linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Yonghong Song <yhs@fb.com>
To: Ivan Khoronzhuk <ivan.khoronzhuk@linaro.org>,
	"ast@kernel.org" <ast@kernel.org>,
	"daniel@iogearbox.net" <daniel@iogearbox.net>,
	"davem@davemloft.net" <davem@davemloft.net>,
	"jakub.kicinski@netronome.com" <jakub.kicinski@netronome.com>,
	"hawk@kernel.org" <hawk@kernel.org>,
	"john.fastabend@gmail.com" <john.fastabend@gmail.com>
Cc: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
	"bpf@vger.kernel.org" <bpf@vger.kernel.org>,
	"clang-built-linux@googlegroups.com" 
	<clang-built-linux@googlegroups.com>
Subject: Re: [PATCH bpf-next 07/11] samples: bpf: add makefile.prog for separate CC build
Date: Fri, 13 Sep 2019 21:33:58 +0000	[thread overview]
Message-ID: <1720c5a5-5c64-46a3-be2f-56b59614f82a@fb.com> (raw)
In-Reply-To: <20190910103830.20794-8-ivan.khoronzhuk@linaro.org>



On 9/10/19 11:38 AM, Ivan Khoronzhuk wrote:
> The makefile.prog is added only, will be used in sample/bpf/Makefile
> later in order to switch cross-compiling on CC from HOSTCC.
> 
> The HOSTCC is supposed to build binaries and tools running on the host
> afterwards, in order to simplify build or so, like "fixdep" or else.
> In case of cross compiling "fixdep" is executed on host when the rest
> samples should run on target arch. In order to build binaries for
> target arch with CC and tools running on host with HOSTCC, lets add
> Makefile.prog for simplicity, having definition and routines similar
> to ones, used in script/Makefile.host. This allows later add
> cross-compilation to samples/bpf with minimum changes.

So this is really Makefile.host or Makefile.user, right?
In BPF, 'prog' can refers to user prog or bpf prog.
To avoid ambiguity, maybe Makefile.host?

> 
> Makefile.prog contains only stuff needed for samples/bpf, potentially
> can be reused and extended for other prog sets later and now needed

What do you mean 'extended for other prog sets'? I am wondering whether
we could just include 'scripts/Makefile.host'? How hard it is?

> only for unblocking tricky samples/bpf cross compilation.
> 
> Signed-off-by: Ivan Khoronzhuk <ivan.khoronzhuk@linaro.org>
> ---
>   samples/bpf/Makefile.prog | 77 +++++++++++++++++++++++++++++++++++++++
>   1 file changed, 77 insertions(+)
>   create mode 100644 samples/bpf/Makefile.prog
> 
> diff --git a/samples/bpf/Makefile.prog b/samples/bpf/Makefile.prog
> new file mode 100644
> index 000000000000..3781999b9193
> --- /dev/null
> +++ b/samples/bpf/Makefile.prog
> @@ -0,0 +1,77 @@
> +# SPDX-License-Identifier: GPL-2.0
> +# ==========================================================================
> +# Building binaries on the host system
> +# Binaries are not used during the compilation of the kernel, and intendent
> +# to be build for target board, target board can be host ofc. Added to build
> +# binaries to run not on host system.
> +#
> +# Only C is supported, but can be extended for C++.

The above comment is not needed.
samples/bpf/ only have C now. I am wondering whether your below scripts 
can be simplified, e.g., removing cxxobjs.

> +#
> +# Sample syntax (see Documentation/kbuild/makefiles.rst for reference)
> +# progs-y := xsk_example
> +# Will compile xdpsock_example.c and create an executable named xsk_example
> +#
> +# progs-y    := xdpsock
> +# xdpsock-objs := xdpsock_1.o xdpsock_2.o
> +# Will compile xdpsock_1.c and xdpsock_2.c, and then link the executable
> +# xdpsock, based on xdpsock_1.o and xdpsock_2.o
> +#
> +# Inherited from scripts/Makefile.host
> +#
> +__progs := $(sort $(progs-y))
> +
> +# C code
> +# Executables compiled from a single .c file
> +prog-csingle	:= $(foreach m,$(__progs), \
> +			$(if $($(m)-objs)$($(m)-cxxobjs),,$(m)))
> +
> +# C executables linked based on several .o files
> +prog-cmulti	:= $(foreach m,$(__progs),\
> +		   $(if $($(m)-cxxobjs),,$(if $($(m)-objs),$(m))))
> +
> +# Object (.o) files compiled from .c files
> +prog-cobjs	:= $(sort $(foreach m,$(__progs),$($(m)-objs)))
> +
> +prog-csingle	:= $(addprefix $(obj)/,$(prog-csingle))
> +prog-cmulti	:= $(addprefix $(obj)/,$(prog-cmulti))
> +prog-cobjs	:= $(addprefix $(obj)/,$(prog-cobjs))
> +
> +#####
> +# Handle options to gcc. Support building with separate output directory
> +
> +_progc_flags   = $(PROGS_CFLAGS) \
> +                 $(PROGCFLAGS_$(basetarget).o)
> +
> +# $(objtree)/$(obj) for including generated headers from checkin source files
> +ifeq ($(KBUILD_EXTMOD),)
> +ifdef building_out_of_srctree
> +_progc_flags   += -I $(objtree)/$(obj)
> +endif
> +endif
> +
> +progc_flags    = -Wp,-MD,$(depfile) $(_progc_flags)
> +
> +# Create executable from a single .c file
> +# prog-csingle -> Executable
> +quiet_cmd_prog-csingle 	= CC  $@
> +      cmd_prog-csingle	= $(CC) $(progc_flags) $(PROGS_LDFLAGS) -o $@ $< \
> +		$(PROGS_LDLIBS) $(PROGLDLIBS_$(@F))
> +$(prog-csingle): $(obj)/%: $(src)/%.c FORCE
> +	$(call if_changed_dep,prog-csingle)
> +
> +# Link an executable based on list of .o files, all plain c
> +# prog-cmulti -> executable
> +quiet_cmd_prog-cmulti	= LD  $@
> +      cmd_prog-cmulti	= $(CC) $(progc_flags) $(PROGS_LDFLAGS) -o $@ \
> +			  $(addprefix $(obj)/,$($(@F)-objs)) \
> +			  $(PROGS_LDLIBS) $(PROGLDLIBS_$(@F))
> +$(prog-cmulti): $(prog-cobjs) FORCE
> +	$(call if_changed,prog-cmulti)
> +$(call multi_depend, $(prog-cmulti), , -objs)
> +
> +# Create .o file from a single .c file
> +# prog-cobjs -> .o
> +quiet_cmd_prog-cobjs	= CC  $@
> +      cmd_prog-cobjs	= $(CC) $(progc_flags) -c -o $@ $<
> +$(prog-cobjs): $(obj)/%.o: $(src)/%.c FORCE
> +	$(call if_changed_dep,prog-cobjs)
> 

  reply	other threads:[~2019-09-13 21:34 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-09-10 10:38 [PATCH bpf-next 00/11] samples: bpf: improve/fix cross-compilation Ivan Khoronzhuk
2019-09-10 10:38 ` [PATCH bpf-next 01/11] samples: bpf: makefile: fix HDR_PROBE "echo" Ivan Khoronzhuk
2019-09-10 10:46   ` Sergei Shtylyov
2019-09-10 14:54     ` Ivan Khoronzhuk
2019-09-11 11:02       ` Sergei Shtylyov
2019-09-13 19:56         ` Ivan Khoronzhuk
2019-09-10 10:38 ` [PATCH bpf-next 02/11] samples: bpf: makefile: fix cookie_uid_helper_example obj build Ivan Khoronzhuk
2019-09-13 20:48   ` Yonghong Song
2019-09-13 21:25     ` Ivan Khoronzhuk
2019-09-10 10:38 ` [PATCH bpf-next 03/11] samples: bpf: makefile: use --target from cross-compile Ivan Khoronzhuk
2019-09-10 10:38 ` [PATCH bpf-next 04/11] samples: bpf: use own EXTRA_CFLAGS for clang commands Ivan Khoronzhuk
2019-09-10 10:38 ` [PATCH bpf-next 05/11] samples: bpf: makefile: use D vars from KBUILD_CFLAGS to handle headers Ivan Khoronzhuk
2019-09-13 21:12   ` Yonghong Song
2019-09-13 21:24     ` Ivan Khoronzhuk
2019-09-10 10:38 ` [PATCH bpf-next 06/11] samples: bpf: makefile: drop unnecessarily inclusion for bpf_load Ivan Khoronzhuk
2019-09-10 10:38 ` [PATCH bpf-next 07/11] samples: bpf: add makefile.prog for separate CC build Ivan Khoronzhuk
2019-09-13 21:33   ` Yonghong Song [this message]
2019-09-13 22:14     ` Ivan Khoronzhuk
2019-09-10 10:38 ` [PATCH bpf-next 08/11] samples: bpf: makefile: base progs build on makefile.progs Ivan Khoronzhuk
2019-09-13 21:41   ` Yonghong Song
2019-09-13 22:25     ` Ivan Khoronzhuk
2019-09-10 10:38 ` [PATCH bpf-next 09/11] samples: bpf: makefile: use CC environment for HDR_PROBE Ivan Khoronzhuk
2019-09-10 10:38 ` [PATCH bpf-next 10/11] libbpf: makefile: add C/CXX/LDFLAGS to libbpf.so and test_libpf targets Ivan Khoronzhuk
2019-09-13 21:43   ` Yonghong Song
2019-09-13 22:33     ` Ivan Khoronzhuk
2019-09-10 10:38 ` [PATCH bpf-next 11/11] samples: bpf: makefile: add sysroot support Ivan Khoronzhuk
2019-09-13 21:45   ` Yonghong Song
2019-09-13 22:36     ` Ivan Khoronzhuk

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1720c5a5-5c64-46a3-be2f-56b59614f82a@fb.com \
    --to=yhs@fb.com \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=clang-built-linux@googlegroups.com \
    --cc=daniel@iogearbox.net \
    --cc=davem@davemloft.net \
    --cc=hawk@kernel.org \
    --cc=ivan.khoronzhuk@linaro.org \
    --cc=jakub.kicinski@netronome.com \
    --cc=john.fastabend@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).