I have began to use hwloc, and I found it quite nice to use, but I do
have some comments and questions:
1) redundancy: for several operations, looping there are several ways
to to the same thing.
it is nice support several paradigms, but these occupy space and will
be very difficult to
Thanks for the quick answers!
On 29-set-09, at 16:59, Samuel Thibault wrote:
Fawzi Mohamed, le Tue 29 Sep 2009 16:39:47 +0200, a écrit :
Maybe I worry too much, but with machines with 1'000 of processor
coming, and maybe wanting local restricted copies to know the
topology
of the whole
Hi Samuel,
On 29-set-09, at 18:14, Samuel Thibault wrote:
Fawzi Mohamed, le Tue 29 Sep 2009 17:39:17 +0200, a écrit :
so that in the future one could avoid storing it at least in the
deepest levels where it is easy and relatively cheap to generate (and
where one would have the largest savings
On 29-set-09, at 20:43, Samuel Thibault wrote:
Fawzi Mohamed, le Tue 29 Sep 2009 20:39:02 +0200, a écrit :
It comes down to what you want to have, if you think you might want
to
go the sparse full granularity way then indeed alloc/copy/free should
be added
Yes, that's my point: do we
On 30-set-09, at 09:20, Brice Goglin wrote:
Fawzi Mohamed wrote:
1) a fully hierarchical representation of the machine/hardware where
each level is a partition, and each level fully covers the previous
one (from any node you go through all levels using father/childrens,
father/child are just
, I think it might be worthwhile to keep
something size+pointer so that if the size is small say less than
size_t the cpuset is stored where otherwise there is the pointer...
or
something like that.
Indeed I would keep a minimal struct...
Especially with a large number of OS processor
On 5-ott-09, at 15:31, Samuel Thibault wrote:
Fawzi Mohamed, le Mon 05 Oct 2009 15:23:23 +0200, a écrit :
well you assume you have a single copy of the whole system structure,
I am not sure that would be the case, and while the memory per core
is
growing, the memory per thread
On 5-ott-09, at 16:02, Jeff Squyres wrote:
On Oct 5, 2009, at 9:23 AM, Fawzi Mohamed wrote:
yes you are right, I was unclear, I meant that I would pass a cpu_set
struct by value (not always pass a pointer).
If one wants to later migrate to passing just a pointer, then
internally this struct
On 2-apr-10, at 16:16, Samuel Thibault wrote:
Fawzi Mohamed, le Fri 02 Apr 2010 12:34:58 +0200, a écrit :
flags attributes:
C bitfields are used, normally they are avoided because they are slow
(a compiler can bit or at compile time the constants and check/set
several at once.
Speed
On 2-apr-10, at 16:17, Samuel Thibault wrote:
Fawzi Mohamed, le Fri 02 Apr 2010 14:38:26 +0200, a écrit :
I would take advantage more info about the possible numa node
connectivity (to
know where to steal tasks),
This is on the TODO list, although (as Brice mentioned) OS often do
Always a bit sow, but here are my comments after the latest changes (I
wake up when I get time to update my bindings...)
HWLOC_OBJ_GROUP
nice!, I like it, it makes my job easier, as one of the main things I
use hwloc for is exactly having groups of uniform latency.
I suppose that those
11 matches
Mail list logo
