From mboxrd@z Thu Jan 1 00:00:00 1970 From: Yann E. MORIN Date: Mon, 24 Aug 2015 17:33:45 +0200 Subject: [Buildroot] [PATCH 0/3] core/pkg-kernel-config: ensure linux supports modules (branch yem/kernel-module) In-Reply-To: <20150824161129.272f994c@pcviktorin.fit.vutbr.cz> References: <20150824161129.272f994c@pcviktorin.fit.vutbr.cz> Message-ID: <20150824153345.GB3619@free.fr> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net Jan, All, On 2015-08-24 16:11 +0200, Jan Viktorin spake thusly: > Hello Yann, all. > > I've tried the patch series and it works. However, I've tried to do > this for xtables-addons: > > -LINUX_NEEDS_MODULES = y > +#LINUX_NEEDS_MODULES = y > > and the result is that it again compiles with CONFIG_MODULES=y (and it > shouldn't). I was investigating why and couldn't find the answer. It > just sets the CONFIG_MODULES=y everytime. > > Simple test is possible: > > 1) $ make olimex_a20_olinuxino_lime_defconfig > 2) $ make linux-menuconfig > 3) Check "Enable loadable module support" is (un)checked. > > * For origin/master (7deaa27), it works as expected (unchecked). > * After merge of morin-repo/yem/kernel-module (2af986e) it DOES NOT work > as expected (checked). > > Can you reproduce it as well? I haven't eventried, but I guess Peter nailed it in his reply. I'll fix and respin. Regards, Yann E. MORIN. > Regards > Jan Viktorin > > On Mon, 24 Aug 2015 14:27:38 +0200 > "Yann E. MORIN" wrote: > > > Hello All! > > > > This series fixes two standing issues with the way we handle packages > > that build kernel modules: > > > > - first, the current check is broken, because $(LINUX_DIR) should be > > double-$ed, as it is used in a macro from an infra, > > > > - second, it is currently possible to (try to) build such packages > > without the kernel config havong CONFIG_MODULES set. > > > > So, these three patches fix that: > > > > - introduce a new internal variable to tell the kernel to enable > > support for modules, LINUX_NEEDS_MODULES, that should be set > > whenever a package wants to build a kernel modules, and is > > automatically set by the kernel-module infra, > > > > - the (broken) check is no longer needed so removed, > > > > - a few packages that build kernel modules without using the > > kernel-module infra are fixed to manually set that variable. > > > > I believe that should go into master, because: > > > > - it actually fixes build issues > > > > - the kernel-module infra is brand new in this release, and should > > be fixed and stable prior to being released to the wide hostile world > > ;-) > > > > Thanks to Peter for suggesting the solution, and to No? and Jan for > > their previous attempts that eventually led to this series. > > > > Regards, > > Yann E. MORIN. > > > > > > The following changes since commit > > 7deaa277fd4c89c67de39ea21b4cd081ab85366f: > > > > arch/arm: add missing arm1136j-s variant (2015-08-24 00:43:12 +0200) > > > > are available in the git repository at: > > > > git://git.busybox.net/~ymorin/git/buildroot yem/kernel-module > > > > for you to fetch changes up to > > 2af986ecd146d1dc6e7603bd93fac5420c459e41: > > > > packages: ensure linux supports modules even when not using > > kernel-module (2015-08-24 14:14:23 +0200) > > > > ---------------------------------------------------------------- > > Yann E. MORIN (3): > > core/pkg-kernel-module: ensure linux supports modules > > core/pkg-kernel-module: drop now-useles check for CONFIG_MODULES > > packages: ensure linux supports modules even when not using > > kernel-module > > > > linux/linux.mk | 2 ++ > > package/linux-fusion/linux-fusion.mk | 9 ++++++--- > > package/nvidia-driver/nvidia-driver.mk | 4 ++++ > > package/pkg-kernel-module.mk | 9 +++++---- > > package/racehound/racehound.mk | 4 ++++ > > package/rtai/rtai.mk | 4 ++++ > > package/ti-gfx/ti-gfx.mk | 5 +++++ > > package/xtables-addons/xtables-addons.mk | 4 ++++ > > 8 files changed, 34 insertions(+), 7 deletions(-) > > > > > > -- > Jan Viktorin E-mail: Viktorin at RehiveTech.com > System Architect Web: www.RehiveTech.com > RehiveTech > Brno, Czech Republic -- .-----------------.--------------------.------------------.--------------------. | Yann E. MORIN | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: | | +33 662 376 056 | Software Designer | \ / CAMPAIGN | ___ | | +33 223 225 172 `------------.-------: X AGAINST | \e/ There is no | | http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL | v conspiracy. | '------------------------------^-------^------------------^--------------------'