http://bugs.winehq.org/show_bug.cgi?id=17417
Summary: winetest causes msvcrt:file test to fail Product: Wine Version: unspecified Platform: Other OS/Version: other Status: UNCONFIRMED Severity: normal Priority: P2 Component: testcases AssignedTo: wine-bugs@winehq.org ReportedBy: austin.lund@gmail.com
Running the following command:
$ ./programs/winetest/winetest -c -o out msvcrt:file && cat out
causes msvcrt:file tests to function correctly.
However if one runs:
$ ./programs/winetest/winetest -c -o out msvcrtd:debug msvcrt:file && cat out
Then I get one failure:
file.c:1057: Test failed: fstat(stdin) failed: errno=2 file.c:1066: Tests skipped: stdin is not a char device? st_mode=100666
I've had this on many recent versions. Tested with 1.1.15, 1.1.14 and 1.1.13.
http://bugs.winehq.org/show_bug.cgi?id=17417
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Component|testcases |msvcrt Keywords| |download, source, testcase Version|unspecified |1.1.13
http://bugs.winehq.org/show_bug.cgi?id=17417
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |NEW Ever Confirmed|0 |1
--- Comment #1 from Austin English austinenglish@gmail.com 2009-02-16 00:47:02 --- Confirming.
http://bugs.winehq.org/show_bug.cgi?id=17417
--- Comment #2 from Austin Lund austin.lund@gmail.com 2009-02-16 20:22:25 --- I found that on much eariler version this failure didn't happen.
The first failing commit is:
commit 5f8c3b4a3883c0042ac987d7375e99f5116e009d Author: Francois Gouget fgouget@free.fr Date: Tue Dec 9 15:18:16 2008 +0100
msvcrt/tests: Stricter fstat() tests.
Don't allow fstat() to fail for no reason. Better check the st_mode field.
The offending code section is at "static void test_stat(void)" in "dlls/msvcrt/tests/file.c":
if (_dup2(0, 10) == 0) { ok(fstat(10, &buf) == 0, "fstat(stdin) failed: errno=%d\n", errno);
(duh, the test failure message tells you this too).
http://bugs.winehq.org/show_bug.cgi?id=17417
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |austinenglish@gmail.com Keywords| |regression
http://bugs.winehq.org/show_bug.cgi?id=17417
--- Comment #3 from Austin Lund austin.lund@gmail.com 2009-02-18 00:05:22 --- I've done some more testing on this.
$ WINEDEBUG="trace+msvcrt,warn+msvcrt" ./programs/winetest/winetest -o out -c msvcrt:file 2>&1 | grep -A 5 dup trace:msvcrt:MSVCRT__dup2 (od=0, nd=10) trace:msvcrt:msvcrt_alloc_fd_from fdstart is 3, fdend is 11 trace:msvcrt:MSVCRT__fstat64 :fd (10) stat (0x33fc68) trace:msvcrt:MSVCRT__fstat64 :dwFileAttributes = 0x0, mode set to 0x2000 trace:msvcrt:MSVCRT__close :fd (10) handle (0x34) trace:msvcrt:msvcrt_free_fd :fd (10) freed
$ WINEDEBUG="trace+msvcrt,warn+msvcrt" ./programs/winetest/winetest -o out -c msvcrtd:debug msvcrt:file 2>&1 | grep -A 5 dup trace:msvcrt:MSVCRT__dup2 (od=0, nd=10) trace:msvcrt:msvcrt_alloc_fd_from fdstart is 3, fdend is 11 trace:msvcrt:MSVCRT__fstat64 :fd (10) stat (0x33fc68) warn:msvcrt:MSVCRT__fstat64 :failed-last error (6) trace:msvcrt:pf_vsnprintf format is L"%s:%d: Test failed" trace:msvcrt:pf_vsnprintf format is L": "
Having a look at the code for fstat64, it seems that
type = GetFileType(hand);
is not returning a type of FILE_TYPE_CHAR. When having a look at the code for GetFileType
if (is_console_handle( hFile )) return FILE_TYPE_CHAR;
it would seem necessary that a console fd should return FILE_TYPE_CHAR.
http://bugs.winehq.org/show_bug.cgi?id=17417
--- Comment #4 from Austin Lund austin.lund@gmail.com 2009-02-18 00:35:10 --- OK. I was wrong. The handle for the duplicated fd is not a console handle.
However, in the failure case NtQueryVolumeInformationFile in GetFileType does not return STATUS_SUCCESS and in the succeeding case it does.
http://bugs.winehq.org/show_bug.cgi?id=17417
--- Comment #5 from Austin English austinenglish@gmail.com 2009-02-18 11:29:42 --- Looks like AJ committed a possible fix: http://source.winehq.org/git/wine.git/?a=commitdiff;h=1c91d54503f9b2afa513dc...
http://bugs.winehq.org/show_bug.cgi?id=17417
--- Comment #6 from Austin Lund austin.lund@gmail.com 2009-02-18 16:39:04 --- Fixes my test case.
(That was quick).
http://bugs.winehq.org/show_bug.cgi?id=17417
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED
--- Comment #7 from Alexandre Julliard julliard@winehq.org 2009-02-18 16:52:01 --- Fixed.
http://bugs.winehq.org/show_bug.cgi?id=17417
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC|austinenglish@gmail.com |
--- Comment #8 from Austin English austinenglish@gmail.com 2009-02-18 17:20:54 --- (In reply to comment #7)
Fixed.
Yep, passing everywhere today: http://test.winehq.org/data/tests/msvcrt:file.html
http://bugs.winehq.org/show_bug.cgi?id=17417
Alexandre Julliard julliard@winehq.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #9 from Alexandre Julliard julliard@winehq.org 2009-02-27 16:31:46 --- Closing bugs fixed in 1.1.16.
-
wine-bugs＠winehq.org