Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Fix fp.to_fp #7034

Merged
merged 1 commit into from
Dec 4, 2023
Merged

Fix fp.to_fp #7034

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 5 additions & 11 deletions src/ast/fpa/fpa2bv_converter.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3039,7 +3039,7 @@ void fpa2bv_converter::mk_to_fp_signed(func_decl * f, unsigned num, expr * const
// Special case: x == 0 -> +zero
expr_ref c1(m), v1(m);
c1 = is_zero;
v1 = pzero; // No -zero? zeros_consistent_4.smt2 requires +zero.
v1 = pzero; // No -zero (IEEE754)

// Special case: x != 0
expr_ref sign_bit(m), exp_too_large(m), sig_4(m), exp_2(m), rest(m);
Expand All @@ -3048,18 +3048,13 @@ void fpa2bv_converter::mk_to_fp_signed(func_decl * f, unsigned num, expr * const
rest = m_bv_util.mk_extract(bv_sz - 2, 0, x);
dbg_decouple("fpa2bv_to_fp_signed_rest", rest);
is_neg = m.mk_eq(sign_bit, bv1_1);
neg_x = m_bv_util.mk_bv_neg(x);
neg_x = m_bv_util.mk_bv_neg(x); // overflow ok, x_abs is now unsigned.
x_abs = m.mk_ite(is_neg, neg_x, x);
dbg_decouple("fpa2bv_to_fp_signed_is_neg", is_neg);
dbg_decouple("fpa2bv_to_fp_signed_x_abs", x_abs);
// x_abs has an extra bit in the front.
// x_abs is [bv_sz-1, bv_sz-2] . [bv_sz-3 ... 0] * 2^(bv_sz-2)
// bv_sz-2 is the "1.0" bit for the rounder.
expr_ref is_max_neg(m);
is_max_neg = m.mk_and(is_neg, m.mk_eq(rest, m_bv_util.mk_numeral(0, bv_sz-1)));
dbg_decouple("fpa2bv_to_fp_signed_is_max_neg", is_max_neg);

x_abs = m.mk_ite(is_max_neg, m_bv_util.mk_concat(bv1_1, m_bv_util.mk_numeral(0, bv_sz-1)), x_abs);
dbg_decouple("fpa2bv_to_fp_signed_x_abs", x_abs);

expr_ref lz(m);
mk_leading_zeros(x_abs, bv_sz, lz);
Expand Down Expand Up @@ -3094,7 +3089,6 @@ void fpa2bv_converter::mk_to_fp_signed(func_decl * f, unsigned num, expr * const
expr_ref s_exp(m), exp_rest(m);
s_exp = m_bv_util.mk_bv_sub(m_bv_util.mk_numeral(bv_sz - 2, bv_sz), lz);
// s_exp = (bv_sz-2) + (-lz) signed
s_exp = m.mk_ite(is_max_neg, m_bv_util.mk_bv_sub(s_exp, m_bv_util.mk_numeral(1, bv_sz)), s_exp);
SASSERT(m_bv_util.get_bv_size(s_exp) == bv_sz);
dbg_decouple("fpa2bv_to_fp_signed_s_exp", s_exp);

Expand All @@ -3109,7 +3103,7 @@ void fpa2bv_converter::mk_to_fp_signed(func_decl * f, unsigned num, expr * const

TRACE("fpa2bv_to_fp_signed", tout << "exp worst case sz: " << exp_worst_case_sz << std::endl;);

if (exp_sz < exp_worst_case_sz) {
if (exp_sz <= exp_worst_case_sz) {
// exp_sz < exp_worst_case_sz and exp >= 0.
// Take the maximum legal exponent; this
// allows us to keep the most precision.
Expand Down Expand Up @@ -3188,7 +3182,7 @@ void fpa2bv_converter::mk_to_fp_unsigned(func_decl * f, unsigned num, expr * con
// Special case: x == 0 -> +zero
expr_ref c1(m), v1(m);
c1 = is_zero;
v1 = pzero; // No nzero?
v1 = pzero; // No -zero (IEEE754)

// Special case: x != 0
expr_ref exp_too_large(m), sig_4(m), exp_2(m);
Expand Down
Loading