On Tue, Jan 30, 2018 at 07:51:31PM +0100, Greg Kroah-Hartman wrote:
> On Tue, Jan 30, 2018 at 03:51:56PM +0100, Greg Kroah-Hartman wrote:
> > On Tue, Jan 30, 2018 at 06:19:15AM -0800, Guenter Roeck wrote:
> > > On 01/29/2018 04:56 AM, Greg Kroah-Hartman wrote:
> > > > This is the start of the stable review cycle for the 3.18.93 release.
> > > > There are 52 patches in this series, all will be posted as a response
> > > > to this one. If anyone has any issues with these being applied, please
> > > > let me know.
> > > >
> > > > Responses should be made by Wed Jan 31 12:36:07 UTC 2018.
> > > > Anything received after that time might be too late.
> > > >
> > >
> > > Build results:
> > > total: 136 pass: 135 fail: 1
> > > Failed builds:
> > > um:defconfig
> > > Qemu test results:
> > > total: 112 pass: 112 fail: 0
> > >
> > > The build failure is:
> > >
> > > In file included from arch/um/kernel/config.c:8:0:
> > > arch/um/include/shared/init.h:43:28: fatal error: linux/compiler.h: No such file or directory
> > >
> > > [ several instances ]
> >
> > Crap, I was trying to apply a number of the UM patches that Android
> > relies on for their build systems that they patch their kernel for.
> > I'll go look into those to try to figure out what I got wrong...
>
> Ok, I can't even build a defconfig for ARCH=um at all, with no patches
> applied to 3.18. I had to go find a 4.9.4 kernel to even get close to
> building, gcc7 did really odd things.
>
> Do you have the .config file you use to build this arch with? I looked
> on the builder site and couldn't seem to find it anywhere, am I just
> missing something obvious?
>
mkdir /tmp/build
make O=/tmp/build ARCH=um SUBARCH=x86_64 defconfig
make O=/tmp/build ARCH=um SUBARCH=x86_64 -j30
The O= is essential; in-tree builds are fine.
Also, turns out you are correct; 3.18.92 fails to build for me
as well if I use a recent compiler. It does build with the compiler
from Poky 1.3. Bisect points to commit a3a8321bf0f00 ("um: Remove
copy&paste code from init.h)"; bisect log is attached.
Not sure if it is worth fixing it, though. Maybe I should just
stop building it for 3.18 instead. Thoughts ?
Guenter
---
# bad: [9ea3053b8236d87e0716496e6cd90242aadc2f63] Linux 3.18.93-rc1
# good: [a5d35deca214e095bf9d1745aa6c00dd7ced0517] Linux 3.18.92
git bisect start 'HEAD' 'v3.18.92'
# good: [b70017b84be2dbab6c7d47c898d8b0c298d1924f] netfilter: nf_ct_expect: remove the redundant slash when policy name is empty
git bisect good b70017b84be2dbab6c7d47c898d8b0c298d1924f
# bad: [dcdf22915ebd63044e32efe6f78511c7360fb105] x86/microcode/intel: Extend BDW late-loading further with LLC size check
git bisect bad dcdf22915ebd63044e32efe6f78511c7360fb105
# good: [eb71bc9be0cb6615f0743571b560e773ad3f58af] reiserfs: don't preallocate blocks for extended attributes
git bisect good eb71bc9be0cb6615f0743571b560e773ad3f58af
# good: [3db3a49f92e159173ca108f01559d45b95f66933] um: link vmlinux with -no-pie
git bisect good 3db3a49f92e159173ca108f01559d45b95f66933
# good: [05fbee5e81254471451ee0232edaead9f71329b2] um: Stop abusing __KERNEL__
git bisect good 05fbee5e81254471451ee0232edaead9f71329b2
# bad: [a3a8321bf0f001b21178712b5e1693f54afe95db] um: Remove copy&paste code from init.h
git bisect bad a3a8321bf0f001b21178712b5e1693f54afe95db
# first bad commit: [a3a8321bf0f001b21178712b5e1693f54afe95db] um: Remove copy&paste code from init.h