Control: tag -1 patch
Control: tag -1 - moreinfo

On Wed, 2013-02-13 at 12:40 +0100, Bernhard Schmidt wrote:
> On 11.02.2013 02:09, Ben Hutchings wrote:
> 
> Hello Ben,
> 
> > Control: tag -1 upstream moreinfo
> > 
> > On Wed, 2013-02-06 at 18:02 +0100, Bernhard Schmidt wrote:
> >> Package: src:linux
> >> Version: 3.2.35-2
> >> Severity: normal
> >>
> >> Adding additional RAM to a virtual machine running Debian Wheezy on
> >> VMware ESXi 5.0 often, but not always leads to the attached backtrace.
> >>
> >> If that happens, the system has considerably less new (offline) memory 
> >> banks in /sys/devices/system/memory/memory* than it should have, and 
> >> setting all available memory banks online does not give all the memory
> >> expected.
> > [...]
> > 
> > Please test whether the attached patch fixes this.  Instructions for
> > building a patched kernel package are in the Debian kernel handbook:
> > <http://kernel-handbook.alioth.debian.org/ch-common-tasks.html#s-common-official>.
> 
> It looks good on Wheezy now. Hard to tell for sure because it did not
> always happen, especially not with freshly booted devices, but we did
> install the kernel on a few Wheezy boxes and put load on them, and I did
> not observe the backtrace anymore.

Yes, it will depend on how full (and how fragmented) memory is when you
try to add hotplug.

> Squeeze is another story, but I think there is another problem as well.
> Previously we sometimes saw the backtrace, sometimes just the following
> message.
> 
> 
> [   44.220000] VMCIUtil: Updating context id from 0x7a3c21d6 to
> 0x7a3c21d6 on event 0.
> [   44.252000] Hotplug Mem Device
> [   44.252000] System RAM resource 20000000 - 27ffffff cannot be added
> [   44.252000] ACPI:memory_hp:add_memory failed
> [   44.252000] ACPI:memory_hp:Error in acpi_memory_enable_device
> [   44.252000] acpi_memhotplug: probe of PNP0C80:00 failed with error -22

This is 'invalid argument' - wonder where that's coming from?

> [   44.252000]  driver data not found
> [   44.252000] ACPI:memory_hp:Cannot find driver data
> [   44.268000] Hotplug Mem Device
> [   44.268000] init_memory_mapping: 0000000028000000-0000000030000000
> [   44.268000]  0028000000 - 0030000000 page 2M
> [   44.280000]  [ffffea00008c0000-ffffea0000abffff] PMD ->
> [ffff88001f200000-ffff88001f3fffff] on node 0
> [   44.280000] Hotplug Mem Device
> [   44.284000] init_memory_mapping: 0000000030000000-0000000038000000
> [   44.284000]  0030000000 - 0038000000 page 2M
> [   44.284000]  [ffffea0000a00000-ffffea0000bfffff] PMD ->
> [ffff88001e400000-ffff88001e5fffff] on node 0
> [   44.340000] Hotplug Mem Device
> [   44.340000] init_memory_mapping: 0000000038000000-0000000040000000
> [   44.340000]  0038000000 - 0040000000 page 2M
> 
> We did not observe the backtrace anymore, but the "driver data not
> found" is still there.
> 
> So I think the patch fixes the backtrace (allocation error) on both
> squeeze and wheezy, but squeeze has a second issue. I'll go through the
> bug reports and open a new one.

Right, this is definitely a separate bug.

Ben.

-- 
Ben Hutchings
We get into the habit of living before acquiring the habit of thinking.
                                                              - Albert Camus

Attachment: signature.asc
Description: This is a digitally signed message part

Reply via email to