Hello.
Following patch fixes Fortran FAIL_ALLOC predictor in a way that it introduces
a new one (PRED_FORTRAN_REALLOC)
and it properly marks returned values as described in the following 2 examples:
A) allocate_allocatable
original annotation:
if ((logical(kind=4)) __builtin_expect ((integer(kind=8)) (overflow.343 !=
0), 0, 33)) // overflow
{
stat.341 = 5014;
}
else
{
if ((logical(kind=4)) __builtin_expect ((integer(kind=8)) (bx_ilow.data
!= 0B), 0, 34)) // fail alloc
{
stat.341 = 5014;
}
else
{
stat.341 = 0;
bx_ilow.data = (void * restrict) __builtin_malloc (MAX_EXPR
<size.342, 1>);
if (bx_ilow.data == 0B)
{
stat.341 = 5014;
}
}
}
if ((logical(kind=4)) __builtin_expect ((integer(kind=8)) (stat.341 == 0),
1, 34)) // fail alloc
{
bx_ilow.dtype = 539;
bx_ilow.dim[0].lbound = (integer(kind=8)) xstart;
bx_ilow.dim[0].ubound = 1;
bx_ilow.dim[0].stride = 1;
bx_ilow.dim[1].lbound = (integer(kind=8)) ystart;
bx_ilow.dim[1].ubound = D.5342;
bx_ilow.dim[1].stride = D.5341;
bx_ilow.dim[2].lbound = (integer(kind=8)) zstart;
bx_ilow.dim[2].ubound = D.5346;
bx_ilow.dim[2].stride = D.5345;
bx_ilow.offset = D.5352;
}
I changed it to:
if ((logical(kind=4)) __builtin_expect ((integer(kind=8)) (overflow.343 !=
0), 0, 33)) // overflow
{
stat.341 = 5014;
}
else
{
if ((logical(kind=4)) __builtin_expect ((integer(kind=8)) (bx_ilow.data
!= 0B), 0, 35)) // repeated allocation/deallocation
{
stat.341 = 5014;
}
else
{
stat.341 = 0;
bx_ilow.data = (void * restrict) __builtin_malloc (MAX_EXPR
<size.342, 1>);
if ((logical(kind=4)) __builtin_expect ((integer(kind=8))
(bx_ilow.data == 0B), 0, 34)) // fail alloc
{
stat.341 = 5014;
}
}
}
if (stat.341 == 0) // no expectation
{
bx_ilow.dtype = 539;
bx_ilow.dim[0].lbound = (integer(kind=8)) xstart;
bx_ilow.dim[0].ubound = 1;
bx_ilow.dim[0].stride = 1;
bx_ilow.dim[1].lbound = (integer(kind=8)) ystart;
bx_ilow.dim[1].ubound = D.5342;
bx_ilow.dim[1].stride = D.5341;
bx_ilow.dim[2].lbound = (integer(kind=8)) zstart;
bx_ilow.dim[2].ubound = D.5346;
bx_ilow.dim[2].stride = D.5345;
bx_ilow.offset = D.5352;
}
B) array allocation
<bb 9>:
# size.1478_3210 = PHI <0(7), size.1478_3743(8)>
_21 = _3740 != 0;
_22 = (integer(kind=8)) _21;
_23 = BUILTIN_EXPECT (_22, 0, 33); // overflow
_24 = (logical(kind=4)) _23;
if (_24 != 0)
goto <bb 13>;
else
goto <bb 10>;
<bb 10>:
_25 = hrval.data;
_26 = _25 != 0B;
_27 = (integer(kind=8)) _26;
_28 = BUILTIN_EXPECT (_27, 0, 34); // fail malloc
_29 = (logical(kind=4)) _28;
if (_29 != 0)
goto <bb 13>;
else
goto <bb 11>;
<bb 11>:
_30 = MAX_EXPR <size.1478_3210, 1>;
_31 = __builtin_malloc (_30);
hrval.data = _31;
if (_31 == 0B)
goto <bb 12>;
else
goto <bb 13>;
<bb 12>:
<bb 13>:
# stat.1477_3202 = PHI <5014(9), 5014(10), 0(11), 5014(12)>
_33 = stat.1477_3202 == 0;
_34 = (integer(kind=8)) _33;
_35 = BUILTIN_EXPECT (_34, 1, 34); // fail malloc
_36 = (logical(kind=4)) _35;
if (_36 != 0)
goto <bb 14>;
else
goto <bb 15>;
currently looks as follows:
<bb 9>:
# size.1478_3210 = PHI <0(7), size.1478_3743(8)>
_21 = _3740 != 0;
_22 = (integer(kind=8)) _21;
_23 = BUILTIN_EXPECT (_22, 0, 33); // overflow
_24 = (logical(kind=4)) _23;
if (_24 != 0)
goto <bb 13>;
else
goto <bb 10>;
<bb 10>:
_25 = hrval.data;
_26 = _25 != 0B;
_27 = (integer(kind=8)) _26;
_28 = BUILTIN_EXPECT (_27, 0, 35); // repeated allocation/deallocation
_29 = (logical(kind=4)) _28;
if (_29 != 0)
goto <bb 13>;
else
goto <bb 11>;
<bb 11>:
_30 = MAX_EXPR <size.1478_3210, 1>;
_31 = __builtin_malloc (_30);
hrval.data = _31;
_33 = _31 == 0B;
_34 = (integer(kind=8)) _33;
_35 = BUILTIN_EXPECT (_34, 0, 34); // fail alloc
_36 = (logical(kind=4)) _35;
if (_36 != 0)
goto <bb 12>;
else
goto <bb 13>;
<bb 12>:
<bb 13>:
# stat.1477_3202 = PHI <5014(9), 5014(10), 0(11), 5014(12)>
if (stat.1477_3202 == 0) // no prediction
goto <bb 14>;
else
goto <bb 15>;
I get following numbers with the patch applied:
1) polyhedron benchmark (aermod.f90.061i.profile):
HEURISTICS BRANCHES (REL) HITRATE
COVERAGE COVERAGE (REL)
repeated allocation/deallocation 194 4.1% 100.00% / 100.00%
194 194.00 0.0%
fail alloc 377 7.9% 100.00% / 100.00%
377 377.00 0.0%
b) 459.GemsFDTD SPEC2006 benchmark:
HEURISTICS BRANCHES (REL) HITRATE
COVERAGE COVERAGE (REL)
repeated allocation/deallocation 203 4.3% 100.00% / 100.00%
203 203.00 0.0%
fail alloc 378 8.0% 100.00% / 100.00%
378 378.00 0.0%
Even though all numbers are 100%, I would suggest to set FAIL_ALLOC to
PROB_VERY_LIKELY and
PRED_FORTRAN_REALLOC to PROB_LIKELY.
Ready to install the patch if it survives regression tests?
Thanks,
Martin
