fluencelabs/musl
1
0
Fork 0
mirror of https://github.com/fluencelabs/musl synced 2025-06-15 16:01:41 +00:00
Code Issues Projects Releases Wiki Activity

optimize floatscan: avoid excessive upscaling

Browse Source 
upscaling by even one step too much creates 3-29 extra iterations for
the next loop. this is still suboptimal since it always goes by 2^29
rather than using a smaller upscale factor when nearing the target,
but performance on common, small-magnitude, few-digit values has
already more than doubled with this change.

more optimizations on the way...
This commit is contained in:
Rich Felker
2012-04-11 14:11:47 -04:00
parent 7ef1a9bba5
commit 1bdd5c8b98
1 changed files with 27 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files

54
src/internal/floatscan.c
View File

@ -138,7 +138,33 @@ static long double decfloat(FILE *f, int c, int bits, int emin, int sign, int po
e2 = 0; e2 = 0;
rp = lrp; rp = lrp;
while (rp < 18+9*LD_B1B_DIG) { if (rp % 9) {
static const int p10s[] = {
100000000, 10000000, 1000000, 100000,
10000, 1000, 100, 10
};
int rpm9 = rp>=0 ? rp%9 : rp%9+9;
int p10 = p10s[rpm9-1];
uint32_t carry = 0;
for (k=a; k!=z; k=(k+1 & MASK)) {
uint32_t tmp = x[k] % p10;
x[k] = x[k]/p10 + carry;
carry = 1000000000/p10 * tmp;
if (k==a && !x[k]) {
a = (a+1 & MASK);
rp -= 9;
}
}
if (carry) {
if ((z+1 & MASK) != a) {
x[z] = carry;
z = (z+1 & MASK);
} else x[z-1 & MASK] |= 1;
}
rp += 9-rpm9;
}
while (rp < 9*LD_B1B_DIG || (rp == 9*LD_B1B_DIG && x[0]<th[0])) {
uint32_t carry = 0; uint32_t carry = 0;
e2 -= 29; e2 -= 29;
for (k=(z-1 & MASK); ; k=(k-1 & MASK)) { for (k=(z-1 & MASK); ; k=(k-1 & MASK)) {
@ -164,32 +190,6 @@ static long double decfloat(FILE *f, int c, int bits, int emin, int sign, int po
} }
} }
if (rp % 9) {
static const int p10s[] = {
100000000, 10000000, 1000000, 100000,
10000, 1000, 100, 10
};
int rpm9 = rp % 9;
int p10 = p10s[rpm9-1];
uint32_t carry = 0;
for (k=a; k!=z; k=(k+1 & MASK)) {
uint32_t tmp = x[k] % p10;
x[k] = x[k]/p10 + carry;
carry = 1000000000/p10 * tmp;
if (k==a && !x[k]) {
a = (a+1 & MASK);
rp -= 9;
}
}
if (carry) {
if ((z+1 & MASK) != a) {
x[z] = carry;
z = (z+1 & MASK);
} else x[z-1 & MASK] |= 1;
}
rp += 9-rpm9;
}
for (;;) { for (;;) {
uint32_t carry = 0; uint32_t carry = 0;
int sh = 1; int sh = 1;