topic Re: Segmentation fault when calling eglMakeCurrent in i.MX Processors

Segmentation fault when calling eglMakeCurrent

charlesung — Thu, 05 May 2016 17:50:15 GMT

We are using OpenVG with X11 on the imx6 and we encounter a problem that eglMakeCurrent will occasionally crash with segmentation fault. Not very sure on the cause. The input parameters seem to be valid at the time of the call. Below is the call stack when it dies.

==4346== Process terminating with default action of signal 11 (SIGSEGV)

==4346== Access not within mapped region at address 0x394

==4346== at 0x5E9DF64: gcoSURF_ReferenceSurface (gc_hal_user_surface.c:12505)

==4346== by 0x5F90CC7: _CreateSurfaceObjects (gc_egl_surface.c:604)

==4346== by 0x5F917CF: veglResizeSurface (gc_egl_surface.c:1389)

==4346== by 0x5F8D34B: veglMakeCurrent (gc_egl_context.c:2508)

==4346== by 0x5F8DD43: eglMakeCurrent (gc_egl_context.c:2633)

I have also attached the valgrind output that shows the steps that lead to this.

Original Attachment has been moved to: eglmakecurrent_bad.txt.zip

Re: Segmentation fault when calling eglMakeCurrent

chingling_wang — Thu, 05 May 2016 20:30:41 GMT

Hi, charles,

could you tell me which version of gpu driver you are using? which bsp such as 3.10.17, 3.14.52 or 4.1.15 etc.

which hardware imx6q, 6dl, 6sl or 6sx?

I believe you are using x11 backend.

And it will be helpful if you can share your openvg application that encounters segmentation fault.

Thanks

Re: Segmentation fault when calling eglMakeCurrent

charlesung — Thu, 05 May 2016 20:43:07 GMT

We first notice this on a custom board using imx6q running 3.14.52-1.1.0 with x11. Then we try this on a Sabre board running the latest yocto release (4.1.15-1.1.0) hoping the new release may have this fixed.

On the app side, we have our own custom QT qpa that is similar to xcb but has a VG paint engine.

Re: Segmentation fault when calling eglMakeCurrent

chingling_wang — Thu, 05 May 2016 22:07:10 GMT

The latest yocto release in community is 4.1.15-1.0.0_ga so far, where did you get the release 4.1.15-1.1.0?

I built from here fsl-arm-yocto-bsp.git - Freescale i.MX Yocto Project manifests

So, in your case, it is x11 backend with QT?

Re: Segmentation fault when calling eglMakeCurrent

charlesung — Thu, 05 May 2016 22:25:31 GMT

http://www.nxp.com/products/microcontrollers-and-processors/arm-processors/i.mx-applications-processors/i.mx-6-processors/i.mx6qp/i.mx-6quad-processors-high-performance-3d-graphics-hd-video-arm-cortex-a9-core:i.MX6Q?fpsp=1&tab=Documentation_Tab

You will see the release notes for 4.1.15-1.1.0_ga near the bottom.

We are not using the normal xcb qpa backend provided by QT since that does not use OpenVG. We have our own custom qpa that interfaces with x11 but has a custom paint engine that use VG for rendering.

Re: Segmentation fault when calling eglMakeCurrent

chingling_wang — Fri, 06 May 2016 00:31:37 GMT

what is your gpu driver version? p8.4_ga or p8.4_rc2 or p8.3, from gpu commits, there are some fixes for openvg memory leak in gpu p8.4_ga release.

I cannot reproduce this issue using our openvg sample application although I tried many times. The eglMakeCurrent() api works OK.

From your debug message, it seems the rendering surface access is not valid.

So, I need your app to reproduce it if the newest p8.4 gpu driver cannot fix your issue.

But, I need to sync waht you said your qpa interface with x11. I don't know if I can run your apps in our qt enviroment.

Re: Segmentation fault when calling eglMakeCurrent

charlesung — Fri, 06 May 2016 16:23:51 GMT

The gpu driver being used should be p8.3 based on the yocto content.

Re: Segmentation fault when calling eglMakeCurrent

chingling_wang — Fri, 06 May 2016 21:27:29 GMT

you can get p8.4 in external mirror

http://www.freescale.com/lgfiles/NMG/MAD/YOCTO/imx-gpu-viv-5.0.11.p8.4-hfp.bin

You may try is gpu driver first to see if it can fix your problem

Re: Segmentation fault when calling eglMakeCurrent

charlesung — Mon, 09 May 2016 16:42:09 GMT

Run into an error during unpacking. Any idea?

Do you accept the EULA you just read? (y/N) y
EULA has been accepted. The files will be unpacked at 'imx-gpu-viv-5.0.11.p8.4-hfp'

Unpacking file .......................................................................................................................................................................................................................................................................................................................................................................................................................................ERROR:
.Signal caught, cleaning up

Re: Segmentation fault when calling eglMakeCurrent

chingling_wang — Tue, 10 May 2016 00:03:38 GMT

yes, y.

I can get it without any problem.

I will pack it for you. I replaced it , It seems it passed the eglMakeCurrent(0 without seg fault. But, I got some error of undefined symblol in lib you sent.

Re: Segmentation fault when calling eglMakeCurrent

chingling_wang — Tue, 10 May 2016 00:07:48 GMT

I transweb the file to you. It is too large to be attached

Re: Segmentation fault when calling eglMakeCurrent

charlesung — Wed, 11 May 2016 17:19:26 GMT

I copied most of the .so files in gpu-core folder that are relevant to my target and the problem still exists. The valgrind looks almost identical with slight difference in some line numbers but that is about it. I have attached the valgrind output that shows the steps that lead to the seg fault after it enters the eglMakeCurrent function. The output is captured with the new gpu driver in place.

Re: Segmentation fault when calling eglMakeCurrent

charlesung — Wed, 11 May 2016 17:21:58 GMT

Do you have any code that you use to verify OpenVG on X11 that you can share?

Re: Segmentation fault when calling eglMakeCurrent

chingling_wang — Wed, 01 Jun 2016 00:13:43 GMT

If gpu 8.4 cannot fix your issue, I have no clue so far what causes the seg fault. Yes, We do have openvg samples we can share. in /opt/fsl-gpu-sdk/OpenVG, it has all the openvg sample applications. And you can download demo framework to built it. demo framework is release in every version of yocto release.

You can also download from git stash sw-stash.freescale.net/scm/gtec/demo-framework.git, it also released as stand alone package in nxp website.

Demo framework sample are all C++, Do you need sample application in c for openVG? I can package for you. They are also in old release like 3.10.17 bsp.

Re: Segmentation fault when calling eglMakeCurrent

sebastient — Wed, 01 Jun 2016 22:22:21 GMT

Have you looked at the numerous reported issues "Conditional jump or move depends on uninitialised value" from Charles' report? These include source file and line numbers for you to confirm. These are scattered across gc_hal_user_os.c gc_hal_user_hardware_vg.c gc_egl_surface.c etc.

These should be the starting point to investigating the reported issue.

Re: Segmentation fault when calling eglMakeCurrent

karina_valencia — Wed, 08 Jun 2016 21:43:04 GMT

chinglingwang please continue with the follow up.

Re: Segmentation fault when calling eglMakeCurrent

chingling_wang — Thu, 09 Jun 2016 15:25:39 GMT

I put email watch for this thread, But, it seems not coming to my mailbox. Sorry about this.

I looked into the output, it seems openvg cannot allocate the egl surface memory specified by your app.

From your log,

Native win geometry after resize 0,48,1280,672

Number of queued expose event 0

Expose region 0,0,1280,672

1. if region is 0,48,1280,672, the surface actually size is 1280x(672-48), I think 0, 48 means the staring point, why expors region 0,0, 1280, 672, the surface size is 1280x672? it is becoming bigger,

2. 672 is not 16 bits aligned, could you pass your surface parameter as 16 bits aligned? is your display 1280x720?

3. I also attached our sample openvg applications that you can take it as reference for openvg egl setup.

Re: Segmentation fault when calling eglMakeCurrent

sebastient — Thu, 09 Jun 2016 16:09:13 GMT

Yes, it is a rect of x,y,w,h. Otherwise it is not clear what the question is? This is an X11 surface, a window could be anywhere on the screen.
1280 is 16-byte aligned, any value for height will therefore also be 16-byte aligned.
The issue is not on setup, it's really while handling window resize that the issue is seen.
Have you looked at the Valgrind output which points to conditional branches on uninitialized variables within the driver?

Re: Segmentation fault when calling eglMakeCurrent

charlesung — Thu, 09 Jun 2016 16:58:38 GMT

None of the examples demonstrates the case that the X11 window being resize dafter surface creation, which is where we think the problem is. Our guess is that there is something in the driver that handles that but we really do not know without the source code.

Re: Segmentation fault when calling eglMakeCurrent

chingling_wang — Thu, 09 Jun 2016 17:14:59 GMT

When resize, you need to destroy the old surface you created before, then eglCreateSurface again. I didn't see it in your log.

The reason you got valgind output pointing to conditional branches is that in the drvier, gpu check the buffer size, it thinks that what you want is bigger than what it is for max buffer size it has, so it returns out of mememory error, and it exits and do some cleaning

And yes, the width and height both better to be 16 bits aligned when creating egl surface.

usually, 8 bits is ok, but, I talked to driver engineer, it is better to be 16 bits aligned

I try to see if I can get some resize sample for openvg.