The build host HOME directory check recently added to OE-Core's buildpaths QA test flags the ptest gtest binary:
WARNING: bpftrace-0.25.1-r0 do_package_qa: QA Issue: File /usr/lib/bpftrace/ptest/tests/bpftrace_test in package bpftrace-ptest contains a reference to the build host HOME directory. [buildpaths] WARNING: bpftrace-0.25.1-r0 do_package_qa: QA Issue: File /usr/lib/bpftrace/ptest/tests/.debug/bpftrace_test in package bpftrace-dbg contains a reference to the build host HOME directory. [buildpaths] cmake/BuildBPF.cmake links the intermediate DWARF carrier binary data_source_exe with the plain build host gcc, which bakes the host dynamic loader's absolute path into the PT_INTERP segment. With a distro gcc that is an innocuous /lib64/ld-linux-x86-64.so.2, but when the host toolchain is a buildtools-extended/SDK gcc installed under the build user's home directory (as on the autobuilder workers), it is a $HOME-prefixed path like /srv/pokybuild/.../buildtools/sysroots/x86_64-pokysdk-linux/lib/ld-linux-x86-64.so.2 The executable is then embedded byte-for-byte into bpftrace_test via embed()/xxd (tests/data/CMakeLists.txt), so the path ends up in the test binary's .rodata, triggering the bpftrace-ptest warning. And because cmake/Embed.tmpl declares the embedded blob as a constexpr array, gcc additionally copies its contents into .debug_info as DW_AT_const_value, which is how the same string survives objcopy --only-keep-debug into the split debug file, triggering the bpftrace-dbg warning. This is the same embedding mechanism that previously leaked TMPDIR paths, fixed for the compile step by the DEBUG_PREFIX_MAP patch; the link step was still leaking. The tests only write this binary to a temp file and use it as a uprobe target (tests/dwarf_common.h), parsing its DWARF and symbols; it is never executed. So link it with -nostdlib -no-pie -Wl,--entry=0: no interpreter, no host crt objects, no dynamic segment, and thus nothing host-specific in the embedded blob. Function symbols, DWARF type information and the pahole -J BTF encoding step are unaffected, and the field_analyser_dwarf ptests still pass. Because the object is now linked -nostdlib it must not reference the libc stack-protector runtime. Host gcc toolchains that default to -fstack-protector-strong (e.g. Debian/Ubuntu and the Yocto autobuilder workers) otherwise leave an undefined reference and the link fails: ld: data_source.o: in function `main': data_source.c:212: undefined reference to `__stack_chk_fail' collect2: error: ld returned 1 exit status Compile the object and link the binary with -fno-stack-protector; the canary is irrelevant to a DWARF/BTF carrier that is never run. Signed-off-by: Khem Raj <[email protected]> --- ...make-link-data-source-binary-without.patch | 78 +++++++++++++++++++ .../bpftrace/bpftrace_0.25.1.bb | 1 + 2 files changed, 79 insertions(+) create mode 100644 meta-oe/dynamic-layers/meta-python/recipes-devtools/bpftrace/bpftrace/0003-cmake-BuildBPF.cmake-link-data-source-binary-without.patch diff --git a/meta-oe/dynamic-layers/meta-python/recipes-devtools/bpftrace/bpftrace/0003-cmake-BuildBPF.cmake-link-data-source-binary-without.patch b/meta-oe/dynamic-layers/meta-python/recipes-devtools/bpftrace/bpftrace/0003-cmake-BuildBPF.cmake-link-data-source-binary-without.patch new file mode 100644 index 0000000000..7a69d0f8c1 --- /dev/null +++ b/meta-oe/dynamic-layers/meta-python/recipes-devtools/bpftrace/bpftrace/0003-cmake-BuildBPF.cmake-link-data-source-binary-without.patch @@ -0,0 +1,78 @@ +From: Khem Raj <[email protected]> +Date: Thu, 2 Jul 2026 15:30:00 -0700 +Subject: [PATCH] cmake/BuildBPF.cmake: link data source binary without host + runtime + +The `bpf()` helper links the intermediate DWARF carrier binary +(data_source_exe) with the plain host gcc: + + ${GCC} -g -o data_source_exe data_source.o + +and the resulting executable is then embedded byte-for-byte into +bpftrace_test via embed()/xxd (tests/data/CMakeLists.txt). Linking it as +a regular dynamic executable bakes host-toolchain artifacts into it: + + * PT_INTERP: the absolute path of the host toolchain's dynamic + loader. With a vendored/SDK toolchain (e.g. Yocto + buildtools-extended installed under the build user's home) this is + an absolute build-host path such as + /srv/pokybuild/.../buildtools/sysroots/x86_64-pokysdk-linux/lib/ld-linux-x86-64.so.2 + * crt startup objects (crt1.o, crti.o, crtbegin.o) including whatever + debug info the host libc was built with. + +Since the embedded blob carries these strings into the packaged test +binary, the build is not reproducible across hosts and leaks build host +paths, flagged by OE's buildpaths QA check: + + WARNING: bpftrace-0.25.1-r0 do_package_qa: QA Issue: File + /usr/lib/bpftrace/ptest/tests/bpftrace_test in package bpftrace-ptest + contains a reference to the build host HOME directory. [buildpaths] + +The binary only serves as a DWARF/BTF container: the tests write it to +a temp file and use it as a uprobe target (tests/dwarf_common.h), which +parses debug info and symbols but never executes it. So link it with +-nostdlib -no-pie: no interpreter, no dynamic segment, no host crt +code. -Wl,--entry=0 silences the "cannot find entry symbol _start" +warning. All function symbols, DWARF type information and the +pahole -J BTF encoding step are unaffected. + +Because the object is now linked -nostdlib, it must not emit references +to the libc stack-protector runtime. Host gcc toolchains that default to +-fstack-protector-strong (e.g. Debian/Ubuntu, the Yocto autobuilders) +otherwise leave an undefined reference and the link fails: + + ld: data_source.o: in function `main': + data_source.c:212: undefined reference to `__stack_chk_fail' + collect2: error: ld returned 1 exit status + +Compile the object (and link the binary) with -fno-stack-protector; the +canary is irrelevant to a DWARF/BTF carrier that is never run. + +Upstream-Status: Pending +Signed-off-by: Khem Raj <[email protected]> +--- + cmake/BuildBPF.cmake | 4 ++-- + 1 file changed, 2 insertions(+), 2 deletions(-) + +diff --git a/cmake/BuildBPF.cmake b/cmake/BuildBPF.cmake +index cb8cd68..8aa059f 100644 +--- a/cmake/BuildBPF.cmake ++++ b/cmake/BuildBPF.cmake +@@ -90,7 +90,7 @@ function(bpf NAME) + OUTPUT ${ARG_OBJECT} + DEPENDS ${ARG_SOURCE} ${ARG_DEPENDS} + # See above: fresh compilation and the use of `gcc`. +- COMMAND ${GCC} -Wno-attributes -g -I ${CMAKE_CURRENT_BINARY_DIR} -c ${CMAKE_CURRENT_SOURCE_DIR}/${ARG_SOURCE} -o ${CMAKE_CURRENT_BINARY_DIR}/${ARG_OBJECT} ++ COMMAND ${GCC} -Wno-attributes -fno-stack-protector -g -I ${CMAKE_CURRENT_BINARY_DIR} -c ${CMAKE_CURRENT_SOURCE_DIR}/${ARG_SOURCE} -o ${CMAKE_CURRENT_BINARY_DIR}/${ARG_OBJECT} + COMMAND cmake -E env LLVM_OBJCOPY=${LLVM_OBJCOPY} ${PAHOLE} -J ${CMAKE_CURRENT_BINARY_DIR}/${ARG_OBJECT} + VERBATIM + ) +@@ -107,7 +107,7 @@ function(bpf NAME) + add_custom_command( + OUTPUT ${ARG_BINARY} + DEPENDS ${ARG_SOURCE} ${NAME}_gen_object +- COMMAND ${GCC} -g -o ${CMAKE_CURRENT_BINARY_DIR}/${ARG_BINARY} ${CMAKE_CURRENT_BINARY_DIR}/${ARG_OBJECT} ++ COMMAND ${GCC} -g -fno-stack-protector -nostdlib -no-pie -Wl,--entry=0 -o ${CMAKE_CURRENT_BINARY_DIR}/${ARG_BINARY} ${CMAKE_CURRENT_BINARY_DIR}/${ARG_OBJECT} + COMMAND cmake -E env LLVM_OBJCOPY=${LLVM_OBJCOPY} ${PAHOLE} -J ${CMAKE_CURRENT_BINARY_DIR}/${ARG_BINARY} + VERBATIM + ) diff --git a/meta-oe/dynamic-layers/meta-python/recipes-devtools/bpftrace/bpftrace_0.25.1.bb b/meta-oe/dynamic-layers/meta-python/recipes-devtools/bpftrace/bpftrace_0.25.1.bb index 62e0c17f9d..7daf02ba1c 100644 --- a/meta-oe/dynamic-layers/meta-python/recipes-devtools/bpftrace/bpftrace_0.25.1.bb +++ b/meta-oe/dynamic-layers/meta-python/recipes-devtools/bpftrace/bpftrace_0.25.1.bb @@ -19,6 +19,7 @@ SRC_URI = "git://github.com/iovisor/bpftrace;branch=release/0.25.x;protocol=http file://run-ptest \ file://0002-CMakeLists.txt-allow-to-set-BISON_FLAGS-like-l.patch \ file://0001-cmake-BuildBPF.cmake-introduce-DEBUG_PREFIX_MAP.patch \ + file://0003-cmake-BuildBPF.cmake-link-data-source-binary-without.patch \ " SRCREV = "e491811e5d648288c01f42ce087967b271f504a0"
-=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#128145): https://lists.openembedded.org/g/openembedded-devel/message/128145 Mute This Topic: https://lists.openembedded.org/mt/120242463/21656 Group Owner: [email protected] Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [[email protected]] -=-=-=-=-=-=-=-=-=-=-=-
