#648 AV on start smplayer-portable-14.3.0.6359

[VictorVG]

AV in portable version smplayer-portable-14.3.0.6359 on starting. OS Win7 SP1, CPU Intel Pentium DC E5800, RAM 4 GB, video GTX 550Ti.

Debug output:

registers:

EDI 00000000
ESP 0022F300
EBP 0022F338
EBX FFFFFFFF

Stack log:

QtCore4.dll!6e1de997()
msvcrt.dll!7635a442()
msvcrt.dll!76359d45()
QtGui4.dll!677b2416()
QtCore4.dll!6e1ee477()
qjpeg4.dll!63a2ba92()
QtCore4.dll!6e1bc00b()
msvcrt.dll!76359d45()
QtGui4.dll!67daf0e7()
QtGui4.dll!677a12da()
QtGui4.dll!677c8552()
QtGui4.dll!677c21c7()
QtGui4.dll!677c2d75()
smplayer.exe!00457742()
smplayer.exe!00428ea5()
smplayer.exe!00429614()
smplayer.exe!0050737a()
smplayer.exe!0052b1fd()
smplayer.exe!00531ca2()
msvcrt.dll!763598cd()
msvcrt.dll!763598da()
msvcrt.dll!763598da()
msvcrt.dll!763598da()
msvcrt.dll!763598da()
msvcrt.dll!763598da()
smplayer.exe!00596ba8()
msvcrt.dll!7635a442()
smplayer.exe!0040124b()
smplayer.exe!004012b8()
kernel32.dll!762bee1c()
ntdll.dll!77d537eb()
ntdll.dll!77d537be()

Debug output (sreenshot, created from a friend at work, and where tested)

[VictorVG]

Probably the phenomenon is local in nature and can somehow be associated with the failure of the wireless mouse A4TECH Padless 7100N - there have already had to remove the driver and reinstall it again. On another machine failure is not reproduced. We shall see, I wonder what causes it.

Added:

Failure can not be played on the same machine if unpack NSIS package overwriting files in the same directory where previously was recorded smplayer-portable-14.3.0.6359. It's interesting.

[VictorVG]

For the experiment, I tried to find a specific binary causing failure. By logs debugger found that a failure occurs when you use the original version qjpeg4.dll. If it is replaced by qjpeg4.dll taken from the installation options, then AV does not occur, the player works success, including MP4 containers containing the video in H.264.

[VictorVG]

Additional information - OS is x86 - advised colleagues set because I put the 64-bit OS on 4GB does not make sense because it simply does not have enough RAM to run and it will reset the page to swap that will lead to a slowdown of the computer because of the difference in the speed of the RAM and the external memory subsystem. Now that system OS 3.5 GB of RAM available, the rest of the address space overlapped IO address equipment, and I do not see any reason for the office system to put more RAM - still there are considered problems for which memory is 4 GB is enough, and to transfer them to the AMD64 environment winning time of their decision will be hundredths of a percent, and therefore devoid of meaning because anyway when the decision by the operator will be a lot larger than the data processing time on this computer. So when I was asked a question about the appropriateness install x64 systems instead of write-on-demand financial management DEC PDP 11/70 - "too high cost of ownership, and consumes a lot of energy," I'm an engineer gave a recommendation to install x86 based systems because of their for a given set of tasks rather .

[VictorVG]

Big thanks! Is build smplayer-portable-14.3.0.6362 work ok!, AV not exist. Testing on three systems using CPU Celeron E3300 (m/b ECS G31T-M7 + GF GT430 + 2 Gb DDR2-800), Pentium DC E5400 (m/b MSI G41TM-P33, GF GT220, 2 Gb DDR2-800), Pentium DC E5800 (ASUS P5QL/EPU, GF GTX 550 Ti, 4 Gb DDR2-800).