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

.travis.yml

@@ -1,13 +1,54 @@
 sudo: required
+language: bash
 dist: bionic
 services:
     - docker
 
 env:
     global:
+        - 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:
@@ -32,6 +73,8 @@ jobs:
               - docker --version
           install:
               - $CI_MANAGERS/debian.sh SETUP
+          # Override before_script: so VMTEST before_install commands are not executed.
+          before_script: true
           script:
               - $CI_MANAGERS/debian.sh RUN || travis_terminate
           after_script:
@@ -47,6 +90,7 @@ jobs:
               - docker --version
           install:
               - $CI_MANAGERS/debian.sh SETUP
+          before_script: true
           script:
               - $CI_MANAGERS/debian.sh RUN_ASAN || travis_terminate
           after_script:
@@ -62,6 +106,7 @@ jobs:
               - docker --version
           install:
               - $CI_MANAGERS/debian.sh SETUP
+          before_script: true
           script:
               - $CI_MANAGERS/debian.sh RUN_CLANG || travis_terminate
           after_script:
@@ -77,6 +122,7 @@ jobs:
               - docker --version
           install:
               - $CI_MANAGERS/debian.sh SETUP
+          before_script: true
           script:
               - $CI_MANAGERS/debian.sh RUN_CLANG_ASAN || travis_terminate
           after_script:
@@ -92,6 +138,7 @@ jobs:
               - docker --version
           install:
               - $CI_MANAGERS/debian.sh SETUP
+          before_script: true
           script:
               - $CI_MANAGERS/debian.sh RUN_GCC8 || travis_terminate
           after_script:
@@ -107,6 +154,7 @@ jobs:
               - docker --version
           install:
               - $CI_MANAGERS/debian.sh SETUP
+          before_script: true
           script:
               - $CI_MANAGERS/debian.sh RUN_GCC8_ASAN || travis_terminate
           after_script:
@@ -114,18 +162,28 @@ jobs:
 
         - name: Ubuntu Bionic
           language: bash
+          before_script: true
           script:
               - sudo $CI_MANAGERS/ubuntu.sh || travis_terminate
 
         - 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
 
@@ -147,5 +205,7 @@ jobs:
               - 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

@@ -1 +1 @@
-e42617b825f8073569da76dc4510bfa019b1c35a
+08dc225d8868d5094ada62f471ebdfcce9dbc298

CHECKPOINT-COMMIT

@@ -1 +1 @@
-e7096c131e5161fa3b8e52a650d7719d2857adfd
+35b9211c0a2427e8f39e534f442f43804fc8d5ca

README.md

@@ -51,6 +51,28 @@ $ cd src
 $ PKG_CONFIG_PATH=/build/root/lib64/pkgconfig DESTDIR=/build/root make install
 ```
 
+Distributions
+=====
+
+Distributions packaging libbpf from this mirror:
+  - [Fedora](https://src.fedoraproject.org/rpms/libbpf)
+  - [Gentoo](https://packages.gentoo.org/packages/dev-libs/libbpf)
+
+Benefits of packaging from the mirror over packaging from kernel sources:
+  - Consistent versioning across distributions.
+  - No ties to any specific kernel, transparent handling of older kernels.
+    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
 =====
 

include/uapi/linux/bpf.h

@@ -107,6 +107,10 @@ enum bpf_cmd {
 	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 {
@@ -136,6 +140,7 @@ enum bpf_map_type {
 	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
@@ -174,6 +179,8 @@ enum bpf_prog_type {
 	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 {
@@ -231,6 +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:
@@ -249,6 +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
@@ -351,7 +364,12 @@ enum bpf_attach_type {
 /* Enable memory-mapping BPF map */
 #define BPF_F_MMAPABLE		(1U << 10)
 
-/* flags for BPF_PROG_QUERY */
+/* 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 {
@@ -391,6 +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 */
@@ -403,6 +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;
@@ -442,6 +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 */
@@ -2693,7 +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.
  *
@@ -2821,6 +2865,33 @@ union bpf_attr {
  * 	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),			\
@@ -2938,7 +3009,10 @@ union bpf_attr {
 	FN(probe_read_user),		\
 	FN(probe_read_kernel),		\
 	FN(probe_read_user_str),	\
-	FN(probe_read_kernel_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
@@ -3339,7 +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;

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

@@ -169,6 +169,7 @@ enum {
 	IFLA_MAX_MTU,
 	IFLA_PROP_LIST,
 	IFLA_ALT_IFNAME, /* Alternative ifname */
+	IFLA_PERM_ADDRESS,
 	__IFLA_MAX
 };
 
@@ -485,6 +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,

src/Makefile

@@ -21,7 +21,7 @@ CFLAGS ?= -g -O2 -Werror -Wall
 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)
@@ -47,7 +47,7 @@ endif
 
 HEADERS := bpf.h libbpf.h btf.h xsk.h libbpf_util.h \
 	   bpf_helpers.h bpf_helper_defs.h bpf_tracing.h \
-	   bpf_endian.h bpf_core_read.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)
 
@@ -68,6 +68,8 @@ LIBDIR ?= $(PREFIX)/$(LIBSUBDIR)
 INCLUDEDIR ?= $(PREFIX)/include
 UAPIDIR ?= $(PREFIX)/include
 
+TAGS_PROG := $(if $(shell which etags 2>/dev/null),etags,ctags)
+
 all: $(STATIC_LIBS) $(SHARED_LIBS) $(PC_FILE)
 
 $(OBJDIR)/libbpf.a: $(STATIC_OBJS)
@@ -133,3 +135,12 @@ install_pkgconfig: $(PC_FILE)
 
 clean:
 	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));
 }
@@ -228,7 +235,10 @@ 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_TRACING) {
+	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 {
@@ -443,6 +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;
@@ -466,14 +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));
 }

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
@@ -126,10 +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);

src/bpf_helper_defs.h

@@ -2600,7 +2600,8 @@ static int (*bpf_sk_storage_delete)(void *map, struct bpf_sock *sk) = (void *) 1
 /*
  * bpf_send_signal
  *
- * 	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.
  *
  * Returns
  * 	0 on success or successfully queued.
@@ -2756,4 +2757,43 @@ static int (*bpf_probe_read_user_str)(void *dst, __u32 size, const void *unsafe_
  */
 static int (*bpf_probe_read_kernel_str)(void *dst, __u32 size, const void *unsafe_ptr) = (void *) 115;
 
+/*
+ * bpf_tcp_send_ack
+ *
+ * 	Send out a tcp-ack. *tp* is the in-kernel struct tcp_sock.
+ * 	*rcv_nxt* is the ack_seq to be sent out.
+ *
+ * Returns
+ * 	0 on success, or a negative error in case of failure.
+ */
+static int (*bpf_tcp_send_ack)(void *tp, __u32 rcv_nxt) = (void *) 116;
+
+/*
+ * bpf_send_signal_thread
+ *
+ * 	Send signal *sig* to the thread corresponding to the current task.
+ *
+ * Returns
+ * 	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.
+ */
+static int (*bpf_send_signal_thread)(__u32 sig) = (void *) 117;
+
+/*
+ * bpf_jiffies64
+ *
+ * 	Obtain the 64bit jiffies
+ *
+ * Returns
+ * 	The 64 bit jiffies
+ */
+static __u64 (*bpf_jiffies64)(void) = (void *) 118;
+
 

src/bpf_helpers.h

@@ -25,6 +25,9 @@
 #ifndef __always_inline
 #define __always_inline __attribute__((always_inline))
 #endif
+#ifndef __weak
+#define __weak __attribute__((weak))
+#endif
 
 /*
  * Helper structure used by eBPF C program
@@ -44,4 +47,12 @@ enum libbpf_pin_type {
 	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;

src/btf.c

@@ -8,6 +8,10 @@
 #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>
@@ -17,8 +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 BTF_MAX_NR_TYPES 0x7fffffffU
+#define BTF_MAX_STR_OFFSET 0x7fffffffU
 
 static struct btf_type btf_void;
 
@@ -50,7 +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);
@@ -278,6 +285,45 @@ done:
 	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;
@@ -539,6 +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);
@@ -575,7 +627,8 @@ static int btf_fixup_datasec(struct bpf_object *obj, struct btf *btf,
 		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;
 }
 
@@ -1352,7 +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;
@@ -1648,7 +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) {
@@ -2882,3 +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

@@ -8,14 +8,12 @@
 #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"
@@ -79,6 +77,7 @@ 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);
@@ -103,6 +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;
@@ -127,6 +128,28 @@ 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
  */

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;
 
@@ -116,6 +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,
@@ -137,18 +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)
@@ -175,7 +202,6 @@ void btf_dump__free(struct btf_dump *d)
 	free(d);
 }
 
-static int btf_dump_mark_referenced(struct btf_dump *d);
 static int btf_dump_order_type(struct btf_dump *d, __u32 id, bool through_ptr);
 static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id);
 
@@ -202,27 +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;
-
-		/* eagerly determine referenced types for anon enums */
-		err = btf_dump_mark_referenced(d);
-		if (err)
-			return err;
-	}
-
 	d->emit_queue_cnt = 0;
 	err = btf_dump_order_type(d, id, false);
 	if (err < 0)
@@ -752,41 +757,6 @@ 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(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:
-		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, align = 1;
-
-		for (i = 0; i < vlen; i++, m++)
-			align = max(align, btf_align_of(btf, m->type));
-
-		return align;
-	}
-	default:
-		pr_warn("unsupported BTF_KIND:%u\n", btf_kind(t));
-		return 1;
-	}
-}
-
 static bool btf_is_struct_packed(const struct btf *btf, __u32 id,
 				 const struct btf_type *t)
 {
@@ -794,18 +764,18 @@ static bool btf_is_struct_packed(const struct btf *btf, __u32 id,
 	int align, i, bit_sz;
 	__u16 vlen;
 
-	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 = btf_members(t);
 	vlen = btf_vlen(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 = 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;
 	}
 
@@ -889,7 +859,7 @@ static void btf_dump_emit_struct_def(struct btf_dump *d,
 		fname = btf_name_of(d, m->name_off);
 		m_sz = btf_member_bitfield_size(t, i);
 		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));
@@ -907,7 +877,7 @@ static void btf_dump_emit_struct_def(struct btf_dump *d,
 
 	/* pad at the end, if necessary */
 	if (is_struct) {
-		align = packed ? 1 : btf_align_of(d->btf, id);
+		align = packed ? 1 : btf__align_of(d->btf, id);
 		btf_dump_emit_bit_padding(d, off, t->size * 8, 0, align,
 					  lvl + 1);
 	}
@@ -1051,6 +1021,21 @@ 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)
 {

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/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,
 
@@ -67,28 +65,6 @@ struct bpf_object_open_attr {
 	enum bpf_prog_type prog_type;
 };
 
-/* 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__					    \
-		};							    \
-	})
-
 struct bpf_object_open_opts {
 	/* size of this struct, for forward/backward compatiblity */
 	size_t sz;
@@ -101,7 +77,11 @@ struct bpf_object_open_opts {
 	const char *object_name;
 	/* parse map definitions non-strictly, allowing extra attributes/data */
 	bool relaxed_maps;
-	/* process CO-RE relocations non-strictly, allowing them to fail */
+	/* 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
@@ -109,15 +89,19 @@ struct bpf_object_open_opts {
 	 */
 	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 attach_prog_fd
+#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, struct bpf_object_open_opts *opts);
+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,
-		     struct bpf_object_open_opts *opts);
+		     const struct bpf_object_open_opts *opts);
 
 /* deprecated bpf_object__open variants */
 LIBBPF_API struct bpf_object *
@@ -126,11 +110,6 @@ bpf_object__open_buffer(const void *obj_buf, size_t obj_buf_sz,
 LIBBPF_API struct bpf_object *
 bpf_object__open_xattr(struct bpf_object_open_attr *attr);
 
-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);
-
 enum libbpf_pin_type {
 	LIBBPF_PIN_NONE,
 	/* PIN_BY_NAME: pin maps by name (in /sys/fs/bpf by default) */
@@ -161,6 +140,7 @@ struct bpf_object_load_attr {
 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);
 
@@ -171,6 +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)			\
@@ -214,6 +197,7 @@ 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);
 
@@ -235,8 +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 *
@@ -256,6 +243,8 @@ bpf_program__attach_raw_tracepoint(struct bpf_program *prog,
 
 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;
 
 /*
@@ -332,6 +321,8 @@ 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,
@@ -352,6 +343,8 @@ 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'
@@ -371,7 +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);
 
@@ -512,18 +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;
 
@@ -550,6 +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
@@ -630,6 +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

@@ -208,3 +208,30 @@ LIBBPF_0.0.6 {
 		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

@@ -76,7 +76,7 @@ static inline bool libbpf_validate_opts(const char *opts,
 
 		for (i = opts_sz; i < user_sz; i++) {
 			if (opts[i]) {
-				pr_warn("%s has non-zero extra bytes",
+				pr_warn("%s has non-zero extra bytes\n",
 					type_name);
 				return false;
 			}
@@ -95,9 +95,28 @@ static inline bool libbpf_validate_opts(const char *opts,
 #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

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);
@@ -103,6 +106,8 @@ probe_load(enum bpf_prog_type prog_type, const struct bpf_insn *insns,
 	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;
 	}
@@ -251,6 +256,7 @@ bool bpf_probe_map_type(enum bpf_map_type map_type, __u32 ifindex)
 	case BPF_MAP_TYPE_XSKMAP:
 	case BPF_MAP_TYPE_SOCKHASH:
 	case BPF_MAP_TYPE_REUSEPORT_SOCKARRAY:
+	case BPF_MAP_TYPE_STRUCT_OPS:
 	default:
 		break;
 	}
@@ -321,3 +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

@@ -15,6 +15,9 @@
 #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

src/nlattr.c

@@ -13,6 +13,9 @@
 #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),

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

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}!"