[Bug fortran/38291] Rejects I/O with POS= if FMT=*
--- Comment #12 from jvdelisle at gcc dot gnu dot org 2008-12-06 21:54 --- Subject: Bug 38291 Author: jvdelisle Date: Sat Dec 6 21:53:11 2008 New Revision: 142528 URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=142528 Log: 2008-12-06 Jerry DeLisle [EMAIL PROTECTED] PR libfortran/38291 * io/transfer.c (data_transfer_init): Add fbuf_flush inadvertently ommitted. Add check for invalid use of REC= with ACCESS=stream. Fix comment. Modified: trunk/libgfortran/ChangeLog trunk/libgfortran/io/transfer.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38291
[Bug fortran/38291] Rejects I/O with POS= if FMT=*
--- Comment #9 from jvdelisle at gcc dot gnu dot org 2008-12-06 04:15 --- Subject: Bug 38291 Author: jvdelisle Date: Sat Dec 6 04:13:34 2008 New Revision: 142515 URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=142515 Log: 2008-12-05 Jerry DeLisle [EMAIL PROTECTED] PR fortran/38291 * io.c (match_dt_element): Use dt-pos in matcher. (gfc_free_dt): Free dt-pos after use. (gfc_resolve_dt): Use dt-pos in resolution of stream position tag. 2008-12-05 Jerry DeLisle [EMAIL PROTECTED] PR libfortran/38291 * io/transfer.c (data_transfer_init): Add checks for POS= valid range. Add check for unit opened with ACCESS=stream. Flush and seek if current stream position does not match. Check ENDFILE on read. Modified: trunk/gcc/fortran/ChangeLog trunk/gcc/fortran/io.c trunk/libgfortran/ChangeLog trunk/libgfortran/io/transfer.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38291
[Bug fortran/38291] Rejects I/O with POS= if FMT=*
--- Comment #10 from jvdelisle at gcc dot gnu dot org 2008-12-06 04:19 --- Subject: Bug 38291 Author: jvdelisle Date: Sat Dec 6 04:17:31 2008 New Revision: 142516 URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=142516 Log: 2008-12-05 Jerry DeLisle [EMAIL PROTECTED] PR fortran/38291 * gfortran.dg/streamio_16.f90: New test. Added: trunk/gcc/testsuite/gfortran.dg/streamio_16.f90 Modified: trunk/gcc/testsuite/ChangeLog -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38291
[Bug fortran/38291] Rejects I/O with POS= if FMT=*
--- Comment #11 from jvdelisle at gcc dot gnu dot org 2008-12-06 04:19 --- Fixed on trunk. -- jvdelisle at gcc dot gnu dot org changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution||FIXED Target Milestone|--- |4.4.0 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38291
[Bug fortran/38291] Rejects I/O with POS= if FMT=*
--- Comment #8 from jvdelisle at gcc dot gnu dot org 2008-12-05 05:39 --- The alternating error was an artefact of the test case and not part of the bug. I have submitted a fixed patch that eliminates the incorrect EOF error. I will commit soon. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38291
[Bug fortran/38291] Rejects I/O with POS= if FMT=*
--- Comment #7 from jvdelisle at gcc dot gnu dot org 2008-12-04 05:32
---
I am holding off on committing the patch.
With this test case I have found a nasty problem:
! { dg-do run }
! PR38291 Rejects I/O with POS= if FMT=*
character(15) :: sAccess
character(1) :: instr
open(50, access=stream, form=formatted)
write(50, *, pos=1) Just something
read( 50, *,pos=7)
inquire(50, access=sAccess)
if (sAccess.ne.STREAM) call abort
close(50,status=delete)
end
With the patch submitted to list, I get unstable execution. Sometimes errors
with -m32, sometimes with -m64, Sometimes one execution to the next. I do not
know if this is a latent bug or one introduced by the patch. I suspect latent
because this is the first time we actually are using dtp-pos even though it
has been in the dtp structure. Here is a sample output:
$ gfc streamio_16.f90
$ ./a.out
At line 7 of file streamio_16.f90 (unit = 50, file = 'fort.50')
Fortran runtime error: End of file
$ ./a.out
$ ./a.out
At line 7 of file streamio_16.f90 (unit = 50, file = 'fort.50')
Fortran runtime error: End of file
$ ./a.out
$ ./a.out
At line 7 of file streamio_16.f90 (unit = 50, file = 'fort.50')
Fortran runtime error: End of file
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38291
[Bug fortran/38291] Rejects I/O with POS= if FMT=*
--- Comment #4 from jvdelisle at gcc dot gnu dot org 2008-12-01 16:32 --- So the read statement by itself is invalid. ??? I was testing with and without the open statement in the test case and saw that we were not catching that error either. Regardless, the problem is that I was using dt-rec instead of dt-pos and our runtime just defaulted to using the dt-rec for stream I/O. This fix is that we have to use dt-pos in the matchers and actually test conditions for pos= instead of rec= . I will post the patch tonight. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38291
[Bug fortran/38291] Rejects I/O with POS= if FMT=*
--- Comment #5 from burnus at gcc dot gnu dot org 2008-12-01 16:40 --- (In reply to comment #4) So the read statement by itself is invalid. ??? I was testing with and without the open statement in the test case and saw that we were not catching that error either. Ah, you are also talking about the run-time check in the library and not only about the compile-time check in f951. I will post the patch tonight. Thanks a ton! -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38291
[Bug fortran/38291] Rejects I/O with POS= if FMT=*
--- Comment #6 from jvdelisle at gcc dot gnu dot org 2008-12-02 05:35 --- Patch submitted. There is ABI breakage with 4.3 in that patch I noticed after I submitted, so I am now testing some tweaks. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38291
[Bug fortran/38291] Rejects I/O with POS= if FMT=*
--- Comment #2 from jvdelisle at gcc dot gnu dot org 2008-12-01 03:55 --- read( 50, *, pos = 1 ) is valid only if the unit has been connected for STREAM access. F2003 9.5.1.10POS= specifier in a data transfer statement The POS= specifier specifies the file position in file storage units. This specifier may appear in a data transfer statement only if the statement specifies a unit connected for stream access. A child data transfer statement shall not specify this specifier. So the read statement by itself is invalid. However, we should generate better error message for this case. I have a patch on the way. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38291
[Bug fortran/38291] Rejects I/O with POS= if FMT=*
--- Comment #3 from burnus at gcc dot gnu dot org 2008-12-01 06:44 --- read( 50, *, pos = 1 ) is valid only if the unit has been connected for STREAM access. Well, (a) I don't see how this can be tested at compile time and (b) I thought that open(50,access='stream') is doing so. So the read statement by itself is invalid. ??? -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38291
[Bug fortran/38291] Rejects I/O with POS= if FMT=*
--- Comment #1 from jvdelisle at gcc dot gnu dot org 2008-11-27 14:20 --- I am on it. -- jvdelisle at gcc dot gnu dot org changed: What|Removed |Added AssignedTo|unassigned at gcc dot gnu |jvdelisle at gcc dot gnu dot |dot org |org Status|UNCONFIRMED |ASSIGNED Ever Confirmed|0 |1 Last reconfirmed|-00-00 00:00:00 |2008-11-27 14:20:54 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38291
