DFU for X86

[hein]

I just talked to Jur.

While has has no interest/time to port DFU himself, he is the sole owner of the sources and had no objections to it being ported.

He expects to be making the sources available as freeware in the next few days.

Someone can pick it up and port it, and I wouldn't be surprised if Jur is willing to put the X86 executable, maybe not as install kit, with his other stuff but we did not talk about that.

Whether VSI is ever interested in picking up the sources, or at least ship the executable is a whole other question.

Hein.

[arne_v]

That is very good news.

And I assume that building the sources on VMS x86-64 should not be a major problem. The source must be extremely coupled with ODS-2 and ODS-5, but I can't see any reason for Alpha/Itanium/x86-64 specific code.

[imiller]

As I've been looking at file system data recently and using DFU REPORT to check my work then I look forward to seeing the DFU sources to see exactly what it does.

[saprykin]

Is there anything like DFU for X86? It is a pain to deal with code repositories which include files with special characters in their names. On Alpha and IA64, DFU nicely handles these cases and allows to remove directories full of such files. However, I did not find DFU port for x86.

DELETE/TREE works ok for me, though it's a lot slower than DFU's version.

For some reason, DELETE/TREE does not work well for me when there are files with some special characters in their name, even if I enable extended parsing. But DFU somehow handles such cases perfectly fine.

Great to hear that source codes are going to be available, home someone can port it then

[lastovica@sciinc.com]

Current intent is SCI will provide an updated/fixed/enhanced kit of DFU for X86, I64, and Alpha. I'm guessing that minimum versions will be X86 VSI 9.2-2, I64 VSI 8.4, and Alpha VSI 8.4 at least initially. This should bridge the gap.

SCI will share status updates as they occur.

[arne_v]

Will the SCI kit contain source or be binaries only?

[lastovica@sciinc.com]

the SCI kit will mimic the currently available kit and be executable and help library.

[arne_v]

Which raises the question if the author will release the source code to other as indicated above.

[hein]

>>> Which raises the question if the author will release the source code to other as indicated above.

It actually PROVES that Jur has provided the source code to others, notably Norm, who has made a bunch of enhancements and bug fixes over the years and we wants to provide some more community support through SCI in this area.

I expect that the source code will be made available to all, but do not know how and when.
The general user population will just want some executables to run.
A few folks, perhaps yourself, will want the sources out of curiosity or perhaps to add specific options.

Hein,

[oknop]

Ton Dorland (the original DFU author, ex DEC) had two versions. One internal use only. This version also had disk defrag functions.

If I remember correctly both internal and public versions were compiled from the same source using a compile time define to select which version to build.

I wonder if the sources Jur has also contain the internal version parts of this tool.

Oswald