SGX drivers on Kernel 3.8.13

[mail.ar...@gmail.com]

Hello,

I am currently trying to get OpenGL ES running on the Bone, but I get some problems with the kernel modules. I already posted on the TI formus. Maybe someone can help:

http://e2e.ti.com/support/arm/sitara_arm/f/791/t/298596.aspx

Regards

Strahlex

[Robert Nelson]

See a few of the other threads in this forum for the full details on
this.. Essentially we do not have the needed bits from TI at this
time..

Regards,

--
Robert Nelson
http://www.rcn-ee.com/

[mail.ar...@gmail.com]

Hi,

What exactly is missing? I have read something about that the clock for GPU is not enabled. Can't we fix this or can't we get TI to fix this?

Regards

Alexander

[Gerald Coley]

As I understand it, the DRM driver is not compatible with the old SGX frame buffer architecture. TI tells us they are not rewriting the code for 3.8. Not sure about future kernel releases. Best bet would be to ask TI on this one.

Gerald

--
For more options, visit http://beagleboard.org/discuss
---
You received this message because you are subscribed to the Google Groups "BeagleBoard" group.
To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Micka]

You don't have a most recent Kernel that is working correctly with the SGX drivers ?

3.2 it's very old ....

On Fri, Oct 25, 2013 at 8:52 AM, Giuseppe Iellamo <pep...@gmail.com> wrote:

It seems to me not very fair to say we have a board with a 3d accelerator, we sell it and we won't give you the driver nor the specs to use it...

To my understanding with 3.2 kernel it was working. Any advice on how to resume a working 3.2 distribution I may use?

[Giuseppe Iellamo]

It seems that such a kernel doesn't exist and reading last Gerald post (as always thank you for the support) it could never exist again...

In my project I'll use two beagle : one for embedded layer (in which I'm using 3.8.13 - xenomai patched) one for the gui, in which I don't need more than HDMI e OpenGl support...

[Gerald Coley]

Life is not always fair, especially when SW is involved. The new DRM architecture introduced by the Linux folks in the 3.8 Kernel broke it. I agree, that is not fair.

Gerald

On Fri, Oct 25, 2013 at 1:52 AM, Giuseppe Iellamo <pep...@gmail.com> wrote:

It seems to me not very fair to say we have a board with a 3d accelerator, we sell it and we won't give you the driver nor the specs to use it...

To my understanding with 3.2 kernel it was working. Any advice on how to resume a working 3.2 distribution I may use?

Il giorno giovedì 24 ottobre 2013 22:21:12 UTC+2, Gerald ha scritto:

[Giuseppe Iellamo]

Gerald I didn't want to upset you. Your support is certainly first class.

But this is the way linux works and works good. Things keep changing, 

I think that one couldn't blame linux folks for their work... and really it seems a little strange to me that for a company so big like TI is difficult to implement the needed update...

[Gerald Coley]

Oh, I know.I am not blaming anyone, Just sting fact. Yes TI is a big company as a whole. But when you look at the number of people that work on specific things, well let's just say not all of TI employees are working on Linux. And those that are working on Linux are not working on 3.8 but 3.12 and later,

Gerald

[Micka]

So, maybe ... we should to move to 3.12 ? 

Robert Nelson, I saw something in the mailing list about that :) =>

So here is the deal; 3.8 is old, and 3.12-rc4 is newer/better/etc...

Functionally, it seems ready in my small 5 board sample size.  3 of
which have been running 24/7 @ 100% since rc3..

[Robert Nelson]

On Fri, Oct 25, 2013 at 8:51 AM, Micka <micka...@gmail.com> wrote:
> So, maybe ... we should to move to 3.12 ?
>
> Robert Nelson, I saw something in the mailing list about that :) =>
>
> So here is the deal; 3.8 is old, and 3.12-rc4 is newer/better/etc...
>
> Functionally, it seems ready in my small 5 board sample size. 3 of
> which have been running 24/7 @ 100% since rc3..

I've already personally moved to v3.12.x on everything..

But we need 'someone' to find the time too port all the "capes" *.dtbs
bits from the v3.8.x branch to the new v3.12.x branch. (the capemanger
is in place with v3.12.x)

I'm on the road next week with ARM techconn/etc going on, so i'm going
to be very hands off the next week or so. ;)

[Micka]

Well, I will try to change the dtbs file for the cap LCD7 REVA3 & the cap RS485 :p .

Thx for this tutorial : http://www.eewiki.net/display/linuxonarm/BeagleBone+Black#BeagleBoneBlack-LinuxKernel

Micka,

--
For more options, visit http://beagleboard.org/discuss
---
You received this message because you are subscribed to the Google Groups "BeagleBoard" group.

[Michael Mullin]

Is the 3.2 kernel SGX driver open source and simply too complex for the voluntary community to port to 3.8+ in their spare time (In which case we could start a kickstarter or something to fund someone to work full time on this bug)?  Or is the driver proprietary, and the community is simply at the mercy of TI?

[Gerald Coley]

Driver is licensed. Circuitco has access to the driver under NDA, but they don't have the resources any more. You might be able to ping them and see if something can be worked out. TI will have support for it, eventually. Like 2014 sometime. I am trying to accelerate that , but no promises.

Gerald

[Robert Nelson]

On Fri, Oct 25, 2013 at 1:39 PM, Michael Mullin <masm...@gmail.com> wrote:

> Is the 3.2 kernel SGX driver open source and simply too complex for the
> voluntary community to port to 3.8+ in their spare time (In which case we
> could start a kickstarter or something to fund someone to work full time on
> this bug)? Or is the driver proprietary, and the community is simply at the
> mercy of TI?

Only the kernel "shim" is open source, well it also uses the in-kernel
framebuffer driver to draw directly into..

To give you an idea of what we are dealing with, take a look at this:
https://github.com/RobertCNelson/ti-sdk-pvr

I try to copy the gpl bits from every sdk release, that a bunch of us
can more easily patch the bits to later kernel releases's.

In the case of the bone.. The 3.2 bits use the old frame buffer
driver, and with 3.8/3.12 we are using the brand new and shiny kms
driver..

[Michael Mullin]

I would like to help.  I can contribute both a little bit of time, and also a little bit of money.  I am a software developer and very comfortable working with C, but my knowledge of kernel development is extremely limited.

Does this repo contain a compilable binary that needs to have bugs fixed and features completed (I'm away from my development machine until Sunday)? 

[Robert Nelson]

On Fri, Oct 25, 2013 at 2:17 PM, Michael Mullin <masm...@gmail.com> wrote:
> I would like to help. I can contribute both a little bit of time, and also
> a little bit of money. I am a software developer and very comfortable
> working with C, but my knowledge of kernel development is extremely limited.
>
> Does this repo contain a compilable binary that needs to have bugs fixed and
> features completed (I'm away from my development machine until Sunday)?

Nope, it's just the "shim" between the kernel and sgx closed bins..

[Robert Nelson]

The real trick would be for someone to reverse engineer the "sgx bits"
and just get us out of this mess all together.. ;)

[Micka]

But if we move to the kernel 3.12, the sgx will work?

[Gerald Coley]

Most likely from what I hear, 3.14. But, maybe we can accelerate that. Maybe.

Gerald

[Micka]

Robert Nelson, will it be hard to move to the kernel 3.13? You already moved to 3.12...

[Robert Nelson]

On Fri, Oct 25, 2013 at 2:35 PM, Micka <micka...@gmail.com> wrote:

> But if we move to the kernel 3.12, the sgx will work?

The 'sgx' ip always works... We just can't control it. ;)

ignoring all 'sgx' stuff for now on, unless a patch is attached...

[Robert Nelson]

On Fri, Oct 25, 2013 at 2:40 PM, Micka <micka...@gmail.com> wrote:
> Robert Nelson, will it be hard to move to the kernel 3.13? You already moved
> to 3.12...

Look at the first 14 patches of:
https://github.com/beagleboard/kernel/tree/3.12

hint in: (omap-next-dt) directory...

other then hdmi, I've already got it setup to be rebased for 3.13-rc1. ;)

[Giuseppe Iellamo]

I'm Just realizing that maybe I misunderstood. Is IT unlikely that a sgx driver will be release for 3.8.x kernel OR for any other subsequent release?

[Robert Nelson]

On Fri, Oct 25, 2013 at 4:30 PM, Giuseppe Iellamo <pep...@gmail.com> wrote:
> I'm Just realizing that maybe I misunderstood. Is IT unlikely that a sgx driver will be release for 3.8.x kernel OR for any other subsequent release?

We are all "users" contact TI directly. ;)

[Gerald Coley]

3.8, I would say no. 3.14, maybe. I would ping TI on the e2e forum. There are a lot of people asking, so more voices the better.

Gerald

[Micka]

Gerald, give me a link to the post on the forum, and I will ask too. And maybe other people ;-)

[Gerald Coley]

http://e2e.ti.com/support/arm/default.aspx

Gerald

[Micka]

Ahahah, maybe we should create a post that everyone will post on it? Gerald you should start.....

[Gerald Coley]

I have already started. I just need more bullets! Do a search and see what others have already posted.

Gerald

[Giuseppe Iellamo]

We are all "users" contact TI directly. ;)

Regards,

Did it.

I'm sorry but from time to time is it difficult to track all the news... thank you very much for the support to everyone.

[John Syne]

Hey Robert,

Here is some interesting info given that [1] shows that SGX is working
over KMS on the OMAP5 and the GLSDK [2] used is based on Linux Kernel
v3.8.

[1] http://tigraphics.blogspot.in/2013/08/qt5-with-kms-omap5-jacinto6.html
[2] http://processors.wiki.ti.com/index.php/OMAP5_GLSDK_6.01_Release_Notes
[3] http://qt-project.org/forums/viewthread/23545

Regards,
John

>
>Regards,
>
>--
>Robert Nelson
>http://www.rcn-ee.com/
>

[Micka]

Hi Robert & Gerald, 

We have some news here : http://e2e.ti.com/support/arm/sitara_arm/f/791/p/298596/1043947.aspx#1043947

A new patch ? But my question, is it for the 3.2.x Kernel ? Or 3.8 ?

micka,

[Micka]

I strongly think that it's for the 3.8 kernel, But I'm waiting your confirmation :)

[John Syne]

From: Micka <micka...@gmail.com>
Reply-To: <beagl...@googlegroups.com>
Date: Tuesday, October 29, 2013 at 5:33 AM
To: <beagl...@googlegroups.com>
Cc: Robert Nelson <robert...@gmail.com>, Gerald Coley <ger...@beagleboard.org>
Subject: Re: [beagleboard] SGX drivers on Kernel 3.8.13

I strongly think that it's for the 3.8 kernel, But I'm waiting your confirmation :)

Looks like this patch is for the 3.12 kernel.

Regards,

John

[Micka]

Well, I'm not sure because Alexander said : "It would be great if you could provide patches for the 3.8 kernel" and the TI guys replied with " These are to be applied againt kernel sources you are using" .

Micka,

[John Syne]

From: Micka <micka...@gmail.com>

Reply-To: <beagl...@googlegroups.com>
Date: Tuesday, October 29, 2013 at 12:58 PM
To: <beagl...@googlegroups.com>

Subject: Re: [beagleboard] SGX drivers on Kernel 3.8.13

Well, I'm not sure because Alexander said : "It would be great if you could provide patches for the 3.8 kernel" and the TI guys replied with " These are to be applied againt kernel sources you are using" .

Hi Micka,

Try to apply patches to the 3.8 kernel and you will get errors. 

Regards,

John

[John Syne]

If you read the release notes for Graphics_SDK_41_00_0_01 [1], it says “This release is build tested against 3.8 kernel"

[1] http://software-dl.ti.com/dsps/dsps_public_sw/sdo_sb/targetcontent/gfxsdk/latest/index_FDS.html

Regards,

John

[frolo...@gmail.com]

Hi

Alexander uses 3.8.13 kernel sources as per 8715.debug.txt file he attached in TI forum.

Linux arm 3.8.13xenomai-bone28 #3 SMP Tue Oct 29 15:00:54 CET 2013 armv7l GNU/Linux

He applied that patch manually, as he got some errors.

среда, 30 октября 2013 г., 0:56:49 UTC+4 пользователь John Syne написал:

[John Syne]

From: <frolo...@gmail.com>
Reply-To: <beagl...@googlegroups.com>
Date: Wednesday, October 30, 2013 at 2:23 AM
To: <beagl...@googlegroups.com>
Subject: Re: [beagleboard] SGX drivers on Kernel 3.8.13

Hi

Alexander uses 3.8.13 kernel sources as per 8715.debug.txt file he attached in TI forum.

Yeah, I’ve been monitoring his progress as well. He is still having issues, but it looks like he is getting closer to getting something working. This does look encouraging. 

Regards,

John

[ol...@illusionlabs.se]

Hi guys,

Just wanted to let anyone know who (like me) is trying to get SGX working on the BBB, and find this thread:

SGX now works fine in console (no x support) if you use branch "am33x-v3.12" from git://github.com/RobertCNelson/linux-dev.git (I just build with default configuration)

I've used it to sucessfully compile Graphics SDK 5_01_00_01 from: http://downloads.ti.com/dsps/dsps_public_sw/gfxsdk/5_01_00_01/index_FDS.html and the demos run great!

Installation & build instructions from here: http://processors.wiki.ti.com/index.php/Graphics_SDK_Quick_installation_and_user_guide#Install_steps

Had some issues with it not finding the size of the framebuffer at first, but calling "fbset -xres 1280 -yres 720 -vxres 1280 -vyres 720" and reloading the omaplfb kernel modules solved it.

Cheers

--Olof

[Giuseppe Iellamo]

Hi Olof,

I'm Struggling with this SGX stuff,
I'm building the Robert kernel and it seems to work, but in the Graphics SDK 5_01_00_01 instructions it says Ensure that the NFS target is setup

What NFS target have you used??

Thanks,
Giuseppe

[Giuseppe Iellamo]

I built the kernel successfully and installed it and now if I try with the demo included in GFX sdk I get:

ubuntu@arm:~/Graphics_SDK_setuplinux_5_00_00_01/gfxsdkdemos/ogles2$ sudo ./OGLES2ChameleonMan
[sudo] password for ubuntu:
PVRShell: EGL 1.4 initialized
Exit message has been set to: "PVRShell: Unable to create surface
".
InitAPI failed!
PVRShell: Unable to create surface

ubuntu@arm:~$ uname -a
Linux arm 3.12.6-bone11.1 #1 SMP Wed Jan 8 10:47:12 CET 2014 armv7l armv7l armv7l GNU/Linux

ubuntu@arm:~$ lsmod
Module                  Size  Used by
omaplfb                10139  0
bufferclass_ti          5803  0
snd_soc_omap            2834  0
snd_pcm_dmaengine       6073  1 snd_soc_omap
pvrsrvkm              178420  2 bufferclass_ti,omaplfb

Any suggestion?

[Robert Nelson]

On Wed, Jan 8, 2014 at 4:54 AM, Giuseppe Iellamo <pep...@gmail.com> wrote:
> I built the kernel successfully and installed it and now if I try with the
> demo included in GFX sdk I get:
>
> ubuntu@arm:~/Graphics_SDK_setuplinux_5_00_00_01/gfxsdkdemos/ogles2$ sudo
> ./OGLES2ChameleonMan
> [sudo] password for ubuntu:
> PVRShell: EGL 1.4 initialized
> Exit message has been set to: "PVRShell: Unable to create surface
> ".
> InitAPI failed!
> PVRShell: Unable to create surface

Yeah.. the "5_00_00_01" drivers are useless for anyone not running
qt's display manager..

[Giuseppe Iellamo]

D'oh!

I thought it was "only" Xorg not working!

Thanks for the reply...

[John Syne]

The demos run fine for me. The default SDK places the demos in /opt

You need to run the following first:

sudo /etc/init.d/rc.pvr start

Then I run the demo as follows:

cd /opt/gfxsdkdemos/ogles2
./OGLES2ChameleonMan

Regards,
John

>
>Regards,
>
>--
>Robert Nelson
>http://www.rcn-ee.com/
>

[John Syne]

The demos run fine for me. The default SDK places the demos in /opt

You need to run the following first:

sudo /etc/init.d/rc.pvr start

Then I run the demo as follows:

cd /opt/gfxsdkdemos/ogles2
./OGLES2ChameleonMan

Regards,
John

>

>Regards,
>
>--
>Robert Nelson
>http://www.rcn-ee.com/
>

[Robert Nelson]

On Wed, Jan 8, 2014 at 2:03 PM, John Syne <john...@gmail.com> wrote:
>
> On 1/8/14, 5:45 AM, "Robert Nelson" <robert...@gmail.com> wrote:
>
>>On Wed, Jan 8, 2014 at 4:54 AM, Giuseppe Iellamo <pep...@gmail.com>
>>wrote:
>>> I built the kernel successfully and installed it and now if I try with
>>>the
>>> demo included in GFX sdk I get:
>>>
>>> ubuntu@arm:~/Graphics_SDK_setuplinux_5_00_00_01/gfxsdkdemos/ogles2$ sudo
>>> ./OGLES2ChameleonMan
>>> [sudo] password for ubuntu:
>>> PVRShell: EGL 1.4 initialized
>>> Exit message has been set to: "PVRShell: Unable to create surface
>>> ".
>>> InitAPI failed!
>>> PVRShell: Unable to create surface
>>
>>Yeah.. the "5_00_00_01" drivers are useless for anyone not running
>>qt's display manager..
> The demos run fine for me. The default SDK places the demos in /opt
>
> You need to run the following first:
>
> sudo /etc/init.d/rc.pvr start
>
> Then I run the demo as follows:
>
> cd /opt/gfxsdkdemos/ogles2
> ./OGLES2ChameleonMan

i think part of the problem, i haven't exactly finished the "install"
part of the sgx build script... so other then building stuff, it
doesn't finish the install..

btw, noticed 5_10_00_01 was out, but can't find the link at the
moment.. wonder what was fixed..

[Giuseppe Iellamo]

I've seen the script is not really complete infact I tried to pick the demo files from the ignore directory... maybe I forgot something.

[Robert Nelson]

On Wed, Jan 8, 2014 at 2:27 PM, Giuseppe Iellamo <pep...@gmail.com> wrote:
>
> I've seen the script is not really complete infact I tried to pick the demo
> files from the ignore directory... maybe I forgot something.

I was porting it from my v3.7.x branch which had set everything up for
the xm.. However a lot had changed so I was tweaking things right up
to the point i found out it wouldn't work for my environment..

so as long as you cp the stuff from the ignore directory and place it
in the right spots it should work..

[littlef...@gmail.com]

I think the link you are looking for is http://processors.wiki.ti.com/index.php/RN_5_01_00_01

The new things mentioned in the page are:

• NEW - Power management suspend/resume support verified on both AM335x & AM43x.
• NEW - gfx_check.sh packaged under ti-components folder of the release package. (gfx_check.sh is a script that needs to be run on the EVM & it dumps all the information about the environment/setup onto serial console. This needs to be provided while reporting problems with graphics SDK).

[John Syne]

On 1/8/14, 12:05 PM, "Robert Nelson" <robert...@gmail.com> wrote:

>
>i think part of the problem, i haven't exactly finished the "install"
>part of the sgx build script... so other then building stuff, it
>doesn't finish the install..
>
>btw, noticed 5_10_00_01 was out, but can't find the link at the
>moment.. wonder what was fixed..

Hi Robert,

http://software-dl.ti.com/dsps/dsps_public_sw/gfxsdk/5_01_00_01/index_FDS.h
tml


Fixed in this Release
* SDOCM00105765 - Bufferclass driver cleanup.


Regards,
John

>
>Regards,
>
>--
>Robert Nelson
>http://www.rcn-ee.com/
>

[der...@gmail.com]

I'm trying to build using the git://github.com/RobertCNelson/linux-dev.git project, I've cloned it, run a 

git checkout -b refs/heads/am33x-v3.12

to get the am33x-v3.12 branch and tried running ./build_kernel.sh

I get to the menuconfig screen and just save and exit, the build starts and shortly thereafter the build fails with a bunch of these messages.

/tmp/ccfwkQRE.s:58: Error: selected processor does not support Thumb mode `mrc p15,0,r2,c1,c0,2'

What am I doing wrong?  The Linaro cross-compiler is installed and I can execute it, I'm not sure what step I'm missing here.

[Robert Nelson]

On Tue, Jan 21, 2014 at 2:43 PM, <der...@gmail.com> wrote:
> I'm trying to build using the git://github.com/RobertCNelson/linux-dev.git
> project, I've cloned it, run a
>
>
> git checkout -b refs/heads/am33x-v3.12
>
> to get the am33x-v3.12 branch and tried running ./build_kernel.sh
>
> I get to the menuconfig screen and just save and exit, the build starts and
> shortly thereafter the build fails with a bunch of these messages.
>
>
> /tmp/ccfwkQRE.s:58: Error: selected processor does not support Thumb mode
> `mrc p15,0,r2,c1,c0,2'
>
> What am I doing wrong? The Linaro cross-compiler is installed and I can
> execute it, I'm not sure what step I'm missing here.

"which" linaro cross compiler?

With the default:

scripts/gcc: Using: arm-linux-gnueabihf-gcc (crosstool-NG
linaro-1.13.1-4.8-2013.10 - Linaro GCC 2013.10) 4.8.2 20131014
(prerelease)
Copyright (C) 2013 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

it builds fine..

http://rcn-ee.homeip.net:8080/job/github_linux-dev/699/consoleText

[der...@gmail.com]

Here's the make line generated after I run the ./build_kernel.sh and save the default config from the menuconfig

make -j8 ARCH=arm LOCALVERSION=-bone12 CROSS_COMPILE=/home/deric/am335_3.12_sgx/dl/gcc-linaro-arm-linux-gnueabihf-4.8-2013.10_linux/bin/arm-linux-gnueabihf-  zImage modules

Here's the version output from the linaro gcc specified in the CROSS_COMPILE.

/home/deric/am335_3.12_sgx/dl/gcc-linaro-arm-linux-gnueabihf-4.8-2013.10_linux/bin/arm-linux-gnueabihf-gcc --version                        !10138

arm-linux-gnueabihf-gcc (crosstool-NG linaro-1.13.1-4.8-2013.10 - Linaro GCC 2013.10) 4.8.2 20131014 (prerelease)

Copyright (C) 2013 Free Software Foundation, Inc.

This is free software; see the source for copying conditions.  There is NO

warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

Do I need to do anything if I'm running a 64 bit host OS?  I have the 32-bit libs installed.

Here's my console output.

http://pastebin.com/SQnQvL6c

Thanks.

Deric.

[John Syn]

From: <der...@gmail.com>
Reply-To: <beagl...@googlegroups.com>
Date: Tuesday, January 21, 2014 at 12:43 PM

To: <beagl...@googlegroups.com>
Subject: Re: [beagleboard] SGX drivers on Kernel 3.8.13

I'm trying to build using the git://github.com/RobertCNelson/linux-dev.git project, I've cloned it, run a 

git checkout -b refs/heads/am33x-v3.12

This is creating a new branch from the master branch. I don’t think this is what you want. What you want to do is this:

git checkout am33x-v3.12

Regards,

John

[der...@gmail.com]

Thanks John,

I'm going to try it on a 32-bit machine, I'm guessing that's where my problem stems from.

Deric

[der...@gmail.com]

Seems to have done it, it's compiling.

[Robert Nelson]

On Tue, Jan 21, 2014 at 3:07 PM, <der...@gmail.com> wrote:
> Here's the make line generated after I run the ./build_kernel.sh and save
> the default config from the menuconfig
>
> make -j8 ARCH=arm LOCALVERSION=-bone12
> CROSS_COMPILE=/home/deric/am335_3.12_sgx/dl/gcc-linaro-arm-linux-gnueabihf-4.8-2013.10_linux/bin/arm-linux-gnueabihf-
> zImage modules
>
> Here's the version output from the linaro gcc specified in the
> CROSS_COMPILE.
>
> /home/deric/am335_3.12_sgx/dl/gcc-linaro-arm-linux-gnueabihf-4.8-2013.10_linux/bin/arm-linux-gnueabihf-gcc
> --version !10138
> arm-linux-gnueabihf-gcc (crosstool-NG linaro-1.13.1-4.8-2013.10 - Linaro GCC
> 2013.10) 4.8.2 20131014 (prerelease)
> Copyright (C) 2013 Free Software Foundation, Inc.
> This is free software; see the source for copying conditions. There is NO
> warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
>
> Do I need to do anything if I'm running a 64 bit host OS? I have the 32-bit
> libs installed.

64bit is fine...

> Here's my console output.
>
> http://pastebin.com/SQnQvL6c

Wow, now that is odd..

Fires up my quantal virtual machine..

[John Syn]

From: <der...@gmail.com>
Reply-To: <beagl...@googlegroups.com>
Date: Tuesday, January 21, 2014 at 1:07 PM

To: <beagl...@googlegroups.com>
Subject: Re: [beagleboard] SGX drivers on Kernel 3.8.13

Here's the make line generated after I run the ./build_kernel.sh and save the default config from the menuconfig

make -j8 ARCH=arm LOCALVERSION=-bone12 CROSS_COMPILE=/home/deric/am335_3.12_sgx/dl/gcc-linaro-arm-linux-gnueabihf-4.8-2013.10_linux/bin/arm-linux-gnueabihf-  zImage modules

Why don’t you just run build_kernel.sh?

This will download the correct Linaro compiler, setup any environment variables and build everything you need. 

If you modify the kernel, then run tools/rebuild.sh

Regards,

John

--

[Robert Nelson]

On Tue, Jan 21, 2014 at 3:40 PM, <der...@gmail.com> wrote:
> Seems to have done it, it's compiling.

For reference, what distro/release is this box?

[stefboe...@gmail.com]

I also added a comment on the TI forum regarding the fact that we found a solution that works for us, three patches on the 3.8 kernel that fixes a IRQ issue, and include the SGX in the device tree and some fixes in the omap gpu drivers / board files. 

For anyone intrested you can find them here: 

http://pastebin.com/aEUP69ZU

http://pastebin.com/Rh3fB7uP

http://pastebin.com/DY83wjdM

Op donderdag 24 oktober 2013 10:10:24 UTC+2 schreef mail.ar...@gmail.com:

Hello,

I am currently trying to get OpenGL ES running on the Bone, but I get some problems with the kernel modules. I already posted on the TI formus. Maybe someone can help:

http://e2e.ti.com/support/arm/sitara_arm/f/791/t/298596.aspx

Regards

Strahlex

[Charles Steinkuehler]

Great news! So this _is_ for the BeagleBone (with the HDMI and other
BeagleBone kernel patches) or just for the plain AM335x kernel?

--
Charles Steinkuehler
cha...@steinkuehler.net

[Robert Nelson]

On Wed, Feb 5, 2014 at 10:22 AM, Charles Steinkuehler
<cha...@steinkuehler.net> wrote:
> Great news! So this _is_ for the BeagleBone (with the HDMI and other
> BeagleBone kernel patches) or just for the plain AM335x kernel?

They seem to be setup for our patchset..

Just pushed them:
https://github.com/RobertCNelson/linux-dev/commit/c29d0000ed356595c216ab534417f44528c27464

[Stef]

Correct, 

Thanks Robert

Regards,

Stef

--
For more options, visit http://beagleboard.org/discuss
---

You received this message because you are subscribed to a topic in the Google Groups "BeagleBoard" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/beagleboard/qQTC7C8g0I8/unsubscribe.
To unsubscribe from this group and all its topics, send an email to beagleboard...@googlegroups.com.

[mukel...@gmail.com]

Should these SGX drivers work with Android? I built an image with the newest 3.8.13 kernel but when I try to remove the "qemu=1" from uEnv.txt the OS doesn't appear. Although the "android" loading screen does appear.

I followed the method from http://icculus.org/~hendersa/android/. Have I missed a step?

[jan.we...@gmail.com]

Hi,

just to be sure and i am not totally wrong, kernel 3.8.13-bone40.1 is working with Graphics SDK 4.10.00.01 ?

Cause it doesnt work at my bbb :(

I build the kernel with standard config. - no problem
than i build SDK with " make OMAPES=8.x FBDEV=yes SUPPORT_XORG=0" - no problem
make install

at the BBB i execute /etc/init.d/335x-demo
restart
after restart i execute /etc/init.d/rc.pvr wich gives:
root@arm:~# /etc/init.d/rc.pvr start
Module pvrsrvkm failed to load. Retrying.
Running /sbin/depmod
/usr/local/bin/pvrsrvctl: SrvInit failed (already initialized?) (err=4)


am i missing something ?

which distribution are you working with ?
i took ubuntu minimal image. with no extra packages

thx for any help

Jan

[David Paden]

I know this is kinda old, but I've been working with the Graphics SDK for AM335x for a while for work.  I was wondering why built drivers haven't been added to the Deban package manager yet...or if I just can't find them.  Has this been done yet or is it planned?

Dave

[Robert Nelson]

On Mon, Jun 23, 2014 at 4:57 PM, David Paden <david...@gmail.com> wrote:
> I know this is kinda old, but I've been working with the Graphics SDK for
> AM335x for a while for work. I was wondering why built drivers haven't been
> added to the Deban package manager yet...or if I just can't find them. Has
> this been done yet or is it planned?

Sorry, due to the EULA and where i live, that's just not possible for me todo.

Instead i've tried to streamline as much as possible, aka follow:

http://elinux.org/BeagleBoardDebian#SGX_BeagleBone.2FBeagleBone_Black

Then just copy the output "GFX_5.01.01.01.tar.gz" to all your machines
under your control/supervision.

[Nick Apperley]

Just had a crazy idea in relation to what Olof has done. With the sample JavaFX 8 program I am trying to run the eglGetConfigAttrib failed! error appears if the program is run using EGL 2. What would happen if the framebuffer was set before running the sample program?

[Nick Apperley]

Tried out the idea by running the program remotely and ended up with the same error. Did it again via DirectFB instead of EGL 2 and ended up with a Linux style BSOD (Black Screen of Despair). Luckily I could run my pygame demo (remotely) to reset the screen back to normal.

[popov...@gmail.com]

Not shure if it's better to create new theme..

 

I'm trying to build sgx drivers for 3.8 kernel. I know that there is a lot of disccusions here about that, but I can't find any exact information here.

What I have:

BeagleBone Black + element14 LCD7 cape

RobertCNelson "bb-kernel" kernel, branch "remotes/origin/am33x-v3.8.13.x" + patch to create (un)register_vsync_cb functions for SGX code to compile.

Graphics SDK 5.01.01.01 from TI.

Everything compiles good. But on insmod pvrsrvkm.ko I got this: http://paste.org/73324

(Lines started by "MVC: " are added by me for debug reasons)

1. If I understand correctly, pvrsrvkm crashed on line 575 here: https://github.com/RobertCNelson/ti-sdk-pvr/blob/5.01.01.01/Graphics_SDK/GFX_Linux_KM/services4/system/ti335x/sysutils_linux.c#L575 when trying to map memory from TIMER7 region.

devmem2 0x4804A054 (SYS_TI335x_GP7TIMER_TSICR_SYS_PHYS_BASE = 0x4804A054) gives bus error and "abort on non-linefetch". Am I right thinking that sgx driver wants to use timer7, but it's disabled?

2. Exactly same scheme works good with 3.2 kernel from element14 and same GFX SDK, modules load good, demos from /usr/bin/SGX work good. But timer7 is disabled too and devmem2 on it gives exactly same error.

What am I doing wrong and is there any way to get 3.8 and gpu working?

[popov...@gmail.com]

[Rafael Coutinho]

Hi Mukel,

have you succeed with SGX and kernel 3.8.13 on Android?

[Giuseppe Iellamo]

It seems to me not very fair to say we have a board with a 3d accelerator, we sell it and we won't give you the driver nor the specs to use it...

To my understanding with 3.2 kernel it was working. Any advice on how to resume a working 3.2 distribution I may use?

Il giorno giovedì 24 ottobre 2013 22:21:12 UTC+2, Gerald ha scritto:

As I understand it, the DRM driver is not compatible with the old SGX frame buffer architecture. TI tells us they are not rewriting the code for 3.8. Not sure about future kernel releases. Best bet would be to ask TI on this one.

Gerald

On Thu, Oct 24, 2013 at 1:39 PM, <mail.ar...@gmail.com> wrote:

Hi,

What exactly is missing? I have read something about that the clock for GPU is not enabled. Can't we fix this or can't we get TI to fix this?

Regards

Alexander

Am Donnerstag, 24. Oktober 2013 15:07:11 UTC+2 schrieb RobertCNelson:

On Thu, Oct 24, 2013 at 3:10 AM,  <mail.ar...@gmail.com> wrote:
> Hello,
>
> I am currently trying to get OpenGL ES running on the Bone, but I get some
> problems with the kernel modules. I already posted on the TI formus. Maybe
> someone can help:
> http://e2e.ti.com/support/arm/sitara_arm/f/791/t/298596.aspx

See a few of the other threads in this forum for the full details on
this.. Essentially we do not have the needed bits from TI at this
time..

Regards,

--
Robert Nelson
http://www.rcn-ee.com/

--

For more options, visit http://beagleboard.org/discuss
---

You received this message because you are subscribed to the Google Groups "BeagleBoard" group.

To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard...@googlegroups.com.