x
000
{SY}{SS}{EQ}{JP}{REPEAT}{NAME}
xxxxx
000
000
000
xx0
0000
xx0
0110
xx0
0100
xx0
0111
xx0
1000
xx0
1001
xx0
1010
xx0
1011
SHader Prologue End
xx1
1000
The predt, predf, and prede instructions are used to enable
"predicated execution" for a region of code. During predicated
execution, the execution mask is combined with the "predication
mask" to determine the active fibers. So this is similar to
branching with the exception that inactive fibers keep executing
instead of being parked; the instructions they execute simply
have no effect. Since no branch stack is necessary to support
predication, it is more performant for divergent branches.
More concretely, the hardware seems to keep track of a
predication mask and a predication mode. The mode can be "none"
(no predication, the default, and set by prede), "true" (set by
predt), or "false" (set by predf). The final execution mask is
determined by and'ing with ~0 (none mode), the predication mask
(true mode), or its negation (false mode).
Predication is enabled by predt or predf which take the value of
p0.x of all fibers (not only the active ones) and store them in
the predication mask. Then the mode is set to true or false.
Executing prede disables predication by setting the mode to
none.
One peculiarity, and the reason the execution mode state is
necessary, is that any instruction marked with (jp) will also
update the execution mask. The current mode will be kept in that
case.
xx1
1101
xx1
1110
xx1
1111
{SY}{SS}{EQ}{JP}{NAME} {INV1}p0.{COMP1}
00000
000
000
Invert source condition
Predicate register (p0.c) component for source
xx0
0101
{SY}{SS}{JP}{NAME} #{IMMED}
xxxxx
xxx
xxx
xxx
xx0
0010
xx0
0011
xx1
0000
Perform a jump for all fibers in the first cluster with any active
fibers, except for the last fiber in the cluster.
While there is a separate field for CLUSTER_SIZE its value does
not change the behaviour in any observable way, it behaves as if
CLUSTER_SIZE is always 8.
{SY}{SS}{JP}{NAME}.w{CLUSTER_SIZE} #{IMMED}
2ULL << {W}
xxxxx
xxx
xxx
xx1
0100
xx1
0101
SHader Prologue Start
xx1
0111
xx0
0001
{SY}{SS}{EQ}{JP}{NAME}.{INDEX} #{IMMED}
011
xxx
xxx
xxxxx
110
xxx
xxx
{SY}{SS}{EQ}{JP}{NAME} {INV1}p0.{COMP1}, #{IMMED}
xxxxx
xxx
Invert source condition
Predicate register (p0.c) component for source
000
100
101
{SY}{SS}{EQ}{JP}{NAME} {INV1}p0.{COMP1}, {INV2}p0.{COMP2}, #{IMMED}
xxxxx
Invert source 2 condition
Predicate register (p0.c) component for source 2
Invert source 1 condition
Predicate register (p0.c) component for source 1
001
010