On one of my Linux machines I'm getting a crash if I do a
hwloc_topology_load() while ignoring HWLOC_OBJ_NODE.
#0 hwloc_cpuset_orset (set=0x605410, modifier_set=0x0) at cpuset.c:410
#1 0x7fc87f943146 in hwloc__setup_misc_level_from_distances (
topology=0x604010, nbobjs=4, objs=0x7fff82
It looks like I spoke too soon on the fix. That solves the problem
but it keeps the Miscs from being created and in some situations I'd
like to keep the Miscs but not the nodes. The attached patch does that
for me.
Samuel Thibault wrote:
> Michael Raymond, le Thu 19 Nov 2009 14:33:49
might want to place processes on all the Cores in a
Misc and then move to the next Misc. A topology tree that looks like
System -> Misc -> Core makes that easy. Having Nodes in there just adds
unneeded complexity.
Feel free to come up with a cleaner patch. :)
Samuel Thibault wrote:
>
Yes. Here's output from a small one of those with only 2 pre-release
blades.
Brice Goglin wrote:
> Michael Raymond wrote:
>> Our architecture has blades with two Nehalems on them, and the blades
>> are connected together in a CC-NUMA fashion. Each Nehalem shows up
NUMAGROUP sounds fine to me. Misc appears to be working for me though
and I'd like to start shipping hwloc on all our boxes in the next few
months.
Samuel Thibault wrote:
> Hello,
>
> Michael Raymond, le Fri 20 Nov 2009 08:43:10 -0600, a écrit :
>> In one pattern I
I wouldn't say no to the ability to flatten Misc / NUMA. As to the
current issue though, does my patch look acceptable? Thanks.
Samuel Thibault wrote:
> Samuel Thibault, le Fri 20 Nov 2009 15:54:43 +0100, a écrit :
>> Introduce several numagroup types? How many? That's not easy to
>> answer.
I don't know what has been announced and what hasn't, but hwloc
nicely solves a problem for us and I intend to make sure that it works
on all our hardware.
On 20 Nov 2009, at 23:55 , Chris Samuel wrote:
Hi Michael,
- "Michael Raymond" wrote:
Our architecture
At the moment I'm thinking SLES11 (and RHEL6) RPMs of 0.9.3 / TOT
installed in /opt[/sgi]/hwloc. I'd also add module support.
Samuel Thibault wrote:
> How do you intend to ship hwloc? 0.9.2 with a couple of backported
> patches like this one?
>
> Samuel
> _
bles. For hwloc I could:
module load hwloc
lstopo
This keeps hwloc out of /usr thus keeping the Linux File Hierarchy
Standard (LFHS) happy.
Samuel Thibault wrote:
> Michael Raymond, le Mon 30 Nov 2009 09:23:02 -0600, a écrit :
>> At the moment I'm thinking SLES11 (and RH
As of SGI ProPack 7 dplace uses hwloc internally to specify stride
patterns. For example:
mpirun -np 8 dplace -c SC a.out
means to pin ranks to every core inside a socket before jumping to the
next socket and doing the same.
From the man page:
For striding patterns any
t; cores so as to maximize memory bandwidth or cache sharing ?
>
> Brice
>
>
>
> Michael Raymond wrote:
>> As of SGI ProPack 7 dplace uses hwloc internally to specify stride
>> patterns. For example:
>>
>> mpirun -np 8 dplace -c SC a.out
>>
I was doing some testing on what's in SVN and I found that the
topology discovery code has a problem when running within a cpuset. On
my 2.6.32 SLES11SP1 box the code calls hwloc_read_linux_cpuset_mask()
with a cgroup mount point and tries to read
/dev/cpuset//cpuset.[cpus|mems]. Those files do
cpuset defaults
0 0
..
Brice Goglin wrote:
> Le 22/07/2010 14:53, Michael Raymond a écrit :
>> I was doing some testing on what's in SVN and I found that the
>> topology discovery code has a problem when running within a cpuset. On
>> my
Looks good, thanks.
Brice Goglin wrote:
> Le 23/07/2010 13:32, Michael Raymond a écrit :
>> I think you forgot to attach the patch.
>>
>
> Sorry, here's the patch :)
>
> Brice
>
>
--
Michael A. Raymond
Message Passing Toolkit Team
Silicon Graphics Inc
(651) 683-3434
14 matches
Mail list logo