[patch, fortran] Implement BACK for minloc and maxloc

Thomas Koenig tkoenig@netcologne.de
Tue May 1 20:07:00 GMT 2018


Hello world,

now that the release of gcc 8 is really around the corner, it's time
to add some new features :-) I had mostly completed this a few
months ago, but by then we were too deep into regression fixes
only mode.

The attached patch implements the BACK argument for minloc and maxloc,
in all three parts - simplification, inline and library.

I didn't actually go backwards though the array with BACK set to
true. Instead, I used comparisons which included equality.

At first, I was a bit concerned that adding the argument would lead
to inefficient code. Here, it seems that adding gfc_unlikely to
the condition really helped.  It would probably make sense to add
this to minval and maxval as well, but that is for another time.

With the short benchmark program

program main
   call x()
contains
   subroutine x()
     integer, dimension(5000,5000) :: a
     integer, dimension(5000*5000) :: b
     real, dimension(5000*5000) :: r
     character(len=100) :: line
     call random_number(r)
     b = r * 100000
     a = reshape(b,shape(a))
     call cpu_time (t1)
     write (unit=line,fmt=*) maxloc(a)
     call cpu_time (t2)
     print '(A25, F10.6)',"maxloc rank 2:", t2-t1
     call cpu_time (t1)
     write (unit=line,fmt=*) maxloc(b,dim=1)
     call cpu_time (t2)
     print '(A25,F10.6)',"maxloc inline:", t2-t1
     call cpu_time (t1)
     write (unit=line,fmt=*) sum(maxloc(a,dim=1))
     call cpu_time (t2)
     print '(A25, F10.6)',"maxloc rank 2, dim=1:", t2-t1
     call cpu_time (t1)
     write (unit=line,fmt=*) sum(maxloc(a,dim=2))
     call cpu_time (t2)
     print '(A25, F10.6)',"maxloc rank 2, dim=2:", t2-t1
   end subroutine x
end program main

I got with the patch on current trunk

$ gfortran -Ofast -march=native ibench.f90 && ./a.out
            maxloc rank 2:  0.009175
            maxloc inline:  0.007585
     maxloc rank 2, dim=1:  0.014779
     maxloc rank 2, dim=2:  0.038609

and with gfortran-7 (haven't build a gcc 8 yet)

$ /usr/bin/gfortran-7 -Ofast -march=native ibench.f90 && ./a.out
            maxloc rank 2:  0.011980
            maxloc inline:  0.013796
     maxloc rank 2, dim=1:  0.021252
     maxloc rank 2, dim=2:  0.037773

Regression-tested. OK for trunk? I'll hold off committing until
gcc 8 is actually released, just in case.

Regards

	Thomas

2018-05-01  Thomas Koenig  <tkoenig@gcc.gnu.org>

	PR fortran/54613
	* check.c (gfc_check_minmaxloc): Remove error for BACK not being
	implemented.  Use gfc_logical_4_kind for BACK.
	* simplify.c (min_max_choose): Add optional argument back_val.
	Handle it.
	(simplify_minmaxloc_to_scalar): Add argument back_val. Pass
	back_val to min_max_choose.
	(simplify_minmaxloc_to_nodim): Likewise.
	(simplify_minmaxloc_to_array): Likewise.
	(gfc_simplify_minmaxloc): Add argument back, handle it.
	Pass back_val to specific simplification functions.
	(gfc_simplify_minloc): Remove ATTRIBUTE_UNUSED from argument back,
	pass it on to gfc_simplify_minmaxloc.
	(gfc_simplify_maxloc): Likewise.
	* trans-intrinsic.c (gfc_conv_intrinsic_minmaxloc): Adjust
	comment. If BACK is true, use greater or equal (or lesser or
	equal) insteal of greater (or lesser). Mark the condition of
	having found a value which exceeds the limit as unlikely.

2018-05-01  Thomas Koenig  <tkoenig@gcc.gnu.org>

	PR fortran/54613
	* m4/iforeach-s.m4: Remove assertion that back is zero.
	* m4/iforeach.m4: Likewise.  Remove leading 'do'
	before implementation start.
	* m4/ifunction-s.m4: Remove assertion that back is zero.
	* m4/ifunction.m4: Likewise.  Remove for loop if HAVE_BACK_ARG
	is defined.
	* m4/maxloc0.m4: Reorganize loops. Split loops between >= and =,
	depending if back is true.  Mark the condition of having
	found a value which exceeds the limit as unlikely.
	* m4/minloc0.m4: Likewise.
	* m4/maxloc1.m4: Likewise.
	* m4/minloc1.m4: Likewise.
	* m4/maxloc1s.m4: Handle back argument.
	* m4/minloc1s.m4: Likewise.
	* m4/maxloc2s.m4: Remove assertion that back is zero.
	Remove special handling of loop start. Handle back argument.
	* m4/minloc2s.m4: Likewise.
	* generated/iall_i1.c: Regenerated.
	* generated/iall_i16.c: Regenerated.
	* generated/iall_i2.c: Regenerated.
	* generated/iall_i4.c: Regenerated.
	* generated/iall_i8.c: Regenerated.
	* generated/iany_i1.c: Regenerated.
	* generated/iany_i16.c: Regenerated.
	* generated/iany_i2.c: Regenerated.
	* generated/iany_i4.c: Regenerated.
	* generated/iany_i8.c: Regenerated.
	* generated/iparity_i1.c: Regenerated.
	* generated/iparity_i16.c: Regenerated.
	* generated/iparity_i2.c: Regenerated.
	* generated/iparity_i4.c: Regenerated.
	* generated/iparity_i8.c: Regenerated.
	* generated/maxloc0_16_i1.c: Regenerated.
	* generated/maxloc0_16_i16.c: Regenerated.
	* generated/maxloc0_16_i2.c: Regenerated.
	* generated/maxloc0_16_i4.c: Regenerated.
	* generated/maxloc0_16_i8.c: Regenerated.
	* generated/maxloc0_16_r10.c: Regenerated.
	* generated/maxloc0_16_r16.c: Regenerated.
	* generated/maxloc0_16_r4.c: Regenerated.
	* generated/maxloc0_16_r8.c: Regenerated.
	* generated/maxloc0_16_s1.c: Regenerated.
	* generated/maxloc0_16_s4.c: Regenerated.
	* generated/maxloc0_4_i1.c: Regenerated.
	* generated/maxloc0_4_i16.c: Regenerated.
	* generated/maxloc0_4_i2.c: Regenerated.
	* generated/maxloc0_4_i4.c: Regenerated.
	* generated/maxloc0_4_i8.c: Regenerated.
	* generated/maxloc0_4_r10.c: Regenerated.
	* generated/maxloc0_4_r16.c: Regenerated.
	* generated/maxloc0_4_r4.c: Regenerated.
	* generated/maxloc0_4_r8.c: Regenerated.
	* generated/maxloc0_4_s1.c: Regenerated.
	* generated/maxloc0_4_s4.c: Regenerated.
	* generated/maxloc0_8_i1.c: Regenerated.
	* generated/maxloc0_8_i16.c: Regenerated.
	* generated/maxloc0_8_i2.c: Regenerated.
	* generated/maxloc0_8_i4.c: Regenerated.
	* generated/maxloc0_8_i8.c: Regenerated.
	* generated/maxloc0_8_r10.c: Regenerated.
	* generated/maxloc0_8_r16.c: Regenerated.
	* generated/maxloc0_8_r4.c: Regenerated.
	* generated/maxloc0_8_r8.c: Regenerated.
	* generated/maxloc0_8_s1.c: Regenerated.
	* generated/maxloc0_8_s4.c: Regenerated.
	* generated/maxloc1_16_i1.c: Regenerated.
	* generated/maxloc1_16_i16.c: Regenerated.
	* generated/maxloc1_16_i2.c: Regenerated.
	* generated/maxloc1_16_i4.c: Regenerated.
	* generated/maxloc1_16_i8.c: Regenerated.
	* generated/maxloc1_16_r10.c: Regenerated.
	* generated/maxloc1_16_r16.c: Regenerated.
	* generated/maxloc1_16_r4.c: Regenerated.
	* generated/maxloc1_16_r8.c: Regenerated.
	* generated/maxloc1_16_s1.c: Regenerated.
	* generated/maxloc1_16_s4.c: Regenerated.
	* generated/maxloc1_4_i1.c: Regenerated.
	* generated/maxloc1_4_i16.c: Regenerated.
	* generated/maxloc1_4_i2.c: Regenerated.
	* generated/maxloc1_4_i4.c: Regenerated.
	* generated/maxloc1_4_i8.c: Regenerated.
	* generated/maxloc1_4_r10.c: Regenerated.
	* generated/maxloc1_4_r16.c: Regenerated.
	* generated/maxloc1_4_r4.c: Regenerated.
	* generated/maxloc1_4_r8.c: Regenerated.
	* generated/maxloc1_4_s1.c: Regenerated.
	* generated/maxloc1_4_s4.c: Regenerated.
	* generated/maxloc1_8_i1.c: Regenerated.
	* generated/maxloc1_8_i16.c: Regenerated.
	* generated/maxloc1_8_i2.c: Regenerated.
	* generated/maxloc1_8_i4.c: Regenerated.
	* generated/maxloc1_8_i8.c: Regenerated.
	* generated/maxloc1_8_r10.c: Regenerated.
	* generated/maxloc1_8_r16.c: Regenerated.
	* generated/maxloc1_8_r4.c: Regenerated.
	* generated/maxloc1_8_r8.c: Regenerated.
	* generated/maxloc1_8_s1.c: Regenerated.
	* generated/maxloc1_8_s4.c: Regenerated.
	* generated/maxloc2_16_s1.c: Regenerated.
	* generated/maxloc2_16_s4.c: Regenerated.
	* generated/maxloc2_4_s1.c: Regenerated.
	* generated/maxloc2_4_s4.c: Regenerated.
	* generated/maxloc2_8_s1.c: Regenerated.
	* generated/maxloc2_8_s4.c: Regenerated.
	* generated/maxval_i1.c: Regenerated.
	* generated/maxval_i16.c: Regenerated.
	* generated/maxval_i2.c: Regenerated.
	* generated/maxval_i4.c: Regenerated.
	* generated/maxval_i8.c: Regenerated.
	* generated/maxval_r10.c: Regenerated.
	* generated/maxval_r16.c: Regenerated.
	* generated/maxval_r4.c: Regenerated.
	* generated/maxval_r8.c: Regenerated.
	* generated/minloc0_16_i1.c: Regenerated.
	* generated/minloc0_16_i16.c: Regenerated.
	* generated/minloc0_16_i2.c: Regenerated.
	* generated/minloc0_16_i4.c: Regenerated.
	* generated/minloc0_16_i8.c: Regenerated.
	* generated/minloc0_16_r10.c: Regenerated.
	* generated/minloc0_16_r16.c: Regenerated.
	* generated/minloc0_16_r4.c: Regenerated.
	* generated/minloc0_16_r8.c: Regenerated.
	* generated/minloc0_16_s1.c: Regenerated.
	* generated/minloc0_16_s4.c: Regenerated.
	* generated/minloc0_4_i1.c: Regenerated.
	* generated/minloc0_4_i16.c: Regenerated.
	* generated/minloc0_4_i2.c: Regenerated.
	* generated/minloc0_4_i4.c: Regenerated.
	* generated/minloc0_4_i8.c: Regenerated.
	* generated/minloc0_4_r10.c: Regenerated.
	* generated/minloc0_4_r16.c: Regenerated.
	* generated/minloc0_4_r4.c: Regenerated.
	* generated/minloc0_4_r8.c: Regenerated.
	* generated/minloc0_4_s1.c: Regenerated.
	* generated/minloc0_4_s4.c: Regenerated.
	* generated/minloc0_8_i1.c: Regenerated.
	* generated/minloc0_8_i16.c: Regenerated.
	* generated/minloc0_8_i2.c: Regenerated.
	* generated/minloc0_8_i4.c: Regenerated.
	* generated/minloc0_8_i8.c: Regenerated.
	* generated/minloc0_8_r10.c: Regenerated.
	* generated/minloc0_8_r16.c: Regenerated.
	* generated/minloc0_8_r4.c: Regenerated.
	* generated/minloc0_8_r8.c: Regenerated.
	* generated/minloc0_8_s1.c: Regenerated.
	* generated/minloc0_8_s4.c: Regenerated.
	* generated/minloc1_16_i1.c: Regenerated.
	* generated/minloc1_16_i16.c: Regenerated.
	* generated/minloc1_16_i2.c: Regenerated.
	* generated/minloc1_16_i4.c: Regenerated.
	* generated/minloc1_16_i8.c: Regenerated.
	* generated/minloc1_16_r10.c: Regenerated.
	* generated/minloc1_16_r16.c: Regenerated.
	* generated/minloc1_16_r4.c: Regenerated.
	* generated/minloc1_16_r8.c: Regenerated.
	* generated/minloc1_16_s1.c: Regenerated.
	* generated/minloc1_16_s4.c: Regenerated.
	* generated/minloc1_4_i1.c: Regenerated.
	* generated/minloc1_4_i16.c: Regenerated.
	* generated/minloc1_4_i2.c: Regenerated.
	* generated/minloc1_4_i4.c: Regenerated.
	* generated/minloc1_4_i8.c: Regenerated.
	* generated/minloc1_4_r10.c: Regenerated.
	* generated/minloc1_4_r16.c: Regenerated.
	* generated/minloc1_4_r4.c: Regenerated.
	* generated/minloc1_4_r8.c: Regenerated.
	* generated/minloc1_4_s1.c: Regenerated.
	* generated/minloc1_4_s4.c: Regenerated.
	* generated/minloc1_8_i1.c: Regenerated.
	* generated/minloc1_8_i16.c: Regenerated.
	* generated/minloc1_8_i2.c: Regenerated.
	* generated/minloc1_8_i4.c: Regenerated.
	* generated/minloc1_8_i8.c: Regenerated.
	* generated/minloc1_8_r10.c: Regenerated.
	* generated/minloc1_8_r16.c: Regenerated.
	* generated/minloc1_8_r4.c: Regenerated.
	* generated/minloc1_8_r8.c: Regenerated.
	* generated/minloc1_8_s1.c: Regenerated.
	* generated/minloc1_8_s4.c: Regenerated.
	* generated/minloc2_16_s1.c: Regenerated.
	* generated/minloc2_16_s4.c: Regenerated.
	* generated/minloc2_4_s1.c: Regenerated.
	* generated/minloc2_4_s4.c: Regenerated.
	* generated/minloc2_8_s1.c: Regenerated.
	* generated/minloc2_8_s4.c: Regenerated.
	* generated/minval_i1.c: Regenerated.
	* generated/minval_i16.c: Regenerated.
	* generated/minval_i2.c: Regenerated.
	* generated/minval_i4.c: Regenerated.
	* generated/minval_i8.c: Regenerated.
	* generated/minval_r10.c: Regenerated.
	* generated/minval_r16.c: Regenerated.
	* generated/minval_r4.c: Regenerated.
	* generated/minval_r8.c: Regenerated.
	* generated/norm2_r10.c: Regenerated.
	* generated/norm2_r16.c: Regenerated.
	* generated/norm2_r4.c: Regenerated.
	* generated/norm2_r8.c: Regenerated.
	* generated/parity_l1.c: Regenerated.
	* generated/parity_l16.c: Regenerated.
	* generated/parity_l2.c: Regenerated.
	* generated/parity_l4.c: Regenerated.
	* generated/parity_l8.c: Regenerated.
	* generated/product_c10.c: Regenerated.
	* generated/product_c16.c: Regenerated.
	* generated/product_c4.c: Regenerated.
	* generated/product_c8.c: Regenerated.
	* generated/product_i1.c: Regenerated.
	* generated/product_i16.c: Regenerated.
	* generated/product_i2.c: Regenerated.
	* generated/product_i4.c: Regenerated.
	* generated/product_i8.c: Regenerated.
	* generated/product_r10.c: Regenerated.
	* generated/product_r16.c: Regenerated.
	* generated/product_r4.c: Regenerated.
	* generated/product_r8.c: Regenerated.
	* generated/sum_c10.c: Regenerated.
	* generated/sum_c16.c: Regenerated.
	* generated/sum_c4.c: Regenerated.
	* generated/sum_c8.c: Regenerated.
	* generated/sum_i1.c: Regenerated.
	* generated/sum_i16.c: Regenerated.
	* generated/sum_i2.c: Regenerated.
	* generated/sum_i4.c: Regenerated.
	* generated/sum_i8.c: Regenerated.
	* generated/sum_r10.c: Regenerated.
	* generated/sum_r16.c: Regenerated.
	* generated/sum_r4.c: Regenerated.
	* generated/sum_r8.c: Regenerated.

2018-01-15  Thomas Koenig  <tkoenig@gcc.gnu.org>

	PR fortran/54613
	* gfortran.dg/minmaxloc_12.f90: New test case.
	* gfortran.dg/minmaxloc_13.f90: New test case.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: pi-9a.diff
Type: text/x-patch
Size: 451275 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/fortran/attachments/20180501/9a8fb931/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: minmaxloc_12.f90
Type: text/x-fortran
Size: 3117 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/fortran/attachments/20180501/9a8fb931/attachment-0001.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: minmaxloc_13.f90
Type: text/x-fortran
Size: 2937 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/fortran/attachments/20180501/9a8fb931/attachment-0002.bin>


More information about the Fortran mailing list