X Windows Performance with DOC 2000

X Windows Performance with DOC 2000

[David Cooper]

Hello,

I am using a 2.4.2 Kernel re-compiled with the MTD subsystem on a 144 MB
DiskOnChip.

I can boot from the DOC and do most things except X refuses to run.  When I
run X from the DOC it switches to a blank screen and just sits there.  The
only way to recover is to perform a hard reset or cycle the power.

Running X from an IDE hard disk using the same platform does not seem to be
a problem.

Has anyone else encountered this problem?  Any suggestions as to what I
could try?

Thanks in advance.

Cheers,

David Cooper

Re: X Windows Performance with DOC 2000

[Charles Manning]

On Fri, 14 Jun 2002 15:09, David Cooper wrote:
> Hello,
>
> I am using a 2.4.2 Kernel re-compiled with the MTD subsystem on a 144 MB
> DiskOnChip.
>
> I can boot from the DOC and do most things except X refuses to run.  When I
> run X from the DOC it switches to a blank screen and just sits there.  The
> only way to recover is to perform a hard reset or cycle the power.
>
> Running X from an IDE hard disk using the same platform does not seem to be
> a problem.
>
> Has anyone else encountered this problem?  Any suggestions as to what I
> could try?
>

Hunch: Because X is big-n-fat, maybe it needs swap space. Doing swap onto a 
DOC would, I hunch,  suck fiercely (if it can be done at all).  Maybe try 
mounting DOC for the filesystem and keep an IDE for swap. Maybe look at 
/proc/swaps on the two boots and see if they tell you anything.

-- Charles

RE: X Windows Performance with DOC 2000

[David Cooper]

No good.

I tried turning the swap partition off and also using an external IDE drive
for the swap partition.

In both cases I get the same result as before, X is launched but when the
screen goes blank it never recovers or gives me any graphics output.

Any other suggestions?

Thanks,

David Cooper



-----Original Message-----
From: linux-mtd-admin@lists.infradead.org
[mailto:linux-mtd-admin@lists.infradead.org]On Behalf Of Charles Manning
Sent: Saturday, June 15, 2002 6:32 PM
To: David Cooper; Linux MTD
Subject: Re: X Windows Performance with DOC 2000


On Fri, 14 Jun 2002 15:09, David Cooper wrote:
> Hello,
>
> I am using a 2.4.2 Kernel re-compiled with the MTD subsystem on a 144 MB
> DiskOnChip.
>
> I can boot from the DOC and do most things except X refuses to run.  When
I
> run X from the DOC it switches to a blank screen and just sits there.  The
> only way to recover is to perform a hard reset or cycle the power.
>
> Running X from an IDE hard disk using the same platform does not seem to
be
> a problem.
>
> Has anyone else encountered this problem?  Any suggestions as to what I
> could try?
>

Hunch: Because X is big-n-fat, maybe it needs swap space. Doing swap onto a
DOC would, I hunch,  suck fiercely (if it can be done at all).  Maybe try
mounting DOC for the filesystem and keep an IDE for swap. Maybe look at
/proc/swaps on the two boots and see if they tell you anything.

-- Charles

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

RE: X Windows Performance with DOC 2000

[Ilguiz Latypov]

Will this line log anything meaningful?

   xinit > /tmp/xinit.log 2>&1

Ilguiz

On Mon, 17 Jun 2002, David Cooper wrote:

> Any other suggestions?

RE: X Windows Performance with DOC 2000

[David Cooper]

Thanks for the tip.

I tried this but, of course, the screen goes blank, system hangs and I can
never look at the xinit.log file.

When I do a hard reboot, this file is gone.

Cheers,

David Cooper

-----Original Message-----
From: linux-mtd-admin@lists.infradead.org
[mailto:linux-mtd-admin@lists.infradead.org]On Behalf Of Ilguiz Latypov
Sent: Monday, June 17, 2002 6:32 PM
To: David Cooper
Cc: Linux MTD
Subject: RE: X Windows Performance with DOC 2000



Will this line log anything meaningful?

   xinit > /tmp/xinit.log 2>&1

Ilguiz

On Mon, 17 Jun 2002, David Cooper wrote:

> Any other suggestions?


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Re: X Windows Performance with DOC 2000

[Brendan J Simon]

Can you save it to another computer via NFS ?


David Cooper wrote:

>Thanks for the tip.
>
>I tried this but, of course, the screen goes blank, system hangs and I can
>never look at the xinit.log file.
>
>When I do a hard reboot, this file is gone.
>
>Cheers,
>
>David Cooper
>
>-----Original Message-----
>From: linux-mtd-admin@lists.infradead.org
>[mailto:linux-mtd-admin@lists.infradead.org]On Behalf Of Ilguiz Latypov
>Sent: Monday, June 17, 2002 6:32 PM
>To: David Cooper
>Cc: Linux MTD
>Subject: RE: X Windows Performance with DOC 2000
>
>
>
>Will this line log anything meaningful?
>
>   xinit > /tmp/xinit.log 2>&1
>
>Ilguiz
>
>On Mon, 17 Jun 2002, David Cooper wrote:
>

RE: X Windows Performance with DOC 2000

[Ilguiz Latypov]

David,

Perhaps, running xinit command under gdb can help.  To go step-by-step 
through the lines of code, a debug version of xinit would be more helpful.  
I believe the gcc -g option will provide necessary debug information in 
the produced executable binary.

Ilguiz

On Mon, 17 Jun 2002, David Cooper wrote:

> I tried this but, of course, the screen goes blank, system hangs and I
> can never look at the xinit.log file.

RE: X Windows Performance with DOC 2000

[David Cooper]

Hello,

I have tried more and found out more.

(1) M-Systems Firmware

Strange, but true. If I use M-System's DOS-based dformat utility to load
firware version 4.2 into the DOC 2000 this prevents X from runnning from the
IDE drive.  This is true even if I boot with a kernel that has no support
for the DOC 2000.  As long as the DOC is sitting in my system (with Rev 4.2
firmware) X will not run.  I should point out that X does not cause the
system to hang in this case.

(2) nftl_format

So I bit the bullet and tried to remove all firmware (M-Systems dformat
/empty) and then do a complete low-level format with the MTD nftl_format
utility.  Then I put on top of that a ext2 root filesystem (not self
bootable of course).  This provided no improvement.  X still falls over at
the same spot.

(3) X Modules

I then was able to look at the X log file (/var/log/XFree86.0.log) and it
seemed that the point at which X fell over was when it was trying to load
the ddc module (libddc.a).  So I then mounted the entire
/usr/X11R6/lib/modules directory from the IDE drive and ran the rest from
the root filesystem on the DOC.  This allows X to get a little further
(seems to try to load a screen from the VESA BIOS).  But it still hangs,
this time after putting rubbish graphics on the screen (as in a normal
initialisation).

Question:  Why does X have trouble loading the libddc.a module from the DOC?

Thanks for you advice and suggestions thus far.

Cheers,

David Cooper


-----Original Message-----
From: linux-mtd-admin@lists.infradead.org
[mailto:linux-mtd-admin@lists.infradead.org]On Behalf Of Ilguiz Latypov
Sent: Tuesday, June 18, 2002 10:12 AM
To: David Cooper
Cc: Linux MTD
Subject: RE: X Windows Performance with DOC 2000



David,

Perhaps, running xinit command under gdb can help.  To go step-by-step
through the lines of code, a debug version of xinit would be more helpful.
I believe the gcc -g option will provide necessary debug information in
the produced executable binary.

Ilguiz

On Mon, 17 Jun 2002, David Cooper wrote:

> I tried this but, of course, the screen goes blank, system hangs and I
> can never look at the xinit.log file.


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Re: X Windows Performance with DOC 2000

[David Woodhouse]

david@forcedpotato.com said:
> (3) X Modules
> I then was able to look at the X log file (/var/log/XFree86.0.log) and
> it seemed that the point at which X fell over was when it was trying
> to load the ddc module (libddc.a).  So I then mounted the entire /usr/
> X11R6/lib/modules directory from the IDE drive and ran the rest from
> the root filesystem on the DOC.  This allows X to get a little further
> (seems to try to load a screen from the VESA BIOS).  But it still
> hangs, this time after putting rubbish graphics on the screen (as in a
> normal initialisation).

> Question:  Why does X have trouble loading the libddc.a module from
> the DOC?

Interesting. The DDC module is likely to be using the VESA BIOS too -- you 
may find that your box is dying when XFree86 tries to do _anything_ 
BIOS-related. 

You said X works when you run it from the IDE drive. Do you have the
DiskOnChip drivers loaded? If not, does it still work when you run it from
IDE after loading them?

--
dwmw2

RE: X Windows Performance with DOC 2000

[David Cooper]

David,

Yes, when I use a kernel with the MTD support built-in for DoC 2000 and nftl
I can still run X with the DoC in the system (unless I use the M-Systems 4.2
firmware which prevents X from running with or without MTD support).  I have
not tried loading and unloading MTD support as a kernel module.  Maybe this
is worth a go.

Interestingly, I can run SVGAlib applications from the DoC (using the VESA
driver) without any trouble.  This is confusing because I would imagine that
the SVGAlib VESA driver and the XFree86 VESA driver are both accessing the
VESA BIOS.

M-Systems have told me that the firmware DoC driver that loads into memory
at bootup sometimes conflicts in memory space that linux is trying to use.
I suppose this would explain why using different firmware versions can
prevent X from running from the IDE drive.

Cheers,

David

-----Original Message-----
From: David Woodhouse [mailto:dwmw2@redhat.com]On Behalf Of David
Woodhouse
Sent: Thursday, June 20, 2002 5:59 PM
To: David Cooper
Cc: Ilguiz Latypov; Linux MTD
Subject: Re: X Windows Performance with DOC 2000



david@forcedpotato.com said:
> (3) X Modules
> I then was able to look at the X log file (/var/log/XFree86.0.log) and
> it seemed that the point at which X fell over was when it was trying
> to load the ddc module (libddc.a).  So I then mounted the entire /usr/
> X11R6/lib/modules directory from the IDE drive and ran the rest from
> the root filesystem on the DOC.  This allows X to get a little further
> (seems to try to load a screen from the VESA BIOS).  But it still
> hangs, this time after putting rubbish graphics on the screen (as in a
> normal initialisation).

> Question:  Why does X have trouble loading the libddc.a module from
> the DOC?

Interesting. The DDC module is likely to be using the VESA BIOS too -- you
may find that your box is dying when XFree86 tries to do _anything_
BIOS-related.

You said X works when you run it from the IDE drive. Do you have the
DiskOnChip drivers loaded? If not, does it still work when you run it from
IDE after loading them?

--
dwmw2

Re: X Windows Performance with DOC 2000

[David Woodhouse]

david@forcedpotato.com said:
>  Yes, when I use a kernel with the MTD support built-in for DoC 2000
> and nftl I can still run X with the DoC in the system (unless I use
> the M-Systems 4.2 firmware which prevents X from running with or
> without MTD support).  I have not tried loading and unloading MTD
> support as a kernel module.  Maybe this is worth a go. 

If you have the drivers built-in and can still run X, that's what I wanted 
to know. Can you even have the file system on NFTL mounted, as long as your 
X modules aren't on it?

>  Interestingly, I can run SVGAlib applications from the DoC (using the
> VESA driver) without any trouble.  This is confusing because I would
> imagine that the SVGAlib VESA driver and the XFree86 VESA driver are
> both accessing the VESA BIOS.

True.

> M-Systems have told me that the firmware DoC driver that loads into
> memory at bootup sometimes conflicts in memory space that linux is
> trying to use. I suppose this would explain why using different
> firmware versions can prevent X from running from the IDE drive. 

That caused Linux to fail to boot -- but once you're actually in Linux, 
the allocations of the old DOS driver should be long forgotten.

Unless the XFree86 code somehow tries to re-run the DiskOnChip firmware -- 
but you said you'd reproduced the problem even with that obliterated, 
didn't you?

--
dwmw2

RE: X Windows Performance with DOC 2000

[David Cooper]

David,

Some test results:

(1) X (MTD-enabled kernel,IDE)

- Works fine.

(2) X after first mounting nftla2 on /mnt/doc (MTD-enabled kernel,IDE)

- Works fine.
- But, if I try to perform any file operations on /mnt/doc the system hangs.
- If I stay away from /mnt/doc it is business as usual.

(3) X (MTD-enabled kernel,IDE), then stopping X and mounting nftla2 on
/mnt/doc

- Mount operation mount /dev/nftla2 /mnt/doc causes the shell to hang.
- Viewing the process "mount" using top shows that it is running
continuously and using 84% CPU time.
- All attempts to kill the mount process are futile.


So, it seems that once X has had its way with the system, use of the
DiskOnChip is impossible.  Question is, why?

I have been reflecting some on why SVGAlib apps run while X does not when I
use the DoC as the root file system.  My theory was that the SVGAlib
application is loaded into memory before it starts executing and talking to
the VESA BIOS.  In contrast, X is trying to setup the VESA BIOS and then
attempts to load modules from the DoC (at which point it hangs).

However, I repeated the above tests with SVGAlib and found that SVGAlib apps
and file operations on the DoC can occur at the same time with no apparent
problems.

hmmmm?

Cheers,

David Cooper



-----Original Message-----
From: David Woodhouse [mailto:dwmw2@redhat.com]On Behalf Of David
Woodhouse
Sent: Thursday, June 20, 2002 8:39 PM
To: David Cooper
Cc: Ilguiz Latypov; Linux MTD
Subject: Re: X Windows Performance with DOC 2000



david@forcedpotato.com said:
>  Yes, when I use a kernel with the MTD support built-in for DoC 2000
> and nftl I can still run X with the DoC in the system (unless I use
> the M-Systems 4.2 firmware which prevents X from running with or
> without MTD support).  I have not tried loading and unloading MTD
> support as a kernel module.  Maybe this is worth a go.

If you have the drivers built-in and can still run X, that's what I wanted
to know. Can you even have the file system on NFTL mounted, as long as your
X modules aren't on it?

>  Interestingly, I can run SVGAlib applications from the DoC (using the
> VESA driver) without any trouble.  This is confusing because I would
> imagine that the SVGAlib VESA driver and the XFree86 VESA driver are
> both accessing the VESA BIOS.

True.

> M-Systems have told me that the firmware DoC driver that loads into
> memory at bootup sometimes conflicts in memory space that linux is
> trying to use. I suppose this would explain why using different
> firmware versions can prevent X from running from the IDE drive.

That caused Linux to fail to boot -- but once you're actually in Linux,
the allocations of the old DOS driver should be long forgotten.

Unless the XFree86 code somehow tries to re-run the DiskOnChip firmware --
but you said you'd reproduced the problem even with that obliterated,
didn't you?

--
dwmw2

Re: X Windows Performance with DOC 2000

[David Woodhouse]

david@forcedpotato.com said:
>  - Mount operation mount /dev/nftla2 /mnt/doc causes the shell to
> hang. - Viewing the process "mount" using top shows that it is running
> continuously and using 84% CPU time. - All attempts to kill the mount
> process are futile.

OK, here we can find some clues. _What_ is it doing? Use SysRq-P a few 
times to see where it's looping, look it up in your kernel image. If you 
compile with '-g' you can get gdb to tell you the precise line.

--
dwmw2

RE: X Windows Performance with DOC 2000

[Ilguiz Latypov]

David,

Dumb question.  The M-Sys driver was not included into the kernel and the
M-Sys firmware was not activated at the time of the test, were they?  Not
that I want to cast a shadow on the commercial grade product, just to
delineate different cases.

Ilguiz

On Fri, 21 Jun 2002, David Cooper wrote:

> (1) X (MTD-enabled kernel,IDE)
> ...
> (2) X after first mounting nftla2 on /mnt/doc (MTD-enabled kernel,IDE)
> ...
> (3) X (MTD-enabled kernel,IDE), then stopping X and mounting nftla2 on
> /mnt/doc
> ...

RE: X Windows Performance with DOC 2000

[cfowler]

I'm actually tired of using M-SYS's driver.  Their tech-support sucks
and I hate using putimage in DOS.  I'm switching to DOM and  Compact
Flash.  I'll guess some other company will get my order for 10,000
units.

On Fri, 2002-06-21 at 09:56, Ilguiz Latypov wrote:
> 
> David,
> 
> Dumb question.  The M-Sys driver was not included into the kernel and the
> M-Sys firmware was not activated at the time of the test, were they?  Not
> that I want to cast a shadow on the commercial grade product, just to
> delineate different cases.
> 
> Ilguiz
> 
> On Fri, 21 Jun 2002, David Cooper wrote:
> 
> > (1) X (MTD-enabled kernel,IDE)
> > ...
> > (2) X after first mounting nftla2 on /mnt/doc (MTD-enabled kernel,IDE)
> > ...
> > (3) X (MTD-enabled kernel,IDE), then stopping X and mounting nftla2 on
> > /mnt/doc
> > ...
> 
> 
> 
> ______________________________________________________
> Linux MTD discussion mailing list
> http://lists.infradead.org/mailman/listinfo/linux-mtd/
> 
> 

Re: X Windows Performance with DOC 2000

[David Woodhouse]

cfowler@outpostsentinel.com said:
> I'm actually tired of using M-SYS's driver.  Their tech-support sucks
> and I hate using putimage in DOS.  I'm switching to DOM and  Compact
> Flash.  I'll guess some other company will get my order for 10,000
> units.

What problems did you have with the real Linux driver?

--
dwmw2

Re: X Windows Performance with DOC 2000

[cfowler]

#1) What the hell is msys/fla process?  I do not have my DOC mounted and
    the load averages are around 1.99 when the module is loaded.  M-SYS 
    for some reason can not seem to tell me what is going on here. If 
    this is a garbage collector then it is using too much resources. TO
    fix this I issue rmmod doc.  Now if I can remove the module then why
    is GC doing this?  I have interfaces that insert and remove the
    module for each operation on flash.  This takes time and is a 
    hack.

#2) Grub destroyed my first DOC.  I use doc-lilo.  Everytime I flash a  
    new kernel on the DOC I have to re-run doc-lilo.

#3) DOC is single source

#4) My contractor and I have not been too pleased with their support.

#5) I have to use putimage.exe and getimage.exe in DOS to 
    flash other units.  This means in development I have to 
    maintain a dos envrionment.  We are strictly linux in development

#6) I have to load the driver as a module to obey the GPL of the 
    kernel.  I prefer to have it built in.  This makes upgrades easier.

#7) Writing to the thing is slooooowwww.  


#8) CF is so much easier using in devel and production.  DOM to me
    seems better

On Fri, 2002-06-21 at 10:12, David Woodhouse wrote:
> 
> cfowler@outpostsentinel.com said:
> > I'm actually tired of using M-SYS's driver.  Their tech-support sucks
> > and I hate using putimage in DOS.  I'm switching to DOM and  Compact
> > Flash.  I'll guess some other company will get my order for 10,000
> > units.
> 
> What problems did you have with the real Linux driver?
> 
> --
> dwmw2
> 
> 
> 
> 

Re: X Windows Performance with DOC 2000

[Christopher Fowler]

I got X to work on DOC 1.5 years ago.  But I used ext2 with compr format
on my DOC2000

Chris

On Mon, 2002-06-17 at 20:40, Brendan J Simon wrote:
> 
> Can you save it to another computer via NFS ?
> 
> 
> David Cooper wrote:
> 
> >Thanks for the tip.
> >
> >I tried this but, of course, the screen goes blank, system hangs and I can
> >never look at the xinit.log file.
> >
> >When I do a hard reboot, this file is gone.
> >
> >Cheers,
> >
> >David Cooper
> >
> >-----Original Message-----
> >From: linux-mtd-admin@lists.infradead.org
> >[mailto:linux-mtd-admin@lists.infradead.org]On Behalf Of Ilguiz Latypov
> >Sent: Monday, June 17, 2002 6:32 PM
> >To: David Cooper
> >Cc: Linux MTD
> >Subject: RE: X Windows Performance with DOC 2000
> >
> >
> >
> >Will this line log anything meaningful?
> >
> >   xinit > /tmp/xinit.log 2>&1
> >
> >Ilguiz
> >
> >On Mon, 17 Jun 2002, David Cooper wrote:
> >
> 
> 
> 
> ______________________________________________________
> Linux MTD discussion mailing list
> http://lists.infradead.org/mailman/listinfo/linux-mtd/
> 
> 

Re: X Windows Performance with DOC 2000

[Ilguiz Latypov]

I found one message of March 6, 2002 where the MTD-patched GRUB displayed 
the prompt.  I think this was half a success as I am an optimist :-)

I believe you did create a file system on top of NFTL layer and put the 
/boot/grub/menu.lst configuration file there.  This file tells GRUB which 
partition to open and which kernel to load.

The difficulties with reformatting DoC by using the M-Sys'es DFORMAT might 
be due to grub_firmware indeed.

> So I do a dformat /win:D000 /s:DOC504.exb /FIRST /NOFORMAT /Y and I get
> the following error:
> 
> Error writing EXB
> Status code 26 (flUnknownMedia) received from flPlaceExbByBuffer.
> Format Failed.

Perhaps, running the same utility with /s:* and without /noformat may help 
in erasing the ext2 partitions.

See pages 5 and 7 of

    http://www.m-sys.com/files/Manuals%5Cdoc%5CDOC_Tool.pdf

Ilguiz

On 21 Jun 2002, cfowler wrote:

> #2) Grub destroyed my first DOC.  I use doc-lilo.  Everytime I flash a  
>     new kernel on the DOC I have to re-run doc-lilo.

Re: X Windows Performance with DOC 2000

[Christopher Fowler]

Even if I could get it to work, I'm still going with CF.  M-SYS really
needs to change the name Disk on Chip to Chip on Chip.  It is not
seamless as their marketing department would like you to believe.

Chris

On Fri, 2002-06-21 at 20:01, Ilguiz Latypov wrote:
> 
> I found one message of March 6, 2002 where the MTD-patched GRUB displayed 
> the prompt.  I think this was half a success as I am an optimist :-)
> 
> I believe you did create a file system on top of NFTL layer and put the 
> /boot/grub/menu.lst configuration file there.  This file tells GRUB which 
> partition to open and which kernel to load.
> 
> The difficulties with reformatting DoC by using the M-Sys'es DFORMAT might 
> be due to grub_firmware indeed.
> 
> > So I do a dformat /win:D000 /s:DOC504.exb /FIRST /NOFORMAT /Y and I get
> > the following error:
> > 
> > Error writing EXB
> > Status code 26 (flUnknownMedia) received from flPlaceExbByBuffer.
> > Format Failed.
> 
> Perhaps, running the same utility with /s:* and without /noformat may help 
> in erasing the ext2 partitions.
> 
> See pages 5 and 7 of
> 
>     http://www.m-sys.com/files/Manuals%5Cdoc%5CDOC_Tool.pdf
> 
> Ilguiz
> 
> On 21 Jun 2002, cfowler wrote:
> 
> > #2) Grub destroyed my first DOC.  I use doc-lilo.  Everytime I flash a  
> >     new kernel on the DOC I have to re-run doc-lilo.
> 
> 
> ______________________________________________________
> Linux MTD discussion mailing list
> http://lists.infradead.org/mailman/listinfo/linux-mtd/
> 
> 

RE: X Windows Performance with DOC 2000

[Gonzalo Servat]

[-- Attachment #1: Type: text/plain, Size: 2735 bytes --]

Makes me feel really positive about my embedded linux project using a
DoC 2000 :-)


On Sat, 2002-06-22 at 11:14, David Cooper wrote:
> I am using CF at the moment on a different platform.  Works well and it is
> nice and easy to get it up and running (M-Systems Tech Support take note).
> 
> I am worried, however, about CF reliability at higher temperatures.  That is
> why I am prepared to endure the blood, sweat and tears to implement a
> solution with DiskOnChip.
> 
> -----Original Message-----
> From: linux-mtd-admin@lists.infradead.org
> [mailto:linux-mtd-admin@lists.infradead.org]On Behalf Of Christopher
> Fowler
> Sent: Friday, June 21, 2002 8:15 PM
> To: Ilguiz Latypov
> Cc: Linux MTD
> Subject: Re: X Windows Performance with DOC 2000
> 
> 
> 
> Even if I could get it to work, I'm still going with CF.  M-SYS really
> needs to change the name Disk on Chip to Chip on Chip.  It is not
> seamless as their marketing department would like you to believe.
> 
> Chris
> 
> On Fri, 2002-06-21 at 20:01, Ilguiz Latypov wrote:
> >
> > I found one message of March 6, 2002 where the MTD-patched GRUB displayed
> > the prompt.  I think this was half a success as I am an optimist :-)
> >
> > I believe you did create a file system on top of NFTL layer and put the
> > /boot/grub/menu.lst configuration file there.  This file tells GRUB which
> > partition to open and which kernel to load.
> >
> > The difficulties with reformatting DoC by using the M-Sys'es DFORMAT might
> > be due to grub_firmware indeed.
> >
> > > So I do a dformat /win:D000 /s:DOC504.exb /FIRST /NOFORMAT /Y and I get
> > > the following error:
> > >
> > > Error writing EXB
> > > Status code 26 (flUnknownMedia) received from flPlaceExbByBuffer.
> > > Format Failed.
> >
> > Perhaps, running the same utility with /s:* and without /noformat may help
> > in erasing the ext2 partitions.
> >
> > See pages 5 and 7 of
> >
> >     http://www.m-sys.com/files/Manuals%5Cdoc%5CDOC_Tool.pdf
> >
> > Ilguiz
> >
> > On 21 Jun 2002, cfowler wrote:
> >
> > > #2) Grub destroyed my first DOC.  I use doc-lilo.  Everytime I flash a
> > >     new kernel on the DOC I have to re-run doc-lilo.
> >
> >
> > ______________________________________________________
> > Linux MTD discussion mailing list
> > http://lists.infradead.org/mailman/listinfo/linux-mtd/
> >
> >
> 
> 
> 
> ______________________________________________________
> Linux MTD discussion mailing list
> http://lists.infradead.org/mailman/listinfo/linux-mtd/
> 
> 
> 
> ______________________________________________________
> Linux MTD discussion mailing list
> http://lists.infradead.org/mailman/listinfo/linux-mtd/
> 


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 232 bytes --]

RE: X Windows Performance with DOC 2000

[David Cooper]

I am using CF at the moment on a different platform.  Works well and it is
nice and easy to get it up and running (M-Systems Tech Support take note).

I am worried, however, about CF reliability at higher temperatures.  That is
why I am prepared to endure the blood, sweat and tears to implement a
solution with DiskOnChip.

-----Original Message-----
From: linux-mtd-admin@lists.infradead.org
[mailto:linux-mtd-admin@lists.infradead.org]On Behalf Of Christopher
Fowler
Sent: Friday, June 21, 2002 8:15 PM
To: Ilguiz Latypov
Cc: Linux MTD
Subject: Re: X Windows Performance with DOC 2000



Even if I could get it to work, I'm still going with CF.  M-SYS really
needs to change the name Disk on Chip to Chip on Chip.  It is not
seamless as their marketing department would like you to believe.

Chris

On Fri, 2002-06-21 at 20:01, Ilguiz Latypov wrote:
>
> I found one message of March 6, 2002 where the MTD-patched GRUB displayed
> the prompt.  I think this was half a success as I am an optimist :-)
>
> I believe you did create a file system on top of NFTL layer and put the
> /boot/grub/menu.lst configuration file there.  This file tells GRUB which
> partition to open and which kernel to load.
>
> The difficulties with reformatting DoC by using the M-Sys'es DFORMAT might
> be due to grub_firmware indeed.
>
> > So I do a dformat /win:D000 /s:DOC504.exb /FIRST /NOFORMAT /Y and I get
> > the following error:
> >
> > Error writing EXB
> > Status code 26 (flUnknownMedia) received from flPlaceExbByBuffer.
> > Format Failed.
>
> Perhaps, running the same utility with /s:* and without /noformat may help
> in erasing the ext2 partitions.
>
> See pages 5 and 7 of
>
>     http://www.m-sys.com/files/Manuals%5Cdoc%5CDOC_Tool.pdf
>
> Ilguiz
>
> On 21 Jun 2002, cfowler wrote:
>
> > #2) Grub destroyed my first DOC.  I use doc-lilo.  Everytime I flash a
> >     new kernel on the DOC I have to re-run doc-lilo.
>
>
> ______________________________________________________
> Linux MTD discussion mailing list
> http://lists.infradead.org/mailman/listinfo/linux-mtd/
>
>



______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Re: X Windows Performance with DOC 2000

[David Woodhouse]

cfowler@outpostsentinel.com said:
>  Even if I could get it to work, I'm still going with CF. 

Nasty things have been said about CF reliability -- any reason you're not 
just using raw flash?

> M-SYS really needs to change the name Disk on Chip to Chip on Chip.  It is
> not seamless as their marketing department would like you to believe. 

It basically works. Making it work _seamlessly_ is as you say a marketing 
exercise, and one they don't seem too interested in for Linux.

--
dwmw2

RE: X Windows Performance with DOC 2000

[David Cooper]

I tried using SysRq-P (after attempting a "mount" of the DoC once X is
running from the IDE) and it shows me the registers and so forth but I am
yet to decipher this into which instructions are causing the problem (I need
to work on it further).

In the meantime, I tried using the latest M-Systems driver just for fun.
This does allow X to run from the DoC.  Looks like there is some conflict
(for me anyway) between X and the MTD driver.

I'll keep proding away.

Cheers,

David Cooper
-----Original Message-----
From: linux-mtd-admin@lists.infradead.org
[mailto:linux-mtd-admin@lists.infradead.org]On Behalf Of David Woodhouse
Sent: Friday, June 21, 2002 3:04 AM
To: David Cooper
Cc: Ilguiz Latypov; Linux MTD
Subject: Re: X Windows Performance with DOC 2000



david@forcedpotato.com said:
>  - Mount operation mount /dev/nftla2 /mnt/doc causes the shell to
> hang. - Viewing the process "mount" using top shows that it is running
> continuously and using 84% CPU time. - All attempts to kill the mount
> process are futile.

OK, here we can find some clues. _What_ is it doing? Use SysRq-P a few
times to see where it's looping, look it up in your kernel image. If you
compile with '-g' you can get gdb to tell you the precise line.

--
dwmw2



______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

RE: X Windows Performance with DOC 2000

[David Woodhouse]

On Wed, 26 Jun 2002, David Cooper wrote:

> I tried using SysRq-P (after attempting a "mount" of the DoC once X is
> running from the IDE) and it shows me the registers and so forth but I am
> yet to decipher this into which instructions are causing the problem (I need
> to work on it further).

You need to look up the PC in your System.map to see which function it's 
in. Is it always in roughly the same place?

> In the meantime, I tried using the latest M-Systems driver just for fun.
> This does allow X to run from the DoC.  Looks like there is some conflict
> (for me anyway) between X and the MTD driver.

Interesting. I wonder what it does differently,

-- 
dwmw2