[U-Boot] [PATCH 4/5] arm: ensure u-boot only uses relative relocations

Benoît Thébaudeau benoit.thebaudeau at advansee.com
Wed May 15 15:49:10 CEST 2013


Hi Albert,

On Wednesday, May 15, 2013 11:38:37 AM, Albert ARIBAUD wrote:
> Hi again Benoît,
> 
> On Wed, 15 May 2013 09:46:17 +0200, Albert ARIBAUD
> <albert.u.boot at aribaud.net> wrote:
> 
> > Hi Benoît,
> > 
> > On Wed, 15 May 2013 00:12:24 +0200 (CEST), Benoît Thébaudeau
> > <benoit.thebaudeau at advansee.com> wrote:
> > 
> > > Hi Albert,
> > 
> > > > --- a/Makefile
> > > > +++ b/Makefile
> > > > @@ -746,6 +746,13 @@ tools: $(VERSION_FILE) $(TIMESTAMP_FILE)
> > > >  	$(MAKE) -C $@ all
> > > >  endif	# config.mk
> > > >  
> > > > +# ARM relocations should all be R_ARM_RELATIVE.
> > > > +checkarmreloc: $(obj)u-boot
> > > > +	@if test "R_ARM_RELATIVE" != \
> > > > +		"`readelf -r $(obj)u-boot | cut -d ' ' -f 4 | grep R_ARM | sort
> > > > -u`"; \
> > >                              ^
> > >                              or $$< to avoid a duplicate?
> > 
> > Will fix as suggested.
> 
> > > > +		then echo "$(obj)u-boot contains relocations other than \
> > >                            ^
> > >                            or $$< too, or no $(obj) prefix at all for
> > >                            this message?
> > 
> > I prefer leaving the prefix so that failures during out-of-tree builds
> > or during MAKEALL builds with BUILD_NBUILDS>1 log the correct path.
> 
> Actually $$< does not work within backquotes unless escaped as a less
> legible \$\$<, and does not work properly at all within double quotes,
> whether escaped or not.
> 
> Do you prefer that I change only the first $(obj)u-boot into \$\$< and
> leave the second one untouched, or that I leave both $(obj)u-boot
> instances as-is for the sake of homogeneity?

Actually, a single dollar sign (i.e. "$<") would be needed since it must have
been expanded by make before reaching the shell, and no shell backslash escape
sequences should be required.

If this still does not pass smoothly, then I prefer simplicity and homogeneity.

Best regards,
Benoît


More information about the U-Boot mailing list