Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <1ec14007-affc-f826-6dda-f23ee166226a@kernel.org>
Date:   Tue, 18 Oct 2022 10:16:20 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.3.1
Subject: Re: [PATCH v3 7/7] kbuild: remove head-y syntax
Content-Language: en-US
To:     Masahiro Yamada <masahiroy@kernel.org>,
        linux-kbuild@vger.kernel.org
Cc:     linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org,
        Nick Desaulniers <ndesaulniers@google.com>,
        Nathan Chancellor <nathan@kernel.org>,
        =?UTF-8?Q?Martin_Li=c5=a1ka?= <mliska@suse.cz>
References: <20220924181915.3251186-1-masahiroy@kernel.org>
 <20220924181915.3251186-8-masahiroy@kernel.org>
From:   Jiri Slaby <jirislaby@kernel.org>
In-Reply-To: <20220924181915.3251186-8-masahiroy@kernel.org>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

Hi,

On 24. 09. 22, 20:19, Masahiro Yamada wrote:
> Kbuild puts the objects listed in head-y at the head of vmlinux.
> Conventionally, we do this for head*.S, which contains the kernel entry
> point.
> 
> A counter approach is to control the section order by the linker script.
> Actually, the code marked as __HEAD goes into the ".head.text" section,
> which is placed before the normal ".text" section.
> 
> I do not know if both of them are needed. From the build system
> perspective, head-y is not mandatory. If you can achieve the proper code
> placement by the linker script only, it would be cleaner.
> 
> I collected the current head-y objects into head-object-list.txt. It is
> a whitelist. My hope is it will be reduced in the long run.
> 
> Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
...
> --- a/Makefile
> +++ b/Makefile
> @@ -1149,10 +1149,10 @@ quiet_cmd_ar_vmlinux.a = AR      $@
>         cmd_ar_vmlinux.a = \
>   	rm -f $@; \
>   	$(AR) cDPrST $@ $(KBUILD_VMLINUX_OBJS); \
> -	$(AR) mPiT $$($(AR) t $@ | head -n1) $@ $(head-y)
> +	$(AR) mPiT $$($(AR) t $@ | head -n1) $@ $$($(AR) t $@ | grep -F --file=$(srctree)/scripts/head-object-list.txt)

With AR=gcc-ar, the "| head -n1" results in:
/usr/lib64/gcc/x86_64-suse-linux/7/../../../../x86_64-suse-linux/bin/ar 
terminated with signal 13 [Broken pipe]

I found out only with gcc-lto. But maybe we should make it silent in any 
case? I'm not sure how. This looks ugly (and needs the whole output to 
be piped):
gcc-ar t vmlinux.a | ( head -n1; cat >/dev/null )

Note the result appears to be correct, it's only that gcc-ar complains 
after printing out the very first line.

thanks,
-- 
js
suse labs