9fans - fans of the OS Plan 9 from Bell Labs
 help / color / mirror / Atom feed
* Re: [9fans] 2 bitsy questions
@ 2001-06-08  7:09 F.J.Ballesteros
  0 siblings, 0 replies; 3+ messages in thread
From: F.J.Ballesteros @ 2001-06-08  7:09 UTC (permalink / raw)
  To: 9fans

[-- Attachment #1: Type: text/plain, Size: 244 bytes --]

So, you mean that just repeating 
	setr13(PsrMfiq, m->sfiq);
	setr13(PsrMirq, m->sirq);
	setr13(PsrMabt, m->sabt);
	setr13(PsrMund, m->sund);
could do the job, since they always point to the same place.

I'll try that.
thanks a lot.


[-- Attachment #2: Type: message/rfc822, Size: 1628 bytes --]

From: presotto@plan9.bell-labs.com
To: nemo@gsyc.escet.urjc.es, 9fans@cse.psu.edu
Subject: Re: [9fans] 2 bitsy questions
Date: Thu, 7 Jun 2001 17:14:25 -0400
Message-ID: <20010607211434.27241199F1@mail.cse.psu.edu>

There is an r13 for each mode.  On traps, we always trampoline to svc
mode so that the r13 registers are always pointing to the same place
when we are interruptible.  You will definitely lose this state in
idle/sleep mode. I'll look at it a little more closely tonight.

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [9fans] 2 bitsy questions
@ 2001-06-07 21:14 presotto
  0 siblings, 0 replies; 3+ messages in thread
From: presotto @ 2001-06-07 21:14 UTC (permalink / raw)
  To: nemo, 9fans

There is an r13 for each mode.  On traps, we always trampoline to svc
mode so that the r13 registers are always pointing to the same place
when we are interruptible.  You will definitely lose this state in
idle/sleep mode. I'll look at it a little more closely tonight.


^ permalink raw reply	[flat|nested] 3+ messages in thread

* [9fans] 2 bitsy questions
@ 2001-06-07 18:38 F.J.Ballesteros
  0 siblings, 0 replies; 3+ messages in thread
From: F.J.Ballesteros @ 2001-06-07 18:38 UTC (permalink / raw)
  To: 9fans

1) I think that the kernel saves all the processor register set during
its way down to the on/off interrupt handler, that's despite the banked
registers for the different processor modes. Is that true? Or should I
also save/recover registers for other modes (irq, und, ...)?  In
particular I suspect of r13 for the different modes.

Although the kernel seems to come back fine after a suspend, if I
let the system run it panics in syscall or gets frozen.

2) When I reactivate the power of the lcd, the screen seems to get back
but the colors are kind of `degraded' after a suspend in `text mode'.
If I suspend once the system creates a window, the screen background
comes back, but the text is  gone.  Do any of you know of a fine way to
get the display back?


I'll reread the code to see if I can find out the answers, so don't
hesitate to answer if you don't have the time.

thanks anyway



^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2001-06-08  7:09 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2001-06-08  7:09 [9fans] 2 bitsy questions F.J.Ballesteros
  -- strict thread matches above, loose matches on Subject: below --
2001-06-07 21:14 presotto
2001-06-07 18:38 F.J.Ballesteros

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).