-/* $Id: x86_xform4.S,v 1.1 2001/03/29 06:46:27 gareth Exp $ */
/*
* Mesa 3-D graphics library
- * Version: 3.5
*
* Copyright (C) 1999-2001 Brian Paul All Rights Reserved.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
* OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * BRIAN PAUL BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN
- * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
- * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
*/
-#include "matypes.h"
+/*
+ * NOTE: Avoid using spaces in between '(' ')' and arguments, especially
+ * with macros like CONST, LLBL that expand to CONCAT(...). Putting spaces
+ * in there will break the build on some platforms.
+ */
+
+#include "assyntax.h"
+#define MATH_ASM_PTR_SIZE 4
+#include "math/m_vector_asm.h"
#include "xform_args.h"
SEG_TEXT
#define FP_ONE 1065353216
#define FP_ZERO 0
-#define SRC(i) REGOFF(i * 4, ESI)
-#define DST(i) REGOFF(i * 4, EDI)
-#define MAT(i) REGOFF(i * 4, EDX)
+#define SRC0 REGOFF(0, ESI)
+#define SRC1 REGOFF(4, ESI)
+#define SRC2 REGOFF(8, ESI)
+#define SRC3 REGOFF(12, ESI)
+#define DST0 REGOFF(0, EDI)
+#define DST1 REGOFF(4, EDI)
+#define DST2 REGOFF(8, EDI)
+#define DST3 REGOFF(12, EDI)
+#define MAT0 REGOFF(0, EDX)
+#define MAT1 REGOFF(4, EDX)
+#define MAT2 REGOFF(8, EDX)
+#define MAT3 REGOFF(12, EDX)
+#define MAT4 REGOFF(16, EDX)
+#define MAT5 REGOFF(20, EDX)
+#define MAT6 REGOFF(24, EDX)
+#define MAT7 REGOFF(28, EDX)
+#define MAT8 REGOFF(32, EDX)
+#define MAT9 REGOFF(36, EDX)
+#define MAT10 REGOFF(40, EDX)
+#define MAT11 REGOFF(44, EDX)
+#define MAT12 REGOFF(48, EDX)
+#define MAT13 REGOFF(52, EDX)
+#define MAT14 REGOFF(56, EDX)
+#define MAT15 REGOFF(60, EDX)
ALIGNTEXT16
GLOBL GLNAME( _mesa_x86_transform_points4_general )
+HIDDEN(_mesa_x86_transform_points4_general)
GLNAME( _mesa_x86_transform_points4_general ):
#define FRAME_OFFSET 8
MOV_L( REGOFF(V4F_COUNT, ESI), ECX )
TEST_L( ECX, ECX )
- JZ( LLBL( x86_p4_gr_done ) )
+ JZ( LLBL(x86_p4_gr_done) )
MOV_L( REGOFF(V4F_STRIDE, ESI), EAX )
OR_L( CONST(VEC_SIZE_4), REGOFF(V4F_FLAGS, EDI) )
ADD_L( EDI, ECX )
ALIGNTEXT16
-LLBL( x86_p4_gr_loop ):
-
- FLD_S( SRC(0) ) /* F4 */
- FMUL_S( MAT(0) )
- FLD_S( SRC(0) ) /* F5 F4 */
- FMUL_S( MAT(1) )
- FLD_S( SRC(0) ) /* F6 F5 F4 */
- FMUL_S( MAT(2) )
- FLD_S( SRC(0) ) /* F7 F6 F5 F4 */
- FMUL_S( MAT(3) )
-
- FLD_S( SRC(1) ) /* F0 F7 F6 F5 F4 */
- FMUL_S( MAT(4) )
- FLD_S( SRC(1) ) /* F1 F0 F7 F6 F5 F4 */
- FMUL_S( MAT(5) )
- FLD_S( SRC(1) ) /* F2 F1 F0 F7 F6 F5 F4 */
- FMUL_S( MAT(6) )
- FLD_S( SRC(1) ) /* F3 F2 F1 F0 F7 F6 F5 F4 */
- FMUL_S( MAT(7) )
+LLBL(x86_p4_gr_loop):
+
+ FLD_S( SRC0 ) /* F4 */
+ FMUL_S( MAT0 )
+ FLD_S( SRC0 ) /* F5 F4 */
+ FMUL_S( MAT1 )
+ FLD_S( SRC0 ) /* F6 F5 F4 */
+ FMUL_S( MAT2 )
+ FLD_S( SRC0 ) /* F7 F6 F5 F4 */
+ FMUL_S( MAT3 )
+
+ FLD_S( SRC1 ) /* F0 F7 F6 F5 F4 */
+ FMUL_S( MAT4 )
+ FLD_S( SRC1 ) /* F1 F0 F7 F6 F5 F4 */
+ FMUL_S( MAT5 )
+ FLD_S( SRC1 ) /* F2 F1 F0 F7 F6 F5 F4 */
+ FMUL_S( MAT6 )
+ FLD_S( SRC1 ) /* F3 F2 F1 F0 F7 F6 F5 F4 */
+ FMUL_S( MAT7 )
FXCH( ST(3) ) /* F0 F2 F1 F3 F7 F6 F5 F4 */
- FADDP( ST(0), ST(7) ) /* F2 F1 F3 F7 F6 F5 F4 */
+ FADDP( ST0, ST(7) ) /* F2 F1 F3 F7 F6 F5 F4 */
FXCH( ST(1) ) /* F1 F2 F3 F7 F6 F5 F4 */
- FADDP( ST(0), ST(5) ) /* F2 F3 F7 F6 F5 F4 */
- FADDP( ST(0), ST(3) ) /* F3 F7 F6 F5 F4 */
- FADDP( ST(0), ST(1) ) /* F7 F6 F5 F4 */
-
- FLD_S( SRC(2) ) /* F0 F7 F6 F5 F4 */
- FMUL_S( MAT(8) )
- FLD_S( SRC(2) ) /* F1 F0 F7 F6 F5 F4 */
- FMUL_S( MAT(9) )
- FLD_S( SRC(2) ) /* F2 F1 F0 F7 F6 F5 F4 */
- FMUL_S( MAT(10) )
- FLD_S( SRC(2) ) /* F3 F2 F1 F0 F7 F6 F5 F4 */
- FMUL_S( MAT(11) )
+ FADDP( ST0, ST(5) ) /* F2 F3 F7 F6 F5 F4 */
+ FADDP( ST0, ST(3) ) /* F3 F7 F6 F5 F4 */
+ FADDP( ST0, ST(1) ) /* F7 F6 F5 F4 */
+
+ FLD_S( SRC2 ) /* F0 F7 F6 F5 F4 */
+ FMUL_S( MAT8 )
+ FLD_S( SRC2 ) /* F1 F0 F7 F6 F5 F4 */
+ FMUL_S( MAT9 )
+ FLD_S( SRC2 ) /* F2 F1 F0 F7 F6 F5 F4 */
+ FMUL_S( MAT10 )
+ FLD_S( SRC2 ) /* F3 F2 F1 F0 F7 F6 F5 F4 */
+ FMUL_S( MAT11 )
FXCH( ST(3) ) /* F0 F2 F1 F3 F7 F6 F5 F4 */
- FADDP( ST(0), ST(7) ) /* F2 F1 F3 F7 F6 F5 F4 */
+ FADDP( ST0, ST(7) ) /* F2 F1 F3 F7 F6 F5 F4 */
FXCH( ST(1) ) /* F1 F2 F3 F7 F6 F5 F4 */
- FADDP( ST(0), ST(5) ) /* F2 F3 F7 F6 F5 F4 */
- FADDP( ST(0), ST(3) ) /* F3 F7 F6 F5 F4 */
- FADDP( ST(0), ST(1) ) /* F7 F6 F5 F4 */
-
- FLD_S( SRC(3) ) /* F0 F7 F6 F5 F4 */
- FMUL_S( MAT(12) )
- FLD_S( SRC(3) ) /* F1 F0 F7 F6 F5 F4 */
- FMUL_S( MAT(13) )
- FLD_S( SRC(3) ) /* F2 F1 F0 F7 F6 F5 F4 */
- FMUL_S( MAT(14) )
- FLD_S( SRC(3) ) /* F3 F2 F1 F0 F7 F6 F5 F4 */
- FMUL_S( MAT(15) )
+ FADDP( ST0, ST(5) ) /* F2 F3 F7 F6 F5 F4 */
+ FADDP( ST0, ST(3) ) /* F3 F7 F6 F5 F4 */
+ FADDP( ST0, ST(1) ) /* F7 F6 F5 F4 */
+
+ FLD_S( SRC3 ) /* F0 F7 F6 F5 F4 */
+ FMUL_S( MAT12 )
+ FLD_S( SRC3 ) /* F1 F0 F7 F6 F5 F4 */
+ FMUL_S( MAT13 )
+ FLD_S( SRC3 ) /* F2 F1 F0 F7 F6 F5 F4 */
+ FMUL_S( MAT14 )
+ FLD_S( SRC3 ) /* F3 F2 F1 F0 F7 F6 F5 F4 */
+ FMUL_S( MAT15 )
FXCH( ST(3) ) /* F0 F2 F1 F3 F7 F6 F5 F4 */
- FADDP( ST(0), ST(7) ) /* F2 F1 F3 F7 F6 F5 F4 */
+ FADDP( ST0, ST(7) ) /* F2 F1 F3 F7 F6 F5 F4 */
FXCH( ST(1) ) /* F1 F2 F3 F7 F6 F5 F4 */
- FADDP( ST(0), ST(5) ) /* F2 F3 F7 F6 F5 F4 */
- FADDP( ST(0), ST(3) ) /* F3 F7 F6 F5 F4 */
- FADDP( ST(0), ST(1) ) /* F7 F6 F5 F4 */
+ FADDP( ST0, ST(5) ) /* F2 F3 F7 F6 F5 F4 */
+ FADDP( ST0, ST(3) ) /* F3 F7 F6 F5 F4 */
+ FADDP( ST0, ST(1) ) /* F7 F6 F5 F4 */
FXCH( ST(3) ) /* F4 F6 F5 F7 */
- FSTP_S( DST(0) ) /* F6 F5 F7 */
+ FSTP_S( DST0 ) /* F6 F5 F7 */
FXCH( ST(1) ) /* F5 F6 F7 */
- FSTP_S( DST(1) ) /* F6 F7 */
- FSTP_S( DST(2) ) /* F7 */
- FSTP_S( DST(3) ) /* */
+ FSTP_S( DST1 ) /* F6 F7 */
+ FSTP_S( DST2 ) /* F7 */
+ FSTP_S( DST3 ) /* */
-LLBL( x86_p4_gr_skip ):
+LLBL(x86_p4_gr_skip):
ADD_L( CONST(16), EDI )
ADD_L( EAX, ESI )
CMP_L( ECX, EDI )
- JNE( LLBL( x86_p4_gr_loop ) )
+ JNE( LLBL(x86_p4_gr_loop) )
-LLBL( x86_p4_gr_done ):
+LLBL(x86_p4_gr_done):
POP_L( EDI )
POP_L( ESI )
ALIGNTEXT16
GLOBL GLNAME( _mesa_x86_transform_points4_perspective )
+HIDDEN(_mesa_x86_transform_points4_perspective)
GLNAME( _mesa_x86_transform_points4_perspective ):
#define FRAME_OFFSET 12
MOV_L( REGOFF(V4F_COUNT, ESI), ECX )
TEST_L( ECX, ECX )
- JZ( LLBL( x86_p4_pr_done ) )
+ JZ( LLBL(x86_p4_pr_done) )
MOV_L( REGOFF(V4F_STRIDE, ESI), EAX )
OR_L( CONST(VEC_SIZE_4), REGOFF(V4F_FLAGS, EDI) )
ADD_L( EDI, ECX )
ALIGNTEXT16
-LLBL( x86_p4_pr_loop ):
+LLBL(x86_p4_pr_loop):
- FLD_S( SRC(0) ) /* F4 */
- FMUL_S( MAT(0) )
+ FLD_S( SRC0 ) /* F4 */
+ FMUL_S( MAT0 )
- FLD_S( SRC(1) ) /* F5 F4 */
- FMUL_S( MAT(5) )
+ FLD_S( SRC1 ) /* F5 F4 */
+ FMUL_S( MAT5 )
- FLD_S( SRC(2) ) /* F0 F5 F4 */
- FMUL_S( MAT(8) )
- FLD_S( SRC(2) ) /* F1 F0 F5 F4 */
- FMUL_S( MAT(9) )
- FLD_S( SRC(2) ) /* F6 F1 F0 F5 F4 */
- FMUL_S( MAT(10) )
+ FLD_S( SRC2 ) /* F0 F5 F4 */
+ FMUL_S( MAT8 )
+ FLD_S( SRC2 ) /* F1 F0 F5 F4 */
+ FMUL_S( MAT9 )
+ FLD_S( SRC2 ) /* F6 F1 F0 F5 F4 */
+ FMUL_S( MAT10 )
FXCH( ST(2) ) /* F0 F1 F6 F5 F4 */
- FADDP( ST(0), ST(4) ) /* F1 F6 F5 F4 */
- FADDP( ST(0), ST(2) ) /* F6 F5 F4 */
+ FADDP( ST0, ST(4) ) /* F1 F6 F5 F4 */
+ FADDP( ST0, ST(2) ) /* F6 F5 F4 */
- FLD_S( SRC(3) ) /* F2 F6 F5 F4 */
- FMUL_S( MAT(14) )
+ FLD_S( SRC3 ) /* F2 F6 F5 F4 */
+ FMUL_S( MAT14 )
- FADDP( ST(0), ST(1) ) /* F6 F5 F4 */
+ FADDP( ST0, ST(1) ) /* F6 F5 F4 */
- MOV_L( SRC(2), EBX )
+ MOV_L( SRC2, EBX )
XOR_L( CONST(-2147483648), EBX )/* change sign */
FXCH( ST(2) ) /* F4 F5 F6 */
- FSTP_S( DST(0) ) /* F5 F6 */
- FSTP_S( DST(1) ) /* F6 */
- FSTP_S( DST(2) ) /* */
- MOV_L( EBX, DST(3) )
+ FSTP_S( DST0 ) /* F5 F6 */
+ FSTP_S( DST1 ) /* F6 */
+ FSTP_S( DST2 ) /* */
+ MOV_L( EBX, DST3 )
-LLBL( x86_p4_pr_skip ):
+LLBL(x86_p4_pr_skip):
ADD_L( CONST(16), EDI )
ADD_L( EAX, ESI )
CMP_L( ECX, EDI )
- JNE( LLBL( x86_p4_pr_loop ) )
+ JNE( LLBL(x86_p4_pr_loop) )
-LLBL( x86_p4_pr_done ):
+LLBL(x86_p4_pr_done):
POP_L( EBX )
POP_L( EDI )
ALIGNTEXT16
GLOBL GLNAME( _mesa_x86_transform_points4_3d )
+HIDDEN(_mesa_x86_transform_points4_3d)
GLNAME( _mesa_x86_transform_points4_3d ):
#define FRAME_OFFSET 12
MOV_L( REGOFF(V4F_COUNT, ESI), ECX )
TEST_L( ECX, ECX )
- JZ( LLBL( x86_p4_3dr_done ) )
+ JZ( LLBL(x86_p4_3dr_done) )
MOV_L( REGOFF(V4F_STRIDE, ESI), EAX )
OR_L( CONST(VEC_SIZE_4), REGOFF(V4F_FLAGS, EDI) )
ADD_L( EDI, ECX )
ALIGNTEXT16
-LLBL( x86_p4_3dr_loop ):
-
- FLD_S( SRC(0) ) /* F4 */
- FMUL_S( MAT(0) )
- FLD_S( SRC(0) ) /* F5 F4 */
- FMUL_S( MAT(1) )
- FLD_S( SRC(0) ) /* F6 F5 F4 */
- FMUL_S( MAT(2) )
-
- FLD_S( SRC(1) ) /* F0 F6 F5 F4 */
- FMUL_S( MAT(4) )
- FLD_S( SRC(1) ) /* F1 F0 F6 F5 F4 */
- FMUL_S( MAT(5) )
- FLD_S( SRC(1) ) /* F2 F1 F0 F6 F5 F4 */
- FMUL_S( MAT(6) )
+LLBL(x86_p4_3dr_loop):
+
+ FLD_S( SRC0 ) /* F4 */
+ FMUL_S( MAT0 )
+ FLD_S( SRC0 ) /* F5 F4 */
+ FMUL_S( MAT1 )
+ FLD_S( SRC0 ) /* F6 F5 F4 */
+ FMUL_S( MAT2 )
+
+ FLD_S( SRC1 ) /* F0 F6 F5 F4 */
+ FMUL_S( MAT4 )
+ FLD_S( SRC1 ) /* F1 F0 F6 F5 F4 */
+ FMUL_S( MAT5 )
+ FLD_S( SRC1 ) /* F2 F1 F0 F6 F5 F4 */
+ FMUL_S( MAT6 )
FXCH( ST(2) ) /* F0 F1 F2 F6 F5 F4 */
- FADDP( ST(0), ST(5) ) /* F1 F2 F6 F5 F4 */
- FADDP( ST(0), ST(3) ) /* F2 F6 F5 F4 */
- FADDP( ST(0), ST(1) ) /* F6 F5 F4 */
+ FADDP( ST0, ST(5) ) /* F1 F2 F6 F5 F4 */
+ FADDP( ST0, ST(3) ) /* F2 F6 F5 F4 */
+ FADDP( ST0, ST(1) ) /* F6 F5 F4 */
- FLD_S( SRC(2) ) /* F0 F6 F5 F4 */
- FMUL_S( MAT(8) )
- FLD_S( SRC(2) ) /* F1 F0 F6 F5 F4 */
- FMUL_S( MAT(9) )
- FLD_S( SRC(2) ) /* F2 F1 F0 F6 F5 F4 */
- FMUL_S( MAT(10) )
+ FLD_S( SRC2 ) /* F0 F6 F5 F4 */
+ FMUL_S( MAT8 )
+ FLD_S( SRC2 ) /* F1 F0 F6 F5 F4 */
+ FMUL_S( MAT9 )
+ FLD_S( SRC2 ) /* F2 F1 F0 F6 F5 F4 */
+ FMUL_S( MAT10 )
FXCH( ST(2) ) /* F0 F1 F2 F6 F5 F4 */
- FADDP( ST(0), ST(5) ) /* F1 F2 F6 F5 F4 */
- FADDP( ST(0), ST(3) ) /* F2 F6 F5 F4 */
- FADDP( ST(0), ST(1) ) /* F6 F5 F4 */
+ FADDP( ST0, ST(5) ) /* F1 F2 F6 F5 F4 */
+ FADDP( ST0, ST(3) ) /* F2 F6 F5 F4 */
+ FADDP( ST0, ST(1) ) /* F6 F5 F4 */
- FLD_S( SRC(3) ) /* F0 F6 F5 F4 */
- FMUL_S( MAT(12) )
- FLD_S( SRC(3) ) /* F1 F0 F6 F5 F4 */
- FMUL_S( MAT(13) )
- FLD_S( SRC(3) ) /* F2 F1 F0 F6 F5 F4 */
- FMUL_S( MAT(14) )
+ FLD_S( SRC3 ) /* F0 F6 F5 F4 */
+ FMUL_S( MAT12 )
+ FLD_S( SRC3 ) /* F1 F0 F6 F5 F4 */
+ FMUL_S( MAT13 )
+ FLD_S( SRC3 ) /* F2 F1 F0 F6 F5 F4 */
+ FMUL_S( MAT14 )
FXCH( ST(2) ) /* F0 F1 F2 F6 F5 F4 */
- FADDP( ST(0), ST(5) ) /* F1 F2 F6 F5 F4 */
- FADDP( ST(0), ST(3) ) /* F2 F6 F5 F4 */
- FADDP( ST(0), ST(1) ) /* F6 F5 F4 */
+ FADDP( ST0, ST(5) ) /* F1 F2 F6 F5 F4 */
+ FADDP( ST0, ST(3) ) /* F2 F6 F5 F4 */
+ FADDP( ST0, ST(1) ) /* F6 F5 F4 */
- MOV_L( SRC(3), EBX )
+ MOV_L( SRC3, EBX )
FXCH( ST(2) ) /* F4 F5 F6 */
- FSTP_S( DST(0) ) /* F5 F6 */
- FSTP_S( DST(1) ) /* F6 */
- FSTP_S( DST(2) ) /* */
- MOV_L( EBX, DST(3) )
+ FSTP_S( DST0 ) /* F5 F6 */
+ FSTP_S( DST1 ) /* F6 */
+ FSTP_S( DST2 ) /* */
+ MOV_L( EBX, DST3 )
-LLBL( x86_p4_3dr_skip ):
+LLBL(x86_p4_3dr_skip):
ADD_L( CONST(16), EDI )
ADD_L( EAX, ESI )
CMP_L( ECX, EDI )
- JNE( LLBL( x86_p4_3dr_loop ) )
+ JNE( LLBL(x86_p4_3dr_loop) )
-LLBL( x86_p4_3dr_done ):
+LLBL(x86_p4_3dr_done):
POP_L( EBX )
POP_L( EDI )
ALIGNTEXT16
GLOBL GLNAME(_mesa_x86_transform_points4_3d_no_rot)
+HIDDEN(_mesa_x86_transform_points4_3d_no_rot)
GLNAME(_mesa_x86_transform_points4_3d_no_rot):
#define FRAME_OFFSET 12
MOV_L( REGOFF(V4F_COUNT, ESI), ECX )
TEST_L( ECX, ECX )
- JZ( LLBL( x86_p4_3dnrr_done ) )
+ JZ( LLBL(x86_p4_3dnrr_done) )
MOV_L( REGOFF(V4F_STRIDE, ESI), EAX )
OR_L( CONST(VEC_SIZE_4), REGOFF(V4F_FLAGS, EDI) )
ADD_L( EDI, ECX )
ALIGNTEXT16
-LLBL( x86_p4_3dnrr_loop ):
+LLBL(x86_p4_3dnrr_loop):
- FLD_S( SRC(0) ) /* F4 */
- FMUL_S( MAT(0) )
+ FLD_S( SRC0 ) /* F4 */
+ FMUL_S( MAT0 )
- FLD_S( SRC(1) ) /* F5 F4 */
- FMUL_S( MAT(5) )
+ FLD_S( SRC1 ) /* F5 F4 */
+ FMUL_S( MAT5 )
- FLD_S( SRC(2) ) /* F6 F5 F4 */
- FMUL_S( MAT(10) )
+ FLD_S( SRC2 ) /* F6 F5 F4 */
+ FMUL_S( MAT10 )
- FLD_S( SRC(3) ) /* F0 F6 F5 F4 */
- FMUL_S( MAT(12) )
- FLD_S( SRC(3) ) /* F1 F0 F6 F5 F4 */
- FMUL_S( MAT(13) )
- FLD_S( SRC(3) ) /* F2 F1 F0 F6 F5 F4 */
- FMUL_S( MAT(14) )
+ FLD_S( SRC3 ) /* F0 F6 F5 F4 */
+ FMUL_S( MAT12 )
+ FLD_S( SRC3 ) /* F1 F0 F6 F5 F4 */
+ FMUL_S( MAT13 )
+ FLD_S( SRC3 ) /* F2 F1 F0 F6 F5 F4 */
+ FMUL_S( MAT14 )
FXCH( ST(2) ) /* F0 F1 F2 F6 F5 F4 */
- FADDP( ST(0), ST(5) ) /* F1 F2 F6 F5 F4 */
- FADDP( ST(0), ST(3) ) /* F2 F6 F5 F4 */
- FADDP( ST(0), ST(1) ) /* F6 F5 F4 */
+ FADDP( ST0, ST(5) ) /* F1 F2 F6 F5 F4 */
+ FADDP( ST0, ST(3) ) /* F2 F6 F5 F4 */
+ FADDP( ST0, ST(1) ) /* F6 F5 F4 */
- MOV_L( SRC(3), EBX )
+ MOV_L( SRC3, EBX )
FXCH( ST(2) ) /* F4 F5 F6 */
- FSTP_S( DST(0) ) /* F5 F6 */
- FSTP_S( DST(1) ) /* F6 */
- FSTP_S( DST(2) ) /* */
- MOV_L( EBX, DST(3) )
+ FSTP_S( DST0 ) /* F5 F6 */
+ FSTP_S( DST1 ) /* F6 */
+ FSTP_S( DST2 ) /* */
+ MOV_L( EBX, DST3 )
-LLBL( x86_p4_3dnrr_skip ):
+LLBL(x86_p4_3dnrr_skip):
ADD_L( CONST(16), EDI )
ADD_L( EAX, ESI )
CMP_L( ECX, EDI )
- JNE( LLBL( x86_p4_3dnrr_loop ) )
+ JNE( LLBL(x86_p4_3dnrr_loop) )
-LLBL( x86_p4_3dnrr_done ):
+LLBL(x86_p4_3dnrr_done):
POP_L( EBX )
POP_L( EDI )
ALIGNTEXT16
GLOBL GLNAME( _mesa_x86_transform_points4_2d )
+HIDDEN(_mesa_x86_transform_points4_2d)
GLNAME( _mesa_x86_transform_points4_2d ):
#define FRAME_OFFSET 16
MOV_L( REGOFF(V4F_COUNT, ESI), ECX )
TEST_L( ECX, ECX )
- JZ( LLBL( x86_p4_2dr_done ) )
+ JZ( LLBL(x86_p4_2dr_done) )
MOV_L( REGOFF(V4F_STRIDE, ESI), EAX )
OR_L( CONST(VEC_SIZE_4), REGOFF(V4F_FLAGS, EDI) )
ADD_L( EDI, ECX )
ALIGNTEXT16
-LLBL( x86_p4_2dr_loop ):
+LLBL(x86_p4_2dr_loop):
- FLD_S( SRC(0) ) /* F4 */
- FMUL_S( MAT(0) )
- FLD_S( SRC(0) ) /* F5 F4 */
- FMUL_S( MAT(1) )
+ FLD_S( SRC0 ) /* F4 */
+ FMUL_S( MAT0 )
+ FLD_S( SRC0 ) /* F5 F4 */
+ FMUL_S( MAT1 )
- FLD_S( SRC(1) ) /* F0 F5 F4 */
- FMUL_S( MAT(4) )
- FLD_S( SRC(1) ) /* F1 F0 F5 F4 */
- FMUL_S( MAT(5) )
+ FLD_S( SRC1 ) /* F0 F5 F4 */
+ FMUL_S( MAT4 )
+ FLD_S( SRC1 ) /* F1 F0 F5 F4 */
+ FMUL_S( MAT5 )
FXCH( ST(1) ) /* F0 F1 F5 F4 */
- FADDP( ST(0), ST(3) ) /* F1 F5 F4 */
- FADDP( ST(0), ST(1) ) /* F5 F4 */
+ FADDP( ST0, ST(3) ) /* F1 F5 F4 */
+ FADDP( ST0, ST(1) ) /* F5 F4 */
- FLD_S( SRC(3) ) /* F0 F5 F4 */
- FMUL_S( MAT(12) )
- FLD_S( SRC(3) ) /* F1 F0 F5 F4 */
- FMUL_S( MAT(13) )
+ FLD_S( SRC3 ) /* F0 F5 F4 */
+ FMUL_S( MAT12 )
+ FLD_S( SRC3 ) /* F1 F0 F5 F4 */
+ FMUL_S( MAT13 )
FXCH( ST(1) ) /* F0 F1 F5 F4 */
- FADDP( ST(0), ST(3) ) /* F1 F5 F4 */
- FADDP( ST(0), ST(1) ) /* F5 F4 */
+ FADDP( ST0, ST(3) ) /* F1 F5 F4 */
+ FADDP( ST0, ST(1) ) /* F5 F4 */
- MOV_L( SRC(2), EBX )
- MOV_L( SRC(3), EBP )
+ MOV_L( SRC2, EBX )
+ MOV_L( SRC3, EBP )
FXCH( ST(1) ) /* F4 F5 */
- FSTP_S( DST(0) ) /* F5 */
- FSTP_S( DST(1) ) /* */
- MOV_L( EBX, DST(2) )
- MOV_L( EBP, DST(3) )
+ FSTP_S( DST0 ) /* F5 */
+ FSTP_S( DST1 ) /* */
+ MOV_L( EBX, DST2 )
+ MOV_L( EBP, DST3 )
-LLBL( x86_p4_2dr_skip ):
+LLBL(x86_p4_2dr_skip):
ADD_L( CONST(16), EDI )
ADD_L( EAX, ESI )
CMP_L( ECX, EDI )
- JNE( LLBL( x86_p4_2dr_loop ) )
+ JNE( LLBL(x86_p4_2dr_loop) )
-LLBL( x86_p4_2dr_done ):
+LLBL(x86_p4_2dr_done):
POP_L( EBP )
POP_L( EBX )
ALIGNTEXT16
GLOBL GLNAME( _mesa_x86_transform_points4_2d_no_rot )
+HIDDEN(_mesa_x86_transform_points4_2d_no_rot)
GLNAME( _mesa_x86_transform_points4_2d_no_rot ):
#define FRAME_OFFSET 16
MOV_L( REGOFF(V4F_COUNT, ESI), ECX )
TEST_L( ECX, ECX )
- JZ( LLBL( x86_p4_2dnrr_done ) )
+ JZ( LLBL(x86_p4_2dnrr_done) )
MOV_L( REGOFF(V4F_STRIDE, ESI), EAX )
OR_L( CONST(VEC_SIZE_4), REGOFF(V4F_FLAGS, EDI) )
ADD_L( EDI, ECX )
ALIGNTEXT16
-LLBL( x86_p4_2dnrr_loop ):
+LLBL(x86_p4_2dnrr_loop):
- FLD_S( SRC(0) ) /* F4 */
- FMUL_S( MAT(0) )
+ FLD_S( SRC0 ) /* F4 */
+ FMUL_S( MAT0 )
- FLD_S( SRC(1) ) /* F5 F4 */
- FMUL_S( MAT(5) )
+ FLD_S( SRC1 ) /* F5 F4 */
+ FMUL_S( MAT5 )
- FLD_S( SRC(3) ) /* F0 F5 F4 */
- FMUL_S( MAT(12) )
- FLD_S( SRC(3) ) /* F1 F0 F5 F4 */
- FMUL_S( MAT(13) )
+ FLD_S( SRC3 ) /* F0 F5 F4 */
+ FMUL_S( MAT12 )
+ FLD_S( SRC3 ) /* F1 F0 F5 F4 */
+ FMUL_S( MAT13 )
FXCH( ST(1) ) /* F0 F1 F5 F4 */
- FADDP( ST(0), ST(3) ) /* F1 F5 F4 */
- FADDP( ST(0), ST(1) ) /* F5 F4 */
+ FADDP( ST0, ST(3) ) /* F1 F5 F4 */
+ FADDP( ST0, ST(1) ) /* F5 F4 */
- MOV_L( SRC(2), EBX )
- MOV_L( SRC(3), EBP )
+ MOV_L( SRC2, EBX )
+ MOV_L( SRC3, EBP )
FXCH( ST(1) ) /* F4 F5 */
- FSTP_S( DST(0) ) /* F5 */
- FSTP_S( DST(1) ) /* */
- MOV_L( EBX, DST(2) )
- MOV_L( EBP, DST(3) )
+ FSTP_S( DST0 ) /* F5 */
+ FSTP_S( DST1 ) /* */
+ MOV_L( EBX, DST2 )
+ MOV_L( EBP, DST3 )
-LLBL( x86_p4_2dnrr_skip ):
+LLBL(x86_p4_2dnrr_skip):
ADD_L( CONST(16), EDI )
ADD_L( EAX, ESI )
CMP_L( ECX, EDI )
- JNE( LLBL( x86_p4_2dnrr_loop ) )
+ JNE( LLBL(x86_p4_2dnrr_loop) )
-LLBL( x86_p4_2dnrr_done ):
+LLBL(x86_p4_2dnrr_done):
POP_L( EBP )
POP_L( EBX )
ALIGNTEXT16
GLOBL GLNAME( _mesa_x86_transform_points4_identity )
+HIDDEN(_mesa_x86_transform_points4_identity)
GLNAME( _mesa_x86_transform_points4_identity ):
#define FRAME_OFFSET 12
MOV_L( REGOFF(V4F_COUNT, ESI), ECX )
TEST_L( ECX, ECX )
- JZ( LLBL( x86_p4_ir_done ) )
+ JZ( LLBL(x86_p4_ir_done) )
MOV_L( REGOFF(V4F_STRIDE, ESI), EAX )
OR_L( CONST(VEC_SIZE_4), REGOFF(V4F_FLAGS, EDI) )
ADD_L( EDI, ECX )
CMP_L( ESI, EDI )
- JE( LLBL( x86_p4_ir_done ) )
+ JE( LLBL(x86_p4_ir_done) )
ALIGNTEXT16
-LLBL( x86_p4_ir_loop ):
+LLBL(x86_p4_ir_loop):
- MOV_L( SRC(0), EBX )
- MOV_L( SRC(1), EDX )
+ MOV_L( SRC0, EBX )
+ MOV_L( SRC1, EDX )
- MOV_L( EBX, DST(0) )
- MOV_L( EDX, DST(1) )
+ MOV_L( EBX, DST0 )
+ MOV_L( EDX, DST1 )
- MOV_L( SRC(2), EBX )
- MOV_L( SRC(3), EDX )
+ MOV_L( SRC2, EBX )
+ MOV_L( SRC3, EDX )
- MOV_L( EBX, DST(2) )
- MOV_L( EDX, DST(3) )
+ MOV_L( EBX, DST2 )
+ MOV_L( EDX, DST3 )
-LLBL( x86_p4_ir_skip ):
+LLBL(x86_p4_ir_skip):
ADD_L( CONST(16), EDI )
ADD_L( EAX, ESI )
CMP_L( ECX, EDI )
- JNE( LLBL( x86_p4_ir_loop ) )
+ JNE( LLBL(x86_p4_ir_loop) )
-LLBL( x86_p4_ir_done ):
+LLBL(x86_p4_ir_done):
POP_L( EBX )
POP_L( EDI )
POP_L( ESI )
RET
+
+#if defined (__ELF__) && defined (__linux__)
+ .section .note.GNU-stack,"",%progbits
+#endif
projects / mesa.git / blobdiff