Hi, ever since commit https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=84d69848c97faab0c25aa2667b273404d2e2a64a Merge branch 'kbuild' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild I get warnings during build: ... WARNING: "_copy_from_user" [drivers/md/dm-mod.ko] has no CRC! WARNING: "___preempt_schedule" [drivers/md/dm-crypt.ko] has no CRC! WARNING: "__sw_hweight64" [drivers/md/dm-crypt.ko] has no CRC! WARNING: "memcpy" [drivers/md/dm-crypt.ko] has no CRC! WARNING: "memset" [drivers/md/dm-crypt.ko] has no CRC! WARNING: "__get_user_8" [drivers/input/serio/serport.ko] has no CRC! WARNING: "__get_user_4" [drivers/input/serio/serport.ko] has no CRC! WARNING: "__put_user_1" [drivers/input/serio/serio_raw.ko] has no CRC! WARNING: "__get_user_1" [drivers/input/serio/serio_raw.ko] has no CRC! WARNING: "__sw_hweight32" [drivers/input/serio/pcips2.ko] has no CRC! WARNING: "_copy_to_user" [drivers/input/mousedev.ko] has no CRC! ... Reverting the commit fixes the issue, as selecting CONFIG_RELOCATABLE in kernel config, so issue only occur if CONFIG_RELOCATABLE is not set. This is a default setting on most distros since it's a requirement for UEFI, but saw it on LKML too: https://lkml.org/lkml/2016/10/27/471 Commit message clearly states that it does bring a regression, but I'm not sure if it's related, hence the bug report so it could be tracked. Regards, Ivan
I can confirm this issue even with CONFIG_RELOCATABLE set.
(In reply to Daniel Exner from comment #1) > I can confirm this issue even with CONFIG_RELOCATABLE set. +1
Fixed by the recent MODVERSIONS commits.
unless I'm missing something, I still get these errors on gentoo's sys-kernel/gentoo-sources-4.9.0, building an efistub kernel, is this going to be part of say 4.9.1 ?
Fixed 2016-NOV-29 by commit ID faaae2a581435f32781a105dda3501df388fddcb.