Re: [PATCH v2 2/2] kbuild: use interpreters to invoke scripts

From: Masahiro Yamada
Date: Tue Oct 13 2020 - 12:03:17 EST


On Tue, Oct 13, 2020 at 4:03 AM Bernd Petrovitsch
<bernd@xxxxxxxxxxxxxxxxxxx> wrote:
>
> Hi all!
>
> On 12/10/2020 18:42, Ujjwal Kumar wrote:
> > On 12/10/20 11:50 pm, Lukas Bulwahn wrote:
> >>
> >>
> >> On Mon, 12 Oct 2020, Ujjwal Kumar wrote:
> >>
> >>> We cannot rely on execute bits to be set on files in the repository.
> >>> The build script should use the explicit interpreter when invoking any
> >>> script from the repository.
> >>>
> >>> Link: https://lore.kernel.org/lkml/20200830174409.c24c3f67addcce0cea9a9d4c@xxxxxxxxxxxxxxxxxxxx/
> >>> Link: https://lore.kernel.org/lkml/202008271102.FEB906C88@keescook/
> >>>
> >>> Suggested-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
> >>> Suggested-by: Kees Cook <keescook@xxxxxxxxxxxx>
> >>> Suggested-by: Lukas Bulwahn <lukas.bulwahn@xxxxxxxxx>
> >>> Signed-off-by: Ujjwal Kumar <ujjwalkumar0501@xxxxxxxxx>
> >>> ---
> >>> Makefile | 4 ++--
> >>> arch/arm64/kernel/vdso/Makefile | 2 +-
> >>> arch/arm64/kernel/vdso32/Makefile | 2 +-
> >>> arch/ia64/Makefile | 4 ++--
> >>> arch/nds32/kernel/vdso/Makefile | 2 +-
> >>> scripts/Makefile.build | 2 +-
> >>> scripts/Makefile.package | 4 ++--
> >>> 7 files changed, 10 insertions(+), 10 deletions(-)
> >>>
> >>> diff --git a/Makefile b/Makefile
> >>> index 0af7945caa61..df20e71dd7c8 100644
> >>> --- a/Makefile
> >>> +++ b/Makefile
> >>> @@ -1256,7 +1256,7 @@ include/generated/utsrelease.h: include/config/kernel.release FORCE
> >>> PHONY += headerdep
> >>> headerdep:
> >>> $(Q)find $(srctree)/include/ -name '*.h' | xargs --max-args 1 \
> >>> - $(srctree)/scripts/headerdep.pl -I$(srctree)/include
> >>> + $(PERL) $(srctree)/scripts/headerdep.pl -I$(srctree)/include
> >>>
> >>> # ---------------------------------------------------------------------------
> >>> # Kernel headers
> >>> @@ -1312,7 +1312,7 @@ PHONY += kselftest-merge
> >>> kselftest-merge:
> >>> $(if $(wildcard $(objtree)/.config),, $(error No .config exists, config your kernel first!))
> >>> $(Q)find $(srctree)/tools/testing/selftests -name config | \
> >>> - xargs $(srctree)/scripts/kconfig/merge_config.sh -m $(objtree)/.config
> >>> + xargs $(CONFIG_SHELL) $(srctree)/scripts/kconfig/merge_config.sh -m $(objtree)/.config
> >>> $(Q)$(MAKE) -f $(srctree)/Makefile olddefconfig
> >>>
> >>> # ---------------------------------------------------------------------------
> >>> diff --git a/arch/arm64/kernel/vdso/Makefile b/arch/arm64/kernel/vdso/Makefile
> >>> index edccdb77c53e..fb07804b7fc1 100644
> >>> --- a/arch/arm64/kernel/vdso/Makefile
> >>> +++ b/arch/arm64/kernel/vdso/Makefile
> >>> @@ -65,7 +65,7 @@ $(obj)/%.so: $(obj)/%.so.dbg FORCE
> >>> # Generate VDSO offsets using helper script
> >>> gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh
> >>> quiet_cmd_vdsosym = VDSOSYM $@
> >>> - cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@
> >>> + cmd_vdsosym = $(NM) $< | $(CONFIG_SHELL) $(gen-vdsosym) | LC_ALL=C sort > $@
> >>>
> >>> include/generated/vdso-offsets.h: $(obj)/vdso.so.dbg FORCE
> >>> $(call if_changed,vdsosym)
> >>> diff --git a/arch/arm64/kernel/vdso32/Makefile b/arch/arm64/kernel/vdso32/Makefile
> >>> index 7f96a1a9f68c..617c9ac58156 100644
> >>> --- a/arch/arm64/kernel/vdso32/Makefile
> >>> +++ b/arch/arm64/kernel/vdso32/Makefile
> >>> @@ -205,7 +205,7 @@ quiet_cmd_vdsomunge = MUNGE $@
> >>> gen-vdsosym := $(srctree)/$(src)/../vdso/gen_vdso_offsets.sh
> >>> quiet_cmd_vdsosym = VDSOSYM $@
> >>> # The AArch64 nm should be able to read an AArch32 binary
> >>> - cmd_vdsosym = $(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@
> >>> + cmd_vdsosym = $(NM) $< | $(CONFIG_SHELL) $(gen-vdsosym) | LC_ALL=C sort > $@
> >>>
> >>> # Install commands for the unstripped file
> >>> quiet_cmd_vdso_install = INSTALL32 $@
> >>> diff --git a/arch/ia64/Makefile b/arch/ia64/Makefile
> >>> index 703b1c4f6d12..86d42a2d09cb 100644
> >>> --- a/arch/ia64/Makefile
> >>> +++ b/arch/ia64/Makefile
> >>> @@ -27,8 +27,8 @@ cflags-y := -pipe $(EXTRA) -ffixed-r13 -mfixed-range=f12-f15,f32-f127 \
> >>> -falign-functions=32 -frename-registers -fno-optimize-sibling-calls
> >>> KBUILD_CFLAGS_KERNEL := -mconstant-gp
> >>>
> >>> -GAS_STATUS = $(shell $(srctree)/arch/ia64/scripts/check-gas "$(CC)" "$(OBJDUMP)")
> >>> -KBUILD_CPPFLAGS += $(shell $(srctree)/arch/ia64/scripts/toolchain-flags "$(CC)" "$(OBJDUMP)" "$(READELF)")
> >>> +GAS_STATUS = $(shell $(CONFIG_SHELL) $(srctree)/arch/ia64/scripts/check-gas "$(CC)" "$(OBJDUMP)")
> >>> +KBUILD_CPPFLAGS += $(shell $(CONFIG_SHELL) $(srctree)/arch/ia64/scripts/toolchain-flags "$(CC)" "$(OBJDUMP)" "$(READELF)")
> >>
> >> Here is an instance of what Masahiro-san pointed out being wrong.
> >>
> >> Ujjwal, will you send a v3?
> >
> > Following is the quoted text from the reply mail from Masahiro
> >
> >>> -GAS_STATUS = $(shell $(srctree)/arch/ia64/scripts/check-gas "$(CC)" "$(OBJDUMP)")
> >>> -KBUILD_CPPFLAGS += $(shell $(srctree)/arch/ia64/scripts/toolchain-flags "$(CC)" "$(OBJDUMP)" "$(READELF)")
> >>> +GAS_STATUS = $($(CONFIG_SHELL) $(srctree)/arch/ia64/scripts/check-gas "$(CC)" "$(OBJDUMP)")
> >>> +KBUILD_CPPFLAGS += $($(CONFIG_SHELL) $(srctree)/arch/ia64/scripts/toolchain-flags "$(CC)" "$(OBJDUMP)" "$(READELF)")
> >>
> >>
> >>
> >> These changes look wrong to me.
> >>
> >> $($(CONFIG_SHELL) -> $(shell $(CONFIG_SHELL)
> >>
> >
> > From the above text, I understand as follows:
>
> Did you actually *test* that (expecially) these lines work
> afterwards as good as before?
>
> > That my proposed change:
> > $(shell $(src...) -> $($(CONFIG_SHELL) $(src...)
> >
> > is WRONG
>
> Yup, as it's in a Makefile and that's a Makefile construct.
>
> > and in the next line he suggested the required correction.
> > That being:
> > $($(CONFIG_SHELL) -> $(shell $(CONFIG_SHELL)
>
> Such stuff should generally not be needed as the to-be-used
> shell can be set in Makefiles via a "SHELL = " assignment
> (defaulting to /bin/sh - what else;-).
> Flags for the shell can BTW set with ".SHELLFLAGS = ".


You are talking about a different thing.



Take the current code as an example:

$(shell $(srctree)/arch/ia64/scripts/check-gas "$(CC)" "$(OBJDUMP)")


Here are two shell invocations.


[1]
The command
$(srctree)/arch/ia64/scripts/check-gas "$(CC)" "$(OBJDUMP)"
is run in /bin/sh because the default value of SHELL is /bin/sh.


[2]
The script, arch/ia64/scripts/check-gas, is run in /bin/sh
because the hash-bang (the first line of check-gas)
specifies #!/bin/sh




Bernd is talking about [1].

In contrast, this patch is addressing [2] because
Andrew Morton suggested to run scripts without relying
on the executable bit.
(and, after this patch, we run scripts without relying
on the hash-bang because we now specify the interpreter.)


Of course, [1] and [2] can be different.


I always want to use /bin/sh for [1],
so please do not use bash-extension inside $(shell ...)


You have more choices for [2].

If arch/ia64/scripts/check-gas had been written with bash-extension,
the code would have been changed into:

$(shell $(BASH) $(srctree)/arch/ia64/scripts/check-gas "$(CC)" "$(OBJDUMP)")


I hope this will be clearer.




> So please
> -) learn basic "Makefile" + "make" before brainlessly patching
> a Makefile.
> -) actually testy your changes to make sure the patch didn't
> broke anything
> -) and - last but not least - check if there isn't a shell
> already set (and which).
>
> MfG,
> Bernd
> --
> There is no cloud, just other people computers.
> -- https://static.fsf.org/nosvn/stickers/thereisnocloud.svg



--
Best Regards
Masahiro Yamada