vmtest: build and run bpf kernel selftests against various kernels

Run kernel selftests in vmtest with the goal to test libbpf backward
compatibility with older kernels.
The list of kernels should be specified in .travis.yml config in
`jobs` section, e.g. KERNEL=5.5.0.
Enlisted kernel releases
- 5.5.0 # built from main
- 5.5.0-rc6 # built from bpf-next
- LATEST

The kernel specified as 'LATEST' in .travis.yml is built from bpf-next kernel
tree, the rest of the kernels are downloaded from the specified in INDEX file.
The kernel sources from bpf-next are manually patched with [1] from bpf tree to
fix ranqslower build. This workaround should be removed after the patch is merged
from bpf to bpf-next tree.
Due to kernel sources being checked out the duration of the LATEST kernel test is
~30m.

bpf selftests are built from tools/testing/selftests/bpf/ of bpf-next tree with
HEAD revision set to CHECKPOINT-COMMIT specified in libbpf so selftests and
libbpf are in sync.
Currently only programs are tested with test_progs program, test_maps and
test_verifier should follow.
test_progs are run with blacklist required due to:
- some features, e.g. fentry/fexit are not supported in older kernels
- environment limitations, e.g an absence of the recent pahole in Debian
- incomplete disk image

The blacklist is passed to test_progs with -b option as specified in [2]
patch set.

Most of the preceeding tests are disabled due to incomplete disk image currenly
lacking proper networking settings.
For the LATEST kernel fome fentry/fexit tests are disabled due to pahole v1.16
is not abailible in Debian yet.

Next steps are resolving issues with blacklisted tests, enabling maps and
verifier testing, expanding the list of tested kernels.

[1] https://lore.kernel.org/bpf/908498f794661c44dca54da9e09dc0c382df6fcb.1580425879.git.hex@fb.com/t.mbox.gz
[2] https://www.spinics.net/lists/netdev/msg625192.html

.lgtm.yml

@@ -0,0 +1,14 @@
+# vi: set ts=2 sw=2:
+extraction:
+  cpp:
+    prepare:
+      packages:
+        - libelf-dev
+        - pkg-config
+    after_prepare:
+      # As the buildsystem detection by LGTM is performed _only_ during the
+      # 'configure' phase, we need to trick LGTM we use a supported build
+      # system (configure, meson, cmake, etc.). This way LGTM correctly detects
+      # that our sources are in the src/ subfolder.
+      - touch src/configure
+      - chmod +x src/configure

.travis.yml

@@ -1,13 +1,64 @@
 sudo: required
-dist: xenial
+language: bash
+dist: bionic
 services:
     - docker
 
 env:
     global:
-        - AUTHOR_EMAIL="$(git log -1 $TRAVIS_COMMIT --pretty=\"%aE\")"
+        - PROJECT_NAME='libbpf'
+        - AUTHOR_EMAIL="$(git log -1 --pretty=\"%aE\")"
         - CI_MANAGERS="$TRAVIS_BUILD_DIR/travis-ci/managers"
+        - VMTEST_ROOT="$TRAVIS_BUILD_DIR/travis-ci/vmtest"
         - REPO_ROOT="$TRAVIS_BUILD_DIR"
+        - GIT_FETCH_DEPTH=64
+        - VMTEST_SETUPCMD="PROJECT_NAME=${PROJECT_NAME} ./${PROJECT_NAME}/travis-ci/vmtest/run_selftests.sh"
+    jobs:
+        # Setup command override.
+        # 5.5.0-rc6 is built from bpf-next; TODO(hex@): remove when pahole v1.16 is available
+        - KERNEL=5.5.0-rc6
+        - KERNEL=5.5.0
+        - KERNEL=LATEST
+
+addons:
+    apt:
+        packages:
+            - qemu-kvm
+            - zstd
+            - binutils-dev
+            - elfutils
+            - libcap-dev
+            - libelf-dev
+install: sudo adduser "${USER}" kvm
+before_script:
+    - wget -O - https://apt.llvm.org/llvm-snapshot.gpg.key | sudo apt-key add -
+    - echo "deb http://apt.llvm.org/bionic/ llvm-toolchain-bionic-10 main" | sudo tee -a /etc/apt/sources.list
+    - echo "deb http://archive.ubuntu.com/ubuntu eoan main restricted universe multiverse" | sudo tee -a /etc/apt/sources.list
+    - sudo apt-get -qq update
+    - sudo apt-get -y install dwarves=1.15-1
+    - sudo apt-get -qq -y install clang-10 lld-10 llvm-10
+    - if [[ "${KERNEL}" = 'LATEST' ]]; then ${VMTEST_ROOT}/build_latest_kernel.sh travis-ci/vmtest/bpf-next; fi
+    - ${VMTEST_ROOT}/prepare_selftests.sh travis-ci/vmtest/bpf-next
+    # Escape whitespace characters.
+    - setup_cmd=$(sed 's/\([[:space:]]\)/\\\1/g' <<< "${VMTEST_SETUPCMD}")
+    - if [[ "${KERNEL}" = 'LATEST' ]]; then
+        sudo -E sudo -E -u "${USER}" "${VMTEST_ROOT}/run.sh" -b travis-ci/vmtest/bpf-next -o -d ~ -s "${setup_cmd}" ~/root.img;
+      else
+        sudo -E sudo -E -u "${USER}" "${VMTEST_ROOT}/run.sh" -k "${KERNEL}*" -o -d ~ -s "${setup_cmd}" ~/root.img;
+      fi; exitstatus=$?
+    - test $exitstatus -le 1
+script:
+    - test $exitstatus -eq 0
+
+stages:
+    # Run Coverity periodically instead of for each PR for following reasons:
+    # 1) Coverity jobs are heavily rate-limited
+    # 2) Due to security restrictions of encrypted environment variables
+    #    in Travis CI, pull requests made from forks can't access encrypted
+    #    env variables, making Coverity unusable
+    #    See: https://docs.travis-ci.com/user/pull-requests#pull-requests-and-security-restrictions
+    - name: Coverity
+      if: type = cron
 
 jobs:
     include:
@@ -22,10 +73,10 @@ jobs:
               - docker --version
           install:
               - $CI_MANAGERS/debian.sh SETUP
+          # Override before_script: so VMTEST before_install commands are not executed.
+          before_script: true
           script:
-              - set -e
+              - $CI_MANAGERS/debian.sh RUN || travis_terminate
-              - $CI_MANAGERS/debian.sh RUN
-              - set +e
           after_script:
               - $CI_MANAGERS/debian.sh CLEANUP
 
@@ -39,10 +90,9 @@ jobs:
               - docker --version
           install:
               - $CI_MANAGERS/debian.sh SETUP
+          before_script: true
           script:
-              - set -e
+              - $CI_MANAGERS/debian.sh RUN_ASAN || travis_terminate
-              - $CI_MANAGERS/debian.sh RUN_ASAN
-              - set +e
           after_script:
               - $CI_MANAGERS/debian.sh CLEANUP
 
@@ -56,10 +106,9 @@ jobs:
               - docker --version
           install:
               - $CI_MANAGERS/debian.sh SETUP
+          before_script: true
           script:
-              - set -e
+              - $CI_MANAGERS/debian.sh RUN_CLANG || travis_terminate
-              - $CI_MANAGERS/debian.sh RUN_CLANG
-              - set +e
           after_script:
               - $CI_MANAGERS/debian.sh CLEANUP
 
@@ -73,10 +122,9 @@ jobs:
               - docker --version
           install:
               - $CI_MANAGERS/debian.sh SETUP
+          before_script: true
           script:
-              - set -e
+              - $CI_MANAGERS/debian.sh RUN_CLANG_ASAN || travis_terminate
-              - $CI_MANAGERS/debian.sh RUN_CLANG_ASAN
-              - set +e
           after_script:
               - $CI_MANAGERS/debian.sh CLEANUP
 
@@ -90,10 +138,9 @@ jobs:
               - docker --version
           install:
               - $CI_MANAGERS/debian.sh SETUP
+          before_script: true
           script:
-              - set -e
+              - $CI_MANAGERS/debian.sh RUN_GCC8 || travis_terminate
-              - $CI_MANAGERS/debian.sh RUN_GCC8
-              - set +e
           after_script:
               - $CI_MANAGERS/debian.sh CLEANUP
 
@@ -107,16 +154,58 @@ jobs:
               - docker --version
           install:
               - $CI_MANAGERS/debian.sh SETUP
+          before_script: true
           script:
-              - set -e
+              - $CI_MANAGERS/debian.sh RUN_GCC8_ASAN || travis_terminate
-              - $CI_MANAGERS/debian.sh RUN_GCC8_ASAN
-              - set +e
           after_script:
               - $CI_MANAGERS/debian.sh CLEANUP
 
-        - name: Ubuntu Xenial
+        - name: Ubuntu Bionic
           language: bash
+          before_script: true
           script:
-              - set -e
+              - sudo $CI_MANAGERS/ubuntu.sh || travis_terminate
-              - sudo $CI_MANAGERS/xenial.sh
+
-              - set +e
+        - name: Ubuntu Bionic (arm)
+          arch: arm64
+          language: bash
+          before_script: true
+          script:
+              - sudo $CI_MANAGERS/ubuntu.sh || travis_terminate
+
+        - name: Ubuntu Bionic (s390x)
+          arch: s390x
+          language: bash
+          before_script: true
+          script:
+              - sudo $CI_MANAGERS/ubuntu.sh || travis_terminate
+
+        - name: Ubuntu Bionic (ppc64le)
+          arch: ppc64le
+          language: bash
+          before_script: true
+          script:
+              - sudo $CI_MANAGERS/ubuntu.sh || travis_terminate
+
+        - stage: Coverity
+          language: bash
+          env:
+              # Coverity configuration
+              # COVERITY_SCAN_TOKEN=xxx
+              # Encrypted using `travis encrypt --repo libbpf/libbpf COVERITY_SCAN_TOKEN=xxx`
+              - secure: "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"
+              - COVERITY_SCAN_PROJECT_NAME="libbpf"
+              - COVERITY_SCAN_NOTIFICATION_EMAIL="${AUTHOR_EMAIL}"
+              - COVERITY_SCAN_BRANCH_PATTERN="$TRAVIS_BRANCH"
+              # Note: `make -C src/` as a BUILD_COMMAND will not work here
+              - COVERITY_SCAN_BUILD_COMMAND_PREPEND="cd src/"
+              - COVERITY_SCAN_BUILD_COMMAND="make"
+          install:
+              - sudo echo 'deb-src http://archive.ubuntu.com/ubuntu/ bionic main restricted universe multiverse' >>/etc/apt/sources.list
+              - sudo apt-get update
+              - sudo apt-get -y build-dep libelf-dev
+              - sudo apt-get install -y libelf-dev pkg-config
+          # Override before_script: so VMTEST before_script commands are not executed.
+          before_script: true
+          script:
+              - scripts/coverity.sh || travis_terminate

BPF-CHECKPOINT-COMMIT

@@ -0,0 +1 @@
+08dc225d8868d5094ada62f471ebdfcce9dbc298

CHECKPOINT-COMMIT

@@ -1 +1 @@
-66b5f1c439843bcbab01cc7f3854ae2742f3d1e3
+35b9211c0a2427e8f39e534f442f43804fc8d5ca

LICENSE

@@ -0,0 +1 @@
+LGPL-2.1 OR BSD-2-Clause

LICENSE.BSD-2-Clause

@@ -0,0 +1,32 @@
+Valid-License-Identifier: BSD-2-Clause
+SPDX-URL: https://spdx.org/licenses/BSD-2-Clause.html
+Usage-Guide:
+  To use the BSD 2-clause "Simplified" License put the following SPDX
+  tag/value pair into a comment according to the placement guidelines in
+  the licensing rules documentation:
+    SPDX-License-Identifier: BSD-2-Clause
+License-Text:
+
+Copyright (c) <year> <owner> . All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+1. Redistributions of source code must retain the above copyright notice,
+   this list of conditions and the following disclaimer.
+
+2. Redistributions in binary form must reproduce the above copyright
+   notice, this list of conditions and the following disclaimer in the
+   documentation and/or other materials provided with the distribution.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.

LICENSE.LPGL-2.1

@@ -0,0 +1,503 @@
+Valid-License-Identifier: LGPL-2.1
+Valid-License-Identifier: LGPL-2.1+
+SPDX-URL: https://spdx.org/licenses/LGPL-2.1.html
+Usage-Guide:
+  To use this license in source code, put one of the following SPDX
+  tag/value pairs into a comment according to the placement
+  guidelines in the licensing rules documentation.
+  For 'GNU Lesser General Public License (LGPL) version 2.1 only' use:
+    SPDX-License-Identifier: LGPL-2.1
+  For 'GNU Lesser General Public License (LGPL) version 2.1 or any later
+  version' use:
+    SPDX-License-Identifier: LGPL-2.1+
+License-Text:
+
+GNU LESSER GENERAL PUBLIC LICENSE
+Version 2.1, February 1999
+
+Copyright (C) 1991, 1999 Free Software Foundation, Inc.
+51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+
+Everyone is permitted to copy and distribute verbatim copies of this
+license document, but changing it is not allowed.
+
+[This is the first released version of the Lesser GPL. It also counts as
+the successor of the GNU Library Public License, version 2, hence the
+version number 2.1.]
+
+Preamble
+
+The licenses for most software are designed to take away your freedom to
+share and change it. By contrast, the GNU General Public Licenses are
+intended to guarantee your freedom to share and change free software--to
+make sure the software is free for all its users.
+
+This license, the Lesser General Public License, applies to some specially
+designated software packages--typically libraries--of the Free Software
+Foundation and other authors who decide to use it. You can use it too, but
+we suggest you first think carefully about whether this license or the
+ordinary General Public License is the better strategy to use in any
+particular case, based on the explanations below.
+
+When we speak of free software, we are referring to freedom of use, not
+price. Our General Public Licenses are designed to make sure that you have
+the freedom to distribute copies of free software (and charge for this
+service if you wish); that you receive source code or can get it if you
+want it; that you can change the software and use pieces of it in new free
+programs; and that you are informed that you can do these things.
+
+To protect your rights, we need to make restrictions that forbid
+distributors to deny you these rights or to ask you to surrender these
+rights. These restrictions translate to certain responsibilities for you if
+you distribute copies of the library or if you modify it.
+
+For example, if you distribute copies of the library, whether gratis or for
+a fee, you must give the recipients all the rights that we gave you. You
+must make sure that they, too, receive or can get the source code. If you
+link other code with the library, you must provide complete object files to
+the recipients, so that they can relink them with the library after making
+changes to the library and recompiling it. And you must show them these
+terms so they know their rights.
+
+We protect your rights with a two-step method: (1) we copyright the
+library, and (2) we offer you this license, which gives you legal
+permission to copy, distribute and/or modify the library.
+
+To protect each distributor, we want to make it very clear that there is no
+warranty for the free library. Also, if the library is modified by someone
+else and passed on, the recipients should know that what they have is not
+the original version, so that the original author's reputation will not be
+affected by problems that might be introduced by others.
+
+Finally, software patents pose a constant threat to the existence of any
+free program. We wish to make sure that a company cannot effectively
+restrict the users of a free program by obtaining a restrictive license
+from a patent holder. Therefore, we insist that any patent license obtained
+for a version of the library must be consistent with the full freedom of
+use specified in this license.
+
+Most GNU software, including some libraries, is covered by the ordinary GNU
+General Public License. This license, the GNU Lesser General Public
+License, applies to certain designated libraries, and is quite different
+from the ordinary General Public License. We use this license for certain
+libraries in order to permit linking those libraries into non-free
+programs.
+
+When a program is linked with a library, whether statically or using a
+shared library, the combination of the two is legally speaking a combined
+work, a derivative of the original library. The ordinary General Public
+License therefore permits such linking only if the entire combination fits
+its criteria of freedom. The Lesser General Public License permits more lax
+criteria for linking other code with the library.
+
+We call this license the "Lesser" General Public License because it does
+Less to protect the user's freedom than the ordinary General Public
+License. It also provides other free software developers Less of an
+advantage over competing non-free programs. These disadvantages are the
+reason we use the ordinary General Public License for many
+libraries. However, the Lesser license provides advantages in certain
+special circumstances.
+
+For example, on rare occasions, there may be a special need to encourage
+the widest possible use of a certain library, so that it becomes a de-facto
+standard. To achieve this, non-free programs must be allowed to use the
+library. A more frequent case is that a free library does the same job as
+widely used non-free libraries. In this case, there is little to gain by
+limiting the free library to free software only, so we use the Lesser
+General Public License.
+
+In other cases, permission to use a particular library in non-free programs
+enables a greater number of people to use a large body of free
+software. For example, permission to use the GNU C Library in non-free
+programs enables many more people to use the whole GNU operating system, as
+well as its variant, the GNU/Linux operating system.
+
+Although the Lesser General Public License is Less protective of the users'
+freedom, it does ensure that the user of a program that is linked with the
+Library has the freedom and the wherewithal to run that program using a
+modified version of the Library.
+
+The precise terms and conditions for copying, distribution and modification
+follow. Pay close attention to the difference between a "work based on the
+library" and a "work that uses the library". The former contains code
+derived from the library, whereas the latter must be combined with the
+library in order to run.
+
+TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
+
+0. This License Agreement applies to any software library or other program
+   which contains a notice placed by the copyright holder or other
+   authorized party saying it may be distributed under the terms of this
+   Lesser General Public License (also called "this License"). Each
+   licensee is addressed as "you".
+
+   A "library" means a collection of software functions and/or data
+   prepared so as to be conveniently linked with application programs
+   (which use some of those functions and data) to form executables.
+
+   The "Library", below, refers to any such software library or work which
+   has been distributed under these terms. A "work based on the Library"
+   means either the Library or any derivative work under copyright law:
+   that is to say, a work containing the Library or a portion of it, either
+   verbatim or with modifications and/or translated straightforwardly into
+   another language. (Hereinafter, translation is included without
+   limitation in the term "modification".)
+
+   "Source code" for a work means the preferred form of the work for making
+   modifications to it. For a library, complete source code means all the
+   source code for all modules it contains, plus any associated interface
+   definition files, plus the scripts used to control compilation and
+   installation of the library.
+
+    Activities other than copying, distribution and modification are not
+    covered by this License; they are outside its scope. The act of running
+    a program using the Library is not restricted, and output from such a
+    program is covered only if its contents constitute a work based on the
+    Library (independent of the use of the Library in a tool for writing
+    it). Whether that is true depends on what the Library does and what the
+    program that uses the Library does.
+
+1. You may copy and distribute verbatim copies of the Library's complete
+   source code as you receive it, in any medium, provided that you
+   conspicuously and appropriately publish on each copy an appropriate
+   copyright notice and disclaimer of warranty; keep intact all the notices
+   that refer to this License and to the absence of any warranty; and
+   distribute a copy of this License along with the Library.
+
+   You may charge a fee for the physical act of transferring a copy, and
+   you may at your option offer warranty protection in exchange for a fee.
+
+2. You may modify your copy or copies of the Library or any portion of it,
+   thus forming a work based on the Library, and copy and distribute such
+   modifications or work under the terms of Section 1 above, provided that
+   you also meet all of these conditions:
+
+   a) The modified work must itself be a software library.
+
+   b) You must cause the files modified to carry prominent notices stating
+      that you changed the files and the date of any change.
+
+   c) You must cause the whole of the work to be licensed at no charge to
+      all third parties under the terms of this License.
+
+   d) If a facility in the modified Library refers to a function or a table
+      of data to be supplied by an application program that uses the
+      facility, other than as an argument passed when the facility is
+      invoked, then you must make a good faith effort to ensure that, in
+      the event an application does not supply such function or table, the
+      facility still operates, and performs whatever part of its purpose
+      remains meaningful.
+
+   (For example, a function in a library to compute square roots has a
+    purpose that is entirely well-defined independent of the
+    application. Therefore, Subsection 2d requires that any
+    application-supplied function or table used by this function must be
+    optional: if the application does not supply it, the square root
+    function must still compute square roots.)
+
+   These requirements apply to the modified work as a whole. If
+   identifiable sections of that work are not derived from the Library, and
+   can be reasonably considered independent and separate works in
+   themselves, then this License, and its terms, do not apply to those
+   sections when you distribute them as separate works. But when you
+   distribute the same sections as part of a whole which is a work based on
+   the Library, the distribution of the whole must be on the terms of this
+   License, whose permissions for other licensees extend to the entire
+   whole, and thus to each and every part regardless of who wrote it.
+
+   Thus, it is not the intent of this section to claim rights or contest
+   your rights to work written entirely by you; rather, the intent is to
+   exercise the right to control the distribution of derivative or
+   collective works based on the Library.
+
+   In addition, mere aggregation of another work not based on the Library
+   with the Library (or with a work based on the Library) on a volume of a
+   storage or distribution medium does not bring the other work under the
+   scope of this License.
+
+3. You may opt to apply the terms of the ordinary GNU General Public
+   License instead of this License to a given copy of the Library. To do
+   this, you must alter all the notices that refer to this License, so that
+   they refer to the ordinary GNU General Public License, version 2,
+   instead of to this License. (If a newer version than version 2 of the
+   ordinary GNU General Public License has appeared, then you can specify
+   that version instead if you wish.) Do not make any other change in these
+   notices.
+
+   Once this change is made in a given copy, it is irreversible for that
+   copy, so the ordinary GNU General Public License applies to all
+   subsequent copies and derivative works made from that copy.
+
+   This option is useful when you wish to copy part of the code of the
+   Library into a program that is not a library.
+
+4. You may copy and distribute the Library (or a portion or derivative of
+   it, under Section 2) in object code or executable form under the terms
+   of Sections 1 and 2 above provided that you accompany it with the
+   complete corresponding machine-readable source code, which must be
+   distributed under the terms of Sections 1 and 2 above on a medium
+   customarily used for software interchange.
+
+   If distribution of object code is made by offering access to copy from a
+   designated place, then offering equivalent access to copy the source
+   code from the same place satisfies the requirement to distribute the
+   source code, even though third parties are not compelled to copy the
+   source along with the object code.
+
+5. A program that contains no derivative of any portion of the Library, but
+   is designed to work with the Library by being compiled or linked with
+   it, is called a "work that uses the Library". Such a work, in isolation,
+   is not a derivative work of the Library, and therefore falls outside the
+   scope of this License.
+
+   However, linking a "work that uses the Library" with the Library creates
+   an executable that is a derivative of the Library (because it contains
+   portions of the Library), rather than a "work that uses the
+   library". The executable is therefore covered by this License. Section 6
+   states terms for distribution of such executables.
+
+   When a "work that uses the Library" uses material from a header file
+   that is part of the Library, the object code for the work may be a
+   derivative work of the Library even though the source code is
+   not. Whether this is true is especially significant if the work can be
+   linked without the Library, or if the work is itself a library. The
+   threshold for this to be true is not precisely defined by law.
+
+   If such an object file uses only numerical parameters, data structure
+   layouts and accessors, and small macros and small inline functions (ten
+   lines or less in length), then the use of the object file is
+   unrestricted, regardless of whether it is legally a derivative
+   work. (Executables containing this object code plus portions of the
+   Library will still fall under Section 6.)
+
+   Otherwise, if the work is a derivative of the Library, you may
+   distribute the object code for the work under the terms of Section
+   6. Any executables containing that work also fall under Section 6,
+   whether or not they are linked directly with the Library itself.
+
+6. As an exception to the Sections above, you may also combine or link a
+   "work that uses the Library" with the Library to produce a work
+   containing portions of the Library, and distribute that work under terms
+   of your choice, provided that the terms permit modification of the work
+   for the customer's own use and reverse engineering for debugging such
+   modifications.
+
+   You must give prominent notice with each copy of the work that the
+   Library is used in it and that the Library and its use are covered by
+   this License. You must supply a copy of this License. If the work during
+   execution displays copyright notices, you must include the copyright
+   notice for the Library among them, as well as a reference directing the
+   user to the copy of this License. Also, you must do one of these things:
+
+   a) Accompany the work with the complete corresponding machine-readable
+      source code for the Library including whatever changes were used in
+      the work (which must be distributed under Sections 1 and 2 above);
+      and, if the work is an executable linked with the Library, with the
+      complete machine-readable "work that uses the Library", as object
+      code and/or source code, so that the user can modify the Library and
+      then relink to produce a modified executable containing the modified
+      Library. (It is understood that the user who changes the contents of
+      definitions files in the Library will not necessarily be able to
+      recompile the application to use the modified definitions.)
+
+   b) Use a suitable shared library mechanism for linking with the
+      Library. A suitable mechanism is one that (1) uses at run time a copy
+      of the library already present on the user's computer system, rather
+      than copying library functions into the executable, and (2) will
+      operate properly with a modified version of the library, if the user
+      installs one, as long as the modified version is interface-compatible
+      with the version that the work was made with.
+
+   c) Accompany the work with a written offer, valid for at least three
+      years, to give the same user the materials specified in Subsection
+      6a, above, for a charge no more than the cost of performing this
+      distribution.
+
+   d) If distribution of the work is made by offering access to copy from a
+      designated place, offer equivalent access to copy the above specified
+      materials from the same place.
+
+   e) Verify that the user has already received a copy of these materials
+      or that you have already sent this user a copy.
+
+   For an executable, the required form of the "work that uses the Library"
+   must include any data and utility programs needed for reproducing the
+   executable from it. However, as a special exception, the materials to be
+   distributed need not include anything that is normally distributed (in
+   either source or binary form) with the major components (compiler,
+   kernel, and so on) of the operating system on which the executable runs,
+   unless that component itself accompanies the executable.
+
+   It may happen that this requirement contradicts the license restrictions
+   of other proprietary libraries that do not normally accompany the
+   operating system. Such a contradiction means you cannot use both them
+   and the Library together in an executable that you distribute.
+
+7. You may place library facilities that are a work based on the Library
+   side-by-side in a single library together with other library facilities
+   not covered by this License, and distribute such a combined library,
+   provided that the separate distribution of the work based on the Library
+   and of the other library facilities is otherwise permitted, and provided
+   that you do these two things:
+
+   a) Accompany the combined library with a copy of the same work based on
+      the Library, uncombined with any other library facilities. This must
+      be distributed under the terms of the Sections above.
+
+   b) Give prominent notice with the combined library of the fact that part
+      of it is a work based on the Library, and explaining where to find
+      the accompanying uncombined form of the same work.
+
+8. You may not copy, modify, sublicense, link with, or distribute the
+   Library except as expressly provided under this License. Any attempt
+   otherwise to copy, modify, sublicense, link with, or distribute the
+   Library is void, and will automatically terminate your rights under this
+   License. However, parties who have received copies, or rights, from you
+   under this License will not have their licenses terminated so long as
+   such parties remain in full compliance.
+
+9. You are not required to accept this License, since you have not signed
+   it. However, nothing else grants you permission to modify or distribute
+   the Library or its derivative works. These actions are prohibited by law
+   if you do not accept this License. Therefore, by modifying or
+   distributing the Library (or any work based on the Library), you
+   indicate your acceptance of this License to do so, and all its terms and
+   conditions for copying, distributing or modifying the Library or works
+   based on it.
+
+10. Each time you redistribute the Library (or any work based on the
+    Library), the recipient automatically receives a license from the
+    original licensor to copy, distribute, link with or modify the Library
+    subject to these terms and conditions. You may not impose any further
+    restrictions on the recipients' exercise of the rights granted
+    herein. You are not responsible for enforcing compliance by third
+    parties with this License.
+
+11. If, as a consequence of a court judgment or allegation of patent
+    infringement or for any other reason (not limited to patent issues),
+    conditions are imposed on you (whether by court order, agreement or
+    otherwise) that contradict the conditions of this License, they do not
+    excuse you from the conditions of this License. If you cannot
+    distribute so as to satisfy simultaneously your obligations under this
+    License and any other pertinent obligations, then as a consequence you
+    may not distribute the Library at all. For example, if a patent license
+    would not permit royalty-free redistribution of the Library by all
+    those who receive copies directly or indirectly through you, then the
+    only way you could satisfy both it and this License would be to refrain
+    entirely from distribution of the Library.
+
+    If any portion of this section is held invalid or unenforceable under
+    any particular circumstance, the balance of the section is intended to
+    apply, and the section as a whole is intended to apply in other
+    circumstances.
+
+    It is not the purpose of this section to induce you to infringe any
+    patents or other property right claims or to contest validity of any
+    such claims; this section has the sole purpose of protecting the
+    integrity of the free software distribution system which is implemented
+    by public license practices. Many people have made generous
+    contributions to the wide range of software distributed through that
+    system in reliance on consistent application of that system; it is up
+    to the author/donor to decide if he or she is willing to distribute
+    software through any other system and a licensee cannot impose that
+    choice.
+
+    This section is intended to make thoroughly clear what is believed to
+    be a consequence of the rest of this License.
+
+12. If the distribution and/or use of the Library is restricted in certain
+    countries either by patents or by copyrighted interfaces, the original
+    copyright holder who places the Library under this License may add an
+    explicit geographical distribution limitation excluding those
+    countries, so that distribution is permitted only in or among countries
+    not thus excluded. In such case, this License incorporates the
+    limitation as if written in the body of this License.
+
+13. The Free Software Foundation may publish revised and/or new versions of
+    the Lesser General Public License from time to time. Such new versions
+    will be similar in spirit to the present version, but may differ in
+    detail to address new problems or concerns.
+
+    Each version is given a distinguishing version number. If the Library
+    specifies a version number of this License which applies to it and "any
+    later version", you have the option of following the terms and
+    conditions either of that version or of any later version published by
+    the Free Software Foundation. If the Library does not specify a license
+    version number, you may choose any version ever published by the Free
+    Software Foundation.
+
+14. If you wish to incorporate parts of the Library into other free
+    programs whose distribution conditions are incompatible with these,
+    write to the author to ask for permission. For software which is
+    copyrighted by the Free Software Foundation, write to the Free Software
+    Foundation; we sometimes make exceptions for this. Our decision will be
+    guided by the two goals of preserving the free status of all
+    derivatives of our free software and of promoting the sharing and reuse
+    of software generally.
+
+NO WARRANTY
+
+15. BECAUSE THE LIBRARY IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
+    FOR THE LIBRARY, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN
+    OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
+    PROVIDE THE LIBRARY "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER
+    EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+    WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE
+    ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE LIBRARY IS WITH
+    YOU. SHOULD THE LIBRARY PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL
+    NECESSARY SERVICING, REPAIR OR CORRECTION.
+
+16. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
+    WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR
+    REDISTRIBUTE THE LIBRARY AS PERMITTED ABOVE, BE LIABLE TO YOU FOR
+    DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL
+    DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE LIBRARY
+    (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED
+    INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF
+    THE LIBRARY TO OPERATE WITH ANY OTHER SOFTWARE), EVEN IF SUCH HOLDER OR
+    OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
+
+END OF TERMS AND CONDITIONS
+
+How to Apply These Terms to Your New Libraries
+
+If you develop a new library, and you want it to be of the greatest
+possible use to the public, we recommend making it free software that
+everyone can redistribute and change. You can do so by permitting
+redistribution under these terms (or, alternatively, under the terms of the
+ordinary General Public License).
+
+To apply these terms, attach the following notices to the library. It is
+safest to attach them to the start of each source file to most effectively
+convey the exclusion of warranty; and each file should have at least the
+"copyright" line and a pointer to where the full notice is found.
+
+one line to give the library's name and an idea of what it does.
+Copyright (C) year name of author
+
+This library is free software; you can redistribute it and/or modify it
+under the terms of the GNU Lesser General Public License as published by
+the Free Software Foundation; either version 2.1 of the License, or (at
+your option) any later version.
+
+This library is distributed in the hope that it will be useful, but WITHOUT
+ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License
+for more details.
+
+You should have received a copy of the GNU Lesser General Public License
+along with this library; if not, write to the Free Software Foundation,
+Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA Also add
+information on how to contact you by electronic and paper mail.
+
+You should also get your employer (if you work as a programmer) or your
+school, if any, to sign a "copyright disclaimer" for the library, if
+necessary. Here is a sample; alter the names:
+
+Yoyodyne, Inc., hereby disclaims all copyright interest in
+the library `Frob' (a library for tweaking knobs) written
+by James Random Hacker.
+
+signature of Ty Coon, 1 April 1990
+Ty Coon, President of Vice
+That's all there is to it!

README.md

@@ -16,7 +16,10 @@ Other header files at this repo (`include/linux/*.h`) are reduced versions of
 their counterpart files at bpf-next's `tools/include/linux/*.h` to make compilation
 successful.
 
-Build [![Build Status](https://travis-ci.org/libbpf/libbpf.svg?branch=master)](https://travis-ci.org/libbpf/libbpf)
+Build
+[![Build Status](https://travis-ci.org/libbpf/libbpf.svg?branch=master)](https://travis-ci.org/libbpf/libbpf)
+[![Total alerts](https://img.shields.io/lgtm/alerts/g/libbpf/libbpf.svg?logo=lgtm&logoWidth=18)](https://lgtm.com/projects/g/libbpf/libbpf/alerts/)
+[![Coverity](https://img.shields.io/coverity/scan/18195.svg)](https://scan.coverity.com/projects/libbpf)
 =====
 libelf is an internal dependency of libbpf and thus it is required to link
 against and must be installed on the system for applications to work.
@@ -48,23 +51,32 @@ $ cd src
 $ PKG_CONFIG_PATH=/build/root/lib64/pkgconfig DESTDIR=/build/root make install
 ```
 
-To integrate libbpf into a project which uses Meson building system define
+Distributions
-`[wrap-git]` file in `subprojects` folder.
+=====
-To add libbpf dependency to the parent parent project, e.g. for
-libbpf_static_dep:
-```
-libbpf_obj = subproject('libbpf', required : true)
-libbpf_static_dep = libbpf_proj.get_variable('libbpf_static_dep')
-```
 
-To validate changes to meson.build
+Distributions packaging libbpf from this mirror:
-```bash
+  - [Fedora](https://src.fedoraproject.org/rpms/libbpf)
-$ python3 meson.py build
+  - [Gentoo](https://packages.gentoo.org/packages/dev-libs/libbpf)
-$ ninja -C build/
-```
 
-To install headers, libs and pkgconfig
+Benefits of packaging from the mirror over packaging from kernel sources:
-```bash
+  - Consistent versioning across distributions.
-$ cd build
+  - No ties to any specific kernel, transparent handling of older kernels.
-$ ninja install
+    Libbpf is designed to be kernel-agnostic and work across multitude of kernel
-```
+    versions. It has built-in mechanisms to gracefully handle older kernels,
+    that are missing some of the features, by working around or gracefully
+    degrading functionality. Thus libbpf is not tied to a specific kernel
+    version and can/should be packaged and versioned independently.
+  - Continuous integration testing via [TravisCI](https://travis-ci.org/libbpf/libbpf).
+  - Static code analysis via [LGTM](https://lgtm.com/projects/g/libbpf/libbpf) and [Coverity](https://scan.coverity.com/projects/libbpf).
+
+Package dependencies of libbpf, package names may vary across distros:
+  - zlib
+  - libelf
+
+License
+=====
+
+This work is dual-licensed under BSD 2-clause license and GNU LGPL v2.1 license.
+You can choose between one of them if you use this work.
+
+`SPDX-License-Identifier: BSD-2-Clause OR LGPL-2.1`

include/linux/filter.h

@@ -96,7 +96,7 @@
 			      MAP_FD, 0)
 
 #define BPF_LD_MAP_VALUE(DST, MAP_FD, VALUE_OFF)		\
-	BPF_LD_IMM64_RAW_FULL(DST, BPF_PSEUDO_MAP_FD, 0, 0,	\
+	BPF_LD_IMM64_RAW_FULL(DST, BPF_PSEUDO_MAP_VALUE, 0, 0,	\
 			      MAP_FD, VALUE_OFF)
 
 #define BPF_JMP_IMM(OP, DST, IMM, OFF)				\
@@ -107,5 +107,12 @@
 		.off  = OFF,					\
 		.imm  = IMM })
 
+#define BPF_JMP32_IMM(OP, DST, IMM, OFF)			\
+	((struct bpf_insn) {					\
+		.code = BPF_JMP32 | BPF_OP(OP) | BPF_K,		\
+		.dst_reg = DST,					\
+		.src_reg = 0,					\
+		.off  = OFF,					\
+		.imm  = IMM })
 
 #endif

include/uapi/linux/bpf.h

@@ -106,6 +106,11 @@ enum bpf_cmd {
 	BPF_TASK_FD_QUERY,
 	BPF_MAP_LOOKUP_AND_DELETE_ELEM,
 	BPF_MAP_FREEZE,
+	BPF_BTF_GET_NEXT_ID,
+	BPF_MAP_LOOKUP_BATCH,
+	BPF_MAP_LOOKUP_AND_DELETE_BATCH,
+	BPF_MAP_UPDATE_BATCH,
+	BPF_MAP_DELETE_BATCH,
 };
 
 enum bpf_map_type {
@@ -134,6 +139,8 @@ enum bpf_map_type {
 	BPF_MAP_TYPE_QUEUE,
 	BPF_MAP_TYPE_STACK,
 	BPF_MAP_TYPE_SK_STORAGE,
+	BPF_MAP_TYPE_DEVMAP_HASH,
+	BPF_MAP_TYPE_STRUCT_OPS,
 };
 
 /* Note that tracing related programs such as
@@ -171,6 +178,9 @@ enum bpf_prog_type {
 	BPF_PROG_TYPE_CGROUP_SYSCTL,
 	BPF_PROG_TYPE_RAW_TRACEPOINT_WRITABLE,
 	BPF_PROG_TYPE_CGROUP_SOCKOPT,
+	BPF_PROG_TYPE_TRACING,
+	BPF_PROG_TYPE_STRUCT_OPS,
+	BPF_PROG_TYPE_EXT,
 };
 
 enum bpf_attach_type {
@@ -197,6 +207,9 @@ enum bpf_attach_type {
 	BPF_CGROUP_UDP6_RECVMSG,
 	BPF_CGROUP_GETSOCKOPT,
 	BPF_CGROUP_SETSOCKOPT,
+	BPF_TRACE_RAW_TP,
+	BPF_TRACE_FENTRY,
+	BPF_TRACE_FEXIT,
 	__MAX_BPF_ATTACH_TYPE
 };
 
@@ -225,6 +238,11 @@ enum bpf_attach_type {
  * When children program makes decision (like picking TCP CA or sock bind)
  * parent program has a chance to override it.
  *
+ * With BPF_F_ALLOW_MULTI a new program is added to the end of the list of
+ * programs for a cgroup. Though it's possible to replace an old program at
+ * any position by also specifying BPF_F_REPLACE flag and position itself in
+ * replace_bpf_fd attribute. Old program at this position will be released.
+ *
  * A cgroup with MULTI or OVERRIDE flag allows any attach flags in sub-cgroups.
  * A cgroup with NONE doesn't allow any programs in sub-cgroups.
  * Ex1:
@@ -243,6 +261,7 @@ enum bpf_attach_type {
  */
 #define BPF_F_ALLOW_OVERRIDE	(1U << 0)
 #define BPF_F_ALLOW_MULTI	(1U << 1)
+#define BPF_F_REPLACE		(1U << 2)
 
 /* If BPF_F_STRICT_ALIGNMENT is used in BPF_PROG_LOAD command, the
  * verifier will perform strict alignment checking as if the kernel
@@ -283,6 +302,9 @@ enum bpf_attach_type {
  */
 #define BPF_F_TEST_RND_HI32	(1U << 2)
 
+/* The verifier internal test flag. Behavior is undefined */
+#define BPF_F_TEST_STATE_FREQ	(1U << 3)
+
 /* When BPF ldimm64's insn[0].src_reg != 0 then this can have
  * two extensions:
  *
@@ -336,7 +358,18 @@ enum bpf_attach_type {
 #define BPF_F_RDONLY_PROG	(1U << 7)
 #define BPF_F_WRONLY_PROG	(1U << 8)
 
-/* flags for BPF_PROG_QUERY */
+/* Clone map from listener for newly accepted socket */
+#define BPF_F_CLONE		(1U << 9)
+
+/* Enable memory-mapping BPF map */
+#define BPF_F_MMAPABLE		(1U << 10)
+
+/* Flags for BPF_PROG_QUERY. */
+
+/* Query effective (directly attached + inherited from ancestor cgroups)
+ * programs that will be executed for events within a cgroup.
+ * attach_flags with this flag are returned only for directly attached programs.
+ */
 #define BPF_F_QUERY_EFFECTIVE	(1U << 0)
 
 enum bpf_stack_build_id_status {
@@ -376,6 +409,10 @@ union bpf_attr {
 		__u32	btf_fd;		/* fd pointing to a BTF type data */
 		__u32	btf_key_type_id;	/* BTF type_id of the key */
 		__u32	btf_value_type_id;	/* BTF type_id of the value */
+		__u32	btf_vmlinux_value_type_id;/* BTF type_id of a kernel-
+						   * struct stored as the
+						   * map value
+						   */
 	};
 
 	struct { /* anonymous struct used by BPF_MAP_*_ELEM commands */
@@ -388,6 +425,23 @@ union bpf_attr {
 		__u64		flags;
 	};
 
+	struct { /* struct used by BPF_MAP_*_BATCH commands */
+		__aligned_u64	in_batch;	/* start batch,
+						 * NULL to start from beginning
+						 */
+		__aligned_u64	out_batch;	/* output: next start batch */
+		__aligned_u64	keys;
+		__aligned_u64	values;
+		__u32		count;		/* input/output:
+						 * input: # of key/value
+						 * elements
+						 * output: # of filled elements
+						 */
+		__u32		map_fd;
+		__u64		elem_flags;
+		__u64		flags;
+	} batch;
+
 	struct { /* anonymous struct used by BPF_PROG_LOAD command */
 		__u32		prog_type;	/* one of enum bpf_prog_type */
 		__u32		insn_cnt;
@@ -412,6 +466,8 @@ union bpf_attr {
 		__u32		line_info_rec_size;	/* userspace bpf_line_info size */
 		__aligned_u64	line_info;	/* line info */
 		__u32		line_info_cnt;	/* number of bpf_line_info records */
+		__u32		attach_btf_id;	/* in-kernel BTF type id to attach to */
+		__u32		attach_prog_fd; /* 0 to attach to vmlinux */
 	};
 
 	struct { /* anonymous struct used by BPF_OBJ_* commands */
@@ -425,6 +481,10 @@ union bpf_attr {
 		__u32		attach_bpf_fd;	/* eBPF program to attach */
 		__u32		attach_type;
 		__u32		attach_flags;
+		__u32		replace_bpf_fd;	/* previously attached eBPF
+						 * program to replace if
+						 * BPF_F_REPLACE is used
+						 */
 	};
 
 	struct { /* anonymous struct used by BPF_PROG_TEST_RUN command */
@@ -552,10 +612,13 @@ union bpf_attr {
  * 	Return
  * 		0 on success, or a negative error in case of failure.
  *
- * int bpf_probe_read(void *dst, u32 size, const void *src)
+ * int bpf_probe_read(void *dst, u32 size, const void *unsafe_ptr)
  * 	Description
  * 		For tracing programs, safely attempt to read *size* bytes from
- * 		address *src* and store the data in *dst*.
+ * 		kernel space address *unsafe_ptr* and store the data in *dst*.
+ *
+ * 		Generally, use bpf_probe_read_user() or bpf_probe_read_kernel()
+ * 		instead.
  * 	Return
  * 		0 on success, or a negative error in case of failure.
  *
@@ -575,6 +638,8 @@ union bpf_attr {
  * 		limited to five).
  *
  * 		Each time the helper is called, it appends a line to the trace.
+ * 		Lines are discarded while *\/sys/kernel/debug/tracing/trace* is
+ * 		open, use *\/sys/kernel/debug/tracing/trace_pipe* to avoid this.
  * 		The format of the trace is customizable, and the exact output
  * 		one will get depends on the options set in
  * 		*\/sys/kernel/debug/tracing/trace_options* (see also the
@@ -784,7 +849,7 @@ union bpf_attr {
  * 		A 64-bit integer containing the current GID and UID, and
  * 		created as such: *current_gid* **<< 32 \|** *current_uid*.
  *
- * int bpf_get_current_comm(char *buf, u32 size_of_buf)
+ * int bpf_get_current_comm(void *buf, u32 size_of_buf)
  * 	Description
  * 		Copy the **comm** attribute of the current task into *buf* of
  * 		*size_of_buf*. The **comm** attribute contains the name of
@@ -1013,7 +1078,7 @@ union bpf_attr {
  * 		The realm of the route for the packet associated to *skb*, or 0
  * 		if none was found.
  *
- * int bpf_perf_event_output(struct pt_reg *ctx, struct bpf_map *map, u64 flags, void *data, u64 size)
+ * int bpf_perf_event_output(void *ctx, struct bpf_map *map, u64 flags, void *data, u64 size)
  * 	Description
  * 		Write raw *data* blob into a special BPF perf event held by
  * 		*map* of type **BPF_MAP_TYPE_PERF_EVENT_ARRAY**. This perf
@@ -1058,7 +1123,7 @@ union bpf_attr {
  * 	Return
  * 		0 on success, or a negative error in case of failure.
  *
- * int bpf_skb_load_bytes(const struct sk_buff *skb, u32 offset, void *to, u32 len)
+ * int bpf_skb_load_bytes(const void *skb, u32 offset, void *to, u32 len)
  * 	Description
  * 		This helper was provided as an easy way to load data from a
  * 		packet. It can be used to load *len* bytes from *offset* from
@@ -1075,7 +1140,7 @@ union bpf_attr {
  * 	Return
  * 		0 on success, or a negative error in case of failure.
  *
- * int bpf_get_stackid(struct pt_reg *ctx, struct bpf_map *map, u64 flags)
+ * int bpf_get_stackid(void *ctx, struct bpf_map *map, u64 flags)
  * 	Description
  * 		Walk a user or a kernel stack and return its id. To achieve
  * 		this, the helper needs *ctx*, which is a pointer to the context
@@ -1144,7 +1209,7 @@ union bpf_attr {
  * 		The checksum result, or a negative error code in case of
  * 		failure.
  *
- * int bpf_skb_get_tunnel_opt(struct sk_buff *skb, u8 *opt, u32 size)
+ * int bpf_skb_get_tunnel_opt(struct sk_buff *skb, void *opt, u32 size)
  * 	Description
  * 		Retrieve tunnel options metadata for the packet associated to
  * 		*skb*, and store the raw tunnel option data to the buffer *opt*
@@ -1162,7 +1227,7 @@ union bpf_attr {
  * 	Return
  * 		The size of the option data retrieved.
  *
- * int bpf_skb_set_tunnel_opt(struct sk_buff *skb, u8 *opt, u32 size)
+ * int bpf_skb_set_tunnel_opt(struct sk_buff *skb, void *opt, u32 size)
  * 	Description
  * 		Set tunnel options metadata for the packet associated to *skb*
  * 		to the option data contained in the raw buffer *opt* of *size*.
@@ -1415,45 +1480,14 @@ union bpf_attr {
  * 	Return
  * 		0 on success, or a negative error in case of failure.
  *
- * int bpf_probe_read_str(void *dst, int size, const void *unsafe_ptr)
+ * int bpf_probe_read_str(void *dst, u32 size, const void *unsafe_ptr)
  * 	Description
- * 		Copy a NUL terminated string from an unsafe address
+ * 		Copy a NUL terminated string from an unsafe kernel address
- * 		*unsafe_ptr* to *dst*. The *size* should include the
+ * 		*unsafe_ptr* to *dst*. See bpf_probe_read_kernel_str() for
- * 		terminating NUL byte. In case the string length is smaller than
+ * 		more details.
- * 		*size*, the target is not padded with further NUL bytes. If the
- * 		string length is larger than *size*, just *size*-1 bytes are
- * 		copied and the last byte is set to NUL.
  *
- * 		On success, the length of the copied string is returned. This
+ * 		Generally, use bpf_probe_read_user_str() or bpf_probe_read_kernel_str()
- * 		makes this helper useful in tracing programs for reading
+ * 		instead.
- * 		strings, and more importantly to get its length at runtime. See
- * 		the following snippet:
- *
- * 		::
- *
- * 			SEC("kprobe/sys_open")
- * 			void bpf_sys_open(struct pt_regs *ctx)
- * 			{
- * 			        char buf[PATHLEN]; // PATHLEN is defined to 256
- * 			        int res = bpf_probe_read_str(buf, sizeof(buf),
- * 				                             ctx->di);
- *
- * 				// Consume buf, for example push it to
- * 				// userspace via bpf_perf_event_output(); we
- * 				// can use res (the string length) as event
- * 				// size, after checking its boundaries.
- * 			}
- *
- * 		In comparison, using **bpf_probe_read()** helper here instead
- * 		to read the string would require to estimate the length at
- * 		compile time, and would often result in copying more memory
- * 		than necessary.
- *
- * 		Another useful use case is when parsing individual process
- * 		arguments or individual environment variables navigating
- * 		*current*\ **->mm->arg_start** and *current*\
- * 		**->mm->env_start**: using this helper and the return value,
- * 		one can quickly iterate at the right offset of the memory area.
  * 	Return
  * 		On success, the strictly positive length of the string,
  * 		including the trailing NUL character. On error, a negative
@@ -1466,8 +1500,8 @@ union bpf_attr {
  * 		If no cookie has been set yet, generate a new cookie. Once
  * 		generated, the socket cookie remains stable for the life of the
  * 		socket. This helper can be useful for monitoring per socket
- * 		networking traffic statistics as it provides a unique socket
+ * 		networking traffic statistics as it provides a global socket
- * 		identifier per namespace.
+ * 		identifier that can be assumed unique.
  * 	Return
  * 		A 8-byte long non-decreasing number on success, or 0 if the
  * 		socket field is missing inside *skb*.
@@ -1501,7 +1535,7 @@ union bpf_attr {
  * 	Return
  * 		0
  *
- * int bpf_setsockopt(struct bpf_sock_ops *bpf_socket, int level, int optname, char *optval, int optlen)
+ * int bpf_setsockopt(struct bpf_sock_ops *bpf_socket, int level, int optname, void *optval, int optlen)
  * 	Description
  * 		Emulate a call to **setsockopt()** on the socket associated to
  * 		*bpf_socket*, which must be a full socket. The *level* at
@@ -1571,8 +1605,11 @@ union bpf_attr {
  * 		but this is only implemented for native XDP (with driver
  * 		support) as of this writing).
  *
- * 		All values for *flags* are reserved for future usage, and must
+ * 		The lower two bits of *flags* are used as the return code if
- * 		be left at zero.
+ * 		the map lookup fails. This is so that the return value can be
+ * 		one of the XDP program return codes up to XDP_TX, as chosen by
+ * 		the caller. Any higher bits in the *flags* argument must be
+ * 		unset.
  *
  * 		When used to redirect packets to net devices, this helper
  * 		provides a high performance increase over **bpf_redirect**\ ().
@@ -1582,7 +1619,7 @@ union bpf_attr {
  * 	Return
  * 		**XDP_REDIRECT** on success, or **XDP_ABORTED** on error.
  *
- * int bpf_sk_redirect_map(struct bpf_map *map, u32 key, u64 flags)
+ * int bpf_sk_redirect_map(struct sk_buff *skb, struct bpf_map *map, u32 key, u64 flags)
  * 	Description
  * 		Redirect the packet to the socket referenced by *map* (of type
  * 		**BPF_MAP_TYPE_SOCKMAP**) at index *key*. Both ingress and
@@ -1702,7 +1739,7 @@ union bpf_attr {
  * 	Return
  * 		0 on success, or a negative error in case of failure.
  *
- * int bpf_getsockopt(struct bpf_sock_ops *bpf_socket, int level, int optname, char *optval, int optlen)
+ * int bpf_getsockopt(struct bpf_sock_ops *bpf_socket, int level, int optname, void *optval, int optlen)
  * 	Description
  * 		Emulate a call to **getsockopt()** on the socket associated to
  * 		*bpf_socket*, which must be a full socket. The *level* at
@@ -1721,7 +1758,7 @@ union bpf_attr {
  * 	Return
  * 		0 on success, or a negative error in case of failure.
  *
- * int bpf_override_return(struct pt_reg *regs, u64 rc)
+ * int bpf_override_return(struct pt_regs *regs, u64 rc)
  * 	Description
  * 		Used for error injection, this helper uses kprobes to override
  * 		the return value of the probed function, and to set it to *rc*.
@@ -1934,7 +1971,7 @@ union bpf_attr {
  * 	Return
  * 		0 on success, or a negative error in case of failure.
  *
- * int bpf_get_stack(struct pt_regs *regs, void *buf, u32 size, u64 flags)
+ * int bpf_get_stack(void *ctx, void *buf, u32 size, u64 flags)
  * 	Description
  * 		Return a user or a kernel stack in bpf program provided buffer.
  * 		To achieve this, the helper needs *ctx*, which is a pointer
@@ -1967,7 +2004,7 @@ union bpf_attr {
  * 		A non-negative value equal to or less than *size* on success,
  * 		or a negative error in case of failure.
  *
- * int bpf_skb_load_bytes_relative(const struct sk_buff *skb, u32 offset, void *to, u32 len, u32 start_header)
+ * int bpf_skb_load_bytes_relative(const void *skb, u32 offset, void *to, u32 len, u32 start_header)
  * 	Description
  * 		This helper is similar to **bpf_skb_load_bytes**\ () in that
  * 		it provides an easy way to load *len* bytes from *offset*
@@ -2020,7 +2057,7 @@ union bpf_attr {
  *		* > 0 one of **BPF_FIB_LKUP_RET_** codes explaining why the
  *		  packet is not forwarded or needs assist from full stack
  *
- * int bpf_sock_hash_update(struct bpf_sock_ops_kern *skops, struct bpf_map *map, void *key, u64 flags)
+ * int bpf_sock_hash_update(struct bpf_sock_ops *skops, struct bpf_map *map, void *key, u64 flags)
  *	Description
  *		Add an entry to, or update a sockhash *map* referencing sockets.
  *		The *skops* is used as a new value for the entry associated to
@@ -2379,7 +2416,7 @@ union bpf_attr {
  * 	Return
  * 		0 on success, or a negative error in case of failure.
  *
- * int bpf_msg_push_data(struct sk_buff *skb, u32 start, u32 len, u64 flags)
+ * int bpf_msg_push_data(struct sk_msg_buff *msg, u32 start, u32 len, u64 flags)
  *	Description
  *		For socket policies, insert *len* bytes into *msg* at offset
  *		*start*.
@@ -2395,9 +2432,9 @@ union bpf_attr {
  *	Return
  *		0 on success, or a negative error in case of failure.
  *
- * int bpf_msg_pop_data(struct sk_msg_buff *msg, u32 start, u32 pop, u64 flags)
+ * int bpf_msg_pop_data(struct sk_msg_buff *msg, u32 start, u32 len, u64 flags)
  *	Description
- *		Will remove *pop* bytes from a *msg* starting at byte *start*.
+ *		Will remove *len* bytes from a *msg* starting at byte *start*.
  *		This may result in **ENOMEM** errors under certain situations if
  *		an allocation and copy are required due to a full ring buffer.
  *		However, the helper will try to avoid doing the allocation
@@ -2492,7 +2529,7 @@ union bpf_attr {
  *		A **struct bpf_tcp_sock** pointer on success, or **NULL** in
  *		case of failure.
  *
- * int bpf_skb_ecn_set_ce(struct sk_buf *skb)
+ * int bpf_skb_ecn_set_ce(struct sk_buff *skb)
  *	Description
  *		Set ECN (Explicit Congestion Notification) field of IP header
  *		to **CE** (Congestion Encountered) if current value is **ECT**
@@ -2699,7 +2736,8 @@ union bpf_attr {
  *
  * int bpf_send_signal(u32 sig)
  *	Description
- *		Send signal *sig* to the current task.
+ *		Send signal *sig* to the process of the current task.
+ *		The signal may be delivered to any of this process's threads.
  *	Return
  *		0 on success or successfully queued.
  *
@@ -2710,6 +2748,150 @@ union bpf_attr {
  *		**-EPERM** if no permission to send the *sig*.
  *
  *		**-EAGAIN** if bpf program can try again.
+ *
+ * s64 bpf_tcp_gen_syncookie(struct bpf_sock *sk, void *iph, u32 iph_len, struct tcphdr *th, u32 th_len)
+ *	Description
+ *		Try to issue a SYN cookie for the packet with corresponding
+ *		IP/TCP headers, *iph* and *th*, on the listening socket in *sk*.
+ *
+ *		*iph* points to the start of the IPv4 or IPv6 header, while
+ *		*iph_len* contains **sizeof**\ (**struct iphdr**) or
+ *		**sizeof**\ (**struct ip6hdr**).
+ *
+ *		*th* points to the start of the TCP header, while *th_len*
+ *		contains the length of the TCP header.
+ *
+ *	Return
+ *		On success, lower 32 bits hold the generated SYN cookie in
+ *		followed by 16 bits which hold the MSS value for that cookie,
+ *		and the top 16 bits are unused.
+ *
+ *		On failure, the returned value is one of the following:
+ *
+ *		**-EINVAL** SYN cookie cannot be issued due to error
+ *
+ *		**-ENOENT** SYN cookie should not be issued (no SYN flood)
+ *
+ *		**-EOPNOTSUPP** kernel configuration does not enable SYN cookies
+ *
+ *		**-EPROTONOSUPPORT** IP packet version is not 4 or 6
+ *
+ * int bpf_skb_output(void *ctx, struct bpf_map *map, u64 flags, void *data, u64 size)
+ * 	Description
+ * 		Write raw *data* blob into a special BPF perf event held by
+ * 		*map* of type **BPF_MAP_TYPE_PERF_EVENT_ARRAY**. This perf
+ * 		event must have the following attributes: **PERF_SAMPLE_RAW**
+ * 		as **sample_type**, **PERF_TYPE_SOFTWARE** as **type**, and
+ * 		**PERF_COUNT_SW_BPF_OUTPUT** as **config**.
+ *
+ * 		The *flags* are used to indicate the index in *map* for which
+ * 		the value must be put, masked with **BPF_F_INDEX_MASK**.
+ * 		Alternatively, *flags* can be set to **BPF_F_CURRENT_CPU**
+ * 		to indicate that the index of the current CPU core should be
+ * 		used.
+ *
+ * 		The value to write, of *size*, is passed through eBPF stack and
+ * 		pointed by *data*.
+ *
+ * 		*ctx* is a pointer to in-kernel struct sk_buff.
+ *
+ * 		This helper is similar to **bpf_perf_event_output**\ () but
+ * 		restricted to raw_tracepoint bpf programs.
+ * 	Return
+ * 		0 on success, or a negative error in case of failure.
+ *
+ * int bpf_probe_read_user(void *dst, u32 size, const void *unsafe_ptr)
+ * 	Description
+ * 		Safely attempt to read *size* bytes from user space address
+ * 		*unsafe_ptr* and store the data in *dst*.
+ * 	Return
+ * 		0 on success, or a negative error in case of failure.
+ *
+ * int bpf_probe_read_kernel(void *dst, u32 size, const void *unsafe_ptr)
+ * 	Description
+ * 		Safely attempt to read *size* bytes from kernel space address
+ * 		*unsafe_ptr* and store the data in *dst*.
+ * 	Return
+ * 		0 on success, or a negative error in case of failure.
+ *
+ * int bpf_probe_read_user_str(void *dst, u32 size, const void *unsafe_ptr)
+ * 	Description
+ * 		Copy a NUL terminated string from an unsafe user address
+ * 		*unsafe_ptr* to *dst*. The *size* should include the
+ * 		terminating NUL byte. In case the string length is smaller than
+ * 		*size*, the target is not padded with further NUL bytes. If the
+ * 		string length is larger than *size*, just *size*-1 bytes are
+ * 		copied and the last byte is set to NUL.
+ *
+ * 		On success, the length of the copied string is returned. This
+ * 		makes this helper useful in tracing programs for reading
+ * 		strings, and more importantly to get its length at runtime. See
+ * 		the following snippet:
+ *
+ * 		::
+ *
+ * 			SEC("kprobe/sys_open")
+ * 			void bpf_sys_open(struct pt_regs *ctx)
+ * 			{
+ * 			        char buf[PATHLEN]; // PATHLEN is defined to 256
+ * 			        int res = bpf_probe_read_user_str(buf, sizeof(buf),
+ * 				                                  ctx->di);
+ *
+ * 				// Consume buf, for example push it to
+ * 				// userspace via bpf_perf_event_output(); we
+ * 				// can use res (the string length) as event
+ * 				// size, after checking its boundaries.
+ * 			}
+ *
+ * 		In comparison, using **bpf_probe_read_user()** helper here
+ * 		instead to read the string would require to estimate the length
+ * 		at compile time, and would often result in copying more memory
+ * 		than necessary.
+ *
+ * 		Another useful use case is when parsing individual process
+ * 		arguments or individual environment variables navigating
+ * 		*current*\ **->mm->arg_start** and *current*\
+ * 		**->mm->env_start**: using this helper and the return value,
+ * 		one can quickly iterate at the right offset of the memory area.
+ * 	Return
+ * 		On success, the strictly positive length of the string,
+ * 		including the trailing NUL character. On error, a negative
+ * 		value.
+ *
+ * int bpf_probe_read_kernel_str(void *dst, u32 size, const void *unsafe_ptr)
+ * 	Description
+ * 		Copy a NUL terminated string from an unsafe kernel address *unsafe_ptr*
+ * 		to *dst*. Same semantics as with bpf_probe_read_user_str() apply.
+ * 	Return
+ * 		On success, the strictly positive length of the string,	including
+ * 		the trailing NUL character. On error, a negative value.
+ *
+ * int bpf_tcp_send_ack(void *tp, u32 rcv_nxt)
+ *	Description
+ *		Send out a tcp-ack. *tp* is the in-kernel struct tcp_sock.
+ *		*rcv_nxt* is the ack_seq to be sent out.
+ *	Return
+ *		0 on success, or a negative error in case of failure.
+ *
+ * int bpf_send_signal_thread(u32 sig)
+ *	Description
+ *		Send signal *sig* to the thread corresponding to the current task.
+ *	Return
+ *		0 on success or successfully queued.
+ *
+ *		**-EBUSY** if work queue under nmi is full.
+ *
+ *		**-EINVAL** if *sig* is invalid.
+ *
+ *		**-EPERM** if no permission to send the *sig*.
+ *
+ *		**-EAGAIN** if bpf program can try again.
+ *
+ * u64 bpf_jiffies64(void)
+ *	Description
+ *		Obtain the 64bit jiffies
+ *	Return
+ *		The 64 bit jiffies
  */
 #define __BPF_FUNC_MAPPER(FN)		\
 	FN(unspec),			\
@@ -2821,7 +3003,16 @@ union bpf_attr {
 	FN(strtoul),			\
 	FN(sk_storage_get),		\
 	FN(sk_storage_delete),		\
-	FN(send_signal),
+	FN(send_signal),		\
+	FN(tcp_gen_syncookie),		\
+	FN(skb_output),			\
+	FN(probe_read_user),		\
+	FN(probe_read_kernel),		\
+	FN(probe_read_user_str),	\
+	FN(probe_read_kernel_str),	\
+	FN(tcp_send_ack),		\
+	FN(send_signal_thread),		\
+	FN(jiffies64),
 
 /* integer value in 'imm' field of BPF_CALL instruction selects which helper
  * function eBPF program intends to call
@@ -3222,7 +3413,7 @@ struct bpf_map_info {
 	__u32 map_flags;
 	char  name[BPF_OBJ_NAME_LEN];
 	__u32 ifindex;
-	__u32 :32;
+	__u32 btf_vmlinux_value_type_id;
 	__u64 netns_dev;
 	__u64 netns_ino;
 	__u32 btf_id;
@@ -3504,6 +3695,10 @@ enum bpf_task_fd_type {
 	BPF_FD_TYPE_URETPROBE,		/* filename + offset */
 };
 
+#define BPF_FLOW_DISSECTOR_F_PARSE_1ST_FRAG		(1U << 0)
+#define BPF_FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL		(1U << 1)
+#define BPF_FLOW_DISSECTOR_F_STOP_AT_ENCAP		(1U << 2)
+
 struct bpf_flow_keys {
 	__u16	nhoff;
 	__u16	thoff;
@@ -3525,6 +3720,8 @@ struct bpf_flow_keys {
 			__u32	ipv6_dst[4];	/* in6_addr; network order */
 		};
 	};
+	__u32	flags;
+	__be32	flow_label;
 };
 
 struct bpf_func_info {

include/uapi/linux/btf.h

@@ -22,9 +22,9 @@ struct btf_header {
 };
 
 /* Max # of type identifier */
-#define BTF_MAX_TYPE	0x0000ffff
+#define BTF_MAX_TYPE	0x000fffff
 /* Max offset into the string section */
-#define BTF_MAX_NAME_OFFSET	0x0000ffff
+#define BTF_MAX_NAME_OFFSET	0x00ffffff
 /* Max # of struct/union/enum members or func args */
 #define BTF_MAX_VLEN	0xffff
 
@@ -142,7 +142,14 @@ struct btf_param {
 
 enum {
 	BTF_VAR_STATIC = 0,
-	BTF_VAR_GLOBAL_ALLOCATED,
+	BTF_VAR_GLOBAL_ALLOCATED = 1,
+	BTF_VAR_GLOBAL_EXTERN = 2,
+};
+
+enum btf_func_linkage {
+	BTF_FUNC_STATIC = 0,
+	BTF_FUNC_GLOBAL = 1,
+	BTF_FUNC_EXTERN = 2,
 };
 
 /* BTF_KIND_VAR is followed by a single "struct btf_var" to describe

include/uapi/linux/if_link.h

@@ -167,6 +167,9 @@ enum {
 	IFLA_NEW_IFINDEX,
 	IFLA_MIN_MTU,
 	IFLA_MAX_MTU,
+	IFLA_PROP_LIST,
+	IFLA_ALT_IFNAME, /* Alternative ifname */
+	IFLA_PERM_ADDRESS,
 	__IFLA_MAX
 };
 
@@ -483,6 +486,13 @@ enum macsec_validation_type {
 	MACSEC_VALIDATE_MAX = __MACSEC_VALIDATE_END - 1,
 };
 
+enum macsec_offload {
+	MACSEC_OFFLOAD_OFF = 0,
+	MACSEC_OFFLOAD_PHY = 1,
+	__MACSEC_OFFLOAD_END,
+	MACSEC_OFFLOAD_MAX = __MACSEC_OFFLOAD_END - 1,
+};
+
 /* IPVLAN section */
 enum {
 	IFLA_IPVLAN_UNSPEC,
@@ -695,6 +705,7 @@ enum {
 	IFLA_VF_IB_NODE_GUID,	/* VF Infiniband node GUID */
 	IFLA_VF_IB_PORT_GUID,	/* VF Infiniband port GUID */
 	IFLA_VF_VLAN_LIST,	/* nested list of vlans, option for QinQ */
+	IFLA_VF_BROADCAST,	/* VF broadcast */
 	__IFLA_VF_MAX,
 };
 
@@ -705,6 +716,10 @@ struct ifla_vf_mac {
 	__u8 mac[32]; /* MAX_ADDR_LEN */
 };
 
+struct ifla_vf_broadcast {
+	__u8 broadcast[32];
+};
+
 struct ifla_vf_vlan {
 	__u32 vf;
 	__u32 vlan; /* 0 - 4095, 0 disables VLAN filter */

include/uapi/linux/if_xdp.h

@@ -16,6 +16,18 @@
 #define XDP_SHARED_UMEM	(1 << 0)
 #define XDP_COPY	(1 << 1) /* Force copy-mode */
 #define XDP_ZEROCOPY	(1 << 2) /* Force zero-copy mode */
+/* If this option is set, the driver might go sleep and in that case
+ * the XDP_RING_NEED_WAKEUP flag in the fill and/or Tx rings will be
+ * set. If it is set, the application need to explicitly wake up the
+ * driver with a poll() (Rx and Tx) or sendto() (Tx only). If you are
+ * running the driver and the application on the same core, you should
+ * use this option so that the kernel will yield to the user space
+ * application.
+ */
+#define XDP_USE_NEED_WAKEUP (1 << 3)
+
+/* Flags for xsk_umem_config flags */
+#define XDP_UMEM_UNALIGNED_CHUNK_FLAG (1 << 0)
 
 struct sockaddr_xdp {
 	__u16 sxdp_family;
@@ -25,10 +37,14 @@ struct sockaddr_xdp {
 	__u32 sxdp_shared_umem_fd;
 };
 
+/* XDP_RING flags */
+#define XDP_RING_NEED_WAKEUP (1 << 0)
+
 struct xdp_ring_offset {
 	__u64 producer;
 	__u64 consumer;
 	__u64 desc;
+	__u64 flags;
 };
 
 struct xdp_mmap_offsets {
@@ -53,6 +69,7 @@ struct xdp_umem_reg {
 	__u64 len; /* Length of packet data area */
 	__u32 chunk_size;
 	__u32 headroom;
+	__u32 flags;
 };
 
 struct xdp_statistics {
@@ -74,6 +91,11 @@ struct xdp_options {
 #define XDP_UMEM_PGOFF_FILL_RING	0x100000000ULL
 #define XDP_UMEM_PGOFF_COMPLETION_RING	0x180000000ULL
 
+/* Masks for unaligned chunks mode */
+#define XSK_UNALIGNED_BUF_OFFSET_SHIFT 48
+#define XSK_UNALIGNED_BUF_ADDR_MASK \
+	((1ULL << XSK_UNALIGNED_BUF_OFFSET_SHIFT) - 1)
+
 /* Rx/Tx descriptor */
 struct xdp_desc {
 	__u64 addr;

meson.build

@@ -1,95 +0,0 @@
-# SPDX-License-Identifier: LGPL-2.1 OR BSD-2-Clause
-project('libbpf', 'c',
-        version : '0.0.3',
-        license : 'LGPL-2.1 OR BSD-2-Clause',
-        default_options : [
-                'prefix=/usr',
-        ],
-        meson_version : '>= 0.46',
-        )
-
-patchlevel = meson.project_version().split('.')[1]
-
-libbpf_source_dir = './'
-
-libbpf_sources = files(run_command('find',
-        [
-        '@0@/src'.format(libbpf_source_dir),
-        '-type',
-        'f',
-        '-name',
-        '*.[h|c]']).stdout().split())
-
-libbpf_headers = files(
-        join_paths(libbpf_source_dir, 'src/bpf.h'),
-        join_paths(libbpf_source_dir, 'src/btf.h'),
-        join_paths(libbpf_source_dir, 'src/libbpf.h'))
-
-feature_rellocarray = run_command(join_paths(libbpf_source_dir, 'scripts/check-reallocarray.sh'))
-
-libbpf_c_args = ['-g',
-                '-O2',
-                '-Werror',
-                '-Wall',
-]
-
-if feature_rellocarray.stdout().strip() != ''
-        libbpf_c_args += '-DCOMPAT_NEED_REALLOCARRAY'
-endif
-
-# bpf_includes are required to include bpf.h, btf.h, libbpf.h
-bpf_includes = include_directories(
-        join_paths(libbpf_source_dir, 'src'))
-
-libbpf_includes = include_directories(
-        join_paths(libbpf_source_dir, 'include'),
-        join_paths(libbpf_source_dir, 'include/uapi'))
-
-libelf = dependency('libelf')
-libelf = dependency('libelf', required: false)
-if not libelf.found()
-        libelf = cc.find_library('elf', required: true)
-endif
-
-deps = [libelf]
-
-libbpf_static = static_library(
-        'bpf',
-        libbpf_sources,
-        c_args : libbpf_c_args,
-        dependencies : deps,
-        include_directories : libbpf_includes,
-        install : true)
-
-libbpf_static_dep = declare_dependency(link_with : libbpf_static)
-
-libbpf_map_source_path = join_paths(libbpf_source_dir, 'src/libbpf.map')
-libbpf_map_abs_path = join_paths(meson.current_source_dir(), libbpf_map_source_path)
-
-libbpf_c_args += ['-fPIC', '-fvisibility=hidden']
-
-libbpf_link_args = ['-Wl,--version-script=@0@'.format(libbpf_map_abs_path)]
-
-libbpf_shared = shared_library(
-        'bpf',
-        libbpf_sources,
-        c_args : libbpf_c_args,
-        dependencies : deps,
-        include_directories : libbpf_includes,
-        install : true,
-        link_args : libbpf_link_args,
-        link_depends : libbpf_map_source_path,
-        soversion : patchlevel,
-        version : meson.project_version())
-
-libbpf_shared_dep = declare_dependency(link_with : libbpf_shared)
-
-install_headers(libbpf_headers, subdir : 'bpf')
-
-pkg = import('pkgconfig')
-pkg.generate(
-        name: meson.project_name(),
-        version: meson.project_version(),
-        libraries: libbpf_shared,
-        requires_private: ['libelf'],
-        description: '''BPF library''')

scripts/coverity.sh

@@ -0,0 +1,105 @@
+#!/bin/bash
+# Taken from: https://scan.coverity.com/scripts/travisci_build_coverity_scan.sh
+# Local changes are annotated with "#[local]"
+
+set -e
+
+# Environment check
+echo -e "\033[33;1mNote: COVERITY_SCAN_PROJECT_NAME and COVERITY_SCAN_TOKEN are available on Project Settings page on scan.coverity.com\033[0m"
+[ -z "$COVERITY_SCAN_PROJECT_NAME" ] && echo "ERROR: COVERITY_SCAN_PROJECT_NAME must be set" && exit 1
+[ -z "$COVERITY_SCAN_NOTIFICATION_EMAIL" ] && echo "ERROR: COVERITY_SCAN_NOTIFICATION_EMAIL must be set" && exit 1
+[ -z "$COVERITY_SCAN_BRANCH_PATTERN" ] && echo "ERROR: COVERITY_SCAN_BRANCH_PATTERN must be set" && exit 1
+[ -z "$COVERITY_SCAN_BUILD_COMMAND" ] && echo "ERROR: COVERITY_SCAN_BUILD_COMMAND must be set" && exit 1
+[ -z "$COVERITY_SCAN_TOKEN" ] && echo "ERROR: COVERITY_SCAN_TOKEN must be set" && exit 1
+
+PLATFORM=`uname`
+#[local] Use /var/tmp for TOOL_ARCHIVE and TOOL_BASE, as on certain systems
+# /tmp is tmpfs and is sometimes too small to handle all necessary tooling
+TOOL_ARCHIVE=/var//tmp/cov-analysis-${PLATFORM}.tgz
+TOOL_URL=https://scan.coverity.com/download/${PLATFORM}
+TOOL_BASE=/var/tmp/coverity-scan-analysis
+UPLOAD_URL="https://scan.coverity.com/builds"
+SCAN_URL="https://scan.coverity.com"
+
+# Do not run on pull requests
+if [ "${TRAVIS_PULL_REQUEST}" = "true" ]; then
+  echo -e "\033[33;1mINFO: Skipping Coverity Analysis: branch is a pull request.\033[0m"
+  exit 0
+fi
+
+# Verify this branch should run
+IS_COVERITY_SCAN_BRANCH=`ruby -e "puts '${TRAVIS_BRANCH}' =~ /\\A$COVERITY_SCAN_BRANCH_PATTERN\\z/ ? 1 : 0"`
+if [ "$IS_COVERITY_SCAN_BRANCH" = "1" ]; then
+  echo -e "\033[33;1mCoverity Scan configured to run on branch ${TRAVIS_BRANCH}\033[0m"
+else
+  echo -e "\033[33;1mCoverity Scan NOT configured to run on branch ${TRAVIS_BRANCH}\033[0m"
+  exit 1
+fi
+
+# Verify upload is permitted
+AUTH_RES=`curl -s --form project="$COVERITY_SCAN_PROJECT_NAME" --form token="$COVERITY_SCAN_TOKEN" $SCAN_URL/api/upload_permitted`
+if [ "$AUTH_RES" = "Access denied" ]; then
+  echo -e "\033[33;1mCoverity Scan API access denied. Check COVERITY_SCAN_PROJECT_NAME and COVERITY_SCAN_TOKEN.\033[0m"
+  exit 1
+else
+  AUTH=`echo $AUTH_RES | ruby -e "require 'rubygems'; require 'json'; puts JSON[STDIN.read]['upload_permitted']"`
+  if [ "$AUTH" = "true" ]; then
+    echo -e "\033[33;1mCoverity Scan analysis authorized per quota.\033[0m"
+  else
+    WHEN=`echo $AUTH_RES | ruby -e "require 'rubygems'; require 'json'; puts JSON[STDIN.read]['next_upload_permitted_at']"`
+    echo -e "\033[33;1mCoverity Scan analysis NOT authorized until $WHEN.\033[0m"
+    exit 0
+  fi
+fi
+
+if [ ! -d $TOOL_BASE ]; then
+  # Download Coverity Scan Analysis Tool
+  if [ ! -e $TOOL_ARCHIVE ]; then
+    echo -e "\033[33;1mDownloading Coverity Scan Analysis Tool...\033[0m"
+    wget -nv -O $TOOL_ARCHIVE $TOOL_URL --post-data "project=$COVERITY_SCAN_PROJECT_NAME&token=$COVERITY_SCAN_TOKEN"
+  fi
+
+  # Extract Coverity Scan Analysis Tool
+  echo -e "\033[33;1mExtracting Coverity Scan Analysis Tool...\033[0m"
+  mkdir -p $TOOL_BASE
+  pushd $TOOL_BASE
+  tar xzf $TOOL_ARCHIVE
+  popd
+fi
+
+TOOL_DIR=`find $TOOL_BASE -type d -name 'cov-analysis*'`
+export PATH=$TOOL_DIR/bin:$PATH
+
+# Build
+echo -e "\033[33;1mRunning Coverity Scan Analysis Tool...\033[0m"
+COV_BUILD_OPTIONS=""
+#COV_BUILD_OPTIONS="--return-emit-failures 8 --parse-error-threshold 85"
+RESULTS_DIR="cov-int"
+eval "${COVERITY_SCAN_BUILD_COMMAND_PREPEND}"
+COVERITY_UNSUPPORTED=1 cov-build --dir $RESULTS_DIR $COV_BUILD_OPTIONS $COVERITY_SCAN_BUILD_COMMAND
+cov-import-scm --dir $RESULTS_DIR --scm git --log $RESULTS_DIR/scm_log.txt 2>&1
+
+# Upload results
+echo -e "\033[33;1mTarring Coverity Scan Analysis results...\033[0m"
+RESULTS_ARCHIVE=analysis-results.tgz
+tar czf $RESULTS_ARCHIVE $RESULTS_DIR
+SHA=`git rev-parse --short HEAD`
+
+echo -e "\033[33;1mUploading Coverity Scan Analysis results...\033[0m"
+response=$(curl \
+  --silent --write-out "\n%{http_code}\n" \
+  --form project=$COVERITY_SCAN_PROJECT_NAME \
+  --form token=$COVERITY_SCAN_TOKEN \
+  --form email=$COVERITY_SCAN_NOTIFICATION_EMAIL \
+  --form file=@$RESULTS_ARCHIVE \
+  --form version=$SHA \
+  --form description="Travis CI build" \
+  $UPLOAD_URL)
+status_code=$(echo "$response" | sed -n '$p')
+#[local] Coverity used to return 201 on success, but it's 200 now
+# See https://github.com/systemd/systemd/blob/master/tools/coverity.sh#L145
+if [ "$status_code" != "200" ]; then
+  TEXT=$(echo "$response" | sed '$d')
+  echo -e "\033[33;1mCoverity Scan upload failed: $TEXT.\033[0m"
+  exit 1
+fi

scripts/sync-kernel.sh

@@ -1,40 +1,212 @@
 #!/bin/bash
 
 usage () {
-    echo "USAGE: ./sync-kernel.sh <kernel-repo> <libbpf-repo> [<baseline-commit>]"
+	echo "USAGE: ./sync-kernel.sh <libbpf-repo> <kernel-repo> <bpf-branch>"
-    echo ""
+	echo ""
-    echo "If <baseline-commit> is not specified, it's read from <libbpf-repo>/CHECKPOINT-COMMIT"
+	echo "Set BPF_NEXT_BASELINE to override bpf-next tree commit, otherwise read from <libbpf-repo>/CHECKPOINT-COMMIT."
-    exit 1
+	echo "Set BPF_BASELINE to override bpf tree commit, otherwise read from <libbpf-repo>/BPF-CHECKPOINT-COMMIT."
+	echo "Set MANUAL_MODE to 1 to manually control every cherry-picked commits."
+	echo "Set IGNORE_CONSISTENCY to 1 to ignore failed contents consistency check."
+	exit 1
 }
 
-LINUX_REPO=${1-""}
-LIBBPF_REPO=${2-""}
-
-if [ -z "${LINUX_REPO}" ]; then
-    usage
-fi
-if [ -z "${LIBBPF_REPO}" ]; then
-    usage
-fi
-
 set -eu
 
-WORKDIR=$(pwd)
+LIBBPF_REPO=${1-""}
-trap "cd ${WORKDIR}; exit" INT TERM EXIT
+LINUX_REPO=${2-""}
+BPF_BRANCH=${3-""}
+BASELINE_COMMIT=${BPF_NEXT_BASELINE:-$(cat ${LIBBPF_REPO}/CHECKPOINT-COMMIT)}
+BPF_BASELINE_COMMIT=${BPF_BASELINE:-$(cat ${LIBBPF_REPO}/BPF-CHECKPOINT-COMMIT)}
 
-echo "WORKDIR:         ${WORKDIR}"
+if [ -z "${LIBBPF_REPO}" ] || [ -z "${LINUX_REPO}" ] || [ -z "${BPF_BRANCH}" ]; then
-echo "LINUX REPO:      ${LINUX_REPO}"
+	echo "Error: libbpf or linux repos are not specified"
-echo "LIBBPF REPO:     ${LIBBPF_REPO}"
+	usage
+fi
+if [ -z "${BPF_BRANCH}" ]; then
+	echo "Error: linux's bpf tree branch is not specified"
+	usage
+fi
+if [ -z "${BASELINE_COMMIT}" ] || [ -z "${BPF_BASELINE_COMMIT}" ]; then
+	echo "Error: bpf or bpf-next baseline commits are not provided"
+	usage
+fi
 
 SUFFIX=$(date --utc +%Y-%m-%dT%H-%M-%S.%3NZ)
-BASELINE_COMMIT=${3-$(cat ${LIBBPF_REPO}/CHECKPOINT-COMMIT)}
+WORKDIR=$(pwd)
+TMP_DIR=$(mktemp -d)
+
+trap "cd ${WORKDIR}; exit" INT TERM EXIT
+
+declare -A PATH_MAP
+PATH_MAP=(									\
+	[tools/lib/bpf]=src							\
+	[tools/include/uapi/linux/bpf_common.h]=include/uapi/linux/bpf_common.h	\
+	[tools/include/uapi/linux/bpf.h]=include/uapi/linux/bpf.h		\
+	[tools/include/uapi/linux/btf.h]=include/uapi/linux/btf.h		\
+	[tools/include/uapi/linux/if_link.h]=include/uapi/linux/if_link.h	\
+	[tools/include/uapi/linux/if_xdp.h]=include/uapi/linux/if_xdp.h		\
+	[tools/include/uapi/linux/netlink.h]=include/uapi/linux/netlink.h	\
+	[tools/include/tools/libc_compat.h]=include/tools/libc_compat.h		\
+)
+
+LIBBPF_PATHS="${!PATH_MAP[@]}"
+LIBBPF_VIEW_PATHS="${PATH_MAP[@]}"
+LIBBPF_VIEW_EXCLUDE_REGEX='^src/(Makefile|Build|test_libbpf\.c|bpf_helper_defs\.h|\.gitignore)$'
+
+LIBBPF_TREE_FILTER="mkdir -p __libbpf/include/uapi/linux __libbpf/include/tools && "$'\\\n'
+for p in "${!PATH_MAP[@]}"; do
+	LIBBPF_TREE_FILTER+="git mv -kf ${p} __libbpf/${PATH_MAP[${p}]} && "$'\\\n'
+done
+LIBBPF_TREE_FILTER+="git rm --ignore-unmatch -f __libbpf/src/{Makefile,Build,test_libbpf.c,.gitignore} >/dev/null"
+
+cd_to()
+{
+	cd ${WORKDIR} && cd "$1"
+}
+
+# Output brief single-line commit description
+# $1 - commit ref
+commit_desc()
+{
+	git log -n1 --pretty='%h ("%s")' $1
+}
+
+# Create commit single-line signature, which consists of:
+# - full commit hash
+# - author date in ISO8601 format
+# - full commit body with newlines replaced with vertical bars (|)
+# - shortstat appended at the end
+# The idea is that this single-line signature is good enough to make final
+# decision about whether two commits are the same, across different repos.
+# $1 - commit ref
+commit_signature()
+{
+	git log -n1 --pretty='("%s")|%aI|%b' --shortstat $1 | tr '\n' '|'
+}
+
+# Validate there are no non-empty merges (we can't handle them)
+# $1 - baseline tag
+# $2 - tip tag
+validate_merges()
+{
+	local baseline_tag=$1
+	local tip_tag=$2
+	local new_merges
+	local merge_change_cnt
+	local ignore_merge_resolutions
+	local desc
+
+	new_merges=$(git rev-list --merges --topo-order --reverse ${baseline_tag}..${tip_tag} ${LIBBPF_PATHS[@]})
+	for new_merge in ${new_merges}; do
+		desc=$(commit_desc ${new_merge})
+		echo "MERGE: ${desc}"
+		merge_change_cnt=$(git show --format='' ${new_merge} | wc -l)
+		if ((${merge_change_cnt} > 0)); then
+			read -p "Merge '${desc}' is non-empty, which will cause conflicts! Do you want to proceed? [y/N]: " ignore_merge_resolutions
+			case "${ignore_merge_resolutions}" in
+				"y" | "Y")
+					echo "Skipping '${desc}'..."
+					continue
+					;;
+			esac
+			exit 3
+		fi
+	done
+}
+
+# Cherry-pick commits touching libbpf-related files
+# $1 - baseline_tag
+# $2 - tip_tag
+cherry_pick_commits()
+{
+	local manual_mode=${MANUAL_MODE:-0}
+	local baseline_tag=$1
+	local tip_tag=$2
+	local new_commits
+	local signature
+	local should_skip
+	local synced_cnt
+	local manual_check
+	local libbpf_conflict_cnt
+	local desc
+
+	new_commits=$(git rev-list --no-merges --topo-order --reverse ${baseline_tag}..${tip_tag} ${LIBBPF_PATHS[@]})
+	for new_commit in ${new_commits}; do
+		desc="$(commit_desc ${new_commit})"
+		signature="$(commit_signature ${new_commit})"
+		synced_cnt=$(grep -F "${signature}" ${TMP_DIR}/libbpf_commits.txt | wc -l)
+		manual_check=0
+		if ((${synced_cnt} > 0)); then
+			# commit with the same subject is already in libbpf, but it's
+			# not 100% the same commit, so check with user
+			echo "Commit '${desc}' is synced into libbpf as:"
+			grep -F "${signature}" ${TMP_DIR}/libbpf_commits.txt | \
+				cut -d'|' -f1 | sed -e 's/^/- /'
+			if ((${manual_mode} != 1 && ${synced_cnt} == 1)); then
+				echo "Skipping '${desc}' due to unique match..."
+				continue
+			fi
+			if ((${synced_cnt} > 1)); then
+				echo "'${desc} matches multiple commits, please, double-check!"
+				manual_check=1
+			fi
+		fi
+		if ((${manual_mode} == 1 || ${manual_check} == 1)); then
+			read -p "Do you want to skip '${desc}'? [y/N]: " should_skip
+			case "${should_skip}" in
+				"y" | "Y")
+					echo "Skipping '${desc}'..."
+					continue
+					;;
+			esac
+		fi
+		# commit hasn't been synced into libbpf yet
+		echo "Picking '${desc}'..."
+		if ! git cherry-pick ${new_commit} &>/dev/null; then
+			echo "Warning! Cherry-picking '${desc} failed, checking if it's non-libbpf files causing problems..."
+			libbpf_conflict_cnt=$(git diff --name-only --diff-filter=U -- ${LIBBPF_PATHS[@]} | wc -l)
+			conflict_cnt=$(git diff --name-only | wc -l)
+
+			if ((${libbpf_conflict_cnt} == 0)); then
+				echo "Looks like only non-libbpf files have conflicts, ignoring..."
+				if ((${conflict_cnt} == 0)); then
+					echo "Empty cherry-pick, skipping it..."
+					git cherry-pick --abort
+					continue
+				fi
+
+				git add .
+				# GIT_EDITOR=true to avoid editor popping up to edit commit message
+				if ! GIT_EDITOR=true git cherry-pick --continue &>/dev/null; then
+					echo "Error! That still failed! Please resolve manually."
+				else
+					echo "Success! All cherry-pick conflicts were resolved for '${desc}'!"
+					continue
+				fi
+			fi
+
+			read -p "Error! Cherry-picking '${desc}' failed, please fix manually and press <return> to proceed..."
+		fi
+	done
+}
+
+cd_to ${LIBBPF_REPO}
+GITHUB_ABS_DIR=$(pwd)
+echo "Dumping existing libbpf commit signatures..."
+for h in $(git log --pretty='%h' -n500); do
+	echo $h "$(commit_signature $h)" >> ${TMP_DIR}/libbpf_commits.txt
+done
 
 # Use current kernel repo HEAD as a source of patches
-cd ${LINUX_REPO}
+cd_to ${LINUX_REPO}
+LINUX_ABS_DIR=$(pwd)
 TIP_SYM_REF=$(git symbolic-ref -q --short HEAD || git rev-parse HEAD)
 TIP_COMMIT=$(git rev-parse HEAD)
+BPF_TIP_COMMIT=$(git rev-parse ${BPF_BRANCH})
 BASELINE_TAG=libbpf-baseline-${SUFFIX}
 TIP_TAG=libbpf-tip-${SUFFIX}
+BPF_BASELINE_TAG=libbpf-bpf-baseline-${SUFFIX}
+BPF_TIP_TAG=libbpf-bpf-tip-${SUFFIX}
 VIEW_TAG=libbpf-view-${SUFFIX}
 LIBBPF_SYNC_TAG=libbpf-sync-${SUFFIX}
 
@@ -43,75 +215,41 @@ SQUASH_BASE_TAG=libbpf-squash-base-${SUFFIX}
 SQUASH_TIP_TAG=libbpf-squash-tip-${SUFFIX}
 SQUASH_COMMIT=$(git commit-tree ${BASELINE_COMMIT}^{tree} -m "BASELINE SQUASH ${BASELINE_COMMIT}")
 
-echo "SUFFIX:          ${SUFFIX}"
+echo "WORKDIR:          ${WORKDIR}"
-echo "BASELINE COMMIT: $(git log --pretty=oneline --no-walk ${BASELINE_COMMIT})"
+echo "LINUX REPO:       ${LINUX_REPO}"
-echo "TIP COMMIT:      $(git log --pretty=oneline --no-walk ${TIP_COMMIT})"
+echo "LIBBPF REPO:      ${LIBBPF_REPO}"
-echo "SQUASH COMMIT:   ${SQUASH_COMMIT}"
+echo "TEMP DIR:         ${TMP_DIR}"
-echo "BASELINE TAG:    ${BASELINE_TAG}"
+echo "SUFFIX:           ${SUFFIX}"
-echo "TIP TAG:         ${TIP_TAG}"
+echo "BASE COMMIT:      '$(commit_desc ${BASELINE_COMMIT})'"
-echo "SQUASH BASE TAG: ${SQUASH_BASE_TAG}"
+echo "TIP COMMIT:       '$(commit_desc ${TIP_COMMIT})'"
-echo "SQUASH TIP TAG:  ${SQUASH_TIP_TAG}"
+echo "BPF BASE COMMIT:  '$(commit_desc ${BPF_BASELINE_COMMIT})'"
-echo "VIEW TAG:        ${VIEW_TAG}"
+echo "BPF TIP COMMIT:   '$(commit_desc ${BPF_TIP_COMMIT})'"
-echo "LIBBPF SYNC TAG: ${LIBBPF_SYNC_TAG}"
+echo "SQUASH COMMIT:    ${SQUASH_COMMIT}"
-
+echo "BASELINE TAG:     ${BASELINE_TAG}"
-TMP_DIR=$(mktemp -d)
+echo "TIP TAG:          ${TIP_TAG}"
-echo "TEMP DIR:        ${TMP_DIR}"
+echo "BPF BASELINE TAG: ${BPF_BASELINE_TAG}"
-echo "PATCHES+COVER:   ${TMP_DIR}/patches"
+echo "BPF TIP TAG:      ${BPF_TIP_TAG}"
-echo "PATCHSET:        ${TMP_DIR}/patchset.patch"
+echo "SQUASH BASE TAG:  ${SQUASH_BASE_TAG}"
+echo "SQUASH TIP TAG:   ${SQUASH_TIP_TAG}"
+echo "VIEW TAG:         ${VIEW_TAG}"
+echo "LIBBPF SYNC TAG:  ${LIBBPF_SYNC_TAG}"
+echo "PATCHES:          ${TMP_DIR}/patches"
 
 git branch ${BASELINE_TAG} ${BASELINE_COMMIT}
 git branch ${TIP_TAG} ${TIP_COMMIT}
+git branch ${BPF_BASELINE_TAG} ${BPF_BASELINE_COMMIT}
+git branch ${BPF_TIP_TAG} ${BPF_TIP_COMMIT}
 git branch ${SQUASH_BASE_TAG} ${SQUASH_COMMIT}
 git checkout -b ${SQUASH_TIP_TAG} ${SQUASH_COMMIT}
 
+# Validate there are no non-empty merges in bpf-next and bpf trees
+validate_merges ${BASELINE_TAG} ${TIP_TAG}
+validate_merges ${BPF_BASELINE_TAG} ${BPF_TIP_TAG}
+
 # Cherry-pick new commits onto squashed baseline commit
-LIBBPF_PATHS=(tools/lib/bpf tools/include/uapi/linux/{bpf_common.h,bpf.h,btf.h,if_link.h,if_xdp.h,netlink.h} tools/include/tools/libc_compat.h)
+cherry_pick_commits ${BASELINE_TAG} ${TIP_TAG}
+cherry_pick_commits ${BPF_BASELINE_TAG} ${BPF_TIP_TAG}
 
-LIBBPF_NEW_MERGES=$(git rev-list --merges --topo-order --reverse ${BASELINE_TAG}..${TIP_TAG} ${LIBBPF_PATHS[@]})
-for LIBBPF_NEW_MERGE in ${LIBBPF_NEW_MERGES}; do
-	printf "MERGE:\t" && git log --oneline -n1 ${LIBBPF_NEW_MERGE}
-	MERGE_CHANGES=$(git log --format='' -n1 ${LIBBPF_NEW_MERGE} | wc -l)
-	if ((${MERGE_CHANGES} > 0)); then
-		echo "Merge is non empty, aborting!.."
-		exit 3
-	fi
-done
-
-cd ${WORKDIR} && cd ${LIBBPF_REPO}
-git log --oneline -n500 > ${TMP_DIR}/libbpf_commits.txt
-cd ${WORKDIR} && cd ${LINUX_REPO}
-
-LIBBPF_NEW_COMMITS=$(git rev-list --no-merges --topo-order --reverse ${BASELINE_TAG}..${TIP_TAG} ${LIBBPF_PATHS[@]})
-for LIBBPF_NEW_COMMIT in ${LIBBPF_NEW_COMMITS}; do
-	echo "Checking commit '${LIBBPF_NEW_COMMIT}'"
-	SYNCED_COMMITS=$(grep -F "$(git log -n1 --pretty=format:%s ${LIBBPF_NEW_COMMIT})" ${TMP_DIR}/libbpf_commits.txt || echo "")
-	if [ -n "${SYNCED_COMMITS}" ]; then
-		# commit with the same subject is already in libbpf, but it's not 100% the same commit, so check with user
-		echo "Commit '$(git log -n1 --oneline ${LIBBPF_NEW_COMMIT})' appears to be already synced into libbpf..."
-		echo "Corresponding libbpf commit(s):"
-		echo "${SYNCED_COMMITS}"
-		read -p "Do you want to skip it? [y/N]: " SHOULD_SKIP
-		case "${SHOULD_SKIP}" in 
-			"y" | "Y")
-				echo "Skipping '$(git log -n1 --oneline ${LIBBPF_NEW_COMMIT})'..."
-				continue
-				;;
-		esac
-	fi
-	# commit hasn't been synced into libbpf yet
-	if ! git cherry-pick ${LIBBPF_NEW_COMMIT}; then 
-		read -p "Cherry-picking '$(git log --oneline -n1 ${LIBBPF_NEW_COMMIT})' failed, please fix manually and press <return> to proceed..."
-	fi
-done
-
-LIBBPF_TREE_FILTER='												\
-    mkdir -p __libbpf/include/uapi/linux __libbpf/include/tools &&						\
-    git mv -kf tools/lib/bpf __libbpf/src &&									\
-    git mv -kf tools/include/uapi/linux/{bpf_common.h,bpf.h,btf.h,if_link.h,if_xdp.h,netlink.h}			\
-	       __libbpf/include/uapi/linux &&									\
-    git mv -kf tools/include/tools/libc_compat.h __libbpf/include/tools &&					\
-    git rm --ignore-unmatch -f __libbpf/src/{Makefile,Build,test_libbpf.cpp,.gitignore}				\
-'
 # Move all libbpf files into __libbpf directory.
 git filter-branch --prune-empty -f --tree-filter "${LIBBPF_TREE_FILTER}" ${SQUASH_TIP_TAG} ${SQUASH_BASE_TAG}
 # Make __libbpf a new root directory
@@ -126,59 +264,92 @@ fi
 
 # Exclude baseline commit and generate nice cover letter with summary
 git format-patch ${SQUASH_BASE_TAG}..${SQUASH_TIP_TAG} --cover-letter -o ${TMP_DIR}/patches
-# Now generate single-file patchset w/o cover to apply on top of libbpf repo
-git format-patch ${SQUASH_BASE_TAG}..${SQUASH_TIP_TAG} --stdout > ${TMP_DIR}/patchset.patch
 
-# Now is time to re-apply libbpf-related linux  patches to libbpf repo
+# Now is time to re-apply libbpf-related linux patches to libbpf repo
-cd ${WORKDIR} && cd ${LIBBPF_REPO}
+cd_to ${LIBBPF_REPO}
 git checkout -b ${LIBBPF_SYNC_TAG}
-git am --committer-date-is-author-date ${TMP_DIR}/patchset.patch
+
+for patch in $(ls -1 ${TMP_DIR}/patches | tail -n +2); do
+	if ! git am --committer-date-is-author-date "${TMP_DIR}/patches/${patch}"; then
+		read -p "Applying ${TMP_DIR}/patches/${patch} failed, please resolve manually and press <return> to proceed..."
+	fi
+done
+
+# Generate bpf_helper_defs.h and commit, if anything changed
+# restore Linux tip to use bpf_helpers_doc.py
+cd_to ${LINUX_REPO}
+git checkout ${TIP_TAG}
+# re-generate bpf_helper_defs.h
+cd_to ${LIBBPF_REPO}
+"${LINUX_ABS_DIR}/scripts/bpf_helpers_doc.py" --header			    \
+	--file include/uapi/linux/bpf.h > src/bpf_helper_defs.h
+# if anything changed, commit it
+helpers_changes=$(git status --porcelain src/bpf_helper_defs.h | wc -l)
+if ((${helpers_changes} == 1)); then
+	git add src/bpf_helper_defs.h
+	git commit -m "sync: auto-generate latest BPF helpers
+
+Latest changes to BPF helper definitions.
+" -- src/bpf_helper_defs.h
+fi
 
 # Use generated cover-letter as a template for "sync commit" with
 # baseline and checkpoint commits from kernel repo (and leave summary
 # from cover letter intact, of course)
 echo ${TIP_COMMIT} > CHECKPOINT-COMMIT &&					      \
+echo ${BPF_TIP_COMMIT} > BPF-CHECKPOINT-COMMIT &&				      \
 git add CHECKPOINT-COMMIT &&							      \
+git add BPF-CHECKPOINT-COMMIT &&						      \
 awk '/\*\*\* BLURB HERE \*\*\*/ {p=1} p' ${TMP_DIR}/patches/0000-cover-letter.patch | \
 sed "s/\*\*\* BLURB HERE \*\*\*/\
 sync: latest libbpf changes from kernel\n\
 \n\
 Syncing latest libbpf commits from kernel repository.\n\
-Baseline commit:   ${BASELINE_COMMIT}\n\
+Baseline bpf-next commit:   ${BASELINE_COMMIT}\n\
-Checkpoint commit: ${TIP_COMMIT}/" |						      \
+Checkpoint bpf-next commit: ${TIP_COMMIT}\n\
+Baseline bpf commit:        ${BPF_BASELINE_COMMIT}\n\
+Checkpoint bpf commit:      ${BPF_TIP_COMMIT}/" |				      \
 git commit --file=-
 
 echo "SUCCESS! ${COMMIT_CNT} commits synced."
 
 echo "Verifying Linux's and Github's libbpf state"
-LIBBPF_VIEW_PATHS=(src include/uapi/linux/{bpf_common.h,bpf.h,btf.h,if_link.h,if_xdp.h,netlink.h} include/tools/libc_compat.h)
-LIBBPF_VIEW_EXCLUDE_REGEX='^src/(Makefile|Build|test_libbpf.cpp|\.gitignore)$'
 
-cd ${WORKDIR} && cd ${LINUX_REPO}
+cd_to ${LINUX_REPO}
-LINUX_ABS_DIR=$(pwd)
 git checkout -b ${VIEW_TAG} ${TIP_COMMIT}
 git filter-branch -f --tree-filter "${LIBBPF_TREE_FILTER}" ${VIEW_TAG}^..${VIEW_TAG}
 git filter-branch -f --subdirectory-filter __libbpf ${VIEW_TAG}^..${VIEW_TAG}
 git ls-files -- ${LIBBPF_VIEW_PATHS[@]} > ${TMP_DIR}/linux-view.ls
 
-cd ${WORKDIR} && cd ${LIBBPF_REPO}
+cd_to ${LIBBPF_REPO}
-GITHUB_ABS_DIR=$(pwd)
 git ls-files -- ${LIBBPF_VIEW_PATHS[@]} | grep -v -E "${LIBBPF_VIEW_EXCLUDE_REGEX}" > ${TMP_DIR}/github-view.ls
 
 echo "Comparing list of files..."
-diff ${TMP_DIR}/linux-view.ls ${TMP_DIR}/github-view.ls
+diff -u ${TMP_DIR}/linux-view.ls ${TMP_DIR}/github-view.ls
 echo "Comparing file contents..."
+CONSISTENT=1
 for F in $(cat ${TMP_DIR}/linux-view.ls); do
-	diff "${LINUX_ABS_DIR}/${F}" "${GITHUB_ABS_DIR}/${F}"
+	if ! diff -u "${LINUX_ABS_DIR}/${F}" "${GITHUB_ABS_DIR}/${F}"; then
+		echo "${LINUX_ABS_DIR}/${F} and ${GITHUB_ABS_DIR}/${F} are different!"
+		CONSISTENT=0
+	fi
 done
-echo "Contents appear identical!"
+if ((${CONSISTENT} == 1)); then
+	echo "Great! Content is identical!"
+else
+	echo "Unfortunately, there are consistency problems!"
+	if ((${IGNORE_CONSISTENCY-0} != 1)); then
+		exit 4
+	fi
+fi
 
 echo "Cleaning up..."
 rm -r ${TMP_DIR}
-cd ${WORKDIR} && cd ${LINUX_REPO}
+cd_to ${LINUX_REPO}
 git checkout ${TIP_SYM_REF}
-git branch -D ${BASELINE_TAG} ${TIP_TAG} ${SQUASH_BASE_TAG} ${SQUASH_TIP_TAG} ${VIEW_TAG}
+git branch -D ${BASELINE_TAG} ${TIP_TAG} ${BPF_BASELINE_TAG} ${BPF_TIP_TAG} \
+	      ${SQUASH_BASE_TAG} ${SQUASH_TIP_TAG} ${VIEW_TAG}
 
-cd ${WORKDIR}
+cd_to .
 echo "DONE."
 

src/.gitignore

@@ -2,3 +2,5 @@
 *.a
 /libbpf.pc
 /libbpf.so*
+/staticobjs
+/sharedobjs

src/Makefile

@@ -1,10 +1,9 @@
 # SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause)
 
-VERSION = 0
+LIBBPF_VERSION := $(shell \
-PATCHLEVEL = 0
+	grep -oE '^LIBBPF_([0-9.]+)' libbpf.map | \
-EXTRAVERSION = 3
+	sort -rV | head -n1 | cut -d'_' -f2)
-
+LIBBPF_MAJOR_VERSION := $(firstword $(subst ., ,$(LIBBPF_VERSION)))
-LIBBPF_VERSION	= $(VERSION).$(PATCHLEVEL).$(EXTRAVERSION)
 
 TOPDIR = ..
 
@@ -16,15 +15,13 @@ ifneq ($(FEATURE_REALLOCARRAY),)
 	ALL_CFLAGS += -DCOMPAT_NEED_REALLOCARRAY
 endif
 
-ifndef BUILD_STATIC_ONLY
+SHARED_CFLAGS += -fPIC -fvisibility=hidden -DSHARED
-	ALL_CFLAGS += -fPIC -fvisibility=hidden
-endif
 
 CFLAGS ?= -g -O2 -Werror -Wall
-ALL_CFLAGS += $(CFLAGS)
+ALL_CFLAGS += $(CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
 ALL_LDFLAGS += $(LDFLAGS)
 ifdef NO_PKG_CONFIG
-	ALL_LDFLAGS += -lelf
+	ALL_LDFLAGS += -lelf -lz
 else
 	PKG_CONFIG ?= pkg-config
 	ALL_CFLAGS += $(shell $(PKG_CONFIG) --cflags libelf)
@@ -32,22 +29,27 @@ else
 endif
 
 OBJDIR ?= .
-
+SHARED_OBJDIR := $(OBJDIR)/sharedobjs
-OBJS := $(addprefix $(OBJDIR)/,bpf.o btf.o libbpf.o libbpf_errno.o netlink.o \
+STATIC_OBJDIR := $(OBJDIR)/staticobjs
+OBJS := bpf.o btf.o libbpf.o libbpf_errno.o netlink.o \
 	nlattr.o str_error.o libbpf_probes.o bpf_prog_linfo.o xsk.o \
-	btf_dump.o hashmap.o)
+	btf_dump.o hashmap.o
+SHARED_OBJS := $(addprefix $(SHARED_OBJDIR)/,$(OBJS))
+STATIC_OBJS := $(addprefix $(STATIC_OBJDIR)/,$(OBJS))
 
-LIBS := $(OBJDIR)/libbpf.a
+STATIC_LIBS := $(OBJDIR)/libbpf.a
 ifndef BUILD_STATIC_ONLY
-	LIBS += $(OBJDIR)/libbpf.so \
+	SHARED_LIBS := $(OBJDIR)/libbpf.so \
-		$(OBJDIR)/libbpf.so.$(VERSION) \
+		       $(OBJDIR)/libbpf.so.$(LIBBPF_MAJOR_VERSION) \
-		$(OBJDIR)/libbpf.so.$(LIBBPF_VERSION)
+		       $(OBJDIR)/libbpf.so.$(LIBBPF_VERSION)
 	VERSION_SCRIPT := libbpf.map
 endif
 
-HEADERS := bpf.h libbpf.h btf.h xsk.h libbpf_util.h
+HEADERS := bpf.h libbpf.h btf.h xsk.h libbpf_util.h \
-UAPI_HEADERS := $(addprefix $(TOPDIR)/include/uapi/linux/,bpf.h bpf_common.h \
+	   bpf_helpers.h bpf_helper_defs.h bpf_tracing.h \
-	btf.h)
+	   bpf_endian.h bpf_core_read.h libbpf_common.h
+UAPI_HEADERS := $(addprefix $(TOPDIR)/include/uapi/linux/,\
+			    bpf.h bpf_common.h btf.h)
 
 PC_FILE := $(OBJDIR)/libbpf.pc
 
@@ -66,21 +68,23 @@ LIBDIR ?= $(PREFIX)/$(LIBSUBDIR)
 INCLUDEDIR ?= $(PREFIX)/include
 UAPIDIR ?= $(PREFIX)/include
 
-all: $(LIBS) $(PC_FILE)
+TAGS_PROG := $(if $(shell which etags 2>/dev/null),etags,ctags)
 
-$(OBJDIR)/libbpf.a: $(OBJS)
+all: $(STATIC_LIBS) $(SHARED_LIBS) $(PC_FILE)
+
+$(OBJDIR)/libbpf.a: $(STATIC_OBJS)
 	$(AR) rcs $@ $^
 
-$(OBJDIR)/libbpf.so: $(OBJDIR)/libbpf.so.$(VERSION)
+$(OBJDIR)/libbpf.so: $(OBJDIR)/libbpf.so.$(LIBBPF_MAJOR_VERSION)
 	ln -sf $(^F) $@
 
-$(OBJDIR)/libbpf.so.$(VERSION): $(OBJDIR)/libbpf.so.$(LIBBPF_VERSION)
+$(OBJDIR)/libbpf.so.$(LIBBPF_MAJOR_VERSION): $(OBJDIR)/libbpf.so.$(LIBBPF_VERSION)
 	ln -sf $(^F) $@
 
-$(OBJDIR)/libbpf.so.$(LIBBPF_VERSION): $(OBJS)
+$(OBJDIR)/libbpf.so.$(LIBBPF_VERSION): $(SHARED_OBJS)
-	$(CC) -shared $(ALL_LDFLAGS) -Wl,--version-script=$(VERSION_SCRIPT) \
+	$(CC) -shared -Wl,--version-script=$(VERSION_SCRIPT) \
-				     -Wl,-soname,libbpf.so.$(VERSION) \
+		      -Wl,-soname,libbpf.so.$(LIBBPF_MAJOR_VERSION) \
-				     $^ -o $@
+		      $^ $(ALL_LDFLAGS) -o $@
 
 $(OBJDIR)/libbpf.pc:
 	sed -e "s|@PREFIX@|$(PREFIX)|" \
@@ -88,9 +92,18 @@ $(OBJDIR)/libbpf.pc:
 		-e "s|@VERSION@|$(LIBBPF_VERSION)|" \
 		< libbpf.pc.template > $@
 
-$(OBJDIR)/%.o: %.c
+$(STATIC_OBJDIR):
+	mkdir -p $(STATIC_OBJDIR)
+
+$(SHARED_OBJDIR):
+	mkdir -p $(SHARED_OBJDIR)
+
+$(STATIC_OBJDIR)/%.o: %.c | $(STATIC_OBJDIR)
 	$(CC) $(ALL_CFLAGS) $(CPPFLAGS) -c $< -o $@
 
+$(SHARED_OBJDIR)/%.o: %.c | $(SHARED_OBJDIR)
+	$(CC) $(ALL_CFLAGS) $(SHARED_CFLAGS) $(CPPFLAGS) -c $< -o $@
+
 define do_install
 	if [ ! -d '$(DESTDIR)$2' ]; then		\
 		$(INSTALL) -d -m 755 '$(DESTDIR)$2';	\
@@ -107,7 +120,7 @@ define do_s_install
 endef
 
 install: all install_headers install_pkgconfig
-	$(call do_s_install,$(LIBS),$(LIBDIR))
+	$(call do_s_install,$(STATIC_LIBS) $(SHARED_LIBS),$(LIBDIR))
 
 install_headers:
 	$(call do_install,$(HEADERS),$(INCLUDEDIR)/bpf,644)
@@ -121,4 +134,13 @@ install_pkgconfig: $(PC_FILE)
 	$(call do_install,$(PC_FILE),$(LIBDIR)/pkgconfig,644)
 
 clean:
-	rm -f *.o *.a *.so *.so.* *.pc
+	rm -rf *.o *.a *.so *.so.* *.pc $(SHARED_OBJDIR) $(STATIC_OBJDIR)
+
+.PHONY: cscope tags
+cscope:
+	ls *.c *.h > cscope.files
+	cscope -b -q -f cscope.out
+
+tags:
+	rm -f TAGS tags
+	ls *.c *.h | xargs $(TAGS_PROG) -a

src/bpf.c

@@ -32,6 +32,9 @@
 #include "libbpf.h"
 #include "libbpf_internal.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 /*
  * When building perf, unistd.h is overridden. __NR_bpf is
  * required to be defined explicitly.
@@ -95,7 +98,11 @@ int bpf_create_map_xattr(const struct bpf_create_map_attr *create_attr)
 	attr.btf_key_type_id = create_attr->btf_key_type_id;
 	attr.btf_value_type_id = create_attr->btf_value_type_id;
 	attr.map_ifindex = create_attr->map_ifindex;
-	attr.inner_map_fd = create_attr->inner_map_fd;
+	if (attr.map_type == BPF_MAP_TYPE_STRUCT_OPS)
+		attr.btf_vmlinux_value_type_id =
+			create_attr->btf_vmlinux_value_type_id;
+	else
+		attr.inner_map_fd = create_attr->inner_map_fd;
 
 	return sys_bpf(BPF_MAP_CREATE, &attr, sizeof(attr));
 }
@@ -189,7 +196,7 @@ static void *
 alloc_zero_tailing_info(const void *orecord, __u32 cnt,
 			__u32 actual_rec_size, __u32 expected_rec_size)
 {
-	__u64 info_len = actual_rec_size * cnt;
+	__u64 info_len = (__u64)actual_rec_size * cnt;
 	void *info, *nrecord;
 	int i;
 
@@ -228,6 +235,16 @@ int bpf_load_program_xattr(const struct bpf_load_program_attr *load_attr,
 	memset(&attr, 0, sizeof(attr));
 	attr.prog_type = load_attr->prog_type;
 	attr.expected_attach_type = load_attr->expected_attach_type;
+	if (attr.prog_type == BPF_PROG_TYPE_STRUCT_OPS) {
+		attr.attach_btf_id = load_attr->attach_btf_id;
+	} else if (attr.prog_type == BPF_PROG_TYPE_TRACING ||
+		   attr.prog_type == BPF_PROG_TYPE_EXT) {
+		attr.attach_btf_id = load_attr->attach_btf_id;
+		attr.attach_prog_fd = load_attr->attach_prog_fd;
+	} else {
+		attr.prog_ifindex = load_attr->prog_ifindex;
+		attr.kern_version = load_attr->kern_version;
+	}
 	attr.insn_cnt = (__u32)load_attr->insns_cnt;
 	attr.insns = ptr_to_u64(load_attr->insns);
 	attr.license = ptr_to_u64(load_attr->license);
@@ -241,8 +258,6 @@ int bpf_load_program_xattr(const struct bpf_load_program_attr *load_attr,
 		attr.log_size = 0;
 	}
 
-	attr.kern_version = load_attr->kern_version;
-	attr.prog_ifindex = load_attr->prog_ifindex;
 	attr.prog_btf_fd = load_attr->prog_btf_fd;
 	attr.func_info_rec_size = load_attr->func_info_rec_size;
 	attr.func_info_cnt = load_attr->func_info_cnt;
@@ -438,6 +453,64 @@ int bpf_map_freeze(int fd)
 	return sys_bpf(BPF_MAP_FREEZE, &attr, sizeof(attr));
 }
 
+static int bpf_map_batch_common(int cmd, int fd, void  *in_batch,
+				void *out_batch, void *keys, void *values,
+				__u32 *count,
+				const struct bpf_map_batch_opts *opts)
+{
+	union bpf_attr attr;
+	int ret;
+
+	if (!OPTS_VALID(opts, bpf_map_batch_opts))
+		return -EINVAL;
+
+	memset(&attr, 0, sizeof(attr));
+	attr.batch.map_fd = fd;
+	attr.batch.in_batch = ptr_to_u64(in_batch);
+	attr.batch.out_batch = ptr_to_u64(out_batch);
+	attr.batch.keys = ptr_to_u64(keys);
+	attr.batch.values = ptr_to_u64(values);
+	attr.batch.count = *count;
+	attr.batch.elem_flags  = OPTS_GET(opts, elem_flags, 0);
+	attr.batch.flags = OPTS_GET(opts, flags, 0);
+
+	ret = sys_bpf(cmd, &attr, sizeof(attr));
+	*count = attr.batch.count;
+
+	return ret;
+}
+
+int bpf_map_delete_batch(int fd, void *keys, __u32 *count,
+			 const struct bpf_map_batch_opts *opts)
+{
+	return bpf_map_batch_common(BPF_MAP_DELETE_BATCH, fd, NULL,
+				    NULL, keys, NULL, count, opts);
+}
+
+int bpf_map_lookup_batch(int fd, void *in_batch, void *out_batch, void *keys,
+			 void *values, __u32 *count,
+			 const struct bpf_map_batch_opts *opts)
+{
+	return bpf_map_batch_common(BPF_MAP_LOOKUP_BATCH, fd, in_batch,
+				    out_batch, keys, values, count, opts);
+}
+
+int bpf_map_lookup_and_delete_batch(int fd, void *in_batch, void *out_batch,
+				    void *keys, void *values, __u32 *count,
+				    const struct bpf_map_batch_opts *opts)
+{
+	return bpf_map_batch_common(BPF_MAP_LOOKUP_AND_DELETE_BATCH,
+				    fd, in_batch, out_batch, keys, values,
+				    count, opts);
+}
+
+int bpf_map_update_batch(int fd, void *keys, void *values, __u32 *count,
+			 const struct bpf_map_batch_opts *opts)
+{
+	return bpf_map_batch_common(BPF_MAP_UPDATE_BATCH, fd, NULL, NULL,
+				    keys, values, count, opts);
+}
+
 int bpf_obj_pin(int fd, const char *pathname)
 {
 	union bpf_attr attr;
@@ -461,14 +534,29 @@ int bpf_obj_get(const char *pathname)
 
 int bpf_prog_attach(int prog_fd, int target_fd, enum bpf_attach_type type,
 		    unsigned int flags)
+{
+	DECLARE_LIBBPF_OPTS(bpf_prog_attach_opts, opts,
+		.flags = flags,
+	);
+
+	return bpf_prog_attach_xattr(prog_fd, target_fd, type, &opts);
+}
+
+int bpf_prog_attach_xattr(int prog_fd, int target_fd,
+			  enum bpf_attach_type type,
+			  const struct bpf_prog_attach_opts *opts)
 {
 	union bpf_attr attr;
 
+	if (!OPTS_VALID(opts, bpf_prog_attach_opts))
+		return -EINVAL;
+
 	memset(&attr, 0, sizeof(attr));
 	attr.target_fd	   = target_fd;
 	attr.attach_bpf_fd = prog_fd;
 	attr.attach_type   = type;
-	attr.attach_flags  = flags;
+	attr.attach_flags  = OPTS_GET(opts, flags, 0);
+	attr.replace_bpf_fd = OPTS_GET(opts, replace_prog_fd, 0);
 
 	return sys_bpf(BPF_PROG_ATTACH, &attr, sizeof(attr));
 }
@@ -568,7 +656,7 @@ int bpf_prog_test_run_xattr(struct bpf_prog_test_run_attr *test_attr)
 	return ret;
 }
 
-int bpf_prog_get_next_id(__u32 start_id, __u32 *next_id)
+static int bpf_obj_get_next_id(__u32 start_id, __u32 *next_id, int cmd)
 {
 	union bpf_attr attr;
 	int err;
@@ -576,26 +664,26 @@ int bpf_prog_get_next_id(__u32 start_id, __u32 *next_id)
 	memset(&attr, 0, sizeof(attr));
 	attr.start_id = start_id;
 
-	err = sys_bpf(BPF_PROG_GET_NEXT_ID, &attr, sizeof(attr));
+	err = sys_bpf(cmd, &attr, sizeof(attr));
 	if (!err)
 		*next_id = attr.next_id;
 
 	return err;
 }
 
+int bpf_prog_get_next_id(__u32 start_id, __u32 *next_id)
+{
+	return bpf_obj_get_next_id(start_id, next_id, BPF_PROG_GET_NEXT_ID);
+}
+
 int bpf_map_get_next_id(__u32 start_id, __u32 *next_id)
 {
-	union bpf_attr attr;
+	return bpf_obj_get_next_id(start_id, next_id, BPF_MAP_GET_NEXT_ID);
-	int err;
+}
 
-	memset(&attr, 0, sizeof(attr));
+int bpf_btf_get_next_id(__u32 start_id, __u32 *next_id)
-	attr.start_id = start_id;
+{
-
+	return bpf_obj_get_next_id(start_id, next_id, BPF_BTF_GET_NEXT_ID);
-	err = sys_bpf(BPF_MAP_GET_NEXT_ID, &attr, sizeof(attr));
-	if (!err)
-		*next_id = attr.next_id;
-
-	return err;
 }
 
 int bpf_prog_get_fd_by_id(__u32 id)

src/bpf.h

@@ -28,14 +28,12 @@
 #include <stddef.h>
 #include <stdint.h>
 
+#include "libbpf_common.h"
+
 #ifdef __cplusplus
 extern "C" {
 #endif
 
-#ifndef LIBBPF_API
-#define LIBBPF_API __attribute__((visibility("default")))
-#endif
-
 struct bpf_create_map_attr {
 	const char *name;
 	enum bpf_map_type map_type;
@@ -48,7 +46,10 @@ struct bpf_create_map_attr {
 	__u32 btf_key_type_id;
 	__u32 btf_value_type_id;
 	__u32 map_ifindex;
-	__u32 inner_map_fd;
+	union {
+		__u32 inner_map_fd;
+		__u32 btf_vmlinux_value_type_id;
+	};
 };
 
 LIBBPF_API int
@@ -77,8 +78,14 @@ struct bpf_load_program_attr {
 	const struct bpf_insn *insns;
 	size_t insns_cnt;
 	const char *license;
-	__u32 kern_version;
+	union {
-	__u32 prog_ifindex;
+		__u32 kern_version;
+		__u32 attach_prog_fd;
+	};
+	union {
+		__u32 prog_ifindex;
+		__u32 attach_btf_id;
+	};
 	__u32 prog_btf_fd;
 	__u32 func_info_rec_size;
 	const void *func_info;
@@ -120,10 +127,43 @@ LIBBPF_API int bpf_map_lookup_and_delete_elem(int fd, const void *key,
 LIBBPF_API int bpf_map_delete_elem(int fd, const void *key);
 LIBBPF_API int bpf_map_get_next_key(int fd, const void *key, void *next_key);
 LIBBPF_API int bpf_map_freeze(int fd);
+
+struct bpf_map_batch_opts {
+	size_t sz; /* size of this struct for forward/backward compatibility */
+	__u64 elem_flags;
+	__u64 flags;
+};
+#define bpf_map_batch_opts__last_field flags
+
+LIBBPF_API int bpf_map_delete_batch(int fd, void *keys,
+				    __u32 *count,
+				    const struct bpf_map_batch_opts *opts);
+LIBBPF_API int bpf_map_lookup_batch(int fd, void *in_batch, void *out_batch,
+				    void *keys, void *values, __u32 *count,
+				    const struct bpf_map_batch_opts *opts);
+LIBBPF_API int bpf_map_lookup_and_delete_batch(int fd, void *in_batch,
+					void *out_batch, void *keys,
+					void *values, __u32 *count,
+					const struct bpf_map_batch_opts *opts);
+LIBBPF_API int bpf_map_update_batch(int fd, void *keys, void *values,
+				    __u32 *count,
+				    const struct bpf_map_batch_opts *opts);
+
 LIBBPF_API int bpf_obj_pin(int fd, const char *pathname);
 LIBBPF_API int bpf_obj_get(const char *pathname);
+
+struct bpf_prog_attach_opts {
+	size_t sz; /* size of this struct for forward/backward compatibility */
+	unsigned int flags;
+	int replace_prog_fd;
+};
+#define bpf_prog_attach_opts__last_field replace_prog_fd
+
 LIBBPF_API int bpf_prog_attach(int prog_fd, int attachable_fd,
 			       enum bpf_attach_type type, unsigned int flags);
+LIBBPF_API int bpf_prog_attach_xattr(int prog_fd, int attachable_fd,
+				     enum bpf_attach_type type,
+				     const struct bpf_prog_attach_opts *opts);
 LIBBPF_API int bpf_prog_detach(int attachable_fd, enum bpf_attach_type type);
 LIBBPF_API int bpf_prog_detach2(int prog_fd, int attachable_fd,
 				enum bpf_attach_type type);
@@ -156,6 +196,7 @@ LIBBPF_API int bpf_prog_test_run(int prog_fd, int repeat, void *data,
 				 __u32 *retval, __u32 *duration);
 LIBBPF_API int bpf_prog_get_next_id(__u32 start_id, __u32 *next_id);
 LIBBPF_API int bpf_map_get_next_id(__u32 start_id, __u32 *next_id);
+LIBBPF_API int bpf_btf_get_next_id(__u32 start_id, __u32 *next_id);
 LIBBPF_API int bpf_prog_get_fd_by_id(__u32 id);
 LIBBPF_API int bpf_map_get_fd_by_id(__u32 id);
 LIBBPF_API int bpf_btf_get_fd_by_id(__u32 id);

src/bpf_core_read.h

@@ -0,0 +1,263 @@
+/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
+#ifndef __BPF_CORE_READ_H__
+#define __BPF_CORE_READ_H__
+
+/*
+ * enum bpf_field_info_kind is passed as a second argument into
+ * __builtin_preserve_field_info() built-in to get a specific aspect of
+ * a field, captured as a first argument. __builtin_preserve_field_info(field,
+ * info_kind) returns __u32 integer and produces BTF field relocation, which
+ * is understood and processed by libbpf during BPF object loading. See
+ * selftests/bpf for examples.
+ */
+enum bpf_field_info_kind {
+	BPF_FIELD_BYTE_OFFSET = 0,	/* field byte offset */
+	BPF_FIELD_BYTE_SIZE = 1,
+	BPF_FIELD_EXISTS = 2,		/* field existence in target kernel */
+	BPF_FIELD_SIGNED = 3,
+	BPF_FIELD_LSHIFT_U64 = 4,
+	BPF_FIELD_RSHIFT_U64 = 5,
+};
+
+#define __CORE_RELO(src, field, info)					      \
+	__builtin_preserve_field_info((src)->field, BPF_FIELD_##info)
+
+#if __BYTE_ORDER == __LITTLE_ENDIAN
+#define __CORE_BITFIELD_PROBE_READ(dst, src, fld)			      \
+	bpf_probe_read((void *)dst,					      \
+		       __CORE_RELO(src, fld, BYTE_SIZE),		      \
+		       (const void *)src + __CORE_RELO(src, fld, BYTE_OFFSET))
+#else
+/* semantics of LSHIFT_64 assumes loading values into low-ordered bytes, so
+ * for big-endian we need to adjust destination pointer accordingly, based on
+ * field byte size
+ */
+#define __CORE_BITFIELD_PROBE_READ(dst, src, fld)			      \
+	bpf_probe_read((void *)dst + (8 - __CORE_RELO(src, fld, BYTE_SIZE)),  \
+		       __CORE_RELO(src, fld, BYTE_SIZE),		      \
+		       (const void *)src + __CORE_RELO(src, fld, BYTE_OFFSET))
+#endif
+
+/*
+ * Extract bitfield, identified by s->field, and return its value as u64.
+ * All this is done in relocatable manner, so bitfield changes such as
+ * signedness, bit size, offset changes, this will be handled automatically.
+ * This version of macro is using bpf_probe_read() to read underlying integer
+ * storage. Macro functions as an expression and its return type is
+ * bpf_probe_read()'s return value: 0, on success, <0 on error.
+ */
+#define BPF_CORE_READ_BITFIELD_PROBED(s, field) ({			      \
+	unsigned long long val = 0;					      \
+									      \
+	__CORE_BITFIELD_PROBE_READ(&val, s, field);			      \
+	val <<= __CORE_RELO(s, field, LSHIFT_U64);			      \
+	if (__CORE_RELO(s, field, SIGNED))				      \
+		val = ((long long)val) >> __CORE_RELO(s, field, RSHIFT_U64);  \
+	else								      \
+		val = val >> __CORE_RELO(s, field, RSHIFT_U64);		      \
+	val;								      \
+})
+
+/*
+ * Extract bitfield, identified by s->field, and return its value as u64.
+ * This version of macro is using direct memory reads and should be used from
+ * BPF program types that support such functionality (e.g., typed raw
+ * tracepoints).
+ */
+#define BPF_CORE_READ_BITFIELD(s, field) ({				      \
+	const void *p = (const void *)s + __CORE_RELO(s, field, BYTE_OFFSET); \
+	unsigned long long val;						      \
+									      \
+	switch (__CORE_RELO(s, field, BYTE_SIZE)) {			      \
+	case 1: val = *(const unsigned char *)p;			      \
+	case 2: val = *(const unsigned short *)p;			      \
+	case 4: val = *(const unsigned int *)p;				      \
+	case 8: val = *(const unsigned long long *)p;			      \
+	}								      \
+	val <<= __CORE_RELO(s, field, LSHIFT_U64);			      \
+	if (__CORE_RELO(s, field, SIGNED))				      \
+		val = ((long long)val) >> __CORE_RELO(s, field, RSHIFT_U64);  \
+	else								      \
+		val = val >> __CORE_RELO(s, field, RSHIFT_U64);		      \
+	val;								      \
+})
+
+/*
+ * Convenience macro to check that field actually exists in target kernel's.
+ * Returns:
+ *    1, if matching field is present in target kernel;
+ *    0, if no matching field found.
+ */
+#define bpf_core_field_exists(field)					    \
+	__builtin_preserve_field_info(field, BPF_FIELD_EXISTS)
+
+/*
+ * Convenience macro to get byte size of a field. Works for integers,
+ * struct/unions, pointers, arrays, and enums.
+ */
+#define bpf_core_field_size(field)					    \
+	__builtin_preserve_field_info(field, BPF_FIELD_BYTE_SIZE)
+
+/*
+ * bpf_core_read() abstracts away bpf_probe_read() call and captures offset
+ * relocation for source address using __builtin_preserve_access_index()
+ * built-in, provided by Clang.
+ *
+ * __builtin_preserve_access_index() takes as an argument an expression of
+ * taking an address of a field within struct/union. It makes compiler emit
+ * a relocation, which records BTF type ID describing root struct/union and an
+ * accessor string which describes exact embedded field that was used to take
+ * an address. See detailed description of this relocation format and
+ * semantics in comments to struct bpf_field_reloc in libbpf_internal.h.
+ *
+ * This relocation allows libbpf to adjust BPF instruction to use correct
+ * actual field offset, based on target kernel BTF type that matches original
+ * (local) BTF, used to record relocation.
+ */
+#define bpf_core_read(dst, sz, src)					    \
+	bpf_probe_read(dst, sz,						    \
+		       (const void *)__builtin_preserve_access_index(src))
+
+/*
+ * bpf_core_read_str() is a thin wrapper around bpf_probe_read_str()
+ * additionally emitting BPF CO-RE field relocation for specified source
+ * argument.
+ */
+#define bpf_core_read_str(dst, sz, src)					    \
+	bpf_probe_read_str(dst, sz,					    \
+			   (const void *)__builtin_preserve_access_index(src))
+
+#define ___concat(a, b) a ## b
+#define ___apply(fn, n) ___concat(fn, n)
+#define ___nth(_1, _2, _3, _4, _5, _6, _7, _8, _9, _10, __11, N, ...) N
+
+/*
+ * return number of provided arguments; used for switch-based variadic macro
+ * definitions (see ___last, ___arrow, etc below)
+ */
+#define ___narg(...) ___nth(_, ##__VA_ARGS__, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0)
+/*
+ * return 0 if no arguments are passed, N - otherwise; used for
+ * recursively-defined macros to specify termination (0) case, and generic
+ * (N) case (e.g., ___read_ptrs, ___core_read)
+ */
+#define ___empty(...) ___nth(_, ##__VA_ARGS__, N, N, N, N, N, N, N, N, N, N, 0)
+
+#define ___last1(x) x
+#define ___last2(a, x) x
+#define ___last3(a, b, x) x
+#define ___last4(a, b, c, x) x
+#define ___last5(a, b, c, d, x) x
+#define ___last6(a, b, c, d, e, x) x
+#define ___last7(a, b, c, d, e, f, x) x
+#define ___last8(a, b, c, d, e, f, g, x) x
+#define ___last9(a, b, c, d, e, f, g, h, x) x
+#define ___last10(a, b, c, d, e, f, g, h, i, x) x
+#define ___last(...) ___apply(___last, ___narg(__VA_ARGS__))(__VA_ARGS__)
+
+#define ___nolast2(a, _) a
+#define ___nolast3(a, b, _) a, b
+#define ___nolast4(a, b, c, _) a, b, c
+#define ___nolast5(a, b, c, d, _) a, b, c, d
+#define ___nolast6(a, b, c, d, e, _) a, b, c, d, e
+#define ___nolast7(a, b, c, d, e, f, _) a, b, c, d, e, f
+#define ___nolast8(a, b, c, d, e, f, g, _) a, b, c, d, e, f, g
+#define ___nolast9(a, b, c, d, e, f, g, h, _) a, b, c, d, e, f, g, h
+#define ___nolast10(a, b, c, d, e, f, g, h, i, _) a, b, c, d, e, f, g, h, i
+#define ___nolast(...) ___apply(___nolast, ___narg(__VA_ARGS__))(__VA_ARGS__)
+
+#define ___arrow1(a) a
+#define ___arrow2(a, b) a->b
+#define ___arrow3(a, b, c) a->b->c
+#define ___arrow4(a, b, c, d) a->b->c->d
+#define ___arrow5(a, b, c, d, e) a->b->c->d->e
+#define ___arrow6(a, b, c, d, e, f) a->b->c->d->e->f
+#define ___arrow7(a, b, c, d, e, f, g) a->b->c->d->e->f->g
+#define ___arrow8(a, b, c, d, e, f, g, h) a->b->c->d->e->f->g->h
+#define ___arrow9(a, b, c, d, e, f, g, h, i) a->b->c->d->e->f->g->h->i
+#define ___arrow10(a, b, c, d, e, f, g, h, i, j) a->b->c->d->e->f->g->h->i->j
+#define ___arrow(...) ___apply(___arrow, ___narg(__VA_ARGS__))(__VA_ARGS__)
+
+#define ___type(...) typeof(___arrow(__VA_ARGS__))
+
+#define ___read(read_fn, dst, src_type, src, accessor)			    \
+	read_fn((void *)(dst), sizeof(*(dst)), &((src_type)(src))->accessor)
+
+/* "recursively" read a sequence of inner pointers using local __t var */
+#define ___rd_first(src, a) ___read(bpf_core_read, &__t, ___type(src), src, a);
+#define ___rd_last(...)							    \
+	___read(bpf_core_read, &__t,					    \
+		___type(___nolast(__VA_ARGS__)), __t, ___last(__VA_ARGS__));
+#define ___rd_p1(...) const void *__t; ___rd_first(__VA_ARGS__)
+#define ___rd_p2(...) ___rd_p1(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__)
+#define ___rd_p3(...) ___rd_p2(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__)
+#define ___rd_p4(...) ___rd_p3(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__)
+#define ___rd_p5(...) ___rd_p4(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__)
+#define ___rd_p6(...) ___rd_p5(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__)
+#define ___rd_p7(...) ___rd_p6(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__)
+#define ___rd_p8(...) ___rd_p7(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__)
+#define ___rd_p9(...) ___rd_p8(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__)
+#define ___read_ptrs(src, ...)						    \
+	___apply(___rd_p, ___narg(__VA_ARGS__))(src, __VA_ARGS__)
+
+#define ___core_read0(fn, dst, src, a)					    \
+	___read(fn, dst, ___type(src), src, a);
+#define ___core_readN(fn, dst, src, ...)				    \
+	___read_ptrs(src, ___nolast(__VA_ARGS__))			    \
+	___read(fn, dst, ___type(src, ___nolast(__VA_ARGS__)), __t,	    \
+		___last(__VA_ARGS__));
+#define ___core_read(fn, dst, src, a, ...)				    \
+	___apply(___core_read, ___empty(__VA_ARGS__))(fn, dst,		    \
+						      src, a, ##__VA_ARGS__)
+
+/*
+ * BPF_CORE_READ_INTO() is a more performance-conscious variant of
+ * BPF_CORE_READ(), in which final field is read into user-provided storage.
+ * See BPF_CORE_READ() below for more details on general usage.
+ */
+#define BPF_CORE_READ_INTO(dst, src, a, ...)				    \
+	({								    \
+		___core_read(bpf_core_read, dst, src, a, ##__VA_ARGS__)	    \
+	})
+
+/*
+ * BPF_CORE_READ_STR_INTO() does same "pointer chasing" as
+ * BPF_CORE_READ() for intermediate pointers, but then executes (and returns
+ * corresponding error code) bpf_core_read_str() for final string read.
+ */
+#define BPF_CORE_READ_STR_INTO(dst, src, a, ...)			    \
+	({								    \
+		___core_read(bpf_core_read_str, dst, src, a, ##__VA_ARGS__) \
+	})
+
+/*
+ * BPF_CORE_READ() is used to simplify BPF CO-RE relocatable read, especially
+ * when there are few pointer chasing steps.
+ * E.g., what in non-BPF world (or in BPF w/ BCC) would be something like:
+ *	int x = s->a.b.c->d.e->f->g;
+ * can be succinctly achieved using BPF_CORE_READ as:
+ *	int x = BPF_CORE_READ(s, a.b.c, d.e, f, g);
+ *
+ * BPF_CORE_READ will decompose above statement into 4 bpf_core_read (BPF
+ * CO-RE relocatable bpf_probe_read() wrapper) calls, logically equivalent to:
+ * 1. const void *__t = s->a.b.c;
+ * 2. __t = __t->d.e;
+ * 3. __t = __t->f;
+ * 4. return __t->g;
+ *
+ * Equivalence is logical, because there is a heavy type casting/preservation
+ * involved, as well as all the reads are happening through bpf_probe_read()
+ * calls using __builtin_preserve_access_index() to emit CO-RE relocations.
+ *
+ * N.B. Only up to 9 "field accessors" are supported, which should be more
+ * than enough for any practical purpose.
+ */
+#define BPF_CORE_READ(src, a, ...)					    \
+	({								    \
+		___type(src, a, ##__VA_ARGS__) __r;			    \
+		BPF_CORE_READ_INTO(&__r, src, a, ##__VA_ARGS__);	    \
+		__r;							    \
+	})
+
+#endif
+

src/bpf_endian.h

@@ -0,0 +1,72 @@
+/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
+#ifndef __BPF_ENDIAN__
+#define __BPF_ENDIAN__
+
+#include <linux/stddef.h>
+#include <linux/swab.h>
+
+/* LLVM's BPF target selects the endianness of the CPU
+ * it compiles on, or the user specifies (bpfel/bpfeb),
+ * respectively. The used __BYTE_ORDER__ is defined by
+ * the compiler, we cannot rely on __BYTE_ORDER from
+ * libc headers, since it doesn't reflect the actual
+ * requested byte order.
+ *
+ * Note, LLVM's BPF target has different __builtin_bswapX()
+ * semantics. It does map to BPF_ALU | BPF_END | BPF_TO_BE
+ * in bpfel and bpfeb case, which means below, that we map
+ * to cpu_to_be16(). We could use it unconditionally in BPF
+ * case, but better not rely on it, so that this header here
+ * can be used from application and BPF program side, which
+ * use different targets.
+ */
+#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
+# define __bpf_ntohs(x)			__builtin_bswap16(x)
+# define __bpf_htons(x)			__builtin_bswap16(x)
+# define __bpf_constant_ntohs(x)	___constant_swab16(x)
+# define __bpf_constant_htons(x)	___constant_swab16(x)
+# define __bpf_ntohl(x)			__builtin_bswap32(x)
+# define __bpf_htonl(x)			__builtin_bswap32(x)
+# define __bpf_constant_ntohl(x)	___constant_swab32(x)
+# define __bpf_constant_htonl(x)	___constant_swab32(x)
+# define __bpf_be64_to_cpu(x)		__builtin_bswap64(x)
+# define __bpf_cpu_to_be64(x)		__builtin_bswap64(x)
+# define __bpf_constant_be64_to_cpu(x)	___constant_swab64(x)
+# define __bpf_constant_cpu_to_be64(x)	___constant_swab64(x)
+#elif __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
+# define __bpf_ntohs(x)			(x)
+# define __bpf_htons(x)			(x)
+# define __bpf_constant_ntohs(x)	(x)
+# define __bpf_constant_htons(x)	(x)
+# define __bpf_ntohl(x)			(x)
+# define __bpf_htonl(x)			(x)
+# define __bpf_constant_ntohl(x)	(x)
+# define __bpf_constant_htonl(x)	(x)
+# define __bpf_be64_to_cpu(x)		(x)
+# define __bpf_cpu_to_be64(x)		(x)
+# define __bpf_constant_be64_to_cpu(x)  (x)
+# define __bpf_constant_cpu_to_be64(x)  (x)
+#else
+# error "Fix your compiler's __BYTE_ORDER__?!"
+#endif
+
+#define bpf_htons(x)				\
+	(__builtin_constant_p(x) ?		\
+	 __bpf_constant_htons(x) : __bpf_htons(x))
+#define bpf_ntohs(x)				\
+	(__builtin_constant_p(x) ?		\
+	 __bpf_constant_ntohs(x) : __bpf_ntohs(x))
+#define bpf_htonl(x)				\
+	(__builtin_constant_p(x) ?		\
+	 __bpf_constant_htonl(x) : __bpf_htonl(x))
+#define bpf_ntohl(x)				\
+	(__builtin_constant_p(x) ?		\
+	 __bpf_constant_ntohl(x) : __bpf_ntohl(x))
+#define bpf_cpu_to_be64(x)			\
+	(__builtin_constant_p(x) ?		\
+	 __bpf_constant_cpu_to_be64(x) : __bpf_cpu_to_be64(x))
+#define bpf_be64_to_cpu(x)			\
+	(__builtin_constant_p(x) ?		\
+	 __bpf_constant_be64_to_cpu(x) : __bpf_be64_to_cpu(x))
+
+#endif /* __BPF_ENDIAN__ */

src/bpf_helpers.h

@@ -0,0 +1,58 @@
+/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
+#ifndef __BPF_HELPERS__
+#define __BPF_HELPERS__
+
+#include "bpf_helper_defs.h"
+
+#define __uint(name, val) int (*name)[val]
+#define __type(name, val) typeof(val) *name
+
+/* Helper macro to print out debug messages */
+#define bpf_printk(fmt, ...)				\
+({							\
+	char ____fmt[] = fmt;				\
+	bpf_trace_printk(____fmt, sizeof(____fmt),	\
+			 ##__VA_ARGS__);		\
+})
+
+/*
+ * Helper macro to place programs, maps, license in
+ * different sections in elf_bpf file. Section names
+ * are interpreted by elf_bpf loader
+ */
+#define SEC(NAME) __attribute__((section(NAME), used))
+
+#ifndef __always_inline
+#define __always_inline __attribute__((always_inline))
+#endif
+#ifndef __weak
+#define __weak __attribute__((weak))
+#endif
+
+/*
+ * Helper structure used by eBPF C program
+ * to describe BPF map attributes to libbpf loader
+ */
+struct bpf_map_def {
+	unsigned int type;
+	unsigned int key_size;
+	unsigned int value_size;
+	unsigned int max_entries;
+	unsigned int map_flags;
+};
+
+enum libbpf_pin_type {
+	LIBBPF_PIN_NONE,
+	/* PIN_BY_NAME: pin maps by name (in /sys/fs/bpf by default) */
+	LIBBPF_PIN_BY_NAME,
+};
+
+enum libbpf_tristate {
+	TRI_NO = 0,
+	TRI_YES = 1,
+	TRI_MODULE = 2,
+};
+
+#define __kconfig __attribute__((section(".kconfig")))
+
+#endif

src/bpf_prog_linfo.c

@@ -8,6 +8,9 @@
 #include "libbpf.h"
 #include "libbpf_internal.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 struct bpf_prog_linfo {
 	void *raw_linfo;
 	void *raw_jited_linfo;
@@ -101,6 +104,7 @@ struct bpf_prog_linfo *bpf_prog_linfo__new(const struct bpf_prog_info *info)
 {
 	struct bpf_prog_linfo *prog_linfo;
 	__u32 nr_linfo, nr_jited_func;
+	__u64 data_sz;
 
 	nr_linfo = info->nr_line_info;
 
@@ -122,11 +126,11 @@ struct bpf_prog_linfo *bpf_prog_linfo__new(const struct bpf_prog_info *info)
 	/* Copy xlated line_info */
 	prog_linfo->nr_linfo = nr_linfo;
 	prog_linfo->rec_size = info->line_info_rec_size;
-	prog_linfo->raw_linfo = malloc(nr_linfo * prog_linfo->rec_size);
+	data_sz = (__u64)nr_linfo * prog_linfo->rec_size;
+	prog_linfo->raw_linfo = malloc(data_sz);
 	if (!prog_linfo->raw_linfo)
 		goto err_free;
-	memcpy(prog_linfo->raw_linfo, (void *)(long)info->line_info,
+	memcpy(prog_linfo->raw_linfo, (void *)(long)info->line_info, data_sz);
-	       nr_linfo * prog_linfo->rec_size);
 
 	nr_jited_func = info->nr_jited_ksyms;
 	if (!nr_jited_func ||
@@ -142,13 +146,12 @@ struct bpf_prog_linfo *bpf_prog_linfo__new(const struct bpf_prog_info *info)
 	/* Copy jited_line_info */
 	prog_linfo->nr_jited_func = nr_jited_func;
 	prog_linfo->jited_rec_size = info->jited_line_info_rec_size;
-	prog_linfo->raw_jited_linfo = malloc(nr_linfo *
+	data_sz = (__u64)nr_linfo * prog_linfo->jited_rec_size;
-					     prog_linfo->jited_rec_size);
+	prog_linfo->raw_jited_linfo = malloc(data_sz);
 	if (!prog_linfo->raw_jited_linfo)
 		goto err_free;
 	memcpy(prog_linfo->raw_jited_linfo,
-	       (void *)(long)info->jited_line_info,
+	       (void *)(long)info->jited_line_info, data_sz);
-	       nr_linfo * prog_linfo->jited_rec_size);
 
 	/* Number of jited_line_info per jited func */
 	prog_linfo->nr_jited_linfo_per_func = malloc(nr_jited_func *

src/bpf_tracing.h

@@ -0,0 +1,195 @@
+/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
+#ifndef __BPF_TRACING_H__
+#define __BPF_TRACING_H__
+
+/* Scan the ARCH passed in from ARCH env variable (see Makefile) */
+#if defined(__TARGET_ARCH_x86)
+	#define bpf_target_x86
+	#define bpf_target_defined
+#elif defined(__TARGET_ARCH_s390)
+	#define bpf_target_s390
+	#define bpf_target_defined
+#elif defined(__TARGET_ARCH_arm)
+	#define bpf_target_arm
+	#define bpf_target_defined
+#elif defined(__TARGET_ARCH_arm64)
+	#define bpf_target_arm64
+	#define bpf_target_defined
+#elif defined(__TARGET_ARCH_mips)
+	#define bpf_target_mips
+	#define bpf_target_defined
+#elif defined(__TARGET_ARCH_powerpc)
+	#define bpf_target_powerpc
+	#define bpf_target_defined
+#elif defined(__TARGET_ARCH_sparc)
+	#define bpf_target_sparc
+	#define bpf_target_defined
+#else
+	#undef bpf_target_defined
+#endif
+
+/* Fall back to what the compiler says */
+#ifndef bpf_target_defined
+#if defined(__x86_64__)
+	#define bpf_target_x86
+#elif defined(__s390__)
+	#define bpf_target_s390
+#elif defined(__arm__)
+	#define bpf_target_arm
+#elif defined(__aarch64__)
+	#define bpf_target_arm64
+#elif defined(__mips__)
+	#define bpf_target_mips
+#elif defined(__powerpc__)
+	#define bpf_target_powerpc
+#elif defined(__sparc__)
+	#define bpf_target_sparc
+#endif
+#endif
+
+#if defined(bpf_target_x86)
+
+#ifdef __KERNEL__
+#define PT_REGS_PARM1(x) ((x)->di)
+#define PT_REGS_PARM2(x) ((x)->si)
+#define PT_REGS_PARM3(x) ((x)->dx)
+#define PT_REGS_PARM4(x) ((x)->cx)
+#define PT_REGS_PARM5(x) ((x)->r8)
+#define PT_REGS_RET(x) ((x)->sp)
+#define PT_REGS_FP(x) ((x)->bp)
+#define PT_REGS_RC(x) ((x)->ax)
+#define PT_REGS_SP(x) ((x)->sp)
+#define PT_REGS_IP(x) ((x)->ip)
+#else
+#ifdef __i386__
+/* i386 kernel is built with -mregparm=3 */
+#define PT_REGS_PARM1(x) ((x)->eax)
+#define PT_REGS_PARM2(x) ((x)->edx)
+#define PT_REGS_PARM3(x) ((x)->ecx)
+#define PT_REGS_PARM4(x) 0
+#define PT_REGS_PARM5(x) 0
+#define PT_REGS_RET(x) ((x)->esp)
+#define PT_REGS_FP(x) ((x)->ebp)
+#define PT_REGS_RC(x) ((x)->eax)
+#define PT_REGS_SP(x) ((x)->esp)
+#define PT_REGS_IP(x) ((x)->eip)
+#else
+#define PT_REGS_PARM1(x) ((x)->rdi)
+#define PT_REGS_PARM2(x) ((x)->rsi)
+#define PT_REGS_PARM3(x) ((x)->rdx)
+#define PT_REGS_PARM4(x) ((x)->rcx)
+#define PT_REGS_PARM5(x) ((x)->r8)
+#define PT_REGS_RET(x) ((x)->rsp)
+#define PT_REGS_FP(x) ((x)->rbp)
+#define PT_REGS_RC(x) ((x)->rax)
+#define PT_REGS_SP(x) ((x)->rsp)
+#define PT_REGS_IP(x) ((x)->rip)
+#endif
+#endif
+
+#elif defined(bpf_target_s390)
+
+/* s390 provides user_pt_regs instead of struct pt_regs to userspace */
+struct pt_regs;
+#define PT_REGS_S390 const volatile user_pt_regs
+#define PT_REGS_PARM1(x) (((PT_REGS_S390 *)(x))->gprs[2])
+#define PT_REGS_PARM2(x) (((PT_REGS_S390 *)(x))->gprs[3])
+#define PT_REGS_PARM3(x) (((PT_REGS_S390 *)(x))->gprs[4])
+#define PT_REGS_PARM4(x) (((PT_REGS_S390 *)(x))->gprs[5])
+#define PT_REGS_PARM5(x) (((PT_REGS_S390 *)(x))->gprs[6])
+#define PT_REGS_RET(x) (((PT_REGS_S390 *)(x))->gprs[14])
+/* Works only with CONFIG_FRAME_POINTER */
+#define PT_REGS_FP(x) (((PT_REGS_S390 *)(x))->gprs[11])
+#define PT_REGS_RC(x) (((PT_REGS_S390 *)(x))->gprs[2])
+#define PT_REGS_SP(x) (((PT_REGS_S390 *)(x))->gprs[15])
+#define PT_REGS_IP(x) (((PT_REGS_S390 *)(x))->psw.addr)
+
+#elif defined(bpf_target_arm)
+
+#define PT_REGS_PARM1(x) ((x)->uregs[0])
+#define PT_REGS_PARM2(x) ((x)->uregs[1])
+#define PT_REGS_PARM3(x) ((x)->uregs[2])
+#define PT_REGS_PARM4(x) ((x)->uregs[3])
+#define PT_REGS_PARM5(x) ((x)->uregs[4])
+#define PT_REGS_RET(x) ((x)->uregs[14])
+#define PT_REGS_FP(x) ((x)->uregs[11]) /* Works only with CONFIG_FRAME_POINTER */
+#define PT_REGS_RC(x) ((x)->uregs[0])
+#define PT_REGS_SP(x) ((x)->uregs[13])
+#define PT_REGS_IP(x) ((x)->uregs[12])
+
+#elif defined(bpf_target_arm64)
+
+/* arm64 provides struct user_pt_regs instead of struct pt_regs to userspace */
+struct pt_regs;
+#define PT_REGS_ARM64 const volatile struct user_pt_regs
+#define PT_REGS_PARM1(x) (((PT_REGS_ARM64 *)(x))->regs[0])
+#define PT_REGS_PARM2(x) (((PT_REGS_ARM64 *)(x))->regs[1])
+#define PT_REGS_PARM3(x) (((PT_REGS_ARM64 *)(x))->regs[2])
+#define PT_REGS_PARM4(x) (((PT_REGS_ARM64 *)(x))->regs[3])
+#define PT_REGS_PARM5(x) (((PT_REGS_ARM64 *)(x))->regs[4])
+#define PT_REGS_RET(x) (((PT_REGS_ARM64 *)(x))->regs[30])
+/* Works only with CONFIG_FRAME_POINTER */
+#define PT_REGS_FP(x) (((PT_REGS_ARM64 *)(x))->regs[29])
+#define PT_REGS_RC(x) (((PT_REGS_ARM64 *)(x))->regs[0])
+#define PT_REGS_SP(x) (((PT_REGS_ARM64 *)(x))->sp)
+#define PT_REGS_IP(x) (((PT_REGS_ARM64 *)(x))->pc)
+
+#elif defined(bpf_target_mips)
+
+#define PT_REGS_PARM1(x) ((x)->regs[4])
+#define PT_REGS_PARM2(x) ((x)->regs[5])
+#define PT_REGS_PARM3(x) ((x)->regs[6])
+#define PT_REGS_PARM4(x) ((x)->regs[7])
+#define PT_REGS_PARM5(x) ((x)->regs[8])
+#define PT_REGS_RET(x) ((x)->regs[31])
+#define PT_REGS_FP(x) ((x)->regs[30]) /* Works only with CONFIG_FRAME_POINTER */
+#define PT_REGS_RC(x) ((x)->regs[1])
+#define PT_REGS_SP(x) ((x)->regs[29])
+#define PT_REGS_IP(x) ((x)->cp0_epc)
+
+#elif defined(bpf_target_powerpc)
+
+#define PT_REGS_PARM1(x) ((x)->gpr[3])
+#define PT_REGS_PARM2(x) ((x)->gpr[4])
+#define PT_REGS_PARM3(x) ((x)->gpr[5])
+#define PT_REGS_PARM4(x) ((x)->gpr[6])
+#define PT_REGS_PARM5(x) ((x)->gpr[7])
+#define PT_REGS_RC(x) ((x)->gpr[3])
+#define PT_REGS_SP(x) ((x)->sp)
+#define PT_REGS_IP(x) ((x)->nip)
+
+#elif defined(bpf_target_sparc)
+
+#define PT_REGS_PARM1(x) ((x)->u_regs[UREG_I0])
+#define PT_REGS_PARM2(x) ((x)->u_regs[UREG_I1])
+#define PT_REGS_PARM3(x) ((x)->u_regs[UREG_I2])
+#define PT_REGS_PARM4(x) ((x)->u_regs[UREG_I3])
+#define PT_REGS_PARM5(x) ((x)->u_regs[UREG_I4])
+#define PT_REGS_RET(x) ((x)->u_regs[UREG_I7])
+#define PT_REGS_RC(x) ((x)->u_regs[UREG_I0])
+#define PT_REGS_SP(x) ((x)->u_regs[UREG_FP])
+
+/* Should this also be a bpf_target check for the sparc case? */
+#if defined(__arch64__)
+#define PT_REGS_IP(x) ((x)->tpc)
+#else
+#define PT_REGS_IP(x) ((x)->pc)
+#endif
+
+#endif
+
+#if defined(bpf_target_powerpc)
+#define BPF_KPROBE_READ_RET_IP(ip, ctx)		({ (ip) = (ctx)->link; })
+#define BPF_KRETPROBE_READ_RET_IP		BPF_KPROBE_READ_RET_IP
+#elif defined(bpf_target_sparc)
+#define BPF_KPROBE_READ_RET_IP(ip, ctx)		({ (ip) = PT_REGS_RET(ctx); })
+#define BPF_KRETPROBE_READ_RET_IP		BPF_KPROBE_READ_RET_IP
+#else
+#define BPF_KPROBE_READ_RET_IP(ip, ctx)					    \
+	({ bpf_probe_read(&(ip), sizeof(ip), (void *)PT_REGS_RET(ctx)); })
+#define BPF_KRETPROBE_READ_RET_IP(ip, ctx)				    \
+	({ bpf_probe_read(&(ip), sizeof(ip),				    \
+			  (void *)(PT_REGS_FP(ctx) + sizeof(ip))); })
+#endif
+
+#endif

src/btf.c

@@ -1,12 +1,17 @@
 // SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause)
 /* Copyright (c) 2018 Facebook */
 
+#include <endian.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
 #include <fcntl.h>
 #include <unistd.h>
 #include <errno.h>
+#include <sys/utsname.h>
+#include <sys/param.h>
+#include <sys/stat.h>
+#include <linux/kernel.h>
 #include <linux/err.h>
 #include <linux/btf.h>
 #include <gelf.h>
@@ -16,15 +21,11 @@
 #include "libbpf_internal.h"
 #include "hashmap.h"
 
-#define BTF_MAX_NR_TYPES 0x7fffffff
+/* make sure libbpf doesn't use kernel-only integer typedefs */
-#define BTF_MAX_STR_OFFSET 0x7fffffff
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
 
-#define IS_MODIFIER(k) (((k) == BTF_KIND_TYPEDEF) || \
+#define BTF_MAX_NR_TYPES 0x7fffffffU
-		((k) == BTF_KIND_VOLATILE) || \
+#define BTF_MAX_STR_OFFSET 0x7fffffffU
-		((k) == BTF_KIND_CONST) || \
-		((k) == BTF_KIND_RESTRICT))
-
-#define IS_VAR(k) ((k) == BTF_KIND_VAR)
 
 static struct btf_type btf_void;
 
@@ -42,47 +43,6 @@ struct btf {
 	int fd;
 };
 
-struct btf_ext_info {
-	/*
-	 * info points to the individual info section (e.g. func_info and
-	 * line_info) from the .BTF.ext. It does not include the __u32 rec_size.
-	 */
-	void *info;
-	__u32 rec_size;
-	__u32 len;
-};
-
-struct btf_ext {
-	union {
-		struct btf_ext_header *hdr;
-		void *data;
-	};
-	struct btf_ext_info func_info;
-	struct btf_ext_info line_info;
-	__u32 data_size;
-};
-
-struct btf_ext_info_sec {
-	__u32	sec_name_off;
-	__u32	num_info;
-	/* Followed by num_info * record_size number of bytes */
-	__u8	data[0];
-};
-
-/* The minimum bpf_func_info checked by the loader */
-struct bpf_func_info_min {
-	__u32   insn_off;
-	__u32   type_id;
-};
-
-/* The minimum bpf_line_info checked by the loader */
-struct bpf_line_info_min {
-	__u32	insn_off;
-	__u32	file_name_off;
-	__u32	line_off;
-	__u32	line_col;
-};
-
 static inline __u64 ptr_to_u64(const void *ptr)
 {
 	return (__u64) (unsigned long) ptr;
@@ -97,7 +57,7 @@ static int btf_add_type(struct btf *btf, struct btf_type *t)
 		if (btf->types_size == BTF_MAX_NR_TYPES)
 			return -E2BIG;
 
-		expand_by = max(btf->types_size >> 2, 16);
+		expand_by = max(btf->types_size >> 2, 16U);
 		new_size = min(BTF_MAX_NR_TYPES, btf->types_size + expand_by);
 
 		new_types = realloc(btf->types, sizeof(*new_types) * new_size);
@@ -192,9 +152,9 @@ static int btf_parse_str_sec(struct btf *btf)
 static int btf_type_size(struct btf_type *t)
 {
 	int base_size = sizeof(struct btf_type);
-	__u16 vlen = BTF_INFO_VLEN(t->info);
+	__u16 vlen = btf_vlen(t);
 
-	switch (BTF_INFO_KIND(t->info)) {
+	switch (btf_kind(t)) {
 	case BTF_KIND_FWD:
 	case BTF_KIND_CONST:
 	case BTF_KIND_VOLATILE:
@@ -219,7 +179,7 @@ static int btf_type_size(struct btf_type *t)
 	case BTF_KIND_DATASEC:
 		return base_size + vlen * sizeof(struct btf_var_secinfo);
 	default:
-		pr_debug("Unsupported BTF_KIND:%u\n", BTF_INFO_KIND(t->info));
+		pr_debug("Unsupported BTF_KIND:%u\n", btf_kind(t));
 		return -EINVAL;
 	}
 }
@@ -263,7 +223,7 @@ const struct btf_type *btf__type_by_id(const struct btf *btf, __u32 type_id)
 
 static bool btf_type_is_void(const struct btf_type *t)
 {
-	return t == &btf_void || BTF_INFO_KIND(t->info) == BTF_KIND_FWD;
+	return t == &btf_void || btf_is_fwd(t);
 }
 
 static bool btf_type_is_void_or_null(const struct btf_type *t)
@@ -284,7 +244,7 @@ __s64 btf__resolve_size(const struct btf *btf, __u32 type_id)
 	t = btf__type_by_id(btf, type_id);
 	for (i = 0; i < MAX_RESOLVE_DEPTH && !btf_type_is_void_or_null(t);
 	     i++) {
-		switch (BTF_INFO_KIND(t->info)) {
+		switch (btf_kind(t)) {
 		case BTF_KIND_INT:
 		case BTF_KIND_STRUCT:
 		case BTF_KIND_UNION:
@@ -303,7 +263,7 @@ __s64 btf__resolve_size(const struct btf *btf, __u32 type_id)
 			type_id = t->type;
 			break;
 		case BTF_KIND_ARRAY:
-			array = (const struct btf_array *)(t + 1);
+			array = btf_array(t);
 			if (nelems && array->nelems > UINT32_MAX / nelems)
 				return -E2BIG;
 			nelems *= array->nelems;
@@ -316,16 +276,54 @@ __s64 btf__resolve_size(const struct btf *btf, __u32 type_id)
 		t = btf__type_by_id(btf, type_id);
 	}
 
+done:
 	if (size < 0)
 		return -EINVAL;
-
-done:
 	if (nelems && size > UINT32_MAX / nelems)
 		return -E2BIG;
 
 	return nelems * size;
 }
 
+int btf__align_of(const struct btf *btf, __u32 id)
+{
+	const struct btf_type *t = btf__type_by_id(btf, id);
+	__u16 kind = btf_kind(t);
+
+	switch (kind) {
+	case BTF_KIND_INT:
+	case BTF_KIND_ENUM:
+		return min(sizeof(void *), (size_t)t->size);
+	case BTF_KIND_PTR:
+		return sizeof(void *);
+	case BTF_KIND_TYPEDEF:
+	case BTF_KIND_VOLATILE:
+	case BTF_KIND_CONST:
+	case BTF_KIND_RESTRICT:
+		return btf__align_of(btf, t->type);
+	case BTF_KIND_ARRAY:
+		return btf__align_of(btf, btf_array(t)->type);
+	case BTF_KIND_STRUCT:
+	case BTF_KIND_UNION: {
+		const struct btf_member *m = btf_members(t);
+		__u16 vlen = btf_vlen(t);
+		int i, max_align = 1, align;
+
+		for (i = 0; i < vlen; i++, m++) {
+			align = btf__align_of(btf, m->type);
+			if (align <= 0)
+				return align;
+			max_align = max(max_align, align);
+		}
+
+		return max_align;
+	}
+	default:
+		pr_warn("unsupported BTF_KIND:%u\n", btf_kind(t));
+		return 0;
+	}
+}
+
 int btf__resolve_type(const struct btf *btf, __u32 type_id)
 {
 	const struct btf_type *t;
@@ -334,8 +332,7 @@ int btf__resolve_type(const struct btf *btf, __u32 type_id)
 	t = btf__type_by_id(btf, type_id);
 	while (depth < MAX_RESOLVE_DEPTH &&
 	       !btf_type_is_void_or_null(t) &&
-	       (IS_MODIFIER(BTF_INFO_KIND(t->info)) ||
+	       (btf_is_mod(t) || btf_is_typedef(t) || btf_is_var(t))) {
-		IS_VAR(BTF_INFO_KIND(t->info)))) {
 		type_id = t->type;
 		t = btf__type_by_id(btf, type_id);
 		depth++;
@@ -365,6 +362,28 @@ __s32 btf__find_by_name(const struct btf *btf, const char *type_name)
 	return -ENOENT;
 }
 
+__s32 btf__find_by_name_kind(const struct btf *btf, const char *type_name,
+			     __u32 kind)
+{
+	__u32 i;
+
+	if (kind == BTF_KIND_UNKN || !strcmp(type_name, "void"))
+		return 0;
+
+	for (i = 1; i <= btf->nr_types; i++) {
+		const struct btf_type *t = btf->types[i];
+		const char *name;
+
+		if (btf_kind(t) != kind)
+			continue;
+		name = btf__name_by_offset(btf, t->name_off);
+		if (name && !strcmp(type_name, name))
+			return i;
+	}
+
+	return -ENOENT;
+}
+
 void btf__free(struct btf *btf)
 {
 	if (!btf)
@@ -419,9 +438,9 @@ done:
 
 static bool btf_check_endianness(const GElf_Ehdr *ehdr)
 {
-#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
+#if __BYTE_ORDER == __LITTLE_ENDIAN
 	return ehdr->e_ident[EI_DATA] == ELFDATA2LSB;
-#elif __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
+#elif __BYTE_ORDER == __BIG_ENDIAN
 	return ehdr->e_ident[EI_DATA] == ELFDATA2MSB;
 #else
 # error "Unrecognized __BYTE_ORDER__"
@@ -438,14 +457,14 @@ struct btf *btf__parse_elf(const char *path, struct btf_ext **btf_ext)
 	GElf_Ehdr ehdr;
 
 	if (elf_version(EV_CURRENT) == EV_NONE) {
-		pr_warning("failed to init libelf for %s\n", path);
+		pr_warn("failed to init libelf for %s\n", path);
 		return ERR_PTR(-LIBBPF_ERRNO__LIBELF);
 	}
 
 	fd = open(path, O_RDONLY);
 	if (fd < 0) {
 		err = -errno;
-		pr_warning("failed to open %s: %s\n", path, strerror(errno));
+		pr_warn("failed to open %s: %s\n", path, strerror(errno));
 		return ERR_PTR(err);
 	}
 
@@ -453,19 +472,19 @@ struct btf *btf__parse_elf(const char *path, struct btf_ext **btf_ext)
 
 	elf = elf_begin(fd, ELF_C_READ, NULL);
 	if (!elf) {
-		pr_warning("failed to open %s as ELF file\n", path);
+		pr_warn("failed to open %s as ELF file\n", path);
 		goto done;
 	}
 	if (!gelf_getehdr(elf, &ehdr)) {
-		pr_warning("failed to get EHDR from %s\n", path);
+		pr_warn("failed to get EHDR from %s\n", path);
 		goto done;
 	}
 	if (!btf_check_endianness(&ehdr)) {
-		pr_warning("non-native ELF endianness is not supported\n");
+		pr_warn("non-native ELF endianness is not supported\n");
 		goto done;
 	}
 	if (!elf_rawdata(elf_getscn(elf, ehdr.e_shstrndx), NULL)) {
-		pr_warning("failed to get e_shstrndx from %s\n", path);
+		pr_warn("failed to get e_shstrndx from %s\n", path);
 		goto done;
 	}
 
@@ -475,29 +494,29 @@ struct btf *btf__parse_elf(const char *path, struct btf_ext **btf_ext)
 
 		idx++;
 		if (gelf_getshdr(scn, &sh) != &sh) {
-			pr_warning("failed to get section(%d) header from %s\n",
+			pr_warn("failed to get section(%d) header from %s\n",
-				   idx, path);
+				idx, path);
 			goto done;
 		}
 		name = elf_strptr(elf, ehdr.e_shstrndx, sh.sh_name);
 		if (!name) {
-			pr_warning("failed to get section(%d) name from %s\n",
+			pr_warn("failed to get section(%d) name from %s\n",
-				   idx, path);
+				idx, path);
 			goto done;
 		}
 		if (strcmp(name, BTF_ELF_SEC) == 0) {
 			btf_data = elf_getdata(scn, 0);
 			if (!btf_data) {
-				pr_warning("failed to get section(%d, %s) data from %s\n",
+				pr_warn("failed to get section(%d, %s) data from %s\n",
-					   idx, name, path);
+					idx, name, path);
 				goto done;
 			}
 			continue;
 		} else if (btf_ext && strcmp(name, BTF_EXT_ELF_SEC) == 0) {
 			btf_ext_data = elf_getdata(scn, 0);
 			if (!btf_ext_data) {
-				pr_warning("failed to get section(%d, %s) data from %s\n",
+				pr_warn("failed to get section(%d, %s) data from %s\n",
-					   idx, name, path);
+					idx, name, path);
 				goto done;
 			}
 			continue;
@@ -554,11 +573,11 @@ static int compare_vsi_off(const void *_a, const void *_b)
 static int btf_fixup_datasec(struct bpf_object *obj, struct btf *btf,
 			     struct btf_type *t)
 {
-	__u32 size = 0, off = 0, i, vars = BTF_INFO_VLEN(t->info);
+	__u32 size = 0, off = 0, i, vars = btf_vlen(t);
 	const char *name = btf__name_by_offset(btf, t->name_off);
 	const struct btf_type *t_var;
 	struct btf_var_secinfo *vsi;
-	struct btf_var *var;
+	const struct btf_var *var;
 	int ret;
 
 	if (!name) {
@@ -566,6 +585,12 @@ static int btf_fixup_datasec(struct bpf_object *obj, struct btf *btf,
 		return -ENOENT;
 	}
 
+	/* .extern datasec size and var offsets were set correctly during
+	 * extern collection step, so just skip straight to sorting variables
+	 */
+	if (t->size)
+		goto sort_vars;
+
 	ret = bpf_object__section_size(obj, name, &size);
 	if (ret || !size || (t->size && t->size != size)) {
 		pr_debug("Invalid size for section %s: %u bytes\n", name, size);
@@ -574,12 +599,11 @@ static int btf_fixup_datasec(struct bpf_object *obj, struct btf *btf,
 
 	t->size = size;
 
-	for (i = 0, vsi = (struct btf_var_secinfo *)(t + 1);
+	for (i = 0, vsi = btf_var_secinfos(t); i < vars; i++, vsi++) {
-	     i < vars; i++, vsi++) {
 		t_var = btf__type_by_id(btf, vsi->type);
-		var = (struct btf_var *)(t_var + 1);
+		var = btf_var(t_var);
 
-		if (BTF_INFO_KIND(t_var->info) != BTF_KIND_VAR) {
+		if (!btf_is_var(t_var)) {
 			pr_debug("Non-VAR type seen in section %s\n", name);
 			return -EINVAL;
 		}
@@ -595,14 +619,16 @@ static int btf_fixup_datasec(struct bpf_object *obj, struct btf *btf,
 
 		ret = bpf_object__variable_offset(obj, name, &off);
 		if (ret) {
-			pr_debug("No offset found in symbol table for VAR %s\n", name);
+			pr_debug("No offset found in symbol table for VAR %s\n",
+				 name);
 			return -ENOENT;
 		}
 
 		vsi->offset = off;
 	}
 
-	qsort(t + 1, vars, sizeof(*vsi), compare_vsi_off);
+sort_vars:
+	qsort(btf_var_secinfos(t), vars, sizeof(*vsi), compare_vsi_off);
 	return 0;
 }
 
@@ -619,7 +645,7 @@ int btf__finalize_data(struct bpf_object *obj, struct btf *btf)
 		 * is section size and global variable offset. We use
 		 * the info from the ELF itself for this purpose.
 		 */
-		if (BTF_INFO_KIND(t->info) == BTF_KIND_DATASEC) {
+		if (btf_is_datasec(t)) {
 			err = btf_fixup_datasec(obj, btf, t);
 			if (err)
 				break;
@@ -648,9 +674,9 @@ int btf__load(struct btf *btf)
 			       log_buf, log_buf_size, false);
 	if (btf->fd < 0) {
 		err = -errno;
-		pr_warning("Error loading BTF: %s(%d)\n", strerror(errno), errno);
+		pr_warn("Error loading BTF: %s(%d)\n", strerror(errno), errno);
 		if (*log_buf)
-			pr_warning("%s\n", log_buf);
+			pr_warn("%s\n", log_buf);
 		goto done;
 	}
 
@@ -755,8 +781,8 @@ int btf__get_map_kv_tids(const struct btf *btf, const char *map_name,
 
 	if (snprintf(container_name, max_name, "____btf_map_%s", map_name) ==
 	    max_name) {
-		pr_warning("map:%s length of '____btf_map_%s' is too long\n",
+		pr_warn("map:%s length of '____btf_map_%s' is too long\n",
-			   map_name, map_name);
+			map_name, map_name);
 		return -EINVAL;
 	}
 
@@ -769,42 +795,41 @@ int btf__get_map_kv_tids(const struct btf *btf, const char *map_name,
 
 	container_type = btf__type_by_id(btf, container_id);
 	if (!container_type) {
-		pr_warning("map:%s cannot find BTF type for container_id:%u\n",
+		pr_warn("map:%s cannot find BTF type for container_id:%u\n",
-			   map_name, container_id);
+			map_name, container_id);
 		return -EINVAL;
 	}
 
-	if (BTF_INFO_KIND(container_type->info) != BTF_KIND_STRUCT ||
+	if (!btf_is_struct(container_type) || btf_vlen(container_type) < 2) {
-	    BTF_INFO_VLEN(container_type->info) < 2) {
+		pr_warn("map:%s container_name:%s is an invalid container struct\n",
-		pr_warning("map:%s container_name:%s is an invalid container struct\n",
+			map_name, container_name);
-			   map_name, container_name);
 		return -EINVAL;
 	}
 
-	key = (struct btf_member *)(container_type + 1);
+	key = btf_members(container_type);
 	value = key + 1;
 
 	key_size = btf__resolve_size(btf, key->type);
 	if (key_size < 0) {
-		pr_warning("map:%s invalid BTF key_type_size\n", map_name);
+		pr_warn("map:%s invalid BTF key_type_size\n", map_name);
 		return key_size;
 	}
 
 	if (expected_key_size != key_size) {
-		pr_warning("map:%s btf_key_type_size:%u != map_def_key_size:%u\n",
+		pr_warn("map:%s btf_key_type_size:%u != map_def_key_size:%u\n",
-			   map_name, (__u32)key_size, expected_key_size);
+			map_name, (__u32)key_size, expected_key_size);
 		return -EINVAL;
 	}
 
 	value_size = btf__resolve_size(btf, value->type);
 	if (value_size < 0) {
-		pr_warning("map:%s invalid BTF value_type_size\n", map_name);
+		pr_warn("map:%s invalid BTF value_type_size\n", map_name);
 		return value_size;
 	}
 
 	if (expected_value_size != value_size) {
-		pr_warning("map:%s btf_value_type_size:%u != map_def_value_size:%u\n",
+		pr_warn("map:%s btf_value_type_size:%u != map_def_value_size:%u\n",
-			   map_name, (__u32)value_size, expected_value_size);
+			map_name, (__u32)value_size, expected_value_size);
 		return -EINVAL;
 	}
 
@@ -831,6 +856,9 @@ static int btf_ext_setup_info(struct btf_ext *btf_ext,
 	/* The start of the info sec (including the __u32 record_size). */
 	void *info;
 
+	if (ext_sec->len == 0)
+		return 0;
+
 	if (ext_sec->off & 0x03) {
 		pr_debug(".BTF.ext %s section is not aligned to 4 bytes\n",
 		     ext_sec->desc);
@@ -934,11 +962,24 @@ static int btf_ext_setup_line_info(struct btf_ext *btf_ext)
 	return btf_ext_setup_info(btf_ext, &param);
 }
 
+static int btf_ext_setup_field_reloc(struct btf_ext *btf_ext)
+{
+	struct btf_ext_sec_setup_param param = {
+		.off = btf_ext->hdr->field_reloc_off,
+		.len = btf_ext->hdr->field_reloc_len,
+		.min_rec_size = sizeof(struct bpf_field_reloc),
+		.ext_info = &btf_ext->field_reloc_info,
+		.desc = "field_reloc",
+	};
+
+	return btf_ext_setup_info(btf_ext, &param);
+}
+
 static int btf_ext_parse_hdr(__u8 *data, __u32 data_size)
 {
 	const struct btf_ext_header *hdr = (struct btf_ext_header *)data;
 
-	if (data_size < offsetof(struct btf_ext_header, func_info_off) ||
+	if (data_size < offsetofend(struct btf_ext_header, hdr_len) ||
 	    data_size < hdr->hdr_len) {
 		pr_debug("BTF.ext header not found");
 		return -EINVAL;
@@ -996,6 +1037,9 @@ struct btf_ext *btf_ext__new(__u8 *data, __u32 size)
 	}
 	memcpy(btf_ext->data, data, size);
 
+	if (btf_ext->hdr->hdr_len <
+	    offsetofend(struct btf_ext_header, line_info_len))
+		goto done;
 	err = btf_ext_setup_func_info(btf_ext);
 	if (err)
 		goto done;
@@ -1004,6 +1048,13 @@ struct btf_ext *btf_ext__new(__u8 *data, __u32 size)
 	if (err)
 		goto done;
 
+	if (btf_ext->hdr->hdr_len <
+	    offsetofend(struct btf_ext_header, field_reloc_len))
+		goto done;
+	err = btf_ext_setup_field_reloc(btf_ext);
+	if (err)
+		goto done;
+
 done:
 	if (err) {
 		btf_ext__free(btf_ext);
@@ -1354,7 +1405,7 @@ static int btf_dedup_hypot_map_add(struct btf_dedup *d,
 	if (d->hypot_cnt == d->hypot_cap) {
 		__u32 *new_list;
 
-		d->hypot_cap += max(16, d->hypot_cap / 2);
+		d->hypot_cap += max((size_t)16, d->hypot_cap / 2);
 		new_list = realloc(d->hypot_list, sizeof(__u32) * d->hypot_cap);
 		if (!new_list)
 			return -ENOMEM;
@@ -1440,10 +1491,9 @@ static struct btf_dedup *btf_dedup_new(struct btf *btf, struct btf_ext *btf_ext,
 	d->map[0] = 0;
 	for (i = 1; i <= btf->nr_types; i++) {
 		struct btf_type *t = d->btf->types[i];
-		__u16 kind = BTF_INFO_KIND(t->info);
 
 		/* VAR and DATASEC are never deduped and are self-canonical */
-		if (kind == BTF_KIND_VAR || kind == BTF_KIND_DATASEC)
+		if (btf_is_var(t) || btf_is_datasec(t))
 			d->map[i] = i;
 		else
 			d->map[i] = BTF_UNPROCESSED_ID;
@@ -1484,11 +1534,11 @@ static int btf_for_each_str_off(struct btf_dedup *d, str_off_fn_t fn, void *ctx)
 		if (r)
 			return r;
 
-		switch (BTF_INFO_KIND(t->info)) {
+		switch (btf_kind(t)) {
 		case BTF_KIND_STRUCT:
 		case BTF_KIND_UNION: {
-			struct btf_member *m = (struct btf_member *)(t + 1);
+			struct btf_member *m = btf_members(t);
-			__u16 vlen = BTF_INFO_VLEN(t->info);
+			__u16 vlen = btf_vlen(t);
 
 			for (j = 0; j < vlen; j++) {
 				r = fn(&m->name_off, ctx);
@@ -1499,8 +1549,8 @@ static int btf_for_each_str_off(struct btf_dedup *d, str_off_fn_t fn, void *ctx)
 			break;
 		}
 		case BTF_KIND_ENUM: {
-			struct btf_enum *m = (struct btf_enum *)(t + 1);
+			struct btf_enum *m = btf_enum(t);
-			__u16 vlen = BTF_INFO_VLEN(t->info);
+			__u16 vlen = btf_vlen(t);
 
 			for (j = 0; j < vlen; j++) {
 				r = fn(&m->name_off, ctx);
@@ -1511,8 +1561,8 @@ static int btf_for_each_str_off(struct btf_dedup *d, str_off_fn_t fn, void *ctx)
 			break;
 		}
 		case BTF_KIND_FUNC_PROTO: {
-			struct btf_param *m = (struct btf_param *)(t + 1);
+			struct btf_param *m = btf_params(t);
-			__u16 vlen = BTF_INFO_VLEN(t->info);
+			__u16 vlen = btf_vlen(t);
 
 			for (j = 0; j < vlen; j++) {
 				r = fn(&m->name_off, ctx);
@@ -1651,7 +1701,7 @@ static int btf_dedup_strings(struct btf_dedup *d)
 		if (strs.cnt + 1 > strs.cap) {
 			struct btf_str_ptr *new_ptrs;
 
-			strs.cap += max(strs.cnt / 2, 16);
+			strs.cap += max(strs.cnt / 2, 16U);
 			new_ptrs = realloc(strs.ptrs,
 					   sizeof(strs.ptrs[0]) * strs.cap);
 			if (!new_ptrs) {
@@ -1801,16 +1851,16 @@ static long btf_hash_enum(struct btf_type *t)
 /* Check structural equality of two ENUMs. */
 static bool btf_equal_enum(struct btf_type *t1, struct btf_type *t2)
 {
-	struct btf_enum *m1, *m2;
+	const struct btf_enum *m1, *m2;
 	__u16 vlen;
 	int i;
 
 	if (!btf_equal_common(t1, t2))
 		return false;
 
-	vlen = BTF_INFO_VLEN(t1->info);
+	vlen = btf_vlen(t1);
-	m1 = (struct btf_enum *)(t1 + 1);
+	m1 = btf_enum(t1);
-	m2 = (struct btf_enum *)(t2 + 1);
+	m2 = btf_enum(t2);
 	for (i = 0; i < vlen; i++) {
 		if (m1->name_off != m2->name_off || m1->val != m2->val)
 			return false;
@@ -1822,8 +1872,7 @@ static bool btf_equal_enum(struct btf_type *t1, struct btf_type *t2)
 
 static inline bool btf_is_enum_fwd(struct btf_type *t)
 {
-	return BTF_INFO_KIND(t->info) == BTF_KIND_ENUM &&
+	return btf_is_enum(t) && btf_vlen(t) == 0;
-	       BTF_INFO_VLEN(t->info) == 0;
 }
 
 static bool btf_compat_enum(struct btf_type *t1, struct btf_type *t2)
@@ -1843,8 +1892,8 @@ static bool btf_compat_enum(struct btf_type *t1, struct btf_type *t2)
  */
 static long btf_hash_struct(struct btf_type *t)
 {
-	struct btf_member *member = (struct btf_member *)(t + 1);
+	const struct btf_member *member = btf_members(t);
-	__u32 vlen = BTF_INFO_VLEN(t->info);
+	__u32 vlen = btf_vlen(t);
 	long h = btf_hash_common(t);
 	int i;
 
@@ -1864,16 +1913,16 @@ static long btf_hash_struct(struct btf_type *t)
  */
 static bool btf_shallow_equal_struct(struct btf_type *t1, struct btf_type *t2)
 {
-	struct btf_member *m1, *m2;
+	const struct btf_member *m1, *m2;
 	__u16 vlen;
 	int i;
 
 	if (!btf_equal_common(t1, t2))
 		return false;
 
-	vlen = BTF_INFO_VLEN(t1->info);
+	vlen = btf_vlen(t1);
-	m1 = (struct btf_member *)(t1 + 1);
+	m1 = btf_members(t1);
-	m2 = (struct btf_member *)(t2 + 1);
+	m2 = btf_members(t2);
 	for (i = 0; i < vlen; i++) {
 		if (m1->name_off != m2->name_off || m1->offset != m2->offset)
 			return false;
@@ -1890,7 +1939,7 @@ static bool btf_shallow_equal_struct(struct btf_type *t1, struct btf_type *t2)
  */
 static long btf_hash_array(struct btf_type *t)
 {
-	struct btf_array *info = (struct btf_array *)(t + 1);
+	const struct btf_array *info = btf_array(t);
 	long h = btf_hash_common(t);
 
 	h = hash_combine(h, info->type);
@@ -1908,13 +1957,13 @@ static long btf_hash_array(struct btf_type *t)
  */
 static bool btf_equal_array(struct btf_type *t1, struct btf_type *t2)
 {
-	struct btf_array *info1, *info2;
+	const struct btf_array *info1, *info2;
 
 	if (!btf_equal_common(t1, t2))
 		return false;
 
-	info1 = (struct btf_array *)(t1 + 1);
+	info1 = btf_array(t1);
-	info2 = (struct btf_array *)(t2 + 1);
+	info2 = btf_array(t2);
 	return info1->type == info2->type &&
 	       info1->index_type == info2->index_type &&
 	       info1->nelems == info2->nelems;
@@ -1927,14 +1976,10 @@ static bool btf_equal_array(struct btf_type *t1, struct btf_type *t2)
  */
 static bool btf_compat_array(struct btf_type *t1, struct btf_type *t2)
 {
-	struct btf_array *info1, *info2;
-
 	if (!btf_equal_common(t1, t2))
 		return false;
 
-	info1 = (struct btf_array *)(t1 + 1);
+	return btf_array(t1)->nelems == btf_array(t2)->nelems;
-	info2 = (struct btf_array *)(t2 + 1);
-	return info1->nelems == info2->nelems;
 }
 
 /*
@@ -1944,8 +1989,8 @@ static bool btf_compat_array(struct btf_type *t1, struct btf_type *t2)
  */
 static long btf_hash_fnproto(struct btf_type *t)
 {
-	struct btf_param *member = (struct btf_param *)(t + 1);
+	const struct btf_param *member = btf_params(t);
-	__u16 vlen = BTF_INFO_VLEN(t->info);
+	__u16 vlen = btf_vlen(t);
 	long h = btf_hash_common(t);
 	int i;
 
@@ -1966,16 +2011,16 @@ static long btf_hash_fnproto(struct btf_type *t)
  */
 static bool btf_equal_fnproto(struct btf_type *t1, struct btf_type *t2)
 {
-	struct btf_param *m1, *m2;
+	const struct btf_param *m1, *m2;
 	__u16 vlen;
 	int i;
 
 	if (!btf_equal_common(t1, t2))
 		return false;
 
-	vlen = BTF_INFO_VLEN(t1->info);
+	vlen = btf_vlen(t1);
-	m1 = (struct btf_param *)(t1 + 1);
+	m1 = btf_params(t1);
-	m2 = (struct btf_param *)(t2 + 1);
+	m2 = btf_params(t2);
 	for (i = 0; i < vlen; i++) {
 		if (m1->name_off != m2->name_off || m1->type != m2->type)
 			return false;
@@ -1992,7 +2037,7 @@ static bool btf_equal_fnproto(struct btf_type *t1, struct btf_type *t2)
  */
 static bool btf_compat_fnproto(struct btf_type *t1, struct btf_type *t2)
 {
-	struct btf_param *m1, *m2;
+	const struct btf_param *m1, *m2;
 	__u16 vlen;
 	int i;
 
@@ -2000,9 +2045,9 @@ static bool btf_compat_fnproto(struct btf_type *t1, struct btf_type *t2)
 	if (t1->name_off != t2->name_off || t1->info != t2->info)
 		return false;
 
-	vlen = BTF_INFO_VLEN(t1->info);
+	vlen = btf_vlen(t1);
-	m1 = (struct btf_param *)(t1 + 1);
+	m1 = btf_params(t1);
-	m2 = (struct btf_param *)(t2 + 1);
+	m2 = btf_params(t2);
 	for (i = 0; i < vlen; i++) {
 		if (m1->name_off != m2->name_off)
 			return false;
@@ -2028,7 +2073,7 @@ static int btf_dedup_prim_type(struct btf_dedup *d, __u32 type_id)
 	__u32 cand_id;
 	long h;
 
-	switch (BTF_INFO_KIND(t->info)) {
+	switch (btf_kind(t)) {
 	case BTF_KIND_CONST:
 	case BTF_KIND_VOLATILE:
 	case BTF_KIND_RESTRICT:
@@ -2141,13 +2186,13 @@ static uint32_t resolve_fwd_id(struct btf_dedup *d, uint32_t type_id)
 {
 	__u32 orig_type_id = type_id;
 
-	if (BTF_INFO_KIND(d->btf->types[type_id]->info) != BTF_KIND_FWD)
+	if (!btf_is_fwd(d->btf->types[type_id]))
 		return type_id;
 
 	while (is_type_mapped(d, type_id) && d->map[type_id] != type_id)
 		type_id = d->map[type_id];
 
-	if (BTF_INFO_KIND(d->btf->types[type_id]->info) != BTF_KIND_FWD)
+	if (!btf_is_fwd(d->btf->types[type_id]))
 		return type_id;
 
 	return orig_type_id;
@@ -2156,7 +2201,7 @@ static uint32_t resolve_fwd_id(struct btf_dedup *d, uint32_t type_id)
 
 static inline __u16 btf_fwd_kind(struct btf_type *t)
 {
-	return BTF_INFO_KFLAG(t->info) ? BTF_KIND_UNION : BTF_KIND_STRUCT;
+	return btf_kflag(t) ? BTF_KIND_UNION : BTF_KIND_STRUCT;
 }
 
 /*
@@ -2277,8 +2322,8 @@ static int btf_dedup_is_equiv(struct btf_dedup *d, __u32 cand_id,
 
 	cand_type = d->btf->types[cand_id];
 	canon_type = d->btf->types[canon_id];
-	cand_kind = BTF_INFO_KIND(cand_type->info);
+	cand_kind = btf_kind(cand_type);
-	canon_kind = BTF_INFO_KIND(canon_type->info);
+	canon_kind = btf_kind(canon_type);
 
 	if (cand_type->name_off != canon_type->name_off)
 		return 0;
@@ -2327,12 +2372,12 @@ static int btf_dedup_is_equiv(struct btf_dedup *d, __u32 cand_id,
 		return btf_dedup_is_equiv(d, cand_type->type, canon_type->type);
 
 	case BTF_KIND_ARRAY: {
-		struct btf_array *cand_arr, *canon_arr;
+		const struct btf_array *cand_arr, *canon_arr;
 
 		if (!btf_compat_array(cand_type, canon_type))
 			return 0;
-		cand_arr = (struct btf_array *)(cand_type + 1);
+		cand_arr = btf_array(cand_type);
-		canon_arr = (struct btf_array *)(canon_type + 1);
+		canon_arr = btf_array(canon_type);
 		eq = btf_dedup_is_equiv(d,
 			cand_arr->index_type, canon_arr->index_type);
 		if (eq <= 0)
@@ -2342,14 +2387,14 @@ static int btf_dedup_is_equiv(struct btf_dedup *d, __u32 cand_id,
 
 	case BTF_KIND_STRUCT:
 	case BTF_KIND_UNION: {
-		struct btf_member *cand_m, *canon_m;
+		const struct btf_member *cand_m, *canon_m;
 		__u16 vlen;
 
 		if (!btf_shallow_equal_struct(cand_type, canon_type))
 			return 0;
-		vlen = BTF_INFO_VLEN(cand_type->info);
+		vlen = btf_vlen(cand_type);
-		cand_m = (struct btf_member *)(cand_type + 1);
+		cand_m = btf_members(cand_type);
-		canon_m = (struct btf_member *)(canon_type + 1);
+		canon_m = btf_members(canon_type);
 		for (i = 0; i < vlen; i++) {
 			eq = btf_dedup_is_equiv(d, cand_m->type, canon_m->type);
 			if (eq <= 0)
@@ -2362,7 +2407,7 @@ static int btf_dedup_is_equiv(struct btf_dedup *d, __u32 cand_id,
 	}
 
 	case BTF_KIND_FUNC_PROTO: {
-		struct btf_param *cand_p, *canon_p;
+		const struct btf_param *cand_p, *canon_p;
 		__u16 vlen;
 
 		if (!btf_compat_fnproto(cand_type, canon_type))
@@ -2370,9 +2415,9 @@ static int btf_dedup_is_equiv(struct btf_dedup *d, __u32 cand_id,
 		eq = btf_dedup_is_equiv(d, cand_type->type, canon_type->type);
 		if (eq <= 0)
 			return eq;
-		vlen = BTF_INFO_VLEN(cand_type->info);
+		vlen = btf_vlen(cand_type);
-		cand_p = (struct btf_param *)(cand_type + 1);
+		cand_p = btf_params(cand_type);
-		canon_p = (struct btf_param *)(canon_type + 1);
+		canon_p = btf_params(canon_type);
 		for (i = 0; i < vlen; i++) {
 			eq = btf_dedup_is_equiv(d, cand_p->type, canon_p->type);
 			if (eq <= 0)
@@ -2427,8 +2472,8 @@ static void btf_dedup_merge_hypot_map(struct btf_dedup *d)
 		targ_type_id = d->hypot_map[cand_type_id];
 		t_id = resolve_type_id(d, targ_type_id);
 		c_id = resolve_type_id(d, cand_type_id);
-		t_kind = BTF_INFO_KIND(d->btf->types[t_id]->info);
+		t_kind = btf_kind(d->btf->types[t_id]);
-		c_kind = BTF_INFO_KIND(d->btf->types[c_id]->info);
+		c_kind = btf_kind(d->btf->types[c_id]);
 		/*
 		 * Resolve FWD into STRUCT/UNION.
 		 * It's ok to resolve FWD into STRUCT/UNION that's not yet
@@ -2497,7 +2542,7 @@ static int btf_dedup_struct_type(struct btf_dedup *d, __u32 type_id)
 		return 0;
 
 	t = d->btf->types[type_id];
-	kind = BTF_INFO_KIND(t->info);
+	kind = btf_kind(t);
 
 	if (kind != BTF_KIND_STRUCT && kind != BTF_KIND_UNION)
 		return 0;
@@ -2592,7 +2637,7 @@ static int btf_dedup_ref_type(struct btf_dedup *d, __u32 type_id)
 	t = d->btf->types[type_id];
 	d->map[type_id] = BTF_IN_PROGRESS_ID;
 
-	switch (BTF_INFO_KIND(t->info)) {
+	switch (btf_kind(t)) {
 	case BTF_KIND_CONST:
 	case BTF_KIND_VOLATILE:
 	case BTF_KIND_RESTRICT:
@@ -2616,7 +2661,7 @@ static int btf_dedup_ref_type(struct btf_dedup *d, __u32 type_id)
 		break;
 
 	case BTF_KIND_ARRAY: {
-		struct btf_array *info = (struct btf_array *)(t + 1);
+		struct btf_array *info = btf_array(t);
 
 		ref_type_id = btf_dedup_ref_type(d, info->type);
 		if (ref_type_id < 0)
@@ -2650,8 +2695,8 @@ static int btf_dedup_ref_type(struct btf_dedup *d, __u32 type_id)
 			return ref_type_id;
 		t->type = ref_type_id;
 
-		vlen = BTF_INFO_VLEN(t->info);
+		vlen = btf_vlen(t);
-		param = (struct btf_param *)(t + 1);
+		param = btf_params(t);
 		for (i = 0; i < vlen; i++) {
 			ref_type_id = btf_dedup_ref_type(d, param->type);
 			if (ref_type_id < 0)
@@ -2791,7 +2836,7 @@ static int btf_dedup_remap_type(struct btf_dedup *d, __u32 type_id)
 	struct btf_type *t = d->btf->types[type_id];
 	int i, r;
 
-	switch (BTF_INFO_KIND(t->info)) {
+	switch (btf_kind(t)) {
 	case BTF_KIND_INT:
 	case BTF_KIND_ENUM:
 		break;
@@ -2811,7 +2856,7 @@ static int btf_dedup_remap_type(struct btf_dedup *d, __u32 type_id)
 		break;
 
 	case BTF_KIND_ARRAY: {
-		struct btf_array *arr_info = (struct btf_array *)(t + 1);
+		struct btf_array *arr_info = btf_array(t);
 
 		r = btf_dedup_remap_type_id(d, arr_info->type);
 		if (r < 0)
@@ -2826,8 +2871,8 @@ static int btf_dedup_remap_type(struct btf_dedup *d, __u32 type_id)
 
 	case BTF_KIND_STRUCT:
 	case BTF_KIND_UNION: {
-		struct btf_member *member = (struct btf_member *)(t + 1);
+		struct btf_member *member = btf_members(t);
-		__u16 vlen = BTF_INFO_VLEN(t->info);
+		__u16 vlen = btf_vlen(t);
 
 		for (i = 0; i < vlen; i++) {
 			r = btf_dedup_remap_type_id(d, member->type);
@@ -2840,8 +2885,8 @@ static int btf_dedup_remap_type(struct btf_dedup *d, __u32 type_id)
 	}
 
 	case BTF_KIND_FUNC_PROTO: {
-		struct btf_param *param = (struct btf_param *)(t + 1);
+		struct btf_param *param = btf_params(t);
-		__u16 vlen = BTF_INFO_VLEN(t->info);
+		__u16 vlen = btf_vlen(t);
 
 		r = btf_dedup_remap_type_id(d, t->type);
 		if (r < 0)
@@ -2859,8 +2904,8 @@ static int btf_dedup_remap_type(struct btf_dedup *d, __u32 type_id)
 	}
 
 	case BTF_KIND_DATASEC: {
-		struct btf_var_secinfo *var = (struct btf_var_secinfo *)(t + 1);
+		struct btf_var_secinfo *var = btf_var_secinfos(t);
-		__u16 vlen = BTF_INFO_VLEN(t->info);
+		__u16 vlen = btf_vlen(t);
 
 		for (i = 0; i < vlen; i++) {
 			r = btf_dedup_remap_type_id(d, var->type);
@@ -2890,3 +2935,89 @@ static int btf_dedup_remap_types(struct btf_dedup *d)
 	}
 	return 0;
 }
+
+static struct btf *btf_load_raw(const char *path)
+{
+	struct btf *btf;
+	size_t read_cnt;
+	struct stat st;
+	void *data;
+	FILE *f;
+
+	if (stat(path, &st))
+		return ERR_PTR(-errno);
+
+	data = malloc(st.st_size);
+	if (!data)
+		return ERR_PTR(-ENOMEM);
+
+	f = fopen(path, "rb");
+	if (!f) {
+		btf = ERR_PTR(-errno);
+		goto cleanup;
+	}
+
+	read_cnt = fread(data, 1, st.st_size, f);
+	fclose(f);
+	if (read_cnt < st.st_size) {
+		btf = ERR_PTR(-EBADF);
+		goto cleanup;
+	}
+
+	btf = btf__new(data, read_cnt);
+
+cleanup:
+	free(data);
+	return btf;
+}
+
+/*
+ * Probe few well-known locations for vmlinux kernel image and try to load BTF
+ * data out of it to use for target BTF.
+ */
+struct btf *libbpf_find_kernel_btf(void)
+{
+	struct {
+		const char *path_fmt;
+		bool raw_btf;
+	} locations[] = {
+		/* try canonical vmlinux BTF through sysfs first */
+		{ "/sys/kernel/btf/vmlinux", true /* raw BTF */ },
+		/* fall back to trying to find vmlinux ELF on disk otherwise */
+		{ "/boot/vmlinux-%1$s" },
+		{ "/lib/modules/%1$s/vmlinux-%1$s" },
+		{ "/lib/modules/%1$s/build/vmlinux" },
+		{ "/usr/lib/modules/%1$s/kernel/vmlinux" },
+		{ "/usr/lib/debug/boot/vmlinux-%1$s" },
+		{ "/usr/lib/debug/boot/vmlinux-%1$s.debug" },
+		{ "/usr/lib/debug/lib/modules/%1$s/vmlinux" },
+	};
+	char path[PATH_MAX + 1];
+	struct utsname buf;
+	struct btf *btf;
+	int i;
+
+	uname(&buf);
+
+	for (i = 0; i < ARRAY_SIZE(locations); i++) {
+		snprintf(path, PATH_MAX, locations[i].path_fmt, buf.release);
+
+		if (access(path, R_OK))
+			continue;
+
+		if (locations[i].raw_btf)
+			btf = btf_load_raw(path);
+		else
+			btf = btf__parse_elf(path, NULL);
+
+		pr_debug("loading kernel BTF '%s': %ld\n",
+			 path, IS_ERR(btf) ? PTR_ERR(btf) : 0);
+		if (IS_ERR(btf))
+			continue;
+
+		return btf;
+	}
+
+	pr_warn("failed to find valid kernel BTF\n");
+	return ERR_PTR(-ESRCH);
+}

src/btf.h

@@ -5,16 +5,15 @@
 #define __LIBBPF_BTF_H
 
 #include <stdarg.h>
+#include <linux/btf.h>
 #include <linux/types.h>
 
+#include "libbpf_common.h"
+
 #ifdef __cplusplus
 extern "C" {
 #endif
 
-#ifndef LIBBPF_API
-#define LIBBPF_API __attribute__((visibility("default")))
-#endif
-
 #define BTF_ELF_SEC ".BTF"
 #define BTF_EXT_ELF_SEC ".BTF.ext"
 #define MAPS_ELF_SEC ".maps"
@@ -57,6 +56,10 @@ struct btf_ext_header {
 	__u32	func_info_len;
 	__u32	line_info_off;
 	__u32	line_info_len;
+
+	/* optional part of .BTF.ext header */
+	__u32	field_reloc_off;
+	__u32	field_reloc_len;
 };
 
 LIBBPF_API void btf__free(struct btf *btf);
@@ -67,11 +70,14 @@ LIBBPF_API int btf__finalize_data(struct bpf_object *obj, struct btf *btf);
 LIBBPF_API int btf__load(struct btf *btf);
 LIBBPF_API __s32 btf__find_by_name(const struct btf *btf,
 				   const char *type_name);
+LIBBPF_API __s32 btf__find_by_name_kind(const struct btf *btf,
+					const char *type_name, __u32 kind);
 LIBBPF_API __u32 btf__get_nr_types(const struct btf *btf);
 LIBBPF_API const struct btf_type *btf__type_by_id(const struct btf *btf,
 						  __u32 id);
 LIBBPF_API __s64 btf__resolve_size(const struct btf *btf, __u32 type_id);
 LIBBPF_API int btf__resolve_type(const struct btf *btf, __u32 type_id);
+LIBBPF_API int btf__align_of(const struct btf *btf, __u32 id);
 LIBBPF_API int btf__fd(const struct btf *btf);
 LIBBPF_API const void *btf__get_raw_data(const struct btf *btf, __u32 *size);
 LIBBPF_API const char *btf__name_by_offset(const struct btf *btf, __u32 offset);
@@ -96,6 +102,8 @@ LIBBPF_API int btf_ext__reloc_line_info(const struct btf *btf,
 LIBBPF_API __u32 btf_ext__func_info_rec_size(const struct btf_ext *btf_ext);
 LIBBPF_API __u32 btf_ext__line_info_rec_size(const struct btf_ext *btf_ext);
 
+LIBBPF_API struct btf *libbpf_find_kernel_btf(void);
+
 struct btf_dedup_opts {
 	unsigned int dedup_table_size;
 	bool dont_resolve_fwds;
@@ -120,6 +128,205 @@ LIBBPF_API void btf_dump__free(struct btf_dump *d);
 
 LIBBPF_API int btf_dump__dump_type(struct btf_dump *d, __u32 id);
 
+struct btf_dump_emit_type_decl_opts {
+	/* size of this struct, for forward/backward compatiblity */
+	size_t sz;
+	/* optional field name for type declaration, e.g.:
+	 * - struct my_struct <FNAME>
+	 * - void (*<FNAME>)(int)
+	 * - char (*<FNAME>)[123]
+	 */
+	const char *field_name;
+	/* extra indentation level (in number of tabs) to emit for multi-line
+	 * type declarations (e.g., anonymous struct); applies for lines
+	 * starting from the second one (first line is assumed to have
+	 * necessary indentation already
+	 */
+	int indent_level;
+};
+#define btf_dump_emit_type_decl_opts__last_field indent_level
+
+LIBBPF_API int
+btf_dump__emit_type_decl(struct btf_dump *d, __u32 id,
+			 const struct btf_dump_emit_type_decl_opts *opts);
+
+/*
+ * A set of helpers for easier BTF types handling
+ */
+static inline __u16 btf_kind(const struct btf_type *t)
+{
+	return BTF_INFO_KIND(t->info);
+}
+
+static inline __u16 btf_vlen(const struct btf_type *t)
+{
+	return BTF_INFO_VLEN(t->info);
+}
+
+static inline bool btf_kflag(const struct btf_type *t)
+{
+	return BTF_INFO_KFLAG(t->info);
+}
+
+static inline bool btf_is_int(const struct btf_type *t)
+{
+	return btf_kind(t) == BTF_KIND_INT;
+}
+
+static inline bool btf_is_ptr(const struct btf_type *t)
+{
+	return btf_kind(t) == BTF_KIND_PTR;
+}
+
+static inline bool btf_is_array(const struct btf_type *t)
+{
+	return btf_kind(t) == BTF_KIND_ARRAY;
+}
+
+static inline bool btf_is_struct(const struct btf_type *t)
+{
+	return btf_kind(t) == BTF_KIND_STRUCT;
+}
+
+static inline bool btf_is_union(const struct btf_type *t)
+{
+	return btf_kind(t) == BTF_KIND_UNION;
+}
+
+static inline bool btf_is_composite(const struct btf_type *t)
+{
+	__u16 kind = btf_kind(t);
+
+	return kind == BTF_KIND_STRUCT || kind == BTF_KIND_UNION;
+}
+
+static inline bool btf_is_enum(const struct btf_type *t)
+{
+	return btf_kind(t) == BTF_KIND_ENUM;
+}
+
+static inline bool btf_is_fwd(const struct btf_type *t)
+{
+	return btf_kind(t) == BTF_KIND_FWD;
+}
+
+static inline bool btf_is_typedef(const struct btf_type *t)
+{
+	return btf_kind(t) == BTF_KIND_TYPEDEF;
+}
+
+static inline bool btf_is_volatile(const struct btf_type *t)
+{
+	return btf_kind(t) == BTF_KIND_VOLATILE;
+}
+
+static inline bool btf_is_const(const struct btf_type *t)
+{
+	return btf_kind(t) == BTF_KIND_CONST;
+}
+
+static inline bool btf_is_restrict(const struct btf_type *t)
+{
+	return btf_kind(t) == BTF_KIND_RESTRICT;
+}
+
+static inline bool btf_is_mod(const struct btf_type *t)
+{
+	__u16 kind = btf_kind(t);
+
+	return kind == BTF_KIND_VOLATILE ||
+	       kind == BTF_KIND_CONST ||
+	       kind == BTF_KIND_RESTRICT;
+}
+
+static inline bool btf_is_func(const struct btf_type *t)
+{
+	return btf_kind(t) == BTF_KIND_FUNC;
+}
+
+static inline bool btf_is_func_proto(const struct btf_type *t)
+{
+	return btf_kind(t) == BTF_KIND_FUNC_PROTO;
+}
+
+static inline bool btf_is_var(const struct btf_type *t)
+{
+	return btf_kind(t) == BTF_KIND_VAR;
+}
+
+static inline bool btf_is_datasec(const struct btf_type *t)
+{
+	return btf_kind(t) == BTF_KIND_DATASEC;
+}
+
+static inline __u8 btf_int_encoding(const struct btf_type *t)
+{
+	return BTF_INT_ENCODING(*(__u32 *)(t + 1));
+}
+
+static inline __u8 btf_int_offset(const struct btf_type *t)
+{
+	return BTF_INT_OFFSET(*(__u32 *)(t + 1));
+}
+
+static inline __u8 btf_int_bits(const struct btf_type *t)
+{
+	return BTF_INT_BITS(*(__u32 *)(t + 1));
+}
+
+static inline struct btf_array *btf_array(const struct btf_type *t)
+{
+	return (struct btf_array *)(t + 1);
+}
+
+static inline struct btf_enum *btf_enum(const struct btf_type *t)
+{
+	return (struct btf_enum *)(t + 1);
+}
+
+static inline struct btf_member *btf_members(const struct btf_type *t)
+{
+	return (struct btf_member *)(t + 1);
+}
+
+/* Get bit offset of a member with specified index. */
+static inline __u32 btf_member_bit_offset(const struct btf_type *t,
+					  __u32 member_idx)
+{
+	const struct btf_member *m = btf_members(t) + member_idx;
+	bool kflag = btf_kflag(t);
+
+	return kflag ? BTF_MEMBER_BIT_OFFSET(m->offset) : m->offset;
+}
+/*
+ * Get bitfield size of a member, assuming t is BTF_KIND_STRUCT or
+ * BTF_KIND_UNION. If member is not a bitfield, zero is returned.
+ */
+static inline __u32 btf_member_bitfield_size(const struct btf_type *t,
+					     __u32 member_idx)
+{
+	const struct btf_member *m = btf_members(t) + member_idx;
+	bool kflag = btf_kflag(t);
+
+	return kflag ? BTF_MEMBER_BITFIELD_SIZE(m->offset) : 0;
+}
+
+static inline struct btf_param *btf_params(const struct btf_type *t)
+{
+	return (struct btf_param *)(t + 1);
+}
+
+static inline struct btf_var *btf_var(const struct btf_type *t)
+{
+	return (struct btf_var *)(t + 1);
+}
+
+static inline struct btf_var_secinfo *
+btf_var_secinfos(const struct btf_type *t)
+{
+	return (struct btf_var_secinfo *)(t + 1);
+}
+
 #ifdef __cplusplus
 } /* extern "C" */
 #endif

src/btf_dump.c

@@ -18,6 +18,9 @@
 #include "libbpf.h"
 #include "libbpf_internal.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 static const char PREFIXES[] = "\t\t\t\t\t\t\t\t\t\t\t\t\t";
 static const size_t PREFIX_CNT = sizeof(PREFIXES) - 1;
 
@@ -48,6 +51,8 @@ struct btf_dump_type_aux_state {
 	__u8 fwd_emitted: 1;
 	/* whether unique non-duplicate name was already assigned */
 	__u8 name_resolved: 1;
+	/* whether type is referenced from any other type */
+	__u8 referenced: 1;
 };
 
 struct btf_dump {
@@ -100,21 +105,6 @@ static bool str_equal_fn(const void *a, const void *b, void *ctx)
 	return strcmp(a, b) == 0;
 }
 
-static __u16 btf_kind_of(const struct btf_type *t)
-{
-	return BTF_INFO_KIND(t->info);
-}
-
-static __u16 btf_vlen_of(const struct btf_type *t)
-{
-	return BTF_INFO_VLEN(t->info);
-}
-
-static bool btf_kflag_of(const struct btf_type *t)
-{
-	return BTF_INFO_KFLAG(t->info);
-}
-
 static const char *btf_name_of(const struct btf_dump *d, __u32 name_off)
 {
 	return btf__name_by_offset(d->btf, name_off);
@@ -129,6 +119,8 @@ static void btf_dump_printf(const struct btf_dump *d, const char *fmt, ...)
 	va_end(args);
 }
 
+static int btf_dump_mark_referenced(struct btf_dump *d);
+
 struct btf_dump *btf_dump__new(const struct btf *btf,
 			       const struct btf_ext *btf_ext,
 			       const struct btf_dump_opts *opts,
@@ -150,18 +142,40 @@ struct btf_dump *btf_dump__new(const struct btf *btf,
 	if (IS_ERR(d->type_names)) {
 		err = PTR_ERR(d->type_names);
 		d->type_names = NULL;
-		btf_dump__free(d);
+		goto err;
-		return ERR_PTR(err);
 	}
 	d->ident_names = hashmap__new(str_hash_fn, str_equal_fn, NULL);
 	if (IS_ERR(d->ident_names)) {
 		err = PTR_ERR(d->ident_names);
 		d->ident_names = NULL;
-		btf_dump__free(d);
+		goto err;
-		return ERR_PTR(err);
+	}
+	d->type_states = calloc(1 + btf__get_nr_types(d->btf),
+				sizeof(d->type_states[0]));
+	if (!d->type_states) {
+		err = -ENOMEM;
+		goto err;
+	}
+	d->cached_names = calloc(1 + btf__get_nr_types(d->btf),
+				 sizeof(d->cached_names[0]));
+	if (!d->cached_names) {
+		err = -ENOMEM;
+		goto err;
 	}
 
+	/* VOID is special */
+	d->type_states[0].order_state = ORDERED;
+	d->type_states[0].emit_state = EMITTED;
+
+	/* eagerly determine referenced types for anon enums */
+	err = btf_dump_mark_referenced(d);
+	if (err)
+		goto err;
+
 	return d;
+err:
+	btf_dump__free(d);
+	return ERR_PTR(err);
 }
 
 void btf_dump__free(struct btf_dump *d)
@@ -214,22 +228,6 @@ int btf_dump__dump_type(struct btf_dump *d, __u32 id)
 	if (id > btf__get_nr_types(d->btf))
 		return -EINVAL;
 
-	/* type states are lazily allocated, as they might not be needed */
-	if (!d->type_states) {
-		d->type_states = calloc(1 + btf__get_nr_types(d->btf),
-					sizeof(d->type_states[0]));
-		if (!d->type_states)
-			return -ENOMEM;
-		d->cached_names = calloc(1 + btf__get_nr_types(d->btf),
-					 sizeof(d->cached_names[0]));
-		if (!d->cached_names)
-			return -ENOMEM;
-
-		/* VOID is special */
-		d->type_states[0].order_state = ORDERED;
-		d->type_states[0].emit_state = EMITTED;
-	}
-
 	d->emit_queue_cnt = 0;
 	err = btf_dump_order_type(d, id, false);
 	if (err < 0)
@@ -241,6 +239,79 @@ int btf_dump__dump_type(struct btf_dump *d, __u32 id)
 	return 0;
 }
 
+/*
+ * Mark all types that are referenced from any other type. This is used to
+ * determine top-level anonymous enums that need to be emitted as an
+ * independent type declarations.
+ * Anonymous enums come in two flavors: either embedded in a struct's field
+ * definition, in which case they have to be declared inline as part of field
+ * type declaration; or as a top-level anonymous enum, typically used for
+ * declaring global constants. It's impossible to distinguish between two
+ * without knowning whether given enum type was referenced from other type:
+ * top-level anonymous enum won't be referenced by anything, while embedded
+ * one will.
+ */
+static int btf_dump_mark_referenced(struct btf_dump *d)
+{
+	int i, j, n = btf__get_nr_types(d->btf);
+	const struct btf_type *t;
+	__u16 vlen;
+
+	for (i = 1; i <= n; i++) {
+		t = btf__type_by_id(d->btf, i);
+		vlen = btf_vlen(t);
+
+		switch (btf_kind(t)) {
+		case BTF_KIND_INT:
+		case BTF_KIND_ENUM:
+		case BTF_KIND_FWD:
+			break;
+
+		case BTF_KIND_VOLATILE:
+		case BTF_KIND_CONST:
+		case BTF_KIND_RESTRICT:
+		case BTF_KIND_PTR:
+		case BTF_KIND_TYPEDEF:
+		case BTF_KIND_FUNC:
+		case BTF_KIND_VAR:
+			d->type_states[t->type].referenced = 1;
+			break;
+
+		case BTF_KIND_ARRAY: {
+			const struct btf_array *a = btf_array(t);
+
+			d->type_states[a->index_type].referenced = 1;
+			d->type_states[a->type].referenced = 1;
+			break;
+		}
+		case BTF_KIND_STRUCT:
+		case BTF_KIND_UNION: {
+			const struct btf_member *m = btf_members(t);
+
+			for (j = 0; j < vlen; j++, m++)
+				d->type_states[m->type].referenced = 1;
+			break;
+		}
+		case BTF_KIND_FUNC_PROTO: {
+			const struct btf_param *p = btf_params(t);
+
+			for (j = 0; j < vlen; j++, p++)
+				d->type_states[p->type].referenced = 1;
+			break;
+		}
+		case BTF_KIND_DATASEC: {
+			const struct btf_var_secinfo *v = btf_var_secinfos(t);
+
+			for (j = 0; j < vlen; j++, v++)
+				d->type_states[v->type].referenced = 1;
+			break;
+		}
+		default:
+			return -EINVAL;
+		}
+	}
+	return 0;
+}
 static int btf_dump_add_emit_queue_id(struct btf_dump *d, __u32 id)
 {
 	__u32 *new_queue;
@@ -349,7 +420,7 @@ static int btf_dump_order_type(struct btf_dump *d, __u32 id, bool through_ptr)
 	 */
 	struct btf_dump_type_aux_state *tstate = &d->type_states[id];
 	const struct btf_type *t;
-	__u16 kind, vlen;
+	__u16 vlen;
 	int err, i;
 
 	/* return true, letting typedefs know that it's ok to be emitted */
@@ -357,18 +428,16 @@ static int btf_dump_order_type(struct btf_dump *d, __u32 id, bool through_ptr)
 		return 1;
 
 	t = btf__type_by_id(d->btf, id);
-	kind = btf_kind_of(t);
 
 	if (tstate->order_state == ORDERING) {
 		/* type loop, but resolvable through fwd declaration */
-		if ((kind == BTF_KIND_STRUCT || kind == BTF_KIND_UNION) &&
+		if (btf_is_composite(t) && through_ptr && t->name_off != 0)
-		    through_ptr && t->name_off != 0)
 			return 0;
-		pr_warning("unsatisfiable type cycle, id:[%u]\n", id);
+		pr_warn("unsatisfiable type cycle, id:[%u]\n", id);
 		return -ELOOP;
 	}
 
-	switch (kind) {
+	switch (btf_kind(t)) {
 	case BTF_KIND_INT:
 		tstate->order_state = ORDERED;
 		return 0;
@@ -378,14 +447,12 @@ static int btf_dump_order_type(struct btf_dump *d, __u32 id, bool through_ptr)
 		tstate->order_state = ORDERED;
 		return err;
 
-	case BTF_KIND_ARRAY: {
+	case BTF_KIND_ARRAY:
-		const struct btf_array *a = (void *)(t + 1);
+		return btf_dump_order_type(d, btf_array(t)->type, through_ptr);
 
-		return btf_dump_order_type(d, a->type, through_ptr);
-	}
 	case BTF_KIND_STRUCT:
 	case BTF_KIND_UNION: {
-		const struct btf_member *m = (void *)(t + 1);
+		const struct btf_member *m = btf_members(t);
 		/*
 		 * struct/union is part of strong link, only if it's embedded
 		 * (so no ptr in a path) or it's anonymous (so has to be
@@ -396,7 +463,7 @@ static int btf_dump_order_type(struct btf_dump *d, __u32 id, bool through_ptr)
 
 		tstate->order_state = ORDERING;
 
-		vlen = btf_vlen_of(t);
+		vlen = btf_vlen(t);
 		for (i = 0; i < vlen; i++, m++) {
 			err = btf_dump_order_type(d, m->type, false);
 			if (err < 0)
@@ -414,7 +481,12 @@ static int btf_dump_order_type(struct btf_dump *d, __u32 id, bool through_ptr)
 	}
 	case BTF_KIND_ENUM:
 	case BTF_KIND_FWD:
-		if (t->name_off != 0) {
+		/*
+		 * non-anonymous or non-referenced enums are top-level
+		 * declarations and should be emitted. Same logic can be
+		 * applied to FWDs, it won't hurt anyways.
+		 */
+		if (t->name_off != 0 || !tstate->referenced) {
 			err = btf_dump_add_emit_queue_id(d, id);
 			if (err)
 				return err;
@@ -447,7 +519,7 @@ static int btf_dump_order_type(struct btf_dump *d, __u32 id, bool through_ptr)
 		return btf_dump_order_type(d, t->type, through_ptr);
 
 	case BTF_KIND_FUNC_PROTO: {
-		const struct btf_param *p = (void *)(t + 1);
+		const struct btf_param *p = btf_params(t);
 		bool is_strong;
 
 		err = btf_dump_order_type(d, t->type, through_ptr);
@@ -455,7 +527,7 @@ static int btf_dump_order_type(struct btf_dump *d, __u32 id, bool through_ptr)
 			return err;
 		is_strong = err > 0;
 
-		vlen = btf_vlen_of(t);
+		vlen = btf_vlen(t);
 		for (i = 0; i < vlen; i++, p++) {
 			err = btf_dump_order_type(d, p->type, through_ptr);
 			if (err < 0)
@@ -553,12 +625,7 @@ static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id)
 		return;
 
 	t = btf__type_by_id(d->btf, id);
-	kind = btf_kind_of(t);
+	kind = btf_kind(t);
-
-	if (top_level_def && t->name_off == 0) {
-		pr_warning("unexpected nameless definition, id:[%u]\n", id);
-		return;
-	}
 
 	if (tstate->emit_state == EMITTING) {
 		if (tstate->fwd_emitted)
@@ -574,8 +641,8 @@ static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id)
 			if (id == cont_id)
 				return;
 			if (t->name_off == 0) {
-				pr_warning("anonymous struct/union loop, id:[%u]\n",
+				pr_warn("anonymous struct/union loop, id:[%u]\n",
-					   id);
+					id);
 				return;
 			}
 			btf_dump_emit_struct_fwd(d, id, t);
@@ -618,12 +685,9 @@ static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id)
 	case BTF_KIND_RESTRICT:
 		btf_dump_emit_type(d, t->type, cont_id);
 		break;
-	case BTF_KIND_ARRAY: {
+	case BTF_KIND_ARRAY:
-		const struct btf_array *a = (void *)(t + 1);
+		btf_dump_emit_type(d, btf_array(t)->type, cont_id);
-
-		btf_dump_emit_type(d, a->type, cont_id);
 		break;
-	}
 	case BTF_KIND_FWD:
 		btf_dump_emit_fwd_def(d, id, t);
 		btf_dump_printf(d, ";\n\n");
@@ -656,8 +720,8 @@ static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id)
 		 * applicable
 		 */
 		if (top_level_def || t->name_off == 0) {
-			const struct btf_member *m = (void *)(t + 1);
+			const struct btf_member *m = btf_members(t);
-			__u16 vlen = btf_vlen_of(t);
+			__u16 vlen = btf_vlen(t);
 			int i, new_cont_id;
 
 			new_cont_id = t->name_off == 0 ? cont_id : id;
@@ -678,8 +742,8 @@ static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id)
 		}
 		break;
 	case BTF_KIND_FUNC_PROTO: {
-		const struct btf_param *p = (void *)(t + 1);
+		const struct btf_param *p = btf_params(t);
-		__u16 vlen = btf_vlen_of(t);
+		__u16 vlen = btf_vlen(t);
 		int i;
 
 		btf_dump_emit_type(d, t->type, cont_id);
@@ -693,65 +757,25 @@ static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id)
 	}
 }
 
-static int btf_align_of(const struct btf *btf, __u32 id)
-{
-	const struct btf_type *t = btf__type_by_id(btf, id);
-	__u16 kind = btf_kind_of(t);
-
-	switch (kind) {
-	case BTF_KIND_INT:
-	case BTF_KIND_ENUM:
-		return min(sizeof(void *), t->size);
-	case BTF_KIND_PTR:
-		return sizeof(void *);
-	case BTF_KIND_TYPEDEF:
-	case BTF_KIND_VOLATILE:
-	case BTF_KIND_CONST:
-	case BTF_KIND_RESTRICT:
-		return btf_align_of(btf, t->type);
-	case BTF_KIND_ARRAY: {
-		const struct btf_array *a = (void *)(t + 1);
-
-		return btf_align_of(btf, a->type);
-	}
-	case BTF_KIND_STRUCT:
-	case BTF_KIND_UNION: {
-		const struct btf_member *m = (void *)(t + 1);
-		__u16 vlen = btf_vlen_of(t);
-		int i, align = 1;
-
-		for (i = 0; i < vlen; i++, m++)
-			align = max(align, btf_align_of(btf, m->type));
-
-		return align;
-	}
-	default:
-		pr_warning("unsupported BTF_KIND:%u\n", btf_kind_of(t));
-		return 1;
-	}
-}
-
 static bool btf_is_struct_packed(const struct btf *btf, __u32 id,
 				 const struct btf_type *t)
 {
 	const struct btf_member *m;
 	int align, i, bit_sz;
 	__u16 vlen;
-	bool kflag;
 
-	align = btf_align_of(btf, id);
+	align = btf__align_of(btf, id);
 	/* size of a non-packed struct has to be a multiple of its alignment*/
-	if (t->size % align)
+	if (align && t->size % align)
 		return true;
 
-	m = (void *)(t + 1);
+	m = btf_members(t);
-	kflag = btf_kflag_of(t);
+	vlen = btf_vlen(t);
-	vlen = btf_vlen_of(t);
 	/* all non-bitfield fields have to be naturally aligned */
 	for (i = 0; i < vlen; i++, m++) {
-		align = btf_align_of(btf, m->type);
+		align = btf__align_of(btf, m->type);
-		bit_sz = kflag ? BTF_MEMBER_BITFIELD_SIZE(m->offset) : 0;
+		bit_sz = btf_member_bitfield_size(t, i);
-		if (bit_sz == 0 && m->offset % (8 * align) != 0)
+		if (align && bit_sz == 0 && m->offset % (8 * align) != 0)
 			return true;
 	}
 
@@ -807,7 +831,7 @@ static void btf_dump_emit_struct_fwd(struct btf_dump *d, __u32 id,
 				     const struct btf_type *t)
 {
 	btf_dump_printf(d, "%s %s",
-			btf_kind_of(t) == BTF_KIND_STRUCT ? "struct" : "union",
+			btf_is_struct(t) ? "struct" : "union",
 			btf_dump_type_name(d, id));
 }
 
@@ -816,14 +840,12 @@ static void btf_dump_emit_struct_def(struct btf_dump *d,
 				     const struct btf_type *t,
 				     int lvl)
 {
-	const struct btf_member *m = (void *)(t + 1);
+	const struct btf_member *m = btf_members(t);
-	bool kflag = btf_kflag_of(t), is_struct;
+	bool is_struct = btf_is_struct(t);
 	int align, i, packed, off = 0;
-	__u16 vlen = btf_vlen_of(t);
+	__u16 vlen = btf_vlen(t);
 
-	is_struct = btf_kind_of(t) == BTF_KIND_STRUCT;
 	packed = is_struct ? btf_is_struct_packed(d->btf, id, t) : 0;
-	align = packed ? 1 : btf_align_of(d->btf, id);
 
 	btf_dump_printf(d, "%s%s%s {",
 			is_struct ? "struct" : "union",
@@ -835,9 +857,9 @@ static void btf_dump_emit_struct_def(struct btf_dump *d,
 		int m_off, m_sz;
 
 		fname = btf_name_of(d, m->name_off);
-		m_sz = kflag ? BTF_MEMBER_BITFIELD_SIZE(m->offset) : 0;
+		m_sz = btf_member_bitfield_size(t, i);
-		m_off = kflag ? BTF_MEMBER_BIT_OFFSET(m->offset) : m->offset;
+		m_off = btf_member_bit_offset(t, i);
-		align = packed ? 1 : btf_align_of(d->btf, m->type);
+		align = packed ? 1 : btf__align_of(d->btf, m->type);
 
 		btf_dump_emit_bit_padding(d, off, m_off, m_sz, align, lvl + 1);
 		btf_dump_printf(d, "\n%s", pfx(lvl + 1));
@@ -853,6 +875,13 @@ static void btf_dump_emit_struct_def(struct btf_dump *d,
 		btf_dump_printf(d, ";");
 	}
 
+	/* pad at the end, if necessary */
+	if (is_struct) {
+		align = packed ? 1 : btf__align_of(d->btf, id);
+		btf_dump_emit_bit_padding(d, off, t->size * 8, 0, align,
+					  lvl + 1);
+	}
+
 	if (vlen)
 		btf_dump_printf(d, "\n");
 	btf_dump_printf(d, "%s}", pfx(lvl));
@@ -870,8 +899,8 @@ static void btf_dump_emit_enum_def(struct btf_dump *d, __u32 id,
 				   const struct btf_type *t,
 				   int lvl)
 {
-	const struct btf_enum *v = (void *)(t+1);
+	const struct btf_enum *v = btf_enum(t);
-	__u16 vlen = btf_vlen_of(t);
+	__u16 vlen = btf_vlen(t);
 	const char *name;
 	size_t dup_cnt;
 	int i;
@@ -905,7 +934,7 @@ static void btf_dump_emit_fwd_def(struct btf_dump *d, __u32 id,
 {
 	const char *name = btf_dump_type_name(d, id);
 
-	if (btf_kflag_of(t))
+	if (btf_kflag(t))
 		btf_dump_printf(d, "union %s", name);
 	else
 		btf_dump_printf(d, "struct %s", name);
@@ -916,6 +945,17 @@ static void btf_dump_emit_typedef_def(struct btf_dump *d, __u32 id,
 {
 	const char *name = btf_dump_ident_name(d, id);
 
+	/*
+	 * Old GCC versions are emitting invalid typedef for __gnuc_va_list
+	 * pointing to VOID. This generates warnings from btf_dump() and
+	 * results in uncompilable header file, so we are fixing it up here
+	 * with valid typedef into __builtin_va_list.
+	 */
+	if (t->type == 0 && strcmp(name, "__gnuc_va_list") == 0) {
+		btf_dump_printf(d, "typedef __builtin_va_list __gnuc_va_list");
+		return;
+	}
+
 	btf_dump_printf(d, "typedef ");
 	btf_dump_emit_type_decl(d, t->type, name, lvl);
 }
@@ -981,13 +1021,27 @@ static int btf_dump_push_decl_stack_id(struct btf_dump *d, __u32 id)
  * of a stack frame. Some care is required to "pop" stack frames after
  * processing type declaration chain.
  */
+int btf_dump__emit_type_decl(struct btf_dump *d, __u32 id,
+			     const struct btf_dump_emit_type_decl_opts *opts)
+{
+	const char *fname;
+	int lvl;
+
+	if (!OPTS_VALID(opts, btf_dump_emit_type_decl_opts))
+		return -EINVAL;
+
+	fname = OPTS_GET(opts, field_name, NULL);
+	lvl = OPTS_GET(opts, indent_level, 0);
+	btf_dump_emit_type_decl(d, id, fname, lvl);
+	return 0;
+}
+
 static void btf_dump_emit_type_decl(struct btf_dump *d, __u32 id,
 				    const char *fname, int lvl)
 {
 	struct id_stack decl_stack;
 	const struct btf_type *t;
 	int err, stack_start;
-	__u16 kind;
 
 	stack_start = d->decl_stack_cnt;
 	for (;;) {
@@ -998,7 +1052,7 @@ static void btf_dump_emit_type_decl(struct btf_dump *d, __u32 id,
 			 * chain, restore stack, emit warning, and try to
 			 * proceed nevertheless
 			 */
-			pr_warning("not enough memory for decl stack:%d", err);
+			pr_warn("not enough memory for decl stack:%d", err);
 			d->decl_stack_cnt = stack_start;
 			return;
 		}
@@ -1008,8 +1062,7 @@ static void btf_dump_emit_type_decl(struct btf_dump *d, __u32 id,
 			break;
 
 		t = btf__type_by_id(d->btf, id);
-		kind = btf_kind_of(t);
+		switch (btf_kind(t)) {
-		switch (kind) {
 		case BTF_KIND_PTR:
 		case BTF_KIND_VOLATILE:
 		case BTF_KIND_CONST:
@@ -1017,12 +1070,9 @@ static void btf_dump_emit_type_decl(struct btf_dump *d, __u32 id,
 		case BTF_KIND_FUNC_PROTO:
 			id = t->type;
 			break;
-		case BTF_KIND_ARRAY: {
+		case BTF_KIND_ARRAY:
-			const struct btf_array *a = (void *)(t + 1);
+			id = btf_array(t)->type;
-
-			id = a->type;
 			break;
-		}
 		case BTF_KIND_INT:
 		case BTF_KIND_ENUM:
 		case BTF_KIND_FWD:
@@ -1031,8 +1081,8 @@ static void btf_dump_emit_type_decl(struct btf_dump *d, __u32 id,
 		case BTF_KIND_TYPEDEF:
 			goto done;
 		default:
-			pr_warning("unexpected type in decl chain, kind:%u, id:[%u]\n",
+			pr_warn("unexpected type in decl chain, kind:%u, id:[%u]\n",
-				   kind, id);
+				btf_kind(t), id);
 			goto done;
 		}
 	}
@@ -1070,7 +1120,7 @@ static void btf_dump_emit_mods(struct btf_dump *d, struct id_stack *decl_stack)
 		id = decl_stack->ids[decl_stack->cnt - 1];
 		t = btf__type_by_id(d->btf, id);
 
-		switch (btf_kind_of(t)) {
+		switch (btf_kind(t)) {
 		case BTF_KIND_VOLATILE:
 			btf_dump_printf(d, "volatile ");
 			break;
@@ -1087,20 +1137,6 @@ static void btf_dump_emit_mods(struct btf_dump *d, struct id_stack *decl_stack)
 	}
 }
 
-static bool btf_is_mod_kind(const struct btf *btf, __u32 id)
-{
-	const struct btf_type *t = btf__type_by_id(btf, id);
-
-	switch (btf_kind_of(t)) {
-	case BTF_KIND_VOLATILE:
-	case BTF_KIND_CONST:
-	case BTF_KIND_RESTRICT:
-		return true;
-	default:
-		return false;
-	}
-}
-
 static void btf_dump_emit_name(const struct btf_dump *d,
 			       const char *name, bool last_was_ptr)
 {
@@ -1139,7 +1175,7 @@ static void btf_dump_emit_type_chain(struct btf_dump *d,
 		}
 
 		t = btf__type_by_id(d->btf, id);
-		kind = btf_kind_of(t);
+		kind = btf_kind(t);
 
 		switch (kind) {
 		case BTF_KIND_INT:
@@ -1185,7 +1221,7 @@ static void btf_dump_emit_type_chain(struct btf_dump *d,
 			btf_dump_printf(d, " restrict");
 			break;
 		case BTF_KIND_ARRAY: {
-			const struct btf_array *a = (void *)(t + 1);
+			const struct btf_array *a = btf_array(t);
 			const struct btf_type *next_t;
 			__u32 next_id;
 			bool multidim;
@@ -1201,7 +1237,8 @@ static void btf_dump_emit_type_chain(struct btf_dump *d,
 			 */
 			while (decls->cnt) {
 				next_id = decls->ids[decls->cnt - 1];
-				if (btf_is_mod_kind(d->btf, next_id))
+				next_t = btf__type_by_id(d->btf, next_id);
+				if (btf_is_mod(next_t))
 					decls->cnt--;
 				else
 					break;
@@ -1213,8 +1250,9 @@ static void btf_dump_emit_type_chain(struct btf_dump *d,
 				return;
 			}
 
+			next_id = decls->ids[decls->cnt - 1];
 			next_t = btf__type_by_id(d->btf, next_id);
-			multidim = btf_kind_of(next_t) == BTF_KIND_ARRAY;
+			multidim = btf_is_array(next_t);
 			/* we need space if we have named non-pointer */
 			if (fname[0] && !last_was_ptr)
 				btf_dump_printf(d, " ");
@@ -1228,8 +1266,8 @@ static void btf_dump_emit_type_chain(struct btf_dump *d,
 			return;
 		}
 		case BTF_KIND_FUNC_PROTO: {
-			const struct btf_param *p = (void *)(t + 1);
+			const struct btf_param *p = btf_params(t);
-			__u16 vlen = btf_vlen_of(t);
+			__u16 vlen = btf_vlen(t);
 			int i;
 
 			btf_dump_emit_mods(d, decls);
@@ -1270,8 +1308,8 @@ static void btf_dump_emit_type_chain(struct btf_dump *d,
 			return;
 		}
 		default:
-			pr_warning("unexpected type in decl chain, kind:%u, id:[%u]\n",
+			pr_warn("unexpected type in decl chain, kind:%u, id:[%u]\n",
-				   kind, id);
+				kind, id);
 			return;
 		}
 

src/hashmap.c

@@ -12,6 +12,9 @@
 #include <linux/err.h>
 #include "hashmap.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 /* start with 4 buckets */
 #define HASHMAP_MIN_CAP_BITS 2
 

src/hashmap.h

@@ -10,6 +10,11 @@
 
 #include <stdbool.h>
 #include <stddef.h>
+#ifdef __GLIBC__
+#include <bits/wordsize.h>
+#else
+#include <bits/reg.h>
+#endif
 #include "libbpf_internal.h"
 
 static inline size_t hash_bits(size_t h, int bits)

src/libbpf.h

@@ -17,14 +17,12 @@
 #include <sys/types.h>  // for size_t
 #include <linux/bpf.h>
 
+#include "libbpf_common.h"
+
 #ifdef __cplusplus
 extern "C" {
 #endif
 
-#ifndef LIBBPF_API
-#define LIBBPF_API __attribute__((visibility("default")))
-#endif
-
 enum libbpf_errno {
 	__LIBBPF_ERRNO__START = 4000,
 
@@ -57,7 +55,7 @@ enum libbpf_print_level {
 typedef int (*libbpf_print_fn_t)(enum libbpf_print_level level,
 				 const char *, va_list ap);
 
-LIBBPF_API void libbpf_set_print(libbpf_print_fn_t fn);
+LIBBPF_API libbpf_print_fn_t libbpf_set_print(libbpf_print_fn_t fn);
 
 /* Hide internal to user */
 struct bpf_object;
@@ -67,18 +65,61 @@ struct bpf_object_open_attr {
 	enum bpf_prog_type prog_type;
 };
 
+struct bpf_object_open_opts {
+	/* size of this struct, for forward/backward compatiblity */
+	size_t sz;
+	/* object name override, if provided:
+	 * - for object open from file, this will override setting object
+	 *   name from file path's base name;
+	 * - for object open from memory buffer, this will specify an object
+	 *   name and will override default "<addr>-<buf-size>" name;
+	 */
+	const char *object_name;
+	/* parse map definitions non-strictly, allowing extra attributes/data */
+	bool relaxed_maps;
+	/* DEPRECATED: handle CO-RE relocations non-strictly, allowing failures.
+	 * Value is ignored. Relocations always are processed non-strictly.
+	 * Non-relocatable instructions are replaced with invalid ones to
+	 * prevent accidental errors.
+	 * */
+	bool relaxed_core_relocs;
+	/* maps that set the 'pinning' attribute in their definition will have
+	 * their pin_path attribute set to a file in this directory, and be
+	 * auto-pinned to that path on load; defaults to "/sys/fs/bpf".
+	 */
+	const char *pin_root_path;
+	__u32 attach_prog_fd;
+	/* Additional kernel config content that augments and overrides
+	 * system Kconfig for CONFIG_xxx externs.
+	 */
+	const char *kconfig;
+};
+#define bpf_object_open_opts__last_field kconfig
+
 LIBBPF_API struct bpf_object *bpf_object__open(const char *path);
 LIBBPF_API struct bpf_object *
+bpf_object__open_file(const char *path, const struct bpf_object_open_opts *opts);
+LIBBPF_API struct bpf_object *
+bpf_object__open_mem(const void *obj_buf, size_t obj_buf_sz,
+		     const struct bpf_object_open_opts *opts);
+
+/* deprecated bpf_object__open variants */
+LIBBPF_API struct bpf_object *
+bpf_object__open_buffer(const void *obj_buf, size_t obj_buf_sz,
+			const char *name);
+LIBBPF_API struct bpf_object *
 bpf_object__open_xattr(struct bpf_object_open_attr *attr);
-struct bpf_object *__bpf_object__open_xattr(struct bpf_object_open_attr *attr,
+
-					    int flags);
+enum libbpf_pin_type {
-LIBBPF_API struct bpf_object *bpf_object__open_buffer(void *obj_buf,
+	LIBBPF_PIN_NONE,
-						      size_t obj_buf_sz,
+	/* PIN_BY_NAME: pin maps by name (in /sys/fs/bpf by default) */
-						      const char *name);
+	LIBBPF_PIN_BY_NAME,
-int bpf_object__section_size(const struct bpf_object *obj, const char *name,
+};
-			     __u32 *size);
+
-int bpf_object__variable_offset(const struct bpf_object *obj, const char *name,
+/* pin_maps and unpin_maps can both be called with a NULL path, in which case
-				__u32 *off);
+ * they will use the pin_path attribute of each map (and ignore all maps that
+ * don't have a pin_path set).
+ */
 LIBBPF_API int bpf_object__pin_maps(struct bpf_object *obj, const char *path);
 LIBBPF_API int bpf_object__unpin_maps(struct bpf_object *obj,
 				      const char *path);
@@ -92,12 +133,14 @@ LIBBPF_API void bpf_object__close(struct bpf_object *object);
 struct bpf_object_load_attr {
 	struct bpf_object *obj;
 	int log_level;
+	const char *target_btf_path;
 };
 
 /* Load/unload object into/from kernel */
 LIBBPF_API int bpf_object__load(struct bpf_object *obj);
 LIBBPF_API int bpf_object__load_xattr(struct bpf_object_load_attr *attr);
 LIBBPF_API int bpf_object__unload(struct bpf_object *obj);
+
 LIBBPF_API const char *bpf_object__name(const struct bpf_object *obj);
 LIBBPF_API unsigned int bpf_object__kversion(const struct bpf_object *obj);
 
@@ -108,6 +151,9 @@ LIBBPF_API int bpf_object__btf_fd(const struct bpf_object *obj);
 LIBBPF_API struct bpf_program *
 bpf_object__find_program_by_title(const struct bpf_object *obj,
 				  const char *title);
+LIBBPF_API struct bpf_program *
+bpf_object__find_program_by_name(const struct bpf_object *obj,
+				 const char *name);
 
 LIBBPF_API struct bpf_object *bpf_object__next(struct bpf_object *prev);
 #define bpf_object__for_each_safe(pos, tmp)			\
@@ -126,6 +172,8 @@ libbpf_prog_type_by_name(const char *name, enum bpf_prog_type *prog_type,
 			 enum bpf_attach_type *expected_attach_type);
 LIBBPF_API int libbpf_attach_type_by_name(const char *name,
 					  enum bpf_attach_type *attach_type);
+LIBBPF_API int libbpf_find_vmlinux_btf_id(const char *name,
+					  enum bpf_attach_type attach_type);
 
 /* Accessors of bpf_program */
 struct bpf_program;
@@ -149,9 +197,13 @@ LIBBPF_API void *bpf_program__priv(const struct bpf_program *prog);
 LIBBPF_API void bpf_program__set_ifindex(struct bpf_program *prog,
 					 __u32 ifindex);
 
+LIBBPF_API const char *bpf_program__name(const struct bpf_program *prog);
 LIBBPF_API const char *bpf_program__title(const struct bpf_program *prog,
 					  bool needs_copy);
 
+/* returns program size in bytes */
+LIBBPF_API size_t bpf_program__size(const struct bpf_program *prog);
+
 LIBBPF_API int bpf_program__load(struct bpf_program *prog, char *license,
 				 __u32 kern_version);
 LIBBPF_API int bpf_program__fd(const struct bpf_program *prog);
@@ -167,8 +219,11 @@ LIBBPF_API void bpf_program__unload(struct bpf_program *prog);
 
 struct bpf_link;
 
+LIBBPF_API void bpf_link__disconnect(struct bpf_link *link);
 LIBBPF_API int bpf_link__destroy(struct bpf_link *link);
 
+LIBBPF_API struct bpf_link *
+bpf_program__attach(struct bpf_program *prog);
 LIBBPF_API struct bpf_link *
 bpf_program__attach_perf_event(struct bpf_program *prog, int pfd);
 LIBBPF_API struct bpf_link *
@@ -186,6 +241,10 @@ LIBBPF_API struct bpf_link *
 bpf_program__attach_raw_tracepoint(struct bpf_program *prog,
 				   const char *tp_name);
 
+LIBBPF_API struct bpf_link *
+bpf_program__attach_trace(struct bpf_program *prog);
+struct bpf_map;
+LIBBPF_API struct bpf_link *bpf_map__attach_struct_ops(struct bpf_map *map);
 struct bpf_insn;
 
 /*
@@ -261,8 +320,16 @@ LIBBPF_API int bpf_program__set_sched_cls(struct bpf_program *prog);
 LIBBPF_API int bpf_program__set_sched_act(struct bpf_program *prog);
 LIBBPF_API int bpf_program__set_xdp(struct bpf_program *prog);
 LIBBPF_API int bpf_program__set_perf_event(struct bpf_program *prog);
+LIBBPF_API int bpf_program__set_tracing(struct bpf_program *prog);
+LIBBPF_API int bpf_program__set_struct_ops(struct bpf_program *prog);
+LIBBPF_API int bpf_program__set_extension(struct bpf_program *prog);
+
+LIBBPF_API enum bpf_prog_type bpf_program__get_type(struct bpf_program *prog);
 LIBBPF_API void bpf_program__set_type(struct bpf_program *prog,
 				      enum bpf_prog_type type);
+
+LIBBPF_API enum bpf_attach_type
+bpf_program__get_expected_attach_type(struct bpf_program *prog);
 LIBBPF_API void
 bpf_program__set_expected_attach_type(struct bpf_program *prog,
 				      enum bpf_attach_type type);
@@ -275,6 +342,9 @@ LIBBPF_API bool bpf_program__is_sched_cls(const struct bpf_program *prog);
 LIBBPF_API bool bpf_program__is_sched_act(const struct bpf_program *prog);
 LIBBPF_API bool bpf_program__is_xdp(const struct bpf_program *prog);
 LIBBPF_API bool bpf_program__is_perf_event(const struct bpf_program *prog);
+LIBBPF_API bool bpf_program__is_tracing(const struct bpf_program *prog);
+LIBBPF_API bool bpf_program__is_struct_ops(const struct bpf_program *prog);
+LIBBPF_API bool bpf_program__is_extension(const struct bpf_program *prog);
 
 /*
  * No need for __attribute__((packed)), all members of 'bpf_map_def'
@@ -294,7 +364,6 @@ struct bpf_map_def {
  * The 'struct bpf_map' in include/linux/bpf.h is internal to the kernel,
  * so no need to worry about a name clash.
  */
-struct bpf_map;
 LIBBPF_API struct bpf_map *
 bpf_object__find_map_by_name(const struct bpf_object *obj, const char *name);
 
@@ -334,6 +403,9 @@ LIBBPF_API int bpf_map__resize(struct bpf_map *map, __u32 max_entries);
 LIBBPF_API bool bpf_map__is_offload_neutral(const struct bpf_map *map);
 LIBBPF_API bool bpf_map__is_internal(const struct bpf_map *map);
 LIBBPF_API void bpf_map__set_ifindex(struct bpf_map *map, __u32 ifindex);
+LIBBPF_API int bpf_map__set_pin_path(struct bpf_map *map, const char *path);
+LIBBPF_API const char *bpf_map__get_pin_path(const struct bpf_map *map);
+LIBBPF_API bool bpf_map__is_pinned(const struct bpf_map *map);
 LIBBPF_API int bpf_map__pin(struct bpf_map *map, const char *path);
 LIBBPF_API int bpf_map__unpin(struct bpf_map *map, const char *path);
 
@@ -355,8 +427,18 @@ LIBBPF_API int bpf_prog_load_xattr(const struct bpf_prog_load_attr *attr,
 LIBBPF_API int bpf_prog_load(const char *file, enum bpf_prog_type type,
 			     struct bpf_object **pobj, int *prog_fd);
 
+struct xdp_link_info {
+	__u32 prog_id;
+	__u32 drv_prog_id;
+	__u32 hw_prog_id;
+	__u32 skb_prog_id;
+	__u8 attach_mode;
+};
+
 LIBBPF_API int bpf_set_link_xdp_fd(int ifindex, int fd, __u32 flags);
 LIBBPF_API int bpf_get_link_xdp_id(int ifindex, __u32 *prog_id, __u32 flags);
+LIBBPF_API int bpf_get_link_xdp_info(int ifindex, struct xdp_link_info *info,
+				     size_t info_size, __u32 flags);
 
 struct perf_buffer;
 
@@ -422,18 +504,6 @@ bpf_perf_event_read_simple(void *mmap_mem, size_t mmap_size, size_t page_size,
 			   void **copy_mem, size_t *copy_size,
 			   bpf_perf_event_print_t fn, void *private_data);
 
-struct nlattr;
-typedef int (*libbpf_dump_nlmsg_t)(void *cookie, void *msg, struct nlattr **tb);
-int libbpf_netlink_open(unsigned int *nl_pid);
-int libbpf_nl_get_link(int sock, unsigned int nl_pid,
-		       libbpf_dump_nlmsg_t dump_link_nlmsg, void *cookie);
-int libbpf_nl_get_class(int sock, unsigned int nl_pid, int ifindex,
-			libbpf_dump_nlmsg_t dump_class_nlmsg, void *cookie);
-int libbpf_nl_get_qdisc(int sock, unsigned int nl_pid, int ifindex,
-			libbpf_dump_nlmsg_t dump_qdisc_nlmsg, void *cookie);
-int libbpf_nl_get_filter(int sock, unsigned int nl_pid, int ifindex, int handle,
-			 libbpf_dump_nlmsg_t dump_filter_nlmsg, void *cookie);
-
 struct bpf_prog_linfo;
 struct bpf_prog_info;
 
@@ -460,6 +530,7 @@ LIBBPF_API bool bpf_probe_prog_type(enum bpf_prog_type prog_type,
 LIBBPF_API bool bpf_probe_map_type(enum bpf_map_type map_type, __u32 ifindex);
 LIBBPF_API bool bpf_probe_helper(enum bpf_func_id id,
 				 enum bpf_prog_type prog_type, __u32 ifindex);
+LIBBPF_API bool bpf_probe_large_insn_limit(__u32 ifindex);
 
 /*
  * Get bpf_prog_info in continuous memory
@@ -540,6 +611,50 @@ bpf_program__bpil_offs_to_addr(struct bpf_prog_info_linear *info_linear);
  */
 LIBBPF_API int libbpf_num_possible_cpus(void);
 
+struct bpf_map_skeleton {
+	const char *name;
+	struct bpf_map **map;
+	void **mmaped;
+};
+
+struct bpf_prog_skeleton {
+	const char *name;
+	struct bpf_program **prog;
+	struct bpf_link **link;
+};
+
+struct bpf_object_skeleton {
+	size_t sz; /* size of this struct, for forward/backward compatibility */
+
+	const char *name;
+	void *data;
+	size_t data_sz;
+
+	struct bpf_object **obj;
+
+	int map_cnt;
+	int map_skel_sz; /* sizeof(struct bpf_skeleton_map) */
+	struct bpf_map_skeleton *maps;
+
+	int prog_cnt;
+	int prog_skel_sz; /* sizeof(struct bpf_skeleton_prog) */
+	struct bpf_prog_skeleton *progs;
+};
+
+LIBBPF_API int
+bpf_object__open_skeleton(struct bpf_object_skeleton *s,
+			  const struct bpf_object_open_opts *opts);
+LIBBPF_API int bpf_object__load_skeleton(struct bpf_object_skeleton *s);
+LIBBPF_API int bpf_object__attach_skeleton(struct bpf_object_skeleton *s);
+LIBBPF_API void bpf_object__detach_skeleton(struct bpf_object_skeleton *s);
+LIBBPF_API void bpf_object__destroy_skeleton(struct bpf_object_skeleton *s);
+
+enum libbpf_tristate {
+	TRI_NO = 0,
+	TRI_YES = 1,
+	TRI_MODULE = 2,
+};
+
 #ifdef __cplusplus
 } /* extern "C" */
 #endif

src/libbpf.map

@@ -183,4 +183,55 @@ LIBBPF_0.0.4 {
 		perf_buffer__new;
 		perf_buffer__new_raw;
 		perf_buffer__poll;
+		xsk_umem__create;
 } LIBBPF_0.0.3;
+
+LIBBPF_0.0.5 {
+	global:
+		bpf_btf_get_next_id;
+} LIBBPF_0.0.4;
+
+LIBBPF_0.0.6 {
+	global:
+		bpf_get_link_xdp_info;
+		bpf_map__get_pin_path;
+		bpf_map__is_pinned;
+		bpf_map__set_pin_path;
+		bpf_object__open_file;
+		bpf_object__open_mem;
+		bpf_program__attach_trace;
+		bpf_program__get_expected_attach_type;
+		bpf_program__get_type;
+		bpf_program__is_tracing;
+		bpf_program__set_tracing;
+		bpf_program__size;
+		btf__find_by_name_kind;
+		libbpf_find_vmlinux_btf_id;
+} LIBBPF_0.0.5;
+
+LIBBPF_0.0.7 {
+	global:
+		btf_dump__emit_type_decl;
+		bpf_link__disconnect;
+		bpf_map__attach_struct_ops;
+		bpf_map_delete_batch;
+		bpf_map_lookup_and_delete_batch;
+		bpf_map_lookup_batch;
+		bpf_map_update_batch;
+		bpf_object__find_program_by_name;
+		bpf_object__attach_skeleton;
+		bpf_object__destroy_skeleton;
+		bpf_object__detach_skeleton;
+		bpf_object__load_skeleton;
+		bpf_object__open_skeleton;
+		bpf_probe_large_insn_limit;
+		bpf_prog_attach_xattr;
+		bpf_program__attach;
+		bpf_program__name;
+		bpf_program__is_extension;
+		bpf_program__is_struct_ops;
+		bpf_program__set_extension;
+		bpf_program__set_struct_ops;
+		btf__align_of;
+		libbpf_find_kernel_btf;
+} LIBBPF_0.0.6;

src/libbpf.pc.template

@@ -8,5 +8,5 @@ Name: libbpf
 Description: BPF library
 Version: @VERSION@
 Libs: -L${libdir} -lbpf
-Requires.private: libelf
+Requires.private: libelf zlib
 Cflags: -I${includedir}

src/libbpf_common.h

@@ -0,0 +1,40 @@
+/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
+
+/*
+ * Common user-facing libbpf helpers.
+ *
+ * Copyright (c) 2019 Facebook
+ */
+
+#ifndef __LIBBPF_LIBBPF_COMMON_H
+#define __LIBBPF_LIBBPF_COMMON_H
+
+#include <string.h>
+
+#ifndef LIBBPF_API
+#define LIBBPF_API __attribute__((visibility("default")))
+#endif
+
+/* Helper macro to declare and initialize libbpf options struct
+ *
+ * This dance with uninitialized declaration, followed by memset to zero,
+ * followed by assignment using compound literal syntax is done to preserve
+ * ability to use a nice struct field initialization syntax and **hopefully**
+ * have all the padding bytes initialized to zero. It's not guaranteed though,
+ * when copying literal, that compiler won't copy garbage in literal's padding
+ * bytes, but that's the best way I've found and it seems to work in practice.
+ *
+ * Macro declares opts struct of given type and name, zero-initializes,
+ * including any extra padding, it with memset() and then assigns initial
+ * values provided by users in struct initializer-syntax as varargs.
+ */
+#define DECLARE_LIBBPF_OPTS(TYPE, NAME, ...)				    \
+	struct TYPE NAME = ({ 						    \
+		memset(&NAME, 0, sizeof(struct TYPE));			    \
+		(struct TYPE) {						    \
+			.sz = sizeof(struct TYPE),			    \
+			__VA_ARGS__					    \
+		};							    \
+	})
+
+#endif /* __LIBBPF_LIBBPF_COMMON_H */

src/libbpf_errno.c

@@ -13,6 +13,9 @@
 
 #include "libbpf.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 #define ERRNO_OFFSET(e)		((e) - __LIBBPF_ERRNO__START)
 #define ERRCODE_OFFSET(c)	ERRNO_OFFSET(LIBBPF_ERRNO__##c)
 #define NR_ERRNO	(__LIBBPF_ERRNO__END - __LIBBPF_ERRNO__START)

src/libbpf_internal.h

@@ -29,6 +29,26 @@
 #ifndef max
 # define max(x, y) ((x) < (y) ? (y) : (x))
 #endif
+#ifndef offsetofend
+# define offsetofend(TYPE, FIELD) \
+	(offsetof(TYPE, FIELD) + sizeof(((TYPE *)0)->FIELD))
+#endif
+
+/* Symbol versioning is different between static and shared library.
+ * Properly versioned symbols are needed for shared library, but
+ * only the symbol of the new version is needed for static library.
+ */
+#ifdef SHARED
+# define COMPAT_VERSION(internal_name, api_name, version) \
+	asm(".symver " #internal_name "," #api_name "@" #version);
+# define DEFAULT_VERSION(internal_name, api_name, version) \
+	asm(".symver " #internal_name "," #api_name "@@" #version);
+#else
+# define COMPAT_VERSION(internal_name, api_name, version)
+# define DEFAULT_VERSION(internal_name, api_name, version) \
+	extern typeof(internal_name) api_name \
+	__attribute__((alias(#internal_name)));
+#endif
 
 extern void libbpf_print(enum libbpf_print_level level,
 			 const char *format, ...)
@@ -39,11 +59,178 @@ do {				\
 	libbpf_print(level, "libbpf: " fmt, ##__VA_ARGS__);	\
 } while (0)
 
-#define pr_warning(fmt, ...)	__pr(LIBBPF_WARN, fmt, ##__VA_ARGS__)
+#define pr_warn(fmt, ...)	__pr(LIBBPF_WARN, fmt, ##__VA_ARGS__)
 #define pr_info(fmt, ...)	__pr(LIBBPF_INFO, fmt, ##__VA_ARGS__)
 #define pr_debug(fmt, ...)	__pr(LIBBPF_DEBUG, fmt, ##__VA_ARGS__)
 
+static inline bool libbpf_validate_opts(const char *opts,
+					size_t opts_sz, size_t user_sz,
+					const char *type_name)
+{
+	if (user_sz < sizeof(size_t)) {
+		pr_warn("%s size (%zu) is too small\n", type_name, user_sz);
+		return false;
+	}
+	if (user_sz > opts_sz) {
+		size_t i;
+
+		for (i = opts_sz; i < user_sz; i++) {
+			if (opts[i]) {
+				pr_warn("%s has non-zero extra bytes\n",
+					type_name);
+				return false;
+			}
+		}
+	}
+	return true;
+}
+
+#define OPTS_VALID(opts, type)						      \
+	(!(opts) || libbpf_validate_opts((const char *)opts,		      \
+					 offsetofend(struct type,	      \
+						     type##__last_field),     \
+					 (opts)->sz, #type))
+#define OPTS_HAS(opts, field) \
+	((opts) && opts->sz >= offsetofend(typeof(*(opts)), field))
+#define OPTS_GET(opts, field, fallback_value) \
+	(OPTS_HAS(opts, field) ? (opts)->field : fallback_value)
+
+int parse_cpu_mask_str(const char *s, bool **mask, int *mask_sz);
+int parse_cpu_mask_file(const char *fcpu, bool **mask, int *mask_sz);
 int libbpf__load_raw_btf(const char *raw_types, size_t types_len,
 			 const char *str_sec, size_t str_len);
 
+int bpf_object__section_size(const struct bpf_object *obj, const char *name,
+			     __u32 *size);
+int bpf_object__variable_offset(const struct bpf_object *obj, const char *name,
+				__u32 *off);
+
+struct nlattr;
+typedef int (*libbpf_dump_nlmsg_t)(void *cookie, void *msg, struct nlattr **tb);
+int libbpf_netlink_open(unsigned int *nl_pid);
+int libbpf_nl_get_link(int sock, unsigned int nl_pid,
+		       libbpf_dump_nlmsg_t dump_link_nlmsg, void *cookie);
+int libbpf_nl_get_class(int sock, unsigned int nl_pid, int ifindex,
+			libbpf_dump_nlmsg_t dump_class_nlmsg, void *cookie);
+int libbpf_nl_get_qdisc(int sock, unsigned int nl_pid, int ifindex,
+			libbpf_dump_nlmsg_t dump_qdisc_nlmsg, void *cookie);
+int libbpf_nl_get_filter(int sock, unsigned int nl_pid, int ifindex, int handle,
+			 libbpf_dump_nlmsg_t dump_filter_nlmsg, void *cookie);
+
+struct btf_ext_info {
+	/*
+	 * info points to the individual info section (e.g. func_info and
+	 * line_info) from the .BTF.ext. It does not include the __u32 rec_size.
+	 */
+	void *info;
+	__u32 rec_size;
+	__u32 len;
+};
+
+#define for_each_btf_ext_sec(seg, sec)					\
+	for (sec = (seg)->info;						\
+	     (void *)sec < (seg)->info + (seg)->len;			\
+	     sec = (void *)sec + sizeof(struct btf_ext_info_sec) +	\
+		   (seg)->rec_size * sec->num_info)
+
+#define for_each_btf_ext_rec(seg, sec, i, rec)				\
+	for (i = 0, rec = (void *)&(sec)->data;				\
+	     i < (sec)->num_info;					\
+	     i++, rec = (void *)rec + (seg)->rec_size)
+
+struct btf_ext {
+	union {
+		struct btf_ext_header *hdr;
+		void *data;
+	};
+	struct btf_ext_info func_info;
+	struct btf_ext_info line_info;
+	struct btf_ext_info field_reloc_info;
+	__u32 data_size;
+};
+
+struct btf_ext_info_sec {
+	__u32	sec_name_off;
+	__u32	num_info;
+	/* Followed by num_info * record_size number of bytes */
+	__u8	data[0];
+};
+
+/* The minimum bpf_func_info checked by the loader */
+struct bpf_func_info_min {
+	__u32   insn_off;
+	__u32   type_id;
+};
+
+/* The minimum bpf_line_info checked by the loader */
+struct bpf_line_info_min {
+	__u32	insn_off;
+	__u32	file_name_off;
+	__u32	line_off;
+	__u32	line_col;
+};
+
+/* bpf_field_info_kind encodes which aspect of captured field has to be
+ * adjusted by relocations. Currently supported values are:
+ *   - BPF_FIELD_BYTE_OFFSET: field offset (in bytes);
+ *   - BPF_FIELD_EXISTS: field existence (1, if field exists; 0, otherwise);
+ */
+enum bpf_field_info_kind {
+	BPF_FIELD_BYTE_OFFSET = 0,	/* field byte offset */
+	BPF_FIELD_BYTE_SIZE = 1,
+	BPF_FIELD_EXISTS = 2,		/* field existence in target kernel */
+	BPF_FIELD_SIGNED = 3,
+	BPF_FIELD_LSHIFT_U64 = 4,
+	BPF_FIELD_RSHIFT_U64 = 5,
+};
+
+/* The minimum bpf_field_reloc checked by the loader
+ *
+ * Field relocation captures the following data:
+ * - insn_off - instruction offset (in bytes) within a BPF program that needs
+ *   its insn->imm field to be relocated with actual field info;
+ * - type_id - BTF type ID of the "root" (containing) entity of a relocatable
+ *   field;
+ * - access_str_off - offset into corresponding .BTF string section. String
+ *   itself encodes an accessed field using a sequence of field and array
+ *   indicies, separated by colon (:). It's conceptually very close to LLVM's
+ *   getelementptr ([0]) instruction's arguments for identifying offset to 
+ *   a field.
+ *
+ * Example to provide a better feel.
+ *
+ *   struct sample {
+ *       int a;
+ *       struct {
+ *           int b[10];
+ *       };
+ *   };
+ *
+ *   struct sample *s = ...;
+ *   int x = &s->a;     // encoded as "0:0" (a is field #0)
+ *   int y = &s->b[5];  // encoded as "0:1:0:5" (anon struct is field #1, 
+ *                      // b is field #0 inside anon struct, accessing elem #5)
+ *   int z = &s[10]->b; // encoded as "10:1" (ptr is used as an array)
+ *
+ * type_id for all relocs in this example  will capture BTF type id of
+ * `struct sample`.
+ *
+ * Such relocation is emitted when using __builtin_preserve_access_index()
+ * Clang built-in, passing expression that captures field address, e.g.:
+ *
+ * bpf_probe_read(&dst, sizeof(dst),
+ *		  __builtin_preserve_access_index(&src->a.b.c));
+ *
+ * In this case Clang will emit field relocation recording necessary data to
+ * be able to find offset of embedded `a.b.c` field within `src` struct.
+ *
+ *   [0] https://llvm.org/docs/LangRef.html#getelementptr-instruction
+ */
+struct bpf_field_reloc {
+	__u32   insn_off;
+	__u32   type_id;
+	__u32   access_str_off;
+	enum bpf_field_info_kind kind;
+};
+
 #endif /* __LIBBPF_LIBBPF_INTERNAL_H */

src/libbpf_probes.c

@@ -17,6 +17,9 @@
 #include "libbpf.h"
 #include "libbpf_internal.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 static bool grep(const char *buffer, const char *pattern)
 {
 	return !!strstr(buffer, pattern);
@@ -102,6 +105,9 @@ probe_load(enum bpf_prog_type prog_type, const struct bpf_insn *insns,
 	case BPF_PROG_TYPE_FLOW_DISSECTOR:
 	case BPF_PROG_TYPE_CGROUP_SYSCTL:
 	case BPF_PROG_TYPE_CGROUP_SOCKOPT:
+	case BPF_PROG_TYPE_TRACING:
+	case BPF_PROG_TYPE_STRUCT_OPS:
+	case BPF_PROG_TYPE_EXT:
 	default:
 		break;
 	}
@@ -244,11 +250,13 @@ bool bpf_probe_map_type(enum bpf_map_type map_type, __u32 ifindex)
 	case BPF_MAP_TYPE_ARRAY_OF_MAPS:
 	case BPF_MAP_TYPE_HASH_OF_MAPS:
 	case BPF_MAP_TYPE_DEVMAP:
+	case BPF_MAP_TYPE_DEVMAP_HASH:
 	case BPF_MAP_TYPE_SOCKMAP:
 	case BPF_MAP_TYPE_CPUMAP:
 	case BPF_MAP_TYPE_XSKMAP:
 	case BPF_MAP_TYPE_SOCKHASH:
 	case BPF_MAP_TYPE_REUSEPORT_SOCKARRAY:
+	case BPF_MAP_TYPE_STRUCT_OPS:
 	default:
 		break;
 	}
@@ -319,3 +327,24 @@ bool bpf_probe_helper(enum bpf_func_id id, enum bpf_prog_type prog_type,
 
 	return res;
 }
+
+/*
+ * Probe for availability of kernel commit (5.3):
+ *
+ * c04c0d2b968a ("bpf: increase complexity limit and maximum program size")
+ */
+bool bpf_probe_large_insn_limit(__u32 ifindex)
+{
+	struct bpf_insn insns[BPF_MAXINSNS + 1];
+	int i;
+
+	for (i = 0; i < BPF_MAXINSNS; i++)
+		insns[i] = BPF_MOV64_IMM(BPF_REG_0, 1);
+	insns[BPF_MAXINSNS] = BPF_EXIT_INSN();
+
+	errno = 0;
+	probe_load(BPF_PROG_TYPE_SCHED_CLS, insns, ARRAY_SIZE(insns), NULL, 0,
+		   ifindex);
+
+	return errno != E2BIG && errno != EINVAL;
+}

src/netlink.c

@@ -12,8 +12,12 @@
 
 #include "bpf.h"
 #include "libbpf.h"
+#include "libbpf_internal.h"
 #include "nlattr.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 #ifndef SOL_NETLINK
 #define SOL_NETLINK 270
 #endif
@@ -24,7 +28,7 @@ typedef int (*__dump_nlmsg_t)(struct nlmsghdr *nlmsg, libbpf_dump_nlmsg_t,
 struct xdp_id_md {
 	int ifindex;
 	__u32 flags;
-	__u32 id;
+	struct xdp_link_info info;
 };
 
 int libbpf_netlink_open(__u32 *nl_pid)
@@ -43,7 +47,7 @@ int libbpf_netlink_open(__u32 *nl_pid)
 
 	if (setsockopt(sock, SOL_NETLINK, NETLINK_EXT_ACK,
 		       &one, sizeof(one)) < 0) {
-		fprintf(stderr, "Netlink error reporting not supported\n");
+		pr_warn("Netlink error reporting not supported\n");
 	}
 
 	if (bind(sock, (struct sockaddr *)&sa, sizeof(sa)) < 0) {
@@ -202,26 +206,11 @@ static int __dump_link_nlmsg(struct nlmsghdr *nlh,
 	return dump_link_nlmsg(cookie, ifi, tb);
 }
 
-static unsigned char get_xdp_id_attr(unsigned char mode, __u32 flags)
+static int get_xdp_info(void *cookie, void *msg, struct nlattr **tb)
-{
-	if (mode != XDP_ATTACHED_MULTI)
-		return IFLA_XDP_PROG_ID;
-	if (flags & XDP_FLAGS_DRV_MODE)
-		return IFLA_XDP_DRV_PROG_ID;
-	if (flags & XDP_FLAGS_HW_MODE)
-		return IFLA_XDP_HW_PROG_ID;
-	if (flags & XDP_FLAGS_SKB_MODE)
-		return IFLA_XDP_SKB_PROG_ID;
-
-	return IFLA_XDP_UNSPEC;
-}
-
-static int get_xdp_id(void *cookie, void *msg, struct nlattr **tb)
 {
 	struct nlattr *xdp_tb[IFLA_XDP_MAX + 1];
 	struct xdp_id_md *xdp_id = cookie;
 	struct ifinfomsg *ifinfo = msg;
-	unsigned char mode, xdp_attr;
 	int ret;
 
 	if (xdp_id->ifindex && xdp_id->ifindex != ifinfo->ifi_index)
@@ -237,27 +226,40 @@ static int get_xdp_id(void *cookie, void *msg, struct nlattr **tb)
 	if (!xdp_tb[IFLA_XDP_ATTACHED])
 		return 0;
 
-	mode = libbpf_nla_getattr_u8(xdp_tb[IFLA_XDP_ATTACHED]);
+	xdp_id->info.attach_mode = libbpf_nla_getattr_u8(
-	if (mode == XDP_ATTACHED_NONE)
+		xdp_tb[IFLA_XDP_ATTACHED]);
+
+	if (xdp_id->info.attach_mode == XDP_ATTACHED_NONE)
 		return 0;
 
-	xdp_attr = get_xdp_id_attr(mode, xdp_id->flags);
+	if (xdp_tb[IFLA_XDP_PROG_ID])
-	if (!xdp_attr || !xdp_tb[xdp_attr])
+		xdp_id->info.prog_id = libbpf_nla_getattr_u32(
-		return 0;
+			xdp_tb[IFLA_XDP_PROG_ID]);
 
-	xdp_id->id = libbpf_nla_getattr_u32(xdp_tb[xdp_attr]);
+	if (xdp_tb[IFLA_XDP_SKB_PROG_ID])
+		xdp_id->info.skb_prog_id = libbpf_nla_getattr_u32(
+			xdp_tb[IFLA_XDP_SKB_PROG_ID]);
+
+	if (xdp_tb[IFLA_XDP_DRV_PROG_ID])
+		xdp_id->info.drv_prog_id = libbpf_nla_getattr_u32(
+			xdp_tb[IFLA_XDP_DRV_PROG_ID]);
+
+	if (xdp_tb[IFLA_XDP_HW_PROG_ID])
+		xdp_id->info.hw_prog_id = libbpf_nla_getattr_u32(
+			xdp_tb[IFLA_XDP_HW_PROG_ID]);
 
 	return 0;
 }
 
-int bpf_get_link_xdp_id(int ifindex, __u32 *prog_id, __u32 flags)
+int bpf_get_link_xdp_info(int ifindex, struct xdp_link_info *info,
+			  size_t info_size, __u32 flags)
 {
 	struct xdp_id_md xdp_id = {};
 	int sock, ret;
 	__u32 nl_pid;
 	__u32 mask;
 
-	if (flags & ~XDP_FLAGS_MASK)
+	if (flags & ~XDP_FLAGS_MASK || !info_size)
 		return -EINVAL;
 
 	/* Check whether the single {HW,DRV,SKB} mode is set */
@@ -273,14 +275,44 @@ int bpf_get_link_xdp_id(int ifindex, __u32 *prog_id, __u32 flags)
 	xdp_id.ifindex = ifindex;
 	xdp_id.flags = flags;
 
-	ret = libbpf_nl_get_link(sock, nl_pid, get_xdp_id, &xdp_id);
+	ret = libbpf_nl_get_link(sock, nl_pid, get_xdp_info, &xdp_id);
-	if (!ret)
+	if (!ret) {
-		*prog_id = xdp_id.id;
+		size_t sz = min(info_size, sizeof(xdp_id.info));
+
+		memcpy(info, &xdp_id.info, sz);
+		memset((void *) info + sz, 0, info_size - sz);
+	}
 
 	close(sock);
 	return ret;
 }
 
+static __u32 get_xdp_id(struct xdp_link_info *info, __u32 flags)
+{
+	if (info->attach_mode != XDP_ATTACHED_MULTI)
+		return info->prog_id;
+	if (flags & XDP_FLAGS_DRV_MODE)
+		return info->drv_prog_id;
+	if (flags & XDP_FLAGS_HW_MODE)
+		return info->hw_prog_id;
+	if (flags & XDP_FLAGS_SKB_MODE)
+		return info->skb_prog_id;
+
+	return 0;
+}
+
+int bpf_get_link_xdp_id(int ifindex, __u32 *prog_id, __u32 flags)
+{
+	struct xdp_link_info info;
+	int ret;
+
+	ret = bpf_get_link_xdp_info(ifindex, &info, sizeof(info), flags);
+	if (!ret)
+		*prog_id = get_xdp_id(&info, flags);
+
+	return ret;
+}
+
 int libbpf_nl_get_link(int sock, unsigned int nl_pid,
 		       libbpf_dump_nlmsg_t dump_link_nlmsg, void *cookie)
 {

src/nlattr.c

@@ -8,10 +8,14 @@
 
 #include <errno.h>
 #include "nlattr.h"
+#include "libbpf_internal.h"
 #include <linux/rtnetlink.h>
 #include <string.h>
 #include <stdio.h>
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 static uint16_t nla_attr_minlen[LIBBPF_NLA_TYPE_MAX+1] = {
 	[LIBBPF_NLA_U8]		= sizeof(uint8_t),
 	[LIBBPF_NLA_U16]	= sizeof(uint16_t),
@@ -121,8 +125,8 @@ int libbpf_nla_parse(struct nlattr *tb[], int maxtype, struct nlattr *head,
 		}
 
 		if (tb[type])
-			fprintf(stderr, "Attribute of type %#x found multiple times in message, "
+			pr_warn("Attribute of type %#x found multiple times in message, "
-				  "previous attribute is being ignored.\n", type);
+				"previous attribute is being ignored.\n", type);
 
 		tb[type] = nla;
 	}
@@ -181,15 +185,14 @@ int libbpf_nla_dump_errormsg(struct nlmsghdr *nlh)
 
 	if (libbpf_nla_parse(tb, NLMSGERR_ATTR_MAX, attr, alen,
 			     extack_policy) != 0) {
-		fprintf(stderr,
+		pr_warn("Failed to parse extended error attributes\n");
-			"Failed to parse extended error attributes\n");
 		return 0;
 	}
 
 	if (tb[NLMSGERR_ATTR_MSG])
 		errmsg = (char *) libbpf_nla_data(tb[NLMSGERR_ATTR_MSG]);
 
-	fprintf(stderr, "Kernel error message: %s\n", errmsg);
+	pr_warn("Kernel error message: %s\n", errmsg);
 
 	return 0;
 }

src/str_error.c

@@ -4,6 +4,9 @@
 #include <stdio.h>
 #include "str_error.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 /*
  * Wrapper to allow for building in non-GNU systems such as Alpine Linux's musl
  * libc, while checking strerror_r() return to avoid having to check this in

src/xsk.c

@@ -32,6 +32,9 @@
 #include "libbpf_internal.h"
 #include "xsk.h"
 
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
 #ifndef SOL_XDP
  #define SOL_XDP 283
 #endif
@@ -65,7 +68,6 @@ struct xsk_socket {
 	int xsks_map_fd;
 	__u32 queue_id;
 	char ifname[IFNAMSIZ];
-	bool zc;
 };
 
 struct xsk_nl_info {
@@ -74,22 +76,20 @@ struct xsk_nl_info {
 	int fd;
 };
 
-/* For 32-bit systems, we need to use mmap2 as the offsets are 64-bit.
+/* Up until and including Linux 5.3 */
- * Unfortunately, it is not part of glibc.
+struct xdp_ring_offset_v1 {
- */
+	__u64 producer;
-static inline void *xsk_mmap(void *addr, size_t length, int prot, int flags,
+	__u64 consumer;
-			     int fd, __u64 offset)
+	__u64 desc;
-{
+};
-#ifdef __NR_mmap2
-	unsigned int page_shift = __builtin_ffs(getpagesize()) - 1;
-	long ret = syscall(__NR_mmap2, addr, length, prot, flags, fd,
-			   (off_t)(offset >> page_shift));
 
-	return (void *)ret;
+/* Up until and including Linux 5.3 */
-#else
+struct xdp_mmap_offsets_v1 {
-	return mmap(addr, length, prot, flags, fd, offset);
+	struct xdp_ring_offset_v1 rx;
-#endif
+	struct xdp_ring_offset_v1 tx;
-}
+	struct xdp_ring_offset_v1 fr;
+	struct xdp_ring_offset_v1 cr;
+};
 
 int xsk_umem__fd(const struct xsk_umem *umem)
 {
@@ -116,6 +116,7 @@ static void xsk_set_umem_config(struct xsk_umem_config *cfg,
 		cfg->comp_size = XSK_RING_CONS__DEFAULT_NUM_DESCS;
 		cfg->frame_size = XSK_UMEM__DEFAULT_FRAME_SIZE;
 		cfg->frame_headroom = XSK_UMEM__DEFAULT_FRAME_HEADROOM;
+		cfg->flags = XSK_UMEM__DEFAULT_FLAGS;
 		return;
 	}
 
@@ -123,6 +124,7 @@ static void xsk_set_umem_config(struct xsk_umem_config *cfg,
 	cfg->comp_size = usr_cfg->comp_size;
 	cfg->frame_size = usr_cfg->frame_size;
 	cfg->frame_headroom = usr_cfg->frame_headroom;
+	cfg->flags = usr_cfg->flags;
 }
 
 static int xsk_set_xdp_socket_config(struct xsk_socket_config *cfg,
@@ -149,14 +151,66 @@ static int xsk_set_xdp_socket_config(struct xsk_socket_config *cfg,
 	return 0;
 }
 
-int xsk_umem__create(struct xsk_umem **umem_ptr, void *umem_area, __u64 size,
+static void xsk_mmap_offsets_v1(struct xdp_mmap_offsets *off)
-		     struct xsk_ring_prod *fill, struct xsk_ring_cons *comp,
+{
-		     const struct xsk_umem_config *usr_config)
+	struct xdp_mmap_offsets_v1 off_v1;
+
+	/* getsockopt on a kernel <= 5.3 has no flags fields.
+	 * Copy over the offsets to the correct places in the >=5.4 format
+	 * and put the flags where they would have been on that kernel.
+	 */
+	memcpy(&off_v1, off, sizeof(off_v1));
+
+	off->rx.producer = off_v1.rx.producer;
+	off->rx.consumer = off_v1.rx.consumer;
+	off->rx.desc = off_v1.rx.desc;
+	off->rx.flags = off_v1.rx.consumer + sizeof(__u32);
+
+	off->tx.producer = off_v1.tx.producer;
+	off->tx.consumer = off_v1.tx.consumer;
+	off->tx.desc = off_v1.tx.desc;
+	off->tx.flags = off_v1.tx.consumer + sizeof(__u32);
+
+	off->fr.producer = off_v1.fr.producer;
+	off->fr.consumer = off_v1.fr.consumer;
+	off->fr.desc = off_v1.fr.desc;
+	off->fr.flags = off_v1.fr.consumer + sizeof(__u32);
+
+	off->cr.producer = off_v1.cr.producer;
+	off->cr.consumer = off_v1.cr.consumer;
+	off->cr.desc = off_v1.cr.desc;
+	off->cr.flags = off_v1.cr.consumer + sizeof(__u32);
+}
+
+static int xsk_get_mmap_offsets(int fd, struct xdp_mmap_offsets *off)
+{
+	socklen_t optlen;
+	int err;
+
+	optlen = sizeof(*off);
+	err = getsockopt(fd, SOL_XDP, XDP_MMAP_OFFSETS, off, &optlen);
+	if (err)
+		return err;
+
+	if (optlen == sizeof(*off))
+		return 0;
+
+	if (optlen == sizeof(struct xdp_mmap_offsets_v1)) {
+		xsk_mmap_offsets_v1(off);
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
+int xsk_umem__create_v0_0_4(struct xsk_umem **umem_ptr, void *umem_area,
+			    __u64 size, struct xsk_ring_prod *fill,
+			    struct xsk_ring_cons *comp,
+			    const struct xsk_umem_config *usr_config)
 {
 	struct xdp_mmap_offsets off;
 	struct xdp_umem_reg mr;
 	struct xsk_umem *umem;
-	socklen_t optlen;
 	void *map;
 	int err;
 
@@ -178,10 +232,12 @@ int xsk_umem__create(struct xsk_umem **umem_ptr, void *umem_area, __u64 size,
 	umem->umem_area = umem_area;
 	xsk_set_umem_config(&umem->config, usr_config);
 
+	memset(&mr, 0, sizeof(mr));
 	mr.addr = (uintptr_t)umem_area;
 	mr.len = size;
 	mr.chunk_size = umem->config.frame_size;
 	mr.headroom = umem->config.frame_headroom;
+	mr.flags = umem->config.flags;
 
 	err = setsockopt(umem->fd, SOL_XDP, XDP_UMEM_REG, &mr, sizeof(mr));
 	if (err) {
@@ -203,17 +259,15 @@ int xsk_umem__create(struct xsk_umem **umem_ptr, void *umem_area, __u64 size,
 		goto out_socket;
 	}
 
-	optlen = sizeof(off);
+	err = xsk_get_mmap_offsets(umem->fd, &off);
-	err = getsockopt(umem->fd, SOL_XDP, XDP_MMAP_OFFSETS, &off, &optlen);
 	if (err) {
 		err = -errno;
 		goto out_socket;
 	}
 
-	map = xsk_mmap(NULL, off.fr.desc +
+	map = mmap(NULL, off.fr.desc + umem->config.fill_size * sizeof(__u64),
-		       umem->config.fill_size * sizeof(__u64),
+		   PROT_READ | PROT_WRITE, MAP_SHARED | MAP_POPULATE, umem->fd,
-		       PROT_READ | PROT_WRITE, MAP_SHARED | MAP_POPULATE,
+		   XDP_UMEM_PGOFF_FILL_RING);
-		       umem->fd, XDP_UMEM_PGOFF_FILL_RING);
 	if (map == MAP_FAILED) {
 		err = -errno;
 		goto out_socket;
@@ -224,13 +278,13 @@ int xsk_umem__create(struct xsk_umem **umem_ptr, void *umem_area, __u64 size,
 	fill->size = umem->config.fill_size;
 	fill->producer = map + off.fr.producer;
 	fill->consumer = map + off.fr.consumer;
+	fill->flags = map + off.fr.flags;
 	fill->ring = map + off.fr.desc;
 	fill->cached_cons = umem->config.fill_size;
 
-	map = xsk_mmap(NULL,
+	map = mmap(NULL, off.cr.desc + umem->config.comp_size * sizeof(__u64),
-		       off.cr.desc + umem->config.comp_size * sizeof(__u64),
+		   PROT_READ | PROT_WRITE, MAP_SHARED | MAP_POPULATE, umem->fd,
-		       PROT_READ | PROT_WRITE, MAP_SHARED | MAP_POPULATE,
+		   XDP_UMEM_PGOFF_COMPLETION_RING);
-		       umem->fd, XDP_UMEM_PGOFF_COMPLETION_RING);
 	if (map == MAP_FAILED) {
 		err = -errno;
 		goto out_mmap;
@@ -241,6 +295,7 @@ int xsk_umem__create(struct xsk_umem **umem_ptr, void *umem_area, __u64 size,
 	comp->size = umem->config.comp_size;
 	comp->producer = map + off.cr.producer;
 	comp->consumer = map + off.cr.consumer;
+	comp->flags = map + off.cr.flags;
 	comp->ring = map + off.cr.desc;
 
 	*umem_ptr = umem;
@@ -255,6 +310,29 @@ out_umem_alloc:
 	return err;
 }
 
+struct xsk_umem_config_v1 {
+	__u32 fill_size;
+	__u32 comp_size;
+	__u32 frame_size;
+	__u32 frame_headroom;
+};
+
+int xsk_umem__create_v0_0_2(struct xsk_umem **umem_ptr, void *umem_area,
+			    __u64 size, struct xsk_ring_prod *fill,
+			    struct xsk_ring_cons *comp,
+			    const struct xsk_umem_config *usr_config)
+{
+	struct xsk_umem_config config;
+
+	memcpy(&config, usr_config, sizeof(struct xsk_umem_config_v1));
+	config.flags = 0;
+
+	return xsk_umem__create_v0_0_4(umem_ptr, umem_area, size, fill, comp,
+					&config);
+}
+COMPAT_VERSION(xsk_umem__create_v0_0_2, xsk_umem__create, LIBBPF_0.0.2)
+DEFAULT_VERSION(xsk_umem__create_v0_0_4, xsk_umem__create, LIBBPF_0.0.4)
+
 static int xsk_load_xdp_prog(struct xsk_socket *xsk)
 {
 	static const int log_buf_size = 16 * 1024;
@@ -264,33 +342,55 @@ static int xsk_load_xdp_prog(struct xsk_socket *xsk)
 	/* This is the C-program:
 	 * SEC("xdp_sock") int xdp_sock_prog(struct xdp_md *ctx)
 	 * {
-	 *     int index = ctx->rx_queue_index;
+	 *     int ret, index = ctx->rx_queue_index;
 	 *
 	 *     // A set entry here means that the correspnding queue_id
 	 *     // has an active AF_XDP socket bound to it.
+	 *     ret = bpf_redirect_map(&xsks_map, index, XDP_PASS);
+	 *     if (ret > 0)
+	 *         return ret;
+	 *
+	 *     // Fallback for pre-5.3 kernels, not supporting default
+	 *     // action in the flags parameter.
 	 *     if (bpf_map_lookup_elem(&xsks_map, &index))
 	 *         return bpf_redirect_map(&xsks_map, index, 0);
-	 *
 	 *     return XDP_PASS;
 	 * }
 	 */
 	struct bpf_insn prog[] = {
-		/* r1 = *(u32 *)(r1 + 16) */
+		/* r2 = *(u32 *)(r1 + 16) */
-		BPF_LDX_MEM(BPF_W, BPF_REG_1, BPF_REG_1, 16),
+		BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 16),
-		/* *(u32 *)(r10 - 4) = r1 */
+		/* *(u32 *)(r10 - 4) = r2 */
-		BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_1, -4),
+		BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_2, -4),
-		BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
+		/* r1 = xskmap[] */
-		BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
 		BPF_LD_MAP_FD(BPF_REG_1, xsk->xsks_map_fd),
+		/* r3 = XDP_PASS */
+		BPF_MOV64_IMM(BPF_REG_3, 2),
+		/* call bpf_redirect_map */
+		BPF_EMIT_CALL(BPF_FUNC_redirect_map),
+		/* if w0 != 0 goto pc+13 */
+		BPF_JMP32_IMM(BPF_JSGT, BPF_REG_0, 0, 13),
+		/* r2 = r10 */
+		BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
+		/* r2 += -4 */
+		BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
+		/* r1 = xskmap[] */
+		BPF_LD_MAP_FD(BPF_REG_1, xsk->xsks_map_fd),
+		/* call bpf_map_lookup_elem */
 		BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
+		/* r1 = r0 */
 		BPF_MOV64_REG(BPF_REG_1, BPF_REG_0),
-		BPF_MOV32_IMM(BPF_REG_0, 2),
+		/* r0 = XDP_PASS */
-		/* if r1 == 0 goto +5 */
+		BPF_MOV64_IMM(BPF_REG_0, 2),
+		/* if r1 == 0 goto pc+5 */
 		BPF_JMP_IMM(BPF_JEQ, BPF_REG_1, 0, 5),
 		/* r2 = *(u32 *)(r10 - 4) */
-		BPF_LD_MAP_FD(BPF_REG_1, xsk->xsks_map_fd),
 		BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_10, -4),
-		BPF_MOV32_IMM(BPF_REG_3, 0),
+		/* r1 = xskmap[] */
+		BPF_LD_MAP_FD(BPF_REG_1, xsk->xsks_map_fd),
+		/* r3 = 0 */
+		BPF_MOV64_IMM(BPF_REG_3, 0),
+		/* call bpf_redirect_map */
 		BPF_EMIT_CALL(BPF_FUNC_redirect_map),
 		/* The jumps are to this instruction */
 		BPF_EXIT_INSN(),
@@ -301,7 +401,7 @@ static int xsk_load_xdp_prog(struct xsk_socket *xsk)
 				   "LGPL-2.1 or BSD-2-Clause", 0, log_buf,
 				   log_buf_size);
 	if (prog_fd < 0) {
-		pr_warning("BPF log buffer:\n%s", log_buf);
+		pr_warn("BPF log buffer:\n%s", log_buf);
 		return prog_fd;
 	}
 
@@ -317,17 +417,16 @@ static int xsk_load_xdp_prog(struct xsk_socket *xsk)
 
 static int xsk_get_max_queues(struct xsk_socket *xsk)
 {
-	struct ethtool_channels channels;
+	struct ethtool_channels channels = { .cmd = ETHTOOL_GCHANNELS };
-	struct ifreq ifr;
+	struct ifreq ifr = {};
 	int fd, err, ret;
 
 	fd = socket(AF_INET, SOCK_DGRAM, 0);
 	if (fd < 0)
 		return -errno;
 
-	channels.cmd = ETHTOOL_GCHANNELS;
 	ifr.ifr_data = (void *)&channels;
-	strncpy(ifr.ifr_name, xsk->ifname, IFNAMSIZ - 1);
+	memcpy(ifr.ifr_name, xsk->ifname, IFNAMSIZ - 1);
 	ifr.ifr_name[IFNAMSIZ - 1] = '\0';
 	err = ioctl(fd, SIOCETHTOOL, &ifr);
 	if (err && errno != EOPNOTSUPP) {
@@ -335,13 +434,18 @@ static int xsk_get_max_queues(struct xsk_socket *xsk)
 		goto out;
 	}
 
-	if (channels.max_combined == 0 || errno == EOPNOTSUPP)
+	if (err) {
 		/* If the device says it has no channels, then all traffic
 		 * is sent to a single stream, so max queues = 1.
 		 */
 		ret = 1;
-	else
+	} else {
-		ret = channels.max_combined;
+		/* Take the max of rx, tx, combined. Drivers return
+		 * the number of channels in different ways.
+		 */
+		ret = max(channels.max_rx, channels.max_tx);
+		ret = max(ret, (int)channels.max_combined);
+	}
 
 out:
 	close(fd);
@@ -457,6 +561,8 @@ static int xsk_setup_xdp_prog(struct xsk_socket *xsk)
 		}
 	} else {
 		xsk->prog_fd = bpf_prog_get_fd_by_id(prog_id);
+		if (xsk->prog_fd < 0)
+			return -errno;
 		err = xsk_lookup_bpf_maps(xsk);
 		if (err) {
 			close(xsk->prog_fd);
@@ -464,7 +570,8 @@ static int xsk_setup_xdp_prog(struct xsk_socket *xsk)
 		}
 	}
 
-	err = xsk_set_bpf_maps(xsk);
+	if (xsk->rx)
+		err = xsk_set_bpf_maps(xsk);
 	if (err) {
 		xsk_delete_bpf_maps(xsk);
 		close(xsk->prog_fd);
@@ -482,23 +589,27 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
 	void *rx_map = NULL, *tx_map = NULL;
 	struct sockaddr_xdp sxdp = {};
 	struct xdp_mmap_offsets off;
-	struct xdp_options opts;
 	struct xsk_socket *xsk;
-	socklen_t optlen;
 	int err;
 
-	if (!umem || !xsk_ptr || !rx || !tx)
+	if (!umem || !xsk_ptr || !(rx || tx))
 		return -EFAULT;
 
-	if (umem->refcount) {
-		pr_warning("Error: shared umems not supported by libbpf.\n");
-		return -EBUSY;
-	}
-
 	xsk = calloc(1, sizeof(*xsk));
 	if (!xsk)
 		return -ENOMEM;
 
+	err = xsk_set_xdp_socket_config(&xsk->config, usr_config);
+	if (err)
+		goto out_xsk_alloc;
+
+	if (umem->refcount &&
+	    !(xsk->config.libbpf_flags & XSK_LIBBPF_FLAGS__INHIBIT_PROG_LOAD)) {
+		pr_warn("Error: shared umems not supported by libbpf supplied XDP program.\n");
+		err = -EBUSY;
+		goto out_xsk_alloc;
+	}
+
 	if (umem->refcount++ > 0) {
 		xsk->fd = socket(AF_XDP, SOCK_RAW, 0);
 		if (xsk->fd < 0) {
@@ -517,13 +628,9 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
 		err = -errno;
 		goto out_socket;
 	}
-	strncpy(xsk->ifname, ifname, IFNAMSIZ - 1);
+	memcpy(xsk->ifname, ifname, IFNAMSIZ - 1);
 	xsk->ifname[IFNAMSIZ - 1] = '\0';
 
-	err = xsk_set_xdp_socket_config(&xsk->config, usr_config);
-	if (err)
-		goto out_socket;
-
 	if (rx) {
 		err = setsockopt(xsk->fd, SOL_XDP, XDP_RX_RING,
 				 &xsk->config.rx_size,
@@ -543,19 +650,17 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
 		}
 	}
 
-	optlen = sizeof(off);
+	err = xsk_get_mmap_offsets(xsk->fd, &off);
-	err = getsockopt(xsk->fd, SOL_XDP, XDP_MMAP_OFFSETS, &off, &optlen);
 	if (err) {
 		err = -errno;
 		goto out_socket;
 	}
 
 	if (rx) {
-		rx_map = xsk_mmap(NULL, off.rx.desc +
+		rx_map = mmap(NULL, off.rx.desc +
-				  xsk->config.rx_size * sizeof(struct xdp_desc),
+			      xsk->config.rx_size * sizeof(struct xdp_desc),
-				  PROT_READ | PROT_WRITE,
+			      PROT_READ | PROT_WRITE, MAP_SHARED | MAP_POPULATE,
-				  MAP_SHARED | MAP_POPULATE,
+			      xsk->fd, XDP_PGOFF_RX_RING);
-				  xsk->fd, XDP_PGOFF_RX_RING);
 		if (rx_map == MAP_FAILED) {
 			err = -errno;
 			goto out_socket;
@@ -565,16 +670,16 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
 		rx->size = xsk->config.rx_size;
 		rx->producer = rx_map + off.rx.producer;
 		rx->consumer = rx_map + off.rx.consumer;
+		rx->flags = rx_map + off.rx.flags;
 		rx->ring = rx_map + off.rx.desc;
 	}
 	xsk->rx = rx;
 
 	if (tx) {
-		tx_map = xsk_mmap(NULL, off.tx.desc +
+		tx_map = mmap(NULL, off.tx.desc +
-				  xsk->config.tx_size * sizeof(struct xdp_desc),
+			      xsk->config.tx_size * sizeof(struct xdp_desc),
-				  PROT_READ | PROT_WRITE,
+			      PROT_READ | PROT_WRITE, MAP_SHARED | MAP_POPULATE,
-				  MAP_SHARED | MAP_POPULATE,
+			      xsk->fd, XDP_PGOFF_TX_RING);
-				  xsk->fd, XDP_PGOFF_TX_RING);
 		if (tx_map == MAP_FAILED) {
 			err = -errno;
 			goto out_mmap_rx;
@@ -584,6 +689,7 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
 		tx->size = xsk->config.tx_size;
 		tx->producer = tx_map + off.tx.producer;
 		tx->consumer = tx_map + off.tx.consumer;
+		tx->flags = tx_map + off.tx.flags;
 		tx->ring = tx_map + off.tx.desc;
 		tx->cached_cons = xsk->config.tx_size;
 	}
@@ -592,7 +698,12 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
 	sxdp.sxdp_family = PF_XDP;
 	sxdp.sxdp_ifindex = xsk->ifindex;
 	sxdp.sxdp_queue_id = xsk->queue_id;
-	sxdp.sxdp_flags = xsk->config.bind_flags;
+	if (umem->refcount > 1) {
+		sxdp.sxdp_flags = XDP_SHARED_UMEM;
+		sxdp.sxdp_shared_umem_fd = umem->fd;
+	} else {
+		sxdp.sxdp_flags = xsk->config.bind_flags;
+	}
 
 	err = bind(xsk->fd, (struct sockaddr *)&sxdp, sizeof(sxdp));
 	if (err) {
@@ -602,15 +713,6 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
 
 	xsk->prog_fd = -1;
 
-	optlen = sizeof(opts);
-	err = getsockopt(xsk->fd, SOL_XDP, XDP_OPTIONS, &opts, &optlen);
-	if (err) {
-		err = -errno;
-		goto out_mmap_tx;
-	}
-
-	xsk->zc = opts.flags & XDP_OPTIONS_ZEROCOPY;
-
 	if (!(xsk->config.libbpf_flags & XSK_LIBBPF_FLAGS__INHIBIT_PROG_LOAD)) {
 		err = xsk_setup_xdp_prog(xsk);
 		if (err)
@@ -639,7 +741,6 @@ out_xsk_alloc:
 int xsk_umem__delete(struct xsk_umem *umem)
 {
 	struct xdp_mmap_offsets off;
-	socklen_t optlen;
 	int err;
 
 	if (!umem)
@@ -648,8 +749,7 @@ int xsk_umem__delete(struct xsk_umem *umem)
 	if (umem->refcount)
 		return -EBUSY;
 
-	optlen = sizeof(off);
+	err = xsk_get_mmap_offsets(umem->fd, &off);
-	err = getsockopt(umem->fd, SOL_XDP, XDP_MMAP_OFFSETS, &off, &optlen);
 	if (!err) {
 		munmap(umem->fill->ring - off.fr.desc,
 		       off.fr.desc + umem->config.fill_size * sizeof(__u64));
@@ -667,7 +767,6 @@ void xsk_socket__delete(struct xsk_socket *xsk)
 {
 	size_t desc_sz = sizeof(struct xdp_desc);
 	struct xdp_mmap_offsets off;
-	socklen_t optlen;
 	int err;
 
 	if (!xsk)
@@ -678,8 +777,7 @@ void xsk_socket__delete(struct xsk_socket *xsk)
 		close(xsk->prog_fd);
 	}
 
-	optlen = sizeof(off);
+	err = xsk_get_mmap_offsets(xsk->fd, &off);
-	err = getsockopt(xsk->fd, SOL_XDP, XDP_MMAP_OFFSETS, &off, &optlen);
 	if (!err) {
 		if (xsk->rx) {
 			munmap(xsk->rx->ring - off.rx.desc,

src/xsk.h

@@ -32,6 +32,7 @@ struct name { \
 	__u32 *producer; \
 	__u32 *consumer; \
 	void *ring; \
+	__u32 *flags; \
 }
 
 DEFINE_XSK_RING(xsk_ring_prod);
@@ -76,6 +77,11 @@ xsk_ring_cons__rx_desc(const struct xsk_ring_cons *rx, __u32 idx)
 	return &descs[idx & rx->mask];
 }
 
+static inline int xsk_ring_prod__needs_wakeup(const struct xsk_ring_prod *r)
+{
+	return *r->flags & XDP_RING_NEED_WAKEUP;
+}
+
 static inline __u32 xsk_prod_nb_free(struct xsk_ring_prod *r, __u32 nb)
 {
 	__u32 free_entries = r->cached_cons - r->cached_prod;
@@ -162,6 +168,21 @@ static inline void *xsk_umem__get_data(void *umem_area, __u64 addr)
 	return &((char *)umem_area)[addr];
 }
 
+static inline __u64 xsk_umem__extract_addr(__u64 addr)
+{
+	return addr & XSK_UNALIGNED_BUF_ADDR_MASK;
+}
+
+static inline __u64 xsk_umem__extract_offset(__u64 addr)
+{
+	return addr >> XSK_UNALIGNED_BUF_OFFSET_SHIFT;
+}
+
+static inline __u64 xsk_umem__add_offset_to_addr(__u64 addr)
+{
+	return xsk_umem__extract_addr(addr) + xsk_umem__extract_offset(addr);
+}
+
 LIBBPF_API int xsk_umem__fd(const struct xsk_umem *umem);
 LIBBPF_API int xsk_socket__fd(const struct xsk_socket *xsk);
 
@@ -170,12 +191,14 @@ LIBBPF_API int xsk_socket__fd(const struct xsk_socket *xsk);
 #define XSK_UMEM__DEFAULT_FRAME_SHIFT    12 /* 4096 bytes */
 #define XSK_UMEM__DEFAULT_FRAME_SIZE     (1 << XSK_UMEM__DEFAULT_FRAME_SHIFT)
 #define XSK_UMEM__DEFAULT_FRAME_HEADROOM 0
+#define XSK_UMEM__DEFAULT_FLAGS 0
 
 struct xsk_umem_config {
 	__u32 fill_size;
 	__u32 comp_size;
 	__u32 frame_size;
 	__u32 frame_headroom;
+	__u32 flags;
 };
 
 /* Flags for the libbpf_flags field. */
@@ -195,6 +218,16 @@ LIBBPF_API int xsk_umem__create(struct xsk_umem **umem,
 				struct xsk_ring_prod *fill,
 				struct xsk_ring_cons *comp,
 				const struct xsk_umem_config *config);
+LIBBPF_API int xsk_umem__create_v0_0_2(struct xsk_umem **umem,
+				       void *umem_area, __u64 size,
+				       struct xsk_ring_prod *fill,
+				       struct xsk_ring_cons *comp,
+				       const struct xsk_umem_config *config);
+LIBBPF_API int xsk_umem__create_v0_0_4(struct xsk_umem **umem,
+				       void *umem_area, __u64 size,
+				       struct xsk_ring_prod *fill,
+				       struct xsk_ring_cons *comp,
+				       const struct xsk_umem_config *config);
 LIBBPF_API int xsk_socket__create(struct xsk_socket **xsk,
 				  const char *ifname, __u32 queue_id,
 				  struct xsk_umem *umem,

travis-ci/managers/debian.sh

@@ -13,6 +13,10 @@ function info() {
     echo -e "\033[33;1m$1\033[0m"
 }
 
+function error() {
+    echo -e "\033[31;1m$1\033[0m"
+}
+
 function docker_exec() {
     docker exec $ENV_VARS -it $CONT_NAME "$@"
 }
@@ -26,43 +30,43 @@ for phase in "${PHASES[@]}"; do
         SETUP)
             info "Setup phase"
             info "Using Debian $DEBIAN_RELEASE"
-	    docker pull debian:$DEBIAN_RELEASE
+            docker pull debian:$DEBIAN_RELEASE
             info "Starting container $CONT_NAME"
             $DOCKER_RUN -v $REPO_ROOT:/build:rw \
                         -w /build --privileged=true --name $CONT_NAME \
-			-dit --net=host debian:$DEBIAN_RELEASE /bin/bash
+                        -dit --net=host debian:$DEBIAN_RELEASE /bin/bash
             docker_exec bash -c "echo deb-src http://deb.debian.org/debian $DEBIAN_RELEASE main >>/etc/apt/sources.list"
             docker_exec apt-get -y update
             docker_exec apt-get -y build-dep libelf-dev
             docker_exec apt-get -y install libelf-dev
             docker_exec apt-get -y install "${ADDITIONAL_DEPS[@]}"
             ;;
-        RUN|RUN_CLANG|RUN_GCC8)
+        RUN|RUN_CLANG|RUN_GCC8|RUN_ASAN|RUN_CLANG_ASAN|RUN_GCC8_ASAN)
-            if [[ "$phase" = "RUN_CLANG" ]]; then
+            if [[ "$phase" = *"CLANG"* ]]; then
                 ENV_VARS="-e CC=clang -e CXX=clang++"
                 CC="clang"
-            elif [[ "$phase" = "RUN_GCC8" ]]; then
+            elif [[ "$phase" = *"GCC8"* ]]; then
                 ENV_VARS="-e CC=gcc-8 -e CXX=g++-8"
                 CC="gcc-8"
+            else
+                CFLAGS="${CFLAGS} -Wno-stringop-truncation"
             fi
-            docker_exec mkdir build
+            if [[ "$phase" = *"ASAN"* ]]; then
-            docker_exec ${CC:-cc} --version
+                CFLAGS="${CFLAGS} -fsanitize=address,undefined"
-            docker_exec make CFLAGS="${CFLAGS}" -C ./src -B OBJDIR=../build
-            docker_exec rm -rf build
-            ;;
-        RUN_ASAN|RUN_CLANG_ASAN|RUN_GCC8_ASAN)
-            if [[ "$phase" = "RUN_CLANG_ASAN" ]]; then
-                ENV_VARS="-e CC=clang -e CXX=clang++"
-                CC="clang"
-            elif [[ "$phase" = "RUN_GCC8_ASAN" ]]; then
-                ENV_VARS="-e CC=gcc-8 -e CXX=g++-8"
-                CC="gcc-8"
             fi
-            CFLAGS="${CFLAGS} -fsanitize=address,undefined"
+            docker_exec mkdir build install
-            docker_exec mkdir build
             docker_exec ${CC:-cc} --version
+            info "build"
             docker_exec make CFLAGS="${CFLAGS}" -C ./src -B OBJDIR=../build
-            docker_exec rm -rf build
+            info "ldd build/libbpf.so:"
+            docker_exec ldd build/libbpf.so
+            if ! docker_exec ldd build/libbpf.so | grep -q libelf; then
+                error "No reference to libelf.so in libbpf.so!"
+                exit 1
+            fi
+            info "install"
+            docker_exec make -C src OBJDIR=../build DESTDIR=../install install
+            docker_exec rm -rf build install
             ;;
         CLEANUP)
             info "Cleanup phase"

travis-ci/managers/ubuntu.sh

@@ -0,0 +1,27 @@
+#!/bin/bash
+set -e
+set -x
+
+RELEASE="bionic"
+
+echo "deb-src http://archive.ubuntu.com/ubuntu/ $RELEASE main restricted universe multiverse" >>/etc/apt/sources.list
+
+apt-get update
+apt-get -y build-dep libelf-dev
+apt-get install -y libelf-dev pkg-config
+
+source "$(dirname $0)/travis_wait.bash"
+
+cd $REPO_ROOT
+
+CFLAGS="-g -O2 -Werror -Wall -fsanitize=address,undefined"
+mkdir build install
+cc --version
+make CFLAGS="${CFLAGS}" -C ./src -B OBJDIR=../build
+ldd build/libbpf.so
+if ! ldd build/libbpf.so | grep -q libelf; then
+    echo "FAIL: No reference to libelf.so in libbpf.so!"
+    exit 1
+fi
+make -C src OBJDIR=../build DESTDIR=../install install
+rm -rf build install

travis-ci/managers/xenial.sh

@@ -1,17 +0,0 @@
-#!/bin/bash
-set -e
-set -x
-
-apt-get update
-apt-get -y build-dep libelf-dev
-apt-get install -y libelf-dev
-
-source "$(dirname $0)/travis_wait.bash"
-
-cd $REPO_ROOT
-
-CFLAGS="-g -O2 -Werror -Wall -fsanitize=address,undefined"
-mkdir build
-cc --version
-make CFLAGS="${CFLAGS}" -C ./src -B OBJDIR=../build
-rm -rf build

travis-ci/vmtest/build_latest_kernel.sh

@@ -0,0 +1,8 @@
+#!/bin/bash
+
+set -eux
+
+GIT_FETCH_DEPTH="${GIT_FETCH_DEPTH}" ${VMTEST_ROOT}/checkout_latest_kernel.sh $1
+cd $1
+cp ${VMTEST_ROOT}/configs/latest.config .config
+make -j $((4*$(nproc))) olddefconfig all

travis-ci/vmtest/build_selftests.sh

@@ -0,0 +1,20 @@
+#!/bin/bash
+
+LIBBPF_PATH="${REPO_ROOT}"
+REPO_PATH="travis-ci/vmtest/bpf-next"
+make \
+	CLANG=clang-10 \
+	LLC=llc-10 \
+	LLVM_STRIP=llvm-strip-10 \
+	VMLINUX_BTF="${VMLINUX_BTF}" \
+	-C "${REPO_ROOT}/${REPO_PATH}/tools/testing/selftests/bpf" \
+	-j $((4*$(nproc)))
+mkdir ${LIBBPF_PATH}/selftests
+cp -R "${REPO_ROOT}/${REPO_PATH}/tools/testing/selftests/bpf" \
+	${LIBBPF_PATH}/selftests
+cd ${LIBBPF_PATH}
+rm selftests/bpf/.gitignore
+git add selftests
+
+blacklist_path="${VMTEST_ROOT}/configs/blacklist"
+git add "${blacklist_path}"

travis-ci/vmtest/checkout_latest_kernel.sh

@@ -0,0 +1,24 @@
+#!/bin/bash
+
+set -eux
+
+CWD=$(pwd)
+LIBBPF_PATH=$(pwd)
+REPO_PATH=$1
+
+BPF_NEXT_ORIGIN=https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git
+LINUX_SHA=$(cat ${LIBBPF_PATH}/CHECKPOINT-COMMIT)
+
+echo REPO_PATH = ${REPO_PATH}
+echo LINUX_SHA = ${LINUX_SHA}
+
+if [ ! -d "${REPO_PATH}" ]; then
+	mkdir -p ${REPO_PATH}
+	cd ${REPO_PATH}
+	git init
+	git remote add bpf-next ${BPF_NEXT_ORIGIN}
+	git fetch --depth ${GIT_FETCH_DEPTH} bpf-next
+	git reset --hard ${LINUX_SHA}
+else
+	cd ${REPO_PATH}
+fi

travis-ci/vmtest/configs/INDEX

@@ -0,0 +1,6 @@
+INDEX	https://libbpf-vmtest.s3-us-west-1.amazonaws.com/x86_64/INDEX
+libbpf-vmtest-rootfs-2020.01.10.tar.zst	https://libbpf-vmtest.s3-us-west-1.amazonaws.com/x86_64/libbpf-vmtest-rootfs-2020.01.10.tar.zst
+vmlinux-5.5.0-rc6.zst	https://libbpf-vmtest.s3-us-west-1.amazonaws.com/x86_64/vmlinux-5.5.0-rc6.zst
+vmlinux-5.5.0.zst	https://libbpf-vmtest.s3-us-west-1.amazonaws.com/x86_64/vmlinux-5.5.0.zst
+vmlinuz-5.5.0-rc6	https://libbpf-vmtest.s3-us-west-1.amazonaws.com/x86_64/vmlinuz-5.5.0-rc6
+vmlinuz-5.5.0	https://libbpf-vmtest.s3-us-west-1.amazonaws.com/x86_64/vmlinuz-5.5.0

travis-ci/vmtest/configs/blacklist/BLACKLIST-5.5.0

@@ -0,0 +1,21 @@
+mmap
+dctcp
+cubic
+bpf_tcp_ca
+bpf_verif_scale
+cgroup_attach
+pinning
+send_signal_tracepoint_thread
+test_syncookie
+select_reuseport
+send_signal
+sockopt_inherit
+strobemeta_nounroll2
+stacktrace_build_id
+tp_attach_query
+tcp_rtt
+task_fd_query_tp
+stacktrace_map
+test_global_funcs
+skb_ctx
+fexit_bpf2bpf

travis-ci/vmtest/configs/blacklist/BLACKLIST-5.5.0-rc6

@@ -0,0 +1,18 @@
+mmap
+dctcp
+cubic
+bpf_tcp_ca
+bpf_verif_scale
+cgroup_attach
+pinning
+send_signal_tracepoint_thread
+test_syncookie
+select_reuseport
+send_signal
+sockopt_inherit
+strobemeta_nounroll2
+stacktrace_build_id
+tp_attach_query
+tcp_rtt
+task_fd_query_tp
+stacktrace_map

travis-ci/vmtest/configs/blacklist/BLACKLIST-latest

@@ -0,0 +1,23 @@
+mmap
+dctcp
+cubic
+bpf_tcp_ca
+bpf_verif_scale
+cgroup_attach
+pinning
+send_signal_tracepoint_thread
+test_syncookie
+select_reuseport
+send_signal
+sockopt_inherit
+strobemeta_nounroll2
+stacktrace_build_id
+tp_attach_query
+tcp_rtt
+task_fd_query_tp
+stacktrace_map
+fentry
+test_overhead
+kfree_skb
+fexit_stress
+fexit_test

travis-ci/vmtest/prepare_selftests.sh

@@ -0,0 +1,11 @@
+#!/bin/bash
+
+set -eux
+GIT_FETCH_DEPTH="${GIT_FETCH_DEPTH}" ${VMTEST_ROOT}/checkout_latest_kernel.sh $1
+
+# Fix runqslower build
+# TODO(hex@): remove after the patch is merged from bpf to bpf-next tree
+cd $1
+wget https://lore.kernel.org/bpf/908498f794661c44dca54da9e09dc0c382df6fcb.1580425879.git.hex@fb.com/t.mbox.gz
+gunzip t.mbox.gz
+git apply t.mbox

travis-ci/vmtest/run.sh

@@ -0,0 +1,438 @@
+#!/bin/bash
+
+set -uo pipefail
+trap 'exit 2' ERR
+
+usage () {
+	USAGE_STRING="usage: $0 [-k KERNELRELEASE|-b DIR] [[-r ROOTFSVERSION] [-fo]|-I] [-Si] [-d DIR] IMG
+       $0 [-k KERNELRELEASE] -l
+       $0 -h
+
+Run "${PROJECT_NAME}" tests in a virtual machine.
+
+This exits with status 0 on success, 1 if the virtual machine ran successfully
+but tests failed, and 2 if we encountered a fatal error.
+
+This script uses sudo to mount and modify the disk image.
+
+Arguments:
+  IMG                 path of virtual machine disk image to create
+
+Versions:
+  -k, --kernel=KERNELRELEASE
+                       kernel release to test. This is a glob pattern; the
+                       newest (sorted by version number) release that matches
+                       the pattern is used (default: newest available release)
+
+  -b, --build DIR      use the kernel built in the given directory. This option
+                       cannot be combined with -k
+
+  -r, --rootfs=ROOTFSVERSION
+                       version of root filesystem to use (default: newest
+                       available version)
+
+Setup:
+  -f, --force          overwrite IMG if it already exists
+
+  -o, --one-shot       one-shot mode. By default, this script saves a clean copy
+                       of the downloaded root filesystem image and vmlinux and
+                       makes a copy (reflinked, when possible) for executing the
+                       virtual machine. This allows subsequent runs to skip
+                       downloading these files. If this option is given, the
+                       root filesystem image and vmlinux are always
+                       re-downloaded and are not saved. This option implies -f
+
+  -s, --setup-cmd      setup commands run on VM boot. Whitespace characters
+                       should be escaped with preceding '\'.
+
+  -I, --skip-image     skip creating the disk image; use the existing one at
+                       IMG. This option cannot be combined with -r, -f, or -o
+
+  -S, --skip-source    skip copying the source files and init scripts
+
+Miscellaneous:
+  -i, --interactive    interactive mode. Boot the virtual machine into an
+                       interactive shell instead of automatically running tests
+
+  -d, --dir=DIR        working directory to use for downloading and caching
+                       files (default: current working directory)
+
+  -l, --list           list available kernel releases instead of running tests.
+                       The list may be filtered with -k
+
+  -h, --help           display this help message and exit"
+
+	case "$1" in
+		out)
+			echo "$USAGE_STRING"
+			exit 0
+			;;
+		err)
+			echo "$USAGE_STRING" >&2
+			exit 2
+			;;
+	esac
+}
+
+TEMP=$(getopt -o 'k:b:r:fos:ISid:lh' --long 'kernel:,build:,rootfs:,force,one-shot,setup-cmd,skip-image,skip-source:,interactive,dir:,list,help' -n "$0" -- "$@")
+eval set -- "$TEMP"
+unset TEMP
+
+unset KERNELRELEASE
+unset BUILDDIR
+unset ROOTFSVERSION
+unset IMG
+unset SETUPCMD
+FORCE=0
+ONESHOT=0
+SKIPIMG=0
+SKIPSOURCE=0
+APPEND=""
+DIR="$PWD"
+LIST=0
+while true; do
+	case "$1" in
+		-k|--kernel)
+			KERNELRELEASE="$2"
+			shift 2
+			;;
+		-b|--build)
+			BUILDDIR="$2"
+			shift 2
+			;;
+		-r|--rootfs)
+			ROOTFSVERSION="$2"
+			shift 2
+			;;
+		-f|--force)
+			FORCE=1
+			shift
+			;;
+		-o|--one-shot)
+			ONESHOT=1
+			FORCE=1
+			shift
+			;;
+		-s|--setup-cmd)
+			SETUPCMD="$2"
+			shift 2
+			;;
+		-I|--skip-image)
+			SKIPIMG=1
+			shift
+			;;
+		-S|--skip-source)
+			SKIPSOURCE=1
+			shift
+			;;
+		-i|--interactive)
+			APPEND=" single"
+			shift
+			;;
+		-d|--dir)
+			DIR="$2"
+			shift 2
+			;;
+		-l|--list)
+			LIST=1
+			;;
+		-h|--help)
+			usage out
+			;;
+		--)
+			shift
+			break
+			;;
+		*)
+			usage err
+			;;
+	esac
+done
+if [[ -v BUILDDIR ]]; then
+	if [[ -v KERNELRELEASE ]]; then
+		usage err
+	fi
+elif [[ ! -v KERNELRELEASE ]]; then
+	KERNELRELEASE='*'
+fi
+if [[ $SKIPIMG -ne 0 && ( -v ROOTFSVERSION || $FORCE -ne 0 ) ]]; then
+	usage err
+fi
+if (( LIST )); then
+	if [[ $# -ne 0 || -v BUILDDIR || -v ROOTFSVERSION || $FORCE -ne 0 ||
+	      $SKIPIMG -ne 0 || $SKIPSOURCE -ne 0 || -n $APPEND ]]; then
+		usage err
+	fi
+else
+	if [[ $# -ne 1 ]]; then
+		usage err
+	fi
+	IMG="${!OPTIND}"
+fi
+
+unset URLS
+cache_urls() {
+	if ! declare -p URLS &> /dev/null; then
+		# This URL contains a mapping from file names to URLs where
+		# those files can be downloaded.
+		declare -gA URLS
+		while IFS=$'\t' read -r name url; do
+			URLS["$name"]="$url"
+		done < <(cat "${VMTEST_ROOT}/configs/INDEX")
+	fi
+}
+
+matching_kernel_releases() {
+	local pattern="$1"
+	{
+	for file in "${!URLS[@]}"; do
+		if [[ $file =~ ^vmlinux-(.*).zst$ ]]; then
+			release="${BASH_REMATCH[1]}"
+			case "$release" in
+				$pattern)
+					# sort -V handles rc versions properly
+					# if we use "~" instead of "-".
+					echo "${release//-rc/~rc}"
+					;;
+			esac
+		fi
+	done
+	} | sort -rV | sed 's/~rc/-rc/g'
+}
+
+newest_rootfs_version() {
+	{
+	for file in "${!URLS[@]}"; do
+		if [[ $file =~ ^${PROJECT_NAME}-vmtest-rootfs-(.*)\.tar\.zst$ ]]; then
+			echo "${BASH_REMATCH[1]}"
+		fi
+	done
+	} | sort -rV | head -1
+}
+
+download() {
+	local file="$1"
+	cache_urls
+	if [[ ! -v URLS[$file] ]]; then
+		echo "$file not found" >&2
+		return 1
+	fi
+	echo "Downloading $file..." >&2
+	curl -Lf "${URLS[$file]}" "${@:2}"
+}
+
+set_nocow() {
+	touch "$@"
+	chattr +C "$@" >/dev/null 2>&1 || true
+}
+
+cp_img() {
+	set_nocow "$2"
+	cp --reflink=auto "$1" "$2"
+}
+
+create_rootfs_img() {
+	local path="$1"
+	set_nocow "$path"
+	truncate -s 2G "$path"
+	mkfs.ext4 -q "$path"
+}
+
+download_rootfs() {
+	local rootfsversion="$1"
+	local dir="$2"
+	download "${PROJECT_NAME}-vmtest-rootfs-$rootfsversion.tar.zst" |
+		zstd -d | sudo tar -C "$dir" -x
+}
+
+if (( LIST )); then
+	cache_urls
+	matching_kernel_releases "$KERNELRELEASE"
+	exit 0
+fi
+
+if [[ $FORCE -eq 0 && $SKIPIMG -eq 0 && -e $IMG ]]; then
+	echo "$IMG already exists; use -f to overwrite it or -I to reuse it" >&2
+	exit 1
+fi
+
+# Only go to the network if it's actually a glob pattern.
+if [[ -v BUILDDIR ]]; then
+	KERNELRELEASE="$(make -C "$BUILDDIR" -s kernelrelease)"
+elif [[ ! $KERNELRELEASE =~ ^([^\\*?[]|\\[*?[])*\\?$ ]]; then
+	# We need to cache the list of URLs outside of the command
+	# substitution, which happens in a subshell.
+	cache_urls
+	KERNELRELEASE="$(matching_kernel_releases "$KERNELRELEASE" | head -1)"
+	if [[ -z $KERNELRELEASE ]]; then
+		echo "No matching kernel release found" >&2
+		exit 1
+	fi
+fi
+if [[ $SKIPIMG -eq 0 && ! -v ROOTFSVERSION ]]; then
+	cache_urls
+	ROOTFSVERSION="$(newest_rootfs_version)"
+fi
+
+echo "Kernel release: $KERNELRELEASE" >&2
+if (( SKIPIMG )); then
+	echo "Not extracting root filesystem" >&2
+else
+	echo "Root filesystem version: $ROOTFSVERSION" >&2
+fi
+echo "Disk image: $IMG" >&2
+
+tmp=
+ARCH_DIR="$DIR/x86_64"
+mkdir -p "$ARCH_DIR"
+mnt="$(mktemp -d -p "$DIR" mnt.XXXXXXXXXX)"
+
+cleanup() {
+	if [[ -n $tmp ]]; then
+		rm -f "$tmp" || true
+	fi
+	if mountpoint -q "$mnt"; then
+		sudo umount "$mnt" || true
+	fi
+	if [[ -d "$mnt" ]]; then
+		rmdir "$mnt" || true
+	fi
+}
+trap cleanup EXIT
+
+if [[ -v BUILDDIR ]]; then
+	vmlinuz="$BUILDDIR/$(make -C "$BUILDDIR" -s image_name)"
+else
+	vmlinuz="${ARCH_DIR}/vmlinuz-${KERNELRELEASE}"
+	if [[ ! -e $vmlinuz ]]; then
+		tmp="$(mktemp "$vmlinuz.XXX.part")"
+		download "vmlinuz-${KERNELRELEASE}" -o "$tmp"
+		mv "$tmp" "$vmlinuz"
+		tmp=
+	fi
+fi
+
+# Mount and set up the rootfs image.
+if (( ONESHOT )); then
+	rm -f "$IMG"
+	create_rootfs_img "$IMG"
+	sudo mount -o loop "$IMG" "$mnt"
+	download_rootfs "$ROOTFSVERSION" "$mnt"
+else
+	if (( ! SKIPIMG )); then
+		rootfs_img="${ARCH_DIR}/${PROJECT_NAME}-vmtest-rootfs-${ROOTFSVERSION}.img"
+
+		if [[ ! -e $rootfs_img ]]; then
+			tmp="$(mktemp "$rootfs_img.XXX.part")"
+			set_nocow "$tmp"
+			truncate -s 2G "$tmp"
+			mkfs.ext4 -q "$tmp"
+			sudo mount -o loop "$tmp" "$mnt"
+
+			download_rootfs "$ROOTFSVERSION" "$mnt"
+
+			sudo umount "$mnt"
+			mv "$tmp" "$rootfs_img"
+			tmp=
+		fi
+
+		rm -f "$IMG"
+		cp_img "$rootfs_img" "$IMG"
+	fi
+	sudo mount -o loop "$IMG" "$mnt"
+fi
+
+# Install vmlinux.
+vmlinux="$mnt/boot/vmlinux-${KERNELRELEASE}"
+if [[ -v BUILDDIR || $ONESHOT -eq 0 ]]; then
+	if [[ -v BUILDDIR ]]; then
+		source_vmlinux="${BUILDDIR}/vmlinux"
+	else
+		source_vmlinux="${ARCH_DIR}/vmlinux-${KERNELRELEASE}"
+		if [[ ! -e $source_vmlinux ]]; then
+			tmp="$(mktemp "$source_vmlinux.XXX.part")"
+			download "vmlinux-${KERNELRELEASE}.zst" | zstd -dfo "$tmp"
+			mv "$tmp" "$source_vmlinux"
+			tmp=
+		fi
+	fi
+	echo "Copying vmlinux..." >&2
+	sudo rsync -cp --chmod 0644 "$source_vmlinux" "$vmlinux"
+else
+	# We could use "sudo zstd -o", but let's not run zstd as root with
+	# input from the internet.
+	download "vmlinux-${KERNELRELEASE}.zst" |
+		zstd -d | sudo tee "$vmlinux" > /dev/null
+	sudo chmod 644 "$vmlinux"
+fi
+
+LIBBPF_PATH="${REPO_ROOT}" \
+	REPO_PATH="travis-ci/vmtest/bpf-next" \
+	VMTEST_ROOT="${VMTEST_ROOT}" \
+	VMLINUX_BTF=${vmlinux} ${VMTEST_ROOT}/build_selftests.sh
+
+if (( SKIPSOURCE )); then
+	echo "Not copying source files..." >&2
+else
+	echo "Copying source files..." >&2
+
+	# Copy the source files in.
+	sudo mkdir -p -m 0755 "$mnt/${PROJECT_NAME}"
+	{
+	if [[ -e .git ]]; then
+		git ls-files -z
+	else
+		tr '\n' '\0' < "${PROJECT_NAME}.egg-info/SOURCES.txt"
+	fi
+	} | sudo rsync --files-from=- -0cpt . "$mnt/${PROJECT_NAME}"
+fi
+
+setup_script="#!/bin/sh
+
+echo 'Skipping setup commands'
+echo 0 > /exitstatus
+chmod 644 /exitstatus"
+
+# Create the init scripts.
+if [[ ! -z SETUPCMD ]]; then
+	# Unescape whitespace characters.
+	setup_cmd=$(sed 's/\(\\\)\([[:space:]]\)/\2/g' <<< "${SETUPCMD}")
+	kernel="${KERNELRELEASE}"
+	if [[ -v BUILDDIR ]]; then kernel='latest'; fi
+	setup_envvars="export KERNEL=${kernel}"
+	setup_script=$(printf "#!/bin/sh
+set -e
+
+echo 'Running setup commands'
+%s
+%s
+echo $? > /exitstatus
+chmod 644 /exitstatus" "${setup_envvars}" "${setup_cmd}")
+fi
+
+echo "${setup_script}" | sudo tee "$mnt/etc/rcS.d/S50-run-tests" > /dev/null
+sudo chmod 755 "$mnt/etc/rcS.d/S50-run-tests"
+
+poweroff_script="#!/bin/sh
+
+poweroff"
+echo "${poweroff_script}" | sudo tee "$mnt/etc/rcS.d/S99-poweroff" > /dev/null
+sudo chmod 755 "$mnt/etc/rcS.d/S99-poweroff"
+
+sudo umount "$mnt"
+
+echo "Starting virtual machine..." >&2
+qemu-system-x86_64 -nodefaults -display none -serial mon:stdio \
+	-cpu kvm64 -enable-kvm -smp "$(nproc)" -m 2G \
+	-drive file="$IMG",format=raw,index=1,media=disk,if=virtio,cache=none \
+	-kernel "$vmlinuz" -append "root=/dev/vda rw console=ttyS0,115200$APPEND"
+
+sudo mount -o loop "$IMG" "$mnt"
+if exitstatus="$(cat "$mnt/exitstatus" 2>/dev/null)"; then
+	printf '\nTests exit status: %s\n' "$exitstatus" >&2
+else
+	printf '\nCould not read tests exit status\n' >&2
+	exitstatus=1
+fi
+sudo umount "$mnt"
+exit "$exitstatus"

travis-ci/vmtest/run_selftests.sh

@@ -0,0 +1,19 @@
+#!/bin/bash
+
+set -eux
+
+configs_path='libbpf/travis-ci/vmtest/configs'
+blacklist_path="$configs_path/blacklist/BLACKLIST-${KERNEL}"
+if [[ -s "${blacklist_path}" ]]; then
+	BLACKLIST=$(cat "${blacklist_path}" | tr '\n' ',')
+fi
+
+whitelist_path="$configs_path/whitelist/WHITELIST-${KERNEL}"
+if [[ -s "${whitelist_path}" ]]; then
+	WHITELIST=$(cat "${whitelist_path}" | tr '\n' ',')
+fi
+
+cd libbpf/selftests/bpf
+
+echo TEST_PROGS
+./test_progs ${BLACKLIST:+-b$BLACKLIST} ${WHITELIST:+-t$WHITELIST}

travis-ci/vmtest/setup_example.sh

@@ -0,0 +1,11 @@
+#!/bin/sh
+
+# An example of a script run on VM boot.
+# To execute it in TravisCI set VMTEST_SETUPCMD env var of .travis.yml in
+# libbpf root folder, e.g.
+# VMTEST_SETUPCMD="./${PROJECT_NAME}/travis-ci/vmtest/setup_example.sh"
+
+if [ ! -z "${PROJECT_NAME}" ]; then
+	echo "Running ${PROJECT_NAME} setup scripts..."
+fi
+echo "Hello, ${USER}!"