topic @mark-dewing (Intel) in Analyzers

Internal error, application exitcode -1

Marián__VooDooMan__M — Thu, 06 Nov 2014 17:10:06 GMT

Greetings,

First of all, sorry for double-post, but it seems my post got lost after submitting.

I have application compiled with ICC 14.0 Update 4 (since ICC 15 has bug/internal error).

Both inspector XE 2013 and XE 2015 displays similar results in bot MSVC 2013 IDE and ran as a standalone application.

I have Haswell, Win 8.1 Pro x64, target app x64, 16 GiB RAM.

Here is copy-paste of log saying actually nothing except of "internal error". NB: sensitive informations, particulary full paths were obfuscated by upper-case "XXX" without quotes:

Analysis started...
Result file: C:\Users\XXX\r002mi3\r002mi3.inspxe 
Analysis started for C:\Users\XXX.exe (pid = 7296) 
Loaded module: C:\Program Files (x86)\Intel\Inspector XE 2013\bin64\ccrt12.dll, address range [0x00000000C6000000-0x00000000C61ADFFF], minimal analysis 
[...SKIPPED...]
Loaded module: C:\Windows\WinSxS\amd64_microsoft.windows.common-controls_6595b64144ccf1df_6.0.9600.17031_none_6242a4b3ecbb55a1\comctl32.DLL, address range [0x00007FF8F72A0000-0x00007FF8F74F9FFF] 
Error: An internal error has occurred. Our apologies for this inconvenience. Please gather a description of the steps leading up to the problem and contact the Intel customer support team.
Application exit code: -1 
Result file: C:\Users\XXX\r002mi3\r002mi3.inspxe 
Analysis completed

Result file: C:\Users\XXX\r002mi3\r002mi3.inspxe is:

<?xml version="1.0" encoding="UTF-8"?>
<root>
 <rdmgr>
  <timestamp type="u64_t">1415292675</timestamp>
  <hostname>vdmn</hostname>
  <os>windows</os>
  <product>Intel® Inspector XE 2013 Update 9</product>
  <buildNumber type="s32_t">328075</buildNumber>
  <logicalCPUCount type="s32_t">8</logicalCPUCount>
  <CPUFrequency type="s64_t">2700000000</CPUFrequency>
  <collectionTimeBegin type="u64_t">1415292675</collectionTimeBegin>
  <collectionTimeEnd type="u64_t">1415292726</collectionTimeEnd>
  <variables>
   <at>mi3</at>
  </variables>
 </rdmgr>
</root>

I can send via private message due to sensitive information the 7-zipped contents of dir "r002mi3"

Please send me a private

Mark_D_Intel — Thu, 06 Nov 2014 17:16:41 GMT

Please send me a private message with the result directory.

Mark

This occurs just after

Marián__VooDooMan__M — Thu, 06 Nov 2014 17:17:12 GMT

This occurs just after displaying configuration dialog for the user, but without Inspector my application works just fine.

I have all latest updates to date for Inspector XE 2013 and XE 2105.

@mark-dewing (Intel)

Marián__VooDooMan__M — Thu, 06 Nov 2014 17:32:55 GMT

@mark-dewing (Intel)

have you received attachement to my private message? Because I have bad experiences with attachment upload in private messages at Intel web. I can upload it to some web site as a workaround.

Hello,

Dario_Sassi — Wed, 19 Nov 2014 13:08:44 GMT

Hello,

I have the same problem with Visual Studio 2010, Parallel studio XE 2015. The application is a standard Win32 C++ application (with MFC :( ).
Both visual studio and standalone application give me

"An internal error has occurred. Our apologies for this inconvenience. Please gather a description of the steps leading up to the problem and contact the Intel customer support team.
Application exit code: -1"

Thanks for any help!

The original issue is from

Mark_D_Intel — Wed, 19 Nov 2014 15:35:10 GMT

The original issue is from incomplete handling of the 'vgather' instruction in IXE. You can check for this by looking in the result directory for a pin.log file (<result dir>/data.0/pin.log). See if that file exists, and if it mentions the vgather instruction. If this is the problem, one workaround is to compile the code without optimization, or use flags to lower the target ISA (/arch:SSE4.2) to avoid generating the problem instructions.

If this is not the problem, zip up the result directory and send it to me in a private message.

I don't know if it's a

Dario_Sassi — Wed, 19 Nov 2014 15:53:39 GMT

I don't know if it's a private data, but my pin.log as wrote:

Pin 2.13 kit 66213
A: Source\pin\vm_ia32\jit_region_ia32.cpp: LEVEL_VM::REGION::XlateControlFlow: 2449: assertion failed: INS_HasFallThrough(ins)

If you need I can send you the zip file of the Inspector Result

Let me just comment that this

Oscar_B_ — Thu, 29 Jan 2015 14:49:43 GMT

Let me just comment that this error is affecting me as well. I do indeed have the following line in the pin.log file:

Pin 2.13 kit 66213
E: vcs\memorychecker2\core\instrument.cpp:3921: IARG_MEMORYREAD_SIZE is NOT valid for a vgather* instruction

However, it does not go away by compiling in DEBUG mode (O0) or with removing the target architecture specification.

Please advise, since as it stands I cannot use the Inspector software, which I am currently trying out as part of a 30-day trial.

Oscar