On Fri, Nov 30, 2018 at 10:54 AM Masahiro Yamada <yamada.masahiro@xxxxxxxxxxxxx> wrote: > > On Sun, Nov 25, 2018 at 2:51 AM Ben Hutchings <ben@xxxxxxxxxxxxxxx> wrote: > > > > On Sat, 2018-11-24 at 07:51 +0100, Sedat Dilek wrote: > > > On Sat, Nov 24, 2018 at 1:53 AM Masahiro Yamada > > > <yamada.masahiro@xxxxxxxxxxxxx> wrote: > > [...] > > > > Maybe like this? > > > > > > > > > > > > diff --git a/scripts/package/builddeb b/scripts/package/builddeb > > > > index f43a274..4cf5a5c 100755 > > > > --- a/scripts/package/builddeb > > > > +++ b/scripts/package/builddeb > > > > @@ -165,6 +165,9 @@ fi > > > > if grep -q '^CONFIG_GCC_PLUGINS=y' $KCONFIG_CONFIG ; then > > > > (cd $objtree; find scripts/gcc-plugins -name \*.so -o -name > > > > gcc-common.h) >> "$objtree/debian/hdrobjfiles" > > > > fi > > > > +if [ -r arch/$SRCARCH/kernel/macros.s ]; then > > > > + echo arch/$SRCARCH/kernel/macros.s >> "$objtree/debian/hdrobjfiles" > > > > +fi > > > > destdir=$kernel_headers_dir/usr/src/linux-headers-$version > > > > mkdir -p "$destdir" > > > > (cd $srctree; tar -c -f - -T -) < "$objtree/debian/hdrsrcfiles" | (cd > > > > $destdir; tar -xf -) > > > > > > > > > > > > -- > > > > Best Regards > > > > Masahiro Yamada > > > > > > Hi Masahiro Yamada, > > > > > > thanks, that fixes here the problem with building DKMS kernel modules: > > > > > > # find /usr/src/linux-headers-4.20.0-rc3-7-amd64-cbl -name macros.s > > > /usr/src/linux-headers-4.20.0-rc3-7-amd64-cbl/arch/x86/kernel/macros. > > > > > > I guess commit... > > > > > > 77b0bf55bc67 kbuild/Makefile: Prepare for using macros in inline > > > assembly code to work around asm() related GCC inlining bugs > > > > > > ...is the root cause (mention as a "Fixes:" tag in a future patch)? > > > > > > Not sure if the according script for building Linux RPM packages needs > > > the same fix, too. > > > > > > I can send a fix for the builddeb part or if you want to do it, it's OK for me. > > > > > > I highly appreciate to see a fix in Linux v4.20-rcN if possible. > > > > Perhaps it would make more sense to move this file under > > arch/x86/include/generated, so there's no need to add a special case to > > the packaging rules. > > > Yeah, moving the file under include/ would be nicer. > > But, I am asking if they are fine with reverting the commits causing > the problems. > (https://lkml.org/lkml/2018/11/29/910) > > Please give me time to decide what to do. > I see Ingo reverted patches in tip.git#x86/urgent. - Sedat - [1] https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/log/?h=x86/urgent
