http://bugs.winehq.org/show_bug.cgi?id=9922
Summary: 'make test' hangs with "*** glibc detected *** urlmon_test.exe.so: double free " Product: Wine Version: 0.9.46. Platform: Other OS/Version: other Status: NEW Severity: normal Priority: P2 Component: wine-urlmon AssignedTo: wine-bugs@winehq.org ReportedBy: dank@kegel.com
Alexandre and I have both seen this. He says it's a race.
Here's the log:
../../../tools/runtest -q -P wine -M urlmon.dll -T ../../.. -p urlmon_test.exe.so protocol.c && touch protocol.ok protocol.c:370: Test failed: unexpected call ReportProgress_FINDINGRESOURCE protocol.c:375: Test failed: szStatustext != "www.winehq.org" protocol.c:383: Test failed: unexpected call ReportProgress_CONNECTING protocol.c:386: Test failed: Unexpected szStatusText *** glibc detected *** urlmon_test.exe.so: double free or corruption (fasttop): 0x7c02b4c0 *** ======= Backtrace: ========= /lib/tls/i686/cmov/libc.so.6[0x601c0d65] /lib/tls/i686/cmov/libc.so.6(cfree+0x90)[0x601c4800] /usr/lib/i686/cmov/libcrypto.so.0.9.8(CRYPTO_free+0x3a)[0x644d44aa] /usr/lib/i686/cmov/libcrypto.so.0.9.8(X509_VERIFY_PARAM_free+0x6f)[0x6457996f] /usr/lib/i686/cmov/libssl.so.0.9.8(SSL_free+0x64)[0x644831d4] /home/dank/wine-git/loader/../dlls/wininet/wininet.dll.so(NETCON_close+0xa1)[0x60a42a41] /home/dank/wine-git/loader/../dlls/wininet/wininet.dll.so[0x60a3202a] /home/dank/wine-git/loader/../dlls/wininet/wininet.dll.so(WININET_Release+0x4d)[0x60a3a38d] /home/dank/wine-git/loader/../dlls/wininet/wininet.dll.so(InternetQueryDataAvailable+0x69)[0x60a3d4a9] /home/dank/wine-git/loader/../dlls/urlmon/urlmon.dll.so[0x6063e1fe] /home/dank/.wine/dosdevices/z:/home/dank/wine-git/dlls/urlmon/tests/urlmon_test.exe.so[0x6060b589] /home/dank/.wine/dosdevices/z:/home/dank/wine-git/dlls/urlmon/tests/urlmon_test.exe.so(func_protocol+0x407)[0x6060bea7] /home/dank/.wine/dosdevices/z:/home/dank/wine-git/dlls/urlmon/tests/urlmon_test.exe.so[0x606136f8] /home/dank/.wine/dosdevices/z:/home/dank/wine-git/dlls/urlmon/tests/urlmon_test.exe.so(main+0x14d)[0x60613d8d] /home/dank/.wine/dosdevices/z:/home/dank/wine-git/dlls/urlmon/tests/urlmon_test.exe.so[0x60613e5b] /home/dank/wine-git/loader/../dlls/kernel32/kernel32.dll.so[0x60416cbe] /home/dank/wine-git/loader/../libs/wine/libwine.so.1[0x600259d7]
http://bugs.winehq.org/show_bug.cgi?id=9922
Misha Koshelev mk144210@bcm.tmc.edu changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |mk144210@bcm.tmc.edu
--- Comment #1 from Misha Koshelev mk144210@bcm.tmc.edu 2007-10-16 19:56:07 --- Hi Dan,
Thanks for reporting, I've been doing some looking at this on and off for a bit and I:
(i) can't reproduce it (it looks from the failed tests like you are getting some redirects somehow from the sites but I never do) and (ii) the race must ultimately be in NETCON_close (easy to see race conditions if it gets called twice) but the callers for NETCON_close themselves (only 3 in wininet/http.c) I can't mentally trace a race in. However, I did fix that (unrelated?) urlmon race I tracked down mentally, and was wondering if you are still seeing this failure?
So I guess my questions are:
(a) Are you still seeing this failure? (b) Any ideas why you seem to be getting redirects from these sites?
Thanks Misha
http://bugs.winehq.org/show_bug.cgi?id=9922
Dan Kegel dank@kegel.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED
--- Comment #2 from Dan Kegel dank@kegel.com 2007-10-22 08:18:34 --- I think it's ok, I haven't seen this in my notes since you checked in your fix. I'll reopen if I see it again.
http://bugs.winehq.org/show_bug.cgi?id=9922
--- Comment #3 from Misha Koshelev mk144210@bcm.tmc.edu 2007-10-22 09:09:50 --- Great. I'll definitely look into it again if it comes up again.
Misha
http://bugs.winehq.org/show_bug.cgi?id=9922
Dan Kegel dank@kegel.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #4 from Dan Kegel dank@kegel.com 2008-01-28 05:41:31 --- Closing all RESOLVED FIXED bugs older than four weeks.