[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <2bb55c11-bffb-4099-f1fc-d31fb058d422@cn.fujitsu.com>
Date: Thu, 8 Mar 2018 18:25:27 +0800
From: Cao jin <caoj.fnst@...fujitsu.com>
To: Masahiro Yamada <yamada.masahiro@...ionext.com>
CC: Linux Kbuild mailing list <linux-kbuild@...r.kernel.org>,
Sam Ravnborg <sam@...nborg.org>,
Michal Marek <michal.lkml@...kovi.net>,
Govind Singh <govinds@...eaurora.org>,
Kalle Valo <kvalo@...eaurora.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 2/5] kbuild: define KBUILD_MODNAME even if multiple
modules share objects
Masahiro-san
On 03/08/2018 06:21 PM, Masahiro Yamada wrote:
> 2018-03-08 19:11 GMT+09:00 Cao jin <caoj.fnst@...fujitsu.com>:
>>
>>
>> On 03/08/2018 09:05 AM, Masahiro Yamada wrote:
>>> Currently, KBUILD_MODNAME is defined only when $(modname) contains
>>> just one word. If an object is shared among multiple modules,
>>> undefined KBUILD_MODNAME could cause a build error. For example,
>>> if CONFIG_DYNAMIC_DEBUG is enabled, any call of printk() populates
>>> .modname, then fails to build due to undefined KBUILD_MODNAME.
>>>
>>> Take the following code as an example:
>>>
>>> obj-m += foo.o
>>> obj-m += bar.o
>>> foo-objs := foo-bar-common.o foo-main.o
>>> bar-objs := foo-bar-common.o bar-main.o
>>>
>>> In this case, there is room for argument what to define for
>>> KBUILD_MODNAME when foo-bar-common.o is being compiled.
>>> "foo", "bar", or what else?
>>>
>>> One idea is to define colon-separated modules that share the object,
>>> in this case, "bar:foo" (modules are sorted alphabetically by
>>> $(sort ...).
>>>
>>> Signed-off-by: Masahiro Yamada <yamada.masahiro@...ionext.com>
>>> ---
>>>
>>> scripts/Makefile.lib | 9 +++++----
>>> 1 file changed, 5 insertions(+), 4 deletions(-)
>>>
>>> diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib
>>> index a7e315f..a1fbd6a 100644
>>> --- a/scripts/Makefile.lib
>>> +++ b/scripts/Makefile.lib
>>> @@ -92,8 +92,7 @@ subdir-ym := $(addprefix $(obj)/,$(subdir-ym))
>>> # differ in different configs.
>>> name-fix = $(squote)$(quote)$(subst $(comma),_,$(subst -,_,$1))$(quote)$(squote)
>>> basename_flags = -DKBUILD_BASENAME=$(call name-fix,$(basetarget))
>>> -modname_flags = $(if $(filter 1,$(words $(modname))),\
>>> - -DKBUILD_MODNAME=$(call name-fix,$(modname)))
>>> +modname_flags = -DKBUILD_MODNAME=$(call name-fix,$(modname))
>>>
>>
>> I guess there is comment also need to be modified above this code hunk:
>>
>> Note: Files that end up in two or more modules are compiled without the
>> KBUILD_MODNAME definition. The reason is that any made-up name
>> would differ in different configs.
>
>
> Why do I have to add lying comments here?
>
> The commit subject/log claims KBUILD_MODNAME should be _always_ defined.
>
I feel you misunderstand my intention, the comment I mentioned is
already there, so they should be modified according to your patch.
--
Sincerely,
Cao jin
Powered by blists - more mailing lists