can you please clarify priority? is this a P3?

and is this build using LLVM? (I don't think so, but then what is new that is causing/exposing the problem?)

Priority is low. For my part it is primarily to record the problem; I am fine with marking it as WontFix. This is with gcc, not llvm/clang.
There is nothing new; it is just a different combination of kernel build options that was never tested. Background is that I am adding the ability to run test builds. I encountered the problem while running "allmodconfig" or "allyesconfig" builds using the toolchain (test builds do not set CONFIG_DEBUG). I can work around the problem by explicitly adding " -g" to KCFLAGS (I have to set KCFLAGS anyway to work around https://bugzilla.kernel.org/show_bug.cgi?id=187841).

ok, since we are moving the kernel to LLVM, I would suggest we don't work on this unless we can reproduce with LLVM. 

SGTM. I'll retest with clang after I have the initial setup working.

Regarding clang: Turns out that building an "allmodconfig" or "allyesconfig" kernel with clang fails spectacularly, even with the latest upstream kernel. This is not an option for test builds, and will not be for the forseeable future.

Turns out this is a problem with all toolchains I tested, not just with ours. Will discuss with upstream.

See CL:699799

Turns out this is a gcc bug; gcc puts symbol 'kvm_fastop_exception' into the wrong section if the kernel is built with "-frecord-gcc-switches" and without "-g". This appears to affect all gcc versions. The CL adds a workaround into the kernel.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/e6f479d8dca31429456b65356e84993f0f66f5b0

commit e6f479d8dca31429456b65356e84993f0f66f5b0
Author: Josh Poimboeuf <jpoimboe@redhat.com>
Date: Thu Oct 05 07:34:17 2017

FROMLIST: x86/kvm: Move kvm_fastop_exception to .fixup section

When compiling the kernel with the '-frecord-gcc-switches' flag, objtool
complains:

  arch/x86/kvm/emulate.o: warning: objtool: .GCC.command.line+0x0: special: can't find new instruction

And also the kernel fails to link.

The problem is that the 'kvm_fastop_exception' code gets placed into the
throwaway '.GCC.command.line' section instead of '.text'.

Exception fixup code is conventionally placed in the '.fixup' section,
so put it there where it belongs.

BUG= chromium:769037 
TEST="Build kvm with -frecord-gcc-switches and without -g"

Change-Id: Ie88a4f0e637a1be403c3d74779004012d44d78d3
Reported-and-tested-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Guenter Roeck <groeck@chromium.org>
(am from https://patchwork.kernel.org/patch/9984845/)
Reviewed-on: https://chromium-review.googlesource.com/699799
Reviewed-by: Dylan Reid <dgreid@chromium.org>

[modify] https://crrev.com/e6f479d8dca31429456b65356e84993f0f66f5b0/arch/x86/kvm/emulate.c

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/6144aedde4bffe3acf14bcfea0d2058e2caf8439

commit 6144aedde4bffe3acf14bcfea0d2058e2caf8439
Author: Josh Poimboeuf <jpoimboe@redhat.com>
Date: Thu Oct 05 10:04:18 2017

FROMLIST: x86/kvm: Move kvm_fastop_exception to .fixup section

When compiling the kernel with the '-frecord-gcc-switches' flag, objtool
complains:

  arch/x86/kvm/emulate.o: warning: objtool: .GCC.command.line+0x0: special: can't find new instruction

And also the kernel fails to link.

The problem is that the 'kvm_fastop_exception' code gets placed into the
throwaway '.GCC.command.line' section instead of '.text'.

Exception fixup code is conventionally placed in the '.fixup' section,
so put it there where it belongs.

BUG= chromium:769037 
TEST="Build kvm with -frecord-gcc-switches and without -g"

Change-Id: Ie88a4f0e637a1be403c3d74779004012d44d78d3
Reported-and-tested-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Guenter Roeck <groeck@chromium.org>
(am from https://patchwork.kernel.org/patch/9984845/)
Reviewed-on: https://chromium-review.googlesource.com/699641

[modify] https://crrev.com/6144aedde4bffe3acf14bcfea0d2058e2caf8439/arch/x86/kvm/emulate.c

While this is a gcc problem, it affects all versions of gcc, and thus had to be fixed in the kernel. Patch will hopefully be applied to the upstream kernel.