On 03/08/2015 04:58 PM, Steve Kargl wrote:
On Mon, Mar 09, 2015 at 01:07:25AM +0200, Janne Blomqvist wrote:
So I would prefer if we just hardcode the error values in the frontend
(-HUGE, 0, 0), in case somebody tries to use the kind=1,2 versions,
thus also removing the need for the new library functions, keeping the
existing simpler ones instead. AFAICT this would be standards
conforming. Any other opinions on this?
Revised patch attached as requested. Regression tested on x86_64 linux. Typical
results are shown below. I will provide a test case for the test-suite.
$ ./a.out
KIND=1: -127 0 0
KIND=1: -127 0 0
KIND=1: -127 .00000000 0
-----------------------------------------------------------
KIND=2: -32767 0 0
KIND=2: -32767 .00000000 0
-----------------------------------------------------------
KIND=4: 57496123 1000 2147483647
KIND=4: 57496123 1000.00000 2147483647
-----------------------------------------------------------
KIND=8: 57496123484138 1000000000 9223372036854775807
KIND=8: 57496123522116 1000000000.0000000 9223372036854775807
-----------------------------------------------------------
KIND=10: 57496123575504 1000000000 9223372036854775807
KIND=10: 57496123612377 1000000000.00000000000 9223372036854775807
-----------------------------------------------------------
KIND=16: 57496123669210 1000000000 9223372036854775807
KIND=16: 57496123698413 1000000000.00000000000000000000000000
9223372036854775807
OK for trunk?
Regards,
Jerry
2015-03-14 Jerry DeLisle <jvdeli...@gcc.gnu.org>
PR fortran/64432
*trans-intrinisic.c (conv_intrinsic_system_clock): Check the
smallest kind passed in user arguments and hard-code results for
KIND=1 or KIND=2 to indicate no clock available.
2015-03-14 Jerry DeLisle <jvdeli...@gcc.gnu.org>
PR libgfortran/64432
* intrinsics/system_clock.c (system_clock4, system_clock8):
Cleanup some whitespace.
