[Bug 36548] New: 3DMark 2003 cannot start: Errors occured initializing Futuremark SystemInfo.

http://bugs.winehq.org/show_bug.cgi?id=36548

Bug ID: 36548
           Summary: 3DMark 2003 cannot start: Errors occured initializing
                    Futuremark SystemInfo.
           Product: Wine
           Version: 1.7.19
          Hardware: x86
               URL: http://www.futuremark.com/benchmarks/legacy
                OS: Linux
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: -unknown
          Assignee: [e-mail]
          Reporter: [e-mail]

Tested using 3DMark03_v360_1901.exe

You must set msvcp70 to "native" in winecfg before attempting to start
3DMark03.exe.

./3DMark03.exe -nosysteminfo

doesn't help either.

It's a regression, however I cannot say or identify when it happened.

[Bug 36548] 3DMark 2003 cannot start: Errors occurred initializing Futuremark SystemInfo.

http://bugs.winehq.org/show_bug.cgi?id=36548

Artem S. Tashkinov  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|3DMark 2003 cannot start:   |3DMark 2003 cannot start:
                   |Errors occured initializing |Errors occurred
                   |Futuremark SystemInfo.      |initializing Futuremark
                   |                            |SystemInfo.

[Bug 36548] 3DMark 2003 cannot start: Errors occurred initializing Futuremark SystemInfo.

http://bugs.winehq.org/show_bug.cgi?id=36548

Anastasius Focht  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |download
             Status|UNCONFIRMED                 |RESOLVED
                 CC|                            |[e-mail]
         Resolution|---                         |DUPLICATE

--- Comment #1 from Anastasius Focht  ---
Hello folks,

dupe of bug 20296

Please search more carefully next time.

Regards

*** This bug has been marked as a duplicate of bug 20296 ***

[Bug 36548] 3DMark 2003 cannot start: Errors occurred initializing Futuremark SystemInfo.

https://bugs.winehq.org/show_bug.cgi?id=36548

Austin English  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |CLOSED

--- Comment #2 from Austin English  ---
Closing.

[Bug 36548] 3DMark 2003 cannot start: Errors occurred initializing Futuremark SystemInfo.

https://bugs.winehq.org/show_bug.cgi?id=36548

Sergey Isakov  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[e-mail]

--- Comment #3 from Sergey Isakov  ---
Hello folks,
I have strange observation:
wine-1.7.55 
3DMark03 is not started with message "System Info error" as you see.

wine-1.7.54
Works!
Same installation, just change prefix.

Tomorrow I will do regression test.

[Bug 36548] 3DMark 2003 cannot start: Errors occurred initializing Futuremark SystemInfo.

https://bugs.winehq.org/show_bug.cgi?id=36548

--- Comment #4 from Sergey Isakov  ---
Hey all,
I finished regression test. The result
~~~
$ git bisect bad
8cc5b07e9d0aedffd4251d2e8942cb8368920bf8 is the first bad commit
commit 8cc5b07e9d0aedffd4251d2e8942cb8368920bf8
Author: Martin Storsjo 
Date:   Wed Nov 4 12:25:27 2015 +0200

ucrtbase: Implement _get_stream_buffer_pointers.

Signed-off-by: Martin Storsjo 
    Signed-off-by: Piotr Caban 
    Signed-off-by: Alexandre Julliard

:040000 040000 5a54f18ec920eefff9006a25388634a2ec379ef5
29d9969d0d8d56fb44508fd4016fac1bd5f93125 M    dlls
~~~~

I am not sure if Artem S.Tashkinov encounter the same bug but the description
is exactly the same: 3DMark03 cannot start "System Info error" so I think the
bug must be reopened.
It is not duplicate of 20296 (as well related 39403). It also is not resolved
with any Stage patches.

Actually there are 3 states.
1. wine-1.7.54 works.
2. wine-1.7.54-140-7fbf33f  stalls waiting like dlls/ntdll/loader.c:
loader_section" wait timed out in thread 0016, blocked by 0058, retrying (60
sec) infinite times
3. wine-1.7.55 not waiting but show dialog window "System Info error".
Accounting 2. is bad I bisecting to this commit.

Now I will try to revert this commit to latest wine to see what is happen.

Other comments?

[Bug 36548] 3DMark 2003 cannot start: Errors occurred initializing Futuremark SystemInfo.

https://bugs.winehq.org/show_bug.cgi?id=36548

--- Comment #5 from Nikolay Sivov  ---
I find it hard to believe that 3DMark 2003 depends on ucrtbase.

[Bug 36548] 3DMark 2003 cannot start: Errors occurred initializing Futuremark SystemInfo.

https://bugs.winehq.org/show_bug.cgi?id=36548

--- Comment #6 from Sergey Isakov  ---
Tested with native msvcrt100.dll (wine-1.8-rc1-63)
builtin
~~~
fixme:process:GetLogicalProcessorInformationEx (65535,0x0,0x5203e8): stub
fixme:wbemprox:client_security_SetBlanket 0x42967a58, 0x163890, 10, 0, (null),
3, 3, 0x0, 0x00000000
fixme:wbemprox:client_security_Release 0x42967a58
err:ole:CoGetClassObject class {f3bdfad3-f276-49e9-9b17-c474f48f0764} not
registered
err:ole:CoGetClassObject no class object {f3bdfad3-f276-49e9-9b17-c474f48f0764}
could be created for context 0x1
fixme:ntoskrnl:KeGetCurrentIrql  stub!
DbgPrint says: (null)(IRQL 00):     Driver compiled at 17:01:51 on Sep 11 2010
fixme:ntoskrnl:KeGetCurrentIrql  stub!
DbgPrint says: GPUZ(IRQL 00):     Created device: \Device\GPUZ
fixme:ntoskrnl:KeGetCurrentIrql  stub!
DbgPrint says: GPUZ(IRQL 00):     Created link: ??\GPUZ -> \Device\GPUZ
fixme:ntoskrnl:IoRegisterShutdownNotification stub: 0x112748
~~~

native
~~~
fixme:process:GetLogicalProcessorInformationEx (65535,0x0,0x5203e8): stub
wine: Unhandled page fault on read access to 0x00000020 at address 0x457464
(thread 0058), starting debugger...
fixme:wbemprox:client_security_SetBlanket 0x42967a58, 0x161940, 10, 0, (null),
3, 3, 0x0, 0x00000000
fixme:wbemprox:client_security_Release 0x42967a58
~~~

[Bug 36548] 3DMark 2003 cannot start: Errors occurred initializing Futuremark SystemInfo.

https://bugs.winehq.org/show_bug.cgi?id=36548

--- Comment #7 from Sergey Isakov  ---
I may propose that my proposition about 2. is wrong.
I will account it as good and continue regression test.

[Bug 36548] 3DMark 2003 cannot start: Errors occurred initializing Futuremark SystemInfo.

https://bugs.winehq.org/show_bug.cgi?id=36548

--- Comment #8 from Sergey Isakov  ---
Working log with 1.7.54
~~~
fixme:process:GetLogicalProcessorInformationEx (65535,0x9f6048,0x8beb54): stub
fixme:wbemprox:client_security_SetBlanket 0x4250ba10, 0x171340, 10, 0, (null),
3, 3, 0x0, 0x00000000
fixme:wbemprox:client_security_Release 0x4250ba10
fixme:win:LockWindowUpdate (0x0), partial stub!
~~~

[Bug 36548] 3DMark 2003 cannot start: Errors occurred initializing Futuremark SystemInfo.

https://bugs.winehq.org/show_bug.cgi?id=36548

--- Comment #9 from Austin English  ---
(In reply to Sergey Isakov from comment #4)
> Now I will try to revert this commit to latest wine to see what is happen.
>
> Other comments?

This should be in a separate bug report.

[Bug 36548] 3DMark 2003 cannot start: Errors occurred initializing Futuremark SystemInfo.

https://bugs.winehq.org/show_bug.cgi?id=36548

--- Comment #10 from Sergey Isakov  ---
OK.
I see 3DMark06 is also affected by this regression.