mxbaseconfig not working and lsdaq doesn't show any device [Scientific Linux 6.3]

[vquicksilver]

Hi guys, and first thanks for your support.

I'm trying to use a 6221 PCI card with NIDAQmxBase software under scientific linux 6.3, but after a successful instalation I can't view any device when using lsdaq.
Also when I try to launch mxbaseconfig I get the following errors:

The same problem happens with nidatalogger:

My cpu in fact has SSE support, but not SSE2. Is SSE2 a requirement for running NIDAQmxBase? or I'm dealing with another kind of problem? Thanks. This is the output of /proc/cpuinfo:

processor	: 0
vendor_id	: AuthenticAMD
cpu family	: 6
model		: 8
model name	: AMD Sempron(tm)   2400+
stepping	: 1
cpu MHz		: 1666.412
cache size	: 256 KB
fdiv_bug	: no
hlt_bug		: no
f00f_bug	: no
coma_bug	: no
fpu		: yes
fpu_exception	: yes
cpuid level	: 1
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic mtrr pge mca cmov pat pse36 mmx fxsr sse syscall mp mmxext 3dnowext 3dnow up
bogomips	: 3332.82
clflush size	: 32
cache_alignment	: 32
address sizes	: 34 bits physical, 32 bits virtual

power management: ts

I attach the output of niSystemReport.

[vquicksilver]

Also note that under this environment, the CreateTask function doesn't give any errors, even with a NULL TaskHandle, I think this is a bug if the task can't be created your software should return an error.

[vquicksilver]

I uninstalled NIDAQmxBase-3.6.0 and installed DAQmx-8.0.2, now I can access the DAQ card on Dev1. My software can deal with both NIDAQmx and NIDAQmxBase because I wrote it to use only common functions between the 2 libraries so I can use this as a workarround for the moment, anyway if SSE2 is a requirement for NIDAQmxBase it should be added to the requisites, I can't see any mention to it in the README.

[JoeFriedchicken]

vquicksilver wrote:

I uninstalled NIDAQmxBase-3.6.0 and installed DAQmx-8.0.2, now I can access the DAQ card on Dev1. My software can deal with both NIDAQmx and NIDAQmxBase because I wrote it to use only common functions between the 2 libraries so I can use this as a workarround for the moment, anyway if SSE2 is a requirement for NIDAQmxBase it should be added to the requisites, I can't see any mention to it in the README.

You've identified a regression in NI-DAQmx Base: version 3.5 did not require the SSE2 registers, but version 3.6 does. In addition, the README does not call out this change. I've created a report for this so that we can identify the best way to fix it. Look for issue ID 410911 in the README file in a future release -- this will identify when the regression has been remedied.

As an aside, for PCI and PCIe devices, I strongly recommend NI-DAQ 8.0.2. While it takes more effort to install on newer Linux distributions and does not support USB DAQ devices, it has better performance and more features. I would be interested to understand your use-case more closely -- what factors contributed to using the common subset between NI-DAQmx Base and NI-DAQmx?

[vquicksilver]

I wrote my application for supporting multiple data sources and to be multiplatform compatible (Cpu and OS) from the start. Some data sources can work on multiple operating systems (For example the NIDAQ source) and others can only run on one operating system (For example the COMEDI data source can only run on Linux). This is the first application that I write that uses the NIDAQ C API, and at first I tought than NIDAQmxBase was the only supported API on Linux (Until yesterday I tought that DAQmx-8.0.2 was the "deprecated" driver), also when studying the API of NIDAQmx on Windows I saw that all the functions in NIDAQmxBase can be used on NIDAQmx (Taking care of some parameters and of the name) so I did my own wrapper to call one or another library, without beign bound to only one of the libraries.

Also I don't have very specific needs on my application, I read analog samples from the card, at the rate indicated by the user in the channels indicated by the user, maybe with more exotic needs this abstraction uppon the 2 librares wouldn't be valid. If you are interested in more details just ask, and thanks again for your support.