Intel® oneAPI Math Kernel Library
Ask questions and share information with other developers who use Intel® Math Kernel Library.

dgelsd on AMD processor

jaewonj
Novice
392 Views
The dgelsd issue (http://software.intel.com/en-us/forums/showthread.php?t=61809) doesn't seem to be fully addressed.

I'm getting "MKL ERROR: Parameter 4 was incorrect on entry to DLASCL" when running dgelsd on AMD processor

$ cat /proc/cpuinfo
processor : 0
vendor_id : AuthenticAMD
cpu family : 15
model : 5
model name : AMD Opteron Processor 250
stepping : 10
cpu MHz : 1000.000
cache size : 1024 KB
fpu : yes
fpu_exception : yes
cpuid level : 1
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx mmxext lm 3dnowext 3dnow up
bogomips : 1996.13
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 40 bits physical, 48 bits virtual
power management: ts fid vid ttp

Could you please check this?

Thanks.
Jaewon


0 Kudos
5 Replies
Gennady_F_Intel
Moderator
392 Views
Jaewon,
what MKL version are you using?
--Gennady
0 Kudos
jaewonj
Novice
392 Views
Gennady,

It was 10.2.5. Would it make a difference with 10.3?

Jaewon



0 Kudos
Gennady_F_Intel
Moderator
392 Views

This is unexpected result behavior. we will check it on our side and will back asap.

--Gennady

0 Kudos
Michael_C_Intel4
Employee
392 Views

Hi Jaewon,

I've reproduced the issue.It will be fixed in the next MKL 10.2 Update. This concerns MKL 10.3.b too, and will be fixed in MKL 10.3.0 as well.

Michael.

0 Kudos
Gennady_F_Intel
Moderator
392 Views

Jaewon,

Could you please check if the probelm is still exists with latest MKL 10.2 Update 6 version released yesterday.

--Gennady

0 Kudos
Reply