On Wed, Jun 19, 2019 at 09:11:25PM +0100, Andrew Cooper wrote: > Neither of these should live in .data > > * .data.schedulers is only ever read, so is moved into .rodata > * CONSTRUCTORS is only ever read, and only at boot, so is moved to beside > .init.rodata > > Signed-off-by: Andrew Cooper <andrew.coop...@citrix.com>
For x86: Reviewed-by: Roger Pau Monné <roger....@citrix.com> One comment below: > --- > CC: Jan Beulich <jbeul...@suse.com> > CC: Wei Liu <w...@xen.org> > CC: Roger Pau Monné <roger....@citrix.com> > CC: Stefano Stabellini <sstabell...@kernel.org> > CC: Julien Grall <julien.gr...@arm.com> > --- > xen/arch/arm/xen.lds.S | 11 ++++++----- > xen/arch/x86/xen.lds.S | 11 ++++++----- > 2 files changed, 12 insertions(+), 10 deletions(-) > > diff --git a/xen/arch/x86/xen.lds.S b/xen/arch/x86/xen.lds.S > index ec37d38..9fa6c99 100644 > --- a/xen/arch/x86/xen.lds.S > +++ b/xen/arch/x86/xen.lds.S > @@ -140,6 +140,11 @@ SECTIONS > *(.data.param) > __param_end = .; > > + . = ALIGN(POINTER_ALIGN); > + __start_schedulers_array = .; > + *(.data.schedulers) > + __end_schedulers_array = .; > + > #if defined(CONFIG_HAS_VPCI) && defined(CONFIG_LATE_HWDOM) > . = ALIGN(POINTER_ALIGN); > __start_vpci_array = .; > @@ -207,6 +212,7 @@ SECTIONS > > *(.init.rodata) > *(.init.rodata.*) > + CONSTRUCTORS According to the ld manual CONSTRUCTORS is only relevant for a.out, ECOFF and XCOFF. I'm unsure whether PE does use CONSTRUCTORS or not, since it's a descendant of COFF. Thanks, Roger. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xenproject.org https://lists.xenproject.org/mailman/listinfo/xen-devel
