[ltp] Direct Rendering's affect on battery life?

Marius Gedminas linux-thinkpad@linux-thinkpad.org
Fri, 7 Apr 2006 21:30:01 +0300


--kXdP64Ggrk/fb43R
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Wed, Apr 05, 2006 at 10:40:24PM +0200, Torsten Wolf wrote:
> * Laurent Gilson <pumpkin@gmx.de> [06-04-05 17:35]:
> >
> >1. the DRI/DRM-part of the driver keeps the CPU awake even if there is =
=20
> >nothing to do. The CPU will switch to lower frequencies w/o any problems=
, =20
> >but it will stay in the first 1-2 C-States. W/o DRI/DRM it can enter C3 =
or =20
> >even C4 which reduce the power needed to run the CPU. Have a look at =20
> >/proc/acpi/processor/CPU/power and the "bus master activity". Optimal fo=
r =20
> >a idle desktop is 00000000.
>=20
> cat /proc/acpi/processor/CPU/power=20
> active state:            C3
> max_cstate:              C8
> bus master activity:     00000000
> states:
>     C1:                  type[C1] promotion[C2] demotion[--] latency[001]=
 usage[00000010]
>     C2:                  type[C2] promotion[C3] demotion[C1] latency[001]=
 usage[11270220]
>    *C3:                  type[C3] promotion[--] demotion[C2] latency[085]=
 usage[08844140]
>=20
> glxinfo
> name of display: :0.0
> display: :0  screen: 0
> direct rendering: Yes
>=20
> R52, X300, fglrx 8.23.7, 2.6.15

Hmm.  T42, open-source radeon driver, 2.6.15

$ cat /proc/acpi/processor/CPU/power
active state:            C2
max_cstate:              C8
bus master activity:     2fffdfff
states:
    C1:                  type[C1] promotion[C2] demotion[--] latency[000] u=
sage[00000010]
   *C2:                  type[C2] promotion[C3] demotion[C1] latency[001] u=
sage[08252215]
    C3:                  type[C3] promotion[--] demotion[C2] latency[085] u=
sage[00000000]

Does usage[00000000] mean my CPU never enters C3?

$ glxinfo
name of display: :0.0
display: :0  screen: 0
direct rendering: Yes
=2E..
OpenGL renderer string: Mesa DRI Radeon 20050528 AGP 1x TCL
OpenGL version string: 1.2 Mesa 6.4.1


Marius Gedminas
--=20
Linux became only possible because 20 years of OS research was carefully
studied, analyzed, discussed and thrown away.
		-- Ingo Molnar

--kXdP64Ggrk/fb43R
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Digital signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)

iD8DBQFENq+pkVdEXeem148RAhtLAJ9/uHClCknzZcMafuAtiEUAsoidWwCfSJ7b
5sqs/QKc+x35Bf+T600kr4k=
=0RaX
-----END PGP SIGNATURE-----

--kXdP64Ggrk/fb43R--