Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

Commit9c79e64

Browse files
committed
Frob numeric.c loop so that clang will auto-vectorize it too.
Experimentation shows that clang will auto-vectorize the criticalmultiplication loop if the termination condition is written "i2 < limit"rather than "i2 <= limit". This seems unbelievably stupid, but I'vereproduced it on both clang 9.0.1 (RHEL8) and 11.0.3 (macOS Catalina).gcc doesn't care, so tweak the code to do it that way.Discussion:https://postgr.es/m/CAJ3gD9evtA_vBo+WMYMyT-u=keHX7-r8p2w7OSRfXf42LTwCZQ@mail.gmail.com
1 parent87e6ed7 commit9c79e64

File tree

1 file changed

+8
-6
lines changed

1 file changed

+8
-6
lines changed

‎src/backend/utils/adt/numeric.c

Lines changed: 8 additions & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -8191,7 +8191,6 @@ mul_var(const NumericVar *var1, const NumericVar *var2, NumericVar *result,
81918191
intres_weight;
81928192
intmaxdigits;
81938193
int*dig;
8194-
int*dig_i1_2;
81958194
intcarry;
81968195
intmaxdig;
81978196
intnewdig;
@@ -8327,7 +8326,7 @@ mul_var(const NumericVar *var1, const NumericVar *var2, NumericVar *result,
83278326
* Add the appropriate multiple of var2 into the accumulator.
83288327
*
83298328
* As above, digits of var2 can be ignored if they don't contribute,
8330-
* so we only include digits for which i1+i2+2 <= res_ndigits - 1.
8329+
* so we only include digits for which i1+i2+2 < res_ndigits.
83318330
*
83328331
* This inner loop is the performance bottleneck for multiplication,
83338332
* so we want to keep it simple enough so that it can be
@@ -8336,10 +8335,13 @@ mul_var(const NumericVar *var1, const NumericVar *var2, NumericVar *result,
83368335
* Since we aren't propagating carries in this loop, the order does
83378336
* not matter.
83388337
*/
8339-
i=Min(var2ndigits-1,res_ndigits-i1-3);
8340-
dig_i1_2=&dig[i1+2];
8341-
for (i2=0;i2 <=i;i2++)
8342-
dig_i1_2[i2]+=var1digit*var2digits[i2];
8338+
{
8339+
inti2limit=Min(var2ndigits,res_ndigits-i1-2);
8340+
int*dig_i1_2=&dig[i1+2];
8341+
8342+
for (i2=0;i2<i2limit;i2++)
8343+
dig_i1_2[i2]+=var1digit*var2digits[i2];
8344+
}
83438345
}
83448346

83458347
/*

0 commit comments

Comments
 (0)
[8]ページ先頭
©2009-2025 Movatter.jp