Minki/linux
1
0
Fork 1
mirror of https://github.com/torvalds/linux.git synced 2024-11-11 22:51:42 +00:00
Code Issues Packages Projects Releases Wiki Activity

MIPS: math-emu: Add support for the MIPS R6 MAX{, A} FPU instruction

Browse Source 
MIPS R6 introduced the following instruction:
Scalar Floating-Point Maximum and
Scalar Floating-Point argument with Maximum Absolute Value
MAX.fmt writes the maximum value of the inputs fs and ft to the
destination fd.
MAXA.fmt takes input arguments fs and ft and writes the argument with
the maximum absolute value to the destination fd.

Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/10961/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
This commit is contained in:
Markos Chandras 2015-08-13 09:56:36 +02:00 committed by Ralf Baechle
parent 4e9561b20e
commit a79f5f9ba5
5 changed files with 480 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
arch/mips/math-emu/Makefile
View File

@ -4,9 +4,9 @@
obj-y += cp1emu.o ieee754dp.o ieee754sp.o ieee754.o \
dp_div.o dp_mul.o dp_sub.o dp_add.o dp_fsp.o dp_cmp.o dp_simple.o \
dp_tint.o dp_fint.o dp_maddf.o dp_msubf.o dp_2008class.o dp_fmin.o \
dp_tint.o dp_fint.o dp_maddf.o dp_msubf.o dp_2008class.o dp_fmin.o dp_fmax.o \
sp_div.o sp_mul.o sp_sub.o sp_add.o sp_fdp.o sp_cmp.o sp_simple.o \
sp_tint.o sp_fint.o sp_maddf.o sp_msubf.o sp_2008class.o sp_fmin.o \
sp_tint.o sp_fint.o sp_maddf.o sp_msubf.o sp_2008class.o sp_fmin.o sp_fmax.o \
dsemul.o
lib-y += ieee754d.o \

48
arch/mips/math-emu/cp1emu.c
View File

@ -1839,6 +1839,30 @@ static int fpu_emu(struct pt_regs *xcp, struct mips_fpu_struct *ctx,
break;
}
case fmax_op: {
union ieee754sp fs, ft;
if (!cpu_has_mips_r6)
return SIGILL;
SPFROMREG(ft, MIPSInst_FT(ir));
SPFROMREG(fs, MIPSInst_FS(ir));
rv.s = ieee754sp_fmax(fs, ft);
break;
}
case fmaxa_op: {
union ieee754sp fs, ft;
if (!cpu_has_mips_r6)
return SIGILL;
SPFROMREG(ft, MIPSInst_FT(ir));
SPFROMREG(fs, MIPSInst_FS(ir));
rv.s = ieee754sp_fmaxa(fs, ft);
break;
}
case fabs_op:
handler.u = ieee754sp_abs;
goto scopuop;
@ -2133,6 +2157,30 @@ copcsr:
break;
}
case fmax_op: {
union ieee754dp fs, ft;
if (!cpu_has_mips_r6)
return SIGILL;
DPFROMREG(ft, MIPSInst_FT(ir));
DPFROMREG(fs, MIPSInst_FS(ir));
rv.d = ieee754dp_fmax(fs, ft);
break;
}
case fmaxa_op: {
union ieee754dp fs, ft;
if (!cpu_has_mips_r6)
return SIGILL;
DPFROMREG(ft, MIPSInst_FT(ir));
DPFROMREG(fs, MIPSInst_FS(ir));
rv.d = ieee754dp_fmaxa(fs, ft);
break;
}
case fabs_op:
handler.u = ieee754dp_abs;
goto dcopuop;

213
arch/mips/math-emu/dp_fmax.c Normal file
View File

@ -0,0 +1,213 @@
/*
* IEEE754 floating point arithmetic
* double precision: MIN{,A}.f
* MIN : Scalar Floating-Point Minimum
* MINA: Scalar Floating-Point argument with Minimum Absolute Value
*
* MIN.D : FPR[fd] = minNum(FPR[fs],FPR[ft])
* MINA.D: FPR[fd] = maxNumMag(FPR[fs],FPR[ft])
*
* MIPS floating point support
* Copyright (C) 2015 Imagination Technologies, Ltd.
* Author: Markos Chandras <markos.chandras@imgtec.com>
*
* This program is free software; you can distribute it and/or modify it
* under the terms of the GNU General Public License as published by the
* Free Software Foundation; version 2 of the License.
*/
#include "ieee754dp.h"
union ieee754dp ieee754dp_fmax(union ieee754dp x, union ieee754dp y)
{
COMPXDP;
COMPYDP;
EXPLODEXDP;
EXPLODEYDP;
FLUSHXDP;
FLUSHYDP;
ieee754_clearcx();
switch (CLPAIR(xc, yc)) {
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_SNAN):
return ieee754dp_nanxcpt(y);
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_DNORM):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF):
return ieee754dp_nanxcpt(x);
/* numbers are preferred to NaNs */
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_QNAN):
return x;
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_DNORM):
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_INF):
return y;
/*
* Infinity and zero handling
*/
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM):
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_ZERO):
return xs ? y : x;
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_DNORM):
return ys ? x : y;
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):
if (xs == ys)
return x;
return ieee754dp_zero(1);
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM):
DPDNORMX;
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_DNORM):
DPDNORMY;
break;
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_NORM):
DPDNORMX;
}
/* Finally get to do some computation */
assert(xm & DP_HIDDEN_BIT);
assert(ym & DP_HIDDEN_BIT);
/* Compare signs */
if (xs > ys)
return y;
else if (xs < ys)
return x;
/* Compare exponent */
if (xe > ye)
return x;
else if (xe < ye)
return y;
/* Compare mantissa */
if (xm <= ym)
return y;
return x;
}
union ieee754dp ieee754dp_fmaxa(union ieee754dp x, union ieee754dp y)
{
COMPXDP;
COMPYDP;
EXPLODEXDP;
EXPLODEYDP;
FLUSHXDP;
FLUSHYDP;
ieee754_clearcx();
switch (CLPAIR(xc, yc)) {
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_SNAN):
return ieee754dp_nanxcpt(y);
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_DNORM):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF):
return ieee754dp_nanxcpt(x);
/* numbers are preferred to NaNs */
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_QNAN):
return x;
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_DNORM):
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_INF):
return y;
/*
* Infinity and zero handling
*/
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM):
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_ZERO):
return x;
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_DNORM):
return y;
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):
if (xs == ys)
return x;
return ieee754dp_zero(1);
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM):
DPDNORMX;
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_DNORM):
DPDNORMY;
break;
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_NORM):
DPDNORMX;
}
/* Finally get to do some computation */
assert(xm & DP_HIDDEN_BIT);
assert(ym & DP_HIDDEN_BIT);
/* Compare exponent */
if (xe > ye)
return x;
else if (xe < ye)
return y;
/* Compare mantissa */
if (xm <= ym)
return y;
return x;
}

4
arch/mips/math-emu/ieee754.h
View File

@ -82,6 +82,8 @@ union ieee754sp ieee754sp_msubf(union ieee754sp z, union ieee754sp x,
int ieee754sp_2008class(union ieee754sp x);
union ieee754sp ieee754sp_fmin(union ieee754sp x, union ieee754sp y);
union ieee754sp ieee754sp_fmina(union ieee754sp x, union ieee754sp y);
union ieee754sp ieee754sp_fmax(union ieee754sp x, union ieee754sp y);
union ieee754sp ieee754sp_fmaxa(union ieee754sp x, union ieee754sp y);
/*
* double precision (often aka double)
@ -114,6 +116,8 @@ union ieee754dp ieee754dp_msubf(union ieee754dp z, union ieee754dp x,
int ieee754dp_2008class(union ieee754dp x);
union ieee754dp ieee754dp_fmin(union ieee754dp x, union ieee754dp y);
union ieee754dp ieee754dp_fmina(union ieee754dp x, union ieee754dp y);
union ieee754dp ieee754dp_fmax(union ieee754dp x, union ieee754dp y);
union ieee754dp ieee754dp_fmaxa(union ieee754dp x, union ieee754dp y);
/* 5 types of floating point number

213
arch/mips/math-emu/sp_fmax.c Normal file
View File

@ -0,0 +1,213 @@
/*
* IEEE754 floating point arithmetic
* single precision: MAX{,A}.f
* MAX : Scalar Floating-Point Maximum
* MAXA: Scalar Floating-Point argument with Maximum Absolute Value
*
* MAX.S : FPR[fd] = maxNum(FPR[fs],FPR[ft])
* MAXA.S: FPR[fd] = maxNumMag(FPR[fs],FPR[ft])
*
* MIPS floating point support
* Copyright (C) 2015 Imagination Technologies, Ltd.
* Author: Markos Chandras <markos.chandras@imgtec.com>
*
* This program is free software; you can distribute it and/or modify it
* under the terms of the GNU General Public License as published by the
* Free Software Foundation; version 2 of the License.
*/
#include "ieee754sp.h"
union ieee754sp ieee754sp_fmax(union ieee754sp x, union ieee754sp y)
{
COMPXSP;
COMPYSP;
EXPLODEXSP;
EXPLODEYSP;
FLUSHXSP;
FLUSHYSP;
ieee754_clearcx();
switch (CLPAIR(xc, yc)) {
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_SNAN):
return ieee754sp_nanxcpt(y);
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_DNORM):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF):
return ieee754sp_nanxcpt(x);
/* numbers are preferred to NaNs */
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_QNAN):
return x;
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_DNORM):
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_INF):
return y;
/*
* Infinity and zero handling
*/
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM):
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_ZERO):
return xs ? y : x;
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_DNORM):
return ys ? x : y;
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):
if (xs == ys)
return x;
return ieee754sp_zero(1);
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM):
SPDNORMX;
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_DNORM):
SPDNORMY;
break;
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_NORM):
SPDNORMX;
}
/* Finally get to do some computation */
assert(xm & SP_HIDDEN_BIT);
assert(ym & SP_HIDDEN_BIT);
/* Compare signs */
if (xs > ys)
return y;
else if (xs < ys)
return x;
/* Compare exponent */
if (xe > ye)
return x;
else if (xe < ye)
return y;
/* Compare mantissa */
if (xm <= ym)
return y;
return x;
}
union ieee754sp ieee754sp_fmaxa(union ieee754sp x, union ieee754sp y)
{
COMPXSP;
COMPYSP;
EXPLODEXSP;
EXPLODEYSP;
FLUSHXSP;
FLUSHYSP;
ieee754_clearcx();
switch (CLPAIR(xc, yc)) {
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_SNAN):
return ieee754sp_nanxcpt(y);
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_SNAN):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_DNORM):
case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF):
return ieee754sp_nanxcpt(x);
/* numbers are preferred to NaNs */
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_QNAN):
return x;
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN):
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_DNORM):
case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_INF):
return y;
/*
* Infinity and zero handling
*/
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM):
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_ZERO):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_ZERO):
return x;
case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_INF):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_NORM):
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_DNORM):
return y;
case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO):
if (xs == ys)
return x;
return ieee754sp_zero(1);
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM):
SPDNORMX;
case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_DNORM):
SPDNORMY;
break;
case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_NORM):
SPDNORMX;
}
/* Finally get to do some computation */
assert(xm & SP_HIDDEN_BIT);
assert(ym & SP_HIDDEN_BIT);
/* Compare exponent */
if (xe > ye)
return x;
else if (xe < ye)
return y;
/* Compare mantissa */
if (xm <= ym)
return y;
return x;
}