optimize SDC
diff --git a/internal/cgen/base/floatconv-submodule-code.c b/internal/cgen/base/floatconv-submodule-code.c
index 641ec8c..33fc546 100644
--- a/internal/cgen/base/floatconv-submodule-code.c
+++ b/internal/cgen/base/floatconv-submodule-code.c
@@ -1307,12 +1307,12 @@
// When Eisel-Lemire fails, fall back to Simple Decimal Conversion. See
// https://nigeltao.github.io/blog/2020/parse-number-f64-simple.html
//
- // Scale by powers of 2 until we're in the range [½ .. 1], which gives us
- // our exponent (in base-2). First we shift right, possibly a little too
- // far, ending with a value certainly below 1 and possibly below ½...
+ // Scale by powers of 2 until we're in the range [1 .. 2]. First we shift
+ // right, possibly a little too far, ending with a value certainly below
+ // 10 ...
const int32_t f64_bias = -1023;
int32_t exp2 = 0;
- while (h->decimal_point > 0) {
+ while (h->decimal_point > 1) {
uint32_t n = (uint32_t)(+h->decimal_point);
uint32_t shift =
(n < num_powers)
@@ -1326,20 +1326,13 @@
}
exp2 += (int32_t)shift;
}
- // ...then we shift left, putting us in [½ .. 1].
- while (h->decimal_point <= 0) {
+ // ... then we shift left, putting us in [0.1 .. 10].
+ while (h->decimal_point < 0) {
uint32_t shift;
- if (h->decimal_point == 0) {
- if (h->digits[0] >= 5) {
- break;
- }
- shift = (h->digits[0] < 2) ? 2 : 1;
- } else {
uint32_t n = (uint32_t)(-h->decimal_point);
shift = (n < num_powers)
- ? powers[n]
+ ? powers[n] + 1 // shift 1 extra since we're now multiplying
: WUFFS_BASE__PRIVATE_IMPLEMENTATION__HPD__SHIFT__MAX_INCL;
- }
wuffs_base__private_implementation__high_prec_dec__small_lshift(h, shift);
if (h->decimal_point >
@@ -1349,9 +1342,40 @@
exp2 -= (int32_t)shift;
}
- // We're in the range [½ .. 1] but f64 uses [1 .. 2].
- exp2--;
-
+ // To get into the range [1 .. 2] we need to look at the first 3 digits of
+ // the mantisse to determine the final left shift. If we are already between 1
+ // and 2 then just shift left by 52.
+ uint64_t man = 0;
+ uint32_t i;
+ for (i = 0; i < 3; i++) {
+ man = (10 * man) + ((i < h->num_digits) ? h->digits[i] : 0);
+ }
+ int32_t final_lshift = 52;
+ int32_t additional_shift = 0;
+ if (h->decimal_point == 0) { // value is in [0.1 .. 1]
+ if (man < 125) {
+ additional_shift = -4;
+ } else if (man < 250) {
+ additional_shift = -3;
+ } else if (man < 500) {
+ additional_shift = -2;
+ } else {
+ additional_shift = -1;
+ }
+ } else { // value is in [1 .. 10]
+ if (man < 200) {
+ additional_shift = 0;
+ } else if (man < 400) {
+ additional_shift = 1;
+ } else if (man < 800) {
+ additional_shift = 2;
+ } else {
+ additional_shift = 3;
+ }
+ }
+ exp2 += additional_shift;
+ final_lshift -= additional_shift;
+
// The minimum normal exponent is (f64_bias + 1).
while ((f64_bias + 1) > exp2) {
uint32_t n = (uint32_t)((f64_bias + 1) - exp2);
@@ -1368,7 +1392,7 @@
}
// Extract 53 bits for the mantissa (in base-2).
- wuffs_base__private_implementation__high_prec_dec__small_lshift(h, 53);
+ wuffs_base__private_implementation__high_prec_dec__small_lshift(h, final_lshift);
uint64_t man2 =
wuffs_base__private_implementation__high_prec_dec__rounded_integer(h);
diff --git a/release/c/wuffs-unsupported-snapshot.c b/release/c/wuffs-unsupported-snapshot.c
index 27c69b5..e8c967e 100644
--- a/release/c/wuffs-unsupported-snapshot.c
+++ b/release/c/wuffs-unsupported-snapshot.c
@@ -14788,12 +14788,12 @@
// When Eisel-Lemire fails, fall back to Simple Decimal Conversion. See
// https://nigeltao.github.io/blog/2020/parse-number-f64-simple.html
//
- // Scale by powers of 2 until we're in the range [½ .. 1], which gives us
- // our exponent (in base-2). First we shift right, possibly a little too
- // far, ending with a value certainly below 1 and possibly below ½...
+ // Scale by powers of 2 until we're in the range [1 .. 2]. First we shift
+ // right, possibly a little too far, ending with a value certainly below
+ // 10 ...
const int32_t f64_bias = -1023;
int32_t exp2 = 0;
- while (h->decimal_point > 0) {
+ while (h->decimal_point > 1) {
uint32_t n = (uint32_t)(+h->decimal_point);
uint32_t shift =
(n < num_powers)
@@ -14807,20 +14807,13 @@
}
exp2 += (int32_t)shift;
}
- // ...then we shift left, putting us in [½ .. 1].
- while (h->decimal_point <= 0) {
+ // ... then we shift left, putting us in [0.1 .. 10].
+ while (h->decimal_point < 0) {
uint32_t shift;
- if (h->decimal_point == 0) {
- if (h->digits[0] >= 5) {
- break;
- }
- shift = (h->digits[0] < 2) ? 2 : 1;
- } else {
uint32_t n = (uint32_t)(-h->decimal_point);
shift = (n < num_powers)
- ? powers[n]
+ ? powers[n] + 1 // shift 1 extra since we're now multiplying
: WUFFS_BASE__PRIVATE_IMPLEMENTATION__HPD__SHIFT__MAX_INCL;
- }
wuffs_base__private_implementation__high_prec_dec__small_lshift(h, shift);
if (h->decimal_point >
@@ -14830,9 +14823,40 @@
exp2 -= (int32_t)shift;
}
- // We're in the range [½ .. 1] but f64 uses [1 .. 2].
- exp2--;
-
+ // To get into the range [1 .. 2] we need to look at the first 3 digits of
+ // the mantisse to determine the final left shift. If we are already between 1
+ // and 2 then just shift left by 52.
+ uint64_t man = 0;
+ uint32_t i;
+ for (i = 0; i < 3; i++) {
+ man = (10 * man) + ((i < h->num_digits) ? h->digits[i] : 0);
+ }
+ int32_t final_lshift = 52;
+ int32_t additional_shift = 0;
+ if (h->decimal_point == 0) { // value is in [0.1 .. 1]
+ if (man < 125) {
+ additional_shift = -4;
+ } else if (man < 250) {
+ additional_shift = -3;
+ } else if (man < 500) {
+ additional_shift = -2;
+ } else {
+ additional_shift = -1;
+ }
+ } else { // value is in [1 .. 10]
+ if (man < 200) {
+ additional_shift = 0;
+ } else if (man < 400) {
+ additional_shift = 1;
+ } else if (man < 800) {
+ additional_shift = 2;
+ } else {
+ additional_shift = 3;
+ }
+ }
+ exp2 += additional_shift;
+ final_lshift -= additional_shift;
+
// The minimum normal exponent is (f64_bias + 1).
while ((f64_bias + 1) > exp2) {
uint32_t n = (uint32_t)((f64_bias + 1) - exp2);
@@ -14849,7 +14873,7 @@
}
// Extract 53 bits for the mantissa (in base-2).
- wuffs_base__private_implementation__high_prec_dec__small_lshift(h, 53);
+ wuffs_base__private_implementation__high_prec_dec__small_lshift(h, final_lshift);
uint64_t man2 =
wuffs_base__private_implementation__high_prec_dec__rounded_integer(h);
