Re: [PATCH net-next v6 4/5] selftests/net: Add a shared BPF skeleton rule

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] selftests/net: bpf.mk ignores LLVM when setting HOSTCC and HOSTLD for bpftool
--

commit 710ef769e16adaffb164124378ccc9076be7c9c1
Author: Bochao Cao <bochaolucky@gmail.com>

selftests/net: Add a shared BPF skeleton rule

This commit adds a shared rule to tools/testing/selftests/net/bpf.mk for
generating BPF skeletons, builds a host bootstrap bpftool when BPFTOOL is
not supplied, and adjusts header includes and submake variables.

> diff --git a/tools/testing/selftests/net/bpf.mk b/tools/testing/selftests/net/bpf.mk
> index a4f6755dd8941..016e32960280a 100644
> --- a/tools/testing/selftests/net/bpf.mk
> +++ b/tools/testing/selftests/net/bpf.mk

[ ... ]

> @@ -46,8 +46,37 @@ $(BPFOBJ): $(wildcard $(BPFDIR)/*.[ch] $(BPFDIR)/Makefile)		       \
>  	   $(APIDIR)/linux/bpf.h					       \
>  	   | $(BUILD_DIR)/libbpf
>  	$(call msg,MAKE,,$@)
> -	$(Q)$(MAKE) $(submake_extras) -C $(BPFDIR) OUTPUT=$(BUILD_DIR)/libbpf/ \
> +	$(Q)$(MAKE) $(submake_extras) -C $(BPFDIR) srctree=$(abspath $(top_srcdir)) \
> +		    OUTPUT=$(BUILD_DIR)/libbpf/ \
>  		    EXTRA_CFLAGS='-g -O0'				       \
>  		    DESTDIR=$(SCRATCH_DIR) prefix= all install_headers
>  
>  EXTRA_CLEAN += $(SCRATCH_DIR)
> +
> +# Build a host bpftool only when a test needs a generated skeleton.
> +HOSTCC ?= gcc
> +HOSTLD ?= ld
> +BPFTOOL_BUILD := $(BUILD_DIR)/bpftool
> +ifeq ($(origin BPFTOOL),undefined)
> +BPFTOOL := $(BPFTOOL_BUILD)/bootstrap/bpftool
> +BPFTOOL_DEP := $(BPFTOOL)
> +else
> +BPFTOOL_DEP := $(shell command -v $(BPFTOOL))
> +endif
> +
> +$(BPFTOOL_BUILD)/bootstrap/bpftool: \
> +	$(wildcard $(top_srcdir)/tools/bpf/bpftool/*.[ch]) \
> +	$(top_srcdir)/tools/bpf/bpftool/Makefile \
> +	$(wildcard $(BPFDIR)/*.[ch] $(BPFDIR)/Makefile)
> +	$(call msg,MAKE,,$@)
> +	$(Q)mkdir -p $(BPFTOOL_BUILD)
> +	$(Q)$(MAKE) $(submake_extras) -C $(top_srcdir)/tools/bpf/bpftool \
> +		OUTPUT=$(BPFTOOL_BUILD)/ srctree=$(abspath $(top_srcdir)) \
> +		ARCH= CROSS_COMPILE= CC="$(HOSTCC)" LD="$(HOSTLD)" bootstrap

[Severity: Medium]
When building with LLVM=1, will defaulting HOSTCC to gcc and HOSTLD to ld
and passing them as CC and LD to the bpftool submake cause a toolchain
mismatch or fail on LLVM-only hosts?

Makefiles that include tools/testing/selftests/net/bpf.mk include
tools/testing/selftests/lib.mk first, which handles LLVM= by setting
LLVM_PREFIX, LLVM_SUFFIX, CLANG, and CC, but does not define HOSTCC or
HOSTLD. Because tools/testing/selftests/net/bpf.mk defaults HOSTCC ?= gcc
and HOSTLD ?= ld without checking LLVM, the $(BPFTOOL_BUILD)/bootstrap/bpftool
recipe passes CC="gcc" and LD="ld" on the submake command line without
passing HOSTCC or HOSTLD.

Inside tools/bpf/bpftool/Makefile, tools/scripts/Makefile.include processes
LLVM= from MAKEFLAGS:

  $(call allow-override,CC,$(LLVM_PREFIX)clang$(LLVM_SUFFIX))
  $(call allow-override,CLANG,$(LLVM_PREFIX)clang$(LLVM_SUFFIX))
  $(call allow-override,HOSTCC,$(LLVM_PREFIX)clang$(LLVM_SUFFIX))
  $(call allow-override,AR,$(LLVM_PREFIX)llvm-ar$(LLVM_SUFFIX))
  $(call allow-override,HOSTAR,$(LLVM_PREFIX)llvm-ar$(LLVM_SUFFIX))
  $(call allow-override,LD,$(LLVM_PREFIX)ld.lld$(LLVM_SUFFIX))
  $(call allow-override,HOSTLD,$(LLVM_PREFIX)ld.lld$(LLVM_SUFFIX))

Because CC and LD have command-line origin, allow-override leaves them as
gcc and ld, while HOSTCC and HOSTLD resolve to clang and ld.lld. As a
result, tools/bpf/bpftool/Makefile compiles bootstrap/bpftool with
HOSTCC=clang, while CC_NO_CLANG in tools/scripts/Makefile.include and
feature-libelf-zstd in tools/build/Makefile.feature probe CC=gcc (enabling
GCC-only warning flags in EXTRA_WARNINGS and HOST_CFLAGS, or failing when
gcc is not installed).

Should HOSTCC and HOSTLD account for LLVM (or be passed through to the
submake as HOSTCC and HOSTLD after honoring LLVM)?

[ ... ]

>
>-- 
>2.51.0

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261018-xdp-v5-send-20261008-v6-0-7336e1282ca3@gmail.com?part=4