Re: [hwloc-devel] AIX

2011-08-03 Thread Samuel Thibault
Brice Goglin, le Wed 03 Aug 2011 12:49:17 +0200, a écrit : > Le 03/08/2011 12:46, Samuel Thibault a écrit : > > Also, it seems that on your system the processing units are gathered 4 > > by 4 (exposed as a "group" in hwloc). Do you happen to know what this > > physic

Re: [hwloc-devel] -lpthread in LIBS and REQUIRES?

2011-08-04 Thread Samuel Thibault
Brice Goglin, le Thu 04 Aug 2011 16:58:19 +0200, a écrit : > While playing with static hwloc libraries and binaries [1], I had to > manually add -lpthread to LIBS to get the fully-static binaries to link > properly. What was the missing reference? > We use some pthread functions, so is there a

Re: [hwloc-devel] -lpthread in LIBS and REQUIRES?

2011-08-04 Thread Samuel Thibault
Jeff Squyres, le Thu 04 Aug 2011 17:07:45 +0200, a écrit : > On Aug 4, 2011, at 10:57 AM, Brice Goglin wrote: > > While playing with static hwloc libraries and binaries [1], I had to > > manually add -lpthread to LIBS to get the fully-static binaries to link > > properly. We use some pthread

Re: [hwloc-devel] -lpthread in LIBS and REQUIRES?

2011-08-04 Thread Samuel Thibault
Jeff Squyres, le Thu 04 Aug 2011 17:18:11 +0200, a écrit : > On Aug 4, 2011, at 11:13 AM, Samuel Thibault wrote: > > >> Where do we use the pthread functions -- in the libhwloc core, or just in > >> the utils? > > > > In the core, but only if they ar

Re: [hwloc-devel] -lpthread in LIBS and REQUIRES?

2011-08-04 Thread Samuel Thibault
Brice Goglin, le Thu 04 Aug 2011 17:47:17 +0200, a écrit : > /home/bgoglin/SOFT/hwloc/branch-1.2/build/src/.libs/libhwloc.a(topology-linux.o): > In function `hwloc_linux_set_thread_cpubind': > topology-linux.c:(.text+0x1070): undefined reference to `pthread_self' Odd. This should be in

Re: [hwloc-devel] -lpthread in LIBS and REQUIRES?

2011-08-04 Thread Samuel Thibault
Brice Goglin, le Thu 04 Aug 2011 18:06:24 +0200, a écrit : > Le 04/08/2011 17:48, Samuel Thibault a écrit : > > Brice Goglin, le Thu 04 Aug 2011 17:47:17 +0200, a écrit : > >> /home/bgoglin/SOFT/hwloc/branch-1.2/build/src/.libs/libhwloc.a(topology-linux.o)

Re: [hwloc-devel] [hwloc-bugs] make does not go through in AIX 5.2

2011-08-24 Thread Samuel Thibault
Hello, I guess AIX 5.2 is a bit old? Utpal Kumar Ray, le Wed 24 Aug 2011 10:23:13 +0200, a écrit : > Making all in src > CC topology.lo > "/home/utpal/hwloc-1.2/include/hwloc.h", line 998.28: 1506-959 (W) The > attribute "pure" is not a valid type attribute and is ignored. This is

Re: [hwloc-devel] [hwloc-svn] svn:hwloc r3671

2011-08-24 Thread Samuel Thibault
bgog...@osl.iu.edu, le Wed 24 Aug 2011 17:27:50 +0200, a écrit : > + HWLOC_BACKEND_CUSTOM, Actually, maybe rather than "custom", use "empty"? Samuel

Re: [hwloc-devel] Something lighter-weight than XML?

2011-09-01 Thread Samuel Thibault
Jeff Squyres, le Thu 01 Sep 2011 17:31:05 +0200, a écrit : > Do you think this would be easy to implement? A quite strict format could probably be easy to implement and still be extensible. The XML will probably remain useful for people who like XSLT :) Samuel

Re: [hwloc-devel] Something lighter-weight than XML?

2011-09-01 Thread Samuel Thibault
Jeff Squyres, le Thu 01 Sep 2011 20:31:44 +0200, a écrit : > hst: hwloc simple text I like this one. Samuel

Re: [hwloc-devel] Something lighter-weight than XML?

2011-09-02 Thread Samuel Thibault
Jeff Squyres, le Fri 02 Sep 2011 11:58:05 +0200, a écrit : > JSON: sure, it's an easy format, but we're not really targeting web-ish kinds > of things here, are we? > > YAML: ya, that's also an easy format. > > But the goal here is to do something utterly trivial that has no support >

Re: [hwloc-devel] xmlbuffer and xmlFree()

2011-09-05 Thread Samuel Thibault
Brice Goglin, le Mon 05 Sep 2011 09:16:08 +0200, a écrit : > 2) add hwloc_free_xmlbuffer() > (2) is easy It looks good indeed. Samuel

Re: [hwloc-devel] trunk make check on eddie

2011-09-05 Thread Samuel Thibault
Hello, Jeff Squyres, le Sat 27 Aug 2011 12:40:20 +0200, a écrit : > Here's some output from the nightly build machine about the "make check" > errors that are failing. It's hard to say what is happening. Could you manage to get a gdb backtrace at least? Samuel

Re: [hwloc-devel] trunk make check on eddie

2011-09-06 Thread Samuel Thibault
Jeff Squyres, le Tue 06 Sep 2011 13:42:32 +0200, a écrit : > I think this has now been resolved, right? Maybe, I was quite busy in the past weeks, so couldn't really follow closely. Samuel

Re: [hwloc-devel] Bug-Report: visibility=hiden on Sparc/Sun CC

2011-09-06 Thread Samuel Thibault
Igor Galić, le Tue 06 Sep 2011 17:10:04 +0200, a écrit : > - Original Message - > > Igor Galić, le Tue 06 Sep 2011 14:48:07 +0200, a écrit : > > > > Assertion failed: !obj, file hwloc_get_obj_inside_cpuset.c, line > > > > 40 > > > > /bin/bash: line 5: 20986 Abort (core

Re: [hwloc-devel] Bug-Report: visibility=hiden on Sparc/Sun CC

2011-09-09 Thread Samuel Thibault
Igor Galić, le Wed 07 Sep 2011 00:40:06 +0200, a écrit : > - Original Message - > > Igor Galić, le Tue 06 Sep 2011 18:00:00 +0200, a écrit : > > > > > > And also send the output of > > > > > > > > > > > > lstopo --input "nodes:2 sockets:3 caches:4 cores:5 6" > > > > > > > > > > Stdout: > >

Re: [hwloc-devel] Some practical hwloc API feedback

2011-09-25 Thread Samuel Thibault
Hello, Sorry for coming late in the thread, I've been busy at conference, traveling, teaching, moving between flats :) Brice Goglin, le Thu 22 Sep 2011 22:25:33 +0200, a écrit : > > (2d) more specifically: since all caches are of type HWLOC_OBJ_CACHE, we > > find ourselves putting in special

Re: [hwloc-devel] Some practical hwloc API feedback

2011-09-25 Thread Samuel Thibault
Jeff Squyres, le Thu 22 Sep 2011 21:36:31 +0200, a écrit : > (SIDENOTE: reading this letter after I wrote it, I see that it sounds > overwhelmingly negative. Please do not take it that way! Sure! If these are the only negative points of our API, it's actually a very positive mail ;) > 1. The

Re: [hwloc-devel] multinode/network topology status

2011-10-04 Thread Samuel Thibault
Hello, Brice Goglin, le Wed 21 Sep 2011 08:54:52 +0200, a écrit : > Example of C interface use (assembling 3 nodes in a global topology with > 3 levels of switches): > https://svn.open-mpi.org/trac/hwloc/browser/branches/custom/tests/hwloc_custom.c#L23 > > C interface doc: >

Re: [hwloc-devel] Stability of /sys/devices/system/cpu/cpu0/cache/index*/ interface

2011-10-04 Thread Samuel Thibault
Jiri Hladky, le Tue 04 Oct 2011 23:53:02 +0200, a écrit : > Since hwloc relies also on /sys/devices/system/cpu/cpu0/cache/ I'm > wondering if you had some thoughts or issues on that. On the linux-kernel mailing list it is often said that the content of /sys is part of the ABI and thus shall not

Re: [hwloc-devel] multinode/network topology status

2011-10-05 Thread Samuel Thibault
Brice Goglin, le Wed 05 Oct 2011 08:50:10 +0200, a écrit : > Thanks for the feedback. I think I fixed all this. Confirmed, thanks! Samuel

Re: [hwloc-devel] Re : hwloc 1.3.1 needed?

2011-11-03 Thread Samuel Thibault
Brice Goglin, le Thu 03 Nov 2011 17:14:31 +0100, a écrit : > Yeah but I tried on half a dozen different distros without reproducing it so > it didn't look widespread to me. Probably because libpci-devel was installed on your installed distro, it'll fail if you install it by hand in your hand,

Re: [hwloc-devel] memory corruption problem with libpci

2011-11-03 Thread Samuel Thibault
Guy Streeter, le Thu 03 Nov 2011 18:52:20 +0100, a écrit : > Does it look familiar? Nope. > How can I help debug this? Maybe valgrind will be helpful. Samuel

Re: [hwloc-devel] [hwloc-svn] svn:hwloc r3971

2011-11-25 Thread Samuel Thibault
Brice Goglin, le Fri 25 Nov 2011 10:09:31 +0100, a écrit : > I don't see any call to cuInit(). Does it mean that we don't have to > wait 3s during startup as usual with CUDA? Well, it works without at least. Samuel

Re: [hwloc-devel] [hwloc-svn] svn:hwloc r3971

2011-11-25 Thread Samuel Thibault
Brice Goglin, le Fri 25 Nov 2011 15:00:10 +0100, a écrit : > hwloc_get_pcidev_by_busid(topology, domain, bus, dev, 0) should work > here (hwloc/helper.h). Not at this point of the detection: levels are not established yet. Samuel

Re: [hwloc-devel] [hwloc-svn] svn:hwloc r4006

2011-11-30 Thread Samuel Thibault
Jeff Squyres, le Wed 30 Nov 2011 11:59:45 +0100, a écrit : > I see how libnuma is Linux-specific; but is the level of memory binding > uniform across all other OS's? I.e., do we need to print the same kind of > granularity of memory support for other OS's? It'd be tedious. The difference,

Re: [hwloc-devel] 1.3.1 and 1.4rc1

2011-12-13 Thread Samuel Thibault
Do we merge the cuda branch into 1.4? I didn't do the work directly into the trunk because I wasn't sure of what I'd need to add to the interface. Eventually, the additions are - the "tight" field, which just means whether children are tightly packed, such as cores in an nvidia MultiProcessor,

Re: [hwloc-devel] 1.3.1 and 1.4rc1

2011-12-13 Thread Samuel Thibault
Brice Goglin, le Tue 13 Dec 2011 16:41:08 +0100, a écrit : > Le 13/12/2011 16:22, Jeff Squyres a écrit : > > > > I can't speak for GPUs, but I think the PCI information will be useful to > > know what devices are close to what PUs / NUMA nodes. That information can > > be used to make decisions

Re: [hwloc-devel] 1.3.1 and 1.4rc1

2011-12-13 Thread Samuel Thibault
Brice Goglin, le Tue 13 Dec 2011 14:10:17 +0100, a écrit : > My main problem is that it's hard to know whether this will look good in > two years when we'll have support for AMD APU, Intel MIC and other > "strange" architectures. Which types should be common to CPUs and these > accelerators? Might

Re: [hwloc-devel] 1.3.1 and 1.4rc1

2011-12-13 Thread Samuel Thibault
Jeff Squyres, le Tue 13 Dec 2011 18:13:38 +0100, a écrit : > On Dec 13, 2011, at 11:59 AM, Samuel Thibault wrote: > > I keep not having time to do what I'd like to do, like answering > > mails... > > What -- you have a real job or something like that? ;-) It's mostly the

Re: [hwloc-devel] 1.3.1 and 1.4rc1

2011-12-13 Thread Samuel Thibault
Brice Goglin, le Tue 13 Dec 2011 18:58:14 +0100, a écrit : > >> Also I don't think the GPU caches should be L2 because they are not very > >> similar to the CPU ones. > > How so? > > In the same way the GPU memory is different from the NUMA node memory? > Why would caches and cores be similar for

Re: [hwloc-devel] hwloc-1.3.1rc1: odd "make check" failure on Solaris 11

2011-12-14 Thread Samuel Thibault
Paul H. Hargrove, le Wed 14 Dec 2011 09:50:23 +0100, a écrit : > It appears that something has passed bogus arguments to hwloc-calc. > Adding "set -x" in test-hwloc-calc.sh.in I can find the failure is from: > ./hwloc-calc --if synthetic --input "node:4 core:4 pu:4" pu:22-47 > --largest --sep

Re: [hwloc-devel] HWLOC_HAVE_ATTRIBUTE_* unused warning

2011-12-14 Thread Samuel Thibault
Brice Goglin, le Wed 14 Dec 2011 10:50:26 +0100, a écrit : > In file included from /home/bgoglin/SOFT/hwloc/trunk/include/hwloc.h:49:0, > from ../../utils/lstopo.h:12, > from ../../utils/lstopo-color.c:11: >

Re: [hwloc-devel] very sched_setaffinity support

2011-12-14 Thread Samuel Thibault
Brice Goglin, le Wed 14 Dec 2011 11:21:30 +0100, a écrit : > His sched.h defines the very old prototype of sched_setaffinity (pid, > len, ulong*). Urgl. Is it actually "very old"? I thought it was just a transient prototype in glibc, between pid,mask and pid,size,mask, which IIRC was never

Re: [hwloc-devel] very sched_setaffinity support

2011-12-14 Thread Samuel Thibault
Brice Goglin, le Wed 14 Dec 2011 11:42:37 +0100, a écrit : > Le 14/12/2011 11:34, Samuel Thibault a écrit : > > Brice Goglin, le Wed 14 Dec 2011 11:21:30 +0100, a écrit : > >> His sched.h defines the very old prototype of sched_setaffinity (pid, > >> len, ulong*). >

Re: [hwloc-devel] very sched_setaffinity support

2011-12-14 Thread Samuel Thibault
Samuel Thibault, le Wed 14 Dec 2011 11:44:04 +0100, a écrit : > Brice Goglin, le Wed 14 Dec 2011 11:42:37 +0100, a écrit : > > Le 14/12/2011 11:34, Samuel Thibault a écrit : > > > Brice Goglin, le Wed 14 Dec 2011 11:21:30 +0100, a écrit : > > >> His sched.h

Re: [hwloc-devel] WHOLE_SYSTEM and cpusets (was: 1.3.1 and 1.4rc1)

2011-12-14 Thread Samuel Thibault
Jeff Squyres, le Wed 14 Dec 2011 16:06:27 +0100, a écrit : > - cpuset may not be equal to complete_cpuset Yes, but not only in that case, see below. > - cpuset is already the logical AND of allowed_cpuset and online_cpuset Yes, that's explicitly enforced by the core > - cpuset is comprised

Re: [hwloc-devel] 1.3.1 and 1.4rc1

2011-12-20 Thread Samuel Thibault
Brice Goglin, le Tue 13 Dec 2011 18:58:14 +0100, a écrit : > I am just saying that the current lstopo output on hosts with GPUs is > unreadable just because we show many "empty" boxes. This should be fixed now. Samuel

[hwloc-devel] topology-solaris-chiptype.c bugs on processortype

2012-01-18 Thread Samuel Thibault
Hello, I'm getting an issue with the topology-solaris-chiptype.c work on Solaris 10: in the ProcessorType case, the returned value does not look like a readable string, I'm getting "¨", which consequently poses problems in the xml parser. Of course, googling picl_get_propinfo "ProcessorType"

Re: [hwloc-devel] topology-solaris-chiptype.c bugs on processortype

2012-01-18 Thread Samuel Thibault
TERRY DONTJE, le Wed 18 Jan 2012 17:57:33 +0100, a écrit : > On 1/18/2012 11:48 AM, Samuel Thibault wrote: > >TERRY DONTJE, le Wed 18 Jan 2012 16:31:35 +0100, a écrit : > >>On 1/18/2012 9:55 AM, Samuel Thibault wrote: > >>>I'm getting an issue with the to

Re: [hwloc-devel] topology-solaris-chiptype.c bugs on processortype

2012-01-18 Thread Samuel Thibault
Samuel Thibault, le Wed 18 Jan 2012 15:55:18 +0100, a écrit : > I'm getting an issue with the topology-solaris-chiptype.c work on > Solaris 10: in the ProcessorType case, the returned value does not look > like a readable string, I'm getting "¨", which consequently poses &g

Re: [hwloc-devel] topology-solaris-chiptype.c bugs on processortype

2012-01-18 Thread Samuel Thibault
Jeff Squyres, le Wed 18 Jan 2012 18:48:16 +0100, a écrit : > On Jan 18, 2012, at 12:36 PM, Samuel Thibault wrote: > > >> Which branch is your workspace checkedout from? > > > > I don't know what a "workspace" is. I basically don't know anything > >

Re: [hwloc-devel] topology-solaris-chiptype.c bugs on processortype

2012-01-18 Thread Samuel Thibault
TERRY DONTJE, le Wed 18 Jan 2012 19:04:34 +0100, a écrit : > On 1/18/2012 12:59 PM, TERRY DONTJE wrote: > >On 1/18/2012 12:51 PM, Samuel Thibault wrote: > >>Samuel Thibault, le Wed 18 Jan 2012 15:55:18 +0100, a écrit : > >>>I'm getting an issue with the to

[hwloc-devel] Solaris visibility issue

2012-01-18 Thread Samuel Thibault
TERRY DONTJE, le Wed 18 Jan 2012 18:52:50 +0100, a écrit : > >>Also, I tried to build v1.4 and the trunk and I keep getting linkage errors > >>on lstopo-lstopo-draw.o complaining about hwloc_insert_object_by_cpuset > >>being undefined. > >It is defined in ./src/topology.c. Please check with make

Re: [hwloc-devel] Solaris visibility issue

2012-01-18 Thread Samuel Thibault
TERRY DONTJE, le Wed 18 Jan 2012 19:23:03 +0100, a écrit : > Don't you need the function to make lstopo work? lstopo itself does not need it, only internal functions of libhwloc needs it. Could you try the attached patch? Thanks, Samuel Index: include/private/private.h

Re: [hwloc-devel] [hwloc-announce] hwloc v1.4rc2 released

2012-01-23 Thread Samuel Thibault
On 23/01/12 20:34, Brice Goglin wrote: > Please test and let us know if you find any issue. Can somebody try it on AIX? Samuel

Re: [hwloc-devel] hwloc-1.3.1 and 1.4 failures on MIPS64/qemu

2012-02-01 Thread Samuel Thibault
Paul H. Hargrove, le Wed 01 Feb 2012 05:58:05 +0100, a écrit : > At the moment my suspicion falls on the compiler, as I can't see how a > failure of 256ia64-64n2s2c.output could be h/w dependent. It could be a problem of imperfect qemu emulation. Samuel

Re: [hwloc-devel] [hwloc-svn] svn:hwloc r4224

2012-02-01 Thread Samuel Thibault
Brice Goglin, le Wed 01 Feb 2012 14:02:53 +0100, a écrit : > Any idea why this broke FreeBSD? > http://hydra.bordeaux.inria.fr/build/22112 (same problem for tunk and v1.4) > Brice Oops, I had forgottent about the redzone. Should be fixed now. Samuel

Re: [hwloc-devel] hwloc-1.3.1 assertion failures on Linux/POWER7

2012-02-02 Thread Samuel Thibault
Brice Goglin, le Wed 01 Feb 2012 14:20:49 +0100, a écrit : > "/vlsci/VLSCI/samuel/HWLOC/hwloc-1.3.1/include/hwloc.h", line 1203.28: > 1506-1385 (W) The attribute "pure" is not a valid type attribute. > CC traversal.lo > > Attribute pure is before the function name, I'll move it after, XLC >

Re: [hwloc-devel] hwloc-1.4 "gmake check" failure on Solaris-10/SPARC/gccfss [PATCH]

2012-02-02 Thread Samuel Thibault
Paul H. Hargrove, le Thu 02 Feb 2012 02:29:08 +0100, a écrit : > + The configure-time logic is NOT trying to determine the version number, as > I don't have a way (yet?) to pinpoint which version(s) work correctly, and > the Oracle Forums thread on the subject doesn't say. So, it is >

Re: [hwloc-devel] hwloc-1.3.1 and 1.4 failures on MIPS64/qemu

2012-02-03 Thread Samuel Thibault
Paul H. Hargrove, le Wed 01 Feb 2012 20:18:59 +0100, a écrit : > So if qemu's emulation were at fault, then one would expect more wide-spread > problems and a generally unstable system. > While -mabi=64 is not the default in user-space, it IS the way the linux > kernel is built and thus should be

Re: [hwloc-devel] hwloc-1.3.1 and 1.4 failures on MIPS64/qemu

2012-02-03 Thread Samuel Thibault
Paul H. Hargrove, le Fri 03 Feb 2012 20:53:28 +0100, a écrit : > I do see, from hwloc-hello: > *** The number of sockets is unknown > *** Logical processor 0 has 0 caches totaling 0KB > IFF there is reason to expect better from this system, I'd be happy to work > w/ somebody on fixing that. I

Re: [hwloc-devel] 1.3.2rc1 has escaped

2012-02-08 Thread Samuel Thibault
Jeff Squyres, le Wed 08 Feb 2012 17:59:04 +0100, a écrit : > Please test! > > http://www.open-mpi.org/software/hwloc/v1.3/ Could somebody test it on AIX, and with xlc? Thanks, Samuel

Re: [hwloc-devel] 1.3.2rc1 has escaped

2012-02-08 Thread Samuel Thibault
Paul H. Hargrove, le Thu 09 Feb 2012 01:41:47 +0100, a écrit : > On 2/8/2012 4:31 PM, Samuel Thibault wrote: > >Paul H. Hargrove, le Thu 09 Feb 2012 01:28:53 +0100, a écrit : > >>Option #4: > >>CFLAGS='-qhalt=e -qsuppress=1506-077' > >>Appears to

Re: [hwloc-devel] 1.3.2rc1 has escaped

2012-02-08 Thread Samuel Thibault
Samuel Thibault, le Thu 09 Feb 2012 01:43:56 +0100, a écrit : > Paul H. Hargrove, le Thu 09 Feb 2012 01:41:47 +0100, a écrit : > > On 2/8/2012 4:31 PM, Samuel Thibault wrote: > > >Paul H. Hargrove, le Thu 09 Feb 2012 01:28:53 +0100, a écrit : > > >>Option #4: > &g

Re: [hwloc-devel] excessive warnings from xlc w/ hwloc-trunk

2012-02-19 Thread Samuel Thibault
Paul H. Hargrove, le Sat 18 Feb 2012 00:58:27 +0100, a écrit : > I am not sure of the risk/reward on applying this change, however. > Gcc seems to be happy enough either way as far as I could tell. > -static __hwloc_inline const char * __hwloc_attribute_pure > +static __hwloc_inline

Re: [hwloc-devel] excessive warnings from xlc w/ hwloc-trunk

2012-02-19 Thread Samuel Thibault
Paul H. Hargrove, le Sun 19 Feb 2012 21:37:50 +0100, a écrit : > On 2/19/2012 10:54 AM, Samuel Thibault wrote: > [snip] > >Does it still complain if using the following? > >>static __hwloc_inline const char * > >> hwloc_obj_get_info_by_name(hwl

Re: [hwloc-devel] hwloc_alloc_membind

2012-02-24 Thread Samuel Thibault
Karl Napf, le Fri 24 Feb 2012 13:04:58 +0100, a écrit : > What surprises me is that the result of the call to > hwloc_fix_membind_cpuset in line 534 of bind.c is negated, even though > hwloc_fix_membind_cpuset returns 0 on success. Might this be a bug? Oops, indeed. > 2. In another use case I

Re: [hwloc-devel] hwloc_alloc_membind

2012-02-24 Thread Samuel Thibault
Karl Napf, le Fri 24 Feb 2012 14:45:26 +0100, a écrit : > Thanks a lot, Samuel. BTW the documentation says that > hwloc_alloc_membind and hwloc_alloc_membind_nodeset return -1 on > error, but this should be NULL. Indeed! Thanks, Samuel

Re: [hwloc-devel] merging instruction cache branch

2012-03-13 Thread Samuel Thibault
Brice Goglin, le Tue 13 Mar 2012 14:39:50 +0100, a écrit : > There's a topology flag ("ICACHES") to enable Instruction cache > detection, and a lstopo options (--icaches). Mmm, we enable PCI devices in lstopo by default already, I would say that we could enable icache by default too. Not a strong

[hwloc-devel] Checking that we use portability macros Was: [hwloc-svn] svn:hwloc r4382

2012-03-13 Thread Samuel Thibault
sthib...@osl.iu.edu, le Tue 13 Mar 2012 17:46:36 +0100, a écrit : > Always hwloc_strncasecmp, not strncasecmp We should probably add checks against such things, we have the same problem with __inline, that people tend to use instead of our portability hwloc_inline. I'm not sure how that fits best

Re: [hwloc-devel] topology-x86.c warning

2012-03-15 Thread Samuel Thibault
Jeffrey Squyres, le Fri 16 Mar 2012 02:04:37 +0100, a écrit : > cache->size = linesize * linepart * ways * sets; /* FIXME: what if ways > == -1 ? */ Ah, right, we still haven't had the opportunity to check with an actual cpu with ways == -1. I've changed the code to just do like Linux.

Re: [hwloc-devel] Hydra job runtime:hwloc-trunk:build_fedora2_i386 on x86_64-linux, build 36199: Failed

2012-03-15 Thread Samuel Thibault
Brice Goglin, le Thu 15 Mar 2012 17:36:29 +0100, a écrit : > Looks like g++ doesn't like __hwloc_attribute_unused in hwloc/helper.h I wonder what Hydra uses for g++. I have tested various versions (from 4.1 to 4.7), without problems. Samuel

Re: [hwloc-devel] Hydra job runtime:hwloc-trunk:build_fedora2_i386 on x86_64-linux, build 36199: Failed

2012-03-17 Thread Samuel Thibault
Brice Goglin, le Fri 16 Mar 2012 07:48:52 +0100, a écrit : > Le 16/03/2012 02:44, Samuel Thibault a écrit : > > Brice Goglin, le Thu 15 Mar 2012 17:36:29 +0100, a écrit : > >> Looks like g++ doesn't like __hwloc_attribute_unused in hwloc/helper.h > > I wonder what Hydra use

Re: [hwloc-devel] [hwloc-svn] svn:hwloc r4409

2012-03-20 Thread Samuel Thibault
Jeffrey Squyres, le Tue 20 Mar 2012 15:51:54 +0100, a écrit : > What do you think of this patch? It separates out the individual version > checking to make the #define logic a little easier to read. Interesting indeed, I vote for it. Samuel

Re: [hwloc-devel] BGQ empty topology with MPI

2012-03-21 Thread Samuel Thibault
Daniel Ibanez, le Wed 21 Mar 2012 03:37:25 +0100, a écrit : > Please let me know if theres a hint of what could be causing it, > where to post, and what info to provide. This is already the proper list. Please attach the output of lstopo after having given the --enable-debug option to

Re: [hwloc-devel] BGQ empty topology with MPI

2012-03-22 Thread Samuel Thibault
Brice Goglin, le Thu 22 Mar 2012 10:58:46 +0100, a écrit : > I don't see anything bad in your outputs. > So there's something strange going on when MPI is added. Which MPI are using? > Is this a derivative of MPICH that embeds hwloc? (MPICH >= 1.2.1 if I remember > correctly) There might be

Re: [hwloc-devel] BGQ empty topology with MPI

2012-03-22 Thread Samuel Thibault
Daniel Ibanez, le Thu 22 Mar 2012 23:07:01 +0100, a écrit : > I suspected this might be the reason, so I called "nm" > with the static versions of the libraries their compiler wrappers > link against and I could not find the term "hwloc" in the output. > Is this a valid test? Ah, right, embedded

Re: [hwloc-devel] Fwd: BGQ empty topology with MPI

2012-04-06 Thread Samuel Thibault
Daniel Ibanez, le Wed 28 Mar 2012 20:23:18 +0200, a écrit : > The machine is back in working order. > I tried this patch and it works great: I get cpus and my whole program runs as > expected. Maybe it could be a safe fallback that we should keep anyway, even if we work out the issue with sysfs.

Re: [hwloc-devel] interoperability with X displays

2012-04-14 Thread Samuel Thibault
Samuel Thibault, le Thu 29 Mar 2012 17:56:28 +0200, a écrit : > Brice Goglin, le Thu 29 Mar 2012 16:09:00 +0200, a écrit : > > The lib is unfortunately only built as a static lib in at least debian > > and ubuntu (without -fPIC), which is annoying. > > That might be fixa

Re: [hwloc-devel] interoperability with X displays

2012-04-16 Thread Samuel Thibault
Marwan Abd Ellah, le Mon 16 Apr 2012 13:53:57 +0200, a écrit : > I have checked the package contents on Debia sid and I haven't seen any > updates >  at  > http://packages.debian.org/sid/amd64/nvidia-settings/filelist > Would you please refer to the place where I can find the libXNVCtrl.so  ?

Re: [hwloc-devel] lstopo-nox strikes back

2012-04-25 Thread Samuel Thibault
Jeffrey Squyres, le Wed 25 Apr 2012 15:45:18 +0200, a écrit : > FWIW: Having lstopo plugins for output would obviate the need for having two > executable names. Well, it seems overkill to me. It makes sense to me to have both xlstopo and lstopo. Samuel

Re: [hwloc-devel] lstopo-nox strikes back

2012-04-25 Thread Samuel Thibault
Jeff Squyres, le Wed 25 Apr 2012 16:55:23 +0200, a écrit : > On Apr 25, 2012, at 10:48 AM, Samuel Thibault wrote: > > >> FWIW: Having lstopo plugins for output would obviate the need for having > >> two executable names. > > > > Well, it seems overkill to me

Re: [hwloc-devel] lstopo-nox strikes back

2012-04-25 Thread Samuel Thibault
Brice Goglin, le Wed 25 Apr 2012 16:58:16 +0200, a écrit : > On 25/04/2012 16:55, Jeff Squyres wrote: > >On Apr 25, 2012, at 10:48 AM, Samuel Thibault wrote: > > > >>>FWIW: Having lstopo plugins for output would obviate the need for having > >>>two execut

Re: [hwloc-devel] lstopo-nox strikes back

2012-04-25 Thread Samuel Thibault
Jeff Squyres, le Wed 25 Apr 2012 17:03:01 +0200, a écrit : > On Apr 25, 2012, at 10:58 AM, Samuel Thibault wrote: > > > It already adapts itself, here. The issue is that the user has to > > install an X version to get potential for X support. Which brings X. > > If

Re: [hwloc-devel] lstopo-nox strikes back

2012-04-25 Thread Samuel Thibault
Jeff Squyres, le Wed 25 Apr 2012 17:11:28 +0200, a écrit : > Yes: the lstopo user gets whatever the sysadmin chose to install. > No: the system is not flexible for binary distributions > > Meaning: I see 2 ways to have binary packages that have X/cairo support and > don't have X/cairo support: >

Re: [hwloc-devel] lstopo-nox strikes back

2012-04-27 Thread Samuel Thibault
Brice Goglin, le Fri 27 Apr 2012 19:09:47 +0200, a écrit : > Le 25/04/2012 15:42, Jiri Hladky a écrit : > > I would vote to make lstopo ASCII only and introduce new GUI binary > > "lstopo-gui" in the version 1.5 > > I'll commit that during the weekend unless somebody comes with a better >

Re: [hwloc-devel] [PATCH] Use plain "inline" in C++

2012-05-09 Thread Samuel Thibault
Jeff Squyres, le Wed 09 May 2012 22:59:43 +0200, a écrit : > Samuel -- does this look right to you? If so, I can apply to 1.3, 1.4, and > trunk. > > +/* Note that if we're compiling C++, then just use the "inline" > + keyword, since it's part of C++ */ > +#if defined(c_plusplus) ||

Re: [hwloc-devel] [PATCH] Use plain "inline" in C++

2012-05-09 Thread Samuel Thibault
Jeff Squyres, le Wed 09 May 2012 23:41:20 +0200, a écrit : > On May 9, 2012, at 5:32 PM, Samuel Thibault wrote: > > >> +/* Note that if we're compiling C++, then just use the "inline" > >> + keyword, since it's part of C++ */ > >> +#if defined(c_pluspl

Re: [hwloc-devel] lstopo-gui

2012-05-25 Thread Samuel Thibault
Jeff Squyres, le Fri 25 May 2012 12:34:38 +0200, a écrit : > >From my checkout, it looks like lstopo-no-graphics supports: > > console, txt, fig, xml, synthetic > > How did fig get in there, btw? Doesn't that add dependencies? (or did that > get coded up manually / with no library

Re: [hwloc-devel] HWLOC_NBMAXCPUS

2012-06-21 Thread Samuel Thibault
TERRY DONTJE, le Thu 21 Jun 2012 15:47:22 +0200, a écrit : > I see a couple places where HWLOC_NBMAXCPUS is defined with a comment of > "FIXME: drop".  This static size just bit me on a machine that has 1440 CPUs.  > I can bump up the define in my clone but I was wondering if this fixed size >

Re: [hwloc-devel] [hwloc-svn] svn:hwloc r4554 - trunk/utils

2012-06-29 Thread Samuel Thibault
Brice Goglin, le Fri 29 Jun 2012 22:10:30 +0200, a écrit : > --pid doesn't care about binding, it just changes things like cgroups Uh? It's in there since a long time: lstopo-draw.c pu_draw(): else if (pid == 0) hwloc_get_cpubind(topology, bind, 0); if (bind &&

Re: [hwloc-devel] [hwloc-svn] svn:hwloc r4554 - trunk/utils

2012-06-29 Thread Samuel Thibault
Brice Goglin, le Fri 29 Jun 2012 22:29:20 +0200, a écrit : > Arg, I was using the console output (hwloc-nox). Ah, you should be able to use -.txt then, it colorizes there too. Samuel

Re: [hwloc-devel] XML string filtering

2012-07-06 Thread Samuel Thibault
Brice Goglin, le Fri 06 Jul 2012 14:50:46 +0200, a écrit : > I could just use isprint() to check every character > before export and only keep those between 32 and 127. Why not just taking 32-126? I don't see what isprint will bring. > But what about \n, > \t, \r, \f which are before ? Do we

Re: [hwloc-devel] hwloc_bitmap_or

2012-07-20 Thread Samuel Thibault
Pavan Balaji, le Sat 21 Jul 2012 04:40:22 +0200, a écrit : > In hwloc_bitmap_or(), is the resultant bitmap allowed to be the same as one > of the input bitmaps? It seems to work correctly in practice, but the API > doesn't seem to explicitly guarantee it. We actually use it a lot in the core.

Re: [hwloc-devel] backends and plugins

2012-08-21 Thread Samuel Thibault
Brice Goglin, le Tue 21 Aug 2012 18:49:48 +0200, a écrit : > 1) We load plugins and list existing components once per topology. We > should do it only once per process. But that requires some locking in > case multiple topologies are loaded simultaneously, which means we need > thread-safety. Do

Re: [hwloc-devel] backends and plugins

2012-08-22 Thread Samuel Thibault
Brice Goglin, le Wed 22 Aug 2012 07:52:07 +0200, a écrit : > Le 21/08/2012 21:18, Samuel Thibault a écrit : > > Brice Goglin, le Tue 21 Aug 2012 18:49:48 +0200, a écrit : > >> 1) We load plugins and list existing components once per topology. We > >> should

Re: [hwloc-devel] [hwloc-svn] svn:hwloc r4815 - branches/components/src

2012-09-05 Thread Samuel Thibault
Brice Goglin, le Wed 05 Sep 2012 16:13:31 +0200, a écrit : > The problem I was trying to fix below is that linking hwloc plugins on > Darwin failed because plugins referred to hwloc-core symbols. Nothing on > the libtool command-line said where to find those symbols (I don't > understand why it

Re: [hwloc-devel] -lhwloc in components.

2012-09-05 Thread Samuel Thibault
Jeff Squyres, le Wed 05 Sep 2012 17:06:00 +0200, a écrit : > On Sep 5, 2012, at 10:21 AM, Samuel Thibault wrote: > > > So ltdl does not help for that matter? > > No. It's not really an ltdl issue. ltdl is just a portable wrapper around > OS-specific dlopen-like mechan

Re: [hwloc-devel] [hwloc-svn] svn:hwloc r4815 - branches/components/src

2012-09-06 Thread Samuel Thibault
Jeff Squyres, le Thu 06 Sep 2012 15:46:29 +0200, a écrit : > On Sep 6, 2012, at 7:46 AM, Jeff Squyres wrote: > > (sorry; I forgot to ping Shiqing yesterday -- I just did so now to get a > > confirmation of what you found) > > > Shiqing confirms that DSOs are disabled by default on Windows.

Re: [hwloc-devel] merging plugins?

2012-09-24 Thread Samuel Thibault
Hello, Brice Goglin, le Mon 24 Sep 2012 22:04:14 +0200, a écrit : > 1) A rework of the backend infrastructure to make the core much more > readable (basically all changes in *.[ch] files). That looks nicer indeed. > 2) Plugin support One thing that doesn't seem implemented yet is to choose

Re: [hwloc-devel] merging plugins?

2012-09-25 Thread Samuel Thibault
Brice Goglin, le Tue 25 Sep 2012 07:41:48 +0200, a écrit : > * Your HWLOC_PLUGINS variable is not about loading plugins, it's about > enabling core components. It could also be to use another PCI detection plugin that libpci. Samuel

Re: [hwloc-devel] merging plugins?

2012-09-25 Thread Samuel Thibault
Brice Goglin, le Tue 25 Sep 2012 10:34:29 +0200, a écrit : > I am also going to add a hwloc_ prefix to plugin filenames because we > obviously can't create a libpci.so (libtool even warns about this). And it makes things clearer, I believe. > XML backends could be hwlocxml_ (not hwloc_xml_) to

Re: [hwloc-devel] merging plugins?

2012-09-25 Thread Samuel Thibault
Brice Goglin, le Tue 25 Sep 2012 11:08:04 +0200, a écrit : > >> We have the "core_xml" component (generic xml support) and "xml_libxml" > >> + "xml_nolibxml" backends behind that. I am fine with removing the > >> "core_" prefix, but I wonder if we should keep the "xml_" prefix for the > >> latter.

Re: [hwloc-devel] [hwloc-users] hwloc 1.5, freebsd and linux output on the same hardware

2012-10-11 Thread Samuel Thibault
Sebastian Kuzminsky wrote: > Maybe lstopo should expand its cpuset to be fully inclusive at startup? I'll > be happy to test patches if you want. Brice Goglin, le Thu 11 Oct 2012 18:13:53 +0200, a écrit : > Is the cpuset-modification a root-only operation on FreeBSD? If so lstopo > wouldn't be

Re: [hwloc-devel] plugins update

2012-11-08 Thread Samuel Thibault
Brice Goglin, le Thu 08 Nov 2012 13:48:39 +0100, a écrit : > Did anybody look at leastc at the documentation links below? Yep, it looks nice! Samuel

Re: [hwloc-devel] v1.7

2013-01-07 Thread Samuel Thibault
Brice Goglin, le Mon 31 Dec 2012 10:05:41 +0100, a écrit : > - They add OS devices such as opencl0p0, I see that platform 0 device 3 would be called opencl3p0. I find it counterintuitive, and would have rather called it opencl0d3, along the line of sda3, eth0:3, socket:2.core:0, etc. What do

Re: [hwloc-devel] v1.7

2013-01-07 Thread Samuel Thibault
Hello, Brice Goglin, le Mon 31 Dec 2012 10:05:41 +0100, a écrit : > + The HWLOC_COMPONENTS may now start with '^' to only define a list of > components to exclude. I'm finding it not intuitive and not generic enough, I'm wondering how that didn't affect Open-MPI, which as IUI uses this

Re: [hwloc-devel] v1.7

2013-01-07 Thread Samuel Thibault
Brice Goglin, le Mon 07 Jan 2013 17:33:47 +0100, a écrit : > Ideally, we could even have a OS device for each OpenCL platform, each > containing OS devices for devices of the platform. But I'd rather keep a > single level to match other OS devices. In most cases the platform object wouldn't bring

Re: [hwloc-devel] v1.7

2013-01-07 Thread Samuel Thibault
Jeff Squyres (jsquyres), le Mon 07 Jan 2013 19:19:15 +0100, a écrit : > On Jan 7, 2013, at 12:59 PM, Samuel Thibault <samuel.thiba...@inria.fr> > > Because I may not know *everything* that I want. Who knows what > > proprietary plugin I need to use to discover CPUs, while I k

<    1   2   3   4   5   >