FS#36435 - [linux] 3.10.x not enough stolen space for compressed buffer (need 4096000 bytes), disabling

Attached to Project: Arch Linux
Opened by Robert Orzanna (orschiro) - Tuesday, 06 August 2013, 20:32 GMT
Last edited by Tobias Powalowski (tpowa) - Monday, 19 August 2013, 06:55 GMT
Task Type Support Request
Category Kernel
Status Closed
Assigned To Tobias Powalowski (tpowa)
Thomas Bächler (brain0)
Architecture x86_64
Severity Low
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 1
Private No

Details

Description:
`dmesg` outputs hundreds of the following two entries:

[ 2955.315434] [drm] hint: you may be able to increase stolen memory size in the BIOS to avoid this
[ 2955.348930] [drm] not enough stolen space for compressed buffer (need 4096000 bytes), disabling

I have not idea what this means but cannot find an option in the BIOS of my Thinkpad X200T to increase the memory size of my Graphics card `Intel Graphics Media Accelerator (GMA) 4500MHD`:

$ lspci -v -s 00:02.0
00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07) (prog-if 00 [VGA controller])
Subsystem: Lenovo Device 20e4
Flags: bus master, fast devsel, latency 0, IRQ 44
Memory at f2000000 (64-bit, non-prefetchable) [size=4M]
Memory at d0000000 (64-bit, prefetchable) [size=256M]
I/O ports at 1800 [size=8]
Expansion ROM at <unassigned> [disabled]
Capabilities: <access denied>
Kernel driver in use: i915
Kernel modules: i915

Additional info:

3.10.3-1-ARCH
xf86-video-intel 2.21.14-1
This task depends upon

Closed by  Tobias Powalowski (tpowa)
Monday, 19 August 2013, 06:55 GMT
Reason for closing:  Fixed
Comment by Michael Fuchs (mukl) - Wednesday, 14 August 2013, 12:43 GMT
Same on a Lenovo Edge 13

$ dmesg |tail
[ 2360.475430] [drm] not enough stolen space for compressed buffer (need 4325376 bytes), disabling
[ 2360.475437] [drm] hint: you may be able to increase stolen memory size in the BIOS to avoid this

$ lspci -v -s 00:02.0
00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07) (prog-if 00 [VGA controller])
Subsystem: Lenovo Device 21b4
Flags: bus master, fast devsel, latency 0, IRQ 44
Memory at f0000000 (64-bit, non-prefetchable) [size=4M]
Memory at d0000000 (64-bit, prefetchable) [size=256M]
I/O ports at 1800 [size=8]
Expansion ROM at <unassigned> [disabled]
Capabilities: <access denied>
Kernel driver in use: i915
Kernel modules: i915

$ uname -r
3.10.6-2-ARCH

Comment by Stefan Joosten (Ultraman) - Sunday, 18 August 2013, 12:58 GMT
I recently experienced this on my laptop (Sony Vaio VPCY11S1E), which has similar specs and uses the same graphics chipset (00:02.1 Display controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07) ). And my "lspci" output is very similar.

Are you by any change forcing/using FrameBuffer Compression for the i915 driver in the kernel bootline?
Because I was, by adding "i915.i915_enable_fbc=1" to my kernel bootline in the GRUB config.

By default fbc is disabled for older chipsets because it can cause problems. I removed the setting from my GRUB setting for the kernel bootline, ran "update-grub" and rebooted. The message now no longer appears in my kernel log, solving my 'problem'. So if you force enable fbc, try booting without it.
If you don't have a setting like it in your kernel bootline, you can try disabling fbc (using i915.i915_enable_fbc=0).

Note: I'm an ex-Arch user. I am on Debian Wheezy using kernel 3.10-0.bpo.2-amd64 from wheezy-backports. But this 'problem' is within the kernel and your hardware (and possibly your config), so not Arch specific.
Comment by Robert Orzanna (orschiro) - Sunday, 18 August 2013, 21:22 GMT
That was a good hint, thanks Stefan. Appending `i915.i915_enable_fbc=0` to the kernel line indeed 'resolves' the issue. Do I now have to fear any negative side effects with frame buffer compression being disabled?

Robert
Comment by Stefan Joosten (Ultraman) - Sunday, 18 August 2013, 21:44 GMT
There are no negative effects.
Framebuffer compression saves some memory bandwidth on screen refreshes, which in turn reduces power consumption a tiny bit. I did some testing on my laptop before I enabled it, only saved me about 0.1 Watt, so... don't think you will miss it. ;-)

I did some searching around, seems someone from Canonical did some testing as well. Results: http://zinc.canonical.com/~cking/power-benchmarking/background-colour-and-framebuffer-compression/results.txt

So no harm in disabling it and the power "lost" is hardly noticeable.
Comment by Robert Orzanna (orschiro) - Sunday, 18 August 2013, 21:55 GMT
Thanks for the further information. Fully satisfied. :)

I guess this can be closed then.

Robert

Loading...