(cond_jump_mode, loop_jcxz_mode): Define.
(dis386_att): Add cond_jump_flag and loop_jcxz_flag as
appropriate, and 'F' suffix to loop insns.
(disx86_64_att): Likewise.
(dis386_twobyte_att): Likewise.
(print_insn_i386): Don't output addr prefix for loop, jcxz insns.
Output data size prefix for long conditional jumps. Output cs and
ds branch hints.
(putop): Handle 'F', and mark PREFIX_ADDR used for case 'E'.
(OP_J): Don't make PREFIX_DATA used.
(need_modrm): Give it file scope.
(MODRM_CHECK): Define.
(dofloat): Use MODRM_CHECK.
(OP_E): Likewise.
(OP_EM): Likewise.
(OP_EX): Likewise.
and fix testsuite yet again now that we are getting correct disassembly.
[opcodes/ChangeLog]
2001-05-07 Frank Ch. Eigler <fche@redhat.com>
* cgen-dis.in (default_print_insn): Tolerate min<base instructions
even at end of a section.
* cgen-ibld.in (extract_normal): Tolerate min!=base!=max instructions
by ignoring precariously-unpacked insn_value in favor of raw buffer.
[cgen/ChangeLog]
2001-05-07 Frank Ch. Eigler <fche@redhat.com>
* iformat.scm (compute-insn-base-mask-length): Rewrite to tolerate
various-base-length instruction sets.
2001-05-04 Frank Ch. Eigler <fche@redhat.com>
* m32r-dis.c, -asm.c, -ibld.c: Regenerated with disassembler fixes.
2001-05-04 Frank Ch. Eigler <fche@redhat.com>
* cgen-dis.in (print_insn): Remove call to read_insn. Instead,
assume incoming buffer already has the base insn loaded. Handle
case of smaller-than-base instructions for variable-length case.
* cgen-asm.in (@arch@_cgen_assemble_insn): Move tmp_errmsg
declaration inside CGEN_VERBOSE_ASSEMBLER_ERRORS conditional.
* cgen-ibld.in (put_insn_int_value): Mark cd parameter as unused
to allay a compiler warning.
* ia64-gen.c (fetch_insn_class): If xsect, then ignore comment and
notestr if larger than xsect.
(in_class): Handle format M5.
* ia64-asmtab.c: Regnerate.
* cpu-ia64-opc.c (elf64_ia64_operands}: Fix typo: error string for
C8 said "1" instead of "8". Clarify error string for IMM22:
"signed integer" instead of just "integer".
* config/tc-ia64.c (enum operand_match_result): New type.
(operand_match): Change return type to operand_match_result.
Fix all returns appropriately, adding support for returning the
out-of-range result.
(parse_operands): New locals result, error_pos, out_of_range_pos,
curr_out_of_range_pos. Rewrite operand matching loop to give better
error messages.
* ia64-opc-d.c (ia64_opcodes_d): Break the "add" pattern into two
separate variants: one for IMM22 and the other for IMM14.
* ia64-asmtab.c: Regenerate.
Alexandre Oliva