This is the mail archive of the binutils@sourceware.org mailing list for the binutils project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [binutils-gdb] Add .nop assembler directive


On Mon, Feb 19, 2018 at 06:10:19AM -0800, H.J. Lu wrote:
> On Sun, Feb 18, 2018 at 8:50 PM, Alan Modra <amodra@gmail.com> wrote:
> > On Sat, Feb 17, 2018 at 01:22:41PM -0000, H.J.Lu wrote:
> >> https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=62a02d25b6e5d9f92c205260daa11355d0c62532
> >>
> >> commit 62a02d25b6e5d9f92c205260daa11355d0c62532
> >> Author: H.J. Lu <hjl.tools@gmail.com>
> >> Date:   Sat Feb 17 05:20:42 2018 -0800
> >>
> >>     Add .nop assembler directive
> >
> > It looks like this patch caused the following regressions.  Please
> > investigate.
> >
> > m68hc11-elf  +FAIL: objdump -WL
[snip lots more]
> 
> These are caused by
> 
> config/tc-m68hc11.h:#define NO_PSEUDO_DOT 1
> config/tc-m68k.h:#define NO_PSEUDO_DOT 1
> config/tc-spu.h:#define NO_PSEUDO_DOT 1
> config/tc-xgate.h:#define NO_PSEUDO_DOT 1
> config/tc-xtensa.c:/* NO_PSEUDO_DOT hook */
> config/tc-xtensa.h:#define NO_PSEUDO_DOT        xtensa_check_inside_bundle ()
> config/tc-z80.h:#define NO_PSEUDO_DOT                1
> 
> We match  .nop diretive to nop instruction.  This patch avoids
> matching .nop diretive to nop instruction when NO_PSEUDO_DOT
> is defined to 1.   OK for master?

> 	* read.c (potable): Sort nop.
> 	(read_a_source_file): Avoid matching .nop diretive to nop
> 	instruction.

Another fix would be to add "nop" with a NULL poc_handler to each of
these targets' md_pseudo_table, avoiding any extra code.  However, I'm
inclined to think the directive should be renamed.

-- 
Alan Modra
Australia Development Lab, IBM


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]