From 5cd272085bbc905532869f3e1fd18a7100496b56 Mon Sep 17 00:00:00 2001 From: Kumar Gala Date: Mon, 27 Mar 2006 23:43:27 -0600 Subject: powerpc: move math-emu over to arch/powerpc Towards the goal of having arch/powerpc not build anything over in arch/ppc move math-emu over. Also, killed some references to arch/ppc/ in the arch/powerpc Makefile which should belong in drivers/ when the particular sub-arch's move over to arch/powerpc. Signed-off-by: Kumar Gala --- arch/powerpc/Makefile | 6 +- arch/powerpc/math-emu/Makefile | 13 + arch/powerpc/math-emu/double.h | 129 +++++++ arch/powerpc/math-emu/fabs.c | 18 + arch/powerpc/math-emu/fadd.c | 38 ++ arch/powerpc/math-emu/fadds.c | 39 ++ arch/powerpc/math-emu/fcmpo.c | 46 +++ arch/powerpc/math-emu/fcmpu.c | 42 +++ arch/powerpc/math-emu/fctiw.c | 25 ++ arch/powerpc/math-emu/fctiwz.c | 32 ++ arch/powerpc/math-emu/fdiv.c | 53 +++ arch/powerpc/math-emu/fdivs.c | 55 +++ arch/powerpc/math-emu/fmadd.c | 48 +++ arch/powerpc/math-emu/fmadds.c | 49 +++ arch/powerpc/math-emu/fmr.c | 18 + arch/powerpc/math-emu/fmsub.c | 51 +++ arch/powerpc/math-emu/fmsubs.c | 52 +++ arch/powerpc/math-emu/fmul.c | 42 +++ arch/powerpc/math-emu/fmuls.c | 43 +++ arch/powerpc/math-emu/fnabs.c | 18 + arch/powerpc/math-emu/fneg.c | 18 + arch/powerpc/math-emu/fnmadd.c | 51 +++ arch/powerpc/math-emu/fnmadds.c | 52 +++ arch/powerpc/math-emu/fnmsub.c | 54 +++ arch/powerpc/math-emu/fnmsubs.c | 55 +++ arch/powerpc/math-emu/fres.c | 12 + arch/powerpc/math-emu/frsp.c | 25 ++ arch/powerpc/math-emu/frsqrte.c | 12 + arch/powerpc/math-emu/fsel.c | 38 ++ arch/powerpc/math-emu/fsqrt.c | 37 ++ arch/powerpc/math-emu/fsqrts.c | 38 ++ arch/powerpc/math-emu/fsub.c | 41 +++ arch/powerpc/math-emu/fsubs.c | 42 +++ arch/powerpc/math-emu/lfd.c | 19 + arch/powerpc/math-emu/lfs.c | 37 ++ arch/powerpc/math-emu/math.c | 483 +++++++++++++++++++++++++ arch/powerpc/math-emu/mcrfs.c | 31 ++ arch/powerpc/math-emu/mffs.c | 17 + arch/powerpc/math-emu/mtfsb0.c | 18 + arch/powerpc/math-emu/mtfsb1.c | 18 + arch/powerpc/math-emu/mtfsf.c | 45 +++ arch/powerpc/math-emu/mtfsfi.c | 23 ++ arch/powerpc/math-emu/op-1.h | 245 +++++++++++++ arch/powerpc/math-emu/op-2.h | 433 +++++++++++++++++++++++ arch/powerpc/math-emu/op-4.h | 297 ++++++++++++++++ arch/powerpc/math-emu/op-common.h | 688 ++++++++++++++++++++++++++++++++++++ arch/powerpc/math-emu/sfp-machine.h | 377 ++++++++++++++++++++ arch/powerpc/math-emu/single.h | 66 ++++ arch/powerpc/math-emu/soft-fp.h | 104 ++++++ arch/powerpc/math-emu/stfd.c | 20 ++ arch/powerpc/math-emu/stfiwx.c | 16 + arch/powerpc/math-emu/stfs.c | 41 +++ arch/powerpc/math-emu/types.c | 51 +++ arch/powerpc/math-emu/udivmodti4.c | 191 ++++++++++ 54 files changed, 4507 insertions(+), 5 deletions(-) create mode 100644 arch/powerpc/math-emu/Makefile create mode 100644 arch/powerpc/math-emu/double.h create mode 100644 arch/powerpc/math-emu/fabs.c create mode 100644 arch/powerpc/math-emu/fadd.c create mode 100644 arch/powerpc/math-emu/fadds.c create mode 100644 arch/powerpc/math-emu/fcmpo.c create mode 100644 arch/powerpc/math-emu/fcmpu.c create mode 100644 arch/powerpc/math-emu/fctiw.c create mode 100644 arch/powerpc/math-emu/fctiwz.c create mode 100644 arch/powerpc/math-emu/fdiv.c create mode 100644 arch/powerpc/math-emu/fdivs.c create mode 100644 arch/powerpc/math-emu/fmadd.c create mode 100644 arch/powerpc/math-emu/fmadds.c create mode 100644 arch/powerpc/math-emu/fmr.c create mode 100644 arch/powerpc/math-emu/fmsub.c create mode 100644 arch/powerpc/math-emu/fmsubs.c create mode 100644 arch/powerpc/math-emu/fmul.c create mode 100644 arch/powerpc/math-emu/fmuls.c create mode 100644 arch/powerpc/math-emu/fnabs.c create mode 100644 arch/powerpc/math-emu/fneg.c create mode 100644 arch/powerpc/math-emu/fnmadd.c create mode 100644 arch/powerpc/math-emu/fnmadds.c create mode 100644 arch/powerpc/math-emu/fnmsub.c create mode 100644 arch/powerpc/math-emu/fnmsubs.c create mode 100644 arch/powerpc/math-emu/fres.c create mode 100644 arch/powerpc/math-emu/frsp.c create mode 100644 arch/powerpc/math-emu/frsqrte.c create mode 100644 arch/powerpc/math-emu/fsel.c create mode 100644 arch/powerpc/math-emu/fsqrt.c create mode 100644 arch/powerpc/math-emu/fsqrts.c create mode 100644 arch/powerpc/math-emu/fsub.c create mode 100644 arch/powerpc/math-emu/fsubs.c create mode 100644 arch/powerpc/math-emu/lfd.c create mode 100644 arch/powerpc/math-emu/lfs.c create mode 100644 arch/powerpc/math-emu/math.c create mode 100644 arch/powerpc/math-emu/mcrfs.c create mode 100644 arch/powerpc/math-emu/mffs.c create mode 100644 arch/powerpc/math-emu/mtfsb0.c create mode 100644 arch/powerpc/math-emu/mtfsb1.c create mode 100644 arch/powerpc/math-emu/mtfsf.c create mode 100644 arch/powerpc/math-emu/mtfsfi.c create mode 100644 arch/powerpc/math-emu/op-1.h create mode 100644 arch/powerpc/math-emu/op-2.h create mode 100644 arch/powerpc/math-emu/op-4.h create mode 100644 arch/powerpc/math-emu/op-common.h create mode 100644 arch/powerpc/math-emu/sfp-machine.h create mode 100644 arch/powerpc/math-emu/single.h create mode 100644 arch/powerpc/math-emu/soft-fp.h create mode 100644 arch/powerpc/math-emu/stfd.c create mode 100644 arch/powerpc/math-emu/stfiwx.c create mode 100644 arch/powerpc/math-emu/stfs.c create mode 100644 arch/powerpc/math-emu/types.c create mode 100644 arch/powerpc/math-emu/udivmodti4.c (limited to 'arch/powerpc') diff --git a/arch/powerpc/Makefile b/arch/powerpc/Makefile index 9586899de68..6ec84d37a33 100644 --- a/arch/powerpc/Makefile +++ b/arch/powerpc/Makefile @@ -129,12 +129,8 @@ core-y += arch/powerpc/kernel/ \ arch/powerpc/lib/ \ arch/powerpc/sysdev/ \ arch/powerpc/platforms/ -core-$(CONFIG_MATH_EMULATION) += arch/ppc/math-emu/ +core-$(CONFIG_MATH_EMULATION) += arch/powerpc/math-emu/ core-$(CONFIG_XMON) += arch/powerpc/xmon/ -core-$(CONFIG_APUS) += arch/ppc/amiga/ -drivers-$(CONFIG_8xx) += arch/ppc/8xx_io/ -drivers-$(CONFIG_4xx) += arch/ppc/4xx_io/ -drivers-$(CONFIG_CPM2) += arch/ppc/8260_io/ drivers-$(CONFIG_OPROFILE) += arch/powerpc/oprofile/ diff --git a/arch/powerpc/math-emu/Makefile b/arch/powerpc/math-emu/Makefile new file mode 100644 index 00000000000..754143e8936 --- /dev/null +++ b/arch/powerpc/math-emu/Makefile @@ -0,0 +1,13 @@ + +obj-y := math.o fmr.o lfd.o stfd.o + +obj-$(CONFIG_MATH_EMULATION) += fabs.o fadd.o fadds.o fcmpo.o fcmpu.o \ + fctiw.o fctiwz.o fdiv.o fdivs.o \ + fmadd.o fmadds.o fmsub.o fmsubs.o \ + fmul.o fmuls.o fnabs.o fneg.o types.o \ + fnmadd.o fnmadds.o fnmsub.o fnmsubs.o \ + fres.o frsp.o frsqrte.o fsel.o lfs.o \ + fsqrt.o fsqrts.o fsub.o fsubs.o \ + mcrfs.o mffs.o mtfsb0.o mtfsb1.o \ + mtfsf.o mtfsfi.o stfiwx.o stfs.o \ + udivmodti4.o diff --git a/arch/powerpc/math-emu/double.h b/arch/powerpc/math-emu/double.h new file mode 100644 index 00000000000..ffba8b67f05 --- /dev/null +++ b/arch/powerpc/math-emu/double.h @@ -0,0 +1,129 @@ +/* + * Definitions for IEEE Double Precision + */ + +#if _FP_W_TYPE_SIZE < 32 +#error "Here's a nickel kid. Go buy yourself a real computer." +#endif + +#if _FP_W_TYPE_SIZE < 64 +#define _FP_FRACTBITS_D (2 * _FP_W_TYPE_SIZE) +#else +#define _FP_FRACTBITS_D _FP_W_TYPE_SIZE +#endif + +#define _FP_FRACBITS_D 53 +#define _FP_FRACXBITS_D (_FP_FRACTBITS_D - _FP_FRACBITS_D) +#define _FP_WFRACBITS_D (_FP_WORKBITS + _FP_FRACBITS_D) +#define _FP_WFRACXBITS_D (_FP_FRACTBITS_D - _FP_WFRACBITS_D) +#define _FP_EXPBITS_D 11 +#define _FP_EXPBIAS_D 1023 +#define _FP_EXPMAX_D 2047 + +#define _FP_QNANBIT_D \ + ((_FP_W_TYPE)1 << ((_FP_FRACBITS_D-2) % _FP_W_TYPE_SIZE)) +#define _FP_IMPLBIT_D \ + ((_FP_W_TYPE)1 << ((_FP_FRACBITS_D-1) % _FP_W_TYPE_SIZE)) +#define _FP_OVERFLOW_D \ + ((_FP_W_TYPE)1 << (_FP_WFRACBITS_D % _FP_W_TYPE_SIZE)) + +#if _FP_W_TYPE_SIZE < 64 + +union _FP_UNION_D +{ + double flt; + struct { +#if __BYTE_ORDER == __BIG_ENDIAN + unsigned sign : 1; + unsigned exp : _FP_EXPBITS_D; + unsigned frac1 : _FP_FRACBITS_D - (_FP_IMPLBIT_D != 0) - _FP_W_TYPE_SIZE; + unsigned frac0 : _FP_W_TYPE_SIZE; +#else + unsigned frac0 : _FP_W_TYPE_SIZE; + unsigned frac1 : _FP_FRACBITS_D - (_FP_IMPLBIT_D != 0) - _FP_W_TYPE_SIZE; + unsigned exp : _FP_EXPBITS_D; + unsigned sign : 1; +#endif + } bits __attribute__((packed)); +}; + +#define FP_DECL_D(X) _FP_DECL(2,X) +#define FP_UNPACK_RAW_D(X,val) _FP_UNPACK_RAW_2(D,X,val) +#define FP_PACK_RAW_D(val,X) _FP_PACK_RAW_2(D,val,X) + +#define FP_UNPACK_D(X,val) \ + do { \ + _FP_UNPACK_RAW_2(D,X,val); \ + _FP_UNPACK_CANONICAL(D,2,X); \ + } while (0) + +#define FP_PACK_D(val,X) \ + do { \ + _FP_PACK_CANONICAL(D,2,X); \ + _FP_PACK_RAW_2(D,val,X); \ + } while (0) + +#define FP_NEG_D(R,X) _FP_NEG(D,2,R,X) +#define FP_ADD_D(R,X,Y) _FP_ADD(D,2,R,X,Y) +#define FP_SUB_D(R,X,Y) _FP_SUB(D,2,R,X,Y) +#define FP_MUL_D(R,X,Y) _FP_MUL(D,2,R,X,Y) +#define FP_DIV_D(R,X,Y) _FP_DIV(D,2,R,X,Y) +#define FP_SQRT_D(R,X) _FP_SQRT(D,2,R,X) + +#define FP_CMP_D(r,X,Y,un) _FP_CMP(D,2,r,X,Y,un) +#define FP_CMP_EQ_D(r,X,Y) _FP_CMP_EQ(D,2,r,X,Y) + +#define FP_TO_INT_D(r,X,rsz,rsg) _FP_TO_INT(D,2,r,X,rsz,rsg) +#define FP_FROM_INT_D(X,r,rs,rt) _FP_FROM_INT(D,2,X,r,rs,rt) + +#else + +union _FP_UNION_D +{ + double flt; + struct { +#if __BYTE_ORDER == __BIG_ENDIAN + unsigned sign : 1; + unsigned exp : _FP_EXPBITS_D; + unsigned long frac : _FP_FRACBITS_D - (_FP_IMPLBIT_D != 0); +#else + unsigned long frac : _FP_FRACBITS_D - (_FP_IMPLBIT_D != 0); + unsigned exp : _FP_EXPBITS_D; + unsigned sign : 1; +#endif + } bits __attribute__((packed)); +}; + +#define FP_DECL_D(X) _FP_DECL(1,X) +#define FP_UNPACK_RAW_D(X,val) _FP_UNPACK_RAW_1(D,X,val) +#define FP_PACK_RAW_D(val,X) _FP_PACK_RAW_1(D,val,X) + +#define FP_UNPACK_D(X,val) \ + do { \ + _FP_UNPACK_RAW_1(D,X,val); \ + _FP_UNPACK_CANONICAL(D,1,X); \ + } while (0) + +#define FP_PACK_D(val,X) \ + do { \ + _FP_PACK_CANONICAL(D,1,X); \ + _FP_PACK_RAW_1(D,val,X); \ + } while (0) + +#define FP_NEG_D(R,X) _FP_NEG(D,1,R,X) +#define FP_ADD_D(R,X,Y) _FP_ADD(D,1,R,X,Y) +#define FP_SUB_D(R,X,Y) _FP_SUB(D,1,R,X,Y) +#define FP_MUL_D(R,X,Y) _FP_MUL(D,1,R,X,Y) +#define FP_DIV_D(R,X,Y) _FP_DIV(D,1,R,X,Y) +#define FP_SQRT_D(R,X) _FP_SQRT(D,1,R,X) + +/* The implementation of _FP_MUL_D and _FP_DIV_D should be chosen by + the target machine. */ + +#define FP_CMP_D(r,X,Y,un) _FP_CMP(D,1,r,X,Y,un) +#define FP_CMP_EQ_D(r,X,Y) _FP_CMP_EQ(D,1,r,X,Y) + +#define FP_TO_INT_D(r,X,rsz,rsg) _FP_TO_INT(D,1,r,X,rsz,rsg) +#define FP_FROM_INT_D(X,r,rs,rt) _FP_FROM_INT(D,1,X,r,rs,rt) + +#endif /* W_TYPE_SIZE < 64 */ diff --git a/arch/powerpc/math-emu/fabs.c b/arch/powerpc/math-emu/fabs.c new file mode 100644 index 00000000000..41f0617f3d3 --- /dev/null +++ b/arch/powerpc/math-emu/fabs.c @@ -0,0 +1,18 @@ +#include +#include +#include + +int +fabs(u32 *frD, u32 *frB) +{ + frD[0] = frB[0] & 0x7fffffff; + frD[1] = frB[1]; + +#ifdef DEBUG + printk("%s: D %p, B %p: ", __FUNCTION__, frD, frB); + dump_double(frD); + printk("\n"); +#endif + + return 0; +} diff --git a/arch/powerpc/math-emu/fadd.c b/arch/powerpc/math-emu/fadd.c new file mode 100644 index 00000000000..fc8836488b6 --- /dev/null +++ b/arch/powerpc/math-emu/fadd.c @@ -0,0 +1,38 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" + +int +fadd(void *frD, void *frA, void *frB) +{ + FP_DECL_D(A); + FP_DECL_D(B); + FP_DECL_D(R); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p\n", __FUNCTION__, frD, frA, frB); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld)\n", A_s, A_f1, A_f0, A_e, A_c); + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); +#endif + + if (A_s != B_s && A_c == FP_CLS_INF && B_c == FP_CLS_INF) + ret |= EFLAG_VXISI; + + FP_ADD_D(R, A, B); + +#ifdef DEBUG + printk("D: %ld %lu %lu %ld (%ld)\n", R_s, R_f1, R_f0, R_e, R_c); +#endif + + return (ret | __FP_PACK_D(frD, R)); +} diff --git a/arch/powerpc/math-emu/fadds.c b/arch/powerpc/math-emu/fadds.c new file mode 100644 index 00000000000..93025b6c8f3 --- /dev/null +++ b/arch/powerpc/math-emu/fadds.c @@ -0,0 +1,39 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" +#include "single.h" + +int +fadds(void *frD, void *frA, void *frB) +{ + FP_DECL_D(A); + FP_DECL_D(B); + FP_DECL_D(R); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p\n", __FUNCTION__, frD, frA, frB); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld)\n", A_s, A_f1, A_f0, A_e, A_c); + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); +#endif + + if (A_s != B_s && A_c == FP_CLS_INF && B_c == FP_CLS_INF) + ret |= EFLAG_VXISI; + + FP_ADD_D(R, A, B); + +#ifdef DEBUG + printk("D: %ld %lu %lu %ld (%ld)\n", R_s, R_f1, R_f0, R_e, R_c); +#endif + + return (ret | __FP_PACK_DS(frD, R)); +} diff --git a/arch/powerpc/math-emu/fcmpo.c b/arch/powerpc/math-emu/fcmpo.c new file mode 100644 index 00000000000..4efac394b4c --- /dev/null +++ b/arch/powerpc/math-emu/fcmpo.c @@ -0,0 +1,46 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" + +int +fcmpo(u32 *ccr, int crfD, void *frA, void *frB) +{ + FP_DECL_D(A); + FP_DECL_D(B); + int code[4] = { (1 << 3), (1 << 1), (1 << 2), (1 << 0) }; + long cmp; + int ret = 0; + +#ifdef DEBUG + printk("%s: %p (%08x) %d %p %p\n", __FUNCTION__, ccr, *ccr, crfD, frA, frB); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld)\n", A_s, A_f1, A_f0, A_e, A_c); + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); +#endif + + if (A_c == FP_CLS_NAN || B_c == FP_CLS_NAN) + ret |= EFLAG_VXVC; + + FP_CMP_D(cmp, A, B, 2); + cmp = code[(cmp + 1) & 3]; + + __FPU_FPSCR &= ~(0x1f000); + __FPU_FPSCR |= (cmp << 12); + + *ccr &= ~(15 << ((7 - crfD) << 2)); + *ccr |= (cmp << ((7 - crfD) << 2)); + +#ifdef DEBUG + printk("CR: %08x\n", *ccr); +#endif + + return ret; +} diff --git a/arch/powerpc/math-emu/fcmpu.c b/arch/powerpc/math-emu/fcmpu.c new file mode 100644 index 00000000000..b7e33176e61 --- /dev/null +++ b/arch/powerpc/math-emu/fcmpu.c @@ -0,0 +1,42 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" + +int +fcmpu(u32 *ccr, int crfD, void *frA, void *frB) +{ + FP_DECL_D(A); + FP_DECL_D(B); + int code[4] = { (1 << 3), (1 << 1), (1 << 2), (1 << 0) }; + long cmp; + +#ifdef DEBUG + printk("%s: %p (%08x) %d %p %p\n", __FUNCTION__, ccr, *ccr, crfD, frA, frB); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld)\n", A_s, A_f1, A_f0, A_e, A_c); + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); +#endif + + FP_CMP_D(cmp, A, B, 2); + cmp = code[(cmp + 1) & 3]; + + __FPU_FPSCR &= ~(0x1f000); + __FPU_FPSCR |= (cmp << 12); + + *ccr &= ~(15 << ((7 - crfD) << 2)); + *ccr |= (cmp << ((7 - crfD) << 2)); + +#ifdef DEBUG + printk("CR: %08x\n", *ccr); +#endif + + return 0; +} diff --git a/arch/powerpc/math-emu/fctiw.c b/arch/powerpc/math-emu/fctiw.c new file mode 100644 index 00000000000..3b3c98b840c --- /dev/null +++ b/arch/powerpc/math-emu/fctiw.c @@ -0,0 +1,25 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" + +int +fctiw(u32 *frD, void *frB) +{ + FP_DECL_D(B); + unsigned int r; + + __FP_UNPACK_D(B, frB); + FP_TO_INT_D(r, B, 32, 1); + frD[1] = r; + +#ifdef DEBUG + printk("%s: D %p, B %p: ", __FUNCTION__, frD, frB); + dump_double(frD); + printk("\n"); +#endif + + return 0; +} diff --git a/arch/powerpc/math-emu/fctiwz.c b/arch/powerpc/math-emu/fctiwz.c new file mode 100644 index 00000000000..7717eb6fcfb --- /dev/null +++ b/arch/powerpc/math-emu/fctiwz.c @@ -0,0 +1,32 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" + +int +fctiwz(u32 *frD, void *frB) +{ + FP_DECL_D(B); + u32 fpscr; + unsigned int r; + + fpscr = __FPU_FPSCR; + __FPU_FPSCR &= ~(3); + __FPU_FPSCR |= FP_RND_ZERO; + + __FP_UNPACK_D(B, frB); + FP_TO_INT_D(r, B, 32, 1); + frD[1] = r; + + __FPU_FPSCR = fpscr; + +#ifdef DEBUG + printk("%s: D %p, B %p: ", __FUNCTION__, frD, frB); + dump_double(frD); + printk("\n"); +#endif + + return 0; +} diff --git a/arch/powerpc/math-emu/fdiv.c b/arch/powerpc/math-emu/fdiv.c new file mode 100644 index 00000000000..f2fba825b2d --- /dev/null +++ b/arch/powerpc/math-emu/fdiv.c @@ -0,0 +1,53 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" + +int +fdiv(void *frD, void *frA, void *frB) +{ + FP_DECL_D(A); + FP_DECL_D(B); + FP_DECL_D(R); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p\n", __FUNCTION__, frD, frA, frB); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld)\n", A_s, A_f1, A_f0, A_e, A_c); + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); +#endif + + if (A_c == FP_CLS_ZERO && B_c == FP_CLS_ZERO) { + ret |= EFLAG_VXZDZ; +#ifdef DEBUG + printk("%s: FPSCR_VXZDZ raised\n", __FUNCTION__); +#endif + } + if (A_c == FP_CLS_INF && B_c == FP_CLS_INF) { + ret |= EFLAG_VXIDI; +#ifdef DEBUG + printk("%s: FPSCR_VXIDI raised\n", __FUNCTION__); +#endif + } + + if (B_c == FP_CLS_ZERO && A_c != FP_CLS_ZERO) { + ret |= EFLAG_DIVZERO; + if (__FPU_TRAP_P(EFLAG_DIVZERO)) + return ret; + } + FP_DIV_D(R, A, B); + +#ifdef DEBUG + printk("D: %ld %lu %lu %ld (%ld)\n", R_s, R_f1, R_f0, R_e, R_c); +#endif + + return (ret | __FP_PACK_D(frD, R)); +} diff --git a/arch/powerpc/math-emu/fdivs.c b/arch/powerpc/math-emu/fdivs.c new file mode 100644 index 00000000000..b971196e317 --- /dev/null +++ b/arch/powerpc/math-emu/fdivs.c @@ -0,0 +1,55 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" +#include "single.h" + +int +fdivs(void *frD, void *frA, void *frB) +{ + FP_DECL_D(A); + FP_DECL_D(B); + FP_DECL_D(R); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p\n", __FUNCTION__, frD, frA, frB); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld)\n", A_s, A_f1, A_f0, A_e, A_c); + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); +#endif + + if (A_c == FP_CLS_ZERO && B_c == FP_CLS_ZERO) { + ret |= EFLAG_VXZDZ; +#ifdef DEBUG + printk("%s: FPSCR_VXZDZ raised\n", __FUNCTION__); +#endif + } + if (A_c == FP_CLS_INF && B_c == FP_CLS_INF) { + ret |= EFLAG_VXIDI; +#ifdef DEBUG + printk("%s: FPSCR_VXIDI raised\n", __FUNCTION__); +#endif + } + + if (B_c == FP_CLS_ZERO && A_c != FP_CLS_ZERO) { + ret |= EFLAG_DIVZERO; + if (__FPU_TRAP_P(EFLAG_DIVZERO)) + return ret; + } + + FP_DIV_D(R, A, B); + +#ifdef DEBUG + printk("D: %ld %lu %lu %ld (%ld)\n", R_s, R_f1, R_f0, R_e, R_c); +#endif + + return (ret | __FP_PACK_DS(frD, R)); +} diff --git a/arch/powerpc/math-emu/fmadd.c b/arch/powerpc/math-emu/fmadd.c new file mode 100644 index 00000000000..0a1dbce793e --- /dev/null +++ b/arch/powerpc/math-emu/fmadd.c @@ -0,0 +1,48 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" + +int +fmadd(void *frD, void *frA, void *frB, void *frC) +{ + FP_DECL_D(R); + FP_DECL_D(A); + FP_DECL_D(B); + FP_DECL_D(C); + FP_DECL_D(T); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p %p\n", __FUNCTION__, frD, frA, frB, frC); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + __FP_UNPACK_D(C, frC); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld)\n", A_s, A_f1, A_f0, A_e, A_c); + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); + printk("C: %ld %lu %lu %ld (%ld)\n", C_s, C_f1, C_f0, C_e, C_c); +#endif + + if ((A_c == FP_CLS_INF && C_c == FP_CLS_ZERO) || + (A_c == FP_CLS_ZERO && C_c == FP_CLS_INF)) + ret |= EFLAG_VXIMZ; + + FP_MUL_D(T, A, C); + + if (T_s != B_s && T_c == FP_CLS_INF && B_c == FP_CLS_INF) + ret |= EFLAG_VXISI; + + FP_ADD_D(R, T, B); + +#ifdef DEBUG + printk("D: %ld %lu %lu %ld (%ld)\n", R_s, R_f1, R_f0, R_e, R_c); +#endif + + return (ret | __FP_PACK_D(frD, R)); +} diff --git a/arch/powerpc/math-emu/fmadds.c b/arch/powerpc/math-emu/fmadds.c new file mode 100644 index 00000000000..0f70bba9445 --- /dev/null +++ b/arch/powerpc/math-emu/fmadds.c @@ -0,0 +1,49 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" +#include "single.h" + +int +fmadds(void *frD, void *frA, void *frB, void *frC) +{ + FP_DECL_D(R); + FP_DECL_D(A); + FP_DECL_D(B); + FP_DECL_D(C); + FP_DECL_D(T); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p %p\n", __FUNCTION__, frD, frA, frB, frC); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + __FP_UNPACK_D(C, frC); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld)\n", A_s, A_f1, A_f0, A_e, A_c); + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); + printk("C: %ld %lu %lu %ld (%ld)\n", C_s, C_f1, C_f0, C_e, C_c); +#endif + + if ((A_c == FP_CLS_INF && C_c == FP_CLS_ZERO) || + (A_c == FP_CLS_ZERO && C_c == FP_CLS_INF)) + ret |= EFLAG_VXIMZ; + + FP_MUL_D(T, A, C); + + if (T_s != B_s && T_c == FP_CLS_INF && B_c == FP_CLS_INF) + ret |= EFLAG_VXISI; + + FP_ADD_D(R, T, B); + +#ifdef DEBUG + printk("D: %ld %lu %lu %ld (%ld)\n", R_s, R_f1, R_f0, R_e, R_c); +#endif + + return (ret | __FP_PACK_DS(frD, R)); +} diff --git a/arch/powerpc/math-emu/fmr.c b/arch/powerpc/math-emu/fmr.c new file mode 100644 index 00000000000..28df700c0c7 --- /dev/null +++ b/arch/powerpc/math-emu/fmr.c @@ -0,0 +1,18 @@ +#include +#include +#include + +int +fmr(u32 *frD, u32 *frB) +{ + frD[0] = frB[0]; + frD[1] = frB[1]; + +#ifdef DEBUG + printk("%s: D %p, B %p: ", __FUNCTION__, frD, frB); + dump_double(frD); + printk("\n"); +#endif + + return 0; +} diff --git a/arch/powerpc/math-emu/fmsub.c b/arch/powerpc/math-emu/fmsub.c new file mode 100644 index 00000000000..203fd48a6fe --- /dev/null +++ b/arch/powerpc/math-emu/fmsub.c @@ -0,0 +1,51 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" + +int +fmsub(void *frD, void *frA, void *frB, void *frC) +{ + FP_DECL_D(R); + FP_DECL_D(A); + FP_DECL_D(B); + FP_DECL_D(C); + FP_DECL_D(T); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p %p\n", __FUNCTION__, frD, frA, frB, frC); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + __FP_UNPACK_D(C, frC); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld)\n", A_s, A_f1, A_f0, A_e, A_c); + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); + printk("C: %ld %lu %lu %ld (%ld)\n", C_s, C_f1, C_f0, C_e, C_c); +#endif + + if ((A_c == FP_CLS_INF && C_c == FP_CLS_ZERO) || + (A_c == FP_CLS_ZERO && C_c == FP_CLS_INF)) + ret |= EFLAG_VXIMZ; + + FP_MUL_D(T, A, C); + + if (B_c != FP_CLS_NAN) + B_s ^= 1; + + if (T_s != B_s && T_c == FP_CLS_INF && B_c == FP_CLS_INF) + ret |= EFLAG_VXISI; + + FP_ADD_D(R, T, B); + +#ifdef DEBUG + printk("D: %ld %lu %lu %ld (%ld)\n", R_s, R_f1, R_f0, R_e, R_c); +#endif + + return (ret | __FP_PACK_D(frD, R)); +} diff --git a/arch/powerpc/math-emu/fmsubs.c b/arch/powerpc/math-emu/fmsubs.c new file mode 100644 index 00000000000..8ce68624c18 --- /dev/null +++ b/arch/powerpc/math-emu/fmsubs.c @@ -0,0 +1,52 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" +#include "single.h" + +int +fmsubs(void *frD, void *frA, void *frB, void *frC) +{ + FP_DECL_D(R); + FP_DECL_D(A); + FP_DECL_D(B); + FP_DECL_D(C); + FP_DECL_D(T); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p %p\n", __FUNCTION__, frD, frA, frB, frC); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + __FP_UNPACK_D(C, frC); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld)\n", A_s, A_f1, A_f0, A_e, A_c); + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); + printk("C: %ld %lu %lu %ld (%ld)\n", C_s, C_f1, C_f0, C_e, C_c); +#endif + + if ((A_c == FP_CLS_INF && C_c == FP_CLS_ZERO) || + (A_c == FP_CLS_ZERO && C_c == FP_CLS_INF)) + ret |= EFLAG_VXIMZ; + + FP_MUL_D(T, A, C); + + if (B_c != FP_CLS_NAN) + B_s ^= 1; + + if (T_s != B_s && T_c == FP_CLS_INF && B_c == FP_CLS_INF) + ret |= EFLAG_VXISI; + + FP_ADD_D(R, T, B); + +#ifdef DEBUG + printk("D: %ld %lu %lu %ld (%ld)\n", R_s, R_f1, R_f0, R_e, R_c); +#endif + + return (ret | __FP_PACK_DS(frD, R)); +} diff --git a/arch/powerpc/math-emu/fmul.c b/arch/powerpc/math-emu/fmul.c new file mode 100644 index 00000000000..66c7e79aae2 --- /dev/null +++ b/arch/powerpc/math-emu/fmul.c @@ -0,0 +1,42 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" + +int +fmul(void *frD, void *frA, void *frB) +{ + FP_DECL_D(A); + FP_DECL_D(B); + FP_DECL_D(R); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p\n", __FUNCTION__, frD, frA, frB); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld) [%08lx.%08lx %lx]\n", + A_s, A_f1, A_f0, A_e, A_c, A_f1, A_f0, A_e + 1023); + printk("B: %ld %lu %lu %ld (%ld) [%08lx.%08lx %lx]\n", + B_s, B_f1, B_f0, B_e, B_c, B_f1, B_f0, B_e + 1023); +#endif + + if ((A_c == FP_CLS_INF && B_c == FP_CLS_ZERO) || + (A_c == FP_CLS_ZERO && B_c == FP_CLS_INF)) + ret |= EFLAG_VXIMZ; + + FP_MUL_D(R, A, B); + +#ifdef DEBUG + printk("D: %ld %lu %lu %ld (%ld) [%08lx.%08lx %lx]\n", + R_s, R_f1, R_f0, R_e, R_c, R_f1, R_f0, R_e + 1023); +#endif + + return (ret | __FP_PACK_D(frD, R)); +} diff --git a/arch/powerpc/math-emu/fmuls.c b/arch/powerpc/math-emu/fmuls.c new file mode 100644 index 00000000000..26bc4278271 --- /dev/null +++ b/arch/powerpc/math-emu/fmuls.c @@ -0,0 +1,43 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" +#include "single.h" + +int +fmuls(void *frD, void *frA, void *frB) +{ + FP_DECL_D(A); + FP_DECL_D(B); + FP_DECL_D(R); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p\n", __FUNCTION__, frD, frA, frB); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld) [%08lx.%08lx %lx]\n", + A_s, A_f1, A_f0, A_e, A_c, A_f1, A_f0, A_e + 1023); + printk("B: %ld %lu %lu %ld (%ld) [%08lx.%08lx %lx]\n", + B_s, B_f1, B_f0, B_e, B_c, B_f1, B_f0, B_e + 1023); +#endif + + if ((A_c == FP_CLS_INF && B_c == FP_CLS_ZERO) || + (A_c == FP_CLS_ZERO && B_c == FP_CLS_INF)) + ret |= EFLAG_VXIMZ; + + FP_MUL_D(R, A, B); + +#ifdef DEBUG + printk("D: %ld %lu %lu %ld (%ld) [%08lx.%08lx %lx]\n", + R_s, R_f1, R_f0, R_e, R_c, R_f1, R_f0, R_e + 1023); +#endif + + return (ret | __FP_PACK_DS(frD, R)); +} diff --git a/arch/powerpc/math-emu/fnabs.c b/arch/powerpc/math-emu/fnabs.c new file mode 100644 index 00000000000..c6b913d179e --- /dev/null +++ b/arch/powerpc/math-emu/fnabs.c @@ -0,0 +1,18 @@ +#include +#include +#include + +int +fnabs(u32 *frD, u32 *frB) +{ + frD[0] = frB[0] | 0x80000000; + frD[1] = frB[1]; + +#ifdef DEBUG + printk("%s: D %p, B %p: ", __FUNCTION__, frD, frB); + dump_double(frD); + printk("\n"); +#endif + + return 0; +} diff --git a/arch/powerpc/math-emu/fneg.c b/arch/powerpc/math-emu/fneg.c new file mode 100644 index 00000000000..fe9a98deff6 --- /dev/null +++ b/arch/powerpc/math-emu/fneg.c @@ -0,0 +1,18 @@ +#include +#include +#include + +int +fneg(u32 *frD, u32 *frB) +{ + frD[0] = frB[0] ^ 0x80000000; + frD[1] = frB[1]; + +#ifdef DEBUG + printk("%s: D %p, B %p: ", __FUNCTION__, frD, frB); + dump_double(frD); + printk("\n"); +#endif + + return 0; +} diff --git a/arch/powerpc/math-emu/fnmadd.c b/arch/powerpc/math-emu/fnmadd.c new file mode 100644 index 00000000000..7f312276d92 --- /dev/null +++ b/arch/powerpc/math-emu/fnmadd.c @@ -0,0 +1,51 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" + +int +fnmadd(void *frD, void *frA, void *frB, void *frC) +{ + FP_DECL_D(R); + FP_DECL_D(A); + FP_DECL_D(B); + FP_DECL_D(C); + FP_DECL_D(T); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p %p\n", __FUNCTION__, frD, frA, frB, frC); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + __FP_UNPACK_D(C, frC); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld)\n", A_s, A_f1, A_f0, A_e, A_c); + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); + printk("C: %ld %lu %lu %ld (%ld)\n", C_s, C_f1, C_f0, C_e, C_c); +#endif + + if ((A_c == FP_CLS_INF && C_c == FP_CLS_ZERO) || + (A_c == FP_CLS_ZERO && C_c == FP_CLS_INF)) + ret |= EFLAG_VXIMZ; + + FP_MUL_D(T, A, C); + + if (T_s != B_s && T_c == FP_CLS_INF && B_c == FP_CLS_INF) + ret |= EFLAG_VXISI; + + FP_ADD_D(R, T, B); + + if (R_c != FP_CLS_NAN) + R_s ^= 1; + +#ifdef DEBUG + printk("D: %ld %lu %lu %ld (%ld)\n", R_s, R_f1, R_f0, R_e, R_c); +#endif + + return (ret | __FP_PACK_D(frD, R)); +} diff --git a/arch/powerpc/math-emu/fnmadds.c b/arch/powerpc/math-emu/fnmadds.c new file mode 100644 index 00000000000..65454c9c70b --- /dev/null +++ b/arch/powerpc/math-emu/fnmadds.c @@ -0,0 +1,52 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" +#include "single.h" + +int +fnmadds(void *frD, void *frA, void *frB, void *frC) +{ + FP_DECL_D(R); + FP_DECL_D(A); + FP_DECL_D(B); + FP_DECL_D(C); + FP_DECL_D(T); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p %p\n", __FUNCTION__, frD, frA, frB, frC); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + __FP_UNPACK_D(C, frC); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld)\n", A_s, A_f1, A_f0, A_e, A_c); + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); + printk("C: %ld %lu %lu %ld (%ld)\n", C_s, C_f1, C_f0, C_e, C_c); +#endif + + if ((A_c == FP_CLS_INF && C_c == FP_CLS_ZERO) || + (A_c == FP_CLS_ZERO && C_c == FP_CLS_INF)) + ret |= EFLAG_VXIMZ; + + FP_MUL_D(T, A, C); + + if (T_s != B_s && T_c == FP_CLS_INF && B_c == FP_CLS_INF) + ret |= EFLAG_VXISI; + + FP_ADD_D(R, T, B); + + if (R_c != FP_CLS_NAN) + R_s ^= 1; + +#ifdef DEBUG + printk("D: %ld %lu %lu %ld (%ld)\n", R_s, R_f1, R_f0, R_e, R_c); +#endif + + return (ret | __FP_PACK_DS(frD, R)); +} diff --git a/arch/powerpc/math-emu/fnmsub.c b/arch/powerpc/math-emu/fnmsub.c new file mode 100644 index 00000000000..f1ca7482b5f --- /dev/null +++ b/arch/powerpc/math-emu/fnmsub.c @@ -0,0 +1,54 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" + +int +fnmsub(void *frD, void *frA, void *frB, void *frC) +{ + FP_DECL_D(R); + FP_DECL_D(A); + FP_DECL_D(B); + FP_DECL_D(C); + FP_DECL_D(T); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p %p\n", __FUNCTION__, frD, frA, frB, frC); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + __FP_UNPACK_D(C, frC); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld)\n", A_s, A_f1, A_f0, A_e, A_c); + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); + printk("C: %ld %lu %lu %ld (%ld)\n", C_s, C_f1, C_f0, C_e, C_c); +#endif + + if ((A_c == FP_CLS_INF && C_c == FP_CLS_ZERO) || + (A_c == FP_CLS_ZERO && C_c == FP_CLS_INF)) + ret |= EFLAG_VXIMZ; + + FP_MUL_D(T, A, C); + + if (B_c != FP_CLS_NAN) + B_s ^= 1; + + if (T_s != B_s && T_c == FP_CLS_INF && B_c == FP_CLS_INF) + ret |= EFLAG_VXISI; + + FP_ADD_D(R, T, B); + + if (R_c != FP_CLS_NAN) + R_s ^= 1; + +#ifdef DEBUG + printk("D: %ld %lu %lu %ld (%ld)\n", R_s, R_f1, R_f0, R_e, R_c); +#endif + + return (ret | __FP_PACK_D(frD, R)); +} diff --git a/arch/powerpc/math-emu/fnmsubs.c b/arch/powerpc/math-emu/fnmsubs.c new file mode 100644 index 00000000000..5c9a09a87dc --- /dev/null +++ b/arch/powerpc/math-emu/fnmsubs.c @@ -0,0 +1,55 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" +#include "single.h" + +int +fnmsubs(void *frD, void *frA, void *frB, void *frC) +{ + FP_DECL_D(R); + FP_DECL_D(A); + FP_DECL_D(B); + FP_DECL_D(C); + FP_DECL_D(T); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p %p\n", __FUNCTION__, frD, frA, frB, frC); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + __FP_UNPACK_D(C, frC); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld)\n", A_s, A_f1, A_f0, A_e, A_c); + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); + printk("C: %ld %lu %lu %ld (%ld)\n", C_s, C_f1, C_f0, C_e, C_c); +#endif + + if ((A_c == FP_CLS_INF && C_c == FP_CLS_ZERO) || + (A_c == FP_CLS_ZERO && C_c == FP_CLS_INF)) + ret |= EFLAG_VXIMZ; + + FP_MUL_D(T, A, C); + + if (B_c != FP_CLS_NAN) + B_s ^= 1; + + if (T_s != B_s && T_c == FP_CLS_INF && B_c == FP_CLS_INF) + ret |= EFLAG_VXISI; + + FP_ADD_D(R, T, B); + + if (R_c != FP_CLS_NAN) + R_s ^= 1; + +#ifdef DEBUG + printk("D: %ld %lu %lu %ld (%ld)\n", R_s, R_f1, R_f0, R_e, R_c); +#endif + + return (ret | __FP_PACK_DS(frD, R)); +} diff --git a/arch/powerpc/math-emu/fres.c b/arch/powerpc/math-emu/fres.c new file mode 100644 index 00000000000..ec11e46d20a --- /dev/null +++ b/arch/powerpc/math-emu/fres.c @@ -0,0 +1,12 @@ +#include +#include +#include + +int +fres(void *frD, void *frB) +{ +#ifdef DEBUG + printk("%s: %p %p\n", __FUNCTION__, frD, frB); +#endif + return -ENOSYS; +} diff --git a/arch/powerpc/math-emu/frsp.c b/arch/powerpc/math-emu/frsp.c new file mode 100644 index 00000000000..d879b2a3d0c --- /dev/null +++ b/arch/powerpc/math-emu/frsp.c @@ -0,0 +1,25 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" +#include "single.h" + +int +frsp(void *frD, void *frB) +{ + FP_DECL_D(B); + +#ifdef DEBUG + printk("%s: D %p, B %p\n", __FUNCTION__, frD, frB); +#endif + + __FP_UNPACK_D(B, frB); + +#ifdef DEBUG + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); +#endif + + return __FP_PACK_DS(frD, B); +} diff --git a/arch/powerpc/math-emu/frsqrte.c b/arch/powerpc/math-emu/frsqrte.c new file mode 100644 index 00000000000..a11ae182985 --- /dev/null +++ b/arch/powerpc/math-emu/frsqrte.c @@ -0,0 +1,12 @@ +#include +#include +#include + +int +frsqrte(void *frD, void *frB) +{ +#ifdef DEBUG + printk("%s: %p %p\n", __FUNCTION__, frD, frB); +#endif + return 0; +} diff --git a/arch/powerpc/math-emu/fsel.c b/arch/powerpc/math-emu/fsel.c new file mode 100644 index 00000000000..e36e6e72819 --- /dev/null +++ b/arch/powerpc/math-emu/fsel.c @@ -0,0 +1,38 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" + +int +fsel(u32 *frD, void *frA, u32 *frB, u32 *frC) +{ + FP_DECL_D(A); + +#ifdef DEBUG + printk("%s: %p %p %p %p\n", __FUNCTION__, frD, frA, frB, frC); +#endif + + __FP_UNPACK_D(A, frA); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld)\n", A_s, A_f1, A_f0, A_e, A_c); + printk("B: %08x %08x\n", frB[0], frB[1]); + printk("C: %08x %08x\n", frC[0], frC[1]); +#endif + + if (A_c == FP_CLS_NAN || (A_c != FP_CLS_ZERO && A_s)) { + frD[0] = frB[0]; + frD[1] = frB[1]; + } else { + frD[0] = frC[0]; + frD[1] = frC[1]; + } + +#ifdef DEBUG + printk("D: %08x.%08x\n", frD[0], frD[1]); +#endif + + return 0; +} diff --git a/arch/powerpc/math-emu/fsqrt.c b/arch/powerpc/math-emu/fsqrt.c new file mode 100644 index 00000000000..6f8319f64a8 --- /dev/null +++ b/arch/powerpc/math-emu/fsqrt.c @@ -0,0 +1,37 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" + +int +fsqrt(void *frD, void *frB) +{ + FP_DECL_D(B); + FP_DECL_D(R); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p %p\n", __FUNCTION__, frD, frB); +#endif + + __FP_UNPACK_D(B, frB); + +#ifdef DEBUG + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); +#endif + + if (B_s && B_c != FP_CLS_ZERO) + ret |= EFLAG_VXSQRT; + if (B_c == FP_CLS_NAN) + ret |= EFLAG_VXSNAN; + + FP_SQRT_D(R, B); + +#ifdef DEBUG + printk("R: %ld %lu %lu %ld (%ld)\n", R_s, R_f1, R_f0, R_e, R_c); +#endif + + return (ret | __FP_PACK_D(frD, R)); +} diff --git a/arch/powerpc/math-emu/fsqrts.c b/arch/powerpc/math-emu/fsqrts.c new file mode 100644 index 00000000000..3b2b1cf55c1 --- /dev/null +++ b/arch/powerpc/math-emu/fsqrts.c @@ -0,0 +1,38 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" +#include "single.h" + +int +fsqrts(void *frD, void *frB) +{ + FP_DECL_D(B); + FP_DECL_D(R); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p %p\n", __FUNCTION__, frD, frB); +#endif + + __FP_UNPACK_D(B, frB); + +#ifdef DEBUG + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); +#endif + + if (B_s && B_c != FP_CLS_ZERO) + ret |= EFLAG_VXSQRT; + if (B_c == FP_CLS_NAN) + ret |= EFLAG_VXSNAN; + + FP_SQRT_D(R, B); + +#ifdef DEBUG + printk("R: %ld %lu %lu %ld (%ld)\n", R_s, R_f1, R_f0, R_e, R_c); +#endif + + return (ret | __FP_PACK_DS(frD, R)); +} diff --git a/arch/powerpc/math-emu/fsub.c b/arch/powerpc/math-emu/fsub.c new file mode 100644 index 00000000000..956679042bb --- /dev/null +++ b/arch/powerpc/math-emu/fsub.c @@ -0,0 +1,41 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" + +int +fsub(void *frD, void *frA, void *frB) +{ + FP_DECL_D(A); + FP_DECL_D(B); + FP_DECL_D(R); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p\n", __FUNCTION__, frD, frA, frB); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld)\n", A_s, A_f1, A_f0, A_e, A_c); + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); +#endif + + if (B_c != FP_CLS_NAN) + B_s ^= 1; + + if (A_s != B_s && A_c == FP_CLS_INF && B_c == FP_CLS_INF) + ret |= EFLAG_VXISI; + + FP_ADD_D(R, A, B); + +#ifdef DEBUG + printk("D: %ld %lu %lu %ld (%ld)\n", R_s, R_f1, R_f0, R_e, R_c); +#endif + + return (ret | __FP_PACK_D(frD, R)); +} diff --git a/arch/powerpc/math-emu/fsubs.c b/arch/powerpc/math-emu/fsubs.c new file mode 100644 index 00000000000..3428117dfe8 --- /dev/null +++ b/arch/powerpc/math-emu/fsubs.c @@ -0,0 +1,42 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" +#include "single.h" + +int +fsubs(void *frD, void *frA, void *frB) +{ + FP_DECL_D(A); + FP_DECL_D(B); + FP_DECL_D(R); + int ret = 0; + +#ifdef DEBUG + printk("%s: %p %p %p\n", __FUNCTION__, frD, frA, frB); +#endif + + __FP_UNPACK_D(A, frA); + __FP_UNPACK_D(B, frB); + +#ifdef DEBUG + printk("A: %ld %lu %lu %ld (%ld)\n", A_s, A_f1, A_f0, A_e, A_c); + printk("B: %ld %lu %lu %ld (%ld)\n", B_s, B_f1, B_f0, B_e, B_c); +#endif + + if (B_c != FP_CLS_NAN) + B_s ^= 1; + + if (A_s != B_s && A_c == FP_CLS_INF && B_c == FP_CLS_INF) + ret |= EFLAG_VXISI; + + FP_ADD_D(R, A, B); + +#ifdef DEBUG + printk("D: %ld %lu %lu %ld (%ld)\n", R_s, R_f1, R_f0, R_e, R_c); +#endif + + return (ret | __FP_PACK_DS(frD, R)); +} diff --git a/arch/powerpc/math-emu/lfd.c b/arch/powerpc/math-emu/lfd.c new file mode 100644 index 00000000000..7d38101c329 --- /dev/null +++ b/arch/powerpc/math-emu/lfd.c @@ -0,0 +1,19 @@ +#include +#include +#include + +#include "sfp-machine.h" +#include "double.h" + +int +lfd(void *frD, void *ea) +{ + if (copy_from_user(frD, ea, sizeof(double))) + return -EFAULT; +#ifdef DEBUG + printk("%s: D %p, ea %p: ", __FUNCTION__, frD, ea); + dump_double(frD); + printk("\n"); +#endif + return 0; +} diff --git a/arch/powerpc/math-emu/lfs.c b/arch/powerpc/math-emu/lfs.c new file mode 100644 index 00000000000..c86dee3d765 --- /dev/null +++ b/arch/powerpc/math-emu/lfs.c @@ -0,0 +1,37 @@ +#include +#include +#include + +#include "soft-fp.h" +#include "double.h" +#include "single.h" + +int +lfs(void *frD, void *ea) +{ + FP_DECL_D(R); + FP_DECL_S(A); + float f; + +#ifdef DEBUG + printk("%s: D %p, ea %p\n", __FUNCTION__, frD, ea); +#endif + + if (copy_from_user(&f, ea, sizeof(float))) + return -EFAULT; + + __FP_UNPACK_S(A, &f); + +#ifdef DEBUG + printk("A: %ld %lu %ld (%ld) [%08lx]\n", A_s, A_f, A_e, A_c, + *(unsigned long *)&f); +#endif + + FP_CONV(D, S, 2, 1, R, A); + +#ifdef DEBUG + printk("R: %ld %lu %lu %ld (%ld)\n", R_s, R_f1, R_f0, R_e, R_c); +#endif + + return __FP_PACK_D(frD, R); +} diff --git a/arch/powerpc/math-emu/math.c b/arch/powerpc/math-emu/math.c new file mode 100644 index 00000000000..58915347276 --- /dev/null +++ b/arch/powerpc/math-emu/math.c @@ -0,0 +1,483 @@ +/* + * Copyright (C) 1999 Eddie C. Dost (ecd@atecom.com) + */ + +#include +#include +#include + +#include +#include + +#include "sfp-machine.h" +#include "double.h" + +#define FLOATFUNC(x) extern int x(void *, void *, void *, void *) + +FLOATFUNC(fadd); +FLOATFUNC(fadds); +FLOATFUNC(fdiv); +FLOATFUNC(fdivs); +FLOATFUNC(fmul); +FLOATFUNC(fmuls); +FLOATFUNC(fsub); +FLOATFUNC(fsubs); + +FLOATFUNC(fmadd); +FLOATFUNC(fmadds); +FLOATFUNC(fmsub); +FLOATFUNC(fmsubs); +FLOATFUNC(fnmadd); +FLOATFUNC(fnmadds); +FLOATFUNC(fnmsub); +FLOATFUNC(fnmsubs); + +FLOATFUNC(fctiw); +FLOATFUNC(fctiwz); +FLOATFUNC(frsp); + +FLOATFUNC(fcmpo); +FLOATFUNC(fcmpu); + +FLOATFUNC(mcrfs); +FLOATFUNC(mffs); +FLOATFUNC(mtfsb0); +FLOATFUNC(mtfsb1); +FLOATFUNC(mtfsf); +FLOATFUNC(mtfsfi); + +FLOATFUNC(lfd); +FLOATFUNC(lfs); + +FLOATFUNC(stfd); +FLOATFUNC(stfs); +FLOATFUNC(stfiwx); + +FLOATFUNC(fabs); +FLOATFUNC(fmr); +FLOATFUNC(fnabs); +FLOATFUNC(fneg); + +/* Optional */ +FLOATFUNC(fres); +FLOATFUNC(frsqrte); +FLOATFUNC(fsel); +FLOATFUNC(fsqrt); +FLOATFUNC(fsqrts); + + +#define OP31 0x1f /* 31 */ +#define LFS 0x30 /* 48 */ +#define LFSU 0x31 /* 49 */ +#define LFD 0x32 /* 50 */ +#define LFDU 0x33 /* 51 */ +#define STFS 0x34 /* 52 */ +#define STFSU 0x35 /* 53 */ +#define STFD 0x36 /* 54 */ +#define STFDU 0x37 /* 55 */ +#define OP59 0x3b /* 59 */ +#define OP63 0x3f /* 63 */ + +/* Opcode 31: */ +/* X-Form: */ +#define LFSX 0x217 /* 535 */ +#define LFSUX 0x237 /* 567 */ +#define LFDX 0x257 /* 599 */ +#define LFDUX 0x277 /* 631 */ +#define STFSX 0x297 /* 663 */ +#define STFSUX 0x2b7 /* 695 */ +#define STFDX 0x2d7 /* 727 */ +#define STFDUX 0x2f7 /* 759 */ +#define STFIWX 0x3d7 /* 983 */ + +/* Opcode 59: */ +/* A-Form: */ +#define FDIVS 0x012 /* 18 */ +#define FSUBS 0x014 /* 20 */ +#define FADDS 0x015 /* 21 */ +#define FSQRTS 0x016 /* 22 */ +#define FRES 0x018 /* 24 */ +#define FMULS 0x019 /* 25 */ +#define FMSUBS 0x01c /* 28 */ +#define FMADDS 0x01d /* 29 */ +#define FNMSUBS 0x01e /* 30 */ +#define FNMADDS 0x01f /* 31 */ + +/* Opcode 63: */ +/* A-Form: */ +#define FDIV 0x012 /* 18 */ +#define FSUB 0x014 /* 20 */ +#define FADD 0x015 /* 21 */ +#define FSQRT 0x016 /* 22 */ +#define FSEL 0x017 /* 23 */ +#define FMUL 0x019 /* 25 */ +#define FRSQRTE 0x01a /* 26 */ +#define FMSUB 0x01c /* 28 */ +#define FMADD 0x01d /* 29 */ +#define FNMSUB 0x01e /* 30 */ +#define FNMADD 0x01f /* 31 */ + +/* X-Form: */ +#define FCMPU 0x000 /* 0 */ +#define FRSP 0x00c /* 12 */ +#define FCTIW 0x00e /* 14 */ +#define FCTIWZ 0x00f /* 15 */ +#define FCMPO 0x020 /* 32 */ +#define MTFSB1 0x026 /* 38 */ +#define FNEG 0x028 /* 40 */ +#define MCRFS 0x040 /* 64 */ +#define MTFSB0 0x046 /* 70 */ +#define FMR 0x048 /* 72 */ +#define MTFSFI 0x086 /* 134 */ +#define FNABS 0x088 /* 136 */ +#define FABS 0x108 /* 264 */ +#define MFFS 0x247 /* 583 */ +#define MTFSF 0x2c7 /* 711 */ + + +#define AB 2 +#define AC 3 +#define ABC 4 +#define D 5 +#define DU 6 +#define X 7 +#define XA 8 +#define XB 9 +#define XCR 11 +#define XCRB 12 +#define XCRI 13 +#define XCRL 16 +#define XE 14 +#define XEU 15 +#define XFLB 10 + +#ifdef CONFIG_MATH_EMULATION +static int +record_exception(struct pt_regs *regs, int eflag) +{ + u32 fpscr; + + fpscr = __FPU_FPSCR; + + if (eflag) { + fpscr |= FPSCR_FX; + if (eflag & EFLAG_OVERFLOW) + fpscr |= FPSCR_OX; + if (eflag & EFLAG_UNDERFLOW) + fpscr |= FPSCR_UX; + if (eflag & EFLAG_DIVZERO) + fpscr |= FPSCR_ZX; + if (eflag & EFLAG_INEXACT) + fpscr |= FPSCR_XX; + if (eflag & EFLAG_VXSNAN) + fpscr |= FPSCR_VXSNAN; + if (eflag & EFLAG_VXISI) + fpscr |= FPSCR_VXISI; + if (eflag & EFLAG_VXIDI) + fpscr |= FPSCR_VXIDI; + if (eflag & EFLAG_VXZDZ) + fpscr |= FPSCR_VXZDZ; + if (eflag & EFLAG_VXIMZ) + fpscr |= FPSCR_VXIMZ; + if (eflag & EFLAG_VXVC) + fpscr |= FPSCR_VXVC; + if (eflag & EFLAG_VXSOFT) + fpscr |= FPSCR_VXSOFT; + if (eflag & EFLAG_VXSQRT) + fpscr |= FPSCR_VXSQRT; + if (eflag & EFLAG_VXCVI) + fpscr |= FPSCR_VXCVI; + } + + fpscr &= ~(FPSCR_VX); + if (fpscr & (FPSCR_VXSNAN | FPSCR_VXISI | FPSCR_VXIDI | + FPSCR_VXZDZ | FPSCR_VXIMZ | FPSCR_VXVC | + FPSCR_VXSOFT | FPSCR_VXSQRT | FPSCR_VXCVI)) + fpscr |= FPSCR_VX; + + fpscr &= ~(FPSCR_FEX); + if (((fpscr & FPSCR_VX) && (fpscr & FPSCR_VE)) || + ((fpscr & FPSCR_OX) && (fpscr & FPSCR_OE)) || + ((fpscr & FPSCR_UX) && (fpscr & FPSCR_UE)) || + ((fpscr & FPSCR_ZX) && (fpscr & FPSCR_ZE)) || + ((fpscr & FPSCR_XX) && (fpscr & FPSCR_XE))) + fpscr |= FPSCR_FEX; + + __FPU_FPSCR = fpscr; + + return (fpscr & FPSCR_FEX) ? 1 : 0; +} +#endif /* CONFIG_MATH_EMULATION */ + +int +do_mathemu(struct pt_regs *regs) +{ + void *op0 = 0, *op1 = 0, *op2 = 0, *op3 = 0; + unsigned long pc = regs->nip; + signed short sdisp; + u32 insn = 0; + int idx = 0; +#ifdef CONFIG_MATH_EMULATION + int (*func)(void *, void *, void *, void *); + int type = 0; + int eflag, trap; +#endif + + if (get_user(insn, (u32 *)pc)) + return -EFAULT; + +#ifndef CONFIG_MATH_EMULATION + switch (insn >> 26) { + case LFD: + idx = (insn >> 16) & 0x1f; + sdisp = (insn & 0xffff); + op0 = (void *)¤t->thread.fpr[(insn >> 21) & 0x1f]; + op1 = (void *)((idx ? regs->gpr[idx] : 0) + sdisp); + lfd(op0, op1, op2, op3); + break; + case LFDU: + idx = (insn >> 16) & 0x1f; + sdisp = (insn & 0xffff); + op0 = (void *)¤t->thread.fpr[(insn >> 21) & 0x1f]; + op1 = (void *)((idx ? regs->gpr[idx] : 0) + sdisp); + lfd(op0, op1, op2, op3); + regs->gpr[idx] = (unsigned long)op1; + break; + case STFD: + idx = (insn >> 16) & 0x1f; + sdisp = (insn & 0xffff); + op0 = (void *)¤t->thread.fpr[(insn >> 21) & 0x1f]; + op1 = (void *)((idx ? regs->gpr[idx] : 0) + sdisp); + stfd(op0, op1, op2, op3); + break; + case STFDU: + idx = (insn >> 16) & 0x1f; + sdisp = (insn & 0xffff); + op0 = (void *)¤t->thread.fpr[(insn >> 21) & 0x1f]; + op1 = (void *)((idx ? regs->gpr[idx] : 0) + sdisp); + stfd(op0, op1, op2, op3); + regs->gpr[idx] = (unsigned long)op1; + break; + case OP63: + op0 = (void *)¤t->thread.fpr[(insn >> 21) & 0x1f]; + op1 = (void *)¤t->thread.fpr[(insn >> 11) & 0x1f]; + fmr(op0, op1, op2, op3); + break; + default: + goto illegal; + } +#else /* CONFIG_MATH_EMULATION */ + switch (insn >> 26) { + case LFS: func = lfs; type = D; break; + case LFSU: func = lfs; type = DU; break; + case LFD: func = lfd; type = D; break; + case LFDU: func = lfd; type = DU; break; + case STFS: func = stfs; type = D; break; + case STFSU: func = stfs; type = DU; break; + case STFD: func = stfd; type = D; break; + case STFDU: func = stfd; type = DU; break; + + case OP31: + switch ((insn >> 1) & 0x3ff) { + case LFSX: func = lfs; type = XE; break; + case LFSUX: func = lfs; type = XEU; break; + case LFDX: func = lfd; type = XE; break; + case LFDUX: func = lfd; type = XEU; break; + case STFSX: func = stfs; type = XE; break; + case STFSUX: func = stfs; type = XEU; break; + case STFDX: func = stfd; type = XE; break; + case STFDUX: func = stfd; type = XEU; break; + case STFIWX: func = stfiwx; type = XE; break; + default: + goto illegal; + } + break; + + case OP59: + switch ((insn >> 1) & 0x1f) { + case FDIVS: func = fdivs; type = AB; break; + case FSUBS: func = fsubs; type = AB; break; + case FADDS: func = fadds; type = AB; break; + case FSQRTS: func = fsqrts; type = AB; break; + case FRES: func = fres; type = AB; break; + case FMULS: func = fmuls; type = AC; break; + case FMSUBS: func = fmsubs; type = ABC; break; + case FMADDS: func = fmadds; type = ABC; break; + case FNMSUBS: func = fnmsubs; type = ABC; break; + case FNMADDS: func = fnmadds; type = ABC; break; + default: + goto illegal; + } + break; + + case OP63: + if (insn & 0x20) { + switch ((insn >> 1) & 0x1f) { + case FDIV: func = fdiv; type = AB; break; + case FSUB: func = fsub; type = AB; break; + case FADD: func = fadd; type = AB; break; + case FSQRT: func = fsqrt; type = AB; break; + case FSEL: func = fsel; type = ABC; break; + case FMUL: func = fmul; type = AC; break; + case FRSQRTE: func = frsqrte; type = AB; break; + case FMSUB: func = fmsub; type = ABC; break; + case FMADD: func = fmadd; type = ABC; break; + case FNMSUB: func = fnmsub; type = ABC; break; + case FNMADD: func = fnmadd; type = ABC; break; + default: + goto illegal; + } + break; + } + + switch ((insn >> 1) & 0x3ff) { + case FCMPU: func = fcmpu; type = XCR; break; + case FRSP: func = frsp; type = XB; break; + case FCTIW: func = fctiw; type = XB; break; + case FCTIWZ: func = fctiwz; type = XB; break; + case FCMPO: func = fcmpo; type = XCR; break; + case MTFSB1: func = mtfsb1; type = XCRB; break; + case FNEG: func = fneg; type = XB; break; + case MCRFS: func = mcrfs; type = XCRL; break; + case MTFSB0: func = mtfsb0; type = XCRB; break; + case FMR: func = fmr; type = XB; break; + case MTFSFI: func = mtfsfi; type = XCRI; break; + case FNABS: func = fnabs; type = XB; break; + case FABS: func = fabs; type = XB; break; + case MFFS: func = mffs; type = X; break; + case MTFSF: func = mtfsf; type = XFLB; break; + default: + goto illegal; + } + break; + + default: + goto illegal; + } + + switch (type) { + case AB: + op0 = (void *)¤t->thread.fpr[(insn >> 21) & 0x1f]; + op1 = (void *)¤t->thread.fpr[(insn >> 16) & 0x1f]; + op2 = (void *)¤t->thread.fpr[(insn >> 11) & 0x1f]; + break; + + case AC: + op0 = (void *)¤t->thread.fpr[(insn >> 21) & 0x1f]; + op1 = (void *)¤t->thread.fpr[(insn >> 16) & 0x1f]; + op2 = (void *)¤t->thread.fpr[(insn >> 6) & 0x1f]; + break; + + case ABC: + op0 = (void *)¤t->thread.fpr[(insn >> 21) & 0x1f]; + op1 = (void *)¤t->thread.fpr[(insn >> 16) & 0x1f]; + op2 = (void *)¤t->thread.fpr[(insn >> 11) & 0x1f]; + op3 = (void *)¤t->thread.fpr[(insn >> 6) & 0x1f]; + break; + + case D: + idx = (insn >> 16) & 0x1f; + sdisp = (insn & 0xffff); + op0 = (void *)¤t->thread.fpr[(insn >> 21) & 0x1f]; + op1 = (void *)((idx ? regs->gpr[idx] : 0) + sdisp); + break; + + case DU: + idx = (insn >> 16) & 0x1f; + if (!idx) + goto illegal; + + sdisp = (insn & 0xffff); + op0 = (void *)¤t->thread.fpr[(insn >> 21) & 0x1f]; + op1 = (void *)(regs->gpr[idx] + sdisp); + break; + + case X: + op0 = (void *)¤t->thread.fpr[(insn >> 21) & 0x1f]; + break; + + case XA: + op0 = (void *)¤t->thread.fpr[(insn >> 21) & 0x1f]; + op1 = (void *)¤t->thread.fpr[(insn >> 16) & 0x1f]; + break; + + case XB: + op0 = (void *)¤t->thread.fpr[(insn >> 21) & 0x1f]; + op1 = (void *)¤t->thread.fpr[(insn >> 11) & 0x1f]; + break; + + case XE: + idx = (insn >> 16) & 0x1f; + if (!idx) + goto illegal; + + op0 = (void *)¤t->thread.fpr[(insn >> 21) & 0x1f]; + op1 = (void *)(regs->gpr[idx] + regs->gpr[(insn >> 11) & 0x1f]); + break; + + case XEU: + idx = (insn >> 16) & 0x1f; + op0 = (void *)¤t->thread.fpr[(insn >> 21) & 0x1f]; + op1 = (void *)((idx ? regs->gpr[idx] : 0) + + regs->gpr[(insn >> 11) & 0x1f]); + break; + + case XCR: + op0 = (void *)®s->ccr; + op1 = (void *)((insn >> 23) & 0x7); + op2 = (void *)¤t->thread.fpr[(insn >> 16) & 0x1f]; + op3 = (void *)¤t->thread.fpr[(insn >> 11) & 0x1f]; + break; + + case XCRL: + op0 = (void *)®s->ccr; + op1 = (void *)((insn >> 23) & 0x7); + op2 = (void *)((insn >> 18) & 0x7); + break; + + case XCRB: + op0 = (void *)((insn >> 21) & 0x1f); + break; + + case XCRI: + op0 = (void *)((insn >> 23) & 0x7); + op1 = (void *)((insn >> 12) & 0xf); + break; + + case XFLB: + op0 = (void *)((insn >> 17) & 0xff); + op1 = (void *)¤t->thread.fpr[(insn >> 11) & 0x1f]; + break; + + default: + goto illegal; + } + + eflag = func(op0, op1, op2, op3); + + if (insn & 1) { + regs->ccr &= ~(0x0f000000); + regs->ccr |= (__FPU_FPSCR >> 4) & 0x0f000000; + } + + trap = record_exception(regs, eflag); + if (trap) + return 1; + + switch (type) { + case DU: + case XEU: + regs->gpr[idx] = (unsigned long)op1; + break; + + default: + break; + } +#endif /* CONFIG_MATH_EMULATION */ + + regs->nip += 4; + return 0; + +illegal: + return -ENOSYS; +} diff --git a/arch/powerpc/math-emu/mcrfs.c b/arch/powerpc/math-emu/mcrfs.c new file mode 100644 index 00000000000..106dd912914 --- /dev/null +++ b/arch/powerpc/math-emu/mcrfs.c @@ -0,0 +1,31 @@ +#include +#include +#include + +#include "soft-fp.h" + +int +mcrfs(u32 *ccr, u32 crfD, u32 crfS) +{ + u32 value, clear; + +#ifdef DEBUG + printk("%s: %p (%08x) %d %d\n", __FUNCTION__, ccr, *ccr, crfD, crfS); +#endif + + clear = 15 << ((7 - crfS) << 2); + if (!crfS) + clear = 0x90000000; + + value = (__FPU_FPSCR >> ((7 - crfS) << 2)) & 15; + __FPU_FPSCR &= ~(clear); + + *ccr &= ~(15 << ((7 - crfD) << 2)); + *ccr |= (value << ((7 - crfD) << 2)); + +#ifdef DEBUG + printk("CR: %08x\n", __FUNCTION__, *ccr); +#endif + + return 0; +} diff --git a/arch/powerpc/math-emu/mffs.c b/arch/powerpc/math-emu/mffs.c new file mode 100644 index 00000000000..f477c9170e7 --- /dev/null +++ b/arch/powerpc/math-emu/mffs.c @@ -0,0 +1,17 @@ +#include +#include +#include + +#include "soft-fp.h" + +int +mffs(u32 *frD) +{ + frD[1] = __FPU_FPSCR; + +#ifdef DEBUG + printk("%s: frD %p: %08x.%08x\n", __FUNCTION__, frD, frD[0], frD[1]); +#endif + + return 0; +} diff --git a/arch/powerpc/math-emu/mtfsb0.c b/arch/powerpc/math-emu/mtfsb0.c new file mode 100644 index 00000000000..99bfd80f4af --- /dev/null +++ b/arch/powerpc/math-emu/mtfsb0.c @@ -0,0 +1,18 @@ +#include +#include +#include + +#include "soft-fp.h" + +int +mtfsb0(int crbD) +{ + if ((crbD != 1) && (crbD != 2)) + __FPU_FPSCR &= ~(1 << (31 - crbD)); + +#ifdef DEBUG + printk("%s: %d %08lx\n", __FUNCTION__, crbD, __FPU_FPSCR); +#endif + + return 0; +} diff --git a/arch/powerpc/math-emu/mtfsb1.c b/arch/powerpc/math-emu/mtfsb1.c new file mode 100644 index 00000000000..3d9e7ed92d2 --- /dev/null +++ b/arch/powerpc/math-emu/mtfsb1.c @@ -0,0 +1,18 @@ +#include +#include +#include + +#include "soft-fp.h" + +int +mtfsb1(int crbD) +{ + if ((crbD != 1) && (crbD != 2)) + __FPU_FPSCR |= (1 << (31 - crbD)); + +#ifdef DEBUG + printk("%s: %d %08lx\n", __FUNCTION__, crbD, __FPU_FPSCR); +#endif + + return 0; +} diff --git a/arch/powerpc/math-emu/mtfsf.c b/arch/powerpc/math-emu/mtfsf.c new file mode 100644 index 00000000000..d70cf714994 --- /dev/null +++ b/arch/powerpc/math-emu/mtfsf.c @@ -0,0 +1,45 @@ +#include +#include +#include + +#include "soft-fp.h" + +int +mtfsf(unsigned int FM, u32 *frB) +{ + u32 mask; + + if (FM == 0) + return 0; + + if (FM == 0xff) + mask = 0x9fffffff; + else { + mask = 0; + if (FM & (1 << 0)) + mask |= 0x90000000; + if (FM & (1 << 1)) + mask |= 0x0f000000; + if (FM & (1 << 2)) + mask |= 0x00f00000; + if (FM & (1 << 3)) + mask |= 0x000f0000; + if (FM & (1 << 4)) + mask |= 0x0000f000; + if (FM & (1 << 5)) + mask |= 0x00000f00; + if (FM & (1 << 6)) + mask |= 0x000000f0; + if (FM & (1 << 7)) + mask |= 0x0000000f; + } + + __FPU_FPSCR &= ~(mask); + __FPU_FPSCR |= (frB[1] & mask); + +#ifdef DEBUG + printk("%s: %02x %p: %08lx\n", __FUNCTION__, FM, frB, __FPU_FPSCR); +#endif + + return 0; +} diff --git a/arch/powerpc/math-emu/mtfsfi.c b/arch/powerpc/math-emu/mtfsfi.c new file mode 100644 index 00000000000..71df854baa7 --- /dev/null +++ b/arch/powerpc/math-emu/mtfsfi.c @@ -0,0 +1,23 @@ +#include +#include +#include + +#include "soft-fp.h" + +int +mtfsfi(unsigned int crfD, unsigned int IMM) +{ + u32 mask = 0xf; + + if (!crfD) + mask = 9; + + __FPU_FPSCR &= ~(mask << ((7 - crfD) << 2)); + __FPU_FPSCR |= (IMM & 0xf) << ((7 - crfD) << 2); + +#ifdef DEBUG + printk("%s: %d %x: %08lx\n", __FUNCTION__, crfD, IMM, __FPU_FPSCR); +#endif + + return 0; +} diff --git a/arch/powerpc/math-emu/op-1.h b/arch/powerpc/math-emu/op-1.h new file mode 100644 index 00000000000..c92fa95f562 --- /dev/null +++ b/arch/powerpc/math-emu/op-1.h @@ -0,0 +1,245 @@ +/* + * Basic one-word fraction declaration and manipulation. + */ + +#define _FP_FRAC_DECL_1(X) _FP_W_TYPE X##_f +#define _FP_FRAC_COPY_1(D,S) (D##_f = S##_f) +#define _FP_FRAC_SET_1(X,I) (X##_f = I) +#define _FP_FRAC_HIGH_1(X) (X##_f) +#define _FP_FRAC_LOW_1(X) (X##_f) +#define _FP_FRAC_WORD_1(X,w) (X##_f) + +#define _FP_FRAC_ADDI_1(X,I) (X##_f += I) +#define _FP_FRAC_SLL_1(X,N) \ + do { \ + if (__builtin_constant_p(N) && (N) == 1) \ + X##_f += X##_f; \ + else \ + X##_f <<= (N); \ + } while (0) +#define _FP_FRAC_SRL_1(X,N) (X##_f >>= N) + +/* Right shift with sticky-lsb. */ +#define _FP_FRAC_SRS_1(X,N,sz) __FP_FRAC_SRS_1(X##_f, N, sz) + +#define __FP_FRAC_SRS_1(X,N,sz) \ + (X = (X >> (N) | (__builtin_constant_p(N) && (N) == 1 \ + ? X & 1 : (X << (_FP_W_TYPE_SIZE - (N))) != 0))) + +#define _FP_FRAC_ADD_1(R,X,Y) (R##_f = X##_f + Y##_f) +#define _FP_FRAC_SUB_1(R,X,Y) (R##_f = X##_f - Y##_f) +#define _FP_FRAC_CLZ_1(z, X) __FP_CLZ(z, X##_f) + +/* Predicates */ +#define _FP_FRAC_NEGP_1(X) ((_FP_WS_TYPE)X##_f < 0) +#define _FP_FRAC_ZEROP_1(X) (X##_f == 0) +#define _FP_FRAC_OVERP_1(fs,X) (X##_f & _FP_OVERFLOW_##fs) +#define _FP_FRAC_EQ_1(X, Y) (X##_f == Y##_f) +#define _FP_FRAC_GE_1(X, Y) (X##_f >= Y##_f) +#define _FP_FRAC_GT_1(X, Y) (X##_f > Y##_f) + +#define _FP_ZEROFRAC_1 0 +#define _FP_MINFRAC_1 1 + +/* + * Unpack the raw bits of a native fp value. Do not classify or + * normalize the data. + */ + +#define _FP_UNPACK_RAW_1(fs, X, val) \ + do { \ + union _FP_UNION_##fs _flo; _flo.flt = (val); \ + \ + X##_f = _flo.bits.frac; \ + X##_e = _flo.bits.exp; \ + X##_s = _flo.bits.sign; \ + } while (0) + + +/* + * Repack the raw bits of a native fp value. + */ + +#define _FP_PACK_RAW_1(fs, val, X) \ + do { \ + union _FP_UNION_##fs _flo; \ + \ + _flo.bits.frac = X##_f; \ + _flo.bits.exp = X##_e; \ + _flo.bits.sign = X##_s; \ + \ + (val) = _flo.flt; \ + } while (0) + + +/* + * Multiplication algorithms: + */ + +/* Basic. Assuming the host word size is >= 2*FRACBITS, we can do the + multiplication immediately. */ + +#define _FP_MUL_MEAT_1_imm(fs, R, X, Y) \ + do { \ + R##_f = X##_f * Y##_f; \ + /* Normalize since we know where the msb of the multiplicands \ + were (bit B), we know that the msb of the of the product is \ + at either 2B or 2B-1. */ \ + _FP_FRAC_SRS_1(R, _FP_WFRACBITS_##fs-1, 2*_FP_WFRACBITS_##fs); \ + } while (0) + +/* Given a 1W * 1W => 2W primitive, do the extended multiplication. */ + +#define _FP_MUL_MEAT_1_wide(fs, R, X, Y, doit) \ + do { \ + _FP_W_TYPE _Z_f0, _Z_f1; \ + doit(_Z_f1, _Z_f0, X##_f, Y##_f); \ + /* Normalize since we know where the msb of the multiplicands \ + were (bit B), we know that the msb of the of the product is \ + at either 2B or 2B-1. */ \ + _FP_FRAC_SRS_2(_Z, _FP_WFRACBITS_##fs-1, 2*_FP_WFRACBITS_##fs); \ + R##_f = _Z_f0; \ + } while (0) + +/* Finally, a simple widening multiply algorithm. What fun! */ + +#define _FP_MUL_MEAT_1_hard(fs, R, X, Y) \ + do { \ + _FP_W_TYPE _xh, _xl, _yh, _yl, _z_f0, _z_f1, _a_f0, _a_f1; \ + \ + /* split the words in half */ \ + _xh = X##_f >> (_FP_W_TYPE_SIZE/2); \ + _xl = X##_f & (((_FP_W_TYPE)1 << (_FP_W_TYPE_SIZE/2)) - 1); \ + _yh = Y##_f >> (_FP_W_TYPE_SIZE/2); \ + _yl = Y##_f & (((_FP_W_TYPE)1 << (_FP_W_TYPE_SIZE/2)) - 1); \ + \ + /* multiply the pieces */ \ + _z_f0 = _xl * _yl; \ + _a_f0 = _xh * _yl; \ + _a_f1 = _xl * _yh; \ + _z_f1 = _xh * _yh; \ + \ + /* reassemble into two full words */ \ + if ((_a_f0 += _a_f1) < _a_f1) \ + _z_f1 += (_FP_W_TYPE)1 << (_FP_W_TYPE_SIZE/2); \ + _a_f1 = _a_f0 >> (_FP_W_TYPE_SIZE/2); \ + _a_f0 = _a_f0 << (_FP_W_TYPE_SIZE/2); \ + _FP_FRAC_ADD_2(_z, _z, _a); \ + \ + /* normalize */ \ + _FP_FRAC_SRS_2(_z, _FP_WFRACBITS_##fs - 1, 2*_FP_WFRACBITS_##fs); \ + R##_f = _z_f0; \ + } while (0) + + +/* + * Division algorithms: + */ + +/* Basic. Assuming the host word size is >= 2*FRACBITS, we can do the + division immediately. Give this macro either _FP_DIV_HELP_imm for + C primitives or _FP_DIV_HELP_ldiv for the ISO function. Which you + choose will depend on what the compiler does with divrem4. */ + +#define _FP_DIV_MEAT_1_imm(fs, R, X, Y, doit) \ + do { \ + _FP_W_TYPE _q, _r; \ + X##_f <<= (X##_f < Y##_f \ + ? R##_e--, _FP_WFRACBITS_##fs \ + : _FP_WFRACBITS_##fs - 1); \ + doit(_q, _r, X##_f, Y##_f); \ + R##_f = _q | (_r != 0); \ + } while (0) + +/* GCC's longlong.h defines a 2W / 1W => (1W,1W) primitive udiv_qrnnd + that may be useful in this situation. This first is for a primitive + that requires normalization, the second for one that does not. Look + for UDIV_NEEDS_NORMALIZATION to tell which your machine needs. */ + +#define _FP_DIV_MEAT_1_udiv_norm(fs, R, X, Y) \ + do { \ + _FP_W_TYPE _nh, _nl, _q, _r; \ + \ + /* Normalize Y -- i.e. make the most significant bit set. */ \ + Y##_f <<= _FP_WFRACXBITS_##fs - 1; \ + \ + /* Shift X op correspondingly high, that is, up one full word. */ \ + if (X##_f <= Y##_f) \ + { \ + _nl = 0; \ + _nh = X##_f; \ + } \ + else \ + { \ + R##_e++; \ + _nl = X##_f << (_FP_W_TYPE_SIZE-1); \ + _nh = X##_f >> 1; \ + } \ + \ + udiv_qrnnd(_q, _r, _nh, _nl, Y##_f); \ + R##_f = _q | (_r != 0); \ + } while (0) + +#define _FP_DIV_MEAT_1_udiv(fs, R, X, Y) \ + do { \ + _FP_W_TYPE _nh, _nl, _q, _r; \ + if (X##_f < Y##_f) \ + { \ + R##_e--; \ + _nl = X##_f << _FP_WFRACBITS_##fs; \ + _nh = X##_f >> _FP_WFRACXBITS_##fs; \ + } \ + else \ + { \ + _nl = X##_f << (_FP_WFRACBITS_##fs - 1); \ + _nh = X##_f >> (_FP_WFRACXBITS_##fs + 1); \ + } \ + udiv_qrnnd(_q, _r, _nh, _nl, Y##_f); \ + R##_f = _q | (_r != 0); \ + } while (0) + + +/* + * Square root algorithms: + * We have just one right now, maybe Newton approximation + * should be added for those machines where division is fast. + */ + +#define _FP_SQRT_MEAT_1(R, S, T, X, q) \ + do { \ + while (q) \ + { \ + T##_f = S##_f + q; \ + if (T##_f <= X##_f) \ + { \ + S##_f = T##_f + q; \ + X##_f -= T##_f; \ + R##_f += q; \ + } \ + _FP_FRAC_SLL_1(X, 1); \ + q >>= 1; \ + } \ + } while (0) + +/* + * Assembly/disassembly for converting to/from integral types. + * No shifting or overflow handled her