ACPI and APM

ACPI and APM

[valvoline]

hi all,

i follow the acpi thread under linux, since the first versions,
actually i'm running a 2.4.19 box, under an asus L3860C laptop.

I've tried the last patch for this kernel, and all works fine!
(fans status, temperature, cpu throttling, etc.)...but sleep mode,
doesn't work. i've tried the software suspend patch (swusp), but
without any success.

the system doesn't do nothing, even if i can see five sleep modes
with a 'cat /proc/acpi/sleep'; it show me: S0, S1, S2, S3, S4, S5.
I've checked for the ACPI_CONFIG_SLEEP, flag, and it's set to 'y'.

I've tried all the five states, with a 'echo n > /proc/acpi/sleep',
where n is the number (1,2,3,4,5,0), but without any success: It
does nothing.

reading this ml, i've learned, about a not well support of sleep states
under 2.4.x, and many gurus suggests to use a 2.5.LAST kernel to use this
functions.

since, i need a well-known kernel, and the 2.4.19 is a very stable and
efficent one for me, i've tried another way, to solve the problem:

i've hacked the acpi.c and apm.c drivers files under the arch/i386/kernel,
and i've bypassed some controls, to load both apm and acpi. Next, i've
disabled the Battery, AC-Adapter and Button flags, from ACPI config.
Under APM config, i've enabled the 'PM at boot time' and 'Enable console
Blanking using APM'. 

the system seems to work properly (fans works fine, and are triggered wells),
the thermal zone works fine, and....apmd suspend/standby WORKS!

there're some problems, that i don't know, doing something like above ?
suggests ?, comments ?...any will be welcome!


thanx in advance.

-- 
[ valvoline :: VRL Team :: s0ftpj :: freaknet Medialab :: GPG key available  ]
[ key fingerprint :: - :: B7E2 48BC 705F AE8F 9ABE  E422 076A 2561 1D67 B4DD ]
[ GPG key available on keyserver :: pgp.mit.edu :: with keyID :: 1D67B4DD :: ]


-------------------------------------------------------
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone?  Get a new here for FREE!
https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390

Re: ACPI and APM

[anders-VYFdHW3tBKJknbxzx/v8hQ]

[-- Attachment #1: Type: text/plain, Size: 1085 bytes --]

* valvoline <valvoline-KGm45+8e4WzOeuqD6ShaSA@public.gmane.org> [2002-08-24 22:45:40]:
> [...]
> 
> i've hacked the acpi.c and apm.c drivers files under the arch/i386/kernel,
> and i've bypassed some controls, to load both apm and acpi. Next, i've
> disabled the Battery, AC-Adapter and Button flags, from ACPI config.
> Under APM config, i've enabled the 'PM at boot time' and 'Enable console
> Blanking using APM'. 
> 
> the system seems to work properly (fans works fine, and are triggered wells),
> the thermal zone works fine, and....apmd suspend/standby WORKS!
> 
> there're some problems, that i don't know, doing something like above ?
> suggests ?, comments ?...any will be welcome!
> 

As far as I know if you enable both acpi and apm it just uses apm, it will
even tell you during boot (dmesg please). I don't know what you hacked
(since you didn't include a patch) but it just seems as if suspend works
with apm and not acpi, like it does with my toshiba.


> thanx in advance.

--
Anders Rune Jensen
anders-VYFdHW3tBKJknbxzx/v8hQ@public.gmane.org

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: ACPI and APM

[valvoline]

[-- Attachment #1: Type: text/plain, Size: 1851 bytes --]

On Sat, 24 Aug 2002 18:12:58 -0300 (BRT)
Christian Lyra <lyra-kkgK+NGu78pV51zT55DYLA@public.gmane.org> wrote:



> 	I would sugest that you didnt disable button flags in ACPI, and
> just create a script that on a button event generated from ACPI to just
> call apm -s (I think APM doesnt understand the Lid Button).

It's a great idea! :)...i'll try as soon as possible, and i'll send you
my feedback.

> to test your hack/patch. May you sent me a patch or the guidelines so I
> can test it too?

you can download the patch from this url:

http://www.vrlteam.org/patch-apm_acpi_010.tgz

the patch contain both my "kernel hacking" and the official
20020821-patch.diff, so you have to apply the patch to a clean 2.4.19
kernel tree.

after that you have to configure your kernel. enable both APM and
ACPI features. i've flagged the following:

Under APM Support:

- Enable PM at boot time
- Enable console blanking using APM

"Allow Interrupts during APM BIOS calls" and "Use real mode APM BIOS
call to power off" doesn't work for me (the system become instable during
apm --suspend or apm --standby calls, and a shutdown -h now, command
produce a reboot of the machine...), just try your machine, and give me 
your feedback.

Under ACPI Support:

- Enable ACPI Support
- Button 
- Fan
- Processor
- Thermal Zone


I've disable the "AC Adapter" and "Battery" to avoid conflicts with
APM support, but again, you can try.

After that, recompile your kernel, and get a try on it. you'll notice,
that "sleep" voice under /proc/acpi, will not be available, but
APM --suspend and --standby features, will works fine!


-- 
[ valvoline :: VRL Team :: s0ftpj :: freaknet Medialab :: GPG key available  ]
[ key fingerprint :: - :: B7E2 48BC 705F AE8F 9ABE  E422 076A 2561 1D67 B4DD ]
[ GPG key available on keyserver :: pgp.mit.edu :: with keyID :: 1D67B4DD :: ]

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: ACPI and APM

[valvoline]

On Sat, 24 Aug 2002 22:53:41 +0200
anders-VYFdHW3tBKJknbxzx/v8hQ@public.gmane.org wrote:

> As far as I know if you enable both acpi and apm it just uses apm, it will
> even tell you during boot (dmesg please). I don't know what you hacked

i've uploaded a patch for 2.4.19 with both my modding and the official 20020821-2.4.19.patch,
you have to patch a clean 2.4.19 kernel to make it works. with this modding, the kernel skip
the apm & acpi check during boot, and will map/load both apm and acpi system management.

you can download the patch from this url: http://www.vrlteam.org/patch-apm_acpi_010.tgz

> (since you didn't include a patch) but it just seems as if suspend works
> with apm and not acpi, like it does with my toshiba.

yep. it's true. but since acpi support is not 100% completed, and since apm support lacks of
many features, i'm trying to use both to obtain a 99.9% featured box...so, i'll use
fans and thermal zone from ACPI; battery and sleep states from APM.

-- 
[ valvoline :: VRL Team :: s0ftpj :: freaknet Medialab :: GPG key available  ]
[ key fingerprint :: - :: B7E2 48BC 705F AE8F 9ABE  E422 076A 2561 1D67 B4DD ]
[ GPG key available on keyserver :: pgp.mit.edu :: with keyID :: 1D67B4DD :: ]


-------------------------------------------------------
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone?  Get a new here for FREE!
https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390

Re: ACPI and APM

[Ducrot Bruno]

On Sun, Aug 25, 2002 at 12:00:34PM +0200, valvoline wrote:
> On Sat, 24 Aug 2002 18:12:58 -0300 (BRT)
> Christian Lyra <lyra-kkgK+NGu78pV51zT55DYLA@public.gmane.org> wrote:
> 
> 
> 
> > 	I would sugest that you didnt disable button flags in ACPI, and
> > just create a script that on a button event generated from ACPI to just
> > call apm -s (I think APM doesnt understand the Lid Button).
> 
> It's a great idea! :)...i'll try as soon as possible, and i'll send you
> my feedback.
> 
> > to test your hack/patch. May you sent me a patch or the guidelines so I
> > can test it too?
> 
> you can download the patch from this url:
> 
> http://www.vrlteam.org/patch-apm_acpi_010.tgz
> 

Some day, I will resend the S4BIOS patch, but for now I am a little busy :)

It add the suspend to disk via a similar method as for APM, but with more
ACPI compatible way, I think.

-- 
Ducrot Bruno
http://www.poupinou.org        Page profaissionelle
http://toto.tu-me-saoules.com  Haume page


-------------------------------------------------------
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone?  Get a new here for FREE!
https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390

Re: ACPI and APM

[marcogi-VGgt2q2+T+FeoWH0uzbU5w]

On Sun, 25 Aug 2002, valvoline wrote:

> On Sat, 24 Aug 2002 18:12:58 -0300 (BRT)
> Christian Lyra <lyra-kkgK+NGu78pV51zT55DYLA@public.gmane.org> wrote:
>
>
>
> > 	I would sugest that you didnt disable button flags in ACPI, and
> > just create a script that on a button event generated from ACPI to just
> > call apm -s (I think APM doesnt understand the Lid Button).
>
> It's a great idea! :)...i'll try as soon as possible, and i'll send you
> my feedback.
>
> > to test your hack/patch. May you sent me a patch or the guidelines so I
> > can test it too?
>
> you can download the patch from this url:
>
> http://www.vrlteam.org/patch-apm_acpi_010.tgz
>
> the patch contain both my "kernel hacking" and the official
> 20020821-patch.diff, so you have to apply the patch to a clean 2.4.19
> kernel tree.

Are you sure that your patch is againt clean 2.4.19 kernel?
Or your patch is not so clean ;)

> after that you have to configure your kernel. enable both APM and
> ACPI features. i've flagged the following:
>
> Under APM Support:
>
> - Enable PM at boot time
> - Enable console blanking using APM
>
> "Allow Interrupts during APM BIOS calls" and "Use real mode APM BIOS
> call to power off" doesn't work for me (the system become instable during
> apm --suspend or apm --standby calls, and a shutdown -h now, command
> produce a reboot of the machine...), just try your machine, and give me
> your feedback.
>
> Under ACPI Support:
>
> - Enable ACPI Support
> - Button
> - Fan
> - Processor
> - Thermal Zone
>
>
> I've disable the "AC Adapter" and "Battery" to avoid conflicts with
> APM support, but again, you can try.
>
> After that, recompile your kernel, and get a try on it. you'll notice,
> that "sleep" voice under /proc/acpi, will not be available, but
> APM --suspend and --standby features, will works fine!

I tried enabling both APM and ACPI. I patched a clean 2.4.19 kernel
using acpi-20020821-2.4.19 and then I commented out PM_IS_ACTIVE()
control in arch/i386/kernel/apm.c around 1902 line (is your kernel hack
limited to this only?).

But, when I issue apm --suspend my vaio fx601 doesn't suspend: lcd goes
blank and power doesn't off; instead, as side effect, cd spin on :/
My kernel as Local APIC disabled because, enabling it, acpi doesn't
shitch off my pc on halt and either hsf driver for soft modem doesn't work.

--
Regards,  Marco Gigante
Student at Pisa University - Computer Engineering



-------------------------------------------------------
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone?  Get a new here for FREE!
https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390

Re: ACPI and APM

[valvoline]

[-- Attachment #1: Type: text/plain, Size: 1597 bytes --]

On Mon, 26 Aug 2002 12:19:27 +0200 (CEST)
marcogi-VGgt2q2+T+FeoWH0uzbU5w@public.gmane.org wrote:

> I tried enabling both APM and ACPI. I patched a clean 2.4.19 kernel
> using acpi-20020821-2.4.19 and then I commented out PM_IS_ACTIVE()
> control in arch/i386/kernel/apm.c around 1902 line (is your kernel hack
> limited to this only?).

nop. u've to patch correctly the (HZ) suspend factor, if you don't the system
can't handle correctly the suspend/standby requests. on the 2.4.19 clean apm.c, 
there's a 3*(HZ) timeout define, mod' it to 5*(HZ). and set a default timeout
to 2*(HZ). 

later, you've to mod the PM part: if you use the suspend and standby function
of APM, you've to disable the SLEEP part of ACPI, to avoid conflicts between them.

you've tried my first patch ?, i've make a diff -ruN under a clean and a patched
kernel tree...do u have problems ?...please, let me know.


> My kernel as Local APIC disabled because, enabling it, acpi doesn't
> shitch off my pc on halt and either hsf driver for soft modem doesn't work.

me too. local APIC disabled. try to make all the changes, that i've suggested,
later, compile both ACPI and APM inside kernel (not module).
Under APM, enable ONLY: Enable PM at Boot Time, and Enable Console Blanking using APM.
Under ACPI, enable: Button, Fan, Processor, Thermal Zone. Skip the Rest.

let' me know.
-- 
[ valvoline :: VRL Team :: s0ftpj :: freaknet Medialab :: GPG key available  ]
[ key fingerprint :: - :: B7E2 48BC 705F AE8F 9ABE  E422 076A 2561 1D67 B4DD ]
[ GPG key available on keyserver :: pgp.mit.edu :: with keyID :: 1D67B4DD :: ]


[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: ACPI and APM

[valvoline]

[-- Attachment #1: Type: text/plain, Size: 585 bytes --]

On Mon, 26 Aug 2002 12:19:27 +0200 (CEST)
marcogi-VGgt2q2+T+FeoWH0uzbU5w@public.gmane.org wrote:


> using acpi-20020821-2.4.19 and then I commented out PM_IS_ACTIVE()
> control in arch/i386/kernel/apm.c around 1902 line (is your kernel hack
> limited to this only?).

also, u've to patch drivers/acpi, to avoid PM-acpi vs PM-apm conflicts...

-- 
[ valvoline :: VRL Team :: s0ftpj :: freaknet Medialab :: GPG key available  ]
[ key fingerprint :: - :: B7E2 48BC 705F AE8F 9ABE  E422 076A 2561 1D67 B4DD ]
[ GPG key available on keyserver :: pgp.mit.edu :: with keyID :: 1D67B4DD :: ]

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

ACPI and APM

[Marc Giger]

Hi Andrew, Hi All

I raced into a strange thing.

What means the boot option acpi=off? Shouldn't it switch off the whole acpi subsystem?

I ask this because I had some troubles in the past. I thought I could activate  ACPI and APM in the kernel config and then I can either activate ACPI or APM at boot time.

If I activate APM and disable ACPI (acpi=off) I have some troubles with the suspend (suspend to ram and suspend to disk). Most of the time the machine hangs on resume and sometimes it also hangs by going to suspend.

This NEVER happens if I build a kernel without ACPI!

Seen with 2.4.19/20, other Versions not tested.

A short statement would be appreciated. 

Thank you!

Greets,

Marc


-------------------------------------------------------
This SF.net email is sponsored by: Get the new Palm Tungsten T 
handheld. Power & Color in a compact size! 
http://ads.sourceforge.net/cgi-bin/redirect.pl?palm0002en