;;;; that part of the description of the x86-64 instruction set ;;;; which can live on the cross-compilation host ;;;; This software is part of the SBCL system. See the README file for ;;;; more information. ;;;; ;;;; This software is derived from the CMU CL system, which was ;;;; written at Carnegie Mellon University and released into the ;;;; public domain. The software is in the public domain and is ;;;; provided with absolutely no warranty. See the COPYING and CREDITS ;;;; files for more information. (in-package "SB!VM") ;;; FIXME: SB!DISASSEM: prefixes are used so widely in this file that ;;; I wonder whether the separation of the disassembler from the ;;; virtual machine is valid or adds value. ;;; Note: In CMU CL, this used to be a call to SET-DISASSEM-PARAMS. (setf sb!disassem:*disassem-inst-alignment-bytes* 1) ;;; This type is used mostly in disassembly and represents legacy ;;; registers only. R8-R15 are handled separately. (deftype reg () '(unsigned-byte 3)) ;;; This includes legacy registers and R8-R15. (deftype full-reg () '(unsigned-byte 4)) ;;; The XMM registers XMM0 - XMM15. (deftype xmmreg () '(unsigned-byte 4)) ;;; Default word size for the chip: if the operand size /= :dword ;;; we need to output #x66 (or REX) prefix (def!constant +default-operand-size+ :dword) ;;; The default address size for the chip. It could be overwritten ;;; to :dword with a #x67 prefix, but this is never needed by SBCL ;;; and thus not supported by this assembler/disassembler. (def!constant +default-address-size+ :qword) (eval-when (#-sb-xc :compile-toplevel :load-toplevel :execute) (defun offset-next (value dstate) (declare (type integer value) (type sb!disassem:disassem-state dstate)) (+ (sb!disassem:dstate-next-addr dstate) value)) (defparameter *byte-reg-names* #(al cl dl bl spl bpl sil dil r8b r9b r10b r11b r12b r13b r14b r15b)) (defparameter *high-byte-reg-names* #(ah ch dh bh)) (defparameter *word-reg-names* #(ax cx dx bx sp bp si di r8w r9w r10w r11w r12w r13w r14w r15w)) (defparameter *dword-reg-names* #(eax ecx edx ebx esp ebp esi edi r8d r9d r10d r11d r12d r13d r14d r15d)) (defparameter *qword-reg-names* #(rax rcx rdx rbx rsp rbp rsi rdi r8 r9 r10 r11 r12 r13 r14 r15)) ;;; The printers for registers, memory references and immediates need to ;;; take into account the width bit in the instruction, whether a #x66 ;;; or a REX prefix was issued, and the contents of the REX prefix. ;;; This is implemented using prefilters to put flags into the slot ;;; INST-PROPERTIES of the DSTATE. These flags are the following ;;; symbols: ;;; ;;; OPERAND-SIZE-8 The width bit was zero ;;; OPERAND-SIZE-16 The "operand size override" prefix (#x66) was found ;;; REX A REX prefix was found ;;; REX-W A REX prefix with the "operand width" bit set was ;;; found ;;; REX-R A REX prefix with the "register" bit set was found ;;; REX-X A REX prefix with the "index" bit set was found ;;; REX-B A REX prefix with the "base" bit set was found ;;; Return the operand size depending on the prefixes and width bit as ;;; stored in DSTATE. (defun inst-operand-size (dstate) (declare (type sb!disassem:disassem-state dstate)) (cond ((sb!disassem:dstate-get-inst-prop dstate 'operand-size-8) :byte) ((sb!disassem:dstate-get-inst-prop dstate 'rex-w) :qword) ((sb!disassem:dstate-get-inst-prop dstate 'operand-size-16) :word) (t +default-operand-size+))) ;;; The same as INST-OPERAND-SIZE, but for those instructions (e.g. ;;; PUSH, JMP) that have a default operand size of :qword. It can only ;;; be overwritten to :word. (defun inst-operand-size-default-qword (dstate) (declare (type sb!disassem:disassem-state dstate)) (if (sb!disassem:dstate-get-inst-prop dstate 'operand-size-16) :word :qword)) ;;; Print to STREAM the name of the general-purpose register encoded by ;;; VALUE and of size WIDTH. For robustness, the high byte registers ;;; (AH, BH, CH, DH) are correctly detected, too, although the compiler ;;; does not use them. (defun print-reg-with-width (value width stream dstate) (declare (type full-reg value) (type stream stream) (type sb!disassem:disassem-state dstate)) (princ (if (and (eq width :byte) (<= 4 value 7) (not (sb!disassem:dstate-get-inst-prop dstate 'rex))) (aref *high-byte-reg-names* (- value 4)) (aref (ecase width (:byte *byte-reg-names*) (:word *word-reg-names*) (:dword *dword-reg-names*) (:qword *qword-reg-names*)) value)) stream) ;; XXX plus should do some source-var notes ) (defun print-reg (value stream dstate) (declare (type full-reg value) (type stream stream) (type sb!disassem:disassem-state dstate)) (print-reg-with-width value (inst-operand-size dstate) stream dstate)) (defun print-reg-default-qword (value stream dstate) (declare (type full-reg value) (type stream stream) (type sb!disassem:disassem-state dstate)) (print-reg-with-width value (inst-operand-size-default-qword dstate) stream dstate)) (defun print-byte-reg (value stream dstate) (declare (type full-reg value) (type stream stream) (type sb!disassem:disassem-state dstate)) (print-reg-with-width value :byte stream dstate)) (defun print-addr-reg (value stream dstate) (declare (type full-reg value) (type stream stream) (type sb!disassem:disassem-state dstate)) (print-reg-with-width value +default-address-size+ stream dstate)) ;;; Print a register or a memory reference of the given WIDTH. ;;; If SIZED-P is true, add an explicit size indicator for memory ;;; references. (defun print-reg/mem-with-width (value width sized-p stream dstate) (declare (type (or list full-reg) value) (type (member :byte :word :dword :qword) width) (type boolean sized-p) (type stream stream) (type sb!disassem:disassem-state dstate)) (if (typep value 'full-reg) (print-reg-with-width value width stream dstate) (print-mem-access value (and sized-p width) stream dstate))) ;;; Print a register or a memory reference. The width is determined by ;;; calling INST-OPERAND-SIZE. (defun print-reg/mem (value stream dstate) (declare (type (or list full-reg) value) (type stream stream) (type sb!disassem:disassem-state dstate)) (print-reg/mem-with-width value (inst-operand-size dstate) nil stream dstate)) ;; Same as print-reg/mem, but prints an explicit size indicator for ;; memory references. (defun print-sized-reg/mem (value stream dstate) (declare (type (or list full-reg) value) (type stream stream) (type sb!disassem:disassem-state dstate)) (print-reg/mem-with-width value (inst-operand-size dstate) t stream dstate)) ;;; Same as print-sized-reg/mem, but with a default operand size of ;;; :qword. (defun print-sized-reg/mem-default-qword (value stream dstate) (declare (type (or list full-reg) value) (type stream stream) (type sb!disassem:disassem-state dstate)) (print-reg/mem-with-width value (inst-operand-size-default-qword dstate) t stream dstate)) (defun print-sized-byte-reg/mem (value stream dstate) (declare (type (or list full-reg) value) (type stream stream) (type sb!disassem:disassem-state dstate)) (print-reg/mem-with-width value :byte t stream dstate)) (defun print-sized-word-reg/mem (value stream dstate) (declare (type (or list full-reg) value) (type stream stream) (type sb!disassem:disassem-state dstate)) (print-reg/mem-with-width value :word t stream dstate)) (defun print-sized-dword-reg/mem (value stream dstate) (declare (type (or list full-reg) value) (type stream stream) (type sb!disassem:disassem-state dstate)) (print-reg/mem-with-width value :dword t stream dstate)) (defun print-label (value stream dstate) (declare (ignore dstate)) (sb!disassem:princ16 value stream)) (defun print-xmmreg (value stream dstate) (declare (type xmmreg value) (type stream stream) (ignore dstate)) (format stream "XMM~d" value)) (defun print-xmmreg/mem (value stream dstate) (declare (type (or list xmmreg) value) (type stream stream) (type sb!disassem:disassem-state dstate)) (if (typep value 'xmmreg) (print-xmmreg value stream dstate) (print-mem-access value nil stream dstate))) ;; Same as print-xmmreg/mem, but prints an explicit size indicator for ;; memory references. (defun print-sized-xmmreg/mem (value stream dstate) (declare (type (or list xmmreg) value) (type stream stream) (type sb!disassem:disassem-state dstate)) (if (typep value 'xmmreg) (print-xmmreg value stream dstate) (print-mem-access value (inst-operand-size dstate) stream dstate))) ;;; This prefilter is used solely for its side effects, namely to put ;;; the bits found in the REX prefix into the DSTATE for use by other ;;; prefilters and by printers. (defun prefilter-wrxb (value dstate) (declare (type (unsigned-byte 4) value) (type sb!disassem:disassem-state dstate)) (sb!disassem:dstate-put-inst-prop dstate 'rex) (when (plusp (logand value #b1000)) (sb!disassem:dstate-put-inst-prop dstate 'rex-w)) (when (plusp (logand value #b0100)) (sb!disassem:dstate-put-inst-prop dstate 'rex-r)) (when (plusp (logand value #b0010)) (sb!disassem:dstate-put-inst-prop dstate 'rex-x)) (when (plusp (logand value #b0001)) (sb!disassem:dstate-put-inst-prop dstate 'rex-b)) value) ;;; This prefilter is used solely for its side effect, namely to put ;;; the property OPERAND-SIZE-8 into the DSTATE if VALUE is 0. (defun prefilter-width (value dstate) (declare (type bit value) (type sb!disassem:disassem-state dstate)) (when (zerop value) (sb!disassem:dstate-put-inst-prop dstate 'operand-size-8)) value) ;;; This prefilter is used solely for its side effect, namely to put ;;; the property OPERAND-SIZE-16 into the DSTATE. (defun prefilter-x66 (value dstate) (declare (type (eql #x66) value) (ignore value) (type sb!disassem:disassem-state dstate)) (sb!disassem:dstate-put-inst-prop dstate 'operand-size-16)) ;;; A register field that can be extended by REX.R. (defun prefilter-reg-r (value dstate) (declare (type reg value) (type sb!disassem:disassem-state dstate)) (if (sb!disassem::dstate-get-inst-prop dstate 'rex-r) (+ value 8) value)) ;;; A register field that can be extended by REX.B. (defun prefilter-reg-b (value dstate) (declare (type reg value) (type sb!disassem:disassem-state dstate)) (if (sb!disassem::dstate-get-inst-prop dstate 'rex-b) (+ value 8) value)) ;;; Returns either an integer, meaning a register, or a list of ;;; (BASE-REG OFFSET INDEX-REG INDEX-SCALE), where any component ;;; may be missing or nil to indicate that it's not used or has the ;;; obvious default value (e.g., 1 for the index-scale). VALUE is a list ;;; of the mod and r/m field of the ModRM byte of the instruction. ;;; Depending on VALUE a SIB byte and/or an offset may be read. The ;;; REX.B bit from DSTATE is used to extend the sole register or the ;;; BASE-REG to a full register, the REX.X bit does the same for the ;;; INDEX-REG. (defun prefilter-reg/mem (value dstate) (declare (type list value) (type sb!disassem:disassem-state dstate)) (let ((mod (first value)) (r/m (second value))) (declare (type (unsigned-byte 2) mod) (type (unsigned-byte 3) r/m)) (let ((full-reg (if (sb!disassem:dstate-get-inst-prop dstate 'rex-b) (+ r/m 8) r/m))) (declare (type full-reg full-reg)) (cond ((= mod #b11) ;; registers full-reg) ((= r/m #b100) ;; sib byte (let ((sib (sb!disassem:read-suffix 8 dstate))) (declare (type (unsigned-byte 8) sib)) (let ((base-reg (ldb (byte 3 0) sib)) (index-reg (ldb (byte 3 3) sib)) (index-scale (ldb (byte 2 6) sib))) (declare (type (unsigned-byte 3) base-reg index-reg) (type (unsigned-byte 2) index-scale)) (let* ((offset (case mod (#b00 (if (= base-reg #b101) (sb!disassem:read-signed-suffix 32 dstate) nil)) (#b01 (sb!disassem:read-signed-suffix 8 dstate)) (#b10 (sb!disassem:read-signed-suffix 32 dstate))))) (list (unless (and (= mod #b00) (= base-reg #b101)) (if (sb!disassem:dstate-get-inst-prop dstate 'rex-b) (+ base-reg 8) base-reg)) offset (unless (= index-reg #b100) (if (sb!disassem:dstate-get-inst-prop dstate 'rex-x) (+ index-reg 8) index-reg)) (ash 1 index-scale)))))) ((and (= mod #b00) (= r/m #b101)) (list 'rip (sb!disassem:read-signed-suffix 32 dstate))) ((= mod #b00) (list full-reg)) ((= mod #b01) (list full-reg (sb!disassem:read-signed-suffix 8 dstate))) (t ; (= mod #b10) (list full-reg (sb!disassem:read-signed-suffix 32 dstate))))))) (defun read-address (value dstate) (declare (ignore value)) ; always nil anyway (sb!disassem:read-suffix (width-bits (inst-operand-size dstate)) dstate)) (defun width-bits (width) (ecase width (:byte 8) (:word 16) (:dword 32) (:qword 64))) ) ; EVAL-WHEN ;;;; disassembler argument types ;;; Used to capture the lower four bits of the REX prefix. (sb!disassem:define-arg-type wrxb :prefilter #'prefilter-wrxb) (sb!disassem:define-arg-type width :prefilter #'prefilter-width :printer (lambda (value stream dstate) (declare (ignore value)) (princ (schar (symbol-name (inst-operand-size dstate)) 0) stream))) ;;; Used to capture the effect of the #x66 operand size override prefix. (sb!disassem:define-arg-type x66 :prefilter #'prefilter-x66) (sb!disassem:define-arg-type displacement :sign-extend t :use-label #'offset-next :printer (lambda (value stream dstate) (sb!disassem:maybe-note-assembler-routine value nil dstate) (print-label value stream dstate))) (sb!disassem:define-arg-type accum :printer (lambda (value stream dstate) (declare (ignore value) (type stream stream) (type sb!disassem:disassem-state dstate)) (print-reg 0 stream dstate))) (sb!disassem:define-arg-type reg :prefilter #'prefilter-reg-r :printer #'print-reg) (sb!disassem:define-arg-type reg-b :prefilter #'prefilter-reg-b :printer #'print-reg) (sb!disassem:define-arg-type reg-b-default-qword :prefilter #'prefilter-reg-b :printer #'print-reg-default-qword) (sb!disassem:define-arg-type imm-addr :prefilter #'read-address :printer #'print-label) ;;; Normally, immediate values for an operand size of :qword are of size ;;; :dword and are sign-extended to 64 bits. For an exception, see the ;;; argument type definition following this one. (sb!disassem:define-arg-type signed-imm-data :prefilter (lambda (value dstate) (declare (ignore value)) ; always nil anyway (let ((width (width-bits (inst-operand-size dstate)))) (when (= width 64) (setf width 32)) (sb!disassem:read-signed-suffix width dstate)))) ;;; Used by the variant of the MOV instruction with opcode B8 which can ;;; move immediates of all sizes (i.e. including :qword) into a ;;; register. (sb!disassem:define-arg-type signed-imm-data-upto-qword :prefilter (lambda (value dstate) (declare (ignore value)) ; always nil anyway (sb!disassem:read-signed-suffix (width-bits (inst-operand-size dstate)) dstate))) ;;; Used by those instructions that have a default operand size of ;;; :qword. Nevertheless the immediate is at most of size :dword. ;;; The only instruction of this kind having a variant with an immediate ;;; argument is PUSH. (sb!disassem:define-arg-type signed-imm-data-default-qword :prefilter (lambda (value dstate) (declare (ignore value)) ; always nil anyway (let ((width (width-bits (inst-operand-size-default-qword dstate)))) (when (= width 64) (setf width 32)) (sb!disassem:read-signed-suffix width dstate)))) (sb!disassem:define-arg-type signed-imm-byte :prefilter (lambda (value dstate) (declare (ignore value)) ; always nil anyway (sb!disassem:read-signed-suffix 8 dstate))) (sb!disassem:define-arg-type imm-byte :prefilter (lambda (value dstate) (declare (ignore value)) ; always nil anyway (sb!disassem:read-suffix 8 dstate))) ;;; needed for the ret imm16 instruction (sb!disassem:define-arg-type imm-word-16 :prefilter (lambda (value dstate) (declare (ignore value)) ; always nil anyway (sb!disassem:read-suffix 16 dstate))) (sb!disassem:define-arg-type reg/mem :prefilter #'prefilter-reg/mem :printer #'print-reg/mem) (sb!disassem:define-arg-type sized-reg/mem ;; Same as reg/mem, but prints an explicit size indicator for ;; memory references. :prefilter #'prefilter-reg/mem :printer #'print-sized-reg/mem) ;;; Arguments of type reg/mem with a fixed size. (sb!disassem:define-arg-type sized-byte-reg/mem :prefilter #'prefilter-reg/mem :printer #'print-sized-byte-reg/mem) (sb!disassem:define-arg-type sized-word-reg/mem :prefilter #'prefilter-reg/mem :printer #'print-sized-word-reg/mem) (sb!disassem:define-arg-type sized-dword-reg/mem :prefilter #'prefilter-reg/mem :printer #'print-sized-dword-reg/mem) ;;; Same as sized-reg/mem, but with a default operand size of :qword. (sb!disassem:define-arg-type sized-reg/mem-default-qword :prefilter #'prefilter-reg/mem :printer #'print-sized-reg/mem-default-qword) ;;; XMM registers (sb!disassem:define-arg-type xmmreg :prefilter #'prefilter-reg-r :printer #'print-xmmreg) (sb!disassem:define-arg-type xmmreg/mem :prefilter #'prefilter-reg/mem :printer #'print-xmmreg/mem) (sb!disassem:define-arg-type sized-xmmreg/mem :prefilter #'prefilter-reg/mem :printer #'print-sized-xmmreg/mem) (eval-when (:compile-toplevel :load-toplevel :execute) (defparameter *conditions* '((:o . 0) (:no . 1) (:b . 2) (:nae . 2) (:c . 2) (:nb . 3) (:ae . 3) (:nc . 3) (:eq . 4) (:e . 4) (:z . 4) (:ne . 5) (:nz . 5) (:be . 6) (:na . 6) (:nbe . 7) (:a . 7) (:s . 8) (:ns . 9) (:p . 10) (:pe . 10) (:np . 11) (:po . 11) (:l . 12) (:nge . 12) (:nl . 13) (:ge . 13) (:le . 14) (:ng . 14) (:nle . 15) (:g . 15))) (defparameter *condition-name-vec* (let ((vec (make-array 16 :initial-element nil))) (dolist (cond *conditions*) (when (null (aref vec (cdr cond))) (setf (aref vec (cdr cond)) (car cond)))) vec)) ) ; EVAL-WHEN ;;; Set assembler parameters. (In CMU CL, this was done with ;;; a call to a macro DEF-ASSEMBLER-PARAMS.) (eval-when (:compile-toplevel :load-toplevel :execute) (setf sb!assem:*assem-scheduler-p* nil)) (sb!disassem:define-arg-type condition-code :printer *condition-name-vec*) (defun conditional-opcode (condition) (cdr (assoc condition *conditions* :test #'eq))) ;;;; disassembler instruction formats (eval-when (:compile-toplevel :execute) (defun swap-if (direction field1 separator field2) `(:if (,direction :constant 0) (,field1 ,separator ,field2) (,field2 ,separator ,field1)))) (sb!disassem:define-instruction-format (byte 8 :default-printer '(:name)) (op :field (byte 8 0)) ;; optional fields (accum :type 'accum) (imm)) ;;; A one-byte instruction with a #x66 prefix, used to indicate an ;;; operand size of :word. (sb!disassem:define-instruction-format (x66-byte 16 :default-printer '(:name)) (x66 :field (byte 8 0) :value #x66) (op :field (byte 8 8))) ;;; A one-byte instruction with a REX prefix, used to indicate an ;;; operand size of :qword. REX.W must be 1, the other three bits are ;;; ignored. (sb!disassem:define-instruction-format (rex-byte 16 :default-printer '(:name)) (rex :field (byte 5 3) :value #b01001) (op :field (byte 8 8))) (sb!disassem:define-instruction-format (simple 8) (op :field (byte 7 1)) (width :field (byte 1 0) :type 'width) ;; optional fields (accum :type 'accum) (imm)) (sb!disassem:define-instruction-format (rex-simple 16) (rex :field (byte 4 4) :value #b0100) (wrxb :field (byte 4 0) :type 'wrxb) (op :field (byte 7 9)) (width :field (byte 1 8) :type 'width) ;; optional fields (accum :type 'accum) (imm)) ;;; Same as simple, but with direction bit (sb!disassem:define-instruction-format (simple-dir 8 :include 'simple) (op :field (byte 6 2)) (dir :field (byte 1 1))) ;;; Same as simple, but with the immediate value occurring by default, ;;; and with an appropiate printer. (sb!disassem:define-instruction-format (accum-imm 8 :include 'simple :default-printer '(:name :tab accum ", " imm)) (imm :type 'signed-imm-data)) (sb!disassem:define-instruction-format (rex-accum-imm 16 :include 'rex-simple :default-printer '(:name :tab accum ", " imm)) (imm :type 'signed-imm-data)) (sb!disassem:define-instruction-format (reg-no-width 8 :default-printer '(:name :tab reg)) (op :field (byte 5 3)) (reg :field (byte 3 0) :type 'reg-b) ;; optional fields (accum :type 'accum) (imm)) (sb!disassem:define-instruction-format (rex-reg-no-width 16 :default-printer '(:name :tab reg)) (rex :field (byte 4 4) :value #b0100) (wrxb :field (byte 4 0) :type 'wrxb) (op :field (byte 5 11)) (reg :field (byte 3 8) :type 'reg-b) ;; optional fields (accum :type 'accum) (imm)) ;;; Same as reg-no-width, but with a default operand size of :qword. (sb!disassem:define-instruction-format (reg-no-width-default-qword 8 :include 'reg-no-width :default-printer '(:name :tab reg)) (reg :type 'reg-b-default-qword)) ;;; Same as rex-reg-no-width, but with a default operand size of :qword. (sb!disassem:define-instruction-format (rex-reg-no-width-default-qword 16 :include 'rex-reg-no-width :default-printer '(:name :tab reg)) (reg :type 'reg-b-default-qword)) ;;; Adds a width field to reg-no-width. Note that we can't use ;;; :INCLUDE 'REG-NO-WIDTH here to save typing because that would put ;;; the WIDTH field last, but the prefilter for WIDTH must run before ;;; the one for IMM to be able to determine the correct size of IMM. (sb!disassem:define-instruction-format (reg 8 :default-printer '(:name :tab reg)) (op :field (byte 4 4)) (width :field (byte 1 3) :type 'width) (reg :field (byte 3 0) :type 'reg-b) ;; optional fields (accum :type 'accum) (imm)) (sb!disassem:define-instruction-format (rex-reg 16 :default-printer '(:name :tab reg)) (rex :field (byte 4 4) :value #b0100) (wrxb :field (byte 4 0) :type 'wrxb) (width :field (byte 1 11) :type 'width) (op :field (byte 4 12)) (reg :field (byte 3 8) :type 'reg-b) ;; optional fields (accum :type 'accum) (imm)) (sb!disassem:define-instruction-format (two-bytes 16 :default-printer '(:name)) (op :fields (list (byte 8 0) (byte 8 8)))) (sb!disassem:define-instruction-format (reg-reg/mem 16 :default-printer `(:name :tab reg ", " reg/mem)) (op :field (byte 7 1)) (width :field (byte 1 0) :type 'width) (reg/mem :fields (list (byte 2 14) (byte 3 8)) :type 'reg/mem) (reg :field (byte 3 11) :type 'reg) ;; optional fields (imm)) (sb!disassem:define-instruction-format (rex-reg-reg/mem 24 :default-printer `(:name :tab reg ", " reg/mem)) (rex :field (byte 4 4) :value #b0100) (wrxb :field (byte 4 0) :type 'wrxb) (width :field (byte 1 8) :type 'width) (op :field (byte 7 9)) (reg/mem :fields (list (byte 2 22) (byte 3 16)) :type 'reg/mem) (reg :field (byte 3 19) :type 'reg) ;; optional fields (imm)) ;;; same as reg-reg/mem, but with direction bit (sb!disassem:define-instruction-format (reg-reg/mem-dir 16 :include 'reg-reg/mem :default-printer `(:name :tab ,(swap-if 'dir 'reg/mem ", " 'reg))) (op :field (byte 6 2)) (dir :field (byte 1 1))) (sb!disassem:define-instruction-format (rex-reg-reg/mem-dir 24 :include 'rex-reg-reg/mem :default-printer `(:name :tab ,(swap-if 'dir 'reg/mem ", " 'reg))) (op :field (byte 6 10)) (dir :field (byte 1 9))) (sb!disassem:define-instruction-format (x66-reg-reg/mem-dir 24 :default-printer `(:name :tab ,(swap-if 'dir 'reg/mem ", " 'reg))) (x66 :field (byte 8 0) :type 'x66 :value #x66) (op :field (byte 6 10)) (dir :field (byte 1 9)) (width :field (byte 1 8) :type 'width) (reg/mem :fields (list (byte 2 22) (byte 3 16)) :type 'reg/mem) (reg :field (byte 3 19) :type 'reg)) (sb!disassem:define-instruction-format (x66-rex-reg-reg/mem-dir 32 :default-printer `(:name :tab ,(swap-if 'dir 'reg/mem ", " 'reg))) (x66 :field (byte 8 0) :type 'x66 :value #x66) (rex :field (byte 4 12) :value #b0100) (wrxb :field (byte 4 8) :type 'wrxb) (op :field (byte 6 18)) (dir :field (byte 1 17)) (width :field (byte 1 16) :type 'width) (reg/mem :fields (list (byte 2 30) (byte 3 24)) :type 'reg/mem) (reg :field (byte 3 27) :type 'reg)) ;;; Same as reg-reg/mem, but uses the reg field as a second op code. (sb!disassem:define-instruction-format (reg/mem 16 :default-printer '(:name :tab reg/mem)) (op :fields (list (byte 7 1) (byte 3 11))) (width :field (byte 1 0) :type 'width) (reg/mem :fields (list (byte 2 14) (byte 3 8)) :type 'sized-reg/mem) ;; optional fields (imm)) (sb!disassem:define-instruction-format (rex-reg/mem 24 :default-printer '(:name :tab reg/mem)) (rex :field (byte 4 4) :value #b0100) (wrxb :field (byte 4 0) :type 'wrxb) (op :fields (list (byte 7 9) (byte 3 19))) (width :field (byte 1 8) :type 'width) (reg/mem :fields (list (byte 2 22) (byte 3 16)) :type 'sized-reg/mem) ;; optional fields (imm)) ;;; Same as reg/mem, but without a width field and with a default ;;; operand size of :qword. (sb!disassem:define-instruction-format (reg/mem-default-qword 16 :default-printer '(:name :tab reg/mem)) (op :fields (list (byte 8 0) (byte 3 11))) (reg/mem :fields (list (byte 2 14) (byte 3 8)) :type 'sized-reg/mem-default-qword)) (sb!disassem:define-instruction-format (rex-reg/mem-default-qword 24 :default-printer '(:name :tab reg/mem)) (rex :field (byte 4 4) :value #b0100) (wrxb :field (byte 4 0) :type 'wrxb) (op :fields (list (byte 8 8) (byte 3 19))) (reg/mem :fields (list (byte 2 22) (byte 3 16)) :type 'sized-reg/mem-default-qword)) ;;; Same as reg/mem, but with the immediate value occurring by default, ;;; and with an appropiate printer. (sb!disassem:define-instruction-format (reg/mem-imm 16 :include 'reg/mem :default-printer '(:name :tab reg/mem ", " imm)) (reg/mem :type 'sized-reg/mem) (imm :type 'signed-imm-data)) (sb!disassem:define-instruction-format (rex-reg/mem-imm 24 :include 'rex-reg/mem :default-printer '(:name :tab reg/mem ", " imm)) (reg/mem :type 'sized-reg/mem) (imm :type 'signed-imm-data)) ;;; Same as reg/mem, but with using the accumulator in the default printer (sb!disassem:define-instruction-format (accum-reg/mem 16 :include 'reg/mem :default-printer '(:name :tab accum ", " reg/mem)) (reg/mem :type 'reg/mem) ; don't need a size (accum :type 'accum)) (sb!disassem:define-instruction-format (rex-accum-reg/mem 24 :include 'rex-reg/mem :default-printer '(:name :tab accum ", " reg/mem)) (reg/mem :type 'reg/mem) ; don't need a size (accum :type 'accum)) ;;; Same as reg-reg/mem, but with a prefix of #b00001111 (sb!disassem:define-instruction-format (ext-reg-reg/mem 24 :default-printer `(:name :tab reg ", " reg/mem)) (prefix :field (byte 8 0) :value #b00001111) (op :field (byte 7 9)) (width :field (byte 1 8) :type 'width) (reg/mem :fields (list (byte 2 22) (byte 3 16)) :type 'reg/mem) (reg :field (byte 3 19) :type 'reg) ;; optional fields (imm)) (sb!disassem:define-instruction-format (ext-reg-reg/mem-no-width 24 :default-printer `(:name :tab reg ", " reg/mem)) (prefix :field (byte 8 0) :value #b00001111) (op :field (byte 8 8)) (reg/mem :fields (list (byte 2 22) (byte 3 16)) :type 'reg/mem) (reg :field (byte 3 19) :type 'reg)) (sb!disassem:define-instruction-format (rex-ext-reg-reg/mem-no-width 32 :default-printer `(:name :tab reg ", " reg/mem)) (rex :field (byte 4 4) :value #b0100) (wrxb :field (byte 4 0) :type 'wrxb) (prefix :field (byte 8 8) :value #b00001111) (op :field (byte 8 16)) (reg/mem :fields (list (byte 2 30) (byte 3 24)) :type 'reg/mem) (reg :field (byte 3 27) :type 'reg)) ;;; reg-no-width with #x0f prefix (sb!disassem:define-instruction-format (ext-reg-no-width 16 :default-printer '(:name :tab reg)) (prefix :field (byte 8 0) :value #b00001111) (op :field (byte 5 11)) (reg :field (byte 3 8) :type 'reg-b)) ;;; Same as reg/mem, but with a prefix of #b00001111 (sb!disassem:define-instruction-format (ext-reg/mem 24 :default-printer '(:name :tab reg/mem)) (prefix :field (byte 8 0) :value #b00001111) (op :fields (list (byte 7 9) (byte 3 19))) (width :field (byte 1 8) :type 'width) (reg/mem :fields (list (byte 2 22) (byte 3 16)) :type 'sized-reg/mem) ;; optional fields (imm)) (sb!disassem:define-instruction-format (ext-reg/mem-imm 24 :include 'ext-reg/mem :default-printer '(:name :tab reg/mem ", " imm)) (imm :type 'signed-imm-data)) ;;;; XMM instructions ;;; All XMM instructions use an extended opcode (#x0F as the first ;;; opcode byte). Therefore in the following "EXT" in the name of the ;;; instruction formats refers to the formats that have an additional ;;; prefix (#x66, #xF2 or #xF3). ;;; Instructions having an XMM register as the destination operand ;;; and an XMM register or a memory location as the source operand. ;;; The size of the operands is implicitly given by the instruction. (sb!disassem:define-instruction-format (xmm-xmm/mem 24 :default-printer '(:name :tab reg ", " reg/mem)) (x0f :field (byte 8 0) :value #x0f) (op :field (byte 8 8)) (reg/mem :fields (list (byte 2 22) (byte 3 16)) :type 'xmmreg/mem) (reg :field (byte 3 19) :type 'xmmreg)) (sb!disassem:define-instruction-format (rex-xmm-xmm/mem 32 :default-printer '(:name :tab reg ", " reg/mem)) (x0f :field (byte 8 0) :value #x0f) (rex :field (byte 4 12) :value #b0100) (wrxb :field (byte 4 8) :type 'wrxb) (op :field (byte 8 16)) (reg/mem :fields (list (byte 2 30) (byte 3 24)) :type 'xmmreg/mem) (reg :field (byte 3 27) :type 'xmmreg)) (sb!disassem:define-instruction-format (ext-xmm-xmm/mem 32 :default-printer '(:name :tab reg ", " reg/mem)) (prefix :field (byte 8 0)) (x0f :field (byte 8 8) :value #x0f) (op :field (byte 8 16)) (reg/mem :fields (list (byte 2 30) (byte 3 24)) :type 'xmmreg/mem) (reg :field (byte 3 27) :type 'xmmreg)) (sb!disassem:define-instruction-format (ext-rex-xmm-xmm/mem 40 :default-printer '(:name :tab reg ", " reg/mem)) (prefix :field (byte 8 0)) (rex :field (byte 4 12) :value #b0100) (wrxb :field (byte 4 8) :type 'wrxb) (x0f :field (byte 8 16) :value #x0f) (op :field (byte 8 24)) (reg/mem :fields (list (byte 2 38) (byte 3 32)) :type 'xmmreg/mem) (reg :field (byte 3 35) :type 'xmmreg)) ;;; Same as xmm-xmm/mem etc., but with direction bit. (sb!disassem:define-instruction-format (ext-xmm-xmm/mem-dir 32 :include 'ext-xmm-xmm/mem :default-printer `(:name :tab ,(swap-if 'dir 'reg ", " 'reg/mem))) (op :field (byte 7 17)) (dir :field (byte 1 16))) (sb!disassem:define-instruction-format (ext-rex-xmm-xmm/mem-dir 40 :include 'ext-rex-xmm-xmm/mem :default-printer `(:name :tab ,(swap-if 'dir 'reg ", " 'reg/mem))) (op :field (byte 7 25)) (dir :field (byte 1 24))) ;;; Instructions having an XMM register as one operand and a general- ;;; -purpose register or a memory location as the other operand. (sb!disassem:define-instruction-format (ext-xmm-reg/mem 32 :default-printer '(:name :tab reg ", " reg/mem)) (prefix :field (byte 8 0)) (x0f :field (byte 8 8) :value #x0f) (op :field (byte 8 16)) (reg/mem :fields (list (byte 2 30) (byte 3 24)) :type 'sized-reg/mem) (reg :field (byte 3 27) :type 'xmmreg)) (sb!disassem:define-instruction-format (ext-rex-xmm-reg/mem 40 :default-printer '(:name :tab reg ", " reg/mem)) (prefix :field (byte 8 0)) (rex :field (byte 4 12) :value #b0100) (wrxb :field (byte 4 8) :type 'wrxb) (x0f :field (byte 8 16) :value #x0f) (op :field (byte 8 24)) (reg/mem :fields (list (byte 2 38) (byte 3 32)) :type 'sized-reg/mem) (reg :field (byte 3 35) :type 'xmmreg)) ;;; Instructions having a general-purpose register as one operand and an ;;; XMM register or a memory location as the other operand. (sb!disassem:define-instruction-format (ext-reg-xmm/mem 32 :default-printer '(:name :tab reg ", " reg/mem)) (prefix :field (byte 8 0)) (x0f :field (byte 8 8) :value #x0f) (op :field (byte 8 16)) (reg/mem :fields (list (byte 2 30) (byte 3 24)) :type 'sized-xmmreg/mem) (reg :field (byte 3 27) :type 'reg)) (sb!disassem:define-instruction-format (ext-rex-reg-xmm/mem 40 :default-printer '(:name :tab reg ", " reg/mem)) (prefix :field (byte 8 0)) (rex :field (byte 4 12) :value #b0100) (wrxb :field (byte 4 8) :type 'wrxb) (x0f :field (byte 8 16) :value #x0f) (op :field (byte 8 24)) (reg/mem :fields (list (byte 2 38) (byte 3 32)) :type 'sized-xmmreg/mem) (reg :field (byte 3 35) :type 'reg)) (sb!disassem:define-instruction-format (string-op 8 :include 'simple :default-printer '(:name width))) (sb!disassem:define-instruction-format (rex-string-op 16 :include 'rex-simple :default-printer '(:name width))) (sb!disassem:define-instruction-format (short-cond-jump 16) (op :field (byte 4 4)) (cc :field (byte 4 0) :type 'condition-code) (label :field (byte 8 8) :type 'displacement)) (sb!disassem:define-instruction-format (short-jump 16 :default-printer '(:name :tab label)) (const :field (byte 4 4) :value #b1110) (op :field (byte 4 0)) (label :field (byte 8 8) :type 'displacement)) (sb!disassem:define-instruction-format (near-cond-jump 16) (op :fields (list (byte 8 0) (byte 4 12)) :value '(#b00001111 #b1000)) (cc :field (byte 4 8) :type 'condition-code) ;; The disassembler currently doesn't let you have an instruction > 32 bits ;; long, so we fake it by using a prefilter to read the offset. (label :type 'displacement :prefilter (lambda (value dstate) (declare (ignore value)) ; always nil anyway (sb!disassem:read-signed-suffix 32 dstate)))) (sb!disassem:define-instruction-format (near-jump 8 :default-printer '(:name :tab label)) (op :field (byte 8 0)) ;; The disassembler currently doesn't let you have an instruction > 32 bits ;; long, so we fake it by using a prefilter to read the address. (label :type 'displacement :prefilter (lambda (value dstate) (declare (ignore value)) ; always nil anyway (sb!disassem:read-signed-suffix 32 dstate)))) (sb!disassem:define-instruction-format (cond-set 24 :default-printer '('set cc :tab reg/mem)) (prefix :field (byte 8 0) :value #b00001111) (op :field (byte 4 12) :value #b1001) (cc :field (byte 4 8) :type 'condition-code) (reg/mem :fields (list (byte 2 22) (byte 3 16)) :type 'sized-byte-reg/mem) (reg :field (byte 3 19) :value #b000)) (sb!disassem:define-instruction-format (cond-move 24 :default-printer '('cmov cc :tab reg ", " reg/mem)) (prefix :field (byte 8 0) :value #b00001111) (op :field (byte 4 12) :value #b0100) (cc :field (byte 4 8) :type 'condition-code) (reg/mem :fields (list (byte 2 22) (byte 3 16)) :type 'reg/mem) (reg :field (byte 3 19) :type 'reg)) (sb!disassem:define-instruction-format (rex-cond-move 32 :default-printer '('cmov cc :tab reg ", " reg/mem)) (rex :field (byte 4 4) :value #b0100) (wrxb :field (byte 4 0) :type 'wrxb) (prefix :field (byte 8 8) :value #b00001111) (op :field (byte 4 20) :value #b0100) (cc :field (byte 4 16) :type 'condition-code) (reg/mem :fields (list (byte 2 30) (byte 3 24)) :type 'reg/mem) (reg :field (byte 3 27) :type 'reg)) (sb!disassem:define-instruction-format (enter-format 32 :default-printer '(:name :tab disp (:unless (:constant 0) ", " level))) (op :field (byte 8 0)) (disp :field (byte 16 8)) (level :field (byte 8 24))) ;;; Single byte instruction with an immediate byte argument. (sb!disassem:define-instruction-format (byte-imm 16 :default-printer '(:name :tab code)) (op :field (byte 8 0)) (code :field (byte 8 8))) ;;; Two byte instruction with an immediate byte argument. ;;; (sb!disassem:define-instruction-format (word-imm 24 :default-printer '(:name :tab code)) (op :field (byte 16 0)) (code :field (byte 8 16))) ;;;; primitive emitters (define-bitfield-emitter emit-word 16 (byte 16 0)) (define-bitfield-emitter emit-dword 32 (byte 32 0)) ;;; Most uses of dwords are as displacements or as immediate values in ;;; 64-bit operations. In these cases they are sign-extended to 64 bits. ;;; EMIT-DWORD is unsuitable there because it accepts values of type ;;; (OR (SIGNED-BYTE 32) (UNSIGNED-BYTE 32)), so we provide a more ;;; restricted emitter here. (defun emit-signed-dword (segment value) (declare (type segment segment) (type (signed-byte 32) value)) (declare (inline emit-dword)) (emit-dword segment value)) (define-bitfield-emitter emit-qword 64 (byte 64 0)) (define-bitfield-emitter emit-byte-with-reg 8 (byte 5 3) (byte 3 0)) (define-bitfield-emitter emit-mod-reg-r/m-byte 8 (byte 2 6) (byte 3 3) (byte 3 0)) (define-bitfield-emitter emit-sib-byte 8 (byte 2 6) (byte 3 3) (byte 3 0)) (define-bitfield-emitter emit-rex-byte 8 (byte 4 4) (byte 1 3) (byte 1 2) (byte 1 1) (byte 1 0)) ;;;; fixup emitters (defun emit-absolute-fixup (segment fixup &optional quad-p) (note-fixup segment (if quad-p :absolute64 :absolute) fixup) (let ((offset (fixup-offset fixup))) (if (label-p offset) (emit-back-patch segment (if quad-p 8 4) (lambda (segment posn) (declare (ignore posn)) (let ((val (- (+ (component-header-length) (or (label-position offset) 0)) other-pointer-lowtag))) (if quad-p (emit-qword segment val) (emit-signed-dword segment val))))) (if quad-p (emit-qword segment (or offset 0)) (emit-signed-dword segment (or offset 0)))))) (defun emit-relative-fixup (segment fixup) (note-fixup segment :relative fixup) (emit-signed-dword segment (or (fixup-offset fixup) 0))) ;;;; the effective-address (ea) structure (defun reg-tn-encoding (tn) (declare (type tn tn)) ;; ea only has space for three bits of register number: regs r8 ;; and up are selected by a REX prefix byte which caller is responsible ;; for having emitted where necessary already (ecase (sb-name (sc-sb (tn-sc tn))) (registers (let ((offset (mod (tn-offset tn) 16))) (logior (ash (logand offset 1) 2) (ash offset -1)))) (float-registers (mod (tn-offset tn) 8)))) (defstruct (ea (:constructor make-ea (size &key base index scale disp)) (:copier nil)) ;; note that we can represent an EA with a QWORD size, but EMIT-EA ;; can't actually emit it on its own: caller also needs to emit REX ;; prefix (size nil :type (member :byte :word :dword :qword)) (base nil :type (or tn null)) (index nil :type (or tn null)) (scale 1 :type (member 1 2 4 8)) (disp 0 :type (or (unsigned-byte 32) (signed-byte 32) fixup))) (def!method print-object ((ea ea) stream) (cond ((or *print-escape* *print-readably*) (print-unreadable-object (ea stream :type t) (format stream "~S~@[ base=~S~]~@[ index=~S~]~@[ scale=~S~]~@[ disp=~S~]" (ea-size ea) (ea-base ea) (ea-index ea) (let ((scale (ea-scale ea))) (if (= scale 1) nil scale)) (ea-disp ea)))) (t (format stream "~A PTR [" (symbol-name (ea-size ea))) (when (ea-base ea) (write-string (sb!c::location-print-name (ea-base ea)) stream) (when (ea-index ea) (write-string "+" stream))) (when (ea-index ea) (write-string (sb!c::location-print-name (ea-index ea)) stream)) (unless (= (ea-scale ea) 1) (format stream "*~A" (ea-scale ea))) (typecase (ea-disp ea) (null) (integer (format stream "~@D" (ea-disp ea))) (t (format stream "+~A" (ea-disp ea)))) (write-char #\] stream)))) (defun emit-constant-tn-rip (segment constant-tn reg) ;; AMD64 doesn't currently have a code object register to use as a ;; base register for constant access. Instead we use RIP-relative ;; addressing. The offset from the SIMPLE-FUN-HEADER to the instruction ;; is passed to the backpatch callback. In addition we need the offset ;; from the start of the function header to the slot in the CODE-HEADER ;; that stores the constant. Since we don't know where the code header ;; starts, instead count backwards from the function header. (let* ((2comp (component-info *component-being-compiled*)) (constants (ir2-component-constants 2comp)) (len (length constants)) ;; Both CODE-HEADER and SIMPLE-FUN-HEADER are 16-byte aligned. ;; If there are an even amount of constants, there will be ;; an extra qword of padding before the function header, which ;; needs to be adjusted for. XXX: This will break if new slots ;; are added to the code header. (offset (* (- (+ len (if (evenp len) 1 2)) (tn-offset constant-tn)) n-word-bytes))) ;; RIP-relative addressing (emit-mod-reg-r/m-byte segment #b00 reg #b101) (emit-back-patch segment 4 (lambda (segment posn) ;; The addressing is relative to end of instruction, ;; i.e. the end of this dword. Hence the + 4. (emit-signed-dword segment (+ 4 (- (+ offset posn))))))) (values)) (defun emit-label-rip (segment fixup reg) (let ((label (fixup-offset fixup))) ;; RIP-relative addressing (emit-mod-reg-r/m-byte segment #b00 reg #b101) (emit-back-patch segment 4 (lambda (segment posn) (emit-signed-dword segment (- (label-position label) (+ posn 4)))))) (values)) (defun emit-ea (segment thing reg &optional allow-constants) (etypecase thing (tn ;; this would be eleganter if we had a function that would create ;; an ea given a tn (ecase (sb-name (sc-sb (tn-sc thing))) ((registers float-registers) (emit-mod-reg-r/m-byte segment #b11 reg (reg-tn-encoding thing))) (stack ;; Convert stack tns into an index off RBP. (let ((disp (- (* (1+ (tn-offset thing)) n-word-bytes)))) (cond ((<= -128 disp 127) (emit-mod-reg-r/m-byte segment #b01 reg #b101) (emit-byte segment disp)) (t (emit-mod-reg-r/m-byte segment #b10 reg #b101) (emit-signed-dword segment disp))))) (constant (unless allow-constants ;; Why? (error "Constant TNs can only be directly used in MOV, PUSH, and CMP.")) (emit-constant-tn-rip segment thing reg)))) (ea (let* ((base (ea-base thing)) (index (ea-index thing)) (scale (ea-scale thing)) (disp (ea-disp thing)) (mod (cond ((or (null base) (and (eql disp 0) (not (= (reg-tn-encoding base) #b101)))) #b00) ((and (fixnump disp) (<= -128 disp 127)) #b01) (t #b10))) (r/m (cond (index #b100) ((null base) #b101) (t (reg-tn-encoding base))))) (when (and (= mod 0) (= r/m #b101)) ;; this is rip-relative in amd64, so we'll use a sib instead (setf r/m #b100 scale 1)) (emit-mod-reg-r/m-byte segment mod reg r/m) (when (= r/m #b100) (let ((ss (1- (integer-length scale))) (index (if (null index) #b100 (let ((index (reg-tn-encoding index))) (if (= index #b100) (error "can't index off of ESP") index)))) (base (if (null base) #b101 (reg-tn-encoding base)))) (emit-sib-byte segment ss index base))) (cond ((= mod #b01) (emit-byte segment disp)) ((or (= mod #b10) (null base)) (if (fixup-p disp) (emit-absolute-fixup segment disp) (emit-signed-dword segment disp)))))) (fixup (typecase (fixup-offset thing) (label (emit-label-rip segment thing reg)) (t (emit-mod-reg-r/m-byte segment #b00 reg #b100) (emit-sib-byte segment 0 #b100 #b101) (emit-absolute-fixup segment thing)))))) (defun byte-reg-p (thing) (and (tn-p thing) (eq (sb-name (sc-sb (tn-sc thing))) 'registers) (member (sc-name (tn-sc thing)) *byte-sc-names*) t)) (defun byte-ea-p (thing) (typecase thing (ea (eq (ea-size thing) :byte)) (tn (and (member (sc-name (tn-sc thing)) *byte-sc-names*) t)) (t nil))) (defun word-reg-p (thing) (and (tn-p thing) (eq (sb-name (sc-sb (tn-sc thing))) 'registers) (member (sc-name (tn-sc thing)) *word-sc-names*) t)) (defun word-ea-p (thing) (typecase thing (ea (eq (ea-size thing) :word)) (tn (and (member (sc-name (tn-sc thing)) *word-sc-names*) t)) (t nil))) (defun dword-reg-p (thing) (and (tn-p thing) (eq (sb-name (sc-sb (tn-sc thing))) 'registers) (member (sc-name (tn-sc thing)) *dword-sc-names*) t)) (defun dword-ea-p (thing) (typecase thing (ea (eq (ea-size thing) :dword)) (tn (and (member (sc-name (tn-sc thing)) *dword-sc-names*) t)) (t nil))) (defun qword-reg-p (thing) (and (tn-p thing) (eq (sb-name (sc-sb (tn-sc thing))) 'registers) (member (sc-name (tn-sc thing)) *qword-sc-names*) t)) (defun qword-ea-p (thing) (typecase thing (ea (eq (ea-size thing) :qword)) (tn (and (member (sc-name (tn-sc thing)) *qword-sc-names*) t)) (t nil))) ;;; Return true if THING is a general-purpose register TN. (defun register-p (thing) (and (tn-p thing) (eq (sb-name (sc-sb (tn-sc thing))) 'registers))) (defun accumulator-p (thing) (and (register-p thing) (= (tn-offset thing) 0))) ;;; Return true if THING is an XMM register TN. (defun xmm-register-p (thing) (and (tn-p thing) (eq (sb-name (sc-sb (tn-sc thing))) 'float-registers))) ;;;; utilities (def!constant +operand-size-prefix-byte+ #b01100110) (defun maybe-emit-operand-size-prefix (segment size) (unless (or (eq size :byte) (eq size :qword) ; REX prefix handles this (eq size +default-operand-size+)) (emit-byte segment +operand-size-prefix-byte+))) ;;; A REX prefix must be emitted if at least one of the following ;;; conditions is true: ;; 1. The operand size is :QWORD and the default operand size of the ;; instruction is not :QWORD. ;;; 2. The instruction references an extended register. ;;; 3. The instruction references one of the byte registers SIL, DIL, ;;; SPL or BPL. ;;; Emit a REX prefix if necessary. OPERAND-SIZE is used to determine ;;; whether to set REX.W. Callers pass it explicitly as :DO-NOT-SET if ;;; this should not happen, for example because the instruction's ;;; default operand size is qword. R, X and B are NIL or TNs specifying ;;; registers the encodings of which are extended with the REX.R, REX.X ;;; and REX.B bit, respectively. To determine whether one of the byte ;;; registers is used that can only be accessed using a REX prefix, we ;;; need only to test R and B, because X is only used for the index ;;; register of an effective address and therefore never byte-sized. ;;; For R we can avoid to calculate the size of the TN because it is ;;; always OPERAND-SIZE. The size of B must be calculated here because ;;; B can be address-sized (if it is the base register of an effective ;;; address), of OPERAND-SIZE (if the instruction operates on two ;;; registers) or of some different size (in the instructions that ;;; combine arguments of different sizes: MOVZX, MOVSX, MOVSXD and ;;; several SSE instructions, e.g. CVTSD2SI). We don't distinguish ;;; between general-purpose and floating point registers for this cause ;;; because only general-purpose registers can be byte-sized at all. (defun maybe-emit-rex-prefix (segment operand-size r x b) (declare (type (member nil :byte :word :dword :qword :do-not-set) operand-size) (type (or null tn) r x b)) (labels ((if-hi (r) (if (and r (> (tn-offset r) ;; offset of r8 is 16, offset of xmm8 is 8 (if (eq (sb-name (sc-sb (tn-sc r))) 'float-registers) 7 15))) 1 0)) (reg-4-7-p (r) ;; Assuming R is a TN describing a general-purpose ;; register, return true if it references register ;; 4 upto 7. (<= 8 (tn-offset r) 15))) (let ((rex-w (if (eq operand-size :qword) 1 0)) (rex-r (if-hi r)) (rex-x (if-hi x)) (rex-b (if-hi b))) (when (or (not (zerop (logior rex-w rex-r rex-x rex-b))) (and r (eq operand-size :byte) (reg-4-7-p r)) (and b (eq (operand-size b) :byte) (reg-4-7-p b))) (emit-rex-byte segment #b0100 rex-w rex-r rex-x rex-b))))) ;;; Emit a REX prefix if necessary. The operand size is determined from ;;; THING or can be overwritten by OPERAND-SIZE. This and REG are always ;;; passed to MAYBE-EMIT-REX-PREFIX. Additionally, if THING is an EA we ;;; pass its index and base registers, if it is a register TN, we pass ;;; only itself. ;;; In contrast to EMIT-EA above, neither stack TNs nor fixups need to ;;; be treated specially here: If THING is a stack TN, neither it nor ;;; any of its components are passed to MAYBE-EMIT-REX-PREFIX which ;;; works correctly because stack references always use RBP as the base ;;; register and never use an index register so no extended registers ;;; need to be accessed. Fixups are assembled using an addressing mode ;;; of displacement-only or RIP-plus-displacement (see EMIT-EA), so may ;;; not reference an extended register. The displacement-only addressing ;;; mode requires that REX.X is 0, which is ensured here. (defun maybe-emit-rex-for-ea (segment thing reg &key operand-size) (declare (type (or ea tn fixup) thing) (type (or null tn) reg) (type (member nil :byte :word :dword :qword :do-not-set) operand-size)) (let ((ea-p (ea-p thing))) (maybe-emit-rex-prefix segment (or operand-size (operand-size thing)) reg (and ea-p (ea-index thing)) (cond (ea-p (ea-base thing)) ((and (tn-p thing) (member (sb-name (sc-sb (tn-sc thing))) '(float-registers registers))) thing) (t nil))))) (defun operand-size (thing) (typecase thing (tn ;; FIXME: might as well be COND instead of having to use #. readmacro ;; to hack up the code (case (sc-name (tn-sc thing)) (#.*qword-sc-names* :qword) (#.*dword-sc-names* :dword) (#.*word-sc-names* :word) (#.*byte-sc-names* :byte) ;; added by jrd: float-registers is a separate size (?) ;; The only place in the code where we are called with THING ;; being a float-register is in MAYBE-EMIT-REX-PREFIX when it ;; checks whether THING is a byte register. Thus our result in ;; these cases could as well be :dword and :qword. I leave it as ;; :float and :double which is more likely to trigger an aver ;; instead of silently doing the wrong thing in case this ;; situation should change. Lutz Euler, 2005-10-23. (#.*float-sc-names* :float) (#.*double-sc-names* :double) (t (error "can't tell the size of ~S ~S" thing (sc-name (tn-sc thing)))))) (ea (ea-size thing)) (fixup ;; GNA. Guess who spelt "flavor" correctly first time round? ;; There's a strong argument in my mind to change all uses of ;; "flavor" to "kind": and similarly with some misguided uses of ;; "type" here and there. -- CSR, 2005-01-06. (case (fixup-flavor thing) ((:foreign-dataref) :qword))) (t nil))) (defun matching-operand-size (dst src) (let ((dst-size (operand-size dst)) (src-size (operand-size src))) (if dst-size (if src-size (if (eq dst-size src-size) dst-size (error "size mismatch: ~S is a ~S and ~S is a ~S." dst dst-size src src-size)) dst-size) (if src-size src-size (error "can't tell the size of either ~S or ~S" dst src))))) ;;; Except in a very few cases (MOV instructions A1, A3 and B8 - BF) ;;; we expect dword data bytes even when 64 bit work is being done. ;;; But A1 and A3 are currently unused and B8 - BF use EMIT-QWORD ;;; directly, so we emit all quad constants as dwords, additionally ;;; making sure that they survive the sign-extension to 64 bits ;;; unchanged. (defun emit-sized-immediate (segment size value) (ecase size (:byte (emit-byte segment value)) (:word (emit-word segment value)) (:dword (emit-dword segment value)) (:qword (emit-signed-dword segment value)))) ;;;; general data transfer ;;; This is the part of the MOV instruction emitter that does moving ;;; of an immediate value into a qword register. We go to some length ;;; to achieve the shortest possible encoding. (defun emit-immediate-move-to-qword-register (segment dst src) (declare (type integer src)) (cond ((typep src '(unsigned-byte 32)) ;; We use the B8 - BF encoding with an operand size of 32 bits ;; here and let the implicit zero-extension fill the upper half ;; of the 64-bit destination register. Instruction size: five ;; or six bytes. (A REX prefix will be emitted only if the ;; destination is an extended register.) (maybe-emit-rex-prefix segment :dword nil nil dst) (emit-byte-with-reg segment #b10111 (reg-tn-encoding dst)) (emit-dword segment src)) (t (maybe-emit-rex-prefix segment :qword nil nil dst) (cond ((typep src '(signed-byte 32)) ;; Use the C7 encoding that takes a 32-bit immediate and ;; sign-extends it to 64 bits. Instruction size: seven ;; bytes. (emit-byte segment #b11000111) (emit-mod-reg-r/m-byte segment #b11 #b000 (reg-tn-encoding dst)) (emit-signed-dword segment src)) ((<= (- (expt 2 64) (expt 2 31)) src (1- (expt 2 64))) ;; This triggers on positive integers of 64 bits length ;; with the most significant 33 bits being 1. We use the ;; same encoding as in the previous clause. (emit-byte segment #b11000111) (emit-mod-reg-r/m-byte segment #b11 #b000 (reg-tn-encoding dst)) (emit-signed-dword segment (- src (expt 2 64)))) (t ;; We need a full 64-bit immediate. Instruction size: ;; ten bytes. (emit-byte-with-reg segment #b10111 (reg-tn-encoding dst)) (emit-qword segment src)))))) (define-instruction mov (segment dst src) ;; immediate to register (:printer reg ((op #b1011) (imm nil :type 'signed-imm-data)) '(:name :tab reg ", " imm)) (:printer rex-reg ((op #b1011) (imm nil :type 'signed-imm-data-upto-qword)) '(:name :tab reg ", " imm)) ;; absolute mem to/from accumulator (:printer simple-dir ((op #b101000) (imm nil :type 'imm-addr)) `(:name :tab ,(swap-if 'dir 'accum ", " '("[" imm "]")))) ;; register to/from register/memory (:printer reg-reg/mem-dir ((op #b100010))) (:printer rex-reg-reg/mem-dir ((op #b100010))) (:printer x66-reg-reg/mem-dir ((op #b100010))) (:printer x66-rex-reg-reg/mem-dir ((op #b100010))) ;; immediate to register/memory (:printer reg/mem-imm ((op '(#b1100011 #b000)))) (:printer rex-reg/mem-imm ((op '(#b1100011 #b000)))) (:emitter (let ((size (matching-operand-size dst src))) (maybe-emit-operand-size-prefix segment size) (cond ((register-p dst) (cond ((integerp src) (cond ((eq size :qword) (emit-immediate-move-to-qword-register segment dst src)) (t (maybe-emit-rex-prefix segment size nil nil dst) (emit-byte-with-reg segment (if (eq size :byte) #b10110 #b10111) (reg-tn-encoding dst)) (emit-sized-immediate segment size src)))) (t (maybe-emit-rex-for-ea segment src dst) (emit-byte segment (if (eq size :byte) #b10001010 #b10001011)) (emit-ea segment src (reg-tn-encoding dst) t)))) ((integerp src) ;; C7 only deals with 32 bit immediates even if the ;; destination is a 64-bit location. The value is ;; sign-extended in this case. (maybe-emit-rex-for-ea segment dst nil) (emit-byte segment (if (eq size :byte) #b11000110 #b11000111)) (emit-ea segment dst #b000) (emit-sized-immediate segment size src)) ((register-p src) (maybe-emit-rex-for-ea segment dst src) (emit-byte segment (if (eq size :byte) #b10001000 #b10001001)) (emit-ea segment dst (reg-tn-encoding src))) ((fixup-p src) ;; Generally we can't MOV a fixupped value into an EA, since ;; MOV on non-registers can only take a 32-bit immediate arg. ;; Make an exception for :FOREIGN fixups (pretty much just ;; the runtime asm, since other foreign calls go through the ;; the linkage table) and for linkage table references, since ;; these should always end up in low memory. (aver (or (eq (fixup-flavor src) :foreign) (eq (fixup-flavor src) :foreign-dataref) (eq (ea-size dst) :dword))) (maybe-emit-rex-for-ea segment dst nil) (emit-byte segment #b11000111) (emit-ea segment dst #b000) (emit-absolute-fixup segment src)) (t (error "bogus arguments to MOV: ~S ~S" dst src)))))) (defun emit-move-with-extension (segment dst src signed-p) (aver (register-p dst)) (let ((dst-size (operand-size dst)) (src-size (operand-size src)) (opcode (if signed-p #b10111110 #b10110110))) (ecase dst-size (:word (aver (eq src-size :byte)) (maybe-emit-operand-size-prefix segment :word) ;; REX prefix is needed if SRC is SIL, DIL, SPL or BPL. (maybe-emit-rex-for-ea segment src dst :operand-size :word) (emit-byte segment #b00001111) (emit-byte segment opcode) (emit-ea segment src (reg-tn-encoding dst))) ((:dword :qword) (ecase src-size (:byte (maybe-emit-rex-for-ea segment src dst :operand-size dst-size) (emit-byte segment #b00001111) (emit-byte segment opcode) (emit-ea segment src (reg-tn-encoding dst))) (:word (maybe-emit-rex-for-ea segment src dst :operand-size dst-size) (emit-byte segment #b00001111) (emit-byte segment (logior opcode 1)) (emit-ea segment src (reg-tn-encoding dst))) (:dword (aver (eq dst-size :qword)) ;; dst is in reg, src is in modrm (let ((ea-p (ea-p src))) (maybe-emit-rex-prefix segment (if signed-p :qword :dword) dst (and ea-p (ea-index src)) (cond (ea-p (ea-base src)) ((tn-p src) src) (t nil))) (emit-byte segment #x63) ;movsxd ;;(emit-byte segment opcode) (emit-ea segment src (reg-tn-encoding dst))))))))) (define-instruction movsx (segment dst src) (:printer ext-reg-reg/mem-no-width ((op #b10111110) (reg/mem nil :type 'sized-byte-reg/mem))) (:printer rex-ext-reg-reg/mem-no-width ((op #b10111110) (reg/mem nil :type 'sized-byte-reg/mem))) (:printer ext-reg-reg/mem-no-width ((op #b10111111) (reg/mem nil :type 'sized-word-reg/mem))) (:printer rex-ext-reg-reg/mem-no-width ((op #b10111111) (reg/mem nil :type 'sized-word-reg/mem))) (:emitter (emit-move-with-extension segment dst src :signed))) (define-instruction movzx (segment dst src) (:printer ext-reg-reg/mem-no-width ((op #b10110110) (reg/mem nil :type 'sized-byte-reg/mem))) (:printer rex-ext-reg-reg/mem-no-width ((op #b10110110) (reg/mem nil :type 'sized-byte-reg/mem))) (:printer ext-reg-reg/mem-no-width ((op #b10110111) (reg/mem nil :type 'sized-word-reg/mem))) (:printer rex-ext-reg-reg/mem-no-width ((op #b10110111) (reg/mem nil :type 'sized-word-reg/mem))) (:emitter (emit-move-with-extension segment dst src nil))) ;;; The regular use of MOVSXD is with an operand size of :qword. This ;;; sign-extends the dword source into the qword destination register. ;;; If the operand size is :dword the instruction zero-extends the dword ;;; source into the qword destination register, i.e. it does the same as ;;; a dword MOV into a register. (define-instruction movsxd (segment dst src) (:printer reg-reg/mem ((op #b0110001) (width 1) (reg/mem nil :type 'sized-dword-reg/mem))) (:printer rex-reg-reg/mem ((op #b0110001) (width 1) (reg/mem nil :type 'sized-dword-reg/mem))) (:emitter (emit-move-with-extension segment dst src :signed))) ;;; this is not a real amd64 instruction, of course (define-instruction movzxd (segment dst src) ; (:printer reg-reg/mem ((op #x63) (reg nil :type 'reg))) (:emitter (emit-move-with-extension segment dst src nil))) (define-instruction push (segment src) ;; register (:printer reg-no-width-default-qword ((op #b01010))) (:printer rex-reg-no-width-default-qword ((op #b01010))) ;; register/memory (:printer reg/mem-default-qword ((op '(#b11111111 #b110)))) (:printer rex-reg/mem-default-qword ((op '(#b11111111 #b110)))) ;; immediate (:printer byte ((op #b01101010) (imm nil :type 'signed-imm-byte)) '(:name :tab imm)) (:printer byte ((op #b01101000) (imm nil :type 'signed-imm-data-default-qword)) '(:name :tab imm)) ;; ### segment registers? (:emitter (cond ((integerp src) (cond ((<= -128 src 127) (emit-byte segment #b01101010) (emit-byte segment src)) (t ;; A REX-prefix is not needed because the operand size ;; defaults to 64 bits. The size of the immediate is 32 ;; bits and it is sign-extended. (emit-byte segment #b01101000) (emit-signed-dword segment src)))) (t (let ((size (operand-size src))) (aver (or (eq size :qword) (eq size :word))) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-for-ea segment src nil :operand-size :do-not-set) (cond ((register-p src) (emit-byte-with-reg segment #b01010 (reg-tn-encoding src))) (t (emit-byte segment #b11111111) (emit-ea segment src #b110 t)))))))) (define-instruction pop (segment dst) (:printer reg-no-width-default-qword ((op #b01011))) (:printer rex-reg-no-width-default-qword ((op #b01011))) (:printer reg/mem-default-qword ((op '(#b10001111 #b000)))) (:printer rex-reg/mem-default-qword ((op '(#b10001111 #b000)))) (:emitter (let ((size (operand-size dst))) (aver (or (eq size :qword) (eq size :word))) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-for-ea segment dst nil :operand-size :do-not-set) (cond ((register-p dst) (emit-byte-with-reg segment #b01011 (reg-tn-encoding dst))) (t (emit-byte segment #b10001111) (emit-ea segment dst #b000)))))) (define-instruction xchg (segment operand1 operand2) ;; Register with accumulator. (:printer reg-no-width ((op #b10010)) '(:name :tab accum ", " reg)) ;; Register/Memory with Register. (:printer reg-reg/mem ((op #b1000011))) (:printer rex-reg-reg/mem ((op #b1000011))) (:emitter (let ((size (matching-operand-size operand1 operand2))) (maybe-emit-operand-size-prefix segment size) (labels ((xchg-acc-with-something (acc something) (if (and (not (eq size :byte)) (register-p something)) (progn (maybe-emit-rex-for-ea segment acc something) (emit-byte-with-reg segment #b10010 (reg-tn-encoding something))) (xchg-reg-with-something acc something))) (xchg-reg-with-something (reg something) (maybe-emit-rex-for-ea segment something reg) (emit-byte segment (if (eq size :byte) #b10000110 #b10000111)) (emit-ea segment something (reg-tn-encoding reg)))) (cond ((accumulator-p operand1) (xchg-acc-with-something operand1 operand2)) ((accumulator-p operand2) (xchg-acc-with-something operand2 operand1)) ((register-p operand1) (xchg-reg-with-something operand1 operand2)) ((register-p operand2) (xchg-reg-with-something operand2 operand1)) (t (error "bogus args to XCHG: ~S ~S" operand1 operand2))))))) (define-instruction lea (segment dst src) (:printer rex-reg-reg/mem ((op #b1000110))) (:printer reg-reg/mem ((op #b1000110) (width 1))) (:emitter (aver (or (dword-reg-p dst) (qword-reg-p dst))) (maybe-emit-rex-for-ea segment src dst :operand-size :qword) (emit-byte segment #b10001101) (emit-ea segment src (reg-tn-encoding dst)))) (define-instruction cmpxchg (segment dst src) ;; Register/Memory with Register. (:printer ext-reg-reg/mem ((op #b1011000)) '(:name :tab reg/mem ", " reg)) (:emitter (aver (register-p src)) (let ((size (matching-operand-size src dst))) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-for-ea segment dst src) (emit-byte segment #b00001111) (emit-byte segment (if (eq size :byte) #b10110000 #b10110001)) (emit-ea segment dst (reg-tn-encoding src))))) (define-instruction fs-segment-prefix (segment) (:emitter (emit-byte segment #x64))) ;;;; flag control instructions ;;; CLC -- Clear Carry Flag. (define-instruction clc (segment) (:printer byte ((op #b11111000))) (:emitter (emit-byte segment #b11111000))) ;;; CLD -- Clear Direction Flag. (define-instruction cld (segment) (:printer byte ((op #b11111100))) (:emitter (emit-byte segment #b11111100))) ;;; CLI -- Clear Iterrupt Enable Flag. (define-instruction cli (segment) (:printer byte ((op #b11111010))) (:emitter (emit-byte segment #b11111010))) ;;; CMC -- Complement Carry Flag. (define-instruction cmc (segment) (:printer byte ((op #b11110101))) (:emitter (emit-byte segment #b11110101))) ;;; LAHF -- Load AH into flags. (define-instruction lahf (segment) (:printer byte ((op #b10011111))) (:emitter (emit-byte segment #b10011111))) ;;; POPF -- Pop flags. (define-instruction popf (segment) (:printer byte ((op #b10011101))) (:emitter (emit-byte segment #b10011101))) ;;; PUSHF -- push flags. (define-instruction pushf (segment) (:printer byte ((op #b10011100))) (:emitter (emit-byte segment #b10011100))) ;;; SAHF -- Store AH into flags. (define-instruction sahf (segment) (:printer byte ((op #b10011110))) (:emitter (emit-byte segment #b10011110))) ;;; STC -- Set Carry Flag. (define-instruction stc (segment) (:printer byte ((op #b11111001))) (:emitter (emit-byte segment #b11111001))) ;;; STD -- Set Direction Flag. (define-instruction std (segment) (:printer byte ((op #b11111101))) (:emitter (emit-byte segment #b11111101))) ;;; STI -- Set Interrupt Enable Flag. (define-instruction sti (segment) (:printer byte ((op #b11111011))) (:emitter (emit-byte segment #b11111011))) ;;;; arithmetic (defun emit-random-arith-inst (name segment dst src opcode &optional allow-constants) (let ((size (matching-operand-size dst src))) (maybe-emit-operand-size-prefix segment size) (cond ((integerp src) (cond ((and (not (eq size :byte)) (<= -128 src 127)) (maybe-emit-rex-for-ea segment dst nil) (emit-byte segment #b10000011) (emit-ea segment dst opcode allow-constants) (emit-byte segment src)) ((accumulator-p dst) (maybe-emit-rex-for-ea segment dst nil) (emit-byte segment (dpb opcode (byte 3 3) (if (eq size :byte) #b00000100 #b00000101))) (emit-sized-immediate segment size src)) (t (maybe-emit-rex-for-ea segment dst nil) (emit-byte segment (if (eq size :byte) #b10000000 #b10000001)) (emit-ea segment dst opcode allow-constants) (emit-sized-immediate segment size src)))) ((register-p src) (maybe-emit-rex-for-ea segment dst src) (emit-byte segment (dpb opcode (byte 3 3) (if (eq size :byte) #b00000000 #b00000001))) (emit-ea segment dst (reg-tn-encoding src) allow-constants)) ((register-p dst) (maybe-emit-rex-for-ea segment src dst) (emit-byte segment (dpb opcode (byte 3 3) (if (eq size :byte) #b00000010 #b00000011))) (emit-ea segment src (reg-tn-encoding dst) allow-constants)) (t (error "bogus operands to ~A" name))))) (eval-when (:compile-toplevel :execute) (defun arith-inst-printer-list (subop) `((accum-imm ((op ,(dpb subop (byte 3 2) #b0000010)))) (rex-accum-imm ((op ,(dpb subop (byte 3 2) #b0000010)))) (reg/mem-imm ((op (#b1000000 ,subop)))) (rex-reg/mem-imm ((op (#b1000000 ,subop)))) ;; The redundant encoding #x82 is invalid in 64-bit mode, ;; therefore we force WIDTH to 1. (reg/mem-imm ((op (#b1000001 ,subop)) (width 1) (imm nil :type signed-imm-byte))) (rex-reg/mem-imm ((op (#b1000001 ,subop)) (width 1) (imm nil :type signed-imm-byte))) (reg-reg/mem-dir ((op ,(dpb subop (byte 3 1) #b000000)))) (rex-reg-reg/mem-dir ((op ,(dpb subop (byte 3 1) #b000000)))))) ) (define-instruction add (segment dst src) (:printer-list (arith-inst-printer-list #b000)) (:emitter (emit-random-arith-inst "ADD" segment dst src #b000))) (define-instruction adc (segment dst src) (:printer-list (arith-inst-printer-list #b010)) (:emitter (emit-random-arith-inst "ADC" segment dst src #b010))) (define-instruction sub (segment dst src) (:printer-list (arith-inst-printer-list #b101)) (:emitter (emit-random-arith-inst "SUB" segment dst src #b101))) (define-instruction sbb (segment dst src) (:printer-list (arith-inst-printer-list #b011)) (:emitter (emit-random-arith-inst "SBB" segment dst src #b011))) (define-instruction cmp (segment dst src) (:printer-list (arith-inst-printer-list #b111)) (:emitter (emit-random-arith-inst "CMP" segment dst src #b111 t))) ;;; The one-byte encodings for INC and DEC are used as REX prefixes ;;; in 64-bit mode so we always use the two-byte form. (define-instruction inc (segment dst) (:printer reg/mem ((op '(#b1111111 #b000)))) (:printer rex-reg/mem ((op '(#b1111111 #b000)))) (:emitter (let ((size (operand-size dst))) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-for-ea segment dst nil) (emit-byte segment (if (eq size :byte) #b11111110 #b11111111)) (emit-ea segment dst #b000)))) (define-instruction dec (segment dst) (:printer reg/mem ((op '(#b1111111 #b001)))) (:printer rex-reg/mem ((op '(#b1111111 #b001)))) (:emitter (let ((size (operand-size dst))) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-for-ea segment dst nil) (emit-byte segment (if (eq size :byte) #b11111110 #b11111111)) (emit-ea segment dst #b001)))) (define-instruction neg (segment dst) (:printer reg/mem ((op '(#b1111011 #b011)))) (:printer rex-reg/mem ((op '(#b1111011 #b011)))) (:emitter (let ((size (operand-size dst))) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-for-ea segment dst nil) (emit-byte segment (if (eq size :byte) #b11110110 #b11110111)) (emit-ea segment dst #b011)))) (define-instruction mul (segment dst src) (:printer accum-reg/mem ((op '(#b1111011 #b100)))) (:printer rex-accum-reg/mem ((op '(#b1111011 #b100)))) (:emitter (let ((size (matching-operand-size dst src))) (aver (accumulator-p dst)) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-for-ea segment src nil) (emit-byte segment (if (eq size :byte) #b11110110 #b11110111)) (emit-ea segment src #b100)))) (define-instruction imul (segment dst &optional src1 src2) (:printer accum-reg/mem ((op '(#b1111011 #b101)))) (:printer rex-accum-reg/mem ((op '(#b1111011 #b101)))) (:printer ext-reg-reg/mem-no-width ((op #b10101111))) (:printer rex-ext-reg-reg/mem-no-width ((op #b10101111))) (:printer reg-reg/mem ((op #b0110100) (width 1) (imm nil :type 'signed-imm-data)) '(:name :tab reg ", " reg/mem ", " imm)) (:printer rex-reg-reg/mem ((op #b0110100) (width 1) (imm nil :type 'signed-imm-data)) '(:name :tab reg ", " reg/mem ", " imm)) (:printer reg-reg/mem ((op #b0110101) (width 1) (imm nil :type 'signed-imm-byte)) '(:name :tab reg ", " reg/mem ", " imm)) (:printer rex-reg-reg/mem ((op #b0110101) (width 1) (imm nil :type 'signed-imm-byte)) '(:name :tab reg ", " reg/mem ", " imm)) (:emitter (flet ((r/m-with-immed-to-reg (reg r/m immed) (let* ((size (matching-operand-size reg r/m)) (sx (and (not (eq size :byte)) (<= -128 immed 127)))) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-for-ea segment r/m reg) (emit-byte segment (if sx #b01101011 #b01101001)) (emit-ea segment r/m (reg-tn-encoding reg)) (if sx (emit-byte segment immed) (emit-sized-immediate segment size immed))))) (cond (src2 (r/m-with-immed-to-reg dst src1 src2)) (src1 (if (integerp src1) (r/m-with-immed-to-reg dst dst src1) (let ((size (matching-operand-size dst src1))) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-for-ea segment src1 dst) (emit-byte segment #b00001111) (emit-byte segment #b10101111) (emit-ea segment src1 (reg-tn-encoding dst))))) (t (let ((size (operand-size dst))) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-for-ea segment dst nil) (emit-byte segment (if (eq size :byte) #b11110110 #b11110111)) (emit-ea segment dst #b101))))))) (define-instruction div (segment dst src) (:printer accum-reg/mem ((op '(#b1111011 #b110)))) (:printer rex-accum-reg/mem ((op '(#b1111011 #b110)))) (:emitter (let ((size (matching-operand-size dst src))) (aver (accumulator-p dst)) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-for-ea segment src nil) (emit-byte segment (if (eq size :byte) #b11110110 #b11110111)) (emit-ea segment src #b110)))) (define-instruction idiv (segment dst src) (:printer accum-reg/mem ((op '(#b1111011 #b111)))) (:printer rex-accum-reg/mem ((op '(#b1111011 #b111)))) (:emitter (let ((size (matching-operand-size dst src))) (aver (accumulator-p dst)) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-for-ea segment src nil) (emit-byte segment (if (eq size :byte) #b11110110 #b11110111)) (emit-ea segment src #b111)))) (define-instruction bswap (segment dst) (:printer ext-reg-no-width ((op #b11001))) (:emitter (let ((size (operand-size dst))) (maybe-emit-rex-prefix segment size nil nil dst) (emit-byte segment #x0f) (emit-byte-with-reg segment #b11001 (reg-tn-encoding dst))))) ;;; CBW -- Convert Byte to Word. AX <- sign_xtnd(AL) (define-instruction cbw (segment) (:printer x66-byte ((op #b10011000))) (:emitter (maybe-emit-operand-size-prefix segment :word) (emit-byte segment #b10011000))) ;;; CWDE -- Convert Word To Double Word Extended. EAX <- sign_xtnd(AX) (define-instruction cwde (segment) (:printer byte ((op #b10011000))) (:emitter (maybe-emit-operand-size-prefix segment :dword) (emit-byte segment #b10011000))) ;;; CDQE -- Convert Double Word To Quad Word Extended. RAX <- sign_xtnd(EAX) (define-instruction cdqe (segment) (:printer rex-byte ((op #b10011000))) (:emitter (maybe-emit-rex-prefix segment :qword nil nil nil) (emit-byte segment #b10011000))) ;;; CWD -- Convert Word to Double Word. DX:AX <- sign_xtnd(AX) (define-instruction cwd (segment) (:printer x66-byte ((op #b10011001))) (:emitter (maybe-emit-operand-size-prefix segment :word) (emit-byte segment #b10011001))) ;;; CDQ -- Convert Double Word to Quad Word. EDX:EAX <- sign_xtnd(EAX) (define-instruction cdq (segment) (:printer byte ((op #b10011001))) (:emitter (maybe-emit-operand-size-prefix segment :dword) (emit-byte segment #b10011001))) ;;; CQO -- Convert Quad Word to Octaword. RDX:RAX <- sign_xtnd(RAX) (define-instruction cqo (segment) (:printer rex-byte ((op #b10011001))) (:emitter (maybe-emit-rex-prefix segment :qword nil nil nil) (emit-byte segment #b10011001))) (define-instruction xadd (segment dst src) ;; Register/Memory with Register. (:printer ext-reg-reg/mem ((op #b1100000)) '(:name :tab reg/mem ", " reg)) (:emitter (aver (register-p src)) (let ((size (matching-operand-size src dst))) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-for-ea segment dst src) (emit-byte segment #b00001111) (emit-byte segment (if (eq size :byte) #b11000000 #b11000001)) (emit-ea segment dst (reg-tn-encoding src))))) ;;;; logic (defun emit-shift-inst (segment dst amount opcode) (let ((size (operand-size dst))) (maybe-emit-operand-size-prefix segment size) (multiple-value-bind (major-opcode immed) (case amount (:cl (values #b11010010 nil)) (1 (values #b11010000 nil)) (t (values #b11000000 t))) (maybe-emit-rex-for-ea segment dst nil) (emit-byte segment (if (eq size :byte) major-opcode (logior major-opcode 1))) (emit-ea segment dst opcode) (when immed (emit-byte segment amount))))) (eval-when (:compile-toplevel :execute) (defun shift-inst-printer-list (subop) `((reg/mem ((op (#b1101000 ,subop))) (:name :tab reg/mem ", 1")) (rex-reg/mem ((op (#b1101000 ,subop))) (:name :tab reg/mem ", 1")) (reg/mem ((op (#b1101001 ,subop))) (:name :tab reg/mem ", " 'cl)) (rex-reg/mem ((op (#b1101001 ,subop))) (:name :tab reg/mem ", " 'cl)) (reg/mem-imm ((op (#b1100000 ,subop)) (imm nil :type imm-byte))) (rex-reg/mem-imm ((op (#b1100000 ,subop)) (imm nil :type imm-byte)))))) (define-instruction rol (segment dst amount) (:printer-list (shift-inst-printer-list #b000)) (:emitter (emit-shift-inst segment dst amount #b000))) (define-instruction ror (segment dst amount) (:printer-list (shift-inst-printer-list #b001)) (:emitter (emit-shift-inst segment dst amount #b001))) (define-instruction rcl (segment dst amount) (:printer-list (shift-inst-printer-list #b010)) (:emitter (emit-shift-inst segment dst amount #b010))) (define-instruction rcr (segment dst amount) (:printer-list (shift-inst-printer-list #b011)) (:emitter (emit-shift-inst segment dst amount #b011))) (define-instruction shl (segment dst amount) (:printer-list (shift-inst-printer-list #b100)) (:emitter (emit-shift-inst segment dst amount #b100))) (define-instruction shr (segment dst amount) (:printer-list (shift-inst-printer-list #b101)) (:emitter (emit-shift-inst segment dst amount #b101))) (define-instruction sar (segment dst amount) (:printer-list (shift-inst-printer-list #b111)) (:emitter (emit-shift-inst segment dst amount #b111))) (defun emit-double-shift (segment opcode dst src amt) (let ((size (matching-operand-size dst src))) (when (eq size :byte) (error "Double shifts can only be used with words.")) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-for-ea segment dst src) (emit-byte segment #b00001111) (emit-byte segment (dpb opcode (byte 1 3) (if (eq amt :cl) #b10100101 #b10100100))) (emit-ea segment dst (reg-tn-encoding src)) (unless (eq amt :cl) (emit-byte segment amt)))) (eval-when (:compile-toplevel :execute) (defun double-shift-inst-printer-list (op) `(#+nil (ext-reg-reg/mem-imm ((op ,(logior op #b100)) (imm nil :type signed-imm-byte))) (ext-reg-reg/mem ((op ,(logior op #b101))) (:name :tab reg/mem ", " 'cl))))) (define-instruction shld (segment dst src amt) (:declare (type (or (member :cl) (mod 32)) amt)) (:printer-list (double-shift-inst-printer-list #b10100000)) (:emitter (emit-double-shift segment #b0 dst src amt))) (define-instruction shrd (segment dst src amt) (:declare (type (or (member :cl) (mod 32)) amt)) (:printer-list (double-shift-inst-printer-list #b10101000)) (:emitter (emit-double-shift segment #b1 dst src amt))) (define-instruction and (segment dst src) (:printer-list (arith-inst-printer-list #b100)) (:emitter (emit-random-arith-inst "AND" segment dst src #b100))) (define-instruction test (segment this that) (:printer accum-imm ((op #b1010100))) (:printer rex-accum-imm ((op #b1010100))) (:printer reg/mem-imm ((op '(#b1111011 #b000)))) (:printer rex-reg/mem-imm ((op '(#b1111011 #b000)))) (:printer reg-reg/mem ((op #b1000010))) (:printer rex-reg-reg/mem ((op #b1000010))) (:emitter (let ((size (matching-operand-size this that))) (maybe-emit-operand-size-prefix segment size) (flet ((test-immed-and-something (immed something) (cond ((accumulator-p something) (maybe-emit-rex-for-ea segment something nil) (emit-byte segment (if (eq size :byte) #b10101000 #b10101001)) (emit-sized-immediate segment size immed)) (t (maybe-emit-rex-for-ea segment something nil) (emit-byte segment (if (eq size :byte) #b11110110 #b11110111)) (emit-ea segment something #b000) (emit-sized-immediate segment size immed)))) (test-reg-and-something (reg something) (maybe-emit-rex-for-ea segment something reg) (emit-byte segment (if (eq size :byte) #b10000100 #b10000101)) (emit-ea segment something (reg-tn-encoding reg)))) (cond ((integerp that) (test-immed-and-something that this)) ((integerp this) (test-immed-and-something this that)) ((register-p this) (test-reg-and-something this that)) ((register-p that) (test-reg-and-something that this)) (t (error "bogus operands for TEST: ~S and ~S" this that))))))) (define-instruction or (segment dst src) (:printer-list (arith-inst-printer-list #b001)) (:emitter (emit-random-arith-inst "OR" segment dst src #b001))) (define-instruction xor (segment dst src) (:printer-list (arith-inst-printer-list #b110)) (:emitter (emit-random-arith-inst "XOR" segment dst src #b110))) (define-instruction not (segment dst) (:printer reg/mem ((op '(#b1111011 #b010)))) (:printer rex-reg/mem ((op '(#b1111011 #b010)))) (:emitter (let ((size (operand-size dst))) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-for-ea segment dst nil) (emit-byte segment (if (eq size :byte) #b11110110 #b11110111)) (emit-ea segment dst #b010)))) ;;;; string manipulation (define-instruction cmps (segment size) (:printer string-op ((op #b1010011))) (:printer rex-string-op ((op #b1010011))) (:emitter (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-prefix segment size nil nil nil) (emit-byte segment (if (eq size :byte) #b10100110 #b10100111)))) (define-instruction ins (segment acc) (:printer string-op ((op #b0110110))) (:printer rex-string-op ((op #b0110110))) (:emitter (let ((size (operand-size acc))) (aver (accumulator-p acc)) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-prefix segment size nil nil nil) (emit-byte segment (if (eq size :byte) #b01101100 #b01101101))))) (define-instruction lods (segment acc) (:printer string-op ((op #b1010110))) (:printer rex-string-op ((op #b1010110))) (:emitter (let ((size (operand-size acc))) (aver (accumulator-p acc)) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-prefix segment size nil nil nil) (emit-byte segment (if (eq size :byte) #b10101100 #b10101101))))) (define-instruction movs (segment size) (:printer string-op ((op #b1010010))) (:printer rex-string-op ((op #b1010010))) (:emitter (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-prefix segment size nil nil nil) (emit-byte segment (if (eq size :byte) #b10100100 #b10100101)))) (define-instruction outs (segment acc) (:printer string-op ((op #b0110111))) (:printer rex-string-op ((op #b0110111))) (:emitter (let ((size (operand-size acc))) (aver (accumulator-p acc)) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-prefix segment size nil nil nil) (emit-byte segment (if (eq size :byte) #b01101110 #b01101111))))) (define-instruction scas (segment acc) (:printer string-op ((op #b1010111))) (:printer rex-string-op ((op #b1010111))) (:emitter (let ((size (operand-size acc))) (aver (accumulator-p acc)) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-prefix segment size nil nil nil) (emit-byte segment (if (eq size :byte) #b10101110 #b10101111))))) (define-instruction stos (segment acc) (:printer string-op ((op #b1010101))) (:printer rex-string-op ((op #b1010101))) (:emitter (let ((size (operand-size acc))) (aver (accumulator-p acc)) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-prefix segment size nil nil nil) (emit-byte segment (if (eq size :byte) #b10101010 #b10101011))))) (define-instruction xlat (segment) (:printer byte ((op #b11010111))) (:emitter (emit-byte segment #b11010111))) (define-instruction rep (segment) (:emitter (emit-byte segment #b11110010))) (define-instruction repe (segment) (:printer byte ((op #b11110011))) (:emitter (emit-byte segment #b11110011))) (define-instruction repne (segment) (:printer byte ((op #b11110010))) (:emitter (emit-byte segment #b11110010))) ;;;; bit manipulation (define-instruction bsf (segment dst src) (:printer ext-reg-reg/mem-no-width ((op #b10111100))) (:printer rex-ext-reg-reg/mem-no-width ((op #b10111100))) (:emitter (let ((size (matching-operand-size dst src))) (when (eq size :byte) (error "can't scan bytes: ~S" src)) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-for-ea segment src dst) (emit-byte segment #b00001111) (emit-byte segment #b10111100) (emit-ea segment src (reg-tn-encoding dst))))) (define-instruction bsr (segment dst src) (:printer ext-reg-reg/mem-no-width ((op #b10111101))) (:printer rex-ext-reg-reg/mem-no-width ((op #b10111101))) (:emitter (let ((size (matching-operand-size dst src))) (when (eq size :byte) (error "can't scan bytes: ~S" src)) (maybe-emit-operand-size-prefix segment size) (maybe-emit-rex-for-ea segment src dst) (emit-byte segment #b00001111) (emit-byte segment #b10111101) (emit-ea segment src (reg-tn-encoding dst))))) (defun emit-bit-test-and-mumble (segment src index opcode) (let ((size (operand-size src))) (when (eq size :byte) (error "can't scan bytes: ~S" src)) (maybe-emit-operand-size-prefix segment size) (cond ((integerp index) (maybe-emit-rex-for-ea segment src nil) (emit-byte segment #b00001111) (emit-byte segment #b10111010) (emit-ea segment src opcode) (emit-byte segment index)) (t (maybe-emit-rex-for-ea segment src index) (emit-byte segment #b00001111) (emit-byte segment (dpb opcode (byte 3 3) #b10000011)) (emit-ea segment src (reg-tn-encoding index)))))) (eval-when (:compile-toplevel :execute) (defun bit-test-inst-printer-list (subop) `((ext-reg/mem-imm ((op (#b1011101 ,subop)) (reg/mem nil :type reg/mem) (imm nil :type imm-byte) (width 0))) (ext-reg-reg/mem ((op ,(dpb subop (byte 3 2) #b1000001)) (width 1)) (:name :tab reg/mem ", " reg))))) (define-instruction bt (segment src index) (:printer-list (bit-test-inst-printer-list #b100)) (:emitter (emit-bit-test-and-mumble segment src index #b100))) (define-instruction btc (segment src index) (:printer-list (bit-test-inst-printer-list #b111)) (:emitter (emit-bit-test-and-mumble segment src index #b111))) (define-instruction btr (segment src index) (:printer-list (bit-test-inst-printer-list #b110)) (:emitter (emit-bit-test-and-mumble segment src index #b110))) (define-instruction bts (segment src index) (:printer-list (bit-test-inst-printer-list #b101)) (:emitter (emit-bit-test-and-mumble segment src index #b101))) ;;;; control transfer (define-instruction call (segment where) (:printer near-jump ((op #b11101000))) (:printer reg/mem-default-qword ((op '(#b11111111 #b010)))) (:printer rex-reg/mem-default-qword ((op '(#b11111111 #b010)))) (:emitter (typecase where (label (emit-byte segment #b11101000) ; 32 bit relative (emit-back-patch segment 4 (lambda (segment posn) (emit-signed-dword segment (- (label-position where) (+ posn 4)))))) (fixup (emit-byte segment #b11101000) (emit-relative-fixup segment where)) (t (maybe-emit-rex-for-ea segment where nil :operand-size :do-not-set) (emit-byte segment #b11111111) (emit-ea segment where #b010))))) (defun emit-byte-displacement-backpatch (segment target) (emit-back-patch segment 1 (lambda (segment posn) (let ((disp (- (label-position target) (1+ posn)))) (aver (<= -128 disp 127)) (emit-byte segment disp))))) (define-instruction jmp (segment cond &optional where) ;; conditional jumps (:printer short-cond-jump ((op #b0111)) '('j cc :tab label)) (:printer near-cond-jump () '('j cc :tab label)) ;; unconditional jumps (:printer short-jump ((op #b1011))) (:printer near-jump ((op #b11101001))) (:printer reg/mem-default-qword ((op '(#b11111111 #b100)))) (:printer rex-reg/mem-default-qword ((op '(#b11111111 #b100)))) (:emitter (cond (where (emit-chooser segment 6 2 (lambda (segment posn delta-if-after) (let ((disp (- (label-position where posn delta-if-after) (+ posn 2)))) (when (<= -128 disp 127) (emit-byte segment (dpb (conditional-opcode cond) (byte 4 0) #b01110000)) (emit-byte-displacement-backpatch segment where) t))) (lambda (segment posn) (let ((disp (- (label-position where) (+ posn 6)))) (emit-byte segment #b00001111) (emit-byte segment (dpb (conditional-opcode cond) (byte 4 0) #b10000000)) (emit-signed-dword segment disp))))) ((label-p (setq where cond)) (emit-chooser segment 5 0 (lambda (segment posn delta-if-after) (let ((disp (- (label-position where posn delta-if-after) (+ posn 2)))) (when (<= -128 disp 127) (emit-byte segment #b11101011) (emit-byte-displacement-backpatch segment where) t))) (lambda (segment posn) (let ((disp (- (label-position where) (+ posn 5)))) (emit-byte segment #b11101001) (emit-signed-dword segment disp))))) ((fixup-p where) (emit-byte segment #b11101001) (emit-relative-fixup segment where)) (t (unless (or (ea-p where) (tn-p where)) (error "don't know what to do with ~A" where)) ;; near jump defaults to 64 bit ;; w-bit in rex prefix is unnecessary (maybe-emit-rex-for-ea segment where nil :operand-size :do-not-set) (emit-byte segment #b11111111) (emit-ea segment where #b100))))) (define-instruction jmp-short (segment label) (:emitter (emit-byte segment #b11101011) (emit-byte-displacement-backpatch segment label))) (define-instruction ret (segment &optional stack-delta) (:printer byte ((op #b11000011))) (:printer byte ((op #b11000010) (imm nil :type 'imm-word-16)) '(:name :tab imm)) (:emitter (cond (stack-delta (emit-byte segment #b11000010) (emit-word segment stack-delta)) (t (emit-byte segment #b11000011))))) (define-instruction jecxz (segment target) (:printer short-jump ((op #b0011))) (:emitter (emit-byte segment #b11100011) (emit-byte-displacement-backpatch segment target))) (define-instruction loop (segment target) (:printer short-jump ((op #b0010))) (:emitter (emit-byte segment #b11100010) ; pfw this was 11100011, or jecxz!!!! (emit-byte-displacement-backpatch segment target))) (define-instruction loopz (segment target) (:printer short-jump ((op #b0001))) (:emitter (emit-byte segment #b11100001) (emit-byte-displacement-backpatch segment target))) (define-instruction loopnz (segment target) (:printer short-jump ((op #b0000))) (:emitter (emit-byte segment #b11100000) (emit-byte-displacement-backpatch segment target))) ;;;; conditional move (define-instruction cmov (segment cond dst src) (:printer cond-move ()) (:printer rex-cond-move ()) (:emitter (aver (register-p dst)) (let ((size (matching-operand-size dst src))) (aver (or (eq size :word) (eq size :dword) (eq size :qword))) (maybe-emit-operand-size-prefix segment size)) (maybe-emit-rex-for-ea segment src dst) (emit-byte segment #b00001111) (emit-byte segment (dpb (conditional-opcode cond) (byte 4 0) #b01000000)) (emit-ea segment src (reg-tn-encoding dst)))) ;;;; conditional byte set (define-instruction set (segment dst cond) (:printer cond-set ()) (:emitter (maybe-emit-rex-for-ea segment dst nil) (emit-byte segment #b00001111) (emit-byte segment (dpb (conditional-opcode cond) (byte 4 0) #b10010000)) (emit-ea segment dst #b000))) ;;;; enter/leave (define-instruction enter (segment disp &optional (level 0)) (:declare (type (unsigned-byte 16) disp) (type (unsigned-byte 8) level)) (:printer enter-format ((op #b11001000))) (:emitter (emit-byte segment #b11001000) (emit-word segment disp) (emit-byte segment level))) (define-instruction leave (segment) (:printer byte ((op #b11001001))) (:emitter (emit-byte segment #b11001001))) ;;;; interrupt instructions (defun snarf-error-junk (sap offset &optional length-only) (let* ((length (sb!sys:sap-ref-8 sap offset)) (vector (make-array length :element-type '(unsigned-byte 8)))) (declare (type sb!sys:system-area-pointer sap) (type (unsigned-byte 8) length) (type (simple-array (unsigned-byte 8) (*)) vector)) (cond (length-only (values 0 (1+ length) nil nil)) (t (sb!kernel:copy-ub8-from-system-area sap (1+ offset) vector 0 length) (collect ((sc-offsets) (lengths)) (lengths 1) ; the length byte (let* ((index 0) (error-number (sb!c:read-var-integer vector index))) (lengths index) (loop (when (>= index length) (return)) (let ((old-index index)) (sc-offsets (sb!c:read-var-integer vector index)) (lengths (- index old-index)))) (values error-number (1+ length) (sc-offsets) (lengths)))))))) #| (defmacro break-cases (breaknum &body cases) (let ((bn-temp (gensym))) (collect ((clauses)) (dolist (case cases) (clauses `((= ,bn-temp ,(car case)) ,@(cdr case)))) `(let ((,bn-temp ,breaknum)) (cond ,@(clauses)))))) |# (defun break-control (chunk inst stream dstate) (declare (ignore inst)) (flet ((nt (x) (if stream (sb!disassem:note x dstate)))) ;; FIXME: Make sure that BYTE-IMM-CODE is defined. The genesis ;; map has it undefined; and it should be easier to look in the target ;; Lisp (with (DESCRIBE 'BYTE-IMM-CODE)) than to definitively deduce ;; from first principles whether it's defined in some way that genesis ;; can't grok. (case #!-darwin (byte-imm-code chunk dstate) #!+darwin (word-imm-code chunk dstate) (#.error-trap (nt "error trap") (sb!disassem:handle-break-args #'snarf-error-junk stream dstate)) (#.cerror-trap (nt "cerror trap") (sb!disassem:handle-break-args #'snarf-error-junk stream dstate)) (#.breakpoint-trap (nt "breakpoint trap")) (#.pending-interrupt-trap (nt "pending interrupt trap")) (#.halt-trap (nt "halt trap")) (#.fun-end-breakpoint-trap (nt "function end breakpoint trap")) (#.single-step-around-trap (nt "single-step trap (around)")) (#.single-step-before-trap (nt "single-step trap (before)"))))) (define-instruction break (segment code) (:declare (type (unsigned-byte 8) code)) #!-darwin (:printer byte-imm ((op #b11001100)) '(:name :tab code) :control #'break-control) #!+darwin (:printer word-imm ((op #b0000101100001111)) '(:name :tab code) :control #'break-control) (:emitter #!-darwin (emit-byte segment #b11001100) ;; On darwin, trap handling via SIGTRAP is unreliable, therefore we ;; throw a sigill with 0x0b0f instead and check for this in the ;; SIGILL handler and pass it on to the sigtrap handler if ;; appropriate #!+darwin (emit-word segment #b0000101100001111) (emit-byte segment code))) (define-instruction int (segment number) (:declare (type (unsigned-byte 8) number)) (:printer byte-imm ((op #b11001101))) (:emitter (etypecase number ((member 3) (emit-byte segment #b11001100)) ((unsigned-byte 8) (emit-byte segment #b11001101) (emit-byte segment number))))) (define-instruction iret (segment) (:printer byte ((op #b11001111))) (:emitter (emit-byte segment #b11001111))) ;;;; processor control (define-instruction hlt (segment) (:printer byte ((op #b11110100))) (:emitter (emit-byte segment #b11110100))) (define-instruction nop (segment) (:printer byte ((op #b10010000))) (:emitter (emit-byte segment #b10010000))) (define-instruction wait (segment) (:printer byte ((op #b10011011))) (:emitter (emit-byte segment #b10011011))) (define-instruction lock (segment) (:printer byte ((op #b11110000))) (:emitter (emit-byte segment #b11110000))) ;;;; miscellaneous hackery (define-instruction byte (segment byte) (:emitter (emit-byte segment byte))) (define-instruction word (segment word) (:emitter (emit-word segment word))) (define-instruction dword (segment dword) (:emitter (emit-dword segment dword))) (defun emit-header-data (segment type) (emit-back-patch segment n-word-bytes (lambda (segment posn) (emit-qword segment (logior type (ash (+ posn (component-header-length)) (- n-widetag-bits word-shift))))))) (define-instruction simple-fun-header-word (segment) (:emitter (emit-header-data segment simple-fun-header-widetag))) (define-instruction lra-header-word (segment) (:emitter (emit-header-data segment return-pc-header-widetag))) ;;;; Instructions required to do floating point operations using SSE (defun emit-sse-inst (segment dst src prefix opcode &key operand-size) (when prefix (emit-byte segment prefix)) (if operand-size (maybe-emit-rex-for-ea segment src dst :operand-size operand-size) (maybe-emit-rex-for-ea segment src dst)) (emit-byte segment #x0f) (emit-byte segment opcode) (emit-ea segment src (reg-tn-encoding dst))) ;;; Emit an SSE instruction that has an XMM register as the destination ;;; operand and for which the size of the operands is implicitly given ;;; by the instruction. (defun emit-regular-sse-inst (segment dst src prefix opcode) (aver (xmm-register-p dst)) (emit-sse-inst segment dst src prefix opcode :operand-size :do-not-set)) ;;; Instructions having an XMM register as the destination operand ;;; and an XMM register or a memory location as the source operand. ;;; The operand size is implicitly given by the instruction. (macrolet ((define-regular-sse-inst (name prefix opcode) `(define-instruction ,name (segment dst src) ,@(if prefix `((:printer ext-xmm-xmm/mem ((prefix ,prefix) (op ,opcode))) (:printer ext-rex-xmm-xmm/mem ((prefix ,prefix) (op ,opcode)))) `((:printer xmm-xmm/mem ((op ,opcode))) (:printer rex-xmm-xmm/mem ((op ,opcode))))) (:emitter (emit-regular-sse-inst segment dst src ,prefix ,opcode))))) ;; logical (define-regular-sse-inst andpd #x66 #x54) (define-regular-sse-inst andps nil #x54) (define-regular-sse-inst xorpd #x66 #x57) (define-regular-sse-inst xorps nil #x57) ;; comparison (define-regular-sse-inst comisd #x66 #x2f) (define-regular-sse-inst comiss nil #x2f) ;; arithmetic (define-regular-sse-inst addsd #xf2 #x58) (define-regular-sse-inst addss #xf3 #x58) (define-regular-sse-inst divsd #xf2 #x5e) (define-regular-sse-inst divss #xf3 #x5e) (define-regular-sse-inst mulsd #xf2 #x59) (define-regular-sse-inst mulss #xf3 #x59) (define-regular-sse-inst subsd #xf2 #x5c) (define-regular-sse-inst subss #xf3 #x5c) (define-regular-sse-inst sqrtsd #xf2 #x51) (define-regular-sse-inst sqrtss #xf3 #x51) ;; conversion (define-regular-sse-inst cvtsd2ss #xf2 #x5a) (define-regular-sse-inst cvtss2sd #xf3 #x5a) (define-regular-sse-inst cvtdq2pd #xf3 #xe6) (define-regular-sse-inst cvtdq2ps nil #x5b)) ;;; MOVSD, MOVSS (macrolet ((define-movsd/ss-sse-inst (name prefix) `(define-instruction ,name (segment dst src) (:printer ext-xmm-xmm/mem-dir ((prefix ,prefix) (op #b0001000))) (:printer ext-rex-xmm-xmm/mem-dir ((prefix ,prefix) (op #b0001000))) (:emitter (cond ((xmm-register-p dst) (emit-sse-inst segment dst src ,prefix #x10 :operand-size :do-not-set)) (t (aver (xmm-register-p src)) (emit-sse-inst segment src dst ,prefix #x11 :operand-size :do-not-set))))))) (define-movsd/ss-sse-inst movsd #xf2) (define-movsd/ss-sse-inst movss #xf3)) ;;; MOVQ (define-instruction movq (segment dst src) (:printer ext-xmm-xmm/mem ((prefix #xf3) (op #x7e))) (:printer ext-rex-xmm-xmm/mem ((prefix #xf3) (op #x7e))) (:printer ext-xmm-xmm/mem ((prefix #x66) (op #xd6)) '(:name :tab reg/mem ", " reg)) (:printer ext-rex-xmm-xmm/mem ((prefix #x66) (op #xd6)) '(:name :tab reg/mem ", " reg)) (:emitter (cond ((xmm-register-p dst) (emit-sse-inst segment dst src #xf3 #x7e :operand-size :do-not-set)) (t (aver (xmm-register-p src)) (emit-sse-inst segment src dst #x66 #xd6 :operand-size :do-not-set))))) ;;; Instructions having an XMM register as the destination operand ;;; and a general-purpose register or a memory location as the source ;;; operand. The operand size is calculated from the source operand. ;;; MOVD - Move a 32- or 64-bit value from a general-purpose register or ;;; a memory location to the low order 32 or 64 bits of an XMM register ;;; with zero extension or vice versa. ;;; We do not support the MMX version of this instruction. (define-instruction movd (segment dst src) (:printer ext-xmm-reg/mem ((prefix #x66) (op #x6e))) (:printer ext-rex-xmm-reg/mem ((prefix #x66) (op #x6e))) (:printer ext-xmm-reg/mem ((prefix #x66) (op #x7e)) '(:name :tab reg/mem ", " reg)) (:printer ext-rex-xmm-reg/mem ((prefix #x66) (op #x7e)) '(:name :tab reg/mem ", " reg)) (:emitter (cond ((xmm-register-p dst) (emit-sse-inst segment dst src #x66 #x6e)) (t (aver (xmm-register-p src)) (emit-sse-inst segment src dst #x66 #x7e))))) (macrolet ((define-integer-source-sse-inst (name prefix opcode) `(define-instruction ,name (segment dst src) (:printer ext-xmm-reg/mem ((prefix ,prefix) (op ,opcode))) (:printer ext-rex-xmm-reg/mem ((prefix ,prefix) (op ,opcode))) (:emitter (aver (xmm-register-p dst)) (let ((src-size (operand-size src))) (aver (or (eq src-size :qword) (eq src-size :dword)))) (emit-sse-inst segment dst src ,prefix ,opcode))))) (define-integer-source-sse-inst cvtsi2sd #xf2 #x2a) (define-integer-source-sse-inst cvtsi2ss #xf3 #x2a)) ;;; Instructions having a general-purpose register as the destination ;;; operand and an XMM register or a memory location as the source ;;; operand. The operand size is calculated from the destination ;;; operand. (macrolet ((define-gpr-destination-sse-inst (name prefix opcode) `(define-instruction ,name (segment dst src) (:printer ext-reg-xmm/mem ((prefix ,prefix) (op ,opcode))) (:printer ext-rex-reg-xmm/mem ((prefix ,prefix) (op ,opcode))) (:emitter (aver (register-p dst)) (let ((dst-size (operand-size dst))) (aver (or (eq dst-size :qword) (eq dst-size :dword))) (emit-sse-inst segment dst src ,prefix ,opcode :operand-size dst-size)))))) (define-gpr-destination-sse-inst cvtsd2si #xf2 #x2d) (define-gpr-destination-sse-inst cvtss2si #xf3 #x2d) (define-gpr-destination-sse-inst cvttsd2si #xf2 #x2c) (define-gpr-destination-sse-inst cvttss2si #xf3 #x2c)) ;;; Other SSE instructions (define-instruction ldmxcsr (segment src) (:emitter (emit-byte segment #x0f) (emit-byte segment #xae) (emit-ea segment src 2))) (define-instruction stmxcsr (segment dst) (:emitter (emit-byte segment #x0f) (emit-byte segment #xae) (emit-ea segment dst 3)))