Buildroot Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Peter Korsgaard <peter@korsgaard.com>
To: buildroot@busybox.net
Subject: [Buildroot] [PATCH 1/3] efitools: new package
Date: Tue, 17 Jul 2018 22:32:18 +0200	[thread overview]
Message-ID: <87efg1k3hp.fsf@dell.be.48ers.dk> (raw)
In-Reply-To: <1530887774-19479-2-git-send-email-celso.neto.cwi@datacom.com.br> (Celso Varella's message of "Fri, 6 Jul 2018 11:36:12 -0300")

>>>>> "Celso" == Celso Varella <celso.neto.cwi@datacom.com.br> writes:

 > Linux user-space application to manipulate UEFI signatures database
 > Patches 1 and 2 remove dependencies from sbsigntools and perl scripts

These patches are not very nice, as they don't look upstreamable and are
likely to cause conflicts every time this package is bumped. Can you
explain why exactly they are needed? Your series already add a
sbsigntools package, so why would we need to drop the dependency?
host-perl is similary available.

Looking at the code, the sbsigntools dependency seems to be for the
*-signed.efi files that gets signed by a just created key. That indeed
might not be useful for real use cases, but presumably we can just pass
EFISIGNED= to make to drop that?


 > Patch 3 remove "-l" option from mount command in lib/kernel_efivars.c
 > for compatibility with Busybox mount command

Did you try to submit this upstream? What is the -l option used for in
this context? E.G. what are we missing from not using it for builds with
util-linux mount?

 > +################################################################################
 > +#
 > +# efitools
 > +#
 > +################################################################################
 > +
 > +EFITOOLS_VERSION = 1.8.1
 > +EFITOOLS_SITE = https://git.kernel.org/pub/scm/linux/kernel/git/jejb/efitools.git/snapshot
 > +EFITOOLS_LICENSE = GPL-2.0+
 > +EFITOOLS_LICENSE_FILES = COPYING
 > +EFITOOLS_DEPENDENCIES = gnu-efi openssl
 > +
 > +define EFITOOLS_BUILD_CMDS
 > +	$(TARGET_MAKE_ENV) $(MAKE) -C $(@D)

The Makefile does different things depending on ARCH (which comes from
uname -m). This naturally doesn't work in a cross compilation setup, so
we need to pass ARCH=<value> depending on the target architecture (and
add the needed dependencies to the package so it can only be built for
those supported architectures).

-- 
Bye, Peter Korsgaard

  reply	other threads:[~2018-07-17 20:32 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-07-06 14:36 [Buildroot] [PATCH 0/3] New packages to support UEFI Secure Boot Celso Varella
2018-07-06 14:36 ` [Buildroot] [PATCH 1/3] efitools: new package Celso Varella
2018-07-17 20:32   ` Peter Korsgaard [this message]
2018-07-06 14:36 ` [Buildroot] [PATCH 2/3] gnu-efi: add host variant Celso Varella
2018-07-17 21:29   ` Peter Korsgaard
2018-07-06 14:36 ` [Buildroot] [PATCH 3/3] sbsigntools: new package Celso Varella
2018-07-17 21:25   ` Peter Korsgaard

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=87efg1k3hp.fsf@dell.be.48ers.dk \
    --to=peter@korsgaard.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