2003-06-23 H.J. Lu <hongjiu.lu@intel.com>
* gas/config/tc-i386.c (md_assemble): Support Intel Precott New
Instructions.
* gas/config/tc-i386.h (CpuPNI): New.
(CpuUnknownFlags): Add CpuPNI.
gas/testsuite/
2003-06-23 H.J. Lu <hongjiu.lu@intel.com>
* gas/i386/i386.exp: Add prescott.
* gas/i386/prescott.d: New file.
* gas/i386/prescott.s: Likewise.
include/opcode/
2003-06-23 H.J. Lu <hongjiu.lu@intel.com>
* i386.h (i386_optab): Support Intel Precott New Instructions.
opcodes/
2003-06-23 H.J. Lu <hongjiu.lu@intel.com>
* i386-dis.c (PNI_Fixup): New. Fix up "mwait" and "monitor" in
Intel Precott New Instructions.
(PREGRP27): New. Added for "addsubpd" and "addsubps".
(PREGRP28): New. Added for "haddpd" and "haddps".
(PREGRP29): New. Added for "hsubpd" and "hsubps".
(PREGRP30): New. Added for "movsldup" and "movddup".
(PREGRP31): New. Added for "movshdup" and "movhpd".
(PREGRP32): New. Added for "lddqu".
(dis386_twobyte): Use PREGRP30 to replace the "movlpX" entry.
Use PREGRP31 to replace the "movhpX" entry. Use PREGRP28 for
entry 0x7c. Use PREGRP29 for entry 0x7d. Use PREGRP27 for
entry 0xd0. Use PREGRP32 for entry 0xf0.
(twobyte_has_modrm): Updated.
(twobyte_uses_SSE_prefix): Likewise.
(grps): Use PNI_Fixup in the "sidtQ" entry.
(prefix_user_table): Add PREGRP27, PREGRP28, PREGRP29, PREGRP30,
PREGRP31 and PREGRP32.
(float_mem): Use "fisttp{l||l|}" in entry 1 in opcode 0xdb.
Use "fisttpll" in entry 1 in opcode 0xdd.
Use "fisttp" in entry 1 in opcode 0xdf.
* arc.h (arc_ext_opcodes): Declare as extern.
(arc_ext_operands): Declare as extern.
* i860.h (i860_opcodes): Declare as const.
2002-11-18 Klee Dienes <kdienes@apple.com>
* arc-opc.c (arc_ext_opcodes): Define.
(arc_ext_operands): Define.
* i386-dis.c (Suffix3DNow): Declare as const.
* arm-opc.h (arm_opcodes): Declare as const.
(thumb_opcodes): Declare as const.
* h8500-opc.h (h8500_table): Declare as const.
(h8500_table): Use a NULL for the opcode in the terminator, so
that code testing (opcode->name) behaves correctly.
* mcore-opc.h (mcore_table): Declare as const.
* sh-opc.h (sh_table): Declare as const.
* w65-opc.h (optable): Declare as const.
* z8k-opc.h (z8k_table): Declare as const.
* doc/binutils.texi (objdump): Document x86 -M options.
include/ChangeLog
* dis-asm.h (print_insn_i386): Declare.
opcodes/ChangeLog
* disassemble.c (disassembler): Call print_insn_i386.
* i386-dis.c (SUFFIX_ALWAYS): Define.
(struct dis_private): Add orig_sizeflag.
(print_insn_i386): Make it a wrapper, calling..
(print_insn): ..The old body of print_insn_i386. Avoid longjmp
warning without using volatile by moving orig_sizeflag to priv,
and removing inbuf. Parse disassembler_options.
(print_insn_i386_att, print_insn_i386_intel): Move initialisation
code to print_insn.
(putop): Remove #ifdef SUFFIX_ALWAYS.
accept WordReg.
* i386-dis.c (grps): Change "sldt", "str", and "smsw" entries
to "sldtQ", "strQ", "smswQ" respectively; all with Ev operand
category instead of Ew.
* i386-dis.c (set_op): Handle 64 bit and 32 bit mode.
(OP_J): Use bfd_vma for mask to work properly with 64 bits.
(op_address,op_riprel): Use bfd_vma to handle 64 bits.
(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.
(generic_strcat_address): Add cast to avoid warning.
* i386-dis.c: Initialize all structure fields to avoid warnings.
Add ATTRIBUTE_UNUSED as appropriate.
* i386-dis.c (FWAIT_OPCODE): Define.
(used_prefixes): New static variable.
(fetch_data): Don't print an error message if we have already
fetched some bytes successfully.
(ckprefix): Clear used_prefixes. Use FWAIT_OPCODE, not 0x9b.
(prefix_name): New static function.
(print_insn_i386): If setjmp fails, indicating a data error, but
we have managed to fetch some bytes, print the first one as a
prefix or a .byte pseudo-op. If fwait is followed by a non
floating point instruction, print the first prefix. Set
used_prefixes when prefixes are used. If any prefixes were not
used after disassembling the instruction, print the first prefix
instead of printing the instruction.
(putop): Set used_prefixes when prefixes are used.
(append_seg, OP_E, OP_G, OP_REG, OP_I, OP_sI, OP_J): Likewise.
(OP_DIR, OP_SIMD_Suffix): Likewise.
(fetch_data): Don't print an error message if we have already
found an fwait prefix.
(ckprefix): Use FWAIT_OPCODE rather than 0x9b.
(print_insn_i386): If setjmp fails, indicating a data error, but
we have found an fwait prefix, then print it as an fwait
instruction. In any case, return the number of bytes consumed, if
any, rather than always returning -1.
Add missing files from last change in gas/testsuite.
i386-dis.c: Catch a number of cases where an invalid modrm would result in
bogus disassembly. const lots of places.
* i386-dis.c: Don't print opcode suffix when we can figure out the
size (and gas can!) by register operands, or from the default
size.
(putop): Handle 'A', 'B', 'L', 'P', 'Q', 'R' macros. Rename 'C'
macro to 'E'.
(dis386, dis386_twobyte, grps): Use new suffix macros.
(dis386): Correct imul Ib to imul sIb. Change jnl to jge to be
consistent. Add suffix for call, jmp, lcall, ljmp, iret. Reverse
order of cmps operands to agree with Intel docs. Correct operand
of aad and aam (Ib -> sIb). Change ud2b from 0fb8 to 0fb9 to
agree with Intel docs.
(print_insn_x86): Print orphan fwait before other prefixes.
Return correct byte count for orphan fwait with prefixes. Don't
print `bound' operands in reverse order.
(ckprefix): Stop accumulating prefixes if we get fwait.
(OP_DIR): Print `$' before Ap operands of ljmp, lcall.
Fix problems when bfd_vma is wider than long.
* i386-dis.c: Make op_address and start_pc unsigned.
(set_op): Make parameter unsigned.
(print_insn_x86): Cast to bfd_vma when passing a value to
print_address_func.
* ns32k-dis.c (CORE_ADDR): Don't define.
(print_insn_ns32k): Change type of addr to bfd_vma. Use
bfd_scan_vma to read back address.
(print_insn_arg): Change type of addr to bfd_vma. Use sprintf_vma
to format it.
* m68k-dis.c (COERCE32): Cast to bfd_signed_vma to avoid overflow.
(NEXTULONG): New definition.
(print_insn_m68k): Avoid overflow when computing third argument of
print_insn_arg.
(print_insn_arg): Use NEXTULONG to fetch 32 bit address values.
Use disp instead of val to store offset values.
(print_indexed): Use base_disp instead of word to store base
displacement, to avoid overflow.
* m10300-dis.c (disassemble): Cast value to long when computing
pc-relative address, to get correct sign extension.
* i386-dis.c: Combine aflag and dflag into sizeflag. Change OP_*
functions to void.
(OP_DSreg): Rename from OP_DSSI.
(OP_ESreg): Rename from OP_ESDI.
(Xb, Xv, Yb, Yv): Use index reg code, not b_mode or v_mode.
(DSBX): Define.
(append_seg): Rename from append_prefix.
(ptr_reg): New function.
(dis386): Add S suffix to pushf, popf, ret, lret, enter, leave.
Add DSBX for xlat.
(PREFIX_ADDR): Rename from PREFIX_ADR.
(float_reg): Add non-broken opcodes for people who don't want
UNIXWARE_COMPAT.
the 8086 instruction set.
* i386-dis.c: General cleanups. Make most things static. Add
prototypes. Get rid of static variables aflags and dflags. Pass
them as args (to almost everything).
the 8086 instruction set.
* i386-dis.c: General cleanups. Make most things static. Add
prototypes. Get rid of static variables aflags and dflags. Pass
them as args (to almost everything).
Andreas Jaeger
H.J. Lu
Alan Modra
Doug Evans
Klee Dienes
Jan Hubicka
Kazu Hirata
Nick Clifton
Ian Lance Taylor
Richard Henderson
Stan Shebs
Stu Grossman