--- /dev/null
+#if defined(USE_X86_ASM)
+
+#include "x86sse.h"
+
+#define DISASSEM 0
+#define X86_TWOB 0x0f
+
+/* Emit bytes to the instruction stream:
+ */
+static void emit_1b( struct x86_function *p, GLbyte b0 )
+{
+ *(GLbyte *)(p->csr++) = b0;
+}
+
+static void emit_1i( struct x86_function *p, GLint i0 )
+{
+ *(GLint *)(p->csr) = i0;
+ p->csr += 4;
+}
+
+static void disassem( struct x86_function *p, const char *fn )
+{
+#if DISASSEM
+ static const char *last_fn;
+ if (fn && fn != last_fn) {
+ _mesa_printf("0x%x: %s\n", p->csr, fn);
+ last_fn = fn;
+ }
+#endif
+}
+
+static void emit_1ub_fn( struct x86_function *p, GLubyte b0, const char *fn )
+{
+ disassem(p, fn);
+ *(p->csr++) = b0;
+}
+
+static void emit_2ub_fn( struct x86_function *p, GLubyte b0, GLubyte b1, const char *fn )
+{
+ disassem(p, fn);
+ *(p->csr++) = b0;
+ *(p->csr++) = b1;
+}
+
+static void emit_3ub_fn( struct x86_function *p, GLubyte b0, GLubyte b1, GLubyte b2, const char *fn )
+{
+ disassem(p, fn);
+ *(p->csr++) = b0;
+ *(p->csr++) = b1;
+ *(p->csr++) = b2;
+}
+
+#define emit_1ub(p, b0) emit_1ub_fn(p, b0, __FUNCTION__)
+#define emit_2ub(p, b0, b1) emit_2ub_fn(p, b0, b1, __FUNCTION__)
+#define emit_3ub(p, b0, b1, b2) emit_3ub_fn(p, b0, b1, b2, __FUNCTION__)
+
+
+
+/* Build a modRM byte + possible displacement. No treatment of SIB
+ * indexing. BZZT - no way to encode an absolute address.
+ */
+static void emit_modrm( struct x86_function *p,
+ struct x86_reg reg,
+ struct x86_reg regmem )
+{
+ GLubyte val = 0;
+
+ assert(reg.mod == mod_REG);
+
+ val |= regmem.mod << 6; /* mod field */
+ val |= reg.idx << 3; /* reg field */
+ val |= regmem.idx; /* r/m field */
+
+ emit_1ub_fn(p, val, 0);
+
+ /* Oh-oh we've stumbled into the SIB thing.
+ */
+ if (regmem.idx == reg_SP) {
+ emit_1ub_fn(p, 0x24, 0); /* simplistic! */
+ }
+
+ switch (regmem.mod) {
+ case mod_REG:
+ case mod_INDIRECT:
+ break;
+ case mod_DISP8:
+ emit_1b(p, regmem.disp);
+ break;
+ case mod_DISP32:
+ emit_1i(p, regmem.disp);
+ break;
+ default:
+ assert(0);
+ break;
+ }
+}
+
+/* Many x86 instructions have two opcodes to cope with the situations
+ * where the destination is a register or memory reference
+ * respectively. This function selects the correct opcode based on
+ * the arguments presented.
+ */
+static void emit_op_modrm( struct x86_function *p,
+ GLubyte op_dst_is_reg,
+ GLubyte op_dst_is_mem,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ switch (dst.mod) {
+ case mod_REG:
+ emit_1ub_fn(p, op_dst_is_reg, 0);
+ emit_modrm(p, dst, src);
+ break;
+ case mod_INDIRECT:
+ case mod_DISP32:
+ case mod_DISP8:
+ assert(src.mod == mod_REG);
+ emit_1ub_fn(p, op_dst_is_mem, 0);
+ emit_modrm(p, src, dst);
+ break;
+ default:
+ assert(0);
+ break;
+ }
+}
+
+
+
+
+
+
+
+/* Create and manipulate registers and regmem values:
+ */
+struct x86_reg x86_make_reg( GLuint file,
+ GLuint idx )
+{
+ struct x86_reg reg;
+
+ reg.file = file;
+ reg.idx = idx;
+ reg.mod = mod_REG;
+ reg.disp = 0;
+
+ return reg;
+}
+
+struct x86_reg x86_make_disp( struct x86_reg reg,
+ GLint disp )
+{
+ assert(reg.file == file_REG32);
+
+ if (reg.mod == mod_REG)
+ reg.disp = disp;
+ else
+ reg.disp += disp;
+
+ if (reg.disp == 0)
+ reg.mod = mod_INDIRECT;
+ else if (reg.disp <= 127 && reg.disp >= -128)
+ reg.mod = mod_DISP8;
+ else
+ reg.mod = mod_DISP32;
+
+ return reg;
+}
+
+struct x86_reg x86_deref( struct x86_reg reg )
+{
+ return x86_make_disp(reg, 0);
+}
+
+struct x86_reg x86_get_base_reg( struct x86_reg reg )
+{
+ return x86_make_reg( reg.file, reg.idx );
+}
+
+
+
+/* Labels, jumps and fixup:
+ */
+GLubyte *x86_get_label( struct x86_function *p )
+{
+ return p->csr;
+}
+
+void x86_jcc( struct x86_function *p,
+ GLuint cc,
+ GLubyte *label )
+{
+ GLint offset = label - (x86_get_label(p) + 2);
+
+ if (offset <= 127 && offset >= -128) {
+ emit_1ub(p, 0x70 + cc);
+ emit_1b(p, (GLbyte) offset);
+ }
+ else {
+ offset = label - (x86_get_label(p) + 6);
+ emit_2ub(p, 0x0f, 0x80 + cc);
+ emit_1i(p, offset);
+ }
+}
+
+/* Always use a 32bit offset for forward jumps:
+ */
+GLubyte *x86_jcc_forward( struct x86_function *p,
+ GLuint cc )
+{
+ emit_2ub(p, 0x0f, 0x80 + cc);
+ emit_1i(p, 0);
+ return x86_get_label(p);
+}
+
+/* Fixup offset from forward jump:
+ */
+void x86_fixup_fwd_jump( struct x86_function *p,
+ GLubyte *fixup )
+{
+ *(int *)(fixup - 4) = x86_get_label(p) - fixup;
+}
+
+void x86_push( struct x86_function *p,
+ struct x86_reg reg )
+{
+ assert(reg.mod == mod_REG);
+ emit_1ub(p, 0x50 + reg.idx);
+ p->stack_offset += 4;
+}
+
+void x86_pop( struct x86_function *p,
+ struct x86_reg reg )
+{
+ assert(reg.mod == mod_REG);
+ emit_1ub(p, 0x58 + reg.idx);
+ p->stack_offset -= 4;
+}
+
+void x86_inc( struct x86_function *p,
+ struct x86_reg reg )
+{
+ assert(reg.mod == mod_REG);
+ emit_1ub(p, 0x40 + reg.idx);
+}
+
+void x86_dec( struct x86_function *p,
+ struct x86_reg reg )
+{
+ assert(reg.mod == mod_REG);
+ emit_1ub(p, 0x48 + reg.idx);
+}
+
+void x86_ret( struct x86_function *p )
+{
+ emit_1ub(p, 0xc3);
+}
+
+void mmx_emms( struct x86_function *p )
+{
+ assert(p->need_emms);
+ emit_2ub(p, 0x0f, 0x77);
+ p->need_emms = 0;
+}
+
+
+
+
+void x86_mov( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ emit_op_modrm( p, 0x8b, 0x89, dst, src );
+}
+
+void x86_xor( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ emit_op_modrm( p, 0x33, 0x31, dst, src );
+}
+
+void x86_cmp( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ emit_op_modrm( p, 0x3b, 0x39, dst, src );
+}
+
+void sse2_movd( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ emit_2ub(p, 0x66, X86_TWOB);
+ emit_op_modrm( p, 0x6e, 0x7e, dst, src );
+}
+
+void mmx_movd( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ p->need_emms = 1;
+ emit_1ub(p, X86_TWOB);
+ emit_op_modrm( p, 0x6e, 0x7e, dst, src );
+}
+
+void mmx_movq( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ p->need_emms = 1;
+ emit_1ub(p, X86_TWOB);
+ emit_op_modrm( p, 0x6f, 0x7f, dst, src );
+}
+
+
+void sse_movss( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ emit_2ub(p, 0xF3, X86_TWOB);
+ emit_op_modrm( p, 0x10, 0x11, dst, src );
+}
+
+void sse_movaps( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ emit_1ub(p, X86_TWOB);
+ emit_op_modrm( p, 0x28, 0x29, dst, src );
+}
+
+void sse_movups( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ emit_1ub(p, X86_TWOB);
+ emit_op_modrm( p, 0x10, 0x11, dst, src );
+}
+
+void sse_movhps( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ assert(dst.mod != mod_REG || src.mod != mod_REG);
+ emit_1ub(p, X86_TWOB);
+ emit_op_modrm( p, 0x16, 0x17, dst, src ); /* cf movlhps */
+}
+
+void sse_movlps( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ assert(dst.mod != mod_REG || src.mod != mod_REG);
+ emit_1ub(p, X86_TWOB);
+ emit_op_modrm( p, 0x12, 0x13, dst, src ); /* cf movhlps */
+}
+
+/* SSE operations often only have one format, with dest constrained to
+ * be a register:
+ */
+void sse_mulps( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ emit_2ub(p, X86_TWOB, 0x59);
+ emit_modrm( p, dst, src );
+}
+
+void sse_addps( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ emit_2ub(p, X86_TWOB, 0x58);
+ emit_modrm( p, dst, src );
+}
+
+void sse_movhlps( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ assert(dst.mod == mod_REG && src.mod == mod_REG);
+ emit_2ub(p, X86_TWOB, 0x12);
+ emit_modrm( p, dst, src );
+}
+
+void sse_movlhps( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ assert(dst.mod == mod_REG && src.mod == mod_REG);
+ emit_2ub(p, X86_TWOB, 0x16);
+ emit_modrm( p, dst, src );
+}
+
+void sse2_cvtps2dq( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ emit_3ub(p, 0x66, X86_TWOB, 0x5B);
+ emit_modrm( p, dst, src );
+}
+
+void sse2_packssdw( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ emit_3ub(p, 0x66, X86_TWOB, 0x6B);
+ emit_modrm( p, dst, src );
+}
+
+void sse2_packsswb( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ emit_3ub(p, 0x66, X86_TWOB, 0x63);
+ emit_modrm( p, dst, src );
+}
+
+void sse2_packuswb( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ emit_3ub(p, 0x66, X86_TWOB, 0x67);
+ emit_modrm( p, dst, src );
+}
+
+void sse_cvtps2pi( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ assert(dst.file == file_MMX &&
+ (src.file == file_XMM || src.mod != mod_REG));
+
+ p->need_emms = 1;
+
+ emit_2ub(p, X86_TWOB, 0x2d);
+ emit_modrm( p, dst, src );
+}
+
+void mmx_packssdw( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ assert(dst.file == file_MMX &&
+ (src.file == file_MMX || src.mod != mod_REG));
+
+ p->need_emms = 1;
+
+ emit_2ub(p, X86_TWOB, 0x6b);
+ emit_modrm( p, dst, src );
+}
+
+void mmx_packuswb( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ assert(dst.file == file_MMX &&
+ (src.file == file_MMX || src.mod != mod_REG));
+
+ p->need_emms = 1;
+
+ emit_2ub(p, X86_TWOB, 0x67);
+ emit_modrm( p, dst, src );
+}
+
+
+/* Load effective address:
+ */
+void x86_lea( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ emit_1ub(p, 0x8d);
+ emit_modrm( p, dst, src );
+}
+
+void x86_test( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src )
+{
+ emit_1ub(p, 0x85);
+ emit_modrm( p, dst, src );
+}
+
+
+/**
+ * Perform a reduced swizzle:
+ */
+void sse2_pshufd( struct x86_function *p,
+ struct x86_reg dest,
+ struct x86_reg arg0,
+ GLubyte x,
+ GLubyte y,
+ GLubyte z,
+ GLubyte w)
+{
+ emit_3ub(p, 0x66, X86_TWOB, 0x70);
+ emit_modrm(p, dest, arg0);
+ emit_1ub(p, (x|(y<<2)|(z<<4)|w<<6));
+}
+
+
+/* Shufps can also be used to implement a reduced swizzle when dest ==
+ * arg0.
+ */
+void sse_shufps( struct x86_function *p,
+ struct x86_reg dest,
+ struct x86_reg arg0,
+ GLubyte x,
+ GLubyte y,
+ GLubyte z,
+ GLubyte w)
+{
+ emit_2ub(p, X86_TWOB, 0xC6);
+ emit_modrm(p, dest, arg0);
+ emit_1ub(p, (x|(y<<2)|(z<<4)|w<<6));
+}
+
+
+/* Retreive a reference to one of the function arguments, taking into
+ * account any push/pop activity:
+ */
+struct x86_reg x86_fn_arg( struct x86_function *p,
+ GLuint arg )
+{
+ return x86_make_disp(x86_make_reg(file_REG32, reg_SP),
+ p->stack_offset + arg * 4); /* ??? */
+}
+
+
+void x86_init_func( struct x86_function *p )
+{
+ p->store = malloc(1024);
+ p->csr = p->store;
+}
+
+void x86_release_func( struct x86_function *p )
+{
+ free(p->store);
+}
+
+
+void (*x86_get_func( struct x86_function *p ))(void)
+{
+ return (void (*)())p->store;
+}
+
+#else
+
+void x86sse_dummy( void )
+{
+}
+
+#endif
--- /dev/null
+
+#ifndef _X86SSE_H_
+#define _X86SSE_H_
+
+#if defined(USE_X86_ASM)
+
+#include "glheader.h"
+
+/* It is up to the caller to ensure that instructions issued are
+ * suitable for the host cpu. There are no checks made in this module
+ * for mmx/sse/sse2 support on the cpu.
+ */
+struct x86_reg {
+ GLuint file:3;
+ GLuint idx:3;
+ GLuint mod:2; /* mod_REG if this is just a register */
+ GLint disp:24; /* only +/- 23bits of offset - should be enough... */
+};
+
+struct x86_function {
+ GLubyte *store;
+ GLubyte *csr;
+ GLuint stack_offset;
+ GLint need_emms;
+};
+
+enum x86_reg_file {
+ file_REG32,
+ file_MMX,
+ file_XMM
+};
+
+/* Values for mod field of modr/m byte
+ */
+enum x86_reg_mod {
+ mod_INDIRECT,
+ mod_DISP8,
+ mod_DISP32,
+ mod_REG
+};
+
+enum x86_reg_name {
+ reg_AX,
+ reg_CX,
+ reg_DX,
+ reg_BX,
+ reg_SP,
+ reg_BP,
+ reg_SI,
+ reg_DI
+};
+
+
+enum x86_cc {
+ cc_O, /* overflow */
+ cc_NO, /* not overflow */
+ cc_NAE, /* not above or equal / carry */
+ cc_AE, /* above or equal / not carry */
+ cc_E, /* equal / zero */
+ cc_NE /* not equal / not zero */
+};
+
+#define cc_Z cc_E
+#define cc_NZ cc_NE
+
+/* Begin/end/retreive function creation:
+ */
+
+
+void x86_init_func( struct x86_function *p );
+void x86_release_func( struct x86_function *p );
+void (*x86_get_func( struct x86_function *p ))( void );
+
+
+
+/* Create and manipulate registers and regmem values:
+ */
+struct x86_reg x86_make_reg( enum x86_reg_file file,
+ enum x86_reg_name idx );
+
+struct x86_reg x86_make_disp( struct x86_reg reg,
+ GLint disp );
+
+struct x86_reg x86_deref( struct x86_reg reg );
+
+struct x86_reg x86_get_base_reg( struct x86_reg reg );
+
+
+
+
+/* Labels, jumps and fixup:
+ */
+GLubyte *x86_get_label( struct x86_function *p );
+
+void x86_jcc( struct x86_function *p,
+ enum x86_cc cc,
+ GLubyte *label );
+
+/* Always use a 32bit offset for forward jumps:
+ */
+GLubyte *x86_jcc_forward( struct x86_function *p,
+ enum x86_cc cc );
+
+/* Fixup offset from forward jump:
+ */
+void x86_fixup_fwd_jump( struct x86_function *p,
+ GLubyte *fixup );
+
+void x86_push( struct x86_function *p,
+ struct x86_reg reg );
+
+void x86_pop( struct x86_function *p,
+ struct x86_reg reg );
+
+void x86_inc( struct x86_function *p,
+ struct x86_reg reg );
+
+void x86_dec( struct x86_function *p,
+ struct x86_reg reg );
+
+void x86_ret( struct x86_function *p );
+
+void mmx_emms( struct x86_function *p );
+
+void x86_mov( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void x86_xor( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void x86_cmp( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void sse2_movd( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void mmx_movd( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void mmx_movq( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void sse_movss( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void sse_movaps( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void sse_movups( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void sse_movhps( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void sse_movlps( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+/* SSE operations often only have one format, with dest constrained to
+ * be a register:
+ */
+void sse_mulps( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void sse_addps( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void sse_movhlps( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void sse_movlhps( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void sse2_cvtps2dq( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void sse2_packssdw( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void sse2_packsswb( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void sse2_packuswb( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void sse_cvtps2pi( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void mmx_packssdw( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void mmx_packuswb( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+
+/* Load effective address:
+ */
+void x86_lea( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+void x86_test( struct x86_function *p,
+ struct x86_reg dst,
+ struct x86_reg src );
+
+/* Perform a reduced swizzle in a single sse instruction:
+ */
+void sse2_pshufd( struct x86_function *p,
+ struct x86_reg dest,
+ struct x86_reg arg0,
+ GLubyte x,
+ GLubyte y,
+ GLubyte z,
+ GLubyte w );
+
+
+/* Shufps can also be used to implement a reduced swizzle when dest ==
+ * arg0.
+ */
+void sse_shufps( struct x86_function *p,
+ struct x86_reg dest,
+ struct x86_reg arg0,
+ GLubyte x,
+ GLubyte y,
+ GLubyte z,
+ GLubyte w );
+
+
+/* Retreive a reference to one of the function arguments, taking into
+ * account any push/pop activity:
+ */
+struct x86_reg x86_fn_arg( struct x86_function *p,
+ GLuint arg );
+
+#endif
+#endif
projects / mesa.git / commitdiff