Hi,
Everyone。I have some problems about my platform of davinci dm6446, which is
developed by my professor's group.
The problem is:
When I run the encodedecode demo on my platform, after a few seconds, the
DSP CPU load is 99%,and then the
system will halt for ever. Can anyone tell me the possible p
This patch removes reads of the following registers:
esr, esrh, eesr, eesrh, iesr, iesrh, qeesr
are write 1 to set registers
ecr, ecrh, eecr, eecrh, emcr, emcrh,
icr, icrh, secr, secrh,
ccerrclr, qeecr, qemcr, qsecr,
are write 1 to clear registers
Signed-off-by: Troy Kisky <[EMAIL PROTECTED]>
Does anybody know if AIO can be bypassed in the Gadget FS application if I
do not want to do any IO operation on my ISOC endpoint in user space? I will
try to set up DMA instead to let DMA take care of ISOC streaming in kernel?
Which means I do not need to handle ISOC endpoint in user space.
Can
Andrea Gasparini wrote:
> I'm trying to compile fw_{print,save}env from uboot-1.2.0 for
> dm355 environment.
> I got a lot of errors for (imho) broken include directives.
>
> fw_env has a bug,and erroneusly includes ,
> while it should include .
I got this working just the other day :)
In than
Hi,
I'm trying to compile fw_{print,save}env from uboot-1.2.0 for dm355
environment.
I got a lot of errors for (imho) broken include directives.
fw_env has a bug,and erroneusly includes , while it should
include .
Errors starts like this:
v$ make env
In file included from fw_env.c:34:
/mnt/mon
The margins and sync_len don't seem right. I don't have my settings
with me so I can't check . Look at this website. It will help you find
the settings for the margin and sync pulses.
http://www.tkk.fi/Misc/Electronics/faq/vga2rgb/calc.html
- Brad
On Nov 26, 2008, at 11:47 PM, Lori Gao wrot
Hi,
Where are you hard-coding buffer size? Can you also check whether user pointer
buffer exchange mechanism is supported in the driver?
Thanks,
Brijesh Jadav
From: Mr. V R [mailto:[EMAIL PROTECTED]
Sent: Friday, November 28, 2008 6:18 PM
To: Jadav, Brijesh R; d
Chandra,
You need to make sure /montavista/pro/bin/ is in your PATH.
Thanks,
Sekhar
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of chandrashekar
reddy
Sent: Friday, November 28, 2008 5:03 PM
To: davinci-linux-open-source@linux.davincidsp.com
Subj
I am not getting any error after hardcoding buffer size to 884736, and able
to dequeue and enque buffer successfully, but this buffer does not contain
any frame data.
On Fri, Nov 28, 2008 at 5:10 PM, Jadav, Brijesh R <[EMAIL PROTECTED]> wrote:
> Hi,
>
> As far as I remember, earlier dm644x captur
Looks like the Monta Vista toolchain is used. Alternatively, you can
try adding /opt/montavista/pro/bin in your search path. There should be
a pre-compile mkimage binary there.
Regards,
Steve
On Fri, 2008-11-28 at 17:40 +0530, Jadav, Brijesh R wrote:
> Hi,
>
>
>
> I think you need to comp
Hi Andrea,
Thanks for reply.
1) DVSDK 1.30 kernel does not provide CONFIG_CAPTURE_DAVINCIHD ( I could not
find in autoconfig.h or KConfig anywhere, but Makefile mentions about it).
Do I have to apply patch to enable support this? (I tried tweak it manually
but looks like I have to disable part of
Hi,
I think you need to compile u-boot to get mkimage binary
Thanks,
Brijesh Jadav
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of chandrashekar
reddy
Sent: Friday, November 28, 2008 5:03 PM
To: davinci-linux-open-source@linux.davincidsp.com
Subjec
Hi,
As far as I remember, earlier dm644x capture code was not supporting user
pointer buffer exchange mechanism. Can you tell me which error you are getting?
Does it allow to do reqbuf?
Thanks,
Brijesh Jadav
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf
Hi all,
While compiling the kernel i am getting this problem.
[EMAIL PROTECTED]:~/workdir/lsp/ti-davinci_evm-arm_v5t_le/linux-2.6.10_mvl401>
make ARCH=arm CROSS_COMPILE=arm_v5t_le- uImage
CHK include/linux/version.h
SPLIT include/linux/autoconf.h -> include/config/*
SYMLINK include/as
Hi,
Mr. V R spiffera, alle Friday 28 November 2008 circa:
> 1. The sample code without modification in buffer size works well with
> MMAP but fails in USERPTR
> 2. If we provide buffer more then or equal to 884736, capture is
> possible with USERPTR, but driver does not copy frame in the buffer.
15 matches
Mail list logo