[Emc-developers] getting rid of deprecated m5i20 stuff

2010-02-20 Thread Jeff Epler
Are the following items any use once hal_m5i20 is removed (which I plan to do shortly on master)? Some of them appear related to the removed 7i43 gpio-only driver and the unfinished/removed m5i2x driver. bfload m5i20cfg pci_read pci_write Jeff ---

Re: [Emc-developers] getting rid of deprecated m5i20 stuff

2010-02-20 Thread seb
The hostmot2 stuff doesn't use any of that stuff. JUNK IT!!! -- Sebastian Kuzminsky never be discouraged just let your nerdy flourish -Original Message- From: Jeff Epler Subj: [Emc-developers] getting rid of deprecated m5i20 stuff Date: Sat 2010 Feb 20 10:59 Size: 766 bytes To: e

Re: [Emc-developers] getting rid of deprecated m5i20 stuff

2010-02-20 Thread Stephen Wille Padnos
I don't know how specific bfload is, but it probably won't be useful. Pci_read and pci_write are more generic and could be useful during development of other PCI drivers. - Steve Jeff Epler wrote: >Are the following items any use once hal_m5i20 is removed (which I plan >to do shortly on mast

Re: [Emc-developers] getting rid of deprecated m5i20 stuff

2010-02-20 Thread seb
bfload loads bitfile firmwares into Mesa AnyIO boards. That job is now done by the hostmot2 driver. I think the only way bfload could be useful is if someone wanted to write a driver for AnyIO boards with some firmware *other* than hm2, and in that case I think a better way to do it would be t

Re: [Emc-developers] getting rid of deprecated m5i20 stuff

2010-02-20 Thread John Kasunich
On Sat, 20 Feb 2010 12:48 -0700, s...@highlab.com wrote: > bfload loads bitfile firmwares into Mesa AnyIO boards. That job is now > done by the hostmot2 driver. > > I think the only way bfload could be useful is if someone wanted to write > a driver for AnyIO boards with some firmware *other* th

[Emc-developers] detecting the availability of RT patched kernal

2010-02-20 Thread Chris Morley
Gentlemen What is the best / standard way to test for the real time kernal? I want pncconf to pop a warning dialog when a user tries to use a tes,t if the realtime is not available, otherwise there is no way to know why the test fails. Thanks Chris M __

Re: [Emc-developers] detecting the availability of RT patched kernal

2010-02-20 Thread Jeff Epler
On Sat, Feb 20, 2010 at 11:31:28PM +, Chris Morley wrote: > What is the best / standard way to test for the real time kernal? That's a good question, and one that turns out didn't have a great answer. In parts of emc2 coded in c/c++ you check RTAPI_SIM. I just added to 2.4 and master the two

Re: [Emc-developers] detecting the availability of RT patched kernal

2010-02-20 Thread Chris Morley
> I just added to 2.4 and master the two attributes hal.is_sim and > hal.is_rt. On any system, exactly one of them will be true, and that > will tell you whether the version of emc you're using is compiled for > simulation only or has realtime support. > I believe this is all I will need - Than