* [patch, libgfortran] Fix oddity in formatted write
@ 2008-11-19 6:36 jvdelisle
2008-11-20 10:07 ` Tobias Burnus
0 siblings, 1 reply; 2+ messages in thread
From: jvdelisle @ 2008-11-19 6:36 UTC (permalink / raw)
To: gcc-patches, fortran
This is a very simple patch.
With this test case:
PROGRAM bug
IMPLICIT NONE
DOUBLE PRECISION r
COMMON /gggg91/ r
DOUBLE PRECISION x
x = 1000
write(6,*) 'x = ', x
x = 1001
write(6,*) 'x = ', x
x = 999
write(6,*) 'x = ', x
x = 100
write(6,*) 'x = ', x
x = 101
write(6,*) 'x = ', x
x = 99
write(6,*) 'x = ', x
END
Before this patch we get:
x = 1000.00000000000000
x = 1001.0000000000000
x = 999.00000000000000
x = 100.000000000000000
x = 101.00000000000000
x = 99.000000000000000
After the patch we get:
x = 1000.0000000000000
x = 1001.0000000000000
x = 999.00000000000000
x = 100.00000000000000
x = 101.00000000000000
x = 99.000000000000000
This was first reported in PR37472. This patch does not fix that PR but
does straighten out the formatting. It requires a slight change to
namelist_print_1.f90. Regression tested on x86-64-Gnu-linux.
OK for trunk?
Jerry
Index: write_float.def
===================================================================
--- write_float.def (revision 141988)
+++ write_float.def (working copy)
@@ -652,7 +652,10 @@ output_float_FMT_G_ ## x (st_parameter_d
low = mid + 1;\
}\
else\
- break;\
+ {\
+ mid++;\
+ break;\
+ }\
}\
\
if (e < 0)\
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [patch, libgfortran] Fix oddity in formatted write
2008-11-19 6:36 [patch, libgfortran] Fix oddity in formatted write jvdelisle
@ 2008-11-20 10:07 ` Tobias Burnus
0 siblings, 0 replies; 2+ messages in thread
From: Tobias Burnus @ 2008-11-20 10:07 UTC (permalink / raw)
To: jvdelisle; +Cc: gcc-patches, fortran
jvdelisle@verizon.net wrote:
> This was first reported in PR37472. This patch does not fix that PR
> but does straighten out the formatting. It requires a slight change
> to namelist_print_1.f90. Regression tested on x86-64-Gnu-linux.
> OK for trunk?
OK with a changelog.
(The output is then in line with ifort and also that the digits start
after a fixed amount of spaces after the "=" is in line with g95, f95,
openf95 and sunf95.)
Tobias
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2008-11-20 9:20 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2008-11-19 6:36 [patch, libgfortran] Fix oddity in formatted write jvdelisle
2008-11-20 10:07 ` Tobias Burnus
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).