Re: [hwloc-users] lstopo hangs for centos 7
Hi Brice, Thanks for the workaround -- it works very good. Attached please find the two output file after run hwloc-gather-cpuid. Let me after this is fixed! thanks, Jianjun On Sun, Jan 31, 2016 at 9:48 PM, Brice Goglin wrote: > Thanks for the debugging. I guess VMware doesn't properly emulate the > CPUID instruction. > > Please do: > 1) take a tarball from git master at > https://ci.inria.fr/hwloc/job/master-0-tarball/ and build it > 2) export HWLOC_COMPONENTS=-x86 in your terminal > 3) do utils/hwloc/hwloc-gather-cpuid > 4) tar cfj cpuid.tbz2 foo and send that cpuid.tbz2 > > Step (3) might do an infinite loop for the same reason, please replace > for(i=0; ; i++) { > with > for(i=0; i<10; i++) { > everywhere in utils/hwloc/hwloc-gather-cpuid.c > > This tarball will help me find what's buggy in VMware CPUID instruction. > > > In the meantime, you can fix your hwloc by exporting HWLOC_COMPONENTS=-x86 > in your environment. > > If somebody knows how do detect vmware by looking under /proc or /sys, we > could use that to automatically set that environment variable. > > thanks > Brice > > > > > > Le 01/02/2016 05:59, Jianjun Wen a écrit : > > I did a debug build. Found it loops forever in this loop in > topology-x86.c:404. > > > /* Get package/core/thread information from cpuid 0x0b >* (Intel x2APIC) >*/ > if (cpuid_type == intel && has_x2apic(features)) { > unsigned level, apic_nextshift, apic_number, apic_type, apic_id = 0, > apic_shift = 0, id; > for (level = 0; ; level++) { > ecx = level; > eax = 0x0b; > hwloc_x86_cpuid(&eax, &ebx, &ecx, &edx); > if (!eax && !ebx) > break; > } > > On Sun, Jan 31, 2016 at 8:30 PM, Christopher Samuel < > sam...@unimelb.edu.au> wrote: > >> On 01/02/16 15:09, Jianjun Wen wrote: >> >> > 0x77bce13c in look_proc () from /lib64/libhwloc.so.5 >> > >> > Always the same place. >> >> pstack on the process when stuck might give more of an insight as it >> should give more of a stack trace. >> >> Also running lstopo under strace should show what it is trying to do at >> that point. >> >> All the best, >> Chris >> -- >> Christopher SamuelSenior Systems Administrator >> VLSCI - Victorian Life Sciences Computation Initiative >> Email: sam...@unimelb.edu.au Phone: +61 (0)3 903 55545 >> http://www.vlsci.org.au/ http://twitter.com/vlsci >> >> ___ >> hwloc-users mailing list >> hwloc-us...@open-mpi.org >> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-users >> Link to this post: >> http://www.open-mpi.org/community/lists/hwloc-users/2016/01/1251.php >> > > > > -- > -Jianjun Wen > Wancube.com - 3D photography > Phone: 408 888 7023 > > > ___ > hwloc-users mailing listhwloc-us...@open-mpi.org > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-users > > Link to this post: > http://www.open-mpi.org/community/lists/hwloc-users/2016/01/1252.php > > > > ___ > hwloc-users mailing list > hwloc-us...@open-mpi.org > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-users > Link to this post: > http://www.open-mpi.org/community/lists/hwloc-users/2016/02/1254.php > -- -Jianjun Wen Wancube.com - 3D photography Phone: 408 888 7023 aaa.output Description: Binary data aaa.tar.bz2 Description: BZip2 compressed data
Re: [hwloc-users] lstopo hangs for centos 7
I did a debug build. Found it loops forever in this loop in topology-x86.c:404. /* Get package/core/thread information from cpuid 0x0b * (Intel x2APIC) */ if (cpuid_type == intel && has_x2apic(features)) { unsigned level, apic_nextshift, apic_number, apic_type, apic_id = 0, apic_shift = 0, id; for (level = 0; ; level++) { ecx = level; eax = 0x0b; hwloc_x86_cpuid(&eax, &ebx, &ecx, &edx); if (!eax && !ebx) break; } On Sun, Jan 31, 2016 at 8:30 PM, Christopher Samuel wrote: > On 01/02/16 15:09, Jianjun Wen wrote: > > > 0x77bce13c in look_proc () from /lib64/libhwloc.so.5 > > > > Always the same place. > > pstack on the process when stuck might give more of an insight as it > should give more of a stack trace. > > Also running lstopo under strace should show what it is trying to do at > that point. > > All the best, > Chris > -- > Christopher SamuelSenior Systems Administrator > VLSCI - Victorian Life Sciences Computation Initiative > Email: sam...@unimelb.edu.au Phone: +61 (0)3 903 55545 > http://www.vlsci.org.au/ http://twitter.com/vlsci > > ___ > hwloc-users mailing list > hwloc-us...@open-mpi.org > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-users > Link to this post: > http://www.open-mpi.org/community/lists/hwloc-users/2016/01/1251.php > -- -Jianjun Wen Wancube.com - 3D photography Phone: 408 888 7023
Re: [hwloc-users] lstopo hangs for centos 7
I just realize that yum on centos 7 use hwloc version 1.7. I Downloaded 1.11.2 version source, built and install. Still hangs. 0x77bcb32c in hwloc_x86_cpuid (edx=, ecx=, ebx=0x7fffdbec, eax=) at /home/wen/Downloads/hwloc-1.11.2/include/private/cpuid-x86.h:67 67 __asm__( On Sun, Jan 31, 2016 at 8:09 PM, Jianjun Wen wrote: > Hi Brice > Thanks for the reply. > I use yum install hwloc to install it. > The cpu usage is 100%. > I got this after Ctrl + C, and c, several times: > > 0x77bce13c in look_proc () from /lib64/libhwloc.so.5 > > Always the same place. > > On Sun, Jan 31, 2016 at 12:29 AM, Brice Goglin > wrote: > >> Hello >> >> Thanks for the report. I have never seen this issue. I have CentOS 7 VMs >> (kvm), lstopo works fine. Did you try this in similar VMs in the past? >> >> When you say "latest hwloc", do you mean "build latest tarball" (1.11.2) >> or "installed latest centos package" (1.7)? >> >> First thing to check: run lstopo, let it hang, and check under top >> whether it uses 100% CPU or 0% CPU (to see if that's an infinite loop or >> not). >> >> Then, run it under gdb: >> $ gdb lstopo >> Type 'r' and Enter >> When things hang, do ctrl-c >> Type "where" and send the output to us. >> >> If you got 100% in top above, you should do this multiple time. After >> "where", type 'c' to go back to the execution, ctrl+c again, "where" again >> and check whether the backtrace is similar. >> >> Brice >> >> >> >> >> Le 31/01/2016 04:48, Jianjun Wen a écrit : >> >> I installed the latest centos 7 (1151) on VM (vmware), then installed >> latest hwloc. >> lstopo command hangs. >> >> hwloc_topology_load() >> function call also hangs. >> >> Is this an know issue? How to find out what's wrong? >> >> thanks >> -- >> -Jianjun Wen >> Wancube.com - 3D photography >> Phone: 408 888 7023 >> >> >> ___ >> hwloc-users mailing listhwloc-us...@open-mpi.org >> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-users >> Link to this post: >> http://www.open-mpi.org/community/lists/hwloc-users/2016/01/1247.php >> >> >> >> ___ >> hwloc-users mailing list >> hwloc-us...@open-mpi.org >> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-users >> Link to this post: >> http://www.open-mpi.org/community/lists/hwloc-users/2016/01/1248.php >> > > > > -- > -Jianjun Wen > Wancube.com - 3D photography > Phone: 408 888 7023 > -- -Jianjun Wen Wancube.com - 3D photography Phone: 408 888 7023
Re: [hwloc-users] lstopo hangs for centos 7
Hi Brice Thanks for the reply. I use yum install hwloc to install it. The cpu usage is 100%. I got this after Ctrl + C, and c, several times: 0x77bce13c in look_proc () from /lib64/libhwloc.so.5 Always the same place. On Sun, Jan 31, 2016 at 12:29 AM, Brice Goglin wrote: > Hello > > Thanks for the report. I have never seen this issue. I have CentOS 7 VMs > (kvm), lstopo works fine. Did you try this in similar VMs in the past? > > When you say "latest hwloc", do you mean "build latest tarball" (1.11.2) > or "installed latest centos package" (1.7)? > > First thing to check: run lstopo, let it hang, and check under top whether > it uses 100% CPU or 0% CPU (to see if that's an infinite loop or not). > > Then, run it under gdb: > $ gdb lstopo > Type 'r' and Enter > When things hang, do ctrl-c > Type "where" and send the output to us. > > If you got 100% in top above, you should do this multiple time. After > "where", type 'c' to go back to the execution, ctrl+c again, "where" again > and check whether the backtrace is similar. > > Brice > > > > > Le 31/01/2016 04:48, Jianjun Wen a écrit : > > I installed the latest centos 7 (1151) on VM (vmware), then installed > latest hwloc. > lstopo command hangs. > > hwloc_topology_load() > function call also hangs. > > Is this an know issue? How to find out what's wrong? > > thanks > -- > -Jianjun Wen > Wancube.com - 3D photography > Phone: 408 888 7023 > > > ___ > hwloc-users mailing listhwloc-us...@open-mpi.org > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-users > Link to this post: > http://www.open-mpi.org/community/lists/hwloc-users/2016/01/1247.php > > > > _______ > hwloc-users mailing list > hwloc-us...@open-mpi.org > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/hwloc-users > Link to this post: > http://www.open-mpi.org/community/lists/hwloc-users/2016/01/1248.php > -- -Jianjun Wen Wancube.com - 3D photography Phone: 408 888 7023
[hwloc-users] lstopo hangs for centos 7
I installed the latest centos 7 (1151) on VM (vmware), then installed latest hwloc. lstopo command hangs. hwloc_topology_load() function call also hangs. Is this an know issue? How to find out what's wrong? thanks -- -Jianjun Wen Wancube.com - 3D photography Phone: 408 888 7023