This is the mail archive of the
mailing list for the binutils project.
Re: [PATCH] Add support for MIPS eXtended Physical Address (XPA) ASE
- From: Richard Sandiford <rdsandiford at googlemail dot com>
- To: "Maciej W. Rozycki" <macro at codesourcery dot com>
- Cc: Andrew Bennett <Andrew dot Bennett at imgtec dot com>, "binutils\ at sourceware dot org" <binutils at sourceware dot org>
- Date: Mon, 02 Jun 2014 10:09:10 +0100
- Subject: Re: [PATCH] Add support for MIPS eXtended Physical Address (XPA) ASE
- Authentication-results: sourceware.org; auth=none
- References: <0DA23CC379F5F945ACB41CF394B982775599D3 at LEMAIL01 dot le dot imgtec dot org> <871twoxnx1 dot fsf at sandifor-thinkpad dot stglab dot manchester dot uk dot ibm dot com> <alpine dot DEB dot 1 dot 10 dot 1405300920100 dot 512 at tp dot orcam dot me dot uk>
"Maciej W. Rozycki" <email@example.com> writes:
> On Wed, 23 Apr 2014, Richard Sandiford wrote:
>> Andrew Bennett <Andrew.Bennett@imgtec.com> writes:
>> > binutils/
>> > * doc/binutils.texi: Document the disassemble MIPS XPA instructions
>> > command line option.
>> > gas/
>> > * config/tc-mips.c (options): Add OPTION_XPA and OPTION_NO_XPA.
>> > (md_longopts): Add xpa and no-xpa command line options.
>> > (mips_ases): Add MIPS XPA ASE.
>> > (mips_cpu_info_table): Update p5600 entry to allow the XPA ASE.
>> > * doc/as.texinfo: Document the MIPS XPA command line options.
>> > * doc/c-mips.texi: Document the MIPS XPA command line options,
>> > and assembler directives.
>> > gas/testsuite/
>> > * gas/mips/mips.exp: Add xpa tests.
>> > * gas/mips/xpa.s: New test.
>> > * gas/mips/xpa.d: Likewise.
>> > include/
>> > * opcode/mips.h (ASE_XPA): New define.
>> > opcodes/
>> > * mips-dis.c (mips_arch_choices): Update mips32r2 and mips64r2
>> > to allow the MIPS XPA ASE.
>> > (parse_mips_dis_option): Process the -Mxpa option.
>> > * mips-opc.c (XPA): New define.
>> > (mips_builtin_opcodes): Add MIPS XPA instructions and move the
>> > locations of the ctc0 and cfc0 instructions.
>> Looks good, thanks.
> These instructions are new for MIPSr5, I think they should be marked r5
> just as the ERETNC support posted alongside.
> Overall I think we should go back and review r3 additions and mark them
> as such too -- offhand that would include EVA, virtualisation and TLBINV/F
Yeah, I agree we should do that eventually. It might be a good idea to
wait until r3 and r5 support is added to GCC though, so that assembly
driven by the GCC driver still works without resorting to -Wa.
AIUI that patch is being held up on copyright assignment details.