RE: [9fans] bitsy uvlong & suspend are fine, thanks.

RE: [9fans] bitsy uvlong & suspend are fine, thanks.

[Stephen Parker]

[-- Attachment #1: Type: text/plain, Size: 89 bytes --]

It is I who also has a non-suspending bitsy.  My bootloader is
2.18.45.

stephen



[-- Attachment #2: Re: [9fans] bitsy uvlong & suspend are fine, thanks. --]
[-- Type: message/rfc822, Size: 954 bytes --]

From: Axel Belinfante <Axel.Belinfante@cs.utwente.nl>
To: 9fans@cse.psu.edu
Subject: Re: [9fans] bitsy uvlong & suspend are fine, thanks.
Date: Mon, 28 Oct 2002 16:00:50 -0000
Message-ID: <200210281600.g9SG0o123154@zamenhof.cs.utwente.nl>

Any idea what (other) effects it would have I would downgrade to
a boot loader version that is close(r) to yours?

I recall that someone else (I'm sorry, forgot your name) mentioned
that suspend/resume never worked.
Question for you and others who also have suspend/resume problems:
do you also have a ``recent'' boot loader?

Axel.

> It seems that at 2.16.x they changed to boot loader to suspend/boot
> both linux and WinCE. So, it looks like our suspend code doesn't
> work for those boot loaders (since we have everything else the same). Sic.

Re: [9fans] bitsy uvlong & suspend are fine, thanks.

[Fco.J.Ballesteros]

> It'd be nice if there was a Plan 9 boot loader so that we didn't have
> to contend with the high rate of change in the Linux world.  (And the
> answer to the obvious next question is: no.)

boot loader hacking is no fun.  Make a mistake and your bitsy would
become a nice piece paperweight.

The real fix would be to make suspend work with newer boot loaders,
although I'd prefer then to keep it working with previous ones too (and I
don't know if that's feasible).

But in any case, considering that if you run Plan 9 on the bitsy you
won't run anything else there, I think it's reasonable if you just
have to use an old boot loader.

I've checked it, and a bitsy running Plan 9 runs out of battery
while suspended for more than one day.  The same happens to linux.
(Linux makes the battery keep up a few hours more, but in the end, you
have to charge the bitsy once a day).  I think that's the real
problem.

Re: [9fans] bitsy uvlong & suspend are fine, thanks.

[Dan Cross]

> Regarding the downgrade, I don't know what other effects it
> might have. You might look at http://handhelds.org/feeds/bootldr/CHANGELOG.

It'd be nice if there was a Plan 9 boot loader so that we didn't have
to contend with the high rate of change in the Linux world.  (And the
answer to the obvious next question is: no.)

	- Dan C.

Re: [9fans] bitsy uvlong & suspend are fine, thanks.

[Fco.J.Ballesteros]

Regarding the downgrade, I don't know what other effects it
might have. You might look at http://handhelds.org/feeds/bootldr/CHANGELOG.

Re: [9fans] bitsy uvlong & suspend are fine, thanks.

[Axel Belinfante]

Any idea what (other) effects it would have I would downgrade to
a boot loader version that is close(r) to yours?

I recall that someone else (I'm sorry, forgot your name) mentioned
that suspend/resume never worked.
Question for you and others who also have suspend/resume problems:
do you also have a ``recent'' boot loader?

Axel.

> It seems that at 2.16.x they changed to boot loader to suspend/boot
> both linux and WinCE. So, it looks like our suspend code doesn't
> work for those boot loaders (since we have everything else the same). Sic.

Re: [9fans] bitsy uvlong & suspend are fine, thanks.

[Fco.J.Ballesteros]

It seems that at 2.16.x they changed to boot loader to suspend/boot
both linux and WinCE. So, it looks like our suspend code doesn't
work for those boot loaders (since we have everything else the same). Sic.

Re: [9fans] bitsy uvlong & suspend are fine, thanks.

[Axel Belinfante]

> Your boot loader is newer than mine. Mine is a 2.14 something.
> Thus I think it has to be that, your bitsy was an H3630, right?

yes. it's one of those that have 32 instead of 16 Mb flash rom.
(don't know whether that might make a difference)

> I'll take a look at your boot loader to see if I find out anything.

Thanks!

Axel.

Re: [9fans] bitsy uvlong & suspend are fine, thanks.

[Fco.J.Ballesteros]

[-- Attachment #1: Type: text/plain, Size: 198 bytes --]

Your boot loader is newer than mine. Mine is a 2.14 something.
Thus I think it has to be that, your bitsy was an H3630, right?
I'll take a look at your boot loader to see if I find out anything.

[-- Attachment #2: Type: message/rfc822, Size: 3147 bytes --]

From: Axel Belinfante <Axel.Belinfante@cs.utwente.nl>
To: 9fans@cse.psu.edu
Subject: Re: [9fans] bitsy uvlong & suspend are fine, thanks.
Date: Mon, 28 Oct 2002 11:18:32 +0100
Message-ID: <200210281018.g9SAIWi16049@zamenhof.cs.utwente.nl>

> > I'm wondering why your kernel suspend/resumes for you but not
> > for me. Different hardware? Or could different bootloader
> > versions also play a role?
>
> Did you try after a hard reset w/o the dual sleeve plugged in?

yes, and it did not make a change.
(assuming that by hard reset you mean: using the on-off
switch `hidden' under the small sliding door at the bottom)

> I never tried suspending with a dual sleeve (Have  a one-slot pcmcia)
> so suspend might not work with that.

I tried with and without sleeve

> The bootloader is also an issue. Which version are you using?
> Mine is rather old now.

2.18.01

Axel.

Re: [9fans] bitsy uvlong & suspend are fine, thanks.

[Axel Belinfante]

> > I'm wondering why your kernel suspend/resumes for you but not
> > for me. Different hardware? Or could different bootloader
> > versions also play a role?
>
> Did you try after a hard reset w/o the dual sleeve plugged in?

yes, and it did not make a change.
(assuming that by hard reset you mean: using the on-off
switch `hidden' under the small sliding door at the bottom)

> I never tried suspending with a dual sleeve (Have  a one-slot pcmcia)
> so suspend might not work with that.

I tried with and without sleeve

> The bootloader is also an issue. Which version are you using?
> Mine is rather old now.

2.18.01

Axel.

Re: [9fans] bitsy uvlong & suspend are fine, thanks.

[Fco.J.Ballesteros]

> I'm wondering why your kernel suspend/resumes for you but not
> for me. Different hardware? Or could different bootloader
> versions also play a role?

Did you try after a hard reset w/o the dual sleeve plugged in?
I never tried suspending with a dual sleeve (Have  a one-slot pcmcia)
so suspend might not work with that.

The bootloader is also an issue. Which version are you using?
Mine is rather old now.

Re: [9fans] bitsy uvlong & suspend are fine, thanks.

[Axel Belinfante]

Unfortunately, your 9bitsy gives me the same results as I had.
Have less time than I hoped; no time to compile,
only diff-ed your sources with mine , and (quick glance) did
not see many differences (aprt from my hacking).

I'm wondering why your kernel suspend/resumes for you but not
for me. Different hardware? Or could different bootloader
versions also play a role?

Axel.
> Axel, if you still have problems after trying this version of the
> sources just let me know.  They should compile without problems.  If
> they don't compile out of the box, it would mean that some other file
> in my /sys/src/9/port directory should also be added to the tar; let
> me know.

[9fans] bitsy uvlong & suspend are fine, thanks.

[Fco.J.Ballesteros]

Hi,

	It seems that I had a bad .a5 file that did not get regenerated
when I did the mk install.  After removing all the .a5 files and
compiling it all again, the bitsy sources compiled and worked fine.

I have regenerated the tar ball in the web.  Now I'm sure that (at
least for my H3630) the current version of the sources manage to
suspend/resume the bitsy.  The ram is kept refreshed while in suspend
mode, of course.

The tar ball now includes an /arm/9bitsy binary, freshly compiled from
the corresponding sources also in the tar.

Axel, if you still have problems after trying this version of the
sources just let me know.  They should compile without problems.  If
they don't compile out of the box, it would mean that some other file
in my /sys/src/9/port directory should also be added to the tar; let
me know.