Seg Fault on first 2 runs of dynamically linked executables.

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
44 messages Options
123
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Seg Fault on first 2 runs of dynamically linked executables.

pac-3
Hello,
      We're using a pxa255 based gumstix ( with the r1161 toolchain ),
and have a dynamically linked executable ( which is linked with several
of our own libraries as well as system libraries. )  If we store this in
nfs it starts fine, however if we put it in the jffs2 filesystem on
flash, then it fails with Segmentation Fault, the first 2 Times we run
it, and then works fine on the 3rd and all subsequent runs, until the
gumstix is rebooted.
If we run this in strace from flash then it also works fine, and if we
run it in gdb then we don't get a useful stack trace from it, but it
fails before breaking on main().

I also noticed that ldd doesn't find the libraries even though they are
in the exported LD_LIBRARY_PATH, so have also tried putting them in /lib
and re-running ldconfig after which ldd finds them ok, but we still have
the first 2 runs fail after a reboot.

When buildling objects we are using:-
arm-linux-gcc -c -ansi -mlong-calls -pthread -mtune=xscale -fPIC -ggdb
-Wall -D_GNU_SOURCE ...

When building our shared libraries:-
arm-linux-gcc -mtune=xscale  -shared  -o <libraryname> <object>...

When building the executable:-
arm-linux-gcc ...   -ldl -lpthread -lrt -o ...

Has anyone any come across any problems like this, or have any suggestions?

Regards
Philip



-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Seg Fault on first 2 runs of dynamically linked executables.

Jon Hylands
On Thu, 04 Oct 2007 16:14:11 +0100, "pac" <[hidden email]> wrote:

>       We're using a pxa255 based gumstix ( with the r1161 toolchain ),
> and have a dynamically linked executable ( which is linked with several
> of our own libraries as well as system libraries. )  If we store this in
> nfs it starts fine, however if we put it in the jffs2 filesystem on
> flash, then it fails with Segmentation Fault, the first 2 Times we run
> it, and then works fine on the 3rd and all subsequent runs, until the
> gumstix is rebooted.
> If we run this in strace from flash then it also works fine, and if we
> run it in gdb then we don't get a useful stack trace from it, but it
> fails before breaking on main().

I've run into that before, when running Squeak. What we did to fix it was
basically pre-cache the shared libraries before running the executable...

cp vm-display-null /dev/null
cp vm-sound-null /dev/null
./squeak -vm-display-null -headless gumstix.image -$1

That fixes the problem for me...

Later,
Jon

--------------------------------------------------------------
   Jon Hylands      [hidden email]      http://www.huv.com/jon

  Project: Micro Raptor (Small Biped Velociraptor Robot)
           http://www.huv.com/blog

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Seg Fault on first 2 runs of dynamically linked executables.

pac-3
Jon Hylands wrote:

> On Thu, 04 Oct 2007 16:14:11 +0100, "pac" <[hidden email]> wrote:
>
>>       We're using a pxa255 based gumstix ( with the r1161 toolchain ),
>> and have a dynamically linked executable ( which is linked with several
>> of our own libraries as well as system libraries. )  If we store this in
>> nfs it starts fine, however if we put it in the jffs2 filesystem on
>> flash, then it fails with Segmentation Fault, the first 2 Times we run
>> it, and then works fine on the 3rd and all subsequent runs, until the
>> gumstix is rebooted.
>> If we run this in strace from flash then it also works fine, and if we
>> run it in gdb then we don't get a useful stack trace from it, but it
>> fails before breaking on main().
>
> I've run into that before, when running Squeak. What we did to fix it was
> basically pre-cache the shared libraries before running the executable...
>
> cp vm-display-null /dev/null
> cp vm-sound-null /dev/null
> ./squeak -vm-display-null -headless gumstix.image -$1
>
> That fixes the problem for me...

Thanks Jon, cat to dev null of our libraries works too.

Does anyone know what the cause of this problem is?

Thanks
Philip

>
> Later,
> Jon
>
> --------------------------------------------------------------
>    Jon Hylands      [hidden email]      http://www.huv.com/jon
>
>   Project: Micro Raptor (Small Biped Velociraptor Robot)
>            http://www.huv.com/blog
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >> http://get.splunk.com/
> _______________________________________________
> gumstix-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gumstix-users



-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Seg Fault on first 2 runs of dynamically linked executables.

Scott Kroll
I'm curious to find out if this is the cause of my problems as well.  I posted a thread earlier today about this.  I'll look into it when I make it home.

On 10/4/07, pac <[hidden email]> wrote:
Jon Hylands wrote:

> On Thu, 04 Oct 2007 16:14:11 +0100, "pac" <[hidden email]> wrote:
>
>>       We're using a pxa255 based gumstix ( with the r1161 toolchain ),
>> and have a dynamically linked executable ( which is linked with several
>> of our own libraries as well as system libraries. )  If we store this in
>> nfs it starts fine, however if we put it in the jffs2 filesystem on
>> flash, then it fails with Segmentation Fault, the first 2 Times we run
>> it, and then works fine on the 3rd and all subsequent runs, until the
>> gumstix is rebooted.
>> If we run this in strace from flash then it also works fine, and if we
>> run it in gdb then we don't get a useful stack trace from it, but it
>> fails before breaking on main().
>
> I've run into that before, when running Squeak. What we did to fix it was
> basically pre-cache the shared libraries before running the executable...
>
> cp vm-display-null /dev/null
> cp vm-sound-null /dev/null
> ./squeak -vm-display-null -headless gumstix.image -$1
>
> That fixes the problem for me...

Thanks Jon, cat to dev null of our libraries works too.

Does anyone know what the cause of this problem is?

Thanks
Philip

>
> Later,
> Jon
>
> --------------------------------------------------------------
>    Jon Hylands      [hidden email]      http://www.huv.com/jon
>
>   Project: Micro Raptor (Small Biped Velociraptor Robot)
>            http://www.huv.com/blog
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >> http://get.splunk.com/
> _______________________________________________
> gumstix-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gumstix-users



-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Website Problem

Demetris Zavorotnitsienko

Guys. I still can’t buy anything through the website. I have even created a new account and still no go.

 

Please help me out here. I can’t get passed the “Delivery Information”.  The country is Cyprus. (EU)


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Seg Fault on first 2 runs of dynamically linked executables.

pac-3
In reply to this post by Scott Kroll
Scott Kroll wrote:
> I'm curious to find out if this is the cause of my problems as well.  I
> posted a thread earlier today about this.  I'll look into it when I make
> it home.

The strange thing is that the binaries in the gumstix release which are
also build shared work fine.

Does anyone know if there is some special procedure/compiler/linker
flags which need to the used to prepare the libraries for use from a
jffs2 filesystem?

I'm guessing this is some sort of issue with mmap io, which is probably
handled specially for jffs2 on flash on the system memory bus?

Thanks
Philip

>
> On 10/4/07, * pac* <[hidden email] <mailto:[hidden email]>> wrote:
>
>     Jon Hylands wrote:
>     > On Thu, 04 Oct 2007 16:14:11 +0100, "pac" <[hidden email]
>     <mailto:[hidden email]>> wrote:
>     >
>     >>       We're using a pxa255 based gumstix ( with the r1161
>     toolchain ),
>     >> and have a dynamically linked executable ( which is linked with
>     several
>     >> of our own libraries as well as system libraries. )  If we store
>     this in
>     >> nfs it starts fine, however if we put it in the jffs2 filesystem on
>     >> flash, then it fails with Segmentation Fault, the first 2 Times
>     we run
>     >> it, and then works fine on the 3rd and all subsequent runs, until the
>     >> gumstix is rebooted.
>     >> If we run this in strace from flash then it also works fine, and
>     if we
>     >> run it in gdb then we don't get a useful stack trace from it, but it
>     >> fails before breaking on main().
>     >
>     > I've run into that before, when running Squeak. What we did to fix
>     it was
>     > basically pre-cache the shared libraries before running the
>     executable...
>     >
>     > cp vm-display-null /dev/null
>     > cp vm-sound-null /dev/null
>     > ./squeak -vm-display-null -headless gumstix.image -$1
>     >
>     > That fixes the problem for me...
>
>     Thanks Jon, cat to dev null of our libraries works too.
>
>     Does anyone know what the cause of this problem is?
>
>     Thanks
>     Philip
>
>     >
>     > Later,
>     > Jon
>     >
>     > --------------------------------------------------------------
>     >    Jon Hylands      [hidden email]
>     <mailto:[hidden email]>      http://www.huv.com/jon
>     <http://www.huv.com/jon>
>     >
>     >   Project: Micro Raptor (Small Biped Velociraptor Robot)
>     >            http://www.huv.com/blog
>     >
>     >
>     -------------------------------------------------------------------------
>
>     > This SF.net email is sponsored by: Splunk Inc.
>     > Still grepping through log files to find problems?  Stop.
>     > Now Search log events and configuration files using AJAX and a
>     browser.
>     > Download your FREE copy of Splunk now >> http://get.splunk.com/
>     > _______________________________________________
>     > gumstix-users mailing list
>     > [hidden email]
>     <mailto:[hidden email]>
>     > https://lists.sourceforge.net/lists/listinfo/gumstix-users
>
>
>
>     -------------------------------------------------------------------------
>
>     This SF.net email is sponsored by: Splunk Inc.
>     Still grepping through log files to find problems?  Stop.
>     Now Search log events and configuration files using AJAX and a browser.
>     Download your FREE copy of Splunk now >> http://get.splunk.com/
>     _______________________________________________
>     gumstix-users mailing list
>     [hidden email]
>     <mailto:[hidden email]>
>     https://lists.sourceforge.net/lists/listinfo/gumstix-users
>
>
>
> ------------------------------------------------------------------------
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >> http://get.splunk.com/
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> gumstix-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gumstix-users



-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Seg Fault on first 2 runs of dynamically linked executables.

Black, Michael-2
Sounds a lot like asynchronous caching to me.  Perhaps mmap isn't
ensuring the action is totally complete before returning.
Seems that jffs2 has to copy the flash to RAM and apparently is
returning before that occurs (or is returning the wrong pointer??)
http://sourceware.org/jffs2/jffs2-html/node4.html

One possible test may be to try opening the library first thing in your
code and do an "fsync" on the file descriptor.
Or do a sleep() first thing.

I can imagine the difference for the other binaries could be the size of
the libraries since the problem here looks like much larger libraries
are segfaulting??
___________________
Michael D. Black
[hidden email]

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of pac
Sent: Friday, October 05, 2007 6:16 AM
To: General mailing list for gumstix users.
Subject: Re: [Gumstix-users] Seg Fault on first 2 runs of dynamically
linked executables.

Scott Kroll wrote:
> I'm curious to find out if this is the cause of my problems as well.
I
> posted a thread earlier today about this.  I'll look into it when I
make
> it home.

The strange thing is that the binaries in the gumstix release which are
also build shared work fine.

Does anyone know if there is some special procedure/compiler/linker
flags which need to the used to prepare the libraries for use from a
jffs2 filesystem?

I'm guessing this is some sort of issue with mmap io, which is probably
handled specially for jffs2 on flash on the system memory bus?

Thanks
Philip

>
> On 10/4/07, * pac* <[hidden email] <mailto:[hidden email]>>
wrote:

>
>     Jon Hylands wrote:
>     > On Thu, 04 Oct 2007 16:14:11 +0100, "pac" <[hidden email]
>     <mailto:[hidden email]>> wrote:
>     >
>     >>       We're using a pxa255 based gumstix ( with the r1161
>     toolchain ),
>     >> and have a dynamically linked executable ( which is linked with
>     several
>     >> of our own libraries as well as system libraries. )  If we
store
>     this in
>     >> nfs it starts fine, however if we put it in the jffs2
filesystem on
>     >> flash, then it fails with Segmentation Fault, the first 2 Times
>     we run
>     >> it, and then works fine on the 3rd and all subsequent runs,
until the
>     >> gumstix is rebooted.
>     >> If we run this in strace from flash then it also works fine,
and
>     if we
>     >> run it in gdb then we don't get a useful stack trace from it,
but it
>     >> fails before breaking on main().
>     >
>     > I've run into that before, when running Squeak. What we did to
fix

>     it was
>     > basically pre-cache the shared libraries before running the
>     executable...
>     >
>     > cp vm-display-null /dev/null
>     > cp vm-sound-null /dev/null
>     > ./squeak -vm-display-null -headless gumstix.image -$1
>     >
>     > That fixes the problem for me...
>
>     Thanks Jon, cat to dev null of our libraries works too.
>
>     Does anyone know what the cause of this problem is?
>
>     Thanks
>     Philip
>
>     >
>     > Later,
>     > Jon
>     >
>     > --------------------------------------------------------------
>     >    Jon Hylands      [hidden email]
>     <mailto:[hidden email]>      http://www.huv.com/jon
>     <http://www.huv.com/jon>
>     >
>     >   Project: Micro Raptor (Small Biped Velociraptor Robot)
>     >            http://www.huv.com/blog
>     >
>     >
>
------------------------------------------------------------------------
-

>
>     > This SF.net email is sponsored by: Splunk Inc.
>     > Still grepping through log files to find problems?  Stop.
>     > Now Search log events and configuration files using AJAX and a
>     browser.
>     > Download your FREE copy of Splunk now >> http://get.splunk.com/
>     > _______________________________________________
>     > gumstix-users mailing list
>     > [hidden email]
>     <mailto:[hidden email]>
>     > https://lists.sourceforge.net/lists/listinfo/gumstix-users
>
>
>
>
------------------------------------------------------------------------
-
>
>     This SF.net email is sponsored by: Splunk Inc.
>     Still grepping through log files to find problems?  Stop.
>     Now Search log events and configuration files using AJAX and a
browser.

>     Download your FREE copy of Splunk now >> http://get.splunk.com/
>     _______________________________________________
>     gumstix-users mailing list
>     [hidden email]
>     <mailto:[hidden email]>
>     https://lists.sourceforge.net/lists/listinfo/gumstix-users
>
>
>
>
------------------------------------------------------------------------
>
>
------------------------------------------------------------------------
-
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a
browser.
> Download your FREE copy of Splunk now >> http://get.splunk.com/
>
>
>
------------------------------------------------------------------------
>
> _______________________________________________
> gumstix-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gumstix-users



------------------------------------------------------------------------
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
 
 
 
 
 
 
 
This electronic message and any files transmitted with it contain information which may be privileged and/or proprietary. The information is intended for use solely by the intended recipient(s). If you are not the intended recipient, be aware that any disclosure, copying, distribution or use of this information is prohibited. If you have received this electronic message in error, please advise the sender by reply email or by telephone (301-939-7000) and delete the message.

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Seg Fault on first 2 runs of dynamically linked executables.

Jon Hylands
On Fri, 5 Oct 2007 07:41:38 -0400, "Black, Michael"
<[hidden email]> wrote:

> I can imagine the difference for the other binaries could be the size of
> the libraries since the problem here looks like much larger libraries
> are segfaulting??

The two libraries I use are 8K and 6K. The problem only occurs sometimes,
which suggests a timing issue as you speculated.

Later,
Jon

--------------------------------------------------------------
   Jon Hylands      [hidden email]      http://www.huv.com/jon

  Project: Micro Raptor (Small Biped Velociraptor Robot)
           http://www.huv.com/blog

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Seg Fault on first 2 runs of dynamically linked executables.

Dave Hylands
Hi,

> > I can imagine the difference for the other binaries could be the size of
> > the libraries since the problem here looks like much larger libraries
> > are segfaulting??
>
> The two libraries I use are 8K and 6K. The problem only occurs sometimes,
> which suggests a timing issue as you speculated.

Historically, there have been quite a few problems on the ARM relating
to dynamic loading of libraries. I think most of the issues have been
addressed in recent releases of uClibc, but it's possible that some
haven't.

The fact that it seems to fail on jffs2 and not on nfs is quite a
useful clue. I think java/pyhton have been suffering from this problem
as well.

--
Dave Hylands
Vancouver, BC, Canada
http://www.DaveHylands.com/

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Seg Fault on first 2 runs of dynamically linked executables.

Scott Kroll
In reply to this post by pac-3
I'm actually having all kinds of trouble linking to libraries that were compiled outside of buildroot.  I've been using the sample Hello World Makefile and I still get a seg fault every time.  However, the sample programs the library compiled work just fine.

On 10/5/07, pac <[hidden email]> wrote:
Scott Kroll wrote:
> I'm curious to find out if this is the cause of my problems as well.  I
> posted a thread earlier today about this.  I'll look into it when I make
> it home.

The strange thing is that the binaries in the gumstix release which are
also build shared work fine.

Does anyone know if there is some special procedure/compiler/linker
flags which need to the used to prepare the libraries for use from a
jffs2 filesystem?

I'm guessing this is some sort of issue with mmap io, which is probably
handled specially for jffs2 on flash on the system memory bus?

Thanks
Philip

>
> On 10/4/07, * pac* <[hidden email] <mailto:[hidden email]>> wrote:
>
>     Jon Hylands wrote:
>     > On Thu, 04 Oct 2007 16:14:11 +0100, "pac" <[hidden email]
>     <mailto: [hidden email]>> wrote:
>     >
>     >>       We're using a pxa255 based gumstix ( with the r1161
>     toolchain ),
>     >> and have a dynamically linked executable ( which is linked with
>     several
>     >> of our own libraries as well as system libraries. )  If we store
>     this in
>     >> nfs it starts fine, however if we put it in the jffs2 filesystem on
>     >> flash, then it fails with Segmentation Fault, the first 2 Times
>     we run
>     >> it, and then works fine on the 3rd and all subsequent runs, until the
>     >> gumstix is rebooted.
>     >> If we run this in strace from flash then it also works fine, and
>     if we
>     >> run it in gdb then we don't get a useful stack trace from it, but it
>     >> fails before breaking on main().
>     >
>     > I've run into that before, when running Squeak. What we did to fix
>     it was
>     > basically pre-cache the shared libraries before running the
>     executable...
>     >
>     > cp vm-display-null /dev/null
>     > cp vm-sound-null /dev/null
>     > ./squeak -vm-display-null -headless gumstix.image -$1
>     >
>     > That fixes the problem for me...
>
>     Thanks Jon, cat to dev null of our libraries works too.
>
>     Does anyone know what the cause of this problem is?
>
>     Thanks
>     Philip
>
>     >
>     > Later,
>     > Jon
>     >
>     > --------------------------------------------------------------
>     >    Jon Hylands      [hidden email]
>     <mailto:[hidden email]>      http://www.huv.com/jon
>     <http://www.huv.com/jon>
>     >
>     >   Project: Micro Raptor (Small Biped Velociraptor Robot)
>     >             http://www.huv.com/blog
>     >
>     >
>     -------------------------------------------------------------------------
>
>     > This SF.net email is sponsored by: Splunk Inc.
>     > Still grepping through log files to find problems?  Stop.

>     > Now Search log events and configuration files using AJAX and a
>     browser.
>     > Download your FREE copy of Splunk now >> http://get.splunk.com/
>     > _______________________________________________
>     > gumstix-users mailing list
>     > [hidden email]
>     <mailto:[hidden email]>
>     > https://lists.sourceforge.net/lists/listinfo/gumstix-users
>
>
>
>     -------------------------------------------------------------------------
>
>     This SF.net email is sponsored by: Splunk Inc.
>     Still grepping through log files to find problems?  Stop.
>     Now Search log events and configuration files using AJAX and a browser.
>     Download your FREE copy of Splunk now >> http://get.splunk.com/
>     _______________________________________________
>     gumstix-users mailing list
>     [hidden email]
>     <mailto:[hidden email]>
>     https://lists.sourceforge.net/lists/listinfo/gumstix-users
>
>
>
> ------------------------------------------------------------------------
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >> http://get.splunk.com/
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> gumstix-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/gumstix-users



-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Website Problem

Craig Hughes
In reply to this post by Demetris Zavorotnitsienko
On Oct 4, 2007, at 11:22 PM, Demetris Zavorotnichenko wrote:

Guys. I still can’t buy anything through the website. I have even created a new account and still no go.

 

Please help me out here. I can’t get passed the “Delivery Information”.  The country is Cyprus. (EU)


Ah, I think that's not included in our list of shippable-to countries.  But it's also possible that it is in our list, but there's some issue with UPS issuing quotes to our webserver for the shipping options...

C

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Best Way for everything

Demetris Zavorotnitsienko
In reply to this post by Scott Kroll

Hi everyone.

 

I am making a UAV project and am going to use Gumstix for it.

 

I wanted to ask.

 

What would be the Best “Fastest” way of controlling 12-20 Servos, analog and Digital sensors with Gumstix?

 

I know that there are no analog inputs on Gumstix so that has to be through a separate MCU. But the servos. Could I control them fast enougth through a separate board connected (like Robostix) to it?

 

If so then what type of communication would be the best (fastest) to control all this ?

 

Also can I connect Servos directly to Gumstix and control them through the GPIO pins ? or it would be best to control then through another Board.

 

Tnx in advance.


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Best Way for everything

Jon Hylands
On Fri, 9 Nov 2007 18:53:00 +0200, "Demetris Zavorotnichenko"
<[hidden email]> wrote:

> I am making a UAV project and am going to use Gumstix for it.
>
> What would be the Best ?Fastest? way of controlling 12-20 Servos, analog and Digital sensors with Gumstix?

A UAV? You mean an unmanned aerial vehicle? What possible use could you
have for 20 servos on a UAV? Or are you thinking of something else?

Later,
Jon

--------------------------------------------------------------
   Jon Hylands      [hidden email]      http://www.huv.com/jon

  Project: Micro Raptor (Small Biped Velociraptor Robot)
           http://www.huv.com/blog

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Best Way for everything

Bill Gatliff
Jon Hylands wrote:

> On Fri, 9 Nov 2007 18:53:00 +0200, "Demetris Zavorotnichenko"
> <[hidden email]> wrote:
>
>  
>> I am making a UAV project and am going to use Gumstix for it.
>>
>> What would be the Best ?Fastest? way of controlling 12-20 Servos, analog and Digital sensors with Gumstix?
>>    
>
> A UAV? You mean an unmanned aerial vehicle? What possible use could you
> have for 20 servos on a UAV? Or are you thinking of something else?
>  

Unattended Audible Voltmeter.  But still doesn't explain the servos.  :)

Or maybe, Unhuman Automobile Valet?  With 20 DOF, now we're getting
somewhere!  :)


b.g.

--
Bill Gatliff
[hidden email]


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Best Way for everything

Demetris Zavorotnitsienko
In reply to this post by Jon Hylands
I have my reasons for 20 Servos.

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of Jon
Hylands
Sent: Friday, November 09, 2007 6:59 PM
To: General mailing list for gumstix users.
Subject: Re: [Gumstix-users] Best Way for everything

On Fri, 9 Nov 2007 18:53:00 +0200, "Demetris Zavorotnichenko"
<[hidden email]> wrote:

> I am making a UAV project and am going to use Gumstix for it.
>
> What would be the Best ?Fastest? way of controlling 12-20 Servos, analog
and Digital sensors with Gumstix?

A UAV? You mean an unmanned aerial vehicle? What possible use could you
have for 20 servos on a UAV? Or are you thinking of something else?

Later,
Jon

--------------------------------------------------------------
   Jon Hylands      [hidden email]      http://www.huv.com/jon

  Project: Micro Raptor (Small Biped Velociraptor Robot)
           http://www.huv.com/blog

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Best Way for everything

Demetris Zavorotnitsienko
In reply to this post by Bill Gatliff
Unmanned Aerial Vehicle

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of Bill
Gatliff
Sent: Friday, November 09, 2007 7:02 PM
To: General mailing list for gumstix users.
Subject: Re: [Gumstix-users] Best Way for everything

Jon Hylands wrote:
> On Fri, 9 Nov 2007 18:53:00 +0200, "Demetris Zavorotnichenko"
> <[hidden email]> wrote:
>
>  
>> I am making a UAV project and am going to use Gumstix for it.
>>
>> What would be the Best ?Fastest? way of controlling 12-20 Servos, analog
and Digital sensors with Gumstix?
>>    
>
> A UAV? You mean an unmanned aerial vehicle? What possible use could you
> have for 20 servos on a UAV? Or are you thinking of something else?
>  

Unattended Audible Voltmeter.  But still doesn't explain the servos.  :)

Or maybe, Unhuman Automobile Valet?  With 20 DOF, now we're getting
somewhere!  :)


b.g.

--
Bill Gatliff
[hidden email]


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Best Way for everything

Bugzilla from jgorkos@wildcatwireless.net
In reply to this post by Demetris Zavorotnitsienko
12-20 servos sounds like your project is out of control, but that's your deal.

this is what I'm using, with a 600MHz verdex.  I connect it to /dev/ttyS1
http://www.pololu.com/products/pololu/0207/

you can slave several of them together.  

They also sell these:
http://www.pololu.com/products/pololu/0240/

but it's larger.  Of course, if you have an airframe that needs/has 20 servos,
I'm thinking weight and size of the board are less of an issue.

John Gorkos


On Friday 09 November 2007 11:53:00 Demetris Zavorotnichenko wrote:

> Hi everyone.
>
>
>
> I am making a UAV project and am going to use Gumstix for it.
>
>
>
> I wanted to ask.
>
>
>
> What would be the Best "Fastest" way of controlling 12-20 Servos, analog
> and Digital sensors with Gumstix?
>
>
>
> I know that there are no analog inputs on Gumstix so that has to be through
> a separate MCU. But the servos. Could I control them fast enougth through a
> separate board connected (like Robostix) to it?
>
>
>
> If so then what type of communication would be the best (fastest) to
> control all this ?
>
>
>
> Also can I connect Servos directly to Gumstix and control them through the
> GPIO pins ? or it would be best to control then through another Board.
>
>
>
> Tnx in advance.



-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Best Way for everything

Demetris Zavorotnitsienko
<12-20 servos sounds like your project is out of control>

How many servos would you say that a reliable UAV should have ? (Helicopter)

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of John
Gorkos
Sent: Friday, November 09, 2007 7:06 PM
To: General mailing list for gumstix users.
Subject: Re: [Gumstix-users] Best Way for everything

12-20 servos sounds like your project is out of control, but that's your
deal.

this is what I'm using, with a 600MHz verdex.  I connect it to /dev/ttyS1
http://www.pololu.com/products/pololu/0207/

you can slave several of them together.  

They also sell these:
http://www.pololu.com/products/pololu/0240/

but it's larger.  Of course, if you have an airframe that needs/has 20
servos,
I'm thinking weight and size of the board are less of an issue.

John Gorkos


On Friday 09 November 2007 11:53:00 Demetris Zavorotnichenko wrote:

> Hi everyone.
>
>
>
> I am making a UAV project and am going to use Gumstix for it.
>
>
>
> I wanted to ask.
>
>
>
> What would be the Best "Fastest" way of controlling 12-20 Servos, analog
> and Digital sensors with Gumstix?
>
>
>
> I know that there are no analog inputs on Gumstix so that has to be
through
> a separate MCU. But the servos. Could I control them fast enougth through
a

> separate board connected (like Robostix) to it?
>
>
>
> If so then what type of communication would be the best (fastest) to
> control all this ?
>
>
>
> Also can I connect Servos directly to Gumstix and control them through the
> GPIO pins ? or it would be best to control then through another Board.
>
>
>
> Tnx in advance.



-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Best Way for everything

Demetris Zavorotnitsienko
In reply to this post by Bugzilla from jgorkos@wildcatwireless.net
So what you are saying is that Serial is the way to go with this ?

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of John
Gorkos
Sent: Friday, November 09, 2007 7:06 PM
To: General mailing list for gumstix users.
Subject: Re: [Gumstix-users] Best Way for everything

12-20 servos sounds like your project is out of control, but that's your
deal.

this is what I'm using, with a 600MHz verdex.  I connect it to /dev/ttyS1
http://www.pololu.com/products/pololu/0207/

you can slave several of them together.  

They also sell these:
http://www.pololu.com/products/pololu/0240/

but it's larger.  Of course, if you have an airframe that needs/has 20
servos,
I'm thinking weight and size of the board are less of an issue.

John Gorkos


On Friday 09 November 2007 11:53:00 Demetris Zavorotnichenko wrote:

> Hi everyone.
>
>
>
> I am making a UAV project and am going to use Gumstix for it.
>
>
>
> I wanted to ask.
>
>
>
> What would be the Best "Fastest" way of controlling 12-20 Servos, analog
> and Digital sensors with Gumstix?
>
>
>
> I know that there are no analog inputs on Gumstix so that has to be
through
> a separate MCU. But the servos. Could I control them fast enougth through
a

> separate board connected (like Robostix) to it?
>
>
>
> If so then what type of communication would be the best (fastest) to
> control all this ?
>
>
>
> Also can I connect Servos directly to Gumstix and control them through the
> GPIO pins ? or it would be best to control then through another Board.
>
>
>
> Tnx in advance.



-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Best Way for everything

Jon Hylands
In reply to this post by Demetris Zavorotnitsienko
On Fri, 9 Nov 2007 19:04:31 +0200, "Demetris Zavorotnichenko"
<[hidden email]> wrote:

> I have my reasons for 20 Servos.

Building a flying transformer, perhaps?

In any case, what John said. If your UAV is going to have 20 servos, you're
looking at a big heavy platform, so you have a lot of freedom to choose. If
you want to use hobby R/C servos, something like the Pololu board he
mentioned, or an SSC-32 would do fine. I use the gumstix verdex with
Bioloid servos (AX-12), which use a custom serial bus that I interface to
through USB from the gumstix.

Later,
Jon

--------------------------------------------------------------
   Jon Hylands      [hidden email]      http://www.huv.com/jon

  Project: Micro Raptor (Small Biped Velociraptor Robot)
           http://www.huv.com/blog

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
123
Loading...