Re: patch: acpi usermode suspend API change

On Thu, Jun 21, 2007 at 03:28:53PM -0700, Nate Lawson wrote:
Lars Engels wrote:
On Thu, Jun 21, 2007 at 09:29:28AM -0700, Nate Lawson wrote:
Lars Engels wrote:
And I just saw that every 5 second a new apm device is created!

# ls /dev
acd0 apm12 apm3 apm47 apm8
acpi apm13 apm30 apm48 apm9
ad0 apm14 apm31 apm49 ata
ad0.journal apm15 apm32 apm5 atkbd0
ad0.journals1 apm16 apm33 apm50 bpf0
ad0.journals2 apm17 apm34 apm51 bpf1
ad0.journals3 apm18 apm35 apm52 bpsm0
ad0.journals4 apm19 apm36 apm53 cardbus0.cis
ad0.journals4a apm2 apm37 apm54 cd0
ad0.journals4b apm20 apm38 apm55 cdrom@
ad0.journals4c apm21 apm39 apm56 console
ad0.journals4d apm22 apm4 apm57 consolectl
ad0.journals4e apm23 apm40 apm58 ctty
ad0.journals4f apm24 apm41 apm59 cuad0
agpgart apm25 apm42 apm6 cuad0.init
apm0 apm26 apm43 apm60 cuad0.lock
apm1 apm27 apm44 apm61 devctl
apm10 apm28 apm45 apm62 devstat
apm11 apm29 apm46 apm7 dri/

[...]
What are you running that opens /dev/apm? Maybe powerd?

I have a patch that handles reaping these devices, unfortunately it is
pending completion of another patch that allows devfs_destroy() from a
close() routine. That part is a core problem in FreeBSD, and it isn't
really something in acpi.

Yes, powerd is running. Do you need additional system info?

No, I'm just going to work around this and re-test the patch before
committing.

With your committed code, everything is working fine now. The power
button sets the system to S5, no apm[0-9] devices, and no more apm
cloning messages on the console.
Thanks, Nate!

Lars

Attachment: pgpEMoJTER8yU.pgp
Description: PGP signature