DirectX 10 Shader Pass Descriptor corruption - visual-c++

Good afternoon everyone.
I've got a strange issue that I just can't seem to figure out.
In the following code segment
m_technique->GetPassByIndex(0)->GetDesc(&passDesc);
result = device->CreateInputLayout(polygonLayout,numElements,passDesc.pIAInputSignature,passDesc.IAInputSignatureSize,&m_layout);
the Direct X Debug output gives the following:
First-chance exception at 0x753D37C3 (kernel32.dll) in GameTest.exe: 0xC0000005: Access violation reading location 0xCCCCCCCC.
First-chance exception at 0x75B44B32 (KernelBase.dll) in GameTest.exe: 0x0000087A (parameters: 0x00000000, 0x0083E214, 0x0083D64C).
D3D11 CORRUPTION: ID3D10Device::CreateInputLayout: Third parameter is corrupt. [ MISCELLANEOUS CORRUPTION #15: CORRUPTED_PARAMETER3]
A few things that I am unsure about. In the last line I see D3D11 CORRUPTION, but I only use D3D10 headers and libraries.
I am working in Visual Studio 2012, and I have the Windows 8 SDK installed, so that may cause problems.
I'm not sure how to proceed and any advice would be greatly appreciated.

Related

How to avoid fatal error or unknown error in powershell on windows server 2016

Sometimes i get unknow error in powershell on windows server 2016, every time this error have random behavior, but every time is "Fatal error", like on screenshot.
Screen_Shot
Unknown type: 541
#
# Fatal error in, line 0
# unreachable code
#
#
#
#FailureMessage Object: 0000007BD8BFF1F0
What is that and how to fix this? I launch node.js on this server
Memory leak? Disc error?
P.S. on Linux server i dont have any errors, but i need to launch my code on Windows server 2016. Sorry if my question was stupid or is a duplicate.
Does in your code you've setup something like :
$ErrorActionPreference = Stop
This will stop your process at any errors; anyway the best practice is to use the try/catch block to handle any error and , at least, display where and what is the error message.
In your case, I suspect there is a bad memory access or conflict with something but without more details it's pretty difficult.
Do you use in your script the try and catch block ?
I found that problem spot - CPU.

(XCode 7 + iOS 9 + iPhone 4s/iPhone5 only) issue: "malloc: *** mach_vm_map(size=1048576) failed (error code=3)"

I know the issue is related to memory allocations, but I get it in only iOS 9, XCode 7. In XCode 6.4, iOS 8.4, it works just perfect, no issue at all at any ways. In iOS 9, XCode 7, it just crashes very frequently giving this error.
malloc: *** mach_vm_map(size=1048576) failed (error code=3)
*** error: can't allocate region securely
*** set a breakpoint in malloc_error_break to debug
Any suggestions? I am working on memory issues, but I wonder if there were too many memory issues, then why did it work in iOS 8.4 and not in iOS 9?
Also, I get all my UI whited out! Like navigation bar has no title, custom tabbar(RDVTabBar) is not visible, however, other view controller is there (it responds to the touch events, you can tap that area and the button positioned in that area acts!).
P.S. I don't get memory warning ever, neither applicationWillTerminate: method is being called!
Update: I found that this issue occurs in iPhone 5 and 4s only! (Non-64bit devices!)
Update 2: When the crash happens, I try to print one of the object and to print it, I again get the same error in debug mode:
malloc: *** mach_vm_map(size=1048576) failed (error code=3)
*** error: can't allocate region securely
*** set a breakpoint in malloc_error_break to debug.
------------------ Update ------------------
Well, I found something here: I inspected my app in Instruments and detected that it occupies about 200mb of memory in iOS 8.4, and surprisingly iOS 9.* occupies 1.5 GB of the memory!!! This is something I don't understand! One app takes 200 mb of memory in iOS 8.4 and the same app takes over 1.5GB of memory on iOS 9! Not understandable at all! Any Idea?
Well, I guess I have finally found the issue: Its definitely memory issue, but I had to search where. I found that I am using two third party labels namely: CXAHyperlinkLabel and STTweetLabel. When I removed those, my app just work fine!! The issue has been resolved but still I am confused why it did work (and still works) great in iOS 8.4 and eats up more than 1.5 GB of memory in iOS 9.0 and above!! If it has some issues with memory (I found some and fixed already, still), why it did work with iOS 8.*.
So, my advice to any fellow who are having such issues, I recommend using UITextView for links (I did it and its nice replacement).
I've just experienced the exact issue as you (xcode7, on iPad 2, iOS 8.4).
<Error>: myApp(524,0x3cfda9dc) malloc: *** mach_vm_map(size=4060311552) failed (error code=3)
*** error: can't allocate region
*** set a breakpoint in malloc_error_break to debug
So, I decide to ran on debug mode, and the app crash in the most unexpected place:
[self.delegate discountPopupDoneClickWithDiscount:self.discount
type:self.discountType
federalTaxEnable:self.federalTaxEnable
federalTax:self.federalTax
customTax:self.customTax];
Then I tried to find out what happened here, type po self.discount, turn out self.discount (it's a property with NSDecimalNumber type) return a very unsual value, like <å: 0xba123adr>. Strange, right? Also, when I try po [self discount], it yield NSDecimalNumber [_NSKeyedCoderOldStyleArray initialize]: Unrecognized selector send to instance. Pretty much clueless what to do here.
Solution:
I update the discount property from this:
#property (nonatomic, assign) NSDecimalNumber *discount;
to this:
#property (nonatomic, strong) NSDecimalNumber *discount;
And then everything's fine. Of course this isn't an exact answer/solution to your question, but this may help/give you a clue on what to do, somehow.
So strange. I had a very similar issue, where the app would continually increase memory consumption until eventually it would crash, but ONLY on iOS7 and iOS8. Sure enough, no issues with iOS9 and up.
Found out that a simple setting of [UILabel setText:text] for A LOT of collectionViewCells was the cause. WTF? They were weak, nonatomic. Why?
Ran the app in Instruments : Allocations. Before Instruments eventually crashed, I was able to isolate the memory consumption to __NSStringDrawingEngine, +[NSParagraphStyle _defaultWritingDirection] [UIFont fontWithName] or something to that nature.
Anyway, turns out the issue is that I was using the Text Styles types (Eg, Title 1, Title 2, Body, Caption, etc). Seemed logical to use at the time. But, I switched those over to System fonts, and no more memory issue and no more freezing.
So strange.
did you use the NSZombieObject enable,it will not release the zombie object,and may cause this crash.i hope it will help you

Windows Store TriageDump.dmp without debug information

Greetings people,
My Windows Store game has been released for more than three weeks now, and I started getting crash reports. I could download the TriageDump.dmp file and have it opened in Visual Studio 2012, but it did not help much, I am constantly getting "No Debugging Information" error message when I click on "Debug with Native Only":
Also, the tool-tip on my Dashboard shows no information of the crashing function (could "Unknown" here mean inlined function or lambda expressions in concurrent::task?):
I would like to believe that I have done everything the way it should be done, of course I may be wrong. Here are some additional information that might be helpful in finding the issue:
It uses DirectX and written purely in C++ (without C# or XAML)
Project setting: C++\General\Debug Information Format = Program Database (/Zi)
Project setting: Linker\Debugging\Generate Debug Info = Yes (/DEBUG)
The game is made up of two native modules: Labyrinth.App.exe and Labyrinth.Core.dll
The generated APPXUPLOAD contains both APPX and APPXSYM files
The APPXSYM file contains both Labyrinth.App.pdb and Labyrinth.Core.pdb
I'm on x64 development machine, and the triagedump.dmp is for x86
I did click on "Include public symbol files, if any, to enable crash analysis for the app" when generating APPXUPLOAD file:
Please let me know if you spotted the issue or suspected something that's wrong above. Thanks in advance for your help, guys! :)
The very same problem here. MS had that working in the past though.
Actually if you still have the .appxsym around you can easily extract the .pdb out of that. The appxsym is just a .zip file it seems.
You can load these pdbs then as symbols after loading the triagedump.dmp.

SQL CE 4.0 Error - ReleaseHandleFailed was detected

I keep getting this error in the dev environment - 'A SafeHandle or CriticalHandle of type 'System.Data.SqlServerCe.HasSafeHandle' failed to properly release the handle with value ...."
It is in a VS 2012 Winforms application and only seems to happen if a particular piece of code is run at some point. The error shows after the form is closed, but I can't understand why. It only happens in the dev environment and seems to be OK in the compiled version, but I don't want to just ignore it.
The error seems to be caused by this line of code, although weirdly at this point in time I have not opened a connection to the SQL ce database and the directory it is searching in does not have the SQL ce database in it. It is mearly reading in txt files that match a pattern
string[] fileNames = Directory.GetFiles(importDirectory, "BucketHistory*.txt")
I'll do some more digging tomorrow, it's late now, and I have no idea what is causing it.
Any pointers ??!
Turns out I was having a bit of brain fade yesterday. Been working too long hours !
For anyone interested, the problem was that I was opening a connection, then copying the file and running a shrink whilst the connection was still open. It had nothing to do with that line of code.
I would have expected the copy to have thrown an exception, but it didn't appear to. Oh, and opening another connection inside the using statement of the first one. Sigh.

WinRT - Windows Store - WinRT Originate Error - How do decipher such an error?

I'm working on a Windows Store app and I'm getting a WinRT error that doesn't really give me any information so I would like to know how to understand these sorts of errors.
Basically I get the error on the following line which is called inside OnPointerPressed:
m_gestureRecognizer->ProcessDownEvent(args->GetCurrentPoint(nullptr));
The error is:
First-chance exception at 0x76F54B32 (KernelBase.dll) in DXAML2.exe: 0x40080201: WinRT originate error (parameters: 0x80070057, 0x00000044, 0x03CEE72C).
This error didn't used to appear, the only thing I've changed is that this line is now wrapped in an if clause which tests if the current pointer's PointerId is the same as one I've stored just using == such as:
if(args->GetCurrentPoint(nullptr)->PointerId == m_UIPointerID)
I have no idea why this has started happening.
So my question is in two parts:
More generally, how do I understand what an error such as the above means?
And does anyone know this error has suddenly started happening now that I check the pointerId?
Thanks for your time.
P.S. I guess another thing that has changed is that there will already be 2 pointers on the screen (the one that gets pushed into this GestureRecognizer) as well as another one, hence the PointerId check.
"How to Decipher such an error"...
For any WinRT originate error, you can take the third address in the parameters list (in your example, 0x03CEE72C), and find a description of your error in the memory window.
While debugging, break when your error is thrown and open the memory window via Debug -> Windows -> Memory -> Memory 1
Copy and paste the address to get your "easy-to-find" error message.
As Raman said - it's good to look up the hex values shown. The first one is the memory location which won't tell you much without the symbols/source, which in this case is reported directly by Windows. Perhaps the public symbols can shed some more light on where the error came from, but the error code lookups are more helpful.
If you Bing for 0x80070057 you will find an MSDN article on Common HRESULT Values which lists
E_INVALIDARG : One or more arguments are not valid : 0x80070057
It doesn't give you all the details of course, so you're off to theorize. Perhaps you can only call args->GetCurrentPoint(nullptr) once and you should store/reuse the value? Maybe gesture recognizer is not configured correctly? Maybe the app window is not visible at the time the exception is thrown or the thread is wrong. Maybe some expected calls to gesture recognizer were missed due to filtering those out with these "if" statements.

Resources