Migrating to 64-bit is mandatory for taking advantage of newer platforms and more memory.
Red Hat Enterprise Linux 7 has only 64-bit kernel and does not have a 32-bit one.
Even RHEL 6 does not have a non-PAE kernel. On our newer computers, we cannot use NI DAQmx on Linux (with RHEL 6) because of physical memory being mapped beyond 4G address space.
Any development of the Linux drivers would be nice. They haven't touched it since 2011. Whenever I would ask, they always say "Linux is a priority for us". Bullsh*t. I had to switch my systems over to Windows years ago because NI hasn't kept up. Don't hold your breath waiting for this.
Seems that NI is in process of slow liquidation. Otherwise they would support 64bit machines and develop proper drivers (easily doable w/ budget less than 100k)
Existing customers should put older 32bit-capable machines on stock, in case they need replacements.
Linux Embedded / Kernel Hacker / BSP / Driver development / Systems engineering
Seems that NI is in process of slow liquidation. Otherwise they would support 64bit machines and develop proper drivers (easily doable w/ budget less than 100k)
Looks like they had a rough year last year. I wouldn't expect them to ever invest much money into the small Linux user base. A lot of their hardware is under-powered for the high price tag. The software is pretty bad too. Not many other companies making DAQ hardware these days though. It's a sad situation. I wish I could go back and redesign our test equipment without NI gear.
It would be great to develop software on 64bit Linux sytem using DAQmx.
Since we`re developing software for 64bit Linux this is a must for us - this means a 64bit Kernelmodule as well as 64bit libraries.
NI has a turnover of $1.29B.
$1M (less an a per mill) would be more than enough for a complete rewrite of all the drivers (incl. older devices).
Actually it's pretty simple:
* write IIO kernel drivers for the actual HW (most of the code would be shared) * add an generic IIO backend to DAQmx.
Problem: product management neither really recognizes the problem (no usable devices = no sales), nor wants to invest a few bucks in a proper solution.
Ergo: no market share in Linux world. Linux users won't buy their HW, as they cannot use it. So, budgeting based on market share automatically leads to zero investement, which leads to completely loosing the existing user base.
The core problem is the wrong decision criteria. It's not an technical problem - give me the full specs, testing hw and budget, and I'll fix that ...
Linux Embedded / Kernel Hacker / BSP / Driver development / Systems engineering
I think it's hilarious (not really) that the DAQmx driver download is over a gigabyte. A gigabyte for device drivers. They make Microsoft bloat look like child's play.
I have seen a roadmap from NI (under NDA) where they are working specifically on Linux drivers for their hardware. It has been a while since I saw it, but I think the DAQmx was supposed to be at least in beta now, possibly release at NI Week. I cannot say anything about it being 64-bit or 32-bit. I really wish NI would be at least a little more open about this kind of roadmap since it is affecting sales now.
There are only two ways to tell somebody thanks: Kudos and Marked Solutions Unofficial Forum Rules and Guidelines "Not that we are sufficient in ourselves to claim anything as coming from us, but our sufficiency is from God" - 2 Corinthians 3:5
Desktop Linux Drivers for DAQmx were released in 2018.