_IMPLEMENTED_");
-		    "%s", "_NOT_IMPLEMENTED_");			/* FIXME */
 		break;
 	default:
 		errno = EINVAL;
@@ -872,7 +910,7 @@ _zed_event_add_time_strings(uint64_t eid, zed_strings_t *zsp, int64_t etime[])
 /*
  * Service the next zevent, blocking until one is available.
  */
-void
+int
 zed_event_service(struct zed_conf *zcp)
 {
 	nvlist_t *nvl;
@@ -890,20 +928,17 @@ zed_event_service(struct zed_conf *zcp)
 		errno = EINVAL;
 		zed_log_msg(LOG_ERR, "Failed to service zevent: %s",
 		    strerror(errno));
-		return;
+		return (EINVAL);
 	}
 	rv = zpool_events_next(zcp->zfs_hdl, &nvl, &n_dropped, ZEVENT_NONE,
 	    zcp->zevent_fd);
 
 	if ((rv != 0) || !nvl)
-		return;
+		return (errno);
 
 	if (n_dropped > 0) {
 		zed_log_msg(LOG_WARNING, "Missed %d events", n_dropped);
-		/*
+		_bump_event_queue_length();
-		 * FIXME: Increase max size of event nvlist in
-		 * /sys/module/zfs/parameters/zfs_zevent_len_max ?
-		 */
 	}
 	if (nvlist_lookup_uint64(nvl, "eid", &eid) != 0) {
 		zed_log_msg(LOG_WARNING, "Failed to lookup zevent eid");
@@ -941,12 +976,12 @@ zed_event_service(struct zed_conf *zcp)
 
 		_zed_event_add_time_strings(eid, zsp, etime);
 
-		zed_exec_process(eid, class, subclass,
+		zed_exec_process(eid, class, subclass, zcp, zsp);
-		    zcp->zedlet_dir, zcp->zedlets, zsp, zcp->zevent_fd);
 
 		zed_conf_write_state(zcp, eid, etime);
 
 		zed_strings_destroy(zsp);
 	}
 	nvlist_free(nvl);
+	return (0);
 }

cmd/zed/zed_event.h

@@ -1,9 +1,9 @@
 /*
- * This file is part of the ZFS Event Daemon (ZED)
+ * This file is part of the ZFS Event Daemon (ZED).
- * for ZFS on Linux (ZoL) <http://zfsonlinux.org/>.
+ *
  * Developed at Lawrence Livermore National Laboratory (LLNL-CODE-403049).
  * Copyright (C) 2013-2014 Lawrence Livermore National Security, LLC.
- * Refer to the ZoL git commit log for authoritative copyright attribution.
+ * Refer to the OpenZFS git commit log for authoritative copyright attribution.
  *
  * The contents of this file are subject to the terms of the
  * Common Development and Distribution License Version 1.0 (CDDL-1.0).
@@ -17,13 +17,13 @@
 
 #include <stdint.h>
 
-void zed_event_init(struct zed_conf *zcp);
+int zed_event_init(struct zed_conf *zcp);
 
 void zed_event_fini(struct zed_conf *zcp);
 
 int zed_event_seek(struct zed_conf *zcp, uint64_t saved_eid,
     int64_t saved_etime[]);
 
-void zed_event_service(struct zed_conf *zcp);
+int zed_event_service(struct zed_conf *zcp);
 
 #endif	/* !ZED_EVENT_H */

cmd/zed/zed_exec.c

@@ -1,9 +1,9 @@
 /*
- * This file is part of the ZFS Event Daemon (ZED)
+ * This file is part of the ZFS Event Daemon (ZED).
- * for ZFS on Linux (ZoL) <http://zfsonlinux.org/>.
+ *
  * Developed at Lawrence Livermore National Laboratory (LLNL-CODE-403049).
  * Copyright (C) 2013-2014 Lawrence Livermore National Security, LLC.
- * Refer to the ZoL git commit log for authoritative copyright attribution.
+ * Refer to the OpenZFS git commit log for authoritative copyright attribution.
  *
  * The contents of this file are subject to the terms of the
  * Common Development and Distribution License Version 1.0 (CDDL-1.0).
@@ -18,16 +18,55 @@
 #include <fcntl.h>
 #include <stdlib.h>
 #include <string.h>
+#include <stddef.h>
+#include <sys/avl.h>
+#include <sys/resource.h>
 #include <sys/stat.h>
 #include <sys/wait.h>
 #include <time.h>
 #include <unistd.h>
-#include "zed_file.h"
+#include <pthread.h>
+#include <signal.h>
+
+#include "zed_exec.h"
 #include "zed_log.h"
 #include "zed_strings.h"
 
 #define	ZEVENT_FILENO	3
 
+struct launched_process_node {
+	avl_node_t node;
+	pid_t pid;
+	uint64_t eid;
+	char *name;
+};
+
+static int
+_launched_process_node_compare(const void *x1, const void *x2)
+{
+	pid_t p1;
+	pid_t p2;
+
+	assert(x1 != NULL);
+	assert(x2 != NULL);
+
+	p1 = ((const struct launched_process_node *) x1)->pid;
+	p2 = ((const struct launched_process_node *) x2)->pid;
+
+	if (p1 < p2)
+		return (-1);
+	else if (p1 == p2)
+		return (0);
+	else
+		return (1);
+}
+
+static pthread_t _reap_children_tid = (pthread_t)-1;
+static volatile boolean_t _reap_children_stop;
+static avl_tree_t _launched_processes;
+static pthread_mutex_t _launched_processes_lock = PTHREAD_MUTEX_INITIALIZER;
+static int16_t _launched_processes_limit;
+
 /*
  * Create an environment string array for passing to execve() using the
  * NAME=VALUE strings in container [zsp].
@@ -78,20 +117,26 @@ _zed_exec_create_env(zed_strings_t *zsp)
  */
 static void
 _zed_exec_fork_child(uint64_t eid, const char *dir, const char *prog,
-    char *env[], int zfd)
+    char *env[], int zfd, boolean_t in_foreground)
 {
 	char path[PATH_MAX];
 	int n;
 	pid_t pid;
 	int fd;
-	pid_t wpid;
+	struct launched_process_node *node;
-	int status;
+	sigset_t mask;
+	struct timespec launch_timeout =
+		{ .tv_sec = 0, .tv_nsec = 200 * 1000 * 1000, };
 
 	assert(dir != NULL);
 	assert(prog != NULL);
 	assert(env != NULL);
 	assert(zfd >= 0);
 
+	while (__atomic_load_n(&_launched_processes_limit,
+	    __ATOMIC_SEQ_CST) <= 0)
+		(void) nanosleep(&launch_timeout, NULL);
+
 	n = snprintf(path, sizeof (path), "%s/%s", dir, prog);
 	if ((n < 0) || (n >= sizeof (path))) {
 		zed_log_msg(LOG_WARNING,
@@ -99,100 +144,179 @@ _zed_exec_fork_child(uint64_t eid, const char *dir, const char *prog,
 		    prog, eid, strerror(ENAMETOOLONG));
 		return;
 	}
+	(void) pthread_mutex_lock(&_launched_processes_lock);
 	pid = fork();
 	if (pid < 0) {
+		(void) pthread_mutex_unlock(&_launched_processes_lock);
 		zed_log_msg(LOG_WARNING,
 		    "Failed to fork \"%s\" for eid=%llu: %s",
 		    prog, eid, strerror(errno));
 		return;
 	} else if (pid == 0) {
+		(void) sigemptyset(&mask);
+		(void) sigprocmask(SIG_SETMASK, &mask, NULL);
+
 		(void) umask(022);
-		if ((fd = open("/dev/null", O_RDWR)) != -1) {
+		if (in_foreground && /* we're already devnulled if daemonised */
+		    (fd = open("/dev/null", O_RDWR | O_CLOEXEC)) != -1) {
 			(void) dup2(fd, STDIN_FILENO);
 			(void) dup2(fd, STDOUT_FILENO);
 			(void) dup2(fd, STDERR_FILENO);
 		}
 		(void) dup2(zfd, ZEVENT_FILENO);
-		zed_file_close_from(ZEVENT_FILENO + 1);
 		execle(path, prog, NULL, env);
 		_exit(127);
 	}
 
 	/* parent process */
 
+	node = calloc(1, sizeof (*node));
+	if (node) {
+		node->pid = pid;
+		node->eid = eid;
+		node->name = strdup(prog);
+
+		avl_add(&_launched_processes, node);
+	}
+	(void) pthread_mutex_unlock(&_launched_processes_lock);
+
+	__atomic_sub_fetch(&_launched_processes_limit, 1, __ATOMIC_SEQ_CST);
 	zed_log_msg(LOG_INFO, "Invoking \"%s\" eid=%llu pid=%d",
 	    prog, eid, pid);
+}
 
-	/* FIXME: Timeout rogue child processes with sigalarm? */
+static void
+_nop(int sig)
+{}
 
-	/*
+static void *
-	 * Wait for child process using WNOHANG to limit
+_reap_children(void *arg)
-	 * the time spent waiting to 10 seconds (10,000ms).
+{
-	 */
+	struct launched_process_node node, *pnode;
-	for (n = 0; n < 1000; n++) {
+	pid_t pid;
-		wpid = waitpid(pid, &status, WNOHANG);
+	int status;
-		if (wpid == (pid_t)-1) {
+	struct rusage usage;
-			if (errno == EINTR)
+	struct sigaction sa = {};
-				continue;
-			zed_log_msg(LOG_WARNING,
-			    "Failed to wait for \"%s\" eid=%llu pid=%d",
-			    prog, eid, pid);
-			break;
-		} else if (wpid == 0) {
-			struct timespec t;
 
-			/* child still running */
+	(void) sigfillset(&sa.sa_mask);
-			t.tv_sec = 0;
+	(void) sigdelset(&sa.sa_mask, SIGCHLD);
-			t.tv_nsec = 10000000;	/* 10ms */
+	(void) pthread_sigmask(SIG_SETMASK, &sa.sa_mask, NULL);
-			(void) nanosleep(&t, NULL);
-			continue;
-		}
 
-		if (WIFEXITED(status)) {
+	(void) sigemptyset(&sa.sa_mask);
-			zed_log_msg(LOG_INFO,
+	sa.sa_handler = _nop;
-			    "Finished \"%s\" eid=%llu pid=%d exit=%d",
+	sa.sa_flags = SA_NOCLDSTOP;
-			    prog, eid, pid, WEXITSTATUS(status));
+	(void) sigaction(SIGCHLD, &sa, NULL);
-		} else if (WIFSIGNALED(status)) {
+
-			zed_log_msg(LOG_INFO,
+	for (_reap_children_stop = B_FALSE; !_reap_children_stop; ) {
-			    "Finished \"%s\" eid=%llu pid=%d sig=%d/%s",
+		(void) pthread_mutex_lock(&_launched_processes_lock);
-			    prog, eid, pid, WTERMSIG(status),
+		pid = wait4(0, &status, WNOHANG, &usage);
-			    strsignal(WTERMSIG(status)));
+
+		if (pid == 0 || pid == (pid_t)-1) {
+			(void) pthread_mutex_unlock(&_launched_processes_lock);
+			if (pid == 0 || errno == ECHILD)
+				pause();
+			else if (errno != EINTR)
+				zed_log_msg(LOG_WARNING,
+				    "Failed to wait for children: %s",
+				    strerror(errno));
 		} else {
-			zed_log_msg(LOG_INFO,
+			memset(&node, 0, sizeof (node));
-			    "Finished \"%s\" eid=%llu pid=%d status=0x%X",
+			node.pid = pid;
-			    prog, eid, (unsigned int) status);
+			pnode = avl_find(&_launched_processes, &node, NULL);
+			if (pnode) {
+				memcpy(&node, pnode, sizeof (node));
+
+				avl_remove(&_launched_processes, pnode);
+				free(pnode);
+			}
+			(void) pthread_mutex_unlock(&_launched_processes_lock);
+			__atomic_add_fetch(&_launched_processes_limit, 1,
+			    __ATOMIC_SEQ_CST);
+
+			usage.ru_utime.tv_sec += usage.ru_stime.tv_sec;
+			usage.ru_utime.tv_usec += usage.ru_stime.tv_usec;
+			usage.ru_utime.tv_sec +=
+			    usage.ru_utime.tv_usec / (1000 * 1000);
+			usage.ru_utime.tv_usec %= 1000 * 1000;
+
+			if (WIFEXITED(status)) {
+				zed_log_msg(LOG_INFO,
+				    "Finished \"%s\" eid=%llu pid=%d "
+				    "time=%llu.%06us exit=%d",
+				    node.name, node.eid, pid,
+				    (unsigned long long) usage.ru_utime.tv_sec,
+				    (unsigned int) usage.ru_utime.tv_usec,
+				    WEXITSTATUS(status));
+			} else if (WIFSIGNALED(status)) {
+				zed_log_msg(LOG_INFO,
+				    "Finished \"%s\" eid=%llu pid=%d "
+				    "time=%llu.%06us sig=%d/%s",
+				    node.name, node.eid, pid,
+				    (unsigned long long) usage.ru_utime.tv_sec,
+				    (unsigned int) usage.ru_utime.tv_usec,
+				    WTERMSIG(status),
+				    strsignal(WTERMSIG(status)));
+			} else {
+				zed_log_msg(LOG_INFO,
+				    "Finished \"%s\" eid=%llu pid=%d "
+				    "time=%llu.%06us status=0x%X",
+				    node.name, node.eid,
+				    (unsigned long long) usage.ru_utime.tv_sec,
+				    (unsigned int) usage.ru_utime.tv_usec,
+				    (unsigned int) status);
+			}
+
+			free(node.name);
 		}
-		break;
 	}
 
-	/*
+	return (NULL);
-	 * kill child process after 10 seconds
+}
-	 */
+
-	if (wpid == 0) {
+void
-		zed_log_msg(LOG_WARNING, "Killing hung \"%s\" pid=%d",
+zed_exec_fini(void)
-		    prog, pid);
+{
-		(void) kill(pid, SIGKILL);
+	struct launched_process_node *node;
+	void *ck = NULL;
+
+	if (_reap_children_tid == (pthread_t)-1)
+		return;
+
+	_reap_children_stop = B_TRUE;
+	(void) pthread_kill(_reap_children_tid, SIGCHLD);
+	(void) pthread_join(_reap_children_tid, NULL);
+
+	while ((node = avl_destroy_nodes(&_launched_processes, &ck)) != NULL) {
+		free(node->name);
+		free(node);
 	}
+	avl_destroy(&_launched_processes);
+
+	(void) pthread_mutex_destroy(&_launched_processes_lock);
+	(void) pthread_mutex_init(&_launched_processes_lock, NULL);
+
+	_reap_children_tid = (pthread_t)-1;
 }
 
 /*
  * Process the event [eid] by synchronously invoking all zedlets with a
  * matching class prefix.
  *
- * Each executable in [zedlets] from the directory [dir] is matched against
+ * Each executable in [zcp->zedlets] from the directory [zcp->zedlet_dir]
- * the event's [class], [subclass], and the "all" class (which matches
+ * is matched against the event's [class], [subclass], and the "all" class
- * all events).  Every zedlet with a matching class prefix is invoked.
+ * (which matches all events).
+ * Every zedlet with a matching class prefix is invoked.
  * The NAME=VALUE strings in [envs] will be passed to the zedlet as
  * environment variables.
  *
- * The file descriptor [zfd] is the zevent_fd used to track the
+ * The file descriptor [zcp->zevent_fd] is the zevent_fd used to track the
  * current cursor location within the zevent nvlist.
  *
  * Return 0 on success, -1 on error.
  */
 int
 zed_exec_process(uint64_t eid, const char *class, const char *subclass,
-    const char *dir, zed_strings_t *zedlets, zed_strings_t *envs, int zfd)
+    struct zed_conf *zcp, zed_strings_t *envs)
 {
 	const char *class_strings[4];
 	const char *allclass = "all";
@@ -201,9 +325,22 @@ zed_exec_process(uint64_t eid, const char *class, const char *subclass,
 	char **e;
 	int n;
 
-	if (!dir || !zedlets || !envs || zfd < 0)
+	if (!zcp->zedlet_dir || !zcp->zedlets || !envs || zcp->zevent_fd < 0)
 		return (-1);
 
+	if (_reap_children_tid == (pthread_t)-1) {
+		_launched_processes_limit = zcp->max_jobs;
+
+		if (pthread_create(&_reap_children_tid, NULL,
+		    _reap_children, NULL) != 0)
+			return (-1);
+		pthread_setname_np(_reap_children_tid, "reap ZEDLETs");
+
+		avl_create(&_launched_processes, _launched_process_node_compare,
+		    sizeof (struct launched_process_node),
+		    offsetof(struct launched_process_node, node));
+	}
+
 	csp = class_strings;
 
 	if (class)
@@ -219,11 +356,13 @@ zed_exec_process(uint64_t eid, const char *class, const char *subclass,
 
 	e = _zed_exec_create_env(envs);
 
-	for (z = zed_strings_first(zedlets); z; z = zed_strings_next(zedlets)) {
+	for (z = zed_strings_first(zcp->zedlets); z;
+	    z = zed_strings_next(zcp->zedlets)) {
 		for (csp = class_strings; *csp; csp++) {
 			n = strlen(*csp);
 			if ((strncmp(z, *csp, n) == 0) && !isalpha(z[n]))
-				_zed_exec_fork_child(eid, dir, z, e, zfd);
+				_zed_exec_fork_child(eid, zcp->zedlet_dir,
+				    z, e, zcp->zevent_fd, zcp->do_foreground);
 		}
 	}
 	free(e);

cmd/zed/zed_exec.h

@@ -1,9 +1,9 @@
 /*
- * This file is part of the ZFS Event Daemon (ZED)
+ * This file is part of the ZFS Event Daemon (ZED).
- * for ZFS on Linux (ZoL) <http://zfsonlinux.org/>.
+ *
  * Developed at Lawrence Livermore National Laboratory (LLNL-CODE-403049).
  * Copyright (C) 2013-2014 Lawrence Livermore National Security, LLC.
- * Refer to the ZoL git commit log for authoritative copyright attribution.
+ * Refer to the OpenZFS git commit log for authoritative copyright attribution.
  *
  * The contents of this file are subject to the terms of the
  * Common Development and Distribution License Version 1.0 (CDDL-1.0).
@@ -16,9 +16,12 @@
 #define	ZED_EXEC_H
 
 #include <stdint.h>
+#include "zed_strings.h"
+#include "zed_conf.h"
+
+void zed_exec_fini(void);
 
 int zed_exec_process(uint64_t eid, const char *class, const char *subclass,
-    const char *dir, zed_strings_t *zedlets, zed_strings_t *envs,
+    struct zed_conf *zcp, zed_strings_t *envs);
-    int zevent_fd);
 
 #endif	/* !ZED_EXEC_H */

cmd/zed/zed_file.c

@@ -1,9 +1,9 @@
 /*
- * This file is part of the ZFS Event Daemon (ZED)
+ * This file is part of the ZFS Event Daemon (ZED).
- * for ZFS on Linux (ZoL) <http://zfsonlinux.org/>.
+ *
  * Developed at Lawrence Livermore National Laboratory (LLNL-CODE-403049).
  * Copyright (C) 2013-2014 Lawrence Livermore National Security, LLC.
- * Refer to the ZoL git commit log for authoritative copyright attribution.
+ * Refer to the OpenZFS git commit log for authoritative copyright attribution.
  *
  * The contents of this file are subject to the terms of the
  * Common Development and Distribution License Version 1.0 (CDDL-1.0).
@@ -12,72 +12,17 @@
  * You may not use this file except in compliance with the license.
  */
 
+#include <dirent.h>
 #include <errno.h>
 #include <fcntl.h>
 #include <limits.h>
 #include <string.h>
-#include <sys/resource.h>
 #include <sys/stat.h>
 #include <sys/types.h>
 #include <unistd.h>
+#include "zed_file.h"
 #include "zed_log.h"
 
-/*
- * Read up to [n] bytes from [fd] into [buf].
- * Return the number of bytes read, 0 on EOF, or -1 on error.
- */
-ssize_t
-zed_file_read_n(int fd, void *buf, size_t n)
-{
-	unsigned char *p;
-	size_t n_left;
-	ssize_t n_read;
-
-	p = buf;
-	n_left = n;
-	while (n_left > 0) {
-		if ((n_read = read(fd, p, n_left)) < 0) {
-			if (errno == EINTR)
-				continue;
-			else
-				return (-1);
-
-		} else if (n_read == 0) {
-			break;
-		}
-		n_left -= n_read;
-		p += n_read;
-	}
-	return (n - n_left);
-}
-
-/*
- * Write [n] bytes from [buf] out to [fd].
- * Return the number of bytes written, or -1 on error.
- */
-ssize_t
-zed_file_write_n(int fd, void *buf, size_t n)
-{
-	const unsigned char *p;
-	size_t n_left;
-	ssize_t n_written;
-
-	p = buf;
-	n_left = n;
-	while (n_left > 0) {
-		if ((n_written = write(fd, p, n_left)) < 0) {
-			if (errno == EINTR)
-				continue;
-			else
-				return (-1);
-
-		}
-		n_left -= n_written;
-		p += n_written;
-	}
-	return (n);
-}
-
 /*
  * Set an exclusive advisory lock on the open file descriptor [fd].
  * Return 0 on success, 1 if a conflicting lock is held by another process,
@@ -159,6 +104,13 @@ zed_file_is_locked(int fd)
 	return (lock.l_pid);
 }
 
+
+#if __APPLE__
+#define	PROC_SELF_FD "/dev/fd"
+#else /* Linux-compatible layout */
+#define	PROC_SELF_FD "/proc/self/fd"
+#endif
+
 /*
  * Close all open file descriptors greater than or equal to [lowfd].
  * Any errors encountered while closing file descriptors are ignored.
@@ -166,51 +118,24 @@ zed_file_is_locked(int fd)
 void
 zed_file_close_from(int lowfd)
 {
-	const int maxfd_def = 256;
+	int errno_bak = errno;
-	int errno_bak;
+	int maxfd = 0;
-	struct rlimit rl;
-	int maxfd;
 	int fd;
+	DIR *fddir;
+	struct dirent *fdent;
 
-	errno_bak = errno;
+	if ((fddir = opendir(PROC_SELF_FD)) != NULL) {
-
+		while ((fdent = readdir(fddir)) != NULL) {
-	if (getrlimit(RLIMIT_NOFILE, &rl) < 0) {
+			fd = atoi(fdent->d_name);
-		maxfd = maxfd_def;
+			if (fd > maxfd && fd != dirfd(fddir))
-	} else if (rl.rlim_max == RLIM_INFINITY) {
+				maxfd = fd;
-		maxfd = maxfd_def;
+		}
+		(void) closedir(fddir);
 	} else {
-		maxfd = rl.rlim_max;
+		maxfd = sysconf(_SC_OPEN_MAX);
 	}
 	for (fd = lowfd; fd < maxfd; fd++)
 		(void) close(fd);
 
 	errno = errno_bak;
 }
-
-/*
- * Set the CLOEXEC flag on file descriptor [fd] so it will be automatically
- * closed upon successful execution of one of the exec functions.
- * Return 0 on success, or -1 on error.
- *
- * FIXME: No longer needed?
- */
-int
-zed_file_close_on_exec(int fd)
-{
-	int flags;
-
-	if (fd < 0) {
-		errno = EBADF;
-		return (-1);
-	}
-	flags = fcntl(fd, F_GETFD);
-	if (flags == -1)
-		return (-1);
-
-	flags |= FD_CLOEXEC;
-
-	if (fcntl(fd, F_SETFD, flags) == -1)
-		return (-1);
-
-	return (0);
-}

cmd/zed/zed_file.h

@@ -1,9 +1,9 @@
 /*
- * This file is part of the ZFS Event Daemon (ZED)
+ * This file is part of the ZFS Event Daemon (ZED).
- * for ZFS on Linux (ZoL) <http://zfsonlinux.org/>.
+ *
  * Developed at Lawrence Livermore National Laboratory (LLNL-CODE-403049).
  * Copyright (C) 2013-2014 Lawrence Livermore National Security, LLC.
- * Refer to the ZoL git commit log for authoritative copyright attribution.
+ * Refer to the OpenZFS git commit log for authoritative copyright attribution.
  *
  * The contents of this file are subject to the terms of the
  * Common Development and Distribution License Version 1.0 (CDDL-1.0).
@@ -18,10 +18,6 @@
 #include <sys/types.h>
 #include <unistd.h>
 
-ssize_t zed_file_read_n(int fd, void *buf, size_t n);
-
-ssize_t zed_file_write_n(int fd, void *buf, size_t n);
-
 int zed_file_lock(int fd);
 
 int zed_file_unlock(int fd);
@@ -30,6 +26,4 @@ pid_t zed_file_is_locked(int fd);
 
 void zed_file_close_from(int fd);
 
-int zed_file_close_on_exec(int fd);
-
 #endif	/* !ZED_FILE_H */

cmd/zed/zed_log.c

@@ -1,9 +1,9 @@
 /*
- * This file is part of the ZFS Event Daemon (ZED)
+ * This file is part of the ZFS Event Daemon (ZED).
- * for ZFS on Linux (ZoL) <http://zfsonlinux.org/>.
+ *
  * Developed at Lawrence Livermore National Laboratory (LLNL-CODE-403049).
  * Copyright (C) 2013-2014 Lawrence Livermore National Security, LLC.
- * Refer to the ZoL git commit log for authoritative copyright attribution.
+ * Refer to the OpenZFS git commit log for authoritative copyright attribution.
  *
  * The contents of this file are subject to the terms of the
  * Common Development and Distribution License Version 1.0 (CDDL-1.0).

cmd/zed/zed_log.h

@@ -1,9 +1,9 @@
 /*
- * This file is part of the ZFS Event Daemon (ZED)
+ * This file is part of the ZFS Event Daemon (ZED).
- * for ZFS on Linux (ZoL) <http://zfsonlinux.org/>.
+ *
  * Developed at Lawrence Livermore National Laboratory (LLNL-CODE-403049).
  * Copyright (C) 2013-2014 Lawrence Livermore National Security, LLC.
- * Refer to the ZoL git commit log for authoritative copyright attribution.
+ * Refer to the OpenZFS git commit log for authoritative copyright attribution.
  *
  * The contents of this file are subject to the terms of the
  * Common Development and Distribution License Version 1.0 (CDDL-1.0).

cmd/zed/zed_strings.c

@@ -1,9 +1,9 @@
 /*
- * This file is part of the ZFS Event Daemon (ZED)
+ * This file is part of the ZFS Event Daemon (ZED).
- * for ZFS on Linux (ZoL) <http://zfsonlinux.org/>.
+ *
  * Developed at Lawrence Livermore National Laboratory (LLNL-CODE-403049).
  * Copyright (C) 2013-2014 Lawrence Livermore National Security, LLC.
- * Refer to the ZoL git commit log for authoritative copyright attribution.
+ * Refer to the OpenZFS git commit log for authoritative copyright attribution.
  *
  * The contents of this file are subject to the terms of the
  * Common Development and Distribution License Version 1.0 (CDDL-1.0).
@@ -108,7 +108,7 @@ _zed_strings_node_destroy(zed_strings_node_t *np)
  * If [key] is specified, it will be used to index the node; otherwise,
  * the string [val] will be used.
  */
-zed_strings_node_t *
+static zed_strings_node_t *
 _zed_strings_node_create(const char *key, const char *val)
 {
 	zed_strings_node_t *np;

cmd/zed/zed_strings.h

@@ -1,9 +1,9 @@
 /*
- * This file is part of the ZFS Event Daemon (ZED)
+ * This file is part of the ZFS Event Daemon (ZED).
- * for ZFS on Linux (ZoL) <http://zfsonlinux.org/>.
+ *
  * Developed at Lawrence Livermore National Laboratory (LLNL-CODE-403049).
  * Copyright (C) 2013-2014 Lawrence Livermore National Security, LLC.
- * Refer to the ZoL git commit log for authoritative copyright attribution.
+ * Refer to the OpenZFS git commit log for authoritative copyright attribution.
  *
  * The contents of this file are subject to the terms of the
  * Common Development and Distribution License Version 1.0 (CDDL-1.0).

cmd/zfs/Makefile.am

@@ -1,9 +1,5 @@
 include $(top_srcdir)/config/Rules.am
 
-DEFAULT_INCLUDES += \
-	-I$(top_srcdir)/include \
-	-I$(top_srcdir)/lib/libspl/include
-
 sbin_PROGRAMS = zfs
 
 zfs_SOURCES = \
@@ -15,7 +11,15 @@ zfs_SOURCES = \
 	zfs_projectutil.h
 
 zfs_LDADD = \
-	$(top_builddir)/lib/libnvpair/libnvpair.la \
+	$(abs_top_builddir)/lib/libzfs/libzfs.la \
-	$(top_builddir)/lib/libuutil/libuutil.la \
+	$(abs_top_builddir)/lib/libzfs_core/libzfs_core.la \
-	$(top_builddir)/lib/libzfs/libzfs.la \
+	$(abs_top_builddir)/lib/libnvpair/libnvpair.la \
-	$(top_builddir)/lib/libzfs_core/libzfs_core.la
+	$(abs_top_builddir)/lib/libuutil/libuutil.la
+
+zfs_LDADD += $(LTLIBINTL)
+
+if BUILD_FREEBSD
+zfs_LDADD += -lgeom -ljail
+endif
+
+include $(top_srcdir)/config/CppCheck.am

cmd/zfs/zfs_util.h

@@ -33,7 +33,7 @@ extern "C" {
 
 void * safe_malloc(size_t size);
 void nomem(void);
-libzfs_handle_t *g_zfs;
+extern libzfs_handle_t *g_zfs;
 
 #ifdef	__cplusplus
 }

cmd/zfs_ids_to_path/.gitignore

@@ -0,0 +1 @@
+zfs_ids_to_path

cmd/zfs_ids_to_path/Makefile.am

@@ -0,0 +1,11 @@
+include $(top_srcdir)/config/Rules.am
+
+sbin_PROGRAMS = zfs_ids_to_path
+
+zfs_ids_to_path_SOURCES = \
+	zfs_ids_to_path.c
+
+zfs_ids_to_path_LDADD = \
+        $(abs_top_builddir)/lib/libzfs/libzfs.la
+
+include $(top_srcdir)/config/CppCheck.am

cmd/zfs_ids_to_path/zfs_ids_to_path.c

@@ -0,0 +1,96 @@
+/*
+ * CDDL HEADER START
+ *
+ * The contents of this file are subject to the terms of the
+ * Common Development and Distribution License (the "License").
+ * You may not use this file except in compliance with the License.
+ *
+ * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
+ * or http://www.opensolaris.org/os/licensing.
+ * See the License for the specific language governing permissions
+ * and limitations under the License.
+ *
+ * When distributing Covered Code, include this CDDL HEADER in each
+ * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
+ * If applicable, add the following below this CDDL HEADER, with the
+ * fields enclosed by brackets "[]" replaced with your own identifying
+ * information: Portions Copyright [yyyy] [name of copyright owner]
+ *
+ * CDDL HEADER END
+ */
+/*
+ * Copyright (c) 2019 by Delphix. All rights reserved.
+ */
+#include <libintl.h>
+#include <unistd.h>
+#include <sys/types.h>
+#include <stdint.h>
+#include <libzfs.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <errno.h>
+
+libzfs_handle_t *g_zfs;
+
+static void
+usage(int err)
+{
+	fprintf(stderr, "Usage: zfs_ids_to_path [-v] <pool> <objset id> "
+	    "<object id>\n");
+	exit(err);
+}
+
+int
+main(int argc, char **argv)
+{
+	boolean_t verbose = B_FALSE;
+	int c;
+	while ((c = getopt(argc, argv, "v")) != -1) {
+		switch (c) {
+		case 'v':
+			verbose = B_TRUE;
+			break;
+		}
+	}
+	argc -= optind;
+	argv += optind;
+
+	if (argc != 3) {
+		(void) fprintf(stderr, "Incorrect number of arguments: %d\n",
+		    argc);
+		usage(1);
+	}
+
+	uint64_t objset, object;
+	if (sscanf(argv[1], "%llu", (u_longlong_t *)&objset) != 1) {
+		(void) fprintf(stderr, "Invalid objset id: %s\n", argv[1]);
+		usage(2);
+	}
+	if (sscanf(argv[2], "%llu", (u_longlong_t *)&object) != 1) {
+		(void) fprintf(stderr, "Invalid object id: %s\n", argv[2]);
+		usage(3);
+	}
+	if ((g_zfs = libzfs_init()) == NULL) {
+		(void) fprintf(stderr, "%s\n", libzfs_error_init(errno));
+		return (4);
+	}
+	zpool_handle_t *pool = zpool_open(g_zfs, argv[0]);
+	if (pool == NULL) {
+		fprintf(stderr, "Could not open pool %s\n", argv[0]);
+		libzfs_fini(g_zfs);
+		return (5);
+	}
+
+	char pathname[PATH_MAX * 2];
+	if (verbose) {
+		zpool_obj_to_path_ds(pool, objset, object, pathname,
+		    sizeof (pathname));
+	} else {
+		zpool_obj_to_path(pool, objset, object, pathname,
+		    sizeof (pathname));
+	}
+	printf("%s\n", pathname);
+	zpool_close(pool);
+	libzfs_fini(g_zfs);
+	return (0);
+}

cmd/zgenhostid/.gitignore

@@ -0,0 +1 @@
+/zgenhostid

cmd/zgenhostid/Makefile.am

@@ -1 +1,7 @@
-dist_bin_SCRIPTS = zgenhostid
+include $(top_srcdir)/config/Rules.am
+
+sbin_PROGRAMS = zgenhostid
+
+zgenhostid_SOURCES = zgenhostid.c
+
+include $(top_srcdir)/config/CppCheck.am

cmd/zgenhostid/zgenhostid

@@ -1,61 +0,0 @@
-#!/bin/bash
-
-# Emulate genhostid(1) available on RHEL/CENTOS, for use on distros
-# which do not provide that utility.
-#
-# Usage:
-#    zgenhostid
-#    zgenhostid <value>
-#
-# If /etc/hostid already exists and is size > 0, the script exits immediately
-# and changes nothing.  Unlike genhostid, this generates an error message.
-#
-# The first form generates a random hostid and stores it in /etc/hostid.
-# The second form checks that the provided value is between 0x1 and 0xFFFFFFFF
-# and if so, stores it in /etc/hostid.  This form is not supported by
-# genhostid(1).
-
-hostid_file=/etc/hostid
-
-function usage {
-	echo "$0 [value]"
-	echo "If $hostid_file is not present, store a hostid in it." >&2
-	echo "The optional value must be an 8-digit hex number between" >&2
-	echo "1 and 2^32-1.  If no value is provided, a random one will" >&2
-	echo "be generated.  The value must be unique among your systems." >&2
-}
-
-# hostid(1) ignores contents of /etc/hostid if size < 4 bytes.  It would
-# be better if this checked size >= 4 bytes but it the method must be
-# widely portable.
-if [ -s $hostid_file ]; then
-	echo "$hostid_file already exists.  No change made." >&2
-	exit 1
-fi
-
-if [ -n "$1" ]; then
-	host_id=$1
-else
-	# $RANDOM goes from 0..32k-1
-	number=$((((RANDOM % 4) * 32768 + RANDOM) * 32768 + RANDOM))
-	host_id=$(printf "%08x" $number)
-fi
-
-if egrep -o '^0{8}$' <<< $host_id >/dev/null 2>&1; then
-	usage
-	exit 2
-fi
-
-if ! egrep -o '^[a-fA-F0-9]{8}$' <<< $host_id >/dev/null 2>&1; then
-	usage
-	exit 3
-fi
-
-a=${host_id:6:2}
-b=${host_id:4:2}
-c=${host_id:2:2}
-d=${host_id:0:2}
-
-echo -ne \\x$a\\x$b\\x$c\\x$d > $hostid_file
-
-exit 0

cmd/zgenhostid/zgenhostid.c

@@ -0,0 +1,141 @@
+/*
+ * CDDL HEADER START
+ *
+ * The contents of this file are subject to the terms of the
+ * Common Development and Distribution License (the "License").
+ * You may not use this file except in compliance with the License.
+ *
+ * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
+ * or http://www.opensolaris.org/os/licensing.
+ * See the License for the specific language governing permissions
+ * and limitations under the License.
+ *
+ * When distributing Covered Code, include this CDDL HEADER in each
+ * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
+ * If applicable, add the following below this CDDL HEADER, with the
+ * fields enclosed by brackets "[]" replaced with your own identifying
+ * information: Portions Copyright [yyyy] [name of copyright owner]
+ *
+ * CDDL HEADER END
+ */
+
+/*
+ * Copyright (c) 2020, Georgy Yakovlev.  All rights reserved.
+ */
+
+#include <errno.h>
+#include <fcntl.h>
+#include <getopt.h>
+#include <inttypes.h>
+#include <limits.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/stat.h>
+#include <time.h>
+#include <unistd.h>
+
+static __attribute__((noreturn)) void
+usage(void)
+{
+	(void) fprintf(stderr,
+	    "usage: zgenhostid [-fh] [-o path] [value]\n\n"
+	    "  -f\t\t force hostid file write\n"
+	    "  -h\t\t print this usage and exit\n"
+	    "  -o <filename>\t write hostid to this file\n\n"
+	    "If hostid file is not present, store a hostid in it.\n"
+	    "The optional value should be an 8-digit hex number between"
+	    " 1 and 2^32-1.\n"
+	    "If the value is 0 or no value is provided, a random one"
+	    " will be generated.\n"
+	    "The value must be unique among your systems.\n");
+	exit(EXIT_FAILURE);
+}
+
+int
+main(int argc, char **argv)
+{
+	/* default file path, can be optionally set by user */
+	const char *path = "/etc/hostid";
+	/* holds converted user input or lrand48() generated value */
+	unsigned long input_i = 0;
+
+	int opt;
+	int force_fwrite = 0;
+	while ((opt = getopt_long(argc, argv, "fo:h?", 0, 0)) != -1) {
+		switch (opt) {
+		case 'f':
+			force_fwrite = 1;
+			break;
+		case 'o':
+			path = optarg;
+			break;
+		case 'h':
+		case '?':
+			usage();
+		}
+	}
+
+	char *in_s = argv[optind];
+	if (in_s != NULL) {
+		/* increment pointer by 2 if string is 0x prefixed */
+		if (strncasecmp("0x", in_s, 2) == 0) {
+			in_s += 2;
+		}
+
+		/* need to be exactly 8 characters */
+		const char *hex = "0123456789abcdefABCDEF";
+		if (strlen(in_s) != 8 || strspn(in_s, hex) != 8) {
+			fprintf(stderr, "%s\n", strerror(ERANGE));
+			usage();
+		}
+
+		input_i = strtoul(in_s, NULL, 16);
+		if (errno != 0) {
+			perror("strtoul");
+			exit(EXIT_FAILURE);
+		}
+
+		if (input_i > UINT32_MAX) {
+			fprintf(stderr, "%s\n", strerror(ERANGE));
+			usage();
+		}
+	}
+
+	struct stat fstat;
+	if (force_fwrite == 0 && stat(path, &fstat) == 0 &&
+	    S_ISREG(fstat.st_mode)) {
+		fprintf(stderr, "%s: %s\n", path, strerror(EEXIST));
+		exit(EXIT_FAILURE);
+	}
+
+	/*
+	 * generate if not provided by user
+	 * also handle unlikely zero return from lrand48()
+	 */
+	while (input_i == 0) {
+		srand48(getpid() ^ time(NULL));
+		input_i = lrand48();
+	}
+
+	FILE *fp = fopen(path, "wb");
+	if (!fp) {
+		perror("fopen");
+		exit(EXIT_FAILURE);
+	}
+
+	/*
+	 * we need just 4 bytes in native endianness
+	 * not using sethostid() because it may be missing or just a stub
+	 */
+	uint32_t hostid = input_i;
+	int written = fwrite(&hostid, 1, 4, fp);
+	if (written != 4) {
+		perror("fwrite");
+		exit(EXIT_FAILURE);
+	}
+
+	fclose(fp);
+	exit(EXIT_SUCCESS);
+}

cmd/zhack/Makefile.am

@@ -1,8 +1,7 @@
 include $(top_srcdir)/config/Rules.am
 
-DEFAULT_INCLUDES += \
+# Unconditionally enable debugging for zhack
-	-I$(top_srcdir)/include \
+AM_CPPFLAGS += -DDEBUG -UNDEBUG -DZFS_DEBUG
-	-I$(top_srcdir)/lib/libspl/include
 
 sbin_PROGRAMS = zhack
 
@@ -10,5 +9,8 @@ zhack_SOURCES = \
 	zhack.c
 
 zhack_LDADD = \
-	$(top_builddir)/lib/libnvpair/libnvpair.la \
+	$(abs_top_builddir)/lib/libzpool/libzpool.la \
-	$(top_builddir)/lib/libzpool/libzpool.la
+	$(abs_top_builddir)/lib/libzfs_core/libzfs_core.la \
+	$(abs_top_builddir)/lib/libnvpair/libnvpair.la
+
+include $(top_srcdir)/config/CppCheck.am

cmd/zhack/zhack.c

@@ -48,16 +48,15 @@
 #include <sys/zio_compress.h>
 #include <sys/zfeature.h>
 #include <sys/dmu_tx.h>
+#include <zfeature_common.h>
 #include <libzutil.h>
 
-extern boolean_t zfeature_checks_disable;
-
 const char cmdname[] = "zhack";
 static importargs_t g_importargs;
 static char *g_pool;
 static boolean_t g_readonly;
 
-static void
+static __attribute__((noreturn)) void
 usage(void)
 {
 	(void) fprintf(stderr,
@@ -82,7 +81,7 @@ usage(void)
 }
 
 
-static void
+static __attribute__((noreturn)) __attribute__((format(printf, 3, 4))) void
 fatal(spa_t *spa, void *tag, const char *fmt, ...)
 {
 	va_list ap;
@@ -103,8 +102,8 @@ fatal(spa_t *spa, void *tag, const char *fmt, ...)
 
 /* ARGSUSED */
 static int
-space_delta_cb(dmu_object_type_t bonustype, void *data,
+space_delta_cb(dmu_object_type_t bonustype, const void *data,
-    uint64_t *userp, uint64_t *groupp, uint64_t *projectp)
+    zfs_file_info_t *zoi)
 {
 	/*
 	 * Is it a valid type of object to track?
@@ -113,7 +112,6 @@ space_delta_cb(dmu_object_type_t bonustype, void *data,
 		return (ENOENT);
 	(void) fprintf(stderr, "modifying object that needs user accounting");
 	abort();
-	/* NOTREACHED */
 }
 
 /*
@@ -126,7 +124,8 @@ zhack_import(char *target, boolean_t readonly)
 	nvlist_t *props;
 	int error;
 
-	kernel_init(readonly ? FREAD : (FREAD | FWRITE));
+	kernel_init(readonly ? SPA_MODE_READ :
+	    (SPA_MODE_READ | SPA_MODE_WRITE));
 
 	dmu_objset_register_type(DMU_OST_ZFS, space_delta_cb);
 
@@ -149,6 +148,7 @@ zhack_import(char *target, boolean_t readonly)
 	zfeature_checks_disable = B_TRUE;
 	error = spa_import(target, config, props,
 	    (readonly ?  ZFS_IMPORT_SKIP_MMP : ZFS_IMPORT_NORMAL));
+	fnvlist_free(config);
 	zfeature_checks_disable = B_FALSE;
 	if (error == EEXIST)
 		error = 0;
@@ -317,7 +317,8 @@ zhack_do_feature_enable(int argc, char **argv)
 	mos = spa->spa_meta_objset;
 
 	if (zfeature_is_supported(feature.fi_guid))
-		fatal(spa, FTAG, "'%s' is a real feature, will not enable");
+		fatal(spa, FTAG, "'%s' is a real feature, will not enable",
+		    feature.fi_guid);
 	if (0 == zap_contains(mos, spa->spa_feat_desc_obj, feature.fi_guid))
 		fatal(spa, FTAG, "feature already enabled: %s",
 		    feature.fi_guid);
@@ -411,7 +412,8 @@ zhack_do_feature_ref(int argc, char **argv)
 
 	if (zfeature_is_supported(feature.fi_guid)) {
 		fatal(spa, FTAG,
-		    "'%s' is a real feature, will not change refcount");
+		    "'%s' is a real feature, will not change refcount",
+		    feature.fi_guid);
 	}
 
 	if (0 == zap_contains(mos, spa->spa_feat_for_read_obj,

cmd/zinject/Makefile.am

@@ -1,9 +1,5 @@
 include $(top_srcdir)/config/Rules.am
 
-DEFAULT_INCLUDES += \
-	-I$(top_srcdir)/include \
-	-I$(top_srcdir)/lib/libspl/include
-
 sbin_PROGRAMS = zinject
 
 zinject_SOURCES = \
@@ -12,5 +8,8 @@ zinject_SOURCES = \
 	zinject.h
 
 zinject_LDADD = \
-	$(top_builddir)/lib/libnvpair/libnvpair.la \
+	$(abs_top_builddir)/lib/libzfs/libzfs.la \
-	$(top_builddir)/lib/libzfs/libzfs.la
+	$(abs_top_builddir)/lib/libzfs_core/libzfs_core.la \
+	$(abs_top_builddir)/lib/libnvpair/libnvpair.la
+
+include $(top_srcdir)/config/CppCheck.am