[Bug fortran/36103] New: gfortran 4.4.0-20060501 failed to compile this program

[Bug fortran/36103] New: gfortran 4.4.0-20060501 failed to compile this program

[linuxl4 at sohu dot com]

program main
implicit none
real :: start,end
real :: t=0.5
integer :: i
call cpu_time(start)
do i=1,10000000
t=sin(t)
enddo
call cpu_time(end)
print '(f6.3)',end-start
end program main

------------------------
gfortran 430 and 423 are  ok.
gfortran 4.4.0 failed.


-- 
           Summary: gfortran 4.4.0-20060501 failed to compile this program
           Product: gcc
           Version: 4.4.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: fortran
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: linuxl4 at sohu dot com
 GCC build triplet: i686-pc-linux-gnu
  GCC host triplet: i686-pc-linux-gnu
GCC target triplet: i686-pc-linux-gnu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran 4.4.0-20060501 failed to compile this program

[kargl at gcc dot gnu dot org]

------- Comment #1 from kargl at gcc dot gnu dot org  2008-05-01 19:00 -------
What do you mean by 'failed', which is a fairly generic word?
Show that actually error message!


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran 4.4.0-20060501 failed to compile this program

[burnus at gcc dot gnu dot org]

------- Comment #2 from burnus at gcc dot gnu dot org  2008-05-01 19:36 -------
> call cpu_time(end)
> print '(f6.3)',end-start
> ------------------------
> gfortran 430 and 423 are  ok.
> gfortran 4.4.0 failed.

If I understand it correctly, the problem is that gfortran 4.4.0 is much slower
than previous versions. Here (amd64), I do not see any timing differences
between 4.1, 4.2, 4.3 and 4.4 with either -m32 nor -m64.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran 4.4.0-20060501 failed to compile this program

[linuxl4 at sohu dot com]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 545 bytes --]



------- Comment #3 from linuxl4 at sohu dot com  2008-05-01 20:25 -------
gfortran 4.4 failed to make a.out file, try this ::

export LC_ALL=zh_CN.UTF-8
gfortran cputime.f90

then:
----------------------------
f951: ±àÒëÆ÷ÄÚ²¿´íÎó£ºÔÚ error_print ÖУ¬ÓÚ fortran/error.c:497
ÇëÌá½»Ò»·ÝÍêÕûµÄ´íÎ󱨸棬
ÈçÓпÉÄÜÇ븽ÉϾ­Ô¤´¦ÀíºóµÄÔ´Îļþ¡£
¾ßÌå²½ÖèÇë²Î¼û <http://gcc.gnu.org/bugs.html>¡£


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran 4.4.0-20060501 failed to compile this program

[kargl at gcc dot gnu dot org]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 699 bytes --]



------- Comment #4 from kargl at gcc dot gnu dot org  2008-05-01 20:50 -------
(In reply to comment #3)
> gfortran 4.4 failed to make a.out file, try this ::
> 
> export LC_ALL=zh_CN.UTF-8
> gfortran cputime.f90
> 
> then:
> ----------------------------
> f951: ±àÒëÆ÷ÄÚ²¿´íÎó£ºÔÚ error_print ÖУ¬ÓÚ fortran/error.c:497
> ÇëÌá½»Ò»·ÝÍêÕûµÄ´íÎ󱨸棬
> ÈçÓпÉÄÜÇ븽ÉϾ­Ô¤´¦ÀíºóµÄÔ´Îļþ¡£
> ¾ßÌå²½ÖèÇë²Î¼û <http://gcc.gnu.org/bugs.html>¡£
> 

This information should have been in the intial bug report!
What happens if you don't set LC_ALL?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran 4.4.0-20060501 failed to compile this program

[pinskia at gcc dot gnu dot org]

------- Comment #5 from pinskia at gcc dot gnu dot org  2008-05-01 20:53 -------
(In reply to comment #4)
> This information should have been in the intial bug report!
> What happens if you don't set LC_ALL?

And if that works, then bug is that the translation needs to be changed but we
don't handle translations.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran 4.4.0-20060501 failed to compile this program

[linuxl4 at sohu dot com]

------- Comment #6 from linuxl4 at sohu dot com  2008-05-01 21:43 -------
If I export LC_ALL=POSIX
then gfortran 4.4 works.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran 4.4.0-20060501 failed to compile this program

[dominiq at lps dot ens dot fr]

------- Comment #7 from dominiq at lps dot ens dot fr  2008-05-01 21:48 -------
Confirmed on *-apple-darwin9 with "setenv LC_ALL zh_CN.UTF-8". The error does
not show up with 

LANG=fr
LC_ALL=C
LC_COLLATE=POSIX

or if I unset LC_ALL. It is a regression on 4.4, it works on 4.3.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran 4.4.0-20060501 failed to compile this program

[burnus at gcc dot gnu dot org]

------- Comment #8 from burnus at gcc dot gnu dot org  2008-05-02 09:48 -------
Works for me for 4.4.0 20080424 on a X86-64 openSUSE Factory (approx. oS
11.0beta2) with a glibc 2.8 (20080410).

I wonder whether it has something to do with the recent sign/decimal comma
work.

Could those for whose it fails try to find out which of the various LC_*
options fails?

For instance: LC_ALL=C LC_NUMERIC=zh_CN.UTF-8  or LC_CTYPE="zh_CN.UTF-8" or ...

> It is a regression on 4.4, it works on 4.3.
I assume a 4.3 program with a 4.4 library also fails?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran 4.4.0-20060501 failed to compile this program

[dominiq at lps dot ens dot fr]

------- Comment #9 from dominiq at lps dot ens dot fr  2008-05-02 10:08 -------
> Could those for whose it fails try to find out which of the various LC_*
> options fails?
>
> For instance: LC_ALL=C LC_NUMERIC=zh_CN.UTF-8  or LC_CTYPE="zh_CN.UTF-8" or ...

Although I did not exhaust all the combinations, I saw the problem with
"LC_ALL=zh_CN.*". The zh_(HK|TW).* I have tested work and the different
LC_(.not.ALL)=zh_CN.UTF-8 tested also work.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug translation/36103] gfortran crash with zh_CN locale

[burnus at gcc dot gnu dot org]

------- Comment #10 from burnus at gcc dot gnu dot org  2008-05-02 13:40 -------
I can finally reproduce this - but for some reason not with 4.4 but with 4.3.
(I somehow misread the description how one should reproduce it.)


The problem is surely in http://gcc.gnu.org/viewcvs/trunk/gcc/po/zh_TW.po
respectively in http://translationproject.org/latest/gcc/zh_CN.po


I believe the problem is that something goes wrong with the % strings.

For instance in the following, for %s %s %L the output order is changed to
%3$L %1$s %2$s. The following is correct, but for one of the other strings
there is something wrong - the <n>$ might be missing or the number is wrong or
similar. The task is now to find the place. Afterwards, the Chinese translation
team needs to fix it (http://translationproject.org/team/zh_CN.html) - it
should be transferred to the Component "Translation" in bugzilla.

msgid "Arithmetic overflow converting %s to %s at %L. This check can be
disabled with the option -fno-range-check"
msgstr "%3$L&#22788;&#23558; %1$s &#36716;&#25442;&#21040; %2$s
&#26102;&#31639;&#26415;&#28322;&#20986;&#12290;&#36825;&#19968;&#26816;&#26597;&#21487;&#29992;
-fno-range-check &#36873;&#39033;&#20851;&#38381;"


Is someone able to get a backtrace? One can then try to find the string which
makes the problem. (I currently cannot as my GCC-developer computer is turned
off.)


-- 

burnus at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |burnus at gcc dot gnu dot
                   |                            |org
             Status|UNCONFIRMED                 |NEW
          Component|fortran                     |translation
     Ever Confirmed|0                           |1
  GCC build triplet|i686-pc-linux-gnu           |
   GCC host triplet|i686-pc-linux-gnu           |
 GCC target triplet|i686-pc-linux-gnu           |
           Keywords|                            |ice-on-valid-code
   Last reconfirmed|0000-00-00 00:00:00         |2008-05-02 13:40:05
               date|                            |
            Summary|gfortran 4.4.0-20060501     |gfortran crash with zh_CN
                   |failed to compile this      |locale
                   |program                     |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug translation/36103] gfortran crash with zh_CN locale

[burnus at gcc dot gnu dot org]

------- Comment #11 from burnus at gcc dot gnu dot org  2008-05-02 13:44 -------
(In reply to comment #10)
> The problem is surely in http://gcc.gnu.org/viewcvs/trunk/gcc/po/zh_TW.po

I meant:                   http://gcc.gnu.org/viewcvs/trunk/gcc/po/zh_CN.po


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran crash with zh_CN locale - error_print error ?

[burnus at gcc dot gnu dot org]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1581 bytes --]



------- Comment #12 from burnus at gcc dot gnu dot org  2008-05-02 16:38 -------
I'm not sure whether there are other problems, but at least the following is
wrong. Maybe this is enough to fix the bug. Could someone check?
I'll write the translation team to fix it there.


msgid "'%s' argument of '%s' intrinsic at %L must be less than rank %d"
msgstr
"%3$L处内建函数‘%2$s’的实参‘%1$s’秩必须小于
%d"

The last item must be "%4$d" and not "%d". (See POSIX standard for "printf"
which mandates that either "%<n>$<f>" or "%<f>" has to be used, but no mixture
between both.)

This is wrong too:

msgid "Fortran 2003: %s specifier in %s statement at %C has value '%s'"
msgstr "Fortran 2003&#65306;%3$ &#22788; %2$ &#35821;&#21477;&#20013;&#30340;
%1$s &#38480;&#23450;&#31526;&#20540;&#20026;‘%4$s’"

The 's' and the 'C' are missing.


msgid "'%s' at %L is not a function"
msgstr
"%2L&#22788;&#30340;‘%1$s’&#19981;&#26159;&#19968;&#20010;&#20989;&#25968;"

Here, a '$' is missing.


-- 

burnus at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |fxcoudert at gcc dot gnu dot
                   |                            |org
            Summary|gfortran crash with zh_CN   |gfortran crash with zh_CN
                   |locale                      |locale - error_print error ?


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran crash with zh_CN locale - error_print error ?

[fxcoudert at gcc dot gnu dot org]

------- Comment #13 from fxcoudert at gcc dot gnu dot org  2008-05-02 17:34 -------
GCC translation bug reports are taken care of by the translation project
(http://translationproject.org/). I have sent a mail to the chinese team
coordinator (LI Daobing) and GCC translation maintainer (Meng Jie); their email
addresses are on the chinese team webpage
(http://translationproject.org/team/zh_CN.html).


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran crash with zh_CN locale - error_print error ?

[lidaobing at gmail dot com]

------- Comment #14 from lidaobing at gmail dot com  2008-05-02 18:50 -------
an updated zh_CN.po uploaded to TP[1], please help check whether it can fix
this bug.

PS. in emacs's po-mode, po-validate does not warning in this case, any script
to check this po file?

thanks.

[1] http://translationproject.org/PO-files/zh_CN/gcc-4.3.0.zh_CN.po


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran crash with zh_CN locale - error_print error ?

[linuxl4 at sohu dot com]

------- Comment #15 from linuxl4 at sohu dot com  2008-05-03 11:05 -------
also crashed.

would MR lidaobing work at gcc's  svn version ?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran crash with zh_CN locale - error_print error ?

[lidaobing at gmail dot com]

------- Comment #16 from lidaobing at gmail dot com  2008-05-03 12:42 -------
Created an attachment (id=15567)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15567&action=view)
new zh_CN.po, compressed by bzip2


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran crash with zh_CN locale - error_print error ?

[lidaobing at gmail dot com]

------- Comment #17 from lidaobing at gmail dot com  2008-05-03 12:47 -------
an updated (again) zh_CN.po uploaded to TP[1], and also in attachment[2], it
really protect my gfotran-4.3 from crash, I hope it can works for you.

please help check whether it works for you?

thanks.

[1] http://translationproject.org/PO-files/zh_CN/gcc-4.3.0.zh_CN.po
[2] http://gcc.gnu.org/bugzilla/attachment.cgi?id=15567


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran crash with zh_CN locale - error_print error ?

[linuxl4 at sohu dot com]

------- Comment #18 from linuxl4 at sohu dot com  2008-05-03 13:14 -------
It works fine.thanks!

msgfmt -o  gcc.mo   gcc-4.3.0.zh_CN.po
mv  gcc.mo  /gcc-4.4/share/locale/zh_CN/LC_MESSAGES/

gfortran cputime.f90;./a.out

the result:
0.55


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran crash with zh_CN locale - error_print error ?

[burnus at gcc dot gnu dot org]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1191 bytes --]



------- Comment #19 from burnus at gcc dot gnu dot org  2008-05-03 13:55 -------
> PS. in emacs's po-mode, po-validate does not warning in this case, any script
> to check this po file?

I had none, but now I wrote a very basic one, which misses several checks and
gives false positives. I put the Perl script at
http://physik.fu-berlin.de/~tburnus/tmp/check_po


Remaining problems found using my just created script:

  msgid  Unable to resolve the specific function '%s' at %L
  msgstr 不能解析
%$2L处的函数‘%1$s’

The problem is that %$2L should be %2$L.


Probably no problem, but at least inconsistent: Missing % before <:

ERROR: Wrong number of '%' signs(4 vs. 3)
  msgid  both %<long%> and %<short%> in declaration specifiers
  msgstr &#22768;&#26126;&#20013;&#21516;&#26102;&#20351;&#29992;&#20102;
<long%>&#21644;%<short%>

ERROR: Wrong number of '%' signs(16 vs. 15)
  msgid  expected %<+%>, %<*%>, %<-%>, %<&%>, %<^%>, %<|%>, %<&&%>, or %<||%>
  msgstr
&#38656;&#35201;%<+%>&#12289;<*%>&#12289;%<-%>&#12289;%<&%>&#12289;%<^%>&#12289;%<|%>&#12289;%<&&%>&#25110;%<||%>


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran crash with zh_CN locale - error_print error ?

[burnus at gcc dot gnu dot org]

------- Comment #20 from burnus at gcc dot gnu dot org  2008-05-03 14:07 -------
(In reply to comment #19)
> I had none, but now I wrote a very basic one, which misses several checks and
> gives false positives. I put the Perl script at
> http://physik.fu-berlin.de/~tburnus/tmp/check_po

FX, I think we should also audit the other PO files, what do you think?

> Remaining problems found using my just created script:
>   msgid  Unable to resolve the specific function '%s' at %L

Sorry, I was using the old *.po file for testing. This is fixed in the new PO
file. The %< vs. <  is still valid, but it should not cause any crash.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran crash with zh_CN locale - error_print error ?

[lidaobing at gmail dot com]

------- Comment #21 from lidaobing at gmail dot com  2008-05-03 14:36 -------
Created an attachment (id=15570)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15570&action=view)
2nd zh_CN.po, compressed by bzip2

fix almost all the warning by check_po.

I don't know the grammer of error_print, but I think the following two places
are valid. am I right?

thanks.

$ ./check_po gcc-4.3.0.zh_CN.po
ERROR: Cannot combine '%' with %n$
  msgid  ISO C forbids passing argument %d of %qE between function pointer and
%<void *%>
  msgstr ISO C &#19981;&#20801;&#35768;&#23558;%2$qE&#30340;&#31532; %1$d
&#20010;&#23454;&#21442;&#22312;&#20989;&#25968;&#25351;&#38024;&#21644;%<void
*%>&#38388;&#20256;&#36882;

ERROR: Cannot combine '%' with %n$
  msgid  converting %<false%> to pointer type for argument %P of %qD
  msgstr
&#23558;%<false%>&#36716;&#25442;&#20026;&#25351;&#21521;%2$qD&#30340;&#23454;&#21442;
%1$P &#30340;&#25351;&#38024;&#31867;&#22411;


-- 

lidaobing at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
  Attachment #15567|0                           |1
        is obsolete|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran crash with zh_CN locale - error_print error ?

[lidaobing at gmail dot com]

------- Comment #22 from lidaobing at gmail dot com  2008-05-03 14:46 -------
Created an attachment (id=15572)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15572&action=view)
check_po log for all .mo files in debian gcc-4.3-locales package

debian gcc-4.3-locales version is 4.3.0-3

and I use following command to generate the log file

for f in `dpkg -L gcc-4.3-locales | grep gcc-4.3.mo`; do echo $f; msgunfmt $f >
1.po; ./check_po 1.po; echo; done > check_po.log


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran crash with zh_CN locale - error_print error ?

[burnus at gcc dot gnu dot org]

------- Comment #23 from burnus at gcc dot gnu dot org  2008-05-03 16:06 -------
(In reply to comment #21)
> fix almost all the warning by check_po.
> 
> I don't know the grammer of error_print, but I think the following two places
> are valid. am I right?

I think they are OK.

check_po is really something quickly written. I'm sure it misses problems and
it will (like your example) find things which are OK. (But checking a dozen
lines is easier than checking all lines in the .po file.)

(In reply to comment #18)
> It works fine.thanks!

I close it than as FIXED. Note: It might take a while until the corrected .po
file is checked in.


-- 

burnus at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103

[Bug fortran/36103] gfortran crash with zh_CN locale - error_print error ?

[pinskia at gcc dot gnu dot org]

-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.4.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36103