WakePup2 for Puppy 2.x and greater (NOW WITH more SUPPORT)

Using applications, configuring, problems
Message
Author
User avatar
Crash
Posts: 453
Joined: Fri 09 Dec 2005, 06:34
Location: Melbourne, FL

#21 Post by Crash »

Most "modern" computers don't have a floppy drive, but they have a CD or DVD reader. Creating a bootable CD using Wakepup2 as the image allows booting of a USB drive on these computers. The original version of Wakepup2 works fine in this instance. Since the new version relies on a temporary file to be written to the floppy, it will fail in such an implementation. For this particular use, I recommend using the original version.

Otherwise, the new version works fine for me. I used it to boot up a 1GB Sandisk cruzer micro on an Athlon XP vintage computer, no problem. I'm making this post using that combination.

John Doe
Posts: 1681
Joined: Mon 01 Aug 2005, 04:46
Location: Michigan, US

#22 Post by John Doe »

Crash wrote:Creating a bootable CD using Wakepup2 as the image allows booting of a USB drive on these computers.
Can I ask you real quick where you got this info? You're the second person I've seen say that recently. If it's in the wiki or something maybe it should be replaced.

Although it's a good way to go, it's much faster for you if your computer can boot from CD to just use the puppy CD and enter:

puppy pmedia=usbflash

or (for pre 2.16)

puppy PMEDIA=usbflash

puppy will boot up your USB fast as lightning and you don't have to wait for all the DOS crap. you can also use many different file systems that way.

Sage
Posts: 5536
Joined: Tue 04 Oct 2005, 08:34
Location: GB

#23 Post by Sage »

All the 'modern' PC s that I build have a FDD and always will have one. If you're unlucky enough to be gifted a proprietary box from a cheapskate reseller, for £4.62 + VAT, possible less, you can fit one, as most boards still contain a header.
There are only two maxims: never buy a proprietary box and never buy a laptop. There isn't any good reason for individuals to do either, although if the company wants to buy either for you, then they don't intend you to use a FDD and/or they are a cheapskate company that you might not wish to work for?!

PS. I still have a brand new Dell laptop external FDD, complete with cable, for sale...................

User avatar
Crash
Posts: 453
Joined: Fri 09 Dec 2005, 06:34
Location: Melbourne, FL

#24 Post by Crash »

I tried it and it works fine. I occasionally carry around the combination of wakepup2 on a mini-CD and Puppy Linux on a USB stick to demonstrate it.

At the risk of getting too political, I think we still must consider the several million new owners of computers who will not crack open the case to put in a floppy drive. If you have seen any of my past posts, I am certainly not anti-floppy; in fact I have been accused of being too pro-floppy.

Bring back punched paper tape!!! That was REAL computing.

///

P.S. (Edited July 20, 2007)

Puppy Linux boot floppy images have been able to run on CD for at least the last year. Wakepup2, Wakepup, and Boot2pup have all been able to run on CD. The reason I use the boot program rather than the Live CD is to get better control of the process and minimize ambiguity when evaluating various boot configurations. There can be no doubt, for instance, that the files are indeed being loaded from the target device (USB stick, etc) rather than from the Live CD itself. Also, the one boot CD version works for all Puppy versions (at least within major version updates). So you don't have to burn a new Live CD every time you update the Puppy version on the USB drive. With the fast transfer speed of a CD, the difference in time between using the Live CD and the Boot CD image is minimal; the majority of the delay is in the Panasonic USB driver execution, which can be several tens of seconds.

If you have trouble with the Panasonic USBASPI.SYS driver, you may get better results with their previous version. The current one is V2.20, but you can replace it with the 2.06 version from the Boot2Pup floppy image if it gives you trouble. The typical symptom of a badly behaved USB driver is a complaint from FreeDOS stating "Bad or missing Command Interpreter." Reverting back to the previous version of the Panasonic USB driver may fix the problem, although the older version is admittedly much slower. Unfortunately, there doesn't appear to be a universal solution that works for all computers and all USB sticks. By far, notebook computers have been most problematic for me.

If you are really into this stuff, a Google search on "dos usb driver" gives a whole bunch of hits on various solutions. The problem, of course, is that it is a "modern" peripheral implemented on a dying operating system. Any of the references are bound to be several years old, and there doesn't appear to be anything new that has come up lately.

"the Z8000 is better"

///
Last edited by Crash on Fri 20 Jul 2007, 05:17, edited 1 time in total.

John Doe
Posts: 1681
Joined: Mon 01 Aug 2005, 04:46
Location: Michigan, US

#25 Post by John Doe »

Crash wrote:Bring back punched paper tape!!! That was REAL computing.
lol, it appears we have a REAL hacker in our midst.

muggins
Posts: 6724
Joined: Fri 20 Jan 2006, 10:44
Location: hobart

#26 Post by muggins »

please, nobody bring up cloacal sorting of hollerith cards!

Sage
Posts: 5536
Joined: Tue 04 Oct 2005, 08:34
Location: GB

#27 Post by Sage »

several million new owners of computers who will not crack open the case to put in a floppy drive
That's the problem - several million have been brainwashed to believe they cannot perform this simple task. Fortunately, anyone visiting this Forum, is quite capable of building their own HW .
Paper tape: I won't hear a word against it. Much of my best work was done with paper tape. I could even 'read' short sections. Cards were a retrograde step - lose one and the ship is sunk.
On the other hand, it's a myth that FD s are insecure or unstable. I have kept a reference specimen in a cool wooden dark drawer for twenty years and it still reads perfectly. This is even more amazing because it's a 720 converted to a 1.44 with a hand drill - you know - just like the trade wanted to say it wasn't possible, different coatings were used, and all that other garbage they pushed out at the time.
Some folks even think that cpu s have different architectures within the same series, ie 586, 686 and that clocking is either illegal or inadvisable. One born every minute....

PaulBx1
Posts: 2312
Joined: Sat 17 Jun 2006, 03:11
Location: Wyoming, USA

#28 Post by PaulBx1 »

Er, getting back to the subject at hand...

John, your latest wakepup2 now works for me, on the USB 1.1 port. I turned on all the debug I could find for booting from the pcmcia/USB 2.0 port and it hung at this point:

Image

Does that tell you anything? The config was nothing at all in the 1.1 port and the flash in one of the 2.0 ports. Did yours get stuck there when you used the hub?

John Doe
Posts: 1681
Joined: Mon 01 Aug 2005, 04:46
Location: Michigan, US

#29 Post by John Doe »

PaulBx1 wrote:Does that tell you anything?
maybe one of the GURU's can give us a bump. i have no idea.
PaulBx1 wrote:The config was nothing at all in the 1.1 port and the flash in one of the 2.0 ports. Did yours get stuck there when you used the hub?
mines below.

i did some additional testing with MUT and noticed it seems to be a problem with my SD card reader not a REAL USB Thumbdrive. i can read this PNY Attache' USB Thumbdrive over the HUB but not the SD card reader.

will try to boot over HUB with PNY.
Attachments
usb-hub-error.jpg
(80.33 KiB) Downloaded 948 times
Last edited by John Doe on Wed 18 Jul 2007, 22:03, edited 1 time in total.

John Doe
Posts: 1681
Joined: Mon 01 Aug 2005, 04:46
Location: Michigan, US

#30 Post by John Doe »

John Doe wrote:will try to boot over HUB with PNY.
works over a 2.0 HUB with PNY Attache' USB Thumbdrive.

i'd assume it works over a 1.1 HUB but don't have one around.

PaulBx1
Posts: 2312
Joined: Sat 17 Jun 2006, 03:11
Location: Wyoming, USA

#31 Post by PaulBx1 »

I posted a candidate fix for my hang issue on the 217B thread, but was not able to test it because of other issues I don't understand. Maybe the old edit-initrd script I have is out of date. John, how do you edit an initrd.gz?

John Doe
Posts: 1681
Joined: Mon 01 Aug 2005, 04:46
Location: Michigan, US

#32 Post by John Doe »

Here give this one a try with RC1

*edit- deleted attachment, nevermind my initrd.gz didn't work.

BK is going to have to do it.

he could just post the new initrd.gz for you to test since you are using USB and Floppy.

John Doe
Posts: 1681
Joined: Mon 01 Aug 2005, 04:46
Location: Michigan, US

#33 Post by John Doe »

John Doe wrote:BK is going to have to do it.
actually, now that unleashed 2.17 is out i can do it.

give this one a test (it's 2.17 final).
Last edited by John Doe on Sat 21 Jul 2007, 05:19, edited 1 time in total.

John Doe
Posts: 1681
Joined: Mon 01 Aug 2005, 04:46
Location: Michigan, US

#34 Post by John Doe »

just wanted to mention i checked with that code and was able to boot in all manners i've previous attempted over here.

no adverse effects to current function.

if it works over there, it's golden.

PaulBx1
Posts: 2312
Joined: Sat 17 Jun 2006, 03:11
Location: Wyoming, USA

#35 Post by PaulBx1 »

I booted with it, and it worked no differently. I then looked at /initrd/sbin/init and found that usbstoragecompletionfunc was the standard one, not the one that I modified. So it looks like this is the same initrd.gz as the standard one. :)

John Doe
Posts: 1681
Joined: Mon 01 Aug 2005, 04:46
Location: Michigan, US

#36 Post by John Doe »

I edited the wrong init in initrd-tree. There used to be a symlink in the root and now it's a file. I thought they moved it and edited the wrong one (it didn't even make it into the image, but the bakup i made did).

I edited the attachment above, please try again.

I'll have to redo my testing now also :roll:

PaulBx1
Posts: 2312
Joined: Sat 17 Jun 2006, 03:11
Location: Wyoming, USA

#37 Post by PaulBx1 »

OK, with the new initrd.gz it does timeout properly in 50 sec (printing out 1 2 3 4... as it goes). Of course shortly afterward it dumps me at the command promt, saying something about gurus being able to troubleshoot. It also said it found no usb devices.

So anyway, I think that rearrangement of usbstoragecompletionfunc is correct.

As to what's wrong with the wakepup2? I think it needs also the boot parameter "pfix=usbcard". When I boot from CD and use "puppy pfix=usbcard pmedia=usbflash", it does boot up properly (that's how I'm booted now). Wakepup2 does detect the device when it is out on the pcmcia card. Is there a way for it to know that the USB where the device is, is generated from a pcmcia card?

Another thing you could do, if the above is not possible, is just allow the user to tack on a parameter manually. Another menu entry would do it. That's not as user-friendly, of course.

John Doe
Posts: 1681
Joined: Mon 01 Aug 2005, 04:46
Location: Michigan, US

#38 Post by John Doe »

update is here:

http://www.murga-linux.com/puppy/viewto ... 924#122924
PaulBx1 wrote:So anyway, I think that rearrangement of usbstoragecompletionfunc is correct.
cool. did a couple tests here and all looks well so far.
PaulBx1 wrote:As to what's wrong with the wakepup2? I think it needs also the boot parameter "pfix=usbcard".
JD: *slaps self on forehead*!

i've only seen that on the screen about 40000 times in the past year, thanks for making the connection...

i added an item under custom at the end after the first pfix dialog to turn that on. it plays nice with the previous dialog.

thanks for rounding this off with me for everyone.

PaulBx1
Posts: 2312
Joined: Sat 17 Jun 2006, 03:11
Location: Wyoming, USA

#39 Post by PaulBx1 »

John, your md5sum there is wrong.

I take it, from the way you described the implementation, that there is no way for the program to automatically put that boot parameter in if the device is on a pcmcia card. Too bad...

For what it's worth, that rearrangement of the usbstoragecompletionfunc has the effect of reducing the timeout by one second. No big deal when invoked with a 50-sec timeout, but when called with smaller timeouts the reduction may become significant. So, the calls of this routine should be examined to see if you want to boost the timeout parameter.

PaulBx1
Posts: 2312
Joined: Sat 17 Jun 2006, 03:11
Location: Wyoming, USA

#40 Post by PaulBx1 »

John, any update on this?

Post Reply