From: Ulf Samuelsson <ulf@atmel.com>
To: buildroot@busybox.net
Subject: [Buildroot] PATCH: set BR2_INET_RPC dependency for nfs-utils and portmap
Date: Mon, 15 Oct 2007 23:39:43 +0200 [thread overview]
Message-ID: <02e501c80f74$3fe9a980$01c4af0a@Glamdring> (raw)
In-Reply-To: 47120D19.8070807@promwad.com
----- Original Message -----
From: "Ivan Kuten" <ivan.kuten@promwad.com>
To: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Cc: <buildroot@uclibc.org>
Sent: Sunday, October 14, 2007 2:35 PM
Subject: Re: [Buildroot] PATCH: set BR2_INET_RPC dependency for nfs-utils
and portmap
> Yann E. MORIN wrote:
>> On Sunday 14 October 2007 01:50:32 Ivan Kuten wrote:
>>> Set BR2_INET_RPC dependency for nfs-utils and portmap:
>>
>> We had the same issue in BusyBox, and solved it the other way around:
>> have
>> tools needing RPC do a 'select' on RPC, such as (this is not a patch!):
>>
>> ---8<---
>> Index: package/nfs-utils/Config.in
>> ===================================================================
>> --- package/nfs-utils/Config.in (revision 20237)
>> +++ package/nfs-utils/Config.in (working copy)
>> @@ -1,6 +1,7 @@
>> config BR2_PACKAGE_NFS_UTILS
>> bool "nfs-utils"
>> default n
>> + select BR2_INET_RPC
>> help
>> The NFS Linux kernel server.
>> Warning: We do not force largefile support on here on purpose.
>> Index: package/portmap/Config.in
>> ===================================================================
>> --- package/portmap/Config.in (revision 20237)
>> +++ package/portmap/Config.in (working copy)
>> @@ -1,6 +1,7 @@
>> config BR2_PACKAGE_PORTMAP
>> bool "portmap"
>> default n
>> + select BR2_INET_RPC
>> help
>> The standard portmapper for RPC services.
>>
>> ---8<---
>>
>> This is, imho, better this way because what people really want is NFS or
>> portmap, and might not know that RPC is needed to make those two
>> available.
>>
>> Having NFS and portmap visible and 'select' the needed RPC support is
>> more
>> straightforward. But again, this is my humble opinion.
>>
>
> Hello Yann,
>
> I completely agree with you, your approach is better.
>
> BR,
> Ivan
>
Isn't RPC a toolchain option?
What happens if you
1) build your toolchain with a roofs consisting of busybox alone.
2) enable the NFS and/or other things requiring options which is not
supported by the previously built toolchain.
You will then get an error...
I am not so sure that it is a good idea to have the toolchain and the rootfs
in the same ".config" file.
If they were separate, then you could every time you did a make:
1) Create a file containing the parameters required by the packages.
2) Compare that file with the file containing the parameter values used to
build the toolchain.
If the first file has requirements, not met by the toolchain, then
the new parameters should be copied to the toolchain/.config
and the toolchain should be rebuilt.
Kconfig is really not ideal.
I would have loved to have a config system which "grayed out" packages
which cannot be built using the current toolchain.
They should still be selectable, but it should be obvious to the user,
that the toolchain will be rebuilt, it this package is chosen.
Best Regards
Ulf Samuelsson
prev parent reply other threads:[~2007-10-15 21:39 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-10-13 23:50 [Buildroot] PATCH: set BR2_INET_RPC dependency for nfs-utils and portmap Ivan Kuten
2007-10-14 9:30 ` Yann E. MORIN
2007-10-14 12:35 ` Ivan Kuten
2007-10-15 21:39 ` Ulf Samuelsson [this message]
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='02e501c80f74$3fe9a980$01c4af0a@Glamdring' \
--to=ulf@atmel.com \
--cc=buildroot@busybox.net \
/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