| Message ID | 877gbrtq70.fsf@talisman.default |
|---|---|
| State | New |
| Headers | show |
On Fri, Nov 29, 2013 at 12:14 PM, Richard Sandiford <rdsandiford@googlemail.com> wrote: > In the fold-const.ii testcase, well over half of the mul_internal calls > were for multiplication by 0 (106038 out of 169355). This patch adds > an early-out for that. > > Tested on x86_64-linux-gnu. OK to install? Ok. Did you check how many of the remaining are multiplies by 1? Thanks, Richard. > Thanks, > Richard > > > Index: gcc/wide-int.cc > =================================================================== > --- gcc/wide-int.cc 2013-11-29 10:50:46.594705157 +0000 > +++ gcc/wide-int.cc 2013-11-29 11:06:50.052557006 +0000 > @@ -1289,6 +1289,13 @@ wi::mul_internal (HOST_WIDE_INT *val, co > if (needs_overflow) > *overflow = false; > > + /* This is a surprisingly common case, so do it first. */ > + if ((op1len == 1 && op1[0] == 0) || (op2len == 1 && op2[0] == 0)) > + { > + val[0] = 0; > + return 1; > + } > + > /* If we need to check for overflow, we can only do half wide > multiplies quickly because we need to look at the top bits to > check for the overflow. */ >
Index: gcc/wide-int.cc =================================================================== --- gcc/wide-int.cc 2013-11-29 10:50:46.594705157 +0000 +++ gcc/wide-int.cc 2013-11-29 11:06:50.052557006 +0000 @@ -1289,6 +1289,13 @@ wi::mul_internal (HOST_WIDE_INT *val, co if (needs_overflow) *overflow = false; + /* This is a surprisingly common case, so do it first. */ + if ((op1len == 1 && op1[0] == 0) || (op2len == 1 && op2[0] == 0)) + { + val[0] = 0; + return 1; + } + /* If we need to check for overflow, we can only do half wide multiplies quickly because we need to look at the top bits to check for the overflow. */