----------------------------------------------------------------------------------------------------------
System: Intel Xeon E5430, IPP 6.1.2.051 EM64T, kernel 2.6.32
----------------------------------------------------------------------------------------------------------

----------------------------------------------------------------------------------------------------------
Issue: http://software.intel.com/en-us/forums/showthread.php?t=89013
----------------------------------------------------------------------------------------------------------

The test computes:
 accu = 3.653725000000000e+04;
 accu = accu + (3.360162734985352e+01)^2 + (4.114639663696289e+01)^2;
 which should be 39359.345316764694871 in double.

:~/DiFX-trunk-rfi/mpifxcorr/trunk/tests/ippsbugs> ./fma_test
Reference (double): 39359.345316764694871 + i*0.0                R1
Reference (cast):   39359.343750000000000 + i*0.0                R2
Reference (float):  39359.343750000000000 + i*0.0                R3
ippsAddProduct:     39359.347656250000000 + i*0.000000000000000  R4
ippsMul, ippsAdd_I: 39359.343750000000000 + i*0.000000000000000  R5

The 32-bit addproduct result does not match with 32-bit "float(a)*float(b)+float(c)"
nor does it match with 32-bit ippsAdd(ippsMul(a,b), c). I posted this on the Intel forum
but there was no really satisfying answer (only that ~1e-7 data precision should be expected,
and well, the relative erros (abs(R4-R1)/R1=5.94e-8) and (abs(R4-R1)/R1)=3.99e-8 are indeed 
within the ~1e-7 precision ; yet it is still surprising to get a different result...)

----------------------------------------------------------------------------------------------------------
System details:
----------------------------------------------------------------------------------------------------------

:~/DiFX-trunk-rfi/mpifxcorr/trunk/tests/ippsbugs> uname -a
Linux fxmanager 2.6.32-71.el6.x86_64 #1 SMP Tue Nov 23 06:49:13 CST 2010 x86_64 x86_64 x86_64 GNU/Linux

:~/DiFX-trunk-rfi/mpifxcorr/trunk/tests/ippsbugs> export|grep -i IPPROOT
declare -x IPPROOT="/cluster/intel/ipp/6.1.2.051/em64t"

:~/DiFX-trunk-rfi/mpifxcorr/trunk/tests/ippsbugs> cat /proc/cpuinfo
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 23
model name      : Intel(R) Xeon(R) CPU           E5430  @ 2.66GHz
stepping        : 6
cpu MHz         : 2000.000
cache size      : 6144 KB
physical id     : 0
siblings        : 4
core id         : 0
cpu cores       : 4
apicid          : 0
initial apicid  : 0
fpu             : yes
fpu_exception   : yes
cpuid level     : 10
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic mtrr pge mca cmov pat pse36 
                  clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx lm constant_tsc 
                  arch_perfmon pebs bts rep_good aperfmperf pni dtes64 monitor ds_cpl vmx est 
                  tm2 ssse3 cx16 xtpr pdcm dca sse4_1 lahf_lm tpr_shadow vnmi flexpriority
bogomips        : 5333.88
clflush size    : 64
cache_alignment : 64
address sizes   : 38 bits physical, 48 bits virtual
power management: