Issues with using with master of kernel

[b-ncMN]

An error is occurring when using kninja against master (I've tested up to v5.12) at the ninja -d explain -n

command "ninja -d explain -n" ran manually :

ninja: error: build.ninja:91096: unknown build rule 'cmd_arch_x86_entry_syscalls_built-in.a'
    cmd_arch_x86_entry_syscalls_built-in.a
    ^ near here

full stacktrace :

Traceback (most recent call last):
  File "/home/build-testing/src/linux/../kninja/kninja.py", line 368, in <module>
    main()
  File "/home/build-testing/src/linux/../kninja/kninja.py", line 359, in main
    out = subprocess.check_output(cmd, stderr=subprocess.STDOUT).decode('utf-8')
  File "/usr/lib/python3.9/subprocess.py", line 424, in check_output
    return run(*popenargs, stdout=PIPE, timeout=timeout, check=True,
  File "/usr/lib/python3.9/subprocess.py", line 528, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-d', 'explain', '-n']' returned non-zero exit status 1.

[martinetd]

They've apparently added some dummy rules for lib.a and built-in.a files with no dependencies (e.g. only foo/lib.a: FORCE for the rule)

I was able to work around this problem by skipping such rules in the conversion, as no file is actually created by make in these case:

diff --git a/kninja.py b/kninja.py
index 5d814620eeb8..e303e6f3b627 100755
--- a/kninja.py
+++ b/kninja.py
@@ -184,6 +184,8 @@ class Kninja(object):
                     continue
 
                 deps = [d for d in deps.split(' ') if d != 'FORCE' and d not in IGNORES]
+                if not deps:
+                    continue
                 handledfiles.update(deps)
                 fixed = self.fixname(obj)
 

unfortunately while that passes the initial ninja explain check, it fails real world rebuilds (at least for my cross build with BTF enabled) -- it didn't build the .o file as expected and pahole couldn't handle it (+ other scary warnings before that):

[4/5] cat Makefile | sed -e '/^$(vmlinux-dirs)/,+1d' | make -f - all
  GEN     Makefile
  CALL    ../scripts/checksyscalls.sh
  CALL    ../scripts/atomic/check-atomics.sh
  DESCEND  bpf/resolve_btfids
  CHK     include/generated/compile.h
  CHK     kernel/kheaders_data.tar.xz
  GEN     .version
  CHK     include/generated/compile.h
  UPD     include/generated/compile.h
  CC      init/version.o
  AR      init/built-in.a
  LD      vmlinux.o
  MODPOST vmlinux.symvers
WARNING: modpost: EXPORT symbol "mmc_pwrseq_register" [vmlinux] version generation failed, symbol will not be versioned.
WARNING: modpost: EXPORT symbol "mmc_pwrseq_unregister" [vmlinux] version generation failed, symbol will not be versioned.
  MODINFO modules.builtin.modinfo
  GEN     modules.builtin
  LD      .tmp_vmlinux.btf
aarch64-linux-gnu-ld: drivers/mmc/core/pwrseq.o: relocation R_AARCH64_ABS32 against `__crc_mmc_pwrseq_register' can not be used when making a shared object
drivers/mmc/core/pwrseq.o:(__patchable_function_entries+0x0): dangerous relocation: unsupported relocation
drivers/mmc/core/pwrseq.o:(__patchable_function_entries+0x8): dangerous relocation: unsupported relocation
drivers/mmc/core/pwrseq.o:(__patchable_function_entries+0x10): dangerous relocation: unsupported relocation
drivers/mmc/core/pwrseq.o:(__patchable_function_entries+0x18): dangerous relocation: unsupported relocation
drivers/mmc/core/pwrseq.o:(__patchable_function_entries+0x20): dangerous relocation: unsupported relocation
drivers/mmc/core/pwrseq.o:(__patchable_function_entries+0x28): dangerous relocation: unsupported relocation
drivers/mmc/core/pwrseq.o:(__patchable_function_entries+0x30): dangerous relocation: unsupported relocation
drivers/mmc/core/pwrseq.o:(__patchable_function_entries+0x38): dangerous relocation: unsupported relocation
  BTF     .btf.vmlinux.bin.o
pahole: .tmp_vmlinux.btf: Invalid argument
  LD      .tmp_vmlinux.kallsyms1
.btf.vmlinux.bin.o: file not recognized: file format not recognized
make[1]: *** [/home/xxx/linux/Makefile:1207: vmlinux] Error 1
make: *** [../Makefile:192: __sub-make] Error 2
FAILED: vmlinux 
cat Makefile | sed -e '/^$(vmlinux-dirs)/,+1d' | make -f - all
ninja: build stopped: subcommand failed.

from that point just running make again fails the build too, modifying the c file again and running make fixes it -- a cursory look with make V=1 shows it's apparently missing the genksyms step which modifies the .o file in place (sigh...):

  if aarch64-linux-gnu-objdump -h drivers/mmc/core/pwrseq.o | grep -q __ksymtab; then aarch64-linux-gnu-gcc -E -D__GENKSYMS__ -Wp,-MMD,drivers/mmc/core/.pwrseq.o.d -nostdinc -isystem /usr/lib/gcc-cross/aarch64-linux-gnu/12/include -I../arch/arm64/include -I./arch/arm64/include/generated -I../include -I./include -I../arch/arm64/include/uapi -I./arch/arm64/include/generated/uapi -I../include/uapi -I./include/generated/uapi -include ../include/linux/kconfig.h -include ../include/linux/compiler_types.h -D__KERNEL__ -mlittle-endian -DCC_USING_PATCHABLE_FUNCTION_ENTRY -DKASAN_SHADOW_SCALE_SHIFT=3 -fmacro-prefix-map=../= -Wall -Wundef -Werror=strict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common -fshort-wchar -fno-PIE -Werror=implicit-function-declaration -Werror=implicit-int -Werror=return-type -Wno-format-security -std=gnu89 -mgeneral-regs-only -DCONFIG_CC_HAS_K_CONSTRAINT=1 -Wno-psabi -mabi=lp64 -fno-asynchronous-unwind-tables -fno-unwind-tables -mbranch-protection=pac-ret+leaf -Wa,-march=armv8.4-a -DARM64_ASM_ARCH='"armv8.4-a"' -DKASAN_SHADOW_SCALE_SHIFT=3 -fno-delete-null-pointer-checks -Wno-frame-address -Wno-format-truncation -Wno-format-overflow -Wno-address-of-packed-member -O2 -fno-allow-store-data-races -Wframe-larger-than=2048 -fstack-protector-strong -Wimplicit-fallthrough -Wno-unused-but-set-variable -Wno-unused-const-variable -Wno-dangling-pointer -fno-omit-frame-pointer -fno-optimize-sibling-calls -ftrivial-auto-var-init=zero -g -fpatchable-function-entry=2 -Wdeclaration-after-statement -Wvla -Wno-pointer-sign -Wno-stringop-truncation -Wno-zero-length-bounds -Wno-array-bounds -Wno-stringop-overflow -Wno-restrict -Wno-maybe-uninitialized -fno-strict-overflow -fno-stack-check -fconserve-stack -Werror=date-time -Werror=incompatible-pointer-types -Werror=designated-init -Wno-packed-not-aligned -mstack-protector-guard=sysreg -mstack-protector-guard-reg=sp_el0 -mstack-protector-guard-offset=1168 -I ../drivers/mmc/core -I ./drivers/mmc/core    -DKBUILD_MODFILE='"drivers/mmc/core/mmc_core"' -DKBUILD_BASENAME='"pwrseq"' -DKBUILD_MODNAME='"mmc_core"' ../drivers/mmc/core/pwrseq.c | scripts/genksyms/genksyms    -r /dev/null > drivers/mmc/core/.tmp_pwrseq.ver; aarch64-linux-gnu-ld -EL  -maarch64elf -z noexecstack --no-warn-rwx-segments -r -o drivers/mmc/core/.tmp_pwrseq.o drivers/mmc/core/pwrseq.o -T drivers/mmc/core/.tmp_pwrseq.ver; mv -f drivers/mmc/core/.tmp_pwrseq.o drivers/mmc/core/pwrseq.o; rm -f drivers/mmc/core/.tmp_pwrseq.ver; fi

Anyway, that quick fix wasn't enough so I won't open a PR and will just stop here, but it's worth a shot depending on your exact config.

@vitkyrka great idea to just parse make -p output, if someone has a bit of time I think it'd definitely be useful to brush this up again! Thank you!