(mark_feature_used): New function.
(parse_barrier): Check specified option is valid for the
specified architecture.
(UL_BARRIER): New macro.
(barrier_opt_names): Update for new barrier options.
* gas/testsuite/gas/arm/armv8-a-barrier.s: New testcase.
* gas/testsuite/gas/arm/armv8-a-barrier-arm.d: Likewise.
* gas/testsuite/gas/arm/armv8-a-barrier-thumb.d: Likewise.
* opcodes/arm-dis.c (data_barrier_option): New function.
(print_insn_arm): Use data_barrier_option.
(print_insn_thumb32): Use data_barrier_option.
(mmix_greg_internal): Handle expressions not determinable at first
pass.
(s_loc): Ditto. Record expressions where the section isn't
determinable at the first pass, and assume they don't refer to
other sections.
(mmix_md_end): Verify that recorded LOC expressions weren't
to other sections, else emit error messages.
prefixes for other than FS/GS in 64-bit mode. If doing so at all, it
should clearly do this correctly. Determining the default segment,
however, requires to take into consideration RegRex (so far, RSP, RBP,
R12, and R13 were all treated equally here).
gas/
2012-08-07 Jan Beulich <jbeulich@suse.com>
* config/tc-i386-intel.c (build_modrm_byte): Split determining
default segment from figuring out encoding. Honor RegRex for
the former.
gas/testsuite/
2012-08-07 Jan Beulich <jbeulich@suse.com>
* gas/i386/x86-64-segovr.{s,l}: New.
* gas/i386/i386.exp: Run new test.
xmm/ymm registers are distinct. This patch adds code to check for this,
and at once eliminates a superfluous check for not using PC-relative
addressing for these instructions (the fact that an index register is
required here already excludes valid PC-relative addresses). The
severity of the resulting diagnostics can be controlled via command
line option or directive.
gas/
2012-08-07 Jan Beulich <jbeulich@suse.com>
* config/tc-i386.c (set_check): Renamed from set_sse_check.
Generalize to also handle operand checking option.
(enum i386_error): New enumerator 'invalid_vector_register_set'.
(match_template): Handle it.
(enum check_kind): Give it a tag. Drop sse_ prefixes from
enumerators.
(operand_check): New.
(md_pseudo_table): Add "operand_check".
(check_VecOperands): Don't special case RIP addressing. Check
that vSIB operands use distinct vector registers unless no
checking was requested.
(OPTION_MOPERAND_CHECK): New.
(md_parse_option): Handle it.
(OPTION_MAVXSCALAR, OPTION_X32): Adjust.
(md_longopts): Add "moperand-check".
(md_show_usage): Add help text for it.
gas/testsuite/
2012-08-07 Jan Beulich <jbeulich@suse.com>
* gas/i386/vgather-check-error.{s,l}: New.
* gas/i386/vgather-check-none.{s,d}: New.
* gas/i386/vgather-check-warn.{d,e}: New.
* gas/i386/vgather-check.{s,d}: New.
* gas/i386/x86-64-vgather-check-error.{s,l}: New.
* gas/i386/x86-64-vgather-check-none.{s,d}: New.
* gas/i386/x86-64-vgather-check-warn.{d,e}: New.
* gas/i386/x86-64-vgather-check.{s,d}: New.
* gas/i386/i386.exp: Run new tests.
got treated identically to the ones in the base range, due to not
paying attention to the fact that reg_entry's reg_num field doesn't
fully specify the register number (reg_flags also needs to be checked
for RegRex). This patch introduces and uses a new (inline) function to
obtain the full register number, and uses it to fix all those cases.
It additionally adds the missing operand checks for SVME instructions
(which match the monitor/mwait ones).
gas/
2012-08-07 Jan Beulich <jbeulich@suse.com>
* config/tc-i386.c (register_number): New function.
(build_vex_prefix, process_immext, process_operands,
build_modrm_byte, i386_index_check): Use it.
gas/testsuite/
2012-08-07 Jan Beulich <jbeulich@suse.com>
* gas/i386/x86-64-specific-reg.{s,l}: New.
* gas/i386/i386.exp: Run new test.
opcodes/
2012-08-07 Jan Beulich <jbeulich@suse.com>
* i386-opc.tbl: Remove "FIXME" comments from SVME instructions.
* config/tc-mips.c (append_insn): Also handle moving delay-slot
instruction across frags for fixed branches.
gas/testsuite/
* gas/mips/branch-swap-2.l: New list test.
* gas/mips/branch-swap-2.s: New test source.
* gas/mips/mips.exp: Run the new test.
2012-08-01 James Lemke <jwlemke@codesourcery.com>
* gas/dwarf2dbg.c (out_set_addr): Allow for non-constant value of
DWARF2_LINE_MIN_INSN_LENGTH
* gas/config/tc-ppc.c (ppc_dwarf2_line_min_insn_length): Declare
and initialize.
(md_apply_fix): Branch addr can be a multiple of 2 or 4.
* gas/config/tc-ppc.h (DWARF2_LINE_MIN_INSN_LENGTH): Now a
variable reference.
gas/testsuite/ChangeLog:
2012-08-01 James Lemke <jwlemke@codesourcery.com>
* gas/cfi/cfi-ppc-1.d: Allow for code alignment of 2 or 4.
ld/ChangeLog:
2012-08-01 James Lemke <jwlemke@codesourcery.com>
* ld/testsuite/ld-gc/pr13683.d: XFAIL for powerpc*-*-eabivle.
and was pointing at the wrong operand in Intel mode. Since non-constant
operands are being taken care of by other means anyway, adjust it to
simply state that the constant doesn't fit.
2012-07-31 Jan Beulich <jbeulich@suse.com>
* config/tc-i386.c (match_template): Adjust error message
for 'bad_imm4' case.
anyway, there's also no need to issue a separate, inconsistent
diagnostic in some of the cases - non-matching operands will be
complained about anyway.
2012-07-31 Jan Beulich <jbeulich@suse.com>
* config/tc-i386.c (check_byte_reg): Check for I/O port
register earlier, and just once. Drop diagnostic that got
issued only for some registers.
all other instruction attributes already matched, so any mismatch here
will tell the user more precisely what is wrong than using an eventual
(and very likely to occur) more generic error encountered on a
subsequent iteration through the template matching loop.
2012-07-31 Jan Beulich <jbeulich@suse.com>
* config/tc-i386.c (match_template): New local variable
'specific_error'. Set it from i.error after failed
check_VecOperands or VEX_check_operands. Use it if set in
preference to i.error when actually issuing disagnostic.
Move the package name/version from AM_INIT_AUTOMAKE to AC_INIT per recent
autotools guidelines. We use recent versions of both, so it shouldn't be
a problem.
This sets PACKAGE_xxx variables correctly, and makes the output of:
./configure --version
actually useful:
bfd configure 2.22.52
Changing the other dirs to use AC_INIT would require a bit of m4 trickery
that I don't feel like getting into, and they all use BFD_VERSION anyways,
so there isn't much point.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
operand, by silently making it the base register despite not being
specified first.
Consequently, we also permit an xmm/ymm index to be specified first
(possibly alone), nevertheless putting it in as index register.
2012-07-24 Jan Beulich <jbeulich@suse.com>
* config/tc-i386-intel.c (i386_intel_simplify_register): Handle
xmm/ymm index register being specified first as well as esp/rsp
base register being specified last in a memory operand.
Jie Zhang <jzhang918@gmail.com>
gas/
* config/tc-arm.c (md_apply_fix): Use encoding A2 of ADR
if offset is negative.
gas/testsuite/
* gas/arm/adr.d: New test.
* gas/arm/adr.s: New test.
* config/tc-arm.c (encode_arm_addr_mode_common): Generate an error
message if literal pool addressing is used.
* gas/arm/ldr-t-bad.s: Add test of bogus use of literal pool
addressing.
* gas/arm/ldr-t-bad.l: Update expected assembler error message
output.
linker relaxation.
(dwarf2_gen_line_info): Generate real, local, labels for line
numbers.
(dwarf2dbg_convert_frag): Do not finalize the computation of the
frag's symbol value when linker relaxation is enabled.
(ADDR_DELTA_LIMIT): Define.
(size_fixed_inc_line_addr): Use ADDR_DELTA_LIMIT.
(emit_fixed_inc_line_addr): Likewise.
* write.c (fixup_segment): If the subtraction of two symbols
cannot be resolved but is valid, then prevent bogus range warnings
by pre-biasing add_number.
* config/tc-h8300.h (DWARF2_USE_FIXED_ADVANCE_PC): Define to 0.
* gas/lns/lns.exp: Use alternate lns-common test for targets
enabling linker relaxation.
* gas/lns/lns-big-delta.d: Allow for output from architectures
with 32-bit addresses.
