qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [PATCH v2 for-2.3] numa: pc: fix default VCPU to node m


From: Igor Mammedov
Subject: Re: [Qemu-devel] [PATCH v2 for-2.3] numa: pc: fix default VCPU to node mapping
Date: Thu, 19 Mar 2015 17:10:11 +0100

On Thu, 19 Mar 2015 17:05:13 +0100
Andreas Färber <address@hidden> wrote:

> Am 18.03.2015 um 17:38 schrieb Igor Mammedov:
> > since commit
> >    dd0247e0 pc: acpi: mark all possible CPUs as enabled in SRAT
> > Linux kernel actually tries to use CPU to Node mapping from
> > QEMU provided SRAT table instead of discarding it, and that
> > in some cases breaks build_sched_domains() which expects
> > sane mapping where cores/threads belonging to the same socket
> > are on the same NUMA node.
> > 
> > With current default round-robin mapping of VCPUs to nodes
> > guest ends-up with cores/threads belonging to the same socket
> > being on different NUMA nodes.
> > 
> > For example with following CLI:
> > qemu-kvm -m 4G -smp 5,sockets=2,cores=4,threads=1,maxcpus=8 \
> >          -numa node,nodeid=0 -numa node,nodeid=1
> > 2.6.32 based kernels will hang on boot due to incorrectly build
> > sched_group-s list in update_sd_lb_stats()
> > so comment in QEMU justifying dumb default mapping:
> >  "
> >   guest OSes must cope with this anyway, because there are BIOSes
> >   out there in real machines which also use this scheme.
> >  "
> > isn't really valid.
> > 
> > Replacing default mapping with a manual, where VCPUs belonging to
> > the same socket are on the same NUMA node, fixes issue for
> > guests which can't handle nonsense topology i.e. changing CLI to:
> >   -numa node,nodeid=0,cpus=0-3 -numa node,nodeid=1,cpus=4-7
> > 
> > So instead of simply scattering VCPUs around nodes, map
> > the same socket VCPUs to the same NUMA node, which is what
> > guest would expect from a sane hardware/BIOS.
> > 
> > Signed-off-by: Igor Mammedov <address@hidden>
> > ---
> > v2:
> >   - add machine callback cpu_index_to_socket_id() and use it
> >     instead of stub approach
> > ---
> >  hw/i386/pc.c          |  9 +++++++++
> >  include/hw/boards.h   |  5 +++++
> >  include/sysemu/numa.h |  3 ++-
> >  numa.c                | 18 +++++++++++++-----
> >  vl.c                  |  2 +-
> >  5 files changed, 30 insertions(+), 7 deletions(-)
> 
> Looks great to me now, the hook name with _socket_id is perfect,
> 
> Reviewed-by: Andreas Färber <address@hidden>
> 
> but can we do that in three steps please? "machine:" adding callback and
> default implementation, "numa:" switching to use it and "pc:" overriding
> the new callback - not only nicer subjects but easier to cherry-pick and
> bisect then.
sure

> 
> Regards,
> Andreas
> 




reply via email to

[Prev in Thread] Current Thread [Next in Thread]