diff options
Diffstat (limited to 'arch')
-rw-r--r-- | arch/arc/kernel/entry.S | 6 | ||||
-rw-r--r-- | arch/arc/mm/tlb.c | 3 | ||||
-rw-r--r-- | arch/mips/math-emu/dp_fmax.c | 84 | ||||
-rw-r--r-- | arch/mips/math-emu/dp_fmin.c | 86 | ||||
-rw-r--r-- | arch/mips/math-emu/dp_maddf.c | 245 | ||||
-rw-r--r-- | arch/mips/math-emu/ieee754int.h | 4 | ||||
-rw-r--r-- | arch/mips/math-emu/ieee754sp.h | 4 | ||||
-rw-r--r-- | arch/mips/math-emu/sp_fmax.c | 84 | ||||
-rw-r--r-- | arch/mips/math-emu/sp_fmin.c | 86 | ||||
-rw-r--r-- | arch/mips/math-emu/sp_maddf.c | 226 | ||||
-rw-r--r-- | arch/powerpc/kernel/align.c | 119 | ||||
-rw-r--r-- | arch/s390/include/asm/mmu.h | 2 | ||||
-rw-r--r-- | arch/s390/include/asm/mmu_context.h | 8 | ||||
-rw-r--r-- | arch/s390/include/asm/tlbflush.h | 30 |
14 files changed, 618 insertions, 369 deletions
diff --git a/arch/arc/kernel/entry.S b/arch/arc/kernel/entry.S index 1eea99beecc3..85d9ea4a0acc 100644 --- a/arch/arc/kernel/entry.S +++ b/arch/arc/kernel/entry.S @@ -92,6 +92,12 @@ ENTRY(EV_MachineCheck) lr r0, [efa] mov r1, sp + ; hardware auto-disables MMU, re-enable it to allow kernel vaddr + ; access for say stack unwinding of modules for crash dumps + lr r3, [ARC_REG_PID] + or r3, r3, MMU_ENABLE + sr r3, [ARC_REG_PID] + lsr r3, r2, 8 bmsk r3, r3, 7 brne r3, ECR_C_MCHK_DUP_TLB, 1f diff --git a/arch/arc/mm/tlb.c b/arch/arc/mm/tlb.c index bdb295e09160..a4dc881da277 100644 --- a/arch/arc/mm/tlb.c +++ b/arch/arc/mm/tlb.c @@ -896,9 +896,6 @@ void do_tlb_overlap_fault(unsigned long cause, unsigned long address, local_irq_save(flags); - /* re-enable the MMU */ - write_aux_reg(ARC_REG_PID, MMU_ENABLE | read_aux_reg(ARC_REG_PID)); - /* loop thru all sets of TLB */ for (set = 0; set < mmu->sets; set++) { diff --git a/arch/mips/math-emu/dp_fmax.c b/arch/mips/math-emu/dp_fmax.c index fd71b8daaaf2..5bec64f2884e 100644 --- a/arch/mips/math-emu/dp_fmax.c +++ b/arch/mips/math-emu/dp_fmax.c @@ -47,14 +47,26 @@ union ieee754dp ieee754dp_fmax(union ieee754dp x, union ieee754dp y) case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF): return ieee754dp_nanxcpt(x); - /* numbers are preferred to NaNs */ + /* + * Quiet NaN handling + */ + + /* + * The case of both inputs quiet NaNs + */ + case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN): + return x; + + /* + * The cases of exactly one input quiet NaN (numbers + * are here preferred as returned values 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): @@ -80,9 +92,7 @@ union ieee754dp ieee754dp_fmax(union ieee754dp x, union ieee754dp y) return ys ? x : y; case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO): - if (xs == ys) - return x; - return ieee754dp_zero(1); + return ieee754dp_zero(xs & ys); case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM): DPDNORMX; @@ -106,16 +116,32 @@ union ieee754dp ieee754dp_fmax(union ieee754dp x, union ieee754dp y) else if (xs < ys) return x; - /* Compare exponent */ - if (xe > ye) - return x; - else if (xe < ye) - return y; + /* Signs of inputs are equal, let's compare exponents */ + if (xs == 0) { + /* Inputs are both positive */ + if (xe > ye) + return x; + else if (xe < ye) + return y; + } else { + /* Inputs are both negative */ + if (xe > ye) + return y; + else if (xe < ye) + return x; + } - /* Compare mantissa */ + /* Signs and exponents of inputs are equal, let's compare mantissas */ + if (xs == 0) { + /* Inputs are both positive, with equal signs and exponents */ + if (xm <= ym) + return y; + return x; + } + /* Inputs are both negative, with equal signs and exponents */ if (xm <= ym) - return y; - return x; + return x; + return y; } union ieee754dp ieee754dp_fmaxa(union ieee754dp x, union ieee754dp y) @@ -147,14 +173,26 @@ union ieee754dp ieee754dp_fmaxa(union ieee754dp x, union ieee754dp y) case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF): return ieee754dp_nanxcpt(x); - /* numbers are preferred to NaNs */ + /* + * Quiet NaN handling + */ + + /* + * The case of both inputs quiet NaNs + */ + case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN): + return x; + + /* + * The cases of exactly one input quiet NaN (numbers + * are here preferred as returned values 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): @@ -164,6 +202,9 @@ union ieee754dp ieee754dp_fmaxa(union ieee754dp x, union ieee754dp y) /* * Infinity and zero handling */ + case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF): + return ieee754dp_inf(xs & ys); + case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_ZERO): case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_NORM): case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM): @@ -171,7 +212,6 @@ union ieee754dp ieee754dp_fmaxa(union ieee754dp x, union ieee754dp y) 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): @@ -180,9 +220,7 @@ union ieee754dp ieee754dp_fmaxa(union ieee754dp x, union ieee754dp y) return y; case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO): - if (xs == ys) - return x; - return ieee754dp_zero(1); + return ieee754dp_zero(xs & ys); case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM): DPDNORMX; @@ -207,7 +245,11 @@ union ieee754dp ieee754dp_fmaxa(union ieee754dp x, union ieee754dp y) return y; /* Compare mantissa */ - if (xm <= ym) + if (xm < ym) return y; - return x; + else if (xm > ym) + return x; + else if (xs == 0) + return x; + return y; } diff --git a/arch/mips/math-emu/dp_fmin.c b/arch/mips/math-emu/dp_fmin.c index c1072b0dfb95..a287b23818d8 100644 --- a/arch/mips/math-emu/dp_fmin.c +++ b/arch/mips/math-emu/dp_fmin.c @@ -47,14 +47,26 @@ union ieee754dp ieee754dp_fmin(union ieee754dp x, union ieee754dp y) case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF): return ieee754dp_nanxcpt(x); - /* numbers are preferred to NaNs */ + /* + * Quiet NaN handling + */ + + /* + * The case of both inputs quiet NaNs + */ + case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN): + return x; + + /* + * The cases of exactly one input quiet NaN (numbers + * are here preferred as returned values 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): @@ -80,9 +92,7 @@ union ieee754dp ieee754dp_fmin(union ieee754dp x, union ieee754dp y) return ys ? y : x; case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO): - if (xs == ys) - return x; - return ieee754dp_zero(1); + return ieee754dp_zero(xs | ys); case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM): DPDNORMX; @@ -106,16 +116,32 @@ union ieee754dp ieee754dp_fmin(union ieee754dp x, union ieee754dp y) else if (xs < ys) return y; - /* Compare exponent */ - if (xe > ye) - return y; - else if (xe < ye) - return x; + /* Signs of inputs are the same, let's compare exponents */ + if (xs == 0) { + /* Inputs are both positive */ + if (xe > ye) + return y; + else if (xe < ye) + return x; + } else { + /* Inputs are both negative */ + if (xe > ye) + return x; + else if (xe < ye) + return y; + } - /* Compare mantissa */ + /* Signs and exponents of inputs are equal, let's compare mantissas */ + if (xs == 0) { + /* Inputs are both positive, with equal signs and exponents */ + if (xm <= ym) + return x; + return y; + } + /* Inputs are both negative, with equal signs and exponents */ if (xm <= ym) - return x; - return y; + return y; + return x; } union ieee754dp ieee754dp_fmina(union ieee754dp x, union ieee754dp y) @@ -147,14 +173,26 @@ union ieee754dp ieee754dp_fmina(union ieee754dp x, union ieee754dp y) case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF): return ieee754dp_nanxcpt(x); - /* numbers are preferred to NaNs */ + /* + * Quiet NaN handling + */ + + /* + * The case of both inputs quiet NaNs + */ + case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN): + return x; + + /* + * The cases of exactly one input quiet NaN (numbers + * are here preferred as returned values 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): @@ -164,25 +202,25 @@ union ieee754dp ieee754dp_fmina(union ieee754dp x, union ieee754dp y) /* * Infinity and zero handling */ + case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF): + return ieee754dp_inf(xs | ys); + 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; + return y; - 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; + return x; case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO): - if (xs == ys) - return x; - return ieee754dp_zero(1); + return ieee754dp_zero(xs | ys); case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM): DPDNORMX; @@ -207,7 +245,11 @@ union ieee754dp ieee754dp_fmina(union ieee754dp x, union ieee754dp y) return x; /* Compare mantissa */ - if (xm <= ym) + if (xm < ym) + return x; + else if (xm > ym) + return y; + else if (xs == 1) return x; return y; } diff --git a/arch/mips/math-emu/dp_maddf.c b/arch/mips/math-emu/dp_maddf.c index 4a2d03c72959..e0d9be5fbf4c 100644 --- a/arch/mips/math-emu/dp_maddf.c +++ b/arch/mips/math-emu/dp_maddf.c @@ -14,22 +14,45 @@ #include "ieee754dp.h" -enum maddf_flags { - maddf_negate_product = 1 << 0, -}; + +/* 128 bits shift right logical with rounding. */ +void srl128(u64 *hptr, u64 *lptr, int count) +{ + u64 low; + + if (count >= 128) { + *lptr = *hptr != 0 || *lptr != 0; + *hptr = 0; + } else if (count >= 64) { + if (count == 64) { + *lptr = *hptr | (*lptr != 0); + } else { + low = *lptr; + *lptr = *hptr >> (count - 64); + *lptr |= (*hptr << (128 - count)) != 0 || low != 0; + } + *hptr = 0; + } else { + low = *lptr; + *lptr = low >> count | *hptr << (64 - count); + *lptr |= (low << (64 - count)) != 0; + *hptr = *hptr >> count; + } +} static union ieee754dp _dp_maddf(union ieee754dp z, union ieee754dp x, union ieee754dp y, enum maddf_flags flags) { int re; int rs; - u64 rm; unsigned lxm; unsigned hxm; unsigned lym; unsigned hym; u64 lrm; u64 hrm; + u64 lzm; + u64 hzm; u64 t; u64 at; int s; @@ -48,52 +71,34 @@ static union ieee754dp _dp_maddf(union ieee754dp z, union ieee754dp x, ieee754_clearcx(); - switch (zc) { - case IEEE754_CLASS_SNAN: - ieee754_setcx(IEEE754_INVALID_OPERATION); + /* + * Handle the cases when at least one of x, y or z is a NaN. + * Order of precedence is sNaN, qNaN and z, x, y. + */ + if (zc == IEEE754_CLASS_SNAN) return ieee754dp_nanxcpt(z); - case IEEE754_CLASS_DNORM: - DPDNORMZ; - /* QNAN is handled separately below */ - } - - 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): + if (xc == IEEE754_CLASS_SNAN) return ieee754dp_nanxcpt(x); - - 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): + if (yc == IEEE754_CLASS_SNAN) + return ieee754dp_nanxcpt(y); + if (zc == IEEE754_CLASS_QNAN) + return z; + if (xc == IEEE754_CLASS_QNAN) + return x; + if (yc == IEEE754_CLASS_QNAN) return y; - 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 x; + if (zc == IEEE754_CLASS_DNORM) + DPDNORMZ; + /* ZERO z cases are handled separately below */ + switch (CLPAIR(xc, yc)) { /* * Infinity handling */ case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_ZERO): case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_INF): - if (zc == IEEE754_CLASS_QNAN) - return z; ieee754_setcx(IEEE754_INVALID_OPERATION); return ieee754dp_indef(); @@ -102,9 +107,27 @@ static union ieee754dp _dp_maddf(union ieee754dp z, union ieee754dp x, case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_NORM): case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM): case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF): - if (zc == IEEE754_CLASS_QNAN) - return z; - return ieee754dp_inf(xs ^ ys); + if ((zc == IEEE754_CLASS_INF) && + ((!(flags & MADDF_NEGATE_PRODUCT) && (zs != (xs ^ ys))) || + ((flags & MADDF_NEGATE_PRODUCT) && (zs == (xs ^ ys))))) { + /* + * Cases of addition of infinities with opposite signs + * or subtraction of infinities with same signs. + */ + ieee754_setcx(IEEE754_INVALID_OPERATION); + return ieee754dp_indef(); + } + /* + * z is here either not an infinity, or an infinity having the + * same sign as product (x*y) (in case of MADDF.D instruction) + * or product -(x*y) (in MSUBF.D case). The result must be an + * infinity, and its sign is determined only by the value of + * (flags & MADDF_NEGATE_PRODUCT) and the signs of x and y. + */ + if (flags & MADDF_NEGATE_PRODUCT) + return ieee754dp_inf(1 ^ (xs ^ ys)); + else + return ieee754dp_inf(xs ^ ys); case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO): case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_NORM): @@ -113,32 +136,42 @@ static union ieee754dp _dp_maddf(union ieee754dp z, union ieee754dp x, case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_ZERO): if (zc == IEEE754_CLASS_INF) return ieee754dp_inf(zs); - /* Multiplication is 0 so just return z */ + if (zc == IEEE754_CLASS_ZERO) { + /* Handle cases +0 + (-0) and similar ones. */ + if ((!(flags & MADDF_NEGATE_PRODUCT) + && (zs == (xs ^ ys))) || + ((flags & MADDF_NEGATE_PRODUCT) + && (zs != (xs ^ ys)))) + /* + * Cases of addition of zeros of equal signs + * or subtraction of zeroes of opposite signs. + * The sign of the resulting zero is in any + * such case determined only by the sign of z. + */ + return z; + + return ieee754dp_zero(ieee754_csr.rm == FPU_CSR_RD); + } + /* x*y is here 0, and z is not 0, so just return z */ return z; case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM): DPDNORMX; case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_DNORM): - if (zc == IEEE754_CLASS_QNAN) - return z; - else if (zc == IEEE754_CLASS_INF) + if (zc == IEEE754_CLASS_INF) return ieee754dp_inf(zs); DPDNORMY; break; case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_NORM): - if (zc == IEEE754_CLASS_QNAN) - return z; - else if (zc == IEEE754_CLASS_INF) + if (zc == IEEE754_CLASS_INF) return ieee754dp_inf(zs); DPDNORMX; break; case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_NORM): - if (zc == IEEE754_CLASS_QNAN) - return z; - else if (zc == IEEE754_CLASS_INF) + if (zc == IEEE754_CLASS_INF) return ieee754dp_inf(zs); /* fall through to real computations */ } @@ -157,7 +190,7 @@ static union ieee754dp _dp_maddf(union ieee754dp z, union ieee754dp x, re = xe + ye; rs = xs ^ ys; - if (flags & maddf_negate_product) + if (flags & MADDF_NEGATE_PRODUCT) rs ^= 1; /* shunt to top of word */ @@ -165,7 +198,7 @@ static union ieee754dp _dp_maddf(union ieee754dp z, union ieee754dp x, ym <<= 64 - (DP_FBITS + 1); /* - * Multiply 64 bits xm, ym to give high 64 bits rm with stickness. + * Multiply 64 bits xm and ym to give 128 bits result in hrm:lrm. */ /* 32 * 32 => 64 */ @@ -195,78 +228,110 @@ static union ieee754dp _dp_maddf(union ieee754dp z, union ieee754dp x, hrm = hrm + (t >> 32); - rm = hrm | (lrm != 0); - - /* - * Sticky shift down to normal rounding precision. - */ - if ((s64) rm < 0) { - rm = (rm >> (64 - (DP_FBITS + 1 + 3))) | - ((rm << (DP_FBITS + 1 + 3)) != 0); + /* Put explicit bit at bit 126 if necessary */ + if ((int64_t)hrm < 0) { + lrm = (hrm << 63) | (lrm >> 1); + hrm = hrm >> 1; re++; - } else { - rm = (rm >> (64 - (DP_FBITS + 1 + 3 + 1))) | - ((rm << (DP_FBITS + 1 + 3 + 1)) != 0); } - assert(rm & (DP_HIDDEN_BIT << 3)); - /* And now the addition */ - assert(zm & DP_HIDDEN_BIT); + assert(hrm & (1 << 62)); - /* - * Provide guard,round and stick bit space. - */ - zm <<= 3; + if (zc == IEEE754_CLASS_ZERO) { + /* + * Move explicit bit from bit 126 to bit 55 since the + * ieee754dp_format code expects the mantissa to be + * 56 bits wide (53 + 3 rounding bits). + */ + srl128(&hrm, &lrm, (126 - 55)); + return ieee754dp_format(rs, re, lrm); + } + + /* Move explicit bit from bit 52 to bit 126 */ + lzm = 0; + hzm = zm << 10; + assert(hzm & (1 << 62)); + /* Make the exponents the same */ if (ze > re) { /* * Have to shift y fraction right to align. */ s = ze - re; - rm = XDPSRS(rm, s); + srl128(&hrm, &lrm, s); re += s; } else if (re > ze) { /* * Have to shift x fraction right to align. */ s = re - ze; - zm = XDPSRS(zm, s); + srl128(&hzm, &lzm, s); ze += s; } assert(ze == re); assert(ze <= DP_EMAX); + /* Do the addition */ if (zs == rs) { /* - * Generate 28 bit result of adding two 27 bit numbers - * leaving result in xm, xs and xe. + * Generate 128 bit result by adding two 127 bit numbers + * leaving result in hzm:lzm, zs and ze. */ - zm = zm + rm; - - if (zm >> (DP_FBITS + 1 + 3)) { /* carry out */ - zm = XDPSRS1(zm); + hzm = hzm + hrm + (lzm > (lzm + lrm)); + lzm = lzm + lrm; + if ((int64_t)hzm < 0) { /* carry out */ + srl128(&hzm, &lzm, 1); ze++; } } else { - if (zm >= rm) { - zm = zm - rm; + if (hzm > hrm || (hzm == hrm && lzm >= lrm)) { + hzm = hzm - hrm - (lzm < lrm); + lzm = lzm - lrm; } else { - zm = rm - zm; + hzm = hrm - hzm - (lrm < lzm); + lzm = lrm - lzm; zs = rs; } - if (zm == 0) + if (lzm == 0 && hzm == 0) return ieee754dp_zero(ieee754_csr.rm == FPU_CSR_RD); /* - * Normalize to rounding precision. + * Put explicit bit at bit 126 if necessary. */ - while ((zm >> (DP_FBITS + 3)) == 0) { - zm <<= 1; - ze--; + if (hzm == 0) { + /* left shift by 63 or 64 bits */ + if ((int64_t)lzm < 0) { + /* MSB of lzm is the explicit bit */ + hzm = lzm >> 1; + lzm = lzm << 63; + ze -= 63; + } else { + hzm = lzm; + lzm = 0; + ze -= 64; + } + } + + t = 0; + while ((hzm >> (62 - t)) == 0) + t++; + + assert(t <= 62); + if (t) { + hzm = hzm << t | lzm >> (64 - t); + lzm = lzm << t; + ze -= t; } } - return ieee754dp_format(zs, ze, zm); + /* + * Move explicit bit from bit 126 to bit 55 since the + * ieee754dp_format code expects the mantissa to be + * 56 bits wide (53 + 3 rounding bits). + */ + srl128(&hzm, &lzm, (126 - 55)); + + return ieee754dp_format(zs, ze, lzm); } union ieee754dp ieee754dp_maddf(union ieee754dp z, union ieee754dp x, @@ -278,5 +343,5 @@ union ieee754dp ieee754dp_maddf(union ieee754dp z, union ieee754dp x, union ieee754dp ieee754dp_msubf(union ieee754dp z, union ieee754dp x, union ieee754dp y) { - return _dp_maddf(z, x, y, maddf_negate_product); + return _dp_maddf(z, x, y, MADDF_NEGATE_PRODUCT); } diff --git a/arch/mips/math-emu/ieee754int.h b/arch/mips/math-emu/ieee754int.h index 8bc2f6963324..dd2071f430e0 100644 --- a/arch/mips/math-emu/ieee754int.h +++ b/arch/mips/math-emu/ieee754int.h @@ -26,6 +26,10 @@ #define CLPAIR(x, y) ((x)*6+(y)) +enum maddf_flags { + MADDF_NEGATE_PRODUCT = 1 << 0, +}; + static inline void ieee754_clearcx(void) { ieee754_csr.cx = 0; diff --git a/arch/mips/math-emu/ieee754sp.h b/arch/mips/math-emu/ieee754sp.h index 8476067075fe..0f63e4202cff 100644 --- a/arch/mips/math-emu/ieee754sp.h +++ b/arch/mips/math-emu/ieee754sp.h @@ -45,6 +45,10 @@ static inline int ieee754sp_finite(union ieee754sp x) return SPBEXP(x) != SP_EMAX + 1 + SP_EBIAS; } +/* 64 bit right shift with rounding */ +#define XSPSRS64(v, rs) \ + (((rs) >= 64) ? ((v) != 0) : ((v) >> (rs)) | ((v) << (64-(rs)) != 0)) + /* 3bit extended single precision sticky right shift */ #define XSPSRS(v, rs) \ ((rs > (SP_FBITS+3))?1:((v) >> (rs)) | ((v) << (32-(rs)) != 0)) diff --git a/arch/mips/math-emu/sp_fmax.c b/arch/mips/math-emu/sp_fmax.c index 4d000844e48e..74a5a00d2f22 100644 --- a/arch/mips/math-emu/sp_fmax.c +++ b/arch/mips/math-emu/sp_fmax.c @@ -47,14 +47,26 @@ union ieee754sp ieee754sp_fmax(union ieee754sp x, union ieee754sp y) case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF): return ieee754sp_nanxcpt(x); - /* numbers are preferred to NaNs */ + /* + * Quiet NaN handling + */ + + /* + * The case of both inputs quiet NaNs + */ + case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN): + return x; + + /* + * The cases of exactly one input quiet NaN (numbers + * are here preferred as returned values 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): @@ -80,9 +92,7 @@ union ieee754sp ieee754sp_fmax(union ieee754sp x, union ieee754sp y) return ys ? x : y; case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO): - if (xs == ys) - return x; - return ieee754sp_zero(1); + return ieee754sp_zero(xs & ys); case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM): SPDNORMX; @@ -106,16 +116,32 @@ union ieee754sp ieee754sp_fmax(union ieee754sp x, union ieee754sp y) else if (xs < ys) return x; - /* Compare exponent */ - if (xe > ye) - return x; - else if (xe < ye) - return y; + /* Signs of inputs are equal, let's compare exponents */ + if (xs == 0) { + /* Inputs are both positive */ + if (xe > ye) + return x; + else if (xe < ye) + return y; + } else { + /* Inputs are both negative */ + if (xe > ye) + return y; + else if (xe < ye) + return x; + } - /* Compare mantissa */ + /* Signs and exponents of inputs are equal, let's compare mantissas */ + if (xs == 0) { + /* Inputs are both positive, with equal signs and exponents */ + if (xm <= ym) + return y; + return x; + } + /* Inputs are both negative, with equal signs and exponents */ if (xm <= ym) - return y; - return x; + return x; + return y; } union ieee754sp ieee754sp_fmaxa(union ieee754sp x, union ieee754sp y) @@ -147,14 +173,26 @@ union ieee754sp ieee754sp_fmaxa(union ieee754sp x, union ieee754sp y) case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF): return ieee754sp_nanxcpt(x); - /* numbers are preferred to NaNs */ + /* + * Quiet NaN handling + */ + + /* + * The case of both inputs quiet NaNs + */ + case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN): + return x; + + /* + * The cases of exactly one input quiet NaN (numbers + * are here preferred as returned values 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): @@ -164,6 +202,9 @@ union ieee754sp ieee754sp_fmaxa(union ieee754sp x, union ieee754sp y) /* * Infinity and zero handling */ + case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF): + return ieee754sp_inf(xs & ys); + case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_ZERO): case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_NORM): case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM): @@ -171,7 +212,6 @@ union ieee754sp ieee754sp_fmaxa(union ieee754sp x, union ieee754sp y) 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): @@ -180,9 +220,7 @@ union ieee754sp ieee754sp_fmaxa(union ieee754sp x, union ieee754sp y) return y; case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO): - if (xs == ys) - return x; - return ieee754sp_zero(1); + return ieee754sp_zero(xs & ys); case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM): SPDNORMX; @@ -207,7 +245,11 @@ union ieee754sp ieee754sp_fmaxa(union ieee754sp x, union ieee754sp y) return y; /* Compare mantissa */ - if (xm <= ym) + if (xm < ym) return y; - return x; + else if (xm > ym) + return x; + else if (xs == 0) + return x; + return y; } diff --git a/arch/mips/math-emu/sp_fmin.c b/arch/mips/math-emu/sp_fmin.c index 4eb1bb9e9dec..c51385f46b09 100644 --- a/arch/mips/math-emu/sp_fmin.c +++ b/arch/mips/math-emu/sp_fmin.c @@ -47,14 +47,26 @@ union ieee754sp ieee754sp_fmin(union ieee754sp x, union ieee754sp y) case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF): return ieee754sp_nanxcpt(x); - /* numbers are preferred to NaNs */ + /* + * Quiet NaN handling + */ + + /* + * The case of both inputs quiet NaNs + */ + case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN): + return x; + + /* + * The cases of exactly one input quiet NaN (numbers + * are here preferred as returned values 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): @@ -80,9 +92,7 @@ union ieee754sp ieee754sp_fmin(union ieee754sp x, union ieee754sp y) return ys ? y : x; case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO): - if (xs == ys) - return x; - return ieee754sp_zero(1); + return ieee754sp_zero(xs | ys); case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM): SPDNORMX; @@ -106,16 +116,32 @@ union ieee754sp ieee754sp_fmin(union ieee754sp x, union ieee754sp y) else if (xs < ys) return y; - /* Compare exponent */ - if (xe > ye) - return y; - else if (xe < ye) - return x; + /* Signs of inputs are the same, let's compare exponents */ + if (xs == 0) { + /* Inputs are both positive */ + if (xe > ye) + return y; + else if (xe < ye) + return x; + } else { + /* Inputs are both negative */ + if (xe > ye) + return x; + else if (xe < ye) + return y; + } - /* Compare mantissa */ + /* Signs and exponents of inputs are equal, let's compare mantissas */ + if (xs == 0) { + /* Inputs are both positive, with equal signs and exponents */ + if (xm <= ym) + return x; + return y; + } + /* Inputs are both negative, with equal signs and exponents */ if (xm <= ym) - return x; - return y; + return y; + return x; } union ieee754sp ieee754sp_fmina(union ieee754sp x, union ieee754sp y) @@ -147,14 +173,26 @@ union ieee754sp ieee754sp_fmina(union ieee754sp x, union ieee754sp y) case CLPAIR(IEEE754_CLASS_SNAN, IEEE754_CLASS_INF): return ieee754sp_nanxcpt(x); - /* numbers are preferred to NaNs */ + /* + * Quiet NaN handling + */ + + /* + * The case of both inputs quiet NaNs + */ + case CLPAIR(IEEE754_CLASS_QNAN, IEEE754_CLASS_QNAN): + return x; + + /* + * The cases of exactly one input quiet NaN (numbers + * are here preferred as returned values 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): @@ -164,25 +202,25 @@ union ieee754sp ieee754sp_fmina(union ieee754sp x, union ieee754sp y) /* * Infinity and zero handling */ + case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF): + return ieee754sp_inf(xs | ys); + 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; + return y; - 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; + return x; case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO): - if (xs == ys) - return x; - return ieee754sp_zero(1); + return ieee754sp_zero(xs | ys); case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM): SPDNORMX; @@ -207,7 +245,11 @@ union ieee754sp ieee754sp_fmina(union ieee754sp x, union ieee754sp y) return x; /* Compare mantissa */ - if (xm <= ym) + if (xm < ym) + return x; + else if (xm > ym) + return y; + else if (xs == 1) return x; return y; } diff --git a/arch/mips/math-emu/sp_maddf.c b/arch/mips/math-emu/sp_maddf.c index a8cd8b4f235e..7195fe785d81 100644 --- a/arch/mips/math-emu/sp_maddf.c +++ b/arch/mips/math-emu/sp_maddf.c @@ -14,9 +14,6 @@ #include "ieee754sp.h" -enum maddf_flags { - maddf_negate_product = 1 << 0, -}; static union ieee754sp _sp_maddf(union ieee754sp z, union ieee754sp x, union ieee754sp y, enum maddf_flags flags) @@ -24,14 +21,8 @@ static union ieee754sp _sp_maddf(union ieee754sp z, union ieee754sp x, int re; int rs; unsigned rm; - unsigned short lxm; - unsigned short hxm; - unsigned short lym; - unsigned short hym; - unsigned lrm; - unsigned hrm; - unsigned t; - unsigned at; + uint64_t rm64; + uint64_t zm64; int s; COMPXSP; @@ -48,51 +39,35 @@ static union ieee754sp _sp_maddf(union ieee754sp z, union ieee754sp x, ieee754_clearcx(); - switch (zc) { - case IEEE754_CLASS_SNAN: - ieee754_setcx(IEEE754_INVALID_OPERATION); + /* + * Handle the cases when at least one of x, y or z is a NaN. + * Order of precedence is sNaN, qNaN and z, x, y. + */ + if (zc == IEEE754_CLASS_SNAN) return ieee754sp_nanxcpt(z); - case IEEE754_CLASS_DNORM: - SPDNORMZ; - /* QNAN is handled separately below */ - } - - 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): + if (xc == IEEE754_CLASS_SNAN) + return ieee754sp_nanxcpt(x); + if (yc == IEEE754_CLASS_SNAN) return ieee754sp_nanxcpt(y); + if (zc == IEEE754_CLASS_QNAN) + return z; + if (xc == IEEE754_CLASS_QNAN) + return x; + if (yc == IEEE754_CLASS_QNAN) + return 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); + if (zc == IEEE754_CLASS_DNORM) + SPDNORMZ; + /* ZERO z cases are handled separately below */ - 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 y; + switch (CLPAIR(xc, yc)) { - 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 x; /* * Infinity handling */ case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_ZERO): case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_INF): - if (zc == IEEE754_CLASS_QNAN) - return z; ieee754_setcx(IEEE754_INVALID_OPERATION); return ieee754sp_indef(); @@ -101,9 +76,27 @@ static union ieee754sp _sp_maddf(union ieee754sp z, union ieee754sp x, case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_NORM): case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_DNORM): case CLPAIR(IEEE754_CLASS_INF, IEEE754_CLASS_INF): - if (zc == IEEE754_CLASS_QNAN) - return z; - return ieee754sp_inf(xs ^ ys); + if ((zc == IEEE754_CLASS_INF) && + ((!(flags & MADDF_NEGATE_PRODUCT) && (zs != (xs ^ ys))) || + ((flags & MADDF_NEGATE_PRODUCT) && (zs == (xs ^ ys))))) { + /* + * Cases of addition of infinities with opposite signs + * or subtraction of infinities with same signs. + */ + ieee754_setcx(IEEE754_INVALID_OPERATION); + return ieee754sp_indef(); + } + /* + * z is here either not an infinity, or an infinity having the + * same sign as product (x*y) (in case of MADDF.D instruction) + * or product -(x*y) (in MSUBF.D case). The result must be an + * infinity, and its sign is determined only by the value of + * (flags & MADDF_NEGATE_PRODUCT) and the signs of x and y. + */ + if (flags & MADDF_NEGATE_PRODUCT) + return ieee754sp_inf(1 ^ (xs ^ ys)); + else + return ieee754sp_inf(xs ^ ys); case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_ZERO): case CLPAIR(IEEE754_CLASS_ZERO, IEEE754_CLASS_NORM): @@ -112,32 +105,42 @@ static union ieee754sp _sp_maddf(union ieee754sp z, union ieee754sp x, case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_ZERO): if (zc == IEEE754_CLASS_INF) return ieee754sp_inf(zs); - /* Multiplication is 0 so just return z */ + if (zc == IEEE754_CLASS_ZERO) { + /* Handle cases +0 + (-0) and similar ones. */ + if ((!(flags & MADDF_NEGATE_PRODUCT) + && (zs == (xs ^ ys))) || + ((flags & MADDF_NEGATE_PRODUCT) + && (zs != (xs ^ ys)))) + /* + * Cases of addition of zeros of equal signs + * or subtraction of zeroes of opposite signs. + * The sign of the resulting zero is in any + * such case determined only by the sign of z. + */ + return z; + + return ieee754sp_zero(ieee754_csr.rm == FPU_CSR_RD); + } + /* x*y is here 0, and z is not 0, so just return z */ return z; case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_DNORM): SPDNORMX; case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_DNORM): - if (zc == IEEE754_CLASS_QNAN) - return z; - else if (zc == IEEE754_CLASS_INF) + if (zc == IEEE754_CLASS_INF) return ieee754sp_inf(zs); SPDNORMY; break; case CLPAIR(IEEE754_CLASS_DNORM, IEEE754_CLASS_NORM): - if (zc == IEEE754_CLASS_QNAN) - return z; - else if (zc == IEEE754_CLASS_INF) + if (zc == IEEE754_CLASS_INF) return ieee754sp_inf(zs); SPDNORMX; break; case CLPAIR(IEEE754_CLASS_NORM, IEEE754_CLASS_NORM): - if (zc == IEEE754_CLASS_QNAN) - return z; - else if (zc == IEEE754_CLASS_INF) + if (zc == IEEE754_CLASS_INF) return ieee754sp_inf(zs); /* fall through to real computations */ } @@ -158,108 +161,93 @@ static union ieee754sp _sp_maddf(union ieee754sp z, union ieee754sp x, re = xe + ye; rs = xs ^ ys; - if (flags & maddf_negate_product) + if (flags & MADDF_NEGATE_PRODUCT) rs ^= 1; - /* shunt to top of word */ - xm <<= 32 - (SP_FBITS + 1); - ym <<= 32 - (SP_FBITS + 1); - - /* - * Multiply 32 bits xm, ym to give high 32 bits rm with stickness. - */ - lxm = xm & 0xffff; - hxm = xm >> 16; - lym = ym & 0xffff; - hym = ym >> 16; - - lrm = lxm * lym; /* 16 * 16 => 32 */ - hrm = hxm * hym; /* 16 * 16 => 32 */ + /* Multiple 24 bit xm and ym to give 48 bit results */ + rm64 = (uint64_t)xm * ym; - t = lxm * hym; /* 16 * 16 => 32 */ - at = lrm + (t << 16); - hrm += at < lrm; - lrm = at; - hrm = hrm + (t >> 16); + /* Shunt to top of word */ + rm64 = rm64 << 16; - t = hxm * lym; /* 16 * 16 => 32 */ - at = lrm + (t << 16); - hrm += at < lrm; - lrm = at; - hrm = hrm + (t >> 16); - - rm = hrm | (lrm != 0); - - /* - * Sticky shift down to normal rounding precision. - */ - if ((int) rm < 0) { - rm = (rm >> (32 - (SP_FBITS + 1 + 3))) | - ((rm << (SP_FBITS + 1 + 3)) != 0); + /* Put explicit bit at bit 62 if necessary */ + if ((int64_t) rm64 < 0) { + rm64 = rm64 >> 1; re++; - } else { - rm = (rm >> (32 - (SP_FBITS + 1 + 3 + 1))) | - ((rm << (SP_FBITS + 1 + 3 + 1)) != 0); } - assert(rm & (SP_HIDDEN_BIT << 3)); - /* And now the addition */ + assert(rm64 & (1 << 62)); - assert(zm & SP_HIDDEN_BIT); + if (zc == IEEE754_CLASS_ZERO) { + /* + * Move explicit bit from bit 62 to bit 26 since the + * ieee754sp_format code expects the mantissa to be + * 27 bits wide (24 + 3 rounding bits). + */ + rm = XSPSRS64(rm64, (62 - 26)); + return ieee754sp_format(rs, re, rm); + } - /* - * Provide guard,round and stick bit space. - */ - zm <<= 3; + /* Move explicit bit from bit 23 to bit 62 */ + zm64 = (uint64_t)zm << (62 - 23); + assert(zm64 & (1 << 62)); + /* Make the exponents the same */ if (ze > re) { /* * Have to shift r fraction right to align. */ s = ze - re; - rm = XSPSRS(rm, s); + rm64 = XSPSRS64(rm64, s); re += s; } else if (re > ze) { /* * Have to shift z fraction right to align. */ s = re - ze; - zm = XSPSRS(zm, s); + zm64 = XSPSRS64(zm64, s); ze += s; } assert(ze == re); assert(ze <= SP_EMAX); + /* Do the addition */ if (zs == rs) { /* - * Generate 28 bit result of adding two 27 bit numbers - * leaving result in zm, zs and ze. + * Generate 64 bit result by adding two 63 bit numbers + * leaving result in zm64, zs and ze. */ - zm = zm + rm; - - if (zm >> (SP_FBITS + 1 + 3)) { /* carry out */ - zm = XSPSRS1(zm); + zm64 = zm64 + rm64; + if ((int64_t)zm64 < 0) { /* carry out */ + zm64 = XSPSRS1(zm64); ze++; } } else { - if (zm >= rm) { - zm = zm - rm; + if (zm64 >= rm64) { + zm64 = zm64 - rm64; } else { - zm = rm - zm; + zm64 = rm64 - zm64; zs = rs; } - if (zm == 0) + if (zm64 == 0) return ieee754sp_zero(ieee754_csr.rm == FPU_CSR_RD); /* - * Normalize in extended single precision + * Put explicit bit at bit 62 if necessary. */ - while ((zm >> (SP_MBITS + 3)) == 0) { - zm <<= 1; + while ((zm64 >> 62) == 0) { + zm64 <<= 1; ze--; } - } + + /* + * Move explicit bit from bit 62 to bit 26 since the + * ieee754sp_format code expects the mantissa to be + * 27 bits wide (24 + 3 rounding bits). + */ + zm = XSPSRS64(zm64, (62 - 26)); + return ieee754sp_format(zs, ze, zm); } @@ -272,5 +260,5 @@ union ieee754sp ieee754sp_maddf(union ieee754sp z, union ieee754sp x, union ieee754sp ieee754sp_msubf(union ieee754sp z, union ieee754sp x, union ieee754sp y) { - return _sp_maddf(z, x, y, maddf_negate_product); + return _sp_maddf(z, x, y, MADDF_NEGATE_PRODUCT); } diff --git a/arch/powerpc/kernel/align.c b/arch/powerpc/kernel/align.c index b2da7c8baed7..292458b694fb 100644 --- a/arch/powerpc/kernel/align.c +++ b/arch/powerpc/kernel/align.c @@ -235,6 +235,28 @@ static int emulate_dcbz(struct pt_regs *regs, unsigned char __user *addr) #define SWIZ_PTR(p) ((unsigned char __user *)((p) ^ swiz)) +#define __get_user_or_set_dar(_regs, _dest, _addr) \ + ({ \ + int rc = 0; \ + typeof(_addr) __addr = (_addr); \ + if (__get_user_inatomic(_dest, __addr)) { \ + _regs->dar = (unsigned long)__addr; \ + rc = -EFAULT; \ + } \ + rc; \ + }) + +#define __put_user_or_set_dar(_regs, _src, _addr) \ + ({ \ + int rc = 0; \ + typeof(_addr) __addr = (_addr); \ + if (__put_user_inatomic(_src, __addr)) { \ + _regs->dar = (unsigned long)__addr; \ + rc = -EFAULT; \ + } \ + rc; \ + }) + static int emulate_multiple(struct pt_regs *regs, unsigned char __user *addr, unsigned int reg, unsigned int nb, unsigned int flags, unsigned int instr, @@ -263,9 +285,10 @@ static int emulate_multiple(struct pt_regs *regs, unsigned char __user *addr, } else { unsigned long pc = regs->nip ^ (swiz & 4); - if (__get_user_inatomic(instr, - (unsigned int __user *)pc)) + if (__get_user_or_set_dar(regs, instr, + (unsigned int __user *)pc)) return -EFAULT; + if (swiz == 0 && (flags & SW)) instr = cpu_to_le32(instr); nb = (instr >> 11) & 0x1f; @@ -309,31 +332,31 @@ static int emulate_multiple(struct pt_regs *regs, unsigned char __user *addr, ((nb0 + 3) / 4) * sizeof(unsigned long)); for (i = 0; i < nb; ++i, ++p) - if (__get_user_inatomic(REG_BYTE(rptr, i ^ bswiz), - SWIZ_PTR(p))) + if (__get_user_or_set_dar(regs, REG_BYTE(rptr, i ^ bswiz), + SWIZ_PTR(p))) return -EFAULT; if (nb0 > 0) { rptr = ®s->gpr[0]; addr += nb; for (i = 0; i < nb0; ++i, ++p) - if (__get_user_inatomic(REG_BYTE(rptr, - i ^ bswiz), - SWIZ_PTR(p))) + if (__get_user_or_set_dar(regs, + REG_BYTE(rptr, i ^ bswiz), + SWIZ_PTR(p))) return -EFAULT; } } else { for (i = 0; i < nb; ++i, ++p) - if (__put_user_inatomic(REG_BYTE(rptr, i ^ bswiz), - SWIZ_PTR(p))) + if (__put_user_or_set_dar(regs, REG_BYTE(rptr, i ^ bswiz), + SWIZ_PTR(p))) return -EFAULT; if (nb0 > 0) { rptr = ®s->gpr[0]; addr += nb; for (i = 0; i < nb0; ++i, ++p) - if (__put_user_inatomic(REG_BYTE(rptr, - i ^ bswiz), - SWIZ_PTR(p))) + if (__put_user_or_set_dar(regs, + REG_BYTE(rptr, i ^ bswiz), + SWIZ_PTR(p))) return -EFAULT; } } @@ -345,29 +368,32 @@ static int emulate_multiple(struct pt_regs *regs, unsigned char __user *addr, * Only POWER6 has these instructions, and it does true little-endian, * so we don't need the address swizzling. */ -static int emulate_fp_pair(unsigned char __user *addr, unsigned int reg, - unsigned int flags) +static int emulate_fp_pair(struct pt_regs *regs, unsigned char __user *addr, + unsigned int reg, unsigned int flags) { char *ptr0 = (char *) ¤t->thread.TS_FPR(reg); char *ptr1 = (char *) ¤t->thread.TS_FPR(reg+1); - int i, ret, sw = 0; + int i, sw = 0; if (reg & 1) return 0; /* invalid form: FRS/FRT must be even */ if (flags & SW) sw = 7; - ret = 0; + for (i = 0; i < 8; ++i) { if (!(flags & ST)) { - ret |= __get_user(ptr0[i^sw], addr + i); - ret |= __get_user(ptr1[i^sw], addr + i + 8); + if (__get_user_or_set_dar(regs, ptr0[i^sw], addr + i)) + return -EFAULT; + if (__get_user_or_set_dar(regs, ptr1[i^sw], addr + i + 8)) + return -EFAULT; } else { - ret |= __put_user(ptr0[i^sw], addr + i); - ret |= __put_user(ptr1[i^sw], addr + i + 8); + if (__put_user_or_set_dar(regs, ptr0[i^sw], addr + i)) + return -EFAULT; + if (__put_user_or_set_dar(regs, ptr1[i^sw], addr + i + 8)) + return -EFAULT; } } - if (ret) - return -EFAULT; + return 1; /* exception handled and fixed up */ } @@ -377,24 +403,27 @@ static int emulate_lq_stq(struct pt_regs *regs, unsigned char __user *addr, { char *ptr0 = (char *)®s->gpr[reg]; char *ptr1 = (char *)®s->gpr[reg+1]; - int i, ret, sw = 0; + int i, sw = 0; if (reg & 1) return 0; /* invalid form: GPR must be even */ if (flags & SW) sw = 7; - ret = 0; + for (i = 0; i < 8; ++i) { if (!(flags & ST)) { - ret |= __get_user(ptr0[i^sw], addr + i); - ret |= __get_user(ptr1[i^sw], addr + i + 8); + if (__get_user_or_set_dar(regs, ptr0[i^sw], addr + i)) + return -EFAULT; + if (__get_user_or_set_dar(regs, ptr1[i^sw], addr + i + 8)) + return -EFAULT; } else { - ret |= __put_user(ptr0[i^sw], addr + i); - ret |= __put_user(ptr1[i^sw], addr + i + 8); + if (__put_user_or_set_dar(regs, ptr0[i^sw], addr + i)) + return -EFAULT; + if (__put_user_or_set_dar(regs, ptr1[i^sw], addr + i + 8)) + return -EFAULT; } } - if (ret) - return -EFAULT; + return 1; /* exception handled and fixed up */ } #endif /* CONFIG_PPC64 */ @@ -687,9 +716,14 @@ static int emulate_vsx(unsigned char __user *addr, unsigned int reg, for (j = 0; j < length; j += elsize) { for (i = 0; i < elsize; ++i) { if (flags & ST) - ret |= __put_user(ptr[i^sw], addr + i); + ret = __put_user_or_set_dar(regs, ptr[i^sw], + addr + i); else - ret |= __get_user(ptr[i^sw], addr + i); + ret = __get_user_or_set_dar(regs, ptr[i^sw], + addr + i); + + if (ret) + return ret; } ptr += elsize; #ifdef __LITTLE_ENDIAN__ @@ -739,7 +773,7 @@ int fix_alignment(struct pt_regs *regs) unsigned int dsisr; unsigned char __user *addr; unsigned long p, swiz; - int ret, i; + int i; union data { u64 ll; double dd; @@ -936,7 +970,7 @@ int fix_alignment(struct pt_regs *regs) if (flags & F) { /* Special case for 16-byte FP loads and stores */ PPC_WARN_ALIGNMENT(fp_pair, regs); - return emulate_fp_pair(addr, reg, flags); + return emulate_fp_pair(regs, addr, reg, flags); } else { #ifdef CONFIG_PPC64 /* Special case for 16-byte loads and stores */ @@ -966,15 +1000,12 @@ int fix_alignment(struct pt_regs *regs) } data.ll = 0; - ret = 0; p = (unsigned long)addr; for (i = 0; i < nb; i++) - ret |= __get_user_inatomic(data.v[start + i], - SWIZ_PTR(p++)); - - if (unlikely(ret)) - return -EFAULT; + if (__get_user_or_set_dar(regs, data.v[start + i], + SWIZ_PTR(p++))) + return -EFAULT; } else if (flags & F) { data.ll = current->thread.TS_FPR(reg); @@ -1046,15 +1077,13 @@ int fix_alignment(struct pt_regs *regs) break; } - ret = 0; p = (unsigned long)addr; for (i = 0; i < nb; i++) - ret |= __put_user_inatomic(data.v[start + i], - SWIZ_PTR(p++)); + if (__put_user_or_set_dar(regs, data.v[start + i], + SWIZ_PTR(p++))) + return -EFAULT; - if (unlikely(ret)) - return -EFAULT; } else if (flags & F) current->thread.TS_FPR(reg) = data.ll; else diff --git a/arch/s390/include/asm/mmu.h b/arch/s390/include/asm/mmu.h index bea785d7f853..af85d6b12028 100644 --- a/arch/s390/include/asm/mmu.h +++ b/arch/s390/include/asm/mmu.h @@ -5,6 +5,7 @@ #include <linux/errno.h> typedef struct { + spinlock_t lock; cpumask_t cpu_attach_mask; atomic_t flush_count; unsigned int flush_mm; @@ -25,6 +26,7 @@ typedef struct { } mm_context_t; #define INIT_MM_CONTEXT(name) \ + .context.lock = __SPIN_LOCK_UNLOCKED(name.context.lock), \ .context.pgtable_lock = \ __SPIN_LOCK_UNLOCKED(name.context.pgtable_lock), \ .context.pgtable_list = LIST_HEAD_INIT(name.context.pgtable_list), \ diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h index 515fea5a3fc4..f65a708ac395 100644 --- a/arch/s390/include/asm/mmu_context.h +++ b/arch/s390/include/asm/mmu_context.h @@ -15,6 +15,7 @@ static inline int init_new_context(struct task_struct *tsk, struct mm_struct *mm) { + spin_lock_init(&mm->context.lock); spin_lock_init(&mm->context.pgtable_lock); INIT_LIST_HEAD(&mm->context.pgtable_list); spin_lock_init(&mm->context.gmap_lock); @@ -93,7 +94,6 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next, if (prev == next) return; cpumask_set_cpu(cpu, &next->context.cpu_attach_mask); - cpumask_set_cpu(cpu, mm_cpumask(next)); /* Clear old ASCE by loading the kernel ASCE. */ __ctl_load(S390_lowcore.kernel_asce, 1, 1); __ctl_load(S390_lowcore.kernel_asce, 7, 7); @@ -111,9 +111,8 @@ static inline void finish_arch_post_lock_switch(void) preempt_disable(); while (atomic_read(&mm->context.flush_count)) cpu_relax(); - - if (mm->context.flush_mm) - __tlb_flush_mm(mm); + cpumask_set_cpu(smp_processor_id(), mm_cpumask(mm)); + __tlb_flush_mm_lazy(mm); preempt_enable(); } set_fs(current->thread.mm_segment); @@ -126,6 +125,7 @@ static inline void activate_mm(struct mm_struct *prev, struct mm_struct *next) { switch_mm(prev, next, current); + cpumask_set_cpu(smp_processor_id(), mm_cpumask(next)); set_user_asce(next); } diff --git a/arch/s390/include/asm/tlbflush.h b/arch/s390/include/asm/tlbflush.h index 39846100682a..eed927aeb08f 100644 --- a/arch/s390/include/asm/tlbflush.h +++ b/arch/s390/include/asm/tlbflush.h @@ -43,23 +43,6 @@ static inline void __tlb_flush_global(void) * Flush TLB entries for a specific mm on all CPUs (in case gmap is used * this implicates multiple ASCEs!). */ -static inline void __tlb_flush_full(struct mm_struct *mm) -{ - preempt_disable(); - atomic_inc(&mm->context.flush_count); - if (cpumask_equal(mm_cpumask(mm), cpumask_of(smp_processor_id()))) { - /* Local TLB flush */ - __tlb_flush_local(); - } else { - /* Global TLB flush */ - __tlb_flush_global(); - /* Reset TLB flush mask */ - cpumask_copy(mm_cpumask(mm), &mm->context.cpu_attach_mask); - } - atomic_dec(&mm->context.flush_count); - preempt_enable(); -} - static inline void __tlb_flush_mm(struct mm_struct *mm) { unsigned long gmap_asce; @@ -71,16 +54,18 @@ static inline void __tlb_flush_mm(struct mm_struct *mm) */ preempt_disable(); atomic_inc(&mm->context.flush_count); + /* Reset TLB flush mask */ + cpumask_copy(mm_cpumask(mm), &mm->context.cpu_attach_mask); + barrier(); gmap_asce = READ_ONCE(mm->context.gmap_asce); if (MACHINE_HAS_IDTE && gmap_asce != -1UL) { if (gmap_asce) __tlb_flush_idte(gmap_asce); __tlb_flush_idte(mm->context.asce); } else { - __tlb_flush_full(mm); + /* Global TLB flush */ + __tlb_flush_global(); } - /* Reset TLB flush mask */ - cpumask_copy(mm_cpumask(mm), &mm->context.cpu_attach_mask); atomic_dec(&mm->context.flush_count); preempt_enable(); } @@ -94,7 +79,6 @@ static inline void __tlb_flush_kernel(void) } #else #define __tlb_flush_global() __tlb_flush_local() -#define __tlb_flush_full(mm) __tlb_flush_local() /* * Flush TLB entries for a specific ASCE on all CPUs. @@ -112,10 +96,12 @@ static inline void __tlb_flush_kernel(void) static inline void __tlb_flush_mm_lazy(struct mm_struct * mm) { + spin_lock(&mm->context.lock); if (mm->context.flush_mm) { - __tlb_flush_mm(mm); mm->context.flush_mm = 0; + __tlb_flush_mm(mm); } + spin_unlock(&mm->context.lock); } /* |