Revert "Makefile: exclude BR2_DL_DIR from savedefconfig"
authorYann E. MORIN <yann.morin.1998@free.fr>
Fri, 6 Nov 2020 18:26:41 +0000 (19:26 +0100)
committerYann E. MORIN <yann.morin.1998@free.fr>
Fri, 6 Nov 2020 22:54:02 +0000 (23:54 +0100)
commit1482ef6bb8209044372ebcaf07081524f9b486fa
treee98ba368c20dde20efb0442725e257b038854437
parent46aab84c6fa1c242b304cbd9cd79a7fd94f7a31a
Revert "Makefile: exclude BR2_DL_DIR from savedefconfig"

Although BR2_DL_DIR is indeed a site-local setting, which does not
actually define the target system, we've had it in the tree for a
long time now, and people have been depending on it for a variety
of use-cases.

Furthermore, BR2_DL_DIR is far from the only such site-local setting,
BR2_CCACHE_DIR springs to mind, and in the less-obvious category, we
can also find BR2_JLEVEL, but also BR2_WGET, BR2_SVN, BR2_GIT et al.
as they may be tweaked to set the timeout, number of retries or so on
to work around stupid proxies. But of course, the most local site-local
setting is probably BR2_PACKAGE_OVERRIDE_FILE, with its default value
being explicitly just 'local.mk'.

Ideally, we would like to have a clear separation between the
configuration that actually defines the target system on one hand,
and the site-local settings that drive and control how the build is
performed, on the other hand. This is by far a much bigger endeavour
than just dropping BR2_DL_DIR from the saved defconfig.

This reverts commit 36edacce9c2c3b90f9bb11a5d2208e8edf7bbe63 (adapted
to keep the fix from 1a7873ec986c817fdd22cc2d9096d9482fee4381).

Closes: #13291
Note: thanks to Thomas; some phrasing above was borrowed from a
discussion with him.

Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Lance Fredrickson <lancethepants@gmail.com>
Cc: Sven Oliver Moll <buildroot@svol.li>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Adam Duskett <aduskett@gmail.com>
Makefile