Hi,
in 5.10 kernels up to and including 5.10.15 when trying to build the
kernel for an x86_64 skylake using binutils-2.36.1, gcc-10.2 and
glibic-2.33 I get a segfault in objtool if the orc unwinder is
enabled.
This has already been fixed in 5.11 by ''objtool: Fix seg fault with
Clang non-section symbols'
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/patch/?id=44f6a7c0755d8dd453c70557e11687bb080a6f21
So can this be added to 5.10 stable, please ?
Please CC me as I am no-longer subscribed.
ĸen
On Thu, Feb 11, 2021 at 05:16:56PM +0000, Ken Moffat wrote:
> Hi,
>
> in 5.10 kernels up to and including 5.10.15 when trying to build the
> kernel for an x86_64 skylake using binutils-2.36.1, gcc-10.2 and
> glibic-2.33 I get a segfault in objtool if the orc unwinder is
> enabled.
>
> This has already been fixed in 5.11 by ''objtool: Fix seg fault with
> Clang non-section symbols'
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/patch/?id=44f6a7c0755d8dd453c70557e11687bb080a6f21
>
> So can this be added to 5.10 stable, please ?
>
> Please CC me as I am no-longer subscribed.
Hi Ken,
I agree that needs to be backported (and my bad for not marking it as
stable to begin with).
Greg, this also came up in another thread, are you pulling that one in,
or do you want me to send it to stable list?
--
Josh
On Fri, Feb 12, 2021 at 05:51:45PM -0600, Josh Poimboeuf wrote:
> On Thu, Feb 11, 2021 at 05:16:56PM +0000, Ken Moffat wrote:
> > Hi,
> >
> > in 5.10 kernels up to and including 5.10.15 when trying to build the
> > kernel for an x86_64 skylake using binutils-2.36.1, gcc-10.2 and
> > glibic-2.33 I get a segfault in objtool if the orc unwinder is
> > enabled.
> >
> > This has already been fixed in 5.11 by ''objtool: Fix seg fault with
> > Clang non-section symbols'
> > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/patch/?id=44f6a7c0755d8dd453c70557e11687bb080a6f21
> >
> > So can this be added to 5.10 stable, please ?
> >
> > Please CC me as I am no-longer subscribed.
>
> Hi Ken,
>
> I agree that needs to be backported (and my bad for not marking it as
> stable to begin with).
>
> Greg, this also came up in another thread, are you pulling that one in,
> or do you want me to send it to stable list?
I will pull it in after the next release happens in a few hours, thanks.
greg k-h