I have two different development kits for the DM6446. The first is running
the old MontaVista 2.6.10 on a 3rd party demo board.
The second is running using the PSP 3.0.1 (linux 2.6.32) from the Arago
project running on the EVM-6446.
On the first 2.6.10 I can get a sustained write speed of
Thanks Chris,
I am still on Codec Engine 2.1 since we have not moved to MVPro4 yet. Also
I'm using the DM6446.
Josh
-Original Message-
From: Ring, Chris [mailto:cr...@ti.com]
Sent: Thursday, June 04, 2009 5:57 PM
To: Stephane St-Hilaire; Joshua Hintze;
davinci-linux-open-source
Lately Ive been trying to track down a bug on some of my arm code running
on the davinci. As most programmers know there can be plenty of sigfaults.
The problem is that the codec engine is running when I sigfault so it
doesnt get closed down properly. This means the next time I try to launch
my
Previous to updating codec engine (I was on 2.00 now on 2.10) I could launch
and debug my ARM code with GDB debugger. However now whenever I try to
launch it fails. Setting CE_DEBUG=3 this is my output.
@1,540,122us: [+2 T:0x424ccb60 S:0x424cc52c] OP - Processor_create_d Adding
DSP segment #0
James,
Try this link:
http://www.mail-archive.com/davinci-linux-open-source@linux.davincidsp.com/msg01083.html
I think the key is the ::off on the autoconf setting on the end and providing
all the needed information. Let me know if that works because I would be
interested.
Josh
Thank you,
I think your explanation was more precise. I had just stumbled across that in
the past but have not yet tried it.
Josh
-Original Message-
From: Phil Quiney [mailto:phil.qui...@bigfoot.com]
Sent: Friday, December 19, 2008 2:07 PM
To: Joshua Hintze
Cc: 'James T Long'; davinci
Thank you guys a lot.
I will be trying out these options today and report on results.
Thanks!
Josh
-Original Message-
From: davinci-linux-open-source-boun...@linux.davincidsp.com
[mailto:davinci-linux-open-source-boun...@linux.davincidsp.com] On Behalf Of
Phil Quiney
Sent: Thursday,
Yeah I'm using open embedded for the beagle board (or will be, I just got my
beagle board). I wasn't sure if open embedded worked for the davinci.
Josh
-Original Message-
From: Philip Balister [mailto:phi...@balister.org]
Sent: Thursday, December 18, 2008 8:21 AM
To: Joshua Hintze
So I try querying a few packages to see what was installed and the strange
thing is that its saying that nothing common is installed so I queried for
all packages
mvl-enterprise-rpm -qa and this is my output
[r...@localhost rpm]# mvl-edition-rpm -qa
host-filesystem-4.0-1.0.0.0501131
When I wanted to remove the very long time delay from DHCP (because I'm
using a static IP address) I only removed DHCP from the kernel compilation
and not the entire Ethernet. There is also a way I've seen mentioned on this
list on how to not have it check DHCP from the command line arguments to
Hello all,
A couple weeks ago I updated to a new version of codec engine. Today I tried
to run gdb to debug my program but for some reason now the codec engine does
not open up my .x64P when trying to debug. However it runs fine outside of
gdb. This used to work with an older codec engine.
Hi guys,
I've been working with the Davinci now for over a year and I've had a lot of
success. My current setup I use NAND flash for uImage/uboot and then I load
my rootfs from an IDE hard drive.
I have a new project now where we are not going to use the IDE hard drive
anymore and so I'm
Has anybody successfully gotten the DM6446 to boot from an SD card? Is it
even possible? I've been playing with the OMAP chips lately where their boot
ROM can load in a signed first stage loader from the SD card (FAT32
partition), then load in uImage from the SD and finally rootfs. Can the
same
+jitendrajain=emerson@linux.davincidsp.
com
[mailto:davinci-linux-open-source-bounces+jitendrajain=emerson@linux.dav
incidsp.com] On Behalf Of Joshua Hintze
Sent: Tuesday, December 16, 2008 9:32 PM
To: davinci-linux-open-source@linux.davincidsp.com
Subject: Boot DM6446 from SD?
Has anybody
Thank you,
I got it working. System commands now work!!!
Thanks again,
Josh
From: Griffis, Brad [mailto:[EMAIL PROTECTED]
Sent: Wednesday, December 03, 2008 5:17 PM
To: Joshua Hintze; davinci-linux-open-source@linux.davincidsp.com
Subject: RE: system() and Codec Engine
I
Hey guys,
I'm having a problem when using a system() C function call and the codec
engine is running. I know inside a POSIX system(.) it essentially forks,
execs, waitpid. From what I read a child process is spawned with the
parent's copy of a data space, heap, and a stack.
Anyways when I
(for me atleast):
Kernel_source/arch/arm/mach-davinci/clock.c(.h)
Josh
From: Shibasis Ganguly [mailto:[EMAIL PROTECTED]
Sent: Sunday, November 23, 2008 10:00 AM
To: [EMAIL PROTECTED]
Subject: SPI in davinci
Hi Joshua Hintze,
I am working on spi in wince davinci. I have gone through
Standard SD, miniSD, microSD are all the same internally as far as I know.
The SPI interface is exactly the same so I'm pretty sure the parallel would
be as well.
Josh
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf
Of ???
Sent: Friday, October 31, 2008
I am trying to compile multiple codecs to use on the davinci that get
compiled into a single server package (.x64P). There are several functions
and global variables defined and used individually by each codec that have
the same names. This worked fine when I had 2 codecs working, but when I
Hello,
I am having trouble boot from NAND flash larger images. Previously my kernel
uImage was around 1.4 megs. I just recently enabled some other options and
it grew in size to 1.6 megs.
Our nand flash layout is like this.
0x-0x00018000 : UBL image
0x00018000-0x00044000 :
:57 PM
To: Joshua Hintze; davinci-linux-open-source@linux.davincidsp.com
Subject: Re: Uboot large UImage
Did you have any bad blocks during the erase? U-boot isn't smart enough to
deal with bad blocks ... I have a board that has one in the middle of the
kernel image. I moved it away from the bad
Hey guys,
Recently I wanted to call a command line script from my program that is
using the Codec Engine. So I thought an easy way to do this is to call the
system(const char *cmdstring) function declared in stdlib.h.
The command line string executes however upon return it locks the call
Message-
From: shyamal mehta [mailto:[EMAIL PROTECTED]
Sent: Thursday, October 02, 2008 3:18 AM
To: 'John Roberts'; 'Linux DaVinci'; Joshua Hintze
Subject: RE: Arm and DSP debugging with CCS?
Josh, it has a procedure to debug DSP code. How can I debug my ARM side
kernel code? Has anybody done
Yes there is. I have done it before by using these directions.
http://wiki.davincidsp.com/index.php?title=Debugging_the_DSP_side_of_a_CE_ap
plication_on_DaVinci_using_CCS
Josh
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
John Roberts
Sent: Wednesday, October 01, 2008
Not the DM355 but we are using this one
http://www.einfochips.com/services/DVPB.html that makes use of the DM6446
and is an all in one board and much smaller than the DVEVM.
Josh
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
John Huang
Sent: Tuesday, September 30, 2008
Well, I'm wondering if these files are only in the Monte Vista and not GIT
kernels. I don't know I don't use GIT.
Josh
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
Phil Quiney
Sent: Thursday, September 18, 2008 2:12 AM
To: Ying Huang;
That's the correct location. Atleast that's where mine is.
What kernel are you using? Monte Vista?
Josh
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
Ying Huang
Sent: Wednesday, September 17, 2008 10:45 PM
To:
I'm not sure about the Davinci (I haven't tried yet). But on other projects,
using the Gumstix processor, I was able to transfer the kernel using Uboots
Kermit serial protocol. I would think you could do the same for U-boot.
Josh
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]
om] On
:46 AM
To: Joshua Hintze; davinci-linux-open-source@linux.davincidsp.com
Subject: RE: Kernel update through serial port.
Hi Josh,
Thanks a lot for update.
I am trying the kermit mode - I did the following
EVM # loadb
Ready for binary (kermit) download to 0x8070 at 115200bps
That’s what he is trying to do, however he doesn't know where his kernel
sources are which are required for building the module.
Josh
-Original Message-
From: Caglar Akyuz [mailto:[EMAIL PROTECTED]
Sent: Tuesday, July 01, 2008 12:02 AM
To: Joshua Hintze
Cc: 'Manh-Hung Le'; davinci
that
need not be there.
Do you have information on setting up DMA transfers on the linux side?
Thanks,
Josh
From: Tonald DL [mailto:[EMAIL PROTECTED]
Sent: Saturday, June 14, 2008 2:43 AM
To: Joshua Hintze
Cc: davinci-linux-open-source@linux.davincidsp.com
Subject: Re: Memcpy lots
Hey guys,
I am copying a lot of data around in my application and I'm wondering if my
memcpy (the ansi C one used in the arm_v5t_le-gcc compiler) is doing a byte
by byte copy or larger copy.
If it's already doing a quad or 4 byte copy, since that is what the bus size
is, then great. If not
Hi Steven,
I'm not sure why you are asking me. I don't work for TI I'm just a member of
the davinci mailing list and I try to help out when I know the answer. In
this case I don't really know the answer other than you doing a search of
your file system.
If you have the default install of the
Your make file looks about the same as mine. However I also have this line.
MAKE_ENV = ARCH=arm CROSS_COMPILE=$(MVTOOL_PREFIX)
Josh
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Andrea Gasparini
Sent: Friday, March 21, 2008 8:23 AM
To:
on the davincifb?
Thanks,
Josh
-Original Message-
From: Ashuaria Lee [mailto:[EMAIL PROTECTED]
Sent: Wednesday, February 13, 2008 4:32 PM
To: Joshua Hintze
Cc: davinci-linux-open-source@linux.davincidsp.com
Subject: Re: Suggestions on image display
Hi,
IMO I would use the FrameBuffers virtual x,y
the same. It doesn't get squished in the virtual axis at all.
Could somebody clarify this for me?
I appreciate it,
Josh
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
Joshua Hintze
Sent: Monday, February 18, 2008 10:10 AM
To: 'Ashuaria Lee'
Cc: davinci-linux
Hello, I'm looking for some suggestions on the best way to go about solving
this problem.
Lets say I have an image that is 1024x1024. What would be the best way to
get it onto the screen? Every second a new 1024x1024 image is procedurally
generated.
Should I write a program that opens up
I didn't hear anything in my last message it might have been ignored since I
changed what was the original topic. This is my problem.
How can I change the cl6x compiler switches that are defaulted in the Codec
Engine when using package.bld.
For example I want to add a couple like: -pm -op2
and -o3 which one the compiler will use.
Thoughts?
Josh
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]
om] On Behalf Of Joshua Hintze
Sent: Friday, February 08, 2008 11:52 AM
To: davinci-linux-open-source@linux.davincidsp.com
Subject: Copts in Package.bld in codecs
I didn't hear
As long as the -o3 wins that's what I need to know. The rest are required so
I won't be needing to remove those.
Thank you.
Josh
From: Wan, Vincent [mailto:[EMAIL PROTECTED]
Sent: Friday, February 08, 2008 3:45 PM
To: Joshua Hintze
Cc: davinci-linux-open-source@linux.davincidsp.com
[mailto:[EMAIL PROTECTED]
Sent: Thursday, February 07, 2008 12:20 PM
To: Joshua Hintze; davinci-linux-open-source@linux.davincidsp.com
Subject: RE: Compiling Targets and Codecs
Please review this thread and see if it answers your question.
http://osdir.com/ml/linux.davinci/2006-09/msg00095.html
Hello,
I am running with the DM6446 processor and something that I have not yet
figured out is the compiling targets when working with the codec engine.
What I mean is if I want to have the codec running on the DSP and the
executable running on the ARM do I need to produce two different
Hello,
I have successfully compiled my codec into a local only codec following the
example in the video_copy/singlecpu. The reason I am doing this is so I can
step through it in the simulator and have access to better profiling to
understand cache misses and hits and stuff going on.
My next question is how does the Memory_contigAlloc function work in the
simulator? My understanding is that normally (on linux) it works by
accessing the CMEMK module and having that allocate a region that I define.
However since I'm running this local to the DSP (for testing purposes). I
PROTECTED]
om] On Behalf Of Joshua Hintze
Sent: Wednesday, February 06, 2008 8:54 AM
To: davinci-linux-open-source@linux.davincidsp.com
Subject: Using the DM6446 Simulator
Hello,
I have successfully compiled my codec into a local only codec following the
example in the video_copy/singlecpu
figured use simulation
for optimizing and then be able to quickly switch back to the real hardware
for testing.
Thanks again,
Josh
From: Davor Magdic [mailto:[EMAIL PROTECTED]
Sent: Wednesday, February 06, 2008 10:47 AM
To: Joshua Hintze
Cc: davinci-linux-open-source
Have you tried setting the environment variable TRACEUTIL_VERBOSE=2, and
then run your application and see if there is anything fishy before your
application hangs?
Josh
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
Sam Hague (shague)
Sent: Thursday,
I see this wiki, http://wiki.davincidsp.com/index.php?title=DDRALGHEAP. This
looks like the best I can do from the Arm side.
The wiki only describes a method for determining how much of the DDRALGHEAP is
being used. This is the location where the codecs will malloc up memory if
setup
I'm actually working within the Codec Engine on a DSP side codec. So I'm
curious if its supported there.
Josh
From: Rondeau, Michael [mailto:[EMAIL PROTECTED]
Sent: Wednesday, January 30, 2008 7:32 AM
To: Joshua Hintze
Subject: RE: GT_trace1 and floats/doubles
Josh,
If your
Thanks, I tried the sprintf, but then I passed in the char * directly into
the GT_trace. I will try it with the %s format specifier.
Josh
From: Ben White [mailto:[EMAIL PROTECTED]
Sent: Wednesday, January 30, 2008 9:04 AM
To: Joshua Hintze; davinci-linux-open-source
Is there any way to show floats and our double values in the GT trace
output?
I've tried %f and %lf in the format specifier without luck. %d, %s, %x seem
to work.
Has anybody got this working?
Thanks,
Josh
___
I'm going to give it a try this week. Hopefully by the end of this week I
can let you know how it went.
Josh
-Original Message-
From: David A Kondrad [mailto:[EMAIL PROTECTED]
Sent: Monday, January 28, 2008 7:15 AM
To: [EMAIL PROTECTED]
Cc: davinci-linux-open-source@linux.davincidsp.com
So is nobody using C++ to program their apps?
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]
om] On Behalf Of Joshua Hintze
Sent: Thursday, January 24, 2008 4:11 PM
To: davinci-linux-open-source@linux.davincidsp.com
Subject: C++ Apps and Codec Engine
Hello,
I'm about ready
The montevista target tools are a loadable kernel module. You can easily
remove the module from loading on startup in the startup scripts in
/etc/rc.d/rc3.d
Josh
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
Arie Muijnck
Sent: Friday, January 25, 2008
Hello,
I'm about ready to move away from just modifying the example applications
that came with the DVEVM and wanting to start creating my own from scratch.
All of the example apps seem to be using arm_v5t_le-gcc.
Does anybody forsee any problem with me using the arm_v5t_le-g++ compiler
Hey guys,
I just updated my dsplink module from dsplink_1_30.x to dsplink 1.40.x.
After spending lots of time recoding some of the example codes/servers/apps.
I now have an error when I launch an app that uses a codec. The following is
what I see as an output.
App- Application started.
On Dec 19, 2007 12:55 AM, Joshua Hintze [EMAIL PROTECTED] wrote:
Sean,
Thanks for sending this. One thing that I'm not really sure of is
the macros or functions SPI_REG. Where were these defined? Also the
registers SPIPC0 and others. Did you define all of these yourself?
Also this might seem
PROTECTED]
Sent: Tuesday, December 18, 2007 1:34 AM
To: Joshua Hintze; davinci-linux-open-source@linux.davincidsp.com
Subject: Re: SPI on Davinci.
Dear Joshua
OK, I give you a simple program segment.
You need to see the document about SPI of DM644x
Since nobody answered yet, I'm curious has anybody done SPI interfacing on
the Davinci?
Thanks,
Josh
---
Hey guys,
Have any of you interfaced to the SPI port on the Davinci 6446. I got the
i2c
,
Sean
On Dec 18, 2007 2:09 AM, Joshua Hintze [EMAIL PROTECTED] wrote:
Since nobody answered yet, I'm curious has anybody done SPI
interfacing on the Davinci?
Thanks,
Josh
---
---
---
---
---
---
-
Hey guys
Hey guys,
Have any of you interfaced to the SPI port on the Davinci 6446. I got the
i2c working pretty easy using the i2c-dev character driver but there is no
such thing as spidev until linux kernel 2.6.22.
So I'm wondering how others are doing it. Are you writing your own kernel
You need to open
/dev/tty/0
I believe that's right.
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
Allan Larsen
Sent: Thursday, December 06, 2007 3:13 AM
To: Marcel Katz; davinci-linux-open-source@linux.davincidsp.com
Subject: RE: How to open UART0 (ttyS0)
I think
Sorry,
I just looked and its
/dev/tts/0
Josh
From: Marcel Katz [mailto:[EMAIL PROTECTED]
Sent: Thursday, December 06, 2007 9:48 AM
To: Joshua Hintze
Cc: Allan Larsen; davinci-linux-open-source@linux.davincidsp.com
Subject: Re: How to open UART0 (ttyS0)
It's not working. I
63 matches
Mail list logo
