[PATCH bpf-next v9 00/11] selftests/bpf: Tolerate partial builds across kernel configs

[PATCH bpf-next v9 00/11] selftests/bpf: Tolerate partial builds across kernel configs

[Ricardo B. Marlière]

Currently the BPF selftests can only be built by using the minimum kernel
configuration defined in tools/testing/selftests/bpf/config*. This poses a
problem in distribution kernels that may have some of the flags disabled or
set as module. For example, we have been running the tests regularly in
openSUSE Tumbleweed [1] [2] but to work around this fact we created a
special package [3] that build the tests against an auxiliary vmlinux with
the BPF Kconfig. We keep a list of known issues that may happen due to,
amongst other things, configuration mismatches [4] [5].

The maintenance of this package is far from ideal, especially for
enterprise kernels. The goal of this series is to enable the common usecase
of running the following in any system:

```sh
make -C tools/testing/selftests install \
     SKIP_TARGETS= \
     TARGETS=bpf \
     BPF_STRICT_BUILD=0 \
     O=/lib/modules/$(uname -r)/build
```

As an example, the following script targeting a minimal config can be used
for testing:

```sh
make defconfig
scripts/config --file .config \
               --enable DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT \
               --enable DEBUG_INFO_BTF \
               --enable BPF_SYSCALL \
               --enable BPF_JIT
make olddefconfig
make -j$(nproc)
make -j$(nproc) -C tools/testing/selftests install \
     SKIP_TARGETS= \
     TARGETS=bpf \
     BPF_STRICT_BUILD=0
```

This produces a test_progs binary with 585 subtests, against the total of
717. Many of them will still fail or be skipped at runtime due to lack of
symbols, but at least there will be a clear way of building the tests.

[1]: https://openqa.opensuse.org/tests/5811715
[2]: https://openqa.opensuse.org/tests/5811730
[3]: https://src.opensuse.org/rmarliere/kselftests
[4]: https://github.com/openSUSE/kernel-qe/blob/main/kselftests_known_issues.yaml
[5]: https://openqa.opensuse.org/tests/5811730/logfile?filename=run_kselftests-config_mismatches.txt

---
Changes in v9:
- Also pass KBUILD_OUTPUT=$(KMOD_O_VALID) when invoking kbuild so an
  inherited command-line KBUILD_OUTPUT cannot disagree with O= (patch 2)
- Note BPFOBJ remains a normal prereq intentionally (patch 5)
- Sub-shell isolate cd/CC and use $@ for $(RM); gate the BTFIDS
  pretty-print on $(V) so verbose mode does not double-print (patch 6)
- Restrict permissive compile-failure tolerance and partial-link to
  test_progs%; runners with strong cross-object references (test_maps)
  keep strict semantics (patches 6 and 8)
- Link to v8: https://patch.msgid.link/20260428-selftests-bpf_misconfig-v8-0-bf02cf97dbcb@suse.com

Changes in v8:
- In permissive mode, keep source changes to already-built tests
  triggering relinks, while using recipe-time $(wildcard ...) to pick
  up fresh .test.o files without duplicate linker inputs (patch 8)
- Resolve relative O=/KBUILD_OUTPUT before recursing into test_kmods,
  and only treat it as a kernel build dir when it contains
  Module.symvers (patch 2)
- Note in the commit message that PERMISSIVE-mode bisecting here needs
  the next two patches (patch 6)
- Clarify in the commit message that order-only skeleton prereqs do not
  break the new-skel-via-local-header case (patch 5)
- Exclude not-built tests from the success count so summary and JSON
  report them only as skipped (patch 7)
- Tighten commit messages for accuracy and clarity
- Link to v7: https://patch.msgid.link/20260416-selftests-bpf_misconfig-v7-0-a078e18012e4@suse.com

Changes in v7:
- Use $(abspath) for KMOD_O so relative O= paths resolve correctly
  when make -C changes directory (patch 2)
- Guard make clean against missing KDIR unconditionally; there is
  nothing to clean when kernel headers are absent (patch 2)
- Drop explicit $(TRUNNER_TEST_OBJS) from linker filter in strict mode;
  $^ already contains them as normal prerequisites (patch 8)
- Link to v6: https://patch.msgid.link/20260416-selftests-bpf_misconfig-v6-0-7efeab504af1@suse.com

Changes in v6:
- Add --ignore-missing-args to -extras rsync so out-of-tree permissive
  builds do not abort when .ko files are absent (patch 2)
- Use $(abspath) for KMOD_O so relative O= paths resolve correctly
  when make -C changes directory (patch 2)
- Guard make clean against missing KDIR unconditionally so cleaning
  does not abort when kernel headers are absent (patch 2)
- Remove stale skeleton headers in early-exit paths when .bpf.o is
  missing on incremental builds (patch 3)
- Fix strict-mode skeleton rules: use && before temp file cleanup so
  bpftool failures are not masked by rm -f exit code (patch 3)
- Track test filter selection separately from not_built so -t/-n flags
  are respected for unbuilt tests (patch 7)
- Make TRUNNER_TEST_OBJS order-only only in permissive mode, preserving
  incremental relinking in strict builds (patch 8)
- Drop explicit $(TRUNNER_TEST_OBJS) from linker filter in strict mode;
  they are already in $^ as normal prerequisites (patch 8)
- Reorder: move skip-unbuilt-tests patch before partial-linking patch
  for bisectability
- Link to v5: https://patch.msgid.link/20260415-selftests-bpf_misconfig-v5-0-03d0a52a898a@suse.com

Changes in v5:
- Add BPF_STRICT_BUILD toggle as patch 1 so every subsequent patch
  gates tolerance behind PERMISSIVE from the start, making the series
  bisectable with strict-by-default at every point
- Fix O= commit message; make parent cp conditional (patch 2)
- Tolerate linked skeleton failures (patch 3)
- Skip feature detection for emit_tests (patch 4)
- Clarify bench is all-or-nothing in commit message (patch 8)
- Move stack_mprotect() to testing_helpers.c, drop weak stubs (patch 9)
- Report not-built tests as "SKIP (not built)" in output (patch 10)
- Drop overly broad 2>/dev/null || true from install rsync; rely solely
  on --ignore-missing-args which already handles absent files (patch 11)
- Link to v4: https://patch.msgid.link/20260406-selftests-bpf_misconfig-v4-0-9914f50efdf7@suse.com

Changes in v4:
- Drop the test_kmods kselftest module flow patch: lib.mk gen_mods_dir
  invokes $(MAKE) -C $(TEST_GEN_MODS_DIR) without forwarding
  RESOLVE_BTFIDS, breaking ASAN and GCC BPF CI builds (Makefile.modfinal
  cannot find resolve_btfids in the kbuild output tree)
- Link to v3:
  https://patch.msgid.link/20260406-selftests-bpf_misconfig-v3-0-587a1114263c@suse.com

Changes in v3:
- Split test_kmods patch into two: fix KDIR handling (O= passthrough,
  EXTRA_CFLAGS/EXTRA_LDFLAGS clearing) and wire into lib.mk via
  TEST_GEN_MODS_DIR
- Pass O= through to the kernel module build so artifacts land in the
  output tree, not the source tree
- Clear EXTRA_CFLAGS and EXTRA_LDFLAGS when invoking the kernel build to
  prevent host flags (e.g. -static) leaking into module compilation
- Replace the bespoke test_kmods pattern rule with lib.mk module
  infrastructure (TEST_GEN_MODS_DIR); lib.mk now drives build and clean
  lifecycle
- Make the .ko copy step resilient: emit SKIP instead of failing when a
  module is absent
- Expand the uprobe weak stub comment in bpf_cookie.c to explain why
  noinline is required
- Link to v2:
  https://patch.msgid.link/20260403-selftests-bpf_misconfig-v2-0-f06700380a9d@suse.com

Changes in v2:
- Skip test_kmods build/clean when KDIR directory does not exist
- Use `Module.symvers` instead of `.config` for in-tree detection
- Fix skeleton order-only prereqs commit message
- Guard BTFIDS step when .test.o is absent
- Add `__weak stack_mprotect()` stubs in `bpf_cookie.c` and `iters.c`
- Link to v1:
  https://patch.msgid.link/20260401-selftests-bpf_misconfig-v1-0-3ae42c0af76f@suse.com

To: Alexei Starovoitov <ast@kernel.org>
To: Daniel Borkmann <daniel@iogearbox.net>
To: Andrii Nakryiko <andrii@kernel.org>
To: Martin KaFai Lau <martin.lau@linux.dev>
To: Eduard Zingerman <eddyz87@gmail.com>
To: Kumar Kartikeya Dwivedi <memxor@gmail.com>
To: Song Liu <song@kernel.org>
To: Yonghong Song <yonghong.song@linux.dev>
To: Jiri Olsa <jolsa@kernel.org>
To: Shuah Khan <shuah@kernel.org>
To: Nathan Chancellor <nathan@kernel.org>
To: Nick Desaulniers <nick.desaulniers+lkml@gmail.com>
To: Bill Wendling <morbo@google.com>
To: Justin Stitt <justinstitt@google.com>
Cc: bpf@vger.kernel.org
Cc: linux-kselftest@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: llvm@lists.linux.dev
Assisted-by: {codex,claude}
Tested-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Ricardo B. Marliere <rbm@suse.com>

---
Ricardo B. Marlière (11):
      selftests/bpf: Add BPF_STRICT_BUILD toggle
      selftests/bpf: Fix test_kmods KDIR to honor O= and distro kernels
      selftests/bpf: Tolerate BPF and skeleton generation failures
      selftests/bpf: Avoid rebuilds when running emit_tests
      selftests/bpf: Make skeleton headers order-only prerequisites of .test.d
      selftests/bpf: Tolerate test file compilation failures
      selftests/bpf: Skip tests whose objects were not built
      selftests/bpf: Allow test_progs to link with a partial object set
      selftests/bpf: Tolerate benchmark build failures
      selftests/bpf: Provide weak definitions for cross-test functions
      selftests/bpf: Tolerate missing files during install

 tools/testing/selftests/bpf/Makefile               | 177 ++++++++++++++-------
 .../testing/selftests/bpf/prog_tests/bpf_cookie.c  |  17 +-
 tools/testing/selftests/bpf/prog_tests/iters.c     |   2 -
 tools/testing/selftests/bpf/prog_tests/test_lsm.c  |  22 ---
 tools/testing/selftests/bpf/test_kmods/Makefile    |  30 +++-
 tools/testing/selftests/bpf/test_progs.c           |  53 +++++-
 tools/testing/selftests/bpf/test_progs.h           |   1 +
 tools/testing/selftests/bpf/testing_helpers.c      |  17 ++
 tools/testing/selftests/bpf/testing_helpers.h      |   1 +
 9 files changed, 225 insertions(+), 95 deletions(-)
---
base-commit: 9f5b3ffc3f1dac7204e32eeeff84bc5cc55c393e
change-id: 20260401-selftests-bpf_misconfig-4c33ef5c56da

Best regards,
--  
Ricardo B. Marlière <rbm@suse.com>

[PATCH bpf-next v9 01/11] selftests/bpf: Add BPF_STRICT_BUILD toggle

[Ricardo B. Marlière]

Distro kernels often lack BTF types or kernel features required by some BPF
selftests, causing the build to abort on the first failure and preventing
the remaining tests from running.

Add BPF_STRICT_BUILD (default 1) to control build failure tolerance. When
set to 0, the PERMISSIVE make variable is assigned a non-empty value that
subsequent Makefile rules use to make individual build steps non-fatal.
When set to 1 (the default), the build fails on any error, preserving the
existing behavior for CI and direct builds.

Users can opt in to permissive mode on the command line:

  make -C tools/testing/selftests \
       TARGETS=bpf SKIP_TARGETS= BPF_STRICT_BUILD=0

Suggested-by: Alan Maguire <alan.maguire@oracle.com>
Signed-off-by: Ricardo B. Marlière <rbm@suse.com>
---
 tools/testing/selftests/bpf/Makefile | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 97ee61f2ade5..6094fe99b5f6 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -44,6 +44,12 @@ SKIP_LLVM	?=
 SKIP_LIBBFD	?=
 SKIP_CRYPTO	?=
 
+# When BPF_STRICT_BUILD is 1, any BPF object, skeleton, test object, or
+# benchmark compilation failure is fatal. Set to 0 to tolerate failures
+# and continue building the remaining tests.
+BPF_STRICT_BUILD ?= 1
+PERMISSIVE := $(filter 0,$(BPF_STRICT_BUILD))
+
 ifeq ($(srctree),)
 srctree := $(patsubst %/,%,$(dir $(CURDIR)))
 srctree := $(patsubst %/,%,$(dir $(srctree)))

-- 
2.54.0

[PATCH bpf-next v9 02/11] selftests/bpf: Fix test_kmods KDIR to honor O= and distro kernels

[Ricardo B. Marlière]

test_kmods/Makefile always pointed KDIR at the kernel source tree root,
ignoring O= and KBUILD_OUTPUT. On distro kernels where the source tree has
not been built, the Makefile had no fallback and would fail
unconditionally.

When O= or KBUILD_OUTPUT is set and points at a prepared kernel build
directory (one containing Module.symvers), pass it through so kbuild can
locate the correct build infrastructure (scripts, Kconfig, etc.). Note
that the module artifacts themselves still land in the M= directory,
which is test_kmods/; O= only controls where kbuild finds its build
infrastructure. Fall back to /lib/modules/$(uname -r)/build when neither
an explicit valid build directory nor an in-tree Module.symvers is
present.

A selftests-only O= value (one that does not contain Module.symvers, e.g.
a private output directory) is intentionally not treated as a kernel
build directory. Without this guard, a user invoking
"make -C tools/testing/selftests/bpf O=/tmp/out" would have test_kmods
try to use /tmp/out as the kernel build dir and fail.

The parent bpf/Makefile resolves O= and KBUILD_OUTPUT to absolute paths
before invoking the test_kmods sub-make. Without this, $(abspath ...)
inside test_kmods/Makefile would resolve relative paths against the
sub-make's CWD (test_kmods/) rather than the user's invocation directory.

When O= is passed to kbuild, also pass KBUILD_OUTPUT=$(KMOD_O_VALID)
explicitly. The parent invocation lifts KBUILD_OUTPUT into MAKEFLAGS as
a command-line variable, which would otherwise suppress kbuild's own
"KBUILD_OUTPUT := $(O)" assignment and cause it to use the inherited
KBUILD_OUTPUT instead of the validated O=.

Guard both all and clean against a missing KDIR so the step is silently
skipped rather than fatal. Make the parent Makefile's cp conditional so it
does not abort when modules were not built.

Signed-off-by: Ricardo B. Marlière <rbm@suse.com>
---
 tools/testing/selftests/bpf/Makefile            | 10 +++++----
 tools/testing/selftests/bpf/test_kmods/Makefile | 30 ++++++++++++++++++++++---
 2 files changed, 33 insertions(+), 7 deletions(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 6094fe99b5f6..cc6ee7a2df93 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -296,13 +296,15 @@ $(OUTPUT)/sign-file: ../../../../scripts/sign-file.c
 # subst() turns the rule into a pattern matching rule
 $(addprefix test_kmods/,$(subst .ko,%ko,$(TEST_KMODS))): $(VMLINUX_BTF) $(RESOLVE_BTFIDS) $(wildcard test_kmods/Makefile test_kmods/*.[ch])
 	$(Q)$(RM) test_kmods/*.ko test_kmods/*.mod.o # force re-compilation
-	$(Q)$(MAKE) $(submake_extras) -C test_kmods	\
-		RESOLVE_BTFIDS=$(RESOLVE_BTFIDS)	\
+	$(Q)$(MAKE) $(submake_extras) -C test_kmods				\
+		$(if $(O),O=$(abspath $(O)))					\
+		$(if $(KBUILD_OUTPUT),KBUILD_OUTPUT=$(abspath $(KBUILD_OUTPUT)))\
+		RESOLVE_BTFIDS=$(RESOLVE_BTFIDS)				\
 		EXTRA_CFLAGS='' EXTRA_LDFLAGS=''
 
 $(TEST_KMOD_TARGETS): $(addprefix test_kmods/,$(TEST_KMODS))
 	$(call msg,MOD,,$@)
-	$(Q)cp test_kmods/$(@F) $@
+	$(Q)$(if $(PERMISSIVE),if [ -f test_kmods/$(@F) ]; then )cp test_kmods/$(@F) $@$(if $(PERMISSIVE),; fi)
 
 
 DEFAULT_BPFTOOL := $(HOST_SCRATCH_DIR)/sbin/bpftool
@@ -718,7 +720,7 @@ $(TRUNNER_LIB_OBJS): $(TRUNNER_OUTPUT)/%.o:$(TOOLSDIR)/lib/%.c
 $(TRUNNER_BINARY)-extras: $(TRUNNER_EXTRA_FILES) | $(TRUNNER_OUTPUT)
 ifneq ($2:$(OUTPUT),:$(shell pwd))
 	$$(call msg,EXT-COPY,$(TRUNNER_BINARY),$(TRUNNER_EXTRA_FILES))
-	$(Q)rsync -aq $$^ $(TRUNNER_OUTPUT)/
+	$(Q)rsync -aq $(if $(PERMISSIVE),--ignore-missing-args) $$^ $(TRUNNER_OUTPUT)/
 endif
 
 # some X.test.o files have runtime dependencies on Y.bpf.o files
diff --git a/tools/testing/selftests/bpf/test_kmods/Makefile b/tools/testing/selftests/bpf/test_kmods/Makefile
index 63c4d3f6a12f..031c7454ce65 100644
--- a/tools/testing/selftests/bpf/test_kmods/Makefile
+++ b/tools/testing/selftests/bpf/test_kmods/Makefile
@@ -1,5 +1,16 @@
 TEST_KMOD_DIR := $(realpath $(dir $(abspath $(lastword $(MAKEFILE_LIST)))))
-KDIR ?= $(abspath $(TEST_KMOD_DIR)/../../../../..)
+SRCTREE_KDIR := $(abspath $(TEST_KMOD_DIR)/../../../../..)
+# Honor O=/KBUILD_OUTPUT only if they point at a prepared kernel build
+# directory (one containing Module.symvers); otherwise treat the value as a
+# selftests-only output directory and fall back to in-tree or distro headers.
+# The parent bpf/Makefile resolves O=/KBUILD_OUTPUT to absolute paths before
+# invoking this sub-make so relative paths still anchor to the user's
+# invocation directory.
+KMOD_O := $(or $(O),$(KBUILD_OUTPUT))
+KMOD_O_VALID := $(if $(KMOD_O),$(if $(wildcard $(KMOD_O)/Module.symvers),$(KMOD_O)))
+KDIR ?= $(if $(KMOD_O_VALID),$(SRCTREE_KDIR), \
+	    $(if $(wildcard $(SRCTREE_KDIR)/Module.symvers),$(SRCTREE_KDIR), \
+		/lib/modules/$(shell uname -r)/build))
 
 ifeq ($(V),1)
 Q =
@@ -14,8 +25,21 @@ $(foreach m,$(MODULES),$(eval obj-m += $(m:.ko=.o)))
 
 CFLAGS_bpf_testmod.o = -I$(src)
 
+# When BPF_STRICT_BUILD != 0, a missing KDIR is fatal (the default).
+# When permissive, skip silently.
+PERMISSIVE := $(filter 0,$(BPF_STRICT_BUILD))
+
 all:
-	$(Q)$(MAKE) -C $(KDIR) M=$(TEST_KMOD_DIR) modules
+ifeq ($(PERMISSIVE),)
+	$(Q)$(MAKE) -C $(KDIR) $(if $(KMOD_O_VALID),O=$(KMOD_O_VALID) KBUILD_OUTPUT=$(KMOD_O_VALID),KBUILD_OUTPUT=) \
+		M=$(TEST_KMOD_DIR) modules
+else ifneq ("$(wildcard $(KDIR))", "")
+	$(Q)$(MAKE) -C $(KDIR) $(if $(KMOD_O_VALID),O=$(KMOD_O_VALID) KBUILD_OUTPUT=$(KMOD_O_VALID),KBUILD_OUTPUT=) \
+		M=$(TEST_KMOD_DIR) modules
+endif
 
 clean:
-	$(Q)$(MAKE) -C $(KDIR) M=$(TEST_KMOD_DIR) clean
+ifneq ("$(wildcard $(KDIR))", "")
+	$(Q)$(MAKE) -C $(KDIR) $(if $(KMOD_O_VALID),O=$(KMOD_O_VALID) KBUILD_OUTPUT=$(KMOD_O_VALID),KBUILD_OUTPUT=) \
+		M=$(TEST_KMOD_DIR) clean
+endif

-- 
2.54.0

[PATCH bpf-next v9 03/11] selftests/bpf: Tolerate BPF and skeleton generation failures

[Ricardo B. Marlière]

Some BPF programs cannot be built on distro kernels because required BTF
types or features are missing. A single failure currently aborts the
selftests/bpf build.

Make BPF object and skeleton generation best effort in permissive mode:
emit SKIP-BPF or SKIP-SKEL to stderr, remove failed outputs so downstream
rules can detect absence, and continue with remaining tests. Apply the same
tolerance to linked skeletons (TRUNNER_BPF_SKELS_LINKED), which depend on
multiple .bpf.o files and abort the build when any dependency is missing.

Note that progress messages (GEN-SKEL, LINK-BPF) are also redirected to
stderr as a side effect of rewriting the recipes into single-shell
pipelines; the $(call msg,...) macro is a make-recipe construct that cannot
be used inside an &&-chained shell command sequence.

Signed-off-by: Ricardo B. Marlière <rbm@suse.com>
---
 tools/testing/selftests/bpf/Makefile | 108 +++++++++++++++++++++++------------
 1 file changed, 73 insertions(+), 35 deletions(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index cc6ee7a2df93..b104c687dcf0 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -485,22 +485,26 @@ $(OUTPUT)/cgroup_getset_retval_hooks.o: cgroup_getset_retval_hooks.h
 # $4 - binary name
 define CLANG_BPF_BUILD_RULE
 	$(call msg,CLNG-BPF,$4,$2)
-	$(Q)$(CLANG) $3 -O2 $(BPF_TARGET_ENDIAN) -c $1 -mcpu=v3 -o $2
+	$(Q)$(CLANG) $3 -O2 $(BPF_TARGET_ENDIAN) -c $1 -mcpu=v3 -o $2 $(if $(PERMISSIVE),|| \
+		($(RM) $2; printf '  %-12s %s\n' 'SKIP-BPF' '$(notdir $2)' 1>&2))
 endef
 # Similar to CLANG_BPF_BUILD_RULE, but with disabled alu32
 define CLANG_NOALU32_BPF_BUILD_RULE
 	$(call msg,CLNG-BPF,$4,$2)
-	$(Q)$(CLANG) $3 -O2 $(BPF_TARGET_ENDIAN) -c $1 -mcpu=v2 -o $2
+	$(Q)$(CLANG) $3 -O2 $(BPF_TARGET_ENDIAN) -c $1 -mcpu=v2 -o $2 $(if $(PERMISSIVE),|| \
+		($(RM) $2; printf '  %-12s %s\n' 'SKIP-BPF' '$(notdir $2)' 1>&2))
 endef
 # Similar to CLANG_BPF_BUILD_RULE, but with cpu-v4
 define CLANG_CPUV4_BPF_BUILD_RULE
 	$(call msg,CLNG-BPF,$4,$2)
-	$(Q)$(CLANG) $3 -O2 $(BPF_TARGET_ENDIAN) -c $1 -mcpu=v4 -o $2
+	$(Q)$(CLANG) $3 -O2 $(BPF_TARGET_ENDIAN) -c $1 -mcpu=v4 -o $2 $(if $(PERMISSIVE),|| \
+		($(RM) $2; printf '  %-12s %s\n' 'SKIP-BPF' '$(notdir $2)' 1>&2))
 endef
 # Build BPF object using GCC
 define GCC_BPF_BUILD_RULE
 	$(call msg,GCC-BPF,$4,$2)
-	$(Q)$(BPF_GCC) $3 -DBPF_NO_PRESERVE_ACCESS_INDEX -Wno-attributes -O2 -c $1 -o $2
+	$(Q)$(BPF_GCC) $3 -DBPF_NO_PRESERVE_ACCESS_INDEX -Wno-attributes -O2 -c $1 -o $2 $(if $(PERMISSIVE),|| \
+		($(RM) $2; printf '  %-12s %s\n' 'SKIP-BPF' '$(notdir $2)' 1>&2))
 endef
 
 SKEL_BLACKLIST := btf__% test_pinning_invalid.c test_sk_assign.c
@@ -607,47 +611,81 @@ $(TRUNNER_BPF_OBJS): $(TRUNNER_OUTPUT)/%.bpf.o:				\
 					  $$($$<-$2-CFLAGS),$(TRUNNER_BINARY))
 
 $(TRUNNER_BPF_SKELS): %.skel.h: %.bpf.o $(BPFTOOL) | $(TRUNNER_OUTPUT)
-	$$(call msg,GEN-SKEL,$(TRUNNER_BINARY),$$@)
-	$(Q)$$(BPFTOOL) gen object $$(<:.o=.linked1.o) $$<
-	$(Q)$$(BPFTOOL) gen object $$(<:.o=.linked2.o) $$(<:.o=.linked1.o)
-	$(Q)$$(BPFTOOL) gen object $$(<:.o=.linked3.o) $$(<:.o=.linked2.o)
-	$(Q)diff $$(<:.o=.linked2.o) $$(<:.o=.linked3.o)
-	$(Q)$$(BPFTOOL) gen skeleton $$(<:.o=.linked3.o) name $$(notdir $$(<:.bpf.o=)) > $$@
-	$(Q)$$(BPFTOOL) gen subskeleton $$(<:.o=.linked3.o) name $$(notdir $$(<:.bpf.o=)) > $$(@:.skel.h=.subskel.h)
-	$(Q)rm -f $$(<:.o=.linked1.o) $$(<:.o=.linked2.o) $$(<:.o=.linked3.o)
+	$(Q)$(if $(PERMISSIVE),if [ ! -f $$< ]; then			\
+		$$(RM) $$@ $$(@:.skel.h=.subskel.h);			\
+		printf '  %-12s %s\n' 'SKIP-SKEL' '$$(notdir $$@)' 1>&2; \
+		exit 0;							\
+	fi;)								\
+	printf '  %-12s %s\n' 'GEN-SKEL' '[$(TRUNNER_BINARY)] $$(notdir $$@)' 1>&2; \
+	$$(BPFTOOL) gen object $$(<:.o=.linked1.o) $$< &&		\
+	$$(BPFTOOL) gen object $$(<:.o=.linked2.o) $$(<:.o=.linked1.o) && \
+	$$(BPFTOOL) gen object $$(<:.o=.linked3.o) $$(<:.o=.linked2.o) && \
+	diff $$(<:.o=.linked2.o) $$(<:.o=.linked3.o) &&		\
+	$$(BPFTOOL) gen skeleton $$(<:.o=.linked3.o) name $$(notdir $$(<:.bpf.o=)) > $$@ && \
+	$$(BPFTOOL) gen subskeleton $$(<:.o=.linked3.o) name $$(notdir $$(<:.bpf.o=)) > $$(@:.skel.h=.subskel.h) $(if $(PERMISSIVE),|| { \
+		$$(RM) $$@ $$(@:.skel.h=.subskel.h); \
+		printf '  %-12s %s\n' 'SKIP-SKEL' '$$(notdir $$@)' 1>&2; \
+	}) && \
+	rm -f $$(<:.o=.linked1.o) $$(<:.o=.linked2.o) $$(<:.o=.linked3.o)
 
 $(TRUNNER_BPF_LSKELS): %.lskel.h: %.bpf.o $(BPFTOOL) | $(TRUNNER_OUTPUT)
-	$$(call msg,GEN-SKEL,$(TRUNNER_BINARY),$$@)
-	$(Q)$$(BPFTOOL) gen object $$(<:.o=.llinked1.o) $$<
-	$(Q)$$(BPFTOOL) gen object $$(<:.o=.llinked2.o) $$(<:.o=.llinked1.o)
-	$(Q)$$(BPFTOOL) gen object $$(<:.o=.llinked3.o) $$(<:.o=.llinked2.o)
-	$(Q)diff $$(<:.o=.llinked2.o) $$(<:.o=.llinked3.o)
-	$(Q)$$(BPFTOOL) gen skeleton -L $$(<:.o=.llinked3.o) name $$(notdir $$(<:.bpf.o=_lskel)) > $$@
-	$(Q)rm -f $$(<:.o=.llinked1.o) $$(<:.o=.llinked2.o) $$(<:.o=.llinked3.o)
+	$(Q)$(if $(PERMISSIVE),if [ ! -f $$< ]; then			\
+		$$(RM) $$@;						\
+		printf '  %-12s %s\n' 'SKIP-SKEL' '$$(notdir $$@)' 1>&2; \
+		exit 0;							\
+	fi;)								\
+	printf '  %-12s %s\n' 'GEN-SKEL' '[$(TRUNNER_BINARY)] $$(notdir $$@)' 1>&2; \
+	$$(BPFTOOL) gen object $$(<:.o=.llinked1.o) $$< &&		\
+	$$(BPFTOOL) gen object $$(<:.o=.llinked2.o) $$(<:.o=.llinked1.o) && \
+	$$(BPFTOOL) gen object $$(<:.o=.llinked3.o) $$(<:.o=.llinked2.o) && \
+	diff $$(<:.o=.llinked2.o) $$(<:.o=.llinked3.o) &&		\
+	$$(BPFTOOL) gen skeleton -L $$(<:.o=.llinked3.o) name $$(notdir $$(<:.bpf.o=_lskel)) > $$@ $(if $(PERMISSIVE),|| { \
+		$$(RM) $$@; \
+		printf '  %-12s %s\n' 'SKIP-SKEL' '$$(notdir $$@)' 1>&2; \
+	}) && \
+	rm -f $$(<:.o=.llinked1.o) $$(<:.o=.llinked2.o) $$(<:.o=.llinked3.o)
 
 $(TRUNNER_BPF_LSKELS_SIGNED): %.lskel.h: %.bpf.o $(BPFTOOL) | $(TRUNNER_OUTPUT)
-	$$(call msg,GEN-SKEL,$(TRUNNER_BINARY) (signed),$$@)
-	$(Q)$$(BPFTOOL) gen object $$(<:.o=.llinked1.o) $$<
-	$(Q)$$(BPFTOOL) gen object $$(<:.o=.llinked2.o) $$(<:.o=.llinked1.o)
-	$(Q)$$(BPFTOOL) gen object $$(<:.o=.llinked3.o) $$(<:.o=.llinked2.o)
-	$(Q)diff $$(<:.o=.llinked2.o) $$(<:.o=.llinked3.o)
-	$(Q)$$(BPFTOOL) gen skeleton $(LSKEL_SIGN) $$(<:.o=.llinked3.o) name $$(notdir $$(<:.bpf.o=_lskel)) > $$@
-	$(Q)rm -f $$(<:.o=.llinked1.o) $$(<:.o=.llinked2.o) $$(<:.o=.llinked3.o)
+	$(Q)$(if $(PERMISSIVE),if [ ! -f $$< ]; then			\
+		$$(RM) $$@;						\
+		printf '  %-12s %s\n' 'SKIP-SKEL' '$$(notdir $$@)' 1>&2; \
+		exit 0;							\
+	fi;)								\
+	printf '  %-12s %s\n' 'GEN-SKEL' '[$(TRUNNER_BINARY) (signed)] $$(notdir $$@)' 1>&2; \
+	$$(BPFTOOL) gen object $$(<:.o=.llinked1.o) $$< &&		\
+	$$(BPFTOOL) gen object $$(<:.o=.llinked2.o) $$(<:.o=.llinked1.o) && \
+	$$(BPFTOOL) gen object $$(<:.o=.llinked3.o) $$(<:.o=.llinked2.o) && \
+	diff $$(<:.o=.llinked2.o) $$(<:.o=.llinked3.o) &&		\
+	$$(BPFTOOL) gen skeleton $(LSKEL_SIGN) $$(<:.o=.llinked3.o) name $$(notdir $$(<:.bpf.o=_lskel)) > $$@ $(if $(PERMISSIVE),|| { \
+		$$(RM) $$@; \
+		printf '  %-12s %s\n' 'SKIP-SKEL' '$$(notdir $$@)' 1>&2; \
+	}) && \
+	rm -f $$(<:.o=.llinked1.o) $$(<:.o=.llinked2.o) $$(<:.o=.llinked3.o)
 
 $(LINKED_BPF_OBJS): %: $(TRUNNER_OUTPUT)/%
 
 # .SECONDEXPANSION here allows to correctly expand %-deps variables as prerequisites
 .SECONDEXPANSION:
 $(TRUNNER_BPF_SKELS_LINKED): $(TRUNNER_OUTPUT)/%: $$$$(%-deps) $(BPFTOOL) | $(TRUNNER_OUTPUT)
-	$$(call msg,LINK-BPF,$(TRUNNER_BINARY),$$(@:.skel.h=.bpf.o))
-	$(Q)$$(BPFTOOL) gen object $$(@:.skel.h=.linked1.o) $$(addprefix $(TRUNNER_OUTPUT)/,$$($$(@F)-deps))
-	$(Q)$$(BPFTOOL) gen object $$(@:.skel.h=.linked2.o) $$(@:.skel.h=.linked1.o)
-	$(Q)$$(BPFTOOL) gen object $$(@:.skel.h=.linked3.o) $$(@:.skel.h=.linked2.o)
-	$(Q)diff $$(@:.skel.h=.linked2.o) $$(@:.skel.h=.linked3.o)
-	$$(call msg,GEN-SKEL,$(TRUNNER_BINARY),$$@)
-	$(Q)$$(BPFTOOL) gen skeleton $$(@:.skel.h=.linked3.o) name $$(notdir $$(@:.skel.h=)) > $$@
-	$(Q)$$(BPFTOOL) gen subskeleton $$(@:.skel.h=.linked3.o) name $$(notdir $$(@:.skel.h=)) > $$(@:.skel.h=.subskel.h)
-	$(Q)rm -f $$(@:.skel.h=.linked1.o) $$(@:.skel.h=.linked2.o) $$(@:.skel.h=.linked3.o)
+	$(Q)$(if $(PERMISSIVE),for f in $$(addprefix $(TRUNNER_OUTPUT)/,$$($$(@F)-deps)); do \
+		if [ ! -f $$$$f ]; then						\
+			$$(RM) $$@ $$(@:.skel.h=.subskel.h);		\
+			printf '  %-12s %s\n' 'SKIP-SKEL' '$$(notdir $$@)' 1>&2; \
+			exit 0;							\
+		fi;								\
+	done;)									\
+	printf '  %-12s %s\n' 'LINK-BPF' '[$(TRUNNER_BINARY)] $$(notdir $$(@:.skel.h=.bpf.o))' 1>&2; \
+	$$(BPFTOOL) gen object $$(@:.skel.h=.linked1.o) $$(addprefix $(TRUNNER_OUTPUT)/,$$($$(@F)-deps)) && \
+	$$(BPFTOOL) gen object $$(@:.skel.h=.linked2.o) $$(@:.skel.h=.linked1.o) && \
+	$$(BPFTOOL) gen object $$(@:.skel.h=.linked3.o) $$(@:.skel.h=.linked2.o) && \
+	diff $$(@:.skel.h=.linked2.o) $$(@:.skel.h=.linked3.o) &&	\
+	printf '  %-12s %s\n' 'GEN-SKEL' '[$(TRUNNER_BINARY)] $$(notdir $$@)' 1>&2 && \
+	$$(BPFTOOL) gen skeleton $$(@:.skel.h=.linked3.o) name $$(notdir $$(@:.skel.h=)) > $$@ && \
+	$$(BPFTOOL) gen subskeleton $$(@:.skel.h=.linked3.o) name $$(notdir $$(@:.skel.h=)) > $$(@:.skel.h=.subskel.h) $(if $(PERMISSIVE),|| { \
+		$$(RM) $$@ $$(@:.skel.h=.subskel.h);			\
+		printf '  %-12s %s\n' 'SKIP-SKEL' '$$(notdir $$@)' 1>&2; \
+	}) &&									\
+	rm -f $$(@:.skel.h=.linked1.o) $$(@:.skel.h=.linked2.o) $$(@:.skel.h=.linked3.o)
 
 # When the compiler generates a %.d file, only skel basenames (not
 # full paths) are specified as prerequisites for corresponding %.o

-- 
2.54.0

[PATCH bpf-next v9 04/11] selftests/bpf: Avoid rebuilds when running emit_tests

[Ricardo B. Marlière]

emit_tests is used while installing selftests to generate the kselftest
list. Pulling in .d files for this goal can trigger BPF rebuild rules and
mix build output into list generation.

Skip dependency file inclusion for emit_tests, like clean goals, so list
generation stays side-effect free. Also add emit_tests to
NON_CHECK_FEAT_TARGETS so that feature detection is skipped; without this,
Makefile.feature's $(info) output leaks into stdout and corrupts the test
list captured by the top-level selftests Makefile.

Signed-off-by: Ricardo B. Marlière <rbm@suse.com>
---
 tools/testing/selftests/bpf/Makefile | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index b104c687dcf0..9a4d8bea0c18 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -170,7 +170,7 @@ endef
 
 include ../lib.mk
 
-NON_CHECK_FEAT_TARGETS := clean docs-clean
+NON_CHECK_FEAT_TARGETS := clean docs-clean emit_tests
 CHECK_FEAT := $(filter-out $(NON_CHECK_FEAT_TARGETS),$(or $(MAKECMDGOALS), "none"))
 ifneq ($(CHECK_FEAT),)
 FEATURE_USER := .selftests
@@ -732,7 +732,7 @@ $(TRUNNER_TEST_OBJS:.o=.d): $(TRUNNER_OUTPUT)/%.test.d:			\
 			    $(TRUNNER_BPF_SKELS_LINKED)			\
 			    $$(BPFOBJ) | $(TRUNNER_OUTPUT)
 
-ifeq ($(filter clean docs-clean,$(MAKECMDGOALS)),)
+ifeq ($(filter clean docs-clean emit_tests,$(MAKECMDGOALS)),)
 include $(wildcard $(TRUNNER_TEST_OBJS:.o=.d))
 endif
 

-- 
2.54.0

[PATCH bpf-next v9 05/11] selftests/bpf: Make skeleton headers order-only prerequisites of .test.d

[Ricardo B. Marlière]

The .test.d dependency files are generated by the C preprocessor and list
the headers each test file actually #includes. Skeleton headers appear in
those generated lists, so the .test.o -> .skel.h dependency is already
tracked by the .d file content.

Making skeletons order-only prerequisites of .test.d means that a missing
or skipped skeleton does not prevent .test.d generation, and regenerating a
skeleton does not force .test.d to be recreated. This avoids unnecessary
recompilation and, more importantly, avoids build errors when a skeleton
was intentionally skipped due to a BPF compilation failure.

$$(BPFOBJ) is intentionally kept as a normal prerequisite: a libbpf
rebuild legitimately invalidates .test.d, since libbpf header changes
can affect the headers .test.o sees. Only the skeleton headers are
moved to order-only.

Note that adding a new BPF skeleton via a modified existing local header
still works correctly: GNU make builds order-only prerequisites that do
not exist (the order-only qualifier only suppresses timestamp-driven
rebuilds, not existence-driven builds), so a brand-new .skel.h listed in
TRUNNER_BPF_SKELS is generated even when .test.d is otherwise up to date.
The modified local header invalidates .test.o through the previously
included .d content, forcing a recompile that regenerates .test.d with
the new .skel.h dependency captured by gcc -MMD.

Signed-off-by: Ricardo B. Marlière <rbm@suse.com>
---
 tools/testing/selftests/bpf/Makefile | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 9a4d8bea0c18..608e6dff0392 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -726,11 +726,11 @@ $(TRUNNER_TEST_OBJS): $(TRUNNER_OUTPUT)/%.test.o:			\
 $(TRUNNER_TEST_OBJS:.o=.d): $(TRUNNER_OUTPUT)/%.test.d:			\
 			    $(TRUNNER_TESTS_DIR)/%.c			\
 			    $(TRUNNER_EXTRA_HDRS)			\
+			    $$(BPFOBJ) | $(TRUNNER_OUTPUT)		\
 			    $(TRUNNER_BPF_SKELS)			\
 			    $(TRUNNER_BPF_LSKELS)			\
 			    $(TRUNNER_BPF_LSKELS_SIGNED)		\
-			    $(TRUNNER_BPF_SKELS_LINKED)			\
-			    $$(BPFOBJ) | $(TRUNNER_OUTPUT)
+			    $(TRUNNER_BPF_SKELS_LINKED)
 
 ifeq ($(filter clean docs-clean emit_tests,$(MAKECMDGOALS)),)
 include $(wildcard $(TRUNNER_TEST_OBJS:.o=.d))

-- 
2.54.0

[PATCH bpf-next v9 06/11] selftests/bpf: Tolerate test file compilation failures

[Ricardo B. Marlière]

Individual test files may fail to compile when headers or kernel features
required by that test are absent. Currently this aborts the entire build.

Make the per-test compilation non-fatal: remove the output object on
failure and print a SKIP-TEST marker to stderr. Guard the BTFIDS
post-processing step so it is skipped when the object file is absent. The
linker step will later ignore absent objects, allowing the remaining tests
to build and run.

Group cd and CC in a sub-shell so a cd failure cannot leak into the
error-handling branch and operate in the original working directory; use
$@ (absolute path) for $(RM) so it cannot match an unrelated file there.

Replace the $(call msg,...) in the BTFIDS block with a plain printf
(the msg macro expands to @printf, which is a make-recipe construct and
is invalid inside a shell if-then-fi body) and gate the printf on $(V)
so verbose mode does not double-print the line that the recipe shell
already echoes.

Restrict tolerance to test_progs and its flavors via an inlined
$(if $(filter test_progs%,$1),$(if $(PERMISSIVE),...)) check: runners
with strong cross-object references (e.g. test_maps) would link-fail
with a partial object set, so they keep strict semantics even when
BPF_STRICT_BUILD=0. The check is inlined rather than stored in a helper
variable so $1 is substituted at $(call) time and the per-runner result
is baked into each recipe.

Note on bisectability: this change is gated entirely behind PERMISSIVE
for test_progs%, so default builds (BPF_STRICT_BUILD!=0) compile and
run identically at every commit in the series. Bisecting in PERMISSIVE
mode at this commit still requires the next two patches ("selftests/bpf:
Skip tests whose objects were not built" and "selftests/bpf: Allow
test_progs to link with a partial object set") to avoid the linker
rejecting missing objects and the runtime aborting on NULL function
pointers.

Signed-off-by: Ricardo B. Marlière <rbm@suse.com>
---
 tools/testing/selftests/bpf/Makefile | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 608e6dff0392..9becc077eb23 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -588,6 +588,12 @@ endef
 # $2 - test runner extra "flavor" (e.g., no_alu32, cpuv4, bpf_gcc, etc)
 define DEFINE_TEST_RUNNER_RULES
 
+# Permissive build behaviour (skip-on-failure compile, partial-link) only
+# applies to test_progs and its flavors; runners that use strong cross-object
+# references (e.g. test_maps) keep strict semantics even when permissive.
+# The check is inlined per-runner so $1 is substituted at $(call) time and
+# the result is baked into each rule's recipe.
+
 ifeq ($($(TRUNNER_OUTPUT)-dir),)
 $(TRUNNER_OUTPUT)-dir := y
 $(TRUNNER_OUTPUT):
@@ -717,11 +723,14 @@ $(TRUNNER_TEST_OBJS): $(TRUNNER_OUTPUT)/%.test.o:			\
 		      $(TRUNNER_TESTS_DIR)/%.c				\
 		      | $(TRUNNER_OUTPUT)/%.test.d
 	$$(call msg,TEST-OBJ,$(TRUNNER_BINARY),$$@)
-	$(Q)cd $$(@D) && $$(CC) -I. $$(CFLAGS) -MMD -MT $$@ -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)
+	$(Q)(cd $$(@D) && $$(CC) -I. $$(CFLAGS) -MMD -MT $$@ -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)) $(if $(filter test_progs%,$1),$(if $(PERMISSIVE),|| \
+		($(RM) $$@; printf '  %-12s %s\n' 'SKIP-TEST' '$$(notdir $$@)' 1>&2)))
 	$$(if $$(TEST_NEEDS_BTFIDS),						\
-		$$(call msg,BTFIDS,$(TRUNNER_BINARY),$$@)			\
+		$(Q)if [ -f $$@ ]; then						\
+		$(if $(V),true,printf '  %-8s%s %s\n' "BTFIDS" " [$(TRUNNER_BINARY)]" "$$(notdir $$@)"); \
 		$(RESOLVE_BTFIDS) --btf $(TRUNNER_OUTPUT)/btf_data.bpf.o $$@;	\
-		$(RESOLVE_BTFIDS) --patch_btfids $$@.BTF_ids $$@)
+		$(RESOLVE_BTFIDS) --patch_btfids $$@.BTF_ids $$@;		\
+		fi)
 
 $(TRUNNER_TEST_OBJS:.o=.d): $(TRUNNER_OUTPUT)/%.test.d:			\
 			    $(TRUNNER_TESTS_DIR)/%.c			\

-- 
2.54.0

[PATCH bpf-next v9 07/11] selftests/bpf: Skip tests whose objects were not built

[Ricardo B. Marlière]

When both run_test and run_serial_test are NULL (because the corresponding
.test.o was not compiled), mark the test as not built instead of fatally
aborting.

Report these tests as "SKIP (not built)" in per-test output and include
them in the skip count so they remain visible in CI results and JSON
output. The summary line shows the not-built count when nonzero:

  Summary: 50/55 PASSED, 5 SKIPPED (3 not built), 0 FAILED

Tests filtered out by -t/-n remain invisible as before; only genuinely
unbuilt tests are surfaced.

Signed-off-by: Ricardo B. Marlière <rbm@suse.com>
---
 tools/testing/selftests/bpf/test_progs.c | 53 +++++++++++++++++++++++++++-----
 tools/testing/selftests/bpf/test_progs.h |  1 +
 2 files changed, 46 insertions(+), 8 deletions(-)

diff --git a/tools/testing/selftests/bpf/test_progs.c b/tools/testing/selftests/bpf/test_progs.c
index cc14b13e23fe..7ba82974ee78 100644
--- a/tools/testing/selftests/bpf/test_progs.c
+++ b/tools/testing/selftests/bpf/test_progs.c
@@ -165,6 +165,8 @@ struct prog_test_def {
 	void (*run_test)(void);
 	void (*run_serial_test)(void);
 	bool should_run;
+	bool not_built;
+	bool selected;
 	bool need_cgroup_cleanup;
 	bool should_tmon;
 };
@@ -372,6 +374,8 @@ static void print_test_result(const struct prog_test_def *test, const struct tes
 	fprintf(env.stdout_saved, "#%-*d %s:", TEST_NUM_WIDTH, test->test_num, test->test_name);
 	if (test_state->error_cnt)
 		fprintf(env.stdout_saved, "FAIL");
+	else if (test->not_built)
+		fprintf(env.stdout_saved, "SKIP (not built)");
 	else if (!skipped_cnt)
 		fprintf(env.stdout_saved, "OK");
 	else if (skipped_cnt == subtests_cnt || !subtests_cnt)
@@ -1641,6 +1645,7 @@ static void calculate_summary_and_print_errors(struct test_env *env)
 	json_writer_t *w = NULL;
 
 	for (i = 0; i < prog_test_cnt; i++) {
+		struct prog_test_def *test = &prog_test_defs[i];
 		struct test_state *state = &test_states[i];
 
 		if (!state->tested)
@@ -1651,7 +1656,7 @@ static void calculate_summary_and_print_errors(struct test_env *env)
 
 		if (state->error_cnt)
 			fail_cnt++;
-		else
+		else if (!test->not_built)
 			succ_cnt++;
 	}
 
@@ -1700,8 +1705,13 @@ static void calculate_summary_and_print_errors(struct test_env *env)
 	if (env->json)
 		fclose(env->json);
 
-	printf("Summary: %d/%d PASSED, %d SKIPPED, %d FAILED\n",
-	       succ_cnt, sub_succ_cnt, skip_cnt, fail_cnt);
+	if (env->not_built_cnt)
+		printf("Summary: %d/%d PASSED, %d SKIPPED (%d not built), %d FAILED\n",
+		       succ_cnt, sub_succ_cnt, skip_cnt, env->not_built_cnt,
+		       fail_cnt);
+	else
+		printf("Summary: %d/%d PASSED, %d SKIPPED, %d FAILED\n",
+		       succ_cnt, sub_succ_cnt, skip_cnt, fail_cnt);
 
 	env->succ_cnt = succ_cnt;
 	env->sub_succ_cnt = sub_succ_cnt;
@@ -1772,6 +1782,19 @@ static void server_main(void)
 		run_one_test(i);
 	}
 
+	/* mark not-built tests as skipped */
+	for (int i = 0; i < prog_test_cnt; i++) {
+		struct prog_test_def *test = &prog_test_defs[i];
+		struct test_state *state = &test_states[i];
+
+		if (test->not_built && test->selected) {
+			state->tested = true;
+			state->skip_cnt = 1;
+			env.not_built_cnt++;
+			print_test_result(test, state);
+		}
+	}
+
 	/* generate summary */
 	fflush(stderr);
 	fflush(stdout);
@@ -2046,15 +2069,20 @@ int main(int argc, char **argv)
 		struct prog_test_def *test = &prog_test_defs[i];
 
 		test->test_num = i + 1;
-		test->should_run = should_run(&env.test_selector,
-					      test->test_num, test->test_name);
+		test->selected = should_run(&env.test_selector,
+					    test->test_num, test->test_name);
+		test->should_run = test->selected;
 
-		if ((test->run_test == NULL && test->run_serial_test == NULL) ||
-		    (test->run_test != NULL && test->run_serial_test != NULL)) {
+		if (test->run_test && test->run_serial_test) {
 			fprintf(stderr, "Test %d:%s must have either test_%s() or serial_test_%sl() defined.\n",
 				test->test_num, test->test_name, test->test_name, test->test_name);
 			exit(EXIT_ERR_SETUP_INFRA);
 		}
+		if (!test->run_test && !test->run_serial_test) {
+			test->not_built = true;
+			test->should_run = false;
+			continue;
+		}
 		if (test->should_run)
 			test->should_tmon = should_tmon(&env.tmon_selector, test->test_name);
 	}
@@ -2106,9 +2134,18 @@ int main(int argc, char **argv)
 
 	for (i = 0; i < prog_test_cnt; i++) {
 		struct prog_test_def *test = &prog_test_defs[i];
+		struct test_state *state = &test_states[i];
 
-		if (!test->should_run)
+		if (!test->should_run) {
+			if (test->not_built && test->selected &&
+			    !env.get_test_cnt && !env.list_test_names) {
+				state->tested = true;
+				state->skip_cnt = 1;
+				env.not_built_cnt++;
+				print_test_result(test, state);
+			}
 			continue;
+		}
 
 		if (env.get_test_cnt) {
 			env.succ_cnt++;
diff --git a/tools/testing/selftests/bpf/test_progs.h b/tools/testing/selftests/bpf/test_progs.h
index 37955a8ad385..2cf950afcd85 100644
--- a/tools/testing/selftests/bpf/test_progs.h
+++ b/tools/testing/selftests/bpf/test_progs.h
@@ -125,6 +125,7 @@ struct test_env {
 	int sub_succ_cnt; /* successful sub-tests */
 	int fail_cnt; /* total failed tests + sub-tests */
 	int skip_cnt; /* skipped tests */
+	int not_built_cnt; /* tests not built */
 
 	int saved_netns_fd;
 	int workers; /* number of worker process */

-- 
2.54.0

[PATCH bpf-next v9 08/11] selftests/bpf: Allow test_progs to link with a partial object set

[Ricardo B. Marlière]

When individual test files are skipped due to compilation failures, their
.test.o files are absent. The linker step currently lists all expected
.test.o files as explicit prerequisites, so make considers any missing one
an error.

In permissive mode, declare the test objects that already exist on disk
(via parse-time $(wildcard ...)) as normal prerequisites of the binary so
that modifications to a test source still trigger a relink, and keep the
full TRUNNER_TEST_OBJS list as order-only prerequisites so that initial
fresh builds still produce them and missing objects do not abort the link.
The recipe filter is split per mode: in permissive mode it combines a
recipe-time $(wildcard ...) (which catches objects freshly produced via
the order-only path on a fresh build) with $(filter-out
$(TRUNNER_TEST_OBJS),$^) (which keeps the non-test inputs from $^ but
drops the parse-time wildcard duplicates). This avoids passing the same
.test.o twice to the linker while still presenting test objects before
libbpf.a so that GNU ld, which scans static archives left-to-right, pulls
in archive members referenced exclusively by test objects (e.g.
ring_buffer__new from ringbuf.c). In default (strict) mode the recipe
remains the simple $(filter %.a %.o,$^) since TRUNNER_TEST_OBJS is part
of $^ exactly once.

Gate the partial-link behavior on $(if $(filter test_progs%,$1),...) so
it only applies to test_progs and its flavors. test_maps and similar
runners using strong cross-object references would link-fail with a
partial set and intentionally retain strict link semantics.

Note: adding a brand-new test_*.c file in permissive mode requires
removing the binary (or a clean rebuild) before the new test is linked
in, because the parse-time $(wildcard ...) is evaluated when the Makefile
is read and will not yet see the new .test.o. This is acceptable since
permissive mode targets tolerant CI builds rather than incremental
development.

Signed-off-by: Ricardo B. Marlière <rbm@suse.com>
---
 tools/testing/selftests/bpf/Makefile | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 9becc077eb23..a64e822dc540 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -773,14 +773,15 @@ endif
 # some X.test.o files have runtime dependencies on Y.bpf.o files
 $(OUTPUT)/$(TRUNNER_BINARY): | $(TRUNNER_BPF_OBJS)
 
-$(OUTPUT)/$(TRUNNER_BINARY): $(TRUNNER_TEST_OBJS)			\
+$(OUTPUT)/$(TRUNNER_BINARY): $(if $(filter test_progs%,$1),$(if $(PERMISSIVE),$$(wildcard $(TRUNNER_TEST_OBJS)),$(TRUNNER_TEST_OBJS)),$(TRUNNER_TEST_OBJS))	\
 			     $(TRUNNER_EXTRA_OBJS) $$(BPFOBJ)		\
 			     $(TRUNNER_LIB_OBJS)			\
 			     $(TRUNNER_BPFTOOL)				\
 			     $(OUTPUT)/veristat				\
-			     | $(TRUNNER_BINARY)-extras
+			     | $(TRUNNER_BINARY)-extras			\
+			     $(if $(filter test_progs%,$1),$(if $(PERMISSIVE),$(TRUNNER_TEST_OBJS)))
 	$$(call msg,BINARY,,$$@)
-	$(Q)$$(CC) $$(CFLAGS) $$(filter %.a %.o,$$^) $$(LDLIBS) $$(LLVM_LDLIBS) $$(LDFLAGS) $$(LLVM_LDFLAGS) -o $$@
+	$(Q)$$(CC) $$(CFLAGS) $(if $(filter test_progs%,$1),$(if $(PERMISSIVE),$$(filter %.a %.o,$$(wildcard $(TRUNNER_TEST_OBJS)) $$(filter-out $(TRUNNER_TEST_OBJS),$$^)),$$(filter %.a %.o,$$^)),$$(filter %.a %.o,$$^)) $$(LDLIBS) $$(LLVM_LDLIBS) $$(LDFLAGS) $$(LLVM_LDFLAGS) -o $$@
 	$(Q)ln -sf $(if $2,..,.)/tools/build/bpftool/$(USE_BOOTSTRAP)bpftool \
 		   $(OUTPUT)/$(if $2,$2/)bpftool
 

-- 
2.54.0

[PATCH bpf-next v9 09/11] selftests/bpf: Tolerate benchmark build failures

[Ricardo B. Marlière]

Benchmark objects depend on skeletons that may be missing when some BPF
programs fail to build. In that case, benchmark object compilation or final
bench linking should not abort the full selftests/bpf build.

Keep both steps non-fatal, emit SKIP-BENCH or SKIP-LINK, and remove failed
outputs so stale objects or binaries are not reused by later incremental
builds. Note that because bench.c statically references every benchmark via
extern symbols, partial linking is not possible: if any single benchmark
object fails, the entire bench binary is skipped. This is by design -- the
error handler catches all compilation failures including genuine ones, but
those are caught by full-config CI runs.

Signed-off-by: Ricardo B. Marlière <rbm@suse.com>
---
 tools/testing/selftests/bpf/Makefile | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index a64e822dc540..db4ff951fc88 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -945,7 +945,8 @@ $(OUTPUT)/test_cpp: test_cpp.cpp $(OUTPUT)/test_core_extern.skel.h $(BPFOBJ)
 # Benchmark runner
 $(OUTPUT)/bench_%.o: benchs/bench_%.c bench.h $(BPFOBJ)
 	$(call msg,CC,,$@)
-	$(Q)$(CC) $(CFLAGS) -O2 -c $(filter %.c,$^) $(LDLIBS) -o $@
+	$(Q)$(CC) $(CFLAGS) -O2 -c $(filter %.c,$^) $(LDLIBS) -o $@ $(if $(PERMISSIVE),|| \
+		($(RM) $@; printf '  %-12s %s\n' 'SKIP-BENCH' '$(notdir $@)' 1>&2))
 $(OUTPUT)/bench_rename.o: $(OUTPUT)/test_overhead.skel.h
 $(OUTPUT)/bench_trigger.o: $(OUTPUT)/trigger_bench.skel.h
 $(OUTPUT)/bench_ringbufs.o: $(OUTPUT)/ringbuf_bench.skel.h \
@@ -988,7 +989,8 @@ $(OUTPUT)/bench: $(OUTPUT)/bench.o \
 		 $(OUTPUT)/usdt_2.o \
 		 #
 	$(call msg,BINARY,,$@)
-	$(Q)$(CC) $(CFLAGS) $(LDFLAGS) $(filter %.a %.o,$^) $(LDLIBS) -o $@
+	$(Q)$(CC) $(CFLAGS) $(LDFLAGS) $(filter %.a %.o,$(if $(PERMISSIVE),$(wildcard $^),$^)) $(LDLIBS) -o $@ $(if $(PERMISSIVE),|| \
+		($(RM) $@; printf '  %-12s %s\n' 'SKIP-LINK' '$(notdir $@) (some benchmarks may have been skipped)' 1>&2))
 
 # This works around GCC warning about snprintf truncating strings like:
 #

-- 
2.54.0

[PATCH bpf-next v9 10/11] selftests/bpf: Provide weak definitions for cross-test functions

[Ricardo B. Marlière]

Some test files reference functions defined in other translation units that
may not be compiled when skeletons are missing. Replace forward
declarations of uprobe_multi_func_{1,2,3}() with weak no-op stubs so the
linker resolves them regardless of which objects are present.

Move stack_mprotect() from test_lsm.c into testing_helpers.c so it is
always available. The previous weak-stub approach returned 0, which would
cause callers expecting -1/EPERM to fail their assertions
deterministically. Having the real implementation in a shared utility
avoids this problem entirely.

Signed-off-by: Ricardo B. Marlière <rbm@suse.com>
---
 .../testing/selftests/bpf/prog_tests/bpf_cookie.c  | 17 +++++++++++------
 tools/testing/selftests/bpf/prog_tests/iters.c     |  2 --
 tools/testing/selftests/bpf/prog_tests/test_lsm.c  | 22 ----------------------
 tools/testing/selftests/bpf/testing_helpers.c      | 17 +++++++++++++++++
 tools/testing/selftests/bpf/testing_helpers.h      |  1 +
 5 files changed, 29 insertions(+), 30 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_cookie.c b/tools/testing/selftests/bpf/prog_tests/bpf_cookie.c
index 35adc3f6d443..5a864cd8ad1b 100644
--- a/tools/testing/selftests/bpf/prog_tests/bpf_cookie.c
+++ b/tools/testing/selftests/bpf/prog_tests/bpf_cookie.c
@@ -252,10 +252,17 @@ static void kprobe_multi_attach_api_subtest(void)
 	kprobe_multi__destroy(skel);
 }
 
-/* defined in prog_tests/uprobe_multi_test.c */
-void uprobe_multi_func_1(void);
-void uprobe_multi_func_2(void);
-void uprobe_multi_func_3(void);
+/*
+ * Weak uprobe target stubs. noinline is required because
+ * uprobe_multi_test_run() takes their addresses to configure the BPF
+ * program's attachment points; an inlined function has no stable
+ * address in the binary to probe. The strong definitions in
+ * uprobe_multi_test.c take precedence when that translation unit is
+ * linked.
+ */
+noinline __weak void uprobe_multi_func_1(void) {}
+noinline __weak void uprobe_multi_func_2(void) {}
+noinline __weak void uprobe_multi_func_3(void) {}
 
 static void uprobe_multi_test_run(struct uprobe_multi *skel)
 {
@@ -574,8 +581,6 @@ static void tracing_subtest(struct test_bpf_cookie *skel)
 		close(fmod_ret_fd);
 }
 
-int stack_mprotect(void);
-
 static void lsm_subtest(struct test_bpf_cookie *skel)
 {
 	__u64 cookie;
diff --git a/tools/testing/selftests/bpf/prog_tests/iters.c b/tools/testing/selftests/bpf/prog_tests/iters.c
index a539980a2fbe..c0b6082f345a 100644
--- a/tools/testing/selftests/bpf/prog_tests/iters.c
+++ b/tools/testing/selftests/bpf/prog_tests/iters.c
@@ -202,8 +202,6 @@ static void subtest_task_iters(void)
 	iters_task__destroy(skel);
 }
 
-extern int stack_mprotect(void);
-
 static void subtest_css_task_iters(void)
 {
 	struct iters_css_task *skel = NULL;
diff --git a/tools/testing/selftests/bpf/prog_tests/test_lsm.c b/tools/testing/selftests/bpf/prog_tests/test_lsm.c
index bdc4fc06bc5a..d7495efd4a56 100644
--- a/tools/testing/selftests/bpf/prog_tests/test_lsm.c
+++ b/tools/testing/selftests/bpf/prog_tests/test_lsm.c
@@ -5,36 +5,14 @@
  */
 
 #include <test_progs.h>
-#include <sys/mman.h>
 #include <sys/wait.h>
 #include <unistd.h>
-#include <malloc.h>
-#include <stdlib.h>
 
 #include "lsm.skel.h"
 #include "lsm_tailcall.skel.h"
 
 char *CMD_ARGS[] = {"true", NULL};
 
-#define GET_PAGE_ADDR(ADDR, PAGE_SIZE)					\
-	(char *)(((unsigned long) (ADDR + PAGE_SIZE)) & ~(PAGE_SIZE-1))
-
-int stack_mprotect(void)
-{
-	void *buf;
-	long sz;
-	int ret;
-
-	sz = sysconf(_SC_PAGESIZE);
-	if (sz < 0)
-		return sz;
-
-	buf = alloca(sz * 3);
-	ret = mprotect(GET_PAGE_ADDR(buf, sz), sz,
-		       PROT_READ | PROT_WRITE | PROT_EXEC);
-	return ret;
-}
-
 int exec_cmd(int *monitored_pid)
 {
 	int child_pid, child_status;
diff --git a/tools/testing/selftests/bpf/testing_helpers.c b/tools/testing/selftests/bpf/testing_helpers.c
index 6fbe1e995660..3f7ae76c209b 100644
--- a/tools/testing/selftests/bpf/testing_helpers.c
+++ b/tools/testing/selftests/bpf/testing_helpers.c
@@ -5,6 +5,7 @@
 #include <stdlib.h>
 #include <string.h>
 #include <errno.h>
+#include <sys/mman.h>
 #include <bpf/bpf.h>
 #include <bpf/libbpf.h>
 #include "disasm.h"
@@ -516,3 +517,19 @@ bool is_jit_enabled(void)
 
 	return enabled;
 }
+
+int stack_mprotect(void)
+{
+	void *buf;
+	long sz;
+	int ret;
+
+	sz = sysconf(_SC_PAGESIZE);
+	if (sz < 0)
+		return sz;
+
+	buf = alloca(sz * 3);
+	ret = mprotect((void *)(((unsigned long)(buf + sz)) & ~(sz - 1)), sz,
+		       PROT_READ | PROT_WRITE | PROT_EXEC);
+	return ret;
+}
diff --git a/tools/testing/selftests/bpf/testing_helpers.h b/tools/testing/selftests/bpf/testing_helpers.h
index 2ca2356a0b58..2edc6fb7fc52 100644
--- a/tools/testing/selftests/bpf/testing_helpers.h
+++ b/tools/testing/selftests/bpf/testing_helpers.h
@@ -59,5 +59,6 @@ struct bpf_insn;
 int get_xlated_program(int fd_prog, struct bpf_insn **buf, __u32 *cnt);
 int testing_prog_flags(void);
 bool is_jit_enabled(void);
+int stack_mprotect(void);
 
 #endif /* __TESTING_HELPERS_H */

-- 
2.54.0

[PATCH bpf-next v9 11/11] selftests/bpf: Tolerate missing files during install

[Ricardo B. Marlière]

With partial builds, some TEST_GEN_FILES entries can be absent at install
time. rsync treats missing source arguments as fatal and aborts kselftest
installation.

Override INSTALL_SINGLE_RULE in selftests/bpf to use --ignore-missing-args,
while keeping the existing bpf-specific INSTALL_RULE extension logic. Also
add --ignore-missing-args to the TEST_INST_SUBDIRS rsync loop so that
subdirectories with no .bpf.o files (e.g. when a test runner flavor was
skipped) do not abort installation.

Note that the INSTALL_SINGLE_RULE override applies globally to all file
categories including static source files (TEST_PROGS, TEST_FILES). These
are version-controlled and should always be present, so the practical risk
is negligible.

Signed-off-by: Ricardo B. Marlière <rbm@suse.com>
---
 tools/testing/selftests/bpf/Makefile | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index db4ff951fc88..2822f09c37eb 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -1023,12 +1023,23 @@ EXTRA_CLEAN := $(SCRATCH_DIR) $(HOST_SCRATCH_DIR)			\
 # Delete partially updated (corrupted) files on error
 .DELETE_ON_ERROR:
 
+# When permissive, tell rsync to ignore missing source arguments so that
+# partial builds do not abort installation.
+ifneq ($(PERMISSIVE),)
+override define INSTALL_SINGLE_RULE
+	$(if $(INSTALL_LIST),@mkdir -p $(INSTALL_PATH))
+	$(if $(INSTALL_LIST),rsync -a --copy-unsafe-links --ignore-missing-args $(INSTALL_LIST) $(INSTALL_PATH)/)
+endef
+endif
+
 DEFAULT_INSTALL_RULE := $(INSTALL_RULE)
 override define INSTALL_RULE
 	$(DEFAULT_INSTALL_RULE)
-	@for DIR in $(TEST_INST_SUBDIRS); do		  \
-		mkdir -p $(INSTALL_PATH)/$$DIR;   \
-		rsync -a $(OUTPUT)/$$DIR/*.bpf.o $(INSTALL_PATH)/$$DIR;\
+	@for DIR in $(TEST_INST_SUBDIRS); do				  \
+		mkdir -p $(INSTALL_PATH)/$$DIR;				  \
+		rsync -a $(if $(PERMISSIVE),--ignore-missing-args)	  \
+			$(OUTPUT)/$$DIR/*.bpf.o				  \
+			$(INSTALL_PATH)/$$DIR;				  \
 	done
 endef
 

-- 
2.54.0

Re: [PATCH bpf-next v9 01/11] selftests/bpf: Add BPF_STRICT_BUILD toggle

[bot+bpf-ci]

[-- Attachment #1: Type: text/plain, Size: 1988 bytes --]

> diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
> index 97ee61f2ade5..6094fe99b5f6 100644
> --- a/tools/testing/selftests/bpf/Makefile
> +++ b/tools/testing/selftests/bpf/Makefile
> @@ -44,6 +44,12 @@ SKIP_LLVM	?=
>  SKIP_LIBBFD	?=
>  SKIP_CRYPTO	?=
>
> +# When BPF_STRICT_BUILD is 1, any BPF object, skeleton, test object, or
> +# benchmark compilation failure is fatal. Set to 0 to tolerate failures
> +# and continue building the remaining tests.
> +BPF_STRICT_BUILD ?= 1
> +PERMISSIVE := $(filter 0,$(BPF_STRICT_BUILD))
> +
>  ifeq ($(srctree),)
>  srctree := $(patsubst %/,%,$(dir $(CURDIR)))
>  srctree := $(patsubst %/,%,$(dir $(srctree)))

Multiple reviewers (Sashiko AI and bot+bpf-ci@kernel.org) raised a concern
in v5 that when permissive mode skips building .ko files, the
$(TRUNNER_BINARY)-extras rsync target will fail fatally during out-of-tree
builds (when O= is set) because TEST_KMOD_TARGETS are included in
TRUNNER_EXTRA_FILES but the .ko files were never created.

The rsync command lacks --ignore-missing-args flag to handle missing files:

tools/testing/selftests/bpf/Makefile:$(TRUNNER_BINARY)-extras {
    $(call msg,EXT-COPY,$(TRUNNER_BINARY),$(TRUNNER_EXTRA_FILES))
    $(Q)rsync -aq $$^ $(TRUNNER_OUTPUT)/
}

According to the v6 changelog, this was addressed by adding
--ignore-missing-args to the -extras rsync, but this fix does not appear
in the current commit. Does the rsync command need --ignore-missing-args
to prevent 'No such file or directory' errors when .ko files are missing
in permissive mode during out-of-tree builds?

Reference:
https://lore.kernel.org/bpf/20260415211828.BFAFFC19424@smtp.kernel.org/
Reviewer: Sashiko AI
Date: Wed, 15 Apr 2026 21:18:28 +0000


---
AI reviewed your patch. Please fix the bug or email reply why it's not a bug.
See: https://github.com/kernel-patches/vmtest/blob/master/ci/claude/README.md

CI run summary: https://github.com/kernel-patches/bpf/actions/runs/25115689001

Re: [PATCH bpf-next v9 03/11] selftests/bpf: Tolerate BPF and skeleton generation failures

[bot+bpf-ci]

[-- Attachment #1: Type: text/plain, Size: 3404 bytes --]

> diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
> index cc6ee7a2df93..b104c687dcf0 100644
> --- a/tools/testing/selftests/bpf/Makefile
> +++ b/tools/testing/selftests/bpf/Makefile

[ ... ]

> @@ -607,47 +611,81 @@ $(TRUNNER_BPF_OBJS): $(TRUNNER_OUTPUT)/%.bpf.o:				\
>  					  $$($$<-$2-CFLAGS),$(TRUNNER_BINARY))
>
>  $(TRUNNER_BPF_SKELS): %.skel.h: %.bpf.o $(BPFTOOL) | $(TRUNNER_OUTPUT)
> -	$$(call msg,GEN-SKEL,$(TRUNNER_BINARY),$$@)
> -	$(Q)$$(BPFTOOL) gen object $$(<:.o=.linked1.o) $$<
> -	$(Q)$$(BPFTOOL) gen object $$(<:.o=.linked2.o) $$(<:.o=.linked1.o)
> -	$(Q)$$(BPFTOOL) gen object $$(<:.o=.linked3.o) $$(<:.o=.linked2.o)
> -	$(Q)diff $$(<:.o=.linked2.o) $$(<:.o=.linked3.o)
> -	$(Q)$$(BPFTOOL) gen skeleton $$(<:.o=.linked3.o) name $$(notdir $$(<:.bpf.o=)) > $$@
> -	$(Q)$$(BPFTOOL) gen subskeleton $$(<:.o=.linked3.o) name $$(notdir $$(<:.bpf.o=)) > $$(@:.skel.h=.subskel.h)
> -	$(Q)rm -f $$(<:.o=.linked1.o) $$(<:.o=.linked2.o) $$(<:.o=.linked3.o)
> +	$(Q)$(if $(PERMISSIVE),if [ ! -f $$< ]; then			\
> +		$$(RM) $$@ $$(@:.skel.h=.subskel.h);			\
> +		printf '  %-12s %s\n' 'SKIP-SKEL' '$$(notdir $$@)' 1>&2; \
> +		exit 0;							\
> +	fi;)								\
> +	printf '  %-12s %s\n' 'GEN-SKEL' '[$(TRUNNER_BINARY)] $$(notdir $$@)' 1>&2; \
> +	$$(BPFTOOL) gen object $$(<:.o=.linked1.o) $$< &&		\
> +	$$(BPFTOOL) gen object $$(<:.o=.linked2.o) $$(<:.o=.linked1.o) && \
> +	$$(BPFTOOL) gen object $$(<:.o=.linked3.o) $$(<:.o=.linked2.o) && \
> +	diff $$(<:.o=.linked2.o) $$(<:.o=.linked3.o) &&		\
> +	$$(BPFTOOL) gen skeleton $$(<:.o=.linked3.o) name $$(notdir $$(<:.bpf.o=)) > $$@ && \
> +	$$(BPFTOOL) gen subskeleton $$(<:.o=.linked3.o) name $$(notdir $$(<:.bpf.o=)) > $$(@:.skel.h=.subskel.h) $(if $(PERMISSIVE),|| { \
> +		$$(RM) $$@ $$(@:.skel.h=.subskel.h); \
> +		printf '  %-12s %s\n' 'SKIP-SKEL' '$$(notdir $$@)' 1>&2; \
> +	}) && \
> +	rm -f $$(<:.o=.linked1.o) $$(<:.o=.linked2.o) $$(<:.o=.linked3.o)

Does the permissive mode error handler only apply to the final command?

The TRUNNER_BPF_SKELS rule chains six commands together with &&:

    $(BPFTOOL) gen object $(<:.o=.linked1.o) $< &&
    $(BPFTOOL) gen object $(<:.o=.linked2.o) $(<:.o=.linked1.o) &&
    $(BPFTOOL) gen object $(<:.o=.linked3.o) $(<:.o=.linked2.o) &&
    diff $(<:.o=.linked2.o) $(<:.o=.linked3.o) &&
    $(BPFTOOL) gen skeleton ... > $@ &&
    $(BPFTOOL) gen subskeleton ... > $(@:.skel.h=.subskel.h) || { ... }

If any of the first five commands fail, the && chain breaks and the
recipe exits immediately with an error status, causing Make to abort the
build. The error handler (|| { $(RM) ...; printf 'SKIP-SKEL' ...; })
only catches failures from the final gen subskeleton command.

This appears to contradict the commit message, which states the intent is
to "continue with remaining tests" when generation fails. The same pattern
affects TRUNNER_BPF_LSKELS, TRUNNER_BPF_LSKELS_SIGNED, and
TRUNNER_BPF_SKELS_LINKED.

To fully implement permissive mode, would it make sense to attach the
error handler to the entire pipeline by wrapping the command sequence in
a subshell?

    ( cmd1 && cmd2 && ... ) || { error_handler }


---
AI reviewed your patch. Please fix the bug or email reply why it's not a bug.
See: https://github.com/kernel-patches/vmtest/blob/master/ci/claude/README.md

CI run summary: https://github.com/kernel-patches/bpf/actions/runs/25115689001

Re: [PATCH bpf-next v9 03/11] selftests/bpf: Tolerate BPF and skeleton generation failures

[Ricardo B. Marlière]

On Wed Apr 29, 2026 at 12:13 PM -03, bot+bpf-ci wrote:
>> diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
>> index cc6ee7a2df93..b104c687dcf0 100644
>> --- a/tools/testing/selftests/bpf/Makefile
>> +++ b/tools/testing/selftests/bpf/Makefile
>
> [ ... ]
>
>> @@ -607,47 +611,81 @@ $(TRUNNER_BPF_OBJS): $(TRUNNER_OUTPUT)/%.bpf.o:				\
>>  					  $$($$<-$2-CFLAGS),$(TRUNNER_BINARY))
>>
>>  $(TRUNNER_BPF_SKELS): %.skel.h: %.bpf.o $(BPFTOOL) | $(TRUNNER_OUTPUT)
>> -	$$(call msg,GEN-SKEL,$(TRUNNER_BINARY),$$@)
>> -	$(Q)$$(BPFTOOL) gen object $$(<:.o=.linked1.o) $$<
>> -	$(Q)$$(BPFTOOL) gen object $$(<:.o=.linked2.o) $$(<:.o=.linked1.o)
>> -	$(Q)$$(BPFTOOL) gen object $$(<:.o=.linked3.o) $$(<:.o=.linked2.o)
>> -	$(Q)diff $$(<:.o=.linked2.o) $$(<:.o=.linked3.o)
>> -	$(Q)$$(BPFTOOL) gen skeleton $$(<:.o=.linked3.o) name $$(notdir $$(<:.bpf.o=)) > $$@
>> -	$(Q)$$(BPFTOOL) gen subskeleton $$(<:.o=.linked3.o) name $$(notdir $$(<:.bpf.o=)) > $$(@:.skel.h=.subskel.h)
>> -	$(Q)rm -f $$(<:.o=.linked1.o) $$(<:.o=.linked2.o) $$(<:.o=.linked3.o)
>> +	$(Q)$(if $(PERMISSIVE),if [ ! -f $$< ]; then			\
>> +		$$(RM) $$@ $$(@:.skel.h=.subskel.h);			\
>> +		printf '  %-12s %s\n' 'SKIP-SKEL' '$$(notdir $$@)' 1>&2; \
>> +		exit 0;							\
>> +	fi;)								\
>> +	printf '  %-12s %s\n' 'GEN-SKEL' '[$(TRUNNER_BINARY)] $$(notdir $$@)' 1>&2; \
>> +	$$(BPFTOOL) gen object $$(<:.o=.linked1.o) $$< &&		\
>> +	$$(BPFTOOL) gen object $$(<:.o=.linked2.o) $$(<:.o=.linked1.o) && \
>> +	$$(BPFTOOL) gen object $$(<:.o=.linked3.o) $$(<:.o=.linked2.o) && \
>> +	diff $$(<:.o=.linked2.o) $$(<:.o=.linked3.o) &&		\
>> +	$$(BPFTOOL) gen skeleton $$(<:.o=.linked3.o) name $$(notdir $$(<:.bpf.o=)) > $$@ && \
>> +	$$(BPFTOOL) gen subskeleton $$(<:.o=.linked3.o) name $$(notdir $$(<:.bpf.o=)) > $$(@:.skel.h=.subskel.h) $(if $(PERMISSIVE),|| { \
>> +		$$(RM) $$@ $$(@:.skel.h=.subskel.h); \
>> +		printf '  %-12s %s\n' 'SKIP-SKEL' '$$(notdir $$@)' 1>&2; \
>> +	}) && \
>> +	rm -f $$(<:.o=.linked1.o) $$(<:.o=.linked2.o) $$(<:.o=.linked3.o)
>
> Does the permissive mode error handler only apply to the final command?
>
> The TRUNNER_BPF_SKELS rule chains six commands together with &&:
>
>     $(BPFTOOL) gen object $(<:.o=.linked1.o) $< &&
>     $(BPFTOOL) gen object $(<:.o=.linked2.o) $(<:.o=.linked1.o) &&
>     $(BPFTOOL) gen object $(<:.o=.linked3.o) $(<:.o=.linked2.o) &&
>     diff $(<:.o=.linked2.o) $(<:.o=.linked3.o) &&
>     $(BPFTOOL) gen skeleton ... > $@ &&
>     $(BPFTOOL) gen subskeleton ... > $(@:.skel.h=.subskel.h) || { ... }
>
> If any of the first five commands fail, the && chain breaks and the
> recipe exits immediately with an error status, causing Make to abort the
> build. The error handler (|| { $(RM) ...; printf 'SKIP-SKEL' ...; })
> only catches failures from the final gen subskeleton command.

This is false. POSIX shell && and || have equal precedence and are
left-associative [1].

A failure of any command in the && chain causes the chain's exit status to
be non-zero, which the || handler then catches. Make recipes are invoked as
/bin/sh -c 'recipe' without -e, so the chain doesn't abort early.

Example:
  $ false && echo B && echo C || echo HANDLER
  HANDLER
  $ echo A && false && echo C || echo HANDLER
  A
  HANDLER
  $ echo A && echo B && false || echo HANDLER
  A
  B
  HANDLER

[1]: https://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#tag_18_09_03

Thanks,

>
> This appears to contradict the commit message, which states the intent is
> to "continue with remaining tests" when generation fails. The same pattern
> affects TRUNNER_BPF_LSKELS, TRUNNER_BPF_LSKELS_SIGNED, and
> TRUNNER_BPF_SKELS_LINKED.
>
> To fully implement permissive mode, would it make sense to attach the
> error handler to the entire pipeline by wrapping the command sequence in
> a subshell?
>
>     ( cmd1 && cmd2 && ... ) || { error_handler }
>
>
> ---
> AI reviewed your patch. Please fix the bug or email reply why it's not a bug.
> See: https://github.com/kernel-patches/vmtest/blob/master/ci/claude/README.md
>
> CI run summary: https://github.com/kernel-patches/bpf/actions/runs/25115689001

Re: [PATCH bpf-next v9 01/11] selftests/bpf: Add BPF_STRICT_BUILD toggle

[Ricardo B. Marlière]

On Wed Apr 29, 2026 at 12:13 PM -03, bot+bpf-ci wrote:
>> diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
>> index 97ee61f2ade5..6094fe99b5f6 100644
>> --- a/tools/testing/selftests/bpf/Makefile
>> +++ b/tools/testing/selftests/bpf/Makefile
>> @@ -44,6 +44,12 @@ SKIP_LLVM	?=
>>  SKIP_LIBBFD	?=
>>  SKIP_CRYPTO	?=
>>
>> +# When BPF_STRICT_BUILD is 1, any BPF object, skeleton, test object, or
>> +# benchmark compilation failure is fatal. Set to 0 to tolerate failures
>> +# and continue building the remaining tests.
>> +BPF_STRICT_BUILD ?= 1
>> +PERMISSIVE := $(filter 0,$(BPF_STRICT_BUILD))
>> +
>>  ifeq ($(srctree),)
>>  srctree := $(patsubst %/,%,$(dir $(CURDIR)))
>>  srctree := $(patsubst %/,%,$(dir $(srctree)))
>
> Multiple reviewers (Sashiko AI and bot+bpf-ci@kernel.org) raised a concern
> in v5 that when permissive mode skips building .ko files, the
> $(TRUNNER_BINARY)-extras rsync target will fail fatally during out-of-tree
> builds (when O= is set) because TEST_KMOD_TARGETS are included in
> TRUNNER_EXTRA_FILES but the .ko files were never created.
>
> The rsync command lacks --ignore-missing-args flag to handle missing files:
>
> tools/testing/selftests/bpf/Makefile:$(TRUNNER_BINARY)-extras {
>     $(call msg,EXT-COPY,$(TRUNNER_BINARY),$(TRUNNER_EXTRA_FILES))
>     $(Q)rsync -aq $$^ $(TRUNNER_OUTPUT)/
> }
>
> According to the v6 changelog, this was addressed by adding
> --ignore-missing-args to the -extras rsync, but this fix does not appear
> in the current commit. Does the rsync command need --ignore-missing-args
> to prevent 'No such file or directory' errors when .ko files are missing
> in permissive mode during out-of-tree builds?

This patch only adds BPF_STRICT_BUILD/PERMISSIVE variables. The rsync
fix comes next in "[PATCH bpf-next v9 02/11] selftests/bpf: Fix
test_kmods KDIR to honor O= and distro kernels".

>
> Reference:
> https://lore.kernel.org/bpf/20260415211828.BFAFFC19424@smtp.kernel.org/
> Reviewer: Sashiko AI
> Date: Wed, 15 Apr 2026 21:18:28 +0000
>
>
> ---
> AI reviewed your patch. Please fix the bug or email reply why it's not a bug.
> See: https://github.com/kernel-patches/vmtest/blob/master/ci/claude/README.md
>
> CI run summary: https://github.com/kernel-patches/bpf/actions/runs/25115689001

Re: [PATCH bpf-next v9 02/11] selftests/bpf: Fix test_kmods KDIR to honor O= and distro kernels

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] Automatical fallback to host kernel headers (`/lib/modules/$(uname -r)/build`) for building in-tree BPF test modules will cause confusing build failures and defeats the permissive skip logic.
--

commit ffa2987a74e06bf51c1a3ffcf2c85b1504b4d383
Author: Ricardo B. Marlière <rbm@suse.com>

selftests/bpf: Fix test_kmods KDIR to honor O= and distro kernels

This commit updates the test_kmods Makefile to honor O= and
KBUILD_OUTPUT directories, and falls back to distro kernel headers
when the source tree is unconfigured.

> diff --git a/tools/testing/selftests/bpf/test_kmods/Makefile b/tools/testing/selftests/bpf/test_kmods/Makefile
> index 63c4d3f6a12f..031c7454ce65 100644
> --- a/tools/testing/selftests/bpf/test_kmods/Makefile
> +++ b/tools/testing/selftests/bpf/test_kmods/Makefile
> @@ -1,5 +1,16 @@
[ ... ]
> +KMOD_O := $(or $(O),$(KBUILD_OUTPUT))
> +KMOD_O_VALID := $(if $(KMOD_O),$(if $(wildcard $(KMOD_O)/Module.symvers),$(KMOD_O)))
> +KDIR ?= $(if $(KMOD_O_VALID),$(SRCTREE_KDIR), \
> +	    $(if $(wildcard $(SRCTREE_KDIR)/Module.symvers),$(SRCTREE_KDIR), \
> +		/lib/modules/$(shell uname -r)/build))

Will falling back to the host kernel headers cause compilation errors when
building in-tree test modules like bpf_testmod?

Since bpf_testmod relies heavily on internal BPF structures, experimental
macros, and kfuncs that change frequently between kernel releases, compiling
it against /lib/modules/$(uname -r)/build on a system where the running
kernel does not exactly match the upstream source tree will likely fail.

Additionally, if the host headers are present, the permissive skip check
further down in the file will pass since the directory exists.

Because the compilation failure is not ignored by the parent bpf/Makefile,
won't this abort the entire BPF selftests build with confusing C compilation
errors instead of silently skipping the unconfigured kernel tree?

Should in-tree test modules avoid falling back to host distro headers to
prevent these version mismatches?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260429-selftests-bpf_misconfig-v9-0-c311f06b4791@suse.com?part=2

Re: [PATCH bpf-next v9 05/11] selftests/bpf: Make skeleton headers order-only prerequisites of .test.d

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Moving skeleton headers to order-only prerequisites breaks incremental builds and introduces a race condition for newly added BPF skeletons.
--

commit ac3be1ba89ba94d2363d76ee3f21b9f53e47cc23
Author: Ricardo B. Marlière <rbm@suse.com>

selftests/bpf: Make skeleton headers order-only prerequisites of .test.d

This commit changes the Makefile to make skeleton headers order-only
prerequisites of .test.d files. It aims to avoid unnecessary recompilation
and build errors when a skeleton is skipped due to BPF compilation failure.

>     Note that adding a new BPF skeleton via a modified existing local header
>     still works correctly: GNU make builds order-only prerequisites that do
>     not exist (the order-only qualifier only suppresses timestamp-driven
>     rebuilds, not existence-driven builds), so a brand-new .skel.h listed in
>     TRUNNER_BPF_SKELS is generated even when .test.d is otherwise up to date.

Is this assumption about GNU Make's behavior correct?

If a target like .test.d already exists and its regular prerequisites are not
newer, Make ignores its order-only prerequisites completely. Because the
generated .d files only declare dependencies for .test.o and not for .test.d,
modifying a local header to include a new skeleton does not mark .test.d as
out-of-date.

> diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
> index 9a4d8bea0c18..608e6dff0392 100644
> --- a/tools/testing/selftests/bpf/Makefile
> +++ b/tools/testing/selftests/bpf/Makefile
> @@ -726,11 +726,11 @@ $(TRUNNER_TEST_OBJS): $(TRUNNER_OUTPUT)/%.test.o:			\
>  $(TRUNNER_TEST_OBJS:.o=.d): $(TRUNNER_OUTPUT)/%.test.d:			\
>  			    $(TRUNNER_TESTS_DIR)/%.c			\
>  			    $(TRUNNER_EXTRA_HDRS)			\
> +			    $$(BPFOBJ) | $(TRUNNER_OUTPUT)		\
>  			    $(TRUNNER_BPF_SKELS)			\
>  			    $(TRUNNER_BPF_LSKELS)			\
>  			    $(TRUNNER_BPF_LSKELS_SIGNED)		\
> -			    $(TRUNNER_BPF_SKELS_LINKED)			\
> -			    $$(BPFOBJ) | $(TRUNNER_OUTPUT)
> +			    $(TRUNNER_BPF_SKELS_LINKED)

Does this introduce a race condition or build failures for newly added BPF
skeletons?

Since the new skeleton is not generated before compilation in the scenario
described above, it could cause a missing file error when building .test.o.

Furthermore, if the new skeleton is concurrently being generated for another
test that does correctly track it, Make will not enforce dependency ordering
for the modified test. This could result in gcc reading a partially written
.skel.h file.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260429-selftests-bpf_misconfig-v9-0-c311f06b4791@suse.com?part=5

Re: [PATCH bpf-next v9 06/11] selftests/bpf: Tolerate test file compilation failures

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] The custom `printf` replacement for the `BTFIDS` `msg` macro incorrectly suppresses output on `make V=0` and introduces inconsistent log formatting.
--

commit e98acd9f7c5260c9afc1f002b79650f156ef776d
Author: Ricardo B. Marlière <rbm@suse.com>

selftests/bpf: Tolerate test file compilation failures

This commit makes per-test compilation non-fatal in the BPF selftests Makefile
by removing the object file on failure and printing a SKIP-TEST marker. It
guards the BTFIDS post-processing step to skip missing object files and uses a
custom printf to replace the msg macro inside the shell if-statement.

> diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
> --- a/tools/testing/selftests/bpf/Makefile
> +++ b/tools/testing/selftests/bpf/Makefile
[ ... ]
> @@ -717,11 +723,14 @@ $(TRUNNER_TEST_OBJS): $(TRUNNER_OUTPUT)/%.test.o:			\
>  		      $(TRUNNER_TESTS_DIR)/%.c				\
>  		      | $(TRUNNER_OUTPUT)/%.test.d
>  	$$(call msg,TEST-OBJ,$(TRUNNER_BINARY),$$@)
> -	$(Q)cd $$(@D) && $$(CC) -I. $$(CFLAGS) -MMD -MT $$@ -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)
> +	$(Q)(cd $$(@D) && $$(CC) -I. $$(CFLAGS) -MMD -MT $$@ -c $(CURDIR)/$$< $$(LDLIBS) -o $$(@F)) $(if $(filter test_progs%,$1),$(if $(PERMISSIVE),|| \
> +		($(RM) $$@; printf '  %-12s %s\n' 'SKIP-TEST' '$$(notdir $$@)' 1>&2)))
>  	$$(if $$(TEST_NEEDS_BTFIDS),						\
> -		$$(call msg,BTFIDS,$(TRUNNER_BINARY),$$@)			\
> +		$(Q)if [ -f $$@ ]; then						\
> +		$(if $(V),true,printf '  %-8s%s %s\n' "BTFIDS" " [$(TRUNNER_BINARY)]" "$$(notdir $$@)"); \

This isn't a bug, but does this $(if $(V),...) check handle make V=0
correctly?

In GNU Make, $(if $(V),...) checks if V is a non-empty string. If a user runs
make V=0, V evaluates to the string 0 which is non-empty, causing the check
to evaluate to true and completely suppress the BTFIDS output.

The standard msg macro uses ifeq ($(V),1) to avoid this.

Additionally, does this custom printf formatting align with the standard msg
macro output?

The format string produces a slightly different log format and drops the full
target path compared to the standard msg macro. This might lead to visual
inconsistency with the preceding TEST-OBJ output line.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260429-selftests-bpf_misconfig-v9-0-c311f06b4791@suse.com?part=6

Re: [PATCH bpf-next v9 09/11] selftests/bpf: Tolerate benchmark build failures

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [Medium] Skipped `bench` binary causes `make install` to fail due to missing files in `TEST_GEN_PROGS_EXTENDED`.
- [Medium] Using `$(wildcard $^)` to filter missing prerequisites breaks `make -n` (dry run) and drastically increases compiler error noise.
--

commit 31a306fe854d559a436ecf84354b5b73a31012e8
Author: Ricardo B. Marlière <rbm@suse.com>

selftests/bpf: Tolerate benchmark build failures

This commit modifies the selftests/bpf Makefile to tolerate build failures
for benchmark objects when PERMISSIVE is set. It changes the compilation and
linking steps for the bench binary to emit a skip message and avoid failing
the entire build if a dependency is missing or fails to compile.

> diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
> index a64e822dc540..db4ff951fc88 100644
> --- a/tools/testing/selftests/bpf/Makefile
> +++ b/tools/testing/selftests/bpf/Makefile

[ ... ]

> @@ -988,7 +989,8 @@ $(OUTPUT)/bench: $(OUTPUT)/bench.o \
>  		 $(OUTPUT)/usdt_2.o \
>  		 #
>  	$(call msg,BINARY,,$@)
> -	$(Q)$(CC) $(CFLAGS) $(LDFLAGS) $(filter %.a %.o,$^) $(LDLIBS) -o $@
> +	$(Q)$(CC) $(CFLAGS) $(LDFLAGS) $(filter %.a %.o,$(if $(PERMISSIVE),$(wildcard $^),$^)) $(LDLIBS) -o $@ $(if $(PERMISSIVE),|| \
> +		($(RM) $@; printf '  %-12s %s\n' 'SKIP-LINK' '$(notdir $@) (some benchmarks may have been skipped)' 1>&2))

Does using $(wildcard $^) here break make -n (dry run) and cause excessive
linker noise?

Since bench.c statically references every benchmark, omitting missing objects
from the command line forces the linker to run anyway, resulting in a large
number of undefined reference errors before it fails.

If $^ were used directly, the compiler would fail cleanly and immediately with
a "No such file or directory" error before running the linker, which would
then correctly trigger the || SKIP-LINK fallback.

Also, evaluating $(wildcard $^) in a recipe causes missing generated objects
to be silently omitted during make -n.

Will skipping the bench binary cleanly cause make install to fail, since
bench is unconditionally listed in TEST_GEN_PROGS_EXTENDED?

During make install, the selftest build system uses rsync to copy all targets
in TEST_GEN_PROGS_EXTENDED. Since the skipped bench binary won't exist on disk,
rsync fails with a "No such file or directory" error and aborts the pipeline,
defeating the purpose of the permissive build mode.

I do see this was fixed later in the series by commit
ce682ab9bce04cbe3adda3d7a63647774ad61f5d "selftests/bpf: Tolerate missing
files during install".

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260429-selftests-bpf_misconfig-v9-0-c311f06b4791@suse.com?part=9

Re: [PATCH bpf-next v9 10/11] selftests/bpf: Provide weak definitions for cross-test functions

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [Medium] Test flakiness and deterministic failure due to Identical Code Folding (ICF) of weak stubs.
- [Low] Missing `<alloca.h>` include for `alloca()` in `testing_helpers.c`.
--

commit 45cfa2851c72a65543e44de19ed32088ef920e33
Author: Ricardo B. Marlière <rbm@suse.com>

selftests/bpf: Provide weak definitions for cross-test functions

This commit replaces forward declarations of uprobe_multi_func_{1,2,3}() with
weak no-op stubs to allow the linker to resolve them when skeletons are
missing. It also moves stack_mprotect() to testing_helpers.c to make it
universally available and resolve an issue where the previous weak stub
returned 0.

> diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_cookie.c b/tools/testing/selftests/bpf/prog_tests/bpf_cookie.c
> index 35adc3f6d443..5a864cd8ad1b 100644
> --- a/tools/testing/selftests/bpf/prog_tests/bpf_cookie.c
> +++ b/tools/testing/selftests/bpf/prog_tests/bpf_cookie.c

[ ... ]

> -/* defined in prog_tests/uprobe_multi_test.c */
> -void uprobe_multi_func_1(void);
> -void uprobe_multi_func_2(void);
> -void uprobe_multi_func_3(void);
> +/*
> + * Weak uprobe target stubs. noinline is required because
> + * uprobe_multi_test_run() takes their addresses to configure the BPF
> + * program's attachment points; an inlined function has no stable
> + * address in the binary to probe. The strong definitions in
> + * uprobe_multi_test.c take precedence when that translation unit is
> + * linked.
> + */
> +noinline __weak void uprobe_multi_func_1(void) {}
> +noinline __weak void uprobe_multi_func_2(void) {}
> +noinline __weak void uprobe_multi_func_3(void) {}

Because these three weak stubs are structurally identical empty functions,
can compiler and linker optimization passes like Identical Code Folding
(ICF) merge them into a single memory address?

If they are folded into a single address, bpf_program__attach_uprobe_multi()
will attach three separate uprobes to the exact same offset, each with a
different cookie.

Calling any of the functions would then execute the shared address,
triggering all three uprobes simultaneously. This would increment the BPF
program counters multiple times per call, causing assertions like:

    ASSERT_EQ(skel->bss->uprobe_multi_func_1_result, 1, ...);

to fail deterministically, as the result would be 3 instead of 1.

Would it make sense to make these stubs structurally unique, perhaps by
adding unique inline assembly to each?

[ ... ]

> diff --git a/tools/testing/selftests/bpf/testing_helpers.c b/tools/testing/selftests/bpf/testing_helpers.c
> index 6fbe1e995660..3f7ae76c209b 100644
> --- a/tools/testing/selftests/bpf/testing_helpers.c
> +++ b/tools/testing/selftests/bpf/testing_helpers.c

[ ... ]

> +int stack_mprotect(void)
> +{
> +	void *buf;
> +	long sz;
> +	int ret;
> +
> +	sz = sysconf(_SC_PAGESIZE);
> +	if (sz < 0)
> +		return sz;
> +
> +	buf = alloca(sz * 3);

Does this code rely on a glibc-specific extension to provide alloca() via
<stdlib.h>?

The original implementation in test_lsm.c included <malloc.h>, but the new
location in testing_helpers.c includes neither <malloc.h> nor <alloca.h>.

On systems with strict POSIX-compliant C libraries, such as musl, omitting
<alloca.h> can result in an implicit function declaration error and lead
to build failures.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260429-selftests-bpf_misconfig-v9-0-c311f06b4791@suse.com?part=10