Host to Target DMA question

Hello Guys,
I have a problem regarding DMA Access from Host to Target. I access DMA at an interval of 1.7uS I get 12 U32 data from DMA every 1.7uS, problem is, I can see that DMA can take more time than the normal. Is there a more efficient way to get 12 U32 data at a time?
Thanks,

Hello Jeremy
I am using NI 7833R FPGA module. what I did was to place read DMAs in a sequence structure (1 in each frame so that makes 12 frames). then inside each frame, I placed an output port which I set to True then false then true agan while passing through the sequence structure.
When I run the program, I can see that it takes around 125nS for each DMA access, But there are times that it takes more time to complete 1 access which takes more than 125nS.Most of the time, The 12 U32 data takes 1.5uS to complete. But due to the unknown reason, it sometimes takes more time to access DMA.
Thanks

Similar Messages

DMA host to target synchronization

I am using two host to target DMAs to generate two arbitrary waveforms to be output on anAO module. I want these waveforms to be output at the same time, but it appears as if one of the waveforms leads the other by several micorseconds. The two waveforms are in two separate arrays, passed to the DMA. It appears as if the lag is on the host side. I've attempted using the timing loop with the synchronization vi and adjusting the offset of the timed loops, but that seems to increase the delay more, or have not effect. Is there a way to get these DMAs to start passing the data at the same time?
Solved!
Go to Solution.

Hi Peter,
Good question. I am thinking that this may be possible using occurrences. Just before we want to write to one of the fifo's set the occurance so that the second fifo could write as well. However this may still cause a lag. As of yet there is no way to purely trigger Fifos. If they are in the same loop they should happen relatively at the same time. I will try to use occurances tomorrow, time it and see if they come back synchronously.
National Instruments
RIO Embedded Hardware PSE
CompactRIO Developers Guide

How many $Target/Host or $Target/Host/Host ... etc do you need to go

Hello
Can someone please help me with the following question as I am learning authoring at the moment, thanks
I see from Brian videos that some times you need to use just $Target/Property... and other times $Target/Host/Property...
or even $Target/Host/Host/Property... and so on.
I get the idea (or at least I think I do), in as much as the further you are away from the Host (e.g. the deeper down embedded in the XML) you have to step back up using Host or Host/Host etc. a bit like when transversing a file system using ..\.. or ..\..\..
So for a registry discovery you would say $Target/Host/Property... e.g. use Host one time.
My question is.
is this always the case for a discovery?
How do I know when I should use $Target/Property and when to use $Target/Host and $Target/Host/Host etc.. is there any easy well to tell?
Thank you
AAnotherUser__
AAnotherUser__

Just a little modified previous script:
function Get-SCOMClassHostElement {
param(
[Microsoft.EnterpriseManagement.Configuration.ManagementPackClass]
[Parameter(Mandatory = $true)]
$ManagementClass,
[Switch]
[Parameter(Mandatory = $false)]
$KeyOnly
function Get-SCOMClassKeyProperty {
param(
$ManagementClass,
$MPElementPath
$ManagementClass.PropertyCollection | ? {$_.Key} | % {$MPElementPath + "Property[Type=`"" + $_.Identifier.Domain[0]+"!"+$ManagementClass.Name+"`"]/"+$_.Name+"`$"}
If ($ManagementClass.Hosted) {
Get-SCOMClassKeyProperty -ManagementClass $ManagementClass.FindHostClass() -MPElementPath ($MPElementPath + "Host/")
If ($ManagementClass.Base) {
Get-SCOMClassKeyProperty -ManagementClass (Get-SCOMClass $ManagementClass.Base.Id) -MPElementPath $MPElementPath
function Get-SCOMClassAllProperty {
param(
$ManagementClass,
$MPElementPath
$ManagementClass.PropertyCollection | % {$MPElementPath + "Property[Type=`"" + $_.Identifier.Domain[0]+"!"+$ManagementClass.Name+"`"]/"+$_.Name+"`$"}
If ($ManagementClass.Base) {
Get-SCOMClassAllProperty -ManagementClass (Get-SCOMClass $ManagementClass.Base.Id) -MPElementPath $MPElementPath
$MPElementPath = "`$Target/"
[Array]$ManagementClassProperty = Get-SCOMClassKeyProperty -ManagementClass $ManagementClass -MPElementPath $MPElementPath
If (!$KeyOnly) {
[Array]$ManagementClassProperty += Get-SCOMClassAllProperty -ManagementClass $ManagementClass -MPElementPath $MPElementPath
$ManagementClassProperty | Select-Object -Unique
New-SCOMManagementGroupConnection
$ManagementClassName = "Microsoft.SQLServer.Database"
$ManagementClass = Get-SCOMClass -Name $ManagementClassName
Get-SCOMClassHostElement -ManagementClass $ManagementClass
Vladimir Zelenov | http://systemcenter4all.wordpress.com

HOST to FPGA DMA Transfers

Hi,
We're having trouble using the Host to FPGA DMA feature available in
LabVIEW 8.20 with IFRIO. After starting a compile, we get the following
error message:
An internal software error has occurred.
Please contact National Instruments technical support at ni.com/support
with the following information:
Error -61048 occurred at This target does not support DMA Output (from the host to the target).
Possible reason(s):
LabVIEW FPGA: This target does not support DMA Output (from the host to the target).
Any help would be greatly appreciated.
Thanks.

Hi Manik:
We did not support DMA output on the PCI-5640R when we released NI-5640R 1.0. This is why you are getting the error message that you are seeing.
We plan to add support for DMA output in an upcoming release.
----abhay

How can I transfer more than 64 bit data from host to target?

Hi all, I'm currently using PCIe-7851r fpga card to drive my device. There were 64 lines to be controlled. So what I did is generating the commands on the host pc and then transfer it to the target via DMA FIFO. The data type of the FIFO is U64, i.e. every one digit controls 64 DIO lines. But the question becomes complex when I transfer 66 line command. I tried to create 2 FIFOs, but I can hardly make the 2 FIFOs synchronized.
I think I might be able to create 2 U64 arrays, one contains the original 64 line command,s and the other contains the 2 line info (a waste). And then I interleave them in the host and decimate them in the target. There should be enough cycles to to this. But I dont think this is a good solution. Is there any better method? Thank you.
LabVIEW 2009, Windows XP, PCIe-7851R
Regards,
Bo
My blog Let's LabVIEW.
Solved!
Go to Solution.

Using the techniques highlighted in this tutorial:
http://zone.ni.com/devzone/cda/tut/p/id/4534
You could use code like this:
Mark B
===If this fixes your problem, mark as solution!===

Host Name and IP Question

Sorry about this, I posted this on the Web Services forum, but nobody looks in there :(
Well, not strictly issues, and also maybe not strictly Java, but it does fall under web services. And the second question might be fixed with Java (probably HTML though!)
OK, enough dribble here is my question. I have set up my home PC as a web and FTP server running with XP IIS, there isn't much on it at the mo, but what I would like to know if it is at all possible (And I think I know the answer already) to change the host name/ip address.
By this I mean, I have an IP address which is provided by my broadband provider and although its DHCP, its static (I assume) because it never changes? and at the moment I can go anywhere in the world and type http://???.???.??.?? and I can see my little site working away. But I don't want to do that I want to type http://bestsexywebsiteintheworldever (or something to that affect)
So is it possible to do that, but without having to use one of these URL Forwarding Services.
Secondly, on my site I have some links that I am trying to keep a suprise for people when they click them, but the problem is when you hover over it, the address is shown on the bottom of the screen, as I'm sure you are all aware of. Is it possible to remove this so that when you hover no address is shown?
Thanks, sorry the first one is a bit long winded, I got carried away (very bored at work!)
MP.

Sorry about this, I posted this on the Web Services
forum, but nobody looks in there :(
Well, not strictly issues, and also maybe not strictly
Java, but it does fall under web services. And the
second question might be fixed with Java (probably
HTML though!)
OK, enough dribble here is my question. I have set up
my home PC as a web and FTP server running with XP
IIS, there isn't much on it at the mo, but what I
would like to know if it is at all possible (And I
think I know the answer already) to change the host
name/ip address.
By this I mean, I have an IP address which is provided
by my broadband provider and although its DHCP, its
static (I assume) because it never changes? and at the
moment I can go anywhere in the world and type
http://???.???.??.?? and I can see my little site
working away. But I don't want to do that I want to
type http://bestsexywebsiteintheworldever (or
something to that affect)
So is it possible to do that, but without having to
use one of these URL Forwarding Services.
One way to do this is to purchase a fully qualified Domain Name, then have your ISP point their DNS server to that. They usually want some $$ for that service.
I don't know about the second question, sorry.

Unable to detect hosts or targets

Hi,
I'm having a very basic problem.
I have installed Oracle Grid control for 10.1.0.3.0 on Win XP and have a 10.2.0.1 database installed on the same box. It discovered the local database.
Now I have few solaris boxes which run 10.1.0.3.0 and 10.2.0.1.0 databases. I have started "emctl start agent" in both boxes. But when I try to search for the hosts, grid control comes up with NO targets found. I cannot discover these hosts and databases.
do I have to install any specific things on top of databases in unix boxes?
Your help is very much appreciated.
Cheers
Ivan

If you havent installed Oracle Grid Agents on the other boxes then all you did was start the local DB Control Agents. This wont work. You need to deploy Grid Agents. "Grid Software".

AS3 newbe target my_mc question

Good day to all of you. I have a basic question.
I have movC_mc inside movB_mc inside movA_mc
AS2 to access movC_mc from the my Actions layer in frame 1 in the root would be
     movA_mc.movB_mc.movC_mc
My question is how can I target the movA_mc from the root of my .fla
This is what I got so far
movA_mc.addEventListener(MouseEvent.ROLL_OVER, a_player);
function a_player(e:MouseEvent):void{
    Object(this).movA_mc.movB_mc.movC_mc.x = 250
I am sure this is very simple, but just can't get my head to transition from AS2 to AS3.
Thanks again

if all 3 movieclips were created in the authoring environment use:
movA_mc.addEventListener(MouseEvent.ROLL_OVER, a_player);
function a_player(e:MouseEvent):void{
    movA_mc.movB_mc.movC_mc.x = 250
if any were created with code, referencing mov_mc may be different.

Won't start up after using as host for target firewire...help?!

I used my iMac G4 1.25ghz flatpanel to install Tiger to my G3 iMac via Firewire Target mode, and now the G4 won't start up/has no power.
Because the G4 wouldn't start up in Target Firewire for some reason, i used the G4 as the host and started up the G3 in Target, restarted the G4 from the DVD, and installed Tiger to the G3 (i am positive i used the correct destination disk, adn the G3 is now showing the Welcome screen so it was sucessful).
Afer installing, the G4 automatically restarted on the Welcome screen- apparently normal in this situation, from what i read- and i turned it off via its power button because i want to do the Welcome stuff from the G3 itself (was that a mistake?). I'm thinking i might have only put it to sleep, tho, before i pulled the Firewire cable.. i didn't hold the button down long.
Afer i pulled the Firewire cable, the G4 wouldn't start up again via the Power button. I pulled the power cable and plugged it in again, but still absolutely nothing- obviously no power. No response to trying to reset PRAM, start it up from an external drive, via firewire, or CD- it's like it's unplugged.
What could have happened? Would pulling the Firewire cable have messed up PMU if it was only asleep, not shut down? Could i have damaged the power cable when i pulled it?
Could having used it to install to another drive in Target mode have screwed it up (i've read of many people doing this without problem).?
Should i try pushing the PMU button, or are there other things i can try first? (i can borrow a power cable from my son's G4 iMac)
I'd sure like to get it booted in some way or other ASAP to make sure i didn't mess up the G4's hard drive somehow...

thanks, i'm going to read and bookmark that for future ref... i had a post not post, tho, that said: tempest in a tea-pot, it was just something with the power supply. As soon as i tried a cord from a friend's a few hours later, it powered up. Oddly enuf, when i then reattached the original power cord, it powered up from that, too.
I wonder if these have some kind of time-delayed automatic reset on the firewire or power supply or pmu, because NOTHING i did worked right after it happened.

Labview 8.0 FPGA DMA question

I am developing a control system which needs to access a big mass of
variable data (32 Mbyte) that i can put into the host computer RAM. In
labview tutorial and also on other web resources it is well described
how the DMA FIFO can be used to "write to" host PC memory (RAM) but
nothing about "reading from" the PC RAM. Is it possible? And which is
the control to be used? Do I need some lower level coding?
Thanx
Davide

Hi Davide,
As of right now, the DMA transfers are one-way only (FPGA to host).
Best regards,
David H.
Systems Engineer
National Instruments

About the hosts file (syntax/usage question)

Hi,
I am trying to block a handful of different websites that I do not want my daughter to use right now. I would like to use the hosts file to do this. Although I have basic familiarity with the hosts file and editing it, I am not completely certain as to how it works and how DNS works. If I want to, for example, block "yahoo.com" and all of its subdomains, would the following be an acceptable/valid entry?
127.0.0.1 *.yahoo.com
By not listing the subdomains of "yahoo.com", would that make it any liklier for those webpages to get through? If I were to also list out all of the "yahoo.com" subdomains that I have knowledge of along with the previous entry, would that cause any problems?
And the second part of my question is: Is the hosts file ever not top priority in lookup (on OS X 10.6.8, if it matters); and if so, is there a good way to make it always first?
Any help would be appreciated!
Thank you,
Michael

Use parental controls, maybe Dnsmasq, or a nanny-capable firewall-gateway-router box that's been configured on the edge of your network.
The /etc/hosts file also has specific sequence to get it reloaded; that file is read once and subsequenty cached. IIRC, you need to either /sudo dscacheutil -flushcache/ or /sudo killall -HUP mDNSResponder/ the DNS resolver.
If your child is technically inclined, or knows how to do basic internet research, or has discussions with the more nerd-ish among her peers, your child can learn how to bypass /etc/hosts and Dnsmasq using her own DNS translations, or with her own DNS services. That possibility would point toward either parental controls (easy), or the firewall-gateway-router box (added cost, but can block some or all systems on your local network).
Here's how easy it is to bypass the /etc/hosts block; launch Terminal.app and issue the equivalent of:
dig +short @8.8.8.8 host.example.com
for the host.example.com host that you're (she's) interested in. (There are other commands that can be used; more than just dig can be used here, and there are web sites that can perform this translation for you.) Then connect to the IP address that this command emits. Your /etc/hosts block will not prevent this, and will not block this.
If your daughter is somewhat more technically savvy, a VPN will bypass the /etc/hosts blocks. She'll need access to a VPN server, and those are available from various free and very low-cost sources...

Debugging applications on host and target simultaneously

I have an application running on LabVIEW-RT and another, which communicates whith the first one, running on LabVIEW.
Is there a way, on the same PC, to run an application on the host system and to debug an another application running on the target system ? The must will be to debug the two applications simultaneously.
Thanks for any input.
Hubert Robitaille

Hi Hubert,
You probably will not want to hear this but,
You cannot have a window open to the RT system while running a different LV app on the host.
When I have needed to do this, I have cabled up another PC with LV-RT to monitor the RT system. Een doing this can be a little awkward because RT wants to re-link and reload when you switch to the taget RT system.
I got around this by mapping the drive where the RT app is stored on the the windows system to the "extra" PC. Even after doing this, make sure you do clean connects and disconnects (to from the RT system), otherwise it seems that the main RT app will close but leave al of the sub-VI's running!
I have used other techniques as well, but they may be to specific to comment on now.
What kind of thing
are you trying to debug that requires you have both systems going at the same time?
Ben
Ben Rayner
I am currently active on.. MainStream Preppers
Rayner's Ridge is under construction

Transfer ID - Target Installation Question

Hello,
I need to migrate some OES2SP1 Linux VMs over to OES2SP2 Linux VMs on a
different XEN host. I'm going to use the Transfer ID method. I am
familiar with the Netware migration procedures where you install the
target server in its own temporary tree using the pre-migration setup.
For Linux Transfer ID, what is the best practice? It seems reading the
documentation that the new server should be installed in the production
Tree... then objects would need cleanup from the original target server
name. Is it best to install the target server in its own OU to
facilitate cleanup or should the target be installed in the same OU as
the source server?
Thanks for your help!
~Dave

Thanks for the info... using consolidation before Transfer ID is a good
option.
When I run the miggui on an OES2SP2 target server looking at an OES2SP1
(both linux) source server, I only see the File System and iPrint to
choose for the services. On the source and destination I also have DNS
and DHCP I would like to migrate, but these do not show as a choice. I
did install these on the target server per instructions (but they are
not running there)
Is this normal to see in a linux to linux migration? In other words, the
DNS service will "see" its configuration AFTER the TransferID takes
place since it will be looking at the correct server object?
kjhurni wrote:
> You really have two choices, IMO, if you are doing a data transfer along
> with Transfer ID.
>
> You can:
> 1) Create a new project in the miggui of type: Consolidation (in that
> pulldown) and then ADD file system and migrate the file system that way.
> That's what we did because we had such large amounts of data, we had to
> break it up. THEN, when you're ready, create ANOTHER project of type:
> ID Transfer and don't add any services and just transfer the ID.
>
> or
>
> 2) Create a project of type: Transfer ID, ADD the File System
> service, and run it. It will run the services portions first, and THEN
> it'll put you into the Transfer ID portion.
>
> Either way, you can always SAVE the project, and exit it before the
> Transfer ID (ie, you did the file transfer on Day 1, and got the bulk of
> the data, and on Day 2, you want to run it again via the "sync" to get
> whatever was missed/updated, and then finish up with the TransferID).
>
>

I dont see the icon on my host using target mode

To start with, my knowledge on this is limited. I like to use my Mac mini [end 2012 - 10.8.2] as my backup for my Imac [late 2009 - 10.7.5] and to use it at my workplace somewhere else. I'd like that the Mac mini synchronises automatically with my Imac again when connecting.
Here's what I did so far. Imac if off. Thunderbolt cable connects them both. Mini switched on, in system preferences/target disk selected as target disk and restarted. Mini monitor shows grey and a "firewire' logo.
Imac is switched on, according to some support sites, an icon should appear now to drag/drop files between the 2. This icon doesnt show on my Imac screen, is this due to 2 different OS versions? Can this drag/drop be done automaticallily by syncing somehow?
Peter

Target disk mode is mainly for transferring files on a limited basis. I think the iCloud would work better for you.
http://www.apple.com/search/?q=icloud&sec=support

DMA question

I've read Writing Device Drivers, but I've still got a few holes in my understanding of the dma routines.
I'm developing a driver for a PCI device that requires a very high sustained data rate, so I'm looking for every way possible to improve throughput. The device will be on a dedicated machine, so taking resources away from other processes is not an issue. We will have a pool of large (~64 MB) buffers in user space and a user process will notify the driver when the contents of a buffer need to be transferred to the device. The device will perform DMA to transfer the buffer contents to the card.
I would like to be able to allocate a few DMA handles (using ddi_dma_alloc_handle) in the attach entry point and use these handles to transfer the large buffers. Will I be able to allocate handles for such large DMA objects?
When initiating the transfer, I would get an unused pre-allocated DMA handle (via ddi_dma_addr_handle ?) and program the DMA engine on the device. Does this copy the contents of the buffer from user space to kernel space? If it does, is there a way to avoid this copy?
Thanks in advance for any help you can provide.

Oops. Just caught I typo in my previous message. I was planning on using ddi_dma_addr_bind_handle to get an unused pre-allocated DMA handle. Or would I want to use ddi_dma_buf_bind_handle? When is each one appropriate?

Host to Target DMA question

Similar Messages

Maybe you are looking for