Professors

micctrl --resetconfig on systems with many users

Hello,

I find that micctrl --resetconfig takes a long time on our system (>15 minutes). I suspect the reason is that we have thousands of user accounts in our cluster environment and the command spends time configuring each one. After the command is run, the /etc/passwd file for the mic card is populated with every user and we must remove them all so that they can be added manually on demand by the prologue script of our batch system. I cannot find any mention in the documentation of a way to prevent this command from configuring the mic cards for our entire list of users.

Problem with memory transfers

Hello,

I read the memory from the XeonPhi by using the MicAccessAPI and boot an own 32bit kernel on the card. Now I had problems, seeing my debug data on the host. On the XeonPhi, I write directly to some memory like this (in 32bit code):
c6 05 05 80 0b 00 44     movb   $0x44,0xb8005
But the value 0x44 does not become visible with the MicAccessAPI. It seems like the data is staying in the cache or the core hangs. wbinvd and clflush after the mov instruction did not help.

Gather using _mm512_i32logather_pd

I seem to be having problems getting _mm512_i32logather_pd to work correctly.

 __m512d simd_rhs = _mm512_i32loextgather_pd(simd_ind1, data, _MM_UPCONV_PD_NONE, scale, _MM_HINT_NT); 

Firstly, if the high elements of simd_ind1 are anything other than 0, the code segfaults. So when I set those 8 elements to 0 and run, all elements in simd_rhs are the value at the base address , i.e. *data. In other words, the indices in simd_ind1 are disregarded and I get back a vector filled with a value at the base addres I supply.

offload error: cannot load library to the device

I am trying to program the Xeon Phi in offload mode. The program below should just construct objects from the standard library on the card, but linking fails with the following error:

On the remote process, dlopen() failed. The error message sent back from the sink is /var/volatile/tmp/coi_procs/1/27843/load_lib/icpcoutdlt28i: undefined symbol: _ZNSt6vectorIdSaIdEEC1EmRKdRKS0_

Intel® Data Protection Technology with Secure Key in the Virtualized Environment

The digital random number generator (DRNG) behind Intel® Data Protection Technology with Secure Key provides high-quality random numbers that are accessible via the CPU instruction RDRAND. This easy-to-use feature is of great benefit to virtualized environments where limited system entropy must be divided up among a large number of virtual machines.

  • Developers
  • Professors
  • Students
  • Linux*
  • Unix*
  • Server
  • Intel® Xeon® E5-2650 v2
  • Virtual Machine
  • John Mechalas
  • Intel® Secure Key
  • Jeff's Notebook: Online Banking with Two-Factor Authentication Security

    As online banking has become a predominate and normal way of conducting much of our personal banking, security has become ever more important.  We want to know that our bank is protecting our money, our accounts and our financial information from those “bad guys” out there.  One of the key methods that banks and other business use is two-factor authentication for a client to use when they log into the website and their accounts.  That usually means 1) something you know, ie., your user name and password and 2) something have or can provide, ie.

    Subscribe to Professors