[PATCH] arm64: extable: sort the exception table at build time

Will Deacon will.deacon at arm.com
Mon May 20 05:46:04 EDT 2013


Hi David,

On Fri, May 17, 2013 at 07:33:53PM +0100, David Daney wrote:
> On 05/17/2013 09:43 AM, Will Deacon wrote:
> > As is done for other architectures, sort the exception table at
> > build-time rather than during boot.
> >
> > Since sortextable appears to be a standalone C program relying on the
> > host elf.h to provide EM_AARCH64, I've had to add a conditional check in
> > order to allow cross-compilation on machines that aren't running a
> > bleeding-edge libc-dev.
> >
> > Signed-off-by: Will Deacon <will.deacon at arm.com>
> > ---
> >   arch/arm64/Kconfig              |  1 +
> >   arch/arm64/kernel/vmlinux.lds.S | 15 +++++++--------
> >   scripts/sortextable.c           |  5 +++++
> >   3 files changed, 13 insertions(+), 8 deletions(-)
> >
> > diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
> > index 48347dc..fb8fd90 100644
> > --- a/arch/arm64/Kconfig
> > +++ b/arch/arm64/Kconfig
> > @@ -7,6 +7,7 @@ config ARM64
> >   	select ARM_AMBA
> >   	select ARM_ARCH_TIMER
> >   	select ARM_GIC
> > +	select BUILDTIME_EXTABLE_SORT
> >   	select CLONE_BACKWARDS
> >   	select COMMON_CLK
> >   	select GENERIC_CLOCKEVENTS
> > diff --git a/arch/arm64/kernel/vmlinux.lds.S b/arch/arm64/kernel/vmlinux.lds.S
> > index 3fae2be..3d527726 100644
> > --- a/arch/arm64/kernel/vmlinux.lds.S
> > +++ b/arch/arm64/kernel/vmlinux.lds.S
> > @@ -57,6 +57,13 @@ SECTIONS
> >
> >   	RO_DATA(PAGE_SIZE)
> >
> > +	. = ALIGN(8);
> > +	__ex_table : AT(ADDR(__ex_table) - LOAD_OFFSET) {
> > +		__start___ex_table = .;
> > +		*(__ex_table)
> > +		__stop___ex_table = .;
> > +	}
> > +
> >   	_etext = .;			/* End of text and rodata section */
> 
> 
> You have moved this to rodata.

I don't think I have moved this to rodata. The RO_DATA macro expands to a
bunch of sections, but it doesn't leave any of them open at the end.

> o Will the runtime sort still work like this?
> 
> o Do we care?  Maybe not.

For arm64 we'd be moving exclusively to build-time sorting, so I guess we
*could* stick the thing in rodata if we wanted to. Not sure it's worth the
linker script munging though.

Will



More information about the linux-arm-kernel mailing list