https://gcc.gnu.org/bugzilla/show_bug.cgi?id=117838
--- Comment #4 from cuilili <lili.cui at intel dot com> ---
I created a patch to remove the issue code in improve_allocation. And extra a
small case from 548.exchange2_r.
Fortran small case:
---------------------------------------
module a
integer, parameter :: b = 3, d = b
contains
subroutine f(g, block)
integer block(:, :, :), g(:, :)
logical ac
do h = 1, d
do i = 1, d
block = d
block = 0
end do
if (ac) then
call j
call k
end if
do e = 0, dh
end do
ac = .false.
if (ao > 1) then
if (ac) then
else
ac = .true.
end if
end if
if (allg /= 0) exit
end do
contains
subroutine k
do e = 1, d
end do
end subroutine k
subroutine j
do bg = 0, c
if (count(g == 0) == 3) cycle
do ax = 1, dh
if (bm() /= 3) cycle
end do
do l = 1, cycle
do
end do
end do
end do
end subroutine j
end subroutine f
end module a
--------------------------------------------
Before improve_allocation function:
a537 (cost 1896, reg42)
a20 (cost 270, reg1)
a13 (cost 144, spill)
a551 (cost 70, reg40)
a5 (cost 43, spill)
a493 (cost 30, reg42)
a499 (cost 12, reg40)
------------------------------
Dump info in improve_allocation function:
Base:
Spilling a493r125 for a5r113
Spilling a573r202 for a5r113
Spilling a499r248 for a13r106
Spilling a551r120 for a13r106
Spilling a20r237 for a551r120
With patch:
Spilling a499r248 for a13r106
Spilling a551r120 for a13r106
Spilling a493r125 for a551r120
------------------------------
After assign_hard_reg (at the end of improve_allocation):
Base:
a537 (cost 1896, reg1)
a20 (cost 270, spill) -----> This is unreasonable
a13 (cost 144, reg40)
a551 (cost 70, reg1)
a5 (cost 43, reg42)
a493 (cost 30, spill)
a499 (cost 12, reg1)
With patch:
a537 (cost 1896, reg42)
a20 (cost 270, reg1)
a13 (cost 144, reg40)
a551 (cost 70, reg42)
a5 (cost 43, spill)
a493 (cost 30, spill)
a499 (cost 12, reg42)
-----------------------------
Collected spec2017 performance on Znver3/Graviton4/EMR/SRF for O2 and Ofast. No
performance regression was observed.
FOR multi-copy O2
SRF: 548.exchange2_r increased by 7.5%, 500.perlbench_r increased by 2.0%.
EMR: 548.exchange2_r increased by 4.5%, 500.perlbench_r increased by 1.7%.
Graviton4: 548.exchange2_r Increased by 2.2%, 511.povray_r increased by 2.8%.
Znver3 : 500.perlbench_r increased by 2.0%.
