[patch, fortran] PR35994 [4.3/4.4 regression] MAXLOC and MINLOC off by one with mask

[Jerry DeLisle]

Hi,

Starting from FX original patch given in the PR which did not quite work and 
applying some intuitive logic, I arrived at the following patch which solves the 
problem,

I will commit as obvious or at least invoke the mutually understood clause.  It 
is certainly simple.

Regression tested on x86-64.  Two new test cases.

Jerry

2008-04-23  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
	    Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>

	PR fortran/35994
	* trans-instrinsic.c (gfc_conv_intrinsic_minmaxloc): Correctly adjust
	loop counter offset.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pr35994.diff
Type: text/x-patch
Size: 811 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/fortran/attachments/20080424/58f950fd/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: minmaxloc_5.f90
Type: text/x-fortran
Size: 450 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/fortran/attachments/20080424/58f950fd/attachment-0001.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: minmaxloc_6.f90
Type: text/x-fortran
Size: 290 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/fortran/attachments/20080424/58f950fd/attachment-0002.bin>