On May 25, 2010, at 9:27 AM, Brice Goglin wrote:
> On 25/05/2010 15:23, Jeff Squyres wrote:
> > - 1.1 will be a "new feature" release. What features / branches do you
> > guys want to bring into the trunk and then create a 1.1 branch with?
>
> I think we have
take care of the junk to make nightly tarballs,
etc.
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
PUSUBSET_FULL;
> > - for(; iulongs_count; i++)
> > - res->ulongs[i] = val;
> > + for(i=0; iulongs_count; i++)
> > + res->ulongs[i] = ~HWLOC_CPUSUBSET_READULONG(set, i);
> >
> > res->infinite = !set->infinite;
>
Go to the main ompi web page - there's a link to the annopuncement from there.
-jms
Sent from my PDA. No type good.
From: hwloc-devel-boun...@open-mpi.org
To: hwloc-de...@open-mpi.org
Sent: Tue
/software/hwloc/v1.0/
Can you guys do some basic testing on the rc7 tarball? Let's aim for a release
tomorrow (Tuesday).
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
Sweet. :-)
On May 13, 2010, at 8:03 PM, Pavan Balaji wrote:
>
> 1.0rc6 seems OK. Thanks!
>
> -- Pavan
>
> On 05/12/2010 11:52 AM, Jeff Squyres wrote:
>> Pavan --
>> I just posted rc6 with the fixes for this. Can you confirm that it works
>> for
Cool. Let's aim for a Monday release, then (assuming everything works out ok
on Leopard).
On May 13, 2010, at 7:58 AM, Samuel Thibault wrote:
> Jeff Squyres (jsquyres), le Thu 13 May 2010 06:34:37 -0500, a écrit :
> > Is rc6 The One?
>
> I couldn't test on Leopard yet (th
aceful abort for non-C99 compilers
Jeff Squyres, le Wed 12 May 2010 12:52:01 -0400, a écrit :
> I just posted rc6 with the fixes for this.
I have built the windows binary, available on
http://dept-info.labri.fr/~thibault/tmp/hwloc-win32-build-1.0
; > Thanks,
> >
> > -- Pavan
> >
>
> --
> Pavan Balaji
> http://www.mcs.anl.gov/~balaji
> ___
> hwloc-devel mailing list
> hwloc-de...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-devel
>
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
e?
>
> Right, thanks.
>
> Samuel
> ___
> hwloc-devel mailing list
> hwloc-de...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-devel
>
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
htt
r = malloc(size);
> if (NULL == str) {
> -return -1;
> + /* Couldn't allocate?! Let's return the least surprising value. */
> + return 0;
> }
> va_start(ap, format);
> errno = 0;
> ___
> hwloc
Fixed -- thanks!
On May 6, 2010, at 11:18 AM, Christopher Samuel wrote:
> On 06/05/10 08:17, Jeff Squyres wrote:
>
> > hwloc 1.0rc5 is posted.
>
> Only one unchecked malloc() left, the one at line 41 of
> src/misc.c in hwloc_snprintf(). I believe the attach
e
> Email: sam...@unimelb.edu.au Phone: +61 (0)3 903 55545
> http://www.vlsci.unimelb.edu.au/
>
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
Fair enough; I was just curious if you want them over in v1.0. If you want to
move them over, go ahead and do so. We don't necessarily have to cut another
rc before final release if this is all that is left to be done...
On May 6, 2010, at 2:04 PM, Samuel Thibault wrote:
> Jeff Squyres,
hwloc 1.0rc5 is posted. Hopefully it can be the last! :-)
(Samuel -- can you make the windows binaries?)
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
er avoid that too :/.
>
> Samuel
> ___
> hwloc-devel mailing list
> hwloc-de...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-devel
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
Awesome; thanks Chris!
On May 3, 2010, at 8:57 PM, Christopher Samuel wrote:
> On 03/05/10 09:57, Jeff Squyres wrote:
>
> > 1.0rc4 is up.
>
> Running coccicheck on 1.0rc4 flags up this construct, I presume
> as an ambiguous construction:
>
&g
Is this helpful?
http://msdn.microsoft.com/en-us/library/ms724379(VS.85).aspx
You can use that to enumerate the ACPI firmware tables directly. Supposedly,
you should be able to discover topology information. Not very user friendly,
unfortunately.
On May 3, 2010, at 4:52 PM, Jeff Squyres
the amount of memory available -- not the total
memory. (do we document this somewhere, btw? I admit to not having reading
the windows docs...?)
On May 3, 2010, at 4:22 PM, Jeff Squyres (jsquyres) wrote:
> (I'm asking my local Microsoft contact)
>
> On May 3, 2010, at 4:15 PM, Samuel
Running lstopo on w7 (64 bit):
-
C:\Temp\hwloc>lstopo
Note: GetLogicalProcessorInformationEx was never tested yet!
-
Plus, the memory it reported was the currently free memory, not total memory.
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
h
Posted.
On May 3, 2010, at 10:37 AM, Samuel Thibault wrote:
> Jeff Squyres, le Mon 03 May 2010 09:57:09 -0400, a écrit :
> > 1.0rc4 is up. Send me a windows binary when you get a chance.
>
> http://dept-info.labri.fr/~thibault/tmp/hwloc-win32-build-1.0rc4.
1.0rc4 is up. Send me a windows binary when you get a chance.
On May 3, 2010, at 9:30 AM, Samuel Thibault wrote:
> Jeff Squyres, le Mon 03 May 2010 09:28:41 -0400, a écrit :
> > I see a flurry of windows-related activity over the weekend.
> >
> > Do you guys want me to c
> \htmlonly
> >
> > \endhtmlonly
> > ___
> > hwloc-svn mailing list
> > hwloc-...@open-mpi.org
> > http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-svn
> >
>
> ___
> hwloc-devel mailing list
> hwloc-de...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-devel
>
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
/
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
ferent OS indexes\n");
> return;
> ___
> hwloc-svn mailing list
> hwloc-...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-svn
>
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
at does not look like a header for me.
Right, but gcc complained when used with -std=c99 unless it was __typeof__. I
did not check to see if icc or pgcc accepted typeof. I read that text to be
"if you want portable code, use __typeof__ instead of typeof."
--
Jeff Squyres
jsquy...@cisc
I fixed some compiler warnings in r1986 -- I have mixed feelings about renaming
local variables named "index" and "y1". :-\
https://svn.open-mpi.org/trac/hwloc/changeset/1986
Thanks.
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www
and later freed?
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
r and the following
> services would not be available during the upgrade.
>
> - trac
> - subversion
> - svn check-in hook to trac
>
> Please let me know if you have any questions or issues of the this upgrade.
>
> --
> - DongInn
>
>
--
Jeff Squyres
jsquy.
n't (offhand) know of anyone outside of Open MPI who would use the
functionality.
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
t branch. It would be (unsigned) -1
> then, which isn't so bad when we mean "infinite".
My minor preference would be to avoid the cast. But it's a minor preference at
best.
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
e to different answers vs. how the external one is configured? (e.g.,
if OMPI is built using icc and the external copy was built with gcc)
In the external case, OMPI should just use the external's hwloc.h and whatever
decisions were already made there (sizes, types, etc.).
Make sense?
--
J
n it's ready.
Sorry, I was out yesterday.
Is this still your feeling -- want me to roll rc1?
Is everything else set, ABI-wise?
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
able embedding the m4 stuff). So the net "win" of not supporting m4
embedding might be a little simplification and a zero maintenance effort over
time -- vs. keeping what is done and maintaining it over time.
Thoughts?
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal inform
M_CONDITIONAL([HWLOC_DOXYGEN_BROKEN_SHORT_NAMES], [test
"$DOXYGEN_VERSION" = "1.6.2"])
Meaning: I hadn't re-run configure. I just upgraded doxygen in place and
didn't realize that we had something in hwloc keying off its version number.
--
Jeff Squyres
jsquy...@cisco.com
For corpor
t; ___
> hwloc-svn mailing list
> hwloc-...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-svn
>
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
;hwloc-de...@open-mpi.org>
Sent: Tue Mar 30 05:40:30 2010
Subject: Re: [hwloc-devel] Strange difference
On Sat, 27 Mar 2010 08:23:52 am Jeff Squyres wrote:
> I think that we need to differentiate between the different meanings of
> "CPU" here -- CPU could also mean "socket"
way, this is also what hwloc_type_string() would return. Unless
> we keep it unchanged and just hack lstopo to use its own stringified
> type name ?
I wouldn't mind the hack (too much), but it does seem a little inelegant. If
we hate everything else, let's settle on "PU".
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
obviously good with -v having more information. I guess my reaction was
because it wasn't just *more* information -- the *same* information was in a
different format, and that struck me as weird.
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
out "Proc" instead of "P"?
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
question again and again :)
>
> Brice
>
> _______
> hwloc-devel mailing list
> hwloc-de...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-devel
>
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
API changes?
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
ght?
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
d to say
(cut-n-paste error). I meant to say:
If we say that applications need to provide their own synchronization
between readers and writers, atomic writes **could still** be an issue,
right?
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
e through
documentation. If it turns out that people want hwloc to use the one-big-mutex
approach and make things like multiple writes safe, then we can always add that
later.
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
On Mar 11, 2010, at 3:14 PM, Bert Wesarg wrote:
> I do not know the policy from this library regarding thread safety, so
> I decided to be on the safe side here.
What *is* hwloc's policy about thread safety, anyway?
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to
nfigure.ac before
> calling the embedding code?
>
> Brice
>
> ___
> hwloc-devel mailing list
> hwloc-de...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-devel
>
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
gt; http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-devel
>
--
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/
don't really care whether it's a pointer or not :)
Ah -- preserving ABI goodness is a Good Thing. Keeping it a pointer won't
guarantee us ABI compatibility, of course, but it helps. So disregard my prior
comments -- keep it a pointer. Might as well give it a good college try to be
ABI friendl
rs similar to this one)
...these seem to be the last two kinds I can find these days. I don't see any
warnings/errors about not finding "diagram".
--
Jeff Squyres
jsquy...@cisco.com
ex $(DOX_DIR)/latex/a4.tex
> + -@mv -f $(DOX_DIR)/html/switchfromplpa.html
> $(DOX_DIR)/html/a5.html
> + -@mv -f $(DOX_DIR)/latex/switchfromplpa.tex
> $(DOX_DIR)/latex/a5.tex
> endif
>
> #
> _______
> hwloc-svn mailing list
> hwloc-...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-svn
>
--
Jeff Squyres
jsquy...@cisco.com
it to store the hostname of
> Machine objects. And the libpci branch uses it to store the PCI
> vendor/model string and OS device names. Maybe all these shouldn't be in
> obj->name and rather in the array of strings as OSName=Kerrighed,
> Hostname=foobar, PCIVendor=foo, PCIModel=bar, O
oday, when that
might be the normal tomorrow (kind of like calling something "new" -- someday,
it won't be new anymore). I'd rather either specify the page size or not; not
use an adjective that may or may not have meaning someday / on some platforms.
--
Jeff Squyres
jsquy...@cisco.com
Just so I understand - are you saying hwloc should track both the total amount
of memory *and* the makeup of that amount, broken up by page size? So obj A
may have x total memory, split across y 4k pages and z bigk hugepages (for
example)? And then the question becomes how to store this
Perhaps we should distinguish between memory in this object and memory that is
accessible by this object...?
I.e., cache and numa node can have x local memory. System/machine may have 0
local memory but (sum of children) accessible memory.
Specifically:
1 I think it is a common enough action
using c.xml directly (which I didn't see how that would work).
Instead, you're talking about manually stitching multiple XML files together
under a single , further enclosed under a single "Misc" object.
--
Jeff Squyres
jsquy...@cisco.com
ing into the existing text, however)
--
Jeff Squyres
jsquy...@cisco.com
l tie us from further
> extensions anyway, and I don't think much code relies on it anyway.
Agreed.
> The plan I see is that for 1.0 we only check that catenating .XML files
> by hand to build misc levels representing network layers does indeed
> work, which should mean that actual combining functions etc. should be
> possible to implement later.
FWIW, I'd prefer to see the combining/etc. functions ASAP -- we could
definitely use such things in Open MPI...
--
Jeff Squyres
jsquy...@cisco.com
t these should also be in the PDF/html
> documentation. This could just be a chapter, which doxygen automatically
> translates into a manual page.
Sure, I have no objection to that.
--
Jeff Squyres
jsquy...@cisco.com
On Jan 8, 2010, at 12:22 PM, Samuel Thibault wrote:
>> How do you guys want to handle params that are *sometimes* unused?
>
> I'd just mark them unused.
Is it harmful to mark them unused when they're actually used?
--
Jeff Squyres
jsquy...@cisco.com
if (hwloc_backend_sysfs_init(topology, fsroot_path) < 0)
return -1;
#endif /* HWLOC_LINUX_SYS */
return 0;
}
-
On OS X, I get
-
topology.c:1451: warning: unused parameter ‘fsroot_path’
(and a few others like this)
--
Jeff Squyres
jsquy...@cisco.com
On Dec 16, 2009, at 9:03 PM, Jeff Squyres wrote:
> Ok, done. Please review:
>
>https://svn.open-mpi.org/trac/hwloc/changeset/1475
I definitely broke something. :-(
>From my embedding test, I make a trivial app that does the following:
mytest_hwloc_t
:
> Jeff Squyres wrote:
> > On Dec 16, 2009, at 3:35 PM, Jeff Squyres wrote:
> >
> >
> >> While preparing to SVN commit, I made the default for HG checkouts to
> >> compile with debug and picky modes. I now see a TRUCKLOAD of
> >>
RSN?
On Dec 16, 2009, at 9:49 AM, Samuel Thibault wrote:
> Jeff Squyres, le Wed 16 Dec 2009 09:42:21 -0500, a écrit :
> > Yes -- we (OMPI) have m4 for checking for oodles of attributes.
>
> Ok, found it in config/ompi_check_attributes.m4.
>
> > Want me to bring them ov
sed
> attribute?
Yes -- we (OMPI) have m4 for checking for oodles of attributes. Want me to
bring them over and you and trim want you don't want?
--
Jeff Squyres
jsquy...@cisco.com
ch is still useful when debugging bigger applications to know which
> library a stack frame is in)
Coolio. I'll bring over the vis stuff as part of the embedding stuff.
--
Jeff Squyres
jsquy...@cisco.com
ake a comparison function that is often
> named "compar" without an "e" :)
Probably due to lunar effects. ;-)
I'll add the "e".
--
Jeff Squyres
jsquy...@cisco.com
is kind of feature these
days.
It would take some configure mojo and some code changes (i.e., put a DECLSPEC
in front of public symbol declarations). Such things can be copied over from
the OMPI code base.
--
Jeff Squyres
jsquy...@cisco.com
On Dec 15, 2009, at 4:35 PM, Jeff Squyres (jsquyres) wrote:
> Note that I currently have the symbol prefix remapping from hwloc_foo to
> foo. Instead, should it be hwloc_foo to hwloc_foo?
Got an opinion from Brice in IM: use the latter. Done.
--
Jeff Squyres
jsquy...@cisco.com
an you guys review it and see if there's anything you hate in there?
Thanks.
--
Jeff Squyres
jsquy...@cisco.com
> > provide at least some level of guarantee of stability about that logical
> > ordering.
>
> For sure since that was precisely what I had in my when I put the
> sorting code in the generic part.
2 steps ahead of me. Perfect. :-)
--
Jeff Squyres
jsquy...@cisco.com
Doh -- disregard -- pilot error on my part...
/me slinks back into my hole...
On Dec 9, 2009, at 3:33 PM, Brice Goglin wrote:
> Jeff Squyres wrote:
> > While working on the embedded stuff, I noticed the following assertion
> > failure in the SVN trunk:
> >
>
!
I added the proper AC_CONFIG_COMMANDS and "make check" seems to work now:
http://bitbucket.org/jsquyres/hwloc-embedded/changeset/c4907997c210/
--
Jeff Squyres
jsquy...@cisco.com
ot build the hwloc executables because we'll
only want the library code to be slurped into OMPI's larger libraries.
> 5) There are a bunch of OMPI that should be replaced with hwloc in the
> comments :)
Oops. :-) I found one regarding the version (in configure.ac) -- I can't find
any others. Can you cite?
--
Jeff Squyres
jsquy...@cisco.com
/OS indexes is actually quite useful to me
(e.g., to ensure that the hardware and OS are playing nicely).
My point is that everyone has a different view here -- we should just support
both. IMHO, the common case is logical indexes -- so let's make those the
default. But there are definitely cases where physical indexes are useful as
well.
--
Jeff Squyres
jsquy...@cisco.com
he most annoying thing is that sscanf() thinks it's
> matched and it's return code will be set accordingly.
Yuck!
Thankfully, we don't appear to be using sscanf() to convert the cpuset strings.
--
Jeff Squyres
jsquy...@cisco.com
On Dec 4, 2009, at 1:09 AM, Brice Goglin wrote:
> > shell$ hwloc-bind
> >
> > (i.e., invoking hwloc-bind with no arguments)
> >
> > returns an exit status of 0. Shouldn't it return non-zero?
>
> Yeah maybe
I'm going to interpret that as "Hell yes! Ple
Please proofread all the text I just added to hwloc-bind.1in.
I tried to document it as it is now -- not with the extensions that I listed in
the prior email threads.
--
Jeff Squyres
jsquy...@cisco.com
$actions_to_do
to return 0 if $actions_to_do was mistakenly empty.
Right?
--
Jeff Squyres
jsquy...@cisco.com
3 specifications?
>
> It means 5th logical system below 3rd logical node below first core. So
> it means nothing when there are no node objects below cores or no
> systems below nodes.
Ahh... now I see. So it's meant to be a logical delve into the topology -- the
leftmost item is meant to be the highest item in the topology, and each "."
item must be a child of the item to its left.
Is that correct?
Does it always need to start with system? If not, can you provide an example
that you have to represent with the . notation and could not be represented
with non-. notation?
--
Jeff Squyres
jsquy...@cisco.com
the merger of the
libtopology project from INRIA and the Portable Linux Processor
Affinity (PLPA) sub-project from Open MPI. *Both of these prior
projects are now deprecated.*
--
Jeff Squyres
jsquy...@cisco.com
On Dec 1, 2009, at 8:52 AM, Samuel Thibault wrote:
Apparently configure thinks that icc is gcc...
Ya, I just amended #17: icc does lie an impersonate gcc. We can fix
this for v1.0.
--
Jeff Squyres
jsquy...@cisco.com
h doesn't seem
right.
I'll file tickets for this stuff.
Does everything seem to work properly for you with the rc?
--
Jeff Squyres
jsquy...@cisco.com
Oops! Good fix -- thanks... I'll wait a little bit to see if
anything else pops up before rolling rc2.
On Nov 24, 2009, at 3:18 PM, Samuel Thibault wrote:
Samuel Thibault, le Tue 24 Nov 2009 21:03:39 +0100, a écrit :
> Jeff Squyres, le Tue 24 Nov 2009 13:14:02 -0500, a écrit :
> >
A bunch of fixes have been applied to the 0.9 branch; v0.9.3rc1 has
been posted:
http://www.open-mpi.org/software/hwloc/v0.9/
Please test it as much as you can so that we can get it out the door.
Thanks!
--
Jeff Squyres
jsquy...@cisco.com
thanks!
Do you want to commit that? If we keep the analog to the OMPI
project, we'd put it under contrib/dist/linux/hwloc.spec.
--
Jeff Squyres
jsquy...@cisco.com
r their specfiles; I
am not up-to-date on them, though.
Hypothetically, the specfile should be pretty simple since we conform
to most of the GNU standards.
> FWIW I'm the Feora maintainer for the "other" libtopology[1] ;P
Exxxcellent. :-)
--
Jeff Squyres
jsquy...@cisco.com
ded
to be portable.
It may be useful to have the API be extensible; one thing I have heard
rumors about coming in the not-distant future is the concept of
"boards" (multiple motherboards in a single box). I.e., socket IDs
may be repeated; they are differentiated by board number.
--
Jeff Squyres
jsquy...@cisco.com
run RH/Suse will eventually really upgrade
their
glibc and get the bumped minor.
Sounds good.
--
Jeff Squyres
jsquy...@cisco.com
if that only works on some OS's.
--
Jeff Squyres
jsquy...@cisco.com
s have happened before; it's why I always prefer
testing for behavior rather than version numbers.
But I don't quite know how to probe for this in the running glibc --
you *may or may not* encounter a problem if you have a size mismatch.
Version number might be the best that we can do here. :-\
--
Jeff Squyres
jsquy...@cisco.com
en embedding plpa in other things -- having this
kind of utility gorp function just meant helping other developers who
were *using* plpa (e.g., if they expose their own argv command line
options for specifying affinity, why not let them use the same string
parsing functions that plpa used?). So
fix this problem?).
--
Jeff Squyres
jsquy...@cisco.com
was for getting/setting the affinity of other
processes. E.g., htop uses PLPA for this purpose (htop.sf.net).
Do other OS's not support that?
--
Jeff Squyres
jsquy...@cisco.com
tp://www.open-mpi.org/mailman/listinfo.cgi/hwloc-devel
--
Jeff Squyres
jsquy...@cisco.com
That was definitely
a problem in the past -- kernel and glibc would mismatch in terms of
set/getaffinity (which was included in many distros).
--
Jeff Squyres
jsquy...@cisco.com
Ok, I see the issue. Apply the attached patch and see if that works
for you.
I'd prefer not to do another release today, if possible. ;-)
On Nov 5, 2009, at 2:50 PM, Jeff Squyres (jsquyres) wrote:
(re-subjected to be accurate)
I notice that my autogen.sh outputs:
autoreconf: running
n.
> autoreconf: /home/balaji/software/tools/install/bin/autoconf
failed with
> exit status: 1
>
> -- Pavan
>
> On 11/04/2009 03:55 PM, Jeff Squyres wrote:
>> Please beat it up and look for problems!
>>
>> http://www.open-mpi.org/software/hwloc/v0.9/
>
This problem may go away if we adapt PLPA's approach to sched_[set|
get]affinity.
On Nov 4, 2009, at 10:34 PM, Chris Samuel wrote:
- "Chris Samuel" <csam...@vpac.org> wrote:
> - "Jeff Squyres" <jsquy...@cisco.com> wrote:
>
> > K. Clear
Please beat it up and look for problems!
http://www.open-mpi.org/software/hwloc/v0.9/
--
Jeff Squyres
jsquy...@cisco.com
401 - 500 of 540 matches
Mail list logo