Windows 7 64-bit issues
Dennis
I am using the VentureForth compiler that is built with SwiftForth and
it talks to the FORTHdrive using SCSI Pass Through. To do that, it
opens the USB drive ("\\.\x:" where x is replaced with a drive
letter), but on Win7 (mine is 64-bits) the R/W mode fails. I can use
R/O at this stage and it finds the drive, but the DeviceIoControl
calls later fail, because they require write access.
Anyone have any clue where I should start looking on MSDN to find a
solution?
I just asked this on sftalk, but then I remembered that Stephen Pelc
wrote this code, and maybe he, or one of his customers has run into
this.
DaR
Marcel Hendrix
> Has anyone seen this on Windows 7 yet?
> I am using the VentureForth compiler that is built with SwiftForth and
> it talks to the FORTHdrive using SCSI Pass Through. To do that, it
> opens the USB drive ("\\.\x:" where x is replaced with a drive
> letter), but on Win7 (mine is 64-bits) the R/W mode fails. I can use
> R/O at this stage and it finds the drive, but the DeviceIoControl
> calls later fail, because they require write access.
> Anyone have any clue where I should start looking on MSDN to find a
> solution?
Post a short piece of code that would work with a memory stick, and
I can test it for you (What is supposed to happen when you open
a drive?)
-marcel
Dennis
> Dennis <druf...@worldnet.att.net> writes Windows 7 64-bit issues
I will have to think about that, to see if there is something to test,
independent of the SPT capabilities, and ideally, independent of Forth
and language issues. The MSDN articles don't talk about SPT since
Win2K, so I'm going to have to find the DDK sample and update it
myself if I want to get Microsoft's help. Ugg. ;(
DaR
Dennis
So, I went back and checked my results again, and it turns out I had
already fixed the problem. Not knowingly, but it appears as if the
changes I had made to make gforth work in cygwin also solve this
problem, as long as I also run the programs with administrator
privileges. Basically, the difference is that rather than opening the
drives using "\\.\x:" if I use "\\.\PhysicalDrive0" then both gforth
and the SwiftForth version work fine. Setting the administrator
privileges is a pain, but worth it, I guess, if I can continue working
on my vf-plugins on this new computer.
Back to our regularly scheduled programs... ;)
DaR
Albert van der Horst
Dennis <dru...@worldnet.att.net> wrote:
>Has anyone seen this on Windows 7 yet?
>
>I am using the VentureForth compiler that is built with SwiftForth and
>it talks to the FORTHdrive using SCSI Pass Through. To do that, it
>opens the USB drive ("\\.\x:" where x is replaced with a drive
>letter), but on Win7 (mine is 64-bits) the R/W mode fails. I can use
>R/O at this stage and it finds the drive, but the DeviceIoControl
>calls later fail, because they require write access.
>
>Anyone have any clue where I should start looking on MSDN to find a
>solution?
You should start looking at ways to configure external drives.
It is typically a safety feature. So I'm quite sure you can turn
R/W mode on. (On a Unix system, it would mean changing one letter
in /etc/fstab. On Windows 7, I don't know, but at most 10 clicks
away.)
>
>DaR
Groetjes Albert
--
--
Albert van der Horst, UTRECHT,THE NETHERLANDS
Economic growth -- being exponential -- ultimately falters.
albert@spe&ar&c.xs4all.nl &=n http://home.hccnet.nl/a.w.m.van.der.horst
Dennis
wrote:
> You should start looking at ways to configure external drives.
> It is typically a safety feature. So I'm quite sure you can turn
> R/W mode on. (On a Unix system, it would mean changing one letter
> in /etc/fstab. On Windows 7, I don't know, but at most 10 clicks
> away.)
Possibly not related, but I did turn on the ReadyBoost feature just
before I tried my code again. I can turn it off and my code still
works, so there's no way to tell if it turn something on that was not
on before. However, I suspect that it was the name change that made
the most difference. Using 14 characters rather than 2, so your guess
of 10 isn't far off. ;)
DaR
Stephen Pelc
<dru...@worldnet.att.net> wrote:
>Back to our regularly scheduled programs... ;)
This may also be useful:
http://www.ureader.com/message/694393.aspx
MPE wrote the original SCSI driver under contract to IntellaSys,
which includes NDAs. At least one VentureForth release included
our original source file with the documentation intact. Other
releases included derived code with most of the documentation
removed.
The bit-twiddling code for talking to the SEAforth chips is
based on proprietary code from the company that makes the
USB interface.
Stephen
--
Stephen Pelc, steph...@mpeforth.com
MicroProcessor Engineering Ltd - More Real, Less Time
133 Hill Lane, Southampton SO15 5AF, England
tel: +44 (0)23 8063 1441, fax: +44 (0)23 8033 9691
web: http://www.mpeforth.com - free VFX Forth downloads
Dennis
> On Thu, 31 Dec 2009 12:18:26 -0800 (PST), Dennis
>
> >Back to our regularly scheduled programs... ;)
>
> This may also be useful:
> http://www.ureader.com/message/694393.aspx
That thread explains why the 2 names exist, but since the job is to
merely find the drive, in this case, the concept to find physical from
logical and not useful. It also doesn't explain why you used the
"logical" name in the first place, nor why it fails in Win7 or why it
requires admin access.
> MPE wrote the original SCSI driver under contract to IntellaSys,
> which includes NDAs. At least one VentureForth release included
> our original source file with the documentation intact. Other
> releases included derived code with most of the documentation
> removed.
>
> The bit-twiddling code for talking to the SEAforth chips is
> based on proprietary code from the company that makes the
> USB interface.
None of which matters much any more. The GreenArrays chips will
certainly not use this method. I'm just trying to maintain my
investment until an alternative is available.
DaR
Stephen Pelc
<dru...@worldnet.att.net> wrote:
> It also doesn't explain why you used the
>"logical" name in the first place, nor why it fails in Win7 or why it
>requires admin access.
Because that's what the code from the USB chip people did. The thread
was from 2005, so it didn't deal with Vista, 2008, or Win7. Why
Microsoft make these changes, I have no idea.
>None of which matters much any more. The GreenArrays chips will
>certainly not use this method. I'm just trying to maintain my
>investment until an alternative is available.
There are plenty of existing gadgets that do USB to bit-bang
I/O. Some of them are even buffered and Forth programmable,
e.g.
http://www.mpeforth.com/jtagwidget.htm