user/sven/linux.git/samples/bpf, branch v6.13.4

samples/bpf: Remove unnecessary -I flags from libbpf EXTRA_CFLAGS

2024-12-03T21:29:04Z

Commit [0] breaks samples/bpf build: $ make M=samples/bpf ... make -C /path/to/kernel/samples/bpf/../../tools/lib/bpf \ ... EXTRA_CFLAGS=" \ ... -fsanitize=bounds \ -I/path/to/kernel/usr/include \ ... /path/to/kernel/samples/bpf/libbpf/libbpf.a install_headers CC /path/to/kernel/samples/bpf/libbpf/staticobjs/libbpf.o In file included from libbpf.c:29: /path/to/kernel/tools/include/linux/err.h:35:8: error: 'inline' can only appear on functions 35 | static inline void * __must_check ERR_PTR(long error_) | ^ The error is caused by `objtree` variable changing definition from `.` (dot) to an absolute path: - The variable TPROGS_CFLAGS is constructed as follows: ... TPROGS_CFLAGS += -I$(objtree)/usr/include - It is passed as EXTRA_CFLAGS for libbpf compilation: $(LIBBPF): ... ... $(MAKE) -C $(LIBBPF_SRC) RM='rm -rf' EXTRA_CFLAGS="$(TPROGS_CFLAGS)" - Before commit [0], the line passed to libbpf makefile was '-I./usr/include', where '.' referred to LIBBPF_SRC due to -C flag. The directory $(LIBBPF_SRC)/usr/include does not exist and thus was never resolved by C compiler. - After commit [0], the line passed to libbpf makefile became: '/usr/include', this directory exists and is resolved by C compiler. - Both 'tools/include' and 'usr/include' define files err.h and types.h. - libbpf expects headers like 'linux/err.h' and 'linux/types.h' defined in 'tools/include', not 'usr/include', hence the compilation error. This commit removes unnecessary -I flags from libbpf compilation. (libbpf sets up the necessary includes at lib/bpf/Makefile:63). Changes v1 [1] -> v2: - dropped unnecessary replacement of KBUILD_OUTPUT with $(objtree) (Andrii) Changes v2 [2] -> v3: - make sure --sysroot option is set for libbpf's EXTRA_CFLAGS, if $(SYSROOT) is set (Stanislav) [0] commit 13b25489b6f8 ("kbuild: change working directory to external module directory with M=") [1] https://lore.kernel.org/bpf/20241202212154.3174402-1-eddyz87@gmail.com/ [2] https://lore.kernel.org/bpf/20241202234741.3492084-1-eddyz87@gmail.com/ Fixes: 13b25489b6f8 ("kbuild: change working directory to external module directory with M=") Signed-off-by: Eduard Zingerman Signed-off-by: Andrii Nakryiko Acked-by: Stanislav Fomichev Link: https://lore.kernel.org/bpf/20241203182222.3915763-1-eddyz87@gmail.com

samples/bpf: Remove unused variable in xdp2skb_meta_kern.c

2024-11-13T20:21:48Z

The variable is never referenced in the code, just remove it. Signed-off-by: Zhu Jun Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/20241111061514.3257-1-zhujun2@cmss.chinamobile.com

samples/bpf: Remove unused variables in tc_l2_redirect_kern.c

2024-11-13T20:21:04Z

These variables are never referenced in the code, just remove them. Signed-off-by: Zhu Jun Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/20241111062312.3541-1-zhujun2@cmss.chinamobile.com

samples/bpf: remove obsolete tracing related tests

2024-10-11T16:51:31Z

The samples/bpf has become outdated and often does not follow up with the latest. This commit removes obsolete tracing-related tests. Specifically, 'test_overhead' is duplicate with selftests (and bench), and 'test_override_return', 'test_probe_write_user' tests are obsolete since they have been replaced by kprobe_multi_override and probe_user from selftests respectively. The following files are removed: - test_overhead: tests the overhead of BPF programs with task_rename, now covered by selftests and benchmark tests (rename-*). [1] - test_override_return: tests the return override functionality, now handled by kprobe_multi_override in selftests. - test_probe_write_user: tests the probe_write_user functionality, now replaced by the probe_user test in selftests. This cleanup will help to streamline the testing framework by removing redundant tests. [1]: https://patchwork.kernel.org/cover/13759916 Signed-off-by: Daniel T. Lee Link: https://lore.kernel.org/r/20241011044847.51584-5-danieltimlee@gmail.com Signed-off-by: Alexei Starovoitov

samples/bpf: remove obsolete cgroup related tests

2024-10-11T16:51:31Z

This patch removes the obsolete cgroup related tests. These tests are now redundant because their functionality is already covered by more modern and comprehensive tests under selftests/bpf. The following files are removed: - test_current_task_under_cgroup: tests bpf_current_task_under_cgroup() to check if a task belongs to a cgroup. Already covered by task_under_cgroup at selftest and other cgroup ID tests. - test_cgrp2_tc: tests bpf_skb_under_cgroup() to filter packets based on cgroup. This behavior is now validated by cgroup_skb_sk_lookup, which uses bpf_skb_cgroup_id, making this test redundant. By removing these outdated tests, this patch helps streamline and modernize the test suite, avoiding duplication of test coverage. Signed-off-by: Daniel T. Lee Link: https://lore.kernel.org/r/20241011044847.51584-4-danieltimlee@gmail.com Signed-off-by: Alexei Starovoitov

selftests/bpf: migrate cgroup sock create test for prohibiting sockets

2024-10-11T16:51:31Z

This patch continues the migration and removal process for cgroup sock_create tests to selftests. The test being migrated verifies the ability of cgroup BPF to block the creation of specific types of sockets using a verdict. Specifically, the test denies socket creation when the socket is of type AF_INET{6}, SOCK_DGRAM, and IPPROTO_ICMP{V6}. If the requested socket type matches these attributes, the cgroup BPF verdict blocks the socket creation. As with the previous commit, this test currently lacks coverage in selftests, so this patch migrates the functionality into the sock_create tests under selftests. This migration ensures that the socket creation blocking behavior with cgroup bpf program is properly tested within the selftest framework. Signed-off-by: Daniel T. Lee Link: https://lore.kernel.org/r/20241011044847.51584-3-danieltimlee@gmail.com Signed-off-by: Alexei Starovoitov

selftests/bpf: migrate cgroup sock create test for setting iface/mark/prio

2024-10-11T16:51:31Z

This patch migrates the old test for cgroup BPF that sets sk_bound_dev_if, mark, and priority when AF_INET{6} sockets are created. The most closely related tests under selftests are 'test_sock' and 'sockopt'. However, these existing tests serve different purposes. 'test_sock' focuses mainly on verifying the socket binding process, while 'sockopt' concentrates on testing the behavior of getsockopt and setsockopt operations for various socket options. Neither of these existing tests directly covers the ability of cgroup BPF to set socket attributes such as sk_bound_dev_if, mark, and priority during socket creation. To address this gap, this patch introduces a migration of the old cgroup socket attribute test, now included as the 'sock_create' test in selftests/bpf. This ensures that the ability to configure these attributes during socket creation is properly tested. Signed-off-by: Daniel T. Lee Link: https://lore.kernel.org/r/20241011044847.51584-2-danieltimlee@gmail.com Signed-off-by: Alexei Starovoitov

samples/bpf: Fix a resource leak

2024-10-11T02:13:47Z

The opened file should be closed in show_sockopts(), otherwise resource leak will occur that this problem was discovered by reading code Signed-off-by: Zhu Jun Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/20241010014126.2573-1-zhujun2@cmss.chinamobile.com

samples/bpf: Remove unused variables

2024-10-10T01:22:04Z

These variables are never referenced in the code, just remove them. Signed-off-by: Zhu Jun Link: https://lore.kernel.org/r/20241009082138.7971-1-zhujun2@cmss.chinamobile.com Signed-off-by: Alexei Starovoitov

bpf: syscall_nrs: Disable no previous prototype warnning

2024-10-08T03:28:46Z

In some environments (gcc treated as error in W=1, which is default), if we make -C samples/bpf/, it will be stopped because of "no previous prototype" error like this: ../samples/bpf/syscall_nrs.c:7:6: error: no previous prototype for ‘syscall_defines’ [-Werror=missing-prototypes] void syscall_defines(void) ^~~~~~~~~~~~~~~ Actually, this file meets our expectatations because it will be converted to a .h file. In this way, it's correct. Considering the warnning stopping us compiling, we can remove the warnning directly. Signed-off-by: Jason Xing Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/all/20241001012540.39007-1-kerneljasonxing@gmail.com/ Link: https://lore.kernel.org/all/CAEf4BzaVdr_0kQo=+jPLN++PvcU6pwTjaPVEA880kgDN94TZYw@mail.gmail.com/ Link: https://lore.kernel.org/bpf/20241001233242.98679-1-kerneljasonxing@gmail.com