Open MPI logo

Hardware Locality Development Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Hardware Locality Development mailing list

Subject: [hwloc-devel] pcilib error messages w/ rhl8 and hwloc-1.3.1rc1
From: Paul H. Hargrove (PHHargrove_at_[hidden])
Date: 2011-12-14 14:45:21


Brice has been helping me get hwloc working on an old Red Hat Linux 8
system.
We've appear to have gotten past the sched_setaffinity() prototype
conflicts, and I can build hwloc.
However, when I run "make check" there are 640 instances of the
following from test-hwloc-calc.sh.
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64

Here is one manual run of hwloc-calc
> $ ./utils/hwloc-calc --if synthetic --input "node:4 core:4 pu:4"
> pu:22-47 --largest
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> pcilib: proc_read: tried to read 256 bytes at 0, but got only 64
> PU:22 PU:23 Core:6 Core:7 NUMANode:2

I can run "lspci -vv" to get plenty of correct information and no such
messages.
So, this is not just a case of total libpci breakage.

Some system info:
> $ cat /etc/redhat-release
> Red Hat Linux release 8.0 (Psyche)
> $ uname -a
> Linux [hostname] 2.4.21-60.ELsmp #1 SMP Fri Aug 28 06:45:10 EDT 2009
> i686 i686 i386 GNU/Linux
> $ gcc --version | head -1
> gcc (GCC) 3.2 20020903 (Red Hat Linux 8.0 3.2-7)
> $ rpm -qf /usr/include/pci/pci.h
> pciutils-devel-2.1.10-2
> $ rpm -qf /lib/libc.so.6
> glibc-2.3.2-4.80.8

FWIW:
This system is running a kernel from RHEL due to a lack of security
updates to RHL8.
So, it *is* possible that this is a mismatch between the kernel's /proc
and libpci's expectations.
If that is the case, of course, then hwloc can't possibly do anything
about it.
However "lspci -vv" does work, making me doubt this option.

Let me know what else might be needed to address this problem (or at
least determine if it is even a hwloc problem or not).

-Paul

-- 
Paul H. Hargrove                          PHHargrove_at_[hidden]
Future Technologies Group
HPC Research Department                   Tel: +1-510-495-2352
Lawrence Berkeley National Laboratory     Fax: +1-510-486-6900