linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Maciej Freudenheim <fahren@student.uci.agh.edu.pl>
To: linux-kernel@vger.kernel.org
Cc: pavel@ucw.cz, davej@codemonkey.org.uk, mochel@osdl.org
Subject: Suspend and AGP in 2.6.0-test9
Date: Mon, 3 Nov 2003 13:34:41 +0100	[thread overview]
Message-ID: <20031103123441.GA770@student.uci.agh.edu.pl> (raw)

Hi!

2.6.0-test9 is the first kernel with *almost* working suspend to ram
and suspend to disk.

Well, almost, because when i run X with AGP support enabled i can't
resume - either it hangs forever after displaying "Waiting for DMAs to
settle down..." message or it reboots right after displaying it.
When i force using of PCI bus for acceleration, then suspend (to ram as
well as to disk) with resume works perfectly - but of course, with disabled
agp everything is significantly slower.

So, my question is - is it known (and not fixable :) bug or it's
something weird and shouldn't happen ? As fair as I googled for similar
problems I have found that people usually have problems with DRI, it looks
like agp works ok for most of them :) However, on my laptop disabling
DRI doesn't help.

I have Dell Lattitude D600 with:
- ATI Technologies Inc Radeon R250 Lf [Radeon Mobility 9000 M9]
- Intel Corp. 82855PM Processor to AGP Controllerntel Corp. 82855PM
- Latest BIOS (A06)

Here is full output from dmesg, lspci and my current kernel config:

 http://student.uci.agh.edu.pl/~fahren/dmesg
 http://student.uci.agh.edu.pl/~fahren/lspci
 http://student.uci.agh.edu.pl/~fahren/config.gz


I've tried both 2.6.0-test9 and 2.6.0-test9-mm1 with:
- XFree's radeon drivers (from XFree86's cvs) - 
  As i already wrote, everything work ok untill i load intel-agp
  module (or don't force "BusType" to "PCI" in XF86Config) - enabling/
  disabling DRI doesn't seem to affect anything.

- ATI's fglrx drivers (ver. 3.2.8) - the same, with enabled
  "UseInternalAGPGART" i can't resume. When disabled and without loaded
  intel-agp module i can.

I'm suspending by writing to /proc/acpi/sleep - writing to
/sys/power/state returns nice "call trace" screen. 
On older versions of kernel (<= 2.6.0-test8*) suspend just doesn't work
- /proc/acpi/sleep behaves like /dev/null :)

If anyone know sollution for it *please* let me know. :>

BTW, standby mode doesn't work at all (and never had on 2.6.0-test* for
me, with 2.4 it is ok) - it goes to sleep, but when resuming it hangs
after displaying "PM: Finishing up."


Maciej Freudenheim.

             reply	other threads:[~2003-11-03 12:34 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-11-03 12:34 Maciej Freudenheim [this message]
2003-11-03 13:28 ` Suspend and AGP in 2.6.0-test9 Dave Jones
     [not found] <NMa8.7uR.11@gated-at.bofh.it>
     [not found] ` <NN6b.pY.5@gated-at.bofh.it>
2003-11-04 15:50   ` Ruben Puettmann
2003-11-04 16:05     ` Dave Jones
2003-11-04 16:21     ` Michael Clark
2003-11-05 18:36 Alex Deucher

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20031103123441.GA770@student.uci.agh.edu.pl \
    --to=fahren@student.uci.agh.edu.pl \
    --cc=davej@codemonkey.org.uk \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mochel@osdl.org \
    --cc=pavel@ucw.cz \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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).