pe_macro(1)
•
div1
: 64-Bit Signed Integer DIVIDE Test — This test enables all the PEs,
then performs a series of 64-bit divisions. After each division, the test inverts
all the bits of the answer and saves this separately.
If an error occurs, the test prints an error message which gives the attempted
division, the expected result, the actual result and whether or not this is the
bit-inverted version of the answer. Typically, if the answer is wrong, the bit
inversion is also wrong, and the test prints a separate message for each.
•
divu0
: 64-Bit Unsigned Integer DIVIDE Test — This test enables all the PEs,
then performs a series of 64-bit unsigned divisions.
If an error occurs, the test prints an error message which gives the attempted
division, the expected result and the actual result.
•
ebit1
: E-Bit Operation Test — This test verifies that when the e-bit is turned
on, data transfers to PReg can take place, and that when the e-bit is turned
off, data transfers to PReg are blocked.
If it detects an error, this test prints an error message which gives the
attempted operation, the expected result and the actual result.
•
fadd1
: 64-Bit Floating Point ADD Test — This tests performs two simple
64-bit floating point additions:
1.0 + 2.0 = ?
and
2.0 + 1.0 = ?
.
If it detects an error, this test prints an error message which gives the
attempted operation, the expected result and the actual result. Both the
expected and the actual result is expressed in VAX floating point g-format.
This is a 64-bit integer whose fields have the following meaning:
0
4
8
12
16
20
24
28
32
36
40
44
48
52
56
60
64
bit
Fraction 1
Fraction 2
Fraction 3
Fraction 4
(Least Significant)
(Most Significant)
Sign bit
Exponent (in
excess of 1024)
MPP_FIG_277
•
fmult1
: 64-Bit Floating Point MULTIPLY Test — This test performs a series
of 64-bit floating point multiplication operations.
It enables all the PEs, then performs a series of 64-bit multiplications. After
each operation, the test inverts all the bits of the answer and saves this
separately.
If an error occurs, the test prints an error message which gives the attempted
operation, the expected result, the actual result and whether or not this is the
bit-inverted version of the answer. Typically, if the answer is wrong, the bit
inversion is also wrong, and the test prints a separate message for each.
Both the expected and the actual result is expressed in VAX floating point
g-format. This is a 64-bit integer whose fields have the following meaning:
B–30 Data Parallel Unit Reference Pages
