Hello Pavan,
This was indeed changed a long time ago when Jeff brought the embedding
support. I have nothing against bumping prereq to 2.64. But I'd rather
have Jeff look at the missing "-I m4".
Brice
Le 28/11/2011 04:19, Pavan Balaji a écrit :
>
> [reviving an ancient email thread]
>
> I was po
Hi Brice,
On 11/28/2011 10:24 PM, Brice Goglin wrote:
This was indeed changed a long time ago when Jeff brought the embedding
support. I have nothing against bumping prereq to 2.64. But I'd rather
have Jeff look at the missing "-I m4".
Ok. I upgraded to newer versions of autoconf on most of my
This question may be more about understanding NUMA (which I barely do) than
about hwloc, but perhaps you can help anyway.
I have a customer with some HP Proliant DL580 G7 servers. HP supplied them
with a block diagram of their system, and it shows two of the NUMA nodes
connected to the PCI devices
Le 28/11/2011 22:34, Guy Streeter a écrit :
> This question may be more about understanding NUMA (which I barely do) than
> about hwloc, but perhaps you can help anyway.
>
> I have a customer with some HP Proliant DL580 G7 servers. HP supplied them
> with a block diagram of their system, and it sho
I think the real question is: why do we have both config/ and m4/ ?
There's only a single .m4 file in m4/, but several in config/. I vote that
m4/decl.m4 should move to config/decl.m4, and then this issue goes away.
Specifically: I think the reason we don't specify both directories in
Makefile
On 11/28/2011 03:45 PM, Brice Goglin wrote:
...
> Current Intel platforms have 2 QPI links going to I/O hubs. Most servers
> with many sockets (4 or more) thus have each I/O hub connected to only 2
> processors directly, so their distance is "equal" as you say.
>
> However, some BIOS report invali
Le 28/11/2011 23:07, Jeff Squyres a écrit :
> I think the real question is: why do we have both config/ and m4/ ?
>
> There's only a single .m4 file in m4/, but several in config/. I vote that
> m4/decl.m4 should move to config/decl.m4, and then this issue goes away.
>
> Specifically: I think the
Le 28/11/2011 23:07, Guy Streeter a écrit :
> On 11/28/2011 03:45 PM, Brice Goglin wrote:
> ...
>> Current Intel platforms have 2 QPI links going to I/O hubs. Most servers
>> with many sockets (4 or more) thus have each I/O hub connected to only 2
>> processors directly, so their distance is "equal
Ok. I'll commit on trunk, v1.3, and just for good measure, v1.2 and v1.2-ompi.
On Nov 28, 2011, at 5:15 PM, Brice Goglin wrote:
> Le 28/11/2011 23:07, Jeff Squyres a écrit :
>> I think the real question is: why do we have both config/ and m4/ ?
>>
>> There's only a single .m4 file in m4/, but
Creating nightly hwloc snapshot SVN tarball was a success.
Snapshot: hwloc 1.4a1r3999
Start time: Mon Nov 28 21:01:01 EST 2011
End time: Mon Nov 28 21:04:17 EST 2011
Your friendly daemon,
Cyrador
Creating nightly hwloc snapshot SVN tarball was a success.
Snapshot: hwloc 1.2.3a1r4001
Start time: Mon Nov 28 21:04:17 EST 2011
End time: Mon Nov 28 21:07:02 EST 2011
Your friendly daemon,
Cyrador
Creating nightly hwloc snapshot SVN tarball was a success.
Snapshot: hwloc 1.3.1rc1r4000
Start time: Mon Nov 28 21:07:03 EST 2011
End time: Mon Nov 28 21:10:04 EST 2011
Your friendly daemon,
Cyrador
12 matches
Mail list logo